Scirus Is a Free, Science-Speciﬁc Search Engine

104 ISSN 1096-7192 S Volume 104, Supplement 2011 Supplement to Molecular Genetics and Metabolism MOLECULAR GENETICS AND METABOLISM Scirus is a free, science-speciﬁc search engine. olecular enetics

Try Scirus at Vol. S100 104S (2011) S1– and etabolism Editor-in-Chief www.Scirus.com Edward R.B. McCabe Management of Phenylketonuria: Current Position

Guest Editors Nenad Blau, Anita MacDonald, and Francjan van Spronsen

FOR ELECTRONIC USE ONLY ELSEVIER

The ofﬁ cial journal of the Available online at www.sciencedirect.com Society for Inherited Metabolic Disorders Molecular Genetics and Metabolism Editor-in-Chief Edward R. B. McCabe University of Colorado, Denver, Colorado / Forward thinking for professionals Editorial Board

Brage Storstein Andresen Gregory A. Grabowski William B. Rizzo University of Southern Denmark Children’s Hospital Medical Center University of Nebraska Odense, Denmark Cincinnati, Ohio Medical Center Rivka Beitner Cheryl R. Greenberg Omaha, Nebraska Bar-Ilan University Children’s Hospital Charles R. Roe Ramat Gan, Israel Winnipeg, Canada Baylor University Medical Center John J. Hopwood Mike Bennett Dallas, Texas Adelaide Children’s Hospital The Children’s Hospital of Philadelphia David S. Rosenblatt North Adelaide, Australia Philadelphia, Pennsylvania McGill University Health Center Jeffrey W. Innis Announcing Montreal, Canada Nenad Blau University of Michigan University Children’s Hospital Medical School Rima Rozen Zurich, Switzerland Montreal Children’s Hospital Endocrinology Advance.com Ann Arbor, Michigan Stephen D. Cederbaum Brendan Lee Quebec, Canada UCLA Mental Retardation Baylor College of Medicine Charles R. Scriver Research Center Houston, Texas McGill University Los Angeles, California Yoichi Matsubara Montreal, Canada Where Endocrinology Pinchas Cohen Tohoku University School of Medicine Ellen Sidransky UCLA School of Medicine Sendai, Japan National Human Genome Professionals Search for Los Angeles, California Roderick R. McInnes Research Institute Jewish General Hospital Robert J. Desnick Bethesda, Maryland Montréal, Quebec, Canada • The latest Endocrinology News Mt. Sinai School of Medicine Flemming Skovby Patricia McLean New York, New York Righospitalet • Articles in Press from 30+ Elsevier journals University College London Copenhagen, Denmark Orly Elpeleg Medical School • Search across 500+ health sciences journals Hebrew University Medical Center London, United Kingdom Mendel Tuchman Jerusalem, Israel Walter L. Miller Children’s National • Selected endocrinology content from Fumio Endo University of California Medical Center Washington, D.C. Kumamoto University School San Francisco, California TheLancet.com of Medicine Maximilian Muenke Jerry Vockley • Meetings and Events Kumamoto, Japan National Human Genome University of Pittsburgh Gregory M. Enns Research Institute Pittsburgh, Pennsylvania • And more... Stanford University Medical Center Bethesda, Maryland Ronald J.A. Wanders Stanford, California Arnold Munnich University Hospital Amsterdam Yoshikatsu Eto Hosp Enfants Malades Amsterdam, The Netherlands Tokyo Jikei University School of Medicine Paris, France David A. Wenger Torayuki Okuyama Tokyo, Japan Jefferson Medical College National Center for Child Health Philadelphia, Pennsylvania William A. Gahl and Development National Human Genome Research Tokyo, Japan Barry Wolf Institute Ferdinando Palmieri Henry Ford Hospital Bethesda, Maryland Universita di Bari Detroit, Michigan Stephen I. Goodman Bari, Italy Philip A. Wood University of Colorado Piero Rinaldo Sanford/Burnham Medical Research Health Science Center Mayo Clinic Institute at Lake Nona Denver, Colorado Rochester, Minnesota Orlando, Florida Managing Editor Founding Editor Linda L. McCabe FORS. P. Bessman ELECTRONIC USE ONLY University of Colorado, Denver, Colorado One click access

Cover photo. Dr. Richard Koch (1921–2011) receiving the 2009 Asbjørn Følling Award at the EPG Symposium in Barcelona, Spain. to essential endocrinology resources. Visit www.EndocrinologyAdvance.com Management of Phenylketonuria: Current Position Guest Editors: Nenad Blau, Anita MacDonald, and Francjan van Spronsen

Dedicated in memory of Dr. Richard Koch (1921–2011) The European Phenylketonuria Group

Volume 104, Supplement 2011

Serono Symposia International Foundation (SSIF) is an independent nonprofit organization dedicated to the Continuing Medical Education (CME) of scientists, physicians, nurses, pharmacists and other healthcare professionals through state-of-the-art clinical and scientific activities designed to enhance patient care and improve healthcare outcomes.

All Serono Symposia International Foundation programs are made possible thanks to unrestricted educational grants received from Centre d’Esclerosi Multiple de Catalunya, ComtecMed, Congrex Sweden, Congrex Switzerland, Cryo-Save, Datanalysis, Esaote, European Society of Endocrinology, Fondazione Humanitas, Fundación IVI, ISFP InternationalFOR SocietyELECTRONIC for Fertility Preservation, ISMH InternationalUSE SocietyONLY of Men’s Health, K.I.T.E., Merck Serono, Sanofi-Aventis, University of Catania, Vall d’Hebron University Hospital Molecular Genetics and Metabolism

Volume 104, Supplement 2011

This journal and the individual contributions contained in it are protected under copyright by Elsevier Inc., and the following terms and conditions apply to their use:

Photocopying

Single photocopies of single articles may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee are required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for nonprofit educational classroom use. For information on how to seek permission visit www.elsevier.com/permissions or call: (+44) 1865 843830 (UK)/(+1) 215 239 3804 (USA).

Derivative Works

Subscribers may reproduce tables of contents or prepare lists of articles including abstracts for internal circulation within their institutions. Permission of the Publisher is required for resale or distribution outside the institution. Permission of the Publisher is required for all other derivative works, including compilations and translations (please consult www.elsevier.com/permissions).

Electronic Storage or Usage

Permission of the Publisher is required to store or use electronically any material contained in this journal, including any article or part of an article (please consult www.elsevier.com/permissions). Except as outlined above, no part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without prior written permission of the Publisher.

Notice

No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of product liability, negligence, or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. Although all advertising material is expected to conform to ethical (medical) standards, inclusion in this publication does not constitute a guarantee or endorsement of the quality or value of such product or of the claims made of it by its manufacturer.

Funding Body Agreements and Policies

Elsevier has established agreements and developed policies to allow authors whose articles appear in journals published by Elsevier, to comply with potential manuscript archiving requirements as specified as conditions of their grant awards. To learn more about existing agreements and policies please visit http://www. elsevier.com/fundingbodies.

Publication information: Molecular Genetics and Metabolism (ISSN 1096-7192) is published monthly by Elsevier (Radarweg 29,1043 NX Amsterdam, The Netherlands). Further information on this journal is available from the Publisher or from the Elsevier Customer Service Department nearest you or from this journal’s website (http://www.elsevier.com/locate/ymgme). Information on other Elsevier products is available through Elsevier’s website (http://www.elsevier.com). USA POSTMASTER: Send change of address to Molecular Genetics and Metabolism, Elsevier Customer Service Department, 3251 Riverport Lane, Maryland Heights, MO 63043, USA.

Orders, claims, and journal inquiries: Please contact the Elsevier Customer Service Department nearest you: St. Louis: Elsevier Customer Service Department, 3251 Riverport Lane, Maryland Heights, MO 63043, USA; phone: (877) 8397126 [toll free within the USA ]; (+1) (314) 4478878 [outside the USA ]; fax: (+1) (314) 4478077; e-mail: [email protected]. Oxford: Elsevier Customer Service Department, The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK; phone: (+44) (1865) 843434; fax: (+44) (1865) 843970; e-mail: [email protected]. Tokyo: Elsevier Customer Service Department, 4F Higashi-Azabu, 1-Chome Bldg,1-9-15 Higashi-Azabu, Minato-ku, Tokyo 106-0044, Japan; phone: (+81) (3) 5561 5037; fax: (+81) (3) 5561 5047; e-mail: [email protected]. Singapore: Elsevier Customer Service Department, 3 Killiney Road, #08-01 Winsland House I, Singapore 239519; phone: (+65) 63490222; fax: (+65) 67331510; e-mail: [email protected].

Author inquiries: For inquiries relating to the submission of articles (including electronic submission where available), please visit this journal’s homepage at http://www.elsevier.com/locate/ymgme. You can track accepted articles at http://www.elsevier.com/trackarticle and set up e-mail alerts to inform you of when an article’s status has changed. Also accessible from here is information on copyright, frequently asked questions, and more. For detailed instructions on the preparation of electronic artwork, please visit http://www.elsevier.com/artworkinstructions. Contact details for questions arising after acceptance of an article, especially those relating to proofs, will be provided by the publisher.

Advertising information: If you are interested in advertising or other commercial opportunities please e-mail [email protected] and your enquiry will be passed to the correct person who will respond to you within 48 hours. FOR ELECTRONICPrinted in the United States of America USE ONLY The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper)

Sponsored supplements and/or commercial reprints: For more information, please contact Elsevier Life Sciences Commercial Sales, Radarweg 29, 1043 NX Amsterdam, The Netherlands; phone: (+31) (20) 4852939/2059; e-mail: [email protected]. Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Volume 104, Supplement 2011

Management of Phenylketonuria: Current Position Guest Editors: Nenad Blau, Anita MacDonald, and Francjan van Spronsen

CONTENTS

Abstract/indexed in: BIOSIS, Chemical Abstracts, Current Contents/Life Sciences, EMBASE, EMBiology, MEDLINER, and Science Citation Index. Also covered in the abstract and citation database SciVerse SCOPUSR. Full text available on SciVerse ScienceDirectR

S1 Editorial Nenad Blau, Anita MacDonald, Francjan van Spronsen

MINIREVIEWS

S2 Diagnosis, classification, and genetics of phenylketonuria and tetrahydrobiopterin (BH4) deficiencies Nenad Blau, Julia B. Hennermann, Ulrich Langenbeck, Uta Lichter-Konecki S10 Nutrition in phenylketonuria A. MacDonald, J.C. Rocha, M. van Rijn, F. Feillet S19 Up to date knowledge on different treatment strategies for phenylketonuria Amaya Bélanger-Quintana, Alberto Burlina, Cary O. Harding, Ania C. Muntau S26 Adult phenylketonuria outcome and management F. Trefz, F. Maillot, K. Motzfeldt, M. Schwarz S31 Follow up of phenylketonuria patients M. Demirkol, M. Giżewska, M. Giovannini, J. Walter S40 The G46S-hPAH mutant protein: A model to study the rescue of aggregation-prone PKU mutations by chaperones João Leandro, Jaakko Saraste, Paula Leandro, Torgeir Flatmark S45 Cognitive, neurophysiological, neurological and psychosocial outcomes in early-treated PKU-patients: A start toward standardized outcome measurement across development F.J. van Spronsen, S.C.J. Huijbregts, A.M. Bosch, V. Leuzzi fi S52 VitaminFOR B12 de ciency ELECTRONIC and phenylketonuria USE ONLY John H. Walter

REGULAR ARTICLES

S55 Speciﬁc prebiotics in a formula for infants with Phenylketonuria Anita MacDonald, Barbara Cochrane, Harm Wopereis, Nik Loveridge S60 The 48-hour tetrahydrobiopterin loading test in patients with phenylketonuria: Evaluation of protocol and inﬂuence of baseline phenylalanine concentration K.Anjema,G.Venema,F.C.Hofstede,E.C.CarbasiusWeber,A.M.Bosch,N.M.TerHorst,C.E.M.Hollak, C.F. Jonkers, M.E. Rubio–Gozalbo, E.M.C. van der Ploeg, M.C. de Vries, R.G. Janssen-Regelink, M.C.H. Janssen, H. Zweers-van Essen, C.C.A. Boelen, N.A.P. van der Herberg-van de Wetering, M.R. Heiner-Fokkema, M. van Rijn, F.J. van Spronsen S64 Does a lower carbohydrate protein substitute impact on blood phenylalanine control, growth and appetite in children with PKU? Hulya Gokmen-Ozel, Carol Ferguson, Sharon Evans, Anne Daly, Anita MacDonald S68 Diurnal variations in blood phenylalanine of PKU infants under different feeding regimes MargreetvanRijn,MariekeHoeksma,PieterJ.J.Sauer,PimModderman,Dirk-JanReijngoud, Francjan J. van Spronsen S73 Neurological complications and behavioral problems in patients with phenylketonuria in a Follow-up Unit María J. González, Alfonso P. Gutiérrez, Rosa Gassió, María E. Fusté, María A. Vilaseca, Jaume Campistol

S80 Optimized loading test to evaluate responsiveness to tetrahydrobiopterin (BH4) in Brazilian patients with phenylalanine hydroxylase deﬁciency Tatiéle Nalin, Ingrid Dalira Schweigert Perry, Angela Sitta, Carmen Regla Vargas, Maria Luiza Saraiva-Pereira, Roberto Giugliani, Nenad Blau, Ida Vanessa Doederlein Schwartz

S86 Phenylalanine hydroxylase deﬁciency: Molecular epidemiology and predictable BH4-responsiveness in South Portugal PKU patients Isabel Rivera, Dina Mendes, Ângela Afonso, Madalena Barroso, Ruben Ramos, Patrícia Janeiro, Anabela Oliveira, Ana Gaspar, Isabel Tavares de Almeida S93 The spectrum of phenylketonuria genotypes in the Armenian population: Identiﬁcation of three novel mutant PAH alleles Natella Kostandyan, Corinne Britschgi, Albert Matevosyan, Alvina Oganezova, Anahit Davtyan, Nenad Blau, Beat Steinmann, Beat Thöny

BRIEF COMMUNICATION

S97 Oxidative stress in phenylketonuric patients C.R. Vargas, M. Wajner, A. Sitta

FOR ELECTRONIC USE ONLY

Language services. Authors who require information about language editing and copyediting services pre- and post-submission please visit http://www.elsevier.com/languageediting or our customer support site at http://epsupport.elsevier.com

For a complete Guide for Authors, please go to http://www/elsevier.com/locate/ymgme

Molecular Genetics and Metabolism has no page charges Molecular Genetics and Metabolism 104 (2011) S1

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Editorial

There is no doubt that the early identification of PKU and prompt on PKU. Highlights of the symposium were the 2011 Asbjørn Følling and continuous intervention prevents mental retardation in most lecture and award presented to Professor Nenad Blau (Zürich, Swit- patients. Dietary phenylalanine restriction remains the mainstay of zerland) for his lifetime work on PKU and BH4 and the SSIF prize treatment, but PKU is an active area of research and new treatment for the best free oral presentation presented to Karen Anjema options are emerging that may lessen the need for stringent dietary (Groningen, The Netherlands) and her colleagues about the first na- measures for patients and their families. In PKU, the recent introduc- tionwide experiences with the sapropterin loading test to discrimi- tion of sapropterin dihydrochloride (tetrahydrobiopterin; BH4) into nate BH4 responsive versus unresponsiveness (see page S60–S63, therapeutic use is likely to ease the burden of dietary care and further this issue). improve outcome in a subset of responsive patients with PKU. There We would like to thank the Serono Symposia International Foun- is also an emerging need for evidence-based international guidelines dation for supporting this event. We believe that specialists in pediat- that provide agreement concerning treatment initiation and determi- ric and adult inherited metabolic diseases, dietitians, nutritionists, nation of the response to therapy with BH4 both by itself and with clinical biochemists, experts in genetics, basic scientists, and profes- dietary treatment. Consensus is also required on target blood Phe sionals within the area of public health have benefited from this concentrations, and clarity on optimized treatment protocols for the Symposium. management of PKU and HPA. For this reason the European Phenylke- tonuria Group (EPG), continuing their series of successful meetings, organized in Lisbon, Portugal the 2011 Annual Symposium dedicated Nenad Blau to PKU. The meeting aims were to review the most important out- Division of Clinical Chemistry and Biochemistry, comes of research in this field and to provide a forum to discuss prac- University Children's Hospital, Steinwiesstrasse 75, tices that enable optimal patient management in clinical practice. As a CH-8032 Zürich, Switzerland consequence of a number of facilitated and structured workshops six Zurich Center for Integrative Human Physiology (ZIHP), Switzerland position papers are included in this Supplement issue of the Molecu- lar Genetics and Metabolism on the following topics: “Diagnosis, clas- Anita MacDonald sification, and genetics of phenylketonuria and tetrahydrobiopterin Birmingham Children's Hospital, Steelhouse Lane, deficiencies”, “Nutrition in phenylketonuria”, “Adult phenylketonuria Birmingham, B4 6NH, UK outcome and management”, “Up to date knowledge on different treatment strategies for phenylketonuria”, "Cognitive, neurophysio- Francjan van Spronsen logical, neurological and psychosocial outcomes in early-treated Section of Metabolic Diseases, Beatrix Children's Hospital, PKU-patients"; and “Follow up of phenylketonuria patients”. These University Medical Center of Groningen, University of Groningen, documents should serve as a basis for a future consensus conference Groningen, The Netherlands

FOR ELECTRONIC USE ONLY

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Minireview Diagnosis, classiﬁcation, and genetics of phenylketonuria and tetrahydrobiopterin (BH4) deﬁciencies

Nenad Blau a,b,c,⁎, Julia B. Hennermann d, Ulrich Langenbeck e, Uta Lichter-Konecki f a University Children's Hospital, Zürich, Switzerland b Zürich Center for Integrative Human Physiology (ZIHP), Zürich, Switzerland c Research Center for Children (RCC), Zürich, Switzerland d Department of Pediatrics, Charité Universitätsmedizin Berlin, Germany e Institute of Human Genetics, University Hospital, Frankfurt/Main, Germany f Center for Neuroscience and Behavioral Medicine, Division of Genetics & Metabolism, Children's National Medical Center, Department of Pediatrics, George Washington University, Medical Center, Washington, DC, USA article info abstract

Article history: This article summarizes the present knowledge, recent developments, and common pitfalls in the diagnosis, Received 8 July 2011 classification, and genetics of hyperphenylalaninemia, including tetrahydrobiopterin (BH4) deficiency. It is a Received in revised form 17 August 2011 product of the recent workshop organized by the European Phenylketonuria Group in March 2011 in Lisbon, Accepted 17 August 2011 Portugal. Results of the workshop demonstrate that following newborn screening for phenylketonuria (PKU), Available online 26 August 2011 using tandem mass-spectrometry, every newborn with even slightly elevated blood phenylalanine (Phe) fi Keywords: levels needs to be screened for BH4 de ciency. Dried blood spots are the best sample for the simultaneous Phenylketonuria measurement of amino acids (phenylalanine and tyrosine), pterins (neopterin and biopterin), and dihydrop- PKU teridine reductase activity from a single specimen. Following diagnosis, the patient's phenotype and individ- BH4 ually tailored treatment should be established as soon as possible. Not only blood Phe levels, but also daily Tetrahydrobiopterin tolerance for dietary Phe and potential responsiveness to BH4 are part of the investigations. Efficiency testing with synthetic BH4 (sapropterin dihydrochloride) over several weeks should follow the initial 24–48-hour screening test with 20 mg/kg/day BH4. The specific genotype, i.e. the combination of both PAH alleles of the patient, helps or facilitates to determine both the biochemical phenotype (severity of PKU) and the responsiveness to BH4. The rate of Phe metabolic disposal after Phe challenge may be an additional useful tool in the interpretation of phenotype–genotype correlation. © 2011 Elsevier Inc. All rights reserved.

Contents

1. Diagnosis ...... S3 1.1. Newborn screening ...... S3 1.2. Differential diagnosis ...... S3 1.3. BH4 loading test ...... S3 1.4. Cerebrospinal ﬂuid investigation ...... S4 2. Classiﬁcation ...... S4 2.1. Phenylalanine loading test ...... S4 2.2. Phenotypes ...... S5 2.3. Blood phenylalanine ...... S5 2.4. PhenylalanineFOR tolerance ELECTRONIC...... USE ONLY S5 2.5. Clinical course ...... S5 2.6. Software and phenylalanine home monitoring device ...... S5

Abbreviations: BH4, tetrahydrobiopterin; CNS, central nervous system; DHPR, dihydropteridine reductase; DBS, dried blood spot; HPA, hyperphenylalaninemia; KOUT, 1st order rate of metabolic disposal; MHP, mild HPA; PAH, phenylalanine hydroxylase; PBW, percent body weight; PKU, phenylketonuria; PROT, net protein synthesis; TMS, tandem mass- spectrometry. ⁎ Corresponding author at: University Children's Hospital, Steinwiesstrasse 75 8032 Zürich, Switzerland. E-mail addresses: [email protected] (N. Blau), [email protected] (J.B. Hennermann), [email protected] (U. Langenbeck), [email protected] (U. Lichter-Konecki).

3. Genetics ...... S6 3.1. PAH mutations and PKU genotypes (incl. databases) ...... S6 3.2. Phenotype–genotype correlation ...... S6 3.3. BH4-responsive mutations and genotypes ...... S7 Acknowledgments...... S7 References ...... S8

1. Diagnosis families, e.g. Turkey or Saudi Arabia [8]. BH4 deficiencies are more severe than PKU with regard to their response to therapy and treatment 1.1. Newborn screening is substantially different. Low-Phe diet is not effective and early substitution with dopamine and serotonin precursors, as well as with the Phenylketonuria (PKU) is identified through national newborn synthetic BH4 (sapropterin dihydrochloride) is crucial for a good out- screening programs [1]. The first efficient test for hyperphenylalani- come. Analysis of DBS or urine for neopterin and biopterin and mea- nemia (HPA) was a bacterial inhibition assay developed by Robert surement of dihydropteridine reductase (DHPR) activity in the DBS is Guthrie [2]. The test was based on Bacillus subtilis, which requires essential for the exact diagnosis and should be performed as early as phenylalanine (Phe) for growth. The Guthrie test was very useful possible. A BH4 loading test and measurement of neurotransmitter for mass screening as the dried blood spot (DBS) can be obtained in metabolites, pterins, and folates in cerebrospinal fluid add further im- the hospital or a doctor's office using a standardized filter paper portant information about the severity of the disease [9]. (“Guthrie card”) and mailed to reference laboratories in an envelope. In patients with BH4 deficiency, the pattern of pterins is identical Tandem mass-spectrometry (TMS) was developed as a fast method in blood, urine, and CSF. The use of DBS on filter paper (Guthrie card) for achieving reliable and quantitative determination of concentra- is, however, more practical and allows measurement of pterins, DHPR tions of amino acids in small volumes of blood or plasma [3]. This activity, and amino acids from a single specimen [10]. It is important method produces a lower rate of false positive results, by measuring to know that patients with classic PKU excrete more pterins in urine levels of both Phe and Tyr and providing the Phe/Tyr ratio, and thus compared with healthy controls and the amount of excreted metabo- requires fewer resources to follow up such cases. In addition, other in- lites is directly proportional to blood Phe levels. Diseases that cause born errors of metabolism can be identified simultaneously. activation of the immune system (elevated neopterin), and antican- All infants should be screened for PKU within the first days of life, cer therapy or rheumatic disease therapy with methotrexate (inhibi- in order to allow timely dietary intervention to protect children with tion of DHPR), may interfere with the analytic procedures. Some PKU from neurological damage. Where screening is carried out in ma- patients with DHPR deficiency show a normal blood or urinary neop- ternity wards, the blood sample is usually obtained between days 2 terin and biopterin profile. Therefore, DHPR activity measurement is and 5; in general, however, screening is carried out mostly between essential in all patients with HPA, regardless of pterin measurements. the ages of 2 and 7 days [4]. In the U.S., samples are typically obtained Fig. 1 shows the algorithm for the diagnostic work-up of elevated at 24–48 h. A commonly used Phe cut-off level for diagnosis of PKU is blood Phe levels. Table 1 summarizes the most important biochemical 120–130 μmol/L (with a Phe/Tyr ratio N2), with TMS employed [5]. parameters used in the differential diagnosis of HPA. Concern has been expressed that screening too early, associated with a shift towards earlier discharge from maternity wards, can provide a false negative result, as there will have been insufficient oppor- 1.3. BH4 loading test tunity for Phe from the diet to build up to diagnostic (and toxic) levels. It is currently accepted that the sensitivity of screening in a The BH4 loading test was initially used to discriminate between healthy neonate is adequate before 24 h of life, especially where the patients with elevated phenylalanine levels due to PAH deficiency screening test involves measurement of Phe/Tyr ratios to increase and patients with elevated Phe levels due to BH4 deficiency (enzyme sensitivity relative to Phe measurement alone [3,6]. However, as the defects in the biosynthesis or regeneration of the cofactor BH4) pretreatment level is often used as a diagnostic parameter for the [11,12]. Thus, the loading test is an additional useful tool for the classification of the PKU phenotype new cut-offs have to be deter- early detection of BH4 deficiencies and it was used in Europe for al- mined for classic and mild PKU and mild HPA when measuring the most 30 years. In addition, this test detects PKU patients responsive Phe level so early. to BH4 administration. Some infants, particularly those born prematurely, may demon- Detection of BH4-responsive PKU patients is important because strate immaturity of enzyme systems involved in amino acid metabo- some PKU patients benefit from oral administration of BH4 (saprop- lism, resulting in a transient elevation of blood Phe to a level sufficient terin dihydrochloride) in that their blood Phe level decreases or to test positive in a PKU screening test. The results of early PKU screen- even normalizes under pharmacological therapy with BH4 [13]. The ing should also be interpreted with caution in sick neonates or in ne- phenomenon of BH4-sensitive PKU was initially described in Japanese onates underFOR parenteral nutrition ELECTRONIC or blood transfusion, and a second patients, and conUSEfirmed in retrospective ONLY and prospective studies with screening test should be sent if it is unclear whether the child had suf- large cohorts of patients [14–16]. Modalities of the BH4 challenge ficient protein intake when the first test was collected. vary in the literature from a 24-hour test with a single administration of BH4 (10–20 mg/kg) to several weeks of administration with daily or weekly monitoring of blood Phe levels [17,18]. 1.2. Differential diagnosis There is general agreement that a reduction on blood Phe of at least 30% in response to BH4 loading indicates a clinically significant About 2% of all Phe level elevations detected by the newborn effect, although in some centers a lower cut-off value may be defined screening are due to disorders in BH4 metabolism, highlighting the for individual patients, or no specific cut-off value may be used [19]. importance of always considering the differential diagnosis for The frequency of BH4-responsiveness is highest in patients with every even slightly elevated blood Phe level [7]. Frequency of BH4 de- mild (non-PKU) HPA, or mild PKU resulting from PAH mutations ficiency is higher in some countries where the rate of consanguineous that allow for residual enzyme activity [15,20]. Conversely, the re- marriages tends to maintain the presence of genetic disorders within sponse rate among patients with classic PKU (little or no residual S4 N. Blau et al. / Molecular Genetics and Metabolism 104 (2011) S2–S9

Fig. 2. Proposed algorithms for the BH4 (sapropterin dihydrochloride) challenge, screening, and initiating treatment in BH4-responsive PKU patients. A) Initial screening test with blood Phe monitoring on the first day and BH4 (sapropterin dihydrochloride) administration (20 mg/kg) on two following days; B) Efficiency testing in BH4-repon- Fig. 1. Diagnostic flow-chart for the laboratory diagnosis of PKU and BH4 deficiencies. sive patients over several weeks with BH4 doses adjusted individually according to Phe Modified according to Opladen et al. [10]. Dried blood spots (DBS) or random urine (U) tolerance and therapeutic blood Phe levels. can be used for the differential diagnosis and depending on the profile of neopterin (Neo), biopterin (Bio), and primapterin (Pri) and dihydropteridine reductase (DHPR) activity in DBS, diagnosis of following BH4 deficiencies can be established: GTP cyclo- dosage of 20 mg/kg is used in 92% of the centers and duration of the hydrolase I (GTPCH) deficiency (low or no detectable neopterin and biopterin), 6- test is quite variable: 24 h (33%), 48 h (24%), 72 h (16%), and about pyruvoyl-tetrahydropterin synthase (PTPS) deficiency (high neopterin and low or no 26% run the test over 1–4 weeks, most of them from the U.S. with fi detectable biopterin), dihydropteridine reductase (DHPR) de ciency (normal neop- test duration over at least 4 weeks. About half of the survey partici- terin and normal or elevated biopterin and no DHPR activity), and pterins-4a- pants defined BH4-responsiveness as both increase in tolerance for carbinolamine dehydratase (PCD) deficiency (elevated neopterin, low-normal biopterin, and elevated primapterin). n: normal. dietary Phe (by a factor of two) and reduction in blood Phe levels (by at least 30%), while the rest use only blood Phe reduction as criterion. PAH activity) is very low. A number of PAH mutations associated with BH4 responsiveness have been identified and genotyping is a useful 1.4. Cerebrospinal fluid investigation additional tool for predicting responsiveness [21,22] (see below). Fig. 2 summarizes the proposed procedure for the BH4 loading test BH4 deficiency influences the synthesis of catecholamines, seroto- in Europe. In newborns the test should be performed before introduc- nin [24] and nitric oxide [25] in the central nervous system (CNS), ing the low-Phe diet and at elevated blood Phe levels (N400 μmol/L). and measurement of their metabolites in cerebrospinal fluid (CSF) is In infant or adult PKU patients on a Phe-restricted diet, the diet needs important for the diagnosis of different forms (severe v. mild) of to be modified by increasing the protein intake (egg or milk powder BH4 deficiencies. Not only the absolute levels of 5-hydroxyindoleace- before and during the test). Table 2 summarizes factors that may po- tic acid and homovanillic acid in CSF, but also differences in the ratios tentially influence the outcome of the test. of neurotransmitter levels provide important diagnostic information An international online survey [23] with 92 participants from 30 relating to the severity and outcome of BH4 deficiency [26]. different countries documented that in 62% of the metabolic centers the BH4 loading test is an integral part of the diagnostic work-up 2. Classification for PKU. The main reason for not using the BH4 test is either relatively high costs or no availability of BH4 in some countries (78%). Most 2.1. Phenylalanine loading test centers use theFOR BH4 test in all age ELECTRONIC groups (79%) and only in about USE ONLY 11% of the metabolic centers pregnant PKU women are tested. A Phenylalanine loading tests were applied since 1956 for the detection of heterozygotes in PKU families [27], until, in the late 1980s, this Table 1 Biochemical parameters in dried blood spots (DBS) used in the differential diagnosis of Table 2 PKU and BH4 deficiencies. Factors potentially affecting the BH4 loading test. Deficiency Neopterin Biopterin Primapterin DHPR activity Phe ■ BH4 dosage (higher sensitivity with 20 mg/kg v. 10 mg/kg) PKU n- n- nn ■ Duration of the test (24–48 h for initial screening v. 4–8 weeks for efficiency) GTPCH nn a ■ Food intake (better GI absorption of BH4 with high calories food) PTPS nn ■ Age (outcome may be different in newborns v. adolescents v. adults) PCD -n n ■ Diet (better response to BH4 when out of diet and on higher blood Phe levels) a DHPR n n- n ■ GI absorption (may be individually different; monitor blood BH4 levels) ■ Genotype a Can be normal in the early neonatal period. N. Blau et al. / Molecular Genetics and Metabolism 104 (2011) S2–S9 S5 approach was replaced by molecular analysis of PAH gene haplotypes More precisely, PAH deficiency may be classified into four different and mutations [28]. Phenylalanine loading tests gained further inter- phenotypes: classic PKU presenting with Phe pretreatment levels est when Guthrie card mass screening uncovered not only the N1200 μmol/L, moderate PKU with Phe pretreatment levels of 900– expected cases of classic PKU but also variants of PKU [29]. Because 1200 μmol/L, mild PKU with Phe pretreatment levels of 600– these variants were initially thought not to require dietary treatment, 900 μmol/L, and mild HPA with Phe pretreatment level b600 μmol/L a reliable discrimination of these phenotypes was needed. For practi- [38,39]. 3 14 cal and ethical reasons, in vivo Hor C isotope studies [30] or inva- Although pretreatment Phe levels are indispensable for pheno- sive testing such as enzyme assays from liver biopsies [31] were not typing PKU, they are dependent on some variables, e. g. on the tim- appropriate. Therefore, Blaskovics [32] developed in the mid-1960s ing of blood Phe measurement, on the neonatal catabolism, and on the standardized three day natural protein loading test with evapo- the diet received at the time of blood sampling. Due to improvement rated milk that was applied at 6 months of age in the U.S. [33] and in newborn screening with an early blood sampling at day three of German [34] PKU Collaborative Studies for classification of PKU and life, patients with PKU are diagnosed much earlier, thus resulting for genotype-phenotype analysis. With this test three principal in significantly lower pretreatment Phe levels. If pretreatment Phe types of Phe blood level response, types 1, 2 and 3, were delineated levels will be used for determining PKU phenotypes in the future, in both studies. Type 1 is characterized by a 72 h Phe beyond this classification needs to be adjusted. In day-to-day-practice, pre- 1200 μmol/L and corresponds to classic PKU. Type 2 is defined by a treatment blood Phe levels have been shown to be used for pheno- spontaneous decline of Phe levels, despite continuation of loading, typing patients with PKU in about 80% of the treatment centers [23]. from above 1200 μmol/L after day 2 to a 72 h Phe levels of 1200– 600 μmol/L. In both studies, about 10% of the patients belong to this 2.4. Phenylalanine tolerance type. In type 3 the plasma Phe fluctuates around 600 μmol/L and the 72 h Phe levels are b600 μmol/L. Clinically, it corresponds to mild Daily Phe tolerance has been established as a stable parameter for HPA. On a free diet, these patients (22% in the U.S., and 13% in the phenotyping the various types of PAH deficiency [29]. Phe tolerance is German study) do not need a low-Phe treatment for normal mental usually determined at the age of 5 years and indicates the amount of outcome [35]. daily Phe intake that a patient can tolerate without an increase of the Major successes of the Blaskovics loading test were (i) the discov- blood Phe level above the upper target range. Three different pheno- ery of the type 2 response as an indicator of mutant enzyme activa- types may be classified by using Phe tolerance: classic PKU with a Phe tion by Phe [36], (ii) the delineation of the type 3 response as the tolerance b20 mg/kg/day, variant PKU with a Phe tolerance of 20– base for the decision to discontinue dietetic treatment, and (iii) its 50 mg/kg/day, and mild HPA with a Phe tolerance N50 mg/kg/day use as a measure of in vivo phenotype for genotype–phenotype anal- [29]. More detailed is the classification into four different phenotypes, ysis. In contrast to its proven scientific value, the test has only limited who defines classic PKU with a Phe tolerance b20 mg/kg/day (250– value for the dietetic treatment of patients with classic and mild PKU 300 mg/day), moderate PKU with a Phe tolerance of 20–25 mg/kg/day because the Phe 72 value will not predict the current and future indi- (350–400 mg/day), mild PKU with a Phe tolerance of 25–50 mg/kg/ vidual dietary requirements (see below) and the patients may mani- day (400–600 mg/day), and mild HPA with patients off diet [38,39]. fest during the test signs of intoxication such as nausea, vomiting, Recently it has been shown that Phe tolerance may be predictable irritability, insomnia and EEG changes. The test is no longer necessary already at the age of 2 years, and that Phe tolerance at age 2, 3, and [37] and has been replaced in practice by predictive molecular and 5 years correlates with that at age 10 years [41]. In contrast, reassess- enzymatic classifications [22,38]. A recent online survey accordingly ment of Phe tolerance may be necessary in adults [42]. Although Phe revealed that only 4% of centers still use it for estimating the pheno- tolerance is a good indicator for the PKU phenotype, its determination type of their patients [23]. may be unreliable if not determined under standardized conditions. In day-to-day-practice, prescribed Phe intake often is much lower 2.2. Phenotypes than the effective Phe intake at home. We therefore recommend determining Phe tolerance under standardized in-patient conditions Depending on the enzyme defect, the genotype and the severity of with precise dietary protocols. In clinical follow-up, Phe tolerance is the disease, different forms of PKU with different clinical phenotypes used for phenotyping patients with PKU in 70% of metabolic centers have been described. Thus, different classifications for PKU pheno- queried [21]. types have been established. PKU may be classified as classic PKU and as variant PKU which includes all milder forms of PKU, (i.e. mod- 2.5. Clinical course erate PKU and mild PKU), as mild HPA or non-PKU HPA, and, additionally, as BH4-responsive PKU [29,39,40].Definition of PKU phenotypes Furthermore, various types of PAH deficiency may be distin- may be essential in establishing treatment options, e.g. new therapeu- guished by the clinical course of the disease. This includes data on tic strategies, in counseling and in prediction of the outcome, and in the outcome (e. g. education, IQ), the maximum blood Phe concentra- pregnancy. Pretreatment blood Phe levels, the individual Phe toler- tions (e. g. during febrile infection, dietary non-compliance), the fluc- ance, and the clinical course of the disease may help to discriminate tuation of blood Phe levels, and the Phe/Tyr ratios [43,44]. Though, in the differentFOR phenotypes of PKU, ELECTRONIC but are not precise parameters and day-to-day-practice, USE clinical course ONLY of the disease is only used in 31% the cut-offs for pretreatment levels collected during the first 24– of metabolic centers to distinguish different PKU forms [23]. 48 h, as they relate to the different types of PKU, have to be newly In day-to-day practice, classification of PKU is essential for choos- defined. ing the optimal treatment. This may suggest a simplified classification scheme based on treatment requirements: a) patients who do need 2.3. Blood phenylalanine strict dietary treatment (PKU), b) patients who do not need any treatment (non-PKU HPA), c) patients who may be treated with BH4 In 1980, for the first time, blood Phe levels were used to discriminate (BH4-responsive PKU). between three different phenotypes of PKU [29]. Still, this classification of the various types of PAH deficiency is used for phenotyping PKU: 2.6. Software and phenylalanine home monitoring device Classic PKU is defined by presenting with Phe pretreatment levels N1200 μmol/L, variant PKU with Phe pretreatment levels of 600– The Blaskovics loading tests at 6 months with their resulting 72 h 1200 μmol/L, and mild HPA with Phe pretreatment levelsb600 μmol/L. Phe values [32] yield indicators of metabolic phenotypes which belong S6 N. Blau et al. / Molecular Genetics and Metabolism 104 (2011) S2–S9 to a bimodal distribution with nadir at about 1,200 and 1,600 μmol/L, it has been shown (in an observational study) that increased Phe respectively [33,34], thus establishing a classification into classic and values, in a considerable part of cases, are the consequence of hidden variant PKU. A more detailed separation of the metabolic phenotypes catabolism (as indicated in the model by diminished PROT values) was established by Güttler and Guldberg [45] through estimating and not of non-compliance [49]. Phe tolerance at 5 years of age (see also above). Neither of these sys- Application of this kinetic model in a user-friendly format (e.g. a tems can, at any age, predict the individual Phe tolerance. This is so-called App) would support self-control with home-monitoring de- made possible, however, by mathematical analysis of the data with a vices [50] and could be trained e.g. in PKU summer camps. Such a de- kinetic model using per cent body water (PBW), net protein synthe- vice is presently in development and according to a recent survey sis (PROT), and 1st order rate of metabolic disposal (KOUT) of Phe [23], 85% would consider home monitoring of Phe useful for both as variable parameters [46]. Both PBW and PROT are age-dependent children and pregnant women. parameters, whereas KOUT is expected not to change with age. With age-specific PBW and PROT data from the literature [47],an 3. Genetics age-independent KOUT would enable the reliable prediction of evolution of Phe tolerance at all ages. Even in adults, from their Blaskovics 3.1. PAH mutations and PKU genotypes (incl. databases) test data, provided the patients, or their data, can be ‘retrieved’ again [48]. As already stated, knowing whether a patient has residual PAH en- By kinetic analysis of the protein loading data of the German PKU zyme activity can be relevant for the therapeutic approach, the likely Study it has been possible to explain the type 2 response as a conse- Phe tolerance, and the expected response to BH4. Delineation of the quence of mutant PAH activation by high Phe levels [36]. Also some mutations of the PAH gene was initiated immediately after the clon- mild cases, originally classified as type 1, were found to be activated ing of the gene in 1983 [28]. Initially the most prevalent mutations by Phe levels to some extent. The distribution of the KOUT estimates, in the Western European population were identified and character- including the maximal KOUT estimates of the cases with activation, is ized with regard to the in vitro residual enzyme activity associated depicted in Fig. 3. This distribution is apparently multimodal and the with the respective mutation [51,52]. Subsequently a ‘Phenylalanine peaks appear to represent classic (a), moderate (b), and mild PKU (c), Hydroxylase Locus Knowledgebase’ PAHdb was created and curated and mild HPA (d), respectively. These findings can be compared to at McGill University [53] (http://www.pahdb.mcgill.ca/) that cata- the classification of Güttler and Guldberg [45] by treating the dietary loged knowledge about PAH alleles and mutations and their charac- tolerance at the target Phe level as the equilibrium state of Phe me- teristics as reported by clinicians and laboratories from around the tabolism. In the kinetic model [46] this corresponds to: world. This database now lists a total of 564 PAH mutations discov- ered world wide as well as the knowledge available about the respec- Target Phe ¼ ðÞðÞintake–PROT mg=day =PBW T KOUT=day ð1Þ tive mutation including the residual enzyme activities of ~200 mutations. Of the 564 mutations 60.5% are missense mutations, With Eq. (1) and taking PBW=0.65 and PROT=2.7 mg Phe per kg 13.5% deletions, 11% splice site mutations, (5.7% silent mutations), body weight and day, the classification of Güttler and Guldberg trans- 5% nonsense mutations, and 1.8% insertions. It was evident early on lates into KOUT values of 0.62 and 1.08 per day for moderate and mild that the majority of the PAH mutations were missense mutations PKU, respectively. The data at KOUT=1.7 (d) corresponds within this not preventing transcription or translation and that the majority of system to an equilibrium Phe level of 110 μmol/L at an intake of the patients are compound heterozygotes, meaning they carry a dif- 120 mg Phe per kg body weight and day (here taken as ‘normal diet’). ferent mutation in each of their alleles. PAH deficiency thus most These calculations implicate that the constant of metabolic dispos- often results from complex interactions of mutant alleles or rapid in- al of Phe is about identical between 6 months and 5 years of age and tracellular destruction of mutant enzyme subunits making genotype/ may therefore be considered as a patient-specific parameter. This is phenotype correlations based on the knowledge about individual mu- relevant for the day-to-day management of PKU children, because tations challenging.

3.2. Phenotype–genotype correlation

The first publication with extensive data on genotype/phenotype correlation in PKU was able to establish that the genotype of the patient correlates with the biochemical phenotype [54]. PKU patients had been tested for 8 mutations of the PAH gene for which the in vitro residual enzyme activity had been determined. This mutation analysis had allowed for complete genotype determination (identifying both mutations) in 104 patients with PAH deficiency. Stringent classification criteria for the biochemical phenotype were applied to determine genotype/phenotype correlation in these 104 patients. FOR ELECTRONICThere was a highly USE significant correlation ONLY between the genotype of the patients and the biochemical phenotype (r=0.74–0.84, pb0.001 depending on the parameter) although the genotype was expressed as the predicted residual enzyme activity of the patient and was calculated as the mean of the combined in vitro residual enzyme activities of both mutant alleles of the patient, which was quite a simplification compared to the real in vivo situation. The goal of the research had been to determine whether genotype analysis after ex- Fig. 3. Distribution of KOUT [per day], the 1st order rate constant of metabolic disposal clusion of BH4 cofactor deficiency could replace more involved clini- of Phe, in the German PKU Study [36], N=157. The data suggest the peaks a, b, c, and d cal testing such as response to a standardized oral protein load at as representing classic, moderate and mild PKU, and mild HPA, respectively. The apparently multimodal distribution implies a frequency in the German population of 15% 6 months of age and Phe tolerance assessment in an inpatient setting and 8% for mild PKU and mild HPA, respectively. Normal KOUT values, determined by over the course of 1–2 weeks at 5 years of age. These clinical tests had i.v. loads, range from 10.6 to 24.9 per day, corresponding to a half-life of 0.7 to 1.6 h. proven to be the most reliable clinical classification criteria in large N. Blau et al. / Molecular Genetics and Metabolism 104 (2011) S2–S9 S7 clinical studies [54] but were cumbersome and in the case of the load- homozygous state indicate that certain mutant homotetramers may ing test exposed the patient to high Phe concentrations. not allow for stable active enzyme even if the mutations themselves In 1998 Guldberg at al. [38] reported the identification of the com- appeared to be mild in vitro. Patients homozygous for these particular plete genotype in 686 patients from 7 European centers. Based on the mutations may occasionally not respond to BH4 or not within 48 h. Fi- Phe tolerance (or in the case of mild HPA the pretreatment level) of nally, for the c.1066-11GNA mutation in the homozygous state different 297 functionally hemizygous patients (patients carrying a null allele phenotypes and the whole spectrum of BH4 responsiveness (negative, and the uncharacterized mutant allele), an arbitrary phenotype cate- positive, and slow) is cataloged. Given that this is a mutation in an in- gory was assigned to each of 105 mutations for which the residual en- tron that creates a cryptic splice acceptor site it is conceivable that the zyme activity was not known. Using these arbitrary categories cryptic splice site does not always come into play and functional en- phenotype predictions for 650 of the patients were made. In 79% of zyme is being translated under certain circumstances creating this the patients the predicted phenotype matched the observed pheno- variability. type and the authors concluded that differences in the phenotype General truths about genotype/phenotype correlations in PKU that classification across centers might have accounted for the geno- have emerged from the data cataloged in the data bases, especially in type–phenotype inconsistencies that were observed. Overall, despite the BIOPKU data base are: the potential shortcomings, this work also led to the conclusion that 1. Mutation combinations that allow for b15% in vitro enzyme activ- the genotype is the main determinant of the biochemical phenotype ity cause classic PKU and do not respond to BH4. Mutation combi- in most patients with PAH deficiency. nations that allow for N20% residual activity responds to BH4. Many more studies were published reporting genotype/phenotype Responders have moderate to mild phenotypes. correlations in different populations and inconsistencies, i.e. genotypes 2. Splice site mutations may cause classic or mild PKU depending on that were associated with several different phenotypes. It was also ‘read through’ (i.e. normal splicing may sometimes occur despite pointed out in numerous publications (first by S. Kaufman [55])and the mutation), and the fact that they have different phenotype as- investigated in vitro [56] that the combination of the two mutant alleles sociations is listed in the available data bases. is important for the residual PAH enzyme activity and that the individ- 3. Specific mild mutation/classic mutation combinations with identical ual mutations of a patient should not be viewed by themselves. It is predicted residual enzyme activity may have different phenotype as- clear that the residual activity of an enzyme that is a homotetramer sociations (due to negative intra-allelic complementation) but the that consists of a combination of different mutant subunits may not phenotype associations of different mutation combinations can be simply be the mean of the activity that each subunit produces by itself found in the available data bases making prediction unnecessary. in vitro because there may be negative intra-allelic complementation 4. The BH4 responsiveness of many mutation combinations (com- between different mutant enzyme subunits [55]. However, a correlation plete genotypes) has been well established multiple times. Pa- between genotype and biochemical phenotype prevailed in the majority tients that have those genotypes may not need to undergo BH4 of the patients reported. Limitations regarding genotype-based prediction responsiveness testing. of the phenotype, however, are that the in vitro residual enzyme activity is not known for all mutations, that negative intra-allelic comple- After assessment of the genotype (both mutations) of a patient mentation may occur, and some exceptions described below may one can now in many cases simply look up the associated phenotype occur. and BH4 responsiveness that is cataloged for this mutation combination in the databases rather than trying to predict it using a simplify- 3.3. BH4-responsive mutations and genotypes ing formula and the in vitro residual enzyme activity of the mutant alleles as was done in the past and had its limitations (see above). As mentioned above a ‘Phenylalanine Hydroxylase Locus Knowl- DNA diagnostic laboratories should include the information cataloged edgebase’ PAHdb exists at McGill University [53] (http://www.pahdb. in the data bases for a specific mutation combination in their reports mcgill.ca/) that has catalogued knowledge about PAH alleles and muta- about PAH mutations to clinicians to allow physicians to make the tions including the associated phenotype. It lists the phenotype associ- best informed decisions based on available information. ated with about 600 allele combinations; however this data is ‘self- In an anonymous survey among 65 US metabolic centers 72% of reported’ by clinicians and laboratories using variable criteria for the de- the centers reported that they classify PKU patients into classic PKU, termination of the phenotype. Another database, the BIOPKU database mild PKU, and mild HPA; 40% were ordering mutation testing on all (www.biopku.org) was created at the University of Zürich. It has catalo- of their patients 53% on some of their patients; 53% of the centers gued the complete genotypes (complete mutation combinations) of order genotype analysis, when they order it, to understand the phe- 730 PKU patients, their phenotypes (based on the highest blood Phe notype of the patient; 72% order to provide the family with genetic in- levels before starting treatment) and their response to BH4. These 730 formation (multiple answers could be chosen); of those who do not patients represent a total of 430 different mutation combinations. order mutation analysis 40% said they had too little information Some of the mutation combinations of the 730 patients listed are not about the meaning of the mutations (what phenotype to expect), always associated with the same phenotype. Mutations for which in- 54% said they do not need mutation analysis to understand the phe- consistencies regarding genotype/phenotype correlation are reported notype of their patients, and 36% said the health insurance of their pa- in this databaseFOR as well as in the ELECTRONIC literature are the R261Q mutation in tients does not USE cover the testing; 20%ONLY of the centers said they order the homozygous state or in combination with the R158Q mutation, mutation analysis on each PKU patient that wants to try BH4 supple- the L48S mutation in the homozygous state or in combination with mentation to see whether the patient should have residual enzyme the R158Q mutation, and the Y414C mutation in combination with activity (which would make it more likely that patient responds to the R408W mutation. However, although different phenotypes are as- BH4). sociated with these mutation combinations their BH4 responsiveness is always the same except in a smaller percentage of L48S homozygotes Acknowledgments (19%) that may not respond while 81% respond. A similar phenomenon was observed in 40% of Turkish patients that were homozygous for the Authors would like to thank Serono Scientific International Foun- E390G mutation and did not respond to BH4 [22] while other homozy- dation (SSIF) for supporting organization of the workshop on “Diag- gous patients and all patients compound heterozygous for this very nosis, classification and genetics of PKU”. This work was supported mild mutation listed in the BIOPKU database did respond to BH4. in part by the Swiss National Science Foundation grant no. 31003A- The observations regarding variability for two mild mutations in the 119982 (to NB). UL is grateful to the German Collaborative Study on S8 N. Blau et al. / Molecular Genetics and Metabolism 104 (2011) S2–S9

PKU for providing the study data whose analysis resulted in Ref. [36] and dopamine in a pediatric reference population, Pediatr. Res. 34 (1993) 10–14. and Fig. 3 of the present communication. [25] G. Zorzi, B. Thöny, N. Blau, Reduced nitric oxide metabolites in CSF of patients with tetrahydrobiopterin deficiency, J. Neurochem. 80 (2002) 362–364. [26] L. Jäggi, M.R. Zurflüh, A. Schuler, A. Ponzone, F. Porta, L. Fiori, M. Giovannini, R. References Santer, G.F. Hoffmann, H. Ibel, U. Wendel, D. Ballhausen, M.R. Baumgartner, N. Blau, Outcome and long-term follow-up of 36 patients with tetrahydrobiopterin [1] J.L. Dhondt, Laboratory diagnosis of phenylketonuria, in: N. Blau (Ed.), PKU and deficiency, Mol. Genet. Metab. 93 (2008) 295–305. BH4: Advances in Phenylketonuria and Tetrahydrobiopterin, SPS Verlagsge- [27] K.W. Driscoll, D.Y. Hsia, W.E. Knox, W. Troll, Detection by phenylalanine toler- sellschaft, Heilbronn, 2006, pp. 161–179. ance tests of heterozygous carriers of phenylketonuria, Nature 178 (1956) [2] R. Guthrie, A. Susi, A simple phenylalanine method for detecting phenylketonuria 1239–1240. in large populations of newborn infants, Pediatrics 32 (1963) 338–343. [28] S.L. Woo, A.S. Lidsky, F. Güttler, T. Chandra, K.J. Robson, Cloned human phenylal- [3] D.H. Chace, J.E. Sherwin, S.L. Hillman, F. Lorey, G.C. Cunningham, Use of anine hydroxylase gene allows prenatal diagnosis and carrier detection of classi- phenylalanine-to-tyrosine ratio determined by tandem mass spectrometry cal phenylketonuria, Nature 306 (1983) 151–155. to improve newborn screening for phenylketonuria of early discharge [29] F. Güttler, Hyperphenylalaninemia: diagnosis and classification of the various specimens collected in the first 24 hours, Clin. Chem. 44 (1998) types of phenylalanine hydroxylase deficiency in childhood, Acta Paediatr. 2405–2409. Scand. Suppl. 280 (1980) 1–80. [4] M. Zaffanello, G. Zamboni, C. Maffeis, L. Tato, Neonatal birth parameters of posi- [30] W.D. Lehmann, R. Fischer, H.C. Heinrich, P. Clemens, R. Grüttner, Metabolic con- tive newborns at PKU screening as predictors of false-positive and positive results version of L-[U-14 C]phenylalanine to respiratory 14CO2 in healthy subjects, phe- at recall-testing, J. Med. Screen 10 (2003) 181–183. nylketonuria heterozygotes and classic phenylketonurics, Clin. Chim. Acta 157 [5] Phenylketonuria (PKU): screening and management, NIH Consens. Statement 17 (1986) 253–266. (2000) 1–33. [31] M.C. Hsieh, H.K. Berry, M.K. Bofinger, P.J. Phillips, M.B. Guilfoile, M.M. Hunt, Com- [6] J.W. Eastman, J.E. Sherwin, R. Wong, C.L. Liao, R.J. Currier, F. Lorey, G. Cunningham, parative diagnostic value of phenylalanine challenge and phenylalanine hydrox- Use of the phenylalanine:tyrosine ratio to test newborns for phenylketonuria ylase activity in phenylketonuria, Clin. Genet. 23 (1983) 415–421. in a large public health screening programme, J. Med. Screen 7 (2000) [32] M.E. Blaskovics, Phenylketonuria: loading studies revisited, in: N. Blau (Ed.), PKU 131–135. and BH4: Advances in Phenylketonuria and Tetrahydrobiopterin, SPS Verlagsge- [7] N. Blau, B. Thöny, R.G.H. Cotton, K. Hyland, Disorders of tetrahydrobiopterin and sellschaft, Heilbronn, 2006, pp. 104–119. related biogenic amines, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle, B. Childs, [33] M.E. O'Flynn, N.A. Holtzman, M. Blaskovics, C. Azen, M.L. Williamson, The diagno- B. Vogelstein (Eds.), The Metabolic and Molecular Bases of Inherited Disease, sis of phenylketonuria: a report from the Collaborative Study of Children Treated McGraw-Hill, New York, 2001, pp. 1725–1776. for Phenylketonuria, Am. J. Dis. Child. 134 (1980) 769–774. [8] N. Blau, F.J. Van Spronsen, H.L. Levy, Phenylketonuria, Lancet 376 (2010) [34] P. Lutz, H. Schmidt, U. Batzler, Study design and description of patients, Eur. J. 1417–1427. Pediatr. 149 (Suppl 1) (1990) S5–S12. [9] N. Longo, Disorders of biopterin metabolism, J. Inherit. Metab. Dis. 32 (2009) [35] J. Weglage, M. Pietsch, R. Feldmann, H.G. Koch, J. Zschocke, G. Hoffmann, A. 333–342. Muntau-Heger, J. Denecke, P. Guldberg, F. Güttler, H. Möller, U. Wendel, K. Ullrich, [10] T. Opladen, B. Abu Seda, A. Rassi, B. Thöny, G.F. Hoffmann, N. Blau, Diagnosis of E. Harms, Normal clinical outcome in untreated subjects with mild hyperpheny- tetrahydrobiopterin deficiency using filter paper blood spots: further develop- lalaninemia, Pediatr. Res. 49 (2001) 532–536. ment of the method and 5 years experience, J. Inherit. Metab. Dis. 34 (2011) [36] U. Langenbeck, P. Burgard, U. Wendel, M. Lindner, J. Zschocke, Metabolic pheno- 819–826. types of phenylketonuria. Kinetic and molecular evaluation of the Blaskovics pro- [11] H.C. Curtius, A. Niederwieser, M. Viscontini, A. Otten, J. Schaub, S. Scheibenreiter, tein loading test, J. Inherit. Metab. Dis. 32 (2009) 506–513. H. Schmidt, Atypical phenylketonuria due to tetrahydrobiopterin deficiency. Di- [37] J. Donlon, H. Levy, C.R. Scriver, Hyperphenylalaninemia: phenylalanine hydroxy- agnosis and treatment with tetrahydrobiopterin, dihydrobiopterin and sepiap- lase deficiency, in: D. Valle, A.L. Beaudet, B. Vogelstein, K.W. Kinzler, S.E. Antonar- terin, Clin. Chim. Acta 93 (1979) 251–262. akis, A. Ballabio (Eds.), The Online Metabolic & Molecular Bases of Inherited [12] A. Ponzone, O. Guardamagna, I. Dianzani, R. Ponzone, G.B. Ferrero, M. Spada, Disease, McGraw-Hill, Montreal, 2008, Ch. 77. R.G.H. Cotton, Catalytic activity of tetrahydrobiopterin in dihydropteridine [38] P. Guldberg, F. Rey, J. Zschocke, V. Romano, B. Francois, L. Michiels, K. Ullrich, reductase deficiency and indications for treatment, Pediatr. Res. 33 (1993) 125–128. G.F.Hoffmann,P.Burgard,H.Schmidt,C.Meli, E. Riva, I. Dianzani, A. Ponzone, [13] A.C. Muntau, W. Röschinger, M. Habich, H. Demmelmair, B. Hoffmann, C.P. J. Rey, F. Güttler, A European multicenter study of phenylalanine hydroxylase Sommerhoff, A.A. Roscher, Tetrahydrobiopterin as an alternative treatment for deficiency: classification of 105 mutations and a general system for genotype- mild phenylketonuria, N. Engl. J. Med. 347 (2002) 2122–2132. based prediction of metabolic phenotype,Am.J.Hum.Genet.63(1998)71–79. [14] C. Bernegger, N. Blau, High frequency of tetrahydrobiopterin-responsiveness [39] P. Guldberg, F. Güttler, Mutations in the phenylalanine hydroxylase gene: among hyperphenylalaninemias: a study of 1919 patients observed from 1988 methods for their characterization, Acta Paediatr. Suppl. 407 (1994) 27–33. to 2002, Mol. Genet. Metab. 77 (2002) 304–313. [40] S. Kure, D.C. Hou, T. Ohura, H. Iwamoto, S. Suzuki, N. Sugiyama, O. Sakamoto, K. [15] B. Fiege, N. Blau, Assessment of tetrahydrobiopterin (BH4)-responsiveness in Fujii, Y. Matsubara, K. Narisawa, Tetrahydrobiopterin-responsive phenylalanine phenylketonuria, J. Pediatr. 150 (2007) 627–630. hydroxylase deficiency, J. Pediatr. 135 (1999) 375–378. [16] H. Levy, A. Milanowski, A. Chakrapani, M. Cleary, P. Lee, F.K. Trefz, C.B. [41] F.J. van Spronsen, M. van Rijn, B. Dorgelo, M. Hoeksma, A.M. Bosch, M.F. Mulder, Whitley, F. Feillet, A.S. Feigenbaum, J.D. Bebchuk, H. Christ-Schmidt, A. J.B. de Klerk, T. de Koning, M.E. Rubio-Gozalbo, M. de Vries, P.H. Verkerk, Phenyl- Dorenbaum, Efficacy of sapropterin dihydrochloride (tetrahydrobiopterin, alanine tolerance can already reliably be assessed at the age of 2 years in patients 6R-BH4) for reduction of phenylalanine concentration in patients with with PKU, J. Inherit. Metab. Dis. 32 (2009) 27–31. phenylketonuria: a phase III randomized placebo-controlled study, Lancet [42] E.L. MacLeod, S.T. Gleason, S.C. van Calcar, D.M. Ney, Reassessment of phenylala- 370 (2007) 504–510. nine tolerance in adults with phenylketonuria is needed as body mass changes, [17] N. Blau, Sapropterin dihydrochloride for phenylketonuria and tetrahydrobiop- Mol. Genet. Metab. 98 (2009) 331–337. terin deficiency, Expert. Rev. Endocrinol. Metab. 5 (2010) 483–494. [43] V. Anastasoaie, L. Kurzius, P. Forbes, S. Waisbren, Stability of blood phenylalanine [18] H. Levy, B. Burton, S. Cederbaum, C. Scriver, Recommendations for evaluation of levels and IQ in children with phenylketonuria, Mol. Genet. Metab. 95 (2008) responsiveness to tetrahydrobiopterin (BH(4)) in phenylketonuria and its use 17–20. in treatment, Mol. Genet. Metab. 92 (2007) 287–291. [44] M. Humphrey, J. Nation, I. Francis, A. Boneh, Effect of tetrahydrobiopterin on Phe/ [19] N. Blau, Defining tetrahydrobiopterin (BH4)-responsiveness in PKU, J. Inherit. Tyr ratios and variation in Phe levels in tetrahydrobiopterin responsive PKU pa- Metab. Dis. 31 (2008) 2–3. tients, Mol. Genet. Metab. (2011), doi:10.1016/j.ymgme.2011.1005.1011. [20] B.K. Burton, D.K. Grange, A. Milanowski, G. Vockley, F. Feillet, E.A. Crombez, V. [45] F. Güttler, P. Guldberg, The influence of mutations on enzyme activity and phe- Abadie, C.O. Harding, S. Cederbaum, D. Dobbelaere, A. Smith, A. Dorenbaum, nylalanine tolerance in phenylalanine hydroxylase deficiency, Eur. J. Pediatr. The response of patients with phenylketonuria and elevated serum phenylala- 155 (1996) S 6–S 10. nine to treatment with oral sapropterin dihydrochloride (6R-tetrahydrobiop- [46] U. Langenbeck, J. Zschocke, U. Wendel, V. Hönig, Modelling the phenylalanine terin): a phaseFOR II, multicentre, open-label, ELECTRONIC screening study, J. Inherit. Metab. Dis. blood level response USE during treatment ofONLY phenylketonuria, J. Inherit. Metab. Dis. 30 (2007) 700–707. 24 (2001) 805–814. [21] M.R. Zurflüh, J. Zschocke, M. Lindner, F. Feillet, C. Chery, A. Burlina, R. Stevens, B. [47] S.J. Fomon, F. Haschke, E.E. Ziegler, S.E. Nelson, Body composition of reference Thöny, N. Blau, Molecular genetics of tetrahydrobiopterin responsive phenylala- children from birth to age 10 years, Am. J. Clin. Nutr. 35 (1982) 1169–1175. nine hydroxylase deficiency, Hum. Mutat. 29 (2008) 167–175. [48] B.K. Burton, L. Leviton, Reaching out to the lost generation of adults with early- [22] S.F. Dobrowolski, C. Heintz, T. Miller, C.R. Ellingson, C.C. Ellingson, I. Özer, G. treated phenylketonuria (PKU), Mol. Genet. Metab. 101 (2010) 146–148. Gökcay, T. Baykal, B. Thöny, M. Demirkol, N. Blau, Molecular genetics and impact [49] U. Langenbeck, M. Ammar, J. Zschocke, E. Solem, I.M. Knerr, J. Herwig, H. Boehles, of residual in vitro phenylalanine hydroxylase activity on tetrahydrobiopterin- Recognizing catabolic states during dietary treatment of phenylketonuria. An ap- responsiveness in Turkish PKU population, Mol. Genet. Metab. 10 (2011) plication of metabolic modelling J Inherit Metab Dis 34 (Suppl 3) (2011) S102. 116–121. [50] U. Wendel, U. Langenbeck, Towards self-monitoring and self-treatment in phe- [23] N. Blau, U. Langenbeck, J.B. Hennermann, U. Lichter Konecki, Diagnosis and nylketonuria—a way to better diet compliance, Eur. J. Pediatr. 155 (Suppl 1) management of PKU: an international survey J Inherit Metab Dis (abstract), (1996) S105–S107. 34 (Suppl 3) (2011) S97. [51] D.S. Konecki, U. Lichter Konecki, The phenylketonuria locus: current knowledge [24] K. Hyland, R.A.H. Surtees, S.J.R. Heales, A. Bowron, D.W. Howells, I. Smith, about alleles and mutations of the phenylalanine hydroxylase gene in various Cerebrospinal fluid concentrations of pterins and metabolites of serotonin populations, Hum. Genet. 87 (1991) 377–388. N. Blau et al. / Molecular Genetics and Metabolism 104 (2011) S2–S9 S9

[52] R.C. Eisensmith, Y. Okano, M. Dasovich, T. Wang, F. Güttler, H. Lou, P. Guldberg, phenotypic heterogeneity in phenylketonuria, N. Engl. J. Med. 324 (1991) U. Lichter-Konecki, D.S. Konecki, E. Svensson, et al., Multiple origins for 1232–1238. phenylketonuria in Europe, Am. J. Hum. Genet. 51 (1992) 1355–1365. [55] S. Kaufman, Phenylketonuria: Biochemical Mechanisms, in: B.W. Agranoff, M.H. [53] L. Hoang, S. Byck, L. Prevost, C.R. Scriver, PAH mutation analysis consortium data- Aprison (Eds.), Advances in Neurochemistry, Plenum Press, New York, 1976, base — a database for disease-producing and other allelic variation at the human pp. 1–132. PAH locus, Nucleic Acids Res. 24 (1996) 127–131. [56] P.J. Waters, How PAH gene mutations cause hyper-phenylalaninemia and why [54]Y.Okano,R.C.Eisensmith,F.Güttler,U.Lichter-Konecki,D.S.Konecki,F.K. mechanism matters: insights from in vitro expression, Hum. Mutat. 21 (2003) Trefz, M. Dasovich, T. Wang, K. Henriksen, H. Lou, et al., Molecular basis of 357–369.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S10–S18

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Minireview Nutrition in phenylketonuria

A. MacDonald a,⁎, J.C. Rocha b, M. van Rijn c, F. Feillet d a The Children's Hospital, Birmingham, UK b Centro de Genética Médica Jacinto de Magalhães, INSA, Porto, Portugal c Beatrix Children's Hospital, University Medical Center Groningen, Groningen, The Netherlands d Centre for Inborn Errors of Metabolism, INSERM U954, Service de Medecine Infantile I, Hôpital d'Enfants, Vandoeuvre les Nancy, France article info abstract

Article history: The same basic principles are used to deliver dietary treatment in PKU that was developed sixty years ago. Received 22 July 2011 Dietary treatment is undoubtedly very successful, but it has gradually evolved and been guided commonly Received in revised form 24 August 2011 by individual experience and expert opinion only. There is little international consensus about dietary prac- Accepted 24 August 2011 tice with improvements in specialist dietary products concentrating on taste and presentation rather than Available online 2 September 2011 nutritional composition. Many areas of dietary treatment have not been rigorously examined. In particular, the amino acid and micronutrient proﬁle of Phenylalanine-free (phe-free) amino acids requires further Keywords: Phenylketonuria study. In different formulations of phe-free amino acids, there are variations in the amino acid patterns as Phenylalanine well the amount of essential and non essential amino acids per 100 g/amino acids. The amount of added ty- Protein rosine and branch chain amino varies substantially, and in PKU speciﬁcally, there is little data about their rel- Amino acids ative absorption rates and bioavailability. In phe-free amino acids, there is evidence suggesting that some of Micronutrients the added micronutrients may be excessive and so the source and amount of each micronutrient should be Growth scrutinized, with a need for the development of international nutritional composition standards exclusively for these products. There is a dearth of data about the life-long phenylalanine tolerance of patients or the nutritional state of adult patients treated with diet. There is a growing need to measure body composition routinely in children with PKU and with the rise in childhood obesity, it is important to measure body fatness and identify those who are at greatest risk of ‘co-morbidities’ of obesity. There is necessity for international collaboration to ensure robust data is collected on many basic aspects of nutritional care to guarantee that diet therapy is delivered to the highest standard. © 2011 Elsevier Inc. All rights reserved.

Contents

1. Introduction ...... S11 2. Protein requirements ...... S11 2.1. Protein requirements for healthy individuals...... S11 2.2. Protein requirements in PKU ...... S11 2.3. The nutritional value of L-amino acids ...... S11 2.3.1. Metabolic control and protein substitute dosage ...... S12 3. Dietary phenylalanine...... S12 3.1. Requirements for phenylalanine in non-PKU ...... S12 3.2. Phenylalanine requirements in PKU ...... S12 3.3. Phenylalanine deﬁciency ...... S13 3.4. AllocationFOR of dietary phenylalanine ELECTRONIC...... USE ONLY S13 4. Growth and body composition...... S13 4.1. Longitudinal growth ...... S13 4.2. Anthropometry and body composition measurements ...... S14 4.3. Overweight and obesity trends in patients with PKU ...... S14 4.4. The long term impact of early nutrition ...... S14 5. Micronutrient status in PKU ...... S14

⁎ Corresponding author at: The Children's Hospital, Birmingham, B4 6NT, UK. Fax: +44 121 333 8021. E-mail address: [email protected] (A. MacDonald).

5.1. The vitamins ...... S14 5.2. The group B vitamins ...... S14 5.3. The minerals ...... S15 6. Recommendations ...... S16 7. Conclusions ...... S16 References ...... S16

1. Introduction requirement. When considering protein requirement, it is assumed energy requirement is met. Phenylketonuria (PKU) is a genetic disorder first treated with a low phenylalanine diet sixty years ago, and more recently with the drug BH4. Diet therapy and nutrition remain a central focus in the 2.2. Protein requirements in PKU treatment of PKU and diet has seen many improvements both in terms of nutritional quality and palatability of specialist dietary prod- Treatment consists of restriction of the essential amino acid phe- ucts. However, there are many areas where either limited knowledge nylalanine by reducing the natural protein intake with concomitant about the nutritional needs of the normal ‘healthy’ population and supplementation of all amino acids except phenylalanine to meet unreliable or under-developed tools to assess nutritional status have protein requirements [7]. Adequate intake of phenylalanine, protein hampered progress in developing diet therapy further. In addition, and energy must be provided to prevent breakdown of body tissues, some areas of diet therapy have not received rigorous examination which can lead to elevated plasma phenylalanine concentrations. and much existing practice is based on years of experience rather The dosage of the phenylalanine-free L-amino acids (phe-free than robust evidence. In this review, we will assess present knowl- amino acids) or protein substitute is determined both by the individ- edge concerning protein, amino acids, vitamin and trace element sta- ual phenylalanine tolerance and the total protein requirement (i.e. tus, in addition to examining the growth and body composition of natural protein+phe-free amino acids=total protein intake). Rec- patients with PKU. ommendations for the optimal amount of phe-free amino acids are based on protein recommendations for healthy individuals and additional factors that may influence protein utilization in PKU. They are 2. Protein requirements founded by: a) protein recommendations for healthy individuals, b) the nutritional and assimilation of phe-free amino acids compared The primary function of dietary protein is to provide amino acids with natural (intact) protein, c) growth and (d) studies investigating for growth, renewing tissue protein and synthesizing specific nitro- metabolic control with differing dosages of phe-free amino acids. gen-containing products [1]. After ingestion, proteins are denatured In PKU, although there is no data to support that a higher dose of in the stomach and pass into the small intestine, where the resultant phe-free amino acids is required when giving amino acids instead of mixture of free amino acids and small peptides is then transported intact protein as the main source, an incremental compensation fac- into the enteral cells and secreted as free amino acids into the blood- tor of 1.2 is used in the Dutch guidelines (unpublished data). This is stream or further metabolized within the gut or liver itself [2]. Protein in line with the USA RDA in which an adjustment is proposed of ap- intake as free amino acids results in a faster rate of appearance in proximately 20% to compensate both for losses due to digestibility plasma compared to intact protein [3,4]. and protein quality for mainly vegetarian diets [8]. In the fasted state, amino acids primarily appear from degradation of body protein as the only storage form; while in the fed state dietary proteins provide extra amino acids resulting in net protein synthesis. 2.3. The nutritional value of L-amino acids When amino acid appearance rate exceeds protein synthesis capacity, excess amino acids are degraded. Body proteins are in a constant state In PKU, the majority of non-phenylalanine protein is supplied by of degradation and (re)synthesis (250 g protein/day in adults), while phe-free amino acids. Studies in healthy adults, investigating the bio- dietary protein intake is much less. Inadequate dietary protein intake, availability of amino acids mixtures compared to natural protein, especially of essential amino acids, results in net protein breakdown. demonstrate differences in the rate of absorption in the gut [9,10]. More recently, there have been discussions on the importance of pro- Amino acids delivered as dietary protein (casein) may support tein quality in addition to quantity especially in people with special whole-body protein metabolism better than ingestion of crystalline needs [5]. L-amino acids, casein hydrolysates or soy protein [11]. Nitrogen losses are lower with intact protein compared with free amino acids. 2.1. Protein requirements for healthy individuals Recently, a new low phenylalanine whey based protein substitute has been developed for PKU, called glycomacroprotein (GMP). It is Three separateFOR terms have beenELECTRONIC used to define protein require- derived from cheeseUSE whey (naturally ONLY low in phenylalanine) and it is ments (biological need, estimated average requirement [EAR] and supplemented with phe-free indispensible amino acids. In PKU, re- recommended dietary intake [RDI]) [6]. Biological need defines the cent short term studies in mice and humans demonstrate a slower ab- quantity that is consumed by the various metabolic pathways. It is sorption, less degradation of L-amino acids and improved protein the protein required for deposition and maintenance of equilibrium retention with glycomacroprotein compared with phe-free amino of N containing substances, with variations for age, growth, pregnan- acids [12,13]. cy, lactation, injury and infection. EAR includes correction for digest- The composition of phe-free amino acids has received little study. ibility and amino acid pattern, the so called Protein Digestibility It is possible they may be deficient in one or more indispensible Corrected Amino Acid Score (PDCAAS) [1]. RDI is for a population, amino acids, their ratios may be sub-optimal or a specific component taking into account variability for both biological need and EAR. of the natural food itself may be essential for differences in L-amino Protein balance is strongly influenced by energy balance. Energy acid utilization. Amino acid patterns of phe-free amino acids designed and/or glucose depletion will result in (especially branched chain) for infants, children and adults are primarily based on that of human amino acids breakdown (gluconeogenesis) to meet minimal glucose milk. S12 A. MacDonald et al. / Molecular Genetics and Metabolism 104 (2011) S10–S18

In different formulations of phe-free amino acids, there are varia- 3. Dietary phenylalanine tions in the amino acid patterns as well the amount of essential and non essential amino acids per 100 g/amino acids, which may inﬂu- In patients with severe PKU, there is a negligible or very minimal ence bioavailability. In PKU, there is very little data comparing their capacity to oxidize phenylalanine. Phenylalanine is an indispensable, relative absorption rates and bioavailability. Also the amount of tyro- aromatic amino acid. It is essential for protein synthesis and to main- sine and the branch chain amino acids (BCAA) added to phe-free tain phenylalanine homeostasis [27] with the remainder being amino acids vary between individual products. This may be particu- hydroxylated to tyrosine. In non PKU, in vivo studies indicate that larly important as the BCAA share the same transporter in the blood 27–41% of L-phenylalanine is converted into tyrosine within 5–8h brain barrier and the gut as phenylalanine [14]. More studies are of intake [28,29]. needed to deﬁne the optimal amino composition of phe-free amino acids. 3.1. Requirements for phenylalanine in non-PKU

Defining any indispensable amino acid requirement is challenging 2.3.1. Metabolic control and protein substitute dosage as no method of assessment is entirely reliable [1]. The traditional Evidence based guidelines on protein requirement for the general technique used to determine amino acid requirements was nitrogen population are still incomplete [1], and so for people with PKU much balance. However, several new methods have been developed, partic- research is required, particularly for adults. In PKU, the few published ularly the indicator amino acid oxidation and breakpoint analysis guidelines (Guidelines of the Medical Research Council 1993; the method, using stable isotopes, and this is now considered the best ap- Ross Metabolic Nutrition Support System) advocate a higher protein proach [30]. For a combination of the aromatic amino acids phenylal- intake than for the general population [15,16]. Even so, a Cochrane re- anine and tyrosine, the WHO 2007 [1] advocate a best estimate for view concluded that there was little evidence to support this recom- adults of 25 mg/kg/day, although estimated mean requirements mendation [17], although increased dosages of phe-free amino acids range from 15 to 39 mg/kg/day [28,30]. For infants and children, the have been shown to improve natural protein (Phenylalanine) toler- WHO requirements for aromatic amino acids are given in Table 1. ance [18–20]. There is some evidence to suggest that tyrosine as a phenylalanine The metabolic and dietary handbooks advise to divide the daily in- sparing effect (up to 78% of dietary phenylalanine need [30]), but take of phe-free amino acids into at least three equal parts, and to the WHO (2007) [1] concluded it was not possible, at present, to combine the intake of natural protein with the amino acid supple- define a specific value for this. The optimal dietary ratio of phenylal- ment. The studies of Monch and MacDonald are convincing on the anine to tyrosine is suggested to be 60:40. positive effect of an even and frequent distribution of phe-free amino acids [19,21,22]. 3.2. Phenylalanine requirements in PKU Although higher dosages of phe-free amino acids appear safe, and dosage may require individual titration according to phenylalanine It is particularly difficult to define phenylalanine needs in PKU. tolerance and metabolic control, this has to be balanced against the Pencharz et al. [31] has proposed that requirements are lower than demanding regimen of administering phe-free amino acids at least those of the normal population by the amount of phenylalanine lost three times daily. They are expensive, particularly if they are being in- via obligatory oxidation [31]. Hence the phenylalanine requirement effectively utilized. for patients with PKU should represent the intake of phenylalanine required for protein synthesis and growth. Obligatory oxidation of phenylalanine is estimated to be approximately 26% of the phenylal- 2.3.1.1. Tyrosine. In PKU, tyrosine is an indispensable amino acid be- anine requirement in adult men [30]. An indicator amino acid oxida- cause it is not supplied endogenously via phenylalanine hydroxyl- tion study showed that the mean phenylalanine requirement for pre- ation or only to a very limited degree; it is important that it is pubertal children aged between 6 and 13 years with classical PKU supplemented in the diet. L-tyrosine is a precursor of several biolog- was 14 mg/kg/day (with an upper 95% confidence interval of the ically active substances, including catecholamine neurotransmitters mean of 19.5 mg/kg/day) [32]. (dopamine, norepinephrine, and epinephrine), thyroxin and mela- In PKU, the individual dietary phenylalanine tolerance is pragmat- nin skin pigments. Quantitatively, the amount of tyrosine needed ically determined and is influenced by many factors (Table 2). It is de- for catecholamine and thyroid hormone synthesis is small [23].Ty- fined as the amount of phenylalanine per kg/body weight that is rosine requirement in children with PKU aged 6 to 9 years was esti- tolerated to achieve plasma phenylalanine concentrations within a mated at 16.3 to 19.2 mg/kg/day [24].TheMRC[15] recommended target range. In PKU, generally phenylalanine tolerance/requirements that phe-free amino acids provide 100–120 mg/kg/day of tyrosine, is highest in early infancy (ranging from 55 mg/kg/day at 0–3 months which is 5 times the amount recommended for non-PKU school of age to 27 mg/kg/day at 12 months [33]). After the age of 1 year, age children [24]. Many widely used phe-free amino acids in Europe there is a slow and steady decline in tolerance per kg/body weight, contain 9 to 11% of their protein equivalent as tyrosine. However, ty- and even from the early days of treating PKU with diet it has been rosine is the least soluble of all the amino acids and if protein substi- recognized that children with classic PKU, on dietary treatment usu- tute is a powderFOR and reconstituted ELECTRONIC as a drink, it is common for ally only tolerate USE between 200 and ONLY 500 mg/day (0.2–0.5 g daily) tyrosine particles to separate, which unintentionally forms waste [34–36]. Patients with a milder form of PKU (untreated blood residueinthecontainer,sotheactualdoseoftyrosinethatsomepa- tients receive is unknown. Table 1 Treated patients with PKU have lower fasting tyrosine concen- Aromatic acid requirements for children [1]. trations, but high post-prandial plasma tyrosine concentrations Age (years) Aromatic amino acids after consuming a dose of their phe-free amino acids. Although no mg/kg/day clinical symptoms have been reported in patients with PKU, with μ 0.5 31 post prandial tyrosine concentrations greater than 200 mol/l, 1–222 some have suggested that the amount of tyrosine added to phe- 3–10 18 free amino acids is excessive [24,25]. A Cochrane review concluded 11–14 17 that no recommendations could be given on tyrosine supplementa- 15–18 16 N tion in PKU [26]. 18 15 A. MacDonald et al. / Molecular Genetics and Metabolism 104 (2011) S10–S18 S13

Table 2 3.4. Allocation of dietary phenylalanine Factors affecting phenylalanine tolerance.

• In vivo phenylalanine hydroxylation rate Phenylalanine comprises 4–6% of all dietary protein containing • Net protein catabolism foods. Practical methods for prescribing phenylalanine intake vary • Non-protein energy ratio throughout the world and it is either allocated by prescribing total • Growth rate daily phenylalanine allowance, where all phenylalanine containing • Age • Gender foods are calculated in the diet, or by phenylalanine exchange sys- • Compliance tems (i.e. pre-calculated, prescribed portions of foods all containing • Dosage of phe-free amino acids the same amount of phenylalanine) [45]. No one method is shown • BH4 treatment to be better than another, and a comparison of blood phenylalanine • Target blood phenylalanine concentration • Pregnancy control in 10 mainly European centers, indicated that blood phenylalanine control was similar in younger children despite differing die- NB: van Spronsen demonstrated no significant correlation between in vivo tary systems [46]. Although many PKU centers advocate that hydroxylation rates and phenylalanine tolerance [41], but this was in a small group of patients, almost all with classical PKU. phenylalanine containing foods are weighed, with time carers or patients are more likely to estimate portion size by ‘eye’ only [47,48]. In addition, controlled studies to develop non-weighed aids to esti- phenylalanine concentrations less than 1000 μmol/l), usually tolerate mate phenylalanine exchanges have been unsuccessful [49]. in excess of 500 mg/day dietary phenylalanine [37–39]. By comparison, in non-PKU, the third US National Health and Nutrition Examina- tion Survey (NHANES 111) demonstrated that mean daily dietary 4. Growth and body composition phenylalanine intakes for all life stages and gender groups was as high as 3.4 g/day [8]. 4.1. Longitudinal growth In PKU, van Spronsen has demonstrated that there is a clear relationship between phenylalanine tolerance at 2 years of age and at In the early years of treating PKU, the desire to achieve normal 10 years of age [40], although it is unknown how this tolerance re- phenylalanine concentrations, using very restrictive diet therapy, lates to older patients with PKU, who may have a differing target led to growth impairment [50] and malnutrition, probably with a sec- blood phenylalanine range. There is clearly a need to evaluate the ondary negative effect on intellectual function [44]. In the 1990's, phenylalanine tolerance of all patients periodically, but particularly poor linear and head circumference growth were reported in German at the times of rapid growth, changes in body composition or treat- children with mean protein intakes ranging from 2.24 g/kg/body ment change. MacLeod demonstrated in a small of group of 8 adults weight until 2 years of age and about 1.98 g/kg/body weight until with PKU, that almost all tolerated additional prescribed phenylala- 6 years (energy intakes were not reported) [51]. Further studies nine intake (by 15–173%) without significantly increasing blood phe- from the Netherlands, Germany and France also reported poor phys- nylalanine concentration [41]. van Rijn also demonstrated in 6 well ical growth, with negative height z-scores, although data on nutrient controlled adults with PKU that they tolerated more phenylalanine intakes were unreported [52–54]. In Dutch children, mean height z- than prescribed [42]. However, the results from these small studies scores deteriorated between birth until 3 years of age [52], and may not reflect the majority of adult patients. there was no relationship between growth and the severity of blood Surprisingly, there have been no longitudinal reports document- phenylalanine control [55]. In France, Dhondt et al. demonstrated im- ing life time phenylalanine intakes in patients adhering to long term proved height z-scores after relaxation of diet at 8 years of age [54]. diet therapy; there have also been few reports documenting the phe- Results from earlier studies in the United States, who had a less strict nylalanine intake of adolescent patients. Also the difference between approach to early management did not report problems with growth prescribed and actual intake of phenylalanine in patients is rarely impairment [56,57]. Moreover, when nutritional recommendations reported. It is possible that some patients are consuming a higher from the UK Medical Research Council Party on PKU [15] were fol- phenylalanine intake than prescribed, but because blood phenylala- lowed physical growth was considered normal [58,59], with higher nine concentrations are within target ranges, dietary phenylalanine protein intakes associated with the best growth [58]. More recently, intake is not measured precisely. MacDonald et al. [43] demonstrated another French centre reported growth retardation despite energy that an unlimited consumption of low phenylalanine foods (including and total protein intakes above the recommend dietary allowances a range of fruits and vegetables) increased prescribed phenylalanine (RDA's) [60]. However, mean dietary zinc intakes were low and 19 intake by 31%, without loss of dietary control, when actual consump- of 20 patients studied had plasma zinc deficiency [60]. There was no tion of all food intake was calculated. direct correlation between zinc intake, status and growth. In the most recent growth study, Huemer et al. demonstrated normal growth patterns with total protein intakes 20 to 40% above the 3.3. Phenylalanine deficiency RDA's [61]. In addition, body composition did not differ significantly from controls [61]. More interestingly, fat free mass correlated with It has beenFOR many years since thereELECTRONIC have been reports of symptom- natural protein USE intake [61] which ONLY was in good accordance with a atic phenylalanine deficiency in PKU. Hanley et al. [44] reviewed the Dutch study. This underlined the importance of the protein quality symptoms associated with deficiency. These include generalized in order to promote an optimal head circumference growth in the amino aciduria, anorexia, listlessness, alopecia, perineal rash, poor first 3 years of life [62]. and variable growth in preschool children and even death. It is possi- Considering the lack of recent data on growth on current dietary ble that young children, in particular, are at risk of phenylalanine de- management and the need to clarify the really impact of a semi-syn- ficiency if they have excellent blood phenylalanine control but refuse thetic diet, the definition of a reliable biochemical parameter to mea- most dietary phenylalanine sources because of food faddiness or sure protein status to classify patients at risk would be helpful. Low feeding difficulties. There is little data examining the ‘actual’ dietary prealbumin concentrations seem to be related with growth restric- intakes of toddlers with PKU. There is also a risk of phenylalanine de- tion in classical PKU [63] while patients with milder forms of PKU ficiency in maternal PKU, particularly during the third trimester when may also be at risk of protein insufficiency, especially if not taking ad- the fetal liver metabolizes the maternal phenylalanine and fetal equate amounts of protein substitute [64]. Guidance on assessment of growth phenylalanine requirements increase. protein status is given in Table 3. S14 A. MacDonald et al. / Molecular Genetics and Metabolism 104 (2011) S10–S18

Table 3 patients with PKU and controls [60,61], while a recent study, using Assessment tools for assessing protein status. Plethysmography, indicated increased body fat percentage especially Assessment tools for assessing protein status in females above 11 years of age, despite no differences between BMI [93]. Dietary records to assess Calculating dietary protein and energy intake from food protein intake records is cheap and easy but unreliable because of differences in compliance, over/underestimating 4.4. The long term impact of early nutrition reported intake, differences in food tables and food composition labels on food packaging Patients with PKU are exposed from an early age to an abnormal Biochemical markers • Hepatic Transport Proteins (total protein/albumin, prealbumin, retinol binding protein) dietary intake. It is unknown if a higher phe-free amino acids intake • Plasma amino acids (prandial and post prandial) in infancy will have any effect on the development of future obesity, • IGF-1 similar to the non-PKU population [94–96]. Any supplementation of Anthropometry Growth, height for age (target height) and head energy intake in early infancy may increase the risk for other meta- circumference b2 years of age bolic disturbances in the later life [97]. Assessment of lean body Densitometry, DXA [68], Bioelectrical impedance mass analysis [86–89] 5. Micronutrient status in PKU

In a low phenylalanine diet, animal proteins act as the vectors for 4.2. Anthropometry and body composition measurements many micronutrients, and to avoid their deficiency, vitamins, minerals and trace minerals must be supplemented in the diet (mainly Comparing growth measurements to a “good reference” is impor- through the phe-free amino acids). In PKU, the micronutrient status tant to determine if patients are receiving adequate nutrition, al- is influenced by many factors: type of diet, amount of natural protein though choosing the most appropriate growth charts is problematic. [animal or vegetable origin], BH4 treatment, metabolic control, mi- When national data is unavailable, care should be taken when com- cronutrient composition of the phe-free amino acids, and frequency paring the growth of PKU patients with reference growth charts as of their dosage. Patients are at risk of developing micronutrient defi- they only represent how a cohort of children grew in a particular ciency if 1) they fail to adhere to their phe-free amino acids, 2) have place and time. However, even the growth data from a specific popu- mild hyperphenylalaninemia, treated with a low protein vegan diet lation may vary widely when different criteria are used to collect the only (without phe-free amino acids) 3) have either relaxed or dataset population and varying cut-offs are applied to define optimal stopped their dietary treatment but fail to eat animal protein foods growth [65]. The WHO (2007) [66] growth charts should be consid- and 4) are treated by BH4 in association with a protein restricted ered as a standard rather than a reference, since they describe growth diet without supplementation with phe-free amino acids. expectations of breast fed and healthy children (aged 0–59 months) under optimal environmental and health conditions [67]. 5.1. The vitamins There is a growing need to measure body composition in children with PKU. Measurement of body composition may improve the ca- Vitamin D status is better in diet adherent than non adherent pa- pacity to tailor nutritional management directly to lean mass as tients with PKU [98]. This is related to the vitamin D content of the weight may be a poor proxy for this [68]. Furthermore, with the rise phe-free amino acids (50–250 IU per 10 g of amino acids equivalent), in childhood obesity in PKU and non-PKU children, it has become in- which, thereby, provides a consistent and regular source of this nutri- creasingly important to measure body fatness in children and identify ent. Interestingly, in a single study in PKU, the bone mineral density those who are at greatest risk of ‘co-morbidities’ of obesity. However, was low in patients on a strict low phenylalanine diet (with a high obesity in pediatrics is not easy to define, given that different criteria and adequate intake of calcium and vitamin D) compared to a relaxed exist [International Obesity Task Force (IOTF); World Health Organi- diet (with lower calcium and vitamin D intakes) [98]. This study sug- zation (WHO); Centers for Disease Control and Prevention (CDC) gests that the observed impairment in peak bone mass is related to and National Data] [69]. abnormalities in bone metabolism [99] rather than calcium or vita- The measurement of body composition may include direct or indi- min D deficiency in PKU [98]. Some studies highlight the role of rect measurements of body fat, lean body mass and bone mass. A osteoclastogenesis in the pathogenesis of bone disease in PKU large body of consistent evidence suggests that body mass index is a [99,100]. practical and valid [70] method of measuring obesity in children over 2 years of age. However, when accurately measured [71], waist 5.2. The group B vitamins circumference appears to be a good predictor of intra-abdominal fat accumulation in adults [72,73], but also in children and adolescents Vitamin B12 deficiency is common in older patients with PKU [74–76]. Abdominal fat should be viewed as a risk factor for later met- [101–103] and this could potentially cause neurological impairment. abolic and heart diseases [77,78], increasing the risk for complications As the main vitamin B12 sources are meat and seafood, patients on di- in adulthood [79]. Waist circumference percentiles are available etary treatment but without adequate vitamin B12 and cobalamin [75,80–84], butFOR there is no agreement ELECTRONIC on the cut-offs to diagnose ab- from either phe-free USE amino acids or aONLY separate vitamin/mineral sup- dominal obesity [85]. plement are likely to have a low intake [104–106]. In contrast, the cobalamin content of phe-free amino acids is high (from 0.65 to

4.3. Overweight and obesity trends in patients with PKU 9.4 μg/10 g of amino acids equivalents), so the risk of vitamin B12 deficiency is minimal if adherence with phe-free amino acids is ade- The first reports of overweight and obesity in PKU appeared in the quate. Consequently, patients at particular risk of cobalamin late 1970's and 80's [56,90]. In later studies, others have reported the deficiency are older patients, who are well known to be poorly com- same tendency of overweight in PKU [57,91]. An Italian study group, pliant with dietary treatment or patients who eat a ‘normal’ diet that in a sample of 97 patients diagnosed by neonatal screening, conclud- contains minimal vitamin B12 sources [101]. As vitamin B12 deficiency ed that early BMI rebound was associated with overweight [92].In causes neurological impairment, it is essential it is detected before the the same study, the prevalence of overweight at 2 years of age was appearance of clinical signs. Therefore, it is important to assess

19.6%, increasing to 24.7% by the age of 8 years. Some studies have the annual vitamin B12 status of all patients, regardless of severity of not shown any signiﬁcant differences in body composition between dietary restriction. However vitamin B12 concentration within the A. MacDonald et al. / Molecular Genetics and Metabolism 104 (2011) S10–S18 S15

Fig. 1. Cross sectional erythrocytes folates (n=136) in a cohort of 74 French PKU patients aged from 3 months to 45 years: demonstrating 25% (n=24) of concentrations were above laboratory reference range (330–1200 nmol/l). reference range does not automatically imply an adequate vitamin intake but an increased mean vitamin B6 intake. The mean dietary vi-

B12 status and therefore measuring serum MMA, or alternatively plas- tamin B6: protein intake of PKU patients was over two fold the ma homocysteine, is advised to diagnose functional vitamin B12 defi- amount of the control group. Mean plasma pyridoxal 5′-phosphate ciency [107]. (PLP) and PLP/total B6 in patients were significantly increased com- In contrast, folate deficiency has not been described in PKU. Phe- pared with control subjects suggesting a reduced turnover of pyri- free amino acids are supplemented with substantial amounts of folic doxine in PKU patients [107]. acid (24 to 130 μg per 10 g of amino acid equivalent), and so the recommended daily intake/reference nutrient intakes can be 5.3. The minerals exceeded well in excess of requirements. This has been demonstrated by high plasma and erythrocyte folate in a large French cohort from The trace mineral status has been widely studied in PKU, but par- one clinic of in one PKU patients (F Feillet personal data: Fig. 1). As ticularly zinc and selenium, as they are found in animal protein foods, there is now concern about the role of excess folate as a potential can- and they are involved in the anti-oxidant system [111]. cer factor [108], it is unknown if the apparent excessive folate supple- Although some studies demonstrate a normal zinc status [112], mentation of phe-free amino acids is safe. Therefore, it is essential the others indicate deficiency in PKU [113,114]. The zinc content of folate content of phe-free amino acids is re-evaluated; and it con- phe-free amino acids varies from 2 mg to 5 mg per 10 g of amino siders the contribution from a low phenylalanine diet. acid equivalent. If a patient only consumed 20 g/daily protein equiva-

There are very few studies examining pyridoxine (vitamin B6) sta- lent from phe-free amino acids, at least 4 to 10 mg/daily of zinc is tus in PKU [102,103,109,110] although it is found in a wide variety of obtained, which should provide an adequate intake. However, despite high protein foods. Compared with control subjects, Prince et al. relatively high dietary intakes, zinc deﬁciency is still described [114]. found that PKU patients had a signiﬁcantly reduced mean protein In a cohort from France, 70 of 146 (48%) measurements of plasma

FOR ELECTRONIC USE ONLY

Fig. 2. Cross sectional plasma selenium concentrations (n=153) in a cohort of 74 French PKU patients aged 0.3 to 45 years. 51% (n=78) of concentrations were below the lower laboratory reference range (0.75–1.15 μmol/l). S16 A. MacDonald et al. / Molecular Genetics and Metabolism 104 (2011) S10–S18

Table 4 • There is a growing need to measure body composition routinely in The status of vitamin and mineral rarely reported in PKU. children with PKU. This may improve the capacity to tailor nutri- Vitamins/minerals Supporting evidence to define status tional management directly to lean mass. • With the rise in childhood obesity, it is important to routinely mea- Fat soluble vitamins Retinol and tocopherol No evidence of deficiency sure body fatness in children and identify those who are at greatest risk of ‘co-morbidities’ of obesity. Water soluble vitamins • There is a need for international consensus on the dietary and nutri- fi Thiamin No evidence of de ciency [110] tional treatment of PKU. Riboflavin Inadequately studied

Minerals Iron Studies either demonstrate no deficiency [123] or 7. Conclusions poor status [124–126]. In iron deficient patients, no correlation has been demonstrated between iron intake and any index of iron status [126]. Nutrition and dietary management in PKU must continue to develop with heightened momentum and in conjunction with the new non-diet treatments. It is important to collect robust evidence about zinc were below the laboratory reference range of b10.7 μmol/l [F the protein, amino acid, micronutrient requirements together with Feillet, personal observation]. These results suggest that the bioavail- full knowledge about the growth and body composition of patients ability of zinc is sub-optimal particularly when the major dietary with PKU. This will provide essential data to act as a benchmark to source is added to phe-free amino acids and not bound by natural compare the outcome of alternative treatments. International stan- protein. In addition, it is well established that zinc can inhibit gut ab- dards should be introduced to regulate the composition of phe-free sorption of copper, particularly because it is used specifically for this amino acids and many current and accepted dietary approaches purpose in Wilson disease [115]. Further studies are required com- need reassessment to ensure that nutritional care is delivered to the paring the bioavailability and efficacy of different sources of zinc sup- highest level. plementation added to phe-free amino acids and their effect on zinc and the status of other trace metals. References Selenium deficiency is, perhaps one of the most commonly reported nutritional problems in PKU [116–120] although in some [1] WHO/FAO/UNU, Protein and amino acid requirements in human nutrition, of the older studies no selenium was added to phe-free amino acids. Report of a joint WHO/FAO/UNU expert consultation, World Health Organ Tech Rep Ser., vol. 935, 2007. In PKU, the amount of selenium in phe-free amino acids varies from [2] H. Daniel, Molecular and integrative physiology of intestinal peptide transport, 6to16μg per 10 g of protein equivalent from the protein substitute. Annu. Rev. Physiol. 66 (2004) 361–384. As the recommended daily requirements/reference nutrient intake [3] M. Daenzer, K.J. Petzke, B.J. Bequette, C.C. Metges, Whole-body nitrogen and μ μ splanchnic amino acid metabolism differ in rats fed mixed diets containing [121] varies from 1 g/kg/day in childhood to 50 g/day in adulthood, casein or its corresponding amino acid mixture, J. Nutr. 131 (2001) 1965–1972. the selenium content of the L-amino acid supplements may be inade- [4] H. Fouillet, F. Mariotti, C. Gaudichon, C. Bos, D. Tome, Peripheral and splanchnic quate. In the French cohort, 51% of plasma selenium concentrations metabolism of dietary nitrogen are differently affected by the protein source in humans as assessed by compartmental modeling, J. Nutr. 132 (2002) 125–133. were below the laboratory reference range using phe-free amino [5] F.E. Viteri, INCAP studies of energy, amino acids, and protein, Food Nutr. Bull. 31 acids with added selenium. [F Feillet, personal observation, Fig. 2]. (2010) 42–53. Moreover, gut absorption of selenium is optimized when its main di- [6] P.J. Reeds, P.J. Garlick, Protein and amino acid requirements and the composition – etary source is animal protein. It is known that carnitine and selenium of complementary foods, J. Nutr. 133 (2003) 2953S 2961S. [7] C. Scriver, Hyperphenylalaninemia: phenylalanine hydroxylase deficiency, in: supplementation improve the anti-oxidant status in PKU patients C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic and Molecular [116,117] and that BH4 therapy which allows a greater intake of nat- Bases of Inherited Disease, McGraw-Hill, New York, 2001, pp. 1667–1724. ural protein improves the selenium status in PKU patients [122]. [8] F.a.N.B. Institute of Medicine of the National Academies, Dietary reference intakes for energy, carbohydrate, fibre, fat, fatty acids, cholesterol, protein and The status of other micronutrients is summarized in Table 4. amino acids, The National Academies Press, Washington DC, 2002 Appendix D-14. 6. Recommendations [9] S.S. Gropper, P.B. Acosta, Effect of simultaneous ingestion of L-amino acids and whole protein on plasma amino acid and urea nitrogen concentrations in humans, JPEN J. Parenter. Enteral Nutr. 15 (1991) 48–53. [10] S.S. Gropper, D.M. Gropper, P.B. Acosta, Plasma amino acid response to ingestion of L-amino acids and whole protein, J. Pediatr. Gastroenterol. Nutr. 16 (1993) • Alternative ‘intact’ protein substitute sources should be explored 143–150. [11] D.I. Officer, E.S. Batterham, D.J. Farrell, Comparison of growth performance and and developed, as they are likely to be associated with improved nutrient retention of weaner pigs given diets based on casein, free amino acids protein utilization and dietary adherence. or conventional proteins, Br. J. Nutr. 77 (1997) 731–744. • Further controlled studies are required examining the ideal dosage [12] D.M. Ney, S.T. Gleason, S.C. van Calcar, E.L. MacLeod, K.L. Nelson, M.R. Etzel, G.M. Rice, J.A. Wolff, Nutritional management of PKU with glycomacropeptide from of phe-free amino acids particularly the relationship between dos- cheese whey, J. Inherit. Metab. Dis. 32 (2009) 32–39. age and improved tolerance of dietary phenylalanine. [13] E.L. MacLeod, M.K. Clayton, S.C. van Calcar, D.M. Ney, Breakfast with glycoma- • The amino acid profile and micronutrient composition of the phe- cropeptide compared with amino acids suppresses plasma ghrelin levels in indi- FOR ELECTRONICviduals with phenylketonuria,USE Mol. Genet. ONLY Metab. 100 (2010) 303–308. free amino acids should be re-examined to develop products that [14] F.J. van Spronsen, M.J. de Groot, M. Hoeksma, D.J. Reijngoud, M. van Rijn, Large better meet the nutritional requirements of patients with PKU. neutral amino acids in the treatment of PKU: from theory to practice, J. Inherit. The introduction of international nutritional standards to regulate Metab. Dis. 33 (2010) 671–676. the composition of phe-free amino acids is necessary. [15] Recommendations on the dietary management of phenylketonuria. Report of Medical Research Council Working Party on Phenylketonuria, Arch. Dis. Child. • Current supplementation with tyrosine may be excessive and the 68 (1993) 426–427. dosage requires further study. [16] P.B. Acosta, S. Yannicelli, Protocol l-phenylketonuria (PKU), in: Ross Laboratories O • Longitudinal data is required about the change in phenylalanine tol- (Ed.), Ross Metabolic Formula System Nutrition Support Protocols, Columbus, 2001. ‘ ’ [17] P. Rutherford, V.J. Poustie, Protein substitute for children and adults with phe- erance throughout life, together with data about the real rather nylketonuria, Cochrane Database Syst. Rev. (2005) CD004731. than prescribed phenylalanine tolerance. [18] E. Kindt, K. Motzfeldt, S. Halvorsen, S.O. Lie, Is phenylalanine requirement in in- • There is little data about the nutritional state of adult patients with fants and children related to protein intake? Br. J. Nutr. 51 (1984) 435–442. [19] A. MacDonald, G. Rylance, P. Davies, D. Asplin, S.K. Hall, I.W. Booth, Administra- PKU on diet. More data is required about their protein requirements tion of protein substitute and quality of control in phenylketonuria: a random- and micronutrient nutritional status in relationship to diet therapy. ized study, J. Inherit. Metab. Dis. 26 (2003) 319–326. A. MacDonald et al. / Molecular Genetics and Metabolism 104 (2011) S10–S18 S17

[20] A. Macdonald, A. Chakrapani, C. Hendriksz, A. Daly, P. Davies, D. Asplin, K. Hall, van den Ploeg, M.E. Rubio-Gozalbo, J.P. Sels, R.C. Sengers, H.W. de Valk, H. I.W. Booth, Protein substitute dosage in PKU: how much do young patients Termeulen, H. Zweers, F.J. van Spronsen, A survey of natural protein intake need? Arch. Dis. Child. 91 (2006) 588–593. in Dutch phenylketonuria patients: insight into estimation or measurement of [21] M.E. Herrmann, H.G. Brosicke, M. Keller, E. Monch, H. Helge, Dependence of the dietary intake, J. Am. Diet. Assoc. 108 (2008) 1704–1707. utilization of a phenylalanine-free amino acid mixture on different amounts of [49] H. Gokmen-Ozel, A. Daly, P. Davies, A. Macdonald, The development of non- single dose ingested. A case report, Eur. J. Pediatr. 153 (1994) 501–503. weighed pilot methods for estimating phenylalanine exchanges in non-phenyl- [22] E. Monch, M.E. Herrmann, H. Brosicke, A. Schoffer, M. Keller, Utilisation of amino ketonuria volunteers, J. Hum. Nutr. Diet. (2011) (July 7; Electronic publication acid mixtures in adolescents with phenylketonuria, Eur. J. Pediatr. 155 (Suppl. 1) ahead of print). (1996) S115–S120. [50] T. Mereu, Adequacy of low-phenylalanine diet, Am. J. Dis. Child. 113 (1967) [23] P.B. Pencharz, J.W. Hsu, R.O. Ball, Aromatic amino acid requirements in healthy 522–523. human subjects, J. Nutr. 137 (2007) 1576S–1578S discussion 1597S-1598S. [51] F. Schaefer, P. Burgard, U. Batzler, A. Rupp, H. Schmidt, G. Gilli, H. Bickel, H.J. [24] R. Bross, R.O. Ball, J.T. Clarke, P.B. Pencharz, Tyrosine requirements in children Bremer, Growth and skeletal maturation in children with phenylketonuria, with classical PKU determined by indicator amino acid oxidation, Am. J. Physiol. Acta Paediatr. 83 (1994) 534–541. Endocrinol. Metab. 278 (2000) E195–E201. [52] P.H. Verkerk, F.J. van Spronsen, G.P. Smit, R.C. Sengers, Impaired prenatal and [25] F.J. van Spronsen, M. van Rijn, J. Bekhof, R. Koch, P.G. Smit, Phenylketonuria: ty- postnatal growth in Dutch patients with phenylketonuria. The National PKU rosine supplementation in phenylalanine-restricted diets, Am. J. Clin. Nutr. 73 Steering Committee, Arch. Dis. Child. 71 (1994) 114–118. (2001) 153–157. [53] J. Weglage, J.H. Bramswig, H.G. Koch, S. Karassalidou, K. Ullrich, Growth in pa- [26] D. Webster, J. Wildgoose, Tyrosine supplementation for phenylketonuria, tients with phenylketonuria, Eur. J. Pediatr. 153 (1994) 537–538. Cochrane Database Syst. Rev. 4 (8) (2010) CD001507. [54] J.L. Dhondt, C. Largilliere, L. Moreno, J.P. Farriaux, Physical growth in patients [27] J. Donlon, H. Levy, C. Scriver, Hyperphenylalanine hydroxylase deficiency, in: A. with phenylketonuria, J. Inherit. Metab. Dis. 18 (1995) 135–137. Beaudet, B. Vogelstein, K.A. Kinzler, S. Antoniadis, A. Ballabio, D. Valle (Eds.), The [55] F.J. van Spronsen, P.H. Verkerk, M. van Houten, G.P. Smit, S.B. van der Meer, H.D. Online Metabolic and Molecular Basis of Inherited Disease, McGraw-Hill, 2008, Bakker, R.C. Sengers, Does impaired growth of PKU patients correlate with the pp. 1–150. Companies.www.ommbid.com. strictness of dietary treatment? National Dutch PKU Steering Committee, Acta [28] A. Basile-Filho, L. Beaumier, A.E. El-Khoury, Y.M. Yu, M. Kenneway, R.E. Gleason, Paediatr. 86 (1997) 816–818. V.R. Young, Twenty-four-hour L-[1-(13)C]tyrosine and L-[3,3-(2)H2]phenylalanine [56] V.A. Holm, R.A. Kronmal, M. Williamson, A.F. Roche, Physical growth in phenyl- oral tracer studies at generous, intermediate, and low phenylalanine intakes to es- ketonuria: II. Growth of treated children in the PKU collaborative study from timate aromatic amino acid requirements in adults, Am. J. Clin. Nutr. 67 (1998) birth to 4 years of age, Pediatrics 63 (1979) 700–707. 640–659. [57] M.A. McBurnie, R.A. Kronmal, V.E. Schuett, R. Koch, C.G. Azeng, Physical growth [29] M. Sanchez, A.E. El-Khoury, L. Castillo, T.E. Chapman, A. Basile Filho, L. Beaumier, of children treated for phenylketonuria, Ann. Hum. Biol. 18 (1991) 357–368. V.R. Young, Twenty-four-hour intravenous and oral tracer studies with L-[1-13C] [58] P.B. Acosta, S. Yannicelli, Protein intake affects phenylalanine requirements phenylalanine and L-[3,3-2H2]tyrosine at a tyrosine-free, generous phenylalanine and growth of infants with phenylketonuria, Acta Paediatr. Suppl. 407 intake in adults, Am. J. Clin. Nutr. 63 (1996) 532–545. (1994) 66–67. [30] G.A. Zello, P.B. Pencharz, R.O. Ball, Phenylalanine flux, oxidation, and conversion [59] P.B. Acosta, S. Yannicelli, B. Marriage, C. Mantia, B. Gaffield, M. Porterfield, M. to tyrosine in humans studied with L-[1-13C]phenylalanine, Am. J. Physiol. 259 Hunt, N. McMaster, L. Bernstein, P. Parton, M. Kuehn, V. Lewis, Nutrient intake (1990) E835–E843. and growth of infants with phenylketonuria undergoing therapy, J. Pediatr. Gas- [31] P.B. Pencharz, R.O. Ball, Different approaches to define individual amino acid re- troenterol. Nutr. 27 (1998) 287–291. quirements, Annu. Rev. Nutr. 23 (2003) 101–116. [60] D. Dobbelaere, L. Michaud, A. Debrabander, S. Vanderbecken, F. Gottrand, D. [32] G. Courtney-Martin, R. Bross, M. Raffi, J.T. Clarke, R.O. Ball, P.B. Pencharz, Phenyl- Turck, J.P. Farriaux, Evaluation of nutritional status and pathophysiology of alanine requirement in children with classical PKU determined by indicator growth retardation in patients with phenylketonuria, J. Inherit. Metab. Dis. 26 amino acid oxidation, Am. J. Physiol. Endocrinol. Metab. 283 (2002) (2003) 1–11. E1249–E1256. [61] M. Huemer, C. Huemer, D. Moslinger, D. Huter, S. Stockler-Ipsiroglu, Growth [33] P.B. Acosta, E. Wenz, M. Williamson, Nutrient intake of treated infants with phe- and body composition in children with classical phenylketonuria: results in nylketonuria, Am. J. Clin. Nutr. 30 (1977) 198–208. 34 patients and review of the literature, J. Inherit. Metab. Dis. 30 (2007) [34] H. Bickel, J. Gerrard, E.M. Hickmans, Influence of phenylalanine intake on phe- 694–699. nylketonuria, Lancet 265 (1953) 812–813. [62] M. Hoeksma, M. Van Rijn, P.H. Verkerk, A.M. Bosch, M.F. Mulder, J.B. de Klerk, T.J. [35] P.B. Acosta, C. Trahms, N.S. Wellman, M. Williamson, Phenylalanine intakes of 1- de Koning, E. Rubio-Gozalbo, M. de Vries, P.J. Sauer, F.J. van Spronsen, The intake to 6-year-old children with phenylketonuria undergoing therapy, Am. J. Clin. of total protein, natural protein and protein substitute and growth of height and Nutr. 38 (1983) 694–700. head circumference in Dutch infants with phenylketonuria, J. Inherit. Metab. Dis. [36] U. Wendel, K. Ullrich, H. Schmidt, U. Batzler, Six-year follow up of phenylalanine 28 (2005) 845–854. intakes and plasma phenylalanine concentrations, Eur. J. Pediatr. 149 (Suppl. 1) [63] G.L. Arnold, C.J. Vladutiu, R.S. Kirby, E.M. Blakely, J.M. Deluca, Protein insufficien- (1990) S13–S16. cy and linear growth restriction in phenylketonuria, J. Pediatr. 141 (2002) [37] R.H. Singh, M.E. Quirk, T.D. Douglas, M.C. Brauchla, BH(4) therapy impacts the 243–246. nutrition status and intake in children with phenylketonuria: 2-year follow- [64] J.C. Rocha, M.F. Almeida, C. Carmona, M.L. Cardoso, N. Borges, I. Soares, G. Salcedo, up, J. Inherit. Metab. Dis. 33 (2010) 689–695. M.R. Lima, I. Azevedo, F.J. van Spronsen, The use of prealbumin concentration as a [38] D.Y. Hsia, M.E. O'Flynn, Hyperphenylalanemia, Metabolism 16 (1967) 99–101. biomarker of nutritional status in treated phenylketonuric patients, Ann. Nutr. [39] A. Boneh, D.E. Francis, M. Humphrey, H.J. Upton, H.L. Peters, Three-year audit of Metab. 56 (2010) 207–211. the hyperphenylalaninaemia/phenylketonuria spectrum in Victoria, J. Paediatr. [65] Z. Mei, C.L. Ogden, K.M. Flegal, L.M. Grummer-Strawn, Comparison of the prev- Child Health 42 (2006) 496–498. alence of shortness, underweight, and overweight among US children aged [40] F.J. van Spronsen, M. van Rijn, B. Dorgelo, M. Hoeksma, A.M. Bosch, M.F. Mulder, 0 to 59 months by using the CDC 2000 and the WHO 2006 growth charts, J. J.B. de Klerk, T. de Koning, M.E. Rubio-Gozalbo, M. de Vries, P.H. Verkerk, Phenyl- Pediatr. 153 (2008) 622–628. alanine tolerance can already reliably be assessed at the age of 2 years in pa- [66] M. de Onis, A. Onyango, E. Borghi, A. Siyam, C. Nishida, J. Siekmann, Develop- tients with PKU, J. Inherit. Metab. Dis. 32 (2009) 27–31. ment of a WHO growth reference for school-aged children and adolescents, [41] E.L. MacLeod, S.T. Gleason, S.C. van Calcar, D.M. Ney, Reassessment of phenylal- Bull. World Health Organ. 85 (2007) 660–667. anine tolerance in adults with phenylketonuria is needed as body mass changes, [67] L. Anna, Enrolment and baseline characteristics in the WHO multicentre growth Mol. Genet. Metab. 98 (2009) 331–337. reference study, Acta Paediatr. Suppl. 450 (2006) 7–15. [42] M. van Rijn, M. Hoeksma, P.J. Sauer, P. Modderman, D.J. Reijngoud, F.J. van [68] J.C. Wells, M.S. Fewtrell, Measuring body composition, Arch. Dis. Child. 91 Spronsen, Adult patients with well-controlled phenylketonuria tolerate inciden- (2006) 612–617. tal additional intake of phenylalanine, Ann. Nutr. Metab. 58 (2011) 94–100. [69] I.A. Lang, R.R. Kipping, R. Jago, D.A. Lawlor, Variation in childhood and [43] A. MacDonald, G. Rylance, S.K. Hall, D. Asplin, I.W. Booth, Factors affecting the adolescent obesity prevalence defined by international and country-specific variation in plasma phenylalanine in patients with phenylketonuria on diet, criteria in England and the United States, Eur. J. Clin. Nutr. 65 (2011) Arch. Dis.FOR Child. 74 (1996) 412– 417.ELECTRONIC143–150. USE ONLY [44] W.B. Hanley, L. Linsao, W. Davidson, C.A. Moes, Malnutrition with early treat- [70] J.J. Reilly, J. Kelly, D.C. Wilson, Accuracy of simple clinical and epidemiological ment of phenylketonuria, Pediatr. Res. 4 (1970) 318–327. definitions of childhood obesity: systematic review and evidence appraisal, [45] K. Ahring, A. Belanger-Quintana, K. Dokoupil, H. Gokmen Ozel, A.M. Lammardo, Obes. Rev. 11 (2010) 645–655. A. MacDonald, K. Motzfeldt, M. Nowacka, M. Robert, M. van Rijn, Dietary man- [71] M.C. Rudolf, J. Walker, T.J. Cole, What is the best way to measure waist circum- agement practices in phenylketonuria across European centres, Clin. Nutr. 28 ference? Int. J. Pediatr. Obes. 2 (2007) 58–61. (2009) 231–236. [72] W. Shen, M. Punyanitya, J. Chen, D. Gallagher, J. Albu, X. Pi-Sunyer, C.E. Lewis, C. [46] K. Ahring, A. Belanger-Quintana, K. Dokoupil, H. Gokmen-Ozel, A.M. Lammardo, Grunfeld, S. Heshka, S.B. Heymsfield, Waist circumference correlates with met- A. MacDonald, K. Motzfeldt, M. Nowacka, M. Robert, M. van Rijn, Blood phenyl- abolic syndrome indicators better than percentage fat, Obesity (Silver Spring) alanine control in phenylketonuria: a survey of 10 European centres, Eur. J. Clin. 14 (2006) 727–736. Nutr. 65 (2011) 275–278. [73] J. Wang, Waist circumference: a simple, inexpensive, and reliable tool that [47] A. Macdonald, P. Davies, A. Daly, V. Hopkins, S.K. Hall, D. Asplin, C. Hendriksz, A. should be included as part of physical examinations in the doctor's office, Am. Chakrapani, Does maternal knowledge and parent education affect blood phe- J. Clin. Nutr. 78 (2003) 902–903. nylalanine control in phenylketonuria? J. Hum. Nutr. Diet. 21 (2008) 351–358. [74] P. Brambilla, P. Manzoni, S. Sironi, P. Simone, A. Del Maschio, B. di Natale, G. [48] M. van Rijn, J. Jansma, A. Brinksma, H.D. Bakker, G.H. Boers, E. Carbasius-Weber, Chiumello, Peripheral and abdominal adiposity in childhood obesity, Int. J. A.C. Douwes, A. van den Herberg, N.M. Ter Horst, J.B. de Klerk, T.J. de Koning, L. Obes. Relat. Metab. Disord. 18 (1994) 795–800. S18 A. MacDonald et al. / Molecular Genetics and Metabolism 104 (2011) S10–S18

[75] P.T. Katzmarzyk, S.R. Srinivasan, W. Chen, R.M. Malina, C. Bouchard, G.S. Berenson, cells in phenylketonuria, J. Inherit. Metab. Dis. (2008) (#121. Electronic publica- Body mass index, waist circumference, and clustering of cardiovascular disease risk tion 17 October). factors in a biracial sample of children and adolescents, Pediatrics 114 (2004) [101] W.B. Hanley, A. Feigenbaum, J.T. Clarke, W. Schoonheyt, V. Austin, Vitamin B12 e198–e205. deficiency in adolescents and young adults with phenylketonuria, Lancet 342 [76] M.I. Goran, C. Lane, C. Toledo-Corral, M.J. Weigensberg, Persistence of pre-diabe- (1993) 997. tes in overweight and obese Hispanic children: association with progressive in- [102] A.M. Hvas, E. Nexo, J.B. Nielsen, Vitamin B12 and vitamin B6 supplementation is sulin resistance, poor beta-cell function, and increasing visceral fat, Diabetes 57 needed among adults with phenylketonuria (PKU), J. Inherit. Metab. Dis. 29 (2008) 3007–3012. (2006) 47–53. [77] A.M. Fredriks, S. van Buuren, M. Fekkes, S.P. Verloove-Vanhorick, J.M. Wit, Are [103] K.H. Schulpis, G.A. Karikas, E. Papakonstantinou, Homocysteine and other vascu- age references for waist circumference, hip circumference and waist-hip ratio lar risk factors in patients with phenylketonuria on a diet, Acta Paediatr. 91 in Dutch children useful in clinical practice? Eur. J. Pediatr. 164 (2005) 216–222. (2002) 905–909. [78] J.C. Wells, M.S. Fewtrell, Is body composition important for paediatricians? Arch. [104] M. Robinson, F.J. White, M.A. Cleary, E. Wraith, W.K. Lam, J.H. Walter, Increased Dis. Child. 93 (2008) 168–172. risk of vitamin B12 deficiency in patients with phenylketonuria on an unrest- [79] N. Cameron, L.L. Jones, P.L. Griffiths, S.A. Norris, J.M. Pettifor, How well do waist ricted or relaxed diet, J. Pediatr. 136 (2000) 545–547. circumference and body mass index reflect body composition in pre-pubertal [105] W.B. Hanley, A.S. Feigenbaum, J.T. Clarke, W.E. Schoonheyt, V.J. Austin, Vitamin children? Eur. J. Clin. Nutr. 63 (2009) 1065–1070. B12 deficiency in adolescents and young adults with phenylketonuria, Eur. J. [80] J.R. Fernandez, D.T. Redden, A. Pietrobelli, D.B. Allison, Waist circumference per- Pediatr. 155 (Suppl. 1) (1996) S145–S147. centiles in nationally representative samples of African–American, European– [106] T.T. Aung, A. Klied, J. McGinn, T. McGinn, Vitamin B12 deficiency in an adult phe- American, and Mexican–American children and adolescents, J. Pediatr. 145 nylketonuric patient, J. Inherit. Metab. Dis. 20 (1997) 603–604. (2004) 439–444. [107] I. Vugteveen, M. Hoeksma, A.L. Monsen, M.R. Fokkema, D.J. Reijngoud, M. van [81] S.E. Messiah, K.L. Arheart, S.E. Lipshultz, T.L. Miller, Ethnic group differences in Rijn, F.J. van Spronsen, Serum vitamin B12 concentrations within reference waist circumference percentiles among U.S. children and adolescents: estimates values do not exclude functional vitamin B12 deficiency in PKU patients of var- from the 1999–2008 national health and nutrition examination surveys, Metab. ious ages, Mol. Genet. Metab. 102 (2011) 13–17. Syndr. Relat. Disord. (2011) (April 5; Electronic publication ahead of print). [108] D.A. Kennedy, S.J. Stern, M. Moretti, I. Matok, M. Sarkar, C. Nickel, G. Koren, Fo- [82] P.T. Katzmarzyk, Waist circumference percentiles for Canadian youth 11–18y of late intake and the risk of colorectal cancer: a systematic review and meta-anal- age, Eur. J. Clin. Nutr. 58 (2004) 1011–1015. ysis, Cancer Epidemiol. 35 (2011) 2–10. [83] H.D. McCarthy, K.V. Jarrett, H.F. Crawley, The development of waist circumfer- [109] A.P. Prince, J.E. Leklem, Vitamin B-6 status of school-aged patients with phenyl- ence percentiles in British children aged 5.0–16.9 y, Eur. J. Clin. Nutr. 55 ketonuria, Am. J. Clin. Nutr. 60 (1994) 262–268. (2001) 902–907. [110] V.M. Kodentsova, O.A. Vrzhesinskaia, S.N. Denisova, V.B. Spirichev, Metabolism [84] J.C. Eisenmann, Waist circumference percentiles for 7- to 15-year-old Australian of vitamins B1 and B2 during phenylketonuria, Vopr. Med. Khim. 45 (1999) children, Acta Paediatr. 94 (2005) 1182–1185. 150–157. [85] P. Zimmet, K.G. Alberti, F. Kaufman, N. Tajima, M. Silink, S. Arslanian, G. Wong, P. [111] K.H. Schulpis, S. Tsakiris, G.A. Karikas, M. Moukas, P. Behrakis, Effect of diet on Bennett, J. Shaw, S. Caprio, The metabolic syndrome in children and adolescents plasma total antioxidant status in phenylketonuric patients, Eur. J. Clin. Nutr. —an IDF consensus report, Pediatr. Diab. 8 (2007) 299–306. 57 (2003) 383–387. [86] U.G. Kyle, I. Bosaeus, A.D. De Lorenzo, P. Deurenberg, M. Elia, J.M. Gomez, B.L. [112] R.M. Fisberg, M.E. Da Silva-Femandes, M. Fisberg, B.J. Schmidt, Plasma zinc, cop- Heitmann, L. Kent-Smith, J.C. Melchior, M. Pirlich, H. Scharfetter, A.M. Schols, per, and erythrocyte superoxide dismutase in children with phenylketonuria, C. Pichard, Bioelectrical impedance analysis—part I: review of principles and Nutrition 15 (1999) 449–452. methods, Clin. Nutr. 23 (2004) 1226–1243. [113] P.B. Acosta, P.M. Fernhoff, H.S. Warshaw, K.M. Hambidge, A. Ernest, E.R. McCabe, [87] U.G. Kyle, I. Bosaeus, A.D. De Lorenzo, P. Deurenberg, M. Elia, J. Manuel Gomez, B. L.J. Elsas II, Zinc and copper status of treated children with phenylketonuria, Lilienthal Heitmann, L. Kent-Smith, J.C. Melchior, M. Pirlich, H. Scharfetter, M.W. JPEN J. Parenter. Enteral. Nutr. 5 (1981) 406–409. J.S.A., C. Pichard, Bioelectrical impedance analysis—part II: utilization in clinical [114] J.R. Barretto, L.R. Silva, M.E. Leite, N. Boa-Sorte, H. Pimentel, A.C. Purificacao, G. practice, Clin. Nutr. 23 (2004) 1430–1453. Carvalho, M.I. Fontes, T. Amorim, Poor zinc and selenium status in phenylketon- [88] M.Y. Jaffrin, Body composition determination by bioimpedance: an update, Curr. uric children and adolescents in Brazil, Nutr. Res. 28 (2008) 208–211. Opin. Clin. Nutr. Metab. Care 12 (2009) 482–486. [115] J. Horvath, P. Beris, E. Giostra, P.Y. Martin, P.R. Burkhard, Zinc-induced copper [89] U.G. Kyle, A. Piccoli, C. Pichard, Body composition measurements: interpretation deficiency in Wilson disease, J. Neurol. Neurosurg. Psychiatry 81 (2010) finally made easy for clinical use, Curr. Opin. Clin. Nutr. Metab. Care 6 (2003) 1410–1411. 387–393. [116] C. Sierra, M.A. Vilaseca, D. Moyano, N. Brandi, J. Campistol, N. Lambruschini, F.J. [90] J.E. White, R.A. Kronmal, P.B. Acosta, Excess weight among children with phenyl- Cambra, R. Deulofeu, A. Mira, Antioxidant status in hyperphenylalaninemia, ketonuria, J. Am. Coll. Nutr. 1 (1982) 293–303. Clin. Chim. Acta 276 (1998) 1–9. [91] P.B. Acosta, S. Yannicelli, R. Singh, S. Mofidi, R. Steiner, E. DeVincentis, E. Jurecki, [117] A. Sitta, C.S. Vanzin, G.B. Biancini, V. Manfredini, A.B. de Oliveira, C.A. Wayhs, L. Bernstein, S. Gleason, M. Chetty, B. Rouse, Nutrient intakes and physical G.O. Ribas, L. Giugliani, I.V. Schwartz, D. Bohrer, S.C. Garcia, M. Wajner, C.R. growth of children with phenylketonuria undergoing nutrition therapy, J. Am. Vargas, Evidence that L-carnitine and selenium supplementation reduces oxi- Diet. Assoc. 103 (2003) 1167–1173. dative stress in phenylketonuric patients, Cell. Mol. Neurobiol. 3 (2011) [92] S. Scaglioni, E. Verduci, L. Fiori, A.M. Lammardo, S. Rossi, G. Radaelli, E. Riva, M. 429–436. Giovannini, Body mass index rebound and overweight at 8 years of age in [118] I. Lombeck, K. Kasperek, L.E. Feinendegen, H.J. Bremer, Serum-selenium concen- hyperphenylalaninaemic children, Acta Paediatr. 93 (2004) 1596–1600. trations in patients with maple-syrup-urine disease and phenylketonuria under [93] M. Albersen, M. Bonthuis, N.M. de Roos, D.A. van den Hurk, E. Carbasius Weber, dieto-therapy, Clin. Chim. Acta 64 (1975) 57–61. M.M. Hendriks, M.G. de Sain-van der Velden, T.J. de Koning, G. Visser, Whole [119] S.S. Gropper, P.B. Acosta, N. Clarke-Sheehan, E. Wenz, M. Cheng, R. Koch, Trace body composition analysis by the BodPod air-displacement plethysmography element status of children with PKU and normal children, J. Am. Diet. Assoc. method in children with phenylketonuria shows a higher body fat percentage, 88 (1988) 459–465. J. Inherit. Metab. Dis. (2010) (Jun 24; Electronic publication ahead of print). [120] G. Darling, P. Mathias, M. O'Regan, E. Naughten, Serum selenium levels in indi- [94] M.F. Rolland-Cachera, M. Deheeger, M. Akrout, F. Bellisle, Influence of macronu- viduals on PKU diets, J. Inherit. Metab. Dis. 15 (1992) 769–773. trients on adiposity development: a follow up study of nutrition and growth [121] Department of Health, Dietary reference values for food, energy and nutrients from 10 months to 8 years of age, Int. J. Obes. Relat. Metab. Disord. 19 (1995) for the United Kingdom, Report on Health and Social Subjects, HMSO, London, 573–578. 1991. [95] S. Scaglioni, C. Agostoni, R.D. Notaris, G. Radaelli, N. Radice, M. Valenti, M. Gio- [122] N. Lambruschini, B. Perez-Duenas, M.A. Vilaseca, A. Mas, R. Artuch, R. Gassio, L. vannini, E. Riva, Early macronutrient intake and overweight at five years of Gomez, A. Gutierrez, J. Campistol, Clinical and nutritional evaluation of phenyl- age, Int. J. Obes. Relat. Metab. Disord. 24 (2000) 777–781. ketonuric patients on tetrahydrobiopterin monotherapy, Mol. Genet. Metab. 86 [96] I. Gunnarsdottir, I. Thorsdottir, Relationship between growth and feeding in in- (Suppl. 1) (2005) S54–S60. fancy and body mass index at the age of 6 years, Int. J. Obes. Relat. Metab. Disord. [123] P.B. Acosta, S. Yannicelli, Plasma micronutrient concentrations in infants under- 27 (2003) 1523FOR–1527. ELECTRONICgoing therapy forUSE phenylketonuria, Biol. ONLY Trace Elem. Res. 67 (1999) 75–84. [97] M.S. Patel, M. Srinivasan, S.G. Laychock, Metabolic programming: role of nutri- [124] B.D. Miranda da Cruz, H. Seidler, K. Widhalm, Iron status and iron supplementation in the immediate postnatal life, J. Inherit. Metab. Dis. 32 (2009) 218–228. tion in children with classical phenylketonuria, J. Am. Coll. Nutr. 12 (1993) [98] D. Modan-Moses, I. Vered, G. Schwartz, Y. Anikster, S. Abraham, R. Segev, O. 531–536. Efrati, Peak bone mass in patients with phenylketonuria, J. Inherit. Metab. Dis. [125] J.L. Bodley, V.J. Austin, W.B. Hanley, J.T. Clarke, S. Zlotkin, Low iron stores in in- 30 (2007) 202–208. fants and children with treated phenylketonuria: a population at risk for iron- [99] I. Roato, F. Porta, A. Mussa, L. D'Amico, L. Fiore, D. Garelli, M. Spada, R. Ferracini, deficiency anaemia and associated cognitive deficits, Eur. J. Pediatr. 152 Bone impairment in phenylketonuria is characterized by circulating osteoclast (1993) 140–143. precursors and activated T cell increase, PLoS One 5 (2010) e14167. [126] P.B. Acosta, S. Yannicelli, R.H. Singh, L.J. Elsas II, S. Mofidi, R.D. Steiner, Iron status [100] F. Porta, I. Roato, A. Mussa, M. Repici, E. Gorassini, M. Spada, R. Ferracini, In- of children with phenylketonuria undergoing nutrition therapy assessed by creased spontaneous osteoclastogenesis from peripheral blood mononuclear transferrin receptors, Genet. Med. 6 (2004) 96–101. Molecular Genetics and Metabolism 104 (2011) S19–S25

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Minireview Up to date knowledge on different treatment strategies for phenylketonuria

Amaya Bélanger-Quintana a,⁎, Alberto Burlina b,1, Cary O. Harding c,2, Ania C. Muntau d,e,3 a Division of Metabolic Diseases, Pediatrics Department, Ramon y Cajal Hospital, Madrid, Spain b Division of Metabolic Diseases, Pediatrics Department, University Hospital of Padua, Padua, Italy c Molecular and Medical Genetics, Oregon Health and Science University, Portland, OR, USA d Department of Inborn Errors of Metabolism, Dr. von Hauner Children's Hospital, Ludwig-Maximilians University, Munich, Germany e Department of Molecular Pediatrics, Dr. von Hauner Children's Hospital, Ludwig-Maximilians University, Munich, Germany article info abstract

Article history: Dietary management for phenylketonuria was established over half a century ago, and has rendered an immense Received 30 May 2011 success in the prevention of the severe mental retardation associated with the accumulation of phenylalanine. Received in revised form 23 July 2011 However, the strict low-phenylalanine diet has several shortcomings, not the least of which is the burden it Accepted 5 August 2011 imposes on the patients and their families consequently frequent dietary non-compliance. Imperfect neurological Available online 16 August 2011 outcome of patients in comparison to non-PKU individuals and nutritional deﬁciencies associated to the PKU diet are other important reasons to seek alternative therapies. In the last decade there has been an impressive effort in Keywords: Phenylketonuria the investigation of other ways to treat PKU that might improve the outcome and quality of life of these patients. Tetrahydrobiopterin These studies have lead to the commercialization of sapropterin dihydrochloride, but there are still many Sapropterin questions regarding which patients to challenge with sapropterin what is the best challenge protocol and what Phenylalanine ammonia lyase could be the implications of this treatment in the long-term. Current human trials of PEGylated phenylalanine Gene therapy ammonia lyase are underway, which might render an alternative to diet for those patients non-responsive to sapropterin dihydrochloride. Preclinical investigation of gene and cell therapies for PKU is ongoing. In this manuscript, we will review the current knowledge on novel pharmacologic approaches to the treatment of phenylketonuria. © 2011 Elsevier Inc. All rights reserved.

Contents

1. Introduction ...... S19 2. Tetrahydrobiopterin (BH4) ...... S20 3. Phenylalanine ammonia lyase (PAL) ...... S22 4. Gene therapy ...... S23 5. Conclusions ...... S23 References ...... S23

1. Introduction

Phenylketonuria (PKU; OMIM 2626000) is one of the most common ⁎ CorrespondingFOR author at: Unidad de EnfermedadesELECTRONIC Metabolicas, Servicio de Pediatría, inborn errors ofUSE metabolism. It is a recessivelyONLY inherited disease caused Hospital Ramon y Cajal, Ctra. Colmenar km 9,1, Madrid 28034, Spain. Fax: +34 913368417. by mutations in the gene encoding the enzyme phenylalanine E-mail addresses: [email protected] (A. Bélanger-Quintana), hydroxylase (PAH; EC 1.14.16.1) [1].Itisalsothefirst metabolic disorder [email protected] (A. Burlina), [email protected] (C.O. Harding), fi [email protected] (A.C. Muntau). in which a toxic agent, phenylalanine (Phe), was identi ed to cause 1 Metabolic Unit, Department of Pediatrics, University Hospital, Via Guistiniani 3, mental retardation, and in which treatment was found to prevent clinical Padova 35128, Italy. symptoms. The dietary management of PKU was established 60 years 2 OHSU, Mailstop L-103, 3181 Sam Jackson Park Rd., Portland, OR 97239, USA. Fax: +1 ago [2], with the first effects of treatment published in 1953 [3].ThePKU 5034946886. diet consists in a restriction of dietary natural protein in order to 3 Departments of Inborn Errors of Metabolism and Molecular Pediatrics, Dr. von Hauner Children's Hospital, Ludwig-Maximilians University, Lindwurmstrasse 4, D-80337 Munich, minimize Phe intake. It requires supplementation with special medical Germany. Fax: +49 8951607952. formulas that supply sufficient essential aminoacids, energy, vitamins

1096-7192/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2011.08.009 S20 A. Bélanger-Quintana et al. / Molecular Genetics and Metabolism 104 (2011) S19–S25 and minerals. To avoid mental retardation, the diet should be started in 2. Tetrahydrobiopterin (BH4) the first weeks of age; therefore, neonatal screening programs are essential in the early identification of these patients. Treatment should Tetrahydrobiopterin is the natural cofactor of the PAH enzyme. The be maintained for life, as hyperphenylalaninemia in adulthood has been lack of BH4 gives rise not only to disfunction of PAH (and therefore associated with attention problems, mood instability and white matter hyperphenylalaninemia), but also to disfunction of brain tyrosine and degeneration leading to seizures and gait disturbances [4,5].During tryptophan hydroxylases, leading to the very severe neurologic pregnancy, moderately high Phe levels in the mother can cause manifestations neurotrasmitter deficiencies in the different BH4 microcephaly, mental retardation and congenital heart defects in the deficient syndromes [25]. In the 1980s, colleagues began to employ fetus [6]. oral BH4 challenges to detect neonates with inherited BH4 deficien- Dietary treatment has been very effective in preventing severe cies [26]. Unfortunately, these short-term challenge protocols did not mental retardation, but it has several shortcomings [5,7]. Even though typically detect individuals with of BH4-responsive PAH deficiency, medical formulas have improved in nutritional quality and palatabil- which requires test protocols of longer duration and higher BH4 doses ity over the years, the severely restrictive diet necessary for the [25]. It was not until 1999 that Kure et al. first showed that in some treatment of PKU still carries risk of associated nutritional deficien- patients with mutations in the PAH gene pharmacological doses of cies. There have been reports of growth retardation and specificdeficits BH4 (20 mg/kg) could reduce blood Phe levels [27]. Since then, 10 to such as calcium, iron, selenium, zinc or vitamin D and B12 deficiencies 60% of patients with a deficiency of the PAH enzyme have been found [8–10]. Signs of osteoporosis may develop at an early age [11]. Besides to be BH4-responsive, depending on the mutations prevalent in the nutritional considerations, the PKU diet imposes a heavy burden, both different populations [28–36]. Initially, only some centers had access economical and social, upon the patient and their families [12].Itis to a non-commercial form of BH4 (Shircks Laboratories, Jona, therefore not surprising that non-compliance, particularly among Switzerland), but since then, sapropterin dihydrocloride was devel- adolescents and adults, is common [13,14]. Most disturbingly, it is oped and commercially launched after regulatory approval in the US evident that even early and continuously treated PKU patients may not in 2007 and in a growing number of countries in Europe and attain their full neurodevelopmental potential. Treated PKU patients worldwide since 2008. typically exhibit normal intellectual quotients (IQ), but there is, still an Not all PAH-deficient patients respond to BH4 and among those IQ gap when compared to their non-PKU siblings or classmates who do, there is a wide range in the extent of the response. In order to [5,15,16]. Treated PKU patients frequently show delays in certain assess whether a PAH-deficient patient is responsive to BH4, an oral neurological functions , which likely contribute to poor results in school challange is employed. Different protocols have been developed. In and long-term employment [17]. Executive function seems to be the general, European centers favor shorter tests, similar to that proposed most consistently affected area in PKU patients [18–20]. Anxiety, by Blau et al. [37]. This test consists of blood Phe monitoring at 0, 8, 16 depression and a low self esteem have also been reported [21]. and 24 h during 2 consecutive days, while receiving a 20 mg/kg/day. It For all these reasons, novel alternative therapies for PKU have been is recommended that basal Phe levels beN400 μmol/L and that blood sought. Due to the strong link between neurological outcome and Phe levels are closely monitored during the previous days in order to blood Phe levels, most new treatment approaches will try aim to avoid false negative or positive results. A decrease of the blood Phe decrease blood Phe levels. However, clinical objectives are no longer levels with respect to the basal value of at least 30% is considered a centered exclusively on blood Phe levels, but also on the impact of positive response. A negative result would be a reduction of less than blood Phe upon brain aminoacid concentrations, early neurodevelop- 20%, while if the reduction is between 20% and 30% a longer response ment and long-term neurologic outcome. Attaining normal growth protocol (1 to 3 weeks), with daily blood Phe concentration and body mass, and avoiding nutritional deficiencies are other measurements, is recommended to establish BH4 responsiveness. In important factors in the consideration of outcome. Finally, the patient's the USA, longer-term challenges are the common. In the protocol psychosocial well-being has become a key point in PKU treatment proposed by Levy et al., 20 mg/kg of BH4 is given daily for 2 to objectives, as other approaches lead to non-compliance and eventually 4 weeks, with blood Phe samples taken weekly [38]. Cutoff Phe levels important clinical manifestations. are similar to European standards, considering positive a reduction of New dietary therapies include more palatable medical formulas, more than 30%. These are general recommendations that unite the large neutral aminoacids supplementation [22] and the development of most common trends followed in the different centers, but different medical foods based upon glycomacropeptide, a naturally low-Phe goals can be set for individual patients. In general, 48-hour tests detect protein [23]. The increasing knowledge of the genetic basis of disease most BH4-responders if multiple blood Phe measurements are taken. and enzymology has allowed for the investigation of novel pharmaco- Short-term challange protocols allow treatment decisions to be logic therapies to directly ameliorate the effects of a mutant enzyme. For made sooner, at a lower cost and with less interference of possible example, pharmacologic chaperones, including the naturally occurring dietary interventions or infectious episodes. On the other hand, longer cofactor. 5,6,7,8-tetrahydrobiopterin (BH4), help stabilize misfolded tests are less likely to miss possible late-responders [39] and can be mutant enzymes and prevent proteolisis. BH4 the cofactor of PAH, has used to assess other benefits of the treatment such as effects upon been used to treat a subset of PKU patients for almost a decade under neurocognitive function. experimental conditions. In the past 2–3 years, BH4 has begun to be Although not generally available, different methods of detecting a commercializedFOR in the form of sapropterin ELECTRONIC dihydrochloride, a synthetic response to BH4 exist.USE The 13C-Phe oxidationONLY test may be a useful and form of BH4, and is beginning to be widely available. Orthotopic liver safe diagnostic tool to assess the effect of BH4 on in vivo enzyme transplantation succesfully cured PKU in a single patient who was activity in the individual patient. In this test, the percentage of expired 13 transplanted because of cryptogenetic cirrhosis [1]. Therapeutic liver CO2 recovered after an oral ingestion of BH4 rises significantly in repopulation following hepatocyte transplant continues to be investi- responsive patients indicating that BH4 improves in vivo PAH enzyme gated in preclinical models [24]. Clinical gene therapy trials are activity [29,40]. underway in several inborn errors of metabolism, and preclinical gene In general, patients with milder phenotypes have a higher transfer experiments continue in PKU animal models. Finally , a novel response rate to BH4 than those with low Phe tolerance [29–36].A enzyme substitution approach that utilizes subcutaneous injection of mutation database (BIOPKU database), which includes all mutations phenylalanine ammonia lyase to metabolize circulating blood Phe, is reported to respond to BH4, is available (www.biopku.org). In Europe, currently in clinical trial. This paper aims to summarize and review the southern countries have the highest predicted response rates based current knowledge on the pharmacological approach to the treatment of on their most prevalent mutations [34]. It has been shown that PKU. missense mutations in the PAH gene induce PAH loss of function by A. Bélanger-Quintana et al. / Molecular Genetics and Metabolism 104 (2011) S19–S25 S21 protein misfolding, destabilization and early degradation [41,42]. BH4 on Phe levels [53]. For this reason in the US, the trend is to use was shown to act as a molecular chaperone by increasing the stability sapropterin at 20 mg/kg/day. However, other studies suggest that each of partially misfolded PAH proteins and by this the effective patient has a dose over which he does not experience any further benefit intracellular concentration of functional PAH enzyme [43–46].Itis [54]. European publications tend to suggest the latter, as the dose is clear, therefore, that for a patient to respond to BH4, sufficient residual titrated to that at which no further tolerance is reached, and patients liver PAH protein must be present to interact with BH4. This explains receive from 5 to 20 mg/kg/day of BH4 [55–60]. The dose per kilo is not why patients with milder forms and milder mutations are more likely age related, but rather it seems affected by the phenotype/genotype of to respond than those with null mutations. It also clarifies why as the the patient: the milder the phenotype, the smaller the BH4 dose needed phenotype worsens, the response to BH4 is often slow and for long-term treatment. As with other drugs, there might be a total incomplete. Using mutations to predict residual activity has been daily dose, unrelated to weight, over which there is no additional effect successful in predicting BH4-responsiveness [36]. However, responses of the medication, but no study has been performed to establish such a have been described across the phenotypic spectrum of PAH measure. The maximum dose for obese and adult patients is being deficiency, and there are also inconsistencies in the response of decided following each center's protocol, maintaining a dose of 20 mg/ patients that share the same genotype, particularly among compound kg/day in some centers, while others use maximum daily doses ranging heterozygous (which are the majority of patients) [47,48]. from 1 g to 1.4 g/day. Few publications indicate that doses higher than Why are there genotype inconsistencies in the response to BH4? One 20 mg/kg/day have been tried [60]. No adverse-effects have been reason is that different BH4 challenge protocols have been employed by reported at these higher doses, but also little additional clinical benefit. It different centers, but these inconsistencies have been described even in must be noted that animal studies revealed that compound heterozy- the same center. Recent work on PAH enzyme kinetics using a technology gous mice carrying both a mild and a severe mutation in the PAH gene, that allows for the analysis of a broad range of substrate and cofactor Pahenu1/2, show an inhibitory effect at doses above 20 mg/kg/day, concentrations on PAH activity showed that in addition to the chaperone whereas this was not true in mice carrying a mild mutation in the effect of BH4, kinetic aspects also have to be taken into account [49]. homozygous state, Pahenu1/1[61]. These investigations demonstrate the importance of the relationship Pharmacokinetic studies suggest a peak blood concentration at between baseline Phe concentrations and the BH4 dose used in the 2–4 h after ingestion, and a half-life of 6 h[54]. Most patients are treated response to a BH4 challenge. In vitro, a given mutation may respond with single daily dosing, but dividing the total amount in 2–3 daily doses differently when these factors are modified. Since most patients are has been suggested to be beneficial for some patients in order to avoid compound heterozygous, this effect would be magnified. This study daily Phe level fluctuations [57]. Again, this idea is supported by animal suggests that patients with the same genotype can have different studies, where the in vivo effect of BH4 was significantly shorter in the responses to BH4 depending on baseline Phe levels or the amount of BH4 compound heterozygous mouse Pahenu1/2 than in the homozygous mild administered [49]. In the future, optimal baseline Phe concentrations or phenotype Pahenu1/1[61]. Taking the tablets with food, rather than the BH4 dose recommendations might have to be individualized. At fasting, has also been proven beneficial [62]. present, recommendations are standard for all individuals but if a After identification of BH4 responsiveness, the diet must be patient with a very suggestive genotype is found to be non-responsive, it progressively modified until the highest Phe tolerance possible is is recommended that testing be repeated while taking careful notice of determined. The first months of treatment require a very close nutritional basal Phe levels and/or using longer duration response tests. follow up, in order to explain the neccesary dietary changes to the family, BH4 challenge to detect BH4-responsive PAH deficiency is effective make sure they implement them and detect any possible food-phobias or at all ages, but there is concern that sensitivity of this testing could be misunderstandings that would lead to nutritional deficiencies in the decreased in newborns. After the first 3 months of age, when long-term. The usual approach to determine the new Phe tolerance is to hepatocytes are sufficiently mature, testing is thought to be reliable. give a small amount of extra phenylalanine per week and closely monitor However, testing during the neonatal period has several advantages. It blood Phe levels. Phenylalanine can be administered as natural food, but helps in the differential diagnosis of PAH and BH4 deficiencies, but some centers prefer initially to give only egg or milk powder, as regarding exclusively PKU, it allows the diagnosis of BH4-responsive calculation of the increase Phe is more straightforward and it avoids the patients early enough to avoid introducing restrictive diets, favoring introduction of foods the patient might like but not be able to tolerate in breastfeeding and natural protein intake (with better protein, fat, the long-term. Special Phe-free formulas should be reduced accordingly vitamins, microelements and immune composition). During the to the new protein requirements [63].Therearesomesuggested neonatal period, only short tests (24 h-long) can be used in order to guidelines as to how to reach the new diet [64], but frequently the avoid delaying treatment in non-responders. Late-responders can be patients’ expectations and each center's experience greatly influence the missed with this protocol; longer term challenges should then be timing and approach to introducing natural foods. In published papers, considered at a later age. Although the few publications regarding this patients with very mild phenotypes and an important reduction in their subject have not mentioned any side-effects of BH4 in very young Phe levels during the testing period can attain a normal unrestricted diet. patients [50–52], there have been no standardized tests to measure However, for patients with more severe phenotypes, a smaller reduction safety in children under 4 years of age. For this reason, the European in their Phe levels or a late treatment response, BH4 administration may authorities do not recommend its use in this age group, and the allow an increase in dietary Phe intake but not fully alleviate the need for centers that doFOR so must do it under ELECTRONIC compassionate use. a restrictive diet USE in the long-term [57] ONLY. The successfully pharmacologically Once a patient is labeled responsive, considerable effort to fine-tune treated patient will show an increase in dietary Phe tolerance. If this does the therapy is yet required. Continued responsiveness over the long- not occur it is probably due to a false positive result of the treatment test term must be documented, as fluctuation in blood Phe concentration and long-term therapy should be reconsidered. However, there is no clear and unintended alteration of dietary Phe intake during a short-term test recommendation for using tolerance as a marker for responsiveness and may yield false positive results. Improved blood Phe control or a clear suggestions vary from a rise in 2 g of daily natural protein, doubling the increase in dietary Phe tolerance confirm that an individual is truly BH4- baseline Phe tolerance or reaching a completely liberalized diet. responsive, but there is no consensus as to what degree of control or After the initial months, patients can be followed with a similar tolerance should be considered as suffient evidence. BH4 treatment routine as that of any other PKU patient. Few centers have a longer must be closely supervised both by the clinician and the dietician during experience with the drug than for a few years [55–60,65,66],but the first year of treatment. benefits of BH4 other than just the increase in tolerance are starting to be First, the appropriate individualized BH4 dose must be decided. noted, such as a reduction in Phe level fluctuation [58,67] improved Some studies suggest that the higher the dose, the greater the reduction growth [58] and improved nutritional status [68,69]. Changes in S22 A. Bélanger-Quintana et al. / Molecular Genetics and Metabolism 104 (2011) S19–S25 neuropsychological or quality of life results, although expected to be benefits of this therapy in very young children or women during good, remain to be documented [70,71]. Large scale studies to evaluate pregnancy and breastfeeding are still to be evaluated [77]. Now that long-term outcomes have been initiated, but results will not be known sapropterin is becoming available worldwide, participation in the for several years. different ongoing registries is essential in order to gather sufficient Possible negative side-effects are the greatest concern of any data to reach solid conclusions. pharmacological therapy. During the use of the previously available non-commercial form of BH4, the only published adverse event was mucositis following sublingual administration [72]. Commercialization 3. Phenylalanine ammonia lyase (PAL) studies for sapropterin were mainly designed to determine treatment efficacy but no severe adverse events attributable to the drug were Enzyme substitution therapy with phenylalanine ammonia lyase reported. [73–75]. In an extension safety study following patients (PAL; E.C.4.3.1.5) is currently under intensive clinical investigation as treated with 20 mg/kg/day for several weeks, it was observed that a possible alternative treatment for PKU. PAL catalyzes the deamina- patients complained more often from headache, nasal discharge and tion of phenylalanine to free ammonia and trans-cinnamic acid gastrointestinal discomfort, mainly nausea and diarrhea. All these (Fig. 1) [78]. In humans, trans-cinnamic acid is safely and rapidly symptoms where reported mild and never required the discontinuation converted to hippuric acid which is then excreted in the urine [79].In of the therapy [53]. Long-term studies need to be conducted to confirm comparison to human phenylalanine hydroxylase (PAH), PAL is these results and try to establish if side-effects are dose dependant and if structurally and catalytically less complex than PAH, physically other factors such as the diet, age or concomitant medication are more stable, and does not require a cofactor. Prior attempts using involved. oral presentations or subcutaneous implantation of PAL-containing The use of BH4 during pregnancy is an attractive option. In women bioreactors were unsuccessful [80–82]. However, simple subcutane- determined to be BH4-responsive, sapropterin can help reach the strict ous injection of the PAL enzyme into Pahenu2 mice, a model of human blood Phe target levels recommended during pregnancy, something all PKU, yielded complete correction of blood phenylalanine concentra- clinicians know is usually not an easy task. In women with the benign tions [83] and this effect was sustained for up to a year with weekly form of PKU, who have never followed a protein-restrictive diet and are enzyme injections. PAL from the algae Anabaena variabilis proved not accustomed to the taste of the special formulas, sapropterin could physically more resistant to proteolytic degradation, less prone to offer the perfect solution to their dilemma. Furthermore, avoidance of aggregation, and catalytically more favorable than PAL from other dietary protein restriction should allow improved nutritional status of species [84]. Attachment of polyethylene glycol polymers to lysine the mother, and hence of the fetus. If BH4 is not widely prescribed side chains exposed on the surface of the enzyme (PEGylation) assist during pregnancy, it is due to the scanty existing knowledge of the to elude immune-mediated detection and elimination of the injected possible teratogenic effect of the medication. A few cases have been enzyme [85]. Based upon these results, BioMarin Corp., Novato, CA has published in which no side-effects for the mother or the baby were chosen to move enzyme substitution therapy with PEGylated noted [76]. However, in this report a very low dose of the medication recombinant Anabaena variabilis PAL (rAvPAL-PEG) into clinical was prescribed (100 to 300 mg/day when the usual dose for an adult trial. In a Phase 1 trial designed to evaluate safety of subcutaneous PKU patient is 800–1400 mg/day) and there is no long-term safety rAvPAL-PEG injection and carried out at seven centers in the US, evidence. The actual recommendation is that diet should be the first twenty-five adults with PKU and blood phenylalanineN600 μM option during pregnancy, but when it cannot be followed and/or blood received a single rAvPAL-PEG injection in escalating dose cohorts. Phe levels are not sufficiently controlled, sapropterin dyhidrocloride No severe adverse events were recorded and in the highest dose treatment can be used. cohort, blood phenylalanine levels decreased significantly after a single Many questions remain to be answered regarding the optimum rAvPAL-PEG injection (unpublished data). A multi-center Phase 2 method for evaluating BH4 responsiveness and regarding the long- clinical trial to assess the safety and efficacy of multiple, repetitive term benefits of the treatment upon neurological function, nutritional rAvPAL-PEG injections is currently underway. Meanwhile, efforts to status and the psychosocial well-being of patients. The possibility of develop an effective orally administered rAvPAL-PEG have also been adverse events after prolonged treatment and the safety and/or initiated [86].

H H H H COO- PAH COO-

+ H + H H3N H3N BH4 4-OH-BH2 OH L-Phe L-Tyr FOR ELECTRONIC USE ONLY H H H COO- PAL COO- + NH3 + H3N H H

L-Phe trans-cinnamic acid

Fig. 1. Differences between phenylalanine hydroxylase and phenylalanine ammonia lyase. PAH requires the presence of its cofactor, BH4, as well as a Fe+ molecule and oxygen. PAL does not need a cofactor. Adapted from [78]. A. Bélanger-Quintana et al. / Molecular Genetics and Metabolism 104 (2011) S19–S25 S23

4. Gene therapy [5] G.M. Enns, R. Koch, V. Brumm, E. Blakely, R. Suter, e. Jurecki, Suboptimal outcome in patients with PKU treated early with diet alone: revisiting the evidence, Mol. Genet. Metab. 101 (2010) 99–109. The cloning of the human and mouse PAH cDNAs was reported in the [6] H.L. Levy, S.E. Waisbren, F. Güttler, W.B. Hanley, R. Matalon, B. Rouse, F.K. Trefz, F. 1980s [87]. Early investigations showed that only reaching about 10% de la Cruz, C.G. Azen, R. Koch, Pregnancy experiences in the woman with mild hyperphenylalaninemia, Pediatrics 112 (2003) 1548–1552. enzyme activity is necessary to attain normal Phe metabolism in PAH- [7] F. Feillet, F.J. van Spronsen, A. MacDonald, F.K. Trefz, M. Demirkol, M. Giovannini, deficient mice [88] and recent gene transfer studies using murine models A. Bélanger-Quintana, N. Blau, Challenges and pitfalls in the management of are promising. A recombinant adeno-associated virus (rAAV) vector is phenylketonuria, Pediatrics 126 (2) (Aug 2010) 333–341 Epub 2010 Jul 12. fi frequently employed which produces a lesser immune response than [8] P.B. Acosta, S. Yannicelli, R. Singh, S. Mo di, R. Steiner, E. DeVincentis, E. Jurecki, L. Bernstein, S. Gleason, M. Chetty, B. Rouse, Nutrient intakes and physical growth of vectors used in early experiments, giving rise to fewer side effects and children with phenylketonuria undergoing nutrition therapy, J. Am. Diet. Assoc. allowing for longer-term successful treatment. In the mouse model, liver 103 (2003) 1167–1173. fi directed gene therapy has allowed for up to one year correction of [9] G.L. Arnold, C.J. Vladutiu, R.S. Kirby, E.m Blakely, J.M. Deluca, Protein insuf ciency and linear growth restriction in phenylketonuria, J. Pediatr. 141 (2002) 243–246. hyperphenylalaninemia [89,90]. However, liver-directed gene therapy [10] D. Dobbelaere, L. Michaud, A. Debrabander, S. Vanderbecken, F. Gottrand, D. Turck, does not lead to a permanent correction of PAH activity. The vector's J.P. Farriaux, Evaluation of nutritional status and pathophysiology of growth genome is not integrated in the hepatocyte's DNA and gradual but retardation in patients with phenylketonuria, J. Inherit. Metab. Dis. 26 (2003) 1–11. continuous hepatocyte regeneration eventually leads to elimination of [11] P. Barat, N. Barthe, I. Redonnet-Vernhet, F. Parrot, The impact of the control of episomal rAAV vector genomes. Reinjection of the same serotype vector serum phenylalanine levels on osteopenia in patients with phenylketonuria, Eur. is not effective, as it is destroyed by an antibody-mediated immune J. Pediatr. 161 (2002) 687–688. [12] E. Simon, M. Schwarz, J. Roos, N. Dragano, M. Geraedts, J. Siegrist, G. Kamp, U. reaction. Wendel, Evaluation of quality of life and description of the sociodemographic Skeletal muscle is an attractive target organ for gene therapy state in adolescent and young adult patients with phenylketonuria (PKU), Health because it does not suffer on-going cell division and it is more easily Qual. Life Outcomes 26 (2008) 6–25. [13] K. Ahring, A. Bélanger-Quintana, K. Dokoupil, H. Gokmen-Ozel, A.M. Lammardo, A. accessible. On the other hand, PAH expression alone in myocites will MacDonald, K. Motzfeldt, M. Nowacka, M. Robert, M. van Rijn, Blood phenylal- not lead to functional Phe hydroxylation; expression of enzymes anine control in phenylketonuria: a survey of 10 European centres, Eur. J. Clin. involved in the BH4 synthetic pathway an local production of BH4 is Nutr. 65 (2011) 275–278. also required. This has been tried with promising results in mice [14] A. MacDonald, H. Gokmen-Ozel, M. van Rijn, P. Burgard, The reality of dietary compliance in the management of phenylketonuria, J. Inherit. Metab. Dis. 33 [91,92]. (2010) 665–670. The improvement in viral vector design has led to human gene therapy [15] S.E. Waisbren, K. Noel, K. Fahrbach, C. Cella, D. Frame, A. Dorenbaum, H. Levy, trials for other inborn errors of metabolism, including α1-antitrypsin Phenylalanine blood levels and clinical outcomes in phenylketonuria: a fi systematic literature review and meta-analysis, Mol. Genet. Metab. 92 (2007) de ciency, Canavan disease, late-infantile neuronal ceroidlipofuscinosis 63–70. and lipoprotein lipase deficiency [93]. Ongoing research, optimizing the [16] R. Gassió, R. Artuch, M.A. Vilaseca, E. Fusté, C. Boix, A. Sans, J. Campistol, Cognitive muscle-directed approach and improving sustainability in liver-directed functions in classic phenylketonuria and mild hyperphenylalaninaemia: experience in a paediatric population, Dev. Med. Child Neurol. 47 (2005) 443–448. gene therapy might lead to human trials in PKU in the next few years. [17] J.K. Gentile, A.E. Ten Hoedt, A.M. Bosch, Psychosocial aspects of PKU: hidden disabilities—a review, Mol. Genet. Metab. 99 (2010) S64–S67. [18] S.E. Christ, S.C. Huijbregts, L.M. de Sonneville, D.A. White, Executive function in early-treated phenylketonuria: profile and underlying mechanisms, Mol. Genet. 5. Conclusions Metab. 99 (2010) S22–S32. [19] R. Sharman, K. Sullivan, R. Young, J. McGill, Biochemical markers associated with Neonatal diagnosis of PKU, early institution of a Phe-restrictive diet executive function in adolescents with early and continuously treated phenylketonuria, Clin. Genet. 75 (2009) 169–174. and the possibility of avoiding devastating brain damage associated to [20] V. Leuzzi, M. Pansini, E. Sechi, F. Chiarotti, C. Carducci, G. Levi, I. Antonozzi, hyperphenylalaninemia was a great success. Nevertheless, the Executive function impairment in early-treated PKU subjects with normal mental difficulties of adhering to a strict diet for life and the presence of development, J. Inherit. Metab. Dis. 27 (2004) 115–125. neurological deficits despite treatment make other therapeutic [21] V.L. Brumm, D. Bilder, S.E. Waisbren, Psychiatric symptoms and disorders in phenylketonuria, Mol. Genet. Metab. 99 (2010) S59–S63. approaches indispensable. The past decade has been a very exciting [22] R. Matalon, K. Michals-Matalon, G. Bhatia, A.B. Burlina, A.P. Burlina, C. Braga, time for investigators, clinicians and PKU patients, as the knowledge L. Fiori, M. Giovannini, E. Grechanina, P. Novikov, J. Grady, S.K. Tyring, F. on the pathogenesis of the disease has been broadened and this has Guttler, Double blind placebo control trial of large neutral aminoacids in treatment of PKU: effect on blood phenylalanine, J. Inherit. Metab. Dis. 30 allowed for the investigation of truly novel therapeutic approaches. In (2007) 153–158. some cases, such as dietary products or tetrahydrobiopterin, these [23] D.M. Ney, S.T. Gleason, S.C. van Calcar, E.L. MacLeod, K.L. Nelson, M.R. Etzel, G.M. novel treatments are already available in the clinical setting. Human Rice, J.A. Wolff, Nutritional management of PKU with glycomacropeptide from cheese whey, J. Inherit. Metab. Dis. 32 (2009) 32–38. trials are underway with rAvPAL-PEG enzyme substitution. Pre- [24] C.O. Harding, K.M. Gibson, Therapeutic liver repopulation for phenylketonuria, J. clinical studies are being done in gene and chaperone therapy. Many Inherit. Metab. Dis. 33 (2010) 681–687. questions regarding existing therapies remain to be answered and [25] K. Hyghland, Tetrahydrobiopterin deficiencies with hyperphenylalaninemia, in: N. fi Blau (Ed.), PKU and BH4. Advances in Phenylketonuria and Tetrahydrobiopterin, SPS much dif cult work has to be done before many of these novel Publications, 2006. technologies reach the patient, but the future of PKU treatment has [26] A. Niederwieser, H.C. Curtius, M. Wang, D. Leupold, Atypical phenylketonuria with never seemed brighter. In the future, different therapies will be defective biopterin metabolism. Monotherapy with tetrahydrobiopterin or sepiapterin, screening and study of biosynthesis in man, Eur. J. Pediatr. 138 available allowing individual treatment depending upon the particular (1982) 110–112. genotype andFOR other personal circumstances ELECTRONIC such as age or pregnancy. [27] S. Kure, D.C. Hou,USE T. Ohura, H. Iwamoto, ONLY S. Suzuki, N. Sugiyama, O. Sakamoto, K. Fujii, Y. Matsubara, K. Narisawa, Tetrahydrobiopterin-responsive phenylalanine hydroxylase deficiency, J. Pediatr. 135 (1999) 375–378. [28] C. Bernegger, N. Blau, High frequency of tetrahydrobiopterin-responsiveness References among hyperphenylalaninemias: a study of 1,919 patients observed from 1988 to 2002, Mol. Genet. Metab. 77 (2002) 304–313. [1] C.R. Scriver, S. Kaufman, Hyperphenylalaninemia: phenylalanine hydroxylase [29] A.C. Muntau, W. Röschinger, M. Habich, H. Demmelmair, B. Hoffmann, C.P. deficiency, in: C.R. Scriver, A.L. Beaudet, D. Valle, B. Vogelstein, W.S. Sly (Eds.), The Sommerhoff, A.A. Roscher, Tetrahydrobiopterin as an alternative treatment for Metabolic and Molecular Bases of Inherited Disease, McGraw-Hill, New York, mild phenylketonuria, N. Engl. J. Med. 347 (2002) 2122–2132. 2001. [30] M. Lindner, R. Steinfeld, P. Burgard, A. Schulze, E. Mayatepek, J. Zschocke, [2] L.I. Woolf, Excretion of conjugated phenylacetic acid in phenylketonuria, Biochem. Tetrahydrobiopterin sensitivity in German patients with mild phenylalanine J. 49 (1951) ix–x. hydroxylase deficiency, Hum. Mutat. 21 (2003) 400. [3] H. Bickel, J. Gerrard, E.M. Hickmans, Influence of phenylalanine intake on [31] L.R. Desviat, B. Pérez, A. Bèlanger-Quintana, M. Castro, C. Aguado, A. Sánchez, M.J. phenylketonuria, Lancet 265 (1953) 812–813. García, M. Martínez-Pardo, M. Ugarte, Tetrahydrobiopterin responsiveness: [4] National institutes of health consensus development conference statement: results of the BH4 loading test in 31 Spanish PKU patients and correlation with phenylketonuria: screening and management, Pediatrics 108 (2000) 972–982. their genotype, Mol. Genet. Metab. 83 (2004) 157–162. S24 A. Bélanger-Quintana et al. / Molecular Genetics and Metabolism 104 (2011) S19–S25

[32] R. Matalon, K. Michals-Matalon, R. Koch, J. Grady, S. Tyring, R.C. Stevens, Response [56] N. Lambruschini, B. Pérez-Dueñas, M.A. Vilaseca, A. Mas, R. Artuch, R. Gassió, L. of patients with phenylketonuria in the US to tetrahydrobiopterin, Mol. Genet. Gómez, A. Gutiérrez, J. Campistol, Clinical and nutritional evaluation of Metab. 86 (2005) S17–S21. phenylketonuric patients on tetrahydrobiopterin monotherapy, Mol. Genet. [33] J.J. Mitchell, B. Wilcken, I. Alexander, C. Ellaway, H. O'Grady, V. Wiley, J. Earl, J. Metab. 86 (2005) S54–S60. Christodoulou, Tetrahydrobiopterin-responsive phenylketonuria: the New South [57] A. Bélanger-Quintana, M.J. García, M. Castro, L.R. Desviat, B. Pérez, B. Mejía, M. Wales experience, Mol. Genet. Metab. 86 (2005) S81–S85. Ugarte, M. Martínez-Pardo, Spanish BH4-responsive phenylalanine hydroxylase- [34] M.R. Zurflüh, J. Zschocke, M. Lindner, F. Feillet, C. Chery, A. Burlina, R.C. Stevens, B. deficient patients: evolution of seven patients on long-term treatment with Thöny, N. Blau, Molecular genetics of tetrahydrobiopterin-responsive phenylalanine tetrahydrobiopterin, Mol. Genet. Metab. 86 (2005) S61–S66. hydroxylase deficiency, Hum. Mutat. 29 (2008) 167–175. [58] F.K. Trefz, D. Scheible, G. Frauendienst-Egger, H. Korall, N. Blau, Long-term [35] S.F. Dobrowolski, K. Borski, C.C. Ellingson, R. Koch, H.L. Levy, E.W. Naylor, A limited treatment of patients with mild and classical phenylketonuria by tetrahydro- spectrum of phenylalanine hydroxylase mutations is observed in phenylketonuria biopterin, Mol. Genet. Metab. 86 (2005) S75–S80. patients in western Poland and implications for treatment with 6R tetrahydro- [59] M.D. Bóveda, M.L. Couce, D.E. Castiñeiras, J.A. Cocho, B. Pérez, M. Ugarte, J.M. Fraga, biopterin, J. Hum. Genet. 54 (2009) 335–359. The tetrahydrobiopterin loading test in 36 patients with hyperphenylalaninae- [36] I. Karacić, D. Meili, V. Sarnavka, C. Heintz, B. Thöny, D.P. Ramadza, K. Fumić,D. mia: evaluation of response and subsequent treatment, J. Inherit. Metab. Dis. 30 Mardesić, I. Barić, N. Blau, Genotype-predicted tetrahydrobiopterin (BH4)- (2007) 812. responsiveness and molecular genetics in Croatian patients with phenylalanine [60] F.K. Trefz, D. Scheible, G. Frauendienst-Egger, Long-term follow-up of patients hydroxylase (PAH) deficiency, Mol. Genet. Metab. 97 (2009) 165–171. with phenylketonuria receiving tetrahydrobiopterin treatment, J. Inherit. Metab. [37] N. Blau, A. Bélanger-Quintana, M. Demirkol, F. Feillet, M. Giovannini, A. MacDonald, F.K. Dis. 9 (2010) [Epub ahead of print]. Trefz, F.J. van Spronsen, Optimizing the use of sapropterin (BH(4)) in the management [61] F.B. Lagler, S.W. Gersting, C. Zsifkovits, A. Steinbacher, A. Eichinger, M.K. Danecka, of phenylketonuria, Mol. Genet. Metab. 96 (2009) 158–163. M. Staudigl, R. Fingerhut, H. Glossmann, A.C. Muntau, New insights into [38] H. Levy, B. Burton, S. Cederbaum, C. Scriver, Recommendations for evaluation of tetrahydrobiopterin pharmacodynamics from Pah enu1/2, a mouse model for responsiveness to tetrahydrobiopterin (BH(4)) in phenylketonuria and its use in compound heterozygous tetrahydrobiopterin-responsive phenylalanine hydrox- treatment, Mol. Genet. Metab. 92 (2007) 287–291. ylase deficiency, Biochem. Pharmacol. 80 (2010) 1563–1571. [39] B. Fiege, L. Bonafé, D. Ballhausen, M. Baumgartner, B. Thöny, D. Meili, L. Fiori, M. [62] D.G. Musson, W.G. Kramer, E.D. Foehr, F.A. Bieberdorf, C.S. Hornfeldt, S.S. Kim, A. Giovannini, N. Blau, Extended tetrahydrobiopterin loading test in the diagnosis of Dorenbaum, Relative bioavailability of sapropterin from intact and dissolved cofactor-responsive phenylketonuria: a pilot study, Mol. Genet. Metab. 86 (2005) sapropterin dihydrochloride tablets and the effects of food: a randomized, open- S91–S95. label, crossover study in healthy adults, Clin. Ther. 32 (2010) 338–346. [40] Y. Okano, K. Takatori, S. Kudo, T. Sakaguchi, M. Asada, M. Kajiwara, T. Yamano, [63] A. Macdonald, K. Ahring, K. Dokoupil, H. Gokmen-Ozel, A.M. Lammardo, K. Effects of tetrahydrobiopterin and phenylalanine on in vivo human phenylalanine Motzfeldt, M. Robert, J.C. Rocha, M. van Rijn, A. Bélanger-Quintana, Adjusting diet hydroxylase by phenylalanine breath test, Mol. Genet. Metab. 92 (2007) 308–314. with sapropterin in phenylketonuria: what factors should be considered? Br. J. [41] A.L. Pey, F. Stricher, L. Serrano, A. Martinez, Predicted effects of missense mutations on Nutr. 5 (2011) 1–8. native-state stability account for phenotypic outcome in phenylketonuria, a paradigm [64] R.H. Singh, E. Jurecki, F. Rohr, Recommendations for personalized diet adjust- of misfolding diseases, Am. J. Hum. Genet. 81 (2007) 1006–1024. ments based on patients response to tetrahydrobiopterin (BH4) in phenylketon- [42] S.W. Gersting, K.F. Kemter, M. Staudigl, D.D. Messing, M.K. Danecka, F.B. Lagler, uria, Top. Clin. Nutr. 23 (2008) 149–157. C.P. Sommerhoff, A.A. Roscher, A.C. Muntau, Loss of function in phenylketonuria is [65] H. Shintaku, S. Kure, T. Ohura, Y. Okano, M. Ohwada, N. Sugiyama, N. Sakura, I. caused by impaired molecular motions and conformational instability, Am. J. Yoshida, M. Yoshino, Y. Matsubara, K. Suzuki, K. Aoki, T. Kitagawa, Long-term Hum. Genet. 83 (2008) 5–17. treatment and diagnosis of tetrahydrobiopterin-responsive hyperphenylalanine- [43] H. Erlandsen, A.L. Pey, A. Gámez, B. Pérez, L.R. Desviat, C. Aguado, R. Koch, S. mia with a mutant phenylalanine hydroxylase gene, Pediatr. Res. 55 (2004) Surendran, S. Tyring, R. Matalon, C.R. Scriver, M. Ugarte, A. Martínez, R.C. Stevens, 425–430. Correction of kinetic and stability defects by tetrahydrobiopterin in phenylke- [66] H. Shintaku, M. Ohwada, K. Aoki, T. Kitagawa, T. Yamano, Diagnosis of tonuria patients with certain phenylalanine hydroxylase mutations, Proc. Natl. tetrahydrobiopterin (BH4) responsive mild phenylketonuria in Japan over the Acad. Sci. U. S. A. 101 (2004) 16903–16908. past 10 years, Ann. Acad. Med. Singapore 37 (2008) 77–82. [44] A.L. Pey, B. Pérez, L.R. Desviat, M.A. Martínez, C. Aguado, H. Erlandsen, A. Gámez, [67] B.K. Burton, H. Bausell, R. Katz, H. Laduca, C. Sullivan, Sapropterin therapy R.C. Stevens, M. Thórólfsson, M. Ugarte, A. Martínez, Mechanisms underlying increases stability of blood phenylalanine levels in patients with BH4-responsive responsiveness to tetrahydrobiopterin in mild phenylketonuria mutations, Hum. phenylketonuria (PKU), Mol. Genet. Metab. 101 (2010) 110–114. Mutat. 24 (2004) 388–399. [68] R.H. Singh, M.E. Quirk, T.D. Douglas, M.C. Brauchla, BH(4) therapy impacts the [45] B. Pérez, L.R. Desviat, P. Gómez-Puertas, A. Martínez, R.C. Stevens, M. Ugarte, nutrition status and intake in children with phenylketonuria: 2-year follow-up, J. Kinetic and stability analysis of PKU mutations identified in BH4-responsive Inherit. Metab. Dis. 33 (2010) 689–695. patients, Mol. Genet. Metab. 86 (2005) S11–S16. [69] M.A. Vilaseca, N. Lambruschini, L. Gómez-López, A. Gutiérrez, J. Moreno, M. Tondo, [46] S.W. Gersting, F.B. Lagler, A. Eichinger, K.F. Kemter, M.K. Danecka, D.D. Messing, M. R.Artuch,J.Campistol,Long-chainpolyunsaturatedfattyacidstatusin Staudigl, K.A. Domdey, C. Zsifkovits, R. Fingerhut, H. Glossmann, A.A. Roscher, A.C. phenylketonuric patients treated with tetrahydrobiopterin, Clin. Biochem. 43 Muntau, Pahenu1 is a mouse model for tetrahydrobiopterin-responsive phenyl- (2010) 411–415. alanine hydroxylase deficiency and promotes analysis of the pharmacological [70] R. Gassió, M.A. Vilaseca, N. Lambruschini, C. Boix, M.E. Fusté, J. Campistol, chaperone mechanism in vivo, Hum. Mol. Genet. 19 (2010) 2039–2049. Cognitive functions in patients with phenylketonuria in long-term treatment with [47] J.B. Hennermann, C. Bührer, N. Blau, B. Vetter, E. Mönch, Long-term treatment tetrahydrobiopterin, Mol. Genet. Metab. 99 (2010) S75–S78. with tetrahydrobiopterin increases phenylalanine tolerance in children with [71] U.R. Somaraju, M. Merrin, Sapropterin dihydrochloride for phenylketonuria, severe phenotype of phenylketonuria, Mol. Genet. Metab. 86 (2005) S86–S90. Cochrane Database Syst. Rev. 16 (2010) CD008005. [48] F.K. Trefz, D. Scheible, H. Götz, G. Frauendienst-Egger, Significance of genotype in [72] T. Opladen, M. Zurflüh, I. Kern, L. Kierat, B. Thöny, N. Blau, Severe mucitis after tetrahydrobiopterin-responsive phenylketonuria, J. Inherit. Metab. Dis. 32 (2009) sublingual administration of tetrahydrobiopterin in a patient with tetrahydro- 22–26. biopterin-responsive phenylketonuria, Eur. J. Pediatr. 164 (2005) 395–396. [49] M. Staudigl, S.W. Gersting, M.K. Danecka, D.D. Messing, M. Woidy, D. Pinkas, K.F. [73] B.K. Burton, D.K. Grange, A. Milanowski, G. Vockley, F. Feillet, E.A. Crombez, V. Kemter, N. Blau, A.C. Muntau, The interplay between genotype, metabolic state Abadie, C.O. Harding, S. Cederbaum, D. Dobbelaere, A. Smith, A. Dorenbaum, The and cofactor treatment governs phenylalanine hydroxylase function and drug response of patients with phenylketonuria and elevated serum phenylalanine to response, Hum. Mol. Genet. (2011) [Epub ahead of print]. treatment with oral sapropterin dihydrochloride (6R-tetrahydrobiopterin): a [50] F. Feillet, C. Chery, F. Namour, A. Kimmoun, E. Favre, E. Lorentz, S.F. Battaglia-Hsu, phase II, multicentre, open-label, screening study, J. Inherit. Metab. Dis. 30 (2007) J.L. Guéant, Evaluation of neonatal BH4 loading test in neonates screened for 700–707. hyperphenylalaninemia, Early Hum. Dev. 84 (2008) 561–567. [74] H.L. Levy, A. Milanowski, A. Chakrapani, M. Cleary, P. Lee, F.K. Trefz, C.B. Whitley, F. [51] M. Bik-Multanowski, J.J. Pietrzyk, B. Didycz, A. Szymczakiewicz-Multanowska, Feillet, A.S. Feigenbaum, J.D. Bebchuk, H. Christ-Schmidt, A. Dorenbaum, Development of a model for assessment of phenylalanine hydroxylase activity in Sapropterin Research Group, Efficacy of sapropterin dihydrochloride (tetrahy- newborns with phenylketonuria receiving tetrahydrobiopterin: a potential for drobiopterin, 6R-BH4) for reduction of phenylalanine concentration in patients practical implementation,FOR Mol. Genet. ELECTRONIC Metab. 94 (2008) 389–390. with phenylketonuria: USE a phase III randomised ONLY placebo-controlled study, Lancet [52] B.K. Burton, D.J. Adams, D.K. Grange, J.I. Malone, E. Jurecki, H. Bausell, K.D. Marra, L. 370 (2007) 504–510. Sprietsma, K.T. Swan, Tetrahydrobiopterin therapy for phenylketonuria in infants [75] F.K. Trefz, B.K. Burton, N. Longo, M.M. Casanova, D.J. Gruskin, A. Dorenbaum, E.D. and young children, J. Pediatr. 158 (2011) 410–415. Kakkis, E.A. Crombez, D.K. Grange, P. Harmatz, M.H. Lipson, A. Milanowski, L.M. [53] P. Lee, E.P. Treacy, E. Crombez, M. Wasserstein, L. Waber, J. Wolff, U. Wendel, A. Randolph, J. Vockley, C.B. Whitley, J.A. Wolff, J. Bebchuk, H. Christ-Schmidt, J.B. Dorenbaum, J. Bebchuk, H. Christ-Schmidt, M. Seashore, M. Giovannini, B.K. Hennermann, Sapropterin Study Group, Efficacy of sapropterin dihydrochloride in Burton, A.A. Morris, Sapropterin Research Group, Safety and efficacy of 22 weeks increasing phenylalanine tolerance in children with phenylketonuria: a phase III, of treatment with sapropterin dihydrochloride in patients with phenylketonuria, randomized, double-blind, placebo-controlled study, J. Pediatr. 154 (2009) Am. J. Med. Genet. A 15 (2008) 2851–2859. 700–707. [54] F. Feillet, L. Clarke, C. Meli, M. Lipson, A.A. Morris, P. Harmatz, D.R. Mould, B. Green, [76] R. Koch, Maternal phenylketonuria and tetrahydrobiopterin, Pediatrics 122 A. Dorenbaum, M. Giovannini, E. Foehr, Sapropterin Research Group, Pharmaco- (2008) 1367–1368. kinetics of sapropterin in patients with phenylketonuria, Clin. Pharmacokinet. 47 [77] F.K. Trefz, A. Belanger-Quintana, Sapropterin dihydrochloride: a new drug and a new (2008) 817–825. concept in the management of phenylketonuria, Drugs Today 46 (2010) 589–600. [55] R. Cerone, M.C. Schiaffino, A.R. Fantasia, M. Perfumo, L. Birk Moller, N. Blau, Long- [78] W. Kim, H. Erlandsen, S. Surendran, R.C. Stevens, A. Gamez, K. Michols-Matalon, term follow-up of a patient with mild tetrahydrobiopterin-responsive phenylke- S.K. Tyring, R. Matalon, Trends in enzyme therapy for phenylketonuria, Mol. Ther. tonuria, Mol. Genet. Metab. 81 (2004) 137–139. 10 (2004) 220–224. A. Bélanger-Quintana et al. / Molecular Genetics and Metabolism 104 (2011) S19–S25 S25

[79] J.A. Hoskins, S.B. Holliday, A.M. Greenway, The metabolism of cinnamic acid by [86] T.S. Kang, L. Wang, C.N. Sarkissian, A. Gámez, C.R. Scriver, R.C. Stevens, Converting healthy and phenylketonuric adults: a kinetic study, Biomed. Mass Spectrom. 11 an injectable protein therapeutic into an oral form: phenylalanine ammonia lyase (1984) 296–300. for phenylketonuria, Mol. Genet. Metab. 99 (2010) 4–9. [80] J.A. Hoskins, G. Jack, H.E. Wade, R.J. Peiris, E.C. Wright, D.J. Starr, J. Stern, Enzymatic [87] S.L. Woo, A.S. Lidsky, F. Güttler, T. Chandra, K.J. Robson, Cloned human control of phenylalanine intake in phenylketonuria, Lancet 1 (1980) 392–394. phenylalanine hydroxylase gene allows prenatal diagnosis and carrier detection [81] S. Safos, T.M. Chang, Enzyme replacement therapy in ENU2 phenylketonuric mice of classical phenylketonuria, Nature 306 (1983) 151–155. using oral microencapsulated phenylalanine ammonia-lyase: a preliminary [88] F.D. Ledley, H.E. Grenett, A.G. DiLella, S.C. Kwok, S.L. Woo, Gene transfer and report, Artif. Cells Blood Substit. Immobil. Biotechnol. 23 (1995) 681–692. expression of human phenylalanine hydroxylase, Science 228 (1985) 77–79. [82] C.M. Ambrus, J.L. Ambrus, C. Horvath, H. Pedersen, S. Sharma, C. Kant, E. Mirand, [89] C.O. Harding, M.B. Gillingham, K. Hamman, H. Clark, E. Goebel-Daghighi, A. Bird, R. Guthrie, T. Paul, Phenylalanine depletion for the management of phenylke- D.D. Koeberl, Complete correction of hyperphenylalaninemia following liver- tonuria: use of enzyme reactors with immobilized enzymes, Science 201 (1978) directed, recombinant AAV2/8 vector-mediated gene therapy in murine phenyl- 837–839. ketonuria, Gene Ther. 13 (2006) 457–462. [83] C.N. Sarkissian, A. Gamez, L. Wang, M. Charbonneau, P. Fitzpatrick, J.F. Lemontt, B. [90] Z. Ding, P. Georgiev, B. Thöny, Administration-route and gender-independent Zhao, M. Vellard, S.M. Bell, C. Henschell, A. Lambert, L. Tsuruda, R.C. Stevens, C.R. long-term therapeutic correction of phenylketonuria (PKU) in a mouse model by Scriver, Preclinical evaluation of multiple species of PEGylated recombinant recombinant adeno-associated virus 8 pseudotyped vector-mediated gene phenylalanine ammonia lyase for the treatment of phenylketonuria, Proc. Natl. transfer, Gene Ther. 13 (2006) 587–593. Acad. Sci. U. S. A. 105 (2008) 20894–20899. [91] Z. Ding, C.O. Harding, A. Rebuffat, L. Elzaouk, J.A. Wolff, B. Thöny, Correction of [84] C.N. Sarkissian, Z. Shao, F. Blain, R. Peevers, H. Su, R. Heft, T.M. Chang, C.R. Scriver, murine PKU following AAV-mediated intramuscular expression of a complete A different approach to treatment of phenylketonuria: phenylalanine degradation phenylalanine hydroxylating system, Mol. Ther. 16 (2008) 673–681. with recombinant phenylalanine ammonia lyase, Proc. Natl. Acad. Sci. U. S. A. 96 [92] A. Rebuffat, C.O. Harding, Z. Ding, B. Thöny, Comparison of adeno-associated (1999) 2339–2344. virus pseudotype 1, 2, and 8 vectors administered by intramuscular injection in [85] A. Gamez, L. Wang, C.N. Sarkissian, D. Wendt, P. Fitzpatrick, J.F. Lemontt, C.R. the treatment of murine phenylketonuria, Hum. Gene Ther. 21 (2010) Scriver, R.C. Stevens, Structure-based epitope and PEGylation sites mapping of 463–477. phenylalanine ammonia-lyase for enzyme substitution treatment of phenylke- [93] I.E. Alexander, S.C. Cunningham, G.J. Logan, J. Christodoulou, Potential of AAV tonuria, Mol. Genet. Metab. 91 (2007) 325–334. vectors in the treatment of metabolic disease, Gene Ther. 15 (2008) 831–839.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S26–S30

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Minireview Adult phenylketonuria outcome and management

F. Trefz a,⁎, F. Maillot b, K. Motzfeldt c, M. Schwarz d a Kreiskliniken Reutlingen GmbH, Reutlingen, Germany b CHRU de Tours, Université François Rabelais, INSERM U921, Tours, France c Oslo University Hospital, Women and Children's Division, Department of Pediatrics and Newborn Screening, Oslo, Norway d Internal Medicine outpatient centre, Kaarst, Germany article info abstract

Article history: The problem to evaluate treatment outcome in adult PKU (phenylketonuric) patients lies in the heterogeneity of the Received 13 July 2011 adult PKU population. This heterogeneity is not only based on the different treatment history of every individual pa- Received in revised form 23 August 2011 tient but also on the different severity of the underlying defect of the enzyme phenylalanine hydroxylase. Recent, Accepted 23 August 2011 partly double blind studies in adult PKU patients further support recommendation for lifelong treatment. However, Available online 26 August 2011 it has become evident that dietary treatment is suboptimal and continuation to adulthood often not accepted. Late detected PKU patients (up to 4–6 years of age) benefit from strict dietary treatment and are able to catch up in Keywords: fi Phenylketonuria intellectual performance. Untreated, severely retarded patients with behavioral changes may bene tfromintro- Hyerphenylalaninemia duction of dietary treatment. However, individual decision is necessary and based on the personal situation of Adult the patient. Sapropterin In early and well treated patients a number of studies have demonstrated that cognitive and neurosychologic tests are different from controls. In addition there is evidence that patients with higher blood phenylalanine (phe) levels demonstratemoreoftenpsychiatricsymptomslikedepression and anxiety. Medical problems are more often observed: there are certain risks as impaired growth, decreased bone mineral density and nutrional deficits probably caused by dietary treatment with an artificial protein substitute and/or missing compliance with an unpleasant diet. The long term risk of a strict dietary treatment must be balanced with the risk of higher blood phe (mean blood phenylalanine N600–900 μmol/L) on cognitive and neuropsychological functions and psychiatric symptoms. Further studies should consider the role of blood phe exposure for brain development in childhood and for brain function in all ages. Besides mean blood phe, fluctuation of blood phe over time is important. Fluctuation of blood phe is decreased by sapropterin treatment in responsive patients which would on the long term may have positive effects on cognitive outcome. Further studies also should include adult PKU patients. © 2011 Elsevier Inc. All rights reserved.

Contents 1. Introduction ...... S27 2. Untreated adult PKU patients ...... S27 3. Treatment in late diagnosed but treated PKU patients ...... S27 4. Outcome in early treated but early discontinued patients ...... S27 5. Treatment results in early treated patients ...... S28 5.1. Cognitive and neuropsychological outcome ...... S28 5.2. Psychiatric findings in early treated adult PKU patients ...... S28 5.3. Nutritional deficiencies and growth ...... S28 5.4. Biochemical outcome parameters ...... S28 5.5. CNS findings in early treated adult PKU patients ...... S29 5.5.1.FOR MRI (magnetic ELECTRONIC resonance imaging) findings...... USE ONLY S29 5.5.2. CNS metabolic findings ...... S29 6. Future management in adult PKU ...... S29 References ...... S29

Abbbreviations: phe, phenylalanine; PAH, phenylalanine hydroxylase; PKU, phenylketonuria; BH4, tetrahydrobioptern. ⁎ Corresponding author at: Kreiskliniken Reutlingen GmbH, Reutlingen, Germany. E-mail addresses: [email protected] (F. Trefz), [email protected] (F. Maillot), [email protected] (K. Motzfeldt), [email protected] (M. Schwarz).

1. Introduction In summary, a general recommendation for a phe restricted diet in severely retarded PKU patients cannot be given. Individual conditions The problem to evaluate treatment outcome in adult PKU (phenylke- should be considered and quality of life should not worsen especially tonuric) patients lies in the heterogeneity of the adult PKU population. in severely retarded patients with behavioral problems. On the other This heterogeneity is not only based on the different treatment history hand in some patients a blood phe level below 900 μmol/L may result of every individual patient but also on the different severity of the under- in reduction of behavioral problems and increase of attention. lying defect of the enzyme phenylalanine hydroxylase (PAH). More than 600 mutations are described so far (http://www.pahdb.mcgill.ca/). 3. Treatment in late diagnosed but treated PKU patients The other problem consists in the fact that there are no prospective studies available and that in later life other modulators as lifestyle and Introduction of newborn screening and early dietary treatment aging become more important when evaluating brain development resulted in the prevention of severe brain damage. However, due to and brain function. Long term dietary treatment also shows develop- screening failures and immigration of patients from countries where ment of other problems like osteopenia or metabolic imbalances. early detection and treatment of PKU is not yet established, there Having these problems in mind it is clear that up to now there are no are a number of untreated PKU's in many European centers. In the evidence based treatment recommendations and only vague recom- last 20 years we saw in one center (Reutlingen, Germany) 10 patients mendations like “lifelong treatment” with very different suggestions missed by newborn screening due to various reasons (Table 2). After for e.g. the blood phe targets as published in a European survey recently introduction of tandem mass spectrometry and a better computerized [1]. sample handling in the screening laboratories one can expect that In this review we would like to describe outcome and manage- missed cases will be minimized in the future. However, there are ment of adult PKU patients excluding maternal PKU but describing still a number of immigrants coming from countries where newborn the different groups of adult PKU patients we see in our metabolic screening is not yet established. outpatient clinic. From early prescreening experiences it is well known that after in- 1. PKU patients never treated, 2. PKU patients late treated (start 1– troduction of dietary treatment in later infancy and childhood partial 6 years of age) 3. Patients early treated but early or with poor compliance reversibility of IQ loss may occur. Catch up of development retarda- and 4. Patients early treated with good compliance until adulthood. tion is especially seen in the first 4–6 years of life [4], but there is a Data are based on literature review and personal clinical experiences. large heterogeneity of treatment success as also shown recently in a review by Grosse [5]. As pointed out in this paper and based on personal experiences 2. Untreated adult PKU patients there is often an impressive improvement of development when introducing a phe restricted diet even in very late detected PKU pa- Patients with untreated PKU are severely retarded in most cases tients. All of these patients benefit from a dietary treatment with and in addition may show challenging behavioral problems phe levels b600 μmol/L beyond infancy and young adulthood. One (Table 1). Treatment studies in untreated PKU's are difficult to per- may speculate that in these patients elevated blood phe levels cause form. Besides ethical issues to conduct such studies measurement of brain dysfunction more likely than in early treated patients with an improvement is difficult. The results are controversial, the study pop- unaffected brain. ulation always small. In some studies severity of behavioral disturbances may be reduced after introduction of a phe restricted diet 4. Outcome in early treated but early discontinued patients [2], Table 1. In a recent placebo controlled study in 17 patients [3] a significant difference between treatment and placebo group using For many years dietary treatment of PKU was terminated at standardized test procedures could not be demonstrated. However, 10 years of age, in some countries even at 6 years of age. Investiga- the positive comments of carers in the treatment group were signifi- tions in patients with early diet termination revealed that there was cant in respect to a positive change than in the placebo group. This in- a loss of intellectual function [6]. Koch and coworkers [7] could dem- dicates a positive effect in individual patients and the authors onstrate that patients continuing the diet until adulthood had a better conclude that “this intervention should be offered to these individ- outcome than those who stopped the diet earlier. In addition in this uals” [3]. According to our personal experiences a close monitoring, collaborative study Koch et al. could demonstrate that treatment not only of the blood phe levels but also of the general health condi- quality also had an influence on outcome: adults who had continued tion of these patients should be performed: excessive weight loss and a dietary control with blood phe levels b15.6 mg/dL had a better IQ development of dystrophy might be a severe side effect after introduction of a low phe diet. We also experienced that treatment of patients living still in the family is easier. Experiences of family members with the diet and a better environment may be reasons Table 2 Neonatal screening failure in 10 patients with phenylketonuria (Trefz, unpublished); for this (Trefz, unpublished). Phe = phenylalanine.

Patient Age at diagnosis Phe-level at diagnosis Reason for not FOR ELECTRONIC(years) USE(mg/dL) ONLYscreening, remarks Table 1 1 1.5 28 Born in Brazil Symptoms and change of symptoms in 6 untreated PKU patients under dietary treatment. 2 9 25 Unclear SIB: self injurious behavior. 3 3.5 32 Mixed blood on screening Adapted from Baumeister and Baumeister [3]. card 4 1.5 24 Lab mistake Patient/symptoms % change Treatment (months) 5 7 21 Russian immigrant 1/SIB −93% 36 67 N20 Russian immigrant 2/Assault/SIB face gouging −91% 26 7 2.5 27 Turkish guest worker −48% family 3/SIB −13% 9 8 0.8 N20 Lab mistake 4/Tantrums −69% 9 91 N20 Lab mistake 5/Stereotypes −54% 3 10 1.5 N20 Lab mistake, infant with 6/Hyperactivity No change 3 Down Syndome and PKU! S28 F. Trefz et al. / Molecular Genetics and Metabolism 104 (2011) S26–S30 than those with higher levels. In addition adults with PKU being off It is difficult to give numbers of the frequency of these symptoms diet had a higher rate of medical problems like eczema, phobias, de- in early and well treated adult PKU patients in comparison to the gen- pression and neurological symptoms. Treatment policy changed so eral population. Most of the studies deal with a heterogeneous sam- that diet for life was suggested [8]. ple including patients who had been treated suboptimally. There is evidence that patients with higher blood phe levels demonstrate 5. Treatment results in early treated patients more often psychiatric symptoms like depression and anxiety. This had clearly been shown in a recent double blind study where 5.1. Cognitive and neuropsychological outcome moods of patients were tested under high and low phenylalanine [15]. Using a profile of “mood states” it could be shown, that under In the following cognitive and neuropsychologic outcomes of high blood phe there was an increase of these symptoms. In addition adult early treated PKU patients are discussed. it could be demonstrated that “fluctuation in tempo during sustained A number of studies have demonstrated that cognitive outcome attention” was higher under increased blood phe levels. Due to the and neurosychologic tests are different from controls. In a meta anal- design of the study these findings are very valuable giving additional ysis by Moyle et al. [9] including 5 studies there were reduced full arguments for a lifelong treatment also for this age group with a tar- scale IQ scores and reduced information processing speed and atten- get level b700–900 μmol/L. tion. In addition in 4 studies inhibitory abilities and motor control were reduced compared to controls. Except for working memory there was no difference to controls (3 studies). In a three year longi- 5.3. Nutritional deficiencies and growth tudinal study from a group in Münster, Germany, Feldman et al. [10] found a significant reduction of full scale IQ compared to a diabetic Since the treatment partly consists of an artificial protein substi- control group (104.9 vs 111.8, pb0.05) but not at baseline. Informa- tute there are concerns about the value of this protein for growth tion processing expressed as time to complete the test was also very and protein status. Some studies have shown that growth is impaired different from the diabetic group (84.9 vs. 67.8 s, p b0.01). There was in early treated PKU children and final height is decreased in compar- evidence that these deficits in information processing were nega- ison to age matched controls [16,17]. A positive correlation between tively correlated with elevated blood phe levels. For children and ad- growth and natural protein intake could be demonstrated in the olescents similar findings were reported by Albrecht et al. [11] from study of Hoeksma et al. [18].Influence of the diet on protein metabo- the group in Heidelberg, Germany, who found in a meta analysis of lism is also shown by decreased protein synthesis expressed by low 20 studies published between 1987 and 2007 an inverse correlation serum prealbumin [19]. between neuropsychological speed in children and adolescents and There is a decreased peak bone mass in young adults [20] and ev- blood phe. However, in adult PKU's “as there is a lack of studies on idence that poor adherence to the diet may influence bone quality at adult PKU's” no such correlation could be found. The authors con- least in children [21]. Others have shown in all age groups decreased clude from these results to keep blood phe in children up to bone mineral density. They also found that docosahexaenoic acid 13 years at b360 μmol/L and in adolescents at b570 μmol/L. Other (DHA) and eicosapentaenoic acid (EPA) and total n-3 fatty acids studies [12,13] were not conclusive to provide target blood phe were significantly diminished in PKU patients compared with healthy levels for adults since the majority of patients investigated had high- controls. DHA, EPA, and total n-3 fatty acids were positively associat- ly increased blood phe levels (usually N1000 μmol/L). However, in ed with bone mineral density [17]. the paper by Channon et al. [13] the diet continuation group scored Absolute and functional vitamin B12 deficiency was described in somewhat better in cognitive function than the discontinuation adult PKU patients opening another risk for brain dysfunction [22,23]. group. Measuring reaction times in a Stroop task Sundermann et al. Our personal experiences support these findings. Many adult PKU's [14] could not show any significant difference in 17 adult PKU pa- have some kind of a “vegan diet” but do not take the amino acid formu- tients with an off diet blood phe level of 1140–1210 μmol/L and an la. Such eating behavior gives them a high risk to suffer from nutritional additional phe load of 100 mg/kg bodyweight to healthy controls. deficiencies without reaching target blood phe levels. Different results were found by ten Hoedt et al. [15] investigating In summary, long-term nutritional follow-up is recommended for nine early treated adult PKU patients in a double blind, placebo con- adult PKU patients, whether they are compliant with the diet or not. trolled study to determine whether mood and sustained reaction is influenced by high blood phe levels (1259 vs 709 μmol/L). The high blood phe group was characterized by impaired sustained attention, 5.4. Biochemical outcome parameters slower reaction times and slower recognition of complex patterns in a visuospatial pattern recognition task. There are different biochemical outcome parameters used to mea- In summary, there are some inconsistent results in early treated sure quality of dietary treatment. Mean blood phe concentrations per adult PKU patients, but most studies show a negative influence of half year, year or lifetime are mostly taken as treatment measure. In a blood phe levels N1000-1200 μmol/L on neuropsychological test bat- meta analysis by Waisbren et al. [24] there was a 1.3 to 4.1-point re- teries and cognitive outcomes. duction in IQ for every 100 μmol/L increase in blood phe in children The effect ofFOR life long exposition ELECTRONIC to high blood phe levels appears 0 to 12 years. Similar USE results were reported ONLY from another group [25]. to be an important issue in adult PKU's. Long term outcome of these Besides mean blood phe fluctuation of blood phe is an important patients is questionned with regard to the occurrence of neurodegen- parameter for outcome [26,27]. In the paper by Burgard et al. [26] it erative disorders (Parkinsonism, Alzheimer's disease or similar cogni- could be shown that the IQ at 9 years of age is more (negatively) cor- tive failure). To this purpose, a long term prospective study including related to the lifelong fluctuation of blood phe than to the mean blood adult PKU's diagnosed through neonatal screening will be launched in phe concentration. Similar results have recently been found by Ana- France in 2011–2012 (more than 200 patients will be enrolled). stasoai etal. who conclude “that stability of blood phe levels may be more important to cognitive functioning than overall exposure to 5.2. Psychiatric findings in early treated adult PKU patients phe in early and continuously treated PKU”[27]. The blood phe/tyrosine ratio may also be used as index of dietary In the review by V.L. Brumm, et al. [12] a number of different psy- control. In a study by Luciana et al. [28] there are advantages to main- chiatric symptoms are described. Symptoms were primarily in the de- taining dietary control for both phe and tyrosine concentrations in pressive category and more frequent in women. adulthood for various neuropsychological parameters. F. Trefz et al. / Molecular Genetics and Metabolism 104 (2011) S26–S30 S29

5.5. CNS findings in early treated adult PKU patients [2] A.A. Baumeister, A.A. Baum eister, Dietary treatment of destructive behavior associated with hyperphenylalaninemia, Clin. Neuropharmacol. 21 (1998) 18–27. [3] P.J. Lee, A. Amos, L. Robertson, B. Fitzgerald, R. Hoskin, M. Lilburn, E. Weetch, G.J. 5.5.1. MRI (magnetic resonance imaging) findings Murphy, Adults with late diagnosed PKU and severe challenging behaviour: a MRI investigations of the brain in PKU patients have gained new in- randomised placebo-controlled trial of a phenylalanine-restricted diet, J. Neurol. Neurosurg. Psychiatry 80 (2009) 631–635. terest after better techniques like diffusion weighted, imaging diffusion [4] F.K. Trefz, S. Cipcic-Schmidt, R. Koch, Final intelligence in late treated patients tensor imaging and “functional” MRI have been applied. By these with phenylketonuria, Eur. J. Pediatr. 159 (Suppl 2) (2000) S145–S148. methods it was possible to better attribute the findings to certain ana- [5] S.D. Grosse, Late-treated phenylketonuria and partial reversibility of intellectual – tomical locations and to quantify them to some extent [29–31].There impairment, Child Dev. 81 (2010) 200 211. [6] M.R. Seashore, E. Friedman, R.A. Novelly, V. Bapat, Loss of intellectual function in was a clear correlation between quality of treatment as e.g. duration children with phenylketonuria after relaxation of dietary phenylalanine restric- of therapy and mean blood phe levels with these changes. However, tion, Pediatrics 75 (1985) 226–232. there is no additional value for treatment monitoring. [7] R. Koch, B. Burton, G. Hoganson, R. Peterson, W. Rhead, B. Rouse, R. Scott, J. Wolff, A.M. Stern, F. Guttler, M. Nelson, F. de la Cruz, J. Coldwell, R. Erbe, M.T. Geraghty, C. Shear, J. Thomas, C. Azen, Phenylketonuria in adulthood: a collaborative study, 5.5.2. CNS metabolic findings J. Inherit. Metab. Dis. 25 (2002) 333–346. [8] National Institutes of Health Consensus Development Conference Statement: In adult PKU patients brain metabolism has also been investigated phenylketonuria: screening and management, October 16–18, 2000. using modern imaging techniques like PET with labeled L-dopamine [9] J.J. Moyle, A.M. Fox, M. Arthur, M. Bynevelt, J.R. Burnett, Meta-analysis of neuro- (FDOPA). [32]. It could be shown that influx and distribution of psychological symptoms of adolescents and adults with PKU, Neuropsychol. – FDOPA is severely impaired and rate of decarboxylation of FDOPA is Rev. 17 (2007) 91 101. [10] R. Feldmann, J. Denecke, M. Grenzebach, J. Weglage, Frontal lobe-dependent markedly reduced. However, even with this metabolic similarity to functions in treated phenylketonuria: blood phenylalanine concentrations and Parkinson disease no special symptoms were observed. The explana- long-term deficits in adolescents and young adults, J. Inherit. Metab. Dis. 28 – tion remains speculative to explain this. (2005) 445 455. fi [11] J. Albrecht, S.F. Garbade, P. Burgard, Neuropsychological speed tests and blood Another nding found recently concerns cerebral protein synthesis phenylalanine levels in patients with phenylketonuria: a meta-analysis, Neurosci. [33]. In adult PKU patients measuring cerebral tyrosine incorporation Biobehav. Rev. 33 (2009) 414–421. rate it could be shown that protein synthesis is reduced with phe levels [12] V.L. Brumm, C. Azen, R.A. Moats, A.M. Stern, C. Broomand, M.D. Nelson, R. Koch, μ Neuropsychological outcome of subjects participating in the PKU adult collabora- above 600 to 800 mol/L. The authors discuss that this impact on cerebral tive study: a preliminary review, J. Inherit. Metab. Dis. 27 (2004) 549–566. protein metabolism would be another argument for lifelong treatment. [13] S. Channon, G. Goodman, S. Zlotowitz, C. Mockler, P.J. Lee, Effects of dietary management of phenylketonuria on long-term cognitive outcome, Arch. Dis. Child. 92 (2007) 213–218. 6. Future management in adult PKU [14] B. Sundermann, B. Pfleiderer, H.E. Möller, W. Schwindt, J. Weglage, J. Lepsien, R.T. Feldmann, Tackling frontal lobe-related functions in PKU through functional Recent (partly double blind) studies in adult PKU patients further sup- brain imaging: a Stroop task in adult patients, J. Inherit. Metab. Dis. 34 (2011) 711–721. port recommendation for lifelong treatment. However, it has become ev- [15] A.E. ten Hoedt, L.M. de Sonneville, B. Francois, N.M. ter Horst, M.C. Janssen, M.E. ident that dietary treatment is suboptimal and continuation to adulthood Rubio-Gozalbo, F.A. Wijburg, C.E. Hollak, A.M. Bosch, High phenylalanine levels often not accepted [34,35]. Therefore other treatment options like phar- directly affect mood and sustained attention in adults with phenylketonuria: a randomised, double-blind, placebo-controlled, crossover trial, J. Inherit. Metab. macologic treatment with sapropterin have to be considered [36,37]. Dis. 34 (2011) 165–171. However, so far there are no long term controlled studies to prove that [16] D. Dobbelaere, L. Michaud, A. Debrabander, S. Vanderbecken, F. Gottrand, D. Turck, J.P. such a treatment is more successful than the diet. Further studies should Farriaux, Evaluation of nutritional status and pathophysiology of growth retardation in patients with phenylketonuria, J. Inherit. Metab. Dis. (2003) 1–11. considertheimpactofbloodpheexposureonbrain development in [17] S. Lage, M. Bueno, F. Andrade, J.A. Prieto, C. Delgado, M. Legarda, P. Sanjurjo, L.J. childhood and brain function at all ages. Mean blood phe concentrations Aldámiz-Echevarría, Fatty acid profile in patients with phenylketonuria and its per half year, year or lifetime are mostly taken as treatment measure. Be- relationship with bone mineral density, J. Inherit. Metab. Dis. (Sep. 10 2010) fl [Epub ahead of print]. sides mean blood phe, uctuation of blood phe over time is important. [18] M. Hoeksma, M. Van Rijn, P.H. Verkerk, A.M. Bosch, M.F. Mulder, J.B. de Klerk, T.J. Burton et al. [38] has shown that fluctuation of blood phe is decreased de Koning, E. Rubio-Gozalbo, M. de Vries, P.J. Sauer, F.J. van Spronsen, The intake by Sapropterin treatment which in the long term may have positive ef- of total protein, natural protein and protein substitute and growth of height and fects on cognitive outcome. head circumference in Dutch infants with phenylketonuria, J. Inherit. Metab. Dis. 28 (2005) 845–854. As discussed, at this time it is difficult to give general treatment rec- [19] J.C. Rocha, M.F. Almeida, C. Carmona, M.L. Cardoso, N. Borges, I. Soares, G. Salcedo, ommendations in adult PKU patients. However, there is no doubt that M.R. Lima, I. Azevedo, F.J. van Spronsen, The use of prealbumin concentration as a late detected PKU patients benefit from strict dietary treatment and biomarker of nutritional status in treated phenylketonuric patients, Ann. Nutr. Metab. 56 (2010) 207–211. are able to catch up in intellectual performance. Untreated, severely re- [20] D. Modan-Moses, I. Vered, G. Schwartz, Y. Anikster, S. Abraham, R. Segev, O. Efrati, tarded patients with behavioral changes may profit from introduction Peak bone mass in patients with phenylketonuria, J. Inherit. Metab. Dis. 30 (2007) of dietary treatment. However, individual decision is necessary and 202–208. [21] F. Porta, A. Mussa, A. Zanin, N.A. Greggio, A. Burlina, M. Spada, Impact of metabolic based on the personal situation of the patient. Long term results in control on bone quality in phenylketonuria and mild hyperphenylalaninemia, J. early treated patients are suboptimal. In addition there are certain Pediatr. Gastroenterol. Nutr. 52 (2011) 345–350. risks as impaired growth, decreased bone mineral density and nutri- [22] A.M. Hvas, E. Nexo, J.B. Nielsen, Vitamin B12 and vitamin B6 supplementation is fi fi needed among adults with phenylketonuria (PKU), J. Inherit. Metab. Dis. 29 tional de cits probably caused by dietary treatment with an arti cial (2006) 47–53. protein substitute and/or missing compliance with an unpleasant diet. [23] I. Vugteveen, M. Hoeksma, A.L. Monsen, M.R. Fokkema, D.J. Reijngoud, M. van The long termFOR risk of a strict dietary ELECTRONIC treatment must be balanced with Rijn, F.J. van Spronsen,USE Serum vitamin B12ONLY concentrations within reference values – μ do not exclude functional vitamin B12 deficiency in PKU patients of various ages, the risk of higher blood phe levels than 600 900 mol/L on cognitive Mol. Genet. Metab. 102 (2011) 13–17. and neuropsychological function and psychiatric symptoms. Saprop- [24] S.E. Waisbren, K. Noel, K. Fahrbach, C. Cella, D. Frame, A. Dorenbaum, H. Levy, terin treatment for responsive patients should lead to better phe con- Phenylalanine blood levels and clinical outcomes in phenylketonuria: a systemat- – trol, less phe fluctuation and should provide a diet with higher ic literature review and meta-analysis, Mol. Genet. Metab. 92 (2007) 63 70. [25] K.S. Viau, H.J. Wengreen, S.L. Ernst, N.L. Cantor, L.V. Furtado, N. Longo, Correlation amount of natural protein for these patients. Further studies must of age-specific phenylalanine levels with intellectual outcome in patients with prove if adults with PKU may also profit from a better treatment. phenylketonuria, J. Inherit. Metab. Dis. (May 10 2011) [Epub ahead of print]. [26] P. Burgard, A. Rupp, D.S. Konecki, F.K. Trefz, H. Schmidt, U. Lichter-Konecki, Phe- nylalanine hydroxylase genotypes, predicted residual enzyme activity and phe- References notypic parameters of diagnosis and treatment of phenylketonuria, Eur. J. Pediatr. 155 (Suppl 1) (1996) S11–S15. [1] F. Feillet, F. van Spronsen, A. MacDonald, F.K. Trefz, M. Demirkol, M. Giovannini, A. [27] V. Anastasoaie, K. Laura, P. Forbes, S. Waisbren, Stability of blood phenylalanine Bélanger-Quintana, N. Blau, Challenges and pitfalls in the management of phenyl- levels and IQ in children with phenylketonuria, Mol. Genet. Metab. 95 (2008) ketonuria, Pediatrics 126 (2010) 333–341. 17–20. S30 F. Trefz et al. / Molecular Genetics and Metabolism 104 (2011) S26–S30

[28] M. Luciana, J. Sullivan, C.A. Nelson, Associations between phenylalanine-to- [34] J.H. Walter, F.J. White, Blood phenylalanine control in adolescents with phenylke- tyrosine ratios and performance on tests of neuropsychological function in ado- tonuria, Int. J. Adolesc. Med. Health 16 (2004) 41–45. lescents treated early and continuously for phenylketonuria, Child Dev. 72 [35] N. Blau, F.J. van Spronsen, H.L. Levy, Phenylketonuria, Lancet 376 (2010) 1417–1427. (2001) 1637–1652. [36] H.L. Levy, A. Milanowski, A. Chakrapani, M. Cleary, P. Lee, F.K. Trefz, B. Chester, C.B. [29] B. Pérez-Dueñas, J. Pujol, C. Soriano-Mas, H. Ortiz, R. Artuch, M.A. Vilaseca, J. Cam- Whitley, F. Feillet, A.S. Feigenbaum, J.D. Bebchuk, H. Christ-Schmidt, A. Dorenbaum, pistol, Global and regional volume changes in the brains of patients with phenyl- for the Sapropterin Research Group*, Efficacy of sapropterin dihydrochloride (tetra- ketonuria, Neurology 66 (2006) 1074–1078. hydrobiopterin, 6R-BH4) for reduction of phenylalanine concentration in patients [30] X.Q. Ding, J. Fiehler, B. Kohlschütter, O. Wittkugel, U. Grzyska, H. Zeumer, K. Ullrich, with phenylketonuria: a phase III randomised placebo-controlledstudy, Lancet 370 MRI abnormalities in normal-appearing brain tissue of treated adult PKU patients, (2007) 504–510. Magn. Reson. Imaging 27 (2008) 998–1004. [37] F.K. Trefz, B.K. Burton, N. Longo, M.M. Casanova, D.J. Gruskin, A. Dorenbaum, E.D. [31] P.J. Anderson, V. Leuzzi, White matter pathology in phenylketonuria, Mol. Genet. Kakkis, E.A. Crombez, D.K. Grange, P. Harmatz, M.H. Lipson, A. Milanowski, L.M. Metab. 99 (Suppl 1) (2010) S3–S9. Randolph, J. Vockley, C.B. Whitley, J.A. Wolff, J. Bebchuk, H. Christ-Schmidt, J.B. [32] C. Landvogt, E. Mengel, P. Bartenstein, H.G. Buchholz, M. Schreckenberger, T. Hennermann, Sapropterin Study Group, Efficacy of sapropterin dihydrochloride Siessmeier, A. Scheurich, R. Feldmann, J. Weglage, P. Cumming, F. Zepp, K. Ullrich, in increasing phenylalanine tolerance in children with phenylketonuria: a Reduced cerebral fluoro-L-dopamine uptake in adult patients suffering from phe- phase III, randomized, double-blind, placebo-controlled study, J. Pediatr. 154 nylketonuria, J. Cereb. Blood Flow Metab. 28 (2008) 824–831. (2009) 700–707. [33] M. Hoeksma, D.J. Reijngoud, J. Pruim, H.W. de Valk, A.M. Paans, F.J. van Spronsen, [38] B.K. Burton, H. Bausell, R. Katz, H. Laduca, C. Sullivan, Sapropterin therapy in- Phenylketonuria: high plasma phenylalanine decreases cerebral protein synthe- creases stability of blood phenylalanine levels in patients with BH4-responsive sis, Mol. Genet. Metab. 96 (2009) 177–182. phenylketonuria (PKU), Mol. Genet. Metab. 101 (2010) 110–114.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S31–S39

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Minireview Follow up of phenylketonuria patients☆,☆☆

M. Demirkol a,⁎,M.Giżewska b, M. Giovannini c, J. Walter d a Div. Nutrition and Metabolism, Children's Hospital, Istanbul Medical Faculty, Istanbul University, Istanbul, Turkey b Department of Pediatrics, Endocrinology, Diabetology, Metabolic Diseases and Cardiology, Pomeranian Medical University, Szczecin, Poland c Department of Pediatrics, San Paolo Hospital, University of Milan, Milan, Italy d Willink Biochemical Genetics Unit, Central Manchester University Hospitals NHS Foundation Trust, Manchester, UK article info abstract

Article history: In recent years our understanding of the follow up policies for PKU has increased substantially. In particular, Received 31 May 2011 we now understand the importance of maintaining control of blood phenylalanine (phe) concentrations life- Received in revised form 31 July 2011 long to achieve the best long-term neuropsychological outcomes. The concordance with the follow up Accepted 7 August 2011 strategy remains a key challenge for the future, especially with respect to adolescents and young adults. The Available online 10 August 2011 recent therapies could ease the burden of the dietary phe restriction for PKU patients and their families. The time may be right for revisiting the guidelines for follow up of PKU in order to address a number of important Keywords: Phenylketonuria issues related to PKU management: promotion of breastfeeding to complementary feeding up to 2 years of age Follow up for prevention of early growth retardation and later overweight development, treatment advancements for Management metabolic control, blood phe and tyr variability, routine screening measures for nutritional biomarkers, Outcome neurocognitive and psychological assessments, bone pathology, understanding the challenges of compliance and transitioning into adulthood as an individual with PKU and addressing unmet needs in this population. © 2011 Elsevier Inc. All rights reserved.

Contents

1. Introduction ...... S32 2. Review of the guidelines and statements for the management of PKU ...... S32 3. To assess outcome quality during the follow up ...... S32 3.1. Monitoring of blood phe concentrations during the treatment ...... S32 3.2. The frequency of blood phe monitoring for each age range ...... S33 3.3. When and how to take the blood samples for phe evaluation ...... S33 3.4. Monitoring of blood tyr concentrations during the treatment ...... S33 4. Nutritional monitoring ...... S34 4.1. Nutritional characteristics of PKU diet and challenges in the follow up ...... S34 4.2. Micronutrient deﬁciencies ...... S34 4.3. Growth monitoring to prevent early growth retardation and later overweight development ...... S35 4.4. Determination of phe tolerance ...... S35 4.5. Evaluation of bone mineral status ...... S35 5. Neurological follow up in PKU ...... S35 6. Keeping to follow up recommendations: from compliance and overcoming adherence barriers to concordance ...... S36 7. Transition in PKU ...... S36 8. ConclusionsFOR...... ELECTRONIC USE ONLY S37

Abbreviations: PKU, phenylketonuria; phe, phenylalanine; tyr, tyrosine; LCPUFA, long-chain polyunsaturated fatty acid; DHA, docosahexaenoic acid. ☆ References to electronic databases: Phenylketonuria, OMIM 262600. ☆☆ Presented at the 3rd European Phenylketonuria Group (EPG) Symposium “Advances and Challenges in PKU”, Serono Symposia International Foundation, March 25–26, 2011 in Lisbon, Portugal. ⁎ Corresponding author. Fax: +90 212 631 1861. E-mail addresses: [email protected] (M. Demirkol), [email protected] (M. Giżewska), [email protected] (M. Giovannini), [email protected] (J. Walter).

Conﬂict of interest ...... S37 References ...... S37

1. Introduction tetrahydrobiopterin responsiveness, all other guidelines focus on nutritional treatment. Phenylketonuria (PKU; OMIM 261600), is an inherited metabolic disease requiring prompt intervention after early diagnosis with neonatal 3. To assess outcome quality during the follow up screening. PKU treatment is recommended for life, but non-compliance with the nutritional management is common. There is a growing body of 3.1. Monitoring of blood phe concentrations during the treatment evidence that neurocognitive, psychosocial, quality of life, growth, nutrition and bone pathology outcomes are suboptimal [1–3]. Although there is no doubt that PKU should be diagnosed early and Management for affected individuals of all ages can be difficult and treated from the very first days of life, there are a number of is enhanced with the teaching and support of an experienced controversies regarding subsequent management policies. There is healthcare team consisting of metabolic physicians, nutritionists, general agreement on the necessity of careful monitoring of phe levels in genetic counselors, social workers, nurses, and psychologists [4]. infancy and early childhood, but recommendations regarding the target Despite normal IQ, early treated children and adolescents have a phe concentrations differ significantly among the clinical centers higher frequency of attention deficit/hyperactivity disorder, de- [2,15,18,19]. Recently, Blau et al. [2] showed the worldwide complexity creased autonomy, and school problems compared to either healthy of the target blood phe concentrations in the follow up of PKU controls or chronically ill peers [2,5–7]. There is some evidence that patients as recommended for treatment policy in 17 countries. The childhood metabolic control correlates with development of these lower target phe levels for PKU patients are in the range of 40 to problems. Further evidence that could serve as a basis for reassess- 130 μmol/L. During life-time period the upper target blood phe ment the criteria of the ‘follow up of PKU patients’ is needed. concentrations are recommended in two to five stages by different countries (Table 1). The upper target phe concentration in PKU patients below 10 years 2. Review of the guidelines and statements for the management of age does not exceed 480 μmol/L except in Denmark (phe concentra- of PKU tion between 7 and 9 years of age b600 μmol/L). The most frequently recommended phe concentration for children below 10 years of age is PKU treatment policies vary not only between different countries between 120 and 360 μmol/L [1,2,20]. This particular strict metabolic worldwide, but also sometimes within an individual country. An control in the first years of life arises from the knowledge about increasing number of national guidelines and recommendations on the significant sensitivity of a young child's brain to neurotoxic influence of management of PKU have emerged that mainly deal with the subjects as hyperphenylalaninemia. Despite such clear recommendations in about the blood phe concentration at which phe restriction should be initiated; 25% of blood samples from patients below 10 years of age, phe the recommended blood phe concentrations at different ages and the concentration is above or below guideline goals [19]. recommended frequency of monitoring phe in blood. There is consensus Studying PKU individuals who were detected by newborn between the guidelines and recommendations that treatment should screening, Smith et al. [21] reported a gradual decrease in IQ, start as early as possible, and monitoring of blood phe concentration and measured by standardized testing (Stanford Binet) that correlated clinical parameters should continue throughout life [8–12]. with increasing increments of blood phe during the first 4 years of life. All guidelines mention the importance of ‘diet for life’ but most of They provide compelling evidence for the importance of keeping the them recommend on the care of children but not on the care of adults. blood phe values below 360 μmol/L in early childhood. Average blood It is of importance that adult patients, both willing and unwilling to be phe levels of 650 μmol/L resulted in a decrement of ~11 IQ points. treated, need a guideline for care and follow up [13–15]. Guidelines Looking for evidence from in vitro studies, investigations of blood– should also address issues of follow up, nutritional risks, neuropsy- brain barrier transport reveal that a blood phe concentration of chological outcome, quality of life, psychosocial support and the 200 μmol/L already saturates the blood–brain barrier transport, inhibit- possibility of various forms of current treatment. With the exception ing the transport of other large neutral amino acids [22]. In the study of of the articles by Levy et al. [16] and Blau et al. [17] dealing with Huijbregts et al. [23] it was found that early and continuously treated

Table 1 Upper target phenylalanine concentrations (μmol/L) as recommended for treatment of phenylketonuria in different countries (N. Blau et al., Lancet 2010). Age groupFOR Two stages ELECTRONIC Three stages USE Four stagesONLY Five stages Australia1 Italy Switzerland Spain USA Hungary Austria Turkey Croatia France Japan Denmark Poland The Netherlands U.K.3 Germany Portugal2

Infancy and young b3501 b360 b300 b360 b360 b360 b240 b240 b240 b300 b240 b300 childhood 0–2y* b360 (b4 y*) Childhood 3–6y* b400 b360 b360 b400 (4–8 y*) Childhood 7–9y* b480 b480 (7-14y*) b600 (8-10y*) Adolescence 10–12 y* b4501 b600 b900 b600 b600 b600 b480 b700 b720 Adolescence 13–15 y* b4802 b720 b600 b900 b600 b900 b600 b7003 Adulthood N16 y* b900 b600 (N14 y*) b1200 b900 b960 b1200 b900 y*: years of age. M. Demirkol et al. / Molecular Genetics and Metabolism 104 (2011) S31–S39 S33 patients aged 7–14 years, whose phe control was poor (blood phe infections, intra- and interday variations in blood phe concentrations, concentration N360 μmol/L), demonstrated a number of neurocognitive energy intake, dosage and timing of protein substitute, intake of natural deficits if compared with the individuals with good phe control and protein and growth velocity, may affect the interpretation of results matched health controls. [25,46]. The British guidelines [10] recommend that blood phe levels The influence of phe on the brain functioning is not only a result of should be monitored at least weekly during infancy through 4 years of constantly elevated phe levels, but also reflects variability of blood phe age, every fortnightly until 10 years of age, and monthly thereafter. which might affect cognitive performance. This variability can be Maintaining blood phe levels of control is difficult in adolescents and related not only to phe intake by itself but also to distribution of young adults with PKU. Patients in these age groups will have to make protein substitute that should be spread evenly through the day choices concerning their phe intakes [47]. Table 2 recommends the [24,25]. Therefore, it is important to advise patients about the frequency of blood phe monitoring in different age groups. necessity to improve their blood phe stability [2]. There is no data indicating that strict treatment is indispensable in 3.3. When and how to take the blood samples for phe evaluation all adolescent and adults with PKU [15,26–28]. Although elevated phe concentrations seem to have smaller influence on brain functioning The British guidelines [10] recommend taking the blood samples beyond early childhood [23], there is evidence that numbers of for phe evaluation in the early morning after an overnight fast at the neuropsychological and neurophysiological dysfunctions can appear time of natural peak levels. Blood is recommended to be taken using in older patients as the consequence of losing metabolic control [29– the Guthrie card. School-aged children should be encouraged to 37]. Constant elevation of blood phe levels can result in deficits in become responsible for their management and obtaining their own executive functions, processing speed and fine motor control. finger prick samples for blood phe measurements [46]. Neurological abnormalities such as tremor and spasticity as well as emotional and psychiatric illnesses can also occur [2,15,26,38–41]. 3.4. Monitoring of blood tyr concentrations during the treatment In PKU adolescents (10 to 15 years of age) the advised phe concentrations fall within the wide range from 40 to 900 μmol/L, with Tyr, a precursor of thyroxine, melanin, and the neurotransmitters the most common range of 120–600 μmol/L [2]. In PKU adults dopamine and norepinephrine, is an essential amino acid in PKU. (N16 years of age) the upper recommended values varied from 450 Reference values of tyr concentrations in plasma vary between 35 and up to 1200 μmol/L (Table 1). Based on a survey of 10 European 102 μmol/L for different age groups and sexes when measured after an centers, Ahring et al. [19] concluded that on average only in 65% of overnight fast [48]. In the nonfasting condition in healthy young samples from patients above 16 years of age phe values were within adults, plasma tyr concentrations vary between 61 and 99 μmol/L [49]. or below guideline goals. These results of Walter et al. [1] are even Adeficiency of the amino acid tyr has been suggested as a cause of worse; they found that in patients aged 15–19 years, the proportion of some of the neuropsychological problems exhibited in PKU. From the samples with phe concentration greater than recommended was 79%. start of PKU treatment in 1954, tyr was added to the protein Limited studies have shown that resumption of low-phe diet or substitutes to a content equal to that found in human milk [50]. The interposing of alternative treatment like sapropterin or large neutral current practice in the monitoring and/or treatment of tyr levels in amino acids can result in improvement in these patients quality of life children with PKU, across 12 countries from Australasia, North [26,27,42–44]. A short-term double-blind placebo control study, using America and Europe showed that tyr monitoring in this population large neutral amino acids in patients with PKU, showed a lowering of has increased over the last 5 years [51]. Supplemental tyr has been blood phe concentration by an average of 39% from baseline [45]. reported to ameliorate symptoms of both attention deficit hyperac- These data clearly show a lack of consensus in the monitoring of tivity disorder and depression in this population [52,53]. blood phe concentrations during the follow up of PKU patients. Investigations into the biochemical markers associated with exec- Relevant questions that need to be considered are: utive function impairment in children with early and continuously

• Should the recommendation for the upper target blood phe concentrations be in two or three stages (infancy and childhood, Table 2 Monitoring treatment and outcome of PKU patients. adolescence/adulthood) for better control during life-time period? • Should the most recommended upper target level for phe of Age group Monitoring blood phe levels Clinical monitoringa,b 360 μmol/L be used until 10 years of age? Infant and young child Weekly Every 1–3 monthsc,d • For PKU adolescents the recommended target Phe levels are 0–2 years of age between 360 and 900 μmol/L. Should we recommend 600 μmol/L Child Weekly–fortnightly Every 3–6 monthsd,e,f 3–9 years of age for this age-period? – e,f • Adolescent Fortnightly monthly Every 6 months Do we need such strict treatment in adulthood as recommended in 10–15 years of age Australia (450 μmol/L) and Portugal (480 μmol/L) when compared Adult Monthly–two monthly Yearlye,f to Hungary, Italy and Switzerland (600 μmol/L) and the U.K. N16 years of age (700 μmol/L), and Poland and Spain (720 μmol/L)? Should the a Body weight, height, head circumference, general status of health, neuropsychological target phe level be relaxed to 900 μmol/L as in Denmark, Japan, assessment; 3-day diet record for evaluation of the intake of macro- and micronutrients Turkey andFOR USA or to 1200 ELECTRONICμmol/L as in Austria, France and and phe-tolerance justUSE before evaluation at outpatientONLY clinics; family therapy when dietary Germany? We need evidence in this period when the adherence recommendations are not followed; when phe levels are within the recommended range, fi in general no additional routine laboratory analysis is necessary. to diet is most dif cult. b A complete fasting profile of all amino acids, vitamins, minerals and trace elements, • The recommended lower target phe levels for PKU patients are in carnitine, calcium and phosphorus metabolism, blood count, ferritin, albumin, the range of 40 to 130 μmol/L in different countries. Do we cholesterol, HDL- and LDL-cholesterol, homocysteine, fatty acids, serum methylmalonic recommend 60 μmol/L for the lower target phe level? acid may be indicated yearly and in individuals with poor compliance. c Promotion of complementary feeding to breastfeeding up to 2 years of age. d Evaluation of body weight, height, head circumference, weight/height, body mass 3.2. The frequency of blood phe monitoring for each age range index on the z-score percentile curves (WHO Anthro, version 3.2.2, January 2011, http://www.who.int/childgrowth/software/en/) for detecting early growth retardation and overweight in the first 5 years. Blood phe concentrations can be misleading unless carried out e Body mass index evaluation on the percentile curve for 2 to 20 years for detecting repeatedly at regular intervals. Some patients will deliberately lower early overweight and obesity. their phe intake just prior to blood sampling. Other factors, such as f Bone density every 2nd year. S34 M. Demirkol et al. / Molecular Genetics and Metabolism 104 (2011) S31–S39 treated PKU remain largely phe-only focused, despite experimental data have been considered as a possible cause, for example the lower showing that a high phe to tyr ratio (phe:tyr) is more strongly associated content of phe compared with the standard infant formulas [64], the with executive function deficit than phe alone. A high phe:tyr ratio is optimal phe:tyr ratio, and the presence of LCPUFAs, particularly hypothesized to lead to a reduction in dopamine synthesis within the arachidonic acid and DHA, which influence positively the growth and brain, which in turn results in the development of executive function the brain functional development [65]. Human milk is a low-protein impairment. Sharman et al. [54] showed that lifetime measures of phe: food (an average content of 0.9–1.0g protein/100ml; 50mg tyr were more strongly associated with executive function than phe- phe/100 ml and 50 mg tyr/100 ml); in contrast, a standard infant only measures. Children with a lifetime phe:tyr less than 6 demon- formula contains two to three times as much phe per 100 ml [3]. The strated normal executive function, whereas children who had a lifetime closer mother-infant bonding from breastfeeding plays an additional phe:tyr above 6, on average, demonstrated clinically impaired executive role on the development. function. The nutritional therapy for PKU requires restricted phe intake, with the majority of protein and other nutrients coming from 4. Nutritional monitoring synthetic medical food. By using a phe free infant formula supplemented with LCPUFA, the risk that infants with PKU will have In PKU severe neurological and functional disorders can be suboptimal DHA status may be minimized. The early DHA status, prevented through an early strict nutritional treatment aimed at irrespective of any specific supplementation, is associated with reducing blood phe to non-toxic levels through a natural-protein- indices of visual maturation [66]. In adult patients with PKU, the restricted diet. The evidence of structural alterations of the brain long-term use of ready-to-drink liquid protein substitutes is associ- tissue [55] and of anomalies of the visual function, evaluated by visual ated with better compliance by lowering blood phe and improving evoked potentials [56], also in well nourished PKU children has raised nutritional biochemical markers [67]. questions concerning the efficacy of this treatment [57,58]. Long-term In dietary schedules of PKU the quantities of permitted natural, dietary guidance and monitoring of the nutritional status of patients unmodified, foods are calculated from a method of phe equivalence with PKU should be part of a follow up program that continues for life. dependent upon how much phe a food contains for a given weight It is a fundamental importance that all professionals, and the families that is not unique in different PKU centers. It leads to difficulties in the themselves, fully understand the principle and the practice of the compliance because of the need of weighing the foods. nutritional therapy. Treatment options for the management of PKU are expanding, with the introduction of new possibilities, such as 4.2. Micronutrient deficiencies large neutral amino acids, sapropterin and, potentially, phe ammonia lyase or gene therapy. However, these options do not always allow a In the patients with PKU lower intakes have been observed for complete relaxation of the patient's diet, and nutritional follow up is, some micronutrients as vitamins A, C and E, selenium, coenzyme Q10, therefore, essential [3]. vitamins B2, B6 and B12, and folates (which can increase homocysteine levels in the blood), iron, zinc, calcium, carnitine, LCPUFAs and, if 4.1. Nutritional characteristics of PKU diet and challenges in the follow sun exposure is not adequate, vitamin D. Special attention should be up given to dietary practices that enhance the absorption of zinc and iron from plant foods [3]. According to some studies, such children may A semi-synthetic diet low in phe and adequate in other nutrients is also have particularly low levels of calcium. Breast-fed infants may used to treat PKU patients. The PKU diet is mainly made up by fruits need supplements of vitamin B12 if the maternal diet is inadequate; and vegetables chosen on the basis of their phe content, variable older infants may need zinc supplements, reliable sources of iron, amounts of low-phe foods, and and phe-free synthetic products vitamins B12 and D. providing essential amino acids in suitable proportions. The PKU diet PKU patients are at risk for functional vitamin B12 deficiency based may be looked upon as a medically prescribed vegan-like diet [3]. This on their diet. A vitamin B12 concentration within the reference range type of dietary regimen provides a high-carbohydrate, low-saturated does not automatically imply a sufficient vitamin B12 status. and long-chain polyunsaturated fat, and low-cholesterol intake Vugteveen et al. [68] recommend measuring serum methylmalonic because of the very low intake of phe-containing animal foods. The acid, or alternatively plasma homocysteine, yearly in all PKU patients nutritional management of PKU is complex and time-consuming, to diagnose functional vitamin B12 deficiency. requiring knowledge of foods and recipes, cooking skills, continuous DHA depletion has been demonstrated in children with PKU and food measurement. The constraint of a diet that is ultimately focused may account for detectable subtle neurological deficits that are not at the threshold of a calculated phe intake bears the risk of nutrient explained by variation in plasma phe concentrations [69]. DHA deficiency. Therefore, the treatment must be monitored by regular supplementation (15 mg/kg body weight) resulted in significantly control of a dietary intake, neurological, physical, intellectual and faster visual evoked potential latencies, indicating more rapid central behavioral development, as well as biochemical monitoring. Table 2 nervous system information processing. In addition, DHA significantly summarizes the recommendations for monitoring treatment and improved outcomes in a test of motor function and coordination. No outcome of PKU patients. changes over time were seen in age-matched healthy controls. The In the PKU populationFOR low rates ELECTRONIC of breastfeeding were found [59]. relationship between USE DHA and domains ONLY related to verbal ability, such In view of the major biological and developmental advantages, human as learning and memory, should be confirmed in a controlled trial breast milk should be given greater consideration in PKU than is [70]. However it is important to prolong DHA supplementation to currently granted. Demand breastfeeding combined with prescribed maintain the functional advantages considering that 3 years after amount of a phe free amino acid formula (phe free amino acid supplementation no biochemical or functional differences between supplement during the second year), the amount adjusted according supplemented and unsupplemented children have been observed to blood phe levels, to complementary feeding should be promoted [71]. Since PKU children are generally healthy and have normal during the first 2 years of life as World Health Organization energy and fatty acid metabolism, childhood populations in general recommends to the healthy children [60–62]. Breastfeeding has a require preformed LCPUFAs to achieve optimal neurological function beneficial effect on the development of PKU children [63]. PKU [72]. Further studies are ongoing aiming at establishing quantitative patients who had been breastfed as infants scored significantly better DHA requirements in children [73]. in IQ (advantage of 14.0 points, p=0.01) than children who had been LCPUFA status was within the reference values in PKU patients formula-fed. Several factors regarding the composition of human milk treated with tetrahydrobiopterin but they did in comparison with M. Demirkol et al. / Molecular Genetics and Metabolism 104 (2011) S31–S39 S35

PKU on protein-restricted diet (pb0.0001). DHA values correlate with 4.5. Evaluation of bone mineral status the index of dietary control only in PKU patients on protein-restricted diet (p=0.002) [74]. PKU itself as well as its low-phe treatment based on dairy- restriction carries the risk of calcium deficiency [84]. Several authors 4.3. Growth monitoring to prevent early growth retardation and later have already shown that in PKU patients low bone mineral density overweight development and high incidence of fractures occur [85–87]. As result of decrease peak bone mass, patients may present not only osteopenia (Z-score Several studies have shown that transient growth retardation is between −1.0 and −2.5) but also clinically relevant osteoporosis (Z- common in children treated for PKU during the first years of life, score b−2.5). Modan-Moses et al. [88] detected compromised bone specially from birth to the age of 3–5 years [75–77]. Studies that mineral density in 42% of 31 adults with PKU. Several potential investigated causes for the retarded height growth in PKU patients contributors to low bone mineral density in PKU individuals should be focused on low blood phe concentrations, a deficiency of tyr or a considered. The main ones being: prolonged restriction of natural deficient intake of total protein. However, the first two suggestions protein with long-standing deficiency in protein, calcium, phosphate, could not be shown to cause growth restriction in length [77]. vitamin D or trace elements, intake of insufficient amount of phe free Theoretically, the conclusion that the total protein intake is not protein supplement, variation of plasma phe concentration with lack related to growth does not exclude the possibility of a relation of dietary compliance and phe levels N1200 μmol/L and low physical between growth and one of the fractions of the protein that PKU activity [88–91]. Another factor is specific lipids profile in patients patients consume (natural protein and the protein substitute). A with PKU with almost total absence of LCPUFA that have been relation with natural protein rather than total protein would favor proposed as potential contributors to bone modeling and remodeling research to further optimize the biological composition of protein process [92,93]. substitutes. Although there is a common agreement regarding the necessity of In opposition to these data, Acosta PB et al. [78] showed that PKU the evaluation of bone mineral status in patients with PKU, the reality children undergoing nutrition management, aged from 2 to 13 years differs among centers. In the study by van Spronsen et al. [81] only 1 of old, presented a normal linear growth. Moreover, the mean body mass 14 PKU clinics evaluated bone density in children below 10 years of index z-scores at the end of the study suggested that many children age. In older patients the frequency of these measures increased, but were overweight. Studies showed that PKU children might weigh still it was not performed in 50% of centers for individuals aged 10–16 more than normal children [78,79]. The rate of overweight at age of and in 41% of centers for older patients. There is no doubt that patients 8 year was around 25% [80], that was comparable to values in normal with PKU should be monitored in terms of their skeletal status and populations [79]. Therefore, even if PKU children are routinely long- body composition. There are still many controversies regarding the term monitored for dietary intake, there are data showing evidence choice of the best monitoring tools. The technique of dual-energy X- for overweight in this population. Metabolic pediatricians and ray absorptiometry (DXA) that is still considered “a gold standard” dietitians should carefully monitor energy intake and quality of has some limitations. Therefore, intensive research is conducted to carbohydrates in PKU diet, physical activity and body weight in PKU find new methods. The use of functional muscle-bone analysis based patients (Table 2). on densitometric parameters seems to be an interesting solution [84]. Early detection of deficits in bone mineral status allows introduction 4.4. Determination of phe tolerance of corrections in dietary plan, making an effort to obtain better compliance and initiating a specific treatment e.g. therapy with Knowledge about the severity of phe hydroxylase deficiency is calcitriol [87]. essential in everyday life of PKU patients. Degree of phe hydroxylase deficiency is usually defined by either pre-treatment phe concentra- 5. Neurological follow up in PKU tion or phe tolerance at 5 years of age. With the worldwide tendency for very early introduction of low-phe diet, pre-treatment phe The outcome for patients with PKU treated from the first weeks of concentration has lost its significance. Therefore, the evaluation of life is generally excellent with children attaining developmental the phe-tolerance at different ages (i.e. the amount of phe that patient milestones and attending normal schools. DQ/IQ assessments identify can tolerate having the phe concentration below the upper limit of the large majority as falling well within the normal range. Whereas recommended range) still has high predictive value in determining there may be little benefit in undertaking formal and time consuming the course of the disease. Phe-tolerance seems to be especially assessments of development in these children those whose control is interesting with reference to the new treatment strategies that poor however are likely to do less well and can show global change the phe tolerance and nowadays are offered to the growing developmental. Early identification of developmental problems to number of patients [81]. The problem that still remains is the most direct appropriate support then becomes important. Significant accurate assessment time of the phe tolerance. Güttler [82] consid- clinical neurological abnormalities are not seen in well controlled ered the age of 5 years the most appropriate time for this purpose. patients with PKU so that formal neurological examination is not Because for many patients it seems too late, some investigators try to usually required. evaluate theFOR phe tolerance in younger ELECTRONIC children. Based on the analysis The situation USE for adults with PKUONLY is less straightforward. Firstly of phe tolerance of 213 early and continuously treated Dutch PKU abnormalities on neurological examination are well described and patients, van Spronsen et al. [81] concluded that phe tolerance can include brisk tendon reflexes, intention and resting tremor. More already reliably be assessed at the age of 2 years and that it is a severe signs such as spasticity are rare but may occur in those off valuable predictor of the tolerance at 10 years of age. According to treatment particularly if they are vitamin B12 deficient. So in adults an these authors, before the age of 2 years, the phe tolerance and annual routine neurological clinical examination would seem appro- metabolic control of PKU are influenced by many factors including priate. Secondly magnetic resonance neuroimaging is almost always growth velocity, infections and energy intake. Therefore, in younger abnormal in adults with blood phe levels consistently above children this parameter does not have strong predictive value. On the 800 μmol/L showing evidence of white matter changes [94]. The other side, Ponzone et al. [83] postulated to evaluate phe tolerance relevance of such abnormalities to clinical manifestations has not even in the first months of life, when the baby's food is not composite been demonstrated in the majority studies and adverse long term and passively received. This information seems to be a reliable consequences have yet to be proven [95]. Similarly the role of predictor of the life-long phe tolerance. magnetic resonance spectroscopy has not been defined [96]. S36 M. Demirkol et al. / Molecular Genetics and Metabolism 104 (2011) S31–S39

Lastly an increasing number of studies have demonstrated subtle report improvements in their ability to concentrate on tasks at work, neurophysiological and neuropsychological abnormalities that may home, or school. be related to blood phe control [41,97–99]. The importance of these to Increased and improved health professional accessibility is being the day to day functioning of individuals may depend on their introduced in novel ways: the provision of clinical services through occupation and social circumstances. It remains to be determined the communications and information technologies ‘telemedicine’ [106], role of different neurofunctional tests outside of research studies. ‘dietetic home visiting’ [107], ‘weekend hospital clinics’ [108], ‘home In conclusion it is not yet possible to make recommendations as to delivery trial of protein-substitute and low protein special foods’ [109]. whether neuroimaging and/or neurophysiological should be part of Concordance is an agreement reached after negotiation between a routine follow up. Hopefully there will be more clarity on this issue in patient and a health care professional that respects the beliefs and the future. wishes of the patient, an alliance in which the health care professionals recognize the primacy of the patient's decisions [110]. It refers to a 6. Keeping to follow up recommendations: from compliance and partnership between the patient/carer and clinician, which involves overcoming adherence barriers to concordance shared, interactive, decision making about treatment goals and strategy, will help enhance compliance. Management and concordance are a task Compliance with treatment is most often satisfactory in infancy for patients and professionals [111,112].Inordertoimprovethe and childhood. However the special diet severely interferes with concordance with recommendations for management guidelines should culturally normal eating habits, particularly in older children and provide an answer not only to the question “what has to be done” but adolescents. This often results in problems keeping to follow up also “how can it be done”. It may also legitimize retrieval of patients, at recommendations. Nutritional treatment of PKU is highly demanding least for follow up so that help can be offered with their treatment if and for patients and families, and is almost impossible without the when they choose to restart or if becomes necessary [15]. support of a therapeutic team trained to work in an interdisciplinary The compliance measures used in the clinical setting include way in a treatment center [100,101]. laboratory methods (percentage of blood phe concentrations within The primary obstacles to better adherence are time constraints and target range, annual mean or median blood phe concentration) and stress associated with food preperation and record-keeping, and the quantitative nonlaboratory methods (percentage of blood samples restrictions imposed on social life. Reasons of the failure of PKU returned by patients and their carers; attendance at outpatient clinics, treatment may be parents' inability to administer a low-phe diet to their self-reported dietary records, protein substitute prescription records). child because of family situations and backgrounds as well as barriers Assessment of blood phe concentrations is perhaps the best overall with language, psychological, and cultural communication [102]. measure, but there is no agreement about the number of phe There is now greater emphasis on training children to take an active concentrations that should be within target range or frequency or role in the daily management of PKU and contribute to decision making timing of measurement [110]. about their own treatment. From early school age onward, practitioners A PKU management strategy that goes beyond simply treating need to begin engaging the PKU individual directly during visits in a blood phe levels can empower and enable people born with PKU to discussion of the impact of the condition on their daily lives. They must achieve similar life goals as those born without PKU. also begin to develop a partnership with their patient, with increasing responsibility being placed on the individual with PKU to both 7. Transition in PKU understand the condition and to manage it. This involves developing increasing independence in maintaining diet records, preparing foods, Whereas in the past the majority of patients with PKU continued taking medication, doing blood tests, and monitoring the results. under the care of their metabolic pediatrician this is no longer the case Meanwhile, parents need to remain involved in a supervisory role. As where adult inborn error services have been established. The transition the individual with PKU moves into adolescence, management of the from pediatric to adult care in patients with PKU is fundamentally a little disorder needs to shift from the parent to the child. This process must be different from those with other chronic non-life threatening conditions facilitated both by the parents and by the health care team. such as diabetes and the principals remain much the same. There are a During adolescence the main challenge is the typical relaxation of the number of publications that are relevant and which summarize the diet, regardless of the guidelines proposed by the medical team. There is important steps in this process and make recommendations as to good an absence of perception of any immediate harm linked with a relaxed practice [113,114]. These can be summarized as follows: diet, and the benefit and pleasure of food becomes more important to • have a written transition policy the individual. Psychological assessment is clearly important during • start early (~12 years of age) adolescence, and it must include an assessment of the cognitive level of • begin to introduce the concept of transition long before it becomes the individual as well as the psychological status of the individual and an issue family [103,104]. Education on the development of autonomy must be • introduce the notion that the young person could be seen by proposed as soon as the child is 16 years old and must be adapted to himself/herself in clinic visits with parents invited to join the each teenager and to each family [105]. In some centers, the parents are session later no longer allowed to accompany the child to the PKU clinic. It is • transfer medical information important that this transition be made without disrupting the parent– FOR ELECTRONIC• be flexible on theUSE timing of events ONLY child relationship. PKU holiday camps can be an excellent solution to • link the process to an appropriate education program improve autonomy and self-confidence for PKU teenagers. • provide opportunities for the young person to meet the adult team In adulthood, men and women with PKU respond differently to • identify a key worker to lead the process elevated phe levels or the stress associated with PKU. This may be the • tailor the process to the individual consequence of the pressure put on young PKU females by the medical • allow a clean break following successful transition. team in order to prevent the maternal PKU syndrome [100]. Many young adults with PKU abandon treatment for their disorder As regards PKU loss of follow up of adult males may not be critical and lose contact with their PKU clinic and their health care team. Social (in view of the lack of evidence of severe long term complications) but support, positive attitudes toward treatment, and manageability have the need to keep women of child bearing age is. Failure to support been shown to predict adherence to medical recommendations in adults female PKU patients both prior and during a pregnancy will likely with PKU. PKU individuals who return to a low-phe diet are able to result in significant and permanent disabilities in their offsprings. It is secure and maintain a job after a lengthy period of unemployment. They therefore important to get it right [115]. M. Demirkol et al. / Molecular Genetics and Metabolism 104 (2011) S31–S39 S37

8. Conclusions [13] J. Pietz, R. Dunckelmann, A. Rupp, D. Rating, H.M. Meinck, H. Schmidt, H.J. Bremer, Neurological outcome in adult patients with early-treated phenylketonuria, Eur. J. Pediatr. 157 (1998) 82–830. Systematic research and clinical observations have steadily [14] J. Weglage, Comments on behaviour in early treated phenylketonuria, Eur. J. increased our scientific knowledge as well as improved the outcome Pediatr. 159 (2000) S94–S95. [15] F.J. van Spronsen, P. Burgard, The truth of treating patients with phenylketonuria in PKU. It is widely appreciated that strict control of blood phe levels after childhood. The need for a new guideline, J. Inherit. Metab. Dis. 31 (2008) in PKU is an absolute requirement to enable individuals with PKU to 673–679. become functional, contributing members of society with normal [16] H. Levy, B. Burton, S. Cederbaum, C. Scriver, Recommendations for evaluation of intelligence quotient (IQ) and social development. responsiveness to tetrahydrobiopterin (BH(4)) in phenylketonuria and its use in treatment, Mol. Genet. Metab. 92 (2007) 287–291. Control of phe levels in PKU requires lifelong management using a [17] N. Blau, A. Bélanger-Quintana, M. Demirkol, F. Feillet, M. Giovannini, F.K. Trefz, diet that restricts intake of phe. The recent therapies could ease the F.J. van Spronsen, Optimizing the use of sapropterin (BH(4)) in the management – burden of the dietary phe restriction for PKU patients and their of phenylketonuria, Mol. Genet. Metab. 96 (2009) 158 163. [18] N. Blau, A. Bélanger-Quintana, M. Demirkol, F. Feillet, M. Giovannini, A. families but these options do not always allow a complete relaxation MacDonald, F.K. Trefz, F.J. van Spronsen, contributing European PKU centres, of the patient's diet. There is significant heterogeneity regarding Management of phenylketonuria in Europe: Survey results from 19 centres, Mol. management strategy among different countries and PKU centers Genet. Metab. 99 (2010) 109–115. [19] K. Ahring, A. Bélanger-Quintana, K. Dokoupil, H. Gokmen-Ozel, A.M. Lammardo, exists. This heterogeneity includes also substantial differences in the A. MacDonald, K. Motzfeld, M. Nowacka, M. Robert, M. van Rijn, Blood follow up rules with regard not only to phe concentrations but also to phenylalanine control in phenylketonuria: a survey of 10 European centres, nutritional, neuropsychological and psychosocial issues. There is a Eur. J. Clin. Nutr. 65 (2011) 275–278. [20] R. Koch, C. Azen, E.G. Friedman, K. Fishler, C. Baumann-Frischling, T. Lin, Care of growing body of evidence that suggests that neurocognitive, adult with phenylketonuria, Eur. J. Pediatr. 155 (1996) S90–S92. psychosocial, quality of life and growth, nutrition, bone pathology [21] I. Smith, M.G. Beasley, A.E. Ades, Intelligence and quality of dietary treatment in are suboptimal. phenylketonuria, Arch. Dis. Child. 65 (1990) 472–478. [22] W.M. Pardridge, Blood–brain barrier carrier-mediated transport and brain Equipping PKU individuals and caregivers with the necessary metabolism of amino acids, Neurochem. Res. 23 (1998) 635–644. practical skills, knowledge, and motivation to successfully produce an [23] S.C. Huijbregts, L.M. de Sonneville, R. Licht, F.J. van Spronsen, P.H. Verkerk, J.A. acceptable low-phe diet that is compatible with a modern life-style is Sergeant, Sustained attention and inhibition of cognitive interference in treated the key to achieving a good follow up and concordance at any age. For phenylketonuria: association with concurrent and lifetime phenylalanine concentrations, Neuropsychologia 40 (2002) 7–15. reasons of possible unknown late sequelae, all patients should be [24] A. MacDonald, G. Rylance, S.K. Hall, D. Asplin, I.W. Booth, Factors affecting the followed up life-long. variation in plasma phenylalanine in patients with phenylketonuria on diet, Arch. Dis. Child. 74 (1996) 412–417. [25] A. MacDonald, G.W. Rylance, D. Asplin, S.K. Hall, I.W. Booth, Does a single plasma Conflict of interest phenylalanine predict quality of control in phenylketonuria, Arch. Dis. Child. 78 (1998) 122–126. [26] B.K. Burton, L. Leviton, Reaching out to the lost generation of adults with early- Dr. Demirkol is a member of the Scientific Advisory Board of treated phenylketonuria (PKU), Mol. Genet. Metab. 101 (2010) 146–148. European Phenylketonuria Group, supported by Merck Serono S.A., [27] R. Gassió, J. Campistol, M.A. Vilaseca, N. Lambruschini, F.J. Cambra, E. Fuste, Do ż adult patients with phenylketonuria improve their quality of life after Geneva, Switzerland. Dr. Gi ewska has received honoraria for introduction/resumption of a phenylalanine-restricted diet? Acta Paediatr. 92 consulting or lecturing from Merck Serono S.A., Geneva, Switzerland (2003) 1474–1478. and Nutricia, Advanced Medical Nutrition, Amsterdam, The Nether- [28] J.T.R. Clarke, R.D. Gates, S.E. Hogan, M. Barrett, G.W. MacDonald, Neuropsycho- fi logical studies in adolescents with phenylketonuria returned to phenylalanine- land. Dr. Giovannini is a member of the Scienti c Advisory Board of restricted diets, Am. J. Ment. Retard. 92 (1997) 255–262. European Phenylketonuria Group, supported by Merck Serono S.A., [29] V.L. Brumm, C. Azen, R.A. Moats, A.M. Stern, C. Broomand, M.D. Nelson, R. Koch, Geneva, Switzerland. Dr. Walter declared no potential conflict of Neuropsychological outcome of subjects participating in the PKU adult collaborative study: a preliminary review, J. Inherit. Metab. Dis. 27 (2004) 549–566. interest. [30] S. Channon, E. German, C. Cassina, P. Lee, Executive functioning, memory, and learning in phenylketonuria, Neuropsychology 18 (2004) 613–620. [31] S. Channon, C. Mockler, P. Lee, Executive functioning and speed of processing in References phenylketonuria, Neuropsychology 19 (2005) 679–686. [32] S. Channon, G. Goodman, S. Zlotowitz, C. Mockler, P.J. Lee, Effects of dietary [1] J.H. Walter, F.J. White, S.K. Hall, A. MacDonald, G. Rylance, A. Boneh, D.E. Francis, management of phenylketonuria on long-term cognitive outcome, Arch. Dis. G.J. Shortland, M. Schmidt, A. Vail, How practical are recommendations for Child. 92 (2007) 213–218. dietary control in phenylketonuria, Lancet 360 (2002) 55–57. [33] R. Feldmann, J. Denecke, M. Grenzebach, J. Weglage, Frontal lobe-dependent functions [2] N. Blau, F.J. van Spronsen, H. Levy, Phenylketonuria, Lancet 376 (2010) in treated phenylketonuria: blood phenylalanine concentrations and long-term 1417–1427. deficits in adolescents and young adults, J. Inherit. Metab. Dis. 28 (2005) 445–455. [3] F. Feillet, C. Agostoni, Nutritional issues in treating phenylketonuria, J. Inherit. [34] V. Leuzzi, S. Rinalduzzi, F. Chiarotti, P. Garzia, G. Trasimeni, N. Accornero, Metab. Dis. 33 (2010) 659–664. Subclinical visual impairment in phenylketonuria. A neurophysiological study [4] J.J. Mitchell, C.R. Scriver, Phenylalanine hydroxylase deficiency, in: R.A. Pagon, (VEP-P) with clinical, biochemical, and neuroradiological (MRI) correlations, J. T.D. Bird, C.R. Dolan, K. Stephens (Eds.), GeneReviews [Internet], University of Inherit. Metab. Dis. 21 (1998) 351–364. Washington, Seattle, Seattle (WA), 1993, [updated 2010 May]. [35] J. Pietz, B. Fatkenheuer, P. Burgard, M. Armbruster, G. Esser, H. Schmidt, [5] E. Simon, M. Schwarz, J. Roos, N. Dragano, M. Geraedts, J. Siegrist, G. Kamp, U. Psychiatric disorders in adult patients with early-treated phenylketonuria, Wendel, Evaluation of quality of life and description of the sociodemographic Pediatrics 99 (1997) 345–350. state in adolescent and young adult patients with phenylketonuria, Health Qual. [36] M. Ris, A. Weber, M. Hunt, H. Berry, S. Williams, N. Leslie, Adult psychosocial Life Outcomes 6 (2008) 25. outcome in early-treated phenylketonuria, J. Inherit. Metab. Dis. 20 (1997) [6] K.M. Antshel, ADHD, learning, and academic performance in phenylketonuria, 499–508. Mol. Genet.FOR Metab. 99 (2010) S52 –ELECTRONICS58. [37] H. Schmidt, IntelligenceUSE and professional ONLY career in young adults treated early for [7] V.L. Brumm, D. Bilder, S.E. Waisbren, Psychiatric symptoms and disorders in phenylketonuria, Eur. J. Pediatr. 155 (1996) S97–S100. phenylketonuria, Mol. Genet. Metab. 99 (2010) S59–S63. [38] R. Cerone, M.C. Schiaffino, S. Di Stefano, E. Veneselli, Phenylketonuria: diet for life [8] National Institutes of Health Consensus Development Conference Statement: or not? Acta Paediatr. 88 (1999) 664–666. phenylketonuria: screening and management, October 16–18, 2000, Pediatrics [39] M. Cleary, J.H. Walter, Assessment of adult phenylketonuria, Ann. Clin. Biochem. 108 (2001) 972–982. 38 (2001) 450–458. [9] P. Burgard, H.J. Bremer, P. Bührdel, P.C. Clemens, E. Mönch, H. Przyrembel, F.K. [40] D.A. White, M.J. Nortz, T. Mandernach, K. Huntington, R.D. Steiner, Deficits in memory Trefz, K. Ullrich, Rationale for the German recommendations for phenylalanine strategy use related to prefrontal dysfunction during early development: evidence level control in phenylketonuria 1997, Eur. J. Pediatr. 158 (1999) 46–54. from children with phenylketonuria, Neuropsychology 15 (2001) 221–229. [10] Report of Medical Research Council Working Party on Phenylketonuria: [41] K. Anjema, M. van Rijn, P.H. Verkerk, J.G. Burgerhof, M.R. Heiner-Fokkema, F.J. recommendations on the dietary management of phenylketonuria, Arch. Dis. van Spronsen, PKU: high plasma phenylalanine concentrations are associated Child. 68 (1993) 426–427. with increased prevalence of mood swings, Mol. Genet. Metab. (Jun. 2 2011) [11] H. Ogier de Baulny, V. Abadie, F. Feillet, L. De Parscau, Managment of PKU and [Epub ahead of print]. HPA, J. Nutr. 137 (2007) 1561–1563. [42] B. Cabalska, N. Duczyńska, J. Borzymowska, K. Zorska, A. Kośłacz-Folga, K. [12] S. Schweitzer-Krantz, P. Burgard, Survey of national guidelines for the treatment Bożkowa, Termination of dietary treatment in phenylketonuria, Eur. J. Pediatr. of phenylketonuria, Eur. J. Pediatr. 159 (2000) S70–S73. 126 (1977) 253–262. S38 M. Demirkol et al. / Molecular Genetics and Metabolism 104 (2011) S31–S39

[43] M. Bik-Multanowski, B. Didycz, R. Mozrzymas, M. Nowacka, L. Kałużny, W. Cichy, [71] C. Agostoni, E. Verduci, N. Massetto, L. Fiori, G. Radaelli, E. Riva, M. Giovannini, B. Schneiberg, J. Amilkiewicz, A. Bilar, M. Giżewska, A. Lange, E. Starostecka, A. Long term effects of long-chain polyunsaturated fats in hyperphenylalaninemic Chrobot, B.I. Wójcicka-Bartłomiejczyk, A. Milanowski, Quality of life in non- children, Arch. Dis. Child. 88 (2003) 582–583. compliant adults with phenylketonuria after resumption of the diet, J. Inherit. [72] B. Koletzko, S. Beblo, H. Demmelmair, W. Müller-Felber, F.L. Hanebutt, Does Metab. Dis. 265 (2008) 630–633. dietary DHA improve neural function in children? Observations in phenylke- [44] R. Koch, M.D. Moseley, S. Yano, M. Nelson, R.A. Moats, Large neutral amino acid tonuria, Prostaglandins Leukot. Essent. Fatty Acids 81 (2009) 159–164. therapy and phenylketonuria: a promising approach to treatment, Mol. Genet. [73] B. Koletzko, S. Beblo, H. Demmelmair, F.L. Hanebutt, Omega-3 LC-PUFA supply Metab. 79 (2003) 110–113. and neurological outcomes in children with phenylketonuria (PKU), J. Pediatr. [45] R. Matalon, K. Michals-Matalon, G. Bhatia, A.B. Burlina, A.P. Burlina, C. Braga, L. Gastroenterol. Nutr. 48 (2009) S2–S7. Fiori, M. Giovannini, E. Grechanina, P. Novikov, J. Grady, S.K. Tyring, F. Guttler, [74] M.A. Vilaseca, N. Lambruschini, L. Gómez-López, A. Gutiérrez, J. Moreno, M. Double blind placebo control trial of large neutral amino acids in treatment of Tondo, R. Artuch, J. Campistol, Long-chain polyunsaturated fatty acid status in PKU: effect on blood phenylalanine, J. Inherit. Metab. Dis. 30 (2007) 153–158. phenylketonuric patients treated with tetrahydrobiopterin, Clin. Biochem. 43 [46] F.J. van Spronsen, M. van Rijn, T. van Dijk, G.P. Smit, D.J. Reijngoud, R. Berger, H.S. (2010) 411–415. Heymans, Plasma phenylalanine and tyrosine responses to different nutritional [75] L.W. van der Schot, W.H. Doesburg, R.C. Sengers, The phenylalanine response conditions (fasting/postprandial) in patients with phenylketonuria: effect of curve in relation to growth and mental development in the first years of life, Acta sample timing, Pediatrics 92 (1993) 570–573. Paediatr. 407 (1994) S68–S69. [47] J.H. Walter, F.J. White, Blood phenylalanine control in adolescents with [76] P.H. Verkerk, F.J. van Spronsen, G.P. Smit, R.C. Sengers, Impaired prenatal and phenylketonuria, Int. J. Adolesc. Med. Health 16 (2004) 41–45. postnatal growth in Dutch patients with phenylketonuria, The National PKU [48] M.D. Armstrong, U. Stave, A study of plasma free amino acid levels. II. Normal Steering Committee, Arch. Dis. Child. 71 (1994) 114–118. values for children and adults, Metabolism 22 (1973) 561–569. [77] F.J. van Spronsen, P.H. Verkerk, M. van Houten, Does impaired growth of PKU [49] D.V. Ashley, D.V. Barclay, F.A. Chauffard, D. Moennoz, P.D. Leathwood, Plasma patients correlate with the strictness of dietary treatment? National Dutch PKU amino acid responses in humans to evening meals of differing nutritional Steering Committee, Acta Paediatr. 86 (1997) 816–818. composition, Am. J. Clin. Nutr. 36 (1982) 143–153. [78] P.B. Acosta, S. Yannicelli, R. Singh, S. Mofidi, R. Steiner, E. DeVincentis, E. Jurecki, [50] H. Bickel, J. Gerard, E.M. Hickmans, The influence of phenylalanine intake on the L. Bernstein, S. Gleason, M. Chetty, B. Rourse, Nutrient intakes and physical chemistry and behaviour of a phenylketonuric child, Acta Paediatr. 43 (1954) 64–77. growth of children with phenylketonuria undergoing nutrition therapy, J. Am. [51] R. Sharman, K.A. Sullivan, R.M. Young, J.J. McGill, Tyrosine monitoring in children Diet. Assoc. 103 (2003) 1167–1173. with early and continuously treated phenylketonuria: results of an international [79] M.A. McBurnie, R.A. Kronmal, R. Koch, C.G. Azen, Physical growth of children practice survey, J. Inherit. Metab. Dis. (Sep. 30 2010). treated for phenylketonuria, Ann. Hum. Biol. 18 (1991) 357–368. [52] J. Posner, D. Gorman, B.J. Nagel, Tyrosine supplements for ADHD symptoms with [80] S. Scaglioni, E. Verduci, L. Fiori, A.M. Lammardo, S. Rossi, G. Radaelli, E. Riva, M. comorbid phenylketonuria, J. Neuropsychiatry Clin. Neurosci. 21 (2009) Giovannini, Body mass index rebound and overweight at 8 years of age in 228–230. hyperphenylalaninemic children, Acta Paediatr. 93 (2004) 1596–1600. [53] D. Webster, J. Wildgoose, Tyrosine supplementation for phenylketonuria, [81] F.J. van Spronsen, K. Ahring, M. Giżewska, PKU — what is daily practice in various Cochrane Database Syst. Rev. 8 (2010) CD001507. centres in europE? J. Inherit. Metab. Dis. 32 (2009) 58–64. [54] R. Sharman, K. Sullivan, R. Young, J. McGill, A preliminary investigation of the [82] F. Güttler, Hyperphenylalaninemia: diagnosis and classification of the various role of the phenylalanine:tyrosine ratio in children with early and continuously types of phenylalanine hydroxylase deficiency in childhood, Acta Paediatr. treated phenylketonuria: toward identification of “safe” levels, Dev. Neuropsy- Scand. 280 (1980) 1–80. chol. 35 (2010) 57–65. [83] A. Ponzone, A. Mussa, F. Porta, In response to F.J. van Spronsen et al., [55] U. Bick, G. Farhendorf, A.G. Ludolph, P. Vasallo, J. Weglage, K. Ullrich, Disturbed Phenylalanine tolerance can already reliably be assessed at the age of 2 years myelination in patients with treated hyperphenylalaninemias: evaluation with in patients with PKU, J. Inherit. Metab. Dis. 32 (2009) 584–586. magnetic resonance imaging, Eur. J. Pediatr. 150 (1991) 185–189. [84] P. Adamczyk, A. Morawiec-Knysak, P. Płudowski, B. Banaszak, J. Karpe, W. [56] R. Korinthenberg, K. Ullrich, F. Fullenkemper, Evoked potentials and electroen- Pluskiewicz, Bone mineral and the muscle-bone relationship in children, cephalography in adolescents with phenylketonuria, Neuropediatrics 19 (1988) adolescents and young adults with phenylketonuria, J. Bone Miner. Metab. 175–178. (August 13 2010), doi:10.1007/s00774-010-0216-x Published online. [57] K.H. Schulpis, C. Namopoulos, A. Scarpalezou, A. Bouloukos, S. Missiou- [85] L.G. Greeves, D.J. Carson, A. Magee, C.C. Patterson, Fractures and phenylketon- Tsagarakis, Serum carnitine level in phenylketonuria children under dietary uria, Acta Paediatr. 86 (1997) 242–244. control in Greece, Acta Paediatr. Scand. 79 (1990) 930–934. [86] J. Zeman, M. Bayer, J. Štépán, Bone mineral density in patients with [58] W.E.M. Lands, Renewed questions about polyunsaturated fatty acid, Nutr. Rev. phenylketonuria, Acta Paediatr. 88 (1999) 1348–1351. 44 (1986) 189–195. [87] B. Perez-Duenas, F.J. Cambra, M.A. Vilaseca, N. Lambruschini, J. Campistol, J.A. [59] C. Agostini, E. Verduci, L. Fiori, E. Riva, M. Giovannini, Breastfeeding rates among Camacho, New approach to osteopenia in phenylketonuric patients, Acta hyperphenylalaninemic infants, Acta Paediatr. 89 (2000) 366–367. Paediatr. 99 (2002) 899–904. [60] WHO, Infant and Young Child Feeding: Model Chapter for Textbooks for Medical [88] D. Modan-Moses, I. Vered, G. Schwartz, Y. Anikster, S. Abraham, R. Segev, O. Students and Allied health Professionals, World Health Organization, Geneva, Efrati, Peak bone mass in patients with phenylketonuria, J. Inherit. Metab. Dis. 30 2009. (2007) 202–208. [61] G. Hüner, T. Baykal, F. Demir, M. Demirkol, Breastfeeding experience in inborn [89] M.P. McMurry, G.M. Chan, C.O. Leonard, S.L. Ernst, Related bone mineral status in errors of metabolism other than phenylketonuria, J. Inherit. Metab. Dis. 28 children with phenylketonuria-relationship to nutritional intake and phenylal- (2005) 457–465. anine control, Am. J. Clin. Nutr. 55 (1992) 997–1004. [62] A. MacDonald, E. Depont, S. Evans, A. Daly, C. Hendriksz, A.A. Charapani, J.M. [90] P. Barat, N. Barthe, I. Redonnet-Vernhet, F. Parrot, The impact of the control of Saudubray, Breastfeeding in IMD, J. Inherit. Metab. Dis. 29 (2006) 299–303. serum phenylalanine levels on osteopenia in patients with phenylketonuria, Eur. [63] E. Riva, C. Agostoni, G. Biasucci, S. Trojan, D. Luotti, L. Fiori, M. Giovannini, Early J. Paediatr. 161 (2002) 687–688. breastfeeding is linked to higher intelligence quotient score in dietary treated [91] S. Yannicelli, D.M. Medeiros, Elevated plasma phenylalanine concentrations may phenylketonuric children, Acta Paediatr. 85 (1996) 56–58. adversely affect bone status of phenylketonuric mouse, J. Inherit. Metab. Dis. 25 [64] N.C.R. Raiha, Milk protein quantity and quality and protein requirements during (2002) 347–361. development, Adv. Pediatr. 36 (1989) 347–348. [92] B.A. Watkins, L. Yong, H.E. Lippman, S. Feng, Modulatory effect of omega-3 [65] L.A. Horrocks, Y.K. Yeo, Health benefits of docosahexaenoic acid (DHA), polyunsaturated fatty acids on osteoblast function and bone metabolism, Pharmacol. Res. 40 (1999) 211–225. Prostaglandins Leukot. Essent. Fatty Acids 68 (2003) 387–398. [66] C. Agostoni, A. Harvie, D.L. McCulloch, C. Demellweek, F. Cockburn, M. [93] S. Lage, M. Bueno, F. Andrate, J.A. Prieto, C. Delgado, M. Legarda, P. Sanjurjo, L.J. Giovannini, G. Murray, R.A. Harkness, E. Riva, A randomised trial of long-chain Aldámiz-Echevarria, Fatty acid profile in patients with phenylketonuria and its polyunsaturated fatty acid supplementation in infants with phenylketonuria, relationship with bone mineral density, J. Inherit. Metab. Dis. (Sept. 10 2010) Dev. Med. Child Neurol. 48 (2006) 207–212. [Epub ahead of print]. [67] H. Gokmen-Ozel, A. MacDonald, A. Daly, K. Hall, L. Ryder, A. Chakrapani, Long- [94] M.A. Cleary, J.H. Walter, J.E. Wraith, J.P. Jenkins, S.M. Alani, K. Tyler, D. Whittle, term efficacyFOR of ‘ready-to-drink’ protein ELECTRONIC substitute in phenylketonuria, J. Hum. Magnetic resonance USE imaging of the brain ONLY in phenylketonuria, Lancet 344 (1994) Nutr. Diet. 22 (2009) 422–427. 87–90. [68] I. Vugteveen, M. Hoeksma, A.L. Monsen, M.R. Fokkema, D.J. Reijngoud, M. van [95] P.J. Anderson PJ, V. Leuzzi, White matter pathology in phenylketonuria, Mol. Rijn, F.J. van Spronsen, Serum vitamin B12 concentrations within reference Genet. Metab. 99 (2010) S3–S9. values do not exclude functional vitamin B12 deficiency in PKU patients of [96] H.E. Moller, K. Ullrich, J. Weglage, In vivo proton magnetic resonance various ages, Mol. Genet. Metab. 102 (2011) 13–17. spectroscopy in phenylketonuria, Eur. J. Pediatr. 159 (2000) S121–S125. [69] B. Koletzko, E. Lien, C. Agostoni, H. Böhles, C. Campoy, I. Cetin, T. Desci, J.W. [97] J. Albrecht, S.F. Garbade, P. Burgard, Neuropsychological speed tests and blood Dudenhausen, C. Dupont, S. Forsyth, I. Hoesli, W. Holzgreve, A. Lapillonne, G. phenylalanine levels in patients with phenylketonuria: a meta-analysis, Putet, N.J. Secher, M. Symonds, H. Szajewska, P. Willatts, R. Uauy, World Neurosci. Biobehav. Rev. 33 (2009) 414–421. Association of Perinatal Medicine Dietary Guidelines Working Group, the roles of [98] C. Dawson, E. Murphy, C. Maritz, H. Chan, C. Ellerton, R.H. Carpenter, R.H. long-chain polyunsaturated fatty acids in pregnancy, lactation and infancy: Lachmann, Dietary treatment of phenylketonuria: the effect of phenylalanine on review of current knowledge and consensus recommendations, J. Perinat. Med. reaction time, J. Inherit. Metab. Dis. 34 (2011) 449–454. 36 (2008) 5–14. [99] J.K. Gentile, A.E. Ten Hoedt, A.M. Bosch, Psychosocial aspects of PKU: hidden [70] S.H. Yi, J.A. Kable, M.L. Evatt, R.H. Singh, A cross-sectional study of docosahex- disabilities—a review, Mol. Genet. Metab. 99 (2010) S64–S67. aenoic acid status and cognitive outcomes in females of reproductive age with [100] F. Feillet, A. MacDonald, D. Hartung (Perron), B. Burton, Outcomes beyond phenylketonuria, J. Inherit. Metab. Dis. 34 (2011) 455–463. phenylalanine: an international perspective, Mol. Genet. Metab. 99 (2010) S79–S85. M. Demirkol et al. / Molecular Genetics and Metabolism 104 (2011) S31–S39 S39

[101] C.S. Camfield, M. Joseph, T. Hurley, K. Campbell, S. Sanderson, P.R. Camfield, [108] A. MacDonald, S. Carver, A. Daly, V. Hopkins, C. Hendriksz, S. Vijay, A. Optimal management of phenylketonuria: a centralized expert team is Chakrapani, Service improvements with PKU clinic, J. Inherit. Metab. Dis. 31 more successful than a decentralized model of care, J. Pediatr. 145 (2004) (2008) 145. 53–57. [109] A. MacDonald, N. Manji, S. Evans, P. Davies, A. Daly, C. Hendriksz, A. Chakrapani, [102] G.M. Olsson, S.M. Montgomery, J. Alm, Family conditions and dietary control in Home delivery of dietary products in inherited metabolic disorders reduces phenylketonuria, J. Inherit. Metab. Dis. 30 (2007) 708–715. prescription and dispensing errors, J. Hum. Nutr. Diet. 19 (2006) 375–381. [103] J. Weglage, B. Funders, B. Wilken, D. Schubert, E. Schmidt, P. Burgard, K. Ullrich, [110] A. MacDonald, H. Gokmen-Ozel, M. van Rijn, P. Burgard, The reality of dietary Psychological and social findings in adolescents with phenylketonuria, Eur. J. compliance in the management of phenylketonuria, J. Inherit. Metab. Dis. 33 Pediatr. 151 (1992) 522–525. (2010) 665–670. [104] J.J. Moyle, A.M. Fox, M. Arthur, M. Bynevelt, J.R. Burnett, Meta-analysis of [111] M.D. Cabana, C.S. Rand, N.R. Powe, A.W. Wu, M.H. Wilson, P.A. Abboud, H.R. neuropsychological symptoms of adolescents and adults with PKU, Neuropsy- Rubin, Why don't physicians follow clinical practice guidelines? A framework for chol. Rev. 17 (2007) 91–101. improvement, JAMA 282 (1999) 1458–1465. [105] M. Szybowska, S. Hewson, B.J. Antle, R. Babul-Hirji, Assessing the informational [112] F. Feillet, F.J. van Spronsen, A. MacDonald, F.K. Trefz, M. Demirkol, M. Giovannini, needs of adolescents with a genetic condition: what do they want to know? J. A. Bélanger-Quintana, N. Blau, Challenges and pitfalls in the management of Genet. Couns. 16 (2007) 201–210. phenylketonuria, Pediatrics 126 (2010) 333–341. [106] J. Dale, I.O. Caramlau, A. Lindenmeyer, S.M. Williams, Peer support telephone [113] The Royal College of Physicians of Edinburgh, Think transition, Developing the calls for improving health, Cochrane Database Syst. Rev. (2008) CD006903. Essential link Between Paediatric and Adult care, 2008. [107] A.MacDonald, P.Davies,A.Daly, V.Hopkins,S.K. Hall, D. Asplin,C.Hendriksz, A. [114] Department of Health UK, Transition: Moving on Well, , 2008. Chakrapani, Does maternal knowledge and parent education affect blood [115] U. Mütze, A. Roth, J.F. Weigel, S. Beblo, C.G. Baerwald, P. Bührdel, W. Kiess, phenylalanine control in phenylketonuria? J. Hum. Nutr. Diet. 21 (2008) Transition of young adults with phenylketonuria from pediatric to adult care, J. 351–358. Inherit. Metab. Dis. 34 (2011) 701–709.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S40–S44

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Minireview The G46S-hPAH mutant protein: A model to study the rescue of aggregation-prone PKU mutations by chaperones

João Leandro a,b,⁎, Jaakko Saraste a, Paula Leandro b,c, Torgeir Flatmark a a Department of Biomedicine, University of Bergen, Jonas Lies vei 91, N-5009 Bergen, Norway b Metabolism and Genetics Group, iMed.UL-Research Institute for Medicines and Pharmaceutical Sciences, Faculty of Pharmacy, University of Lisbon, Av. Prof. Gama Pinto, 1649-003 Lisbon, Portugal c Department of Biochemistry and Human Biology, Faculty of Pharmacy, University of Lisbon, Av. Prof. Gama Pinto, 1649-003 Lisbon, Portugal article info abstract

Article history: Phenylketonuria (PKU), the most common inborn error of metabolism, is caused by dysfunction of the liver Received 26 July 2011 enzyme phenylalanine hydroxylase (PAH), with more than 550 PAH gene mutations identiﬁed to date. A large Accepted 26 July 2011 number of these mutations result in mutant forms of the enzyme displaying reduced stability, increased Available online 31 July 2011 propensity to aggregate, and accelerated in cellulo degradation. Loss or reduction of human PAH activity results in hyperphenylalaninemia (HPA) which, if untreated, results in severe mental retardation and Keywords: impaired cognitive development. Until now, strict low phenylalanine diet has been the most effective therapy, Phenylketonuria Human phenylalanine hydroxylase but as a protein misfolding disease PKU is a good candidate for treatment by natural/chemical/ Conformational disease, Aggregation-prone pharmacological chaperones. The natural cofactor of human PAH, (6R)-L-erythro-5,6,7,8-tetrahydrobiopterin Chemical/pharmacological chaperones (BH4), has already been approved for oral treatment of HPA, giving a positive response in mild forms of the disease showing considerable residual enzymatic activity. In the case of the most severe forms of PKU, ongoing studies with chemical and pharmacological chaperones to rescue misfolded mutant proteins from aggregation and degradation are providing promising results. The PKU mutation G46S is associated with a severe form of the disease, resulting in an aggregation-prone protein. The human PAH mutant G46S is rapidly degraded in the cellular environment and, in vitro (upon removal of its stabilizing fusion partner maltose binding protein (MBP)) self-associates to form higher-order oligomers/ﬁbrils. Here, we present an in vitro experimental model system to study the modulation of G46S aggregation by chemical/pharmacological chaperones, which may represent a useful approach to study the rescue of other severe PKU mutations by chemical/pharmacological chaperones. © 2011 Elsevier Inc. All rights reserved.

Contents

1. Phenylketonuria as a conformational disease ...... S40 2. Rescuing the function of mutant forms of human PAH by chemical and pharmacological chaperones ...... S41 3. The G46S PAH mutation: An aggregation-prone PKU mutation ...... S41 4. The G46S PAH as a model protein to study the ability of chaperones to rescue aggregation-prone PKU mutations ...... S42 5. Conclusions ...... S42 Conﬂict of interest ...... S43 Acknowledgments ...... S43 References ...... S43 FOR ELECTRONIC USE ONLY

1. Phenylketonuria as a conformational disease

The majority of human genetic diseases are caused by missense mutations resulting in single amino acid substitutions in the expressed ⁎ Corresponding author at: Metabolism and Genetics Group, iMed.UL, Faculty of proteins and their abnormal folding [1]. The misfolded proteins can Pharmacy, University of Lisbon, Av. Prof. Gama Pinto, 1649-003 Lisbon, Portugal. Fax: +351 217946491. be recognized by the cellular protein quality control machinery and E-mail address: [email protected] (J. Leandro). targeted for degradation, thus lowering the concentration of the protein

1096-7192/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2011.07.024 J. Leandro et al. / Molecular Genetics and Metabolism 104 (2011) S40–S44 S41 at its destination (loss-of-function). Alternatively, they can form of both wild-type and several human mutant PAH proteins [7,24,25]. aggregates which present a cytotoxic function (gain-of-function). This mechanism has been conﬁrmed using Pahenu1 mice harboring the Since in both cases the misfolded protein species trigger a pathological V106A mutation associated with a mild HPA phenotype [26].By enu2 condition, they are now recognized as new therapeutic targets. contrast, administration of BH4 to Pah mice with the F263S The autosomal recessive disorder phenylketonuria (PKU; OMIM mutation associated with a severe PKU phenotype did not restore L- 261600) is caused by mutations in the PAH gene encoding for Phe levels [26]. phenylalanine hydroxylase (PAH; EC 1.14.16.1). Mammalian PAH is Using a ﬂuorescence-based high-throughput thermal stability an iron dependent homotetrameric enzyme catalyzing the hydroxyl- assay Pey et al. [27] screened a library of N1000 organic compounds. ation of L-phenylalanine (L-Phe) into L-tyrosine (L-Tyr) in the presence Two of them, Compound III (3-amino-2-benzyl-7-nitro-4-(2-quino- of the cofactor (6R)-L-erythro-5,6,7,8-tetrahydrobiopterin (BH4) and lyl)-1,2-dihydroisoquinolin-1-one) and Compound IV (5,6-dimethyl- dioxygen. It assembles as a dimer of dimers and each monomer 3-(4-methyl-2-pyri-dinyl)-2-thioxo-2,3-dihydrothienol[2,3-d]pyri- comprises a N-terminal regulatory domain, a catalytic domain, and a midin-4(1H)-one), enhanced the thermal stability of the WT PAH

C-terminal region which includes the dimerization motif and the (ΔT0.5 of 13.6 and 7.2 °C, respectively) [27]. Further in cellulo studies tetramerization domain [2–4]. Mutations in human PAH lead to a wide with mild forms of PKU (I65T, R68S and R261Q mutations) spectrum of clinical and biochemical phenotypes, ranging from demonstrated that these compounds significantly increased the classical or severe PKU to mild PKU and mild hyperphenylalaninemia enzyme activity and steady-state protein levels in transiently (HPA) [5,6]. About 2/3 of the more than 550 mutations identified to transfected cells, as well as the activity of WT PAH in mouse liver date are missense mutations [6]. Analysis of the mutant proteins [27]. The effect of these compounds has also been evaluated on the expressed in vitro in different eukaryotic and prokaryotic cells or in structurally and functionally related tyrosine and tryptophan hydrox- cell free systems have demonstrated that only a small fraction of ylase, where Compound III demonstrated a large stabilizing effect on mutant proteins present changes in their affinity for the substrate or mutant R202H of tyrosine hydroxylase, a protein variant with reduced the cofactor [7,8]. When transiently expressed in eukaryotic cells, the stability and residual activity associated with L-DOPA-responsive majority shows reduced half-lives, and when overexpressed in dystonia [28]. Escherichia coli they demonstrate a high tendency to form soluble The natural osmolytes glycerol and trimethylamine N-oxide and insoluble aggregates [9–13]. The tetrameric form of recombinant (TMAO) have been shown to rescue the catalytic activity of the mild human wild-type (WT) PAH shows only marginal stability [14], forms of PKU associated with the mutations I65T, R261Q and V388M explaining the high frequency of misfolding human PAH mutations as well as the severe form R270K mutation [29,30]. Expression of the associated with PKU/HPA. In addition, co-expression studies of several mutant proteins in E. coli in the presence of 1% glycerol resulted in PKU mutant proteins in E. coli with bacterial chaperonins (GroES/ decreased formation of aggregated forms and increased recovery of GroEL) led to a considerable increase in the amount of functional tetramers (R261Q, R270K and V388M) or dimers (I65T) [30]. protein and residual activity unveiling also the deleterious mutational Moreover, when synthesized in the presence of glycerol both the effects on stability and folding properties of human PAH [10,13]. In V388M and I65T dimers exhibited more structured forms resembling vitro studies with molecular chaperones (Hsp70/40 and Hsp90) also a native-like state as measured by far-UV circular dichroism. These points to a likely role of the chaperone machinery in vivo [15]. In silico results are compatible with rescued folding of intracellular human predictions of the energetic impact of mutations on the native-state PAH, resulting from stabilization of folding intermediates or native- stability of human PAH have also provided evidence in support of like states by glycerol [30,31]. decreased protein stability as the molecular basis of PKU [16], Although considerable progress has been made in the search for connecting it with the growing number of genetic diseases classified PAH chaperones and the understanding of the underlying stabilizing as conformational disorders with a loss-of-function. mechanisms, the advances in the case of the most severe forms of the disease have been modest. In general, the evaluation of in vitro 2. Rescuing the function of mutant forms of human PAH by stabilizing effects of chaperones has been hampered by the high levels chemical and pharmacological chaperones of aggregated forms.

In conformational disorders the misfolded protein represents the 3. The G46S PAH mutation: An aggregation-prone PKU mutation therapeutic target. Thus, small molecules able to affect the protein folding reaction or to stabilize folding intermediates and native-like The G46S PAH mutation belongs to the group of PKU mutations states, are potential candidates as therapeutic compounds. Such which result in a propensity of the protein to self-associate and form molecules include enzyme cofactors and inhibitors, as well as soluble/insoluble aggregates when over-expressed in prokaryotic pharmacological and chemical chaperones. Phenylketonuria, which systems, and its rapid degradation when expressed in eukaryotic cells is a paradigm of inherited metabolic diseases, provides an example in [32]. It is a relatively common mutation in Scandinavia (i.e. ~10% this new area of therapeutic intervention. relative frequency for PKU alleles in Southeast Norway [32,33]) and is In 1999 Kure and co-workers [17] ﬁrst described the natural associated with a severe form of the disease [32]. cofactor BH4 as a potential therapeutic agent, and after many years of The G46S mutation is localized in the regulatory domain of PAH research a stableFOR form of the humanELECTRONIC PAH natural cofactor is now that contains anUSE ACT module (domain ONLY which name derives from available for the treatment of PKU [18]. Mild forms, with a con- aspartokinase, chorismate mutase, and TyrA (prephenate dehydro- siderable residual enzyme activity, correlate with the best response to genase)) [34]. The mutation is located at the start of α-helix 1 (A47–

BH4 intake, whereas patients with a severe form of PKU, presenting E57), in a ﬁve residue (L41–G46) loop stabilized by a network of two inactive alleles, generally respond poorly [19,20]. When mild PKU hydrogen bonds and a salt-bridge (K42–E44). When the mutant mutations are expressed in a coupled in vitro transcription-translation protein was transiently expressed in human embryonic kidney system in the presence of BH4, the proteins display an increased (HEK293) cells, the level of immunoreactive protein was only ~3% catalytic activity and stability [8]. Thus, the underlying molecular of the WT protein, at identical PAH mRNA levels [32]. However, its mechanism of BH4-responsiveness is intrinsically related to the expression as a fusion protein in E. coli, coupled to the maltose binding inhibitory regulation of human PAH by its cofactor. Upon binding protein (MBP-pep(Xa)-G46S-PAH), gave rise to a metastable tetra- BH4, in the absence of the substrate, the PAH protein assumes a more meric form characterized by a near normal catalytic efﬁciency [15,32]. closed conformation, which protects it from proteolytic degradation Cleavage of the fusion partner by the restriction protease factor Xa

[21–23]. It is now well accepted that BH4 acts as an in cellulo stabilizer lead to destabilization and self-association of the G46S tetramer (and S42 J. Leandro et al. / Molecular Genetics and Metabolism 104 (2011) S40–S44 dimer) [15], and the formation of higher-order oligomers and large twisted non-amyloid ﬁbrils (Fig. 1). The self-association process has been well characterized in terms of the intrinsic physico-chemical properties of the mutant protein, as well as extrinsic factors (e.g. ionic strength, temperature, pH, protein concentration, phosphorylation state and deamidation), thus offering a well controlled in vitro system to evaluate the effect of chaperones [15].

4. The G46S PAH as a model protein to study the ability of chaperones to rescue aggregation-prone PKU mutations

The mutation G46S is considered a non-responsive BH4 mutation [35]. This conclusion was confirmed by our in vitro light scattering analyses showing that the cofactor did not inhibit the self-association of the mutant protein upon factor Xa-mediated cleavage of the metastable MBP fusion protein [15]. Moreover, the chemical chaperones glycerol, TMAO and (−)-epigallocatechin gallate (EGCG), the major polyphenol in green tea, gave variable effects (Fig. 2). Glycerol and TMAO have been shown to rescue the enzyme activity of mild and severe (R270K) PKU mutations, when added to the culture medium during protein expression in E. coli [30]. In case of the R270K mutation, a 6.5-fold increase in enzymatic activity with 5% glycerol, and a 4.6-fold increase with 5 mM TMAO were observed. Concomi- tantly, the presence of the natural osmolytes reduced the amount of aggregates and increased the recovery of enzyme tetramer. For example, 5% glycerol decreased the amount of R270K mutant aggregates (from 65 to 4%) resulting in corresponding increase in the tetrameric fraction (from 30 to 95%) [30]. By contrast, in the study of the rescue of the G46S protein, only glycerol showed a concentration-dependent inhibition of self-association, whereas TMAO promoted the process (Fig. 2). Higher concentrations of TMAO (N10 mM) have been found to have a negative effect on the enzyme activity of some PKU mutations [30]. Additionally, the green tea polyphenol EGCG gave a similar effect as the osmolyte TMAO, and did not inhibit fibril formation (Fig. 2), in contrast to what has been reported for certain amyloidogenic proteins [36,37]. The proposed pharmacological chaperone of aromatic amino acid hydroxylases, Compound III [27,28], partly inhibits the self-association process of the G46S mutant and disrupts the formation of fibrils (Fig. 3). Thus, this compound (or a derivative thereof) seems to be a promising candidate to rescue severe and aggregation-prone PKU mutations.

Fig. 2. The effect of chemical chaperones on the self-association of G46S PAH was followed by the change in light scattering following cleavage of the metastable,

tetrameric MBP-pep(Xa)-G46S-PAH fusion protein by the restriction protease factor Xa. (A) Effect of glycerol on the self-association of G46S PAH: control (●), 1% glycerol (■), 2.5% glycerol (▲) and 5% glycerol (◊). (B) Effect of trimethylamine N-oxide (TMAO) on the self-association of G46S PAH: control (●), 5 mM TMAO (■), 150 mM TMAO (▲) and FOR ELECTRONIC250 mM TMAO (◊). (C)USE Effect of (−)-epigallocatechin ONLY gallate (EGCG) on the self- association of G46S PAH (expressed as molar ratios of G46S PAH fusion protein:EGCG): 1:0 control (●), 1:0.1 (■) and 1:1 (▲). The chemical structure of the compounds is represented inside each panel, respectively. Data taken from [15].

5. Conclusions

Although promising, a therapy of PKU based on chaperoning of misfolded mutant proteins may be characterized by its mutant Fig. 1. Electron micrograph showing negatively stained ﬁbrils of tetrameric G46S PAH. ﬁ The protein was negatively stained with aqueous uranyl acetate and analyzed in JEOL speci city and compound-dependency. Consequently, a panoply of 1230 electron microscope operated at 80 kV. Scale bar: 500 nm. experimental approaches including prokaryotic and eukaryotic J. Leandro et al. / Molecular Genetics and Metabolism 104 (2011) S40–S44 S43

Fig. 3. The effect of pharmacological chaperones on the self-association of G46S PAH was followed by the change in light scattering after cleavage of the metastable, tetrameric MBP- pep(Xa)-G46S-PAH fusion protein by the restriction protease factor Xa in the absence of any added compound (●), in the presence of 0.83% dimethyl sulfoxide (DMSO) (▲) or 100 μM 3-amino-2-benzyl-7-nitro-4-(2-quinolyl)-1,2-dihydroisoquinolin-1-one (in 0.83% DMSO) (■). Error bars represent mean±SD, (n=3). The electron micrographs on the right correspond to negatively stained G46S PAH in the absence of any added compound after 3 h of cleavage of its MBP fusion partner by factor Xa (upper panel), and in the presence of the pharmacological chaperone Compound III, 3-amino-2-benzyl-7-nitro-4-(2-quinolyl)-1,2-dihydroisoquinolin-1-one in 0.83% DMSO at t=3 h of cleavage (lower panel). The proteins were negatively stained with aqueous uranyl acetate and examined in JEOL 1230 electron microscope operated at 80 kV. Scale bar: 500 nm. Data taken and ﬁgure adapted from [15].

expression systems, mouse models and in vitro assays will be [10] A. Gámez, B. Pérez, M. Ugarte, L.R. Desviat, Expression analysis of phenylketonuria mutations. Effect on folding and stability of the phenylalanine hydroxylase necessary to screen and identify the most promising chaperones and protein, J. Biol. Chem. 275 (2000) 29737–29742. elucidate their underlying mechanisms. In this context, our studies on [11] T. Gjetting, M. Petersen, P. Guldberg, F. Güttler, In vitro expression of 34 naturally the aggregation-prone G46S mutant could be a useful model in occurring mutant variants of phenylalanine hydroxylase: correlation with metabolic phenotypes and susceptibility toward protein aggregation, Mol. systematic in vitro studies of misfolded, aggregation-prone PKU/HPA- Genet. Metab. 72 (2001) 132–143. associated mutant proteins. [12] S.W. Gersting, K.F. Kemter, M. Staudigl, D.D. Messing, M.K. Danecka, F.B. Lagler, C.P. Sommerhoff, A.A. Roscher, A.C. Muntau, Loss of function in phenylketonuria is caused by impaired molecular motions and conformational instability, Am. J. Conflict of interest Hum. Genet. 83 (2008) 5–17. [13] A.L. Pey, L.R. Desviat, A. Gámez, M. Ugarte, B. Pérez, Phenylketonuria: genotype- fl phenotype correlations based on expression analysis of structural and functional The authors declare that there are no con icts of interest. mutations in PAH, Hum. Mutat. 21 (2003) 370–378. [14] R. Kleppe, K. Uhlemann, P.M. Knappskog, J. Haavik, Urea-induced denaturation of human phenylalanine hydroxylase, J. Biol. Chem. 274 (1999) 33251–33258. Acknowledgments [15] J. Leandro, N. Simonsen, J. Saraste, P. Leandro, T. Flatmark, Phenylketonuria as a protein misfolding disease: the mutation pG46S in phenylalanine hydroxylase This work was supported by Fundação para a Ciência e a promotes self-association and fibril formation, Biochim. Biophys. Acta 1812 (2011) 106–120. Tecnologia, Portugal, grant SFRH/BD/19024/2004 and the University [16] A.L. Pey, F. Stricher, L. Serrano, A. Martínez, Predicted effects of missense mutations of Bergen, Norway. on native-state stability account for phenotypic outcome in phenylketonuria, a paradigm of misfolding diseases, Am. J. Hum. Genet. 81 (2007) 1006–1024. [17] S. Kure, D.C. Hou, T. Ohura, H. Iwamoto, S. Suzuki, N. Sugiyama, O. Sakamoto, K. References Fujii, Y. Matsubara, K. Narisawa, Tetrahydrobiopterin-responsive phenylalanine hydroxylase deficiency, J. Pediatr. 135 (1999) 375–378. [1] N. Gregersen, P. Bross, S. Vang, J.H. Christensen, Protein misfolding and human [18] H.L. Levy, A. Milanowski, A. Chakrapani, M. Cleary, P. Lee, F.K. Trefz, C.B. Whitley, F. disease, Annu. Rev. Genomics Hum. Genet. 7 (2006) 103–124. Feillet, A.S. Feigenbaum, J.D. Bebchuk, H. Christ-Schmidt, A. Dorenbaum, Efficacy

[2] T. Flatmark, R.C. Stevens, Structural insight into the aromatic amino acid of sapropterin dihydrochloride (tetrahydrobiopterin, 6R-BH4) for reduction of hydroxylases and their disease-related mutant forms, Chem. Rev. 99 (1999) phenylalanine concentration in patients with phenylketonuria: a phase III 2137–2160. randomised placebo-controlled study, Lancet 370 (2007) 504–510. [3] F. Fusetti, H. Erlandsen, T. Flatmark, R.C. Stevens, Structure of tetrameric human [19] C. Bernegger, N. Blau, High frequency of tetrahydrobiopterin-responsiveness phenylalanine hydroxylase and its implications for phenylketonuria, J. Biol. Chem. among hyperphenylalaninemias: a study of 1,919 patients observed from 1988 to 273 (1998) 16962–16967. 2002, Mol. Genet. Metab. 77 (2002) 304–313. [4] B. Kobe, I.G. Jennings, C.M. House, B.J. Michell, K.E. Goodwill, B.D. Santarsiero, R.C. [20] A.C. Muntau, W. Röschinger, M. Habich, H. Demmelmair, B. Hoffmann, C.P. Stevens, R.G. Cotton, B.E. Kemp, Structural basis of autoregulation of phenylal- Sommerhoff, A.A. Roscher, Tetrahydrobiopterin as an alternative treatment for anine hydroxylase, Nat. Struct. Biol. 6 (1999) 442–448. mild phenylketonuria, N. Engl. J. Med. 347 (2002) 2122–2132. [5] N. Blau, F.J. van Spronsen, H.L. Levy, Phenylketonuria, Lancet 376 (2010) 1417–1427. [21] L.J. Mitnaul, R. Shiman, Coordinate regulation of tetrahydrobiopterin turnover and [6] C.R. Scriver,FOR M. Hurtubise, D. Konecki, ELECTRONIC M. Phommarinh, L. Prevost, H. Erlandsen, R. phenylalanine USE hydroxylase activity in ratONLY liver cells, Proc. Natl. Acad. Sci. U.S.A. 92 Stevens, P.J. Waters, S. Ryan, D. McDonald, C. Sarkissian, PAHdb 2003: what a (1995) 885–889. locus-speciﬁc knowledgebase can do, Hum. Mutat. 21 (2003) 333–344. [22] A.L. Pey, A. Martinez, Tetrahydrobiopterin for patients with phenylketonuria, [7] H. Erlandsen, A.L. Pey, A. Gámez, B. Pérez, L.R. Desviat, C. Aguado, R. Koch, S. Lancet 370 (2007) 462–463. Surendran, S. Tyring, R. Matalon, C.R. Scriver, M. Ugarte, A. Martínez, R.C. Stevens, [23] T. Solstad, A.J. Stokka, O.A. Andersen, T. Flatmark, Studies on the regulatory Correction of kinetic and stability defects by tetrahydrobiopterin in phenylke- properties of the pterin cofactor and dopamine bound at the active site of human tonuria patients with certain phenylalanine hydroxylase mutations, Proc. Natl. phenylalanine hydroxylase, Eur. J. Biochem. 270 (2003) 981–990. Acad. Sci. U.S.A. 101 (2004) 16903–16908. [24] C. Aguado, B. Pérez, M. Ugarte, L.R. Desviat, Analysis of the effect of [8] A.L. Pey, B. Pérez, L.R. Desviat, M.A. Martínez, C. Aguado, H. Erlandsen, A. Gámez, tetrahydrobiopterin on PAH gene expression in hepatoma cells, FEBS Lett. 580 R.C. Stevens, M. Thórólfsson, M. Ugarte, A. Martínez, Mechanisms underlying (2006) 1697–1701. responsiveness to tetrahydrobiopterin in mild phenylketonuria mutations, [25] B. Pérez, L.R. Desviat, P. Gómez-Puertas, A. Martínez, R.C. Stevens, M. Ugarte,

Hum. Mutat. 24 (2004) 388–399. Kinetic and stability analysis of PKU mutations identiﬁed in BH4-responsive [9] E. Bjørgo, P.M. Knappskog, A. Martínez, R.C. Stevens, T. Flatmark, Partial patients, Mol. Genet. Metab. 86 (Suppl. 1) (2005) S11–S16. characterization and three-dimensional-structural localization of eight mutations [26] S.W. Gersting, F.B. Lagler, A. Eichinger, K.F. Kemter, M.K. Danecka, D.D. Messing, M. in exon 7 of the human phenylalanine hydroxylase gene associated with phenyl- Staudigl,K.A.Domdey,C.Zsifkovits,R.Fingerhut,H.Glossmann,A.A.Roscher, ketonuria, Eur. J. Biochem. 257 (1998) 1–10. A.C. Muntau, Pahenu1 is a mouse model for tetrahydrobiopterin-responsive S44 J. Leandro et al. / Molecular Genetics and Metabolism 104 (2011) S40–S44

phenylalanine hydroxylase deficiency and promotes analysis of the pharmaco- [32] H.G. Eiken, P.M. Knappskog, J. Apold, T. Flatmark, PKU mutation G46S is associated logical chaperone mechanism in vivo, Hum. Mol. Genet. 19 (2010) 2039–2049. with increased aggregation and degradation of the phenylalanine hydroxylase [27] A.L. Pey, M. Ying, N. Cremades, A. Velazquez-Campoy, T. Scherer, B. Thöny, J. Sancho, enzyme, Hum. Mutat. 7 (1996) 228–238. A. Martínez, Identification of pharmacological chaperones as potential therapeutic [33] J. Zschocke, J.P. Mallory, H.G. Eiken, N.C. Nevin, Phenylketonuria and the peoples agents to treat phenylketonuria, J. Clin. Invest. 118 (2008) 2858–2867. of Northern Ireland, Hum. Genet. 100 (1997) 189–194. [28] A.C. Calvo, T. Scherer, A.L. Pey, M. Ying, I. Winge, J. McKinney, J. Haavik, B. Thöny, A. [34] J. Siltberg-Liberles, A. Martínez, Searching distant homologs of the regulatory ACT Martínez, Effect of pharmacological chaperones on brain tyrosine hydroxylase domain in phenylalanine hydroxylase, Amino Acids 36 (2009) 235–249. and tryptophan hydroxylase 2, J. Neurochem. 114 (2010) 853–863. [35] M.D. Boveda, M.L. Couce, D.E. Castineiras, J.A. Cocho, B. Perez, M. Ugarte, J.M. Fraga, [29] P. Leandro, M.C. Lechner, I. Tavares de Almeida, D. Konecki, Glycerol increases the The tetrahydrobiopterin loading test in 36 patients with hyperphenylalaninae- yield and activity of human phenylalanine hydroxylase mutant enzymes mia: evaluation of response and subsequent treatment, J. Inherit. Metab. Dis. 30 produced in a prokaryotic expression system, Mol. Genet. Metab. 73 (2001) (2007) 812. 173–178. [36] J. Bieschke, J. Russ, R.P. Friedrich, D.E. Ehrnhoefer, H. Wobst, K. Neugebauer, E.E. [30] C. Nascimento, J. Leandro, I. Tavares de Almeida, P. Leandro, Modulation of the Wanker, EGCG remodels mature α-synuclein and amyloid-β fibrils and reduces activity of newly synthesized human phenylalanine hydroxylase mutant proteins cellular toxicity, Proc. Natl. Acad. Sci. U.S.A. 107 (2010) 7710–7715. by low-molecular-weight compounds, Protein J. 27 (2008) 392–400. [37] D.E. Ehrnhoefer, J. Bieschke, A. Boeddrich, M. Herbst, L. Masino, R. Lurz, S. Engemann, [31] P. Leandro, C.M. Gomes, Protein misfolding in conformational disorders: rescue of A. Pastore, E.E. Wanker, EGCG redirects amyloidogenic polypeptides into unstruc- folding defects and chemical chaperoning, Mini Rev. Med. Chem. 8 (2008) 901–911. tured, off-pathway oligomers, Nat. Struct. Mol. Biol. 15 (2008) 558–566.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S45–S51

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Minireview Cognitive, neurophysiological, neurological and psychosocial outcomes in early-treated PKU-patients: A start toward standardized outcome measurement across development

F.J. van Spronsen a,⁎, S.C.J. Huijbregts b,c, A.M. Bosch d, V. Leuzzi e,f a Section of Metabolic Diseases, Beatrix Children's Hospital, University Medical Center of Groningen, University of Groningen, Groningen, The Netherlands b Clinical Child and Adolescent Studies, Leiden University, The Netherlands c Leiden Institute for Brain and Cognition, Leiden University, The Netherlands d Department of Pediatrics, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands e Dept of Paediatrics, Sapienza Università di Roma, Italy f Dept of Child Neurology and Psychiatry, Sapienza Università di Roma, Italy article info abstract

Article history: The aim of this paper is to provide a concise summary of findings from outcome studies in early-treated Received 31 July 2011 phenylketonuria (PKU). The paper should not be considered as an extensive review of the many different Received in revised form 28 September 2011 outcome measures that have been used in PKU-research, but as an attempt to integrate such findings so Accepted 29 September 2011 that they will be of additional value for day to day monitoring of PKU-patients and may direct future research Available online 6 October 2011 to fill the present gaps of knowledge. Neurological, neuropsychological, neurophysiological, neuroimaging, quality fi Keywords: of life, and psychosocial ndings will be discussed in the context of their potential contributions to lifelong Phenylalanine follow-up and treatment of PKU-patients being summarized in statements. Neurophysiological © 2011 Elsevier Inc. All rights reserved. Neurological Neuropsychological Quality of life Psychosocial

Contents

1. Introduction ...... S45 2. Neurocognitive outcome ...... S46 2.1. Characterizing the EF-deﬁcit in treated PKU ...... S46 2.2. Assessment of EF in PKU ...... S47 3. Neurological outcome ...... S47 4. Behavior and quality of life ...... S47 5. Neurophysiological and neuroimaging alterations and outcome ...... S48 6. Neurological outcome ...... S49 7. Statements ...... S49 8. Conclusion...... S49 References ...... S49 FOR ELECTRONIC USE ONLY

1. Introduction Abbreviations: ANT, Amsterdam Neuropsychological Tasks; BAEP, brainstem auditory evoked potentials; BRIEF, Behavior Rating Inventory of Executive Function; CANTAB, Cambridge Neuropsychological Test Automated Battery; EEG, electroencephalography; For Phenylketonuria (PKU; MIM# 261600), past decades have EF, executive functioning; EP, Evoked Potentials; ERP, Event-Related Potentials; HRQoL, shown important progress in treatment strategies and outcome [1]. Health related quality of life; MRI, magnetic resonance imaging; MRS, magnetic resonance This journal has recently published a comprehensive review about spectroscopy; PKU, phenylketonuria; PET, positron emission tomography; VEP, Visual psychological and neuropathological ﬁndings in PKU-patients [2]. Evoked Potentials; WM, white matter. ⁎ Corresponding author. Fax: +31 50 3614235. The focus of the papers in that special issue of Molecular Genetics E-mail address: [email protected] (F.J. van Spronsen). and Metabolism was a description of the many different outcomes

1096-7192/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2011.09.036 S46 F.J. van Spronsen et al. / Molecular Genetics and Metabolism 104 (2011) S45–S51 measured in PKU. Studies generally focused on only one or two spe- continuous monitoring of cognitive outcome. However, matters are cific outcome measurements. The focus of the present paper is a somewhat more complicated, mainly because it is difficult to sepa- first step toward integration of the different findings of studies rate processing speed and EF in most neuropsychological tasks: a using different outcome measures. The ultimate goal is to develop a processing speed deficit becomes more evident when a task be- standardized set of instruments that can be used in day to day clinical comes more complex (i.e. when more executive control is required) practice to monitor cognitive and social functioning throughout because of the requirement of more brain regions communicating development of children into adulthood in relation to treatment with each other (and of more communication between neurons parameters. We realize that, with the present state of knowledge, within brain regions). If Phe has affected white matter integrity which is derived from studies using very different samples, designs [15], it is unlikely this has happened in one specificareaofthe and methodologies, this is a very ambitious goal. Therefore, a number brain, and therefore, there will be additional slowing of task perfor- of the statements at the end of this paper concern theoretical reflec- mance with each new brain region involved. Involvement of many tions, or are meant to draw attention to gaps in the current knowledge, different brain regions is typical of executive function (EF-) tasks. rather than concrete recommendations for long-term monitoring of For example, performance of a task requiring inhibitory control gen- PKU-patients. In this paper we will focus on neurocognitive outcome, erally involves dorsolateral prefrontal cortex, the occipital cortex with an emphasis on studies into executive control, and on psychiatric (for perception of the stimuli), and a number of motor areas (e.g. and behavioral outcomes, emphasizing quality of life. Furthermore, primary and premotor cortex); a working memory task also involves attention will be paid to studies that have used more direct measures these areas, as well as parietal regions [16–18]. EF-tasks which also of brain functioning in PKU, including clinical neurological issues and incorporate a motivational or emotional component additionally in- measures those that have used electroencephalography (EEG), magnetic volve temporal, orbitofrontal and subcortical regions [19].Ifawhite- resonance imaging (MRI), magnetic resonance spectroscopy (MRS), and matter related processing speed deficit underlies cognitive dysfunction positron emission tomography (PET). All these issues will be discussed in in treated PKU, one would thus expect much greater processing speed relation to (dietary) treatment parameters. deficits when EF are required than when they are not. Consequently, it will be insufficient for monitoring of PKU-patients to have them perform 2. Neurocognitive outcome simple processing speed tasks, as many problems will only become apparent when there is a demand for (integration of) multiple aspects of Studies on IQ are relatively simple, relatively uniform, and ref- cognition. erence data are widely available. It is already clear from many Historically, the prefrontal cortex has been put at the center of all studies that the earlier the start of the diet, the better the outcome. EF [20], and although this notion has become slightly dated with the Studies compared IQ data in patients who started treatment before increasing knowledge of brain networks, it is still a very important 3 weeks with patients who started treatment between 3 and 6 weeks structure for EF, a structure that is always involved in executive oper- up to 3 months, and found higher IQ-scores for those who had started ations, a structure containing dopaminergic neurons that appear to be earlier, indicating the importance of taking into account the time more sensitive to dopamine precursor reductions than those else- of start of treatment when studying neurocognitive outcomes in where in the brain [21,22], and the brain structure that matures the PKU [3,4]. Both the duration of high Phe concentrations before latest of all brain regions [23–25], thereby prolonging the period treatment and the absolute Phe concentrations during that time throughout which it is likely to be sensitive to higher Phe levels and may be important. lower monoamine levels. Thus, evidence appears to favor regular From data on IQ, we cannot decide on the exact Phe concentration EF-assessments in order to monitor cognitive development and func- that is safe not to treat [5], or the age at which treatment can be safely tioning. As a dependent measure, one might opt to use reaction stopped. There are data supporting the idea that IQ does not deteriorate times/processing speed, but this would be sufficiently informative when treatment is stopped after 8, 10, and 12 years of age and Phe is only in EF-tasks with multiple levels of complexity. within target range until that age [6–8], while the study of Koch et al. indicates that treatment cannot be stopped too early to prevent clinical 2.1. Characterizing the EF-deficit in treated PKU problems in adults [9]. However, results of the Koch et al. study are not entirely comparable with the results of the other studies showing EF has generally been used as an umbrella term encompassing a no deterioration, as this study compared patients who had stopped di- wide range of (combinations of) abilities such as planning, organiza- etary treatment at 6 years with patients that had not stopped at all tion, strategy use, cognitive flexibility, inhibitory control, and working [9]. Thus, based on these data, the question as to whether IQ is a suitable memory [14]. A number of efforts have been made to model or define measurement to determine whether treatment should be continued the EF-deficit that is typical of PKU-patients [13,14]. Such approaches until or beyond 10–12 years of age cannot be answered [10]. A recent have been very useful in that they narrowed down findings from a myr- meta-analysis by Waisbren et al. showed that until 18 years there is a iad of EF-tasks to a number of core concepts, some of them impaired in negative effect of high blood Phe concentrations on IQ [11], indicating treated PKU and others intact. In general, inhibitory control of prepotent that treatment should at least be continued until the end of adolescence. responding and the manipulation- and monitoring aspect of working It is disputable, however, whether IQ is the most informative mea- memory appear to be the most strongly affected executive functions in sure to assess neurocognitiveFOR outcome ELECTRONIC in PKU, as an IQ-score is made treated PKU, particularly USE when they areONLY required simultaneously [26– up of scores on several different tasks, measuring different cognitive 29]. It should be noted, however, that the characterization of EF-deficits domains, and good or above-average scores on some tasks can mask in PKU within theoretical frameworks has particularly been based on poor or below-average scores on other tasks. Although it could simply studies with children who were on diet at the time of neuropsychological be argued to focus on subtest-performance then, evidence indicates testing. Studies with adolescents and adults show mixed results. Some ap- that specific impairments are present in treated PKU that go beyond pear to indicate that PKU-control differences are smaller for older on-diet what can be picked up by IQ-tests. For PKU, especially processing groups [30], but others indicate more serious EF-problems for the older speed and cognitive control/executive functioning (EF) appear to be groups [31]. The main complicating factor in finding out whether deficits affected [12–14]. EF refers to regulatory higher-order cognitive abili- increase or decrease with age, and whether the abovementioned theoret- ties allowing integration and processing of information across a range ical frameworks also apply to EF in older PKU-populations, is that these of cognitive domains, sensory modalities, and response modalities. patients often have discontinued or at least relaxed dietary treatment One test for processing speed and one for EF, which are repeatable regimens. Therefore, when the goal is the assemble a compendium across different ages, may thus appear a possible solution in light of of instruments to monitor PKU patients across different ages, one F.J. van Spronsen et al. / Molecular Genetics and Metabolism 104 (2011) S45–S51 S47 of the first steps would have to be neuropsychological testing of children with early-treated PKU had significantly lower Metacognition older patients who have maintained strict treatment after entering Index scores than healthy controls [43]. Moreover, there is a lack of ev- adolescence. This neuropsychological testing should be performed idence showing associations between the BRIEF-domains and laborato- using the same conceptually informed instruments that have been ry measurements of EF [14]. Many operations in day-to-day life appear used for child studies. Another consideration before deciding upon to require multiple EF simultaneously or EF in combination with other which tests should be used for longitudinal monitoring of cogni- cognitive abilities, and this appears to be reflected in the type of ques- tive outcome, is that there is evidence for non-EF deficits in treated tions asked in the BRIEF. Therefore, the BRIEF can, at present, not replace PKU as well. For instance, studies have shown abnormalities in (computerized) neuropsychological tasks in long-term monitoring of perceptual processing, vigilance and sustained attention deficits, PKU-patients. It may, however, serve as an additional measure in this motor control, and visual-spatial abilities (for an overview, see monitoring process, as it likely to represent problems in daily life better [32]). Although there are studies showing that such deficits become than laboratory tests. much more evident when there is also an executive component in- Despite its shortcomings, there are advantages as well. One such volved [33], this might indicate that either working memory or in- advantage is that it also has an Emotional Control-scale. EF in daily hibitory control could also be combined with non-executive life will often contain some emotional, motivational, or affective com- demands in follow-up/monitoring tasks. ponents. This type of EF can again be investigated using questionnaires and laboratory tasks, both of which have hardly been done in PKU. 2.2. Assessment of EF in PKU Empirically-validated dual-pathway models, comprising both cognitive and emotional control (and for which neuroanatomical and neurobio- Whereas it may be argued that IQ-tests also reflect several EF or EF logical underpinnings have been identified) could be the best possible in combination with other cognitive abilities, it appears that several theoretical framework to capture all functional deficits observed in computerized test batteries such as the Amsterdam Neuropsychological PKU [44,45]. Tasks (ANT) [34] and Cambridge Neuropsychological Test Automated Battery (CANTAB) [35,36] include tasks that more clearly measure 3. Neurological outcome inhibitory control, working memory, a combination of both, and a combination of either with specific other cognitive abilities. The test battery In contrast with the poor neurological outcome in untreated PKU that has been used most frequently in PKU-research is the ANT [37], patients especially with spasticity and epilepsy, there are very few but recently studies have appeared using the CANTAB [38].Perfor- reports on neurological issues in early and continuously treated mance by PKU-patients of ANT-tasks has consistently been related PKU patients. Problems observed in early treated PKU patients are to different indices of dietary control. Such data is not yet available brisk reflexes and tremor that may develop in poorly treated as for the CANTAB, although there are some indications that largely well as well-treated PKU patients especially after adolescence [56], the same cognitive domains are affected and that there are differ- and more severe neurological problems seem to be due to vitamin ences between Phe high and Phe low subjects [38]. More research B12 deficiency rather than high Phe concentrations, while a combi- is necessary on outcomes from both batteries in relation to clinical nation of anesthesia and vitamin B12 deficiency may lead to further parameters and in relation to each other, before a decision can be deterioration of the patient [57–59]. No one knows the weight yet made on which tests could be used to continuously monitor cogni- of brisk reflexes at the longer term as a parameter for neurological tive outcome in PKU-patients. If tests from both batteries are indeed outcome. Do patients with longer or more exaggerated brisk reflexes strongly related, indicating they measure the same cognitive abilities, it develop any motor (“pyramidal”) impairment in the end? Therefore, would be preferable to use only one of the batteries across centers, as at present, we just have to follow this clinical sign thoroughly. some subtle differences will undoubtedly remain present. Moreover, there is still a need for comparable tests that can be administered to 4. Behavior and quality of life the youngest PKU patients (aged 0–5 years old), and there is no clarity yet on the validity of the computerized tasks in relation to day-to-day There is limited knowledge about possible psychiatric problems life EF. in PKU, and there is a particular lack of such data in relation to the In recent years, a number of questionnaires have been developed neurobiology of PKU. Some findings, however, have been sufficiently measuring EF in daily life, with the Behavior Rating Inventory of Execu- replicated to render them important enough to take into consideration tive Function (BRIEF) [39] as their best-known exponent. It has been when assembling a standardized outcome battery. These include the suggested that this questionnaire could be used in regular assessments presence of internalizing behavior problems, specifically a proneness throughout development and during adult life of PKU-patients [40].The to depression, anxiety and phobic tendencies [46]. For aggression or BRIEF is a standardized informant report with versions for 2–5year- antisocial behavior little evidence exists showing lower-than-normal olds (five scales: Inhibit, Shift, Emotional Control, Working Memory, levels. Attention Deficit Hyperactivity Disorder-type behaviors have Plan/Organization), 5–18 year-olds (eight scales: the five scales also been observed in PKU, but evidence has particularly been provided for present for the youngest group plus Initiate, Organization of Materials, attention problems rather than hyperactivity [47]. and Monitor), and for adults (nine scales: Monitor from the BRIEF for Health related quality of life (HRQoL) is a subjective measure of an 5–18 year oldsFOR is divided into Self-Monitor ELECTRONIC and Task Monitor). An over- individual's satisfaction USE or happiness ONLY regarding domains of life that all composite score can be generated from all lists, as well as a number affect health or are affected by health. The effects of having PKU on of broader EF-indices: for 2–5: Inhibitory Self-Control, Flexibility, and the HRQoL of patients can be assessed in cohort studies with validated Emergent Metacognition, and for 5–18 year-olds and adults Behavioral questionnaires. A number of these studies have been published. First in Regulation and Metacognition. To date not many PKU-studies have in- 1992 Weglage et al. evaluated 34 early treated adolescents with a corporated the BRIEF, and unfortunately, results have not been very normal IQ, with a personality inventory and a biographic inventory consistent. Whereas one might expect the most pronounced differences [48]. Patients demonstrated less autonomy, a more negative self with controls in the inhibitory control and working memory domains, description, less extraversion and impulsiveness, a feeling of not Anderson et al. identified impairment on subscales shifting (which being quite healthy, and more grave and a higher level of dependency may be considered a combination of inhibitory control and working from their families. The HRQoL of 37 patients with PKU between 3 memory) and monitoring [41], while Sharman et al. reported impair- and 18 years of age was evaluated in 2002 by Landolt et al. [49].Their ments on a number of subscales: initiation, working memory, planning, parents completed a generic questionnaire, the TNO-AZL Questionnaire organization, and monitoring [42]. Antshel and Waisbren reported that for Children's Health-Related Quality of Life, and the Child Behavior S48 F.J. van Spronsen et al. / Molecular Genetics and Metabolism 104 (2011) S45–S51

Checklist. The only deviation in the PKU group was a reduction of posi- transient (Leuzzi V, unpublished data) or persistent [67,68] Phe varia- tive emotions, and psychological adjustment in patients with PKU was tions, Tyrosine, Tryptophan [15],andL-DOPA[61,69] supplementation. better than in a healthy reference group. The authors concluded that a On the contrary, early and continuous administration of Long-Chain normal HRQoL is possible for patients with PKU. The same was found PUFA improved VEP P100 latency at the age of 12 months [70,71]. in a study of the course of life, sociodemographic outcomes and Event-Related Potentials (ERP) allow for the measurement of brain health-related quality of life in 32 early treated patients with PKU age activity from 1 m to the next and intercept many aspects of attention 18 to 30 years [50]. Patients completed the Course of Life questionnaire and perception, which appear to operate on a temporal scale. A few (a validated questionnaire evaluating accomplishment of developmen- cross-sectional studies have been performed in the last decade in PKU tal tasks and achievement of developmental milestones), the generic patients using different paradigms and tasks [72–75]. Relevant results RAND-36 Health Survey, and the generic cognitive scale of the TNO- were: a) the derangement of ERP generation in late-treated patients; b) AZL Adult Quality of Life questionnaire. The results of the Course of the impairment of early sensory processing in early-treated on and off Life and Health-Related Quality of Life questionnaires were comparable diet patients; c) problems with response inhibition; d) the impairment to controls, except that a higher percentage received special education of a number of components involved in selective attention, which were in primary school. Their educational attainment, however, was compa- influenced by high Phe value; e) scarce accuracy in the task execution, rable to that of their peers. A study of 67 early treated patients, who that improves with age in PKUs (but not in the controls). Future per- completed the Profile of Quality of Life in the Chronically Ill, a question- spectives in this field should involve the assessment of the outcome naire on quality of life and social status, also reported the outcome to be of neurophysiological alterations and the exploration of the linkage in the same range as in the controls [51]. Schooling and professional between neuropsychological and neurophysiological alterations. career corresponded approximately to the control collective. However, Brain white matter (WM) alterations on T2-weighted and Flair evaluation of the social state of PKU patients revealed a tendency MRI sequences were initially described in a few patients suffering toward lower or delayed autonomy, and a low rate of forming normal from neurological deterioration after diet discontinuation. After- adult relationships in which to have children. wards, similar alterations were also detected in over 90% of early In a very recent study seventy two early treated patients demon- treated PKU patients without symptoms of neurological deteriora- strated a normal HRQoL at the time of transfer from pediatric to tion (for a review, see [15]). Most of them were young adult off adult care, and a marital status compared to the general population, diet patients. The fact that WM alterations can be reversed by a strict but fewer patients with PKU had children [52]. diet over a few months [76] strongly points to the causative role of Together, these studies do not demonstrate an absolute negative ef- (CNS) Phe. However, other factors are probably involved as suggested fect of having PKU on the HRQoL of the patients, but there are trends to- by a) the occurrence of WM variation (improvement or worsening) in ward delayed autonomy and possibly a lower rate of forming adult patients who did not change their Phe values, and b) the wide variability relationships and having children. In our experience, if we report the of WM involvement under similar value of blood and CNS Phe [77,78]. normal HRQoL results to our patients, they frequently feel that it does Most of the studies find a weak or no direct link with mental develop- not do justice to the efforts they make to lead a “normal” life. ment and functioning, even if some relations between the degree of In 2010, Gentile et al. proposed the possibility that early- and WM alteration and neuropsychological disorders were found [79].How- well-treated patients experience hidden (subtle) disabilities, that ever, also in this case the variability in clinical outcome (i.e. impairment are not detected with the generic questionnaires but which do affect in patients without WM alteration) suggests the contributing role of ad- their daily lives [53]. As reported in the above section, there are many ditive unknown factors. Moreover, the hypothesis that WM abnormality reports of deficits in executive functioning in patients with PKU. resultsinanimpairmentofneuraltransmissionwasnotconfirmed by Possibly, these executive functioning deficits may lead to psychosocial neurophysiological studies [58]. Only a few data are available on the out- deficits that are not always visible (including social difficulties and come of WM MRI alterations as detected by serial examinations. Prelim- psychosocial problems, such as forming interpersonal relationships, inary results suggest that WM alterations are correlated with lifetime achieving autonomy, attaining educational goals, and having healthy and concurrent Phe level, age (with a marked inter-individual variabili- emotional development). Furthermore, it was recently demonstrated ty), and sex (females showing a greater vulnerability than males). with a randomized controlled trial that high phenylalanine values In conclusion, results to date from MRI examinations and neuro- have a strong negative effect on the mood of patients, perceived by physiological studies do not seem to significantly improve the predic- the patients themselves as well as by relatives and friends [37].This tive power of clinical and biochemical examinations in relation to may be expected to affect social functioning of the patients. neuropsychological and psychosocial outcomes. Therefore, these should, for now, be used for research purposes or to explore specific 5. Neurophysiological and neuroimaging alterations and outcome clinical conditions. Future topics for these powerful techniques should include the identification of a set of preclinical neuro-anatomic or/and Owing to their impressive temporal and spatial resolution, respec- neuro-physiologic alterations, which are predictive of the clinical and tively, neurophysiological and neuroimaging techniques are excellent behavioral outcomes. tools for the non-invasive in vivo study of the central nervous system MRS and PET studies have increased our understanding of brain functions and structure in PKU subjects. Several papers published in neuropathology. Especially the studies of Pietz et al. have learned us the 80s and 90sFOR focused on the signal ELECTRONIC transmission inside the central that large neutral aminoUSE acids may decreaseONLY brain Phe concentrations nervous system and were based on the recording of latency and am- and improve EEG outcome, either due to a decrease of the brain Phe plitude of early components of multimodal Evoked Potentials (EP). The concentrations or an increase of one or more of the other large neutral most consistent result was the increase of the latency of the main com- amino acids [80]. The studies by Weglage et al. and the discussion af- ponent (P100) of Visual Evoked Potentials (VEP), which was reported in terwards have learned us that still many issues of brain Phe measure- 9 out of 12 studies [54–65], and affected mainly patients older than 10. ment are unsolved [81,82]. First, it has to be proven that consecutive Clinical consequences and prognostic value of this alteration, if any, re- MRS measurements result in comparable brain Phe concentrations main to be elucidated. Only a few longitudinally designed studies have using the same MRS in the same institute. This has been done recently been performed so far. A maturational lag of some brainstem auditory [83]. Next, some centers with proven experience within this field need evoked potentials (BAEP) and VEP components with mainly postnatal to compare their data, probably by measuring their patients in the var- development was detected during the first year of life in early treated ious institutes. These studies will help us to further understand the PKU infants and was not associated with any concomitant developmen- role of brain Phe concentrations. In contrast to MRS, PET studies do tal disorder [66]. In older patients EP latencies were not affected by not involve concentrations but processes, such as energy F.J. van Spronsen et al. / Molecular Genetics and Metabolism 104 (2011) S45–S51 S49 metabolism (which informs on activity of various cerebral regions) 4. Further studies should especially be aimed at validation of MRS [84], protein metabolism [85,86], and transport across the blood–brain results as obtained in various centers by various apparatus. barrier [87]. 5. It might be interesting to use brisk reflexes as outcome measure to develop the safe metabolic target especially in adolescent and adult 6. Neurological outcome patients. 6. Regular monitoring, whether this concerns neuropsychological, Given all these possibilities of measures we tend to forget the behavioral or neurological measurements, should be evaluated pure clinical data of brain function, clinical neurological evaluation. not only by clinicians and researchers, but also by the patients In contrast with the poor neurological outcome in untreated PKU and their families themselves in order to increase the number of patients especially with spasticity and epilepsy, only some papers potential benefits for them (cf. [92]). address real neurological issues in early and continuously treated PKU patients. Problems observed in early treated PKU patients are brisk reflexes and tremor that may develop in PKU patients especially 8. Conclusion after adolescence [88–91]. These are not known to have clear clinical relevance, but data relating this kind of outcome to metabolic control This paper summarizes the present knowledge of today's most im- are scarce. Such data might also help to establish the metabolic targets portant measures of outcome in PKU patients. Notwithstanding the pro- of treatment. Therefore, future studies can further take into account the gress in knowledge, there is an evident need for integration at various possible existence of relation between brisk reflexes and treatment levels. Whereas in the past research was performed in different insti- parameters either at present or in history. tutes that were often unaware of each others' activities, efforts should be made to integrate knowledge in future multicenter research pro- 7. Statements grams. PKU-patients are likely to benefit most from a more integrated approach both in and across clinical and research settings. Such a trans- The authors developed the following statements that may help lational approach will increase our understanding of PKU-pathophysi- determining the instruments that can be used in long-term monitoring ology and benefit the development of new treatment strategies. of PKU-patients:

References 1. IQ tests are not the optimal instrument to monitor cognitive out-

come of early and continuously treated patients. [1] N. Blau, F.J. van Spronsen, H.L. Levy, Phenylketonuria, Lancet 376 (2010) 2. In monitoring neuropsychological outcome of PKU patients at vari- 1417–1427. ous ages, the executive functions ‘working memory’ and ‘inhibition [2] D.A. White, S. Waisbren, F.J. van Spronsen, Final commentary: a new chapter, Mol. – of prepotent responding’ should feature, preferably in combination Genet. Metab. 99 (Suppl. 1) (2010) S106 S107. [3] P. Burgard, F. Rey, A.A. Rupp, V. Abadie, J. Rey, Neuropsychologic functions of early with each other, or in combination with other aspects of cognition, treated patients with phenylketonuria, on and off diet: results of a cross-national and such as sustained attention, high perceptual demands and motor cross-sectional study, Pediatr. Res. 41 (1997) 368–374. control. [4] I. Smith, M.G. Beasley, A.E. Ades, Intelligence and quality of dietary treatment in phenylketonuria, Arch. Dis. Child. 65 (1990) 472–478. 3. There is a great demand for measurement of social information [5] F.J. van Spronsen, Mild hyperphenylalaninemia: to treat or not to treat, J. Inherit. processing and emotional control in early and continuously treated Metab. Dis. 34 (2011) 651–656. PKU patients, and measurement of these should be part of long- [6] J.M. Saudubray, F. Rey, H. Ogier, V. Abadie, J.P. Farriaux, J. Ghisolfi, P. Guibaud, J. Rey, M. Vidailhet, Intellectual and school performances in early-treated classical term monitoring of PKU patients. PKU patients. The French collaborative study, Eur. J. Pediatr. 146 (Suppl. 1) 4. The BRIEF may be a valuable instrument for monitoring daily life EF (1987) A20–A22. and emotional control of PKU patients, but it is unlikely, at present [7] P. Burgard, Development of intelligence in early treated phenylketonuria, Eur. J. Pediatr. 159 (Suppl. 2) (2000) S74–S79. or in the future, that it can replace (computerized) laboratory tasks [8] C.G. Azen, R. Koch, E.G. Friedman, S. Berlow, J. Coldwell, W. Krause, R. Matalon, E. in this process. McCabe, M. O'Flynn, R. Peterson, et al., Intellectual development in 12-year-old 5. Studies on HRQoL with generic validated questionnaires may be in- children treated for phenylketonuria, Am. J. Dis. Child. 145 (1991) 35–39. adequate to detect hidden deficits so that development of a PKU spe- [9] R. Koch, B. Burton, G. Hoganson, R. Peterson, W. Rhead, B. Rouse, R. Scott, J. Wolff, A.M. Stern, F. Guttler, M. Nelson, F. de la Cruz, J. Coldwell, R. Erbe, M.T. Geraghty, cific questionnaire is needed to evaluate the HRQoL of PKU patients. C. Shear, J. Thomas, C. Azen, Phenylketonuria in adulthood: a collaborative study, 6. MRS and PET studies may help us to understand the cerebral patho- J. Inherit. Metab. Dis. 25 (2002) 333–346. physiology in PKU, but do not appear to be of added value in longitu- [10] F.J. van Spronsen, P. Burgard, The truth of treating patients with phenylketonuria after childhood: the need of a new guideline, J. Inherit. Metab. Dis. 31 (2008) dinal monitoring of patients. 673–679. [11] S.E. Waisbren, K. Noel, K. Fahrbach, C. Cella, D. Frame, A. Dorenbaum, H. Levy, The following statements are more hypothetical, and are partic- Phenylalanine blood levels and clinical outcomes in phenylketonuria: a sys- ularly aimed at eliciting more research activities so that they might ul- tematic literature review and meta-analysis, Mol. Genet. Metab. 92 (2007) 63–70. timately become important for longitudinal monitoring of PKU-patients [12] J. Albrecht, S.F. Garbade, P. Burgard, Neuropsychological speed tests and blood as well: phenylalanine levels in patients with phenylketonuria: a meta-analysis, Neurosci. Biobehav. Rev. 33 (2009) 414–421. FOR ELECTRONIC[13] S. Huijbregts, L.USE de Sonneville, F.J. van ONLY Spronsen, R. Licht, J. Sergeant, The neuro- 1. Conventional brain MRI examination as well as neurophysiological psychological profile of early and continuously treated phenylketonuria: orienting, studies should be restricted to PKU patients presenting unusual vigilance, and maintenance versus manipulation-functions of working memory, clinical symptoms such as neurological and/or mental deterioration, Neurosci. Biobehav. Rev. 26 (2002) 697–712. tremor, epilepsy, etc. [14] S.E. Christ, S.C.J. Huijbregts, L.M.J. De Sonneville, D.A. White, Executive function in early-treated Phenylketonuria: profile and underlying mechanisms, Mol. Genet. 2. Stimulating topics for the future research in these fields are: a) the ex- Metab. 99 (2010) S22–S32. ploration of the neurophysiological and/or neuroimaging correlates of [15] P.J. Andersen, V. Leuzzi, White matter pathology in phenylketonuria, Mol. Genet. – neuropsychological alterations in PKU patients; b) the identification Metab. 99 (2010) S3 S9. [16] R.G. Schlösser, G. Wagner, H. Sauer, Assessing the working memory network: of a set of preclinical neuro-anatomical and/or neuro-physiological al- studies with functional magnetic resonance imaging and structural equation terations reflecting the individual vulnerability to Phe. modeling, Neuroscience 139 (2006) 91–103. 3. Mood is affected by high Phe values. A validated measure for mood [17] H.M. de Bie, M. Boersma, S. Adriaanse, D.J. Veltman, A.M. Wink, S.D. Roosendaal, F. Barkhof, C.J. Stam, K.J. Oostrom, H.A. Delemarre-van de Waal, E.J. Sanz-Arigita, and well-being needs to be developed and may well prove a strong Resting-state networks in awake five- to eight-year old children, Hum. Brain measure of therapeutic effects in patients with PKU. Mapp. (2011) (Epub ahead of print). S50 F.J. van Spronsen et al. / Molecular Genetics and Metabolism 104 (2011) S45–S51

[18] A.R. Laird, S.B. Eickhoff, K. Li, D.A. Robin, D.C. Glahn, P.T. Fox, Investigating the [46] I. Smith, J. Knowles, Behaviour in early treated phenylketonuria: a systematic review, functional heterogeneity of the default mode network using coordinate-based Eur. J. Pediatr. 159 (2000) S89–S93. meta-analytic modelling, J. Neurosci. 29 (2009) 14496–14505. [47] J.R. Wiersema, J.J. van der Meere, H. Roeyers, State regulation and response inhi- [19] K.N. Ochsner, How thinking controls feeling: a social cognitive neuroscience ap- bition in children with ADHD and children with early- and continuously treated proach, in: E. Harmon-Jones, P. Winkielman (Eds.), Social Neuroscience: Integrating phenylketonuria: an event-related potential comparison, J. Inherit. Metab. Dis. biological and psychological explanations of social behavior, The Guildford Press, 28 (2005) 831–843. New York, 2007, pp. 106–133. [48] J. Weglage, B. Funders, B. Wilken, D. Schubert, E. Schmidt, P. Burgard, K. Ulrich, [20] M. Kane, R. Engle, The role of prefrontal cortex in working-memory capacity, executive Psychological and social findings in adolescents with phenylketonuria, Eur. J. attention, and general fluid intelligence: an individual-differences perspective, Pediatr. 151 (7) (1992) 522–525. Psychon. Bull. Rev. 9 (2002) 637–671. [49] M.A. Landolt, J.M. Nuoffer, B. Steinmann, A. Superti-Furga, Quality of life and psy- [21] M. Bannon, E. Bunney, R. Roth, Mesocortical dopamine neurons: rapid transmitter chologic adjustment in children and adolescents with early treated phenylketon- turnover compared to other brain catecholamine systems, Brain Res. 218 (1981) uria can be normal, J. Pediatr. 140 (5) (2002 May) 516–521. 376–382. [50] A.M. Bosch, W. Tybout, F.J. van Spronsen, H.W. de Valk, F.A. Wijburg, M.A. [22] S.Y. Tam, R.H. Roth, Mesoprefrontal dopaminergic neurons: can tyrosine availability Grootenhuis, The course of life and quality of life of early and continuously influence their functions? Biochem. Pharmacol. 53 (1997) 441–453. treated Dutch patients with phenylketonuria, J. Inherit. Metab. Dis. 30 (2007) [23] E.A. Crone, C. Wendelken, S. Donohue, L. van Leijenhorst, S.A. Bunge, Neurocognitive 29–34. development of the ability to manipulate information in working memory, Proc. [51] E. Simon, M. Schwarz, J. Roos, N. Dragano, M. Geraedts, J. Siegrist, G. Kamp, U. Natl. Acad. Sci. U. S. A. 103 (2006) 9315–9320. Wendel, Evaluation of quality of life and description of the sociodemographic [24] T. Paus, M. Keshavan, J.N. Giedd, Why do many psychiatric disorders emerge during state in adolescent and young adult patients with phenylketonuria (PKU), Health adolescence? Nat. Rev. Neurosci. Biobehav. 9 (2008) 947–957. Qual. Life Outcomes 6 (2008) 25. [25] D. Yurgelun-Todd, Emotional and cognitive changes during adolescence, Curr. [52] U. Mütze, A. Roth, J.F. Weigel, S. Beblo, C.G. Baerwald, P. Bührdel, W. Kiess, Transition Opin. Neurobiol. 17 (2007) 251–257. of young adults with phenylketonuria from pediatric to adult care, J. Inherit. Metab. [26] A. Diamond, M. Prevor, G. Callendar, D. Druin, Prefrontal cortex cognitive deficits Dis. 34 (2011) 701–709. in children treated early and continuously for PKU, Monogr. Soc. Res. Child Dev. [53] J.K. Gentile, A.E. ten Hoedt, A.M. Bosch, Psychosocial aspects of PKU: hidden disabilities 62 (1997) 1–206. —a review, Mol. Genet. Metab. (Suppl. 1) (2010) S64–S67. [27] A. Diamond, L. Briand, J. Fossella, L. Gehlbach, Genetic and neurochemical modulation [54] R. Korinthemberg, K. Ullrich, F. Fullenkemper, Evoked potentials and electroen- of prefrontal cognitive functions in children, Am. J. Psychiatry 161 (2004) 125–132. cephalography in adolescents with phenylketonuria, Neuropediatrics 19 (1988) [28] S. Huijbregts, L. de Sonneville, R. Licht, J. Sergeant, F. van Spronsen, Inhibition of 175–178. prepotent responding and attentional flexibility in treated phenylketonuria, [55] M. Giovannini, R. Valsasina, R. Villani, A. Ducati, E. Riva, A. Landi, R. Longhi, Pattern Dev. Neuropsychol. 22 (2002) 481–499. reversal visual evoked potentials in phenylketonuria, J. Inherit. Metab. Dis. 11 [29] S. Huijbregts, L. de Sonneville, R. Licht, F. van Spronsen, P. Verkerk, J. Sergeant, (1988) 416–421. Sustained attention and inhibition of cognitive interference in treated phenylketonuria: [56] A.C. Ludolph, K. Ullrich, S. Nedjat, H. Masur, U. Bick, Neurological outcome in 22 associations with concurrent and lifetime phenylalanine concentrations, treated adolescents with hyperphenylalaninemia. A clinical and electrophysio- Neuropsychologia 40 (2002) 7–15. logical study, Acta Neurol. Scand. 85 (1992) 243–248. [30] J.J. Moyle, A.M. Fox, M. Arthur, M. Bynevelt, J.R. Burnett, Meta-analysis of neuro- [57] H.C. Lou, P.B. Toft, J. Andresen, I. Mikkelsen, B. Olsen, F. Guttler, S. Wieslander, psychological symptoms of adolescents and adults with PKU, Neuropsychol. Henriksen, J. Inherit. Metab. Dis. 15 (1992) 687–695. Rev. 17 (2007) 91–101. [58] M.A. Cleary, J.H. Walter, J.E. Wraith, et al., Magnetic resonance imaging of the [31] D. White, M. Nortz, T. Mandernach, K. Huntington, R. Steiner, Deficits in brain in phenylketonuria, Lancet 344 (1994) 87–90. memory strategy use related to prefrontal dysfunction during early develop- [59] V. Leuzzi, F. Cardona, I. Antonozzi, A. Loizzo, Visual, auditory and somatosensorial ment: evidence from children with phenylketonuria, Neuropsychology 15 evoked potentials in early and late treated adolescents with phenylketonuria, J. (2001) 221–229. Clin. Neurophysiol. 11 (1994) 603–606. [32] D. Janzen, M. Nguyen, Beyond executive function: non-executive cognitive abilities [60] S.J. Jones, G. Turano, A. Kriss, F. Shawkat, B. Kendall, A.J. Thompson, Visual evoked in individuals with PKU, Mol. Genet. Metab. 99 (Suppl. 1) (2010) S47–S51. potentials in phenylketonuria: association with brain MRI, dietary state, and IQ, J. [33] S.C. Huijbregts, L.M. de Sonneville, F.J. van Spronsen, I.E. Berends, R. Licht, P.H. Neurol. Neurosurg. Psychiatry 59 (1995) 260–265. Verkerk, J.A. Sergeant, Motor function under lower and higher controlled processing [61] K. Ullrich, J. Weglage, C. Oberwittler, M. Pietsch, B. Funders, H. van Eckhardstein, demands in early and continuously treated phenylketonuria, Neuropsychology 17 J.P. Colombo, Effect of L-DOPA on pattern visual evoked potentials (P-100) and (2003) 369–379. neuropsychological tests in untreated adult patients with phenylketonuria, J. Inherit. [34] L.M.J. de Sonneville, Amsterdam Neuropsychological Tasks: A computer aided Metab. Dis. 17 (1994) 349–352. assessment programme, in: B.P.L.M. den Brinker, P.J. Beek, A.N. Brand, F.J. Maarse, [62] J. Pietz, R. Kreis, H. Schmidt, U.K. Meyding-Lamadé, A. Rupp, C. Boesch, Phenylke- L.J.M. Mulder (Eds.), Cognitive ergonomics, clinical assessment and computer assisted tonuria: findings at MR imaging and localized in vivo H-1 MR spectroscopy of the learning: computers in psychology, Swets and Zeitlinger, Lisse, The Netherlands, brain in patients with early treatment, Radiology 201 (1996) 413–420. 1999, pp. 187–203. [63] A.C. Ludolph, U. Vetter, K. Ullrich, Studies of multimodal evoked potentials in [35] B.J. Sahakian, R.G. Morris, J.L. Evenden, A. Heald, R. Levy, M. Philpot, T.W. Robbins, treated phenylketonuria: the patter of vulnerability, Eur. J. Pediatr. 155 (1996) A comparative study of visuospatial memory and learning in Alzheimer-type S64–S68. dementia and Parkinson's disease, Brain 111 (1988) 695–718. [64] V. Leuzzi, S. Rinalduzzi, F. Chiarotti, P. Garzia, G. Trasimeni, N. Accornero, Subclinical [36] M. Luciana, Practitioner review: computerized assessment of neuropsychological visual impairment in phenylketonuria. A neurophysiologic study (VEP-P) with function in children: clinical and research applications of the Cambridge Neuro- clinical, biochemical, and neuroradiological (MRI) correlations, J. Inherit. Metab. psychological Testing Automated Battery (CANTAB), J. Child Psychol. Psychiatry Dis. 21 (1998) 351–364. 44 (2003) 649–663. [65] G.V. McDonnel, T.F. Esmonde, D.R. Hadden, J.I. Morrow, A Neurological evaluation [37] A.E. ten Hoedt, L.M. de Sonneville, B. Francois, N.M. ter Horst, M.C. Janssen, M.E. of adult phenylketonuria in Norhern Ireland, Eur. Neurol. 39 (1998) 38–43. Rubio-Gozalbo, F.A. Wijburg, C.E. Hollak, A.M. Bosch, High phenylalanine levels [66] F. Cardona, V. Leuzzi, I. Antonozzi, P. Benedetti, A. Loizzo, The development of directly affect mood and sustained attention in adults with phenylketonuria: a auditory and visual potentials in early treated phenylketonuric children, EEG randomised, double-blind, placebo-controlled, crossover trial, J. Inherit. Metab. Clin. Neurophysiol. 80 (1991) 8–15. Dis. 34 (2011) 165–171. [67] S.M. Pueschel, L. Fogelson-Doyle, B. Kammerer, Y. Matsumiya, Neurophysiological, [38] M. Bik-Multanowski, J.J. Pietrzyk, R. Mozrzymas, Routine use of CANTAB system psychological, and nutritional investigations during discontinuation of the for detection of neuropsychological deficits in patients with PKU, Mol. Genet. phenylalanine-restricted diet in children with classic phenylketonuria, J. Ment. Metab. 102 (2011) 210–213. Defic. Res. 27 (1983) 61–67. [39] G. Gioia, P. Isquith, S. Guy, L. Kenworthy, Behavior Rating Inventory of Executive [68] H.C. Lou, P.B. Toft, J. Andresen, B. Olsen, P. Guldberg, F. Guttler, Unchanged MRI of Function, Psychological Assessment Resources, Odessa, FL, 2000. myelin in adolescents with PKU supplied with non-phe essential amino acids [40] S. Waisbren, D.A. White, Screening for cognitive and social–emotional problems after diet relaxation, Acta Paediatr. 83 (1994) 1312–1314. in individualsFOR with PKU: tools for use inELECTRONIC the metabolic clinic, Mol. Genet. Metab. [69] K. Ullrich, J. Weglage, USE C. Oberwittler, M. Pietsch,ONLY B. Funders, H.N. Eckhardstein, Ef- 99 (2010) S96–S99. fect of L-Dopa on visual evoked potentials and neuropsychological tests in adult [41] V. Anderson, P. Anderson, E. Northam, R. Jacobs, O. Mikiewicz, Relationships between phenylketonuria patients, Eur. J. Pediatr. 155 (1996) S74–S77. cognitive and behavioral measures of executive function in children with brain disease, [70] S. Beblo, H. Reinhardt, A.C. Muntau, W. Mueller-Felber, A.A. Roscher, B. Koletzko, Child Neuropsychol. 8 (2002) 231–240. Fisch oil supplementation improves visual evoked potentials in children with [42] R. Sharman, K.R. Sullivan, R.K.A. Young, J.R.M. McGill, Biochemical markers asso- phenylketonuria, Neurology 57 (2001) 1488–1491. ciated with executive function in adolescents with early and continuously treated [71] C. Agostoni, E. Verducci, N. Massetto, G. Radaelli, E. Riva, M. Giovannini, Plasma phenylketonuria, Clin. Genet. 75 (2009) 169–174. long-chain polyunsatured fatty acids and neurodevelopment through the first [43] K.M. Antshel, S.E. Waisbren, Timing is everything: executive functions in chil- 12months of life in phenylketonuria, Dev. Med. Child Neurol. 45 (2003) dren exposed to elevated levels of phenylalanine, Neuropsychology 17 (2003) 257–261. 458–468. [72] V. Leuzzi, S. Seri, A. Cerquiglini, C.I. Carducci, Ca. Carducci, I. Antonozzi, Derangement [44] J. Sonuga-Barke, Psychological heterogeneity in AD/HD: a dual pathway model of of the dopaminergic system in phenylketonuria: study of the event-related potential behaviour and cognition, Behav. Brain Res. 130 (2002) 29–36. (P300), J. Inherit. Metab. Dis. 23 (2000) 317–320. [45] A.L. Krain, A.M. Wilson, R. Arbuckle, F.X. Castellanos, M.P. Milham, Distinct neural [73] R.M. Henderson, D.L. McCulloch, A.M. Herbert, P.H. Robinson, M.J. Taylor, Visual mechanisms of risk and ambiguity: a meta-analysis of decision-making, Neuro- event-related potentials in children with phenylketonuria, Acta Paediatr. 89 image 32 (2006) 477–484. (2000) 52–57. F.J. van Spronsen et al. / Molecular Genetics and Metabolism 104 (2011) S45–S51 S51

[74] J.J. Moyle, A.M. Fox, M. Bynevelt, M. Arthur, J.R. Burnett, Event-related potentials [83] R. Kreis, K. Zwygart, C. Boesch, J.M. Nuoffer, Reproducibility of cerebral phenylalanine elicited during a visual Go-Nogo task in adults with phenylketonuria, Clin. Neuro- levels in patients with phenylketonuria determined by 1H-MR spectroscopy, Magn. physiol. 117 (2006) 2154–2160. Reson. Med. 62 (2009) 11–16. [75] L.M.J. de Sonneville, Huijbregts, F.J. van Spronsen, P.H. Verkerk, J.A. Sergeant, R. [84] J. Pietz, A. Rupp, F. Ebinger, D. Rating, E. Mayatepek, C. Boesch, R. Kreis, Cerebral Licht, Event-related potential correlates of selective processing in early- and con- energy metabolism in phenylketonuria: findings by quantitative In vivo 31P MR tinuously-treated children with phenylketonuria: effects of concurrent phenylal- spectroscopy, Pediatr. Res. 53 (2003) 654–662. anine level and dietary control, Mol. Genet. Metab. 99 (2010) S10–S17. [85] A.M. Paans, J. Pruim, G.P. Smit, G. Visser, A.T. Willemsen, K. Ullrich, Neurotrans- [76] M.A. Cleary, J.H. Walter, J.E. Wraith, F. White, K. Tyler, J.P. Jenkins, Magnetic resonance mitter positron emission tomographic-studies in adults with phenylketonuria, a imaging in phenylketonuria: reversal of cerebral white matter change, Eur. J. Pediatr. pilot study, Eur. J. Pediatr. 155 (Suppl. 1) (1996) S78–S81. 127 (1995) 251–255. [86] M. Hoeksma, D.J. Reijngoud, J. Pruim, H.W. de Valk, A.M. Paans, F.J. van Spronsen, [77] V. Leuzzi, M. Tosetti, D. Montanaro, et al., The pathogenesis of the white matter ab- Phenylketonuria: high plasma phenylalanine decreases cerebral protein synthesis, normalities in phenylketonuria. A multimodal 3.0 tesla MRI and magnetic reso- Mol. Genet. Metab. 96 (2009) 177–182. nance spectroscopy (1H MRS) study, J. Inherit. Metab. Dis. 30 (2007) 209–216. [87] C. Landvogt, E. Mengel, P. Bartenstein, H.G. Buchholz, M. Schreckenberger, T. [78] V. Leuzzi, F. Chiarotti, J. Walter, F. Mercante, P. Burgard, The outcome of the white Siessmeier, A. Scheurich, R. Feldmann, J. Weglage, P. Cumming, F. Zepp, K. Ullrich, matter alteration in early treated phenylketonuric (PKU) patients, J. Inherit. Reduced cerebral fluoro-L-dopamine uptake in adult patients suffering from phe- Metab. Dis. 33 (suppl 1) (2010) S113. nylketonuria, J. Cereb. Blood Flow Metab. 28 (2008) 824–831. [79] P.J. Anderson, S.J. Wood, D.E. Francis, L. Coleman, L. Warwick, S. Casanelia, V.A. [88] W.B. Hanley, A. Feigenbaum, J.T.R. Clarke, W. Schoonheyt, V. Austin, Vitamin B12 Anderson, A. Boneh, Neuropsychological functioning in children with early-treated deficiency in adolescents and young adults with phenylketonuria, Eur. J. Pediatr. phenylketonuria: impact of white matter abnormalities, Dev. Med. Child Neurol. 46 155 (1) (1996) 145–147. (2004) 230–238. [89] A.J. Thompson, I. Smith, D. Brenton, B.D. Youl, G. Rylance, D.C. Davidson, Neurological [80] J. Pietz, R. Kreis, A. Rupp, E. Mayatepek, D. Rating, C. Boesch, H.J. Bremer, Large deterioration in young adults with phenylketonuria, Lancet 336 (1990) 602–605. neutral amino acids block phenylalanine transport into brain tissue in patients [90] P. Lee, I. Smith, A. Piesowicz, D. Brenton, Spastic paraparesis after anaesthesia, with phenylketonuria, J. Clin. Invest. 103 (1999) 1169–1178. Lancet 353 (1999) 554. [81] J. Weglage, D. Wiedermann, J. Denecke, R. Feldmann, H.G. Koch, K. Ullrich, E. [91] J. Pietz, R. Dunckelmann, A. Rupp, D. Rating, H.M. Meinck, H. Schmidt, H.J. Bremer, Harms, H.E. Möller, Individual blood–brain barrier phenylalanine transport deter- Neurological outcome in adult patients with early-treated phenylketonuria, Eur. J. mines clinical outcome in phenylketonuria, Ann. Neurol. 50 (2001) 463–467. Pediatr. 157 (1998) 824–830. [82] J. Pietz, A. Rupp, P. Burgard, C. Boesch, R. Kreis, No evidence for individual blood– [92] D. Bodin, J.T. Beetar, K.O. Yeates, K. Boyer, A.N. Colvin, S.A. Mangeot, A survey of brain barrier phenylalanine transport to influence clinical outcome in typical parent satisfaction with pediatric neuropsychological evaluations, Clin. Neuropsychol. phenylketonuria patients, Ann. Neurol. 52 (2002) 382–383. 21 (2007) 884–898.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S52–S54

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Minireview

Vitamin B12 deficiency and phenylketonuria John H. Walter ⁎

Inherited Metabolic Medicine, Manchester Academic Health Sciences Centre, Central Manchester University Hospitals NHS Foundation Trust, Manchester, United Kingdom Willink Biochemical Genetics Unit, Genetic Medicine, 6th Floor, St. Mary's Hospital, CMFT, Oxford Rd., Manchester, M13 9WL, United Kingdom article info abstract

Article history: The literature regarding the vitamin B12 status of patients with phenylketonuria was reviewed. Adequate Received 2 July 2011 amounts of B12 are provided in products used in dietary treatment; however, a number of case reports and Received in revised form 17 July 2011 cohort studies document deficiency in those who have discontinued taking amino acid, mineral and vitamin Accepted 18 July 2011 supplements but who continue to eat only very limited amounts of natural protein. Symptoms and signs of B Available online 22 July 2011 12 deficiency are variable but severe deficiency may cause serious neurological disease. Nitrous oxide anaesthesia is a particular risk. It is recommended that plasma total homocysteine and plasma or urinary methylmalonate Keywords: fi Phenylketonuria should be routinely measured, as they are more sensitive markers of de ciency than serum B12 concentrations. fi PKU Functional B12 de ciency can occur in the presence of a normal B12 concentration.

Contents

1. Introduction ...... S52

2. PKU and vitamin B12 ...... S53

3. Neurological disease in PKU—is it related to B12 deﬁciency? ...... S53

4. Prevention of B12 deﬁciency in PKU ...... S53 5. Nitrous oxide...... S53 6. Conclusion ...... S54 Con. ﬂict of interest statement ...... S54 References...... S54

1. Introduction occurs in the lysosome (release of cobalamin from protein), the cytoplasm (methylation) and in mitochondria (adenosylation). Total

Vitamin B12 (cobalamin) is essential for the function of only two body stores of B12 are estimated to be about 2.5 mg in adults with half enzymes: as methylcobalamin it is required for normal activity of of this being present in the liver. The recommended daily intake for methionine synthase and as adenosylcobalamin for methylmalonyl children is 0.7 μg per day, for adults and adolescents 2 μg per day and

CoA mutase. Cobalamin is a complex molecule made up of a cobalt and during pregnancy and lactation 2.6 μg per day. Functional vitamin B12 a corrin ring [1]. Digestion and absorption require the presence of an R deficiency can occur for a number of reasons, including the following: binder protein, intrinsic factor and then uptake into enterocytes via the cubulin receptor. Once absorbed cobalamin circulates in the plasma • poor dietary intake (vegans, bowel disease etc.) combined withFOR either of 2 carrier proteins:ELECTRONIC transcobalamin (TCI) and • defects in absorption USE (PA, Imerslund ONLY Grasbeck syndrome), haptocorrins (also known as TCII and TCIII). Cobalamin saturated • defects in transport (TC1 deficiency) transcobalamin (holotranscobalamin) is the biologically active form • defects in cellular processing (Cbl A-F) and constitutes between 6% and 20% of total plasma vitamin B12.By Biochemical consequences of vitamin B deficiency include the contrast haptocorrin has no other known function apart from binding 12 following: metabolically inert forms of B12. Intracellular processing of cobalamin • reduced purine synthesis which disturbs nucleotide synthesis, integrity and transcription ⁎ Willink Biochemical Genetics Unit, Genetic Medicine, 6th Floor, St. Mary's Hospital, • — CMFT, Oxford Rd., Manchester, M13 9WL, United Kingdom. Fax: +44 161 701 2303. reduced methyl donors which has an adverse effect on DNA E-mail address: [email protected]. methylation and on non-genomic methylation. The latter has

consequences for numerous neurometabolic pathways, for example meat, poultry, shellﬁsh, milk and milk products, all foods that are high the production of myelin basic protein and membrane phospholipids. in protein and therefore disallowed in the PKU diet. For those patients • increased homocysteine, with adverse effects on endothelial who have stopped taking any amino acids, mineral and vitamin function and damage to DNA within the central nervous system supplement but have continued on a low protein intake then they are

• increased methylmalonic acid likely to be at risk from vitamin B12 deficiency. This was further supported by the data presented by Robinson et al., fi Vitamin B12 de ciency has also been shown to disturb a number of who found that in a cohort of 83 patients with PKU, the lowest levels of cytokines in growth factors including TNF Alfa, nerve growth factor, vitamin B12 were in those on an unrestricted diet who were not taking IL6 and epidermal growth factor. amino acids supplement as compared to those who either were on a Laboratory abnormalities are variable but may include: relaxed diet (but still taking supplements) or under strict dietary treatment [6]. • low plasma vitamin B12 (where diet and/or absorption are abnormal), Hvas et al. reported a case series of 31 adult patients whose ages • a macrocytic megaloblastic anaemia, ranged between 18 and 43 years [7]. None had overt clinical vitamin fi • increased blood homocysteine and B12 de ciency but 13 had at least 1 neurological symptom including 9 • an increase in urine and plasma methylmalonate. who had paraesthesia. Twenty four (77%) had biochemical evidence of low vitamin B12, nine (29%) had plasma homocysteine above In view of these profound biochemical abnormalities that can 12 μmol/L, and eleven (39%) were found to take less than the daily occur, and their consequences within the central and peripheral recommended vitamin B12. Patients with at least one neurological nervous system it is unsurprising that neurological manifestations symptom were reported to have had lower serum vitamin B12 associated with vitamin B12 are common. These deficiencies include concentrations although this did not reach statistical significance. peripheral neuropathy, subacute combined degeneration of the cord Vugteveen et al. studied 75 patients with PKU whose ages ranged (posterior and lateral column involvement) and a spastic paraparesis between 1 to 37 years [8]. They found that 67 of patients had vitamin in the most severe forms, autonomic dysfunction, optic atrophy and B12 concentration within the reference range, however, despite this, psychiatric illness. In infancy cerebral atrophy and encephalopathy 10 (15%) had either increased concentrations of methylmalonate or can occur. Neurological findings can include symmetrical distal homocysteine. Of the remaining 8 patients who had low vitamin B12 sensory manifestations and ataxia, diminished vibration sense and concentrations, 1 patient had an increased level of serum methylmalonate proprioception, exaggerated, diminished or absent limb reflexes [2]. and 1 of homocysteine; however 6 had normal concentrations of both. The authors felt that either serum methylmalonate or homocysteine were

2. PKU and vitamin B12 reliable indicators of vitamin B12 deficiency and superior to measuring serum vitamin B12 alone and recommended that these should be Vitamin B12 is not involved with phenylalanine catabolism nor is monitored regularly in PKU. Holotranscobalamin has been shown to be the metabolism of vitamin B12 affected by high blood phenylalanine the best indicator of B12 deficiency, at least in the elderly [9],butis levels; however, despite this, vitamin B12 deficiency does appear to be currently not routinely available. relatively common in patients with PKU.

In 1981 Farriaux et al. reported a patient with PKU who had a 3. Neurological disease in PKU—is it related to B12 deficiency? macrocytic anaemia and megaloblastic bone marrow and who also had methylmalonic acid in the urine and a low normal serum B12 level In adults with PKU who have stopped treatment, minor neurological [3]. abnormalities are relatively common and include brisk reflexes, resting Hanley et al. reported an 18 year old woman who was poorly or intention tremor, possibly an increased frequency of depression and compliant with treatment for her PKU who developed a slowly MR brain abnormalities. It is possible although unproven that some of progressive spastic paraparesis [4]. She was found to have a megaloblas- these manifestations might be related to B12 deficiency. tic anaemia and a low serum of vitamin B12 of 65.8 pmol/L (reference More severe neurological disease appears to be very much less range 150–700). She showed a rapid improvement in her anaemia frequent in PKU. Thompson et al. described 7 patients between the ages following oral vitamin B12 but there was only an incomplete resolution of of 13 and 25 who developed severe late neurological illness and who her neurological disease. These authors also looked at other patients and were poorly compliant with diet [10]. However in their series they did reported on 37 adults and adolescents with PKU. They found that 6 (16%) not measure or report vitamin B12, homocysteine or methylmalonate had a low vitamin B12 level of less than 150 pmol/L, a further 6 (16%) had levels. Consequently it is not possible to know whether these clinical borderline levels of vitamin B12 between 150 and 200 pmol/L. Only 1 manifestations could have been related to B12 deficiency. patient had a low haemoglobin (although 8 had increased MCVs of

N94 fl). They did not measure methylmalonate nor homocysteine. None 4. Prevention of B12 deficiency in PKU of these individuals had neurological abnormalities on clinical examination. On the basis of their studies they recommended that patients In order to prevent vitamin B12 deficiency in PKU, it is important with PKU should have routine measurement of plasma vitamin B12 and that those on significant natural protein restriction take vitamin B12 fi folate, urine methylmalonateFOR and ELECTRONIC total plasma homocysteine. supplements. VitaminUSE B12 is included, ONLY in suf cient amounts, in Aung et al. reported a 19 year old male with PKU who presented commercial products for both children and adults with PKU and with tiredness, poor memory and poor concentration [5].On therefore those who are taking these regularly in the prescribed examination he had a pale smooth red tongue but the neurological quantities, should not become vitamin B12 deficient. examination was normal. This individual had stopped diet at the age of 14 and had been lost to follow up. On investigation he was found to 5. Nitrous oxide have a megaloblastic macrocytic anaemia, a serum vitamin B12 of 125 pg/mL (reference range 200–900), a normal serum folate and a Finally it is important to remember that nitrous oxide forms a normal Schilling test. He was started on oral vitamin B12 and after one complex with the cobalt in methylcobalamin and results in deactiva- month his anaemia had resolved. The authors hypothesised that tion of methionine synthase. Individuals who have low or marginal of stopping or relaxing the PKU diet would predispose to vitamin B12 vitamin B12 levels are therefore at risk from nitrous oxide anaesthesia. deficiency because of poor dietary intake. This conclusion seemed This is clearly a risk factor in some patients with PKU as exemplified in quite plausible since the main source of vitamin B12 is from eggs, the case report by Lee et al. of a patient with PKU who at the age of 14 S54 J.H. Walter / Molecular Genetics and Metabolism 104 (2011) S52–S54 had a myringoplasty under general anaesthesia using nitrous oxide References [11]. Following this procedure he became irritable and disorientated [1] B. Krautler, Vitamin B12: chemistry and biochemistry, Biochem. Soc. Trans. 33 and had episodes of urinary incontinence. On examination he had a (2005) 806–810.

Parkinsonian facies, a resting tremor, lower limb hypotonia and an [2] E. Reynolds, Vitamin B12, folic acid, and the nervous system, Lancet Neurol. 5 extensor plantar response with sustained ankle clonus. His phenylal- (2006) 949–960. μ [3] J.P. Farriaux, D. Desombre-Denys, M.A. Charles-Bassi, J.L. Dhondt, Le traitement de anine was raised at 1490 mol/L, haemoglobin 8.4 g/dl with an MCV of la phenylcetonurie, Semin. Hop. Paris 57 (1981) 356–360. ﬂ b 114.9 , and vitamin B12 level of 110 pg/mL. [4] W.B. Hanley, A. Feigenbaum, J.T. Clarke, W. Schoonheyt, V. Austin, Vitamin B12 deﬁciency in adolescents and young adults with phenylketonuria, Lancet 342 (1993) 997.

6. Conclusion [5] T.T. Aung, A. Klied, J. McGinn, T. McGinn, Vitamin B12 deficiency in an adult phenylketonuric patient, J. Inherit. Metab. Dis. 20 (1997) 603–604. In conclusion, individuals with PKU who are off diet and not taking [6] M. Robinson, F.J. White, M.A. Cleary, E. Wraith, W.K. Lam, J.H. Walter, Increased risk of vitamin B12 deficiency in patients with phenylketonuria on an unrestricted supplement are at risk from vitamin B12 deficiency. It is unclear or relaxed diet, J. Pediatr. 136 (2000) 545–547. whether vitamin B12 deficiency has been a major causative factor in [7] A.M. Hvas, E. Nexo, J.B. Nielsen, Vitamin B12 and vitamin B6 supplementation is those with severe neurological deterioration. Serum vitamin B levels needed among adults with phenylketonuria (PKU), J. Inherit. Metab. Dis. 29 12 (2006) 47–53. are an insensitive marker of deficiency and anaemia may not always [8] I. Vugteveen, M. Hoeksma, A.L. Monsen, M.R. Fokkema, D.J. Reijngoud, M. van Rijn, be present. It is recommended that plasma or urine methylmalonate F.J. van Spronsen, Serum vitamin B12 concentrations within reference values do not exclude functional vitamin B12 deficiency in PKU patients of various ages, Mol. should be measured along with total homocysteine since functional – fi Genet. Metab. 102 (2011) 13 17. B12 de ciency can occur in the presence of a normal serum B12 [9] E. Valente, J.M. Scott, P.M. Ueland, C. Cunningham, M. Casey, A.M. Molloy, concentration. It is important to ensure adequate intake of vitamin B12 Diagnostic accuracy of holotranscobalamin, methylmalonic acid, serum cobala- and continue long-term follow up in all adults with PKU. min, and other indicators of tissue vitamin B12 status in the elderly, Clin. Chem. 57 (2011) 856–863. [10] A.J. Thompson, I. Smith, D. Brenton, B.D. Youl, G. Rylance, D.C. Davidson, B. Kendall, Conflict of interest statement A.J. Lees, Neurological deterioration in young adults with phenylketonuria, Lancet 336 (1990) 602–605. [11] P. Lee, I. Smith, A. Piesowicz, D. Brenton, Spastic paraparesis after anaesthesia, The author has declared no conflicts of interest. Lancet 353 (1999) 554.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S55–S59

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Speciﬁc prebiotics in a formula for infants with Phenylketonuria

Anita MacDonald a,⁎, Barbara Cochrane b, Harm Wopereis c, Nik Loveridge d a Birmingham Children's Hospital, Birmingham, UK b Royal Hospital for Sick Children, Glasgow, UK c Danone Research, Center for Specialised Nutrition, Wageningen, The Netherlands d Nutricia Advanced Medical Nutrition, Liverpool, UK article info abstract

Article history: Objective: This exploratory study investigated the influence of adding a patented, specific mixture of prebiotic Received 20 July 2011 oligosaccharides (scGOS/lcFOS [9:1 ratio], Danone Research) to a protein substitute suitable for infants with Received in revised form 10 September 2011 Phenylketonuria (PKU); PKU Anamix Infant (Nutricia). Accepted 10 September 2011 Design: This was an 8-week open-label, single-arm, pilot intervention study in 9 infants (8-week median Available online 16 September 2011 age) diagnosed with PKU. On study entry, infants were prescribed PKU Anamix Infant to replace an infant phenylalanine-free protein substitute without prebiotics (IPS). Blood phenylalanine concentrations were moni- Keywords: tored and stool samples analyzed for pH/bacterial groups. Infant nutrition disorders Infant formula Results: PKU Anamix infant was well tolerated and accepted with no adverse events reported. Overall, plasma Nutrition therapy phenylalanine and tyrosine concentrations were maintained within target ranges throughout the study (120– Phenylketonuria 360 μmol/l phenylalanine, 30–100 μmol/l tyrosine). All infants exhibited microbiota dominated by bifidobacteria Prebiotics (median 58.97% at Week 8), although no statistically significant change from baseline was observed at study Fibers endpoint. No infants showed abnormally high levels of Clostridium histolyticum/lituseburense or potentially pathogenic enterobacteriaceae at any point during the study. A significant reduction in median stool pH versus baseline was observed at Week 4 (pH reduced from 6.79 to 5.83), but this significance was not present at Week 8(pH=6.61). Conclusions: PKU Anamix Infant maintains phenylalanine control in line with established IPS without prebiotics and maintains levels of bifidobacteria and lowers stool pH. In exclusively breast-fed infants the latter two factors have been associated with a reduced risk of infection and may be of particular importance in infants with PKU. © 2011 Elsevier Inc. All rights reserved.

1. Introduction intake of between 362 and 464 ml/day of breast milk [1] (providing approximately 3–4 g of intact protein), which is substantially lower A life-long low phenylalanine diet remains the first-line treatment than infants on a normal diet. Infants with moderate to severe PKU in the management of the inherited metabolic disorder Phenylketon- receive up to 75% of their protein requirements (excluding phenylal- uria (PKU). In the UK, the diet primarily consists of: 1) a measured but anine), along with an equivalent percentage of their vitamins and restricted allocation of phenylalanine, providing typically less than minerals, from their IPS. 10 g/day of intact protein, 2) a phenylalanine-free protein substitute Breast milk is widely recognized as the gold-standard for infant consisting of all other L-amino acids, vitamins and trace minerals, feeding, providing all essential nutrients for growth and development and 3) allowance of foods naturally low in phenylalanine that are [2]. Compared with formula-fed infants, breast-fed infants have an given without measurement. In infants with PKU, phenylalanine tol- enhanced immune system and reduced risk of infection [3,4].Itis erance can vary considerably dependent on the individual, but it has therefore advantageous that the composition of IPS continues evolv- been reportedFOR that neonates (bELECTRONIC6 months) with PKU have a mean ing to achieve USE a nutritional composition ONLY comparable to that of human breast milk. Breast milk naturally contains prebiotic ingredients (e.g. oligosaccharides), which has led to the supplementation Abbreviations: Infant Protein substitute (IPS), refers to an infant formula without phenylalanine for infants with phenylketonuria; PB, refers to prebiotic oligosaccharides/ of standard infant formulae with components that have similar ef- prebiotics; XP Analog LCP, is an infant phenylalanine-free protein substitute without fects [5,6]. Dietary prebiotics are defined as “a selectively fermented prebiotics and all Subjects were taking this before changing to study formula. It is referred ingredient that results in specific changes, in the composition to as IPS without prebiotics; PKU Anamix Infant, is a prebiotic containing infant and/or activity of the gastrointestinal microbiota, thus conferring phenylalanine-free protein substitute under evaluation (i.e. the study formula). benefit(s) upon host health” [7,8]. ⁎ Corresponding author at: Birmingham Children's Hospital, Steelhouse Lane, Bir- mingham, B4 6NH, UK. Fax: +44 121 333 8021. If an infant is unable to exclusively receive breast milk, the compo- E-mail address: [email protected] (A. MacDonald). sition of the intestinal microbiota can be influenced in two ways: by

1096-7192/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2011.09.015 S56 A. MacDonald et al. / Molecular Genetics and Metabolism 104 (2011) S55–S59 ingesting living health-promoting bacteria that are able to survive the concentrations were monitored weekly throughout the study. Care- GI tract and colonize the colon (probiotic approach) [9] or by dietary givers, who had received training from specialist nurses, collected ingredients that are able to reach the colon and selectively stimulate blood spots from infants at standardized times, always before the first health-promoting bacteria (prebiotic approach) [10,11]. Prebiotic ol- feed in the morning. These were subsequently analyzed by the partici- igosaccharides therefore, as an important component of breast milk, pating hospital using tandem mass spectrometry. are of particular importance in any infant formula. With highly re- Stool samples were collected at Weeks −1 (extended baseline pe- stricted intakes of breast milk and standard infant formula being the riod), 4 and 8 by the caregiver using specialized vials/scoops and cornerstone of the management of PKU, it is important to identify stored at −20 °C. Two samples were collected, if possible, for each whether infants with PKU are at a higher risk of developing a less time point. Samples were subsequently packed in insulated con- than optimal intestinal microbiota, through decreased intake of pre- tainers containing solid CO2 (i.e. dry ice) and posted to Danone Re- biotic oligosaccharides. search B.V. (Wageningen, The Netherlands) where they were Studies have demonstrated that standard infant formula supple- analyzed for fecal pH and dominant bacterial groups: bifidobacteria, mented with short-chain galactooligosaccharides (scGOS) and long- lactobacilli–enterococci, bacteroides distasonis/fragilis, Clostridium chain fructooligosaccharides in the ratio 9:1 increases bifidobacteria histolyticum/lituseburense, E. rectale/C.coccoides group and a subset of and lactobacilli levels and decreases fecal pH [12–16]. Clinical studies enterobacteriaceae (E. coli, Shigella, Salmonella, Klebsiella) by means have also reported that these changes to the gastrointestinal microbiota of fluorescence in situ hybridization (FISH) as described by Knol et may be associated with better gastrointestinal transit (stool frequency al. [27]. In addition, caregivers recorded stool frequency, size (small, and consistency) and a reduced risk of infection [12,17–25]. It therefore, moderate, large or huge), appearance (hard lumps, lumpy/allantoid, appears appropriate that prebiotic components, i.e. scGOS/lcFOS (9:1 allantoid/cracked, allantoid/smooth, soft blobs, fluffy pieces, watery) ratio), are added to an IPS. and consistency (firm/hard, formed, soft/unformed, semi-liquid and Little is known about the GI function and bowel habits of infants watery) at Weeks −1, 1, 4 (no intake recorded) and 8. This was com- with amino acid disorders, although MacDonald et al. reported that pleted using an adapted stool chart that combined the Bristol Stool caregivers of children with PKU perceived them to have more issues Scale [28] with an independent chart from the Children's Hospital, with constipation, diarrhea and stomach pains [26], possibly associated Birmingham, UK (Murphy, personal communication). Any adverse with a high intake of high osmolar protein substitutes. While these ef- events or serious adverse events were recorded by the investigator fects alone is undesirable, abnormal bowel function can disrupt feeding, using a standard clinical trial report form. which may ultimately affect metabolic control. Therefore, understand- Due to the rarity of the condition and the exploratory nature of the ing the potential influence of gut modulating ingredients could ulti- study, sample size calculations were not performed and only descrip- mately contribute to improved control in metabolic patients. tive statistics were undertaken. These analyses were performed on Due to the rarity of PKU, there has been no research into the effica- the intent to treat population (ITT), which included all subjects who cy and tolerability of added prebiotic oligosaccharides to an IPS. Here completed the changeover to the IPS with PB. Comparisons for statis- we report the findings from the first exploratory study to investigate tical significance were performed using Sigma Stat Statistical Analysis the tolerability and efficacy of an IPS with prebiotics (PKU Anamix In- Software: student's t-tests were applied to normally distributed data, fant: Nutricia). The study objectives were to evaluate any effects on the and Wilcoxon Signed Rank Tests applied to non-parametric data. blood phenylalanine control, gastro-intestinal (GI) tolerance (i.e. stool Ethical approval was obtained from the South Birmingham ethical frequency and characteristics) and GI microbiota of infants with PKU, committee. Prior to enrolment in the study, written informed consent who changed from an IPS without prebiotics to PKU Anamix Infant was obtained from the primary caregiver of each subject. (Advanced Medical Nutrition, Liverpool). 3. Results 2. Methods In total, nine infants were recruited and completed the study, six of This was an 8-week, open-label, pilot intervention study in infants whom were female. The median age at baseline was 7.86 weeks (range aged between 4 weeks and 6 months with a diagnosis of PKU requir- 7.14–19.43). At study entry (in addition to IPS without PB), infants re- ing a low phenylalanine diet. A one week extended baseline period ceived their intact protein from either breast milk (n=3), a combina- was included in the study design; during this time infants entered tion of breast milk and infant formula containing prebiotics (n=3), the study, but remained on an IPS without prebiotics. After this ex- solely infant formula containing prebiotics (n=2), or standard infant tended baseline period, subjects transferred onto PKU Anamix Infant. formula without prebiotics (n=1). Overall, the daily amount of breast Patients were recruited from two specialist PKU centers in the UK milk/formula consumed remained similar throughout the study. There (Birmingham Children's Hospital and the Royal Hospital for Sick Chil- appeared to be a trend towards infants receiving a greater amount of dren, Glasgow). Inclusion criteria included satisfactory blood phenylal- PKU Anamix Infant at Week 8 than at baseline (associated with increas- anine control for ≥2 weeks, whilst receiving IPS without prebiotics and ing energy requirements and infant growth), with mean pre-study pro- either breast milk or commercial infant formula (with or without prebi- tein intakes from IPS without prebiotics of 8.71 g of protein equivalent otics) in amounts dependent on plasma phenylalanine concentrations per day (g/PE/day) (range 4.33–11.7 g/day), and mean intakes of PKU and individual toleranceFOR to natural protein.ELECTRONIC Exclusion criteria included Anamix Infant at weekUSE 8 of 11.3 g PE/day ONLY (range 6.9–14.9 g PE/day). A the presence of a serious concurrent illness, intake of solids, abnormally similar pattern was exhibited in standard infant formula intake, with low birth weight (b2000 g), prematurity (b37 weeks gestation), treat- mean pre-study protein intakes from standard formula of 4.16 g PE/day ment with antibiotics b2 weeks prior to study entry, or investigator (range 3.64 g–4.83 g PE/day) and mean week 8 intakes of 3.88 g PE/day concerns relating to the willingness or ability of the subject to comply (range 0.8–4.8 g PE/day). Mean protein intakes from breast milk cannot with protocol requirements. The study objectives were to investigate be included in these figures as only values of “minutes on breast” are phenylalanine control (efficacy) and tolerability of the PKU Anamix In- available for those subjects. One infant (subject 3) initiated solids fant and to observe any effect on gastrointestinal microbiota and stool during Week 8 of the trial, at 20 weeks of age. consistency. Values for all primary and secondary assessments at each study On entry into the study, infants were prescribed PKU Anamix Infant visit are summarized in Table 1.Overall,phenylalanineandtyro- for a total of 8 weeks (Weeks 0–8), to replace the previous IPS without sine blood concentrations were maintained within target ranges PB (Weeks −1–0). Three day intake diaries were completed by care- (120–360 μmol/l phenylalanine, 30–100 μmol/l tyrosine) through- givers during weeks −1, 4 and 8. Blood phenylalanine and tyrosine out the study (Table 1). Median weight-for-age Z-scores remained A. MacDonald et al. / Molecular Genetics and Metabolism 104 (2011) S55–S59 S57 similar whilst on study product (−0.88 at Week 0, to −0.62 at Week 8). PKU Anamix Infant was well tolerated and well accepted. No adverse events or serious adverse events were reported, and full prescriptions of feeds were consistently taken (recorded formula intake measured against prescribed intake: 95% at Week 4 and 98% at Week 8).

3.1. Bacterial assessments

No statistically significant change from baseline was observed in either bifidobacteria or lactobacilli–enterococci levels at study endpoint (Fig. 1, Table 1). However, there was an indication that bifidobacteria levels were markedly increased in two subjects who recorded very low concentrations of bifidobacteria at baseline (3.6% and 6.7% at baseline, increased by 54.8% and 27.9%, respectively, (Table 2)). One of Fig. 1. Changes in fecal bifidobacteria levels (median% of total bacteria, 95% CI) com- – these infants (subject 8) received solely standard infant formula with- pared with typical concentrations found in healthy, breast-fed infants [26,29 31]. out prebiotics prior to the study (Table 2). Group mean lactobacilli–enterococci levels were reduced through- Results confirm that this formula was well tolerated and maintained phe- out the study (Table 1), but the overall values were likely to be signif- nylalanine control. The safety of adding prebiotics (scGOS/lcFOS [9:1 icantly influenced by one infant who presented with high lactobacilli– ratio]) to infant formulae has been well documented in previous clinical enterococci levels at baseline (25.0%). Without this outlier, mean studies [12–15,30,31]. The results from this study confirmed these find- group values would have been significantly lower both at study initia- ings, with no adverse events or gastrointestinal complications (e.g. diar- tion and at study end (group mean at initiation of 3.64% including out- rhea) and low levels of bacterial groups containing potential pathogens lier versus 0.97% not including outlier and study end of 2.36% and (Clostridium histolyticum/lituseburense and enterobacteriaceae). 0.88% respectively). The total group median value did not significantly No significant changes in bifidobacteria or lactobacilli–enterococci change throughout the study period. concentrations were observed overall. The lack of statistical significance No infants showed abnormally high levels of Clostridium histolyticum/ was expected, as this was (a) an exploratory pilot study with a small lituseburense or enterobacteriaceae containing potential pathogens [29]) population and therefore lacked the statistical power to detect such dif- at any point during the study. The infant who recorded the highest con- ferences, and (b) only one infant was on a prebiotic-free diet on entry centrations of Clostridium histolyticum/lituseburense and enterobacteria- into the study. Despite this, comparisons with historical data show bifi- ceae at baseline (9.2% and 1.05%, respectively) was receiving infant dobacteria and lactobacilli–enterococci levels at the end of the study to formula without prebiotics prior to the study, and showed a marked de- be similar to those reported for healthy breast-fed infants (60–75% crease in both groups at Week 8 (to 2.8% and b0.1% [below detection and 0.8–4%, respectively) [29,32–34], and greater than those reported limit], Table 2) accompanied by a marked increase in bifidobacteria for infants on infant formula without prebiotics (48.0%–53.4% and (27.9%, Table 2). 0.4%) [30,31]. This, in turn, may be advantageous by reducing the risk of infection [3,4,23–25]. 3.2. Stool characteristics It is interesting to note that the single infant that did not receive a prebiotic-containing diet prior to the study reported the greatest de- Asignificant reduction in median stool pH versus baseline was ob- crease in the group of Clostridium histolyticum/lituseburense (including served at Week 4 (p=0.002; Table 1, Fig. 2), but this was not observed potentially pathogenic species such as C. perfringens and C. difficile) at Week 8. Overall analysis of the primary caregiver description of stool and a subset of enterobacteriaceae (E. coli, Shigella, Salmonella and Kleb- consistency, type and frequency showed that moderate/normal sized siella), together with the second-greatest increase in bifidobacteria ‘fluffy’ stools were maintained throughout the study, but that stool con- levels. This is in line with previously published reports [16,23]. Howev- sistency improved. Mean categorical variable analysis showed an over- er, as this is a single patient observation, trends cannot be surmised and all transition from semi-liquid stools at baseline to soft/unformed stools the results require confirmation in further studies. at Week 8. This effect occurred in four of the nine infants. Typically, the fecal pH of breast-fed infants is more acidic than that of formula-fed infants (5.5 compared with 7.5), which is associated 4. Discussion with reduced growth of intestinal pathogens [35]. Stool pH was significantly reduced during the study at week 4, which is consistent with This is the first study to investigate an infant protein substitute findings from previous studies [12,13] but this effect is diminished at containing prebiotic oligosaccharides (scGOS/lcFOS [9:1 ratio]). week 8. However, the reduction of pH does not reflect the results of

Table 1 Primary and secondaryFOR assessments for theELECTRONIC total study population (median and mean value±SD, n=9) by studyUSE visit. ONLY Assessment Week 0 (study formula Week 4 Week 8 not yet commenced)

Median Mean Median Mean Median Mean

Blood phenylalanine, μmol/l 174.5 187.6±102.4 205 195.6±120.0 185 205.8 ±151.6 Blood tyrosine, μmol/l 85 90.7±51.3 102 102.5±30.9 97.5 96.4±36.1 Biﬁdobacteria,% 68.23 53.67±29.23 64.82 66.54±13.56 58.97 61.81 ±18.30 Lactobacilli–enterococci,% 0.99 3.64±8.02 1.09 1.79±2.72 0.93 1.64±2.64 Clostridium histolyticum/lituseburense,% 0.54 1.52±2.91 0.49 1.14±1.75 0.47 0.74±0.86 Enterobacteriaceae,% 0.58 0.61±0.30 0.93 0.91±0.68 0.46 0.54±0.30 Bacteroides distasonis/fragilis,%** b0.1 b0.1 b0.1 b0.1 b0.1 b0.1 E. rectale group 5.10 7.36±6.08 2.91 4.82±4.70 2.51 4.52±4.52 Stool pH 6.76 6.79±0.55 5.63* 5.83±0.98* 6.68 6.61±0.92

ITT population; *pb0.05 versus baseline; **All measurements below the detection limit (0.1% of total bacterial count). S58 A. MacDonald et al. / Molecular Genetics and Metabolism 104 (2011) S55–S59

Table 2 Changes from baseline at study endpoint (Week 8) for bacterial assessments in each individual.

Study Pre-study feedinga Biﬁdobacteria Lactobacilli–enterococci Clostridium histolyticum/lituseburense Enterobacteriaceae subset E. rectale/C. coccoides group subject (%) (%) group (%) (%) (%)

1 Formula −1.16 +0.47 −0.22 −0.66 −5.22 2 Breast-fed +formula −6.20 −0.37 0 −0.14 +2.35 3 Formula −5.48 +0.08 −0.02 +0.35 +2.26 4 Breast-fed +54.78 −16.42 +0.29 +0.48 −10.30 5 Breast-fed +formula +14.13 −0.25 −0.53 −0.07 −8.69 6 Breast-fed +13.04 +0.11 −1.04 +0.13 −1.77 7 Breast fed +formula −29.87 −0.7 −0.14 −0.56 −3.62 8 Formula (no prebiotic) +27.91 −0.66 −6.43 −0.95 −0.84 9 Breast-fed +6.11 −0.23 +0.34 −0.08 +0.20 Mean +8.14 −1.99 −0.78 −0.07 −2.85

a Formula represents standard infant formula containing prebiotics, unless indicated. the other study parameters (bifidobacteria and lactobacilli–enterococci intestinal microbiota cannot be concluded from such a sample size. concentrations) and so conclusions based on this result alone may not These issues should be addressed in larger, controlled studies to con- be drawn; it is possible that the differences between pre-study formula firm the hypotheses generated by this exploratory study. intakes and variations in baseline microbiota levels may have influenced this outcome. Stool consistency also appeared to improve during 5. Conclusions the study. This is unlikely to be due to the introduction of solids, as only fi one infant was given solids and this was during the nal week of the The exploratory data from this pilot study in infants with PKU provide study. However, in neonates stool consistency would be expected to evidence that PKU Anamix Infant (containing prebiotics scGOS/lcFOS change naturally over time. [9:1 ratio]) was well tolerated and supported good phenylalanine con- There are a number of limitations to this study. This was an ex- trol. This formula may help maintain levels of bifidobacteria and lower ploratory study only, with a small population, but this was con- stool pH. The inclusion of this scGOS/lcFOS (9:1 ratio) prebiotic blend strained by the numbers of infants diagnosed with PKU, and due to in infant formulae for non-PKU infants may contribute to the develop- the nature of the diagnosis and young age of infants at recruitment ment of an effective defence system against infection [20,24].Theseben- it restricted imposing highly selective entry criteria. In this trial, it efits would be particularly welcome in infants with PKU, and this initial would have been ethically unacceptable to include a control group exploratory data indicates that further investigation is warranted in trials of infants who had deliberately avoided breast milk to ensure a with larger numbers, to fully evaluate the health benefits of prebiotics in fl prebiotic-free diet. The ability to in uence pre-feeding regimens for an infant protein substitute. this study group was also not possible. Infants with PKU have commenced feeding either using breast milk or standard infant formula fl for a few days at least, before the diagnosis of PKU is confirmed. Con ict of interest Therefore, play-of-chance may have influenced the overall outcomes, particularly considering the variations in pre-study feeding regimens. The study, data collection and data analysis were sponsored by In addition, although the number of breast feeds remained relatively Nutricia Advanced Medical Nutrition. Alpha-Plus Medical Communi- constant throughout the study, it was reported through “time spent cations Ltd provided editorial support with the production of the on breast” which unfortunately does not provide exact amounts of manuscript, which was supported by Nutricia. breast milk consumed. Ideally, larger studies in this area would en- deavor to use more accurate measures of expressed breast milk to “What is already known on this topic?” provide more definitive results for interpretation. In addition, the open-label nature of this study limited interpretation of the results • to historical comparisons. Finally, the amount of breast milk tolerated Human milk oligosaccharides are known to help promote the defi fi varied from individual to individual, impacting on the intake of prebi- velopment of bene cial microbiota (e.g. bi dobacteria) in infants. • otics during the study. This also introduces a variable that can sub- Clinical studies have shown that supplementation of standard in- fi stantially affect the outcomes in such a small population, and fant formula with a speci c mixture of prebiotic oligosaccharides fi whether infants with PKU are at risk of developing a less healthy can provide similar bene ts. • Prebiotics are now included in most standard infant formulae, but have not been evaluated in special formulae for amino acid disorders such as PKU.

FOR ELECTRONIC“What this study USE adds” ONLY

• This is the ﬁrst study investigating the use of an infant phenylalanine- free protein substitute containing prebiotics (scGOS/lcFOS [9:1 ratio]) in PKU. • The infant phenylalanine-free protein substitute containing prebiotics (scGOS/lcFOS [9:1 ratio]) was well tolerated and maintained phenylalanine control. • The infant phenylalanine-free protein substitute containing prebiotics (scGOS/lcFOS [9:1 ratio]) maintained an intestinal microbiota comparable to breast-fed infants, which might confer health-related beneﬁts Fig. 2. Changes in fecal pH (median, 95% CI). and support metabolic control. A. MacDonald et al. / Molecular Genetics and Metabolism 104 (2011) S55–S59 S59

References [20] E. Bruzzese, M. Volpicelli, F. Salvini, et al., Early administration of GOS/FOS prevents intestinal and respiratory infections, J. Pediatr. Gastroenterol. Nutr. 42 [1] L. McCabe, A.E. Ernest, M.R. Neifert, et al., The management of breast feeding among (2006) E95. infants with phenylketonuria, J. Inherit. Metab. Dis. 12 (4) (1989) 467–474. [21] S. Arslanoglu, G.E. Moro, G. Boehm, Early supplementation of prebiotic oligosac- [2] WHO, Breastfeeding Last accessed on 26 November, 2010 at, http://www.who. charides protects formula-fed infants against infections during the first 6 months int/topics/breastfeeding/en/cited; Available from:. of life, J. Nutr. 137 (11) (Nov 2007) 2420–2424. [3] P.F. Chien, P.W. Howie, Breast milk and the risk of opportunistic infection in infancy [22] M.E. Bongers, F. de Lorijn, J.B. Reitsma, et al., The clinical effect of a new infant for- in industrialized and non-industrialized settings, Adv. Nutr. Res. 10 (2001) 69–104. mula in term infants with constipation: a double-blind, randomized cross-over [4] P.W. Howie, J.S. Forsyth, S.A. Ogston, et al., Protective effect of breast feeding trial, Nutr. J. 6 (2007) 8. against infection, BMJ 300 (6716) (Jan 6 1990) 11–16. [23] J. Knol, G. Boehm, M. Lidestri, et al., Increase of faecal bifidobacteria due to dietary [5] L. Bode, Recent advances on structure, metabolism, and function of human milk oligosaccharides induces a reduction of clinically relevant pathogen germs in the oligosaccharides, J. Nutr. 136 (8) (Aug 2006) 2127–2130. faeces of formula-fed preterm infants, Acta Paediatr. Suppl. 94 (449) (Oct 2005) [6] C. Kunz, S. Rudloff, W. Baier, et al., Oligosaccharides in human milk: structural, 31–33. functional, and metabolic aspects, Annu. Rev. Nutr. 20 (2000) 699–722. [24] G. Boehm, J. Jelinek, J. Knol, et al., Prebiotics and immune responses, J. Pediatr. [7] ISAPP, 6th Meeting of the International Scientific Association of Probiotics and Gastroenterol. Nutr. 39 (Suppl 3) (Jun 2004) S772–S773. Prebiotics, 2008, London, Ontario. [25] G. Boehm, J. Jelinek, B. Stahl, et al., Prebiotics in infant formulas, J. Clin. Gastroen- [8] M. Roberfroid, G.R. Gibson, L. Hoyles, et al., Prebiotic effects: metabolic and health terol. 38 (6 Suppl) (Jul 2004) S76–S79. benefits, Br. J. Nutr. 104 (Suppl 2) (Aug 2010) S1–S63. [26] E.A. Mevissen-Verhage, J.H. Marcelis, M.N. de Vos, W.C. Harmsen-van Ameron- [9] R. Fuller, Probiotics in man and animals, J. Appl. Bacteriol. 66 (1989) 365–378. gen, J. Verhoef, Bifidobacterium, Bacteroides, and Clostridium spp. in fecal sam- [10] G.R. Gibson, M.B. Roberfroid, Dietary modulation of the human colonic micro- ples from breast-fed and bottle-fed infants with and without iron supplement, biota: introducing the concept of prebiotics, J. Nutr. 125 (1995) 1401–1412. J. Clin. Microbiol. 25 (1987) 285–289. [11] G.R. Gibson, H.M. Probert, J.V. Loo, R.A. Rastall, M.B. Roberfroid, Dietary modula- [27] J. Knol, P. Scholtens, C. Kafka, et al., Colon microflora in infants fed formula with tion of the human colonic microbiota: updating the concept of prebiotics, Nutr. galacto- and fructo-oligosaccharides: more like breast-fed infants, J. Pediatr. Gas- Res. Rev. 17 (2004) 259–275. troenterol. Nutr. 40 (1) (Jan 2005) 36–42. [12] G. Moro, I. Minoli, M. Mosca, et al., Dosage-related bifidogenic effects of galacto- [28] S.J. Lewis, K.W. Heaton, Stool form scale as a useful guide to intestinal transit time, and fructooligosaccharides in formula-fed term infants, J. Pediatr. Gastroenterol. Scand. J. Gastroenterol. 32 (9) (Sep 1997) 920–924. Nutr. 34 (3) (Mar 2002) 291–295. [29] H.J. Harmsen, A.C. Wildeboer-Veloo, G.C. Raangs, et al., Analysis of intestinal flora [13] A.M. Bakker-Zierikzee, M.S. Alles, J. Knol, et al., Effects of infant formula contain- development in breast-fed and formula-fed infants by using molecular identifica- ing a mixture of galacto- and fructo-oligosaccharides or viable Bifidobacterium tion and detection methods, J. Pediatr. Gastroenterol. Nutr. 30 (1) (Jan 2000) animalis on the intestinal microflora during the first 4 months of life, Br. J. Nutr. 61–67. 94 (5) (Nov 2005) 783–790. [30] H. Schmelzle, S. Wirth, H. Skopnik, et al., Randomized double-blind study of the [14] G.E. Moro, B. Stahl, S. Fanaro, et al., Dietary prebiotic oligosaccharides are detect- nutritional efficacy and bifidogenicity of a new infant formula containing partially able in the faeces of formula-fed infants, Acta Paediatr. Suppl. 94 (449) (Oct hydrolyzed protein, a high beta-palmitic acid level, and nondigestible oligosac- 2005) 27–30. charides, J. Pediatr. Gastroenterol. Nutr. 36 (3) (Mar 2003) 343–351. [15] G.E. Moro, F. Mosca, V. Miniello, et al., Effects of a new mixture of prebiotics on faecal [31] G. Boehm, G. Moro, Structural and functional aspects of prebiotics used in infant flora and stools in term infants, Acta Paediatr. Suppl. 91 (441) (Sep 2003) 77–79. nutrition, J. Nutr. 138 (9) (Sep 2008) 1818S–1828S. [16] G. Boehm, B. Stahl, J. Jelinek, et al., Prebiotic carbohydrates in human milk and [32] R.I. Mackie, A. Sghir, H.R. Gaskins, Developmental microbial ecology of the neona- formulas, Acta Paediatr. Suppl. 94 (449) (Oct 2005) 18–21. tal gastrointestinal tract, Am. J. Clin. Nutr. 69 (5) (May 1999) 1035S–1045S. [17] S. Fanaro, G. Boehm, J. Garssen, et al., Galacto-oligosaccharides and long-chain [33] M. Haarman, J. Knol, Quantitative real-time PCR assays to identify and quantify fructo-oligosaccharides as prebiotics in infant formulas: a review, Acta Paediatr. fecal Bifidobacterium species in infants receiving a prebiotic infant formula, Suppl. 94 (449) (Oct 2005) 22–26. Appl. Environ. Microbiol. 71 (5) (May 2005) 2318–2324. [18] P.A. Scholtens, P. Alliet, M. Raes, et al., Fecal secretory immunoglobulin A is increased [34] M. Haarman, J. Knol, Quantitative real-time PCR analysis of fecal Lactobacillus in healthy infants who receive a formula with short-chain galacto-oligosaccharides species in infants receiving a prebiotic infant formula, Appl. Environ. Microbiol. and long-chain fructo-oligosaccharides, J. Nutr. 138 (6) (Jun 2008) 1141–1147. 72 (4) (Apr 2006) 2359–2365. [19] G. Puccio, C. Cajozzo, F. Meli, et al., Clinical evaluation of a new starter formula for [35] C. Van Limpt, A. Crienen, A. Vriesema, et al., Effect of colonic short chain fatty infants containing live Bifidobacterium longum BL999 and prebiotics, Nutrition 23 acids, lactate and pH on the growth of common gut pathogens, Pediatr. Res. 56 (1) (Jan 2007) 1–8. (2004) 487.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S60–S63

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

The 48-hour tetrahydrobiopterin loading test in patients with phenylketonuria: Evaluation of protocol and inﬂuence of baseline phenylalanine concentration☆

K. Anjema a,⁎, G. Venema a, F.C. Hofstede b, E.C. Carbasius Weber b, A.M. Bosch c, N.M. Ter Horst c, C.E.M. Hollak c, C.F. Jonkers c, M.E. Rubio–Gozalbo d, E.M.C. van der Ploeg d, M.C. de Vries e, R.G. Janssen-Regelink e, M.C.H. Janssen e, H. Zweers-van Essen e, C.C.A. Boelen f, N.A.P. van der Herberg-van de Wetering f, M.R. Heiner-Fokkema a, M. van Rijn a, F.J. van Spronsen a a Beatrix Children's Hospital, University Medical Centre Groningen, The Netherlands b Wilhelmina Children's Hospital, University Medical Centre Utrecht, The Netherlands c Academic Medical Centre, Amsterdam, The Netherlands d Maastricht University Medical Centre, The Netherlands e Radboud University Nijmegen Medical Centre, The Netherlands f Leiden University Medical Centre, The Netherlands article info abstract

Article history: Background: The 24- and 48-hour tetrahydrobiopterin (BH4) loading test (BLT) performed at a minimum Received 31 July 2011 baseline phenylalanine concentration of 400 μmol/l is commonly used to test phenylketonuria patients for Received in revised form 17 September 2011 BH4 responsiveness. This study aimed to analyze differences between the 24- and 48-hour BLT and the Accepted 17 September 2011 necessity of the 400 μmol/l minimum baseline phenylalanine concentration. Available online 23 September 2011 Methods: Data on 186 phenylketonuria patients were collected. Patients were supplemented with phenylalanine if phenylalanine was b400 μmol/l. BH4 20 mg/kg was administered at T=0 and T=24. Blood samples were Keywords: taken at T=0, 8, 16, 24 and 48 h. Responsiveness was defined as ≥30% reduction in phenylalanine concentration Phenylketonuria ≥ PKU at 1timepoint. Tetrahydrobiopterin Results: Eighty-six (46.2%) patients were responsive. Among responders 84% showed a ≥30% response at BH4 T=48. Fifty-three percent had their maximal decrease at T=48. Fourteen patients had ≥30% phenylalanine Sapropterin decrease not before T=48. A ≥30% decrease was also seen in patients with phenylalanine concentrations Phenylalanine b400 μmol/l. Conclusion: In the 48-hour BLT, T=48 seems more informative than T=24. Sampling at T=32, and T=40 may have additional value. BH4 responsiveness can also be predicted with baseline blood phenylalanine b400 μmol/l, when the BLT is positive. Therefore, if these results are confirmed by data on long-term BH4 responsiveness, we advise to first perform a BLT without phenylalanine loading and re-test at higher phenylalanine concentrations when no response is seen. Most likely, the 48-hour BLT is a good indicator for BH4 responsiveness, but comparison with long term responsiveness is necessary. © 2011 Elsevier Inc. All rights reserved.

1. Introduction the treatment is highly successful in preventing neurological damage, it is a major burden for patients and caregivers. The cornerstone of treatment of Phenylketonuria (PKU; MIM Recently, tetrahydrobiopterin (BH4) was introduced as a new treat- 261600) is the life-long dietary restriction of phenylalanine (Phe) ment option, but only for BH4-responsive patients. Usually, these are [1]. Dietary treatmentFOR lowers the bloodELECTRONIC Phe concentration. Although patients with a milder USE PAH deficiency [2]ONLY, although BH4 responsiveness has also been reported in a few patients with more severe PAH deficiencies [3]. Unfortunately, determining BH4 responsiveness is not explicit. Abbreviations: BH4, Tetrahydrobiopterin; BLT, BH4-loading test; Phe, Phenylalanine; PAH, Phenylalanine-hydroxylase; PKU, Phenylketonuria. Genotyping can be useful [4], but it tends to overestimate BH4 respon- ☆ Financial disclosures: Dr. FJ van Spronsen is a member of the scientific advisory siveness [5]. Selection based on either Phe concentrations and/or toler- board for Merck Serono and Danone. He has received several grants for scientific pro- ance in day to day practice is also unreliable [6]. Therefore, BH4 loading jects from both Merck Serono and Danone. tests are used to discriminate between responders and non-responders. ⁎ Corresponding author at: Department of metabolic disorders, Beatrix Children's Different protocols of the BH4 loading test exist around the world. In Hospital, University Medical Centre Groningen, CA80, PO Box 30.001, 9700 RB Groningen, “ The Netherlands. Europe, the 48-hour test, developed by the European working group E-mail address: [email protected] (K. Anjema). for Phenylketonuria”, is commonly used. In this protocol a minimum

Phe level of 400 μmol/l is essential. A ≥30% decrease of the Phe Table 1 concentration compared to the baseline Phe concentration suggests Demographic and clinical details of the b30% and ≥30% response group. long-term BH4 responsiveness [7], but data on long-term response b30% response ≥30% response need to prove the results. N=100 N=86 P BH4 (Sapropterin) was approved by the EMEA for PKU patients Gender M/F 47/53 39/47 0.822 from four years of age, at the end of 2008 [8]. The Netherlands was – – fi Age at BH4 loading 14.7 (0.5 46.0) 12.1 (3.6 35.2) 0.013 one of the rst countries in Europe where BH4 (Sapropterin) became Baseline Phe (μmol/l)* 659 (203–1544) 479 (203–1181) 0.000 available. We took the opportunity to test for BH4 responsiveness Mean Phe (μmol/l)** using a standardized national protocol based largely on the recom- b12 years 397 (137–920) 302 (142–566) 0.045 mendations of the European working group [7]. We would like to ≥12 years 564 (159–1466) 336 (224–759) 0.000 Phe at diagnosis (μmol/l)† 0.000 address two important questions to optimize the BH4 loading test 1. b600 3 34 What are the differences in the results between the 24- and 48-hour 600–1200 14 33 BH4 loading test? 2. Does the BH4 loading test require a minimum N1200 66 18 μ Phe level of 400 mol/l at baseline? *Phe concentration at T=0 of the BH4 loading test. **Mean Phe concentration in one year prior to the BH4 loading test (min. 4 samples), value missing in N=15 and 2. Methods N=13 respectively. †First Phe concentration in hospital, missing in N=17 and N=1, respectively. P-values in bold are ≤0.05. 2.1. Patients of 94.5%. In Fig. 1 the prevalence of response rates is shown in pa- Subjects were all PAH-deficient patients who were treated with a tients with a ≥30% response. protein restricted diet and supplementation of amino acids, and also underwent the 48-hour BH4 loading test as standard patient care. 3.1. Duration of the test Except for ten patients all of the others were above the age of 4 years. Data included are from patients from six of the eight University Medical In patients responding to BH4, Fig. 2A shows the percentages of Centers in the Netherlands that started to use the national 48-hour BH4 patients with a ≥30% Phe decrease at the particular time points. loading test protocol. The medical ethical committee of the University When only one time point (either T=8, T=16, T=24 or T=48) Medical Center of Groningen concluded that their approval was not was analyzed versus baseline, 48 (56%), 31 (36%), 28 (33%) and 14 required, since the 48-hour BH4 loading test is performed as standard (16%) patients would have been missed, respectively. Most patients patient care. (53.5%) had their maximal individual decrease at T=48 (Fig. 2B). Fourteen patients (16%) had a ≥30% response at T=48 and none be- 2.2. Procedure fore, they were regarded to as slow-responders. Table 2 presents the demographic and clinical details of normal responders versus slow- All participating patients with plasma Phe concentrations below responders. Additionally, three patients showed a response at T=8 400 μmol/l where supplemented with Phe (L-Phe in powder, a protein and T=16, but not subsequently. rich supplement such as milk powder or an increase in natural protein intake) until the end of the test. The test consisted of two doses of 3.2. Baseline Phe concentration and Phe response 20 mg/kg body weight BH4 at T=0 and T=24 hours just after blood samples were taken. Blood samples on filter paper were taken at Before starting the BH4 loading test, 50% of all patients had Phe T=0, 8, 16, 24 and 48 h. Patients who (later) appeared to have devel- concentrations b400 μmol/l. All 92 patients were loaded with Phe oped a fever during the test or during the preceding day as well (median 275 mg Phe, range 40–2000 mg). Despite Phe loading, 32% as patients who missed a dose of BH4 were excluded. Responsiveness of these patients had a Phe concentration b400 μmol/l at the time of was defined as a 30% or more reduction in Phe concentration at 1 or the test. Out of the patients who were not loaded with Phe, 11% of more moment(s) compared to baseline (T=0). 94 patients had a Phe concentration b400 μmol/l at baseline. Blood Phe concentrations from dried blood specimens were mea- Fig. 3 shows that a decrease of ≥30% is not restricted to patients sured according to current quantitative methods used in the various with plasma Phe concentrations above 400 μmol/l. Seventy-four per- centers. These included methods such as high-performance liquid cent of patients with a baseline Phe concentration b400 μmol/l (both chromatography with fluorescence and tandem mass spectrometric detection.

2.3. Statistical analysis

All descriptive statistics were shown as medians with ranges. For comparing categorical data the chi square test was used. For continuous data the MannFOR–Whitney U test wasELECTRONIC used. A p-value≤0.05 was consid- USE ONLY ered to be statistically signiﬁcant. Statistics were performed with PASW statistics 18.0; SPSS, Inc., Chicago, IL, USA.

3. Results

BH4 loading test data of 186 patients were collected over a period of 17 months following the introduction of Sapropterin in the Nether- lands. The median age of the patients was 13.5 years (range 0.5– 46.0 years). Eighty-six out of 186 patients (46.2%) had a ≥30% Phe decrease compared to T=0. Demographic and clinical details of the patients with b30% and ≥30% response are shown in Table 1. The Fig. 1. Distribution of maximal blood Phe reductions, in 86 patients responding to BH4 range of responses on BH4 included an increase of 27.3% to a decrease loading (20 mg/kg, twice) during the 48-hour BH4 loading test. S62 K. Anjema et al. / Molecular Genetics and Metabolism 104 (2011) S60–S63

Fig. 3. Maximal reduction in blood Phe concentration (%) in response to BH4 in the 48-hour BH4 loading test compared to the blood Phe concentration at the start of the loading test (T = 0).

also shows that the higher the Phe concentration, the lower the chance of BH4 responsiveness.

4. Discussion

This is one of the first papers addressing issues that are important to establish the predictive value of the BH4 loading test [3,6,9–11]. Many aspects deserve further attention to optimize the 24 and 48- hour BH4 loading test. The most important findings of the present study were that the second half of the 48-hour BH4 loading test seems more informative than the first 24 h, and that 400 μmol/l is at least not a necessary prerequisite in all patients to adequately test BH4 responsiveness. As expected, sampling moments later than 24 h seem to be impor- Fig. 2. A. Percentages of responders with ≥30% Phe concentration reduction at specific tant. We found a considerable number of patients who responded to time points. B. Moment of maximal reduction of blood Phe concentration (%) in patients BH4 at 48 hours rather than before (Fig. 2A), and therefore the 24- responding to BH4 loading of (20 mg/kg, twice) during the 48-hour BH4 loading test. hour test is likely to miss patients that could benefit from BH4. This is in line with the report of Fiege et al., showing that 48 h seem to Phe-loaded and not Phe-loaded patients) showed a ≥30% response. be useful to detect BH4-responsiveness in more severe phenotypes Baseline Phe concentrations of patients with a positive response ranged and “slow responders” [9]. Furthermore, in most patients the largest from 203 to 1181 μmol/l. The 203 μmol/l baseline Phe level in one response is seen at T=48. It has to be taken into account that 48 h patient was due to a combination of mild PKU and strict treatment might not be long enough to detect even slower responders [12]. Nev- as can be learned from the pre-treatment Phe level (740 μmol/l at ertheless, it has to be considered that an early response might be a day 10) and the present Phe intake of 450 mg (+500 mg extra predicting factor for the long-term response. However, it would be Phe-supplemented during the test) at the age of 14 years. Fig. 3 fair to expect that in most of these patients, later samples would also show a decrease of ≥30%. It would be interesting to see whether samples during the second half of the 48-hour loading test (e.g. Table 2 Demographic and clinical details of patients responding to BH4 during the first 24 h T=32 and T=40) would be more informative than earlier blood versus patients responding at T=48 (slow-responders). samples. This would also select fast responders/metabolizers and therefore possibly the samples at T=8 and T=16 could be left out 24 h-responders Slow-responders as the test would be easier for patients and parents if the number of FORN=72 ELECTRONIC N=14 P sampling moments USE could be reduced, ONLY especially during nighttime. Gender M/F 32/40 7/7 0.702 However, it is not advisable to leave out the T=24-hour sample. Age at BH4 loading 11.8 (3.6–35.2) 15.3 (5.5–28.6) 0.094 In the neonatal BH4 loading test, it has been considered that tests Baseline Phe (μmol/l)* 461 (203–969) 540 (352–1181) 0.017 performed with baseline Phe concentrations b400 μmol/l were inad- Mean Phe (μmol/l)** b12 years 298 (142–539) 462 (276–566)† 0.048 equate to interpret BH4 responsiveness [13]. Hence, Phe supplementa- ≥12 years§ 323 (224–759) 414 (233–617)‡ 0.725 tion in case of Phe concentrations b400 μmol/l was shown mandatory. Phe at diagnosis (μmol/l)§§ - However, data of the present study show that BH4 responsiveness can b 600 33 1 be observed in patients N4 years of age, when baseline Phe is – 600 1200 26 7 b μ N1200 12 6 400 mol/l, as already reported elsewhere [6,9,10].Thisisinline with the study by Staudigl et al. [14], showing that some mutations *Phe concentration at T=0 of the BH4 loading test. **Mean Phe concentration in one appear BH4 responsive, especially at lower Phe concentrations. In year prior to the BH4 loading test (min. 4 samples). §Value missing in N=8 and N=5 respectively. §§First Phe concentration in hospital, missing in N=1 and N=0, contrast, Staudigl's paper also shows that in some mutations BH4 re- respectively. †N=4 patients. ‡N=5 patients. P-values in bold are ≤0.05. sponsiveness is observed in case of higher Phe concentrations. K. Anjema et al. / Molecular Genetics and Metabolism 104 (2011) S60–S63 S63

Our data may change practice to load every patient with Phe be- Acknowledgments fore performing a BH4 loading test. Knowing the impact of this change, future studies have to confirm our data before such a change We are very thankful to all the patients and families who partici- can become practice. pated in this study. Also we would like to thank the analysts of the Some aspects need to be addressed when the method and results metabolic laboratories for analyzing the phenylalanine samples. of this study are considered. First of all, each population has its own genetic background. Therefore, when results from different papers References are discussed, the PKU genotype of each population should be taken into account. Second, quite unexpectedly, a large number of patients [1] N. Blau, F.J. van Spronsen, H.L. Levy, Phenylketonuria, Lancet 376 (2010) 1417–1427. b μ [2] S. Kure, D.C. Hou, T. Ohura, H. Iwamoto, S. Suzuki, N. Sugiyama, O. Sakamoto, K. had Phe concentrations 400 mol/l at baseline, despite the Phe supple- Fujii, Y. Matsubara, K. Narisawa, Tetrahydrobiopterin-responsive phenylalanine mentation and the monitoring of Phe levels during the period of Phe hydroxylase deficiency, J. Pediatr. 135 (1999) 375–378. loading. Due to the procedure, patients completed the BH4 loading [3] B. Fiege, N. Blau, Assessment of tetrahydrobiopterin (BH4) responsiveness in phenylketonuria, J. Pediatr. 150 (2007) 627–630. test before baseline Phe concentrations were known. Third, for con- [4] M.R. Zurfluh, J. Zschocke, M. Lindner, F. Feillet, C. Chery, A. Burlina, R.C. Stevens, B. venience of the patients and their families and financial reasons, the Thony, N. Blau, Molecular genetics of tetrahydrobiopterin-responsive phenylala- T =32 and T = 40 blood samples were not included in our protocol. nine hydroxylase deficiency, Hum. Mutat. 29 (2008) 167–175. [5] I. Karacic, D. Meili, V. Sarnavka, C. Heintz, B. Thony, D.P. Ramadza, K. Fumic, D. As discussed above, it probably would have been better to incorpo- Mardesic, I. Baric, N. Blau, Genotype-predicted tetrahydrobiopterin (BH4)- rate additional blood sampling at T = 32 and T =40 as advised by responsiveness and molecular genetics in Croatian patients with phenylalanine hy- the European working group on PKU [7]. Additionally, the ultimate droxylase (PAH) deficiency, Mol. Genet. Metab. 97 (2009) 165–171. test of BH4 responsiveness is the long-term experience in patients, [6] V. Leuzzi, C. Carducci, C. Carducci, F. Chiarotti, C. Artiola, T. Giovanniello, I. Antonozzi, The spectrum of phenylalanine variations under tetrahydrobiopterin load in sub- showing adequate increase in Phe tolerance and/or adequate de- jects affected by phenylalanine hydroxylase deficiency, J. Inherit. Metab. Dis. 29 crease in plasma Phe concentration. Due to the limited time for (2006) 38–46. follow-up we are currently collecting data of the patients who con- [7] N. Blau, A. Belanger-Quintana, M. Demirkol, F. Feillet, M. Giovannini, A. MacDonald, F.K. Trefz, F.J. van Spronsen, Optimizing the use of sapropterin (BH(4)) in the man- tinued BH4 treatment after a positive BH4 loading test. agement of phenylketonuria, Mol. Genet. Metab. 96 (2009) 158–163. [8] Sapropterin dihydrochloride (Kuvan), European summary of product characteristics. Available at, http://www.medicines.org.uk/EMC/medicine/21362/SPC/ Kuvan+100+mg+soluble+tablets/. 5. Conclusion [9] B. Fiege, L. Bonafe, D. Ballhausen, M. Baumgartner, B. Thony, D. Meili, L. Fiori, M. Giovannini, N. Blau, Extended tetrahydrobiopterin loading test in the diagnosis A follow-up of 48 h in the BH4 loading test seems to predict the of cofactor-responsive phenylketonuria: a pilot study, Mol. Genet. Metab. 86 (Suppl 1) (2005) S91–S95. BH4 responsiveness more reliably when compared to 24 h. To further [10] J.J. Mitchell, B. Wilcken, I. Alexander, C. Ellaway, H. O'Grady, V. Wiley, J. Earl, J. improve the 48-hour BH4 loading test we suggest to include addition- Christodoulou, Tetrahydrobiopterin-responsive phenylketonuria: the New al blood samples between T=24 and T=48 h (e.g. T=32 and South Wales experience, Mol. Genet. Metab. 86 (Suppl 1) (2005) S81–S85. [11] M. Lindner, G. Gramer, S.F. Garbade, P. Burgard, Blood phenylalanine concentra- T=40). This will improve the knowledge on BH4 response in the sec- tions in patients with PAH-deficient hyperphenylalaninaemia off diet without ond 24 h of the BH4 loading test. and with three different single oral doses of tetrahydrobiopterin: assessing re- Furthermore, it seems that BH4 responsiveness can be predicted sponsiveness in a model of statistical process control, J. Inherit. Metab. Dis. 32 – by BH4 loading test even when the baseline blood Phe concentration (2009) 514 522. [12] J.B. Nielsen, K.E. Nielsen, F. Guttler, Tetrahydrobiopterin responsiveness after ex- is b400 μmol/l. If these results are confirmed by other data and data tended loading test of 12 Danish PKU patients with the Y414C mutation, J. Inherit. on long-term BH4 responsiveness, it could be advised to first perform Metab. Dis. 33 (2010) 9–16. a BH4 loading test without Phe loading and re-test at higher Phe con- [13] N. Blau, L. Bonafé, M.E. Blaskovics, Disorders of phenylalanine and tetrahydrobiopterin, in: N. Blau, M. Duran, M. Blaskovics, K.M. Gibson (Eds.), Physician's Guide to centrations when no response is seen. the Laboratory Diagnosis of Metabolic Disease, Springer, Heidelberg, 2002, Probably the BH4 loading test is a good indicator for BH4 respon- pp. 89–106. siveness, but a comparison with long term treatment results is neces- [14] M. Staudigl, S.W. Gersting, M.K. Danecka, D.D. Messing, M. Woidy, D. Pinkas, K.F. Kemter, N. Blau, A.C. Muntau, The interplay between genotype, metabolic state sary to really answer this question. Other issues still need to be solved and cofactor treatment governs phenylalanine hydroxylase function and drug re- to optimize this testing procedure. sponse, Hum. Mol, Genet, 2011.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S64–S67

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Does a lower carbohydrate protein substitute impact on blood phenylalanine control, growth and appetite in children with PKU?

Hulya Gokmen-Ozel a, Carol Ferguson b, Sharon Evans a, Anne Daly a, Anita MacDonald a,⁎ a Birmingham Children's Hospital, Steelhouse Lane, Birmingham, B4 6NH, UK b Royal Victoria Inﬁrmary, Queen Victoria Road, Newcastle upon Tyne, NE1 4LP, UK article info abstract

Article history: Background: In children with phenylketonuria (PKU), it is possible that high carbohydrate protein substitutes Received 23 July 2011 may adversely affect blood phenylalanine control. We evaluated if a low carbohydrate, ‘ready-to-drink’ protein Received in revised form 10 September 2011 substitute would impact on short term blood phenylalanine control, weight and appetite in children with PKU Accepted 12 September 2011 aged 3–10 years. Available online 16 September 2011 Methods: This was a 3-part, 5-week randomised, controlled, crossover study in which two different carbohydrate/ protein-equivalent ratios in protein substitute [control protein substitute (CPS) median 1:1; trial protein substi- Keywords: tute (TPS) 0.5:1] were compared. The effects on feeding behaviour, weight change and phenylalanine concentra- Liquid protein substitutes Carbohydrate tions were studied. Fourteen children (12 boys; median age 6.3 y, range 3 to 9.7 y) with PKU on diet were Protein substitute recruited from 2 treatment centres. Phenylalanine Results: Phenylalanine control did not deteriorate with TPS and remained unchanged between pre-study and CPS Phenylketonuria (p=0.783). No statistical differences were noted in energy intake between the two study parts. Any changes in Appetite weight were similar between the two groups and there was limited change in feeding behaviour. Conclusion: This study suggests that the carbohydrate/protein-equivalent ratio of protein substitutes can be reduced to 0.5:1 with no loss of blood phenylalanine control or adverse effect on weight gain in children with PKU. © 2011 Elsevier Inc. All rights reserved.

1. Introduction substitutes have relatively high carbohydrate profiles with carbohydrate/protein-equivalent ratios of greater than 1:1. This was partly to In children with phenylketonuria (PKU), there is evidence that ad- assist with disguising the taste of L-amino acids, supplementing a low equate energy intake is important in maintaining satisfactory blood phenylalanine diet with additional energy, as well as promoting anabo- phenylalanine control [1]. Traditionally, most phenylalanine-free pro- lism. Adequate energy intake is associated with stimulation of net protein substitutes are supplemented with carbohydrate+/−fat and tein deposition [3], and in turn will help reduce the oxidation of they provide over 30% of energy intake in young children [2], but phenylalanine. Carbohydrate is considered more efficient than fat [4]. there is an increasing trend for newer protein substitutes with a One of the potential advantages of less carbohydrate added to pro- lower carbohydrate profile to be produced for PKU. The impact of tein substitute is that children may compensate for any reduction in this remains unstudied in children with PKU less than 10 years of age. carbohydrate intake by eating more energy from natural foods, and It is possible, that a reduction in the carbohydrate content of protein sub- this is probably preferable to patients. In turn, it is known that feeding stitutes could adversely affect blood phenylalanine control in children problems are common in young children with PKU [2]. It is suggested with PKU, but no published studies have explored this. that the energy content of protein substitutes may be a causative factor The ideal composition of protein substitute has not been defined for by suppressing appetite, and reducing energy intake from natural foods. children aged 3FOR–10 years with PKU ELECTRONIC and there are no European stan- Slowness to feed, poorUSE appetite, dislike ONLY of sweet foods and a limited va- dards to regulate macronutrient composition. Conventional protein riety of food intake are typical feeding issues. In the UK, low carbohydrate ‘ready-to-drink’ protein substitutes suitable for children with PKU from the age of 3 years have been available since 2008. The aim of this short-term randomised controlled, crossover study was to evaluate if a lower energy (low in car- Abbreviations: PKU, Phenylketonuria; CPS, Control protein substitute; TPS, Trial bohydrate and fat) ‘ready-to-drink’ protein substitute would improve protein substitute; UK, United Kingdom; MREC, The Multi-Centre Research Ethics appetite and energy intake from natural foods without loss of blood Committee. ⁎ Corresponding author. phenylalanine control or weight deterioration in a group of children E-mail address: [email protected] (A. MacDonald). aged 3 to 10 years of age with PKU.

2. Materials and methods Powder (n=1). The CPS protein substitutes were taken as a paste (n=12) or drink (n=2). 2.1. Subjects 2.4. Trial protein substitute (TPS) Fourteen children (13 Caucasian; 1 Asian origin) with PKU were recruited from 2 UK centres; Birmingham Children's Hospital The ‘ready-to-drink’, liquid low-carbohydrate, low-fat protein (n=12) and Newcastle General Infirmary (n=2). There were 12 substitute was ‘PKU Cooler’ (Vitaflo International) with a carbohy- boys and 2 girls, with a median age of 6.3 years (range 3 to drate content of 5.9 g/100 ml. All the children drank TPS directly 9.7 years). All subjects followed a low phenylalanine diet comprising: from the pouches. The nutritional composition is described else- (1) a dietary phenylalanine allocation using a 50 mg exchange system; where [6]. (2) a phenylalanine-free protein substitute; and (3) low-phenylalanine foods, with fruits and vegetables containing up to 75 mg/100 g permit- 2.5. Assessment of plasma phenylalanine concentrations ted in normal quantities. The median number of 50 mg phenylalanine exchanges was 5 per day (range 3 to 30 daily). This was equivalent to Once daily finger prick blood samples (fasting pre-breakfast at 5 g/day of natural protein. standardised times) were taken on day 5–7 of the pre-study period There were four inclusion criteria: 1) aged 3 to 10 years; 2) main- and on days 12–14 in part A and B. These daily bloods were collected tenance of plasma phenylalanine concentrations within recom- over 3 days for each trial period to allow for any variability in daily mended ranges for age group by MRC Working Group [5] for 70% of plasma phenylalanine concentrations. Blood phenylalanine specimen the six month period prior to entry to the study; 3) taking a protein collection and analysis has been described elsewhere [7]. substitute with added carbohydrate and 4) parental ability to take skin puncture blood specimens at home by thumb prick. Exclusion 2.6. Assessment of dietary intake criteria included subjects taking at baseline either a liquid protein substitute or a protein substitute with added fat. The Multi-Centre Re- In the pre-study period (days 5–7), and part A and B (days 12–14) search Ethics Committee (MREC) of West Midlands approved the maternal carers recorded food intake and quantity of protein substi- study. Written informed content was obtained from all caregivers tute consumed over 3 days. Nutritional analysis of food intake was and written assent from all children. calculated using the Microdiet computer programme based on McCance and Widdowson's ‘The Composition of Foods’ with supple- – 2.2. Study design mentary data provided by the Royal Society of Chemistry [8 12] and from food manufacturers. This was a 3 part, 5-week randomised, controlled, crossover study in which 2 different carbohydrate/protein-equivalent ratios in protein 2.7. Assessment of feeding behaviour substitute were compared. Subjects took either higher carbohydrate control protein substitutes (CPS) (with a median carbohydrate/pro- A prospective questionnaire, collecting data about the evening tein-equivalent ratio of 1:1, range 0.7:1 to 1.6:1) for 14 days compared meal for the pre-trial, and study part A and B was collected. Informa- with a lower carbohydrate liquid protein substitute (TPS) with a carbo- tion on time spent eating, appetite, and percentage of meal eaten was hydrate/protein-equivalent ratio of 0.5:1 for 14 days. The order was recorded. This was completed by the maternal caregiver for 3 days in – – randomised. Their effects on feeding behaviour, appetite, weight the pre-trial (days 5 7) and in part A and B (days 12 14). They also change and blood phenylalanine concentrations were compared. The completed a caregiver stress rating (1=not stressed [maximum same quantity of protein equivalent was prescribed for each child in score of 3 over 3 days]; 5=stressed [maximum score of 15 over 3 both parts of the study. The protein substitute was taken in three or days]) associated with mealtime. In addition, they recorded the fi four doses at individualised standard timings throughout the day. time taken to consume the protein substitute, any practical dif cul- Each type of protein substitute was taken for 2 weeks, preceded by ties and negative behaviour associated with taking protein substitute, a 7-day, pre-trial to provide baseline data. The order of study part A and a caregiver stress rating score for protein substitute administra- and B was randomised by computer generated random number tion, scored in the same way as feeding behaviour. sequences. 2.8. Assessment of anthropometry • Pre-study: The subjects took their usual protein substitute in their prescribed dose for 7 days. Anthropometric data were collected on day 0 and day 7 of the pre- • Part A: The subjects took their control (usual) protein substitute trial period, and again on day 14 of study part A and B. Weight was (CPS) (median carbohydrate/protein equivalent ratio of 1:1) in measured to the nearest 10 g using portable electronic scales. Length their regular dose, individually standardised for 14 days. was recorded on day 0 of the pre-trial period to the nearest 1 mm • Part B: The subjects took the low carbohydrate, ‘ready-to-drink’ using a Harpenden Stadiometer. All anthropometric measurements protein substitute (TPS) whose carbohydrate/protein equivalent were converted into z scores, which denote units of standard devia- ratio was 0.5:1.FOR They took equal ELECTRONIC protein equivalent to the amount tion from the median.USE ONLY provided by the control protein substitute [g/d protein equivalent median CPS: 56.1 (range 37.8 to 74.5); TPS median: 60 (range 40 2.9. Statistics to 70); g/kg/d protein equivalent median CPS median: 2.38 (range 1.35 to 2.88); TPS median: 2.5 (range 1.62 to 3)]. Non-parametric tests for crossover studies (adapted from Wilcoxon Signed Rank tests) were used to detect differences in plasma phenylalanine concentrations between the 2 trial and control periods. Energy 2.3. Control protein substitutes (CPS) and macronutrient intakes from natural foods during 3 study periods were compared using the nonparametric Friedman test. The CPS were PKU Gel (Vitaflo International) (n=10), a combination of XP Maxamaid (SHS International) and Aminogran Food Sup- 2.9.1. Power calculation plement Powder (UCB Pharma) (n=3) and a combination of XP Assuming as with past data [8–12] that phenylalanine day-to-day Maxamum (SHS International) and Aminogran Food Supplement variation is approximately ±14%, 14 subjects should detect a 10% S66 H. Gokmen-Ozel et al. / Molecular Genetics and Metabolism 104 (2011) S64–S67

Table 1 3.5. Feeding behaviour Anthropometric measurements, including change in weight during the trial.

Median Min–Max More of the evening meal (observed and documented by the main caregiver) was eaten (TPS: 92%, CPS 75%) when the TPS protein sub- Weight (kg) pre-trial 24 15.4 to 38.8 ﬁ Weight change on TPS (kg) compared with 0 −1 to 0.7 stitute was taken but the difference was not signi cant. The median pre-trial weight length of time taken to eat the evening meal was similar for all 3 study Weight change on CPS (kg) compared with 0.05 −0.6 to 0.9 parts (pre-trial, 20 min; TPS and CPS, 18 min). There was an improve- pre-trial weight ment in carer rating stress score associated with giving the evening Height (cm) pre-trial 118.3 95.9 to 144 − Weight for age z score at trial baseline −0.38 −1.96 to 1.23 meal for the TPS compared with pre-trial (median change: TPS, 1; Height for age z score at trial baseline 0.45 −1.62 to 2.71 CPS, 0) but the difference was not signiﬁcant.

3.6. Behaviour with protein substitute difference in median phenylalanine concentrations between the 2 study periods with a 90% power for establishing the equivalence of For both the TPS and CPS, 64% (9 of 14) of all children took the the periods in median phenylalanine within +/−7%. protein substitute in 5 min or less. However, 29% (n=4) consumed the TPS quicker than the CPS; and one child (7%) took between 21 3. Results and 30 min with both types of protein substitute. All of the children taking the TPS drank this independently. With the CPS, 86% (n=12) 3.1. Carbohydrate:protein equivalent ratios of protein substitutes took it as a paste from a spoon; [58% were spoon fed by parents (n=7/12)]; 14% (n=2) reconstituted the powder into a drink and On the CPS, 10 children had a carbohydrate to protein-equivalent so children drank this independently. Eighty six percent (n=12) of ratio of 1:1; three children a ratio of 1.6:1 and one child a ratio of the TPS caregivers reported that it was easy to give the TPS compared 0.7:1 compared with a ratio of 0.5:1 on the TPS. with 57% (n=8) in the CPS group. The median accumulative carer 3- day rating stress score for administering the protein substitute for the CPS was 7.5 (range 3–15) compared with 3 (range 3–9) in the TPS 3.2. Plasma phenylalanine concentrations group (higher score associated with more stress). This score remained unchanged in 8 (57%) children, improved in 5 (36%) but de- Plasma phenylalanine control on TPS was similar to the pre-study teriorated in 1 (7%) on the TPS. and CPS (p=0.783) [median blood phenylalanine concentrations pre-study were: 240 (range 150 to 420 μmol/l), on CPS: 275 (range 150 to 590 μmol/l) and TPS: 230 (range 41 to 710 μmol/l)]. 4. Discussion

We have shown in this study that the carbohydrate/protein-equiva- 3.3. Anthropometry (Table 1) lent ratio can be reduced to 0.5:1 without any loss of phenylalanine control in children aged 3 to 10 years of age. There was a concern that Three (21%) of subjects lost weight in each trial part compared children in this age group would not counterbalance the lower carbohy- with baseline weight although there was no median weight change drate profile of the TPS by eating more energy from natural foods. Howev- for the TPS compared with 0.05 kg weight gain in the CPS (Table 1). er, interestingly, although energy intake was reduced by approximately 100 kcal/day from the lower carbohydrate TPS, whilst taking the TPS, al- 3.4. Dietary intake (Table 2) though not significant, there was some suggestion that there was improvement in the energy intake consumed from natural foods (a Energy intake from protein substitute only was reduced on the median of 71 kcal/day). Furthermore any weight changes of the children TPS. Median energy intake on TPS was 372 kcal/day (range 248 to were similar between the two parts of the study. We did not demonstrate 434) and CPS was 479 kcal/day (range 308 to 550) (pb0.001). any improvement in appetite. Although the overall trend was that median energy, protein and fat The lack of apparent effect of the lower carbohydrate profile on intake were higher from natural foods in the TPS compared with CPS pe- blood phenylalanine control of the TPS in this age group of children riod (energy intake: 1338 kcal/day and 1267 kcal/day; protein intake: was similar to the results we reported in older patients in an earlier 8.5 g/day and 7.8 g/day; fat intake: 56 g/day and 47 g/day respectively), study [6]. In the latter study, we administered a protein substitute there was no significant difference between two periods. Median carbo- with the same carbohydrate/protein-equivalent ratio (0.5:1). Con- hydrate intakes were similar in TPS and CPS study periods (196 g/day versely, in a further study examining protein substitute dosage, and 205 g/day respectively) (Table 2). when the natural food intake was maintained at a constant intake

Table 2 Dietary energy andFOR macronutrient intake fromELECTRONIC diet only comparing pre-study, with CPS and TPS. USE ONLY Median energy and macronutrient intake from diet during 1) pre-study; 2) CPS Median difference of energy and macronutrients intake from and 3) TPS (min–max)a diet between 1) pre-study and CPS and 2) TFS and pre-study (min–max)a

Pre-study CPS TPS pb CPS-pre study TPS-Pre study pb

Energy (kcal/day) 1204 (736–2850) 1267 (737–3231) 1338 (609–2812) 0.526 −26 (−591 to 381) 55 (−598 to 900) 0.272 Protein (g/day) 7.3 (4.6–21.9) 7.8 (3.8–26.2) 8.5 (4.3–20.6) 0.424 0.78 (−1.8 to 4.2) 0.08 (−1.4 to 6.0) 0.683 Fat (g/day) 49 (27–174) 47 (26–236) 56 (20–171) 0.607 −0.88 (−43 to 62) 2.70 (−27 to 53) 0.300 Carbohydrate (g/day) 201 (122–336) 205 (109–353) 196 (105–457) 0.395 −11.1 (−103 to 98) −0.3 (−86 to 202) 0.300

TPS: trial protein substitute. CPS: control protein substitute. The data excludes energy and macronutrient intake provided by protein substitutes. a Friedman test. b Wilcoxon test. H. Gokmen-Ozel et al. / Molecular Genetics and Metabolism 104 (2011) S64–S67 S67 and protein substitute was reduced, blood phenylalanine concentra- Conflict of interests tions significantly increased which correlated with the reduction in energy and carbohydrate intake from protein substitute [1]. The im- Anita MacDonald is in receipt of research grants/posts from pact on blood phenylalanine control of a totally carbohydrate-free Nutricia and Vitaflo International, and is a member of the Nutri- protein substitute still remains to be determined. cia IMD Advisory Board, Sapropterin PKU Advisory Board and Eu- The main limitation with this study was the small subject num- ropean Nutrition Expert Panel (ENEP); Sharon Evans is a bers, so the results may not be representative of all children. Howev- research dietitian. Her salary has been partly funded by research er, there are few published studies on protein substitutes in children grants from Vitaflo International and Nutricia. However, no with PKU with greater numbers. This is mainly because children are member of the research team received direct payments to con- unwilling to change protein substitutes and whilst almost all duct this trial. recruited children (except one) drank the new TPS with ease, those who declined to participate in this study were reluctant to try any References new protein substitute. Furthermore, the PKU mutations and therefore, severity of this group of patients were unknown, although the [1] A.MacDonald,A.Chakrapani,C.Hendriksz,A.Daly,P.Davies,D.Asplin,K.Hall, I.W. Booth, Protein substitute dosage in PKU: how much do young patients median allocated phenylalanine intake was only 250 mg/day. This need? Arch. Dis. Child. 91 (2006) 588–593. study, although controlled, was only for 5 weeks duration and a lon- [2] A. MacDonald, G. Rylance, D. Asplin, G. Harris, I.W. Booth, Abnormal feeding behaviours ger term study is required. However, since trial completion, all but in phenylketonuria, J. Hum. Nutr. Diet. 10 (1997) 163–170. [3] D.J. Millward, J.P. Rivers, Protein and amino acid requirements in the adult one child has remained on the TPS and there has been maintenance human, J. Nutr. 116 (1986) 2559–2561. of growth, weight z score and blood phenylalanine control one year [4] H.N. Munro, E.D. Downie, Relationship of liver composition to intensity of protein post trial in all children. Many caregivers have anecdotally reported metabolism in different mammals, Nature 203 (1964) 603–604. [5] Medical Research Council (MRC) Working Party on Phenylketonuria, Recommen- that their children's appetites have improved long term, but this is dations on the dietary management of phenylketonuria, Arch. Dis. Child. 68 unmeasured and other confounding factors could account for this. (1993) 426–427. In conclusion, this short term study would suggest that the carbo- [6] A. MacDonald, M. Lilburn, P. Davies, S. Evans, A. Daly, S.K. Hall, C. Hendriksz, A. Chakrapani, P. Lee, Ready to drink' protein substitute is easier is for people with hydrate/protein-equivalent ratio of protein substitutes can be re- phenylketonuria, J. Inherit. Metab. Dis. 29 (2006) 526–531. duced to 0.5:1 without loss of blood phenylalanine control or [7] A. MacDonald, M. Lilburn, B. Cochrane, P. Davies, A. Daly, D. Asplin, S.K. Hall, A. Cousins, adverse effect on weight gain in children over the age of three A. Chakrapani, P. Robinson, P. Lee, A new, low-volume protein substitute for teenagers – years. However, it is important that this study is repeated over a lon- and adults with phenylketonuria, J. Inherit. Metab. Dis. 27 (2004) 127 135. [8] B. Holland, I.D. Unwin, D.H. Buss, Cereals and Cereal Products, Royal Society of ger time period, in a larger cohort of children with PKU using a range Chemistry and Ministry of Agriculture, Fisheries and Food, Cambridge, 1998. of carbohydrate/protein-equivalent ratios, examining both the im- [9] B. Holland, I.D. Unwin, D.H. Buss, Milk Products and Eggs, Royal Society of Chemistry pact on feeding behaviour and blood phenylalanine control. In prac- and Ministry of Agriculture, Fisheries and Food, Cambridge, 1989. [10] B. Holland, A.A. Welch, I.D. Unwin, D.H. Buss, A.A. Paul, D.A.T. Southgate, McCance and tice, there is an increasing trend for manufacturers to minimise the Widdowson's The Composition of Foods, Royal Society of Chemistry and Ministry of carbohydrate content of protein substitutes designed for children Agriculture, Fisheries and Food, Cambridge, 1991. with PKU. It is essential that there is data to demonstrate that this [11] B. Holland, I.D. Unwin, D.H. Buss, Fruits and Nuts, Royal Society of Chemistry and Ministry of Agriculture, Fisheries and Food, Cambridge, 1991. practice does not adversely affect blood phenylalanine control or [12] B. Holland, I.D. Unwin, D.H. Buss, Vegetables, Herbs and Spices, Royal Society of growth. Chemistry and Ministry of Agriculture, Fisheries and Food, Cambridge, 1991.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S68–S72

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Diurnal variations in blood phenylalanine of PKU infants under different feeding regimes

Margreet van Rijn a,⁎, Marieke Hoeksma a, Pieter J.J. Sauer b, Pim Modderman c, Dirk-Jan Reijngoud c,d, Francjan J. van Spronsen a,d a Department of Pediatrics, Section of Metabolic Diseases, Beatrix Children's Hospital, University Medical Center of Groningen, University of Groningen, The Netherlands b Department of Pediatrics, Beatrix Children's Hospital, University Medical Center of Groningen, University of Groningen, The Netherlands c Research Laboratory of Paediatrics, Beatrix Children's Hospital, University Medical Center of Groningen, University of Groningen, The Netherlands d Center for Liver, Digestive and Metabolic Diseases, University Medical Center of Groningen, University of Groningen, The Netherlands article info abstract

Article history: In phenylketonuria (PKU) patients, diurnal fluctuations of blood phenylalanine (Phe) are different from Received 30 June 2011 healthy individuals. Until now this pattern has been studied in PKU patients over one year of age. Received in revised form 10 August 2011 Objective: The aim of this observational study was to investigate diurnal patterns in PKU infants under one Accepted 10 August 2011 year of age receiving both the natural protein and Phe-free formula at the same time or in an alternating Available online 16 August 2011 feeding scheme. Methods: In 7 PKU infants (aged 3–8 months), diurnal variations in blood Phe concentrations were recorded: Keywords: Phenylketonuria on day A they received natural protein and Phe-free formula combined in each feeding; on day B they received Phenylalanine these in an alternating feeding scheme. The number of feedings, total protein, and energy intake was similar Diurnal fluctuation on both study days. Blood samples were taken before each feeding. Dietary Phe distribution Results: The means (±SD) of the difference between the individual minimum and maximum blood Phe concentrations were 81(±50)μmol/L and 104(±26)μmol/L on days A and B, respectively (n.s.). Fifty and 30% of the samples were below target range for age (120 μmol/L), while only 3% and 6% were above target range (360 μmol/L) on days A and B respectively (n.s.). Conclusion: Both feeding regimes, i.e. the natural protein and Phe-free formula combined in each feeding or alternating, resulted in comparable diurnal fluctuations of blood Phe concentrations. © 2011 Elsevier Inc. All rights reserved.

1. Introduction [2,4,5]. However, in PKU infants below one year of age, no data are available. Treatment of phenylketonuria (PKU, McKusick 261600) aims at Infants with PKU consume a more or less ﬁxed amount of Phe blood phenylalanine (Phe) concentrations within ranges that are (natural protein) and Phe free formula. In breastfeeding natural thought acceptable to prevent brain damage to the patient [1].In protein, natural protein and Phe-free formula is either given in the most centers the target range for young children is quite small, resulting same feeding or in an alternating feeding scheme [8–10]. Based on our in a relatively high risk for blood Phe concentrations out of the target previous study, and the assumption that feeding times were frequent range. Phe concentrations beyond target range will result in adjustment enough to prevent large diurnal variation in Phe concentrations, of the diet as the amounts of natural protein and Phe-free protein advising the alternating scheme had become common practice [10]. substitute contribute largely to the blood Phe concentration. However, theoretically, the alternating feeding scheme might result in Other factors that inﬂuence the blood Phe concentration include different diurnal variations of blood Phe concentration. growth velocity,FOR intercurrent illness, ELECTRONIC residual enzyme (phenylalanine Therefore, the USE aim of the present ONLY study was to investigate the hydroxylase) activity and energy intake [2–7]. Some reports concluded diurnal variations in blood Phe concentrations in PKU infants in that there is a large diurnal variation, especially in young children relation to different feeding regimes.

2. Methods and materials

Abbreviations: PKU, phenylketonuria; PHE, phenylalanine; TYR, tyrosine. 2.1. Patients ⁎ Corresponding author at: Department of Pediatrics, Section of Metabolic Diseases, Beatrix Children's Hospital, University Medical Center Groningen, P.O. Box 30.001, 9700 RB Groningen, The Netherlands. Fax: +31 503611704. The study was carried out in seven PKU infants aged three to eight E-mail address: [email protected] (M. van Rijn). months detected by neonatal screening and treated from time of

1096-7192/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2011.08.010 M. van Rijn et al. / Molecular Genetics and Metabolism 104 (2011) S68–S72 S69 diagnosis in the Beatrix Children's Hospital of the University Medical both days A and B were tested for significance using the paired Center of Groningen. Patients had blood Phe concentrations of Student t test. Differences in the number of blood Phe concentration N400 μmol/L at time of diagnosis. Patients were free from co-morbidity out of the target range of 120–360 μmol/L on days A and B were also and feeding problems. Patients were either taking breast feeding and tested for significance using the paired Student t test. The relationship Phe free formula, or normal formula and Phe free formula. between the severity of the disease (defined as the blood Phe After diagnosis, parents of PKU infants were instructed how to concentration at the time of diagnosis and the Phe tolerance) and the sample blood on filter paper from heel punctures and to send the resulting mean and ranges of blood Phe concentrations on days A and bloodspots to the laboratory for measurement of blood Phe B, was tested for significance with univariate regression analysis concentration. Based on the Phe concentration, intake of Phe was (ANOVA). This test was also applied to test for significant relationship adjusted. Frequency of sampling was adapted to age and blood Phe between the first sample of the day and the number of blood Phe concentrations and varied from almost daily in the first week after concentration out of the target range. Statistical analysis was done diagnosis, to once a week when the Phe concentration had stabilized, using SPSS version 12 (Chicago, IL, USA). Statistical significance was usually at 1–2 months. To be informed about the frequency of blood assumed at Pb0.05. Phe concentrations out of target range due to the diurnal variation, parents performed one or two blood Phe day profiles when the infant 3. Results had a stable feeding pattern, mostly when infants were between three and eight months of age. Data of all patients but one in whom PKU was diagnosed in our Informed consent was obtained from the parents to perform Phe center between 2004 and 2006, could be compared. Table 1 shows the day profiles during the different feeding regimes. clinical data of all seven patients. At the time of diagnosis (days 6–9) blood Phe concentration varied from 847 μmol/L to 2329 μmol/L. 2.2. Design of the study Patient 1 was diagnosed with a blood Phe concentration of 433 μmol/L, 24 h after birth as a sib of a PKU patient. Six patients were born after All patients that will be presented were studied twice. The studies uncomplicated pregnancies of 37–42 weeks; patient 6 was born at 36+5 were performed at home under strict supervision of a dietician (MR). weeks. Growth parameters of all patients for height, weight and head On day A, they received the natural protein (breast milk, normal circumference at time of the first testing day are given in Table 1. formula or fruit and vegetables) and the Phe-free formula combined In Table 2 Phe intake per day and per feeding, and the difference in all feedings. Infants receiving breast milk got Phe-free formula just between minimum and maximum blood Phe concentrations observed before being breastfed; the infants receiving normal formula got a on both days A and B are shown together with age at time of testing. mixture of normal formula and Phe-free formula. On day B, they Age varied between 12 and 33 weeks at study time, while the time received the natural protein and Phe-free formula at different feeding period between days A and B varied from one to five weeks. Individual times, i.e. Phe-free formula and natural protein were given in an − − Phe tolerance varied between 22 and 44 mg Phe·kg 1·day 1 on day alternating scheme. Feeding frequency depended on the feeding A and between 16 and 42 mg Phe·kg−1·day−1 on day B. habits of the infant at the time of the studied day. At 3–7 months of Fig. 1 shows the individual blood Phe concentrations at feeding age the prescribed diet contained 2.0 g protein kg−1 day−1 (as times on days A and B. On day A and day B the median difference natural and Phe-free formula combined) and 100 kcal·kg−1·day−1 between minimum and maximum blood Phe concentration was on both study days. When tested at 7–8 month of age the prescribed 79 μmol/L and 104 μmol/L respectively. Mean blood Phe concentra- diet contained 2.0 g protein·kg−1·day−1 (as natural and Phe-free tions and differences between minimum and maximum were not formula combined), while energy intake was reduced to − − significantly different between both days A and B (p=0.140 and 90 kcal·kg 1·day 1. Phe content of each feeding and each day was p=0.184 respectively). Fig. 1 also shows that on day A, 16 samples calculated in mg Phe per kg bodyweight using the ZIS-food calculation were below target and one sample was above the target range. On day computer program based on NEVO with supplementary data of B, 10 and 2 samples were below and above the target range manufacturer [11]. Natural protein sources that were applied were respectively. No significant differences were observed in the numbers breast milk, Nutrilon 1® (Nutricia, Zoetermeer, The Netherlands), of blood Phe concentrations out of target range on days A and B fruits and vegetables. Phe free formula was applied as PKU-1 mix® (p=0.253). (Milupa Friedrichsdorf Germany). Bloodspots on filter paper (type Univariate regression analysis (ANOVA) showed no significant 2992, Scheicher and Schuell, ‘s Hertogenbosch, The Netherlands) were relationships between the blood Phe concentration at time of taken before each feeding. Samples were sent by mail to the diagnosis, the Phe tolerance at days A and B and the resulting mean laboratory for further analysis. and ranges of blood Phe concentrations. 2.3. Analyses of blood Phe concentrations 4. Discussion Blood Phe concentrations were determined in eluates of 1/8 ' disks punched from dried blood spots. Disks were eluted at room Our results showed that in PKU infants under one year of age the μ w fi temperatureFOR for 30 min in 150 ELECTRONICL TCA solution (6.6% /v TCA in different tested USE feeding regimes ONLY did not result in signi cantly H2O) containing norvaline as internal standard, after which the debris different diurnal variations of blood Phe concentrations and numbers was precipitated by centrifugation for 5 min at 14,000 rpm. Ten of samples out of the target range. microliters of the supernatant was transferred to a clean reaction vial Before discussing the results in more detail, first a few methodolog- and prepared for high-performance liquid chromatography analysis ical issues need to be addressed. To study the diurnal variation of blood by the AccQ-Tag® method according to the manufacturer's protocol Phe concentration in PKU infants, we used the results of samples that (Waters, Breda, The Netherlands). parents took routinely at home. The intake might have been less strictly controlled than in a clinical situation. However the advantage of 2.4. Statistics performing the sampling at home was considered of more importance as performing the study at home reflects more the actual daily situation Mean, SD and the difference between the minimum and maximum and enables minimal disturbance of feeding habits of the infant. To of the measured blood Phe concentrations on days A and B of all control the intake the supervision of the dietician was strict and the individuals were calculated. Differences in these parameters between intake was recorded by the parents precisely. S70 M. van Rijn et al. / Molecular Genetics and Metabolism 104 (2011) S68–S72

Table 1 Patient characteristics.

Sex Age at diagnosis Phe at diagnosis Weight Length Head-circumference (days) μmol/L Z-scorea Z-scorea Z-scorea

Patient 1 f 1 433 1.23 0.48 0.8 Patient 2 m 6 2329 1.30 1.03 0.84 Patient 3 f 8 983 0.35 −0.71 −0.39 Patient 4 m 9 1698 −2.04 −2.26 −2.08 Patient 5 f 7 910 −1.41 −1.44 −0.81 Patient 6 f 6 847 0.33 1.17 −0.15 Patient 7 m 7 1774 −0.81 −0.12 −0.91

a At 3 months of age, analysis performed with the computer program of the Dutch Growth Foundation (Growth Analyzer version 3.5).

We observed that the range between the minimum and maximum [14]. In that time frame treatment aimed at normal blood Phe blood Phe concentrations on both days A and B was not significantly concentrations of 35–100 μmol/L [15,16]. The study of Smith et al. different and alternating intake of Phe did not result in significant showed that blood Phe concentration below 120 μmol/L during five differences in blood Phe concentrations. A remarkable number of months in the first two years of age was related to a slightly lower IQ samples were found to be below the minimum target Phe concentra- at four years of age [17]. It, however, should be taken into account that tion of 120 μmol/L, irrespective of the distribution of Phe intake over this study was based on historical blood Phe measurements, and the day. In Fig. 1, we see that especially those infants with a blood Phe measurements were usually performed at about 9 AM with the concentration in the morning below or just above 120 μmol/L showed highest Phe concentrations of a day. As a consequence, the Phe blood Phe concentrations below the target range during the rest of the concentrations reported in that study might reflect clearly lower Phe day. Regression analysis showed significant relationship between the concentrations during the rest of the day. In later years no significant first sample of the day and the number of measurements below relation was found between blood Phe concentrations ≤120 μmol/L at 120 μmol/L on that day. Thus, low absolute Phe concentrations at the young age and growth impairment [18]. In the study of Hoeksma et al. start of the day rather than large fluctuations seemed responsible for natural protein intake is positively related with head circumference the blood Phe concentrations below the target range. Various factors, growth [19]. Larger studies with data of blood Phe measurement with including unbalanced intake of natural protein, protein substitute and improved accuracy are needed to prove the safety of blood Phe energy, as well as age, and morning blood Phe concentration influence concentrations b120 μmol/L. Until these data are available, possible the diurnal course of the blood Phe concentration [2–5,12,13]. consequences of too low blood Phe concentrations should be taken The present study was the first to report on the diurnal variations in into consideration when treatment strategies are concerned. infants below one year of age. In line with study of MacDonald et al. that showed larger fluctuations in children compared to adolescents and 5. Conclusion adults, one might have expected considerable large fluctuations in our population [5]. Diurnal patterns of blood Phe concentrations were comparable under In PKU treatment most attention is primarily focused on different feeding regimes, i.e. the natural protein and Phe-free formula prevention of blood Phe concentrations above the target. Only few combined in each feeding or in an alternating feeding scheme, in PKU studies addressed possible consequences of too low (≤120 μmol/L) infants under one year of age. Our results also showed that blood Phe blood Phe concentrations. When dietary Phe restriction was started as concentrations below the target range of various guidelines were treatment of PKU patients, severe growth retardation was observed frequently observed at this young age. In infants, blood Phe concentration

Table 2 Age, number of feedings, Phe intake and blood Phe ranges on both study days A and B, in infants receiving either the natural protein and Phe-free formula combined in each feeding (A) or in an alternating feeding scheme (B).

Patient Age Body weight Natural protein Phe intake Number of feedings Phe intake p. feeding/kg Difference in blood Phe (weeks) (kg) /g/day mg/kg/day (μmol/L)a

Study A 1 21 7.5 5.6 33 4 8.4 28 2 27 8.9 4.3 22 6 3.3–5.3 103 3 32 8.5 6 32 5 5.2–7.3 135 4 19 5.3 3.7 31 5 6.2 128 5 18 5.5 5.1 42 4 10.1–11.9 89 6 13 5.7 5.6 44 5 8.8 19 7 14FOR 5.7 ELECTRONIC 3.7 29 5 USE 5.8 ONLY 67 Mean 81 SD 50

Study B 1 18 7.2 5.9 37 4 0–15 70 2 24 8.5 2.9 16 6 0–7.8 104 3 33 8.6 5.5 29 4 0–17.7 91 4 24 6.1 4.4 32 5 0–14.4 156 5 21 6.3 5.1 37 4 0–17.6 104 6 12 5.5 5.1 42 5 0–14 113 7 16 6 4.4 33 5 0–11 97 Mean 105 SD 26

a Difference between lowest and highest blood Phe concentration within 1 day. M. van Rijn et al. / Molecular Genetics and Metabolism 104 (2011) S68–S72 S71 Day A 480

420

360

patient 1 300 patient 2 patient 3 240 patient 4 patient 5 180 patient 6 patient 7

120

60 Blood Phe concentration (µmol/L)

0 23456789101112131415161718192021222324 time of the day (h)

Day B 420

360

300 patient 1 patient 2 240 patient 3 patient 4 patient 5 180 patient 6 patient 7 120

60 Blood Phe concentration(µmol/L) 0 234 5 6 7 8 9 101112131415161718192021222324 time of the day (h)

Fig. 1. Course of the individual blood Phe concentrations at days A and B. The dashed lines represent the upper and lower limits of the age-adjusted target range of blood Phe concentration. on the morning after an overnight fast appeared to be a good predictor of [6] F.J. van Spronsen, T. van Dijk, G.P. Smit, M. van Rijn, D.J. Reijngoud, R. Berger, H.S. Heymans, Phenylketonuria: plasma phenylalanine responses to different distributions the chance that blood Phe concentration will be below the target range, of the daily phenylalanine allowance over the day, Pediatrics 97 (1996) 839–844. irrespective of the feeding scheme. [7] F.J. van Spronsen, M. van Rijn, T. van Dijk, G.P. Smit, D.J. Reijngoud, R. Berger, H.S. Heymans, Plasma phenylalanine and tyrosine responses to different nutritional conditions (fasting/postprandial) in patients with phenylketonuria: effect of Acknowledgment sample timing, Pediatrics 92 (1993) 570–573. [8] P.B. Acosta, S. Yannicelli, Protocol l-phenylketonuria (PKU), Ross Metabolic This research was supported by the Beatrix Children's Hospital Formula System Nutrition Support Protocols, Ross Laboratories Columbus, OH, Foundation. 2001. [9] V. Shaw, M. Lawson, Clinical Paediatric Dietetics, 2nd ed. Blackwell Publishing, Oxford UK, 2001. References FOR ELECTRONIC[10] M. van Rijn, J.USE Bekhof, T. Dijkstra, P.G. ONLY Smit, P. Moddermam, F.J. van Spronsen, A different approach to breast-feeding of the infant with phenylketonuria, Eur. J. [1] N. Blau, F.J. van Spronsen, H.L. Levy, Phenylketonuria, Lancet 23 (2010) Pediatr. 162 (2003) 323–326. 1417–1427. [11] Dutch Food Composition Table NEVO, NEVO Foundation The Netherlands [2] F. Guttler, E.S. Olesen, E. Wamberg, Diurnal variations of serum phenylalanine in Nutrition Centre, The Hague, 2001. phenylketonuric children on low phenylalanine diet, Am. J. Clin. Nutr. 22 (1969) [12] F.J. van Spronsen, T. van Dijk, G.P. Smit, M. van Rijn, D.J. Reijngoud, R. Berger, H.S. 1568–1570. Heymans, Phenylketonuria: plasma phenylalanine responses to different distri- [3] D.L. Farquhar, F. Steven, A. Westwood, Preliminary report on inverse diurnal butions of the daily phenylalanine allowance over the day, Pediatrics 97 (1996) variation of phenylalanine: implications in maternal phenylketonuria, Hum. Nutr. 839–844. Appl. Nutr. 39 (1985) 224–226. [13] M.R. Crone, F.J. van Spronsen, K. Oudshoorn, J. Bekhof, G. van Rijn, P.H. Verkerk, [4] A. MacDonald, G.W. Rylance, D. Asplin, S.K. Hall, I.W. Booth, Does a single plasma Behavioural factors related to metabolic control in patients with phenylketonuria, phenylalanine predict quality of control in phenylketonuria? Arch. Dis. Child. 78 J. Inherit. Metab. Dis. 28 (2005) 627–637. (1998) 122–126. [14] anonymous, Malnutrition with early treatment of phenylketonuria, Nutr. Rev. 29 [5] A. MacDonald, G. Rylance, S.K. Hall, D. Asplin, I.W. Booth, Factors affecting the (1971) 11–13. variation in plasma phenylalanine in patients with phenylketonuria on diet, Arch. [15] D.M. Gregory, D. Sovetts, C.L. Clow, C.R. Scriver, Plasma free amino acid values in Dis. Child. 74 (1996) 412–417. normal children and adolescents, Metabolism 35 (1986) 967–969. S72 M. van Rijn et al. / Molecular Genetics and Metabolism 104 (2011) S68–S72

[16] C.R. Scriver, D.M. Gregory, D. Sovetts, G. Tissenbaum, Normal plasma free amino strictness of dietary treatment? National Dutch PKU Steering Committee, Acta acid values in adults: the inﬂuence of some common physiological variables, Paediatr. 86 (1997) 816–818. Metabolism 34 (1985) 868–873. [19] M. Hoeksma, M. Van Rijn, P.H. Verkerk, A.M. Bosch, M.F. Mulder, J.B. de Klerk, T.J. [17] I. Smith, M.G. Beasley, A.E. Ades, Intelligence and quality of dietary treatment in de Koning, E. Rubio-Gozalbo, M. de Vries, P.J. Sauer, F.J. van Spronsen, The intake of phenylketonuria, Arch. Dis. Child. 65 (1990) 472–478. total protein, natural protein and protein substitute and growth of height and [18] F.J. van Spronsen, P.H. Verkerk, M. van Houten, G.P. Smit, S.B. van der Meer, H.D. head circumference in Dutch infants with phenylketonuria, J. Inherit. Metab. Dis. Bakker, R.C. Sengers, Does impaired growth of PKU patients correlate with the 28 (2005) 845–854.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S73–S79

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Neurological complications and behavioral problems in patients with phenylketonuria in a Follow-up Unit

María J. González a,b, Alfonso P. Gutiérrez a,b, Rosa Gassió a,b, María E. Fusté b,c, María A. Vilaseca b, Jaume Campistol a,b,d,⁎ a Neuropediatrics Department, Hospital Universitari Sant Joan de Déu, Barcelona, Spain b PKU Follow-up Unit, Hospital Universitari Sant Joan de Déu, Barcelona, Spain c Psychology Department, Hospital Universitari Sant Joan de Déu, Barcelona, Spain d Centre for Biomedical Research on Rare Diseases (CIBERER), Institute of Health Carlos III, Spain article info abstract

Article history: Objective: To investigate the relationship between neurological complications, neuroradiological findings, and Received 30 May 2011 behavioral problems, age at diagnosis and dietary control along the follow-up of the PKU patients in our Received in revised form 14 July 2011 metabolic unit. Accepted 14 July 2011 Design: Retrospective study of the PKU patients diagnosed and controlled in our unit from 1985 to 2010. Available online 23 July 2011 Methods: Registry of patients in a database with 50 items filled in by review of the clinical histories. Statistical study of the data (SPSS, 19.0 version). Keywords: Results: 121 patients were included (median age: 16.0, range 1 month–46 years). 76% of them were diagnosed Phenylketonuria PKU through neonatal screening. 12.4% had mild-PKU, 19% moderate-PKU and 68.6% classic-PKU. 88.4% of patients Dietary control were treated with a protein-restricted diet, and 11.6% with BH4. 97.7% of the early diagnosed patients had Neurological complications normal IQ, while 46.3% of late diagnosed patients had mental retardation, 28.5% were borderline and 25% had Behavioral problems normal IQ. In early diagnosed patients, there was a significantly negative correlation between IQ [mean (SD) Brain magnetic resonance 100 (11.1)] and the index of dietary control during the first six years of life [median (range) 310 (105–992)] and that of the immediately past year [348 (106–1127)] (pb0.0001). The proportion of patients with late diagnosis and neurological and behavioral problems was significantly higher than that of the early diagnosed ones (pb0.001). The proportion of early diagnosed patients with neurological and behavioral problems who had good, intermediate or poor dietary control during the first 6 years of life and the immediately past year was significantly different (pb0.001). Conclusions: The results show the impact of early diagnosis and good dietary treatment on the IQ and on the percentage of neurological complications and behavioral problems in PKU patients. © 2011 Elsevier Inc. All rights reserved.

1. Introduction alternative treatment with BH4 also succeeds in decreasing high Phe concentrations in patients who respond to this therapy [5]. Early Phenylketonuria (PKU) is an inborn error of phenylalanine (Phe) treatment of PKU prevents severe neurological damage [1]. However, metabolism caused by a deficiency of phenylalanine hydroxylase despite appropriate treatment, some patients may present a slight (PAH), the hepatic enzyme which synthesizes tyrosine from Phe, decrease in intelligence in comparison with the general population using tetrahydrobiopterin (BH4) as a cofactor. Phe accumulation in [6,7] as well as specificdeficits of executive functions, especially when plasma and tissues and a decrease in tyrosine synthesis seem to be metabolic control is poor [8,9]. Behavioral problems have also been involved in the pathogenesis of the disease [1,2]. Classical therapy of described in patients with early treatment [10–12]. Quality of dietary PKU consistsFOR of Phe intake restriction, ELECTRONIC which means a diet with low control is determinant USE in the disease ONLY prognosis [7,13]. natural protein content supplemented with a special Phe-free The neurological and metabolic evaluation of wide series of formula, enriched with tyrosine and the other amino acids and patients may allow better understanding of the relationship between micronutrients (vitamins, minerals, oligoelements and essential fatty dietary control and the neurological outcome. The recently created acids in some of them) to prevent nutritional deficiencies [3,4].An registry of Spanish PKU patients [14] led us to do a retrospective neurological, behavioral and metabolic study of the whole series of patients diagnosed and followed-up in our metabolic unit from its ⁎ Corresponding author at: Neuropediatrics Department, Hospital Sant Joan de Déu, creation in 1985 until 2010. Passeig Sant Joan de Déu 2, 08095-Esplugues (Barcelona), Spain. Fax: +34 93 280 36 26. Our objective was to investigate the relationship between E-mail address: [email protected] (J. Campistol). neurological complications and behavioral problems, age at diagnosis

1096-7192/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2011.07.015 S74 M.J. González et al. / Molecular Genetics and Metabolism 104 (2011) S73–S79 and dietary control along the follow-up of the PKU patients in our and Babinski's sign) were especially focused. Tremor was evaluated with metabolic unit. the WHIGET Tremor Rating Scale. Development evaluation was performed at one, and three years of 2. Material and methods age, and intelligence quotient was evaluated at the age of 6, 12 and 18 years. Patients were evaluated annually by the unit psychologist 2.1. Patients until the age of 6 and at 12 and 18 years of age. The brain magnetic resonance imaging (MRI) had not routinely A retrospective, descriptive study of the PKU patients diagnosed and been performed, but only in selected groups of patients for research or controlled in the Follow-up Unit of the Hospital Sant Joan de Déu medical reasons. (Reference Center for PKU in Catalonia) from its creation in 1985 until Behavioral problems include internalizing symptoms such as 2010 was performed. The inclusion criteria were: a) PKU patients anxiety, depressed mood, and low self-esteem, and externalizing diagnosed and/or followed-up in our center, with PAH deficiency symptoms such as hyperactivity, impulsivity and aggressiveness. confirmed by differential diagnosis and/or genetic analysis of the PAH gene, and b) PKU patients with plasma Phe concentrations before 2.4.3. Biochemical methods treatment above 360 μmol/L. The exclusion criteria were: a) PKU patients Plasma and dried blood spot Phe concentrations were analyzed by ion with late diagnosis who refused treatment and follow-up; b) foreign exchange chromatography (Biochrom 30, Pharmacia Biotech). The patients diagnosed in our center who returned to their country and are not frequency of Phe measurements varied from weekly (b2monthofage) controlled at present in our unit; and c) patients who died owing to causes or biweekly (2 months–4 years old) to monthly (N4yearsofage). unrelated with PKU. PKU patients were classified according to plasma Phe concentrations before treatment (mild PKU: 360–600 μmol/L, moderate 2.4.4. Index of dietary control (IDC) PKU: 600–1200 μmol/L, and classical PKUN1200 μmol/L) [15]. We calculated the IDC as the Phe data reduction in half-year medians and the mean of all medians throughout the patient's life [5].We 2.2. Patient's registry examined the registry for the IDC of the first 6 years of life and the IDC of the immediately past year. We considered the IDC of the first 6 years of life An Excel database was created with 50 patient issues including to be good with Phe b360 μmol/L. The IDC of the immediately past year clinical data (age, sex, age at diagnosis and treatment), neurological depended on patient age: in patients below 10 years of age b360 μmol/L complications (epilepsy, tremor, clumsiness, spasticity, and mental was considered to be good and from 10 to 18 years b480 μmol/L was retardation), neuroradiological findings, and behavioral problems considered also good. In patients younger than 10 years of age an IDC from (attention deficit, impulsiveness, hyperactivity, anxious and depres- 360 to 480 μmol/L was intermediate, and it also was in patients older than sive mood, phobias, aggressiveness and low self-esteem), metabolic 10 years of age an IDCb600 μmol/L. Control was deemed poor when the data (dietary control) and mutations. The data were recorded IDC was higher than these values. following a review of the patient clinical histories. 2.5. Statistical study 2.3. Ethical issues The statistical study was performed with the SPSS program Adult patients or the parents (when patients were younger than (version 19.0). Pearson chi-square test was applied to search for 18 years of age or had mental retardation) signed an informed consent association between categorical variables. ANOVA with Bonferroni agreement in accord with the Helsinki Declaration of 1964, revised in correction was used to compare quantitative variables when the Edinburgh in 2000. Our hospital ethics committee approved the study. number of data was higher than 30, the distribution was normal and the variances were not significantly different (Levene test). Other- 2.4. Methods wise, Kruskal–Wallis test was used. Pearson test was used for correlations between quantitative data. Statistical significance was 2.4.1. General intelligence measurement accepted for pb0.05. Developmental quotient (DQ) was calculated with the Brunet– Lezine Scale for measuring Psychomotor Development in Early 3. Results Infancy in patients younger than 4 years of age. Intelligence quotient (IQ) was evaluated with the Kaufman Bit Intelligence Test (K-Bit) 3.1. Patients (patients from 4 to 6 years), Wechsler Intelligence Scale for Children- Revised (WISC-R) or IV (WISC-IV) (patients from 6 to 16 years 3.1.1. Characteristics of the patients 11 months), and Wechsler Adult Intelligence Scale Third Edition We evaluated data from 121 patients who fulfilled selection criteria. (WAIS-III) (patients older than 17 years). Twelve patients were excluded (3 dead patients; 7 late diagnosed adults Level of intelligence was classified as normal range (DQ/IQN80), who refused treatment and follow-up, and 2 foreigners who had returned borderline (DQ/IQ of 70–79) and mental retardation (DQ/IQb69), to their countries). Of the included patients, 46% (56) were males and 54% according to theFOR rules of the test manuals ELECTRONIC[16–20]. Mental retardation (65) females. Ages rangedUSE from 1 mon thONLY to 46 years (median 16.0 years). level was classified according to DSM-IV-TR criteria [21]. The age distribution was as follows: below 6 years: 32/121 (26.4%); from 6 to 11 years: 16/121 (13.2%); from 12 to 18 years: 19/121 (15.7%); and 2.4.2. Neurological complications, neuroradiological study, and adults: 54/121 (44.6%). behavioral problems Neurological complications were registered by review of the 3.1.2. Diagnosis patients' clinical histories, and behavioral problems were recorded The newborn screening program started in Catalonia in 1970 but had based on information from patients, parents and teachers. only a universal coverture in 1985. Of the whole series, 76% (92) of Patients were controlled by the neuropediatrician every 6 months patients were early diagnosed (during the first two months of life) and during the first year of life, annually for the first two years of life, and every 24% (29) were late diagnosed. The present age of early diagnosed 2 years after the age of three. In the neurological examination altered signs patients was a median 11.0 years (range: 1 month to 39 years), and that such as tremor, clumsiness and pyramidal signs (hiperreflexia, spasticity, of late diagnosed patients was 34.0 (15 to 46 years). M.J. González et al. / Molecular Genetics and Metabolism 104 (2011) S73–S79 S75

3.1.3. PKU classification When we considered the DQ/IQ related to the age at diagnosis, a 12.4% (15) were mild PKU, 19% (23) were moderate PKU and 68.6% significantly different distribution was observed among patients with (83) were classic PKU. early diagnosis (before one month and from one to two months of age) and late diagnosis (from two months to one year and older than one year – b 3.1.4. Genetic study of age) (Kruskal Wallis; P 0.0001) (Fig. 1). From 242 alleles, 177 mutations were found. Those with higher allelic frequency were summarized in Table 1. 3.2.2. Relationship between DQ/IQ and dietary control in early diagnosed patients 3.1.5. Treatment Asignificantly negative correlation was observed between DQ/IQ All patients were treated with Phe-restricted diet at diagnosis. and the IDC of the immediately past year and that of the first 6 years of Fifteen patients (controlled elsewhere at that moment) abandoned life (Pearson correlation test: r=−0.478, and r=−0.435, respectively, the diet after the age of 6–10 years and all of them resumed the diet in pb0.0001). 1985, when they started to be controlled in our unit. Seven of them The DQ/IQ was significantly different in patients with good, were early diagnosed. intermediate and poor IDC in the first 6 years of life (Fig. 2) and in In 2003 a BH4-loading test was performed on 64 patients, and 13 the immediately past year (ANOVA, Bonferroni correction pb0.0001). of them who responded to BH4 were subsequently treated with this cofactor. Afterwards one more patient was treated at diagnosis. Therefore, 14 patients were treated at the moment of the registry with 3.3. Neurological complications, neuroradiological findings, and behavioral doses ranging from 6 to 18 mg/kg per day of BH4, while 107 patients problems were treated with diet and formula. 3.3.1. Characteristics of the neurological complications, neuroradiological findings, and behavioral problems 3.1.6. Dietary control Characteristics of the neurological complications, such as epilepsy, In patients with early diagnosis, the IDC for the first 6 years of life was: tremor and clumsiness were examined. The proportion of patients median (range) 310 (105–992)μmol/L and that for the immediately past with these problems was summarized in Table 2. year was 348 (104–1127)μmol/L. Fifty six patients (61%) had good dietary Generalized or focal epilepsy was present in 9 of the 29 late control, 22 patients (24%) had intermediate dietary control, and 14 diagnosed patients (31%), with favorable response to antiepileptic patients (15%) had poor dietary control either during the first 6 years of drugs, except for two cases. In the 30% of late diagnosed patients life or during the last year. epilepsy started with flexion spasms. Only one early diagnosed In patients with late diagnosis the IDC for the immediately past patient with epilepsy family history had seizures. year was 433 (260–1247) μmol/L. Seventeen patients (59%) has good Tremor was predominantly postural and kinetic, and was limited dietary control, 4 patients (14%) intermediate, and 8 patients (27%) to the hands. poor dietary control. Clumsiness was present in 11% of the early diagnosed PKU patients No significant differences were observed regarding sex in any age who showed difficulties in gross motor functions, in throwing or group. catching a ball, in coordinating different limbs or in motor imitation skills. Four of the 29 late diagnosed patients (13.8%) did not walk and 3.2. Cognitive evaluation were wheelchair-bound. Spasticity was present in 8/29 late diagnosed patients (27.6%). 3.2.1. Development and intelligence quotient related to the age at diagnosis Brain magnetic resonance imaging (MR) was only performed in a The DQ/IQ of the early diagnosed patients was mean 100 (SD 11.1). Of limited number of early diagnosed patients (N=28), for special these, 97.7% had normal DQ/IQ, one patient was borderline, and only one research studies or owing to clinical suspicion of abnormalities patient with Turner syndrome had slight mental retardation. Three because of poor metabolic control. An abnormal image was detected patients could not be evaluated. in 17 patients (60.7%) consisting of increased signal in periventricular The IQ of the late diagnosed patients was mean 62 (SD 21.6). Of posterior white matter in T2 sequences. In the case of late diagnosed these, 25% had severe mental retardation, 3.5% moderate mental patients, MR was performed in 17 patients, and an abnormal white retardation, 17.8% slight mental retardation, 28.5% were borderline matter with cerebral (5 patients) and cerebellar (2 patients) atrophy and 25% had normal IQ. One patient could not be evaluated. was found in 12 of them (70.6%) (Table 2). Behavioral problems were found in 28.3% of early diagnosed patients and in the 86.2% of late diagnosed patients (Table 2).

Table 1 Spectrum of the most common mutations in the PAH gene (in bold are those of major 3.3.2. Neurological complications, neuroradiological findings, and allelic frequency). behavioral problems in relation with the age at diagnosis Mutations in theFOR PAH gene DetectedELECTRONIC alleles Allelic frequency The proportion USE of patients with ONLY late diagnosis and neurological fi N and behavioral problems, and abnormal neuroradiological ndings IVS10-11 G A 26 14.6% fi V388M 13 7.3% was signi cantly higher than that of early diagnosed patients IVS4 +5 GNT 12 6.7% (Pearson chi-square test, pb0.0001) (Tables 2 and 3). We could not S349P 11 6.2% consider individual complications because of the low number of I65T 9 5.0% affected patients with early diagnosis. R158Q 9 5.0% R261Q 6 3.3% R243Q 6 3.3% fi R261X 6 3.3% 3.3.3. Neurological complications, neuroradiological ndings, and R408W 5 2.8% behavioral problems in patients with early diagnosis in relation to sex L48S 5 2.8% No significant differences were observed in the proportion of patients R243X 5 2.8% with early diagnosis and neurological complications or behavioral Y414C 4 2.2% problems in terms of sex (44 males/48 females) (Pearson chi-square test). S76 M.J. González et al. / Molecular Genetics and Metabolism 104 (2011) S73–S79

Fig.1. Development or intelligence quotient (DQ/IQ) related to the age at diagnosis of the PKU patients (Kruskal–Wallis test, pN0.0001). Figure legend: The length of the boxes indicates the interquartile space (P25–P75); the horizontal line in the box represents the median (P50) and the whiskers indicate the adjacent values, i.e. the maximum and minimum values of the distribution, which may not be considered abnormal. The circles indicate the outliers.

3.3.4. Neurological complications, neuroradiological findings, and 4. Discussion behavioral problems in patients with early diagnosis in relation to dietary control We previously performed several neurological and neuropsycho- The proportion of patients with neurological complications was logical studies involving selected groups of PKU patients controlled in significantly different in patients with good, intermediate and poor our unit [22–26], but we had never evaluated the neurological and IDC in the first 6 years of life (Pearson chi-square test p=0.007), and behavioral outcome of the whole series of patients either with early or those in the immediately past year (p=0.001). late diagnosis. The recent register of the Spanish PKU patients [14] led The proportion of patients with behavioral problems was also us to include all patients diagnosed and controlled in our unit in a significantly different in patients with good, intermediate and poor database, and to specifically analyze the neurological complications IDC in the first 6 years of life (Pearson chi-square test: p=0.007) and and behavioral problems of the whole series. Moreover, we completed those in the immediately past year (pb0.0001). Neurological and the general information with new data regarding the dietary control behavioral alterations could not be individually considered because of during the first 6 years of life, the patients who abandoned the diet, the low number of affected patients. and specific information about the complications found in individual Considering early diagnosed patients who abandoned the diet at the patients and their basic schooling and higher education. age of 8–10 years (N=7) and resumed after a period free of dietary There is a general consensus in the literature on the decrease in the control, the proportion of patients with neurological complications and intellectual capacity of early treated PKU patients [6,27–29]. In spite of behavioral problems was not significantly different compared with having an intelligence quotient within the normal limits, these those from a group of early diagnosed patients with similar ages patients show a slight decrease in intellectual capacity compared (N=15) who continued treatment through their lives. with identical control groups, including their healthy brothers and sisters. Moreover, they show some deficits, especially in executive 3.4. Schooling functions [8,9,30,31]. All these deficits are related with the early start of treatment and the quality and the duration of the dietary control In the group of patients with early diagnosis and of school age (35 [32–34]. Our previous results, involving 37 early diagnosed patients patients from 6 toFOR 18 years), 82.8% (29/35)ELECTRONIC attended ordinary school and and 29 healthy, ageUSE and sex-matched ONLY subjects [8], and the present 14.3% (5/35) ordinary school with support. Only one patient attended a study with 92 patients, both found a mean IQ of 100 in PKU patients, special school. Among the 25 early diagnosed patients older than 18 years which is significantly lower than that found in our control group 36% (9/25) had completed schooling. (IQ=111) [8]. Only one patient with combined PKU and Turner Among late diagnosed patients, 7% (2/29) attended ordinary syndrome showed slight mental retardation, and a further one had school, 24% (7/29) ordinary school with support, 48% (14/29) attend borderline intelligence. or attended a special school, and 21% (6/29) had completed schooling. Conversely, in our study late diagnosed patients showed a mean IQ of Among the 54 patients older than 18 years (with early or late 62, with 46.3% of patients with severe to slight mental retardation, 28.5% diagnosis), 15% (8/54) attend or attended university. At present 2 of with borderline intelligence and 25% with normal IQ. Two of these late them are dieticians (one with late diagnosis), one is an engineer, one diagnosed patients even achieved university studies (PAH genotype studied political science, one is a business manager, one is a chemist P279fsdelC-V388M) [35]. This proportion of mentally retarded patients and one studied Catalan philology. A further late diagnosed patient is rather low compared with what is found in the literature (96–98% of studied and is now working as a social worker. untreated PKU patients with IQ lower than 50 [36], and 84% in another M.J. González et al. / Molecular Genetics and Metabolism 104 (2011) S73–S79 S77

Fig. 2. Development or intelligence quotient in early diagnosed PKU patients with good, intermediate and poor IDC in the first 6 years of life. series [37]). A possible explanation for this may be the higher proportion authors (25%) [36,41,42], and this was also the case with tremor and of mild mutations in the Mediterranean countries [38,39], which allows clumsiness [42,25]. However, even a few patients with early diagnosis a better outcome in some untreated patients. A further reason is that and acceptable dietary treatment showed tremor and clumsiness. The most of our late diagnosed patients (65%) assumed the Phe-restricted only early diagnosed patient with seizures had familial antecedents of diet with rather good control, and this might have had a beneficial effect epilepsy, and the only one with slight mental retardation also suffered on their intellectual capacity even in late diagnosed patients [40]. Turner syndrome. Another relevant question in our study was the similar IQ between A high prevalence of abnormal white matter MR imaging has already patients diagnosed and treated during the first month and those been described in the literature, especially in patients with poor diagnosed and treated during the second month of life. The IQ sharply metabolic control [43]. These lesions have been associated with high descended after the second month and kept on decreasing after the concentrations of Phe in plasma and central nervous system by MR with first year of life (Fig. 1). spectroscopy [44], and they seem to be quite reversible by optimization The quality of dietary control, especially during the first 6 years of of the metabolic control. We found these abnormal patterns in a high life, further determined the DQ/IQ of early treated patients, as has percentage of early and late diagnosed patients in the present series already been observed in previous studies with smaller series [8,13]. (Table 2), although most of these patients are asymptomatic. These Similar results were found by other authors [33]. results should be interpreted with caution in the case of early diagnosed Regarding neurological complications, we focused on epilepsy, patients because of the limited patients sample (N=28). Moreover, the tremor, clumsiness and mental retardation. The proportion of late brain MR had been performed for special research studies or owing to diagnosed patients with epilepsy (26%) was similar than that of other the clinical suspicion of abnormalities because of poor metabolic control. Nevertheless we found abnormal volumetric studies in this population [26]. Behavioral problems, extensively described in PKU patients, include Table 2 internalizing symptoms such as anxiety, depressed mood, and low self- Neurological complications and neuroradiological findings in early and late diagnosed PKU patients. Table 3 Neurological complications Early diagnosis Late diagnosis Behavioral alterations in early and late diagnosed PKU patients. Number of patients Number of patients Behavioral alterations Early diagnosis Late diagnosis FOR (%)ELECTRONIC(%) USE ONLY Number of patients Number of patients Total patients=92 Total patients =29 (%) (%) Epilepsy 1 (1.1) 9 (31.0) Total patients=92 Total patients =29 Tremor 11(12.0) 27 (93.1) Clumsiness 10 (10.9) 26 (89.7) Attention deficit 21 (22.8) 20 (69.0) Spasticity – 8 (27.6) Impulsiveness 16 (17.4) 10 (34.5) Mental retardation 1 (1.1) 13 (46.3) Hyperactivity 19 (20.7) 13 (44.8) Anxious mood 7 (7.6) 14 (48.3) MR findings Depressed mood 7 (7.6) 14 (48.3) Altered brain MRa 17/28(60.7) 12/17(70.6) Phobias 3 (3.3) 7 (24.1) Total % of patients with neurological 23 (25) 28 (96.6) Low self-esteem 11(12.0) 17 (58.6) complications and altered MR Aggressiveness – 8 (27.6) Total % of patients with behavioral 26 (28.3) 25 (86.2) a Brain MR was only performed in a reduced number of PKU patients with early problems diagnosis (N=28) and late diagnosis (N=17). S78 M.J. González et al. / Molecular Genetics and Metabolism 104 (2011) S73–S79 esteem [7,45,46], and externalizing symptoms such as hyperactivity and [11] B.A. Stemerdink, A.F. Kalverboer, J.J. van der Meere, et al., Behaviour and school achievement in patients with early and continuously treated phenylketonuria, J. impulsivity [11,47,48,12]. Although half of our late diagnosed patients Inherit. Metab. Dis. 23 (2000) 548–562. presented one or more of these symptoms, only a very few of the early [12] V.L. Brumm, D. Bilder, S.E. Waisbren, Psychiatric symptoms and disorders in diagnosed patients showed them. The externalizing symptoms, includ- phenylketonuria, Mol. Genet. Metab. 99 (2010) S59–S63. fi [13] M.A. Vilaseca, N. Lambruschini, L. Gómez-López, et al., Quality of dietary control in ing attention de cit, impulsivity and hyperactivity, are the most phenylketonuric patients and its relationship with general intelligence, Nutr. prevalent ones in our series. We did not find a different proportion of Hosp. 25 (2010) 60–66. males and females with behavioral symptoms in our series, in contrast [14] J. Campistol, M.J. González, A.P. Gutiérrez, M.A. Vilaseca, and the Collaborative to the results of other authors [11]. Group of the Spanish Follow-up Units, treatment and control of phenylketonuric patients: results of the Collaborative Group of the Spanish Follow-up Units. Med. The early diagnosed patients who abandoned dietary treatment at Clin. (Barcelona) (in press). the age of 8–10 years and resumed it after a diet-free period did not [15] J. Campistol, R. Gassió, R. Artuch, M.A. Vilaseca, Neurocognitive function in mild – show a higher proportion of neurological complications or behavioral hyperphenylalaninemia, Dev. Med. Child Neurol. 53 (2011) 405 408. [16] D. Wechsler, WISC-R. Escala de Inteligencia de Wechsler para niños — Revisada, troubles compared with patients who continued treatment through- TEA ediciones, Madrid, 1994. out life. [17] D. Wechsler, WISC-IV. Escala de inteligencia de Wechsler para niños — IV, 2ª ed. As regards schooling, even early diagnosed patients showed a TEA edits, Madrid, 2007. [18] D. Wechsler, WAIS-III. Escala de inteligencia de Wechsler para adultos — III, TEA higher incidence of school problems compared with control groups edits, Madrid, 1999. [7,11,49,50]. We had previously performed a study in a limited group [19] A.S. Kaufman, N.L. Kaufman, K-bit, test breve de inteligencia de Kaufman, TEA of 26 PKU patients [22] and demonstrated that they presented with edits, Madrid, 2000. – fi [20] O. Brunet, I. Lezine, Brunet Lezine scale for measuring psychomotor development more learning dif culties than controls, probably related to the in early infancy, Psymtec, Madrid, 1997. disturbed cognitive functions observed in PKU patients. The index of [21] American Psychiatric Association, Diagnostic and statistical manual of mental dietary control of the immediately past 6 months showed a close disorders –ivtext revision, DSM-IV TR, Barcelona, Masson, 2001. [22] R. Gassió, J. Campistol, M.A. Vilaseca, N. Lambruschini, F.J. Cambra, E. Fusté, Do adult relationship with school performance [18], which points to the patients with phenylketonuria improve their quality of life after introduction/ importance of good dietary control in PKU. In the present study, 82.8% resumption of a phenylalanine-restricted diet? Acta Paediatr. 92 (2003) 1474–1478. of early diagnosed patients attended ordinary school, 14.3% attended [23] R. Gassió, E. Fusté, A. López-Sala, R. Artuch, M.A. Vilaseca, J. Campistol, School to ordinary school with support, and only one patient needed a special performance in early and continuously treated phenylketonuria, Pediatr. Neurol. 33 (2005) 267–271. school. Moreover, of the total group, 15% of adult patients had [24] R. Gassió, R. Artuch, M.A. Vilaseca, E. Fusté, C. Colomé, J. Campistol, Cognitive functions attended university and are now working in jobs related with their and the antioxidant system in phenylketonuric patients, Neuropsychology 22 (2008) – studies. 426 431. [25] B. Pérez-Dueñas, J. Valls-Solé, E. Fernández-Alvarez, et al., Characterization of tremor in phenylketonuric patients, J. Neurol. 252 (2005) 1328–1334. [26] B. Pérez-Dueñas, J. Pujol, C. Soriano-Mas, et al., Global and regional volume 4.1. Conclusions changes in the brains of patients with phenylketonuria, Neurology 66 (2006) 1074–1078. [27] S.E. Waisbren, K. Noel, K. Fahrbach, et al., Phenylalanine blood levels and clinical Our results demonstrate the impact of early diagnosis and good outcomes in phenylketonuria: a systematic literature review and meta-analysis, dietary treatment on IQ and on the incidence of neurological Mol. Genet. Metab. 92 (2007) 63–70. complications and behavioral problems in PKU patients. [28] J.J. Moyle, A.M. Fox, M. Arthur, M. Bynevelt, J.R. Burnett, Meta-analysis of neuropsychological symptoms of adolescents and adults with PKU, Neuropsychol. Rev. 17 (2007) 91–101. [29] K. De Roche, M. Welsh, Twenty-five years of research on neurocognitive outcomes Acknowledgments in early-treated phenylketonuria: intelligence and executive functions, Dev. Neuropsychol. 33 (2008) 474–504. We very much appreciate the close collaboration of the Metabolic [30] M.C. Welsh, B.F. Pennington, S. Ozonoff, Neuropsychology of early-treated phenylketonuria: specific executive function deficits, Child Dev. 61 (1990) Nutrition staff, the skillful technical assistance of Juan Moreno, and the 1697–1713. collaboration of the PKU patients and their families in the study. [31] A. Diamond, M.B. Prevor, G. Callender, D. Druin, Prefrontal cortex cognitive deficits in children treated early and continuously for PKU, Monogr. Soc. Res. Child Dev. 62 (1997). [32] I. Smith, M.G. Beasley, A.E. Ades, Effect on intelligence of relaxing the low References phenylalanine diet in phenylketonuria, Arch. Dis. Child. 66 (1991) 311–316. [1] C.R. Scriver, S. Kaufman, Hyperphenylalaninemias: phenylalanine hydroxylase [33] P. Burgard, Development of intelligence in early treated phenylketonuria, Eur. J. deficiency, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle, B. Childs, B. Vogelstein Pediatr. 159 (Suppl. 2) (2000) S74–S79. (Eds.), The metabolic and molecular basis of inherited diseases, 8th ed., McGraw [34] S.E. Waisbren, Comments on cognition and intelligence in phenylketonuria, Eur. J. Hill, New York, 2001, pp. 1667–1724, Chap. 77. Pediatr. 159 (Suppl. 2) (2000) S80–S81. [2] J. Campistol, N. Lambruschini, L. Gómez-López, A. Gutiérrez, E. Fusté, M.A. [35] J. Mallolas, M.A. Vilaseca, J. Campistol, et al., Clinical, biomedical, neurological and Vilaseca, Hiperfenilalaninemias, in: P. Sanjurjo, A. Baldellou (Eds.), Diagnóstico y molecular study of 11 patients with new mutations in PAH gene, Rev. Neurol. 31 tratamiento de las enfermedades metabólicas hereditarias, 3ª edición, Editorial (2000) 907–910. Ergon, 2010, pp. 423–439, Cap 29. [36] A. Tourian, J.B. Sidbury, Phenylketonuria and hyperphenylalaninemia, in: J.B. [3] H. Przyrembel, H.J. Bremer, Nutrition, physical growth, and bone density in Stanbury, J.B. Wyngaarden, D.S. Fredrickson, J.L. Goldstein, M.S. Brow (Eds.), The treated phenylketonuria, Eur. J. Pediatr. 159 (Suppl. 2) (2000) S129–S135. metabolic basis of inherited disease, McGraw-Hill, New York, 1982, pp. 270–286. [4] M. Giovannini, E. Verduci, E. Salvatici, L. Fiori, E. Riva, Phenylketonuria: dietary [37] D.B. Pitt, D.M. Danks, The natural history of untreated phenylketonuria over and therapeutic challenges, J. Inherit. Metab. Dis. 30 (2007) 145–152. 20 years, J. Paediatr. Child Health 27 (1991) 189–190. [5] N. Blau, H. Erlandsen, The metabolic and molecular bases of tetrahydrobiopterin- [38] L.R. Desviat, B. Perez, M.J. Garcia, et al., Relationship between mutation genotype responsive phenylalanineFOR hydroxylase ELECTRONIC deficiency, Mol. Genet. Metab. 82 (2004) and biochemical USE phenotype in a hetero ONLYgeneous Spanish phenylketonuria 101–111. population, Eur. J. Hum. Genet. 5 (1997) 196–202. [6] I. Smith, M.G. Beasley, A.E. Ades, Intelligence and quality of dietary treatment in [39] J. Mallolas, M.A. Vilaseca, J. Campistol, et al., Mutational spectrum of phenylal- phenylketonuria, Arch. Dis. Child. 65 (1990) 472–478. anine hydroxylase deficiency in the population resident in Catalonia: genotype– [7] J. Weglage, M. Pietsch, B. Fünders, H.G. Koch, K. Ullrich, Deficits in selective and phenotype correlation, Hum. Genet. 105 (1999) 468–473. sustained attention processes in early treated children with phenylketonuria — [40] R. Koch, K. Moseley, J. Ning, A. Romstad, P. Guldberg, F. Guttler, Long-term result of impaired frontal lobe functions, Eur. J. Pediatr. 155 (1996) 200–2004. beneficial effects of the phenylalanine-restricted diet in late-diagnosed in- [8] R. Gassió, R. Artuch, M.A. Vilaseca, et al., Cognitive functions in classic dividuals with phenylketonuria, Mol. Genet. Metab. 67 (1999) 148–155. phenylketonuria and mild hyperphenylalaninemia: experience in a paediatric [41] I. Smith, D.P. Brenton, Hyperphenylalaninaemias, in: J. Fernandes, J.M. Saudubray, population, Dev. Med. Child Neurol. 47 (2005) 443–448. G. van den Berghe (Eds.), Inborn metabolic diseases, Springer-Verlag, Berlin- [9] S.E. Christ, S.C. Huijbregts, L.M. de Sonneville, D.A. White, Executive function in Heidelberg-New York, 1996, pp. 147–160. early-treated phenylketonuria: profile and underlying mechanisms, Mol. Genet. [42] D.P. Brenton, J. Pietz, Adult care in phenylketonuria and hyperphenylalaninaemia: Metab. 99 (Suppl. 1) (2010) S22–S32. the relevance of neurological abnormalities, Eur. J. Pediatr. 159 (Suppl. 2) (2000) [10] J. Pietz, B. Fätkenheuer, P. Burgard, M. Armbruster, G. Esser, H. Schmidt, S114–S120. Psychiatric disorders in adult patients with early-treated phenylketonuria, [43] J.P. Anderson, V. Leuzzi, White matter pathology in phenylketonuria, Mol. Genet. Pediatrics 99 (1997) 345–350. Metab. 99 (2010) S3–S9. M.J. González et al. / Molecular Genetics and Metabolism 104 (2011) S73–S79 S79

[44] A.J. Thompson, S. Tillotson, I. Smith, B. Kendall, S.G. Moore, D.P. Brenton, Brain MRI [48] G.L. Arnold, C.J. Vladutiu, C.C. Orlowski, E.M. Blakely, J. DeLuca, Prevalence of changes in phenylketonuria. Associations with dietary status, Brain 116 (1993) stimulant use for attentional dysfunction in children with phenylketonuria, J. 811–821. Inherit. Metab. Dis. 27 (2004) 137–143. [45] J. Pietz, R. Dunckelmann, A. Rupp, et al., Neurological outcome in adult patients with [49] J. Zeman, A. Pijackova, J. Behulova, O. Urge, D. Saligova, J. Hyanek, early-treated phenylketonuria, Eur. J. Pediatr. 157 (1998) 824–830. Intellectual and school performance in adolescents with phenylketonuria [46] I. Smith, J. Knowles, Behaviour in early treated phenylketonuria: a systematic review, according to their dietary compliance, Eur. J. Pediatr. 155 (Suppl. 1) (1996) Eur. J. Pediatr. 159 (Suppl. 2) (2000) S89–S93. S56–S58. [47] K.M. Antshel, ADHD, learning, and academic performance in phenylketonuria, [50] J.K. Gentile, A.E. Ten Hoedt, A.M. Bosch, Psychosocial aspects of PKU: hidden Mol. Genet. Metab. 99 (2010) S52–S58. disabilities — a review, Mol. Genet. Metab. 99 (2010) S64–S67.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S80–S85

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Optimized loading test to evaluate responsiveness to tetrahydrobiopterin (BH4) in Brazilian patients with phenylalanine hydroxylase deﬁciency

Tatiéle Nalin a, Ingrid Dalira Schweigert Perry b, Angela Sitta c, Carmen Regla Vargas c,d,e, Maria Luiza Saraiva-Pereira d,f, Roberto Giugliani a,d,g,h, Nenad Blau i,j, Ida Vanessa Doederlein Schwartz a,d,g,⁎ a Post-Graduation Program in Medicine: Medical Sciences, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil b Department of Internal Medicine, Universidade Federal do Rio Grande do Sul. Food and Nutrition Research Center of Hospital de Clínicas de Porto Alegre, Porto Alegre, Brazil c Post-Graduation Program in Biochemistry, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil d Medical Genetics Service, Hospital de Clínicas de Porto Alegre, Porto Alegre, Brazil e School of Pharmacy, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil f Department of Biochemistry, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil g Department of Genetics, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil h INAGEMP — Instituto Nacional de Genética Médica Populacional, Porto Alegre, Brazil i Division of Clinical Chemistry, University Children's Hospital, Zürich, Switzerland j Zürich Center for Integrative Human Physiology (ZIHP), Zürich, Switzerland article info abstract

Article history: Introduction: Recent studies showed that phenylalanine (Phe) plasma concentrations may decrease in some Received 27 July 2011 patients with hyperphenylalaninemia (HPA) due to phenylalanine hydroxylase (PAH) deficiency, after the Received in revised form 13 September 2011 administration of tetrahydrobiopterin (BH4). Accepted 13 September 2011 Objective: To determine responsiveness to a single dose of BH4 administered according to an innovative Available online 20 September 2011 protocol using a combined Phe and BH4 loading test in Brazilian phenylketonuria (PKU) patients. Methods: Patient age should be ≥4 years, and median Phe plasma concentration ≤600 μmol/L when following Keywords: dietary restrictions. Participants received a simple Phe loading test using 100 mg/kg L-Phe (Test 1) and a com- Inborn errors of metabolism Hyperphenylalaninemia bined Phe+BH4 loading test using 100 mg/kg L-Phe and 20 mg/kg/BH4 (Test 2). Blood samples were collected Phenylketonuria at baseline and 3, 11 and 27 h after Phe ingestion (T0, T1, T2 and T3). Responsiveness was defined as: criterion Phenylalanine A: plasma Phe reduction of ≥30% at T1 and T2 for Tests 1 and 2; criterion B: plasma Phe reduction of ≥30% at Tetrahydrobiopterin T1 and T3 for Tests 1 and 2; and criterion C: at least 30% difference of the areas under the Phe curve for Tests 1 and 2. Results: Eighteen patients (median age 12 yrs; 8 classical PKU; 10 mild PKU) participated in the study. Six patients (2 classical PKU; 4 mild PKU) were classified as responsive according to at least one of the criteria. Responsiveness was concordant when criteria A+B we compared with criterion C (kappa=0.557; p=0.017).

Of the patients whose genotype was available (n=16), six had data about BH4-responsiveness genotypes described in the literature, which were in agreement with our ﬁndings.

Conclusion: The comparison of simple Phe loading and combined Phe+BH4 loading seems to be an optimal

1. Introduction The standard PKU treatment is based on a Phe-restricted diet and FOR ELECTRONICingestion of a Phe-free USE amino acid-rich ONLY metabolic formula that supplies Phenylketonuria (PKU) or hyperphenylalaninemia due to phenyl- the daily protein requirements of a patient [1,2]. Because of the toxic alanine hydroxylase (PAH) deficiency is an inborn error of amino acid effects of high Phe levels, this condition, if left untreated, may lead to metabolism characterized by the persistent increase of phenylalanine neurological impairment, mental retardation and behavioral disorders (Phe) plasma concentration. PAH converts Phe into tyrosine (Tyr), in [3,4]. the presence of its cofactor tetrahydrobiopterin (BH4) [1]. The necessary dietary restriction and the associated difficulty to adhere to the treatment [4–6], have motivated the search for new PKU management strategies [7]. Since the publication of the study ⁎ Corresponding author at: Serviço de Genética Médica, Hospital de Clínicas de Porto fi Alegre, Rua Ramiro Barcelos 2350, 90035-903 Porto Alegre, RS, Brazil. by Kure et al. [8], who described the rst case of patients with PKU E-mail address: [email protected] (I.V.D. Schwartz). whose Phe levels decreased after BH4 administration, several studies

have been conducted to confirm the efficacy and safety of this medica- 2.2.2. Combined Phe and BH4 loading (Test 2) tion, which has already been approved by the FDA and the EMEA. In the second week, evaluation was conducted using the protocol Patients are usually evaluated to check their responsiveness and, in described by Blau et al. [24], with a modification, as Phe and Tyr levels case results suggest that they are responsive, that is, that Phe levels were not analyzed 7 h after Phe loading. The initial phases of Test 2 will decrease after BH4 administration, BH4 supplementation is initiat- (collection at T0, Phe loading, food ingestion, and blood collection ed. Studies, however, have used different protocols to evaluate re- at T1) were similar to those described for Test 1. In addition, immedi- sponsiveness: BH4 doses are different and may range from 10 to ately after collection at T1 bloods sample, a single dose of 20 mg/kg 20 mg/kg/day in a single dose or distributed along the day; test BH4 (sapropterin dihydrochloride, KUVAN®, Merck Serono) was evaluation times range from some hours to weeks or even months; administered orally and samples were collected at 8 h (T2) and 24 h the cut-off point of Phe variation defined to determine responsiveness (T3) after BH4 ingestion. Time points T0 and T1 of Tests 1 and 2 also varies, and the criterion most frequently adopted is a decrease of were, therefore, equivalent to each other, whereas T2 and T3 were

30% in Phe levels 24 h after BH4 administration. In addition, different differed in that BH4 administration was included in Test 2. diets are used during tests: normal diets, Phe-restricted diet, or even L-Phe and BH4 were dissolved in orange juice before administration. a Phe-loading diet using, for example, powder milk or L-Phe [7,9–22]. Patients were told to fast for at least 1 h before all blood collections.

This study describes responsiveness to a single dose of BH4 in a Phe and Tyr plasma concentrations were measured using tandem sample of Brazilian patients with PKU and good metabolic control. mass spectrometry (MS/MS) in the Laboratory of Inborn Errors of For that purpose, an innovative protocol with a single Phe plus a Metabolism of SGM/HCPA, as described by Rashed et al. [25]. All mea- combined Phe and BH4 loading tests was used. surements were made in duplicate, and the mean of the two measurements was calculated. In Test 2 samples, the levels of BH4 (total 2. Material and methods biopterins) were also measured, according to the method of Opladen et al. [26]. This study included patients with PKU seen in the Outpatient

Metabolic Disorder Treatment Clinic of the Medical Genetics Service 2.3. Responsiveness to BH4 of Hospital de Clínicas de Porto Alegre (ATDM-SGM/HCPA), Porto

Alegre, Brazil. At the time this study was conducted, 68 patients Patients were defined as responsive to BH4 if they met at least one with different phenotypes were followed up in the ATDM-SGM/HCPA of the criteria listed below: and 64 of them underwent dietary treatment. Criterion A: this criterion used Phe values at T1 [3 h after Phe loading This study was approved by the Ethics in Research Committee of HCPA, and all patients or their guardians signed a written informed in Tests 1 (1T1) and 2 (2T1)] and at T2 [11 h after Phe loading in consent term. Tests 1 (1T2) and 2 (2T2) and 8 h after BH4 administration in Test 2]. The following equation was used for calculations: [((2T2− 2.1. Patients 2T1)/2T1)×100]−[((1T2−1T1)/ 1T1)×100]. Individuals were responsive if the values found corresponded to a Patients included in the study should be aged ≥4yearsand reduction of ≥30% in Phe levels in Test 2. should be under dietary treatment; median Phe plasma levels Criterion B: this criterion used Phe values at T1 and T3 [27 h after should be ≤600 μmol/L in the 12 months before the start of the Phe loading in Tests 1 (1T3) and 2 (2T3) and 24 h after BH4 ad- study. The Phe cut-off point adopted ensured that patients with ministration in Test 2]. The following equation was used for calcu- the mild form of the disease, good metabolic control, or both lations: [((2T3−2T1)/2T1)×100]−[((1T3−1T1)/1T1)×100]. were also included in the test. A previous trial conducted by our study team [23] adopted a different protocol and different inclu- Individuals were responsive if the values found corresponded to a reduction of ≥30% in Phe levels in Test 2. sion criteria to evaluate responsiveness to BH4, and most patients had the classical form of the disease and inadequate metabolic Criterion C: this criterion used the percentage difference of the control, as their Phe plasma levels had to be ≥360 μmol/L in all value found for the area under the Phe curve in Tests 1 (AUC1) measurements during the previous 12 months. and 2 (AUC2). The following equation was used for calculations: Exclusion criteria were: pregnancy, clinical signs suggestive of [((AUC2−AUC1)/AUC1)×100]. Individuals were responsive if liver disease; use of levodopa; allergy to any component of BH4;median the difference was ≥30%, as long as the Test 1 area was greater N μ Phe level 600 mol/L in the measurements made in the 12 months than the Test 2 area. before inclusion in the study; irregular follow-up in the ATDM-SGM/ HCPA in the same 12 months; and probable non-compliance with To compare the classification of responsiveness, four additional study procedures according to evaluations made by the authors. criteria were used, as described below: PKU types were defined according to Nalin et al. [6], and the patients were classified as having classical PKU or mild PKU. Criterion D: only the Phe values at T1 and T2 of Test 2 were used. The following equation was used for calculations: [((2T2 −2T1)/2T1)× 2.2. Single PheFOR and combined Phe +ELECTRONICBH4 loading tests 100]. Individuals USE were responsive ONLY if the values found corresponded to a reduction of ≥30% in Phe levels in time point 2. The patients were asked to come to two visits in HCPA and to stay Criterion E: only the Phe values at T1 and 3 of Test 2 were used. The under evaluation for 27 h each time; the two visits were made at a following equation was used for calculations: [((2T3−2T1)/2T1)× one-week interval. 100]. Individuals were responsive if the values found corresponded to a reduction of ≥30% in Phe levels in time point 3. 2.2.1. Simple Phe loading (Test 1) Criterion F: this criterion was used by the authors in a previ- In the first week, after overnight fasting, blood was collected to ous study [23] to evaluate the responsiveness of 5 patients measure Phe and Tyr plasma concentrations (T0). After that, patients fi ingested 100 mg/kg of L-Phe and resumed their usual diet (Phe- also included in this study, and was de ned as a reduction ≥ restricted diet and supplementation with Phe-free metabolic formula). of 30% in Phe levels 8 h after simple BH4 loading (single- Blood for Phe and Tyr was then collected at 3 (T1), 11 (T2) and 27 h dose of BH4 at 20 mg/kg, without a concomitant load of Phe or (T3) after Phe loading. L-Phe). S82 T. Nalin et al. / Molecular Genetics and Metabolism 104 (2011) S80–S85

Criterion G: this criterion was used by the authors in a previous 3. Results study [23] to evaluate the responsiveness of 5 patients also included in this study, and was defined as a reduction of ≥30% in Phe levels Eighteen patients (11 girls and 7 boys) from 16 nonrelated families were included in the study. Parental consanguinity was found in 1/16 24 h after simple BH4 loading. (6.25%) family. Median patient age was 12 (interquartile range: 10– 16.5) years; minimum age was 6, and maximum, 31 years. 2.4. Dietary intake of Phe Mean Phe levels in Tests 1 and 2 at T0 were 255±138 and 333± μ Phe intake was evaluated using food recalls on the day before and 173 mol/L (n=18; p=0.342), and at T1, 730±221 and 790±310 fi on the first day of Tests 1 and 2, which totaled, therefore, two recalls (n=18; p=1.0). In both Tests 1 and 2 there were signi cant increases b for each Test. Dietary Phe intake was calculated using the nutrition- in Phe levels between T0 and T1 (p 0.01). There were no statistically fi support software NutriBase (NB7), Clinical Edition. All patients signi cant differences in Phe:Tyr ratios when the same time points received instructions to keep the same dietary Phe prescription that were compared between Tests 1 and 2. they followed before the beginning of the study. BH4 plasma levels in Test 2 increased after L-Phe administration (T1) and BH4 loading (T2) (data not reported).

2.5. Genotype 3.1. Response to BH4 Genotypes of patients 1 to 17 (Table 1) were established previously and retrieved from their medical charts. Patient 18 was the only patient Six patients (33.3%) were responsive, four according to criterion A, who had not been genotyped at the time of the study. three according to criterion B, and four according to criterion C (Table 2). The comparison of BH4-responsiveness according to criterion used is shown in Table 3. In the calculation of the Kappa Index, 2.6. Statistical analysis the data reported by Giugliani et al. [23] were not included because of the small sample size (5/18 patients included in this study). The Statistical Package for Social Sciences 18.0 (SPSS® Inc., Chicago, Kappa was 0.557 when criteria A+B were compared with criterion IL) was used for statistical analysis. Data were described using absolute C (p=0.017). For the other criteria, Kappa values indicated a weak and relative frequencies. The Stata program was used to calculate the agreement. area under the Phe curve. Continuous variables were expressed as For responsive patients, mean Phe level variations between time mean±standard deviation or median and interquartile range. The points T1 and T2 and between time points T1 and T3 were different Shapiro–Wilk test was used to analyze variable distribution. only in Test 2 (p=0.002 and 0.011, respectively), that difference The Student t test was used to compare Phe intake between the was not found for non-responsive patients (Table 2). The analysis different food recalls, the area under the Phe curve, and the difference of area under the curve (AUC) revealed a statistically signiﬁcant in Phe levels between the collection time points for responsive pa- difference between Tests 1 and 2 mean values for responsive patients versus non-responsive patients. Phe plasma concentrations tients (pb0.01), but not for non-responsive patients (p=0.242) and Phe:Tyr ratios were analyzed using Generalized Estimating (Table 2).

Equations and Bonferroni correction. The comparison between BH4 Fig. 1 shows the Phe plasma concentration values found in this responsiveness criteria was made using kappa statistics. study according to collection time point in Tests 1 and 2 for a BH4- The level of signiﬁcance was set at 5%. responsive patient (Fig. 1A) and a non-responsive patient (Fig. 1B).

Table 1 Genotype, PKU phenotype for each patient and comparison of responsiveness in this study with ﬁndings in the literature.

a c Patient Type of Genotype Responsiveness to BH4 Responsiveness to BH4 Protocol used in the literature PKU (our study) (number of patients included)b Allele 1 Allele 2

1 Classical p.I65T p.R408W R R (2) BH4(10 mg) — 24 h d 2 Classical p.V388M p.V388M R R (1) — Slow R (1) BH4(20 mg) — 24 h 3 Classical p.I65T p.R176X NR Nd Nd 4 Classical p.R252W p.R261Q NR Nd Nd

5 Classical p.165T p.I65T NR NR (1) Phe(100 mg)/BH4(20 mg) — 8h 6 Mild p.E390G p.R408Q R Nd Nd 7 Classical p.R261X p.R176X NR Nd Nd 8 Mild p.R408W c.165delT NR Nd Nd 9FOR Mild p.R408W ELECTRONIC c.165delT NR USE Nd ONLY Nd 10 Mild p.E390G p.A104D R Nd Nd 11 Mild p.R408W p.L348V NR Nd Nd 12 Mild p.R408W p.L348V NR Nd Nd 13 Mild p.A300S p.L311P R Nd Nd d 14 Classical p.R408W p.R408W NR NR(13) Slow R (1) BH4(20 mg) — 24 h (13)/Other (1)

15 Mild p.R261Q p.V388M R R (1) BH4(10 mg) — 8h

16 Mild p.L249F p.V388M NR NR (1) BH4(20 mg) — 24 h 17 Classical c.165delT ? NR –– 18 Mild ? ? NR ––

PKU: phenylketonuria; Nd: not described; R: Responsive; NR: non-responsive; Siblings: 8 and 9; 11 and 12. a To be classiﬁed as responsive, patient had to obtain a positive result according to at least one of the criteria used (criteria A, B and C). b Data retrieved from BIOPKUdb http://www.bh4.org/BH4DatabasesBioHPA-PAH.asp. c Loading protocol used in other PKU patients with the same genotype (amount of BH4 per kg of current weight; time used for responsiveness criterion). d Patients with a Phe reduction of 20 to 30% after BH4 loading. T. Nalin et al. / Molecular Genetics and Metabolism 104 (2011) S80–S85 S83

Table 2

BH4 responsiveness: Phe values and corresponding areas under the curve (n=18 patients).

Patients Test 1 Test 2 Area under Phe curve Criterion A Criterion B Critérion C (%) (%) (%) Phe (μmol/L) Phe (μmol/L) AUC1 AUC2

T1 T2 T3 T1 T2 T3

Responsive 1 852 1032 949 1010 823 816 373.1 289.8 −39.6 −30.6 −22.3 2 1193 1314 849 939 895 639 435.2 297.2 −14.8 −3.1 −31.7 6 588 474 136 570 144 74,8 133.8 27.9 −55.3 −10 −79.1 10 522 372 113 612 127 58,4 103.1 19.3 −50.5 −12.1 −81.3 13 744 602 550 748 516 189 207.1 145.2 −11.9 −48.6 −29.9 15 657 822 589 819 498 355 281.6 153.1 −64.3 −46.3 −45.6 ⁎ ⁎ ⁎ ⁎⁎ ⁎⁎ μ±sd 759±242 769±358 531±349 783±175 500±324 355±312 255.6±132.0 155.4±120.9 −39.4±21.7 −25.1±19.5 −48.3±25.8

Non-responsive 3 1095 750 706 991 946 808 253.2 327.3 27 17 29.3 4 702 555 462 496 553 390 185.9 186.6 32.4 12.8 0.4 5 517 427 530 652 661 494 161.1 223.4 18.7 −26.7 38.6 7 816 775 617 634 696 542 263.5 239.5 14.8 9.9 −9.1 8 367 425 390 699 726 715 152.9 263.0 −11.9 −3.9 72.0 9 544 626 486 742 736 730 216.4 265.8 −15.8 9 22.8 11 836 655 613 608 513 418 225.7 172.6 6 −4.5 −23.5 12 991 925 717 1095 855 660 311.9 280.7 −15.2 −12 −10.0 14 711 562 296 587 421 197 171.3 123.6 −7.3 −8 −27.8 16 789 640 446 714 637 422 206.6 206.4 8.1 2.5 −0.1 17 790 810 799 1810 1787 1398 293.2 608.1 −3.8 −23.9 107.4 18 422 350 227 489 304 208 111.7 93.1 −20.8 −11.2 −16.7 μ±sd 715±220 625±172 524±174 793±367 736±375 582±323 212.8±60.0 249.2±131.0 2.6±17.8 −3.2±14 15.3±41.1

Phe: phenylalanine; Test 1: simple Phe loading; Test 2: combined Phe+BH4 loading; Siblings: 8 and 9; 11 and 12. Criterion A: [((2T2−2T1)/2T1)×100]−[((1T2−1T1)/1T1)×100]. Criterion B: [((2T3−2T1)/2T1)×100]−[((1T3−1T1)/1T1)×100]. Criterion C: [((AAC2−AAC1)/AAC1)×100]. To be classiﬁed as responsive, patient had to obtain a positive result according to at least one of the criteria used (criteria A, B and C).

T1 — time point 1: Phe level 3 h after Phe loading; T2 — time point 2: Phe level 11 h after Phe loading and 8 h after BH4 loading (Test 2); T3 — time point 3: Phe level 27 h after Phe loading and 24 h after BH4 loading (Test 2); AUC1: Area under the curve in Test 1; AUC2: Area under the curve in Test 2. ⁎ Phe variation between T2 and T1 and between T3 and T1 was statistically signiﬁcant (p=0.002 and 0.011). ⁎⁎ Statistically signiﬁcant difference (pb0.01).

3.2. Association between genotype and responsiveness to BH4 (Table 1) this type of loading tend to spontaneously decrease in 24 h even when BH4 is not been administered [20,32]. Therefore, BH4 respon- Data about the genotype, severity of PKU and BH4-responsiveness siveness would have to be conﬁrmed by comparing Phe plasma levels are shown in Table 1. after the Phe+BH4 loading test and after the simple Phe loading test. We have found in the literature only two studies performing the sim-

3.3. Dietary intake ple Phe loading test and the combined Phe+BH4 loading test in the

The patients ingested in average 705.56±356.77 and 608.89± 405.51 mg (mean±SD) of Phe per day, according to dietary recalls Table 3 ﬁ 1 and 2 of Test 1 (p=0.446), and 672.78±406.17 and 572.78± BH4 responsiveness according to criteria used for classi cation in this study.

329.47 mg (mean±SD) of Phe per day according to dietary recalls 1 Patients Criterion Criterion Criterion Criterion Criterion Criterion Criterion and 2 of Test 2 (p=0.281). There were no statistically significant dif- A B C D E Fa Ga ferences in Phe intake between Tests 1 (657.2±379.6 mg Phe/day) 1 R R NRNRNRNRR and 2 (622.7±368 mg Phe/day) (p=0.564). 2 NR NR R NR R –– 3 NRNRNRNRNRR R –– 4. Discussion 4 NRNRNRNRNR 5 NRNRNRNRNR –– 6 RNRRRR –– fi –– In this study, patients with PKU that might potentially bene t 7 NRNRNRNRNR –– from the use of BH4 were identified by comparing their Phe levels 8 NRNRNRNRNR FOR ELECTRONIC USE ONLY–– after a simple Phe loading test and after a combined Phe+BH load- 9 NRNRNRNRNR 4 –– ing test. Although no consensus has been reached in the literature 10RNRRRR 11 NR NR NR NR R R R about the most adequate method to identify these individuals, Phe 12 NR NR NR NR R NR NR plasma levels should be elevated at the time of BH4 administration 13 NR R NR R R –– to induce an increase in PAH activity and, consequently, to potentiate 14 NR NR NR NR R –– 15RRRRR–– the effect of BH4 [27]. This recommendation adds complexity to the investigation of BH responsiveness in patients with good adherence 16 NR NR NR NR R NR NR 4 17 NR NR NR NR NR –– to dietary treatment. The increase of Phe levels is often achieved by 18 NR NR NR R R –– increasing the ingestion of dietary Phe [8,9,20,22,28–30], which may R=responsive; NR=non-responsive. raise ethical and psychological issues. A single Phe dose using L-Phe The criteria are defined in the Material and methods section of this study. may also be administered, with the advantage that the patient's diet Siblings: 8 and 9; 11 and 12. does not have to be changed [19,31–33]; however, Phe levels after a Data reported by Giugliani et al. [23]. S84 T. Nalin et al. / Molecular Genetics and Metabolism 104 (2011) S80–S85

Although the protocol here described has some limitations, e.g. the need for patients to attend the treatment center twice (two consecutive weeks) and to undergo a higher number of blood collections, the superiority of this protocol when compared with others already described in the literature can only be assessed if comparative studies are carried out. This underscores the need for determining the sensitivity, speciﬁcity, as well as positive and negative predictive values of these tests, especially if we take into consideration that, as our

ﬁndings clearly show, patients diagnosed as BH4-responsive according to one given protocol may not be diagnosed as such in another protocol. In this sense, one of the future goals of the authors is to

test simple BH4 loading in all patients included in the present study.

4.1. Comments on dietary ingestion and Phe levels during the Tests

Dietary Phe ingestion did not change in the two recalls in the same Test, and there was also no variation in the comparison of total Phe amount ingested in Test 1 and in Test 2, which suggests that the fall in

Phe levels in responsive individuals was secondary to BH4 administration. The variation of Phe plasma concentrations along the study revealed an elevation at time point T1 in comparison with time point T0, that is, 3 h after Phe loading, which demonstrates, therefore, that L-Phe was absorbed by the patient. The collection point 3 h after Phe loading was used because it has been described in the literature as the point at which Phe plasma levels peak predominantly [24]. However, in 9 patients in this study, the highest Phe plasma concentration values occurred at least, in one of the Tests, 11 h after Phe loading, which has also been reported by other studies [19,20]. Phe plasma concentrations at time points 0 and 1 in the two Tests Fig. 1. (A) Phe variation in a patient classified as responsive to BH4 according to collection time points. (B) Phe variation in a patient classified as non-responsive to BH4 did not differ from each other, which emphasizes their comparability. according to collection time points. 4.2. Conclusions same patient. The one conducted by Desviat et al. [20] included six BH has emerged as a new treatment for patients with mostly patients with Phe level at diagnosis below 360 μmol/L, but classified 4 milder forms of PKU and may substantially improve their quality of their responsiveness only according to the combined Phe+BH4 test; life. Numerous positive findings, including increases in Phe tolerance, and the one conducted by Ponzone et al. [32], which included seven have been reported in association with the use of this medication. patients with different forms of PKU, but did not establish clear However, no consensus has been reached about the best method criteria for the classification of responsiveness. The analysis of Phe and criteria to define responsiveness to BH . The validation of curves in both studies suggests that Phe plasma levels reach similar 4 methods and criteria for this purpose is fundamental to optimize values at the end of the simple Phe and combined Phe+BH4 loading the treatment with BH4 also in terms of cost and effectiveness. Our tests; although, Phe levels seem to decrease faster when BH4 is used. data suggest that, in responsive individuals, Phe levels decrease faster The protocol to evaluate responsiveness in this study is innovative after Phe+BH loading than after simple Phe loading, and confirm for two reasons: (1) it included the use of a simple Phe and a combined 4 that the comparison between simple Phe and combined Phe+BH4 Phe+BH4 loading test for 18 patients with PKU and good metabolic loading is valid to evaluate responsiveness. Moreover, because of control; (2) it defined BH4 responsiveness parameters based on the the wide range of variation of responsiveness classification for each comparison of results of these two tests. According to the strategy patient, more than one criterion should be used to establish a defini- adopted in this study, 33.3% (6/18) of the patients were classified as tion, and these criteria should take into consideration the comparison responsive, which is in agreement with data in the literature [17– between the values obtained in single Phe and combined Phe+BH 19,23,30,34]. Of the responsive patients, two had the classical PKU 4 loading tests. (n=2/8; 25.5%) and four, the mild PKU (n=4/10; 40%), which confirmed that responsiveness is greater among individuals with milder forms of PKU. Moreover, of the 16 patients for whom genotypes Acknowledgments were available,FOR six had data about ELECTRONIC BH4-responsiveness genotypes USE ONLY described in the literature. The authors thank the following: Halfway House, Statistic Unit, FIPE There was also an important discordance between the seven criteria and the staff of the Medical Genetics Service at HCPA. They also thank used to compare the definition of responsiveness. In the group of 18 the Brazilian Coordinating Agency for Advanced Training of Graduate individuals, only one was responsive and six were non-responsive Personnel (CAPES) and Merck Serono for their support and collabora- according to all criteria. If criteria 4 and 5, which consider the results tion in this study. This work was supported in part by the Swiss National of combined Phe+BH4 loading only, had not been used, a greater num- Science Foundation grant no. 3100A0-1199852/1 (to NB). ber of patients would have concordant results for non-responsiveness (n=10). The use of criteria 4 and 5 independently seems to lead to a References greater number of diagnoses of patients responsive to BH4;thismay be explained by the fact that these criteria do not take into consider- [1] C. Scriver, S. Kaufman, Hyperphenylalaninemia: phenylalanine hydroxylase deficiency, in: C. Scriver, A. Beaudet, W. Sly, D. Valle, B. Childs, B. Vogelstein (Eds.), ation that Phe may be excreted spontaneously and not due to BH4 The metabolic and molecular bases of inherited disease, McGraw-Hill, New action. York, 2001, pp. 1667–1724. T. Nalin et al. / Molecular Genetics and Metabolism 104 (2011) S80–S85 S85

[2] National Institutes of Health Consensus Development Conference Statement: [19] B. Perez-Duenas, M.A. Vilaseca, A. Mas, N. Lambruschini, R. Artuch, L. Gomez, J. phenylketonuria: screening and management, October 16–18, 2000 Pediatrics, Pineda, A. Gutierrez, M. Mila, J. Campistol, Tetrahydrobiopterin responsiveness 108, 2001, pp. 972–982. in patients with phenylketonuria, Clin. Biochem. 37 (2004) 1083–1090. [3] C.N. Sarkissian, A. Gamez, C.R. Scriver, What we know that could influence future [20] L.R. Desviat, B. Perez, A. Belanger-Quintana, M. Castro, C. Aguado, A. Sanchez, M.J. treatment of phenylketonuria, J. Inherit. Metab. Dis. 32 (2009) 3–9. Garcia, M. Martinez-Pardo, M. Ugarte, Tetrahydrobiopterin responsiveness: results [4] N. Blau, F.J. van Spronsen, H.L. Levy, Phenylketonuria, Lancet 376 (2010) 1417–1427. of the BH4 loading test in 31 Spanish PKU patients and correlation with their geno- [5] J.H. Walter, F.J. White, S.K. Hall, A. MacDonald, G. Rylance, A. Boneh, D.E. Francis, type, Mol. Genet. Metab. 83 (2004) 157–162. G.J. Shortland, M. Schmidt, A. Vail, How practical are recommendations for die- [21] N. Blau, A. Belanger-Quintana, M. Demirkol, F. Feillet, M. Giovannini, A. MacDonald, tary control in phenylketonuria? Lancet 360 (2002) 55–57. F.K. Trefz, F. van Spronsen, Management of phenylketonuria in Europe: survey [6] T.Nalin,I.Perry,L.Refosco,C.Netto,C.Souza,T.Vieira,P.Picon,I.Schwartz,Fenilcetonúria results from 19 countries, Mol. Genet. Metab. 99 (2010) 109–115. no Sistema Único de Saúde: Avaliação de Adesão ao Tratamento em um Centro de [22] F.K. Trefz, B.K. Burton, N. Longo, M.M. Casanova, D.J. Gruskin, A. Dorenbaum, E.D. Atendimento do, Rio Grande do Sul Revista HCPA 30 (2010) 222–232. Kakkis, E.A. Crombez, D.K. Grange, P. Harmatz, M.H. Lipson, A. Milanowski, L.M. [7] K.T. Friedrich, K.B. Barbara, L. Nicola, C. Mercedes Martinez-Pardo, J.G. Daniel, D. Randolph, J. Vockley, C.B. Whitley, J.A. Wolff, J. Bebchuk, H. Christ-Schmidt, J.B. Alex, D.K. Emil, A.C. Eric, K.G. Dorothy, H. Paul, H.L. Mark, M. Andrzej, R. Linda Hennermann, Efficacy of sapropterin dihydrochloride in increasing phenylalanine Marie, V. Jerry, B.W. Chester, A.W. Jon, B. Judith, C.-S. Heidi, B.H. Julia, Efficacy of tolerance in children with phenylketonuria: a phase III, randomized, double- sapropterin dihydrochloride in increasing phenylalanine tolerance in children blind, placebo-controlled study, J. Pediatr. 154 (2009) 700–707. with phenylketonuria: a phase III, randomized, double-blind, placebo-controlled [23] L. Giugliani, A. Sitta, C.R. Vargas, L.C.S. da Silva, T. Nalin, M.L. Saraiva-Pereira, R. study, J. Pediatr. 154 (2009) 700–707.e701. Giugliani, I.V.D. Schwartz, Responsividade à tetrahidrobiopterina em pacientes [8] S. Kure, D.C. Hou, T. Ohura, H. Iwamoto, S. Suzuki, N. Sugiyama, O. Sakamoto, K. brasileiros com deficiência de fenilalanina hidroxilase, J. Pediatr. 87 (2011) Fujii, Y. Matsubara, K. Narisawa, Tetrahydrobiopterin-responsive phenylalanine 245–251. hydroxylase deficiency, J. Pediatr. 135 (1999) 375–378. [24] N. Blau, B. Fiege, F. Trefz, Tetrahydrobiopterin-responsive phenylalanine hydroxy- [9]A.C.Muntau,W.Roschinger,M.Habich,H.Demmelmair,B.Hoffmann,C.P. lase deficiency: diagnosis, treatment, genetics, and international BIOPKU database, Sommerhoff, A.A. Roscher, Tetrahydrobiopterin as an alternative treatment in: N. Blau, B. Thfny (Eds.), Pterins, Folates and Neurotransmitters in Molecular for mild phenylketonuria, N. Engl. J. Med. 347 (2002) 2122–2132. Medicine, Verlagsgesellschaft mbH, Heilbronn,Germany, 2004, pp. 132–142. [10] L.J. Spaapen, M.E. Rubio-Gozalbo, Tetrahydrobiopterin-responsive phenylalanine [25] M.S. Rashed, P.T. Ozand, M.P. Bucknall, D. Little, Diagnosis of inborn errors of me- hydroxylase deficiency, state of the art, Mol. Genet. Metab. 78 (2003) 93–99. tabolism from blood spots by acylcarnitines and amino acids profiling using auto- [11] N. Blau, H. Erlandsen, The metabolic and molecular bases of tetrahydrobiopterin- mated electrospray tandem mass spectrometry, Pediatr. Res. 38 (1995) 324–331. responsive phenylalanine hydroxylase deficiency, Mol. Genet. Metab. 82 (2004) [26] T. Opladen, B. Abu Seda, A. Rassi, B. Thony, G.F. Hoffmann, N. Blau, Diagnosis of tetra- 101–111. hydrobiopterin deficiency using filter paper blood spots: further development of the [12] B. Fiege, L. Bonafe, D. Ballhausen, M. Baumgartner, B. Thony, D. Meili, L. Fiori, M. method and 5 years experience, J. Inherit. Metab. Dis. 34 (2011) 819–826. Giovannini, N. Blau, Extended tetrahydrobiopterin loading test in the diagnosis [27] N. Blau, B.K. Burton, B. Thöny, F.J.v. Sprosen, S. Waisbren, Phenylketonuria and of cofactor-responsive phenylketonuria: a pilot study, Mol. Genet. Metab. 86 BH4 Deficiencies, UNI-MED, Bremen, 2010. (2005) S91–S95. [28] J.B. Nielsen, K.E. Nielsen, F. Guttler, Tetrahydrobiopterin responsiveness after ex- [13] G. Gramer, P. Burgard, S.F. Garbade, M. Lindner, Effects and clinical significance of tended loading test of 12 Danish PKU patients with the Y414C mutation, J. Inherit. tetrahydrobiopterin supplementation in phenylalanine hydroxylase-deficient Metab. Dis. 33 (2010) 9–16. hyperphenylalaninaemia, J. Inherit. Metab. Dis. 30 (2007) 556–562. [29] B.K. Burton, H. Bausell, R. Katz, H. Laduca, C. Sullivan, Sapropterin therapy in- [14] U. Langenbeck, Classifying tetrahydrobiopterin responsiveness in the hyperphe- creases stability of blood phenylalanine levels in patients with BH4-responsive nylalaninaemias, J. Inherit. Metab. Dis. 31 (2008) 67–72. phenylketonuria (PKU), Mol. Genet. Metab. 101 (2010) 110–114. [15] G. Gramer, S.F. Garbade, N. Blau, M. Lindner, Pharmacokinetics of tetrahydrobiop- [30] A. Bélanger-Quintana, M.J. García, M. Castro, L.R. Desviat, B. Pérez, B. Mejía, M. terin following oral loadings with three single dosages in patients with phenylke- Ugarte, M. Martínez-Pardo, Spanish BH4-responsive phenylalanine hydroxylase- tonuria, J. Inherit. Metab. Dis. 32 (2009) 52–57. deficient patients: evolution of seven patients on long-term treatment with tetra- [16] C.O. Harding, New era in treatment for phenylketonuria: pharmacologic therapy hydrobiopterin, Mol. Genet. Metab. 86 (2005) 61–66. with sapropterin dihydrochloride, Biologics 4 (2010) 231–236. [31] A. Baldellou Vázquez, M.I. Salazar García-Blanco, M.P. Ruiz-Echarri Zalaya, C. [17] B.K. Burton, D.K. Grange, A. Milanowski, G. Vockley, F. Feillet, E.A. Crombez, V. Campos Calleja, L. Ruiz Desviat, M. Ugarte Pérez, Tratamiento de la hiperfenilala- Abadie, C.O. Harding, S. Cederbaum, D. Dobbelaere, A. Smith, A. Dorenbaum, ninemia por déficit de fenilalanina hidroxilasa con tetrahidrobiopterina. ¿Cuándo The response of patients with phenylketonuria and elevated serum phenylalanine y cómo? Anales de Pediatría 64 (2006) 146–152. to treatment with oral sapropterin dihydrochloride (6R-tetrahydrobiopterin): a [32] A. Ponzone, F. Porta, A. Mussa, A. Alluto, S. Ferraris, M. Spada, Unresponsiveness to phase II, multicentre, open-label, screening study, J. Inherit. Metab. Dis. 30 (2007) tetrahydrobiopterin of phenylalanine hydroxylase deficiency, Metabolism 59 (2010) 700–707. 645–652. [18] H.L. Levy, A. Milanowski, A. Chakrapani, M. Cleary, P. Lee, F.K. Trefz, C.B. Whitley, F. [33] J.J. Mitchell, B. Wilcken, I. Alexander, C. Ellaway, H. O'Grady, V. Wiley, J. Earl, J. Feillet, A.S. Feigenbaum, J.D. Bebchuk, H. Christ-Schmidt, A. Dorenbaum, Efficacy of Christodoulou, Tetrahydrobiopterin-responsive phenylketonuria: the New sapropterin dihydrochloride (tetrahydrobiopterin, 6R-BH4) for reduction of phenyl- South Wales experience, Mol. Genet. Metab. 86 (2005) S81–S85. alanine concentration in patients with phenylketonuria: a phase III randomised [34] B. Fiege, N. Blau, Assessment of tetrahydrobiopterin (BH4) responsiveness in phe- placebo-controlled study, Lancet 370 (2007) 504–510. nylketonuria, J. Pediatr. 150 (2007) 627–630.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S86–S92

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Phenylalanine hydroxylase deﬁciency: Molecular epidemiology and predictable

BH4-responsiveness in South Portugal PKU patients Isabel Rivera a,b,⁎, Dina Mendes a, Ângela Afonso a, Madalena Barroso a, Ruben Ramos a, Patrícia Janeiro c, Anabela Oliveira d, Ana Gaspar c, Isabel Tavares de Almeida a,b a Metabolism and Genetics Group, iMed.UL-Research Institute for Medicines and Pharmaceutical Sciences, Faculty of Pharmacy, University of Lisbon, Portugal b Department of Biochemistry and Human Biology, Faculty of Pharmacy, University of Lisbon, Portugal c Department of Paediatrics, Santa Maria Hospital, Lisbon, Portugal d Department of Medicine, Santa Maria Hospital, Lisbon, Portugal article info abstract

Article history: Hyperphenylalaninemia (HPA, OMIM #261600), which includes phenylketonuria (PKU), is caused by Received 2 June 2011 mutations in the gene encoding phenylalanine hydroxylase (PAH), being already described more than 600 Received in revised form 27 July 2011 different mutations. Genotype–phenotype correlation is a useful tool to predict the metabolic phenotype, to Accepted 28 July 2011 establish the better tailored diet and, more recently, to assess the potential responsiveness to BH therapy, a Available online 31 July 2011 4 current theme on PKU field. The aim of this study was the molecular analysis of the PAH gene, evaluation of genotype–phenotype Keywords: Phenylketonuria relationships and prediction of BH4-responsiveness in the HPA population living in South Portugal. We Hyperphenylalaninemia performed the molecular characterization of 83 HPA patients using genomic DNA extracted from peripheral Tetrahydrobiopterin blood samples or Guthrie cards. PAH mutations were scanned by PCR amplification of exons and related Genotype intronic boundaries, followed by direct sequence analysis. Intragenic polymorphisms were determined by Phenotype PCR-RFLP analysis. The results allowed the full characterization of 67 patients. The mutational spectrum encompasses 34 distinct mutations, being the most frequent IVS10nt-11GNA (14.6%), V388M (10.8%), R261Q (8.2%) and R270K (7.6%), which account for 46% of all mutant alleles. Moreover, 12 different haplotypes were identified and most mutations were associated with a single one. Notably, more than half of the 34 mutations belong to the group

of more than 70 mutations already identiﬁed in BH4-responsive patients, according to BIOPKU database. Fifty

one different genotypic combinations were found, most of them in single patients and involving a BH4- responsive mutation. In conclusion, a signiﬁcant number (30–35%) of South Portugal PKU patients may

potentially beneﬁt from BH4 therapy which, combined with a less strict diet, or eventually in special cases as monotherapy, may contribute to reduce nutritional deﬁciencies and minimize neurological and psychological dysfunctions. © 2011 Elsevier Inc. All rights reserved.

1. Introduction PAH is responsible for the conversion, in the presence of the cofactor

tetrahydrobiopterin (BH4) and dioxygen, of phenylalanine into tyrosine, Hyperphenylalaninemia (HPA, OMIM #261600), which includes which becomes an essential amino acid when the hydroxylating activity phenylketonuria (PKU), is the most common inborn error of amino acid is absent or impaired. Then a simultaneous elevation in phenylalanine metabolism with an average incidence among Caucasian and Oriental levels can be observed altogether with an impairment in tyrosine ones Asian populations of 1 per 10,00 newborns, while the mutant allele and both events contribute to the features characterizing the metabolic frequency in theFOR population is polymorphic ELECTRONIC (~0.01). This pathology and clinical phenotypes USE displayed by theONLY patients. displays autosomal recessive inheritance and its cause is multifactorial: PKU and related HPA represent the paradigm of a genetic disease that mutations in the gene encoding phenylalanine hydroxylase (PAH; EC can be treated. For the last 50 years, the traditional treatment has been a 1.14.16.1) and exposition to dietary phenylalanine are both necessary phenylalanine restricted diet for all lifelong with all the problems and sufﬁcient conditions to trigger it [1]. associated, namely nutritional deﬁcits and socialization troubles, just to mention two different aspects [1–3]. However, some new approaches are being tried, as the use of large neutral amino acids and macroglycopeptide, ⁎ Corresponding author at: Metabolism and Genetics Group, iMed.UL-Research as well as enzyme replacement therapy with phenylalanine ammonia Institute for Medicines and Pharmaceutical Sciences, Faculty of Pharmacy, University of Lisbon, Av. Prof. Gama Pinto, 1649-003 Lisbon, Portugal. Fax: +351 21 7946491. lyase, but these therapeutic roads are not yet fully effective [3]. E-mail address: [email protected] (I. Rivera). Additionally, more sophisticated approaches, like chaperone and gene

1096-7192/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2011.07.026 I. Rivera et al. / Molecular Genetics and Metabolism 104 (2011) S86–S92 S87 therapies, are still far in the horizon. Recently, however, a new therapy is 2.2. Genotype analysis becoming widely used, the supplementation with pharmacological doses of the PAH cofactor, which can alleviate, and in special cases to avoid, the Genomic DNA was isolated from peripheral blood samples or from diet burden [4–7]. The molecular basis of BH4 action is not fully Guthrie cards according to a salting-out procedure (Puregene Cell and understood, but the effect upon PAH gene expression was discarded. Tissue kit, Gentra Systems, Minneapolis, MN, USA). After PCR

BH4 can act as a chemical chaperone helping the stabilization of some amplification of individual exons and related intronic boundaries, mutant forms or, by increasing the cofactor concentration, it can promote PAH gene (GenBank accession no. AF404777) was scanned for the enzymatic activity of low-affinity mutants [8,9]. Accordingly, BH4 mutations by direct sequence analysis, using the ABI Prism BigDye seems to display a multifactorial mechanism of action. Terminator Cycle Sequencing Ready Reaction Kit, in an ABI PRISM 310 PAH gene located in the long arm of chromosome 12 (12q24.1), covers Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). ~100 kb of genomic DNA and is structured in 13 exons separated by introns. The messenger is 2448 bp long, being translated into 452 amino 2.3. Haplotype analysis acid polypeptides, which are assembled onto functional homotetramers [1]. More than 600 different mutations have already been described Mini-haplotypes were established after PCR-RFLP analysis of the (PAHdb, www.pahdb.mcgill.ca), being the majority missense ones, and intragenic bi-allelic polymorphisms BglII, PvuIIa, PvuIIb, MspI and every population displays a mutational spectrum characterized by a XmnI and of the multi-allelic VNTR system at HindIII site at the 3′ reduced number of prevalent and public mutations and a large number of region of PAH gene [14]. Haplotype numbering followed the rules of private pathogenic alterations. Moreover, each mutation impairs PAH Eisensmith and Woo [15]. activity in a specific manner and, as a consequence of most patients being compound heterozygotes, phenotypes range from the most severe form of 2.4. Calculation of homozigosity (j) classic PKU to the non-PKU hyperphenylalaninemia status.Moreover, several polymorphic markers (RFLP, VNTR and STR) have been defined Homozigosity (j) at the PAH locus in a given population is 2 within the PAH gene. These intragenic polymorphisms have proven to be a determined by j=∑xi , where xi is the frequency of the ith allele. useful guide to mutation detection and, besides, have been extensively In our population, where ascertainment of mutations was not 100%, used in molecular anthropology studies to improve our understanding of each of the uncharacterized alleles was defined as having a frequency the ancestral migratory movements which underlie the present geo- of 1/N, where N is the total number of mutant chromosomes graphic distribution of the most frequent PAH gene mutant alleles [10,11]. investigated. As PAH enzyme displays a hepatic expression and liver needle biopsies are no longer justified in PKU patients, the primary source of information 2.5. Relative residual PAH activity and genotype–phenotype correlation concerning residual enzyme activity relies on in vitro expression and analysis of recombinant mutant proteins [12]. The data thus obtained is Relative residual PAH activity was calculated from data compiled then used to estimate the patients' phenotype. Genotype–phenotype from PAHdb which displays values calculated from in vitro expression correlations are difficult to realize in most inherited metabolic disorders, of recombinant mutant proteins. PAH activity is defined as the average but in PKU and related HPA they revealed to be a strong and reliable sum of activities of both individual mutant alleles, and expressed as predictive tool [12,13]. Accordingly, the characterization of each patient the percentage of the wild-type enzyme. genotype can greatly help to predict the metabolic phenotype, to establish the better tailored diet and, more recently, to assess the potential 3. Results responsiveness to BH4 therapy. Herein is presented the molecular analysis of the PAH gene, the 3.1. Mutation analysis genotype–phenotype relationships evaluated and the rate of BH4- responsiveness predicted in the HPA population living in South Portugal. This study involved the molecular characterization of 83 HPA patients living in South Portugal, corresponding to 158 mutant alleles. The results revealed a mutational spectrum encompassing 34 distinct 2. Materials and methods mutations which were distributed along the PAH gene sequence (Table 1). Most mutations were nucleotide substitutions corresponding 2.1. Patients and phenotypic classification to 27 missense (79.6%), 3 nonsense (8.8%) and 2 at splicing sites (5.8%); additionally, two deletions were found. Two mutations displayed a Eighty-three patients, encompassing four pairs of siblings and all relative frequency N10% (IVS10nt-11GNA and p.V388M); a group of five living in the Southern region of Portugal, were investigated. Most mutations had a frequency between 3.8 and 8.2% (p.R261Q, p.R270K, p. patients were detected by the newborn screening program, running in P281L, p.I65T and p.R158Q); another group of seventeen mutations had Portugal since 1979, and whose current cut-off value is 180 μMof a frequency in the range 1.3–2.5% and the remaining eight mutations phenylalanine. Diagnosis was confirmed after exclusion of BH4 deficien- were present in only one mutant allele (0.6% each). The majority of the cy by evaluating urinary pterin levels and erythrocyte dihydropteridine mutations were situated in the catalytic domain (85%), while two were reductase activity.FOR ELECTRONIClocated in the regulatoryUSE domain (6%)ONLY and three in the tetramerization Metabolic phenotype was assigned to each patient according to pre- domain (9%). treatment blood phenylalanine concentrations and, when available, A detection rate of 89.9% was achieved, with complete genotyping of dietary tolerance at 5 years of age. Patients were classified as having 67 patients, while in the remaining sixteen individuals only one classic PKU (pre-treatment Phe levels N1200 μM and Phe tolerance causative mutation was identified. Among the fully genotyped patients, b20 mg/kg/day), moderate PKU (pre-treatment Phe levels between 900 only 14 (20.9%) were homozygous, the majority displaying compound and 1200 μM and Phe tolerance between 20 and 25 mg/kg/day), mild heterozigosity for two different mutations. Among the 14 homozygous PKU (pre-treatment Phe levels between 600 and 900 μMandPhe patients, three individuals carried the IVS10nt-11GNA mutation, two tolerance above 25 mg/kg/day) and non-PKU HPA if they keep their Phe individuals carried the p.R261Q mutation and another two the p.R270K levels below 600 μMonafreediet. allele. The remaining seven patients harbored p.I65T, p.R176X, p.E280K, This study was approved by the local Ethics Committee and p.P281L, p.L311P, p.T323del and p.V388M mutations in homozigosity. informed consents were obtained from the patients or from their Moreover, it was interesting to note the wide array of genotypes parents, who were also enrolled in the study. (Table 2). Among the 67 fully genotyped patients we could observe 51 S88 I. Rivera et al. / Molecular Genetics and Metabolism 104 (2011) S86–S92

Table 1 Characterization of PAH mutations identiﬁed in 83 South Portuguese HPA patients.

Sequence variation PAH mutation Alleles Allele frequency PAH activity Mini-Haplotype (RFLP.VNTR) BH4-responsiveness (n) (%) (%)

c.1066-10GNA IVS10nt-11GNA 23 14.6 b1 6.7 − c.1162GNA p.V388M 17 10.8 27 1.7 + c.782GNA p.R261Q 13 8.2 27 1.8 + c.809GNA p.R270K 12 7.6 b1 1.8 − c.842CNT p.P281L 10 6.3 b1 1.8/2.3/16.3 − c.194TNC p.I65T 8 5.1 27 9.8/57.8 + c.473GNA p.R158Q 5 3.8 10 4.3 − c.526CNT p.R176X 4 2.5 b1 1.8 − c.754CNT p.R252W 4 2.5 b1 1.8 − c.967_969delACA p.T323del 4 2.5 b1 1.7/1.8 − c.745CNT p.L249F 3 1.9 ? 1.8 ? c.781CNT p.R261X 3 1.9 b1 5.8 − c.932TNC p.L311P 3 1.9 b1 3.8 − c.204ANT p.R68S 2 1.3 76 1.8 + c.527GNT p.R176L 2 1.3 42 4.3 + c.593_614del22 p.Y198_E205NCfs 2 1.3 0 5.8 − c.728GNA p.R243Q 2 1.3 10 1.8 + c.727CNT p.R243X 2 1.3 b1 7.8 − c.805ANC p.I269L 2 1.3 63 4.3 + c.838GNA p.E280K 2 1.3 b1 1.8 − c.898GNT p.A300S 2 1.3 ? 1.8 + c.1045TNC p.S349P 2 1.3 b1 4.3 − c.1169ANG p. E390G 2 1.3 75 2.3 + c.1241ANG p.Y414C 2 1.3 38 4.3 + c.533ANG p.E178G 1 0.6 ? 1.7 + c.688GNA p.V230I 1 0.6 63 5.9 + c.722GNA p.R241H 1 0.6 23 5.9 + c.926CNT p.A309V 1 0.6 70 1.7 + c.965CNG p.A322G 1 0.6 75 1.9 + c.1042CNG p.L348V 1 0.6 33 9.8 + c.965CNG p.A403V 1 0.6 N70 1.8 + c.1222CNT p.R408W 1 0.6 b1 1.8 − c.1243GNA p.D415N 1 0.6 72 1.8 + c.1315+1GNA IVS12nt+1GNA 1 0.6 b1 2.3 − Unknown 16 10.1 Total 158 100.0

different genotypic combinations: 2 genotypes were found in 3 patients frequent are haplotype 6.7 (13.7%), haplotype 4.3 (10.4%) and haplotype and 12 genotypes were detected in 2 patients, while the remaining 9.8 (4.6%), reflecting the presence of the frequent mutations IVS10nt- genotypic combinations were observed in single patients. Interestingly, 11GNA, p.R158Q and p.I65T, respectively (see Table 1). the majority of the genotypes involve a BH4-responsive mutation. All haplotypes but two were observed either among mutant or Moreover, and besides the pathogenic mutations, six different normal chromosomes (data not shown). However, haplotypes 6.7 and single nucleotide polymorphisms (SNPs) were also identified among 9.8 were only identified as background of mutant alleles, IVS10nt- our patients, namely IVS5nt-54GNA (7.0%), p.Q232Q or rs1126758 11GNA the first and p.I65T and p.L348V the second. (13.3%), p.V245V or rs1042503 (9.5%), p.T328T (0.6%), p.L385L or rs772897 (3.2%) and p.Y414Y or rs1801152 (1.3%). Each of these silent mutations has already been described, except p.T328T found 3.3. Mutation–haplotype correlation for the first time in this study in a single allele. The evaluation of the homozigosity rate in the South Portuguese Almost all the mutations were found associated with a single HPA population revealed a high genetic heterogeneity (j=0.059), haplotype (Table 1), corresponding to the one usually linked to them, similar to the one observed among other South European populations which indicates that the hypothesis concerning their origins may also be (Sicily j=0.06) or ethnically mixed populations in Germany and The applied to the Portuguese population. The three exceptions are p.I65T, p. Netherlands [16]FOR. ELECTRONICP281L and IVS10nt-11G USENA. Most mutant ONLY alleles coding for p.I65T were found associated with the predominant haplotype 9.8 and single alleles with haplotypes 1.8 and 57.8. Concerning mutation p.P281L the majority 3.2. Haplotype analysis of the alleles were expressed on the background of haplotype 1.8, while two alleles were associated with haplotypes 2.3 and 16.3. The characterization of the haplotypic background underlying the The most interesting case was the mutant allele IVS10nt-11GNA, PKU locus in the Portuguese HPA patients was achieved in 153 mutant always associated with haplotype 6.7 except in a gipsy patient chromosomes. The analysis revealed the presence of 12 diverse displaying a 34.7 haplotypic background; this situation has already haplotypes, with the HindIII VNTR polymorphic site displaying hetero- been observed among the Spanish PKU population, where the geneity of haplotypes 1 and 5. It is interesting to note the high prevalence IVS10nt-11GNA mutation carried by patients with gipsy ancestry of haplotype 1 (14.3% for VNTR 7, 41.8% for VNTR 8 and 0.7% for VNTR 9) was always associated with haplotype 34.7. which is prevalent among mutant chromosomes with 18 different The PAH mutation database displays only one entry [17] for p. mutations associated with it. Among the remaining haplotypes, the most T323del mutation, with no haplotypic association. The study of I. Rivera et al. / Molecular Genetics and Metabolism 104 (2011) S86–S92 S89

Table 2 Genotypic and phenotypic data of 63 South Portuguese HPA patients.

Patient Allele 1 Allele 2 Pre-treatment Phe levels (μM) Clinical phenotypes

1 p.P281L IVS10nt-11GNA N1210a Classic PKU 2 p.P281L IVS10nt-11GNA 3329 Classic PKU 3 p.P281L IVS10nt-11GNA 587 Classic PKU 4 IVS10nt-11GNA IVS10nt-11GNA 2155 Classic PKU 5 IVS10nt-11GNA IVS10nt-11GNA 1622a Classic PKU 6 IVS10nt-11GNA IVS10nt-11GNA 1826 Classic PKU 7 p.I65T p.R261Q 230 Non-PKU HPA 8 p.I65T p.R261Q 1198 Moderate PKU 9 p.I65T p.P281L 1186 Moderate PKU 10 p.I65T p.P281L 581 Moderate PKU 11* p.R243X IVS10nt-11GNA 4806a Classic PKU 12* p.R243X IVS10nt-11GNA N1210a Classic PKU 13 p.R261Q p.R261Q 1513 Moderate PKU 13 p.R261Q p.R261Q 1017 Moderate PKU 15 p.R261Q p.V388M 999 Moderate PKU 16 p.R261Q p.V388M 1501 Moderate PKU 17* p.R261Q p.R408W 968a Moderate PKU 18* p.R261Q p.R408W 908a Moderate PKU 19* p.R261X p.I269L 297 Non-PKU HPA 20* p.R261X p.I269L 50b Non-PKU HPA 21 p.R270K p.R270K 847a Classic PKU 22 p.R270K p.R270K 1678 Classic PKU 23 p.R270K IVS10nt-11GNA 2276 Classic PKU 24 p.R270K IVS10nt-11GNA 1495 Classic PKU 25* p.A322G p.S349P 266 Non-PKU HPA 26* p.A322G p.S349P 255b Non-PKU HPA 27 IVS10nt-11GNA p.V388M 1090 Moderate PKU 28 IVS10nt-11GNA p.V388M 1210a Classic PKU 29 p.V388M p.Y414C 654a Mild PKU 30 p.V388M p.Y414C 478 Mild PKU 31 p.I65T p.I65T 1065 Moderate PKU 32 p.I65T p.R243Q 968a Moderate PKU 33 p.R68S p.E178G 278 Non-PKU HPA 34 p.R158Q p.Y198_E205NCfs 1210 Classic PKU 35 p.R158Q p.R270K 1332 Classic PKU 36 p.R158Q p.P281L 1271 Classic PKU 37 p.R158Q p.R176L 266 Non-PKU HPA 38 p.R158Q p.R261Q 1483 Classic PKU 39 p.R176L p.E390G 272 Non-PKU HPA 40 p.R176X p.R176X N1210 Classic PKU 41 p.R176X IVS10nt-11GNA N1210a Classic PKU 42 p.R176X p.V388M 1146 Mild PKU 43 p.Y198_E205NCfs p.R243X 1392 Classic PKU 44 p.V230I p.P281L 272 Non-PKU HPA 45 p.R241H p.R261Q 599 Moderate PKU 46 p.R243Q p.V388M 1108 Moderate PKU 47 p.L249F p.R261X 1174 Moderate PKU 48 p.L249F p.A309V 288 Non-PKU HPA 49 p.L249F p.V388M 793a Moderate PKU 50 p.R252W p.T323del 1816a Classic PKU 51 p.R252W IVS10nt-11GNA 1798a Classic PKU 52 p.R252W p.V388M 1500a Classic PKU 53 p.R261Q p.L311P 1412 Classic PKU 54 p.R261Q IVS10nt-11GNA 920 Classic PKU 55 p.I269L IVS10nt-11GNA 242 Non-PKU HPA 56 p.R270K p.L348V 847 Moderate PKU 57 p.R270K p.S349P 1210 Classic PKU 58 p.R270K p.E390G 605 Non-PKU HPA 59 p.R270K p.D415N 212 Non-PKU HPA 60 p.E280K p.E280K 800 Classic PKU 61 p.P281L p.P281L 1029 Classic PKU 62FOR p.A300S ELECTRONIC IVS10nt-11GNA 284USE ONLY Non-PKU HPA 63 p.L311P p.L311P 1683a Classic PKU 64 p.T323del p.T323del 1210 Classic PKU 65 p.T323del p.V388M 1090a Classic PKU 66 IVS10nt-11GNA p.A403V 316 Non-PKU HPA 67 p.V388M p.V388M 1254 Moderate PKU

a Patients born before the implementation of the Newborn Screening Program in Portugal, in 1980, and accordingly diagnosed on the basis of altered phenotype. b These patients, siblings of the index cases, were missed on the newborn screening test and were later detected during family studies.

Bercovich and colleagues [18] gives no information on the haplotypic As in other populations, the most frequent SNPs were strongly background of this deletion. The present study revealed that it is associated with specific haplotypes: p.Q232Q with haplotypes 3, 4 and always expressed on the background of haplotype 1. 7; p.V245V exclusively with haplotype 4; and p.L385L with haplotypes 3 S90 I. Rivera et al. / Molecular Genetics and Metabolism 104 (2011) S86–S92 and 7. The IVS5nt-54GNA and p.Y414Y polymorphisms were always large number of different people who populated the south-western part observed in linkage disequilibrium with haplotype 1.8. of the Iberian Peninsula, that later became Portugal. This heterogeneity is a common feature of South European populations [20–22] in contrast 3.4. Genotype–phenotype correlation with the North-eastern ones, especially in Poland [13]. The mutational spectrum involves 34 mutations, most of them falling According to the phenotypic classification indicated in the Methods on the category of missense type. Only seven transversions were fi section, out of the 83 patients under study, 36 were classified as classic observed, which con rms the higher transition rate over transversion N fi fi PKU (43.4%), 25 as moderate PKU (30.1%), 5 as mild PKU (6.0%) and 17 as rate [23]. Moreover, eight C T transitions were identi ed, ve of them non-PKU HPA (20.5%). Table 2 lists the 67 fully genotyped patients along locatedinCpGsites,knownasmutationalhotspots[24].CpGmethylation fi with their respective phenotype and a good genotype–phenotype has already been identi ed as accounting for recurrence of p.R408W correlation could be observed. mutation [25]. Additionally, it is interesting to note that the number of fi Concerning the homozygous patients, those carrying severe or null mild mutations exceeds the severe ones: 16 mutations are classi ed as mutations [p.R176X, p.E280K, p.P281L, p. L311P, p.T323del (all n=1), severe, while 18 mutations display considerable residual activity, ranging p.R270K (n=2) and IVS10nt-11GNA (n=3)] were classified as classic from moderate until high enzymatic activities (Table 1). fi PKU, while those harboring mutations with some residual enzymatic All the mutations identi ed in our patients have already been activity [p.I65T and p.V388M (both n=1) and p.R261Q (n=2)] described in other populations, apart from p.R270K. Indeed, this mutation fi displayed moderate PKU. has only been identi ed in Portuguese patients or in foreign individuals The nine cases where a genotype–phenotype inconsistency was (USA) but with Portuguese ancestry. Notably, any Spanish patient was observed corresponded to compound heterozygous patients, all of them detected carrying this mutation, which may denote a local origin. carrying p.R261Q or p.V388M mutations. Both transitions are recognized Another peculiar case is the p.T323del mutation found in our PKU as inducing negative inter-allelic complementation [19] and, accordingly, population. Previously, it had been detected in only a single individual patients displayed phenotypes more severe than expected. The only during the study of a heterogeneous USA population [17],butrecently fi exception was a patient harboring the genotype p.I65T/p.R261Q and two alleles were identi ed among the Ashkenazi Jewish population showing a much milder phenotype, non-PKU HPA. living in Israel [18]. Portuguese population involves a well-represented Jewish community and, besides, it has a long tradition of emigration. Accordingly, we may postulate that those patients could probably have a 3.5. Genotype and BH4-responsiveness Portuguese ancestry, because this mutation is well represented among our PKU population (4 alleles). One major goal of this study was to evaluate the rate of BH4- The correlation between each mutation and its haplotypic responsiveness among South Portugal PKU patients, aiming the background allowed us to confirm that the hypothesis underlying fi rationale identi cation of those displaying a potential positive each mutation origin(s) could be applied to our population and also to response to this novel pharmacologic therapy. trace the very origin of each mutation. Effectively, some mutations are Notably, more than half (I65T, R68S, R176L, E178G, V230I, R241H, well represented among all populations, though associated with R243Q, R261Q, I269L, A300S, A309V, A322G, L348V, V388M, E390G, different haplotypes. This is the case of p.S349P mutation which is fi A403V, Y414C and D415N) of the 34 mutations identi ed among detected on the background of haplotype 1 in North and Center South Portugal HPA patients belong to the group of more than 70 Europe and haplotype 4 in North Africa populations [26]. All the fi mutations already identi ed in BH4-responsive patients, according to Portuguese patients carrying this mutant allele displayed the BIOPKU database (www.biopku.org), and the majority lies in the haplotype 4.3, thus putting in evidence the strong colonization of catalytic domain of the protein. South Portugal by the Moors between VIII and XII centuries. The analysis of the allelic data allowed anticipating that 62.7% of the Additionally, we could determine that the single p.R408W mutant fully genotyped patients are potential responders to BH4 treatment allele was associated with haplotype 1.8, thus reflecting a Celtic rather (Table 3). More than half (52.4%) of the patients display the appropriate than Slavic origin [27,28]. metabolic phenotype, moderate or mild PKU, showing good perspectives Genotype–phenotype correlation is the cornerstone in most for an in vivo responsiveness. A minority (11.9%) display classic PKU, and studies on metabolic diseases. Nowadays, it relies on in vitro their answer to BH4 supplementation can only be disclosed by a expression analysis (IVE) of recombinant mutant proteins, using physiological challenge, because all these patients carry in heterozigosity several different systems [29]. These studies revealed that the the p.R261Q and p.V388M mutations, known for their potential induction mutations displaying extreme enzymatic activities in vitro, either of negative inter-allelic complementation, as previously referred. Finally, null/reduced or elevated, are always associated with the most severe and notably, the remaining 15 patients have a non-PKU phenotype, are or milder forms of the disease, respectively. Problems arise with already on a free diet and accordingly should not need BH4 mutant enzymes associated with moderate levels of residual activity supplementation. [16,30]. Accordingly, it has been suggested that IVE may ultimately be important in discriminating between mutations which allow leeway 4. Discussion for variability in the enzymatic and metabolic phenotypes and those FOR ELECTRONICthat do not [29] . USE ONLY Molecular characterization of the HPA population living in South In this study, genotype–phenotype relationships were evaluated Portugal confirmed its high genetic heterogeneity, originating on the taking into consideration data from in vitro expression of recombinant mutant proteins, and PAH activity was defined as the average sum of Table 3 activities of both individual mutant alleles. The results obtained Anticipation of BH4-responsiveness in South Portugal HPA patients. revealed a good correlation between the genotype and the observed

Number of Frequency Phenotype Frequency phenotype, with only nine inconsistent cases. Moreover, most of these patients (%) (%) patients were born before the implementation of the newborn

Potential 42 62.7 Classic PKU 11.9 screening program in Portugal and, accordingly, were only detected responders Moderate/mild PKU 52.4 after the appearance of an abnormal phenotype. All the discordant Non-PKU HPA 35.7 cases involve two specific mutations, p.R261Q and p.V388M, already Potential non- 25 37.3 Classic PKU 100.0 described by several authors [19,31–33] as responsible for negative responders interallelic complementation. As both mutations are not located in the I. Rivera et al. / Molecular Genetics and Metabolism 104 (2011) S86–S92 S91 tetramerization domain of PAH, it was suggested that they could molecular genetics in Croatian patients with phenylalanine hydroxylase (PAH) deficiency,Mol.Genet.Metab.97(2009)165–171. affect the interactions between the subunit interfaces, namely upon [7] C.O. Harding, New era in treatment for phenylketonuria: pharmacologic therapy the conformational changes occurring at the dimer interface upon with sapropterin dihydrochloride, Biologics 4 (2010) 231–236. activation by L-phenylalanine [19]. The most striking inconsistency [8] H. Erlandsen, R.C. Stevens, A structural hypothesis for BH4 responsiveness in patients with mild forms of hyperphenylalaninaemia and phenylketonuria, J. Inherit. Metab. case was the patient carrying the p.I65T/p.R261Q genotype, who Dis. 24 (2001) 213–230. should display a moderate phenotype and effectively is a non-PKU [9] A.P. Pey, B. Pérez, L.R. Desviat, M.A. Martínez, C. Aguado, H. Erlandsen, A. Gámez, HPA individual, requiring no diet. Previous co-expression of these two R.C. Stevens, M. Thórólfsson, M. Ugarte, A. Martínez, Mechanisms underlying mutations also revealed problems, hampering the study of interallelic responsiveness to tetrahydrobiopterin in mild phenylketonuria mutations, Hum. Mutat. 24 (2004) 388–399. complementation [19]. [10] B. Pérez, L.R. Desviat, M. Ugarte, Analysis of the phenylalanine hydroxylase gene in fi The evaluation of responsiveness to BH4 treatment in South Portugal the Spanish population: mutation pro le and association with intragenic – patients revealed encouraging results. Effectively, two thirds of the polymorphic markers, Am. J. Hum. Genet. 60 (1997) 95 102. [11] M.G. Mirisola, F. Cali, A. Gloria, P. Schinocca, M. D'Amato, G. Cassara, G. De Leo, L. Palillo, individuals harbor genotypes conferring a potential positive response to C. Meli, V. Romano, PAH gene mutations in the Sicilian population: association with this pharmacologic therapy. However, one must stress that false minihaplotypes and expression analysis, Mol. Genet. Metab. 74 (2001) 353–361. [12] S.F. Dobrowolski, C. Heintz, T. Miller, C. Ellingson, C. Ellingson, I. Ozer, G. Gokçay, expectations need to be avoided and, accordingly, the decision of BH4 T.Baykal,B.Thony,M.Demirkol,N.Blau,Moleculargeneticsandimpactof treatment must also take into account some other issues, such as the residual in vitro phenylalanine hydroxylase activity on tetrahydrobiopterin possibility of negative inter-allelic complementation between some responsiveness in Turkish PKU population, Mol. Genet. Metab. 10 (2011) mutant alleles in compound heterozygous patients, as previously 116–121. [13] S.F. Dobrowolski, K. Borski, C.E. Ellingson, R. Koch, H.L. Levy, E.W. Naylor, A limited referred. Indeed, among our PKU population almost half of the patients spectrum of phenylalanine hydroxylase mutations is observed in phenylketonuria carry on their genotypes one allele (p.I65T, p.R261Q and p.V388M) patients in western Poland and implications for treatment with 6R tetrahydrobiopterin, known to induce negative inter-allelic complementation. J. Hum. Genet. 54 (2009) 335–339. [14] I. Rivera, P. Leandro, U. Lichter-Konecki, I.T. De Almeida, M.C. Lechner, Population Moreover, and recently, we became aware that some missense genetics of hyperphenylalaninaemia resulting from phenylalanine hydroxylase mutations are actually splicing mutations, inducing worse phenotypes deficiency in Portugal, J. Med. Genet. 35 (1998) 301–304. than the expected ones. Already in 1999, Ellingsen and co-workers [34] [15] R.C. Eisensmith, S.L.C. Woo, Update listing of haplotypes at the human phenylalanine – highlighted the fact that aberrant transcripts are a frequent consequence hydroxylase (PAH) locus, Am. J. Hum. Genet. 51 (1992) 1445 1448. [16] P. Guldberg, F. Rey, J. Zschocke, V. Romano, B. François, L. Michiels, K. Ullrich, G.F. of exonic point mutations, which can alter enhancer or silencer splicing Hoffmann, P. Burgard, H. Schmidt, C. Meli, E. Riva, I. Dianzani, A. Ponzone, J. Rey, F. sites. Additionally, the putative missense Y204C mutation revealed to be Guttler, A European multicenter study of phenylalanine hydroxylase deficiency: classification of 105 mutations and a general system for genotype-based indeed a splicing mutation and accordingly its nomenclature was – N prediction of metabolic phenotype, Am. J. Hum. Genet. 63 (1998) 71 79. changed to p.EX6-96A G [35]. This fact may explain some of the missing [17] R.C. Eisensmith, D. Martinez, A. Kuzmin, A. Goltsov, S.L.C. Woo, Molecular basis of genotype–phenotype correlation observed in some patients, and a re- phenylketonuria in a heterogeneous US population, Am. J. Hum. Genet. 57 (1995) examination of a large number of disease-causing mutations should be A163-(926). [18] D. Berkovich, A. Elimelech, T. Yardeni, S. Korem, J. Zlotogora, N. Gal, N. Goldstein, B. taken into consideration. Vilenski, R. Segev, S. Avraham, R. Loewenthal, G. Schwartz, Y. Anikster, A mutation Very recently, Staudigl and colleagues [36] published important analysis of the phenylalanine hydroxylase (PAH) gene in the Israeli population, data concerning the influence of substrate and cofactor concentra- Ann. Hum. Genet. 72 (2008) 305–309. [19] J. Leandro, C. Nascimento, I. Tavares de Almeida, P. Leandro, Co-expression of tions, in the presence of certain genotypes, on enzyme function and on different subunits of human phenylalanine hydroxylase: evidence of negative response to BH4 treatment. Indeed, BH4 supply can display two interallelic complementation, Biochim. Biophys. Acta 1762 (2006) 544–550. different effects, kinetic at a short-term and chaperone at long-term, [20] P. Guldberg, V. Romano, N. Ceratto, P. Bosco, M. Ciuna, A. Indelicato, F. Mollica, C. Meli, M. Giovannini, E. Riva, et al., Mutational spectrum of phenylalanine hydroxylase and individual mutations may shift these effects. Moreover, the initial deficiency in Sicily: implications for diagnosis of hyperphenylalaninemia in Southern phenylalanine levels revealed to be important, also. These data are Europe, Hum. Mol. Genet. 2 (1993) 1703–1707. particularly relevant once a significant number of South Portugal HPA [21] B. Perez, L.R. Desviat, M. De Lucca, M. Ugarte, Spectrum and origin of phenylketonuria – patients harbor mutations whose enzymatic activity strongly depends mutations in Spain, Acta Paediatr. Scand. Suppl. 407 (1994) 34 36. [22] I. Dianzani, S. Giannattasio, L. de Sanctis, C. Alliaudi, P. Lattanzio, C.D. Vici, A. Burlina, on cofactor and/or substrate concentrations, namely p.I65T, p.R261Q M. Burroni, G. Sebastio, F. Carnevale, et al., Characterization of phenylketonuria and p.Y414C [36]. So, the present study may benefit from this newly alleles in the Italian population, Eur. J. Hum. Genet. 3 (1995) 294–302. gained knowledge, enabling us to design more personalized screening [23] D.W. Collins, T.H. Jukes, Rates of transition and transversion in coding sequences since the human–rodent divergence, Genomics 20 (1994) 386–396. loading tests, according to each patient genotype. This approach will [24] D.N. Cooper, M. Krawczak, S.E. Antonorakis, The nature and mechanisms of allow then a more safe and correct assessment of BH4 responsiveness, human gene mutation, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The avoiding false negative responses. Metabolic and Molecular Bases of Inherited Disease, McGraw-Hill, New York, – fi 2000, pp. 343 377. In conclusion, a signi cant number of Portuguese PKU patients are [25] B.C. Murphy, C.R. Scriver, S.M. Singh, CpG methylation accounts for a recurrent likely to benefit from BH4 therapy which, combined with a less strict mutation (c.1222CNT) in the human PAH gene, Hum. Mutat. 27 (2006) 975. diet, or eventually in special cases as monotherapy, may contribute to [26] M. Weinstein, R.C. Eisensmith, V. Abadie, S. Avigad, S. Lyonnet, G. Schwartz, A. Munnich, fi S.L.C. Woo, Y. Shiloh, A missense mutation, S349P, completely inactivates phenylalanine reduce nutritional de ciencies and minimize neurological and hydroxylase in north African Jews with phenylketonuria, Hum. Genet. 90 (1993) psychological dysfunctions. Additionally, and also importantly, it 645–649. may contribute to a better quality of life of these patients. [27] R.C. Eisensmith, Y. Okano, M. Dasovich, T. Wang, F. Güttler, H. Lou, P. Guldberg, U. Lichter-Konecki, D.S. Konecki, E. Svensson, et al., Multiple origins for phenylketonuria in Europe, Am. J. Hum. Genet. 51 (1992) 1355–1365. References FOR ELECTRONIC[28] E. Treacy, S. Byck,USE C. Clow, C.R. Scriver, ONLY‘Celtic’ phenylketonuria chromosomes found? Evidence in two regions of Quebec Province, Eur. J. Hum. Genet. 1 (1993) [1] C.R. Scriver, The PAH gene, phenylketonuria, and a paradigm shift, Hum. Mutat. 28 220–228. (2007) 831–845. [29] P.J. Waters, How PAH gene mutations cause hyper-phenylalaninemia and why [2] F.K. Treftz, N. Blau, Potential role of tetrahydrobiopterin in the treatment of mechanism matters: insights from in vitro expression, Hum. Mutat. 21 (2003) maternal phenylketonuria, Pediatrics 112 (2003) 1566–1569. 357–369. [3] F.J. van Spronsen, G.M. Enns, Future treatment strategies in phenylketonuria, Mol. [30] E. Kayaalp, E. Treacy, P.J. Waters, S. Byck, P. Nowacki, C.R. Scriver, Human Genet. Metab. 99 (2010) S90–S95. phenylalanine hydroxylase mutations and hyperphenylalaninemia phenotypes: a [4] S. Kure, D.C. Hou, T. Ohura, H. Iwamoto, S. Suzuki, N. Sugiyama, O. Sakamoto, K. Fujii, metanalysis of genotype–phenotype correlations, Am. J. Hum. Genet. 61 (1997) Y. Matsubara, K. Narisawa, Tetrahydrobiopterin-responsive phenylalanine deficien- 1309–1317. cy, J. Pediatr. 135 (1999) 375–378. [31] P. Leandro, I. Rivera, M.C. Lechner, T. de Almeida I, D. Konecki, The V388M [5] G. Gramer, P. Burgard, S.F. Garbade, M. Lindner, Effects and clinical significance of mutation results in a kinetic variant form of phenylalanine hydroxylase, Mol. tetrahydrobiopterin supplementation in phenylalanine hydroxylase-deficient Genet. Metab. 69 (2000) 204–212. hyperphenylalaninaemia, J. Inherit. Metab. Dis. 30 (2007) 556–562. [32] A. Gámez, B. Pérez, M. Ugarte, L.R. Desviat, Expression analysis of phenylketonuria [6] I. Karacic, D. Meili, V. Sarnavka, C. Heintz, B. Thöny, D.P. Ramadza, K. Fumic, D. Mardesic, mutations. Effect on folding and stability of the phenylalanine hydroxylase I. Baric, N. Blau, Genotype-predicted tetrahydrobiopterin (BH4)-responsiveness and protein, J. Biol. Chem. 275 (2000) 29737–29742. S92 I. Rivera et al. / Molecular Genetics and Metabolism 104 (2011) S86–S92

[33] P.J. Waters, C.R. Scriver, M.A. Parniak, Homomeric and heteromeric interactions [35] S. Ellingsen, P.M. Knappskog, H.G. Eiken, Phenylketonuria splice mutation between wild-type and mutant phenylalanine hydroxylase subunits: evaluation (EXON6nt-96ANG) masquerading as missense mutation (Y204C), Hum. Mutat. of two-hybrid approaches for functional analysis of mutations causing 9 (1997) 88–90. hyperphenylalaninemia, Mol. Genet. Metab. 73 (2001) 230–238. [36] M. Staudigl, S.W. Gersting, M.K. Danecka, D.D. Messing, M. Woidy, D. Pinkas, K.F. [34] S. Ellingsen, P.M. Knappskog, J. Apold, H.G. Eiken, Diverse PAH transcripts in Kemter, N. Blau, A.C. Muntau, The interplay between genotype, metabolic state, lymphocytes of PKU patients with putative nonsense (G272X, Y356X) and and cofactor treatment governs phenylalanine hydroxylase function and drug missense (P281L, R408Q) mutations, FEBS Lett. 457 (1999) 505–508. response, Hum. Mol. Genet. 20 (2011) 2628–2641.

FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S93–S96

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

The spectrum of phenylketonuria genotypes in the Armenian population: Identiﬁcation of three novel mutant PAH alleles

Natella Kostandyan a, Corinne Britschgi c,d, Albert Matevosyan a, Alvina Oganezova a, Anahit Davtyan b, Nenad Blau c,d, Beat Steinmann e,⁎, Beat Thöny c,d,⁎⁎ a Republic Center of Medical Genetics of Yerevan State Medical University, Armenia b Laboratory Diagnostic Center of Yerevan State Medical University, Armenia c Division of Clinical Chemistry and Biochemistry, Department of Pediatrics, University of Zürich, Switzerland d Research Center for Children (RCC), Department of Pediatrics, University of Zürich, Switzerland e Division of Metabolic Diseases, Department of Pediatrics, University of Zürich, Switzerland article info abstract

Article history: We present the spectrum of phenylalanine hydroxylase (PAH) gene mutations upon investigating 35 index Received 22 June 2011 patients identified with hyperphenylalaninemia in Armenia. One patient was diagnosed with dihydropter- Received in revised form 9 August 2011 idine reductase (DHPR) deficiency, whereas all other 34 and their 6 affected siblings presented with mild or Accepted 9 August 2011 classical phenylketonuria (PKU). By analyzing all 13 exons plus exon–intron boundaries of the PAH gene, we Available online 12 August 2011 identified two mutant alleles in 23 PKU patients, three mutations in 1, only one mutation in 5, and no mutation in 5 PKU patients. The most prevalent mutation was the well defined splicing error in intron 10, Keywords: N N N Hyperphenylalaninemia c.1066-11G A (17/68 alleles). The three alterations, c.836C T (p.Pro279Leu) in exon 7, c.1129T G Mutation analysis (p.Tyr377Asp) in exon 11, and c.1244ANT (p.Asp415Val) in exon 12, have not been reported in the PAH DHPR deficiency locus database (http://www.pahdb.mcgill.ca) and, thus, might be specific for the culturally homogenous Armenian population. © 2011 Elsevier Inc. All rights reserved.

1. Introduction number of enzymes, including the tyrosine and tryptophan hydroxylases, responsible for the rate limiting steps in catecholamines and

Phenylketonuria (PKU; OMIM ID: 261600) is an autosomal serotonin biosynthesis, respectively. As a consequence, BH4 deﬁciency recessive disease caused by mutations in the phenylalanine hydrox- requires not only peripheral control of hyperphenylalaninemia, but ylase (PAH) gene encoding the hepatic PAH enzyme (EC 1.14.16.1) [1]. also oral therapy with dopamine and serotonin precursors to restore PAH is responsible for the conversion of phenylalanine to tyrosine in monoamine neurotransmitters biosynthesis in the central nervous an oxygen and tetrahydrobiopterin (BH4)-cofactor dependent reac- system. tion. When PAH enzymatic activity is reduced, phenylalanine The PAHdb locus database (www.pahdb.mcgill.ca), initiated in accumulates in the blood and brain and leads to mental retardation. 1990 as a catalog of variant alleles of the human PAH gene [3], lists Besides mutations in the PAH gene, which account for 97–99% of all more than 560 PAH mutations today, and continues to grow as novel hyperphenylalaninemias, rare variants are caused by autosomal mutations are detected with similar such investigations in many recessive mutations in BH4 cofactor-metabolizing genes, i.e. in formerly unstudied countries and in different ethnic populations. For GCH1, PTS or QDPR, encoding the GTP cyclohydrolase I, 6-pyruvoylte- instance, the Korean [4], Southern Italian [5], Out-of-African in South trahydropterin synthase, and dihydropteridine reductase (DHPR; EC East England [6], Israeli [7], Western Poland [8], Croatian [9], Turkish 1.6.99.7, OMIM ID: 261630), respectively [2]. Differential diagnosis of [10], Indian [11], and Iranian populations [12] have provided reports hyperphenylalaninemiasFOR based ELECTRONIC on urinary or dried blood pterin to this effect inUSE the last couple of years.ONLY analyses and DHPR enzyme activity measurements on dried blood The incidences of PKU in formerly reported populations average spots is critical to rule out cofactor deﬁciency, as BH4 is essential for a approximately 1:10,000 in European–Caucasians. However, the number varies with a higher incidence in e.g. Ireland (1:4500) [13], Turkey (1:4000) [14], and Sicily (1:2700) [15], and a much lower ⁎ Correspondence to: B. Steinmann, Division of Metabolic Diseases, Department of incidence is reported in the Finnish (below 1:100,000) [16], Chinese Pediatrics, University of Zürich, Steinwiesstrasse 75, CH-8032 Zürich, Switzerland. (1:15,000 to 1:100,500) [17,18], Japanese (1:70,000) [19] and Korean ⁎⁎Correspondence to: B. Thöny, Division of Clinical Chemistry and Biochemistry, Department of Pediatrics, University of Zürich, Switzerland. Fax: +41 44 266 7169. (1:41,000) [4] populations (for an overview see also Ref. [20]). E-mail addresses: [email protected] (B. Steinmann), However, not all populations have been studied and, since neonatal [email protected] (B. Thöny). screening for PKU was introduced in Armenia only in April 2008, the

1096-7192/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2011.08.006 S94 N. Kostandyan et al. / Molecular Genetics and Metabolism 104 (2011) S93–S96 incidence for the ~3.1 million inhabitants, which has a nearly- DHPR deficiency; his values obtained from filter paper blood were as homogenous population with approximately 98% being ethnic follows: neopterin 0.38 nmol/g Hb (normal 0.31–4.45 nmol/g Hb), Armenian, remains unknown (2011 Official Armenian Census Statis- biopterin 0.41 nmol/g Hb (0.15–2.91 nmol/g Hb), DHPRb0.05 mU/mg tics at http://en.wikipedia.org/wiki/Armenian_people). Hb (1.8–3.8 mU/mg Hb), and he was found to be compound Here we present the spectrum of mutations in 35 Armenian heterozygous for the two known pathogenic mutations c.445GNA families identified with hyperphenylalaninemia. We found the well (p.Gly149Arg) and c.449ANG (p.Tyr150Cys) in QDPR, respectively [2]. defined splicing error in intron 10 of the PAH gene, c.1066-11GNA, as Whereas the p.Gly149Arg exchange affects protein dimerization and the most prevalent mutation, plus three previously unreported PAH was associated with the severe, central form of DHPR deficiency, the p. mutations, which might be specific to the culturally homogenous Tyr150Cys mutant was described to have reduced but significant Armenian population. residual enzyme activity under in vitro expression conditions, leading to an intermediate type of DHPR deficiency (for further mutation 2. Subjects and methods information see also www.biopku.org). Among the remaining 34 patients with classical PKU, two mutant 2.1. Patients PAH alleles were detected in 23 subjects, three mutations in 1 (patient #10), only one mutation in 5 (patient #s 4, 11, 16, 21, 24), and no All 35 index patients with hyperphenylalaninemia, known to the mutation in 5 (patient #s 1, 19, 27, 29, 30; for more details see also Republic Center of Medical Genetics in Yerevan, and their parents Table 1). The prevalence of PAH mutations, in decreasing frequency, and siblings were contacted and investigated for this study in was the c.1066-11GNA splicing mutation with 17 of the total 68 alleles February and March of 2010; five of them were detected by the (including the 3rd mutation in case #10), 4 for the alleles c.842CNT recently introduced neonatal screening at the Laboratory Diagnostic and c.782GNA, 3 for the alleles c.47_48delCT and c.727CNT, 2 for the Center in Yerevan by the use of Enzolve's enzymatic phenylketon- alleles c.143TNC and c.754GNA, and 1 for the remaining alleles uria method (Enzolve Technologies, Ireland). Some of the patients together with the three novel mutant PAH alleles: c.836CNT were under strict low-phenylalanine dietary treatment, whereas (p.Pro279Leu), c.1129TNG (p.Tyr377Asp), and c.1244ANT (p.Asp415Val). others followed a less stringent vegetarian diet. This study was These findings will be discussed brieflyhere. approved by the ethics committee of the Yerevan State Medical The c.1066-11GNA splicing mutation was described as a major cause University and informed consent was obtained from all parents or for PKU in parts of Southern Europe [22], and is the second most legal representatives. common mutant allele reported in the PAHdb with an overall percentage of 7.4%. In the limited number of Armenian PKU patients analyzed in this 2.2. Metabolic and genetic analyses study, this mutation was found to be present in 17 out of 68 alleles analyzed, i.e. in 25% of all PAH mutant alleles. Although clearly more Phenylalanine and pterin concentrations were measured from dried genetic data on PKU patients from Armenia (and from other countries in blood spots using a routine tandem mass spectrometry method for Asia) are needed to confirm such a high prevalence of this particular phenylalanine and an HPLC method for pterins [21], respectively. DNA PAH-splicing mutation, the term “Mediterranean PKU mutation” might was isolated from peripheral blood using the EZ1 DNA Blood Kit plus the be expanded to “Southern Eurasian PKU mutation”, since this mutant extraction device BIOROBOT® EZ1 (Qiagen Inc.). Genetic analysis allele was found to have a very high frequency at least among the included PCR amplification and “Sanger” genomic DNA sequencing Turkish [10],Iranian[12] and the here reported Armenian populations. carried out on a GeneAmp PCR System 9700 and a 3130xl Genetic Homozygous mutations in the investigated PKU families are rare; Analyzer (Applied Biosystems Inc.), respectively. All 13 exons of the PAH this may be explained by the fact that consanguinity, in the prevalent gene and 7 exons of the QDPR gene (encoding DHPR), plus their exon– Christian population of Armenia, is the exception despite the fact that intron boundaries were analyzed. The corresponding primers for PCR the inhabitants live in a population isolate. Of the four patients born amplification of genomic DNA isolated from peripheral blood samples, into a consanguineous union, only case #31 is homozygous for the as well as for DNA sequence analysis, are listed in the supplementary most frequent c.1066-11GNA allele; no mutation was detected on Tables S1 and S2. The standard amplification conditions we used can be both alleles for case #30, and the two other patients, cases #8 and obtained upon request. Reference accession numbers for normal alleles #11, were compound heterozygous. were ENSG00000171759.2 for PAH and ENSG00000151552.6 for QDPR. In 10 subjects, only one or no mutation was detectable, despite Mutation designation was according to the official mutation nomen- that they were clearly affected by PKU as determined by clinical clature (http://www.hgvs.org/mutnomen/). To validate nomenclature findings, i.e. grossly elevated blood phenylalanine values, and normal of mutations, we used the program Mutalyzer 2.0 β-2 (http://www. biopterin levels. In a comparable study performed recently among the mutalyzer.nl/2.0/). Mutations were confirmed by carrier analysis of Turkish population, where 588 hyperphenylalaninemic patients were parents and/or siblings, if available (for details see Table 1). genotyped [10], 29 subjects with only one mutant allele could be identified by PAH-exon sequencing, plus 13 patients with no 3. Results and discussion mutations in the PAH gene (unpublished observation). It is assumed that functional alterations are present in mainly intronic regions, Prior to theFOR initiation of this study, ELECTRONIC nothing was known about the which represent overUSE 98% of the PAH ONLYgene sequence length. incidence and the spectrum of mutations in patients with hyperphe- As mentioned before, the three alterations, c.836CNT (p.Pro279- nylalaninemia in Armenia. We initially collected DNA from 36 Leu) in exon 7, c.1129TNG (p.Tyr377Asp) in exon 11, and c.1244ANT subjects reported with hyperphenylalaninemia; however, upon re- (p.Asp415Val) in exon 12, have not been reported in the PAH locus assessing blood phenylalanine concentrations, one subject could not (http://www.pahdb.mcgill.ca)andthe“1000 genome project” be confirmed as hyperphenylalaninemic (case #23). For the remain- (http://browser.1000genomes.org/index.html) databases, and might ing 35 patients with confirmed hyperphenylalaninemia and their be specific for the Armenian population. For functional prediction family members, we continued with routine differential diagnosis by regarding disease causing potential of these three novel mutant PAH pterin and DHPR activity measurements on dried blood spots, and alleles, we performed in silico testing using the PolyPhen and performed DNA mutation analysis. The results from genotyping and MutationTaster programs (see also http://genetics.bwh.harvard.edu/ the clinical-genetic findings are summarized in Table 1. pph2/ and http://www.mutationtaster.org). All three novel alter- From the 35 patients with hyperphenylalaninemia, 34 had normal ations were predicted to be probably damaging with scores of 1.000 BH4 metabolism, whereas 1 patient, case #14, was diagnosed with for p.Pro279Leu and p.Tyr377Asp, and a score of 0.963 for N. Kostandyan et al. / Molecular Genetics and Metabolism 104 (2011) S93–S96 S95

Table 1 Spectrum of Armenian PKU patients.

Case # (a) PAH Genotype Clinical Genetic Findings

Paternal Allele Maternal Allele Consanguinity Affected / total siblings NB-screening 1 no mutation detected no mutation detected 0/1 2 c.1066 − 11G > A, splicing c.809G > A, p.Arg270Lys, missense 0/0 3 c.1066 − 11G > A, splicing c.781C > T, p.Arg261X, nonsense 0/0 4 no mutation detected c.1066 − 11G > A, splicing 0/0 5 c.782G > A, p.Arg261Gln, missense c.1066 − Α11G > A, splicing 0/1 6 c.47_48delCT, p.S16XfsX1, deletion c.1244A > T, p.Asp415Val, missense 0/0 7 c.842C > T, p.Pro281Leu, missense c.168+5G > C, splicing 1/3 8 c.1066 − 11G > A, splicing c.1129T > G, p.Tyr377Asp, missense 0/0 9 c.1066 − 11G > A, splicing c.1066 − 11G > A, splicing 0/1 10 c.143T > C, p.Leu48Ser, missense and c.782G > A, p.Arg261Gln, missense 0/0 c.1066 − 11G > A, splicing (b) 11 no mutation detected c.441 + 5G > T, splicing 2° 0/1 12 c.728G > A, p.Arg243Gln, missense c.842C > T, p.Pro281Leu, missense 0/1 13 c.47_48delCT, p.Ser16XfsX1, deletion c.1066 − 11G > A, splicing 0/1 15 c.727C>T, p.Arg243X, nonsense c.842C > T, p.Pro281Leu, missense 0/1 16 c.47_48delCT, p.Ser16XfsX1, deletion no mutation detected 0/1 17 c.1169A > G, p.Glu390Gly, missense c.754C > T, p.Arg252Trp, missense 0/1 18 c.1208C > T, p.Ala403Val, missense c.1066 − 11G > A, splicing 0/0 19 no mutation detected no mutation detected 0/1 20 c.1066 − 11G > A, splicing c.1244A > T, p.Asp415Val, missense 1/1 21 c.782G > A, p.Arg261Gln, missense no mutation detected 0/0 22 c.1089delG, p.Lys363AsnfsX37, deletion c.838G > A, p.Glu280Lys, missense 1/1 24 no mutation detected (c) c.1066 − 11G > A, splicing 0/1 25 c.143T > C, p.Leu48Ser, missense c.836C > T, p.Pro279Leu, missense 1/1 26 c.842C > T, p.Pro281Leu, missense c.1222C > T, p.Arg408Trp, missense 0/1 27 no mutation detected no mutation detected 0/1 28 c.809G > A, p.Arg270Lys, missense c.727C > T, p.Arg243X, nonsense 0/1 29 no mutation detected no mutation detected 1/5 30 no mutation detected no mutation detected 3° 1/1 31 c.1066 − 11G > A, splicing c.1066 − 11G > A, splicing 2.5° 0/3 32 c.1066 − 11G > A, splicing c.1200 − 2A > G, splicing 0/1 33 c.1159T > C, p.Tyr387His, missense c.754C > T, p.Arg252Trp, missense 0/1 34 c.728G > A, p.Arg243Gln, missense c.1066 − 11G > A, splicing 0/2 35 c.592_613del22, p.Tyr198SerfsX136, deletion c.782G > A, p.Arg261Gln, missense 0/0 36 c.727C > T, p.Arg243X, nonsense c.1066 − 11G > A, splicing 0/1 (a) Note that case # 14 was diagnosed as DHPR deficiency and case # 23 was not confirmed as PKU; both these cases are not listed in Table 1 (see text for details). (b) Three mutations were found in this patient; the maternal mutation was confirmed by analyzing the mother’s DNA, however the paternal DNA was not available. (c) No DNA was available from this patient, but DNA analyzed from the mother and from the non-affected sibling revealed that they were carrier for c.1066-11G>A.

Key: No mutation detected Novel mutation

Consanguinity ࿻: absent (࿻): uncertain 2°: parents are 2nd degree cousins §: families # 6 and # 20 are related, i.e. both mothers are sisters

Affected/Total Siblings 0/1: the single sibling is not affected 1/1: the single sibling is affected 1/5: one among 5 siblings is affected

Newborn screening +: detected in Yerevan in the time period from April 15th 2008 to March 2010 (one additional patient was detected until the end of 2010 but not included in this study) (+): detected in Moscow ࿻: not screened

FOR ELECTRONIC USE ONLY p.Asp415Val. Also, no SNPs in these altered genomic regions were respectively. Both are moderately retarded and display behavioral found. Furthermore, although they are not in a homozygous state, one problems and aggressiveness. Unfortunately, it was not possible to can draw some indirect conclusions regarding the effect of the amino explore their phenylalanine tolerance. From these observations it can be acid substitution on the PAH enzyme based on the clinical and concluded that the p.Pro279Leu exchange in PAH is expected to lead to a chemical ﬁndings in these subjects. mostly inactive mutant protein, as the second allele in these patients,

Case #25 with the newly identified PAH allele (p.Pro279Leu) has a p.Leu48Ser, is known to be responsible for a mild, BH4-responsive PKU second mild allele with known residual PHA activity (p.Leu48Ser). The phenotype (see also www.biopku.org). patient and his younger brother were detected by newborn screening in Case #8 was diagnosed at 1 year 10 months and is compound Moscow at the ages of 6 days and 2 months, respectively. Because the heterozygous with the new mutant allele c.1129TNG(p.Tyr377Asp)in Armenian parents had no Russian citizenship, the costs for the amino acid combination with the c.1066-11GNA splicing mutation that is known to mixtures were not covered and treatment was stopped. At the ages of 6.5 lead to classical, BH4 non-responsive PKU. Although his parents are and 5.5 years, their phenylalanine levels were 20.5 mg/dl and 18.3 mg/dl, consanguineous (2.5° cousins), he is compound heterozygous. The S96 N. Kostandyan et al. / Molecular Genetics and Metabolism 104 (2011) S93–S96 patient is severely retarded and cannot adhere to a diet. His phenylal- locus: evidence in French-Canadians and a catalog of mutations, Am. J. Hum. Genet. 46 (1990) 970–974. anine values are between 20 and 30 mg/dl; phenylalanine tolerance [4] D.H. Lee, S.K. Koo, K.S. Lee, Y.J. Yeon, H.J. Oh, S.W. Kim, S.J. Lee, S.S. Kim, J.E. Lee, I. Jo, could not be determined. From this observation, it is expected that the S.C. Jung, The molecular basis of phenylketonuria in Koreans, J. Hum. Genet. 49 newly identified mutant p.Tyr377Asp leads to an inactive PAH protein. (2004) 617–621. [5] A. Daniele, G. Cardillo, C. Pennino, M.T. Carbone, D. Scognamiglio, A. Correra, A. The mothers of cases #6 and #20 are sisters and both carry the newly Pignero, G. Castaldo, F. Salvatore, Molecular epidemiology of phenylalanine identified mutant allele c.1244ANT (p.Asp415Val). Case #6 was detected hydroxylase deficiency in Southern Italy: a 96% detection rate with ten novel at age 3 years because both his maternal cousins were diagnosed with mutations, Ann. Hum. Genet. 71 (2007) 185–193. PKU and mild mental retardation. Phenylalanine tolerance was approx- [6] P. Hardelid, M. Cortina-Borja, A. Munro, H. Jones, M. Cleary, M.P. Champion, Y. Foo, C.R. Scriver, C. Dezateux, The birth prevalence of PKU in populations of European, imately 275 mg/day. Case #20 is severely retarded and lives at home South Asian and sub-Saharan African ancestry living in South East England, Ann. while his older brother is in a foster home. The patient is not on a Hum. Genet. 72 (2008) 65–71. controlled diet and his blood phenylalanine levels are between 15 and [7] D. Bercovich, A. Elimelech, T. Yardeni, S. Korem, J. Zlotogora, N. Gal, N. Goldstein, B. Vilensky, R. Segev, S. Avraham, R. Loewenthal, G. Schwartz, Y. Anikster, A 25 mg/dl. Since patients #6 and #20 are compound heterozygous with a mutation analysis of the phenylalanine hydroxylase (PAH) gene in the Israeli second allele, c.47_48delCT and c.1066-11GNA, respectively, both known population, Ann. Hum. Genet. 72 (2008) 305–309. [8] S.F. Dobrowolski, K. Borski, C.C. Ellingson, R. Koch, H.L. Levy, E.W. Naylor, A limited to be responsible for classical, BH4 non-responsive PKU, it is expected that fi spectrum of phenylalanine hydroxylase mutations is observed in phenylketonuria the newly identi ed amino acid exchange p.Asp415Val leads to an patients in western Poland and implications for treatment with 6R tetrahydro- inactive PAH protein. biopterin, J. Hum. Genet. 54 (2009) 335–339. The incidence of PKU in Armenia is not known as neonatal screening to [9] I. Karacic, D. Meili, V. Sarnavka, C. Heintz, B. Thony, D.P. Ramadza, K. Fumic, D. fi Mardesic, I. Baric, N. Blau, Genotype-predicted tetrahydrobiopterin (BH4)- detect this disorder was introduced on April 15th, 2008. For nancial and responsiveness and molecular genetics in Croatian patients with phenylalanine logistic reasons, currently this program covers only the ten birth clinics in hydroxylase (PAH) deficiency, Mol. Genet. Metab. 97 (2009) 165–171. Yerevan. By the end of 2010, a total of 6 subjects with hyperphenylala- [10] S.F. Dobrowolski, C. Heintz, T. Miller, C. Ellingson, C. Ellingson, I. Ozer, G. Gokcay, T.Baykal,B.Thony,M.Demirkol,N.Blau,Moleculargeneticsandimpactof ninemia were detected among 49,657 newborns screened, which residual in vitro phenylalanine hydroxylase activity on tetrahydrobiopterin corresponds roughly to half of all children born in the whole country. responsivenessinTurkishPKUpopulation,Mol.Genet.Metab.102(2011) The calculated incidence based on these numbers is approximately 116–121. 1:8,300. However, during the same time period, 7 symptomatic infants [11] M.D. Bashyam, A.K. Chaudhary, E.C. Reddy, A.R. Devi, G.R. Savithri, R. Ratheesh, L. Bashyam, E. Mahesh, D. Sen, R. Puri, I.C. Verma, S. Nampoothiri, S. Vaidyanathan, and toddlers with PKU were diagnosed who were born outside of Yerevan, M.D. Chandrashekar, P. Kantheti, Phenylalanine hydroxylase gene mutations in which indicates that the incidence may be far more frequent. phenylketonuria patients from India: identification of novel mutations that affect – In summary, we have investigated 35 patients diagnosed with PAH RNA, Mol. Genet. Metab. 100 (2010) 96 99. [12] S. Zare-Karizi, S.M. Hosseini-Mazinani, Z. Khazaei-Koohpar, S.M. Seifati, B. hyperphenylalaninemia in Armenia and analyzed the spectrum of PAH Shahsavan-Behboodi, M.T. Akbari, J. Koochmeshgi, Mutation spectrum of and QDPR gene mutations thereby identifying the well defined phenylketonuria in Iranian population, Mol. Genet. Metab. 102 (2011) 29–32. splicing error in intron 10, the c.1066-11GNA allele, as the most [13] J. Zschocke, J.P. Mallory, H.G. Eiken, N.C. Nevin, Phenylketonuria and the peoples of Northern Ireland, Hum. Genet. 100 (1997) 189–194. prevalent mutation (25%), and found three novel alterations that [14] I. Ozalp, T. Coskun, A. Tokatli, H.S. Kalkanoglu, A. Dursun, S. Tokol, G. Koksal, M. might be specific for the culturally homogenous Armenian population. Ozguc, R. Kose, Newborn PKU screening in Turkey: at present and organization for Supplementary data to this article can be found online at doi:10. future, Turk. J. Pediatr. 43 (2001) 97–101. [15] P. Guldberg, V. Romano, N. Ceratto, P. Bosco, M. Ciuna, A. Indelicato, F. Mollica, C. 1016/j.ymgme.2011.08.006 Meli, M. Giovannini, E. Riva, et al., Mutational spectrum of phenylalanine hydroxylase deficiency in Sicily: implications for diagnosis of hyperphenylalani- – Acknowledgments nemia in southern Europe, Hum. Mol. Genet. 2 (1993) 1703 1707. [16] P. Guldberg, K.F. Henriksen, I. Sipila, F. Guttler, A. de la Chapelle, Phenylketonuria in a low incidence population: molecular characterisation of mutations in Finland, We thank Véronique Wettstein and Bettina Abu Seda for excellent J. Med. Genet. 32 (1995) 976–978. technical work, Christineh Sarkissian and Gábor Mátyás for valuable [17] J.Y. Zhan, Y.F. Qin, Z.Y. Zhao, Neonatal screening for congenital hypothyroidism and phenylketonuria in China, World J. Pediatr. 5 (2009) 136–139. comments, and JoAnne Locher for editorial help. This study was [18] J. Jiang, X. Ma, X. Huang, X. Pei, H. Liu, Z. Tan, L. Zhu, A survey for the incidence of supported by the “Stiftung für Kinder mit angeborenen Gebrechen” phenylketonuria in Guangdong, China, Southeast Asian J. Trop. Med. Public Health and made possible by the “Partnership Program of the Arabkir 34 (Suppl 3) (2003) 185. ” [19] K. Aoki, M. Ohwada, T. Kitagawa, Long-term follow-up study of patients with Hospital Yerevan with the University Children's Hospital in Zurich . phenylketonuria detected by the newborn screening programme in Japan, J. This work was supported in part by The Swiss National Science Inherit. Metab. Dis. 30 (2007) 608. Foundation grant no. 3100A0-1199852/1 (to NB and BT). [20] J. Donlon, H.L. Levy, C.R. Scriver, Hyperphenylalanine: phenylalanine hydroxylase deficiency, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle, B. Childs, K.W. Kinzler (Eds.), The Online Metabolic and Molecular Bases of Inherited Diseases, References McGraw-Hill, New York, 2010. [21] T. Opladen, B. Abu Seda, A. Rassi, B. Thony, G.F. Hoffmann, N. Blau, Diagnosis of [1] N. Blau, F.J. van Spronsen, H.L. Levy, Phenylketonuria, Lancet 376 (2010) 1417–1427. tetrahydrobiopterin deficiency using filter paper blood spots: further [2] B. Thöny, N. Blau, Mutations in the BH4-metabolizing genes GTP cyclohydrolase I, development of the method and 5 years experience, J. Inherit. Metab. Dis. 6-pyruvoyl-tetrahydropterin synthase, sepiapterin reductase, carbinolamine-4a- 34 (2011) 819–826. dehydratase, and dihydropteridine reductase, Hum. Mutat. 27 (2006) 870–878. [22] B. Dworniczak, C. Aulehla-Scholz, L. Kalaydjieva, K. Bartholome, K. Grudda, J. [3] S.W. John, R. Rozen, C.R. Scriver, R. Laframboise, C. Laberge, Recurrent mutation, Horst, Aberrant splicing of phenylalanine hydroxylase mRNA: the major cause for gene conversion, or recombination at the human phenylalanine hydroxylase phenylketonuria in parts of southern Europe, Genomics 11 (1991) 242–246. FOR ELECTRONIC USE ONLY Molecular Genetics and Metabolism 104 (2011) S97–S99

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism

journal homepage: www.elsevier.com/locate/ymgme

Brief Communication Oxidative stress in phenylketonuric patients

C.R. Vargas a,b,c,⁎, M. Wajner a,b, A. Sitta a,b a Programa de Pós-Graduação em Ciências Biológicas:Bioquímica, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil b Serviço de Genética Médica, Hospital de Clínicas de Porto Alegre, Porto Alegre, RS, Brazil c Programa de Pós-Graduação em Ciências Farmacêuticas, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil article info abstract

Article history: Phenylketonuria is the most frequent disturbance of amino acid metabolism. Untreated patients present Received 30 May 2011 mental retardation whose pathophysiology is not completely established. In this work we discuss the Received in revised form 11 July 2011 oxidative stress in phenylketonuric patients. Several studies have shown reduction in antioxidant defenses, Accepted 11 July 2011 possibly due to dietary restriction of nutrients with antioxidant properties and increase in oxidative damage Available online 21 July 2011 to biomolecules, probably secondary to increased formation of reactive species. Therefore, antioxidants could be considered an adjuvant therapy in phenylketonuria. Keywords: Phenylketonuria © 2011 Elsevier Inc. All rights reserved. Oxidative stress Antioxidants

1. Introduction amino acids are precursors of neurotransmitters [3]. Phe metabolites accumulated in this disease could contribute to its pathophysiology, Phenylketonuria (PKU) is an autosomal recessive disorder of being the phenylacetate considered the most toxic component [4]. amino acid metabolism caused by deficiency of L-phenylalanine-4- In recent years, the role of oxidative stress in neuronal damage in hydroxylase, which is a liver-specific enzyme that catalyzes the PKU in animal models and in patients, which would be caused by the hydroxylation of L-phenylalanine (Phe) to tyrosine. Untreated accumulation of metabolites or by an alteration of the antioxidant patients present elevated levels of Phe and its metabolites phenylpyr- system has been investigated [5–10]. Oxidative damage is defined as uvate, phenylacetate and phenyllactate in biological fluids. Treatment an imbalance between the production of oxygen reactive species and for PKU patients consists of restriction of Phe intake, through natural- the decrease in the antioxidant defenses in favor of the former [11]. protein-restricted diets supplemented with phenylalanine-free amino So, in the present work we aim to discuss the parameters of oxidative acid mixtures enriched with trace elements, vitamins, and minerals. stress in PKU patients. Retarded development and intellectual impairment are the most important clinical features presented by untreated PKU patients 2. Oxidative biomarkers in PKU patients whose pathophysiology is not completely understood [1]. It has been postulated that the neurotoxicity of PKU could be Oxidative damage to proteins, lipids and DNA has been investi- attributed to the high concentration of Phe, which shares the same gated in PKU patients. It was verified that malondialdehyde and transport system with other neutral amino acids, difficulting its thiobarbituric acid-reactive species (TBARS) are increased in plasma passage across cell membranes and blood brain barrier (BBB). The from PKU patients at diagnosis and during dietary treatment, plasma and intracellular imbalance may lead to the formation of demonstrating lipid peroxidation in this disease [12–17]. Besides, it abnormal proteins, resulting in a dendritic proliferation and defective showed an increase in carbonyl content and a decrease in sulfhydryl myelination [1]. Moreover, it was shown that high Phe concentrations groups, two protein oxidative damage biomarkers, in treated PKU inhibit the synthesisFOR of brain proteins ELECTRONIC[2] and disrupt the transport of patients. It was suggestedUSE that the PheONLY and/or its metabolites probably tyrosine and tryptophan through the BBB, leading to a reduction in the induced lipid and protein damage, since PKU patients with late synthesis of serotonin, dopamine and norepinephrine, once these diagnosis present the higher levels in these biomarkers [15]. Furthermore, Schulpis and coworkers [18] reported augmented levels of 8-hydroxy-2-deoxyguanosine, a product of DNA oxidative damage Abbreviations: BBB, blood brain barrier; GSH-Px, glutathione peroxidase; GSH, in serum of poorly controlled PKU patients (with high Phe levels), glutathione; Phe, phenylalanine; PKU, phenylketonuria; Q10, ubiquinone-10; TAR, total which were positively correlated with plasma Phe concentrations. antioxidant reactivity; TBARS, thiobarbituric acid-reactive species. Another study showed that Phe induces DNA damage in vivo and in ⁎ Corresponding author at: Serviço de Genética Médica, HCPA, Rua Ramiro Barcelos, 2350 CEP 90035-903, Porto Alegre, RS, Brasil. Fax: +55 51 33598010. vitro in a dose-dependent manner in human leukocytes, indicating E-mail address: [email protected] (C.R. Vargas). that DNA damage in PKU seems to be strongly correlated with the

Table 1 strictly adhered to the diet (Table 1) and that there is a significant Altered oxidative stress parameters and antioxidants in blood from PKU patients at negative correlation between lipid peroxidation and L-carnitine diagnosis and during treatment with restricted-protein diet plus amino acid formula plasma levels, as well as a significant positive correlation between (not containing Phe, L-carnitine and selenium). TAR and plasma L-carnitine in this patients [14]. Since the main Parameter Diagnosis Treatment Reference dietary sources of L-carnitine are milk and red meat, secondary L- Lipid peroxidation Increased Increased [14–17,25] carnitine deficiency may be a consequence of the protein-restricted Protein oxidation NM Increased [15,25] diet [24]. Sitta and coworkers [25] verified that the supplementation DNA damage NM Increased [19] of PKU patients with L-carnitine and selenium was capable to reverse TAR levels Decreased Decreased [14–17] GSH levels NM Decreased [15] the lipid peroxidation and the protein oxidative damage, reinforcing GSH-Px activity Decreased Decreased [15–17,25] the importance of these antioxidant compounds in PKU. L-carnitine levels NM Decreased [14,25] Selenium levels NM Decreased [25] 5. Concluding remarks NM: not measured PKU patients at diagnosis moment and during dietary therapy are susceptible to oxidative damage caused by an increase in free radical blood Phe concentrations [19]. Table 1 describes the parameters of production and by a depletion in antioxidant capacity. The data from oxidative damage to lipids, proteins and DNA in biological fluids from literature suggest that oxidative stress may contribute to the PKU patients at diagnosis (late) and during dietary treatment neurological disturbance in this disease and, therefore, the adminis- (restricted protein diet supplemented with essential amino acids tration of antioxidants should be considered when treating phenyl- formula without Phe, L-carnitine and selenium). ketonuric patients.

3. Antioxidant defenses in PKU patients Acknowledgments

It has been demonstrated that restricted diet to which PKU patients This work was supported in part by grants from CNPq, FAPERGs are submitted leads to a decrease in the antioxidant capacity. Selenium and FIPE/HCPA-Brazil. deficiency in PKU was verified by different authors [5,13,20,21]. Also, erythrocyte glutathione peroxidase (GSHPx) activity is significantly References lower in PKU children since selenium is the cofactor of this enzyme [12]. Decreased selenium levels in PKU were associated to a worsened [1] C.R. Scriver, S. Kaufman, Hyperphenylalaninemia: phenylalanine hydroxylase deficiency, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic performance on the Conner's Continuous Performance Test and the and Molecular Bases of Inherited Disease, McGraw-Hill, New York, 2001, decreased activity of GSHPx was associated with neurological pp. 1667–1724. disturbances in PKU patients [5,13]. Selenium deficiency in these [2] M. Hoeksma, D.J. Reijngoud, J. Pruim, H.W. de Valk, A.M. Paans, F.J. van Spronsen, Phenylketonuria: high plasma phenylalanine decreases cerebral protein synthe- patients was also correlated with the decrease of glutathione (GSH) in sis, Mol. Genet. Metab. 96 (2009) 177–182. erythrocytes from these individuals [16,20].Itisimportantto [3]S.Y.Tam,R.H.Roth,Mesoprefrontaldopaminergic neurons: can tyrosine emphasize that GSH is an endogenous antioxidant compound, which availability influence their functions? Biochem. Pharmacol. 53 (1997) 441–453. [4] S. Kaufman, An evaluation of the possible neurotoxicity of metabolites of is the substrate of the enzyme GSHPx. The decrease in GSH levels could phenylalanine, J. Pediatr. 114 (1989) 895–900. be secondary to an inhibition by Phe on amino acid transport system [5] C. Sierra, M.A. Vilaseca, D. Moyano, N. Brandi, J. Campistol, N. Lambruschini, F.J. for cysteine and glycine or by an increased reactive species production. Cambra, R. Deulofeu, A. Mira, Antioxidant status in hyperphenylalaninemia, Clin. Chim. Acta 276 (1998) 1–9. These observations led to supplementation of selenium, vitamins and [6] R. Artuch, C. Colome, C. Sierra, N. Brandi, N. Lambruschini, J. Campistol, D. Ugarte, trace elements to Phe-free mixtures for PKU patients. M.A. Vilaseca, A longitudinal study of antioxidant status in phenylketonuric In what concerns to non-enzymatic antioxidant defenses, it was patients, Clin. Biochem. 3 (2004) 198–203. verified that the total antioxidant status is decreased in plasma of [7] Y. Sanayama, H. Nagasaka, M. Takayanagi, T. Ohura, O. Sakamoto, T. Ito, M. Ishige- Wada, H. Usui, M. Yoshino, A. Ohtake, T. Yorifuji, H. Tsukahara, S. Hirayama, T. hyperphenylalaninemic and PKU patients, probably attributed to the Miida, M. Fukui, Y. Okano, Experimental evidence that phenylalanine is strongly low levels of selenium, α-tocopherol and uric acid presented by these associated to oxidative stress in adolescents and adults with phenylketonuria, – patients [20]. The quality of the tissue antioxidants was evaluated by Mol. Genet. Metab. 103 (2011) 220 225. [8] K.H. Schulpis, M. Papastamataki, H. Stamou, I. Papassotiriou, A. Margeli, The effect measuring the total antioxidant reactivity (TAR), which was of diet on total antioxidant status, ceruloplasmin, transferrin and ferritin serum decreased in plasma of PKU patients, indicating a deficient tissue levels in phenylketonuric children, Acta Paediatr. 99 (2010) 1565–1570. capacity to modulate the damage associated to an increased free [9] T.B. Moraes, F. Zanin, A. da Rosa, A. de Oliveira, J. Coelho, F. Petrillo, M. Wajner, C.S. – Dutra-Filho, Lipoic acid prevents oxidative stress in vitro and in vivo by an acute radicals production [14 17]. hyperphenylalaninemia chemically-induced in rat brain, J. Neurol. Sci. 15 (2010) 89–95. 4. Antioxidant compounds in PKU treatment [10] C.G. Fernandes, G. Leipnitz, B. Seminotti, A.U. Amaral, A. Zanatta, C.R. Vargas, C.S. Dutra Filho, M. Wajner, Experimental evidence that phenylalanine provokes oxidative stress in hippocampus and cerebral cortex of developing rats, Cell. Mol. Ubiquinone-10 (Q10) is a lipophilic antioxidant that prevents lipid Neurobiol. 30 (2009) 317–326. peroxidation in blood and tissues by limiting the formation of lipid [11] B. Halliwell, J.M.C. Gutteridge, Oxidative stress: adaptation, damage, repair and FOR ELECTRONICdeath, in: B. Halliwell,USE J.M.C. Gutteridge ONLY (Eds.), Free Radicals in Biology and peroxyl radicals and by reducing other natural antioxidant com- Medicine, Oxford University Press, Oxford, 2007, pp. 246–350. pounds, specifically the conversion of tocopheroxyl to tocopherol [12] C. Colome, R. Artuch, M.A. Vilaseca, C. Sierra, N. Brandi, N. Lambruschini, F.J. [12]. This compound is decreased in serum of PKU patients, since the Cambra, J. Campistol, Lipophilic antioxidants in patients with phenylketonuria, – diets are devoid of poultry, meat, soybean products and nuts which Am. J. Clin. Nutr. 77 (2003) 185 188. [13] R. Gassió, R. Artuch, M.A. Vilaseca, E. Fusté, R. Colome, J. Campistol, Cognitive fi are sources of Q10 [6,12,13,22]. Colomé and coworkers [12] veri ed functions and the antioxidant system in phenylketonuric patients, Neuropsych. – that the low Q10 values in PKU patients are associated with high levels 22 (2008) 426 431. of plasma malondialdehyde, suggesting an important role of this [14] A. Sitta, A.G. Barschak, M. Deon, J.F. De Mari, A.B. Barden, C. Vanzin, G.B. Biancini, I. V.D. Schwartz, M. Wajner, C.R. Vargas, L-Carnitine blood levels and oxidative substance in the prevention of lipid peroxidation. stress in treated phenylketonuric patients, Cell Mol. Neurobiol. 29 (2009) L-Carnitine exerts an important role by facilitating the mitochon- 211–218. drial transport of fatty acids and is also a scavenger of free radicals, [15] A. Sitta, A.G. Barschak, M. Deon, A.T. Barden, G.B. Biancini, P.R. Vargas, C.F. de Souza, C. Netto, M. Wajner, C.R. Vargas, Effect of short- and long-term exposition protecting cells from oxidative damage [23]. It was demonstrated that to high phenylalanine blood levels on oxidative damage in phenylketonuric blood L-carnitine levels are decreased in treated PKU patients who patients, Int. J. Dev. Neurosci. 27 (2009) 243–247. C.R. Vargas et al. / Molecular Genetics and Metabolism 104 (2011) S97–S99 S99

[16] A. Sitta, A.G. Barschak, M. Deon, T. Terroso, R. Pires, R. Giugliani, C.S. Dutra-Filho, [21] B.C. Wilke, M. Vidailhet, A. Favier, C. Guillemin, V. Ducros, J. Arnaud, M.J. Richard, M. Wajner, C.R. Vargas, Investigation of oxidative stress parameters in treated Selenium, glutathione peroxidase (GSH-Px) and lipid peroxidation products phenylketonuric patients, Metab. Brain Dis. 21 (2006) 287–296. before and after selenium supplementation, Clin. Chim. Acta 207 (1992) [17] L.R. Sirtori, C.S. Dutra-Filho, D. Fitarelli, A. Sitta, A. Haeser, A.G. Barschak, M. 137–142. Wajner, D.M. Coelho, S. Llesuy, A. Belló-Klein, R. Giugliani, M. Deon, C.R. Vargas, [22] R. Artuch, M.A. Vilaseca, J. Moreno, N. Lambruschini, F.J. Cambra, J. Campistol, Oxidative stress in patients with phenylketonuria, Biochim. Biophys. Acta 1740 Decreased serum ubiquinone-10 concentrations in phenylketonuria, Am. J. Clin. (2005) 68–73. Nutr. 70 (1999) 892–895. [18] K.H. Schulpis, S. Tsakiris, J. Traeger-Synodinos, I. Papassotiriou, Low total [23] I. Gülçin, Antioxidant and antiradical activities of L-carnitine, Life Sci. 78 (2006) antioxidant status is implicated with high 8-hydroxy-2-deoxyguanosine serum 803–811. concentrations in phenylketonuria, Clin. Biochem. 38 (2005) 239–242. [24]K.H.Schulpis,C.Nounopoulos,A.Scarpalezou,A.Bouloukos,S.Missiou- [19] A. Sitta, V. Manfredini, L. Biasi, R. Treméa, I.V.D. Schwartz, M. Wajner, C.R. Vargas, Tsagarakis, Serum carnitine level in phenylketonuric children under dietary Evidence that DNA damage is associated to phenylalanine blood levels in control in Greece, Acta Paediatr. Scand. 79 (1990) 930–934. leukocytes from phenylketonuric patients, Mutat. Res. 679 (2009) 13–16. [25] A. Sitta, C.S. Vanzin, G.B. Biancini, V. Manfredini, A.B. de Oliveira, C.A.Y. Wayhs, G.O. [20] M.M. Van Bakel, G. Printzen, B. Wermuth, U.N. Wiesmann, Antioxidant and Ribas, L. Giugliani, I.V.D. Schwartz, D. Bohrer, S.C. Garcia, M. Wajner, C.R. Vargas, thyroid hormone status in selenium-deﬁcient phenylketonuric and hyperpheny- Evidence that L-carnitine and selenium supplementation reduces oxidative stress lalaninemic patients, Am. J. Clin. Nutr. 72 (2000) 976–981. in phenylketonuric patients, Cell. Mol. Neurobiol. 31 (2011) 429–436.

FOR ELECTRONIC USE ONLY