Mucopolysaccharidosis IVA) Identiﬁes 39 Novel GALNS Mutations

YMGME-05730; No. of pages: 11; 4C: 4, 5, 8 Molecular Genetics and Metabolism xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism

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

Molecular testing of 163 patients with Morquio A (Mucopolysaccharidosis IVA) identiﬁes 39 novel GALNS mutations

A. Morrone a,b,1,K.L.Tyleec,1,M.Al-Sayedd, A.C. Brusius-Facchin e, A. Caciotti a, H.J. Church c,M.J.Collf, K. Davidson g, M.J. Fietz h, L. Gort f,M.Hegdei,F.Kubaskie,L.Lacerdaj,F.Laranjeiraj, S. Leistner-Segal e, S. Mooney k,S.Pajaresf, L. Pollard l,I.Ribeiroj,R.Y.Wangm,N.Millerg,⁎ a Molecular and Cell Biology Laboratory, Pediatric Neurology Unit and Laboratories, Meyer Children's Hospital, Florence, Italy b Department of Neurosciences, Psychology, Pharmacology and Child Health, University of Florence, Florence Italy c Willink Biochemical Genetics, Central Manchester University Hospitals NHS Foundation Trust, Saint Mary's Hospital Oxford Road, Manchester, UK d Department of Medical Genetics, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia e Laboratório de Genética Molecular, Serviço de Genética Médica, Hospital de Clínicas de Porto Alegre (HCPA), Porto Alegre, Brazil f Sección de Errores Congénitos del Metabolismo-IBC, Servicio de Bioquímica y Genética Molecular, Hospital Clínic, CIBERER, IDIBAPS, Barcelona, Spain g BioMarin Pharmaceutical Inc., Novato, CA, USA h SA Pathology, Women's and Children's Hospital, North Adelaide, SA, Australia i Emory Genetics Laboratory, Emory University School of Medicine, Atlanta, GA, USA j Unidade de Bioquímica Genética, Centro de Genética Médica Jacinto Magalhães (CGMJM) do Centro Hospitalar do Porto (CHP), Porto, Portugal k The Buck Institute for Research on Aging, Novato, CA, USA l Biochemical Genetics Laboratory, Greenwood Genetic Center, Greenwood, SC, USA m Children's Hospital of Orange County, Orange, CA, USA article info abstract

Article history: Morquio A (Mucopolysaccharidosis IVA; MPS IVA) is an autosomal recessive lysosomal storage disorder Received 18 January 2014 caused by partial or total deficiency of the enzyme galactosamine-6-sulfate sulfatase (GALNS; also known as Received in revised form 11 March 2014 N-acetylgalactosamine-6-sulfate sulfatase) encoded by the GALNS gene. Patients who inherit two mutated Accepted 12 March 2014 GALNS gene alleles have a decreased ability to degrade the glycosaminoglycans (GAGs) keratan sulfate and chon- Available online xxxx droitin 6-sulfate, thereby causing GAG accumulation within lysosomes and consequently pleiotropic disease. GALNS mutations occur throughout the gene and many mutations are identified only in single patients or fami- Keywords: fi MPS IVA lies, causing dif culties both in mutation detection and interpretation. In this study, molecular analysis of 163 pa- Morquio A tients with Morquio A identified 99 unique mutations in the GALNS gene believed to negatively impact GALNS Mucopolysaccharidosis type IVA protein function, of which 39 are previously unpublished, together with 26 single-nucleotide polymorphisms. GALNS Recommendations for the molecular testing of patients, clear reporting of sequence findings, and interpretation Lysosomal storage disorder of sequencing data are provided. Mutation © 2014 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

1. Introduction (KS) and chondroitin-6-sulfate (C6S) in lysosomes and results in prominent skeletal and connective tissue abnormalities. In addition to skele- Morquio A syndrome (also known as mucopolysaccharidosis type IV tal and connective tissue abnormalities, patients also experience joint A or MPS IVA) is a member of a group of inherited metabolic disorders hypermobility [2], muscle weakness, and pulmonary and cardiac mani- collectively termed mucopolysaccharidoses (MPSs). An MPS disorder festations of the disease, all of which can result in reduced endurance is caused by a deficiency of one of the 11 different lysosomal enzymes and impact both quality of life and mortality. Compared with patients required for the degradation of mucopolysaccharides or glycosamino- with other MPS disorders, those with Morquio A often show joint glycans (GAGs). Morquio A is caused by a deficiency of galactosamine- hypermobility in contrast to joint stiffness. Morquio A often shows 6-sulfatase (N-acetyl-galactosamine-6-sulfate sulfatase; GALNS). prominent spine involvement [50], but patients are typically reported GALNS deficiency leads to the accumulation of the GAGs keratan sulfate to be intellectually normal. The clinical presentation of Morquio A varies among patients in both the specificfeaturesobservedandtheirseverity, making any single metric an incomplete description of disease burden ⁎ Corresponding author at: BioMarin Pharmaceutical Inc., 105 Digital Dr., Novato, CA [17,18,34,57]. 94949, USA. E-mail address: [email protected] (N. Miller). Morquio A is a rare disorder, with incidence estimated to range from 1 Co-first authors. 1 in 76,000 to 1 in 640,000 live births in different populations [18].The

http://dx.doi.org/10.1016/j.ymgme.2014.03.004 1096-7192/© 2014 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Please cite this article as: A. Morrone, et al., Molecular testing of 163 patients with Morquio A (Mucopolysaccharidosis IVA) identifies 39 novel GALNS mutations, Mol. Genet. Metab. (2014), http://dx.doi.org/10.1016/j.ymgme.2014.03.004 2 A. Morrone et al. / Molecular Genetics and Metabolism xxx (2014) xxx–xxx incidence has been reported as one in 76,000 in Northern Ireland, one in separate PCR reactions amplify the GALNS exons and short stretches 640,000 in Western Australia, one in 450,000 in the Netherlands, and of the adjacent sequence, followed by sequencing of the individual one in 450,000 in Portugal [30,37,38,43,44]. More accurate data about PCR products. However, standard sequencing approaches can yield mis- the incidence of Morquio A can be obtained in the future if screening leading results when one patient allele contains deletions or point mu- of newborns is introduced, as this occurs for other lysosomal storage tations that eliminate PCR primer binding sites, resulting in apparent disorders such as Pompe and Fabry diseases [25,35,49]. homozygosity due to allele dropout [27]. Parental testing can confirm Diagnosis of Morquio A begins with clinical suspicion, followed by biallelic inheritance and indicate cases where deletion/duplication screening tests (which are sometimes omitted if there is a known family testing is necessary or uniparental isodisomy (UPD) may be a possibility history). The Morquio A diagnostic algorithm recommends a GALNS en- [8]. However, not all mutations can be detected by standard molecular zyme activity assay performed in leukocytes or fibroblasts as the gold approaches, and sometimes the clinical significance of newly detected standard for diagnosis of Morquio A [65]; diagnosis of Morquio A can sequence alterations will not be clear. For example, standard GALNS be supported by molecular analysis of the GALNS gene [65]. Screening sequencing approaches do not sequence deep within most introns, tests that may also be used for Morquio A are urinary GAG analysis and the functional consequences of deep intronic mutations can be and/or enzyme activity analysis performed on dried blood spots. difficult to assess. Molecular analysis reports should clearly state Urinary GAG analysis measures either the total accumulation of all uri- which conclusions are possible for any detected sequence alterations. nary GAGs (quantitative assay) or the relative abundance of each of the In this study, we report the molecular testing that led to the identi- GAGs (qualitative assay). It is recommended to perform both quantita- fication of genetic lesions in GALNS in 163 patients with Morquio A. tive and qualitative urinary GAG analyses in parallel, because quantita- We identified 99 mutations believed to be disease associated or that tive GAGs are not always elevated in Morquio A patients and both tests are likely disease associated, of which 39 are previously unpublished. are susceptible to false-negative results due to low KS excretion Of the 39 novel mutations, 25 are missense mutations, six are nonsense (relative to other GAGs) in teenagers and adults [59,63–65]. Enzyme mutations, five are small deletions, two are intronic mutations that like- assays performed on dried blood spot samples are an alternative screen- ly affect splicing sites, and one is a large deletion-duplication. We also ing tool [7] but are not recommended for Morquio A diagnosis where propose guidelines for the interpretation and reporting of GALNS muta- alternatives exist, since assay robustness and sample quality are poten- tions based on the supporting information available. Using these classi- tial concerns [65]. A liquid chromatography/tandem mass spectrome- fication guidelines, 16 of the 39 novel changes are “disease associated” try-based approach may also be used to measure levels of keratanase and 23 are “likely disease associated”. In addition to the 39 novel chang- II-digested mono- and di-sulfated KS disaccharides, providing a means es believed to be disease associated or that are likely disease associated, to measure KS both quantitatively and qualitatively at the same time two novel changes were detected that are classified as “variants of un- [19,29,41,53,56]. A diagnosis of Morquio A is established if GALNS en- known significance ”. We also identified 26 separate single-nucleotide zyme activity is markedly decreased in the fibroblasts or leukocytes polymorphisms (SNPs). Finally, we provide guidelines and recommen- and control enzymes display wild-type activity [65]. Additional refer- dations for molecular testing and highlight situations where inaccurate ence enzyme measurements are critical to confirm sample integrity genotyping conclusions may occur. and exclude other disorders, such as MPS VI (caused by loss of arylsulfatase B activity; patients with Morquio A have been 2. Materials and methods misdiagnosed with MPS VI), Morquio B (caused by a deficiency of β- galactosidase due to mutations in GLB1; patients with Morquio B have Individuals genotyped in this study had received a diagnosis of been misdiagnosed with Morquio A), multiple sulfatase deficiency (mu- Morquio A prior to and independent of any molecular testing results; tations in the SUMF1 gene result in reduced activity of multiple sulfa- and thus, no power calculations were performed or required for the tases, including GALNS), and mucolipidoses types II/III (leads to purposes of this study. All detected mutations in the GALNS gene were mislocalization of GALNS and other lysosomal enzymes in some checked against reported SNPs. At three of the six institutions reporting tissues). novel alterations in GALNS, novel alterations were shown to be absent The GALNS gene is approximately 50 kb long and contains 14 exons, from the GALNS sequences of at least 100 unaffected individuals; at producing a 2339-bp mRNA that encodes a 522-amino acid protein [36, the remaining three institutions, this was not routinely performed for 55]. The protein structure of the human GALNS protein has recently GALNS alterations identified in patients with a diagnosis of Morquio A. been solved [47]. The GALNS active site is a large trench containing a The results reported here 1) have been reviewed and approved by a catalytic formylglycine aldehyde, derived from a cysteine residue by duly constituted ethics committee (Hospital Clínic, Barcelona, Spain; action of the formylglycine-generating enzyme (FGE) [9,12,13,47].The and Hospital de Clínicas de Porto Alegre, Porto Alegre, Brazil), 2) are GALNS protein is found as a homodimer [45] and is described as occur- from samples referred as diagnostic samples, which do not require ring in a multiprotein complex with other lysosomal enzymes [1,45]. ethical approval at this institution (Willink Biochemical Genetics Unit, The GALNS mutations that cause Morquio A are very heterogeneous Manchester, United Kingdom), or 3) are from retrospective case reports and are detected throughout the gene [60]. Even the most frequently that do not require ethics committee approval at these institutions detected mutations are relatively uncommon [60]; however, founder (Unidade de Bioquímica Genética, Porto, Portugal; Children's Hospital of effects can greatly alter GALNS allele frequencies in individual popula- Orange County, Orange, California, United States; King Faisal Specialist tions [24,65,66]. DNA methylation at CpG sites occurs in every exon Hospital and Research Center, Riyadh, Saudi Arabia; and SA Pathology, but one, and inappropriate repair is thought to lead to transition muta- North Adelaide, South Australia, Australia). tions at these sites [58]. Multiple introns contain Alu repetitive elements, which can undergo recombination and lead to large deletions 2.1. Patient material and molecular analysis and/or rearrangements (http://genome.ucsc.edu/; February 2009 as- sembly; [31]). This mutational heterogeneity can lead to difficulties in Genomic DNA was isolated from fibroblasts or peripheral blood cells interpretation of molecular testing results, as novel mutations/variants using standard protocols. GALNS exons and adjacent intron regions were of unknown significance may be detected relatively frequently. amplified by PCR reactions and then sequenced; in some cases, a single- Molecular analysis can confirm the Morquio A diagnosis and aid in strand conformation polymorphism assay was also performed, all using genetic counseling by detecting causative mutations in the GALNS standard protocols (see Supplementary methods in Appendix A). gene. Morquio A is an autosomal recessive disorder, so for disease to Molecular testing was performed at the laboratories included in this occur, both GALNS alleles must carry mutations that decrease or elimi- publication, and no data from BioMarin Pharmaceutical–sponsored nate GALNS enzyme activity. In standard DNA sequencing approaches, studies are reported.

The DNA and protein sequence numbering was based on the GALNS occurrence in the molecular analysis of patients with Morquio A. Here, cDNA sequence (GenBank entry NM_000512.4), with the sequence po- we have created and used a GALNS sequence alteration categorization sition +1 corresponding to the A of the initial ATG in the reference se- and reporting system consistent with guidelines from the Human Ge- quence. Novel mutations are defined as nucleotide changes in the nome Variation Society and Emory Genetics Laboratory (Table 1). This GALNS gene that have not previously been described and published; reporting system aims to communicate succinctly what can reasonably they are classified by the supporting evidence available per Table 1. be concluded about a detected GALNS sequence alteration. For example, reasonably confident statements can be made about sequence alter- 2.2. GALNS enzyme assay and urinary GAG testing ations that have clear-cut predicted effects on the GALNS protein and alterations that have been detected multiple times in independent fam- GALNS enzyme activity was measured in leukocytes or cultured fi- ilies, while more cautious statements are appropriate for missense alter- broblasts using either fluorogenic [69] or radiolabeled substrates [20]; ations detected in a single patient. We believe that this reporting system assays were performed according to the manufacturer's protocols. is useful both in publishing GALNS alterations and in communicating Total urinary GAGs were quantified using the Alcian blue or GALNS sequence results to clinicians. dimethylmethylene blue (DMB) method [42]. Qualitative GAG analysis for the detection of KS was carried out by one dimensional or two 3.1.1. Missense alleles dimensional low voltage electrophoresis of extracted GAGs, or by thin Among the 25 novel missense changes detected, the most fre- layer chromatography [21,22]. quently detected novel allele was the missense change c.347GNT (p.Gly116Val), detected 14 times in seven patients. In every case, 3. Results molecular testing indicated that patients were homozygous for the c.347GNT (p.Gly116Val) allele; of these seven patients, the ethnicity 3.1. Detected alterations in GALNS of six was ‘Asian-multiethnic’ and one was Norwegian (but possibly of Asian origin). Where assayed, patients homozygous for the c.347GNT In this study, molecular testing was performed on 163 patients with (p.Gly116Val) change were diagnosed with Morquio A at three years of Morquio A (Supp. Table S1). A total of 99 predicted sequence alterations age or younger and had substantial reductions in GALNS enzyme activity suspected of negatively impacting GALNS function were identified, (6% or less of wild-type). While the specific c.347GNT (p.Gly116Val) of which 39 are previously unpublished: 25 missense changes, six change has not previously been published, c.346GNA (p.Gly116Ser) nonsense changes, five small deletions (one in-frame deletion, three is a previously reported mutation associated with a severe growth frameshifts, and one deletion that encompasses exonic and intronic phenotype [14,47,58,60]. In addition to c.347GNT (p.Gly116Val), the sequences), one complex large deletion-duplication, and two intronic novel alterations c.107TNG (p.Leu36Arg), c.497ANG (p.His166Arg), changes affecting consensus splice sites (Figs. 1, 2, Table 2,Supp. c.865ANG (p.Asn289Asp), c.1138ANG (p.Arg380Gly), and c.1259CNG Table S2; information from all patients in Supp. Table S1). Six of these (p.Pro420Arg) also affect amino acid residues where other missense novel changes were detected in more than one patient: c.347GNT mutations have previously been reported [54,60,62,67]. (p.Gly116Val, found in seven patients), c.107TNG (p.Leu36Arg, found Three novel missense changes that directly impact the active site in three patients), c.602GNA (p.Gly201Glu, found in three patients, of of the GALNS enzyme [47] were identified: c.242CNT (p.Pro81Leu), whom two are siblings), c.422GNA (p.Trp141Ter, found in two pa- c.118GNA (p.Asp40Asn) and c.865ANG (p.Asn289Asp) (Fig. 2). The tients), c.242CNT (p.Pro81Leu, found in two patients), and c.1155CNA change c.242CNT (p.Pro81Leu) alters the middle residue of the FGE (p.Tyr385Ter, found in two patients); all other novel mutations were recognition motif and may reduce or eliminate the FGE-dependent only detected in a single patient. In addition to the 39 novel GALNS posttranslational generation of the GALNS catalytic residue. Consistent alterations associated with Morquio A, we also detected two missense with this interpretation, a patient homozygous for the change variants of unknown significance (Table 2, Supp. Table S1, Supp. c.242CNT (p.Pro81Leu) had substantially reduced levels of GALNS en- Table S2) and 26 separate SNPs (Supp. Table S3). zyme activity. The GALNS active site contains a Ca2+ ion coordinated The sequence alterations in GALNS associated with Morquio A are by oxygen atoms from five residues [47]. The overall negative charge co- numerous, heterogeneous, and are mostly missense mutations ([18, ordinating the calcium ion would change with c.118GNA (p.Asp40Asn) 60]; unpublished results). While some sequence alterations have and c.865ANG (p.Asn289Asp), likely with deleterious effects on Ca2+ clear-cut impacts on GALNS enzyme function (e.g., nonsense mutations, binding. Additionally, the primary active site residue Arg83 [47] is adja- large deletions, or frameshifts), the interpretation of detected missense cent to an amino acid affected by the novel missense change c.251CNA mutations can be more difficult. This poses a challenge to those who in- (p.Ala84Glu), which would substitute a bulky charged residue for a terpret sequence alterations in GALNS, since the detection of novel or small nonpolar residue and may impact the structure and charge poorly characterized missense mutations is a relatively common of the enzyme active site. A patient homozygous for the change

Table 1 Suggested GALNS mutation reporting guidelines.

Mutation disease associated Mutation likely disease associated Variant of unknown signiﬁcance (VUS) Likely benign variant

Known to cause disease and found in Disease diagnosis conﬁrmed Ambiguous/unclear effect on protein Nucleotide change present at high more than one family in association and Rare in population frequency in population with enzyme defect unlikely to be benign variant Reported in single individual/family or or and one of below: with inadequate clinical or segregation accurate published evidence clear-cut effect of mutation on protein: • Likely to effect splicing information characterizes as benign • Frameshift or polymorphism • Nonsense • Nucleotide change in trans to known mutation • Catalytic residue or mutation that is disease associated • Large deletions/insertions or • Initiation/termination codon changed • Reported in homozygous state in single individual/family or • Signiﬁcantly alters characterized active site residue

Mutation reporting based on recommendations of Emory Genetics Laboratory [40,46] and the Human Genome Variation Society [10] (http://www.hgvs.org/mutnomen/).

c.758+4a>t Intronic c.120+1g>c

deletion in Intron X duplication encompassing Intron V to Intron IX Deletions p.Val427SerfsTer13 p.Leu372SerfsTer6 c.405_422+1del p.Pro357ArgfsTer21 p.Glu477_Gln485del

Nonsense p.Trp141Ter Mutations p.Arg251Ter p.Glu126Ter p.Tyr209Ter p.Tyr385Ter p.Gln414Ter 189 107 141 211 253 300 344 380 414 454 494 522

40 82

Amino Acids 1 I 41 82 III IV V VI VII VIII IX X XI XII XIII XIV II 107 141 189 212 253 300 335 380 415 455 495

* p.Val16Glu * p.Gly116Val p.Asn289Asp p.Arg380Gly p.Ala492Thr p.Leu36Arg p.His145Tyr p.Gly415Val p.Gly500Ser p.Asp40Asn

p.Phe156Leu p.Ile416Thr p.Cys507Phe

p.Val48Gly * p.His166Arg p.Pro420Arg Missense p.Glu51Lys p.Gly201Glu p.Pro81Leu Mutations p.Leu214Pro p.Ala84Glu p.Phe216Ser p.Leu91Pro p.Thr235Lys p.Ser264Thr

* Mutation may impact GALNS active site Allele classification, as per Table 1: locations of Alu elements between exons Disease-associated Likely disease-associated Gray-filled exons indicate detected DNA methylation (Tomatsu et al., 2004)

Fig. 1. Location of 39 novel mutations detected in the GALNS gene. Exons are depicted as boxes and labeled with roman numerals. Novel alleles are color coded by their classiﬁcation per Table 1.

Fig. 2. Location of novel Morquio A missense mutations mapped on the protein structure of the human GALNS protein. Novel missense mutations identiﬁed in this study are mapped on the human GALNS protein structure (pdb:4FDI). Mutations identiﬁed were mapped to the structure and visualized using UCSF Chimera (http://www.cgl.ucsf.edu/chimera). The protein chain is represented as a ribbon showing beta sheets as light purple and alpha helices as light red, surrounded by a transparent solvent excluded surface. Mutation positions are represented as spheres for the wild-type side chains. Note that position 16 is in the N-terminal disordered region. Mutations colored in orange are predicted to affect GALNS active site primary residues; all others are colored in blue.

(p.Pro81Leu), c.118GNA (p.Asp40Asn) and c.865ANG (p.Asn289Asp) change c.47TNA (p.Val16Glu), which mutation predicting programs do are all predicted to greatly alter amino acid residues characterized as not strongly associate with disease. The c.47TNA (p.Val16Glu) allele “primary active site residues” [47] and so can also be predicted to affect was documented in a patient with a less severe Morquio A growth phe- protein function. notype, but with accompanying detection of KS and elevation of urinary This study identified 25 novel changes predicted to result in mis- GAGs, together with a demonstrated GALNS enzyme activity defect in sense mutations in the GALNS protein either associated or likely to be both dried blood spots and leukocytes and normal activity of the control associated with Morquio A disease; additionally, the novel missense enzymes β-galactosidase, α-iduronidase, arylsulfatase B, iduronate sul- changes c.[937ANG; 977GNC] (p.[Thr313Ala; Trp326Ser]) were found fatase, and β-glucuronidase; further, this patient had a pedigree indicat- in cis in one allele of a patient with Morquio A (both in trans with the ing closely related parents. The novel splice site alterations reported known allele c.740GNA (p.Gly247Asp)) and are both currently classi- here are c.120+1GNC and c.758+4ANT. While the c.120 + 1GNC fied as “variants of unknown significance”. For a subset of these novel mutation is predicted to be a clear splicing defect, the c.758+4A N T missense changes, previous studies have identified Morquio A patients change is less well characterized as a splicing defect and we recom- with different missense changes at the same amino acid residue: mend, as for all changes categorized as “likely disease associated”,that p.Leu36Arg is a novel change, but p.Leu36Pro has been described [60], if it should be identified in a second family, a defect in GALNS enzyme p.Gly116Val (p.Gly116Ser [14,52]), p.Phe156Leu (p.Phe156Cys and activity be demonstrated. p.Phe156Ser [4,60,66]), p.His166Arg (p.His166Gln [58,60]), p.Asn289Asp (p.Asn289Ser [62]), p.Arg380Gly (p.Arg380Ser and 4.2. Challenges in interpretation p.Arg380Thr [32,52,60]), and p.Pro420Arg (p.Pro420Ser [67]). These previously published missense changes at the same amino acid residues The high proportion of detected novel changes emphasizes the het- may strengthen the link between these novel changes and disease. erogeneity in GALNS mutations. This heterogeneity creates challenges in Currently, the following mutations have only been described in the interpretation of patient genotypes as many patients will carry one family and changes at these amino acids have not been previous- novel or poorly characterized mutations. For example, some patients ly noted but are associated with a GALNS enzyme defect in the with identified deleterious mutations in both GALNS alleles also had ad- patient: c.47TNA (p.Val16Glu), c.120+1GNC, c.143TNG (p.Val48Gly), ditional GALNS sequence variants detected (Supp. Table S3). These var- c.151GNA (p.Glu51Lys), c.251CNA (p.Ala84Glu), c.272TNC iants have been reported as SNPs even if an eventual role in modulating (p.Leu91Pro), c.433CNT (p.His145Tyr), c.466TNC (p.Phe156Leu), phenotype (i.e. aberrant splicing induction) cannot be excluded. Given c.497ANG (p.His166Arg), c.641TNC (p.Leu214Pro), c.647TNC the potential complexity and ambiguity in interpreting the functional (p.Phe216Ser), c.704CNA (p.Thr235Lys), c.758+4ANT, c.791GNC consequences of a mutation, we recommend standardized molecular (p.Ser264Thr), c.865ANG (p.Asn289Asp), c.1138ANG (p.Arg380Gly), testing reports that clearly state what conclusions can be drawn c.1244GNT (p.Gly415Val), c.1247TNC (p.Ile416Thr), c.1259CNG (Table 1). It is recommended that physicians collect as much as possible (p.Pro420Arg), c.1474GNA (p.Ala492Thr), c.1498GNA (p.Gly500Ser), of a patient's clinical phenotype data and laboratory screening results and c.1520GNT (p.Cys507Phe); these changes are classified as “likely together with the age of diagnosis; however, as evidenced by our expe- disease associated,” and it is recommended that when found in a second rience (see Supp. Table S1), this is often not practically possible. To aid in family, patients should be verified to have a GALNS enzyme activity de- the interpretation of sequencing data, it is important for detected muta- fect to confirm a diagnosis of Morquio A. Of particular note is the novel tions to be reported to available mutation databases, such as the Human

Please cite this article as: A. Morrone, et al., Molecular testing of 163 patients with Morquio A (Mucopolysaccharidosis IVA) identiﬁes 39 novel GALNS mutations, Mol. Genet. Metab. (2014), http://dx.doi.org/10.1016/j.ymgme.2014.03.004 6 laect hsatcea:A orn,e l,Mlclrtsigo 6 ainswt oqi McplschrdssIA identi IVA) (Mucopolysaccharidosis A Morquio with patients 163 of testing Molecular al., et Morrone, A. as: article this cite Please AN uain,Ml ee.Mtb (2014), Metab. Genet. Mol. mutations, GALNS .Mroee l oeua eeisadMtbls x 21)xxx (2014) xxx Metabolism and Genetics Molecular / al. et Morrone A. http://dx.doi.org/10.1016/j.ymgme.2014.03.004

Table 2 Genotypes of patients with Morquio A carrying novel GALNS mutations identiﬁed in this study.

Allele 1 Allele 2 GALNS enzyme activity

1 c.47 TNAp.Val16Glu Likely disease associated c.47 TNAp.Val16Glu Brazilian F 10 Yes 9 HCPA 2 c.107TNG p.Leu36Arg Disease associated c.107TNG p.Leu36Arg Asian-multiethnic F 2 Yes 6 Willink 3 c.107TNG p.Leu36Arg Disease associated c.107TNG p.Leu36Arg Asian-multiethnic Yes 4 Willink 4 c.107TNG p.Leu36Arg Disease associated c.107TNG p.Leu36Arg Asian-multiethnic F 12 Yes 5 Willink 5 c.118GNA p.Asp40Asn Likely disease associated c.118GNA p.Asp40Asn Asian-multiethnic F Willink 6 c.120+1gNc- Likely disease associated c.120+1gNc- Middle Eastern M 0.5 Yes 3 Willink 7 c.151GNA p.Glu51Lys Likely disease associated c.151GNA p.Glu51Lys Brazilian F 4 Yes 0.6 HCPA – 8 c.242CNT p.Pro81Leu Disease associated c.242CNT p.Pro81Leu Turkish M Yes 6 Willink xxx 9 c.251CNAp.Ala84Glu Likely disease associated c.251CNAp.Ala84Glu Portuguese M 4 Yes 0 CGMJM 10 c.272TNC p.Leu91Pro Likely disease associated c.272TNC p.Leu91Pro Portuguese F 3 Yes 0 CGMJM 11 c.347GNT p.Gly116Val Disease associated c.347GNT p.Gly116Val Asian-multiethnic Willink 12 c.347GNT p.Gly116Val Disease associated c.347GNT p.Gly116Val Asian-multiethnic F 3 Yes 3 Willink 13 c.347GNT p.Gly116Val Disease associated c.347GNT p.Gly116Val Asian-multiethnic M 3 Yes 2 Willink 14 c.347GNT p.Gly116Val Disease associated c.347GNT p.Gly116Val Asian-multiethnic F Willink 15 c.347GNT p.Gly116Val Disease associated c.347GNT p.Gly116Val Asian-multiethnic F 0.08 Yes 2 Willink 16 c.347GNT p.Gly116Val Disease associated c.347GNT p.Gly116Val Asian-multiethnic M 1 Yes 5 Willink 17 c.347GNT p.Gly116Val Disease associated c.347GNT p.Gly116Val Norwegian M Yes Willink 18 c.376GNT p.Glu126Ter Disease associated c.865ANG p.Asn289Asp Australian 4.4 Yes 1.7 SA Path 19 c.422GNA p.Trp141Ter Disease associated c.422GNA p.Trp141Ter UK M 3 Yes 6 Willink 20 c.433CNT p.His145Tyr Likely disease associated c.433CNT p.His145Tyr Canadian 4.7 Yes SA Path 21 c.466TNC p.Phe156Leu Likely disease associated c.466TNC p.Phe156Leu Middle Eastern F 2 Yes KFSH&RC ﬁ

s3 novel 39 es 22 c.497ANG p.His166Arg Likely disease associated c.497ANG p.His166Arg UK F Yes Willink 23 c.602GNA p.Gly201Glu Disease associated c.602GNA p.Gly201Glu Middle Eastern Yes Willink 24 c.602GNA p.Gly201Glu Disease associated c.602GNA p.Gly201Glu Middle Eastern 5.2 SA Path 25 c.602GNA p.Gly201Glu Disease associated c.602GNA p.Gly201Glu Middle Eastern Yes SA Path 26 c.627CNG p.Tyr209Ter Disease associated c.627CNG p.Tyr209Ter F 4 Yes Willink Allele 1 Allele 2 GALNS enzyme activity

Case Nucleotide change Predicted effect on protein Classiﬁcation of novel Nucleotide change Predicted effect Reported country/ M/F Age at diagnosis Reported % Wild-type Reporting number allele(s) per Table 1 on protein ethnicity (years) as low? institute N N laect hsatcea:A orn,e l,Mlclrtsigo 6 ainswt oqi McplschrdssIA identi IVA) (Mucopolysaccharidosis A Morquio with patients 163 of testing Molecular al., et Morrone, A. as: article this cite Please AN uain,Ml ee.Mtb (2014), Metab. Genet. Mol. mutations, GALNS 27 c.641 T C p.Leu214Pro Likely disease associated c.641 T C p.Leu214Pro Turkish Yes 4 Willink 28 c.647TNC p.Phe216Ser Likely disease associated c.647TNC p.Phe216Ser Middle Eastern Yes 8 Willink 29 c.791GNC p.Ser264Thr Likely disease associated c.791GNC p.Ser264Thr UK M 2 Yes 4 Willink 30 c.1070delC p.Pro357ArgfsTer21 Disease associated c.1070delC p.Pro357ArgfsTer21 Middle Eastern M 1 KFSH&RC 31 c.1114delC p.Leu372SerfsTer6 Disease associated c.1114delC p.Leu372SerfsTer6 Portuguese M 10 Yes 0 CGMJM 32 c.1138ANG p.Arg380Gly Likely disease associated c.1138ANG p.Arg380Gly Cape Verdean M 3 Yes 0 CGMJM 33 c.1244GNT p.Gly415Val Likely disease associated c.1244GNT p.Gly415Val Portuguese M 15 Yes 0 CGMJM 34 c.1247TNC p.Ile416Thr Likely disease associated c.1247TNC p.Ile416Thr Asian-multiethnic M Yes Willink 35 c.1259CNG p.Pro420Arg Likely disease associated c.1259CNG p.Pro420Arg Asian-multiethnic F Yes Willink 36 c.1429_1455del27 p.Glu477_Gln485del Disease associated c.1429_1455del27 p.Glu477_Gln485del Middle Eastern F 13 KFSH&RC 37 c.1474GNA p.Ala492Thr Likely disease associated c.1474GNA p.Ala492Thr Middle Eastern M 9 Yes KFSH&RC 38 c.1498GNA p.Gly500Ser Likely disease associated c.1498GNA p.Gly500Ser Moroccan F 11 Yes 0 CGMJM 39 c.143TNGp.Val48Gly Likely disease associated c.604delG p.Glu202LysfsTer115 UK M Yes Willink 40 c.242CNT p.Pro81Leu Disease associated c.1156CNT p.Arg386Cys Other M Yes Willink 41 c.[566+?_1003-?dup; complex del-dup: duplication from Diseaseassociated NF NF MiddleEastern F 12 KFSH&RC 1139+?_1140-?del] intron 5 to intron 9, deletion in intron 10 42 c.751CNT p.Arg251Ter Disease associated c.268CNT p.Arg90Trp UK F 4 Yes 6 Willink

43 c.1171ANG p.Met391Val Disease associated c.405_422+1 - French Canadian 48 SA Path xxx (2014) xxx Metabolism and Genetics Molecular / al. et Morrone A. 44 c.925GNA p.Gly309Arg Disease associated c.422GNA p.Trp141Ter UK M Willink http://dx.doi.org/10.1016/j.ymgme.2014.03.004 45 c.1171ANG p.Met391Val Likely disease associated c.704CNA p.Thr235Lys French Canadian 8.9 Yes SA Path 46 c.860CNT p.Ser287Leu Likely disease associated c.758+4aNt- Greek M Yes Willink 47 c.740GNA p.Glu247Asp VUS, VUS c.[937ANG; 977GNC] p.[Thr313Ala; Trp326Ser] Australian 1.1 Yes 0.8 SA Path 48 c.752GNA p.Arg251Gln Disease associated c.1155CNA p.Tyr385Ter UK F Yes Willink 49 c.752GNA p.Arg251Gln Disease associated c.1155CNA p.Tyr385Ter UK M Yes Willink 50 c.776GNA p.Arg259Gln Likely disease associated c.1520GNT p.Cys507Phe Hispanic Yes 7 CHOC 51 c.850 TNG p.Phe284Val Disease associated c.1275delA p.Val427SerfsTer13 UK M 5 Yes 4 Willink 52 c.860CNT p.Ser287Leu Disease associated c.1240CNT p.Gln414Ter New Zealander 1.8 Yes 0 SA Path

Novel alleles in bold. Where data were not provided to the lab for the given patient, the field is left blank. The field “Classification of novel allele(s) per Table 1” refers to the novel allele(s), if any; when the novel alleles differ, in this study both happened to share the same classification. CGMJM, Unidade de Bioquímica Genética, Centro de Genética Médica Jacinto Magalhães (CGMJM) do Centro Hospitalar do Porto (CHP), Porto, Portugal; CHOC, Children's Hospital of Orange County, Orange, California, United States; HCPA, Laboratório de Genética Molecular, Serviço de Genética Médica, Hospital de Clínicas de Porto Alegre (HCPA), Porto Alegre, Brazil; KFSH&RC, Department of Medical Genetics, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia; SA Path,SA Pathology, Women's and Children's Hospital, North Adelaide, South Australia, Australia; Willink, Willink Biochemical Genetics Unit, Department of Genetic Medicine, Saint Mary's Hospital, Manchester, United Kingdom. – xxx fi s3 novel 39 es 7 8 A. Morrone et al. / Molecular Genetics and Metabolism xxx (2014) xxx–xxx

Gene Mutation Database, because independent reports that associate a function, testing of the patient alone can still result in molecular under- particular GALNS variant with Morquio A increase confidence in its diagnosis since cases of UPD will not be detected; a case of Morquio A association with disease. resulting from UPD has been reported [8]. We recommend caution in using software programs that predict the The ideal molecular testing approach outlined in Fig. 3 aims to consequences of amino acid substitutions on protein structure. Two of address potential errors when genotyping patients with Morquio A. the more well-known software algorithms for predicting the effects of When possible, both parents should be genotyped to confirm biallelic nonsynonymous SNPs are SIFT and PolyPhen-2. SIFT assumes that inheritance of the causative mutations, allowing detection of situations changes in conserved regions of a protein are more deleterious than in where (1) deletions or polymorphisms at PCR primer sites could result other regions so the availability of suitable homologs is particularly in allele dropout and misinterpretation of patient sequencing results, important. SIFT has a reported true-positive prediction rate of 69% (2) multiple alterations may be present in cis, or (3) homozygosity re- using the Swiss-Prot database, with a false-positive rate of 20% [39]. sults from UPD (Fig. 3). Inaccurate and incomplete molecular diagnoses PolyPhen-2 makes a prediction based on 11 assessments (eight based resulting from the absence of parental genotyping are potential con- on sequence and three on structure), with reported true-positive cerns for many autosomal recessive disorders. In patients with MPS VI, predictions rates of 92% and 73% on HumDiv and HumVar datasets, re- multiple examples of patients with two deleterious mutations in cis spectively, and a false-positive rate of 20% [1]. As with all statistical have been reported [23]. At one molecular testing center, parental probabilities, the predictions are more likely to be true for the entirety genotyping of patients with 40 different autosomal recessive disorders of a dataset than for any specific mutation, warranting further caution revealed that of 75 apparently homozygous patients, four were incor- in their use for the interpretation of individual novel mutations in a rectly assessed as homozygotes due to allele dropout and two were gen- diagnostic setting. uine homozygotes but resulted from UPD [27]. Additionally, parental Molecular testing approaches have limitations in mutation detection genotyping can aid in differentiating alleles that are genuinely associat- in patients with Morquio A. In contrast to MPS I—for which it is common ed with disease from benign polymorphisms, as recently shown in MPS to identify both causative mutations in IDUA (up to 95% of the time in VI [68]. However, we recognize that circumstances do not always allow one study of 85 MPS I families [3])—in Morquio A, a second mutation for parental testing to be performed. In this study, relatively few is not identified in up to 15% of patients [60]. The lower mutation detec- patients had parental testing performed and we recognize that this is tion rate in patients with Morquio A compared with those with MPS I the challenging reality we face when molecularly testing patients: it is may be due to a greater frequency of large deletions, duplications, or not always feasible to test other family members or perform additional other rearrangements in GALNS than in IDUA, where large deletions analyses. have yet to be reported. In addition to limitations in mutation detection, there are limitations 4.3. Challenges in genotype/phenotype correlations in what can be concluded from molecular testing of the patient alone. For example, an inaccurate finding of homozygosity can occur when It is often desirable to make specific genotype/phenotype correla- one patient allele lacks a PCR primer binding site due to deletions, tions between specific alleles and disease severity in Morquio A pa- point mutations, or SNPs, causing that exon to “drop out” from PCR am- tients, but we believe that this practice contains two potential sources plification; since only one patient exon will produce a PCR product and of error: first, the arbitrary classification of a complex disease state yield sequence data, any mutations present in that exon's PCR product based on a few metrics (or even a single metric) and second, the consid- will falsely appear to be homozygous. Apparent homozygosity due to eration of a single mutation divorced from its context in the patient's ge- allele dropout has been reported in patients with MPS VI [61],cysticfi- notype. Morquio A manifests along a broad spectrum of clinical features brosis [11,16], and familial hypercholesterolemia [26]. Even in patients and severity due to both the pleiotropic nature of the disease and the who are genuinely homozygous for a mutation that affects GALNS varied molecular consequences of its many distinct mutations; this

Initial sequencing of Parental Sequencing and Patient Genotype Patients Alone Deletion/Duplication Testing

A Heterozygous for Possible Parental Genotypes: ? * * & * ? detected mutation or WT

( ) Point mutation and

Patient * Heterozygous for Possible Parental Genotypes: & * ( ) large deletion detected mutation

( ) Point mutation

B Possible Parental Genotypes: & * ( * ) and large deletion or * Possible Parental Genotypes: * & * * Homozygous Patient * * Single mutation detected: Uniparental Possible Parental Genotypes: & * appears homozygous * isodisomy * C Possible Parental Genotypes: Mutations in trans * * & * * * or WT WT * ? Mutations in cis Patient Possible Parental Genotypes: & * **? Appears to be * compound heterozygote *mutation? unidentified mutation

Fig. 3. Potential complexity of patient genotypes. For three examples (apparently heterozygous, apparently homozygous, and apparently a compound heterozygote), initial genotyping results illustrate how parental genotyping results can conﬁrm the initial genotype or reveal unanticipated complexity.

Please cite this article as: A. Morrone, et al., Molecular testing of 163 patients with Morquio A (Mucopolysaccharidosis IVA) identifies 39 novel GALNS mutations, Mol. Genet. Metab. (2014), http://dx.doi.org/10.1016/j.ymgme.2014.03.004 A. Morrone et al. / Molecular Genetics and Metabolism xxx (2014) xxx–xxx 9 can create difficulties in succinctly and accurately describing the disease activity. Also, the lack of detection of the second mutation means that state. The metrics most commonly used to classify Morquio A severity conclusive carrier testing is not available for all family members. All of are patient height and growth rates. However, a patient of a given these are reminders that while molecular testing provides diagnostic height can present with other symptoms (e.g., pulmonary or cardiac and genetic counseling information, enzyme activity testing of GALNS, manifestations) of greater or lesser severity than might be expected; along with other enzymes, remains the standard for diagnosis of consequently, the level of dysmorphism does not fully define the dis- Morquio A. ease severity [17,33,34,57]. A single, multivariate metric of disease severity does not currently exist, but weight, height, relative growth Conflicts of interest statement rate, and age at diagnosis are key markers [18]. Concentrations of KS measured in blood and urine, particularly in younger than pre- Dr Fietz has received travel support and honoraria from BioMarin adolescent patients, have been found to correlate with Morquio A Pharmaceutical. Drs Church, Morrone, and Tylee have received consul- clinical status, suggesting their use in assessing disease severity and tant fees and limited travel support from BioMarin Pharmaceutical prognosis [14,17,17,29,56,59,60], and additional Morquio A biomarkers and provide a diagnostic service for MPS for samples from Turkey that have also been proposed [29]. is funded by BioMarin. Dr Pollard is a paid employee of the Greenwood Furthermore, we believe that the possibility exists for unexpected Genetic Center, which has contracts with BioMarin. Dr Wang holds a fi- interactions between alleles, particularly between missense alleles. nancial interest in BioMarin. Dr Mooney serves as a consultant for Most mutations in GALNS are missense mutations [60], with conse- BioMarin Pharmaceutical. Mrs Davidson and Dr Miller are employees quences that can potentially include reduction in GALNS enzyme activ- of BioMarin. Drs Al-Sayed, Brusius-Facchin, Caciotti, Coll, Gort, Kubaski, ity, disruption of protein–protein interactions with other lysosomal Lacerda, Laranjeira, Leistner-Segal, Pajares, and Ribeiro declare no hydrolases [45], or GALNS protein destabilization and degradation. potential conflicts of interest. Because the GALNS enzyme is found as a homodimer [45,58],aMorquio A patient expressing two missense mutations could have three distinct Acknowledgments GALNS dimers (aa, ab, and bb), each potentially with distinct enzymatic activity levels and protein interactions. Moreover, there are examples of We thank Drs E. Pinto, H. Ribeiro, E. Silva, S. Rocha, C. Caseiro, C. pseudodeficiency alleles from other lysosomal storage disorders; these Ferreira, and F. Pinto (all of the Unidade de Bioquímica Genética, Centro are ordinarily benign partial loss-of-function mutations that can poten- de Genética Médica Jacinto Magalhães [CGMJM] do Centro Hospitalar do tially cause disease when combined with other loss-of-function muta- Porto [CHP], Porto, Portugal) and Dr S. Alves (Unidade de Investigação, tions [6,15,51,70]. A complex example from the lysosomal storage INSA, Porto, Portugal) for collaboration in molecular characterization, en- disorder GM1 gangliosidosis describes an ordinarily benign polymor- zymatic assays, GAG analysis, and cell culture; Dr A. Gaspar (Consulta de phism that increases the strength of a deleterious partial loss-of- Doenças Metabólicas, Hospital S. Maria, Centro Hospitalar de Lisboa function allele when combined in cis [5]. While no examples of Norte, Lisboa, Portugal), Dr L. Diogo (Consulta de Doenças Metabólicas, pseudodeficiency alleles or complex allelic interactions are currently Centro Hospitalar Universitário de Coimbra, Coimbra, Portugal), Dr S. described in Morquio A, we believe that the possibility warrants caution Sequeira (Consulta de Doenças Metabólicas, Hospital D. Estefânia, Centro and the analysis of mutations in the context of complete and accurate Hospitalar de Lisboa Central, Lisboa, Portugal), and Dr I. Quelhas (Consulta patient genotypes. de Desenvolvimento, Hospital Padre Américo, Centro Hospitalar Tâmega e Sousa, Penafiel, Portugal); Dr Roberto Giugliani, coordinator of the 4.4. Implications for molecular testing in clinical practice Instituto Nacional de Genética Médica Populacional of Brazil (INAGEMP) and the MPS Brazil Network; Drs J. Jarque, H. Sellés, and A.M. Valle (all From a clinical and counseling perspective, the precise definition of of the Sección de Errores Congénitos del Metabolismo-IBC, Servicio de both causative mutations for Morquio A is particularly important. In ad- Bioquímica y Genética Molecular, Hospital Clínic, CIBERER, IDIBAPS, dition to assisting in the prediction of phenotype, the detection of both Barcelona, Spain); and Dr Emanuela Izzo and Dr Matthew Mealiffe of mutations is required for molecular prenatal diagnosis as well as carrier BioMarin Pharmaceutical Inc, Novato, California, United States. Medical testing for family members. Therefore, in patients heterozygous for two writing support was provided by Dr Karl Zawadzki and Dr Sue Currie of known or probable mutations, it is very important to obtain parental Health Interactions, with funding provided by BioMarin Pharmaceutical DNA samples to confirm biallelic inheritance of these mutations. For pa- Inc. tients appearing homozygous for a mutation, parental samples are also necessary to show that both parents do indeed carry the mutation, Appendix A. Supplementary data confirming heterozygosity, or to indicate cases where further deletion/ duplication or UPD studies are required. Unfortunately, the precise def- Supplementary data to this article can be found online at http://dx. inition of large deletions or duplications using standard Sanger sequenc- doi.org/10.1016/j.ymgme.2014.03.004. ing can often prove particularly difficult, especially when the deletion spans either end of the GALNS gene. Therefore, these studies are rarely References performed as part of the initial sequence studies and in many cases will not be completed once a diagnosis has been “confirmed” by enzyme [1] I.A. Adzhubei, S. Schmidt, L. Peshkin, V.E. Ramensky, A. Gerasimova, P. Bork, A.S. Kondrashov, S.R. Sunyaev, A method and server for predicting damaging missense studies and the detection of a single mutation. mutations, Nat. Methods 7 (2010) 248–249. Furthermore, there is a signi ficant proportion of cases (up to 15%) in [2] R. Aslam, A.C. van Bommel, C.J. Hendriksz, A. Jester, Subjective and objective assess- which the second causative mutation is not detected by standard Sanger ment of hand function in Mucopolysaccharidosis IVa patients, JIMD Rep. 9 (2013) – sequencing [60]. Many of these may be due to heterozygous deletions/ 59 65. [3] C.E. Beesley, C.A. Meaney, G. Greenland, V. Adams, A. Vellodi, E.P. Young, B.G. duplications that are masked by a normal sequence in the other allele. Winchester, Mutational analysis of 85 Mucopolysaccharidosis type I families: fre- Therefore, more extensive mutation studies are required in an attempt quency of known mutations, identification of 17 novel mutations and in vitro – to define the second mutation. As for cases where deletions cause ap- expression of missense mutations, Hum. Genet. 109 (2001) 503 511. [4] S. Bunge, W.J. Kleijer, A. Tylki-Szymanska, C. Steglich, M. Beck, S. Tomatsu, S. Fukuda, parent homozygosity, further analysis is labor intensive and may rarely B.J. Poorthuis, B. Czartoryska, T. Orii, A. Gal, Identification of 31 novel mutations in be carried out. In these cases, prenatal diagnosis using mutation studies the N-acetylgalactosamine-6-sulfatase gene reveals excessive allelic heterogeneity is not feasible and must rely on enzyme studies, which are currently among patients with Morquio A syndrome, Hum. Mutat. 10 (1997) 223–232. [5] A. Caciotti, T. Bardelli, J. Cunningham, A. D'Azzo, E. Zammarchi, A. Morrone, available in only a limited number of centers and can potentially Modulating action of the new polymorphism L436F detected in the GLB1 gene of prove problematic in cases with considerable residual GALNS enzyme a type-II GM1 gangliosidosis patient, Hum. Genet. 113 (2003) 44–50.

[6] A. Caciotti, M.A. Donati, A. Boneh, A. d'Azzo, A. Federico, R. Parini, D. Antuzzi, T. Clawson, G.P. Barber, D. Haussler, W.J. Kent, The UCSC Genome Browser database: Bardelli, D. Nosi, V. Kimonis, E. Zammarchi, A. Morrone, Role of beta-galactosidase extensions and updates 2013, Nucleic Acids Res. 41 (2013) D64–D69. and elastin binding protein in lysosomal and nonlysosomal complexes of patients [32] A.M.Montaño,K.Sukegawa,Z.Kato,R.Carrozzo,P.DiNatale,E.Christensen,K.O.Orii, with GM1-gangliosidosis, Hum. Mutat. 25 (2005) 285–292. T.Orii,N.Kondo,S.Tomatsu,Effectof‘attenuated’ mutations in mucopolysaccharidosis [7] M.V. Camelier, M.G. Burin, J. De Mari, T.A. Vieira, G. Marasca, R. Giugliani, IVA on molecular phenotypes of N-acetylgalactosamine-6-sulfate sulfatase, J. Inherit. Practical and reliable enzyme test for the detection of mucopolysaccharidosis IVA Metab. Dis. 30 (2007) 758–767. (Morquio Syndrome type A) in dried blood samples, Clin. Chim. Acta 412 (2011) [33] A.M. Montaño, S. Tomatsu, A. Brusius, M. Smith, T. Orii, Growth charts for patients 1805–1808. affected with Morquio A disease, Am. J. Med. Genet. A 146A (2008) 1286–1295. [8] S. Catarzi, L. Giunti, F. Papadia, O. Gabrielli, R. Guerrini, M.A. Donati, M. Genuardi, A. [34] A.M. Montaño, S. Tomatsu, G.S. Gottesman, M. Smith, T. Orii, International Morquio Morrone, Morquio A syndrome due to maternal uniparental isodisomy of the A Registry: clinical manifestation and natural course of Morquio A disease, J. Inherit. telomeric end of chromosome 16, Mol. Genet. Metab. 105 (2012) 438–442. Metab. Dis. 30 (2007) 165–174. [9] M.P. Cosma, S. Pepe, I. Annunziata, R.F. Newbold, M. Grompe, G. Parenti, A. Ballabio, [35] K. Nakamura, K. Hattori, F. Endo, Newborn screening for lysosomal storage disor- The multiple sulfatase deficiency gene encodes an essential and limiting factor for ders, Am. J. Med. Genet. C Semin. Med. Genet. 157C (2011) 63–71. the activity of sulfatases, Cell 113 (2003) 445–456. [36] Y. Nakashima, S. Tomatsu, T. Hori, S. Fukuda, K. Sukegawa, N. Kondo, Y. Suzuki, N. [10] J.T. den Dunnen, S.E. Antonarakis, Mutation nomenclature extensions and sugges- Shimozawa, T. Orii, Mucopolysaccharidosis IV A: molecular cloning of the human tions to describe complex mutations: a discussion, Hum. Mutat. 15 (2000) 7–12. N-acetylgalactosamine-6-sulfatase gene (GALNS) and analysis of the 5′-flanking [11] A. Diana, R. Tesse, A.M. Polizzi, T. Santostasi, A. Manca, G. Leonetti, M. Seia, L. Porcaro, region, Genomics 20 (1994) 99–104. L. Cavallo, A large deletion causes apparent homozygosity for the D1152H [37] J. Nelson, Incidence of the mucopolysaccharidoses in Northern Ireland, Hum. Genet. mutation in the cystic fibrosis transmembrane regulator (CFTR) gene, Gene 497 101 (1997) 355–358. (2012) 90–92. [38] J. Nelson, J. Crowhurst, B. Carey, L. Greed, Incidence of the mucopolysaccharidoses in [12] T. Dierks, B. Schmidt, L.V. Borissenko, J. Peng, A. Preusser, M. Mariappan, K. von Western Australia, Am. J. Med. Genet. A 123A (2003) 310–313. Figura, Multiple sulfatase deficiency is caused by mutations in the gene encoding [39] P.C. Ng, S. Henikoff, Accounting for human polymorphisms predicted to affect the human C(alpha)-formylglycine generating enzyme, Cell 113 (2003) 435–444. protein function, Genome Res. 12 (2002) 436–446. [13] T. Dierks, B. Schmidt, K. von Figura, Conversion of cysteine to formylglycine: a pro- [40] S. Ogino, M.L. Gulley, J.T. den Dunnen, R.B. Wilson, Association for Molecular tein modification in the endoplasmic reticulum, Proc. Natl. Acad. Sci. U. S. A. 94 Pathology Training and Education Committtee, Standard mutation nomenclature (1997) 11963–11968. in molecular diagnostics: practical and educational challenges, J. Mol. Diagn. 9 [14] V.C. Dung, S. Tomatsu, A.M. Montaño, G. Gottesman, M.B. Bober, W. Mackenzie, M. (2007) 1–6. Maeda, G.A. Mitchell, Y. Suzuki, T. Orii, Mucopolysaccharidosis IVA: correlation [41] T. Oguma, S. Tomatsu, O. Okazaki, Analytical method for determination of disaccha- between genotype, phenotype and keratan sulfate levels, Mol. Genet. Metab. 110 rides derived from keratan sulfates in human serum and plasma by high- (2013) 129–138. performance liquid chromatography/turbo-ionspray ionization tandem mass [15] V. Gieselmann, A.L. Fluharty, T. Tonnesen, K. Von Figura, Mutations in the spectrometry, Biomed. Chromatogr. 21 (2007) 356–362. arylsulfatase A pseudodeficiency allele causing metachromatic leukodystrophy, [42] G. Panin, S. Naia, R. Dall'Amico, L. Chiandetti, F. Zachello, C. Catassi, L. Felici, G.V. Am.J.Hum.Genet.49(1991)407–413. Coppa, Simple spectrophotometric quantification of urinary excretion of glycosami- [16] F.M. Hantash, A. Rebuyon, M. Peng, J.B. Redman, W. Sun, C.M. Strom, Apparent noglycan sulfates, Clin. Chem. 32 (1986) 2073–2076. homozygosity of a novel frame shift mutation in the CFTR gene because of a large [43] R. Pinto, C. Caseiro, M. Lemos, L. Lopes, A. Fontes, H. Ribeiro, E. Pinto, E. Silva, S. Rocha, deletion, J. Mol. Diagn. 11 (2009) 253–256. A. Marcão, I. Ribeiro, L. Lacerda, G. Ribeiro, O. Amaral, M.C. Sá Miranda, Prevalence of [17] P. Harmatz, K.E. Mengel, R. Giugliani, V. Valayannopoulos, S.P. Lin, R. Parini, N. lysosomal storage diseases in Portugal, Eur. J. Hum. Genet. 12 (2003) 87–92. Guffon, B.K. Burton, C.J. Hendriksz, J. Mitchell, A. Martins, S. Jones, N. Guelbert, A. [44] B.J. Poorthuis, R.A. Wevers, W.J. Kleijer, J.E. Groener, J.G. de Jong, S. van Weely, K.E. Vellodi, C. Hollak, P. Slasor, C. Decker, The Morquio A Clinical Assessment Program: Niezen-Koning, O.P. van Diggelen, The frequency of lysosomal storage diseases in baseline results illustrating progressive, multisystemic clinical impairments in The Netherlands, Hum. Genet. 105 (1999) 151–156. Morquio A subjects, Mol. Genet. Metab. 109 (2013) 54–61. [45] A.V. Pshezhetsky, M. Potier, Association of N-acetylgalactosamine-6-sulfate sulfatase [18] C.J. Hendriksz, P. Harmatz, M. Beck, S. Jones, T. Wood, R. Lachman, C.G. Gravance, T. Orii, with the multienzyme lysosomal complex of beta-galactosidase, cathepsin A, and S. Tomatsu, Review of clinical presentation and diagnosis of Mucopolysaccharidosis neuraminidase. Possible implication for intralysosomal catabolism of keratan IVA, Mol. Genet. Metab. 110 (2013) 54–64. sulfate, J. Biol. Chem. 271 (1996) 28359–28365. [19] J.P. Hintze, S. Tomatsu, T. Fujii, A.M. Montaño, S. Yamaguchi, Y. Suzuki, M. Fukushi, T. [46] C.S. Richards, S. Bale, D.B. Bellissimo, S. Das, W.W. Grody, M.R. Hegde, E. Lyon, B.E. Ishimaru, T. Orii, Comparison of liquid chromatography-tandem mass spectrometry Ward, Molecular Subcommittee of the ACMG Laboratory Quality Assurance and sandwich ELISA for determination of keratan sulfate in plasma and urine, Committee, ACMG recommendations for standards for interpretation and reporting Biomark. Insights 6 (2011) 69–78. of sequence variations: Revisions 2007, Genet. Med. 10 (2008) 294–300. [20] J.J. Hopwood, H. Elliott, Detection of Morquio A syndrome using radiolabelled sub- [47] Y. Rivera-Colón, E.K. Schutsky, A.Z. Kita, S.C. Garman, The structure of human GALNS strates derived from keratan sulphate for the estimation of galactose 6-sulphate reveals the molecular basis for Mucopolysaccharidosis IV A, J. Mol. Biol. 423 (2012) sulphatase, Clin. Sci. (Lond.) 65 (1983) 325–331. 736–751. [21] J.J. Hopwood, J.R. Harrison, High-resolution electrophoresis of urinary glycosamino- [48] L. Romdhane, R. Kefi, H. Azaiez, N. Ben Halim, K. Dellagi, S. Abdelhak, Founder glycans: an improved screening test for the mucopolysaccharidoses, Anal. Biochem. mutations in Tunisia: implications for diagnosis in North Africa and Middle East, 119 (1982) 120–127. Orphanet J. Rare Dis. 7 (2012) (52-1172-7-52). [22] R. Humbel, C. Marchal, M. Fall, Diagnosis of Morquio's disease: a simple chromato- [49] C.R. Scott, S. Elliott, N. Buroker, L.I. Thomas, J. Keutzer, M. Glass, M.H. Gelb, F. graphic method for the identification of keratosulfate in urine, J. Pediatr. 81 Turecek, Identification of infants at risk for developing Fabry, Pompe, or (1972) 107–108. mucopolysaccharidosis-I from newborn blood spots by tandem mass spectrometry, [23] L. Karageorgos, D.A. Brooks, A. Pollard, E.L.Melville,L.K.Hein,P.R.Clements,D. J. Pediatr. 163 (2013) 498–503. Ketteridge,S.J.Swiedler,M.Beck,R.Giugliani,P.Harmatz,J.E.Wraith,N. [50] G.A. Solanki, K.W. Martin, M.C. Theroux, C. Lampe, K.K. White, R. Shediac, C.G. Guffon,E.LeaoTeles,M.C.SaMiranda,J.J.Hopwood,Mutationalanalysisof Lampe, M. Beck, W.G. Mackenzie, C.J. Hendriksz, P.R. Harmatz, Spinal involvement 105 Mucopolysaccharidosis type VI patients, Hum. Mutat. 28 (2007) 897–903. in mucopolysaccharidosis IVA (Morquio-Brailsford or Morquio A syndrome): [24] Z. Kato, S. Fukuda, S. Tomatsu, H. Vega, T. Yasunaga, A. Yamagishi, N. Yamada, A. presentation, diagnosis and management, J. Inherit. Metab. Dis. 36 (2013) 339– 355. Valencia, L.A. Barrera, K. Sukegawa, T. Orii, N. Kondo, A novel common missense [51] G.H. Thomas, “Pseudodeficiencies” of lysosomal hydrolases, Am. J. Hum. Genet. 54 mutation G301C in the N-acetylgalactosamine-6-sulfate sulfatase gene in (1994) 934–940. MucopolysaccharidosisIVA,Hum.Genet.101(1997)97–101. [52] S. Tomatsu, T. Dieter, I.V. Schwartz, P. Sarmient, R. Giugliani, L.A. Barrera, N. [25] P. Labrousse, Y.H. Chien, R.J. Pomponio, J. Keutzer, N.C. Lee, V.R. Akmaev, T. Scholl, W. Guelbert, R. Kremer, G.M. Repetto, M.A. Gutierrez, T. Nishioka, O.P. Serrato, A.M. L. Hwu, Genetic heterozygosity and pseudodeficiency in the Pompe disease Montaño, S. Yamaguchi, A. Noguchi, Identification of a common mutation in newborn screening pilot program, Mol. Genet. Metab. 99 (2010) 379–383. mucopolysaccharidosis IVA: correlation among genotype, phenotype, and keratan [26] E. Laios, K. Glynou, Allelic drop-out in the LDLR gene affects mutation detection in sulfate, J. Hum. Genet. 49 (2004) 490–494. familial hypercholesterolemia, Clin. Biochem. 41 (2008) 38–40. [53] S. Tomatsu, T. Fujii, M. Fukushi, T. Oguma, T. Shimada, M. Maeda, K. Kida, Y. Shibata, [27] M.L. Landsverk, G.V. Douglas, S. Tang, V.W. Zhang, G.L. Wang, J. Wang, L.J. Wong, H. Futatsumori, A.M. Montaño, R.W. Mason, S. Yamaguchi, Y. Suzuki, T. Orii, Diagnostic approaches to apparent homozygosity, Genet. Med. 14 (2012) Newborn screening and diagnosis of mucopolysaccharidoses, Mol. Genet. Metab. 877–882. 110 (2013) 42–53. [28] S.Laradi,T.Tukel,S.Khediri,J.Shabbeer,M.Erazo,L.Chkioua,M.Chaabouni, [54] S. Tomatsu, S. Fukuda, A. Cooper, J.E. Wraith, P. Ferreira, P. Di Natale, P. Tortora, A. S. Ferchichi, A. Miled, R.J. Desnick, Mucopolysaccharidosis type IV: N- Fujimoto, Z. Kato, N. Yamada, K. Isogai, A. Yamagishi, K. Sukegawa, Y. Suzuki, N. acetylgalactosamine-6-sulfatase mutations in Tunisian patients, Mol. Genet. Metab. Shimozawa, N. Kondo, W.S. Sly, T. Orii, Fourteen novel mucopolysaccharidosis IVA 87 (2006) 213–218. producing mutations in GALNS gene, Hum. Mutat. 10 (1997) 368–375. [29] L.A. Martell, R.L. Cunico, J. Ohh, W. Fulkerson, R. Furneaux, E.D. Foehr, Validation of [55] S. Tomatsu, S. Fukuda, M. Masue, K. Sukegawa, T. Fukao, A. Yamagishi, T. Hori, H. an LC-MS/MS assay for detecting relevant disaccharides from keratan sulfate as a Iwata, T. Ogawa, Y. Nakashima, Morquio disease: isolation, characterization and ex- biomarker for Morquio A syndrome, Bioanalysis 3 (2011) 1855–1866. pression of full-length cDNA for human N-acetylgalactosamine-6-sulfate sulfatase, [30] P.J. Meikle, J.J. Hopwood, A.E. Clague, W.F. Carey, Prevalence of lysosomal storage Biochem. Biophys. Res. Commun. 181 (1991) 677–683. disorders, JAMA 281 (1999) 249–254. [56] S. Tomatsu, A.M. Montaño, T. Oguma, V.C. Dung, H. Oikawa, T.G. de Carvalho, M.L. [31] L.R. Meyer, A.S. Zweig, A.S. Hinrichs, D. Karolchik, R.M. Kuhn, M. Wong, C.A. Sloan, K. Gutierrez, S. Yamaguchi, Y. Suzuki, M. Fukushi, K. Kida, M. Kubota, L. Barrera, T. R. Rosenbloom, G. Roe, B. Rhead, B.J. Raney, A. Pohl, V.S. Malladi, C.H. Li, B.T. Lee, K. Orii, Validation of keratan sulfate level in mucopolysaccharidosis type IVA by liquid Learned, V. Kirkup, F. Hsu, S. Heitner, R.A. Harte, M. Haeussler, L. Guruvadoo, M. chromatography-tandem mass spectrometry, J. Inherit. Metab. Dis. 33 (Suppl. 3) Goldman, B.M. Giardine, P.A. Fujita, T.R. Dreszer, M. Diekhans, M.S. Cline, H. (2010) S35–S42.

[57] S. Tomatsu, A.M. Montaño, H. Oikawa, M. Smith, L. Barrera, Y. Chinen, M.M. Thacker, [64] C.B. Whitley, M.D. Ridnour, K.A. Draper, C.M. Dutton, J.P. Neglia, Diagnostic test for W.G. Mackenzie, Y. Suzuki, T. Orii, Mucopolysaccharidosis type IVA (Morquio A mucopolysaccharidosis. I. Direct method for quantifying excessive urinary glycos- disease): clinical review and current treatment, Curr. Pharm. Biotechnol. 12 aminoglycan excretion, Clin. Chem. 35 (1989) 374–379. (2011) 931–945. [65] T.C. Wood, K. Harvey, M. Beck, M.G. Burin, Y.H. Chien, H.J. Church, V. D'Almeida, O.P. [58] S. Tomatsu, T. Nishioka, A.M. Montaño, M.A. Gutierrez, O.S. Pena, K.O. Orii, W.S. Sly, van Diggelen, M. Fietz, R. Giugliani, P. Harmatz, S.M. Hawley, W.L. Hwu, D. S. Yamaguchi, T. Orii, E. Paschke, S.G. Kircher, A. Noguchi, Mucopolysaccharidosis Ketteridge, Z. Lukacs, N. Miller, M. Pasquali, A. Schenone, J.N. Thompson, K. Tylee, IVA: identification of mutations and methylation study in GALNS gene, J. Med. C. Yu, C.J. Hendriksz, Diagnosing mucopolysaccharidosis IVA, J. Inherit. Metab. Dis. Genet. 41 (2004) e98. 36 (2013) 293–307. [59] S. Tomatsu, K. Okamura, T. Taketani, K.O. Orii, T. Nishioka, M.A. Gutierrez, S. Velez- [66] N. Yamada, S. Fukuda, S. Tomatsu, V. Muller, J.J. Hopwood, J. Nelson, Z. Kato, A. Castrillon, A.A. Fachel, J.H. Grubb, A. Cooper, M. Thornley, E. Wraith, L.A. Barrera, R. Yamagishi, K. Sukegawa, N. Kondo, T. Orii, Molecular heterogeneity in Giugliani, I.V. Schwartz, G.S. Frenking, M. Beck, S.G. Kircher, E. Paschke, S. mucopolysaccharidosis IVA in Australia and Northern Ireland: nine novel mutations Yamaguchi, K. Ullrich, K. Isogai, Y. Suzuki, T. Orii, N. Kondo, M. Creer, A. Noguchi, including T312S, a common allele that confers a mild phenotype, Hum. Mutat. 11 Development and testing of new screening method for keratan sulfate in (1998) 202–208. mucopolysaccharidosis IVA, Pediatr. Res. 55 (2004) 592–597. [67] J. Ye, H.L. Lei, H.W. Zhang, W.J. Qiu, L.S. Han, Y. Wang, X.Y. Li, X.F. Gu, Analysis of [60] S. Tomatsu, A.M. Montaño, T. Nishioka, M.A. Gutierrez, O.M. Peña, G.G. Tranda GALNS gene mutation in thirty-eight Chinese patients with mucopolysaccharidosis Firescu, P. Lopez, S. Yamaguchi, A. Noguchi, T. Orii, Mutation and polymorphism type IVA, Zhonghua Er Ke Za Zhi 51 (2013) 414–419. spectrum of the GALNS gene in mucopolysaccharidosis IVA (Morquio A), Hum. [68] A. Zanetti, E. Ferraresi, L. Picci, M. Filocamo, R. Parini, C. Rosano, R. Tomanin, M. Mutat. 26 (2005) 500–512. Scarpa, Segregation analysis in a family at risk for the Maroteaux-Lamy syndrome [61] G.R. Villani, M. Grosso, G. Pontarelli, A. Chierchia, R. Sessa, M. Sibilio, G. Parenti, P. Di conclusively reveals c.1151G N A (p.S384N) as to be a polymorphism, Eur. J. Hum. Natale, Large deletion involving exon 5 of the arylsulfatase B gene caused apparent Genet. 17 (2009) 1160–1164. homozygosity in a mucopolysaccharidosis type VI patient, Genet. Test. Mol. [69] H. Zhao, O.P. Van Diggelen, R. Thoomes, J. Huijmans, E. Young, T. Mazurczak, W.J. Biomarkers 14 (2010) 113–120. Kleijer, Prenatal diagnosis of Morquio disease type A using a simple fluorometric [62] Z. Wang, W. Zhang, Y. Wang, Y. Meng, L. Su, H. Shi, S. Huang, Mucopolysaccharidosis enzyme assay, Prenat. Diagn. 10 (1990) 85–91. IVA mutations in Chinese patients: 16 novel mutations, J. Hum. Genet. 55 (2010) [70] J. Zlotogora, G. Bach, Deficiency of lysosomal hydrolases in apparently healthy 534–540. individuals, Am. J. Med. Genet. 14 (1983) 73–80. [63] C.B. Whitley, K.A. Draper, C.M. Dutton, P.A. Brown, S.L. Severson, L.A. France, Diag- nostic test for mucopolysaccharidosis. II. Rapid quantification of glycosaminoglycan in urine samples collected on a paper matrix, Clin. Chem. 35 (1989) 2074–2081.