Haematology and Blood Transfusion Hamatologie und Bluttransfusion 35 Rolf Neth Elena Frolova Robert C. Gallo Melvyn F. Greaves Boris V. Afanasiev and Elena Elstner (Eds.) Modern Trends in Human IX New Results in Clinical and Biological Research Including Pediatric

Organized on behalf of the Deutsche Gesellschaft fUr Hamatologie und Onkologie, Wilsede, June 17-21,1990

With 194 Figures and 47 Tables

Springer-Verlag Berlin Heidelberg New York London Paris Tokyo Hong Kong Barcelona Budapest Dr. Rolf Neth Universitats-Krankenhaus Eppendorf, Zentrum fiir Knochenmark-Transplantation Martinistr. 52, 2000 Hamburg 20, FRO

Dr. Elena Frolova Shemyakin Institute of Bioorganic Chemistry, Miklukho-Maklaya Str. 16/10, 117871 Moscow, Russia

Dr. Robert C. Gallo National Cancer Institute, Laboratory of Tumor Cell Biology, Bethesda, MD 20205, USA

Dr. Melvyn F. Greaves Leukemia Research Fund Centre, Institute of Cancer Research, Chester Beatty Laboratories, Fulham Road, London SW 3 6 JB, UK

Dr. Boris V. Afanasiev N. N. Petrov Institute of Oncology, Department of Bone Marrow Transplantation, St. Petersburg, Russia

Dr. Elena Elstner Medizinische Fakultat der Humboldt-Universitat Berlin, Klinik fUr Innere Medizin, Abteilung Hamatologie, Schumannstr. 20/21, 1040 Berlin, FRO

ISBN-13: 978-3-540-54360-2 e-ISBN-13: 978-3-642-76829-3 DOl: 10.007/978-3-642-76829-3

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer-Verlag. Violations are liable for prosecution under the German Copyright Law. © Springer-Verlag Berlin Heidelberg 1992

The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Product liability: The publishers cannot guarantee the accuracy of any information about dosage and applica• tion contained in this book. In every individual case the user must check such information by consulting the relevant literature. Typesetting: Fotosatz-Service Kiihler, Wiirzburg, FRG 27/3145/-54321 0 - Printed on acid-free paper Volga Wilsede Meeting I Moskau / Volga 13.6.-15.6.1990 "Molecular Factors of Hematopoiesis"

Organisation: Elena Frolova Gregory Dolganov Shemyakin Institute of Bioorganic Chemistry VI. Miklukho-Maklaya 16/10 Moscow

Neva Wilsede Meeting I St. Petersburg 14.6.-17.6.1990 "Clinical Aspects"

Organisation: Boris Afanasiev Peter Knyazev Bone Marrow Transplantation Institute of Oncology Leningradskay str. 68 St. Petersburg

Wilsede Joint Meeting Hamburg 16.6.1990 "Signal Transfer Pathway"

Organisation: Wolfram Ostertag Heinrich Pette Institut fUr Experimentelle Virologie und Immunologie an der Vniversihit Hamburg Martinistr. 52 2000 Hamburg 20

Local Organisation: Klaus Mannweiler Heinrich Pette Institut fUr Experimentelle Virologie und Immunologie an der Vniversitiit Hamburg Martinistr. 52 2000 Hamburg 20

Rolf Neth Zentrum fiir Knochenmark-Transplantation V niversi tii ts-Krankenha us E ppendorf Martinistr. 52 2000 Hamburg 20

Johannes Schubert St. Joseph Hospital Wiener Str. 1 2850 Bremerhaven v Ladies and Gentlemen, Dear Friends

In 1988, after the last Wilsede meeting, Den v6lligen Gegensatz zum Antlitz Elena Frolova and her friends had the bildet das Wort Maske. marvelous idea of organizing a Volga Wilsede meeting. Today, I am very Pavel Florenski 1882-1937 in: pleased that this idea became true. "Die Ikonostase"

Every day, we dream of freedom and peace. A small part of this dream has today come into our hands.

Everybody here has done his best to come together for good science and in friendship.

I am sure that this Volga Wilsede meeting and the Neva Wilsede Meeting will be• come part of the international tradition of good science in a peaceful human atmosphere.

For us this door IS open: let us go through.

Moscow, 13 June 1990 Rolf Neth

VII

Dr. Rolf Neth asked me to say a few The upper picture which looks like a words about my last work, which I called Super Nova is a cell ofamouse fibroplast "Alexander Maximov's world". I could which Alexander Maximov in his time think that many of you had to learn or to could not see like this, as the electronic study his experiments and results, when microscope was'nt invented then. This is they were students. Well, I am not com• the contemporary level of this work to• petent enough to give a valuation of his gether with the circled Wilsede motif work or meaning in histology. But the over the tower. few things I read about his work and life If you look closer and I could help gave me a great deal of respect and everybody with my magnifying glass, you admiration. Specially how he drew and discover that Maximov's two blood pic• painted the results of his experiments in tures are drawn in the silhouette of the such a filligran and precise way. Emhof (up side down) and the Granary. Amongst scientists, often the work of a The Center shows the portrait of colleague is not always considered with Alexander Maximov about the time as a great admiration rather more with a feel• young man in St. Petersburg and it shows. ing of respect or competition. So, when I close to his heart the entrance to his labo• put up some kind of a monument for ratory in the Military Academy. Alexander Maximov, it is the painters To end now, I sum up the key idea may admiration for a scientist, who probably be called 5 seconds of inspiration and five drew much better than I ever would do in weeks of rather hard work. Alexander histology. Maximov as a tower surrounded by the Now to my work: There is the real and real buildings St. Petersburg, Berlin and spiritual world of Alexander Maximov. Chicago and on the other hand sur• The polarisation of the idea is seen on the rounded by his spiritual buildings, the right side which shows the three stations blood pictures in the silhouette of in his life. Wilsedes granary and the Emhof. So I thank you for accompanying First: St. Petersburg, where he worked my explanations of Alexander Maximov at the Military Medical Acad• from st. Petersburg to us here in emy for some years. Wilsede. Second: In Berlin, the Charite Hospital, where he worked like his fa• Michel Weidemann mous colleagues Arinkin and Botkin before. Maximov's world Third: And last station of his life was Etching 40 x 60 by Michel Weidemann the University of Chicago where he had to emigrate with heavy heart after october revolution. On the left side you see two blood pictures which have been drawn by Alexander Maximov.

IX Scientific and private discussions on the Volga and around Uglich

Xl XII Scientific and private discussions in and around "The Emhor'

Photographs: R. Valz (Hamburg)

XIII XIV Contents

K. Munk: In Memoriam Clinical Aspects Dr. Mildred Scheel ...... XXI D. Pinkel: A. Ullrich and J. Schlessinger: Bone Marrow Transplantation Growth Factor Receptors: in Children with Acute Leukemia: Role in Normal Mitogenic Signalling A 1990 View...... 3 and Oncogenesis ...... XXIII M. Eder, O. G. Ottmann, T. E. Hansen-Hagge, C. R. Bartram, V. T. Ivanov: In Memoriam S. Gillis, D. Hoelzer, and A. Ganser: Prof. Yuri Ovchinnikov ...... XXXIII Proliferation-Including Effects V. T. Ivanov and V. A. Nesmeyanov: of Recombinant Human Interleukin-7 Yuri Ovchinnikov ...... XXXV and Interleukin-3 in B-Lineage Acute Lymphoblastic Leukemia...... 11 J. D. Rowley: Ovchinnikov Lecture K. Welte, C. Zeidler, A. Reiter, J. Rosler, Human Leukemia Genes: Search for the T. Pietsch, T. Hecht, and H. Riehm: Villains ...... XL Effects of Granulocyte Colony• R. C. Gallo and L. S. Nerurkar: Stimulating Factor in Children with Ovchinnikov Lecture Severe Congenital Neutropenia . . . .. 16 Human Retroviruses: Linkage N. J. Simbirtseva: to Leukemia and AIDS. . XLVII Interleukin-1 Production in Patients with Nonlymphocytic Leukemia Pastor Bode Lecture ...... LXX and Myelodysplastic Syndromes . . .. 22 N. A. Mitchison: Pastor Bode Lecture: J. Boguslawska-Jaworska, J. Pisarek, Escape from the Red Queen . . . . . LXXIII A. Chybicka, B. Kazanowska, J. Armata, Frederick Stohlman Jr. Memorial W. Balwierz, H. Bubala, T. Jackowska, D. Michalewska, M. Ochocka, Lecture ...... LXXXI U. Radwanska, R. Rokicka Milewska, S. Broder: Clinical Research Using M. Rytlewska, S. Skomra, G. Sladkowska, 3' -Azido-2',3' -Dideoxythymidine (AZT) J. Wachowiak, M. Wieczorek, E. Zelenay, and Related Dideoxynucleosides D. SOilta-Jakimczuk, and A. Wozniak: in the Therapy of AIDS ..... LXXXIII The Results of Treatment of Acute Lymphoblastic Leukemia Relapses F. Grosveld, M. Antoniou, M. Berry, in the Polish Children's E. de Boer, N. Dillon, J. Ellis, P. Fraser, Leukemia/ Study Group 28 D. Greaves, O. Hanscombe, J. Hurst, A. Petrakova and J. Stary: M. Lindenbaum, V. Mignotte, Long-Term Follow-up Study S. Philipsen, S. Pruzina, J. Strouboulis, in Childhood Acute Lymphoblastic D. Talbot, and D. Whyatt: The Human ...... 32 f3-Globin Locus Control Region. . . .. CV T. Miyashita, S. Mizutani, M. Asada, List of Participants ...... CXIX J. Fujimoto, T. Inaba, and T. Furukawa: Monoclonal Blast Cell Proliferation Wilsede Scholarship Holders ... CXXVII in Transient Myeloproliferative Acknowledgements...... CXXIX Disorder ...... 33 xv K. Palbczi, K. Natonek, E. Pbcsik, M. Waechter, E. Elstner, J. Maciejewski, B. Kottan, R. Mihalik, M. Benczur, H.-D. Volk, and R. lhle: J. Demter, and G. G. Petrlmyi: In Vitro Effect of GM-CSF and IFN-y The Comparison of the Expression on the Establishment of Stromal Layer of Activation Antigens on Peripheral and Hemopoiesis in Human Dexter Blood Mononuclear Cells Culture ...... 97 in Chronic Lymphocytic Leukemia A. Y. Zaritskey and o. V. Strizhak: and in Hairy Cell Leukemia...... 38 The Effect of Bone Marrow Fibroblast V. A. Almazov, E.l. Podoltzeva, and Stromal Cell-Conditioned Media on Hemopoietic Cells in Culture. . .. 101 E. V. Morozova, X. O. Sitskaya, V. M. Kravzova, and B. V. Afanasiev: N. L.Samoilina, A. L. Medvinsky, Multiple Myeloma: Immunodeficient, 0.1. Gan, and T. E. Manakova: Osteolytic, Renal, and Amyloid Hemopoietic Stem Cells Syndromes...... 41 in Embryogenesis of the Mouse 106

B. V. Afanasiev, E.l. Podoltzeva, Po-min Chen, Huo-l Hsiao, E. V. Morozova, V. A. Almazov, Jenn-long Su, J. Liu, and Ling-ling Yang: L. S. Zubarovskaya, and V. M.Kravzova: Induction of Differentiation Prognostic Factors in Patients of the Human Promyelocytic Cell Line (HL-60) by Conditioned Medium with Multiple Myeloma ...... 50 of Ceathea letifera-Stimulated A. G. Zedginidze: Mononuclear Cells...... 110 Cytogenetic Disorders During Tumor L. R. Rohrschneider: Molecular Progression in Leukemia ...... 56 Mechanisms in Myeloid Lineage Development ...... 115 N. N. Mamaev, V. A. Pavlova, and o. V. Marinets: New Cytogenetic M. Ranson and J. T. Gallagher: Aspects of Myelodysplasia and Acute Proteoglycans in Cellular Recognition Leukemia ...... 59 and .secretory Functions in the Haemopoietic System...... 121 T. E. Hansen-Hagge, S. Yokota, S. H. Gohla, R. A. Zeman, M. Bogel, J. W. G. Janssen, and C. R. Bartram: E. Jurkiewicz, S. Schrum, H.D. Haubeck, Detection of Minimal Residual H.Schmitz, G.Hunsmann, andR.D.Neth: Leukemia by Polymerase Chain Modification of the In Vitro Replication Reactions ...... 65 of the Human Immunodeficiency Virus HIV-1 by TPSg, a Polysaccharide J. Monaselidze, Z. Chanchalashvili, Fraction Isolated from the Cupressaceae I. Kalandadze, D. Khachidze, Thuja occidentalis L. (Arborvitae) ... 140 and I. Topuridze: Structural Changes in Blast Cell Chromatin and DNA in Patients with Acute Leukemia After Molecular Genetics Cytosar and Irradiation Treatment .. 72 M. F. Greaves, J. Brown, L. Fina, D. Robertson, D. Delia, Cell Biology and H. V. M olgaard: Molecular Cloning and Expression of CD 34: J. E. Dick: Animal Models of Normal A Haemopoietic Progenitor-Associated and Leukemic Human Hematopoiesis. 77 Cell Surface Glycoprotein ...... 153 N. M. Gough: Molecular Genetics M. Akashi and H. P. Koeffler: of the Human GM-CSF Receptor . .. 159 Colony Stimulating Factors: E.l. Frolova, D. V. Smirnov, Regulation of Production ...... 83 G. M. Dolganov, 1. A. Mazo, N. S. Bystrov, and 1. V. Severtsova: M. Y. Gordon, D. Clarke, C. R. Dowding, Structural Organization of the Cytokine and M. Siczkowski: Adhesive Interactions Gene Cluster on Human in the Regulation of Haemopoiesis 93 Chromosome 5 ...... 166

XVI S. A. Nedospasov, A. N. Shakhov, A. G. Tatosyan, M. S. Shtutman, D. V. Kuprash, I. A. Udalova, L. Z. Topol, N. P. Kisseljova, M. M. Azizov, T. M. Seregina, E. A. Musatkina, and G. I. Deichman: M.I. Mekshenkov, and R.L. Turetskaya: Selective Integration of Rous Sarcoma Genes Encoding Tumor Necrosis Virus Genome in High- and Low- Factors: Genome Organization, Metastatic Transformed Cell Lines 260 Polymorphism, and Expression . . 175 P. G. Knyazev, O. M. Serova, A. Miyajima: Molecular Characterization I. F. Nikiforova, V. I. Babenko, of the Mouse Interleukin-3 Receptor 185 A. A. Goltzov, G. F. Pluzhnikova, o. V. Plutalov, U. A. Berlin, J. H. Signal Transduction Pierce: and E. I. Scwartz: Molecular Mechanism Through Foreign Growth Factor of Alteration of H-rasI Oncogene Receptors and Oncogenes Transfected in Human Breast Carcinomas: G to T into Interleukin-3-Dependent Transversion in 12th Codon Hematopoietic Cells ...... 191 of the Undeleted Allele in the Case M. Dean, M. B. White, B. Gerrard, of the Loss of the Other Gene Allele 262 L. Krueger, V. Baranov, N. Kapronov, M.lanuzzui, G. Sebastio, M. Leppert, and J. Amos: Genetic Analysis of Ion Immunology Transport...... 194 H.-G. Ljunggren, P. Hoglund, C. Ohlen, and K. Kiirre: Natural Killer Cell Virology Mediated Defence Directed Selectively Against Target Cells Lacking MHC F. Wendling, M. Souyri, I. Vigon, Class I Gene Products ...... 267 S. Gisselbrecht, and P. Tambourin: The Molecular Biology of the B. Van den Eynde, B. Lethe, A. Van Pel, Myeloproliferative Leukemia Virus 201 and T. Boon: Tumor Rejection Antigens and Immune Surveillance ...... 279 W. Kolch, R. H. Bassin, and U. R. Rapp: Raf Function is Required P. J. Robinson: for Proliferation of NIHj3T3 Cells Glycophosphatidylinositol-Anchored and Transformation by Nonnuclear Membrane Proteins as Co receptors Oncogenes ...... 208 in T-Cell Activation ...... 286 C. Zechel, H. Peters, U. Schleenbecker, V. A. Nesmeyanov, S. V. Khaidukov, A. Anders, and F. Anders: erbB*": R. L. Komal'eva, M. V. Sumaroka, an "Ignition Spark" for the Xiphorus I. S. Litvinov, E. G. Dorodnikh, Melanoma Machinery? ...... 213 T. I. Valyakina, A. Malakhov, and K. Brisken: The Molecular U. Lorenz, P. Beimling, J. Cao, Mechanism of Muramyl Peptides' and K. Moelling: Analysis of Human Biological Activity ...... 290 Retroviral Regulatory Proteins Tax and Tat...... 235 S. M. Kang, M. Grilli, A. C. Tran, and M. J. Lenardo: A Novel Nuclear M. Schwab: Human Neuroblastoma: Factor Binds to the NF-KB Motif Paradigm for a Tumor with Oncogene in the Interleukin-2 Enhancer . . . .. 294 Amplification and Loss of a Putative Tumor Suppressor Gene ...... 241 D. Cosman: Soluble Cytokine Receptors as Immunomodulators ...... 302 M. E. M. Campbell and S. McCorkindale: Activation of Gene Expression by Human Herpes Virus 6 ...... 251 Early Life and Evolution C. Leib-Mosch, M. Bachmann, E.-M. Geigl, R. Brack-Werner, T. Werner, S. Ohno: Immunological Self-Nonself V. Erfie, and R. Hehlmann: Discrimination and Numerous Peptide Expression of S 71-Related Sequences Fragments Shared by Unrelated in Human Cells ...... 256 Proteins ...... 311

XVII S. N. Rodin, A. Y. Rzhetsky, J. Schwietering and P. J. Plath: and A. A. Zharkikh: Frozen Temporal Pattern in Growing Multigene Families: The Problem Systems ...... 329 of Molecular Recapitulation ...... 316 U. Krause: On Some Concepts S. N. Rodin, Y. G. Matushkin, in Modeling Nonlinear Phenomena ... 336 and J. S. Krushkal: Repeated Intragenome "Parasites" as a Factor in Molecular Coevolution . 323 Subject Index ...... 343

XVIII Dr. Mildred Scheel Memorial Lecture Hamburg, June 18, 1988

Munk, Klaus In Memoriam Dr. Mildred Scheel

Anders, Fritz A Biologist's View of Human Being Hamburg, June 16, 1990

Ullrich, Axel Growth Factor Receptors: Role in Normal Mitogenic Signalling and Oncogenesis

XIX In Memoriam Dr. Mildred Scheel

Klaus Munk

The Wilsede Meeting is also supported by her efforts were directed towards en• the Wilsede Fellowship Programme of couraging people to contribute money for the Dr. Mildred Scheel Stiftung, which is this crucial purpose. She developed many part' of the Deutsche Krebshilfe. Since significant ideas for organizing cancer that Foundation was established by Dr. prevention, early diagnosis and treatment Mildred Scheel, it is appropriate that we that was applicable on a large scale. She should reflect and comment on the great initiated the establishment of the first five contribution which she made to cancer cancer centers in this country. Once they prevention, treatment and research. were functioning successfully, she was Who was Mildred Scheel? What were able to convince the Government to as• her ideas and what did she achieve with sume full responsibility for maintaining her Foundation? them. She then prepared to launch new Mildred Scheel was born in Cologne in undertakings. It became apparent to 1932, daughter of a physician and radio• people that she had unique qualities that logist. She studied medicine and spe• enabled her to initiate new ideas for cialized in radiology. Later, she married fighting cancer, and this added signifi• Mr. Walter Scheel before he was appoin• cantly to her personal success. She also ted Minister for Foreign Affairs. When supported in particular the treatment of Mr. Scheel subsequently became Presi• childhood cancer in many hospitals, and dent of the Federal Republic of Ger• initiated the psychosocial after-care of many, she became the "First Lady" of patients and their families. In addition, this country. No doubt this helped her to she aided individuals who were economi• . fulfil her noble ambition to contribute to cally affected by having cancer. the fight against cancer. As a conse• The Dr. Mildred Scheel Stiftung was quence of this, she founded the Deutsche established to promote and support can• Krebshilfe in 1974. From that time on, all cer research. It supports a great number

Haematology and Blood Transfusion VoL 35 XXI Modern Trends in Human Leukemia IX R. Neth et aL (Eds.) © Springer-Verlag Berlin Heidelberg 1992 of research projects in many institutes remember her ever missing a meeting of and provides a fellowship programme for the board or the scientific councils of the scientists to work and study at insti• Deutsche Krebshilfe or the Dr. Mildred tutions abroad. Included in that pro• Scheel Stiftung, until the last few weeks of gramme is the Wilsede Fellowship Pro• her life. During those meetings she lis• gramme. The Dr. Mildred Scheel Stiftung tened carefully to the experts, although is now an important body in the Federal sometimes she came to her own conclu• Republic of Germany for the granting of sions when she was convinced that a fellowships. Many of Mildred Scheel's particular step forward had to be made. initiatives were not broadly accepted at She never lost her enthusiasm for helping first, but through her continued energy others, even when she realized what they are now accepted as common prac• would be the consequence of her own tices in the oncological field in this illness. She always seemed to be positive country. in her attitude and could always stimulate When she had a particular goal in sight, others with her spirit and her personality. no obstacles could prevent her from re• She could have done so much more in the aching it. Yet, for all her tenacity, future and she is sadly missed by all of us. Mildred Scheel was a warm, loving and We all will always remember her with sensitive person who had special under• great devotion. standing for cancer patients, together The Mildred Scheel Memorial Lectures with a human touch. She was always very are our tribute. The second lecture will be hard-working and enthusiastic, and held by Axel Ullrich, a classic molecular stimulating for all of us. None of those biologist, who has made important con• who, like myself, had worked with her in tributions to the understanding of cancer the Foundation for over 10 years can in the field of molecular biology.

XXII Growth Factor Receptors: Role in Normal Mitogenic Signalling and Oncogenesis

A. Ullrich 1 and J. Schlessinger z

Growth factors, differentiation factors, and IV RTKs, respectively. The tyrosine and polypeptide hormones are cru• kinase domain of the latter is interrupted cial components of the regulatory sys• by hydrophilic insertion sequences of tem that coordinates development of varying length. The availability of RTK multicellular organisms. Many of these cDNA clones has made it possible to factors mediate their pleiotropic actions initiate detailed structure-function ana• by binding to and activating cell surface lyses of the mechanisms of action of R TK receptors with an intrinsic protein ty• family members. Numerous mutants of rosine kinase (PTK) activity. Figure 1 insulin, epidermal growth factor (EGF), presents a schematic representation of the platelet-derived growth factor (PDGF), known growth factor receptors that bear insulin-like growth factor 1 (IGF-l), PTK activity. Growth factor receptors colony-stimulating factor 1 (CSF-l), and with PTK activity, or receptor tyrosine other receptors have been characterized kinases (RTKs), have a similar molecular in regard to their biological and biochem• topology. All possess a large, glycosy• ical properties. This has led to the es• lated, extracellular, ligand-binding tablishment of a receptor domain func• domain, a single hydrophobic transmem• tion map and model for R TK-mediated brane region, and a cytoplasmic domain signal generation (Fig. 2). which contains a PTK catalytic domain Ligand binding to the extracellular (Hanks et aI., 1988; Yarden and Ullrich domain of the receptor results in con• 1988, Schlessinger 1988; Williams 1989). formational change and subsequent Primary sequence homology and dis• oligomerization [Schlessinger 1988]. Re• tinct structural characteristics of different ceptor oligomerization is a universal R TKs allow the classification of these phenomenon among growth factor re• receptors into subclasses (Fig. 1). The ceptors. It has been detected in living structural features characteristic of the cells, in isolated membranes, and in pre• four subclasses include two cysteine-rich parations of solubilized and purified re• repeat sequences in the extracellular ceptors [Schlessinger 1986; Yarden and domain of monomeric subclass I recep• Schlessinger 1985, 1987 a, b; Cochet et aI., tors, disulfide-linked heterotetrameric 1988]. It may be induced by either mono• Ci.zfJz structures with similar cysteine-rich meric ligands, such as EGF, which cause sequences in subclass II RTKs, and five receptor oligomerization by inducing or three immunoglobulin-like repeats in conformational changes [Greenfield et ai. the extracellular domains of subclass III 1989] resulting in receptor-receptor inter• actions [Lax et ai. 1990] or by bivalent ligands, such as PDGF and CSF-l, which mediate dimerization of neighboring re• 1 Max-Planck-Institut fiir Biochemie, Am Klopferspitz 18A, 8033 Martinsried, FRG. ceptors [Seifert et ai. 1989; Heldin et ai. 2 Department of Pharmacology, New York 1989; Hammacher et ai. 1989]. Oligo• University Medical Center, 550 First merized growth factor receptors possess Avenue, New York, NY 10016, USA. elevated PTK activity [Yarden and

Haematology and Blood Transfusion Vol. 35 XXIII Modern Trends in Human Leukemia IX R. Neth et al. (Eds.) © Springer-Verlag Berlin Heidelberg 1992 II III IV Receptor phosphorylation releases an internal constraint b~ stabilizing a con• formation that is competent to interact with and phosphorylate cellular substrates [Honegger et al. 1988a, b]. The recent cys observation that phosphorylation of EGF and insulin receptors can occur by intermolecular cross-phosphorylation both in vitro and in living cells [Honegger cys et al. 1989, 1990; Ballotti et al. 1989; Lammers et al. 1990] further supports the importance of receptor oligomerization in the process of receptor activation. PTK The chain of events that is initiated by tyrosine phosphorylation of cellular substrates is still poorly understood. Several R TK sub states of potential bi• EGF·R Insulin·R PDGF·R-A FGF·R HER2/neu IGF·j·R PDGF·R-B fig ological importance have recently been HER3/c·erbB·3 IRR CSF-1-R bek Xmrk c-kit identified (Figure 3). Both PDGF and EGF can induce tyrosine phosphoryl• Fig.1. Schematic representation of receptor ation of phospholipase Cy (PLC-y) in tyrosine kinase subclasses. For details, see Ullrich and Schlessinger (1990) vitro and in living cells [Margolis et al. 1989; Meisenhelder et al. 1989; Wahl et al. 1989]. In addition, PLC-y was observed to associate with the activated receptor Schlessinger 1987 a, b; B6ni-Schnetzler kinases in a ligand- and kinase-dependent and Pilch 1987], which leads to phos• manner [Margolis et al. 1989, 1990a; phorylation of tyrosine residues of the Kumjian et al. 1989]. However, growth receptor polypeptide chain and of cel• factor-induced inositol triphosphate lular substrates. (IP 3) generation appears not to be the

DOMAIN FUNCTION

Extracellular Ligand binding Dimerization

Transmembrane Membrane anchor

Juxtamembrane Signal control

ATP binding (Lys 721) Substrate binding Tyrosin e Kinase

Carboxy-term inal tail Signal regulation Fig. 2. Proposed structure-function topology formed by subdomains I and III. The symbols of the EGF receptor. Subdomains II and IV Sand R within the PTK domain represent (stippled) represent the cysteine-rich regions of proposed interaction sites for substrates and the extracellular domain. Most of the struc• regulatory factors [Ullrich and Schlessinger, tural determinants that define EGF binding 1990] affinity are proposed to be located in the cleft

XXIV PKC ---- I I I I PlC'y Ptdlns-3K GAP raj o ~ 8 ~ ~ ~ DAG IP3 Ptdlns(3)P ~ras B ! j tCa2+ -V Fig. 3. Receptor-mediated multiple signalling of enzymatic functions by aSSOCiatIOn, and pathways. Direct phosphorylation (black dots Thr/Ser phosphorylation (white dot on sym• on symbols) of substrates, PLC-y, PtdIns-3 K, bol) of substrates [Ullrich and Schlessinger, GAP, and raj leads to secondary events, 1990 ]Ptdlns-3K: phosphatidylinositol 3-ki• including enzymatic activation and metabolite nase; GAP: GTPase-activating protein; Ptd• formation (DAG, IP3, PtdIns(3)P), activation Ins(3)P: phosphatidylinositol 3-phosphate sole mechanism leading to the initiation able model systems not only for studying of DNA synthesis [Downing et al. 1989], the mechanisms of oncogenesis but also which is compatible with the notion that for the analysis of normal structure• the phosphatidylinositol (PI) signalling function relationships for these signal• pathway does not play an essential role in transmitter molecules. Constitutive acti• the mitogenic response [Lopez-Rivas et vation of R TK signalling functions can al.1987; L'Allemain et al.1989; Margolis be achieved in a number of ways. For et al. 1990b]. example, in the cases ofv-erb-B and v-kit, Other RTK substrates that have re• deletion of the extracellular binding cently been identified include PI kinase domain eliminates the negative control and the ras binding protein GAP [Kaplan that this structure normally exerts on the et al. 1987; Varticovski et al. 1989; Molloy cytoplasmic domain. Even point muta• et al. 1989] (Fig. 3). Similarly, it has been tions within the extracellular domain suggested that the cora! protooncogene can lead to intracellular activation, as in product becomes phosphorylated in re• sponse to PDGF receptor activation [Morrison et al. 1989]. Intriguingly, all proteins identified thus far as R TK tar• gets are either components of second messenger pathways, protooncogene products, or factors that regulate the activity of protooncogene products. Amplified Signal The importance of allosteric regulation of receptor activation and signal trans• J1 Transformation duction is further emphasized by the fact that a large variety of structural alter• Fig. 4. Transformation by receptor amplific• ations found in RTK-derived oncogene ation. Schematic representation of proposed transformation model by autocrine stimul• products lead to constitutive kinase acti• ation of overexpressed receptor tyrosine vation and, consequently, subversion of kinases. Ligand (black dots) is activating re• molecular control mechanisms and alter• ceptors in the plasma membrane of a tumor ation of receptor signals. Thus, trans• cell, resulting in an amplified transforming forming RTK derivatives serve as valu- signal xxv the case of v-fms mutations at residues Despite the presence of an intact extra• 301 and 374 [Woolford et al. 1988; Rous• cellular domain, these mutations render sel et al. 1988] (Fig. 4). These mutations the EGF receptor competent for mito• appear to induce and stabilize a confor• genic and transforming signalling with• mational change equivalent to that trig• out autophosphorylation. RTK-derived gered by ligand binding and possibly oncogenes possess other structural le• dimerization. Another dramatic effect of sions such as cytoplasmic point muta• a single point mutation is exemplified by tions, deletions, and C-terminal trunca• the Val/Glu conversion in the neu trans• tions which appear to enhance and membrane domain [Bargmann et al. modulate the transforming signal [Kha• 1986], which suggests that this part of the zaie et al. 1988; Woolford et al. 1988]. putative receptor is involved in an overall For human cancer, activating RTK conformational alteration that occurs mutations appear to be of minor im• upon interaction with the yet unidentified portance. The most common cellular le• ligand. In this case, the transmembrane sion found in human cancers involves mutation results in constitutive receptor autocrine activation in conjunction with oligomerization [Weiner et al. 1989]. receptor overexpression (Fig. 4). Many Another type of structural alteration has tumors and tumor cell lines have been been identified in the EG F receptor / erb-B found to coexpress growth factors and system and involves mutations in the their receptors, including TG F -C(, PDGF• PTK core region [Massoglia et al. 1990]. A, PDGF-B, acidic fibroblast growth

-

Fig. 5. HER2/neu gene amplification in mam• mammary carcinoma tumors [Slamon et al. mary carcinoma. Southern blot hybridization 1987] analysis of chromosomal DNA from primary

XXVI "0 HER ~ c -o A 8 U EGF TGF-oc

200- HER-A

HER-8

NIH-3T3 97-

b a

Fig. 6a, b. Cell transformation by EGF re• man EGF receptor (a) were stimulating with ceptor overexpression. NIH-3T3 cell lines EGF or TGF-o: and tested for their ability to HER-A and HER-B overexpressing the hu- grow in soft agar (b)

factor (FGF), basic FGF, and their spe• altered p 185HER2/neu [Hudziak et al. cific R TKs. Thus, autocrine receptor ac• 1987]. Analogous experiments with the tivation represents yet another scenario EGF receptor indicated that autocrine of subversion of normal growth control. stimulation of the overexpressed receptor For mammary and ovarian carcinoma, was essential to achieve a transforming extensive studies have demonstrated a effect [Di Fiore et al. 1987; Velu et al. direct correlation between the extent of 1987; Riedel et al. 1988] (Fig. 6). overexpression of p 185HER2/neu and a On the basis of these findings, patient's prognosis, a result which strong• strategies involving antireceptor anti• ly suggests a critical role for this EGF bodies were designed for the treatment of receptor-like R TK in tumor progression mammary and ovarian carcinoma. and perhaps even tumor initiation Monoclonal antibodies, such as the anti• [Slamon et al. 1989] (Fig. 5). This possi• HER2/neu antibody 4D5, are able to bility is further supported by efficient interfere with autocrine activation of the induction of mammary carcinoma in receptor, which results in inhibition of mice by an activated neu gene product tumor cell growth in tissue culture and [Muller et al. 1988] and transformation of nude mouse models (Ullrich et aI., NIH-3T3 cells by overexpression of un- unpublished).

XXVII In principle, every receptor with PTK singer J (1989) Phospholipase C-y, a sub• activity has oncogenic potential. One can strate for PDGF r~ceptor kinase, is not anticipate that many more types of ac• phosphorylated on tyrosine residues dur• tivating mutations, as well as specific ing the mitogenic response to CSF-1. instances of R TK overexpression, will be EMBO J 8:3345-3350 8. Greenfield C, Hils I, Waterfield MD, detected in animal and human tumors. Federwisch M, Wollmer A, Blundell TL, The molecular identification and charac• McDonald N (1989) EGF binding induces terization of these mutants will not only a conformational change in the external provide important insights into funda• domain of its receptor. EMBO J 8:4115- mental mechanisms underlying receptor 4124 activation and normal growth control, 9. Hammacher A, Mellstrom K, Heldin CH, but may also enhance our understanding Westermark B (1989) Isoform-specific in• of oncogenesis and open new avenues for duction of actin reorganization by PDGF diagnosis and therapy. suggests that the functionally active re• ceptor is a dimer. EMBO J 8:2489-2495 10. Hanks SK, Quinn AM, Hunter T (1988) The protein kinase family: conserved fea• References tures and deduced phylogeny of the cataly• tic domains. Science 241:45-52 1. Ballotti R, Lammers R, Scimeca JC, Dull 11. Heldin C-H, Ernlund A, Rorsman C, T, Schlessinger J, Ullrich A, Van Ob• Ronnstrand L (1989) Dimerization of the berghen E (1989) Intermolecular trans• B type PDGF-receptor occurs after ligand phosphorylation between insulin recep• binding and is closely associated with tors and EGF-insulin receptor chimerae. receptor kinase activation. J Bioi Chern EMBO J 8:3303-3309 264:8905-8912 2. Bargmann CI, Hung MC, Weinberg RA 12. Honegger A, Dull TJ, Bellot F, Van Ob• (1986) Multiple independent activations berghen E, Szapary D, Schmidt A, Ullrich of the neu oncogene by a point mutation A, Schlessinger J (1988a) Biological activ• altering the transmembrane domain of ities of EGF-receptor mutants with indi• p 185. Cell 45:649-657 vidually altered autophosphorylation 3. B6ni-Schnetzler M, Pilch PF (1987) Mech• sites. EMBO J 7:3045-3052 anism of EGF-receptor autophosphory• 13. Honegger A, Dull TJ, Szepary D, Komori• lation and high affinity binding. Proc ya A, Kris R, Ullrich A, Schlessinger J Natl Acad Sci USA 84:7832-7836 (1988 b) Kinetic parameters of the protein 4. Cochet C, Kashles 0, Chambaz EM, tyrosine kinase activity of EGF-receptor Borrello I, King CR, Schlessinger J (1988) mutants with individually altered auto• Demonstration of epidermal growth phosphorylation sites. EMBO J 7:3053- factor-induced receptor dimerization in 3060 living cells using a chemical covalent cross• 14. Honegger AM, Kris RM, Ullrich A, linking agent. J Bioi Chern 263: 3290- Schlessinger J (1989) Evidence that auto• 3295 phosphorylation of solubilized EGF• 5. Di Fiore PP, Pierce JH, Fleming TP, receptors is mediated by intermolecular Hazan R, Ullrich A, King CR, Schlessin• cross phosphorylation. Proc Natl Acad Sci ger J, Aaronson SA (1987) Overexpression USA 86:925-929 of the human EGF receptor confers an 15. Honegger AM, Schmidt A, Ullrich A, EGF-dependent transformed phenotype Schlessinger J (1990) Evidence for EGF to NIH 3T3 cells. Cell 51:1063-1070 induced intermolecular autophosphory• 6. Dionne C, Crumley G, Bellot F, Kaplow J, lation of the EG F -receptor in living cells. Searfoss G, Ruta M, Burgess W, Jaye M, Mol Cell BeioI10:4035-4044 Schlessinger J (1990) Cloning and ex• 16. Hudziak RM, Schlessinger J, Ullrich A pression of two distinct high-affinity re• (1987) Increased expression ofthe putative ceptors cross-reacting with acidic and growth factor receptor p 185HER2 causes basic fibroblast growth factors. EMBO J transformation and tumorigenesis of NIH 9:2685-2692 3T3 cells. Proc Natl Acad Sci USA 7. Downing JR, Margolis BL, Zilberstein A, 84:7159-7163 Ashmun RA, Ullrich A, Sherr CJ, Schles-

XXVIII 17. Kaplan DR, Whitman M, Schaffhausen Swiss 3 T3 mouse cells. Proc Nat! Acad Sci B, Pallas DC, White M, Cutley L, Roberts USA 84:5768-5772 TM (1987) Common elements in growth 27. Margolis BL, Rhee SG, Felder S, Lyall R, factor stimulation and oncogenic trans• Levitski A, Ullrich A, Zilberstein A, formation: 85 Kd phosphorylation and Schlessinger J (1989) EGF induces phos• phosphatidylinositol kinase activity. Cell phorylation of phospholipase C-II: a po• 50: 1021-1029 tential mechanism for EGF-receptor sig• 18. Khazaie K, Dull TJ, Graf T, Schlessinger nalling. Cell 57: 1102 -11 07 J, Ullrich A, Beug H, Vennstrom B (1988) 28. Margolis B, Bellot F, Honegger AM, Ull• Truncation of the human EGF receptor rich A, Schlessinger J, Zilberstein A leads to differential transforming poten• (1990a) Tyrosine kinase activity is essen• tials in primary avian fibroblasts and ery• tial for the association of phospholipase throblasts. EMBO J 7: 3061-3071 C-y with EGF-receptor. Mol Cell BioI 19. Kornbluth S, Paulson KE, Hanafusa H 10:435-441 (1988) Novel tyrosine kinase identified by 29. Margolis B, Zilberstein A, Franks C, Fel• phospho tyrosine antibody screening of der S, Kreamer S, Ullrich A, Rhee SG, cDNA.libraries. Mol Cell BioI 8:5541 Skorecki K, Schlessinger J (1990b) Effect 5544 of phospholipase c-y overexpression on 20. Kraus MH, Issing W, Miki T, Popescu PDGF-induced second messengers and NC, Aaronson SA (1989) Isolation and mitogenesis. Science 248:607-610 characterization of ERBB 3, a third mem• 30. Massoglia S, Gray A, Dull TJ, Munemitsu ber of the ERBB/epidermal growth factor S, Kung H-J, Schlessinger J, Ullrich A receptor family: evidence of overex• (1990) Epidermal growth factor receptor pression in a subset of human mammary cytoplasmic domain mutations trigger tumors. Proc Natl Acad Sci USA ligand-independent transformation. Mol 86:9193-9197 Cell BioI 10: 3048 - 3055 21. Kumjian DA, Wahl MI, Rhee SG, Daniel 31. Meisenhelder J, Suh P-G, Rhee SG, Hun• TO (1989) Platelet-derived growth factor ter T (1989) Phospholipase C-y is a sub• (PDGF) binding promotes physical as• strate for the PDGF and EGF receptor sociation of PDGF receptor with phos• protein-tyrosine kinases in vivo and in pholipase C. Proc Natl Acad Sci USA vitro. Cell 57: 1109-1122 86:8232-8236 32. Molloy CJ, Bottaro DP, Fleming TP, 22. L' Allemain G, Seuwen K, Velu T, Pou• Marshall MS, Gibbs JB, Aaronson SA yssegur J (1989) Signal transduction in (1989) PDGF induction of tyrosine phos• hamster fibroblasts overexpressing the phorylation of GTPase activating protein. human EGF-receptor. Growth Factors Nature 342:711-714 1 :311-321 33. Morrison DK, Kaplan DR, Escobedo JA, 23. Lammers R, Van Obberghen E, Ballotti R, Rapp UR, Roberts TM, Williams L T Schlessinger J, Ullrich A (1990) Trans• (1989) Direct activation of the serine/thre• phosphorylation as a possible mechanism onine kinase activity through tyrosine for insulin and EGF receptor activation. phosphorylation by the PDGF f3 receptor. BioI Chem 265:16886-16890 Cell 58:649-657 24. Lax I, Mitra AK, Stroud RM, Ravera C, 34. Muller WJ, Sinn E, Pattengale PK, Wal• Givol D, Hurwitz DR, Ullrich A, Schles• lace R, Leder P (1988) Single-step induc• singer J (1990) EGF-induced oligomeri• tion of mammary adenocarcinoma in zation of soluble, extracellular, ligand transgenic mice bearing the activated c• binding domain of EGF receptor BioI neu oncogene. Cell 54: 105 -1 09 Chem (in press) 35. Riedel H, Massoglia S, Schlessinger J, 25. Lee PL, Johnson DE, Cousens LS, Fried Ullrich A (1988) Ligand activation of VA, Williams L T (1989) Purification and overexpressed epidermal growth factor re• complementary DNA cloning of a re• ceptors transforms NIH 3 T 3 mouse fi• ceptor for basic fibroblast growth factor. broblasts. Proc Nat! Acad Sci USA Science 245:57-60 85: 1477 -1481 26. Lopez-Rivas A, Mendoza SA, Nanberg E, 36. Roussel MF, Downing JR, Rettenmier Sinnett-Smith J, Rozengurt E (1987) Ca +2 CW, Sherr CJ (1988) A point mutation in mobilizing actions of PDGF differ from the extracellular domain of the human those of bombesin and vasopressin in CSF-1 receptor (c-fms proto-oncogene

XXIX product) activates its transforming poten• lingham MC, Merlino GT, Pastan I, Lowy tial. Cell 55:979-988 OR (1987) Epidermal growth factor• 37. Ruta M, Howk R, Ricca G, Drohan W, dependent transformation by a human Zabelshansky M, Laureys G, Barton DE, EGF receptor protooncogene. Science Francke U, Schlessinger J, Givol D (1988) 237: 1408-1410 A novel protein tyrosine kinase gene 46. Wahl M, Nishibe S, Suh P-G, Rhee SG, whose expression is modulated during Carpenter G (1989) Epidermal growth endothelial cell differentiation. Oncogene factor stimulates tyrosine phosphoryla• 3:9-15 tion of phospholipase C-II independently 38. Ruta M, Burgess W, Givol D, Epstein J, of receptor internalization and extracel• Neiger N, Kaplow J, Crumley G, Dionne lular calcium. Proc Nat! Acad Sci USA C, Jaye M, Schlessinger J (1989) Receptor 86:1568-1572 for acidic FGF is related to the tyrosine 47. Weiner DB, Liu J, Cohen JA, Williams kinase encoded by the fms like gene (fig). WV, Greene M (1989) A point mutation in Proc Natl Acad Sci USA 86:8722-8726 the neu oncogene mimics ligand induction 39. Schlessinger J (1986) Allosteric regulation of receptor aggregation. Nature 339: 230- of the epidermal growth factor receptor 231 kinase. J Cell Bioi 103:2067-2072 48. Williams LT (1989) Signal transduction by 40. Schlessinger J (1988) Signal transduction the PDGF-receptor. Science 243: 1564- by allosteric receptor oligomerization. 1570 Trends Biochem Sci 13:443-447 49. Wittbrodt J, Adam D, Malitschek B, 41. Seifert RA, Hart CE, Phillips PE, For• Maueler B, Raulf F, Telling A, Robertson strom JHW, Ross R, Murray MJ, Bowen• SM, Schartl M (1989) Novel putative Pope DF (1989) Two different subunits receptor kinase encoded by the melanoma• associate to create isoform-specific inducing T4 locus in Xiphophorus. Na• PDGF-receptors. J BioI Chern 264: 8771- ture 341 :415-421 8778 50. Woolford JW, McAuliffe A, Rohrschneider 42. Shier P, Watt VM (1989) Primary struc• LR (1988) Activation of the feline c-fms ture of a putative receptor for a ligand of proto oncogene: multiple alterations are the insulin family. J BioI Chern required to generate a fully transformed 264: 14605-14608 phenotype. Cell 55:965-977 42a. Slamon DJ, Clark GM, Wong SG, Levin 51. Yarden Y, Schlessinger J (1985) EGF WJ, Ullrich A, McGuire WL (1987) Hu• receptor self-phosphorlyation is mediated man breast cancer: Correlation of relapse by an intermolecular allosteric process. In: and survival with amplification of the Growth factors in biology and medicine. HER-2/neu oncogene. Science 235:177- Pitman, London, pp 23-45 182 52. Yarden Y, Schlessinger J (1987a) Self• 43. Slamon OJ, Godolphin W, Jones LA, Holt phosphorylation of epidermal growth fac• JA, Wong SG, Keith DE, Levin WJ, tor receptor: Evidence of a model of Stuart SG, Udove J, Ullrich A, Press MF intermolecular allosteric activation. Bio• (1989) Studies of the HER-2/neu proto• chemistry 26: 1434-1442 oncogene in human breast and ovarian 53. Yarden Y, Schlessinger J (1987b) Epi• cancer. Science 244: 707 - 712 dermal growth factor induces rapid, 44. Varticovski L, Oruker B, Morrison 0, reversible aggregation of the purified Cantley L, Roberts T (1989) The colony epidermal growth factor receptor. Bio• stimulating factor-1 receptor associates chemistry 26:1443-1451 with and activates phosphatidylinositol-3 54. Yarden Y, Ullrich A (1988) Growth factor kinase. Nature 342:699-702 receptor tyrosine kinases. Annu Rev Bio• 45. Velu TJ, Beguinot L, Vass WC, Wil- chern 57:443-478

xxx Yuri Ovchinnikov Memorial Lecture Moscow, June 13, 1990

V. T. Ivanov: In Memoriam Yuri Ovchinnikov

V. T. Ivanov and V. A. Nesmeyanov: Prof. Yuri Ovchinnikov

Jannet D. Rowley: Human Leukemia Genes: Search for the Villains

Robert C. Gallo: Human Retroviruses: Linkage to Leukemia and AIDS

XXXI In memoriam Prof. Yuri Ovchinnikov

V. T. Ivanov

Who was Yuri Ovchinnikov? What were chemist were reached. Here, I would like his ideas and what did he achieve during to offer the reader a glipse of Ovchin• his dynamic although brief carrier? Why nikov as a human being. He was born in do we remember him at this Wilsede 1934 in Moscow; in 1952 he entered the meeting? I do not think that these ques• Chemical Department of Moscow Uni• tions really ought to be answered, since versity and in 1957, after graduation, he Yuri Ovchinnikov belonged to the elite of became a professional researcher. Many international scientific community, being facets of his unique personality showed extremely well known not only for the up already in these early years. He had a results of his own research but also as an phenomenal memory and was a leading outstanding leader of chemical and bio• actor at the University studio. logical sciences in the Soviet Union and Yuri loved sports. He was a University a champion of international scientific champion in free-style wrestling, and was collaboration. also a keen swimmer and cross-country A separate chapter of this book de• skier. These skills he maintained for scribes the path along which the main many years to come. It seems that the achievements of Ovchinnikov as a bio- famous Robert Woodward has little

Fig. 1. Moments of relaxation (1970): Yuri Ovchinnikov (left) and Nobel prize winner Prof. R. Woodward (USA)

Haematology and Blood Transfusion Vol. 35 XXXIII Modern Trends in Human Leukemia IX R. Neth et al. (Eds.) © Springer· Verlag Berlin Heidelberg 1992 Fig. 2. Yuri Ovchinnikov lecturing (1972)

Fig. 4. At Yaroslavl airport (1982)

of the characteristic Yuri's qualities and feelings which are well reflected in Figs. 2, 3. A favorite among many of those who knew Yuri well is Fig. 4, where we see his Fig. 3. Deep in thought (1975) firm stand, the boyish mischief, the in• vincible optimism and the belief that "we shall overcome", whatever the ob• chance of surviving in the billiard game stacles. with Yuri shown in Fig. 1. Indeed, he overcame much, but one Being a loving husband and father and thing proved too hard even for him: generally a very warm personality, Ov• disease, the disease which could have chinnikov had a talent for collecting been defeated if we knew more about teams of young enthusiastic people and the hemopoietic process. Maybe this is helping them in their research. one more reason for dedicating the Volga A clear vision, high motivation, the Wilsede meeting to the memory of Yuri burden of responsibility - these are some Ovchinnikov.

XXXIV Prof. Yuri Ovchninnikov

V. T. Ivanov and V. A. Nesmeyanov 1

Yuri A. Ovchinnikov started his carrier The subject under study depsipeptide within the precincts of Moscow Univer• antibiotics, atypical peptides containing sity at the Chemical Department under hydroxy and amino acid residues. The Professor Yu. A. Arbuzov. The project of problems of synthesis of the optically his masters' degree (1957) provided active N-methylated amino acids, rever• material for the first publication on a new sible protection ofthe hydroxyl function technique for the synthesis ofpyrrolidine of hydroxy acids, and cyclization oflinear and thiophan derivatives. By that time, depsipeptides were rapidly solved and the gifted student had already shown a compounds with the structures proposed disposition toward synthetic organic in the 1940s by Swiss researchers for chemistry. It was at this period that his antibiotics enniatins A and B were pre• belief took shape that the chemistry of pared. However, the samples obtained living organisms was by far the most were devoid of antimicrobial activity and attractive area for an organic chemist to their physicochemical properties differed enter. Therefore, having begun his post• much from those of the naturally reliably graduate course at the Chemical Depart• confirmed, it remained to conclude that ment, Y. A. Ovchinnikov readily accep• the formulae proposed for enniatins A ted an invitation to participate in the and B were incorrect. Several alternative project on the complete synthesis of an structures differing in ring size were sug• important group of antibiotics, the tetra• gested and accordingly synthesized. Two cyclines. While working toward his doc• of them were indistinguishable from the torate, Yuri Ovchinnikov met M. M. natural enniatins A and B, which meant a Shemyakin, the leader of the project. solution to structural problems. The joint work led to a long-lasting Later (1964-1970), Ovchinnikov and collaboration between the two scientists, his colleagues performed a series of ele• whose contribution to the foundation gant syntheses of some other naturally and advancement of physicochemical occurring depsipeptides (sporidesmolides biology in the USSR was outstanding. I -IV, angolide, serratamolide, esperin, After finished his postgraduate course, beauvericin). he was awarded a D. Sc. in Ovchinnikov joined the Institute for 1966 for the synthesis of natural de• Chemistry of Natural Products of the psipeptides and their analogs. USSR Academy of Sciences, set up not In 1967, Shemyakin, Ovchinnikov, and long ago. Here, Professor Shemyakin their team formulated the original (so• proposed that he go into peptide called topochemical) principle of trans• chemistry. formation of biologically active peptides: novel molecules can be designed by such deep structural modifications as reversal of the acylation direction and the con•

1 Shemyakin Institute of Bioorganic Chemis• figuration of asymmetric centers, replace• try, USSR Academy of Sciences, Moscow, ment of ester bonds by amide bonds and USSR. vica versa, cyclization oflinear molecules,

Haematology and Blood Transfusion Vol. 35 xxxv Modern Trends in Human Leukemia IX R. Neth et al. (Eds.) © Springer-Verlag Berlin Heidelberg 1992 etc. The conditions favorable for retain• Several laboratories outside the USSR ing the original stereoelectronic para• were about to get similar results, but "the meters and, consequently, biological pro• train had already gone." perties of the molecule were found. Ideas Step-by-step protein compounds, the from this pioneer research were taken up major working bodies of any living sys• by many laboratories and served to create tem, began to occupy the prominent novel highly active pep tides (hormones, place in Ovchinnikov's research activity. antibiotics, neuropeptides, enzyme sub• The 1970s witnessed a series of studies on strates, and inhibitors). the primary structures of porcine aspar• The experience accumulated during tate aminotransferase, and toxins from this synthetic work served as a basis for the venoms of cobra, bee, scorpion, etc. the next and the culiminative step in As a result, more than 20 structures were studying the depsipeptide antibiotics. added to international data banks and Bearing in mind the recently discovered atlases of protein structures. Inspired by ability of valinomycin and enniatins to these advances, Ovchinnikov and his induce permeability of lipid membranes group tackled the deciphering of the to alkali metals ions, Y. A. Ovchinnikov primary structure of E. coli DNA• and his colleagues undertook a study of dependent RNA polymerase, a key tran• the physicochemical basis of the pheno• scription enzyme investigated in many menon. It appeared that valinomycin laboratories. Ovchinnikov had a very binds potassium ions in solution, yielding strong team, but even for them the pro• stable complexes, and shows a unique blem seemed extremely difficult, since KINa-selectivity of complex formation RNA polymerase is built of several sub• unsurpassed in nature. Enniatins bind units, among them two very large 13- and virtually all alkali and alkali-earth cat• 13' -subunits (each over 1300 amino acid ions, though with a lower selectivity. residues). Indeed, after rapid sequencing These complexes are the ion-transporting of the a-subunit (over 300 amino acids) it species, and selectivity of ion binding is became clear that analysis of the 13- and the origin of the selectivity of transmem• 13' -subunits exclusively by conventional brane ion transport. Further, the three• methods of protein chemistry could take dimensional structures of the free anti• many years. A decision was made to biotics and their complexes were estab• utilize the methods of genetic engineering lished. It was shown for the first time that and to analyze the sequences of genes such sophisticated structures can be re• coding for the subunits. In those days, solved not only by X-ray analysis but also such an approach was new for this coun• in solution by spectral methods. The try, and elsewhere it was at the early bound ion appeared to reside always in stages of development. the center of the depsipeptide molecular Genes for large subunits of DNA• cavity and be kept in place by ion-dipole dependent RNA polymerase form the so• interactions with the carbonyl oxygens. called operon rpo Be and contain about The size of the valinomycin cavity is 10000 base pairs. They were isolated, limited by a bracelet-like system of six inserted into plasmids, and sequenced. intramolecular hydrogen bonds that ac• Structures of peptides of large subunits counts for its inability to adapt to smaller were detected in parallel and indepen• sized ions such as sodium or lithium. dently. That was of use: when the struc• Enniatin structures are more flexible, tural analysis of genes was completed and which enables adjustment of the cavity to the structures of corresponding proteins the size of the bound ion. The molecular were derived according to the genetic periphery of both valinomycin and en• code, they appeared to coincide with the niatin complexes is fully hydrophobic, peptide structures and, consequently, which allows them to migrate freely were determined correctly. Soon after across lipid zones of the membrane. that, other laboratories reported the gene

XXXVI fragments but not the complete gene. It Though bacteriorhodopsin functioned as is worthwhile noting that the structures a light-dependent pro top pump, from the of these fragments contained errors. On• viewpoint of the primary photochemical ly the combined use of the methods of properties it was very similar to rhodop• protein and nucleotide chemistry pro• sin. At the same time, bacteriorhodopsin vided reliable results. is more readily available in large amounts The structural analysis of RNA poly• and has a simpler structure than the merase served as a basis for a thorough visual rhodopsin, the main effort was investigation of the mechanism of action initially directed to that protein. It was of the enzyme, for numerous genetic and also considered that bacteriorhodopsin biochemical studies. was (and still is) an ideal model for That was in the late 1970s. More and structure-functional analysis of mem• more laboratories outside the USSR were brane proteins. Simaltaneously with successfully applying genetic engineering Prof. G. Khorana of the USA, the Nobel methods to microbiological synthesis of prize winner, Ovchinnikov succeeded in practically important proteins. Yuri Ov• determining the amino acid sequence of chinnikov was the first in the USSR to bacteriorhodopsin, is was the first time assess the prospects. He united en• that the chemical structure of the mem• thusiasts and headed the work on improv• brane protein had been deciphered ing the methods of chemical synthesis (1987). Ovchinnikov and his team were and directed mutagenesis of DNA to then pioneers in solving the struc• create microorganisms producing alien ture of rhodopsin from bovine retina peptides and proteins. As a result, strains (1981). producing an opioid neuropeptide, Research into the topography of poly• leucine-enkephalin (1979), the antiviral peptide chains of these proteins in native and antitumour human protein inter• membranes and elucidation of the struc• feron-cx 2 (1981), and the precursor of ture of their active sites and disposition of human insulin, proinsulin (1983), were functionally important groups were the obtained. next steps in this project. Using a variety Despite these advances of Yuri Ov• of approaches including chemical modi• chinnikov in genetic engineering and bio• fication, enzymatic treatment, and im• technology, the bioorganic chemistry of munochemical methods, Yuri Ovchin• peptides and proteins was always his nikov and his colleagues demonstrated major interest and devotion. that the two rhodopsins are arranged in In the mid-1970s, he, N. Abdulaev, the membrane in a similar way - as seven and a group of colleagues focused their· extended protein segments spanning the interest on the molecular mechanisms of membrane's width and connected with photoreception. By that time, a series of each other on the two sides of the mem• substantial discoveries had been made brane by short peptide links. that paved the way for solving the prob• In the mid-1980s, Y. Ovchinnikov and lem of how light energy is transformed V. Lipkin focused their attention on the into the electric energy of the nerve studies of other proteins involved in impulse by rhodopsin, a well-known transmission and amplification of the light-sensitive protein from the animal visual cascade - transducin and cyclic retina. GMD phosphodiesterase. In 1985, the Soon after wards, there appeared data primary structures of the y- and cx• on the membrane protein - bacteriorhod• subunits of transducin from bovine re• opsin - found in microorganisms living in tinal rods were sequenced. Interestingly, salt lakes. The protein was given that the y-subunit is characterized by the two name because of its similarity to the adjoining cysteine residues also connec• visual rhodopsin (the presence of the ted by a disulfide bridge. The residues are bound retinal, light-sensitivity, etc.). apparently involved in the formation of

XXXVII the transducin-photoactivated rhodopsin Yuri Ovchinnikov, together with Eu• complex. gene Sverdlov and their groups of resear• An exciting page in the scientific bio• chers, obtained novel data on the regions graphy of Yuri Ovchinnikov was his last of the human genome encoding the sys• project, devoted to studies of the system tems of active ion transport that seem to of active ion transport, i.e., Na,K• be of general biological significance. A transporting adenosine triphosphatase family of at least five genes was defined in and related proteins. In the late 1970s, the human genome coding for several Ovchinnikov initiated research into the isoforms of the Na,K-ATPase catalytic structure of Na,K-ATPase. At the subunit as well as other structurally beginning, oligomeric organization of the similar ion-transporting ATPases. functionally active complex in the na• The discovery of the multigene family tive membrane was unraveled and the gave rise to new concepts on regulation of asymmetric arrangement of the subunits the active ion transport through changes described. Further progress depended in the activity of the appropriate genes. upon determination of the amino acid This was supported by experiments on sequence of the subunits. Around 1985- the expression level of various genes for 1986, Ovchinnikov's team completed Na,K-ATPase in healthy and patholog• studies of the nucleotide sequences of ical human tissues. Thus, ideas on the genes for subunits and amino acid se• mechanisms of genetic regulation of ion• quences of their polypeptide chains, transporting enzymes received a solid which led to the complete primary struc• foundation. ture of Na,K-ATPase from pig kidney Lately, the problems of immunology outer medulla. Some research centers and hematology attracted the attention of outside the USSR were also working Yuri Ovchinnikov, who believed that intensively in these areas. The teams of chemistry and biology should do more to S. Numa (Japan) and A. Schwartz (USA) help solving medical problems in the simultaneously reported amino acid se• USSR. Intense investigations of naturally quences of similar enzymes from other occurring regulators of immunity and sources. hemopoiesis have been started at the However, the approach chosen by Ov• Shemyakin Institute. Some presentations chinnikov extended far beyond the pri• at this symposium deal with these mary structure determination. Comple• problems. mented by spectroscopic and molecular Above, we have outlined the scientific modelling studies, it resulted in the first interests of Yuri Ovchinnikov, who was detailed model of the Na,K-ATPase also in the driving seat in leading the spatial structure. Here, the ex-subunit chemical and biological scientific com• (1016 amino acid residues) forms seven munities of his country. Ovchinnikov transmembrane segments and the major could not imagine how the science could portion of its hydrophilic region accom• evolve without intensive international co• modating the catalytic site is located operation. He excellently presented the inside the cell. The fJ-subunit (302 re• advances of the Institute, and promoted sidues) spans the membrane once and the scientific contacts, giving impetus to a main part of its polypeptide chain forms series of bilateral symposia such as an extracellular glycosylated domain. USSR-FRO, USSR-USA, France• As for the Na,K-ATPase active site, USSR, Sweden-USSR, and Italy• Ovchinnikov and his team employing USSR in various fields of physicochem• affinity modification by A TP analog suc• ical biology, many of which have now ceeded in identifying an unknown com• became a tradition. The remarkable sym• ponent of the catalytic site, thus experi• posia on Frontiers in Bioorganic Chemis• mentally confirming its dynamic changes try and Molecular Biology in Tashkent during enzyme functioning. (1980) and Moscow-Alma-Ata (1984)

XXXVIII were also organized and presided over by numerous works will inspire many gener• him. ations of bioorganic chemists to come, Of Yuri Ovchinnikov occupies a pro• providing the key to solving a diversity of minent place in the world's scientific problems and demonstrating again and heritage. We can only guess at what his again the beauty and the attractive power further endeavors would have been, if he of the world of science. were still alive. It is our hope that this

XXXIX Yuri Ovchinnikov Lecture

Human Leukemia Genes: Search for the Villains

J. D. Rowley

This Ovchinnikov Lecture provides an letions, that were often uniquely as• occasion to review our progress in a sociated with a particular type of leuke• central area of cancer research, namely mia, lymphoma, or sarcoma provided the genetic changes that occur within the clear evidence that these rearrangements cancer cell that are critically involved in were critically involved in malignant the transformation of a normal to a transformation [1-3]. About 70 recur• malignant cell. To concentrate on genes ring translocations as well as many non• to the exclusion of cell biology would be random deletions and other structural too narrow and shortsighted a perspec• abnormalities are listed in the chapter on tive. Nonetheless, I am convinced that structural chromosome changes in neo• until we have isolated the genes that are plasia included in Human Gene Mapping centrally involved in at least some of the 10 [4]. The evidence for the presence of malignant processes in different cell types recurring chromosome abnormalities in a we will be unable to answer the funda• wide variety of human was the mental questions about malignant trans• result of 30 years of painstaking chromo• formation. More importantly, we will be some analysis by my cytogenetic col• unable to answer the questions with pre• leagues around the world. cision. I will limit my consideration to Second, and I believe an even more those changes that have been detected by powerful force acting within the general analyzing the karyotypic pattern of scientific community to reassess the role human cancer cells using chromosome of karyotypic alterations, was the identi• banding, and in particular to those found fication of the genes involved in some of in leukemia. the chromosome rearrangements and the We are living in a golden age of the discovery that some of these genes were biomedical sciences. Increasingly sophis• the human counterparts of the viral onco• ticated instruments and creative scientific genes [5]. In a sense, each group of strategies allow remarkably precise un• investigators gave the other scientific val• derstanding of some aspects of cancer idity. The fact that oncogenes were direct• biology. It is clear that during the course ly involved in chromosome transloca• of the last three decades, the scientific tions demonstrated that both transloca• community's assessment of the role of tions and oncogenes were critically in• chromosome changes in the complex pro• volved in human cancer. cess of malignant transformation has The genetic changes that occur in changed from considering them to be different types of malignant cells are merely trivial epiphenomena to recogniz• quite varied, and clearly several different ing their fundamental involvement at changes occur in the same cell as it is least for some tumors. This change in altered from a normal to a fully malig• attitude has occurred for at least two nant cell. Cytogenetic analysis has been reasons. First, the demonstration of spec• the key to defining at least two major ific recurring chromosome rearrange• categories of rearrangements, namely re• ments, including translocations and de- curring translocations and consistent de-

XL Haematology and Blood Transfusion Vol. 35 Modern Trends in Human Leukemia IX R. Neth et al. (Eds.) © Springer-Verlag Berlin Heidelberg 1992 letions. One of the first translocations, cell tumors have provided the essential identified in 1972, was the 9; 22 translo• DNA probes to clone several dozen cation in chronic myeloid leukemia [6]. translocations [11-13]. There are now at least 70 recurring trans• Fortunately, the rapid progress being locations that have been detected in made in mapping the human genome, human malignant cells. The identifica• coupled with major advances in working tion of consistent chromosome deletions effectively with large pieces of DNA, has has been equally important because it has already made important contributions to provided the absolutely essential infor• the successful mapping of some of the mation regarding the chromosome loca• recurring translocations in the acute tions of the genes that are involved in leukemias and sarcomas. The use of cos• cancer. I submit that the retinoblastoma mids or yeast artificial chromosomes gene would not have been cloned, or at (YAC) as probes to screen much larger least not yet, if cytogeneticists had not segments of DNA for rearrangements identified deletions of the long arm of provides a new strategy for the analysis of chromosome 13, and specifically of band these chromosome abnormalities. We 13 q 14, in patients with constitutional have used these probes in in situ hy• chromosome abnormalities who had a bridization with biotin labeling of the high incidence of retinoblastoma [7]. This DNA and detection with a fluorescein is not to detract from the careful and isothiocyanate-(FITC)-tagged avidin• exciting work of many scientists is actually antiavidin conjugate. The focus of our cloning the gene, but at least they knew research has been the analysis of chromo• where to look [8]. This triumph has now some translocations involving band been joined by the recent cloning of the 11 q 23. This band is of great interest DCC (deleted in colorectal carcinomas) because it is affected in a large number of gene on chromosome 18; the fact that a different recurring rearrangements. The gene important in the transformation of translocations may occur in either acute colorectal cells was located on chromo• lymphoblastic or acute myeloid leuke• some 18 was the result of cytogenetic mias, especially of the monoblastic or analysis of colon cancer cells that re• myelomonocytic subtype. Finally, about vealed that loss of chromosome 18 was a two-thirds of chromosome abnormalities recurring abnormality [9-10]. in leukemia cells of children under 1 year I must acknowledge that it has been a of age involve 11 q 23, regardless of the source of great disapointment to me that morphological classification of the we have progressed so slowly in cloning leukemia. most of the genes located at the break• We have used a series of cosmid probes points in the recurring translocations or as well as a yeast clone containing two inversions in human leukemia. This em• YACs to map the llq23 breakpoint in phasizes the fact that knowing the lo• four different translocations, nail}ely cation of the breakpoint is very helpful t(4; 11), t(6; 11), t(9; 11), and t(ll; 19) in selecting the genes to use as probes [14]. The cosmid probes were isolated by for these rearrangements. However, a Evans et al. [15] and they were mapped to chromosome band contains at least five the region 11 q 22 to 11 q 25 by Lichter et million base pairs and the likelihood that al. [16]. The yeast clone with the YACs the DNA probe that you "pull off the was identified using polymerase chain shelf' is at the breakpoint and can detect reaction primers specific for the CD3G a rearrangement on Southern blot analy• gene. We showed that this yeast clone sis is vanishingly small. The lymphoid contains two Y ACs of 320 and 275 kb leukemias and are the major that differ only because of a 45 kb dele• exceptions to this slow progress, because tion in one of the YACs; the deletion is in the immunoglobulin genes in B cell the end opposite the CD3G/CD3G com• tumors and the T cell receptor genes in T plex. With the use of cosmid probes we

XLI obtained essentially the same results in all other laboratories. There was evidence four translocations. The cosmid probes for rearrangement adjacent to one of the 3.16,23.20, 1.16,4.13, ZB6, and CD3D IGH constant regions in each case. In all remained on chromosome 11. The four of the cases, this rearranged band cosmid probes XH5, XB1, ZC9, PBGD, has been cloned; all showed a rearrange• 9.4, ZA 7, THY 1,8.5, SRPR, XB2, ETS1 ment with sequences from chromosome 23.2, and 5.8 all were translocated to the 19. Three of the breakpoints on chromo• other chromosome. Seven cosmid probes some 19 were within 170 bp of each other; (XH5, XB1, ZC9, PBGD, 9.4, ZA 7, and the fourth lay 19 kb centromeric. Overall, THY 1) were deleted in one t (9: 11) pa• a region of about 35 kb surrounding these tient, presumably simultaneously with breakpoints has been cloned and the translocation. mapped. The CD3G YACs localized only to A cluster of CG-containing restriction chromosome 11 in normal cells. How• sites was found close to the cluster of ever, in addition to labeling the normal breakpoints on chromosome 19. These chromosome 11, the Y ACs were split in "CpG islands" are usually associated cells with the four translocations; thus with the 5' ends of genes. The presence of one portion remained on chromosome 11 a CpG island adjacent to the cluster of and the other was translocated to the t(14; 19) breakpoints was confirmed by other chromosome. There was no label• sequencing. Probes from this region were ing of any other chromosomes in these used in Northern blot experiments, which cells. Thus the breakpoint in these trans• detected a 2.1-2.3 kb transcript in many locations, which occur in both lympho• hematopoietic cell lines. S 1 protection blastic and myeloid leukemia, is within experiments confirmed this result and the same 320 kb region of human DNA. showed that transcription occurred in a We have no evidence, at present, of direction away from the breakpoint to• whether the break involves the same ward the telomere. segment in the different translocations. The BCL3 cDNA was cloned and Thus the use of Y AC clones provides a sequenced [18]. A basic protein of 446 new strategy for screening large pieces of amino acids and a molecular weight of DNA and for focusing intensive mole• 46741 is predicted, which shows a remar• cular analysis only on the segment that is kable structure. The N-terminus is high• shown cytogenetically to be of interest. ly enriched in proline (25 %), and the C• YACs will also be of great benefit in terminus in proline (23 %) and serine defining the genetic boundaries of (28 %). Almost the entire remainder of chromosome deletions. These probes also the protein (about half) is made up of provide powerful tools for detecting seven tandem repeats of 33-37 amino these same rearrangements in interphase acids. Comparison with proteins in the cells. available data bases showed significant Using more conventional techniques, a homology to the Drosophila Notch pro• colleague of mine, Dr. Timothy McKei• tein. The homology is in the region of the than, has cloned the translocation break• repeats. Notch has six repeats with clear point found in some patients with B similarity to the repeat in BCL3. These cell chronic lymphatic leukemia [17]. The repeats have been found in three ad• translocation involves the immunoglo• ditional proteins - namely, lin-12 of bulin heavy chain locus (IGH) located at Caenorhabditis elegans (six repeats), chromosome band 14q32 and a previ• cdc 10 in Schizosaccharomyces pombe ously unknown gene that we have called (two) and SW16 in Saccharomyces BCL3 on chromosome 19 (band 19q13). cerevisiae (two) [reviewed in 18]; the re• They cloned the translocation breakpoint peats are generally referred to in the from two of our patients as well as from literature as cdc10 repeats. The role of several others from material provided by this motif is not known.

XLII Total RNA from two patients with this control point, in 01, is known as chronic lymphocytic leukemia (CLL) and "start". Much more is known about the the t(14; 19) - one with a break on function of the other required gene, cdc2, chromosome 19 close to BCL3 and one encoding a protein kinase which is highly with a break more than 25 kb away - was conserved among eukaryotes and is re• hybridized on Northern blots to deter• quired both for start and for mitosis. mine the level of BCL3 expression in cells Little is known about the function of the containing the t(14; 19). The samples cdc10 protein. Recently, a specific anti• were compared with total RNA from the body to cdc10 was shown to detect a peripheral blood of two other patients protein of similar size in mammalian with CLL, as well as with three cell lines cells, suggesting that, like cdc2, the pro• derived from the prolymphocytic variant tein may be conserved throughout the of CLL; none of the cell lines contain the eukaryotes. t(4; 19). In addition, five other hema• SW16 is one of several genes known to topoietic cell lines were examined. The be required for transcription of the HO level of message in the two CLL samples gene, which encodes the endonuclease with a t(14; 19) was higher than that which initiates mating type switching in found in any other sample examined. By Saccharomyces cerevisiae. HO is acti• hybridization to blots containing various vated immediately after commitment to quantities of RNA, the two t(14; 19) start, and a particular repeated sequence samples were found to contain 5-7 times in the 5' flanking region of the gene has and 10-15 times the level of message been shown to be responsible for cell cycle present in the CLL cell line with the control of its transcription. SW16 and greatest quantity of message [18]. The SW14 (whose sequence has not yet been message present in the cells with the reported) are the only genes known to be t(14; 19) was identical in size to that specific for this control element. These present in normal hematopoietic cells, as two genes appear to be at least partially would be expected from the fact that the interchangeable since neither single mu• translocation breakpoints occur up• tation is lethal, but double mutations stream of the transcription start site. The are nonviable. While the function of apparent normality of the message sug• SW16 strongly suggests that it is a nu• gests that the increased message level clear trans-activating protein, it has not results from increased transcription and been directly shown to interact with DNA not from an increased message stability or even to be a nuclear protein. arising from changes in the structure of If the cdc 10 motif is involved in the transcript itself. protein-protein interactions, there may The known functions of the other be little commonality in function between proteins containing the cdc 10 motif may the two yeast proteins and the two inver• offer a clue to the function of BCL3; tebrate proteins. Nevertheless, there are a unfortunately, however, the divergent few plausible models in which the pro• structure and function of these proteins teins could have related functions. For makes it difficult to image a common role example, BCL3, cdc10, and SW16 may for the motif. Notch (in Drosophila) and be peripheral membrane proteins which lin-12 (in the nematode Caenorhabditis) interact with the cytoplasmic domains of are transmembrane proteins involved in transmembrane proteins and are invol• cell lineage determination. On the other ved in signal transduction. According to hand, the two yeast proteins are not this interpretation, the ancestor to the transmembrane proteins and they share linl2 and Notch genes could have resulted functions involved in control of the cell from the fusion of two genes in evo• cycle. cdc 10 is one of two genes in Schizo• lution - one encoding a transmembrane saccharomyces pombe known to be re• protein, and the other, a cdc 10-related quired for commitment to the cell cycle; protein.

XLIII The increased levels of BCL3 message answers regarding the function of these following mitogenic stimulation and the genes in normal cells; ~ow are they altered homology of the gene to cell cycle control by the chromosome rearrangements, and genes suggest that abnormally large how does this relate to malignant trans• quantities of the protein present in CLLs formation? The questions are endless. with the t(14; 19) may lead to an in• The answers will provide insights into cell creased proliferative rate in these cells. biology that have very profound impli• This superficially seems inconsistent with cations. Within the next decade or two, the very low mitotic rate of CLL cells. we should be able to define the major Perhaps this mitotic rate, while low, is genetic abnormalities in many types of nevertheless greater than that of normal cancer and to identify the specific changes CD 5 + B lymphocytes. Alternatively, a in the tumor cells of many patients. For subpopulation of CLL cells, perhaps most leukemias, lymphomas, and sar• those present in pseudofollicular growth comas, unique chromosome changes are centers in lymph nodes, may show an often associated with a particular subtype abnormally high rate of proliferation. of these neoplasms. One of the major reasons to concen• Cloning of the genes involved in these trate on cloning the genes involved in chromosome changes will provide spec• rearrangements is that the consistent ific DNA markers that will have diagnos• chromosome changes pinpoint the loca• tic importance. For some solid tumors, tion of the genes whose functions are on the other hand, current evidence sug• critical in the growth potential of that cell gests that deletions of the same chromo• type. The chromosome changes that we some region may occur in different types concentrate on are present in all of the of tumors, such as the deletions of 13 q in malignant cells; thus they are not random retinoblastoma, osteosarcoma (not sec• events affecting one or a few cells in the ondary to radiation for retinoblastoma), involved tissue. Moreover, they are breast cancer, and lung cancer. The dele• clonal in origin and are derived from a tion of the same region does not neces• single cell in which the intial chromosome sarily imply that the same gene is involved change occurred. These changes are som• or that the change within the gene is atic mutations in individuals who other• identical, witness the fact that two differ• wise virtually always have a normal ky• ent translocations in band 22q 11 involve rotype in their uninvolved cells. These different genes, namely, the lambda light observations provide the evidence that chain gene in the 8; 22 translocation in cancer is a genetic disease. This notion Burkitt's lymphoma and the BCR gene in seems self-evident today, but it was not chronic myelocytic leukemia (CML); fur• generally accepted several decades ago thermore, the breakpoints within BCR in when many of us began working in PhI-positive leukemia are also somewhat cytogenetics. Clearly, I am using "gene• variable. tic" in a special way, referring to changes The multistep process of malignant in genes within the affected cell, not in the transformation is complex. In the leuke• more usual sense of a constitutional gene• mias and lymphomas, we often see spec• tic disease such as hemophilia or color ific chromosome translocations com• blindness. bined with loss or gain of particular I will conclude with some comments chromosome segments. Some combina• regarding the longer-term potential im• tion of alterations in dominantly acting pact of. discovering new genes via proto-oncogenes and in recessively acting chromosome rearrangements. Once these tumor-suppressor genes certainly act genes are identified, many previously synergistically to enhance the malignant unknown, BCR or BCL3 for example, phenotype. they become the focus of very active In the future, the precise definition of investigation. Scientists try to find the genetic changes in the malignant cells

XLIV of a patient will be used to select the most ation, I must understand how its ho• appropriate treatment for cells with these mology to portions of the cdeJO and genetic defects. This treatment will be less Notch genes might provide clues as to its toxic for the normal cells in the patient. functions in both normal and malignant Moreover, this genetic profile may allow cells. As more genes involved in transloc• monitoring of the patient's course and ations and deletions are defined, many of early detection of relapse. These same us in cancer research will continue to have genetic markers may be used to detect the to "go back to school" to be able to involvement of other tissues such as bone incorporate the knowledge provided by marrow, spleen, or lymph nodes. These molecular geneticists and cell biologists changes in treatment strategies will clear• into our concepts of carcinogenesis. The ly benefit the patient. Of more general golden age in biology and in medicine are scientific importance, however, will be nourishing one another as never before in the identification of dozens of genes, history. many hitherto unknown, that can be used to study the complex process of the Acknowledgements. My colleagues, Drs. regulation of cell growth and differenti• Manuel Diaz, Michelle Le Beau, and ation. This development may be the most Timothy McKeithan, have been generous significant result of our success in under• with their review of this manuscript. I standing the genetic changes that occur in acknowledge the expert secretarial as• cancer cells. sistance of Ms Felecia Stokes. The re• I would like to conclude with a more search described has been supported by personal note based on my continuing the Department of Energy Contract No. amazement at the interrelatedness of the DE-FG02-86ER60408 and by United biomedical sciences. This should be no States Public Health Service Grant CA surprise to me, but it is. Many inves• 42557. tigators have found that a successful system for carrying out some function in primitive organisms has evolved and then this system is used repeatedly with vary• References ing modifications as the organisms become more complex. As a cytogeneti• 1. Rowley JD (1989) Principles of molecular cist, I had to learn something about the cell biology of cancer: chromosomal ab• cell cycle and DNA replication, about normalities. In: de Vita VT, Hellman S, chromosome structure, and about vari• Rosenberg SA (eds) Cancer: principles ous agents that can alter both of these. and practice of oncology, 3rd edn. Lippin• More recently, I have had to become an cott, Philadelphia, pp 81-97 amateur tumor virologist at least with 2. Heim S, Mitelman F (1987) Cancer regard to the action of viral oncogenes cytogenetics. Liss, New York 3. Mitelman F (1988) Catalog of chromo• and their cellular counterparts, the proto• some aberrations in cancer. Liss, New oncogenes. With the cloning of transloc• York ations, especially some of the recent ones, 4. Trent JM, Kaneko Y, Mitelman F (1989) a knowledgeable cancer cytogeneticist Human gene mapping 10: report of the must understand cell cycle control genes Committee on Structural Chromosome in yeast (cdeJO and SWI4 and 6) and Changes in Neoplasia. Cytogenet Cell developmentally regulated genes in Genet 51:533-562 Drosophila (Notch for example) and 5. Bishop JM (1987) The molecular genetics of cancer. Science 235:305-311 nematodes (lin-IO, glp-I). I have already 6. Rowley JD (1973) A new consistent described in some detail, the cloning of chromosomal abnormality in chronic the BeL3 gene by McKeithan et al. [17, myelogenous leukemia identified by quin• 18]. If I am to understand the possible icrine fluorescence and Giemsa staining. roles this gene plays in B cell transform- Nature 243:290-293

XLV 7. Francke U (1976) Retinoblastoma and in tumour pathogenesis. Trends Genet chromosome 13. Cytogenet Cell Genet 4:300-304 16:131-134 14. Rowley JD, Diaz MO, Espinosa R (1990) 8. Friend SH, Bernards R, Rogel S, et al. Mapping chromosome band 11 q 23 in (1986) A human DNA segment with pro• human acute leukemia: identification of perties of the gene that predisposes to 11 q 23 breakpoints with a yeast artificial retinoblastoma. Nature 323:643-646 chromosome. Proc Nat! Acad Sci USA 9. Fearon ER, Cho KR, Nigro JM, et al. 87:9358-9362 (1990) Identification of a chromosome 15. Evans GA, Lewis K, Rothenberg BE 18 q gene that is altered in colorectal (1989) High efficiency vectors for cosmic cancers. Science 247:49-56 microcloning and genomic analysis. Gene 10. Mulleris M, Salmon RJ, Zafrani B, et al. 79:9-20 (1985) Consistent deficiencies of chromo• 16. Lichter P, Tang C-JC, Call K, et al. (1990) some 18 and of the short arm of chromo• High-resolution mapping of human some 17 in eleven cases of human large chromosome 11 by in situ hybridization bowel cancer: a possible recessive deter• with cosmid clones. Science 247:64-69 minism. Ann Genet (Paris) 28:206-213 17. McKeithan TW, Rowley JD, Shows TB, 11. Leder P, Battey J, Lenoir G, et al. (1983) Diaz MO (1987) Cloning of the chromo• Translocations among antibody genes in some translocation breakpoint junction of human cancer. Science 222:765-771 the t(14; 19) in chronic lymphocytic leuke• 12. Croce CM, Nowell PC (1986) Molecular mia. Proc Nat! Acad Sci 84:9257-9260 genetics of human B cell neoplasia. Adv 18. Ohno H, Takimoto G, McKeithan TW ImmunoI38:245-274 (1990) The candidate proto-oncogene bcl- 13. Rabbitts TH, Boehm T, Mengle-Gaw L 3 is related to genes implicated in cell (1988) Chromosomal abnormalities in lineage determination and cell cycle con• lymphoid tumours; mechanisms and role trol. Cell 60: 991-997

XLVI Yuri Ovchinnikov Lecture Human Retroviruses: Linkage to Leukemia and AIDS

R. C. Gallo 1 and L. S. Nerurkar 1

Introduction and Background public health measures. In retrospect, we should have remembered that the last This review will discuss interactions of great pandemic that affected the United retroviruses with the cells of the hema• States, Europe and the world was only topoietic system. Such interactions have about 70 years ago. It was the great been studied in the past as tools to insert influenza epidemic of 1918-1920. And if genes in the cells to study their regulation one reviews the history of microbiology, or to study cellular and molecular basis of there were often periods where epidemics transformation in vitro. The emphasis of disappeared and mysteriously reap• this review will be on viruses which cause peared after more than 60 or 70 years, or diseases, particularly in man. even for 100 or 200 years. Perhaps we A decade ago there was no general were overconfident in thinking that acceptance of the concept that genes were epidemics belonged to the past: an critical to leukemias and lymphomas or epidemic or pandemic of the acquired to disorders of hematopoietic cells. In a immunodeficiency syndrome (AIDS) as somewhat analogous way there was also now been with us for a decade. There was a general feeling that viruses did not cause also a feeling that pandemic diseases were human cancers, and that retroviruses, in not possible unless the causative agents or particular, did not exist in human beings. the microbes were casually transmissible. We now know that viruses, either directly We now know that we have a pandemic of or indirectly, either as a cofactor or as a AIDS and the agent is not casually trans• direct cause, playa role in more than 40 % missible, but transmissible only by close of human cancers. We have also learned contact and with the exchange of body that human retroviruses do exist and in fluids. mUltiple types. The failure to appreciate the coming of During the 1970s, there was also a these events was probably because of the feeling in the United States that serious or failure to remember some of the lessons fatal, epidemic or pandemic diseases were of past medical history; that often there things of the past. Infectious diseases that are major changes in diseases following would become global epidemics were no some major changes in the society. The longer a problem for the so-called "in• major changes in the post-World War II dustrialized nations." Such diseases were era were: a great increase in air travel; the really a problem for the less-privileged use of blood and blood products, often nations. We had preventive and curative going from one nation to another; the measures like vaccines and antibiotics in insane habit of intravenous drug abuse; addition to better sanitary conditions and and the increase in sexual contacts. All these things made it possible to transmit something that was remote or rare so that

1 Laboratory of Tumor Cell Biology, Na• it become relatively common and global. tional Cancer Institute, National Institutes Comparing the epidemics of the past, the of Health, Bethesda, MD 20892, USA. AIDS epidemic is not particularly novel,

Haematology and Blood Transfusion Vol. 35 XLVII Modern Trends in Human Leukemia IX R. Neth et al. (Eds.) © Springer-Verlag Berlin Heidelberg 1992 nor is the response to the epidemic by the virus type 1 (HIV-1) or AIDS virus public as is often portrayed by the media. causing the epidem\c we now face was The only novel feature of this epidemic is found in 1983 by Barre-Sinoussi et al. at the nature of the microbes that are caus• the Institute Pasteur [4] and established ing the epidemic. The novel properties of by our laboratory as the cause of AIDS in these microbes are: they are newly dis• early 1984 with many isolations of the covered (but are not new); they are mi• virus and the development of the blood crobes that are often difficult to find test. The related virus from West Africa, because they do not replicate much or do called HIV -2, is neither as pathogenic as not infect many target cells; they are only HIV-1 nor is it spreading like HIV-1. It produced by the infected cells, primarily appears to be almost limited to West during the proliferating phase of the cell's Africa. life cycle. In fact, this is true for all human The technology developed in the 1970s, retroviruses. particularly, the sensitive assays for re• For the same reasons that they are verse transcriptase (RT), was crucial for difficult to find, the viruses are difficult to the discovery of human retroviruses. The transmit. Almost always, they have a very discovery of RT, by Temin [5] and inde• long latency period. This is an important pendently by Baltimore [6], was quickly characteristic and has allowed their trans• extended by the finding of a similar mission to become global as they are enzyme in human leukemic cells from present in the host from the time of unusual cases by Gallo and his colleagues infection until death (lifelong infection) during the 1970s [7]. Enzymes from at and during that period can be transmitted least four or five patients which had the to others. This is a major difference properties of the viral enzyme were between retroviruses and other viruses, partially purified. More important was which we tend to think are transmitted the development of synthetic template while a person is sick or in the early phase primers, e.g., synthetic polymers (oligo• of incubation, which can be a few days, a dT-poly-A and oligo-dG-poly-C) that few weeks, or at the most a few months. made the assays for these enzymes spe• The very long latency period of retrovir• cific and sensitive. This improved the uses means that it may be several years or detection of retroviruses several-fold several decades from the time of infection compared to the electron microscopic before the first manifestation of disease method used for decades. Also, the assay will be noted. They often cause serious using R T is much simpler and cheaper diseases, e.g., central nervous system dis• and can be done continuously while the ease, malignancies, and immune defi• culture is ongoing. Electron microscopy ciency. These agents have thus become does not offer that possibility. Retrovir• increasingly important because of the uses, including those affecting humans, serious and often fatal consequence of complete their replication cycle much their infection. more efficiently during the proliferating phase of that cell's life cycle. R T assays performed continually on the cells in Discovery of Human Retroviruses . culture can reveal shorHerm viral repli• cation which otherwise may be missed There are four human retroviruses well by electron microscopic techniques. characterized by now [1-3]. Human T• The second important technology was lymphotropic (leukemia) virus type I the ability to grow human T cells, parti• (HTLV-I) was found by Gallo and cularly with interleukin-2 (IL-2), dis• coworkers in the late 1970s and first covered by Morgan, Ruscetti and Gallo reported in 1980. Its relative, HTLV-II, in 1976 [8]. Developments in the field of was also found in our laboratory a year or immunology such as monoclonal anti• two later. The human immunodeficiency bodies have allowed defining subsets of

XLVIII lymphocytes by surface markers or by opment of a malignancy. More recently, other assays to understand different func• we have other animal models, particular• tions of T-cell subtypes. A third factor ly the simian models, in which lenti• contributing to the discovery of human retroviruses have been isolated that are retroviruses is the fact that they spread closer to the humans [10]. globally in the 1960s and the 1970s, and became much more common. We believe Morphology. The size and shape of that this may have increased the chances HTLVs and HIVs are roughly the same. of detecting and isolating them consider• However, the core structure of the leuke• ably. And the last point which is worth mia viruses is much different from that of mentioning was the perseverance in look• the AIDS virus. The latter is much more ing for them, even though most scien-. condensed and cylindrical in shape com• tists did not think they existed. pared to that of the leukemia viruses (see Fig. 1).

Classification of Human Retroviruses Biological Properties. The HTLVs and HIVs show many parallels in their biolog• Human retroviruses belong to two en• ical characteristics. Both viruses infect tirely different subclasses which differ in CD4 + T lymphocytes but they vary in the their morphology, some aspects of their consequence of infection. The overall genomic organization, and some aspects effect depends on the extent of virus of their biology. The HTLV s belong to replication and on the functions of some the more classic type of animal retrovir• of the genes the virus carries. HIV-1 uses known in most species as type C or infection kills the CD4 + T cells. HIV-2 oncorna retroviruses, whereas the HIVs essentially behaves in a similar manner. belong to the category known as lenti• On the other hand, HTLV-I and HTLV• retroviruses. "Lenti" is not an accurate II, like most animal retroviruses, have no term, as it means slow. HIV does not lytic activity on their target cells, but can replicate slowly compared to HTLVs. alter the function of that cell. Some HTLVs are much more slowly replicating infected T cells become immortalized in viruses, and thus the class names can be vitro and may contribute in the same misnomers. Until the discovery of HIVs, manner to the development of leukemia lenti-retroviruses were only known to in vivo. Both classes of viruses remain occur in ungulates, the hoofed animals latent in the patient or in the cell for their like horses, sheep, cows, and goats [9]. lifetime. Another feature both HIVs and One has to be careful in not drawing too HTLVs have in common is the tight much of an analogy between HIV and control of the DNA-integrated provirus. these ungulate lenti-retroviruses. How• Following the infection of a CD4 + T cell ever, there are some common character• by these viruses there is integration of istics, e.g., they infect cells of the macro• their DNA forms into the host chromo• phage lineage and morphologically their somes, but the DNA forms do not induce core structures appear similar. But there expression of RNA or proteins. So an are major differences in other aspects. infected cell will have no viral RNA or For example, some of the ungulate lenti• viral protein immediately after infection. retroviruses can be transmitted casually. This means that the immune system can• The visna virus of sheep is thought to be not find the infected cells. This is one way transmitted by fomites in crowded sheep these viruses escape the immune system. that are herded together in a closed Other mechanisms which allow these vir• environment. None of the ungulate lenti• uses to escape are, for example, by infect• retroviruses target CD4 + T lymphocytes ing the brain and by undergoing consider• and they are not known to be associated able genomic variation from isolate to with the increased frequency of the devel- isolate (particularly in the case of HIV).

IL cells infects other cells. In this manner, the HIV -infected host who has other chronic infections is more likely to spread this virus.

Modes of Transmission. The HIVs and the HTLVs have common modes of trans• mission. They are transmitted by blood or sex and from mother to child. For HTLV-I, males chiefly get infected from their mothers, and women chiefly get infected from their male sexual contacts. The mode of transmission from mother to child is in utero, transplacentaly as well as by milk or in the actual birth HTLV-I HTLV-II process. Blood transfusions and the use of blood products are, of course, also modes of transmission of both viruses, the major target cell being the CD 4 + T cell [11].

Analogous Animal Retroviruses. The clos• est relative of the human leukemia vir• uses, called simian T-cell leukemia/lym• phoma viruses, are found in African monkeys [12, 13]. They are not found at all in New World monkeys, i.e., in mon• keys from the North American continent and those from Asia [14]. But the viruses in African monkeys are closer to the human viruses than to those from Asian HIV-I monkeys. Similarly, the closest relative of HIV-2 HIVs are also found in African monkeys. Fig. 1. Morphological structures of HTLV-I, No relatives of HIV have been found in HTLV-II, HIV-1, and HIV-2. The top panels New World or Asian monkeys. Because represent the budding from the cell mem• of the fact that the closest relatives ofall branes, the bottom panels show the cross section of the mature virons human retroviruses are present in African primates, the ancestral origin of these viruses is almost certainly African. That does not mean that the recent epidemic of In addition, HIV destroys the cells of the AIDS came from Africa. As far as one immune system which are crucial in the can tell, the epidemic of AIDS began immune surveillance itself, thereby escap• almost simultaneously in parts of Central ing the immune attack. Africa, some of the Caribbean islands, When the T cells are immune-sti• particularly Haiti, and the United States, mulated, perhaps by another infection, and perhaps in Europe. the viral genes become active along with a variety of other cellular genes and viral proteins are expressed on the cell surface. Geographical Distribution of HTLV-I This allows the immune system to see the infected cell. Such immune clearance may HTLV-I transmission is extremely tightly be too late. The virus released from such controlled and if one did not have a

L handle on the virus (virus isolation or For example, we do not know very much virus detection using probes), the diseases about infection by HTLV-I in the Soviet it causes, e.g., leukemia or neurological Union. diseases, could be mistakenly thought to be genetically inherited. HTLV-I is en• demic in Subsaharan Africa. It is not Nature of the Diseases present in all parts of Subsaharan Africa, Caused by HTLV-I but seems to be restricted to certain tribes or geographical areas and is not casually Leukemia. The picture of the first patient transmitted. HTLV-I is also endemic in from whom a retrovirus was isolated is the Caribbean basin, including the north• given in Fig. 2. This patient was a young ern part of South America, Central black male and came from the southeast• America, most of the Caribbean islands, ern part of the United States. He had no and parts of the southeastern United interesting past history, either medical, States. Some Caribbean islands do not familial, or occupational. He developed a have any HTLV -I. It depends on where in severe acute T-cell malignancy of the Africa the island inhabitants have their CD4 + T lymphocytes. The skin manifes• origin. If the ancestral tribe is positive, tation in this disease is due to infiltrates then the descendants in some Caribbean of leukemic cells in the skin, which is a islands are positive. Similarly, if the common feature in this disease [15, 16]. ancestral tribe is negative, then the de• Frequently, there is high blood calcium, scendants in another Caribbean island which can lead to death of the individual. are negative for the most part. HTLV-I is Liberation of some lymphokines is sug• also endemic in the southern islands of gested as a possible molecular mechanism Japan in Shikoku, Kyushu, Okinawa, for high blood calcium [17]. There is also and other neighboring islands. Seroepi• an increased incidence of opportunistic demiologic studies have suggested that infections and slight immune impairment clusters ofHTLV-I, or a virus like it, are can be observed in infected people. How• observed in some villages in Spain and in ever, when a disease begins to develop, southeastern Italy in a region called the course is very rapid. It resembles the Apulia. Manzari of Rome and Varnier of chronic myelogenous leukemia going Genoa believe that the virus in Apulia in into blast crisis. Death usually follows in southeast Italy is endemic. It is not a less than 6 months. coastal introduction from outside in re• These manifestations of the disease are cent times; rather, it is found in the peo• common, occurring in about 70 %- ple living in the interior hills. Recent mo• 80 % of people who have leukemia with lecular analysis studies of some of the HTLV-1. Another 20%-30% show a isolates from that region indicate that it is more chronic course, and the diagnosis is not the classic HTLV-I, but may be of chronic lymphocytic leukemia of a another retrovirus related to HTLV-I. CD4 + T-cell type, mixed cell lymphoma, There have been clusters of HTLV-I• or histiocytic lymphoma of a CD4+ T• related leukemia reported in Amsterdam cell type. So, in any CD4 + T-cell malig• and London in migrating populations nancy, one has to consider the possibil• from the Caribbean. The rate of develop• ity of HTLV-I, and particularly if the dis• ing leukemia after HTLV-I infection is ease is as aggressive as described above. identical in populations which have mi• grated and in the nonmigrating popul• Neurologic Disease. It is now known that ation, indicating that no other environ• HTLV-I also causes a serious and fatal mental factor is needed for the cause of progressive neurologic disease which is leukemia, at least as far as the epidemi• similar to multiple sclerosis but can be ology can determine [11]. For a great distinguished from it. There is some part of the world, we have very little data. confusion in this area because some labo-

LI Fig. 2. The first patient with T-cell leukemia caused by HTLV-I

LII ratories have reported HTLV-lor a close• Table 1. Diseases caused by or associated ly related virus as being involved in with human retroviruses multiple sclerosis itself. The data are not 1. Adult T-cell leukemia (ATL) consistent from laboratory to laboratory. 2. Occasional other T 4 leukemias( More evidence is required to implicate lymphomas HTLV-I or a relative of HTLV-I in 3. Tropical spastic paraparesis (TSP) playing a definitive role in multiple or HTLV-associated myeloneuropathy sclerosis. However, the neurologic dis• (HAM) ease that has been called tropical spastic 4. Mild immune impairment paraparesis or Jamaican neuropathy, and 5. Polymyositis sometimes misdiagnosed as multiple 6. Rheumatoid arthritis-like disease (?) sclerosis or a variant of multiple sclerosis, 7. Retinitis (?) is certainly linked to HTLV-I [18], 8. B-ce1llymphocytic leukemia (B-CLL), indirect (?) although the disease mechanism is not 9. AIDS progression, possible role as understood. The HTLV-I-associated dis• co factors ease differs from multiple sclerosis in that 10. Guillain-Barre syndrome it does not have exacerbations and re• 11. Chronic lung disease missions like multiple sclerosis: it is pro• 12. M-proteinamia gressive. It is characterized by incon• 13. Chronic renal failure tinence of the bladder, impotency in males, loss of bowel function and spastic• ity of the lower extremities. The disease can occur rapidly after infection with against tax and env gene products [20]. HTLV -I. It appears to depend on the This has led to the speCUlation that the dose of the virus. immune response to the virus produces There is a recent report of a Frenchman an autoimmune disease. Recent reports who received a transfusion with HTLV -1- indicate that HTLV-I is also involved or positive blood, developed the neurologic linked epidemiologically to other diseases disease in 5 weeks, and transmitted the listed in Table 1. virus to his wife during that period. This implies that all of the blood supply should Genome of HTLVs. Like any retrovirus, be tested for HTLV -1 as well as for HIV HTLV-I has long terminal repeat se• [19]. However, the neurologic disease quences at each end. These sets of nucleo• could take many years to develop and tides are involved in regulation of the there is some indication that genetic fac• viral gene expression and form sites of tors are important. There are some re• covalent attachment to cellular sequences ports from Japan showing an HLA class 2 on each side of the integrated provirus. association and that certain patterns have Like any animal retrovirus, it has the an increased frequency of developing the three genes for structural proteins, the neurologic disease. A known fact is that gag gene for the viral core proteins, the the virus is not found in the central pol gene for the enzymes, including RT, nervous system tissues, e.g., brain cells or and the env gene for the envelope (Fig. 3). cells of the spinal cord, but only in the These give the retrovirus the ability to cerebrospinal fluid. reproduce itself. The other known facts are that people When the molecular analyses of who develop the neurologic disease have HTLV -1 and HTLV -II were completed, it a very high titer of antibody, much higher became evident that they have new genes than the healthy carriers or the leukemic present at the 3' end of the genome. patients. Even more interesting are the Originally, one of the genes was called tat, recent results of Jacobson, McFarlin, and but with the revised terminology, this their coworkers, who describe high levels gene is now called tax. The tax makes a of cytotoxic T lymphocytes reactive 40000-dalton protein localized in the

LIII / rex ~ LTR gag pol LTR HTLV-I,II • env • --- I-tax-_-

• /r~ I • vif .v~vpu LTR gag _g/-tat_1 LTR HIV-1 _ ••••.,I. ••• P.O~I.... I-rev-._ .vpr env -nef ___ rev L TR gag vif vpr 1-- --... nef HIV-2 _.IIIIIi"'~ __ .PO~I __':_ .. ---tat_._ vpx env LTR Fig. 3. Genomic structures of human retroviruses . - nucleus or the infected cells. The rex is the ment, fusion of the viral envelope with the second gene in the 3' region of HTLV-I cell membrane occurs, followed by emp• and HTLV-II. These genes are coded tying of the viral core into the cytoplasm from two segments of the genome and are of the cell. The viral RNA is transcribed products of doubly spliced messenger in the cytoplasm, with formation of the RNAs. This phenomenon (double or double-stranded linear DNA, which then even triple splicing) was new in human enters the nucleus and integrates into the retrovirology. It was soon realized that chromosomal DNA. the protein products of these genes are With most animal retroviruses, after absolutely essential for the replication of provirus integration into a permissive HTLV-I and HTLV-II. They are also target cell, virus replication and its ex• essential for the biological activity of pression start immediately. There are these viruses. sufficient cellular factors, and sufficient The products of the gag, pol, and env viral and cellular machinery allowing genes are formed from unspliced or singly quick transcription of the DNA provirus, spliced messenger RNA molecules. This to reform viral RNA in the nucleus. The is similar to what was known among viral RNA then traverses to the cy• animal retroviruses. toplasm and assembles at the cell mem• brane with viral proteins that have been Replication Cycle of HTLVs. The repli• formed by translation of unspliced or cation cycle ofHTLVs can be divided in• singly spliced messenger RNAs in the to two parts (Fig. 4). The first part, like cytoplasm. The viral proteins are pro• any animal retrovirus, involves a phase cessed, particularly by cleavage through of attachment to the cell membrane. The viral and cellular proteases (the former receptors for HTLV-I or HTLV-II are for viral core proteins, the latter for vi• unknown, as for most of the animal ral envelope proteins). After assembly, retroviruses. However, the chromosomal budding and release of the newly formed site of the HTLV-I receptor has been virions completes the replication cycle of determined [21]. Following the attach- the viruses.

LIV Attachment Penetration Budding

rr:======~,~======::::;') Processing, Assembly, Maturation Protease &RNA Reverse t Transcriptase I~DNA \ , ._--AAAA\ ~-, ,-'.-' '-' RNA

Fig.4. Life cycle of HTLV-I and HTLV-II

HTLV-I and HTLV-II have intro• made are the messenger RNAs that are duced a new complexity into our under• doubly spliced, i.e., the messenger RNAs standing of the replication cycle, and that for rex and tax. But once the REX complexity relates to the events which protein is made, the formation of the take place in the nucleus. In order to have unspliced RNAs or the singly spliced successful transcription of the DNA pro• RNAs for the viral structural proteins is virus to viral RNA, first there is the favored. This is an interesting mechanism expression of an early gene product. This because once the REX protein is made, it phenomenon, although known in some down-regulates its own expression. It also DNA viruses, was newly discovered in down-regulates tax and allows the for• retroviruses. The first genes to be ex• mation of the viral proteins so that there pressed are tax and rex (Table 2). What is a sudden release of virus during this turns on the expression of tax and rex is narrow window in which these human unknown, but the tax gene product retroviruses have to complete their (TAX) is essential for the early transcrip• cycle. This mechanism is evident even in tional events to make the viral RNA. The HIV but not in the lenti-retroviruses of function of the rex gene product (REX) is animals. This may suggest a convergent not only newly observed in retrovirology, evolution of mechanisms for infection but it has introduced some new mechan• of human T cells by two entirely dif• isms into all of molecular biology. The ferent classes of human retroviruses. REX protein is involved in removal or transport of the messenger RNAs for the Mechanism of Leukemogenesis. TAX viral structural proteins, i.e., the messen• protein plays an important role in leuke• ger RNAs that are unspliced or singly mogenesis. It acts in trans and is involved spliced. In other words, in the absence of in the mechanism of transcription of viral REX, the only messenger RNAs that are RN A. TAX protein also activates cellular

LV Table 2. Accessory genes of human retroviruses Immunogenicity Size Cellular Function Replication localization competence of ( - ) mutants vif + p23 Cytoplasm/inner Infectivity ± membrane tat + p14 Nucleus/nucleolus Transcriptional and post-transcriptional activation rev + p19 Nucleus/nucleolus Expression of structural proteins, modulation of transcription nef ++ p27 Cytoplasm Negative regulator ++ vpr + pi8 Nucleus Rapid viral growth (?) ++ vpu + pi5 Cytoplasm/membrane Assembly and release (?) ++ (HIV-1) vpx + p15 Cytoplasm ? ++ (HIV-2) genes indirectly. It complexes to some stages ofleukemia. At least this is the way cellular proteins and transcriptional fac• we think about it today. The TAX protein tors that are involved in the turning on of is also involved in turning on other genes important for T-cell proliferation cellular genes, e.g., the c-fos proto• such as those for IL-2 and IL-2 receptor oncogene. The development of adult T• (lL-2R) [22, 23]. It is somewhat ironic cell leukemia (ATL) by HTLV-I is sum• that the protein (IL-2) used to grow T marized in Fig. 5. Perhaps about one• cells to isolate the virus is the very protein third ofT cells may be infected by HTLV• that the virus uses or turns on in its first I, but only a small fraction expresses the

Healthy Carrier o State 6 0 Polyclonal 8 0 Wo ~T 'cell Expansion Activation.'r ~ '\. ~~_~L-2R, T:X ~* 2nd Event@88 o o~~ %1 ~ O@ ~07o--Q-~~-~ -*88 9.\ _ 6~' e.r JE $ ~y Expression 0 8 ~UnOIOgiCal ATL o Clearance 8 HTlVMI infected cell ~ IL-2R o IL-2 * TAX Fig. 5. Development of adult T -cell leukemia ,. HTLV-I env protein

LVI virus. The immune system cannot see the others [25] that feline leukemia virus can cells which do not express the virus and be transmitted horizontally and cause cannot attack them. immune deficiency as well as leukemia. At some stage, the tax gene is turned Essex, in his epidemiologic studies in the on. What exactly leads to the turning on early 1980s, highlighted the greater im• of tax is unknown, but once it occurs portance of this feline virus in immune genes for other viral proteins can be suppression than in causing leukemia, turned on. The tax gene also turns on the whereas Gallo suspected from the expe• IL-2 and IL-2R genes. The IL-2R has a riences with HTLVs a possible involve• complex structure and is made of differ• ment of a retrovirus in AIDS. These ent polypeptides. The high-affinity poly• experiences were: studies of HTLV-I peptide ofIL-2R that binds best to IL-2 is epidemiology showed that the AIDS activated by tax. virus was, like HTLV-I, endemic in Cen• This may lead to autocrine and para• tral Africa; the causative agent, like the crine phenomena allowing polyclonal T• HTLVs, targeted CD4 + T cells; the cell expansion. At this stage it is not a modes of transmission by sex, blood, and malignancy but it can be documented in the maternal/fetal route were similar; many people infected by HTLV -I. The AIDS was associated with immunosup• immune system attacks and clears the pression and the HLTVs can be immune proliferating cells expressing viral pro• suppressive (although modestly); HTLV• teins. The cycles of appearance and II had just been discovered, providing clearance of virus-expressing cells occur impetus to the idea of there being more for years and maybe for decades. The human retroviruses. All these things led virus continues to increase the expansion to thoughts that a new human retrovirus of proliferating T cells. As estimated existed perhaps derived from a mutation recently in Japan, 3%-5% of the in• or a recombinant change in an HTLV-I fected individuals will be able to develop emerging from Africa, moving to Haiti monoclonal expansion of a T cell within and then to the United States. This was their lifetime, most likely mediated by the notion that led people, ourselves and another as yet unknown genetic event. scientists in Paris, to look for a new This event could be an accident, a muta• retrovirus. However, ironically, we soon tion, or a rearrangement, but appears to learned that, though AIDS is caused by a be a chance event ultimately leading to retrovirus, the virus is not a variant of true leukemia. A third genetic event HTLV-I or a recombinant with HTLV-I, which leads to the blast crisis may be but is due to a different category of necessary, analogous to chronic myelo• human retrovirus(es) that simply has genous leukemia. There is no complete (have) many properties in common, agreement on specific chromosomal although with a much different genomic changes to account for the second or the organization as well as classification. third event. There are some that are There are several components of the common, but not consistent. overall pathogenesis of AIDS, the major one being the immune deficiency with mv-1 and AIDS opportunistic infections. Because of the lifestyle of the individuals there is an The idea that AIDS might be caused by a increased incidence of infection with real CD 4 + T-cell lymphotropic retrovirus pathogens which include mycobacteria, came from discussions between R. Gallo herpesviruses, HTLVs, and hepatitis and and M. Essex and his colleagues in Bos• papilloma viruses. In addition, there is ton who had worked on feline leukemia infection of the brain in 40 % - 50 % of virus. Discovered in the 1960s by infected people. Subsequent to infection W. Jarrett et al. [24] in Scotland, it was of the brain, there is a thinking disorder shown by W. Jarrett, O. Jarrett, and and some acute psychosis. The develop-

LVII ment of two types of tumors is very HTLV-II and are absolutely essential to common (Kaposi's sarcoma and B-cell the replication of HIVs and probably lymphoma) and must be thought of as critical to the biological ability ofHIVs to involving mechanisms distinct from the cause AIDS or other manifestations. other manifestation of AIDS. Many other genes are discovered in the genomes of the AIDS viruses: vi!, which is Immunodeficiency. The essence of the essential for cell-free infection by these AIDS problem is immune suppression viruses; vpr, which was recently found in and immune deficiency. Part of the envel• our laboratory to be essential for infec• ope of HI V, the gp 120 molecule, interacts tion of primary human macro phages, but with the CD4 molecule. This interaction not T cells; vpu, whose function is not yet has been described as being much tighter well known; nef, which is controversial as and with much greater affinity than many to whether it does nothing to virus repli• antigen-antibody interactions. cation or slightly down-regulates it; and The CD4 molecule is expressed on the at least two more genes discovered in surface of cells that are important for the the last year or so, particularly by Hasel• immune system, including T helper lym• tine and his colleagues, whose functions phocytes, peripheral blood monocytes are not yet well understood (Table 2). [26, 27], and cells of the macrophage The early steps of the life cycle of HIVs lineage such as microglial cells of the are the same as those of animal retrovir• brain [28], Langerhans cells of skin [29], uses (Fig. 6) and involve attachment and which are widely distributed in the body, penetration of the virus into the target and the follicular dendritic cells of the cell. It is known that the CD4 molecule is germinal center of the lymph nodes; this the receptor or at least part of the re• allows the AIDS virus, immediately upon ceptor for HIV. Once HIV penetrates the infection, to alter the most pivotal cells of CD4 + Tcell, RNA to DNA transcription the immune system. An idea to use the and DNA integration into the cell follow. soluble CD4 in therapy of people who are Actually, it is not established whether infected has already been launched and HIV provirus integrates into macro• animal systems are being investigated for phage. Like HTLV-I and HTLV-II, fol• that purpose. To use CD4 as a molecular lowing integration, there is a silent or decoy to bind virus before it finds CD4 latent period even for HIVs. After T-cell on the cell surface seems to be the most activation due to any stimuli, expression rational approach to the therapy of this of the DNA provirus takes place to form disease. Unexpectedly, however, CD4 is viral RNA and viral proteins. The pro• rapidly excreted by humans, and so the ducts of the tat and rev genes have the results have been extremely disappoint• same kind offunctional corollary as those ing. Much research in the United States for tax and rex of HTLVs. They act as and Europe is focused on modification of regulatory switches in the replication CD4; for example, Genentech's use of cycle. In the long silent period after immunoglobulin attached to the CD4 infection nothing happens to the cell if molecule seems to prolong the half-life of virus is not expressed. But if that cell is CD4, diminishing its rate of excretion. immune-stimulated the replication cycle is completed. The virus comes out in a Genomes of HIVs. The genomes of HIV-1 burst, and the cell dies. So, contrary to the and HIV -2 are significantly more com• notions of some, that retroviruses can not plex than that of HTLV-I or HTLV-II be cytopathic and cytolytic, HIV is cer• (Fig. 3). In addition to the three genes tainly cytopathic. Actually, the earlier that all retroviruses have, gag, pol, and work of Howard Temin with avian retro• env, HIVs have two regulatory genes viruses also showed the cytopathic and called tat and rev which are analogous to even cytolytic nature of some of those the tax and rex genes of HTLV-I and viruses.

LVIII Attachment Penetration Budding vi' vpu

Processing, Assembly, Maturation 89-* RNA Protease Reverse ~t_ Transcriptase t~DNA .. 1 __ ., ,~. .__--AAAA\

Fig. 6. Life cycle of HIV-1

Mechanisms of mv Pathogenesis regulation of IL-2 expression in unin• fected T cells when this protein binds to Role of HIV and mv Proteins. The CD4 molecules. There are still other question is often asked "Why do the indirect mechanisms that could permit CD4 + T cells become depleted in the depletion of T cells (see Table 3). AIDS?" HIV may be involved in direct killing of infected T cells. HIV also has Escape from the Immune System. The the capacity to form multinucleated giant infection of the macrophages by HIV cells. When the virus is forming, the shows a very unique feature, namely, envelope protein gp 120 is on the cell infectious whole virus in vesicles inside surface. If there are uninfected cells the cytoplasm of the cell [26]. This hap• nearby expressing the CD 4 molecule, pens in only a small fraction of all there will be binding and fusion of the two macrophages. cells; and this can give rise to fusion of The important question, however, is: literally several cells together. Such cells "What if the immune system attacks this have aberrant function and die prema• cell and destroys it?" Would there be a turely. Based on the laboratory observa• release of more infectious virus? In labo• tions, one can speculate on other ways ratory studies the answer is yes. Ifwe take which could account for the CD4 + T cells an infected macrophage and we break it depletion. Extrapolation of these to the in manually or by attack from cytotoxic T vivo situations may still be remote. It is cells, more infectious virions are indeed important to mention that the gp 120 falls released. Therefore, immune therapy that off the virus easily. In vitro studies show kills infected macrophage must also con• that the gpi20 can interfere with T-cell sider the need for a direct antiviral attack, activation. It can also lead to the down- for example, azidothymidine or neutraliz-

LlX Table 3. Mechanisms of CD4 + T-cell deple• still some variation. And this, in time, has tion in AIDS been shown to have ~iologic significance [31]. For instance, in one virus strain 1. Direct killing by HIV following immune stimulation and virus expression there are many virions with minor dif• . 2. Cell death following syncytia formation ferences. At time zero, one variant may 3. Decreased IL-2 production predominate and the neutralizing anti• 4. Cell-mediated cytotoxicity against body could neutralize almost all of this uninfected cells mediated by free virus variant. Some time later, another gp120 complexed to CD4 and minor variant, perhaps with as little as antibodies against this complex one amino acid change in the envelop, 5. Some viral protein products inhibit may emerge, and this may not be neu• T-cell proliferation tralized by the original antibody. We have 6. Another virus, HHV -6, upon replication is T4 cell-lytic; HHV-6 is common been able to study this in a laboratory in HI V-infected people and may replicate worker who, while mass-producing the more in them virus in another laboratory, was infected 7. Defective antigen presentation leads by accident. One can follow such a person to lower T4 cell proliferation in time. This seems to be an important 8. Inappropriate release of certain cytokines, way by which this vii-us continues to e.g., tumor necrosis factor-IX, escape the immune system. Variant• can decrease T-cell proliferation specific antibody develops and can neu• 9. The gp 120-specific class II -restricted tralize the virus; new minor variants then cytotoxic lymphocytes can lyse activated emerge, but are not neutralized. One (1a +), autologous, uninfected T4 lymphocytes. The CD4 receptor-mediated would expect the variation to occur in the uptake of gp120 is a critical event for region of the neutralizing epitope which this lytic process. This mechanism could exists in the hypervariable region of the allow destruction of a large number of envelope. This is true not only with activated lymphocytes responding to many neutralizing antibody, but with cellular pathogens immunity as well. In addition to variation in this region, we have seen mutation in completely distant regions, e.g., in the transmembrane region of the envelope ing antibodies. Neutralizing antibodies protein gp41. Such a mutation also af• against the HIV-1 work by complexing to fects the interaction of antibody with this a certain region of gp120 and blocking site. More likely the mutations at a dis• entry into the cell. tance bring about conformational changes [32, 33]. Virus Variation. Another important ques• tion is "Why does the HIV -infected per• AIDS and Cofactors. In a small study of son continue to spread the virus?" We homosexuals in Trinidad, Bartholomew know that HIV varies from person to et al. studied dual infections with HIV person. We discovered in 1984 the hetero• and HTLVs and concluded that HTLV-I geneity of HIV for the first time [30]. We may be a cofactor in AIDS [34]. There are found that no two viruses were the same, additional reports now that agree with and the variation was up to 4 % -15 % in this, from Japan in hemophiliacs and the genomes of different HIV variants. from New Jersey with drug addicts [35]. The variation was predominantly in the HTLV -I is not needed to get AIDS at all, envelope region. We also showed later but the rate of progression may be ac• that within anyone virus isolate there are celerated in the presence of HTLV -I. minor variants. That is to say, if you Several mechanisms are possible. HTLV• isolate the virus from one person with I can lead itself to minor T-cell impair• AIDS, although most of the viral par• ment. The TAX protein of HTLV-I can ticles will be very closely related, there is also activate HIV if the cells are infected

LX 1.0 80 % of all people infected with this virus are seropositive. Theq:fore, in most people, it obviously causes no problem. It is the cause of roseola in babies, which is Uninfected not a very serious disease. Adults who have antiviral antibodies and cellular

~_"""HHV-6 immunity can control the replication of HIV·' the virus. It is possible that in AIDS with immune impairment, there is increased o 2 4 6 8 10 12 replication of this virus. If so, one must DAYS POST-INFECTION consider damage to the immune system Fig. 7. Killing of CD4+ T cells by HHV-6 by direct killing of T cells by this virus. In addition, this herpesvirus can activate the HIV genome. It has a gene which makes a with both viruses. In addition, HIV-1- protein that can trans-activate the ex• infected T cells can be activated by the pression of HIV [39], analogously to the simple interactions of HTLV -I with the HTLV-I TAX protein. cell membrane. We have also shown that Finally, we recently showed that this HTLV-I and HIV can form mixed virus human herpesvirus is, as far as we know, particles, which gives HIV the ability to the only biological agent existing natu• affect cell types it normally could not rally that turns on the CD4 gene at the affect [36]. transcriptional level, and to our knowled• The new herpesvirus, human herpes• ge this is the first time it is known that one virus type 6 (HHV -6), which we dis• virus can turn on the receptor of another covered and isolated in 1986 from B cells [40]. In CD8+ T cells and in some [37], actually principally infects CD4 + T epithelial cells, infection by HHV-6 turns cells. This herpesvirus can also kill T cells on CD4 so that they can become targets (Fig. 7) [38]. In the United States, 70%- for HIV infection. All these aspects of

Fig. Sa-c. Spindle cells of Kaposi's sarcoma. medium plus HTLV-II CM (20% v/v). (From a Culture in standard medium (RPMI 1640 [42], with permission; Copyright 1988 by the plus FCS 15 %). b Standard medium plus American Association for the Advancement of endothelial cell growth factor (ECGS) Science) 30 Ilg/ml and heparin 45 Ilg/ml. c Standard

LXI Fig. 9 A, B. Angiogenesis in chick chorioallan• result. (From [42], with permIssIOn; Copy• toic membrane. A Fixed (0.00125 % glutaral• right 1988 by the American Association for dehyde) cells gave a negative result. B Meta• the Advancement of Science) bolically active cells gave a strongly positive

HHV -6 lead us to propose that this virus sexuals, one can speculate that HIV infec• may contribute to the impairment of the tion plays some role. It is not known immune system in people already im• whether any new or unknown viruses mune suppressed by HIY. play a role in AIDS-related Kaposi's sarcoma. We started to explore the possi• AIDS-Related Kaposi's Sarcoma. Since bility of other virus(es) but did not find there has been a great increase in the any. In the process, we developed a incidence of Kaposi's sarcoma in HIV• system for studying Kaposi's sarcoma infected people, more so in male homo- [41,42].

LXII Fig. 10. Lesion in nude mouse induced by Arrow indicates positive result. The left side Kaposi's sarcoma spindle cells (4 x 106 meta• was injected with fixed cells bolically active cells injected subcutaneously).

The important thing that came out of they release a number of cytokines that our studies over the last few years is that have powerful angiogenic activity, which we have a system in the laboratory for is a key feature of Kaposi's sarcoma. studying Kaposi's sarcoma. We can grow Figure 9 shows angiogenic activity re• the spindle cells which are believed to be leased by the spindle cells grown in the the tumor cells of Kaposi's sarcoma. culture tested in the normal chick Figure 8 shows the spindle cells derived chorioallantoic membrane [41]. One can from a person with Kaposi's sarcoma take either the intact spindle cells or the which were grown for several months in concentrate of factors released from them culture. We have several such cell cultures and apply it to the membrane. Distinct now. These spindle cells have been ana• angiogenic activity is observed in both lyzed in collaboration with Judah Folk• instances. man and his associates from Harvard U ni• Mote interestingly, these spindle cells, versity [43]. They have the properties of when put into a nude mouse, cause a primitive smooth muscle cells of vascular tumor similar to human Kaposi's sar• origin, as well as some properties of coma to develop (Fig. 10). The lesion endothelial cells. We think then that the develops near the site of inoculation of precursor cell of Kaposi's sarcoma is a the spindle cells within 10 days. When the mesenchymal, primitive precursor of cells spindle cells regress, the lesion dies out. of the blood vessel walls. Althoughwe We examined the lesion histologically. It could not find any virus, particularly appears like early Kaposi's sarcoma with HIV-1, in these cells, it was found that blood vessel proliferation, fibroblasts,

LXIII LTA pol vpr vpu env LTR HIV -l ~TI----=g,---ag=------"'--r------'OOJ....V-:7i tT'-.,r.'~~t~~~~~~~-'J~--'~

I I I I rey I ifi net I 23456 789

tat r-I..______---~---I

Exon -l hon·3 (noncoding.

~hon - 2

N·t8rminuti 1 Met Glu Pro Val A$p Pro Arg leu Glu Pro Trp LY$ His Pro Gly Ser Gin PrO lV5 Tl'1 r 2Q

HYDROPHOBIC > AC IOIC REG IO N. FUNCTIONAL. DOMAIN £?I

21 Ala Cys. Thr Ali" Cv~ Tvr CY$ Lys Lys Cys Cys Phe His Cys Gi n

CY$TE,INE RICH STR UCTURAlIFUNCT IONAl/DOMAIN CO!'l5ll!lrved

41 Lys Al a L.eu Gly lie S8' Tyr Gly Arg" lys LYi5- Arg Arg Gin Pro Gin 60 BA$IC / NUCLEAR I.OCAI.IZATION OOMA IN

-+-- EICon -2-, I"" Exon ·3 ~ 61 Gly Ser Gin Thr HIs. Gin Val S8r leu S8r Lv!;: Gin Pro Th r Sur Gin Ser Arg G1V Asp 80

CEll BINDING REGION hon - 3~ 81 PrO Th r Gly Pro Lvs. Gil"! 86 C-terminu$ Fig.H. Different structural and functional domains of the TAT molecule

infiltration with leukocytes, and spindle (nanograms) by HIV-1-infected T cells cells. The conclusion is that the spindle and acts as a growth factor for the spindle cells secrete factors that are responsible cells [45]. for the early lesion of Kaposi's sarcoma The TAT molecule has different [42]. We have evaluated the cytokines it regions which are responsible for differ• makes. It appears that IL-1 and basic ent activities (Fig. 11). We think one fibroblast growth factor are the most region is particularly important for the important ones. These molecules can growth-promoting activity on the spindle have angiogenic activity and promote cells [45]. This small protein of 10000 growth of fibroblasts and endothelial daltons is very complex. It has a region cells directly or indirectly. In addition to that is important for the trans-activation these, other factors such as granulocyte• activity of the virus and a basic domain macrophage colony-stimulating factor, important for the nuclear localization. tumor growth factor-p, IL-6, and low levels of acidic fibroblast growth factor and platelet-derived growth factor are Human Retroviruses and Tumorigenesis also detected [44]. The manner in which we succeeded in The direct effects of a retrovirus like growing the spindle cells is interesting in HTLV-I where the virus infects its target itself. We grew the spindle cells by using cell, can immortalize that cell, and makes lymphokine(s) made by chronically ac• it abnormal have been discussed earlier. tivated CD4 + T cells. The major active We find the viral sequences in every cell in lymphokine for this effect is currently the sample place, indicating their clonal being purified in our laboratory and is the derivation from the original transformed most potent growth factor for AIDS cell. HTLV -I can thus be called a directly Kaposi's sarcoma spindle cells. In ad• acting tumor virus. We refer to HIV as dition, we have found that the TAT having indirect effects that can lead to the protein is released in very small amounts increased possibility of tumor develop-

LXIV DIRECT EFFECT References

@- 1. Gallo RC (1986) The first human retro• Virus virus. Sci Am 255: 88 8-8Cell A Cell A 2. Gallo RC (1987) The AIDS virus. Sci Am 256:46 3. Gallo RC, Montagnier L (1988) AIDS in 11\ 1988. Sci Am 259:41 4. Barre-Sinoussi F, Chermann JC, Rey F, et al. (1983) Isolation of a T-Iymphotropic retrovirus from a patient at risk for ac• 888Cell A Cell A Cell A quired immune deficiency syndrome (AIDS). Science 220:868 TUMOR 5. Temin H, Mizutani S (1970) RNA• dependent DNA polymerase in virions of INDIRECT EFFECT Rous sarcoma virus. Nature 226:1211 "B 6. Baltimore D (1970) RNA-dependent DNA polymerase in virions of RNA Virus@-8-8,00- tumor viruses. Nature 226:1209 CeliA I " o + 0 0 o 0 0 0 7. Sarngadharan MG, Sarin PS, Reitz MS, Cell B Gallo RC (1972) Reverse transcriptase activity of human acute leukemic cells: l purification of the enzyme, response to ~ Trigger to Proliferate AMV 70S RNA, and characterization of ~---~ the DNA product. Nature 240:67 TUMOR 8. Morgan DA, Ruscetti FW, Gallo RC (1976) Selective in vitro growth of T Fig. 12. Direct and indirect mechanisms of lymphocytes from normal human bone tumor induction by human retroviruses marrows. Science 193: 1007 9. Narayan 0, Clements JE (1990) Len• tiviruses. In: Field BN eta!. (eds) Virology, Lentiviruses, 2nd edn. Raven, New York, ment. HIV probably increases the possi• pp1571-1589 bility of Kaposi's sarcoma developing in 10. Gardner MB, Luciw P, Lerche N, Marx P at least two ways: (1988) Non-human primate retrovirus iso• lates and AIDS. In: Perk K (ed) Advances 1) it infects T cells and releases TAT in veterinary science and comparative protein; medicine: immunodeficiency disorders 2) its proteins activate immune cells (T and retroviruses, vol 32. Academic, San cells and B cells) which release Diego, pp 171-226 lymphokines. 11. Blattner WA (1989) Retroviruses. In: Evans AS (ed) Virus infections of humans, 3rd edn. Plenum, New York, pp 545-592 Some of these lymphokines can have an 12. Homma T, Kanki PJ, King NW Jr, et a!. effect on the primitive mesenchymal cell (1984) Lymphoma in macaques: Associ• that has lineage to smooth muscle and ation with virus of human T lympho• endothelium and which is the precursor trophic family. Science 225:716 of the spindle cell of Kaposi's sarcoma. 13. Hunsmann G, Schneider J, Schmitt J, This cell in turn releases a series of Yamamato N (1983) Detection of serum cytokines that act to form a complex antibodies to adult T-cell leukemia virus in mixed tumor that we call Kaposi's sar• non-human primates and in people from Africa. Int J Cancer 32: 329 coma. In summary, human retroviruses 14. Miyoshi I, Yoshimoto S, Fujishita M, et can induce tumors, directly or indirectly, al. (1982) Natural adult T-cell leukemia in addition to their suppressive effects on virus infection in Japanese monkeys. Lan• the immune system and abnormal effects cet ii:658 on the nervous system (Fig. 12).

LXV 15. Poiesz BJ, Ruscetti FW, Gazdar AF, et al. 27. Gartner S, Markovits P, Markovits D, (1980) Detection and isolation of type C Betts R, Popovic M (1986) Virus isolation retrovirus particles from fresh and cul• from and identification of HTLV• tured lymphocytes of a patient with cuta• III/LAV-producing cells in brain tissue neous T-cell lymphoma. Proc Natl Acad from a patient with AIDS. JAMA Sci USA 77:7415 256:2365 16. Poiesz BJ, Ruscetti FW, Reitz MS, et al. 28. Koenig S, Gendelman HE, Orenstein JM, (1981) Isolation ofa new type C retrovirus et al. (1986) Detection of AIDS virus in (HTL V) in primary uncultured cells of a macrophages in brain tissue from AIDS patient with Sezary T-cell leukemia. Na• patients with encephalopathy. Science ture 294:268 233: 1089 17. Tschachler E, Robert-Guroff M, Gallo 29. Rappersberger K, Gartner S, Schenk P, et RC, Reitz MS Jr (1989) Human T• al. (1988) Langerhans' cells are an actual lymphotropic virus I-infected T cells con• site of HIV -1 replication. Intervirology stitutively express lymphotoxin in vitro. 29: 185 Blood 73: 194 30. Hahn BH, Gonda MA, Shaw GM, et al. 18. Gessain A, Barin F, Vernant JC, et al. (1985) Genomic diversity of the acqui• (1985) Antibodies to human T-Iym• red immune deficiency syndrome virus photropic virus type-I in patients with tro• HTLV-III: different viruses exhibit pical spastic paraparesis. Lancet ii: 407 greatest divergence in their envelope 19. Gout 0, Baulac M, Gessain A, et al. (1990) genes. Proc Natl Acad Sci USA 82:4813 Rapid development of myelopathy after 31. Hahn BH, Shaw GM, Taylor ME, et al. HTLV-I infection acquired by transfusion (1986) Genetic variation in HTL V• during cardiac transplantation. N Engl J III/LA V over time in patients with AIDS Med 322:383 or at risk for AIDS. Science 232: 1548 20. Jacobson S, Shida H, McFarlin DE, Fauci 32. Robert-Guroff M, Reitz MS Jr, Robey AS, Koenig S (1990) Circulating CD 8 + WG, Gallo RC (1986) In vitro generation cytotoxic T lymphocytes specific for of an HTLV-III variant by neutralizing HTLV-I pX in patients with HTLV-I antibody. J Immunol 137: 3306 associated neurological disease. Nature 33. Reitz MS Jr, Wilson C, Naugle C, Gallo 348:245 RC, Robert-Guroff M (1988) Generation 21. Sommerfelt MA, Williams BP, Chapham of a neutralization-resistant variant of PR, et al. (1988) Human T cell leukemia HIV-l is due to selection for a point viruses use a receptor determined by mutation in the envelope gene. Cell 54:57 human chromosome 17. Science 242: 1557 34. Bartholomew C, Blattner W, Cleghorn F 22. Inoue J, Seiki M, Taniguchi T, et al. (1986) (1987) Progression to AIDS in homo• Induction of interleukin 2 receptor gene sexual men co-infected with HIV and expression by p40x encoded by human T• HTLV-I in Trinidad. Lancet ii:1469 cell leukemia virus type 1. EMBO J 5: 2883 35. Robert-Guroff M, Gallo RC (1991) The 23. Cross SL, Feinberg MB, Wolf JB, et al. interaction of human T-cell leukemia and (1987) Regulation of the human inter• human immunodeficiency retroviruses. In leukin-2 receptor alpha chain promoter: Srivastava R, Ram BP, Tyle P (eds) Mole• activation of a nonfunctional promoter cular mechanisms of immune regulation, by the transactivator gene of HTLV-I. VCH, New York, pp 233-249 Cell 49:47 36. Lusso P, Lori F, Gallo RC (1990) CD4- 24. Jarrett WFH, Crawford EM, Martin WB, independent infection by human immuno• Davie F (1964) Virus-like particles as• deficiency virus type 1 after phenotypic sociated with leukaemia (lymphosar• mixing with human T-cell leukemia vir• coma). Nature 202:567 uses. J Virol 64: 6341 25. Jarrett W, Jarrett 0, Mackey L, et al. 37. Salahuddin SZ, Ablashi DV, Markham (1973) Horizontal transmission of leuke• PD, et al. (1986) Isolation of a new virus, mia virus and leukemia in the cat. J Natl HBL V, in patients with lymphoprolifera• Cancer Inst 51: 833 tive disorders. Science 234: 596 26. Gartner S, Markovits P, Markovits D, 38. Lusso P, Ensoli B, Markham PD, et al. et al. (1986) The role of mononuclear (1989) Productive dual infection of human phagocytes in HTL V- III/LA V infection. CD4+ T lymphocytes by HIV-l and Science 233 :215 HHV-6. Nature 337:370

LXVI 39. Ensoli B, Lusso P, Schachter F, et al. Kaposi's sarcoma-derived cells after long• (1989) Human herpes virus-6 increases term culture in vitro. Science 242:430 HIV-1 expression in coinfected T cells via 43. Weich MA, Salahuddin SZ, Gill P, Naka• nuclear factors binding to the HIV-1 en• mura S, Gallo RC, Folkman J (1991) hancer. EMBO J 8:3019 AIDS-Kaposi's sarcoma-derived cells in 40. Lusso P, DeMaria A, Malnati M, et al. long-term culture express and synthesize (1991) Induction of CD4 and suscepti• smooth muscle a-actin. Am J Pathol bility to HIV-1 infection in human CD8+ 139 : 1251 T lymphocytes by human herpesvirus 6. 44. Ensoli B, Nakamura S, Salahuddin SZ, et Nature 349: 533 al. (1989) AIDS-Kaposi's sarcome-derived 41. Nakamura S, Salahuddin SZ, Biberfeld P, cells express cytokines with autocrine and et al. (1988) Kaposi's sarcoma cells: long• paracrine growth effects. Science 243: 223 term culture with growth factor from 45. Ensoli B, Barillari G, Salahuddin SZ, retrovirus-infected CD4 + T cells. Science Gallo RC, Wong-Staal F (1990) Tat pro• 242:426 tein of HIV -1 stimulates growth of cells 42. Salahuddin SZ, Nakamura S, Biberfeld P, derived from Kaposi's sarcoma lesions of et al. (1988) Angiogenic properties of AIDS patients. Nature 345: 84

LXVII Pastor Bode Lecture Wilsede, June 20, 1990

N. Avrion Mitchinson: Escape from the Red Queen

LXIX "Heidepastor" Wilhelm Bode

Michel Weidmann (Aquarell)

LXX Haematology and Blood Transfusion Vol. 35 Modern Trends in Human Leukemia IX R. Neth et al. (Eds.) © Springer-Verlag Berlin Heidelberg 1992 Wilhelm Bode was born on October 20, speaker, or anything else other than a 1860, in Luneburg, in Germany, as the servant of the true Teaching! Take care third of ten children. His father was a what is said from this, your pulpit, and teacher at the seminary in Luneburg. watch jealously that it not be profaned! There is a legend that the father, on the The parish that requires nothing of its night of the birth of each of his sons, pastor is asleep; but the one that requires took the child and, placing hin in the much is alive... The first premise for wide pocket of his coat, diaper and all, beneficial cooperation is an unreserved walked up the nearby Kalkberg hill. and mutual give and take between us. We There, he held up the child facing the shouldn't say; 'Here are the parishioners, town spread out peacefully below, and and there is the pastor,' but rather 'We said, "Behold, my son! This is your home together!' This is may task." country! Remain true to it, hold it dear to Pastor Bode tried to make lessons in your heart, and protect it!" the schools more interesting, and checked Wilhelm, the third son, inherited his to see that the teachers really began father's inclination towards nature and lessons on time. He himself organized his enthusiasm for it, and indeed for programsfor official school celebrations; everything unspoiled and free, more he arranged for the children to be pro• strongly than any of the other children. vided with their school books, with the Wilhem often accompanied his father on exception of Bible and psalm book, at the the long excursions he habitually took in cost of the school district. He advocated the school holidays. And he showed from putting an atlas and a book on science the first a particular love of nature lore and languages in the hands of each pupil; and science. and he considered sports, gymnastics, to The youth grew up in a tightly knit be one of the most important subjects. It family. The father's favorite saying was, was Pastor Bode, with his farmers, who in "The rich man is not the one with many 1888 founded the first savings and loan possessions, but the one with few needs." bank in the Luneburg Heath. Later, a In 1880, upon fishing school, the 19-year• cooperative for the insurance of farm old Wilhelm started theological studies, animals, the Kuhkasse or "cow found," first in Gottingen, then in Strasbourg. was added. The cooperative purchase of Even as a student, his interests ranged feed and fertilizer was organized, fol• wide: zoology, botany, history, and Ger• lowed by a water cooperative, which used man romantic literature fascinated him in wind power to provide the village with addition to his main subject of study. For running water. several weeks during one holiday, he This was the one side of his work: even joined a wandering circus. practical, active help so that the people After completing his theological entrusted to him could improve their studies, he was assigned to the parish in standard of living. Egestorf, six miles to the east of Wilsede Pastor Bode's other side seems marked Hill. This was to become the scene of his by a sort of natural piety: his sermons life's work. His first serom, on August 15, breathe the air of freedom and nature. 1886, was delivered on the theme We His passion for the heath is not an together. "You are my parishioners," he ideology; rather, it was a part of his said, "and I am your pastor; and if two pastoral teaching to win each of his people are going to live together, and take farmers to an appreciation of the land up housekeeping together, it is a good that he farmed. On a walking trip with his thing for each to have a clear notion of the father, the young Bode had passed from rights and duties that each has towards Egestorf via Aue and Radenbach to Wil• the other." "Do not demand that I de• sede, through the untouched natural monstrate all the social graces," he went beauty of the open heath with its juniper on, "or that I be worldly wise, or a flashy bushes, with the lustrous dark green of

LXXI the bordering pine woods. At one point, formed, enthusiastically erecfed an inn, his father said, "My son, if a man could the "Inn at the Heath Museum." Pastor preserve this landscape for future gener• Bode wrote the advertising pamphlets ations, he would have accomplished a himself, and argued: "No paved road! No great work, a good work." nickelodeon!" Many years later, when Bode found As Pastor Bode learned that a dance that a considerable parcel of land, the hall was to be built on Wilsede Hill, he Totengrund, was to be sold and used for managed to delay the sale of the land. construction, he tried to prevent the sale. District Counsellor Ecker from Win• After many fruitless attempts, he found a sen/Luhe, a member of the Nature valuable ally in Professor Thomsen from Park Society, sent the author of Kosmos, Munster: this man was prepared to supp• Dr. Curt Floericke, to look over the ly funds for the purchase of the Toten• situation. Impressed, Floericke wrote a grund, thus saving it from the develop• decisive report. Ecker, as representative ment that threatened it. After very dif• in the Prussian Legislature, succeeded in ficult negotiations, Bode succeeded in arranging for public funds to be made purchasing the Totengrund in 1906 for available: the Nature Park Society was the sum of 6000 marks. This piece ofland then able to purchase this parcel as well. was to become the seed from which the When Pastor Bode died on June 10, Liineburg Heath Nature Park was to 1927, he was mourned by large numbers grow. of people. It was his request that his ashes Bode carried out his next project in be scattered to the winds from the top of cooperation with a Herr DagefOrde, a the Wilsede hill. This wish was granted teacher from Tangendorf. This teacher him. had assembled an extensive anthropolog• ical collection which filled the school• Acknowledgment. I thank Griffin Ander• house to overflowing. On the initiative of sen for the English translation. Pastor Bode, a piece of land was pur• chased in Wilsede. Thus, Bode became Hanne-Lore Neth one of the founders of the Wilsede Heath Museum Society. Dageforde acquired (quite cheaply, as it was going to be torn down) a fine old farmhouse in Hanstedt Literatur dating from 1750. This house was disas• 1. Walter Groll: "Durch die Liineburger sembled, and then rebuilt on the lot in Heide". Verlag Hans Christians, Hamburg, Wilsede. It opened on August 15, 1907, as 1977 "The Old House", or in the North Ger• 2. Walter Brauns: "Der Heidepastor". Verlag man idiom, "dat ole Huus." Tourism des Vereins Natursehutzpark e. v., Stutt• increased. The Society, only recently gart und Ham burg. 1983

LXXII Pastor Bode Lecture

Escape from the Red Queen

N. A. Mitchison 1

Immunological diseases follow a charac• driven by presentation of a self-peptide teristically fluctuating course of relapse by the HLA molecule. Not only does this and remission, such as is illustrated in provide an attractively simple picture of Fig. 1. This is true most obviously of how the disease develops, but it also diseases such as rheumatoid arthritis points the way forward to new modes of which have a major autoimmune compo• treatment. From the tight associations nent, but it holds equally well for chronic spring the present flurry of excitement infectious diseases such as leprosy in concerning HLA-blocking peptides and which hypersensitivity plays an import• monoclonal antibodies. ant part. It is likely, but not definitely In comparison, beneficial HLA genes established, that the fluctuations reflect have suffered neglect. This seems a pity, if an imperfect balance between opposing only because it makes sense to try to forces within the immune system, and understand what makes a patient get that these in turn reflect the activity of better. The negative associations between opposing control genes. Among such HLA and disease seem on the whole to be control genes, those of the major his• weaker than the positive ones, although tocompatibility complex (MHC) are like• this has not been categorically estab• ly to. be the most important. lished. Another reason for neglect is that Studies on the MHC and disease have it is less obvious how an MHC gene could tended to focus on detrimental genes, that inhibit an immune response, in the way is, those which are positively associated that these beneficial genes seem to do. with the disease in question, predispose for it, and presumably act as causal fac• tors. Some of the autoimmune diseases are Delrimenlal tightly associated with particular HLA gene action: genes, such as is ankylosing spondylitis 110 be discouraged (and certain forms of reactive arthritis) with HLA-B27. For others the tightness of the association has become apparent only as seemingly unrelated predisposing genes have been discovered to share se• quences in common. Thus an epitope shared between HLA-DR1 and HLA• Dw4 explains well why both of these genes predispose for rheumatoid arthritis Beneficial gene action: [1]. The existence of a tight association 1to be encouraged suggests that the disease process may be Fig. 1. The disease pattern of relapse and 1 Deutsches Rheuma Forschungszentrum remission, characteristic of immunological dis• Berlin, Am K1einen Wannsee 5, W-1000 eases, suggests that opposing activities operate Berlin 39, FRG. within the immune system

Haematology and Blood Transfusion Vol. 35 LXXIII Modern Trends in Human Leukemia IX R. Neth et al. (Eds.) © Springer-Verlag Berlin Heidelberg 1992 One has to think seriously about sup• life in Venezuela. Since then leprosy has pressive activity and suppressor cells, become an arena for testing ideas about and those are subjects that immunolo• genetic control of suscepbility to chronic gists have learned to be cautious about. infectious disease, and more recent pre• Nevertheless they are the subject of this sentations of the topic are available [5]. paper. Rather than go over the whole list of inhibitory genes in this and other areas Inhibitory MHC Genes again in detail, it seems more useful on the present occasion to offer the follow• Examples of inhibition of immune re• ing generalizations. sponses by MHC genes are not hard to find. Our recent survey lists some 20 in 1. As evidence of an immunoinhibitory mouse and man [2]. What that listing did effect, a negative association between not include are the significant but only HLA and an autoimmune disease (or moderately impressive negative associ• other immunological disease, such as ations between HLA and autoimmune allergy) is equivalent to a positive as• disease that have often been recorded, sociation with an infectious disease. usually as by-products of surveys aimed 2. Evidence of such associations often principally at verifying positive associ• springs initially from population sur• ations. Figure 2 gives an example, show• veys. Such data need eventually to be ing the apparently beneficial effects of supported by the stronger evidence three HLA-DR genes on rheumatoid that multiple-case family studies can arthritis in a recent study [3]. provide. In the mouse, studies on If I had to choose just one example of a panels of recombinant inbred mice disease study, it would be the joint work provide the equivalent of family carried out by groups from Leyden and studies in man. It should be noted, Caracas on HLA and lepromatous lep• however, that families with multiple rosy [4]. Not only does this contain cases tend to have a high susceptibility byautiful data, but also it provides an background, and that this will dimin• amusing sidelight on paternity and family ish the impact of protective genes (I am

Relative risk

DR2 .54

OR3 .38

OR7 .43

o 1 0 20 30 40 % Fig. 2. Negative associations between class II HLA genes and rheumatoid arthritis, detected in a UK survey

LXXIV grateful to H. O. McDevitt for point• but they do share certain features in ing this out to me). common. Both seem to be secondary 3. As mentioned above, many immuno• class II genes, with relatively few T logical diseases have strong positive cells restricted by them, with relatively HLA associations. This will tend to low expression, and with relatively low produce negative associations for polymorphism (none of these features other HLA genes, and the more fre• are definitively documented, unfortu• quent such a gene is in the study pop• nately). Function seems to have flip• ulation, the more likely is this nega• ped from one gene to another dur• tive association to reach significance. ing the evolutionary divergence of the 4. Although a negative MHC association two species. may be taken as prima facie evidence of immunoinhibition in an immuno• One of the important issues of the day logical disease (and likewise a posi• in this area is the claim that the presence tive one in an infectious disease), de• of asp-57 in HLA-DQ protects against tailed immunological study would be insulin-dependent diabetes. The most re• needed to substantiate the claim. This cent publication on this subject amounts might involve exploring the possibility to a vigorous rejection of this claim on the of relieving the inhibition by in vitro basis of a segregation study in multiple• (or in the mouse by in vivo) treatment case families [6], but the qualification with anti-MHC monoclonal anti• about this type of study, mentioned in bodies, or other procedures. point 2 above, means that judgement 5. The clear-cut inhibitory MHC genes should be suspended. I doubt if we have have all so far turned out to belong to yet heard the last word on this matter. class II. This is surprising, in view of Finally let me mention the recent study the fact that T cells able to mediate an that raised my interest in the present issue inhibitory effect often have the CD8 [7]. It showed that substitution ofH-2Ab phenotype. One can think of explan• for H-2Ak strongly inhibits the response ations, for instance, along the lines of of mice to F liver antigen and does so the phenomenon known in inbred mice more weakly for their response to Thyl where, when an active H-2K allele (in a antigen. This provides a promising sys• cytotoxic response) is replaced by an tem for further study of mechanism. inactive one, the previously inactive H- 2D allele become active (and vice versa). But the absence of class I genes Mechanisms of Inhibition still seems odd, and perhaps further research will change the picture. A comprehensive conceptual framework 6. No MHC gene has been found to within which to consider mechanisms of mediate inhibition exclusively. All the inhibition is much needed, and the main genes which inhibit a response act purpose of this paper is to present one in positively in others (this statement re• the outline form shown in Fig. 3. In doing quires some qualification as regards so, I gladly acknowledge the benefit de• HLA-DQ in man, where most of the rived from discussions with C. S. David evidence for a positive effect comes and H. O. McDevitt, and from the from in vitro studies with cloned T cells trenchant commentary of Nepom [6]. of "helper" phenotype). This classification begins with a dis• 7. Nevertheless a certain bias in the loc• tinction between intracellular and inter• ation of inhibitory activity within the cellular mechanisms, so that only in the MHC is evident, both in mouse and former does the inhibitory MHC product man. This is respectively towards H- operate within the same cell as the posi• 2 E and HLA-DQ. These genes are not, tive MHC (immune response, Ir) prod• of course, homologous in evolution, uct. Intercellular mechanisms corre-

LXXV Intramolecular

Intracellular / Intermolecular / " egative selection Intercellular / "(" independent") Positive selection Network (anti.id) Competitive/inhibitory Iymphokines (THl.TH2)

Fig. 3. A classification of immunoinhibitory hibitory molecule or repertoire IS shaded mechanisms mediated by class II major his• darker tocompatibility complex molecules. The in· spond roughly to what Nepom designates monstrating loss of inhibition when the as "independent" inhibitory activity, a dose of antigen is increased. We do not category whose existence he questions. known for certain whether the allo-MHC Intracellular mechanisms further sub• antigen in this experiment is presented by divide into intra- and intermolecular self-class II molecules in the usual way, ones. Intramolecular inhibition, for in• but if it is then the gene(s) which encode stance, might involve an amino acid sub• them could be regarded as inhibitory for stitution in the cd helix inhibiting the the anti-Thyl response. In summary, the activity of an Ir gene previously defined known instances of intracellular, inter• by substitution in the !X2 helix. This is a molecular inhibition boil down to anti• makeshift, as eventually the combination genic competition. would be designated simply as a new The intercellular mechanisms subdiv• neutral allele. On the other hand, the ide into those that result from negative or category of intracellular, intermolecular positive selection. In negative selection inhibitions has several examples, mostly the product of one MHC gene inhibits the involving class I genes in the mouse. activity of another by deleting a part of its These I have discussed some time ago [8]. repertoire. The ikon for this category in The phenomenon of competition in the Fig. 2 depicts the T cell repertoire as antiviral cytotoxic response between H- subdivided into four parts according to 2K and H-2D mentioned above belongs their restriction elements (e.g. H-2Aa, H- here, as also does the competition be• 2Ah, H-2E", H-2Eh); the part restricted tween heterozygous alleles at H-2K or H- by the inhibitory gene is shaded darker, 2D that has been noted in the same type and there are holes in the other parts of of response. More relevant to our present the repertoire. This phenomenon is now discussion of class II inhibitory genes, familiar in the context of superantigens, perhaps, is the competitive suppression such as H-2E, or the mls product. It has that my group has studied in the anti• not yet been identified (to the best of my Thyl response [9]. There the presence of knowledge) as a mechanism of inhibition an allogeneic MHC molecule on a Thyl resulting from class II allelic substitution, antigenic cell (but not on a neighboring although that may well be disclosed in the cell) can inhibit the response. That this is future. truly a competitive inhibition is strongly Suppression mediated by positive supported by our (unpublished) data de- selection is the most challenging category,

LXXVI for this is our old friend the suppressor T Now that we have this classification, cell revived. In the formal sense used here, are we yet in a position to assign any of inhibition results from positive selection the known immunoinhibitory effects to when the product of one MHC class II their correct slot within it? For the effects gene enables a group ofT cells to develop which matter in human disease, of the that are able to inhibit the activity of type shown in Fig. 2, the answer is, not another group of T cells that would yet. For mouse models some assignments otherwise perform a response. The ikon can be made to the category of positive shows the part of the repertoire that is selection and others to that of negative restricted by the inhibitory gene as biased selection. V-gene usage provides an im• towards inhibition, while the rest of the portant clue to the operation of negative repertoire is biased away from that activ• selection, and enhancement of the re• ity (note the patches of shading). As sponse by anti-class II antibody does mentioned in the figure, current theory is likewise for positive selection. But it must that inhibitory T cells could operate in be emphasized that assignments made on alternative ways. One would be, by en• these bases are only provisional, because gaging in inhibitory anti-idiotypic recog• it is always hard to exclude the possibili• nition, thus involving Jerne's network. ty that some additional mechanism is An alternative would be the secretion of operating. inhibitory or competitive lymphokines, This discussion has focussed on ways such as do TH 1 and TH 2 cells in the of carving up CD4 class II-restricted T mouse [10]. Other possibilities could be cells, the main regulatory compartment cited, such as antigen-specific suppressor of the immune system and arguably the factors, but these seem too remote to be single most numerous and most impor• included in the discussion. tant group of lymphocytes. They can be This is not the place to discuss these last subdivided according to restriction ele• alternatives in detail. Inhibition via the net• ment, involvement in the network, lym• work has a long history and much exper• phokine secretion profile, and positive imental support (see my reviews [11,12]) versus negative effect (and also according and can be regarded from various points to markers such as CD45R which dis• of view. As K. Rajewsky has pointed out criminate between naive and memory to me, it could be no more than a form of cells). Some of these are lineage markers mopping up, needed only to prevent while others are not, and one needs to use inescapable network interactions getting ones wits when mixing the two [14]. It out of control. Or, as I. R. Cohen pro• seems to me that the alignment of these poses, self-macromolecules such as heat• various characteristics is one of the most shock proteins or myelin basic protein important items on the agenda of cellular could induce a positive response within immunology. the immune system, which would nor• mally be contained by anti-idiotypic sup• pression, but which would on occasion Why Are Immunoinhibitory Genes break out in the form of autoimmune so Frequent? disease. As regards lymphokines, the evidence in the mouse is convincing of It is hard to believe that autoimmune mutual inhibition mediated by y-in• disease occurs with sufficient frequency, terferon produced by TH 1 cells, and or in a young enough age group, to have interleukin (IL)-4 and IL-10 produced by had much evolutionary impact. As we TH2 cells. In man the position is less have argued elsewhere [15], the driving clear; perhaps atopy and its control by force is more likely to have been the therapeutic vaccination may offer the hypersensitivity induced by chronic infec• best example of inhibitory T cell activity tion. Most or all of the major tropical [13]. diseases are associated with hypersensi-

LXXVII tivity, and nowhere is this more con• elimination from the MHC gene pool. It spicuous than in leprosy. In that disease is tempting to suppose that reduced class immunopathological mechanisms are II expression may provide a mechanistic most threatening in borderline cases. It is signal for suppression within the immune as though individuals at either end of the system, thus dosing the evolutionary spectrum are protected: at the tubercu• circle. loid pole (and in the much larger number All this is of course highly speculative. of individuals who are infected but never The value of the evolutionary arguments show clinical symptoms) the immune is that they focus attention on particular system functions in its usual protective mechanisms, and also that they identify mode, while at the lepromatous pole its the need for particular types of im• protective functions are inhibited and the munoepidemiological data. parasite becomes free to mUltiply. Im• munoinhibitory genes may thus occur in human populations in the developed New Therapies: Combatting world largely as a result of past selection or Enhancing Immunoinhibition for inhibition of infection-associated hypersensitivity. The proof of these ideas about inhibitory The association noted above between activity is whether they lead on to new immunoinhibitory activity of MHC forms of therapy. In this context three genes, low expression, and low polymor• lines of current research look particularly phism now begins to make sense. The promising. The first two concern chronic predominant activity of MHC genes is infectious diseases in which immunoin• positive where they function as immune hibitory activity has long been suspected response genes. Such genes are driven to of preventing recovery, and where a novel become intensely polymorphic, as a result form of therapy offers hope of breaking of what the evolutionary biologists have through that barrier. The third concerns come to call "the Red Queen strategy." the opposite problem, autoimmune dis• By this is meant that anyone species lives ease in which the lack of adequate im• wifhin an environment provided by other munoinhibitory activity may help cause species, and as one evolves so must the the disease, and where a novel form of others. The final result is a great deal of therapy might rectify that defect. evolutionary change but little real pro• This year a group from the Rockefeller gress, just as in Lewis Carrol's Through University collaborating with local re• the Looking Glass where Alice and the searchers in Addis Ababa published their Red Queen hold hands and run, without results on sublesional administration of getting anywhere. Nowhere does this IL-2 in leprosy [16]. This is the first trial of apply with greater force than in the co• lymphokine treatment in a chronic infec• evolution of immune response genes in tious disease, and it used the lymphokine the host and the antigen genes in parasites at something approaching physiological to which they are opposed. This ceases to concentrations (far less than has been apply to MHC genes in respect of their used in cancer trials). The results were inhibitory function. In that case the inter• encouraging, as judged by the histolog• ests of the host and the parasite coincide; ical response determined in skin biopsies, the Red Queen, so to speak, comes out of and treatment with other lymphokines is play. We can therefore expect im• planned. From the point of view ex• munoinhibition to associate with dimin• pression above, treatment of this sort ished polymorphism. The association carries great promise as well as some with diminished expression may occur hazard. If patients are to be shifted along because, on balance, such genes prove the spectrum towards the tuberculoid less valuable in an evolutionary sense; pole, it is essential that they be moved out they may even be on their way to total of the intermediate zone of hypersensitiv-

LXXVIII ity and not into it; that will require careful References patient selection. While these results do not provide direct support for the TH 1- 1. Wordsworth BP, Lanchbury JSS, Sakkas TH2 concept, they are at least compat• LI, Welsh KI, Panayi GS, Bell JI (1989) ible with it. HLA-DR4 subtype frequencies in rheumatoid arthritis indicate that DRB 1 is Last year there appeared a full report the major susceptibility locus within the on the treatment of cutaneous leish• HLA class II region. Proc Nat! Acad Sci maniasis by immunotherapy in the form USA 86:10049-10053 of vaccination with bacille Calmette• 2. Oliveira DGB, Mitchison NA (1989) Im• Guerin (BCG) plus killed leishmania mune suppression genes. Clin Exp Im• organisms [17], a form of treatment that munoI75:167:177 has also been applied in leprosy. Results 3. JaraquemadaD,OllierW,AwadJ, Young as good as those of conventional chemo• A, Silman A, Roitt 1M, Corbett M, Hay F, therapy were obtained. That reports inc• Cosh JA, Maini RN (1986) HLA and ludes a detailed and thoughtful dis• rheumatoid arthritis: a combined analysis of 440 British patients. Ann Rheum Dis cussion of the possible mode of action; 45:627-636 once again, while many other possibilities 4. Van Eden W, Gonzalez NM, de Vries RR, remain open, an interpretation in terms ConvitJ, van Rood JJ (1985) HLA-linked of competing lymphokines seems at• control of predisposition to lepromatous tractive. leprosy. J Infect Dis 151 :9-14 The third attractive line of research is 5. Li SG, de Vries RRP (1989) HLA-DQ . After long debate and molecules may be products of an immune much hesitation, we are about to witness suppression gene responsible for Myco• bacterium leprae specific nonresponsive• the first gene therapy trials in man, start• ness. Int J Leprosy 57:445-556 ing probably with cancer and with con• 6. Nepom GT (1990) HLA and type 1 dia• genital enzyme deficiencies. A strong case betes. Immunol Today 11(9):314-315 can be made for following these with 7. Mitchison NA, Simon K (1990) Dominant trials in the hemoglobinopathies. If all reduced responsiveness controlled by H- goes well, it would seem reasonable to 2 (Kb) Ab. A new pattern evoked by Thy-1 consider such therapy for cases of auto• antigen and F liver antigen. Immunogene• immunity which have proved refractory tics 32: 104-109 to other forms of treatment. The type of 8. Mitchison NA (1980) Regulation of the immune response to cell surface antigens. gene that one would wish to implant are In: Pernis B, Vogel HJ (eds) Regulatory T those shown in Fig. 2, or possibly an asp- lymphocytes. Academic, New York, 57 HLA-DQ if the doubts mentioned pp147-157 above can be resolved. I am well aware of 9. Lake P, Mitchison NA, Clark EA, Khor• the difficulties: how to protect an implan• shidi M, Nakashima I, Bromberg JS, ted allogeneic major transplantation anti• Brunswick MR, Szensky T, Sainis KB, gen, for example; and, for those genes Sunshine GH, Favila-Castillo L, Woody which operate their inhibitory effect IN, Lebwohl D (1989) The regulation of through positive selection, how to obtain antibody responses to antigens of the cell surface: studies with Thy-1 and H-2 anti• expression in thymic epithelium. But with gens. In: Reif AE, Schlessinger M (eds) the technologies that are becoming avail• Cell surface antigen Thy-l: immunology, able these obstacles do not seem insuper• neurology and therapeutic applications. able. Now many be the time to start a Dekker, New York, pp 367-394 serious research effort towards that goal. 10. Mosmann TR, Coffman RL (1989) TH1 and TH2 cells: different patterns of lym• phokine secretion lead to different func• tional properties. Annu Rev Immunol 7:145 11. Mitchison NA, Oliveira DBG (1986) Epirestriction and a specialised subset ofT

LXXIX helper cells are key factors in the regu• evolution of the supressor· T celi' system. lation of T suppressor cells. In: Cinader B, Parasitol Today 2:312-313 Miller RG (eds) Progress in immunology, 16. Kaplan G, Kiessling R, Tekelmariam S, vol 6. Academic, London, pp 326-334 Hancock G, Sheftel G, Job CK, Converse 12. Mitchison NA (1989) Is genes in the P, OttenhoffTHM, Becx-Bleumenink M, mouse. In: Melchers F (ed) Progress in Dietz M, Cohn ZA (1990) The reconsti• immunology, vol 7. Springer, Berlin Hei• tution of cell mediated immunity in the delberg New York, pp 845-852 cutaneous lesions of lepromatous leprosy 13. Plaut M (1990) Antigen-specific lymph• by recombinant interleukin-2. J Exp Med okine secretory patterns in atopic disease. 169:893-908 J ImmunoI144:4497-4500 17. Convit J, Castellanos PC, Rondon A, 14. Mitchison NA (1988) Suppressor activity Pinardi ME, Ulrich M, Castes M, Bloom as a composite property. Scand J Immunol B, Garcia L (1987) Immunotherapy versus 28:271-276 in localsed cutaneous leish• 15. Mitchison NA, Oliveira DBG (1986) maniasis. Lancet 1 :401-405 Chronic infection as a major force in the

LXXX Frederick Stohlman Jr. Memorial Lecture

Wilsede, June 21, 1978 Wilsede, June 18, 1984 Moloney, William C.: Memorial Tribute to Gallo, Robert c.: Dr. Frederick Stohlman Introduction for Peter H. Duesberg Gallo, Robert c.: Cellular and Virological Duesberg, Peter H.: Are Activated Proto• Studies Directed to the Pathogenesis of the Onc Genes Cancer Genes? Human Myelogenous Leukemias Pinkel, Donald: Treatment of Childhood Wilsede, June 22, 1986 Acute Lymphocytic Leukemia Vogt, Peter K.: Introduction for Harald zur Hausen Wilsede, June 18, 1980 Harald zur Hausen: Viruses in Human Tumors Klein, George: The Relative Role of Viral Transformation and Specific Moore, Malcolm A. S.: Cytogenetic Changes in the Development Introduction for Donald Metcalf of Murine and Human Lymphomas Metcalf, Donald: Hemopoietic Growth Kaplan, Henry S.: On the Biology Factors and Oncogenes in Myeloid and Immunology of Hodgkin's Disease Leukemia Development

Wilsede, June 21, 1982 Wilsede, June 19, 1988 Greaves, Melvyn, F: Immunobiology Rowley, Janet D.: Molecular Analysis of of Lymphoid Malignancy Rearrangements in Philadelphia (PhI) Chromosome-Positive Leukemia Thomas, E. Donnall: Bone Marrow Transplantation in Leukemia Moore, Malcolm A. S.: Interactions Between Hematopoietic Crowth Factors: The Clinical Role of Combination Biotherapy Wilsede, June 17, 1984 Klein, Jan: Wilsede, June 17, 1990 Introduction for N. Avrion Mitchison Broder, Samuel: Clinical Research Using 3'• Mitchison, N. AVI'ion: Repertoire Azido-2',3'-Dideoxythymydine (AZT) and purging by medium-concentration Related Dideoxynucleosides in the Therapy self-macromolecules is the major factor of AIDS determining helper and suppressor Grosveld, Frank: The Human p-Globin repertoire differences Locus Control Region

Haematology and Blood Transfusion Vol. 35 LXXXI Modern Trends in Human Leukemia IX R. Neth et al. (Eds.) © Springer-Verlag Berlin Heidelberg 1992 Frederick Stohlman Jr. Memorial Lecture Clinical Research Using 3'-Azido-2' ,3'-Dideoxythymidine (AZT) and Related Dideoxynucleosides in the Therapy of AIDS

S. Broder!

Introduction another group that cannot form phos• phodiester linkages. Pathogenic retroviruses play an etiolog• 3'-Azido-2',3'-dideoxythymidine (also ic role in the acquired immune deficien• called zidovudine, 3'-azido-3'-deoxy• cy syndrome (AIDS) and its related thymidine, azidothymidine, or AZT), the disorders. While no cure for diseases first of these compounds to be tested caused by these agents is available, we are clinically, reduced the morbidity and now in an era in which therapy against mortality associated with severe HIV pathogenic retroviruses is a practical re• infection [27, 28]. Volberding et al. [29] ality. Therapies against the etiologic reported that AZT is effective in delaying agent of AIDS [1- 3] were made pos• progression to fulminant AIDS in as• sible by the discovery that a retrovirus, ymptomatic patients infected with HIV. now called the human immunodeficiency Other dideoxynucleosides are now in virus (HIV), caused the disorder [4-6]. various stages of clinical testing. Other This discovery and the ability to grow the substances that act at various stages of virus in large quantities enabled the de• HIV's replicative cycle also have been velopment of in vitro techniques to find shown to block replication in vitro, and drugs that inhibit the replication of HIV some are undergoing clinical testing [7, [7, 8].' Substances that acted against HIV 31-59]. This article reviews certain clin• in vitro could then be identified for ical applications of one antiretroviral further research, and the orderly develop• agent, AZT, and discusses the status of ment of drugs thus proceeded. The work several related compounds. It also ad• was in part a outgrowth of early research dresses other approaches to antiretroviral on animal retroviral systems in a number treatment. While an ultimate cure for oflaboratories [9-12]. No one person or AIDS will require further basic research, group can take full credit for these dis• the knowledge already at hand might covieries, and a great debt is owed to make a major impact against the death many scientists who pioneered this re• and suffering from this disease in the search. More recently, our group and coming decade. During the decade of the other groups have observed that certain nineties, AIDS is expected to increase, members of a class of compounds called and the disease is likely to become a dideoxynucleosides are potent inhibitors major cause of death in men, women, and in vitro of the replication of HIV in children. In some parts of the world, human T cells [8, 13-26]. In all the infant and child mortality could be as compounds, the hydroxy (-OH) group much as 30 % greater than what one in the 3'-position on the sugar ring is would have expected [60]. replaced by a hydrogen atom (-H) or

1 National Cancer Institute, Bethesda, Mary• land 20892, USA.

Haematology and Blood Transfusion Vol. 35 LXXXIII Modern Trends in Human Leukemia IX R. Neth et al. (Eds.) © Springer-Verlag Berlin Heidelberg 1992 Potential Mechanisms mammalian DNA polymerase alpha is of Action Against HIV relatively resistant to. the effects of these drugs (inhibition constants, 100- Although there are substantial difffer• 230 11M) [12, 61, 63, 70], which is one ences among the dideoxynucleosides, we reason for their selective antiretroviral can make several general comments activity in cells that can phosphorylate about their mechanisms of action. Each them. However, mammalian DNA poly• drug likely inhibits reverse transcriptase merase gamma, found in mitochondria, but must first be activated to a 5'-tri• and DNA polymerase beta are also sensi• phosphate form by various enzymes of a tive to these compounds (inhibition target cell [61-65]. It is the triphosphate constants, 0.016-0.4 11M and 2.6- form that is active against HIV [61, 66]. 70 11M, respectively), and this may be a The activation process, called anabolic basis for drug toxicity [12, 61, 63, 70]. It phosphorylation, involves a series of en• should always be borne in mind that zymes (kinases) [61-65, 67]. AZT and additional mechanisms of activity and other dideoxynucleosides, as triphos• toxicity might be at work. It is possible phates, exert their antiretroviral activity that AZT works through an as yet uni• at the level of reverse transcriptase (viral dentified intermediate. DNA polymerase) [12, 61, 66, 68, 69]. AZT and related dideoxynucleosides Reverse transcriptase is essential in the have activity against certain retroviruses, replicative cycle of HIV. A great deal is including HIV type 2, human T-cell lym• now known about the overall structure of photropic virus type I, animal lenti• the polymerase domain of HIV reverse viruses, and murine retroviruses [9-12, transcriptase, about the active site, and 68, 74-78]. (Some congeners also have in about the secondary structure within the vitro activity in an animal model of active site (e.g., see [69]. When HIV enters hepatitis B virus [79] that, although a a target cell, this enzyme makes a comple• DNA virus, replicates through an RNA mentary strand DNA copy of the viral intermediate using a reverse transcrip• genomic RNA and then catalyzes the tase-like DNA polymerase [80].) This production of a second, positive strand conservation of activity suggests that DNA copy. The genetic information of the sensitivity of reverse transcriptase HIV is thus encoded in a double-stranded to these agents (as triphosphates) is form of DNA. Two mechanisms may linked to an essential feature of the viral contribute to the effect of AZT and other enzyme. Nevertheless, two groups re• dideoxynucleosides on reverse transcrip• cently reported that certain isolates of tase. First, as triphosphates, they com• HIV from patients who had taken AZT pete with the cellular deoxynucleoside-5'• for 6 months or longer had reduced sen• triphosphates that are essential substrates sitivity to AZT in vitro [81, 82]. Some of for the formation of proviral DNA by the known mutations would be likely to reverse transcriptase with inhibition affect the charge or alpha helix content of constants generally in the range of 0.005- the catalytic site, or at least the probable 0.21lM [61, 70-73]. These concent• catalytic site, of reverse transcriptase. rations can be attained in cells exposed to Larder et al. [83] previously showed that the drugs [61-65]. Second, such 5'• the induction in vitro of certain mut• triphosphatases act as chain terminators ations in reverse transcriptase by site• in the synthesis of proviral DNA. directed mutagenesis could make the en• Because of the 3' modification of these zyme less sensitive to inhibition of AZT compounds, once viral reverse transcrip• triphosphate and phosphonoformate tase adds them to a growing chain of viral (but also reduce enzymatic activity). DNA, the DNA is elongated by exactly However, HIV isolated from patients one residue and then terminated [66]. In receiving long-term AZT therapy retains contrast to HIV reverse transcriptase, its in vitro sensitivity to inhibition by

LXXXIV most other dideoxynucleosides (such as [61]. AZT-MP binds efficiently to thy• 2' ,3' -dideoxycytidine) and phosphono• midylate kinase, but comes off slowly, formate [82]. Also, preliminary studies thus tying up the enzyme. In certain T-cell suggest that the sensitivity of the reverse lines exposed to very high concentrations transcriptase obtained from these resist• of AZT (50-200 1lM), decreased phos• ant viral isolates to 3'-azido-2',3'• phorylation of thymidine and decreased dideoxythymidine-5' -triphosphate levels of thymidine-5'-triphosphate (the (AZT -TP) did not change [82]. normal DNA building block that com• These preliminary findings suggest that petes with AZT- TP for reverse tran• while changes in reverse transcriptase scriptase) have been reported [61, 85]. may account for the loss of sensitivity to This may result from the inhibition of AZT, additional studies are needed to thymidylate kinase by AZT-MP (inhi• clarify these issues. Many clinical inves• bition constant, 8.6 1lM) [61]. However, tigators believe that the emergence of decreased concentrations of thymidine- AZT-insensitive isolates is a marker of 5'-triphosphate have not been found in impending clinical progression, but the other cell lines or with lower concent• clinical importance of the reduced viral rations of AZT [85, 86], and whether sensitivity to AZT is not fully known. The thymidine kinase inhibition contributes potential problem, however, underscores to the bone marrow toxicity induced by the urgency for additional experimental AZT is uncertain. Furman et al. [61] therapeutic agents and for regimens that originally reported that the levels of employ multiple agents. deoxycytidine-5' -triphosphate, another building block of DNA, fell in the pre• sence of AZT, but later they concluded Biochemical Pharmacology of AZT that the finding was due to a technical error [87]. It should be stressed that Unlike most nucleosides that enter cells murine cells handle AZT in a very differ• by specialized transport systems, AZT ent way from human cells. can ynter mammalian cells by passive, nonfacilitated diffusion [84]. Once inside, Pharmacokinetics of AZT the drug is phosphorylated to a triphos• phate form by a series of kinases that In culture, AZT inhibits new HIV infec• usually phosphorylate thymidine [51]. 3'• tion of lymphocytes at levels of 1-5 IlM Azido-2' ,3' -dideoxythymidine-5' -mono• (even under conditions' of high multiplic• phosphate (AZT-MP) is first produced ity of infection) [8]. The initial pharmaco• by thymidine kinase, and two addi• kinetic studies demonstrated that AZT tional phosphates are then added by the absorbed well orally (average oral sequential action of thymidylate ki• bioavailability, 63 %) and that peak nase and nucleotide diphosphate kinase plasma levels of3-4 IlM are attained 30- to form 3'-azido-2',3'-dideoxythymidine- 90 min after an oral dose of 200 mg is 5'-triphosphate (AZT-TP), the active ingested [27, 88, 89]. The peak plasma moiety [61, 72, 73]. The addition of a concentration is proportional to the second phosphate group to AZT-MP by amount administered, over a wide range thymidylate kinase is probably the rate• of doses [89]. The serum half-life is only limiting step for this process [61]. The 1.1 h [27, 89, 90]. For this reason, a dosing reaction occurs much more slowly (rela• every 4 h was chosen for phase II testing tive maximal velocity, 0.3%) than the in patients with advanced disease and phosphorylation of thymidine-5'-mono• then adopted as the recommended phosphate, the usual substrate for this en• schedule. It should be stressed, however, zyme [61], and human cells exposed to that there is not yet enough information AZT accumulate relatively high levels to determine the optimal dose or dosing of AZT-MP, but low levels of AZT-TP schedules for AZT. Less frequent dosing

LXXXV schedules and total daily doses work, but after a dose of AZT has been admin• further research is needed. It is likely that istered, levels in cerebrospinal fluid are total doses of approximately 600 mg per approximately 60 % of those in plasma day using an interval of eight hours be• (range, 10%-156%) [27, 38, 89, 99], tween doses would be adequate in most which indicates that AZT can enter the adults. The levels of intracellular AZT-TP brain by diffusion from the cerebrospinal (the activity moiety of AZT) have a half• fluid or possibly through capillaries in the life of about 3 h [61]. Thus, effective anti• brain. The clinical improvement that HIV activity may theoretically be at• occurs in patients with HIV dementia tained with an oral dosing interval of 8 or who are given AZT suggests that the drug 12 h. Alternatively, for the optimal thera• reaches the site of viral replication in the peutic benefit, it may be necessary to central nervous system [28, 99]. However, maintain constant plasma drug levels. AZT may not cross the blood-brain bar• Resolving these issues will require careful rier in all species [100]. controlled trials. Evidence suggests that cells belonging Approximately 15 % - 20 % of an ad• to the monocyte-macrophage series are ministered dose of AZT is excreted un• the most important target cells of HIV changed in the urine, and 75 % is meta• infection in the brain [101, 102]. These bolized by hepatic glucuronidation to nonproliferating cells have lower levels of form 3'-azido-2',3'-dideoxy-5'-glucuron• kinases than lymphocytes [103, 104] and ylthymidine, an apparently inert metab• dideoxynucleosides may not be efficiently olite that is also excreted into the urine phosphorylated in them. One study in• [90]. The enzyme or enzymes respon• dicated that AZT is poorly phosphory• sible for the glucuronidation of AZT lated in peripheral blood monocytes and may be inhibited by other compounds macrophages and does not protect these that share this pathway, and such com• cells against infection by the lym• pounds may prolong the half-life of the phadenopathy-associated virus strain drug [91]. In this regard, probenecid of HIV in vitro [104]. This study is not inhibits both hepatic glucuronidation correct. Two subsequent studies demon• and renal excretion and thus reduces the strated that low concentrations of AZT total body clearance of AZT by 65 %. and other dideoxynucleosides profound• Other drugs that undergo hepatic glu• ly inhibited the replication of a mono• curonidation and may, theoretically, in• cytotropic strain of HIV in monocytes hibit the metabolism of AZT include and macro phages [105, 106]. These re• nonsteroidal anti-inflammatory agents, sults are consistent with the observation narcotic analgesics, and sulfonamide that dementia induced by HIV may be at antibiotics. Until the interactions of such least temporarily reversed by AZT [21, drugs with AZT have been carefully in• 99, 107]. The potent activity of AZT in vestigated, clinicians should be aware monocytes and macrophages can perhaps that they may affect the metabolism of be explained by the observation that such AZT, and AZT may affect theirs. Finally, cells have very low levels of thymidine-5'• the metabolism of AZT could be slower triphosphate, the normal nucleotide that in patients with severe hepatic disease. It competes with AZT -TP at the level of should be stressed that different nu• reverse transcriptase [105]. Thus, the cleoside analogs will show individual ratio of AZT -TP to thymidine-5' -tri• pharmacokinetic profiles [27 - 30, 32, 63, phosphate may actually be higher in 89-98]. monocytes than it is in T cells [105]. Because HIV can infect cells in the Additional studies indicate that granu• central nervous system and cause demen• locyte-monocyte colony-stimulating fac• tia, antiretroviral agents used in the treat• tor, which stimulates the replication ment of AIDS should be capable of of HIV in monocytes [108, 109], in• penetrating the brain. Three to four hours creases the entry of AZT into these cells

LXXXVI and potentiates its actlVlty against more than 6 weeks of such prophylaxis HIV [109]. However, it is important to during the 52 weeks after their entry into use caution in extrapolating these data the study. Eliminating these patients to clinical applications. from the analysis does not affect the basis conclusion of the trial. Subsequent studies have suggested that patients with Clinical Application of AZT AIDS who receive AZT in conjunction with prophylactic therapy for Pneumocy• In the initial clinical studies of AZT at the stis pneumonia may have a lower mor• National Cancer Institute and Duke Uni• tality than those who receive AZT alone versity Medical Center, patients with [112]. However, it is important to stress AIDS or AIDS-related complex had im• that the precise role of Pneumocystis munologic, virologic, and clinical impro• pneumonia prophylaxis in patients who vement during 6 weeks of therapy [27,88]. are receiving antiretroviral therapy has (Three of the 19 patients from the original not been defined. phase I study, each of whom had AIDS• The phase II study also showed that related complex or Kaposi's sarcoma patients receiving AZT had a temporary when they entered the trial, were alive 3.5 increase in their CD4 + lymphocyte years after the initiation of therapy.) counts (average, 80/mm3), fewer oppor• Also, several patients with HIV dementia tunistic infections, and an average weight who were given AZT had substantial . gain of about 0.5 kg, as compared with improvement in their intellectual func• those receiving placebo [28]. Further• tion, accompanied in some by a normal• more, the results confirmed an observa• ization in the pattern of use of cerebral tion in the phase I study; some patients glucose (as assessed by positron emission with cognitive dysfunction induced by tomography) [99, 100]. On the basis of HIV improved when given AZT [99, 107]. these results, Wellcome Research Labo• Finally, patients who received AZT had a ratories began a multicenter, random• decreased viral load as compared with the ized, controlled trial of AZT in Feb• placebo group in assessments made by ruary 1986 among 282 patients with measuring levels of serum HIV p 24 anti• AIDS (after their first episode of Pneu• gen [113]. The ability to isolate HIV from mocystis carinii pneumonia) or severe cultured lymphocytes, however, was not AIDS-related complex [29, 111]. The trial affected, although there was a delay in the demonstrated a reduced mortality in the appearance of HIV in cultures [114]. patients receiving AZT. By September Although serum HIV p 24 antigen levels 1986, 19 patients taking placebo but only are an experimental clinical measure of one taking the drug had died [28]. After HIV replication, they may be affected by 36 weeks, 39.3 % of those taking placebo antibodies to p 24 and other factors. had died compared with 6.2 % of those Better methods of assessing viral load are taking AZT, and after 52 weeks the urgently needed. The polymerase chain cumulative mortality in the patients reaction will likely prove useful in this treated with AZT was still only 10.3 % regard [115]. On the basis of this trial's (no comparable figure is available for the results, AZT was approved for the treat• placebo group, because most were given ment of severe HIV infection in most AZT after September 1986, thus ending countries. In the United States, it was the control arm) [28]. Suppressive pro• approved in March 1987 for patients who phylaxis for Pneumocystis pneumonia have had P. carinii pneumonia or whose was not a formal protocol option in the 7- CD4+ cell count is below 200/mm3 . month randomized trial, although a few Evidence from the New York State De• patients randomly distributed between partment of Health indicates that the the two arms received it. Also, 19 of the survival of patients given a diagnosis of 144 originally assigned to AZT received AIDS in 1987 increased substantially

LXXXVII over that of patients whose disease was or low (or low-normal) serum folic acid diagnosed in previous years. It is exceed• or vitamin B12 lev.els before therapy ingly likely that the widespread use of begins [27, 67, 111]. Pending further AZT contributed to this trend [116]. The study, vitamin-replacement therapy may dose used in the phase II study is now be useful in patients with low levels of known to be higher than is 'necessary for these vitamins. The platelet count is gen• optimal effects. erally spared until late in the course of In both the phase I and phase II trials, AZT therapy. In fact, the drug can actu• it became apparent that the increase in ally induce increased platelet counts in the number of CD4+ lymphocytes in• patients with thrombocytopenia induced duced by AZT may be transient [27, 29, by HIV [122, 123]. 117], particularly in patients with fulmi• Other toxic effects of AZT include nant AIDS, whose CD4 count often nausea, vomiting, myalgias, myositis returns to baseline after 16-20 weeks of (particularly in patients who receive the therapy [28]. In this trend, the contri• drug for more than a year), headaches, butions of direct drug toxicity, altered abnormalities of liver function, and host defense mechanisms, and changes in bluish nail pigmentation [27, 111, 124- viral sensitivity to AZT are unclear. It is 126]. Very high doses can cause anxiety, highly probable that no antiretroviral confusion, and tremulousness [27, 117, agent will work to maximal advantage if 127]. These symptoms occasionally de• the host immune response is severely velop in patients receiving the current damaged. Some patients have had late recommended dosage. Finally, a few pa• increases in levels of HIV p 24 antigen tients have had seizure, an encephalo• even while receiving a constant dose of pathy similar to Wernicke's or Stevens• AZT [118]. In addition, AZT has a num• Johnson syndrome [128-130]. Thus, ber of toxic effects; the most frequent is although AZT decreases morbidity and suppression of bone marrow cells, and mortality among patients with severe anemia is its most frequent manifestation HIV infection, its use can be associated [27,28, 111, 117]. An increase in the mean with substantial toxicity, particularly in corpuscular volume of erythrocytes often those with advanced disease. occurs before frank anemia [27,111], but Slightly more than 2 years after AZT the dose should not be modified on the was first observed to inhibit the replica• basis of this measure. Patients receiving tion of HIV in vitro, it was approved by AZT may have megaloblastic changes in the Food and Drug Administration for bone marrow, maturational arrest of the treatment of AIDS. Because of this erythrocyte lineage, or hypoplastic (rare• extraordinarily rapid development, a ly aplastic) changes. Hypoplastic changes number of questions regarding its use can occur without an increase in mean remain unanswered. In AIDS, as in per• corpuscular volume [119,120]. Bone mar• haps no other condition, the line between row toxicity occurs more frequently in approved and experimental therapy is patients with established AIDS, and in difficult to draw. the phase II study, 45% of the patients Physicians frequently ask whether who had P. carinii pneumonia required AZT should be administered early in the transfusions or a reduction in dose during course of HIV infection. The drug ap• the first 6 months of AZT therapy [111]. pears to be relatively well tolerated at this In a subsequent open trial, only 21 % of stage [131], and recent results of a ran• the patients with AIDS could complete 6 domized trial have indicated that HIV• months of full-dose AZT treatment with• seropositive patients with less than out a reduction in dose or the interrup• 500 CD4 cellsjmm3 who were given AZT tion of therapy [121]. Marrow toxicity is had less frequent progression to severe also more frequent in patients with un• AIDS-related complex or AIDS than derlying anemia, low CD4+ cell counts, those given placebo [29]. As we learn

LXXXVIII more about the factors that make a Neurologic dysfunction and high-grade progression to AIDS highly likely (e.g., bacterial infections are much more evi• high serum HIV p 24 antigen levels or P2- dent, for example [137], and the patterns microglobulinemia) [132, 133], it may be of drug toxicity may differ. Ongoing possible to target AZT therapy to pa• studies at the National Cancer Institute tients who are at high risk. At present, we suggest that administration of AZT by cannot say that starting AZT early in an continuous intravenous infusion can re• asymptomatic phase of HIV infection verse certain neurologic symptoms as• provides a survival advantage over sociated with HIV in children with AIDS waiting until more symptomatic disease [138]. In some patients, the intelligence supervienes. quotients returned to what they had been In considering early intervention with before the disease developed [138]. How• AZT, it is of particular concern that the ever, the problems of bone marrow sup• drug may be carcinogenic or mutagenic pression limit that treatment. In both [134]. Its long-term effects are unknown. adults and children who have dementia Rodents exposed to high doses of AZT associated with AIDS, considerable bone for long periods can develop vaginal marrow suppression may be tolerable if neoplasms (principally nonmetastasizing antiretroviral therapy can reverse major squamous cell carcinomas). Male rodents neurologic deficits. Another unresolved did not develop tumors. The implications issue is whether AZT can prevent new of these sex- and site-specific tumorigenic HIV infection if it is given at the time of effects in rodents are not clear at this viral exposure. Kittens can be protected time, but the results provide a warning against feline leukemia virus (a retro• against complacency when using this· virus) by the administration of AZT at class of drug. It is worth emphasizing the time of infection [78]. Also, fetal mice again that AIDS itself makes the develop• can be protected against retroviral infec• ment of certain cancers more likely, and tion by the administration of AZT to AZT may be associated with the higher their mothers [139]. A short course of incidence of cancers in patients whose AZT at the time of exposure (e.g., after a immunosurveillance mechanisms are dis• serious needle-stick or laboratory acci• turbed, simply because it increases their dent) may therefore be useful. However, longevity. This has occurred in certain because it is mutagenic (and carcinogenic immunodeficiency disorders of child• in rodents) and can induce chromosomal hood, in which advances in the treatment abnormalities [134], its use in such a of infections have allowed patients to setting cannot be recommended except in survive longer [135]. Lymphomas have an approved protocol. developed in a significant subset of the Finally, even in patients for whom original phase I patients between 1 and 3 AZT is recommended, there is much to years after AZT therapy began. learn. The available evidence suggests but The use of AZT in children with AIDS does not prove that patients should con• is an area that is only now being inves• tinue to receive therapeutic doses for as tigated. The high incidence in certain long as they can tolerate the drug. AZT cities of cord blood samples that are was determined to be effective because it seropositive for HIV (for example, 1 in 80 lowered the risk of opportunistic infec• newborns in New York City are seroposi• tions and prolonged life. The late decline tive [116,136]) indicates that the number in the CD4+ cell counts is thus not an of children with AIDS will grow in the indication to stop therapy. However, it is near future. AIDS is dramatically alter• not yet clear how patients who have ing the landscape of obstetrical and hematologic toxicity, evidence of clinical pediatric care in many countries. The progression, or increased levels of serum manifestations of HIV infection in chil• HIV p 24 antigen while receiving AZT dren can differ from those in adults. should be managed. Certain factors that

LXXXIX stimulate bone marrow, such as erythro• 5'-triphosphate [14]. However, its anabo• poietin, granulocyte-stimulating factors, lic phosphorylation and activity may be or granulocyte-monocyte colony-stimu• affected by other nucleosides (e.g. thy• lating factors [109, 110], may reduce midine) [143]. ddC is exceptionally po• the suppression of bone marrow as• tent and the optimal dose to avoid neuro• sociated with AZT, and these approaches pathy is still under study. Total doses of need further research in an academic 2 mg per day or less are being tested in center. adults. It is worth emphasizing that this entire Biochemical Pharmacology class of nucleoside analogs represents a of Other Antiretroviral new area of clinical research. These drugs Dideoxynucleosides Including have antiretroviral activity; however, 2' ,3'-Dideoxyinosine (Didanosine) they also have considerable potential for side effects. These drugs should be ad• As noted, a number of dideoxynucleo• ministered only by physicians who are sides other than AZT have antiretro• well versed in their properties. viral activity in vitro [13 - 26], and several ddA and its immediate metabolite ddI studies of such agents are now enrolling are purine analogs with in vitro activity patients. There are substantial differences against HIV [13] and, unlike AZT, rela• in the rate at which human cells phos• tively little toxicity against bone marrow phorylate these compounds and in their precursor cells [144]. Within cells, ddA enzymatic pathways. These differences can be phosphorylated to its active 5'• are profoundly important to their antiret• triphosphate moiety [64]. It is also sus• roviral activities. 2' ,3'-Dideoxycytidine ceptible to deamination by adenosine (ddC) is, for example much more potent deaminase and forms ddI [64]. In human than 2',3'-dideoxythymidine (ddT) in plasma and cell extracts, this conversion most human cells because of differences occurs almost instantaneously [26, 64]. in its phosphorylation [31, 61-65, 67]. As mentioned, ddI has potent in vitro Since their rates of phosphorylation activity against HIV [13] because it can be differ between species [68], one cannot metabolized in human cells to form ddA- draw conclusions about their activities 5'-triphosphate through a complex series in human cells on the basis of their per• of reactions [65]. Interestingly, ddI uses formance in animal cells. 2',3'-Di• the enzyme 5'-nucleotidase to undergo deoxyadenosine (ddA), 2' ,3'-dideoxy• the initial phosphorylation it needs for inosine (ddI), and ddC are three com• activation and ultimate salvage back to pounds with potent activity against ddATP. Thus, for many purposes the two HIV in human T cells and monocytes in drugs can be considered identical. Once vitro under study [13, 14, 105] in clinical ddA and ddI are converted to ddA-5'• trials. One of the most active dideoxy• triphosphate in cells, they remain there nucleosides is ddC, a pyrimidine analog for a relatively long time - their intra• [13, 14]. Unlike many cytidine analogs, it cellular half-life is more than 12 h [145]. is resistant to deamination by the ubiqui• Thus, even with their short plasma half• tous enzyme cytidine deaminase [62, 63]. life, they may be clinically effective when It is, therefore, stable in plasma and administered relatively infrequently (e.g., bioavailable after oral administration. every 8-24 h). Unlike AZT or ddC, ddA After entering a cell, ddC is phosphory• and ddI undergo solvolysis (cleavage) in lated by a set of enzymes that usually acid reactions to form a purine base and phosphorylate deoxycytidine [62,63,141, dideoxyribose [64]. This may lower their 142]. Thus, ddC is activated by a different capacity for oral absorption, and they pathway than AZT. Also, ddC does not must be used with antacids or buffers. affect the levels of its competing dideoxy• High concentrations of the free purine nucleoside triphosphate, deoxycytidine- base of ddA, adenine, have been reported xc to cause renal damage [146]. The free base thymidylate kinase [148], whether it in• of ddI, hypoxanthine, does not have duces bone marrow toxicity remains un• similar toxicity, and it may, therefore, be determined. The dose-limiting toxicity is preferable for oral administration. Pre• peripheral neuropathy. Finally, a 3'• liminary results from a phase I trial and substituted uridine analog, 3'-azido-2',3'• ddI suggest that it is an active antiretro• deoxyuridine, which appears to be ac• viral drug [9]. Some patients have now tivated by the same enzymes that phos• received this drug for more than 2 years. phorylate AZT, has some anti-HIV activ• The major side effects to data have been a ity in vitro [19, 149]. All these dideoxy• reversible peripheral neuropathy and nucleosides have an intact oxacyclopen• acute pancreatitis. In some cases, the tane (sugar) ring. However, several acyc• pancreatitis may be lethal. It appears that lic compounds (adenallene, cytallene, and a prior history of pancreatic disease is risk a phosphonyl-methylethyl purine deriva• factor for this complication. Significant tive) also have activity against HIV in diarrhea and hypokalemia may occur. vitro as single agents. Such compounds These side effects seem to be dose related. provide new relation between structure Patients who have advanced disease or and activity and may be of value in who are debilitated have an increased risk developing a new class of anti-HIV for toxicity. At doses less than 8 mg/kg agents. per day, serious side effects are signifi• cantly less common than at higher doses. Clinical Research In an average adult, total doses should with Dideoxycytidine not exceed 500 mg per day. It is possible that even lower doses eventually will be Although AZT can prolong the lives of found active. The use of alcohol is con• patients with AIDS, there are some limi• traindicated in patients receiving ddI due tations to its use. The hematologic toxicity to the possibility of pancreatitis. of AZT is not inextricably linked to its Certain analogs of ddA (e.g., 2',3'• antiviral effect, and we can expect that dideoxy-2' -fluoro-ara-adenosine) are re• other agents will be worth exploring or sistant to acid hydrolysis [21] and, there• have different patterns of toxicity. In fore, may have better bioavailability vitro testing and studies of animal toxi• than ddA. Also, certain 2-halogen-sub• cology can provide clues as to which stituted forms of ddA are resistant to drugs are likely to have favorable thera• deamination [26], and they may be direct• peutic results. Ultimately, however, the ly phosphorylated and not follow the ddI issues can be resolved only by testing in pathway. Whether clinical studies with patients, and an effort is now under way such forms will produce compounds su• to test several of these agents in patients perior to ddA or ddI is not known; ddI with AIDS or related conditions. The has potent in vitro activity against HIV in first to be studied clinically (after AZT) its own right [13], and the issue of acid was ddC, which has potent activity instability can be addressed by simple against HIV in vitro at concentrations of measures such as the buffering of gastric 0.01-0.5 11M, depending on the viral secretion. dose used in the assay system [13, 69, 70]. Like AZT, a number of analogs of ddT It is well absorbed when given orally, and have been tested for activity against HIV peak levels of 0.1-0.2 11M can be attain• in vitro. Many were inactive, but a few ed after the oral administration of blocked the replication of HI V in human 0.03 mg/kg body weight [30]. Like AZT, T cells [14,16,17,22,147]. An unsatu• ddC has a half-life of slightly more than rated form of ddT called 2' ,3'-didehydro- 1 h. It differs from AZT in that it is 2',3'-dideoxythymidine is about as active excreted by the kidneys [30,150]. Finally, as AZT on a molar basis [16, 17]. Unlike ddC penetrates at least partially the AZT, it does not affect the activity of cerebrospinal fluid [30, 150].

XCI Both the initial study of ddC [30] and a way. Scientists at Hoffmann-LaRoche subsequent trial [:1:-51] found evidence of have begun studying a fluorinated ver• clinical activity against HIY. Nearly all sion of ddC which may spare mito• the patients who received daily doses of chondrial DNA polymerase. between 0.06 and 0.54 mg/kg had de• A continuation of the study of Merigan creased levels of serum HIV p24 antigen et al. [151] and a separate study organized [30,151], and most had small increases in by M. Gottlieb and W. Soo (personal the number of CD 4 + cells by week 2 [30]. communication) have shown that many Furthermore, some had an increase in patients can tolerate lower doses of ddC antigen-induced T-cell proliferation in (0.03 mg/kg per day) for 6 months or vitro [30]. The decrease in levels of p24 more; mild, readily reversible neuropathy antigen persisted in some patients for at developed in a minority of patients. This least several weeks after the drug was dose of ddC was associated with a decline withdrawn. In others, however, the im• in HIV p24 antigen levels and an increase munologic and virologic values moved in the number of CD4+ lymphocytes in toward baseline after several weeks de• most patients. Since the toxicity of ddC is spite the continued administration of strikingly different from that of AZT, ddC [30]. One purpose of these studies combining the two agents may reduce was to define the dose-limiting toxic overall toxicity. To test this approach, a effects of ddC. In a number of patients, group of patients with AIDS or AIDS• particularly those receiving higher doses, related complex followed a regimen alter• maculovesicular cutaneous eruptions, nating AZT (200 mg every 4 h) and ddC aphthous oral ulcerations, fever, and (0.09 or 0.18 mg/kg per day) therapy in 7- malaise developed after 1-4 weeks of day periods [30, 155]. It was hoped that therapy [30, 151, 152]. These symptoms neuropathy would not occur or would usually resolved in 1-2 weeks even with occur later with the intermittent admini• continued therapy. However, after stration of ddC. Preliminary results sug• several months of continuous therapy gest that the toxicity of both agents can be with daily doses of 0.06 mg/kg or more, significantly reduced. Some patients have most patients had a painful sensory now tolerated the regimen for more than motor peripheral neuropathy (involving 36 months [155] (unpublished data). mainly the feet) that became the dose• Overall, the patients had an average limiting toxic effect [30,151]. This neuro• increase of more than 70 CD 4 + pathy appeared earlier, was more severe, cells/mm3 at week 22, sustained decreases and lasted longer when the highest doses in serum p24 antigen levels, and a mean were tested; some patients receiving the weight gain of 5 kg (not caused by fluid highest doses still had persistent, moder• retention) [30, 155]. It is interesting to ate sensory loss and pain a year after the note that on low-dose or intermittent drug was discontinued [151,153]. Neuro• dosing regimens, once patients pass the 6- toxicity resolved much more quickly, month mark without neuropathy, they however, in patients receiving lower doses may have a significant probability of [30, 151, 153]. One metabolic product of avoiding serious neuropathy on con• ddC in human cells is dideoxycytidine tinued administration of ddC. Next to diphosphate choline [62], which could AZT, ddC has been given to patients conceivably contribute to the neuro• longer than any other dideoxynucleoside. pathy. Alternatively, the neuropathy may It is probable that ddC will find its best result from an inhibitory effect of ddC-5'• use as part of a combination regimen triphosphate on mitochondrial DNA with AZT. polymerase gamma (inhibition constant, 0.16 J.1M) [63, 154]. Thus, a search for ddC congeners which would not affect mitochondrial DNA synthesis is under

XCII Other Anti-HIV Agents in Preclinical monas endotoxin or ricin selectivity kill and Clinical Development cells expressing HIV envelope proteins in vitro [164, 165]. In patients, such agents This article has focused on the use of might selectively kill cells that can repli• dideoxynucleosides as antiviral agents, in cate HIV without being killed by the virus part because they are bioavailable after (e.g., macrophages). Unfortunately, CD4 oral administration and because data does not necessarily bind to the gp120 of from several studies support their virusta• primary isolates (as opposed to labora• tic activity in vitro. This is, however, by tory isolates) with high affinity. This may no means the only approach being inves• mean that very high doses ofCD4 need to tigated for the treatment of AIDS. The be used. genome and replicative cycle of HIV are Recently, low-molecular-weight dex• very complex, and several stages of repli• tran sulfate (7000-8000) was found to cation may, therefore, be potential targets inhibit the infectivity of T cells by HIV for antiretroviral therapy [31, 67, 156, [36-38]. This polyanionic polysaccharide 157]. Already, a number of agents that also appears to inhibit the initial bind• may act at various stages have been ing step [38]. A phase I/ll trial of orally defined. Although an extensive review of administered dextran sulfate suggested these other approaches is beyond the that it had little toxicity but also little scope of this article, a few points are effect on the number of CD4+ cells or worth stressing. Certain agents under serum p24 antigen levels [166]. However, study appear to act by inhibiting the dextran sulfate has since been found to be initial binding of HIV to its CD4 gly• very poorly absorbed when given orally, coprotein receptor on target cells [36-38, and studies of intravenous dextran sulfate 41, 42, 52-56, 158-160]. Using mole• are needed in order to assess this agent. cular biologic techniques, several groups Other molecules in this class are worth recently reported truncated soluble forms studying by parenteral administration. ofCD4 that lack the transmembrane and Recent advances in our understanding cytoplasmic domains [52-56]. At con• of the biochemistry of HIV replication centrations of 2-20 Ilg/ml, these forms have made the testing of new approaches inhibited the binding of HIV to T cells, to therapy possible. For example, anti• the formation of syncytia, and the infec• sense phosphorothioate oligodeoxynu• tion ofT cells [52 - 56]. A potential advan• cleo tides, which can. bind to specific tage of this approach is that soluble CD4 segments of the HIV genome, have is likely to inhibit, to some degree, all sequence-specific inhibitory effects that forms of HIV that use CD4 as the cell may result from the arrest of translation receptor. Also, agents that act at the cell after its hybridization to messenger RNA surface may block cell death induced by [48]. Interestingly, such compounds may syncytia, which can occur even when the also inhibit the replication of HIV in a target cell is not infected by HIV [38,161, manner that is not sequence specific [47]. 162]. Phase I trials of recombinant CD4 Alteration of the sugar moiety of viral are now under way. Second-generation glycoproteins (e.g., by inhibitors of versions of CD4 (such as CD4-im• trimming glucosidases) reduces the in• munoglobulin hybrid proteins) retain fectivity of the resulting viruses [57, 58]. their activity against HIV in vitro, but In addition to dideoxynucleosides, other may gain other desirable properties, such agents may act at the level of reverse as a longer circulating half-life [163]. A transcriptase. In particular, phosphono• phase I trial of such CD4-immu• formate, a pyrophosphate analog with no globulin hybrids is now under way activity against herpesvirus, has activity at the National Cancer Institute and at against HIV in vitro. Several pilot trials other academic centers. Also, forms of suggest that this drug can reduce serum recombinant CD4 linked to Pseudo- HIV p24 antigen levels in patients with

XCIII HIV infection [167, 168]. However, no effects. For example, the nucleoside ana• reliable oral formulation is available, log ribavirin inhibits the phosphorylation and this remains one drawback of this of AZT in vitro and blocks its activity drug. against HIV [50]. Ribavirin, however, There is a growing interest in develop• increases the phosphorylation of purine ing drugs that inhibit the protease ofHIY. analogs such as ddA through complex During the next few years, it is likely that mechanisms involving its ability to in• several protease inhibitors will enter clin• hibit inosine monophosphate dehydro• ical trials. genase [180]. Ribavirin can be given or• Several agents that act at different ally and may in theory potentiate the anti• stages of viral replication (e.g., in• HIV effects of ddA or ddl. Unfortu• terferon-a) have synergy with AZT in nately, one cannot predict from first vitro [34, 36, 40] and this could theoreti• principles whether this kind of potenti• cally result in better treatment in patients. ation would be good or bad. Once again, Interferon-a may be particularly interest• only carefully controlled clinical trials ing in this regard, because it also has a can resolve this issue. These in vitro direct antitumor effect against cutaneous observations should alert clinicians to the Kaposi's sarcoma [169-172]. In a similar possibility of unexpected interactions vein, the antiherpes drug acyclovir, which among agents, and they are an argument has little activity against HIV alone, can against ad hoc experimentation with anti• potentiate the anti-HIV activity of AZT retroviral therapies outside approved in vitro [14, 31]. In a pilot clinical trial, clinical trials. patients with AIDS or AIDS-related com• plex tolerated these drugs together for Conclusion 10-30 weeks [32]. A theoretical advan• tage of the regimen is that suppressing the In this article a number of new therapeu• replication of herpesvirus may second• tic agents and strategies have been discus• arily reduce the replication of HIV since sed. We now have at hand a number of a product of herpesvirus, ICPO (infected• approaches that can inhibit the replica• cell protein), can increase the initiation of tion of HIV in vitro. These approaches, HIV transcription [173, 174]. The pos• as well as a number of additional develop• sible suppression of human herpesvirus ments which are in the offing, can be 6 (human B-cell lymphotropic virus), expected to induce clinical improvement which can infect lymphoid cells [175], and prolong life even in patients with may also be relevant. (In a related fash• advanced AIDS. The progress against the ion, adenovirus enzyme-immunoassay mortality caused by AIDS is noteworthy product can also amplify HIV transcrip• in its own right, but there have been a tion [173]. Certain dideoxynucleosides number of advances that have improved can inhibit the replication of adenovirus the quality of life. For example, the [176] and thus may conceivably reduce incidence of dementias ascribable to the replication of HIV) Acyclovir has AIDS has been noted to have decreased been reported to be at least additive with after the introduction of AZT [181]. AZT in inhibiting the replication of HIV infection is probably a lifelong Epstein-Barr virus, and it could theoreti• process. It now appears highly likely that cally benefit patients infected with both a complete latency phase does not exist. that virus and HIV [177 -179]. Whether Rather, many, if not all patients, have AZT and acyclovir together offer a thera• circulating infectious HIV particles pre• peutic advantage over AZT alone is not sent in their plasma even when the disease yet clear. Only properly controlled clin• is clinically quiescent. Thus, it is perhaps ical trials can answer this point. unrealistic to expect a single drug to Not all combinations of anti-HIV provide therapy for all patients. The drugs have synergistic or even additive experiences with cancer therapy, as well

XCIV as the experiences with other serious principles of scientific drug development infections, suggest that a combination of and controlled trials are maintained. drugs will produce superior clinical out• come and less toxicity than any single therapy used alone [182]. Combination References therapy may also delay or prevent the emergence of viral resistance. Just as in 1. Gottlieb MS, Schroff R, Schanker HM, the treatment of certain leukemias or et al. (1981) Pneumocystis carinii pneu• advanced bacterial diseases, optimal monia and mucosal candidiasis in previ• therapy against HIV may require at least ously healthy homosexual men: evidence three different phases: induction, con• of a new acquired cellular immunodefi• ciency. N Engl M Med 305:1425-1431 It solidation, and maintenance. is worth 2. Masur H, Michelis MA, Greene JB, et al. noting that the drugs and biological ag• (1981) An outbreak of community ac• ents, as well as the relevant doses of such quired Pneumocystis carinii pneumonia: drugs and agents, may vary in each phase. initial manifestation of cellular immune At present, the only formally approved dysfunction. N Engl Med 305: 1431- antiretroviral agents are AZT and ddI. 1438 AZT has been proven to reduce morbidity 3. Siegal FP, Lopez C, Hammer GS, et al. and mortality above and beyond any effect (1981) Severe acquired immunodefi• of aerosolized pentamidine in severe cases ciency in male homosexuals, manifested by chronic perianal ulcerative herpes of AIDS [183]. Nevertheless, several vi• simplex lesions. N Engl J Med 305: rustatic drugs in the same general family 1439-1444 are being tested in patients, and it seems 4. Barre-Sinoussi F, Chermann JC, Rey F, highly probable that AZT is not the only et al. (1983) Isolation of a T-Iym• agent which eventually will prove effec• photropic retrovirus from a patient at tive against HIV. As with a number of risk for acquired immune deficiency syn• other therapies used in life-threatening drome (AIDS). Science 220:868-871 disorders, AZT may have a relatively low 5. Gallo RC, Salahuddin SZ, Popovic M, et therapeutic index in some patients. al. (1984) Frequent detection and isola• tion of cytopathic retroviruses (HTLV• Therefore, it is very important that clini• III) from patients with AIDS and at risk cians pay close attention to its clinical for AIDS. Science 224:500-503 pharmacology and to the specific patient 6. Popovic M, Sarngadharan MG, Read E, responses that occur following initial Gallo RC (1984) Detection isolation, and therapy. As new experimental agents are continuous production of cytopathic ret• tested and become more widely available, roviruses (HTLV-III) from patients with it is important that careful adherence to AIDS and pre-AIDS. Science 224:497- the principles of clinical trials be a major 500 priority if we are to succeed in the mission 7. Mitsuya H, Popovic M, Yarchoan R, Matsushita S, Gallo RC, Broder S (1984) of developing better therapeutic options. Suramin protection of T-cells in vitro As simpler assays to measure plasma against infectivity and cytopathic effect drug levels become available [184, 185], ofHTLV-III. Science 226:172-174 their results conceivably may provide 8. Mitsuya H, Weinhold KJ, Furman PA, et useful data in the optimal management of al. (1985) 3'-Azido-3'-deoxythymidine HIV infections. A number of studies are (BW A509U): an antiviral agent that now under way to test whether various inhibits the infectivity and cytopathic agents should be administered to patients effect of human T-Iymphotropic virus with early HIV infections and to explore type III/lymphadenopathy-associated vi• rus in vitro. Proc Natl Acad Sci USA other therapeutic regimens. In the com• 82:7096-7100 ing decade, it seems highly probable that 9. Ostertag W, Roesler G, Krieg CJ, et al. major advances will occur against the (1974) Induction of endogenous virus death and suffering caused by HIV, but and of thymidine kinase by bromo• this progress can be ensured only if the deoxy-uridine in cell cultures transform-

XCV ed by Friend virus. Proc Natl Acad Sci replication by 3' -azido-2',3' -dideoxyuri• USA 71 :4980-4985 dine (CS-87) [Abs}ract]. J Clin Biochem 10. De Clercq E (1979) Suramin: a potent [Suppl] 11 D: 74 inhibitor of the reverse transcriptase of 20. Kim C-H, Marquez VE, Broder S, Mit• RNA tumor viruses. Cancer Lett 8:9-22 suya H, Driscoll JS (1987) Potential anti• 11. Furmanski P, Bourguignon GJ, Bolles AIDS drugs: 2',3'-dideoxycytidine ana• CS, Corombos JD, Das MR (1980) Inhi• logues. J Med Chern 30:862-866 bition by 2',3'-dideoxythymidine of ret• 21. Marquez VE, Tseng CK, Kelly JA, et al. roviral infection of mouse and human (1987) 2',3'-Dideoxy-2'-fluoro-ara-A: an cells. Cancer Lett 8: 307 - 315 acid-stable purine nucleoside active 12. Waqar MA, Evans' MJ, Manly KF, against human immunodeficiency virus Hughes RG, Huberman JA (1984) (HIV). Biochem Pharmacol 36:2719- Effects of 2',3'-dideoxynucleosides on 2722 mammalian cells and viruses. J Cell 22. Herdewijn P, Balzarini J, de Clercq E, et Physiol 121 :402-408 al. (1987) 3'-Substituted 2',3'-dideoxy• 13. Mitsuya H, Broders S (1986) Inhibition nucleoside analogues as potential anti• of the in vitro infectivity and cytopathic HIV (HTLV -III/LA V) agents. J Med effect on human T-Iymphotropic virus Chern 30:1270-1278 ype III/lymphadenopathy-associated vir• 23. Balzarini J, Robins MJ, Zou RM, us (HTLV-III/LAV) by 2',3'-dideoxy• Herdewijn P, de C1ercq E (1987) The nucleosides. Proc Nat! Acad Sci USA 2',3' -dideoxyriboside of 2,6-diaminopu• 83:1911-1915 rine and its 2',2' -didehydro derivative 14. Mitsuya H, Matsukura M, Broder S inhibit the deamination of 2',3' -dideoxy• (1987) Rapid in vitro screening systems adenosine, an inhibitor of human for assessing activity of agents against immunodeficiency virus (HIV) repli• HTLV -III/LAV. In: Broder S (ed) AIDS: cation. Biochem Biophys Res Commun modern concepts and therapeutic chal• 145:277-283 lenges. Dekker, New York, pp 303-333 24. Baba M, Pauwels R, Balzarini J, Herde• 15. Balzarini J, Pauwels R, Herdewijn P, et wijn P, de Clercq E (1987) Selective al. (1986) Potent and selective anti• inhibition of human immunodeficiency HTLV-III/LAV activity of 2',3'-dide• virus (HIV) by 3'-azido-2',3'-dideoxy• oxycytidine, the 2',3' unsaturated deri• guanosine in vitro. Biochem Biophys vative of 2',3' -dedeoxycytidine. Biochem Res Comrnun 145:1080-1086 Biophys Res Commun 140:735-742 25. Herdewijn P, Pauwels R, Baba M, Balza• 16. Hamamoto Y, Nakashima H, Matsui T, rini J, de Clercq E (1987) Synthesis and Matsuda A, Ueda T, Yamamoto N anti-HIV activity of various 2'- and 3'• (1987) Inhibitory effect of 2',3'-dide• substituted 2' ,3'.-dideoxyadenosines: a hydro-2',3'-dideoxynucleosides on infec• structure-activity analysis. J Med Chern tivity, cytopathic effects, and replica• 30:2131-2137 tion of human immunodeficiency virus. 26. Haertle T, Carrera CJ, Wasson DB, Antimicrob Agents Chemother 31:907- Sowers LC, Richman DD, Carson DA 910 (1988) Metabolism and anti-human im• 17. Balzarini J, Kang G-J, Dalal M, et al. munodeficiency virus-l activation of 2- (1987) The anti-HTLV-III (anti-HIV) halo-2' ,3' -dideoxyadenosine derivatives. and cytotoxic activity of 2' ,3' -didehydro- J Bioi Chern 263:5870-5875 2',3'-dideoxyribonucleosides: a compa• 27. Yarchoan R, Klecker RW, Weinhold KJ, rison with their parenteral 2',3'-dide• et al. (1986) Administration of 3'-azido oxyribonucleosides. Mol Pharmacol 32: 3'-deoxythymidine, an inhibitor of 162-167 HTLV -III/LAV replication to patients 18. Lin T-S, Schinazi RF, PrusoffWH (1987) with AIDS or AIDS-related complex. Potent and selective in vitro activity of 3'• Lancet 1 :575-580 deoxythymidin-2' -ene (3-deoxy-2',3' -di• 28. Fischl MA, Richman DD, Grieco MH, et dehydrothymidine) against human im• al. (1987) The efficacy of azidothymidine munodeficiency virus. Biochem Phar• (AZT) in the treatment of patients with macoI36:2713-2718 AIDS and AIDS-related complex: a 19. Schinazi RF, Chu C-K, Ahn M-K, et al. double-blind, placebo-controlled trial. N (1987) Selective in vitro inhibition of Engl J Med 317:185-191 human immunodeficiency virus (HIV) 29. Volberding P A, Lagakos SW, Koch MA,

XCV! et al. (1990) Zidovudine in asymptomatic 39. Ho DD, Hartshorn KL, Rota TR, et al. human immunodeficiency virus infec• (1985) Recombinant human interferon tion: a controlled trial in persons with alfa-A suppresses HTLV-III replication fewer than 500 CD4-positive cells per in vitro. Lancet 1 :602-604 cubic millimeter. N EnglJ Med 322: 941- 40. Hartshorn KL, Vogt MW, Chou T-C, et 949 al. (1987) Synergistic inhibition of human 30. Yarchoan R, Perno CF, Thomas RV, et immunodeficiency virus in vitro by al. (1988) Phase I studies of 2',3'• azidothymidine and recombinant A in• dideoxycytidine in severe human immu• terferon. Antimicrob Agents Chemother nodeficiency virus infection as a single 31: 168-172 agent and alternating with zidovudine 41. Matsushita S, Robert-GuroffM, Rusche (AZT). Lancet 1 :76-81 J, et al. (1988) Characterization of a 31. Mitsuya H, Broders S (1987) Strategies human immunodeficiency virus neu• for antiviral therapy in AIDS. Nature tralizing monoclonal antibody and 325:773-778 mapping of the neutralizing epitope. J 32. Surbone A, Yarchoan R, McAtee N, et ViroI62:2107-2141 al. (1988) Treatment of the acquired 42. Pert CB, Hill JM, Ruff MR, et al. (1986) immunodeficiency syndrome (AIDS) Octapeptides deduced from the neuro• and AIDS-related complex with a regi• peptide receptor-like pattern of antigen men of 3'-azido-2',3'-dideoxythymidine T4 in brain potently inhibit human im• (azidothymidine or zidovudine) and acy• munodeficiency virus receptor binding clovir: a pilot study. Ann Intern Med and T-cell infectivity. Proc Natl Acad Sci 108: 534-540 USA 83:9254-9258 33. Montefiori DC, Mitchell WM (1987) 43. Wetterberg L, Alexius B, Saaf J, Antiviral activity of mismatched double• Siinnerborg A, Britton S, Pert C (1987) stranded RNA against human immuno• Peptide T in treatment of AIDS. Lancet deficiency virus in vitro. Proc Natl Acad 1: 159 Sci USA 84:2985-2989 44. Barnes DH (1987) Debate over potential 34. Mitchell WM, Montefiori DC, Robinson AIDS drug. Science 237: 128 -130 WE Jr, Strayer DR, Carter WA (1987) 45. Sandstrom EG, Kaplan JC, Byington Mismatched double-sized RNA (ampli• RE, Hirsch MS (1985) Inhibition of gen) reduces concentration ofzidovudine human T celllymphotropic virus type II (azidothymidine) required for in vitro in vitro by phosphonoformate. Lancet inhibition of human immunodeficiency 1: 1480-1482 virus. Lancet 1 :890-892 46. Sarin PS, Taguchi Y, Sun D, Thornton A, 35. Carter W A, Strayer DR, Brodsky I, et al. Gallo RC, Oberg B (1985) Inhibition of (1987) Clinical, immunological, and viro• HTLV-III/LA V replication by foscarnet. logical effects of ampligen, a mismat• Biochem Pharmacol 34:4075-4079 ched double-stranded RNA, in patients 47. Matsukura M, Shinozuka K, Zon G, et with AIDS or AIDS-related complex. al. (1987) Phosphorothioate analogs of Lancet 1:1286-1292 oligodeoxynucleotides: inhibitors of re• 36. Ueno R, Kuno S (1987) Dextran sul• plication and cytopathic effects of hu• phate, a potent anti-HIV agent in vitro man immunodeficiency virus. Proc Nat! having synergism with zidovudine. Lan• Acad Sci USA 84:7706-7710 cet 1: 1379 48. Matsukura M, Shinozuka K, Zon G, et 37. Ito M, Baba M, Sato A, Pauwels R, de al. (1988) Phosphorothioate analogs of Clercq E, Shigeta S (1987) Inhibitory oligodeoxynucleotide inhibit viral repli• effect of dextran sulfate and heparin on cation of human immunodeficiency virus the replication of human immunodefi• (HIV): inhibition of de novo infection in ciency virus (HIV) in vitro. Antiviral uninfected cells and regulation of viral Res 7:361-367 expression in chronically infected cells 38. Mitsuya H, Looney DJ, Kuno S, Ueno (Abstr). Clin Res 36:463A R, Wong-Staal F, Broder S (1988) Dex• 49. McCormick JB, Getchell JP, Mitchell tran sulfate suppression of viruses in SW, Hicks DR (1984) Ribavirin sup• the HIV family: inhibition of virion presses replication of lymphadenopathy• binding to CD4+ cells. Science 240: associated virus in cultures of human 646-649 adult T lymphocytes. Lancet 2: 1367- 1369

XCVII 50. Vogt MW, Hartshorn KL, Furman PA, (1986) Initial studies on the cellular et al. (1987) Ribavirin antagonizes the pharmacology of 2',3'-dideoxycytidine, effect of azidothymidine on HIV repli• an inhibitor ofHTLV-III infectivity. Bio• cation. Science 235: 1367 -1379 chern Pharmacol 35:2065-2068 51. Anand R, Moore 1, Feorino P, Curran 1, 63. Starnes MC, Cheng YC (1987) Cellular Srinivasan A (1986) Rifabutine inhibits metabolism of 2',3' -dideoxycytidine, a HTLV-III. Lancet 1:97-98 compound active against human immu• 52. Smith DH, Byrn RA, Marsters SA, nodeficiency virus in vitro. 1 BioI Chern Gregory T, Groopman JE, Capon Dl 262:988-991 (1987) Blocking of HIV I infectivity by a 64. Cooney DA, Ahluwalia G, Mitsuya H, et soluble, secreted form of CD4 antigen. al. (1987) Initial studies on the cellular Science 238: 1704-1709 pharmacology of 2',3' -dideoxyadeno• 53. Fisher RA, Bertonis 1M, Meier W, et al. sine, an inhibitor of HTLV-III infec• (1988) HIV infection is blocked in vitro tivity. Biochem Pharmacol 36: 1765- by recombinant soluble CD4. Nature 1768 331:76-78 65. Ahluwalia G, Cooney DA, Mitsuya H, et 54. Hussey RE, Richardson NE, Kowalski al. (1987) Initial studies on the cellular M, et al. (1988) A soluble CD4 protein pharmacology of 2',3'-dideoxyinosine, selectively inhibits HIV replication and an inhibitor of HIV infectivity. Biochem syncytium formation. Nature 331 :78-81 Pharmacol 36: 3797 - 3800 55. Deen KC, McDouga1JS, Inacker R, et al. 66. Mitsuya H, larrett RF, Matsukura M, et (1988) A soluble form of CD4 (T4) al. (1987) Long-term inhibitor of hu• protein inhibits AIDS virus infection. man T Iymphotropic virus type III/lym• Nature 331 :82-84 phadenopathy-associated virus (human 56. Traunecker A, Luke W, Karjalainen K immunodeficiency virus) DNA syn• (1988) Soluble CD4 molecules neutralize thesis and RNA expression in T cells pro• human immunodeficiency virus type 1. tected by 2',3' -dideoxynucleosides in Nature 331:84-86 vitro. Proc Natl Acad Sci USA 84:2033- 57. Walker BD, Kowalski M, Goh WC, et al. 2037 (1987) Inhibition of human immunode• 67. Yarchoan R, Broder S (1987) Develop• ficiency virus syncytium formation and ment of antiretroviral therapy for the virus replication by castano spermine. acquired immunodeficiency syndrome Proc Nat! Acad Sci USA 84:8120-8124 and related disorders: a progress report. 58. Gruters RA, Neefjes 11, Tersmette M, et N Engl 1 Med 316:557-564 al. (1987) Interference with HIV-induced 68. Dahlberg lE, Mitsuya H, Blam SB, syncytium formation and viral infectivity Broder S, Aaronson SA (1987) Broad by inhibitors of trimming glucosidase. spectrum antiretroviral activity of 2',3'• Nature 330:74-77 dideoxynucleosides. Proc Nat! Acad Sci 59. Hayashi S, Phadtare S, Zemlicka 1, Mat• USA 84:2469-2473 sukura M, Mitsuya H, Broder S (1988) 69. Barber AM, Hizi A, Maizel lV, Hughes Adenallene and cytallene: acyclic-nu• SH (1990) HIV-1 reverse transcriptase: cleoside analogues that inhibit replica• structure predictions for the polymerase tion and cytopathic effect of human domain. AIDS Res Hum Retroviruses immunodeficiency virus (HI V) in vitro. 6: 1061-1072 Proc Nat! Acad Sci USA 85:6127-6131 70. Mitsuya H, Dahlberg lE, Spigelman Z, 60. Chin 1 (1990) Current and future dimen• et al. (1988) 2',3'-Dideoxynuc1eosides: sions of the HIV / AIDS pandemic in broad spectrum antiretroviral activity women and children. Lancet 336: 221- and mechanism of action. In: Bolognesi 224 D (ed) Human retroviruses, cancer, and 61. Furman PA, Fyfe lA, St Clair MH, et al. AIDS: approaches to prevention and (1986) Phosphorylation of 3'-azido-3'• therapy. Liss, New York, pp407-421 deoxythymidine and selective interaction 71. Hoa Z, Dalal M, Cooney DA, et al. of the 5'-triphosphate with human im• (1987) A comparison of 2',3'-di• munodeficiency virus reverse transcrip• deoxynuc1eoside 5'-triphosphates as in• tase. Proc Natl Acad Sci USA 83: 8333- hibitors of retroviral reverse transcrip• 8337 tases (Abstr). Proc Am Assoc Cancer 62. Cooney DA, Dalal M, Mitsuya H, et al. Res 28:323

XCVIII 72. Vrang L, Bazin H, Remaud G, Chattop• zidovudine (AZT) isolated during pro• adhyaya J, Oberg B (1987) Inhibition of longed therapy. Science 243:1731~1734 the reverse transcriptase from HIV by 3'• 83. Larder BA, Purifoy DJM, Powell KL, azido-3' -deoxythymidine triphosphate Darby G (1987) Site-specific mutagenesis and its threo analogue. Antiviral Res of AIDs virus reverse transcriptase. Na• 7:139~149 ture 327:716~717 73. Cheng YC, Dutschman GE, Bastow KF, 84. Zimmerman TP, Mahony WB, Prus KL Sarngadharan MG, Ting RY (1987) (1987) 3'-Azido-3'-deoxythymidine: an Human immunodeficiency virus reverse unusual nucleoside analogue that perme• transcriptase: general properties and its ates the membrane of human erythro• interactions with nucleoside triphosphate cytes and lymphocytes by nonfacilitated analogs. J Bioi Chern 262:2187~2189 diffusion. J Bioi Chern 262:5748~5754 74. Richman DD (1987) Dideoxynucleosides 85. Frick LW, Nelson DJ, St Clair MH, are less inhibitory in vitro against human Furman PA, Krenitsky TA (1988) Effects immunodeficiency virus type 2 (HIV-2) of 3'-azido-3'-deoxythymidine on the de• than against HIV-1. Antimicrob Agents oxynucleotide triphosphate pools of cul• Chemother 31: 1879~ 1881 tured human cells. Biochem Biophys 75. Mitsuya H, Broder S (1988) Inhibition of Res Commun 154:124~129 infectivity and replication of HIV-2 and 86. Hao Z, Cooney DA, Hartman NR, et al. SIV in helper T-cells by 2',3'-dideoxy• (1988) Factors determining the activity nucleosides in vitro. AIDS Res Hum of 2',3'-dideoxynucleosides in suppress• Retroviruses 4: 107 ~ 113 ing human immunodeficiency virus in 76. Matsushita S, Mitsuya H, Reitz MS, vitro. Mol Pharmacol 34:431 ~435 Broder S (1987) Pharmacological inhi• 87. Harrington JA, Miller WH, Spector T bition of in vitro infectivity of human T (1987) Effector studies of 3'-azido• lymphotropic virus type 1. J Clin Invest thymidine nucleotides with human ribo• 80:394~400 nucleotide reductase. Biochem Pharma• 77. Ruprecht RM, O'Brien LG, Rossoni LD, col 36:3757~3761 Nusinoff-Lehrman S (1986) Suppression 88. Yarchoan R, Broder S (1987) Strategies of mouse viraemia and retroviral disease for the pharmacological intervention by 3' -azido-3' -deoxythymidine. Nature against HTLV-III/LA V. In: Broder S 323:467~469 (ed) AIDS: modern concepts and thera• 78. Tavares L, Roneker C, Johnston K, peutic challenges. Dekker, New York, Lehrman SN, de Noronha F (1987) 3'• pp 335~360 Azido 3'-deoxythymidine in feline leuke• 89. KleckerRW Jr, CollinsJM, YarchoanR, mia virus-infected cats: a model for et al. (1987) Plasma and cerebrospinal therapy and prophylaxis of AIDS. Can• fluid pharmacokinetics of 3' -azido-3'• cer Res 47:3190~3194 deoxythymidine: a novel pyrimidine ana• 79. Kassianides C, Hoofnagle J, Miller RH, log with potential application for the et al. (1989) Inhibition of duck hepatitis B treatment of patients with AIDS and virus replication by 2',3' -dideoxycyti• related diseases. Clin Pharmacol Ther dine: a potent inhibitor of reverse trans• 41:407~412 criptase. Gastroenterology 97: 1275~ 1280 90. Blum MR, Liao SH, Good SS, de Mir• 80. Summers J, Mason WS (1982) Replica• anda P (1988) Pharmacokinetics and tion of the genome of a hepatitis B-like bioavailability of zidovudine in humans. virus by reverse transcription of an RNA Am J Med [SuppI2A] 85:189~194 intermediate. Cell 29:403~415 91. De Miranda P, Good SS, Blum MR, 81. Zimmerman F, Biessert L, von Briesen H, Thomas RV, Yarchoan R, Broder S et al. (1988) Development of HIV• (1986) The effect of probenecid on the variants with higher resistance against pharmacokinetic disposition of azidothy• HIV under treatment with AZT (Abstr). midine (AZT) (Abstr). 2nd International 4th International Conference on AIDS, Conference on AIDS, June 23~25, Paris June 12~16, Stockholm, vol 2. Swedish 92. Klecker RW Jr, Collins JM, Yarchoan Ministry of Health and Social Affairs, RC, et al. (1988) Pharmacokinetics of Stockholm, p 180 2',3' -dideoxycytidine in patients with 82. Larder BA, Darby G, Richman DD AIDS and related disorders. J Clin Phar• (1989) HIV with reduced sensitivity to macol 28: 837 ~842

Ie 93. Blum MR, Liao SH, de Miranda P (1982) 105. Perno C-F, Yarchoan R, Cooney DA, et Overview of acyclovir pharmacokinetic al. (1988) Inhibition of human immuno• disposition in adults and children. Am J deficiency virus (HIV -l/HTLV IIIBa - L) Med [SuppllA] 73:186-192 replication in fresh and cultured human 94. Furman P A, de Miranda P, St Clair MH, peripheral blood monocytes/macro• Elion GB (1981) Metabolism of acyclovir phages by azidothymidine and related in virus-infected and uninfected cells. 2',3'-dideoxynucleosides. J Exp Med 168: Antimicrob Agents Chemother 20:518- 1111-1125 524 106. Skinner MA, Matthews TJ, Greenwall 95. Laskin OL, Longstreth JA, Hart CC, et TK, Bolognesi DP, Hebdon M (1988) al. (1987) Ribavirin disposition in high• AZT inhibits HIV -1 replication in mono• risk patients for acquired immunodefi• cytes. J AIDS 1:162-163 ciency syndrome. Clin Pharmacol Ther 107. Schmitt FA, Bigley JW, McKinnis R, et 41:546-555 al. (1988) Neuropsychological outcome 96. Crumpacker C, Bubley G, Lucey D, of zidovudine (AZT) treatment of pa• Hussey S, Connor J (1986) Ribavirin tients with AIDS and AIDS-related com• enters cerebrospinal fluid. Lancet 2:45- plex. N Engl J Med 319:1573-1578 46 108. Koyanagi Y, O'Brien WA, Zhao JQ, 97. Yarchoan R, Mitsuya H, Thomas RV, et Golde DW, Gasson JC, Chen ISY (1988) al. (1989) In vivo activity against HIV Cytokines alter production of HIV-l and favorable toxicity profile of 2',3'• from primary mononuclear phagocytes. dideoxyinosine. Science 245:412-415 Science 241:1673-1675 98. Hartman NR, Yarchoan R, Kelly JA, 109. Perno C-F, Yarchoan R, Cooney DA, et Cooney DA, Broder S, Johns DG (1989) al. (1989) Replication of human im• Cellular and clinical pharmacokinetics of munodeficiency virus in monocytes: 2',3'-dideoxyadenosine and 2',3'-dide• granulocyte/macrophage colony-stimu• oxyinosine (Abstr). 5th Internatio• lating factor (GM-CSF) potentiates nal Conference on AIDS, June 4-9, viral production yet enhances the anti• Montreal viral effect mediated by 3'-azido-2',3'• 99. Yarchoan R, Thomas RV, Grafman J, et dideoxythymidine AZT and other di• al. (1988) Long-term administration of deoxynucleoside congeners of thymi• 3'-azido-2',3'-dideoxythymidine to pa• dine. J Exp Med 169:933-951 tients with AIDS-related neurological 110. Brunetti A, Berg C, di Chiro G, et al. disease. Ann Neurol [Suppl] 23:S82- (1989) Reversal of brain metabolic ab• S87 normalities following treatment of AIDS 100. Ellison S, Terasaki T, Pardridge WM dementia complex with 3'-azido-2',3'• (1988) AZT and dideoxynucleosides do dideoxythymidine (AZT, zidovudine): A not cross the blood-brain barrier (Abstr). PET-FDG study. J Nucl Med 30:581- Clin Res 36:117A 590 101. Gartner S, Markovits P, Markovitz DM, 111. Richman DD, Fischl MA, Grieco MH, et Kaplan MH, Gallo RC, Popovic M al. (1987) The toxicity of azidothymidine (1986) The role of mononuclear pha• (AZT) in the treatment of patients with gocytes in HTLV - III/LA V infection. AIDS and AIDS-related complex: a Science 233:215-219 double-blind, placebo-controlled trial. N 102. Koenig S, Gendelman HE, Orenstein Engl J Med 317:192-197 JM, et al. (1986) Detection of AIDs virus 112. Andrews JC, McManus M, Rogers G in macrophages in brain tissue from (1988) AZT Collaborative Working AIDS patients with encephalopathy. Group. Clinical benefits of concurrent Science 233:1089-1093 administration of zidovudine and ther• 103. Stadecker MJ, Unanue ER (1979) The apy for the suppression of Pneumocy• regulation of thymidine secretion by ma• stis carinii. 4th International Conference crophages. J ImmunoI123:568-571 on AIDS, June 12-16, Stockholm, vol 1. 104. Richman DD, Kornbluth RS, Carson Swedish Ministry of Health and Social DA (1987) Failure of dideoxynucleosides Affairs, Stockholm, p 412 to inhibit human immunodeficiency virus 113. Chaisson RE, Allain J-P, Leuther M, replication in cultured human macro• Volberding PA (1986) Significant phages. J Exp Med 166:1144-1149 changes in HIV antigen level in the serum c of patients treated with azidothymidine. with zidovudine therapy of AIDs and N Engl J Med 315:1610-1611 ARC. N Engl J Med 318:708 114. Makuch RW, Parks WP (1988) Statistical 125. Furth PA, Kazakis AM (1987) Nail pig• methods for the analysis of HIV-l core mentation changes associated with azido• polypeptide antigen data in clinical thymidine (zidovudine). Ann Intern Med studies. AIDS Res Hum Retroviruses 107:350 4:305-316 126. Melamed AJ, Muller RJ, Gold JWM, et 115. Kwok S, Mack DH, Mullis KB, et al. al. (1987) Possible zidovudine-induced (1987) Identification of human immuno• hepatotoxicity. JAMA 258:2063 deficiency virus sequences by using in 127. Bach MC (1987) Possible drug interac• vitro enzymatic amplication and oli• tion during therapy with azidothymidine gomer cleavage detection. J Virol 1: and acyclovir for AIDS. N Engl J Med 1690-1694 316:547 116. AIDS in New York State: through 1988. 128. Hagler DN, Frame PT (1986) Azidothy• State of New York Department of midine neurotoxicity. Lancet 2:1392- Health, Albany 1393 117. Yarchoan R, Broder S (1987) Prelimi• 129. Davtyan DG, Vinters HV (1987) Wer• nary results on the use of dideoxynu• nicke's encephalopathy in AIDS pa• cleosides in the therapy of AIDS. In: tient treated with zidovudine. Lancet Chanock RM, Lerner RA, Brown F, 1 :919-920 Ginsberg HS (eds) Vaccine 87. Cold 130. Creagh-Kirk T, Doi P, Andrews E, et al. Spring Harbor Press, Cold Spring Har• (1988) Survival experience among pa• bor, pp 214-224 tients with AIDS receiving zidovudine: 118. Reiss P, Lange JMA, Boucher CA, Dan• follow-up of patients in a compassionate ner SA, Goudsmit J (1988) Resumption plea program. JAMA 260:3009-3015 of HIV antigen production during con• 131. De Wolf F, Lange JMA, Goudsmit J, et tinuous zidovudine treatment. Lancet al. (1988) Effect of zidovudine on serum 1 :421 human immunodeficiency virus antigen 119. Gill PS, Rarick M, Brynes RK, Causey levels in symptom-free subjects. Lancet D, Loureiro C, Levine AM (1987) Azido• 1 :372-376 thymidine associated with bone marrow 132. Allain J-P, Laurian Y, Paul DA, et al. failure in the acquired immunodeficiency (1987) Long-term evaluation of HIV . syndrome (AIDS). Ann Intern Med antigen and antibodies to p24 and gp41 107: 502-505 in patients with hemophilia: potential 120. Walker RE, Parker RI, Kovacs JA, et al. clinical importance. N Engl J Med (1988) Anemia and erythropoiesis in pa• 317:1114-1121 tients with the acquired immunodefi• 133. Moss AR, Bacchetti P, Osmond D, et al. ciency syndrome (AIDS) and Kaposi (1988) Seropositivity for HIV and the sarcoma treated with zirdovudine. Ann development of AIDS or AIDS-related Intern Med 108:372-376 condition: three-year follow up of the 121. Dournon E, Matheron S, Rozenbaum W, San Francisco General Hospital cohort. et al. (1988) Effects of zidovudine in 365 Br Med J 296:745-750 consecutive patients with AIDS or 134. Retrovir Capsules Package Insert (Zido• AIDS-related complex. Lancet 2: 1297- ·vudine). Burroughs-Wellcome, Research 1302 Triangle Park, N.C. 122. Gottlieb MS, Wolfe PR, Chafey S (1987) 135. Waldmann TA, Misiti J; Nelson DL, Response of AIDS-related thrombocyt• Kraemer KH (1983) Ataxia telangiec• openia to intravenous and oral azido• tasis: a multisystem disease with im• thymidine (3' -azido-3' -deoxythymidine). munodeficiency impaired organ matura• AIDS Res Hum Retroviruses 3: 109-114 tion x-ray hypersensitivity, and a high 123. Hymes KB, Greene JB, Karpatkin S incidence of neoplasia. Ann Intern Med (1988) The effect of azidothymidine on 99:367-379 HIV-related thrombocytopenia. N Engl J 136. Hoff R, Berardi VP, Weiblen BJ, Med 318:516-517 Mahoney-Trout L, Mitchell ML, Grady 124. Bessen LJ, Greene JB, Louie E, Seitz• GF (1988) Seroprevalence of human im• man P, Weinberg H (1988) Severe munodeficiency virus among childbear• polymyositis-like syndrome associated ing women: estimation by testing samples

CI of blood from newborns. N Engl J Med 147. Greengrass CW, Hoople DW, Street SD, 318:525-530 et al. (1989) 1-(3-Cyano-2,3-deoxy-beta• 137. Parks WP, Scott GB (1985) Pediatric D-erythro-pentofuranosyl)thymine (cy• AIDS: a disease spectrum causally as• anothymidine): synthesis and antivi• sociated with HTLV-III infection. Can• ral evaluation against human immuno• cer Res 45 :4659s-4661 s deficiency virus. J Med Chern 32:618- 138. Pizzo PA, Eddy J, Falloon J, et al. (1988) 622 Effect of continuous intravenous infu• 148. August EM, Lin T-S, Marongiu ME, sion ofzidovudine (AZT) in children with Gao YS, Qian H-Y, Prusoff WH (1988) symptomatic HIV infection. N Engl J Cellular pharmacology of 2',3'-dide• Med 319:889-896 oxy-2',3'-didehydrothymidine and 2',3'• 139. Sharpe AH, Jaenisch R, Ruprecht RM dideoxy-2,3' -didehydrocytidine in H 9 (1987) Retroviruses and mouse embryos: cells. 4th International Conference on a rapid model for neurovirulence and AIDS, June 12-16, Stockholm, vol1. transplacental antiviral therapy. Science Swedish Ministry of Health and Social 236:1671-1674 Affairs, Stockholm, p 220 140. Groopman JE, Mitsuyasu RT, DeLeo 149. Balzarini J, Baba M, Pauwels R, Herde• MJ, Oette DH, Golde DW (1987) Effect wijn P, de Clercq E (1988) Anti-retrovirus of recombinant human granulocyte• activity of 3'-fluoro- and 3'-azido• macrophage colony-stimulating factor substituted pyrimidine 2',3'-dideoxynu• on myelopoiesis in the acquired immuno• cleoside analogues. Biochem Pharmacol deficiency syndrome. N Engl J Med 37:2847-2856 317:593-598 150. Kelly JA, Litterst CL, Roth JS, et al. 141. Kierdaszuk B, Bohman C, Eriksson S (1987) The disposition and metabolism (1988) Phosphorylation of the antiviral of 2',3' -dideoxycytidine, an in vitro in• nucleoside 2',3'-dideoxycytidine by pure hibitor of human T -lymphotropic virus human deoxycytidine kinase (Abstr). 4th type III infectivity, in mice and monkeys. International Conference on AIDS, June Drug Metab Dispos 15:595-601 12-16, Stockholm, vol 1. Swedish Minis• 151. Merigan TC, Skowron G, Bozzette S, et try of Health and Social Affairs, Stock• al. (1989) Circulating p 24 antigen levels holm, p 223 and responses to dideoxycytidine in 142. Balzarini J, Pauwels R, Baba M, et al. human immunodeficiency virus (HIV) (1988) The in vitro and in vivo anti• infections: a phase I and II study. Ann retrovirus activity, and intracellular Intern Med 110:189-194 metabolism of 3'-azido-2',3'-dideoxy• 152. McNeely MC, Yarchoan R, Broder S, thymidine and 2',3'-dideoxycytidine Lawley TJ (1989) Dermatologic compli• are highly dependent on the cell species. cations associated with administration of Biochem Pharmacol 37:897-903 2',3' -dideoxycytidine in patients with 143. Balzarini J, Cooney DA, Dalal M, et al. human immunodeficiency virus infec• (1987) 2',3'-Dideoxycytidine: regulation tion. J Am Acad Dermatol 21: 1213- of its metabolism and anti-retroviral po• 1217 tency by natural pyrimidine nucleosides 153. Dubinsky RM, Dalakas M, Yarchoan R, and by inhibitors of pyrimidine nucleo• Broders S (1988) Follow-up of neuro• tide synthesis. Mol Pharmacol 32:798- pathy of 2',3'-dideoxycytidine. Lancet 806 1:832 144. Ganser A, Greher J, Volkers B, Stas• 154. Chen CH, Cheng YC (1989) Delayed zewski A, Hoelzer D (1988) Azidothy• cytotoxicity and selective loss of mito• midine in the treatment of AIDS. N Engl chondrial DNA in cells treated with anti• J Med 318:250-251 human immunodeficiency virus (HIV) 145. Ahluwalia G, Johnson MA, Fridland A, compound 2',3'-dideoxycytidine (ddc) Cooney DA, Broder S, Johns DG (1988) (Abstr). FASEB J 3:A1282 Cellular pharmacology of the anti-HIV 155. Yarchoan R, Thomas RV, Pluda J, et al. agent 2',3'-dideoxyadenosine (Abstr). (1988) Long term treatment of severe Proc. Am Assoc Cancer Res 29: 349 human immunodeficiency virus (HIV) 146. Lindblad G, Jonsson G, Falk J (1973) infection with an alternating regimen of Adenine toxicity: a three week intraven• two dideoxynucleosides (Abstr). Clin Res ous study in dogs. Acta Pharmacol Tox• 36:450A icol (Copenh) 32:246-256 ell 156. Hirsch MS, Kaplan JC (1987) Treatment AIDS, June 12-16, Stockholm, vol 2. of human immunodeficiency virus infec• Swedish Ministry of Health and Social tions. Antimicrob Agents Chemother Affairs, Stockholm, p 163 31 :839-843 168. Bergdahl S, Sonnerborg A, Albert J, et al. 157. De Clercq E (1987) Perspectives for the (1988) Antiviral effect against HIV in chemotherapy of AIDS. Anticancer Res patients with AIDS-related complex 7:1023-1038 given intermittent LV. foscarnet. 4th In• 158. Dalgleish AG, Beverley PC, Clapham ternational Conference on AIDS, June PR, Crawford DH, Greaves MF, Weiss 12-16, Stockholm, vol 2. Swedish Minis• RA (1984) The CD4 (T4) antigen is an try of Health and Social Affairs, Stock• essential component of the receptor for holm, p 163 the AIDS retrovirus. Nature 312: 763- 169. Krown SE, Real FX, Cunningham• 767 Rundles S, et al. (1983) Preliminary ob• 159. Klatzmann D, Champagne E, Chamaret servations on the effect of recombinant S, et al. (1984) T-lymphocyte T4 mole• leukocyte A interferon in homosexual cule behaves as the receptor for human men with Kaposi's sarcoma. N Engl J retrovirus LAV. Nature 312:767-768 Med 308:1071-1076 160. Ashkenazi A, Presta LG, Marsters SA, 170. Groopman JE, Gottlieb MS, Goodman Camarrato TR, Rosenthal KA, Fandly J, et al. (1984) Recombinant alpha 2 BM, Capon DJ (1990) Mapping the CD4 interferon therapy for Kaposi's sarcoma binding site for human immunodefi• associated with the acquired immunode• ciency virus by alanine-scanning muta• ficiency syndrome. Ann Intern Med genesis. Proc Nat! Acad Sci USA 87: 100:671-676 7150-7154 171. Gelmann EP, Longo D, Lane HC, et al. 161. Lifson JD, Reyes GR, McGrath MS, (1987) Combination chemotherapy of Stein BS, Engleman EG (1986) AIDS disseminated Kaposi's sarcoma in pa• retrovirus induced cytopathology: giant tients with acquired immunodeficiency cell formation and involvement of CD4 syndrome. Am J Med 82:456-462 antigen. Science 232: 1123-1127 172. Lane HC, Kovacs JA, Feinberg J, et al. 162. Lifson JD, Feinberg MB, Reyes GR, et (1988) Anti-retroviral effects of inter• al. (1986) Induction of CD4 dependent feron-alpha in AIDS-associated Kaposi's cell fusion by the HTLV III/LAV enve• sarcoma. Lancet 2: 1218 -1222 lope glycoprotein. Nature 323 :725-728 173. Nabel GJ, Rice SA, Knipe DM, Balti• 163. Capon JD, Chamow SM, Mordenti J, et more D (1988) Alternative mechanisms al. (1989) Designing CD4 immuno• for activation of human immunodefi• adhesins for AIDS therapy. Nature 337: ciency virus enhancer in T cells. Science 525-531 239: 1299-1302 164. Chaudhary VK, Mizukami T, Fuerst TR, 174. Mosca JD, Bednarik DP, Raj NB, et al. et al. (1988) Selective killing of HIV• (1987) Activation of human immunode• infected cells by recombinant human ficiency virus by herpesvirus infection: CD4 Pseudomonas exotoxin hybrid pro• identification of a region within the long tein. Nature 335:369-372 terminal repeat that responds to a trans• 165. Till MA, Ghetie V, Gregory T, et al. acting factor encoded by herpes simplex (1988) HIV-infected cells are killed by virus 1. Proc Nat! Acad Sci USA 84: rCD4-ricin A chain. Science 242:1166- 7408-7412 1168 175. Salahuddin SZ, Ablashi DV, Markham 166. Abrams DI, Kuno S, Wong R, et al. PD, et al. (1986) Isolation of a new virus, (1989) Oral dextran sulfate (UA001) in HBL V, in patients with lymphoprolifera• the treatment of the acquired immunode• tive disorders. Science 234: 596-601 ficiency syndrome (AIDS) and AIDS• 176. Van der Vliet PC, Kwant MM (1981) related complex. Ann Intern Med 110: Role of DNA polymerase y in adenovirus 183-188 DNA replication: mechanism of inhi• 167. Jacobson MA, Crowe S, Levy J, et al. bition by 2',3'-dideoxynucleoside 5'• (1988) Beneficial effect of intermittent triphosphates. Biochemistry 20: 2628- intravenous (IV) foscarnet (PF A) ther• 2632 apy on HIV infection in patients with 177. Lin J-C, Zhang Z-X, Smith MC, Biron AIDS. 4th International Conference on K, Pagano JS (1988) Anti-human im-

em munodeficiency virus agent 3'-azido-3'• dementia complex after introduction of deoxythymidine inhibits replication of· zidovudine treatment. Br Med J 299: Epstein-Barr virus. Antimicrob Agents 819-821 \ Chemother 32:265-267 182. DeVita VT, Schein PS (1973) The use of 178. Birx DL, Redfield RR, Tosato G (1986) drugs in combination for the treatment of Defective regulation of Epstein-Barr cancer: rationale and results. N Engl J virus infection in patients with acquired Med 288:998-1006 immunodeficiency syndrome (AIDS) or 183. Montgomery AB, Leoung GS, Wardlaw AIDS-related disorders. N Engl J Med LA, et al. (1989) Effect of zidovudine on 314:874-879 mortality rates and Pneumocystis carinii 179. Yarchoan R, Redfield RR, Broder S (PCP) incidence in AIDS and ARC pa• (1986) Mechanisms ofB cell activation in tients on aerosol pentamidine (Abstr). patients with acquired immunodeficiency Am Rev Respir Dis [Suppl] 139:A250 syndrome and related disorders. Contri• 184. Henry K, Chinnock BJ, Quinn RP, Flet• bution of antibody-producing B cells, of cher CV, de Miranda P, Balfour HH Jr Epstein-Barr virus-infected B cells, and (1988) Concurrent zidovudine levels in of immunoglobulin production induced semen and serum determined by radioim• by human T celllymphotropic virus, type munoassay in patients with AIDS or III/lymphadenopathy-associated virus. J AIDS-related complex. JAMA 259: Clin Invest 78 :439-447 3023-3026 180. Baba M, Pauwels R, Balzarini J, Herde• 185. Puckett LD, Orf JW, Hostager BS, Gol• wijn R, de Clercq E, Desmyter J (1987) den so ph CR, Donner JE, Heinke CJ Ribavirin antagonizes inhibitory effects (1988) Characteristics of a two hour RIA of pyrimidine 2',3'-dideoxynucleosides for monitoring of the levels of AZT in but enhances inhibitory effects of purine patients' serum (Abstr). 4th Interna• 2',3'-dideoxynucleosides on replication tional Conference on AIDS, June 12- of human immunodeficiency virus in 16, Stockholm, vol 2. Swedish Ministry vitro. Antimicrob Agents Chemother of Health and Social Affairs, Stockholm, 31:1613-1617 p177 181. Portegies P, de Gans J, Lange JMA, et al. (1989) Declining incidence of AIDS

elv Frederick Stohlman Jr. Memorial Lecture

The Human fJ-Globin Locus Control Region *

F. Grosveld, M. Antoniou, M. Berry, Ernie de Boer, N. Dillon, J. Ellis, P. Fraser, D. Greaves, O. Hanscombe, J. Hurst, M. Lindenbaum, V. Mignotte, S. Philipsen, S. Pruzina, J. Strouboulis, D. Talbot, and D. Whyatt 1

Introduction thalassaemia subgroups according to the type of gene affected. In a related con• The human p-globin gene cluster spans a dition, hereditary persistence of foetal region of70 kilo bases (kb) containing five haemoglobin (HPFH), y-globin gene ex• developmentally regulated genes in the pression and hence HbF (fetal haemo• order 5'-B,yGYA,15,P-3' (Fig. 1). The hae• globin) production persist into adult life. matopoietic tissue in the early stages of These clinically important diseases pro• human development is the embryonic vide natural models for the study of the yolk sac and the a-globin gene is ex• regulation of globin gene regulation dur• pressed. This is switched to the y-globin ing development. Most interesting in genes in the foetal liver and the 15- and p• terms of transcription are the promoter globin genes in adult bone marrow mutations and deletions. The 15P• (Fig. 2; for review, see [9]). High levels of thalassaemias and a number of the these genes are expressed in circulating HPFHs are associated with an elevated red blood cells (RBC), giving rise to 90 % expression of the y-genes in adult life as a of the total soluble protein. RBC are result of deletions of varying size. Analy• derived from a pluripotent stem cell sis of these deletions has suggested that which can differentiate along alternate they act over considerable distances, to pathways to erythrocytes, platelets, influence differential gene expression granulocytes, macrophages and lympho• within the human f3-globin domain. cytes. During the transition to erythro• blasts which have lost the capacity to proliferate, the f3-globin genes become The Locus Control Region transcriptionally activated achieving messenger RNA (mRNA) levels of more The existance of a region that activates than 25000 copies per cell. the entire p-globin gene cluster first A large number of structural defects became apparent from the study of a have been documented in the f3-globin heterozygous yp-thalassaemia (Fig. 1) gene locus (for review, see [9, 44]). These [31]. This patient contained one deletion defects are responsible for a hetero• allele which lacked 100 kb, eliminating geneous group of genetic diseases collec• the entire upstream region but not the p• tively known as the p-thalassaemias, globin gene [57], which was shown to be which are classified into p, 15P, y15P, etc. completely normal [31, 64]. The other allele was expressed in the patient and it was therefore not a lack of trans-acting factors which silenced the mutant * This work was supported by the MRC (UK). chromosome but an important control 1 Laboratory of Gene Structure and Ex• region had be missing. A set of develop• pression, National Institute for Medical mentally stable, hypersensitive sites, 5' Research, The Ridgeway, Mill Hill, London HS 1, 2, 3 and 4, were shown to be present NW71AA, UK. upstream in the deleted region, and these

Haematology and Blood Transfusion Vol. 35 CV Modern Trends in Human Leukemia IX R. Neth et al. (Eds.) © Springer· Verlag Berlin Heidelberg 1992 5'HS 3'HS 432 1 2

______~ ~ ~ ~ ~~L_ __~~~~L_ ____~~L_~~L_ ____~~5~ ------Dutch 'Y\3 thaI ------Spanish 'Y\3 thaI -<------DNase sensitivity + Fig. 1. Schematic representation of the 13- The two arrows downstream of the p-gene are globin locus. Boxes indicate the different genes 3'HSl and 2. The black bars represent two in which are all transcribed from left to right. The vivo deletions which eliminate the function of vertical arrows indicate DNase hypersensitive the LCR, The horizontal arrows indicate low sites. The four arrows mark the LCR contain• ( - ) and high ( + ) sensitivity of the chromatin ing 5'HS1, 2, 3 and 4 upstream of the e-gene. to DNaseI digestion

EMBRYO FOETAL

HUMAN

L A

~maj& in MOUSE

b1rth Fig. 2. Schematic representation of the devel• in human and mouse. The curves refer to the opmental expression patterns of globin genes f3-like genes only

CVI were potential candidates for such a locus are used. This indicates that the interac• control region (LCR; Fig. 1) [17, 27, 60]. tion between the LCR and the promoter Linkage of this region to a cloned p• is dominant except when the promoter is globin gene resulted in erythroid-specific suppressed [10, 12, 18, 20, 49, 54]. In high-level expression of the gene in trans• agreement with the deletion observed in a genic mice and in tissue culture cells. This Hispanic yfj-thalassaernia (Fig. 1) [13], the expression is dependent on the copy num• main activity of the LCR is associated ber of the trans gene and independent of with HS2, 3 and 4 [10, 18, 20, 54, 61], the integration site in the host genome [3, which can each activate a linked trans• 27], a phenomenon which had not previ• gene, independent of the site of integra• ously been observed in trans gene ex• tion. A number of protein binding sites pression. This posed two questions: How have been mapped to these fragments, in is independence of the site of integration particular, sites for two erythroid-specific achieved? And how is the LCR involved factors and several ubiquitously expressed in globin gene switching? factors. A number of the binding sites are Position independence and copy num• present in all three active sites (Fig. 3) [43, ber dependence can theoretically be ex• 45, 55]. One of the shared factors is the plained by at least two independent erythroid/megakaryocyte-specific factor mechanisms; either positive activation by GATA-l [38,48], which has been shown the LCR is always achieved, overriding to be essential for erythroid development position effects that could be present, or [42]. Deletion of GATA-l binding sites (and) the region contains a locus border prevents erythroid-specific induction of element(s) (LBE) that insulate it from the p-glo bin gene [11], and the protein has neighbouring regions. Matrix attachment been shown to have transcriptional acti• sites (MAR) [22, 30] or "A" elements [4, vation properties [38]. However, the pre• 53] could be LBEs, and we initially specu• sence of GATA-l binding sites per se is lated these were part of the LCR in insufficient to give position-independent addition to activating sequences [27]. expression, e.g. the flanking regions of However, preliminary experiments indi• the human p-globin gene contain at least cate that this is not the case and that such six GATA-l binding sites [11, 63], but do a border may be located further up• not confer integration site-independent stream. The latter is based on the fact that expression [32, 34, 58]. However, all three the DNaseI sensitivity of chromatin is active 5' HS contain two closely spaced strongly decreased in the sequences 25- GAT A-l sites in opposite orientations. 30 kb 5' to the LCR (Fig. 1) [19, 31, 57]. This arrangement is also observed in the At least 150 kb of chromatin in the 3' chicken p-globin enhancer, which ap• direction is sensitive under the control of pears to provide position-independent the LCR [19], suggesting that such se• expression [47]. Possibly an inverted dou• quences are not present for a considerable ble GATA-l site is a key component in distance 3' (Fig. 1). The position indepen• erythroid-specific, position-independent dence we observe is therefore due to a activation and GAT A-l can interact with dominant activation of transcription by itself or one of the other GAT A proteins the LCR, perhaps by creating very stable to achieve this [66]. Classical enhancer interactions between the LCR and the activity is only associated with 5'HS2 [41, genes. Consequently, positive position 61], and not with the others. Dissection of effects would only be present in the the HS2 showed that a number of pro• background and only become apparent in teins are bound to the core fragment situations where the linked gene is (Fig. 3) [55]. Attention has been focused suppressed [12] (see below for dis• on a double consensus sequence for the cussion). Position effects are not ob• jun/fos family of DNA binding proteins served at low levels of expression when which appeared crucial for HS2 activity part of the LCR or mutations in the LCR [41, 52, 55]. Several pieces of evidence

eVIl Hae Xba ubiquitous proteins [56]. This suggests that a combination of erythroid-specific SITE 2 ~ and ubiquitous factors may be required NFE2 G~~ J I J-BP jun H-BP to render the fJ-globin gene independent of its site of integration. The (abundant) ubiquitous factors shared by the three HS Hph Fnu of the LCR which have been studied to date are Spl and TEF-2 [23, 65], but a SITE 3 ~ simple multimerized combination of a ~~ ~~II GATAJ GATAJ GATAJI I GATA-l and a SplJTEF-2 binding site is J-BP J-BP not functional (S. Philipsen, unpublished results). We therefore think that other, as yet less well characterised factors may be Sac Ava involved in LCR function. SITE 4 ~ I I~~ NFE2 GATAJ J-BP The LCR and Disease Fig. 3. Summary of factor binding sites to the minimal fragment of 5'HS2, 3 and 4, which The discovery, characterization and provide position-independent expression in mapping of the LCR has enabled the transgenic mice. Individual factors are de• pursuit of two novel approaches to the scribed in the text. Black boxes indicate erythroid-specific factors; open boxes indicate study of globin-related diseases. Firstly it ubiquitous factors. GT indicates a GT-rich allows high-level expression of disease motif [43] genes such as the fJs gene which is respon• sible for sickle cell disease. By linking this gene in combination with human (J.• globin genes several laboratories have show that the functional activator which succeeded in producing transgenic mice iriteracts with the junJfos binding site is which show sickle cell disease [26, 50, 59]. NF-E2, originally described as interact• High levels of human haemoglobin Scan ing with another erythroid-specific gene, be obtained in mice and the RBC of these that for porphobilinogen deaminase [39, mice show a pronounced change in shape 40]. Two NF-E2 molecules and at least when deoxygenated (Fig. 5). We are pre• one other protein binding at two non• sently improving this model for two rea• equivalent sites are involved [56]. How• sons, firstly, to study the effects of sickle ever, the presence of this double NF-E2 cell disease on the progression of infec• sequence alone is insufficient to provide tion by different malaria strains, and high levels of expression [55] and when secondly, to be able to study the pro• the junJfos binding site is removed from gression of sickle cell disease and the the 300 bp core fragment, HS2 retains the treatment thereof by new protocols, in ability to activate a linked fJ-globin gene particular the development of gene in a copy number dependent fashion, therapy. The latter has been given new albeit at low levels (Fig. 4) [56]. We there• hope by the mapping of the minimal fore conclude that the 5'HS2 NF-E2 elements that give the full activity of the region has strong enhancer activity but LCR. The LCR can now be incorporated that it is not necessarily required to into retroviral vectors to develop therapy obtain position-independent globin gene protocols and preliminary experiments activation. (F. Meyer, personal communication) in• All the other factors which have been dicate that the LCR will provide high shown to interact with LCR sequences, levels of expression in this context in including the factors H-BP and J-BP, are mice.

CVIII construct 613 13

fetus (coPY noJ

M~mJ

10 613 100

c 2 ~ 6 ..:: . .. 20

10 20 )0 40 50 60 10 20 30 copy number copy number

y - - 0,2456 • 0.2003. R · 1.00 y - 3 , 3eo~ • 4,0755. R - 0 .98

Fig. 4. S 1 nuclease analysis ofHS2 constructs probe and a mouse a-globin probe as an containing the NF-E2 sites (13) ornot (/),13) in internal control. The % expression is given as transgenic mice. Foetal liver RNA (day 13.5) the total H up-globin signal divided by the total was assayed using a mixed probe S 1 nuclease mouse a-globin signal (adjusted for specific experiment using the 5' human fJ-globin probe activities). This was plotted against the copy and the mouse pmaj probe [56]. Specific activ• number. The line represents the result of a ities were 10: 1 for Hu p : M p. Protected prod• linear regression analysis on the data points. ucts are indicated on the left. The 200 series of The R value, the correlation coefficient, indi• transgenic mice contains the /),13 construct cates very high correlation with a straight line and the 300 series contains the 13 construct. (R = 1). The dashed line in the /),13 graph Copy numbers are shown in parentheses. represents the minimal level that can be measu• Lower pnael depicts a quantitation experiment red accurately of the Sl protection analysis using the HuP5'

Developmental Regulation Those clustered around the distal' of the fJ-Globin Locus CAA T box appear to result in the loss of factor binding sites [21, 36], suggesting Genetics. The study of globin gene that this region may contain a binding site switching has been assisted by the charac• for a negative regulator. For example, a terization of deletions and point muta• 13-bp deletion which removes the distal tions which affect expression of the y CAAT box results in a very strong HPFH and fJ-genes. Point mutations in the y• (60%) [25]. Interestingly, a recently de• promoters have been linked to HPFH scribed Japanese HPFH (20%) is as• phenotypes and these can be divided into sociated with a point mutation in the two groups (Fig. 6) . CAA T sequence of the distal CAA T box

CIX ex ex LCR o ell

Fig. 5. Sickle cell disease in transgenic mice. panel shows sickled cells from one of the The top line shows the arrangement of genes transgenic mice [26]; the bottom panel shows and the LCR that was injected into fertilized control non transgenic red blood cells mouse eggs to obtain transgenic mice. The top

GIT C TAT A T

clCCTUc - AAAITATC - c!GTLc~- AGCCTTGCCT~ AGCCTT&---iG:~mM.'i--{~MW-iP~Jl-..lr-- ; :: ; ; : : : -202 I I -175 -161 I -117 I -198 i -Ide -11!S - J 96

Fig. 6. Summary of mutations occurring in the y-globin promoter resulting in HPFH pheno• types (see [44]) ex [21] which reduces affinity for the tran· (>10kb). Of these, the AyJP• scription factor CP 1. The -117 muta• thalassaemias all have deletions which tion associated with Greek HPFH extend into the region of y transcription. (40%) has been reported to cause re· They are uninformative for competition duced binding of the erythroid-specific models because enhancers found near the factor NF-E3 [36]. These findings suggest deletion breakpoints may be responsible a model for y silencing in which factors for the high level, pancellular y expression binding to the distal CAAT box (at observed in the deletion HPFHs [1, 15]. -115) compete for interaction with fac• Some increased y expression is also ob• tors bound to upstream promoter se• served in the J p- and Dutch p• quences preventing the proximal CAAT thalassaemias, but the broad range of box (at - 87) from forming such interac• values between patients with the same tions. The distal CAAT box is located deletion and the heterocellular distri• outside the normal optimal position for bution of y-protein among the red cells CAAT elements, and this is likely to suggest that the increase in y expression is prevent it from functioning as an effective not solely at the transcriptional level. positive promoter element. One would This is supported by the observation expect this type of silencing mechanism to that non transcriptional defects (e. g. depend on the topology of the promoter RNA processing) in heterozygous f3-tha• region and it is also likely to be affected lassaemias cause elevated levels of )1- by the creation of extra factor binding chains (up to 5 %). Selection of a small sites in the upstream sequences. Such sites proportion of cells expressing y is a likely may partially bypass the competition mechanism for this increase. Deletion of between the proximal and distal CAAT the J-gene (which is normally expressed boxes, resulting in suboptimal transcrip• at only 2 % - 3 % of the level of p) also tion. Indeed, a second group of muta• does not seem to be required for the tions, upstream of -150, result in new )I-expression observed in the 1J f3-tha• or improved binding sites for transcrip• lassaemias, since it is intact in Dutch tion factors, e.g. Spl [16,28] and GATA- p-thalassaemia, which has a very similar 1 [35, 37]. phenotype. Instead, the requirement ap• Activation of y transcription in the pears to be a minimum size of deletion nondeletion HPFHs is associated with (> 10 kb). Probably these large deletions down regulation of the p-gene. The reduc• perturb the chromatin structure of the tion in p expression (to around 60% in locus, resulting in a small increase in y Southern Italian HPFH) is approximate• transcription which is further amplified ly equivalent to the rise in expression of by the chain imbalance. In conclusion, the cis-linked y-globin gene, with only a the genetic data show that strong down• slight reduction in overall transcriptional regulation of the p-gene can result from output from the locus [24, 62]. This an increase of y-gene transcription, while suggests that competition is taking place there does not seem to be any significant between the genes and that this is tightly link between transcription of the p-gene linked to the process of transcription. and silencing of the )I-genes in adult life. However, a drastic reduction or loss of p transcription due to point mutations and deletions in the p-promoter does Transgenic Mice. Attempts to study not significantly increase y expression switching of globin genes have also made (less than 5 %; Fig. 7) [44]. Clearly, a y• use of transgenic mice as a model system. gene exerts a negative effect on the p-gene Mice do not possess separate foetal (coupled to transcription) but this effect globin genes but instead switch directly is not reciprocal. Some p-gene deletions from embryonic to adult p-globin ex• show higher levels of y expression (Fig. 7) pression at 11-13 days of gestation but these deletions are always large (Fig. 2).

CXI I I I I I E yGyA1jI13 Ii 13 f50 KB f f f f ------~~~' ____ _J.L_~.~~DL_~~~~·~ ______iI~1------13 THALASSAEMIA _ TURKISH (2.7-3.3) IND I AN (NA) - _ BLACK (7) CZECH/CANADI AN (3.3-5.7) _ ____ DUTCH(4-1 0)

____ SICILI AN (5- 13) lil3 THALASSAEMI A BLACK (NA) _ HbLEPORE MACEDONIAN (7-14) GREEK ( I. 1-1.6) JAPANESE (7-6) SPANISH (5- 13)

HPFH HPFH 2 (20-30) HPFH I (20 - 30)

SOU THERN ITALIAN (29-32) I ND I AN (17-25)

Ayol3 THAL ASSAEMIA ______------BLACK 7 MALAYS (10-13)I AN 2 (NA)

CHINESE (9-16) MALAYSI AN 1 (NA) GERMAN (9.8-12.5) INDIAN(lO-I7) _ ( - - ___

______------TURKISH (10 7 - 13.5) CANTONESE (19. 1- 19.6)

Fig. 7. Schematic representation of the differ• the size of the deletion; numbers in parentheses ent deletions occurring in the p-globin locus in indicate the levels of y-globin expression in thalassaemias and HPFHs. Black bars indicate heterozygotes

The developmental regulation of the The human y-transgene without the human e-gene has been analysed in both LCR is expressed like the mouse em• embryonic stem cells and transgenic bryonic genes [7, 32]. It was initially mice. In mice the e-gene is expressed at reported that linkage to the LCR resulted high levels during the embryonic stage in y expression at all developmental only when linked to the LCR and is com• stages and that the y-gene was silenced in pletely silenced thereafter [33, 46, 51]. adult mice only when the fi-gene was also Based on the studies by Cao et al. dele• present. This appeared to support a com• tion mutants lacking the - 200 to - 300 petition model where the fi-gene is re• promoter region show a small increase in quired for silencing of the y-gene [2, 14]. e expression in adult transgenic mice but However, a different result was obtained the low level indicates that other se• when the single y-gene experiments were quences may also be involved in silencing carried out on animals carrying only one e (P. Fraser, unpublished observations). or two copies of the LCR-y-gene con-

CXII EMBRYO FOETUS y

IblLCR I ~ I Ii I bl_~(:R __ _ ++ + ++ ~5KD -- -- IO.5KO ------• o.SKD ------I 2.SKO

Ibl LCR I Ii I ~ 1u..~(:R __ _ ++ + ++ J.SKb -- -- B.SKb ------.7.SKO ------12.5KO Fig. 8. Microlocus (IlLCR) yp and py con• the distance from a promoter to a 5' and 3' structs [29]. Genes are represented as shaded LCR. These distances are indicated by dotted boxes. All genes are in the same trancriptional lines below the constructs. Plus and minus orientation, 5' to 3', with respect to each other symbols indicate high, medium, and very low and the LCR. The dotted LCR lines indicate levels of expression the situation in multicopy animals to illustrate struct. y expression persisted in the early LCR blocks this expression [2, 14, 29], foetal liver, but was silenced at adult supporting the idea that competition stages, independent of the presence of the plays a role in preventing premature 13 f3-gene [12]. Transcription of the LCR• expression. However, when the order is linked y-gene can therefore also be reversed and the f3-gene is placed in the blocked completely by stage-specific first position, it is expressed at a level negative regulators acting on the se• similar to that observed for the f3-gene in quences immediately flanking the gene, the absence of the 1'- or rJ.-gene (Fig. 8) and this removes the basis of the argu• [29]. Silencing of the f3-gene at the em• ment that the f3-gene would be needed for bryonic stage is therefore not caused by y silencing. The elements responsible for y reciprocal competition only, but relative silencing have not yet been identified but distance between the LCR and the genes the mutations associated with the nonde• (i.e. position and polarity) is also letion HPFHs suggest that at least the important. sequences around the distal CAAT box Polarity in the locus has long been are likely to be involved (see above). The suggested by the fact that the genes are availability of a transgenic mouse model arranged in the order of their expression for (-gene silencing should allow this to during development. The order of the be tested and possibly lead to novel ap• genes is conserved among mammals but proaches for treating thalassaemia and there is some divergence in the other sickle cell anaemia. If )I-gene expression vertebrate loci. In chicken, the embryonic were understood at the level of the tran• e- and Q-genes are located at opposite scription factors, it might be possible to ends of the locus, with the adult f3-genes develop novel therapies that could speci• between them. However, it is important fically interfere with the adult suppres• to note that the chicken f3-globin LCR sion of the )I-gene and alleviate the clini• may have been split as part of an e cal problems associated with severe chain translocation such that part of it is imbalance or sickling. located between the 13- and e-genes [8, 47] Linkage of the adult f3-gene to the LCR and that the e-gene contributes only 20 % results in inappropriate expression at the of the total embryonic haemoglobin com• embryonic stage [2, 3, 14,29,33], albeit at pared to 80 % for Q [5]. a low level. Placing a y-gene or a human rJ.• The data reviewed above indicate that globin gene between the f3-gene and the developmental regulation of the human

CXIlI l e are negat ive regula tors LCR £ YG yA I) ~ Embryo 0 ~ f§] !'Sl ~ ••• • Embryo • rl LCR Ie' l l YG yA I) ~ Foetus I~ E§l ~ E§l ~ ••• • Foe tus • rli rl rl i LCR Ie' ~'@' l l I) ~ Adult I ~ I~ I ~ ~ !S3 ••• • Adult • Fig. 9. Model for stage-specific regulation of tors silencing the gene. The location of these is the genes of the p-globin locus. Solid lines not accurate and there may be more than one indicate activation of genes by the LCR. The factor for each gene symbol e indicates stage-specific negative fac-

,8-globin locus is a complex process which on the effective volume in which these centres around developmentally specific elements operate. This effect would be suppressors and the polarity of the locus most pronounced if the LCR and the (Figs. 9, 10). The earliest gene to be genes were all present on one structural activated, the a-gene, is also the one chromatin loop several times the distance closest to the LCR. The y- and ,8-genes between the LCR and the genes. The fact may be suppressed by competition with a; that the LCR controls DNase hypersen• alternatively, or in addition, the y- and,8- sitivity of the ,8-globin locus over at least genes may bind embryonic stage-specific 150-kb [19] suggests that the entire ,8- factors which keep their promoters locus may be present on one very large suppressed. The a-promoter is silenced in chromatin loop. If we assume that to be the foetal liver by one or more suppressor the case, the frequency of interactions factors, negating its competitive ability between any of the promoters with the (Fig. 9). As a result the y-genes are ex• LCR would be proportional to their pressed, and they in turn keep expression effective concentration relative to the of the ,8-gene suppressed by competition. LCR (Fig. 10). On basis of ring closure The y-genes are switched off during the probabilities with naked DNA, the effec• period around birth, again by stage• tive concentration of two points on the specific negative regulators, and as a DNA will be related to the volume of a consequence the ,8-gene is activated and sphere and will be proportional to the expressed in the bone marrow. We pro• power of3/2 of the distance. Applying the pose that loop formation between regu• rule to the ,8-locus, the ,8-gene is twice as latory elements is the crucial parameter to far as the Gy-gene from the HS 2 enhancer explain the suppression of the late genes of the LCR. Therefore, the ,8-gene by the early genes at early stages but not occupies an approximately eight-fold lar• vice versa. ger volume relative to the HS-2 enhancer The frequency of interaction between of the LCR than the Gy-gene, which the promoters and the LCR will depend should give it a three-fold lower fre-

CXIV " "

------....-:

Fig.tO. Schematic representation of the rela• tation the LCR is shown as a fixed point in tive volumes occupied by the Gy- and fJ-genes the centre of the sphere. Only half the fJ• relative to the LCR. For simplicity of presen- globin gene outer sphere is shown quency of interaction with the LCR References (Fig. 10). This effect will work in favour of the proximal gene, decreasing the 1. Anagnou NP, Perez-Stable C, Gelinas R, affinity differences required for compe• Costantini F, Liapaki K, Constan• tition, but it will work against the distal topoulou M, Costeas T, Moschonas N, gene. Distal genes would be incapable of Stamatoyannopoulos G (1990) DNA se• quences residing 3' of the breakpoint of the suppressing upstream genes under similar HPFH-3 deletion can modify the develop• circumstances unless the downstream mental regulation of the fetal Ay globin gene promoter increased its affinity by gene. Clin Res 38: 301 A several orders of magnitude relative to 2. Behringer RR, Ryan TM, Palmiter RD, the upstream gene. The transgenic mouse Brinster RL, Townes TM (1990) Human data on the expression of the j3-globin y- to fJ-globin gene switching in transgenic gene at the embryonic and foetal/adult mice. Genes Dev 4:380-389 stages argue strongly against this possi• 3. Blom van Assendelft G, Hanscombe 0, bility. Instead, the problem is solved by Grosveld F, Greaves DR (1989) The fJ• globin domain control region activates local suppression of the upstream pro• homologous and heterologous promoters moters to allow expression from the in a tissue-specific manner. Cell 56:969- downstream gene (Fig. 9). Experiments 977 to substantiate or disprove this prediction 4. Bonifer C, Vidal M, Grosveld F, Sippel are presently in progress. AE (1990) Tissue-specific and position-

CXV independent expression of the complete two DNA binding proteins. Nucleic Acids gene domain for chicken lysozyme in Res 18:5685 transgenic mice. EMBO J 9:2843-2848 17. Forrester W, TaRegawa S, Papayan• 5. Brown J, Ingram V (1974) Structural nopoulou T, Stamatoyannopoulos G, Studies on Chick Embryonic Hemo• Groudine M (1987) Evidence for a locus globins. J BioI Chern 249:3960-3972 activation region: the formation of 6. Cao S, Gutman PD, Dave HPG, Schech• developmentally stable hypersensitive ter AJ (1989) Identification of a transcrip• sites in globin-expressing hybrids. Nu• tional silencer in the 5' -flanking region of cleic Acids Res 15:10159-10177 the human e-globin gene. Proc Nat! Acad 18. Forrester W, Novak U, Gelinas R, Sci USA 86:5306-5309 Groudine M (1989) Molecular analysis of 7. Chada K, Magram J, Costantini F (1986) the human p-globin locus activation An embryonic pattern of expression of a region. Proc Nat! Acad Sci USA 86: 5439- human fetal globin gene in transgenic 5443 mice. Nature 319:685-689 19. Forrester W, Epner E, Driscoll C, Enver T, 8. Choi O-R, Engel JD (1988) Develop• Brice M, Papayannopoulou T, Groudine mental regulations of p-globin gene M (1990) A deletion of the human pglobin switching. Cell 55: 17 - 26 locus activation region causes a major 9. Collins FS, Weissman SM (1984) The alteration in chromatin structure and rep• molecular genetics of human hemoglobin. lication across the entire p globin locus. Prog Acid Res Mol Bioi 31 :315-462 Genes Dev 4:1637-1649 10. Collis P, Antoniou M, Grosveld F (1990) 20. Fraser P, Hurst J, Collis P, Grosveld F Definition of the minimal requirements (1990) DNaseI hypersensitive sites 1, 2 and within the human p-globin gene and the 3 of the human p-globin dominant control dominant control region for high level region directs position-independent ex• expression. EMBO J 9:233-240 pression. Nucleic Acids Res 18: 3503- 11. de Boer E, Antoniou M, Mignotte V, Wall 3508 L, Grosveld F (1988) The human p-globin 21. Fucharoen S, Shimiza K, Fukumaki M promoter; nuclear protein factors and (1990) A novel C-T transition within the erythroid specific induction of transcrip• distal CCAAT motif of the Gy globin gene tion. EMBO J 7:4203-4212 in the Japanese HPFH: Implication of 12. Dillon N, Grosveld F (1991) Human y• factor binding in elevated fetal globin globin genes silenced independently of expression. Nucleic Acids Res 18: 5245 other genes in the p-globin locus. Nature 22. Gasser S, Laemmli U (1986) Cohabitation 350:252-254 of scaffold binding regions with upstream/ 13. Driscoll C, Dobkin C, Alter B (1989) enhancer elements of three developmen• Gamma/delta/beta thalassemia due to a de tally regulated genes of D. melanogaster. novo mutation deleting the 5' p-globin Cell 46: 521-530 gene locus activating region hypersensitive 23. Gidoni D, Kadonaga JT, Barrera-Saldana sites. Proc Nat! Acad Sci USA 86:7470- H, Takahashi K, Chambon P, Tjian R 7474 (1985) Bidirectional SV 40 transcription 14. Enver T, Raich N, Ebens AJ, Papayan• mediated by tandem Sp1 binding interac• nopoulou T, Costantini F, Stamatoyan• tions. Science 230:511-514 nopoulos G (1990) Developmental regu• 24. Giglioni B, Casini C, Mantovani R, Merli lation of human fetal-to-adult globin S, Comp P, Ottolenghi S, Saglio G, gene switching in transgenic mice. Nature Camaschella C, Mazza U (1984) A mole• 344:309-313 cular study of a family with Greek heredi• 15. Feingold E, Forget B (1989) The break• tary persistence of fetal hemoglobin and p• point of a large deletion causing hereditary thalassaemia. EMBO J 11: 2641-2645 persistence of fetal hemoglobin occurs 25. Gilman J, Mishima N, Wen X, Stoming T, within an erythroid DNA domain remote Lobel J, Huisman T (1988) Distal CCAAT from the p globin gene cluster. Blood box deletion in the Ay globin gene of two 74:2178-2186 black adolescents with elevated fetal Ay 16. Fischer K, Nowock J (1990) The T to C globin. Nucleic Acids Res 18:10635- substitution at - 198 ofthe Ay globin gene 10642 associated with the British form of HPFH 26. Greaves DR, Fraser P, Vidal MA, Hedges generates overlapping recognition sites for MJ, Ropers D, Luzzatto L, Grosveld F

CXVI (1990) A transgenic mouse model of sickle DNA-binding factor. Nature 338:435- cell disorder. Nature 343:183-185 438 27. Grosveld F, Blom van Assendelft G, 38. Martin D, Orkin S (1990) Transcriptional Greaves D, Kollias G (1987) Position• activation and DNA binding by the independent high level expression of the erythroid factor GF-1/NF-E1/Eryf 1. human p-globin gene in transgenic mice. Genes Dev 4:1886-1898 Cell 51 :975-985 39. Mignotte V, Eleouet EF, Raich N, Romeo 28. Gumicio D, Rood K, Gray T, Riordan M, PH (1989) Cis- and transacting elements Sartor C, Collins F (1988) Nuclear pro• involved in the regulation of the erythroid teins that bind the human I' globin gene promoter of the human porphobilinogen promoter: Alterations in binding pro• deaminase gene. Proc Nat! Acad Sci USA duced by point mutations associated with 86:6548-6552 hereditary persistence of fetal hemoglobin. 40. Mignotte V, Wall L, deBoer E, Grosveld Mol Cell Bioi 8:5310-5322 F, Romeo P-H (1989) Two tissue-specific 29. Hanscombe 0, Whyatt D, Fraser P, Yan• factors bind the erythroid promoter of the noutsos N, Greaves D, Grosveld F (1991) human porphobilinogen deaminase gene. Importance of globin gene order for cor• Nucleic Acids Res 17:37-54 rect developmental expression. Genes Dev 41. Ney PA, Sorrentino BP, Lowrey CH, 5: 1387 -1394 Nienhuis AW (1990) Inducibility of the 30. Jarman A, Higgs D (1988) Nuclear scaf• HS II enhancer depends on binding of an fold attachment sites in the human globin erythroid specific nuclear protein. Nucleic gene complexes. EMBO J 7:3337-3344 Acids Res 18:6011-6017 31. Kioussis D, Vanin E, deLange T, Flavell 42. Pevny L, Simon MC, Robertson E, Klein RA, Grosveld F (1983) p-globin gene WH, Tsai S, D'Agati V, Orkin SH, Cos• inactivation by DNA translocation in 1'• tantini F (1991) Erythroid differentiation thalassaemia. Nature 306:662-666 in chimaeric mice blocked by a targeted 32. Kollias G, W righton N, Hurst J, Grosveld mutation in the gene for transcription F (1986) Regulated expression of human factor GATA-1. Nature 349:257-260 Ay_, p_, and hybrid yp-globin genes in 43. Philipsen S, Talbot D, Fraser P, Grosveld transgenic mice: manipulation of the de• F (1990) The p-globin dominant control velopmental expression patterns. Cell region: hypersensitive site 2. EMBO J 46:89-94 9:2159-2167 33. tindenbaum M, Grosveld F (1990) An in 44. Poncz M, Henthorn P, Stoeckert C, Surrey vitro globin gene switching model based S (1989) Globin Gene Expression in Her• on differentiated embryonic stem cells. editary Persistence of Fetal Hemoglobin Genes Dev 4:2075-2085 and t5p Thalassaemia. Oxford University 34. Magram J, Chada K, Costantini F (1985) Press Developmental regulation of a cloned 45. Pruzina S, Hanscombe 0, Whyatt D, adult p-globin gene in transgenic mice. Grosveld F, Philipsen S (1991) Hypersen• Nature 315:338-340 sitive site 4 of the human p-globin locus 35. Mantovani R, Malgaretti N, Nicolls N, control region. Nucleic Acids Res Ronchi A, Giglioni B, Ottolenghi S (1988) 19:1413-1419 The effects of HPFH mutations in the 46. Raich N, Enver T, Nakamoto B, Joseph• human y globin promoter on binding of son B, Papayannopoulou T, Stamatoyan• ubiquitous and erythroid specific nuclear nopoulos G (1990) Autonomous develop• factors. Nucleic Acids Res 16:7783-7797 mental control of human embryonic 36. Mantovani R, Superti-Fuga G, Gilman J, globin switching in transgenic mice. Ottolenghi S (1989) The deletion of the Science 250:1147-1149 distal CCAAT box region of the AI' globin 47. Reitman M, Lee E, Westphal H, Felsen• gene in black HPFH abolishes the binding feld G (1990) Site independent expression of the erythroid specific protein NFE 3 and of the chicken pA globin gene in transgenic of the CCAAT displacement protein. Nu• mice. Nature 348:749-752 cleic Acids Res 17:6681-6691 48. Romeo PH, Prandini MH, Joulin V, Vig• 37. MartinD, Tsai S, Orkin S (1989) Increased notte V, Prenant W, Valnchenker W, Mar• I' globin expression in a non deletion guerie G, Uzan G (1990) Megakaryocytic HPFH mediated by an erythroid-specific and erythrocytic lineages share specific

CXVII transcription factors. Nature 334:447- result of a 100 kpb deletion in the human 449 fJ-globin cluster. Nucleic Acids Res 49. Ryan TM, Behringer RR, Martin NC, 14:7017 -7029 Townes TM, Palmiter RD, Brinster RL 58. Townes T, Lingrel J, Chen H, Brinster R, (1989) A single erythroid specific DNaseI Palmiter R (1985) Erythroid-specific ex• super-hypersensitive site activates high pression of human fJ-globin genes in trans• levels of human fJ-globin gene expression genic mice. EMBO J 4:1715-1723 in transgenic mice. Genes Dev 3: 314-323 59. Trudel M, Saadaen N, Gare M-C, 50. Ryan T, Townes T, Reilly M, Asahura T, Bardakdjian-Michau J, Blouquit Y, Palmiter R, Brinster R, Behringer R (1990) Guerquin-Kern J-L, Rouyer-Fessard P, A single erythroid specific DNaseI super• Vidaud D, Pachnis A, Romeo P-H, hypersensitive site activates high levels of Beuzard Y, Costantini F (1991) Towards a human fJ-globin gene expression in trans• transgenic mouse model of sickle cell dis• genic mice. ease: Hemoglobin SAD. EMBO J 51. Shih D, Wall R, Shapiro (1990) Develop• 10:3157-3165 mentally regulated and erythroid-specific 60. Tuan D, Solomon W, Li Q, London I expression of the human embryonic fJ• (1985) The "fJ-like-globin" gene domain in globin gene in transgenic mice. Nucleic human erythroid cells. Proc Natl Acad Sci Acids Res 18: 5465-5472 USA 82:6384-6388 52. Sorrentino BP, Ney PA, Bodine DM, 61. Tuan D, Soloman W, London I, Lee DP Nienhaus AW (1990) A 46base pair en• (1989) An erythroid-specific develop• hancer sequence within the locus activat• mental-stage-independent enhancer far ing regionis required to induced ex• upstream of the human "beta-like globin" pression of the y-globin gene during genes. Proc Natl Acad Sci 86:2554- erythroid differentiation. Nucleic Acids 2558 Res 18:2721-2731 62. Weatherall DJ, Clegg JB (1981) The thal• 53. Stief A, Winter DM, Stratling WH, Sippel assaemia syndromes. Blackwell, Oxford AE (1989) A nuclear DNA attachment 63. Wall L, deBoer E, Grosveld F (1988) The element mediates elevated and position• human fJ-globin gene 3' enhancer contains independent gene activity. Nature multiple binding sites for an erythroid 341 :343-345 specific induction of transcription. Genes 54. Talbot D, Collis P, Antoniou M, Vidal M, Dev 2:1089-1100 Grosveld F, Greaves DR (1989) A domi• 64. Wright S, deBoer E, Rosenthal A, Flavell nant control region from the human fJ• RA, Grosveld FG (1984) DNA sequences globin locus conferring integration site required for regulated expression of the fJ• independent gene expression. Nature globin genes in murine erythroleukaemia 338:352-355 cells. Phil Trans R Soc Lond B307:271- 55. Talbot D, Philipsen S, Fraser P, Grosveld 282 F (1990) Detailed analysis of the site 3 65. Xiao J, Davidson I, Macchi M, Rosales R, region of the human fJ-globin dominant Vigneron M, Staub A, Chambon P (1987) control region. EMBO J 9:2169-2177 In vitro binding of several cell-specific and 56. Talbot D, Grosveld F (1991) The 5' HS2 ubiquitous nuclear proteins to the GT-I of the globin locus control region func• motif of the SV -40 enhancer. Genes Dev tions through the interaction of a multi• 1:794-807 meric complex binding at two functionally 66. Yamamoto M, Ko L, Leonard M, Beug H, distinct NF-E2 binding sites. EMBO J Orkin S, Engel J (1990) Acitivity and 10:1391-1398 tissue-specific expression of the transcrip• 57. Taramelli R, Kioussis D, Vanin E, Bar• tion factor NF-El multigene family. tram K, Groffen J, Hurst J, Grosveld FG Genes Dev 4:1650-1662 (1986) yi5fJ-thalassaemias 1 and 2 are the

CXVIIl List of Participants

Australia Dr. Fritz Anders Genetisches Institut Dr. Nicholas Gough der Justus-Liebig-Universitiit, Walter and Eliza Hall Institute of Medical Heinrich-Buff-Ring 58-62, 6300 Giessen Research, Post Office Royal Melbourne Hospital, Parkville Victoria 3050 Dr. Jens Atzpodien Dr. Paula Stapleton Medizinische Hochschule Hannover, CMRF, P.O. Box 61, Camperdown, Abt. Hiimatologie u. Onkologie, NSW 2050 Zentrum Innere Medizin und Dermatologie, Postfach 61 01 80, 3000 Hannover

Belgium Dr. C. R. Bartram Universitiitskinderklinik Ulm, Postfach 3880, Dr. Thierry Boon 7900 Ulm Ludwig Institute for Cancer Research, Brussels Branch, 74 Avenue Hippocrate, Dr. Stefan Burdach UCL 74.59, 1200 Brussels Bone Marrow Transplant Program, Dept. of Pediatric Hematology and Oncology, Zentrum fUr Kinderheilkunde der Universitiit, Canada Moorenstr. 5, 4000 Diisseldorf 1 Dr. Sam Benchimol Dr. Hans G. Drexler The Ontario Cancer Institute, DSM - Deutsche Sammlung 500 Sherbourne Street, Toronto M4X 1K9 von Mikroorganismen u. Zellkulturen, Dr. Allan Bernstein Mascheroder Weg 1 b, 3300 Braunschweig Mount Sinai Hospital Research Institute, 600 University Avenue, Dr. Volker Erfle Toronto/Ontario M5G 1X5 Gesellschaft fUr Strahlen- u. Umweltforschung mbH Miinchen, Dr. John Dick Abt. f. Molekulare Zellpathologie, Department of Genetics Research Institute, Ingolstiidter Landstr. 1, 8042 Neuherberg Hospital for Sick Children, 555 University Avenue, Toronto/Ontario Dr. Sven Gohla R&D Department, Lab. Face Care, Czechoslovakia Unnastr. 48, 2000 Hamburg 20 Dr. Alena Petrakova Dr. Thomas Graf Faculty of Pediatrics, Charles University, Europiiisches Laboratorium V. uvalu 84, 150 18 Prag 5 - Motol fUr Mo1ekularbiologie, EMBL, Postfach 102209, Meyerhofstr. 1, 6900 Heidelberg Federal Republic of Germany Dr. Klaus Harbers Dr. Annerose Anders Heinrich-Pette-Institut fUr Experimentelle Genetisches Institut Virologie und Immunologie der Justus-Liebig-U niversitiit, an der Universitiit Hamburg, Martinistr. 52, Heinrich-Buff-Ring 58-62, 6300 Giessen 2000 Hamburg 20

exIX Dr. R. Hehlmannn Dr. Peter Jorg Plath III. Med. Klinik Mannheim der Universitat Universitat Bremen, \ Heidelberg, Wiesbadener Str. 7-11, Angewandte und Physikalische Chernie, 6800 Mannheim Bibliothekstr. NW2, Postfach 330440, 2800 Bremen 33 Dr. F. Herrmann Dept. of Hematology/Oncology, Dr. Helga Rilbsamen Waigmann University of Freiburg, Hugstetter Str. 55, Chemotherapeutisches F orschungsinstitut, 7800 Freiburg Georg Speyer Haus, Paul-Ehrlich-Str. 42-44, Dr. Dieter Hoelzer 6000 Frankfurt am Main 70 Zentrum der Inneren Medizin, Abt. Hamatologie, Theodor Stem Kai 7, Dr. Johannes Schubert 6000 Frankfurt 70 St. Joseph Hospital, Wiener Str. 1, 2850 Bremerhaven Dr. Hans-Jochen Illiger Stadt. Kliniken Oldenburg, Klinik fUr Innere Dr. Manfred Schwab Medizin II, Onkologie/Hamatologie, Deutsches Krebsforschungszentrum, Dr. Eden Str. 10,2900 Oldenburg Institut fUr experimentelle Pathologie, 1m Neuenheimer Feld 280, 6900 Heidelberg Dr. Ursula Just Heinrich-Pette-lnstitut fUr Experimentelle Dr. Carol Stocking Virologie und Immunologie, Martinistr. 52, Heinrich-Pette-Institut, Martinistr. 52, 2000 Hamburg 20 2000 Hamburg 20 Dr. Hartmut Kirchner Medizinische Hochschule Hannover, Dr. Axel Ullrich Abt. Hamatologie u. Onkologie, Max-Planck-Institut fUr Biochemie, Zentrum Innere Medizin und Dermatologie, 8033 Martinsried b. Miinchen Postfach 6101 80, 3000 Hannover 61 Dr. Ch. Wagener Dr. Reinhard Kurth Abt. Klinische Chemie, Paul-Ehrlich-Str. 51-59, 6070 Langen 1 II. Med. Klinik, Universitats-Krankenhaus Eppendorf, Martinistr. 52, 2000 Hamburg 20 Dr. Ulrich Krause Bibliothekstr. MCH 6220, Postfach 330440, Dr. Karl Welte 2800 Bremen 33 Kinderklinik d. Medizinischen Hochschule Dr. Klaus Mannweiler Hannover, Konstany Gutschow Str. 8, Heinrich-Pette-Institut fUr Experimentelle 3000 Hannover 61 Virologie und Immunologie an der Universitat Hamburg, Martinistr. 52, 2000 Hamburg 20 France Dr. Roland Mertelsmann Klinikum d. Albert-Ludwigs-Universitat Dr. Jean Claude Cherman Freiburg, Abt. Innere Medizin, Hugstetter Str., Unite de Recherches INSERM U 322 sur les 7800 Freiburg Retrovirus et Maladies Associees, Campus Universitaire de Luminy B.P. 33, Dr. Klaus Munk 13273 Marseille Cedex 9 Deutsches Krebsforschungszentrum, 1m Neuenheimer Feld 280, 6900 Heidelberg Dr. Bernard Malissen Dr. Rolf Neth Centre d'Immunologie, INSERM-CNRS de Medizinische Klinik, Zentrum fUr Marseilles Luminy Case 906, Knochenmarktransplantation, 13288 Marseilles Cedex 9 Universitats-Krankenhaus Eppendorf, Martinistr. 52, 2000 Hamburg 20 Dr. Diane Mathis Laboratoire de Genetique Moleculaire des Dr. Wolfram Ostertag Eucaryotes du CNRS Unite 184 de Biologie Heinrich-Pette-lnstitut f. Experimentelle Moleculaire et de Genetique de I'INSERM, Virologie u. Immunologie, Martinistr. 52, Faculte de Medecine, 11, Rue Humann, 2000 Hamburg 20 67085 Strasbourg Cedex cxx Dr. Francoise Wendling Hungaria Institut Cochin de Genetique Moleculaire, Unite de Recherches en Immunologie Dr. Katalin Paloczi et Oncologie des Maladies Retrovirales, National Institute of Haematology and Blood INSERM U 152, Transfusion, Daroczi u. 24, 1502 Budapest 27, Rue du Faubourg St. Jacques, 75014 Paris Dr. Guy de The Hong Kong Research Director CNRS, Centre National de la Recherche Scientifique, Dr. Li Chong Chan Universite Claude Bernard, Rue G. Paradin, Haematology Section, Dept. of Pathology, 69372 Lyon Cedex 8 Queen Mary Hospital, Pokfulam, Hong Kong

German Democratic Republic Israel

Dr. Elena Elstner Dr. Alpha Peled Universitatsklinik fUr lnnere Medizin The Weizmann Institute of Science, "Theodor Brugsch" des Bereiches Medizin Department of Chemical Immunology, (Charite) der Humboldt Universitat zu Berlin, 76100 Rehovot 1040 Berlin Dr. Michael Fleischhacker Humboldt Universitat zu Berlin, Bereich Italy Medizin der Humboldt Universitat zu Berlin (Charite), Abt. Hamatologie, Dr. Andrea Biondi Schumannstr. 20, DDR 1040 Berlin Clinica Pediatrica Universita Milano, H. S. Gerardo, v. Donizetti 106, Dr. Hartmut Goldschmidt 20052 Monza (MI) Bereich Medizin (Charite), Humboldt-Universitat Berlin, Dr. Federico Caligaris-Cappio Klinik f. Innere Medizin, Sip. Sc. Biomediche e Oncologie Umana Schumannstr. 20/21, Berlin 1040 Sezione Clinic a, Via Genova 3, 10126 Torino Dr. Angelika Muller Dr. Domenico Delia Universitats-Kinderklinik "Jussuf Abrahim" Division OSA, Istituto Nazionale Tumori, der FSU Jena, Kochstr. 2, DDR 6900 Jena Via G. Venezian 1, 20133 Milano Dr. E. Schulze Dr. Guiseppe Saglio Klinik fUr Innere Medizin der KMU Leipzig, Dip. Scienze Biomediche e Oncologie Umana Abt. Hamatologie/Onkologie, Sezione Clinica, Via Genova 3, 10126 Torino Johannisallee 32, 7010 Leipzig Dr. Raffaella Schiro Dr. Sommer Clinica Pediatrica, E. O. Monza, Akademie der Wissenschaften der DDR, Via Donizetti 106, 20052 Monza Forschungsbereiche Biowissenschaften und Medizin, Robert-Riissle-Str. 10, 1115 Berlin-Buch Japan Dr. Michael Strauss Akademie der Wissenschaften der DDR, Tasuku Honjo Forschungsbereich Biowissenschaften Kyoto University Faculty of Medicine, und Medizin, Robert-Riissle-Str. 10, Yoshida, Sakyo-Ku, Kyoto 1115 Berlin-Buch Dr. Toshiyuki Miyashita Dr. Wo((gang Walther Department of Virology, The National Zentralinstitut fur Krebsforschung, Children's Medical Research Center 3-35-31, Lindenberger Weg 80, 1115 Berlin Taishido, Setagaya-ku, Tokyo 154 Dr. Felix Zintl Dr. [sao Miyoshi Universitats-Kinderklinik, Kochstr. 2, Kochi Medical School Okohcho, Nankoku, 6900 Jena Kochi 78-51

CXXI Dr. Shuki Mizutani Sweden Division of Virology, The National Children's Medical Research Center, 3-35-31 Taishido, Prof Kenneth Nilsson Setagaya-ku, Tokyo Prof. of Cellular Pathology, Uppsala University, Department of Pathology, Akademiska Sjukhuset, 751 85 Uppsala

Netherlands United Kingdom Dr. Machteld van der Feltz Dr. Daniel den Hoed Kliniek, afd. Celkweek, Dr. T. Michael Dexter Groene Hilledijk 301, 3075 EA Rotterdam Paterson Institute for Cancer Research, Christie Hospital and Holt Radium Dr. G. Grosveld Institute, Wilmslow Road, Department of Cell Biology and Genetics, Manchester M20 9BX Erasmus University, P.O.B. 1738, 3000 DR Rotterdam Dr. Amanda Fisher HTIG, ICRF, Courtauld Bldg., 91, Riding House Street, Dr. Peter Laird Department of Molecular Genetics, London WIP 8BT The Netherlands Cancer Institute, Dr. Anthony Ford Plesmanlaan 121, 1066 CX Amsterdam Leukaemia Research Fund Centre, Institute of Cancer Research, Chester Beatty Laboratories, Fulham Road, London SW3 6JB Poland Dr. J. Gallagher Dr. Jolanta Pisarek Paterson Institute for Cancer Research, Department of Children Hematology Christie Hospital & Holt Radium Inst., and Oncology, Faculty of Medicine, Wilmslow Road, Manchester M20 98X Bujwida 44, 50-345 Wroclaw Dr. M. Y. Gordon LRF Centre, Institute of Cancer Research, Dr. Monika Plodziszewska 237, Fulham Road, London SW3 6JB Medical Department, Institut Hematology, ul. Chocimska 5, 00-957 Warszawa Dr. Melvyn F. Greaves Leukemia Research Fund Center Institute Dr. W. Wiktor-Jedrzejczak of Cancer Research, Fulham Road, Department of Immunology CSK WAM, London SW 3 6JB 00909 Warsaw 60 Dr. Frank Grosveld National Institute for Medical Research, The Ridgeway Mill Hill, London NW7 1AA Peoples Rep. of China Dr. Ian Kerr Imperial Cancer Research Fund Lab., Dr. Gao Hui Bao P.O. Box 123, Lincoln's Inn Fields, Shanghai Second Medical University, London WC2A 3PX 280 South Chongquing Road, Shanghai Dr. Andrew T. Lister Imperial Cancer Research Fund, Dept. of Medical Oncology, Rep. of China T. Bartholomew's Hospital, London EC1A 7BE

Dr. Po-Min-Chen Dr. Matthias Merkenschlager Medical Oncology, Dept. of Medicine HTIG (ICRF), 91 Riding House Street, Veterans General Hospital Shi-pai, Taipei London W1P8BT

CXXII Dr. AVI'ion MitchisOI1 Dr. Elena Frolova Deutsches Rheumaforschungszentrum Berlin Shemyakin Inst. of Bioorganic Chemistry, Robert Koch Institut, Haus II Miklukho-Maklaya Str. 16/10, Nordufer 20, 1000 Berlin 39 117871 Moscow Dr. Gareth Morgan Dr. T. N. Golovina Leukaemia Research Fund Centre, Shemyakin Inst. of Bioorganic Chemistry, Institute of Cancer Research, USSR Academy of Sciences, Chester Beatty Laboratories, Miklukho-Maklaya 16/10, Moscow 117871 Fulham Road, London SW3 6JB Tatiana Goriatscheva Dr. Martin RaIf Bone Marrow Transplantation, University College London, Dept. Scientific Research, Institute Medawar Building, London WC1 E 6BT of Oncology, Leningradskay St. 68, Pesoschnay-2, Leningrad 188646 Dr. Peter Robinson Dr.K. P. Hanson Transplantation Biology Section, N. N. Petrov Institute of Oncology, Clinical Research Centre, Watford Road, Pesochny-2, Leningradskay St. 68, Harrow 188646 Leningrad Dr. Hans Stauss Dr. Peter Knyazev University College London, Dept. of Biology, N. N. Petrov Institute of Oncology, Medawar Building, Gower Street, Pesoschny-2, Leningrad London WCl E 6BT Dr. G. Korobko Dr. Judit Szollar Shemyakin Institute of Bioorganic Chemistry, King George Hospital, Eastern Avenue, UI. Miklukho-Maklaya 16/10, I1ford, Essex 1G 2 7RL Moscow 117871 Dr. Rose Zamoyska Dr. B. A. Lapin TIU Biology Dept., Medawar Building, Director of Institute of Experimental Gower Street, London WCiE 6BT Pathology and Therapy, Gora Trapezia, Suchumi Dr. Mikhail Markelov Shemyakin Inst. of Bioorganic Chcmistry, USSR Acad. of Science. UdSSR Miklukho Maklaya 16/10, Moscow 117871 Dr. V A. Ncsmeyanov Dr. Boris Afanasiev Shemyakin Inst. of Bioorganic Chemistry. Bone Marrow Transplantation, USSR Academy of Sciences, Dept. Scientific Research Institute Miklukho-Maklaya 16/10, Moscow 117871 of Oncology, Leningradskay St. 68, Dr. Sergey Rodin Leningrad Sector of Molecular Evolution. Theoretical Dr. V. A. Almazov Dept.. Institute of Cytology & Genetics, Director, Inst. of Cardiology, USSR Academy of Sciences, Siberian Branch, Parkhomenko str. 15, St. Petersburg 630090 Novosibirsk-90 Dr. Elena Sadovnikova Mrs. Budkovskaya National Scientific Hematological Center, Shemyakin Institute of Bioorganic Chemistry, Laboratory of Physology of Hemopoesis, Miklukho-Maklya Str. 16/10, Novosykovsky proezd. 4, Moscow 125d167 117871 Moscow Dr. Strelkov Dr. Gregory Dolganov Shemyakin Institute of Bioorganic Chemistry, M. Maklya 16/10, Shemyakin Institute USSR Academy of Sciences, of Bioorganic Chemistry, Miklukho Maklaya 16/10, Moscow 1117871 USSR Academy of Sciences, Moscow V 437 Marina Wolodina Dr. Alexander Friedenstein Bone Marrow Transplantation, Camaleya Institute for Epidemiology Dept. Scientific Research, and Microbiology, Academy of Medical Institute of Oncology, Leningradskay St. 68, Sciences, Camalaya Str. 18, D-98 Moscow Pesoschnay 2, Leningrad 188646

CXXlII USA Dr. Michael J. Lenardo Laboratory of Immunology NIAID, Dr. Perry Blackshear National Institutes of Health, Building 10, Duke University Medical Center, Room 11N311, Bethesda, MD 20892 Diabetes and Metabolism Section, Dr. Kenneth McCredie Dept. of Medicine, Box 3897, The University of Texas, Durham/North Carolina 27710 M.D. Anderson Cancer Center, Dr. Samuel Broder 1515 Holcombe, Houston, TX 77030 National Cancer Institute, National Institute Dr. Atsushi Miyajima of Health, Bethesda, Maryland 20892 DNAX Research Institute of Molecular Dr. Yin-ming A. Chen and Cellular Biology Inc., Dept. of Cancer Biology, Harvard School 901 California Avenue, Palo Alto, of Public Health, 665, Huntington Ave, California 94304-11 04 Boston, MA OF2115 Dr. Malcolm Moore Dr. D. Cosman Memorial Sloan-Kettering, Cancer Center, Immunex, 51, University Str., 1275 New York NY 10021 Seattle, WA 98104 Dr. Stephen Nimer Dr. Michael Dean UCLA School of Medicine, P.O. Box B, NCI-FCRF, National Cancer Div. Hematology-Oncology, Institute, Frederick Cancer Research Facility, CHS 42-121, Los Angeles, CA 90035 Frederick, MD 21701 Dr. Makio Ogawa Dr. R. Gale VA Medical Center, 109 Bee Street, UCLA Dept. of Medicine, Charleston, SC 29403 Hematology/Oncology, Los Angeles, CA 9004 Dr. Susumu Ohno Beckman Research Institute of the City Dr. Robert C. Gallo of Hope, 1450 East Duarte Road, Department of Health & Human Services, Department of Theoretical Biology, National Institutes of Health, National Cancer Duarte, Ca 91010 Institute BId. 37, Room 6All, Bethesda, Maryland 20892 Dr. J. Pierce National Cancer Institute, National Institutes Dr. Sang-Mo Kang of Health, Bethesda, Maryland 20892 Howard Hughes Medical Institute/Howard Medical School/N.I.H., 1 Cloister Ct, Dr. Linda Park Bethesda, MD 20814 Research and Development Corporation, 51, University Street, Seattle, Dr. Gordon Keller Washington 98101 National Jewish Center, 1400 Jackson St., Denver, CO 80206 Dr. Donald Pinkel Dr. Werner Kirsten The University of Texas, NCr Frederick Cancer Research Facility, MD Anderson Cancer Center, Bldg. 427, Room 10, 1515 Holcombe Boulevard, Frederick, MD 21701-1013 Houston, Texas Dr. Phillip KoejJler Dr. U/f Rapp University of California, Los Angeles National Cancer Institute, Viral Pathology Division of Hematology/Oncology, Section, Frederick, Maryland 21701 11-240 Factor Building, 10833 Le Conte Avenue, Dr. Sue Goo Rhee Los Angeles, CA 90024-1678 Head, Section on Signal Transduction, Dept. of Health and Human Services, Dr. Gisela Kramer National Institutes of Health, Dept. Chemistry, University of Texas, Building 3, Room 122, Austin TX 78712 Bethesda, Maryland 20892

CXXIV Dr. Larry Rohrschneider Dr. John Shively Fred Hutchinson Cancer Research Center, Beckman Research Institute. City of Hope. 1124, Columbia Street, Seattle, Duarte. California 91010-0269 Washington 98104 Dr. Michael Rudnicki Dr. Dick Smith Whitehead Institute for Biomedical Research, Lofstrand Labs. Ltd., 7961, Cessna Ave., 9, Cambridge Center, Cambridge, MA 02142 Gaithersbury, Maryland 20879 Dr. Donald A. Rowley Dr. Philippe Soriano Dept. of Pathology, University of Chicago, Institute for Molecular Genetics, 950 E. 54th Street, Chicago, Ill. 60637 Baylor College of Medicine, Dr. Janet D. Rowley One Baylor Plaza, Houston, TX 77030 Section of Hematology/Oncology. University of Chicago. 5841 S. Maryland. Dr. Inder Verma Box 420. Chicago. IL 60637 The Salk Institute, P.O.B. 85800, San Diego, CA 92138 Dr. Ruth Ruprecht Dana Farber Cancer Institute. Dr. John H. Wilson Harvard Medical School. 44 Binney Street. Yerna and Marrs McLean, Boston MA Department of Biochemistry, Dr. Hans Schreiber Baylor College of Medicine, The University of Chicago. One Baylor Plaza, Houston, TX 77030 Department of Pathology, 5841 South Maryland Avenue. Box 414, Chicago, Illinois 60637

cxxv Wilsede Scholarship Holders

Dr. Ayad Atrah Dr. Erik Engel LRF Centre for Childhood Leukaemia, Max Brauer Allee 52, 2000 Hamburg 50, Department of Haematology & Oncology, FRG Institute of Child Health, 30 Guilford Street, London WC1N 1EH, UK Dr. Ortwin Faff Abt. f. Molek. Zellpathologie/GSF, Manuela Bachmann Ingolstiidter LandstraBe 1, 8042 Neuherberg, GSF Miinchen, Abt. Molek. Zellpathologie, FRG Ingolstiidter Landstr. 1,8042 Neuherberg, FRG Dr. John Hancock Department of Haematology, Royal Free Dr. Carsten Bokemeyer Hospital, Pond Street, Hampstead, London Med. Hochschule Hannover, Konstanty• NW3 2QG, UK Gutschow-Str. 8, 3000 Hannover 61, FRG Dr. Stefan Burdach Dr. Thomas Hansen-Hagge Kinderklinik u. Poliklinik, Heinrich Heine Universitiitskinderklinik Ulm, Postfach Universitiit, Abt. f. Piidiatrische 3880, 7900 Ulm, FRG Hiimatologie und Onkologie, Moorenstr. 5, 4000 Diisseldorf 1, FRG Frau Dr. A. Hartmann Stiidtisches Krankenhaus Kaiserslautem, Dr. Moyra Campbell 6750 Kaiserslautem, FRG LRF Virus Centre, Department of Veterinary Pathology, University of Dr. Cordula Holtfreter Glasgow Veterinary School, Bearsden Road, Lab. f. expo Hiimatologie und Onkologie d. Glasgow G61 1QH, UK Kinderklinik Diisseldorf, Moorenstr. 5, 4000 Diisseldorf 1, FRG Dr. Peter Carey Department of Medicine (Haematology), Dr. Klaus Josten Royal Victoria Infirmary, Queen Victoria St. George's Hospital, Dept. of Hematology, Road, Newcastle upon Tyne 1NE1 4LP, UK Cranmer Terrace, London SW17 ORE, UK Dr. N. A. Cicco Jahnstr. 27, 7800 Freiburg i. Br., FRG Dr. Urte Kyas Abt. f. Piidiatrische Hiimatologie/Onkologie, Dr. Dagmar Dilloo Kinderklinik d. Med. Hochschule Hannover, KMT Station, Abt. fUr Piidiatrische Konstanty-Gutschow-Str. 8, 3000 Hannover Hiimatologie u. Onkologie, Med. 61, FRG Einrichtungen der Heinrich Heine Universitiit, Moorenstr. 5, 4000 Diisseldorf, Hans Labonte FRG Montabaurer Str. 8, 5431 Welschneudorf, FRG Dr. Ulrich Duhrsen Abt. Onkologie u. Hiimatologie, II. Med. Dr. M. Lubbert Klinik, Universitiits-Krankenhaus Abt. Innere Medizin 1., Klinikum der Eppendorf, Martinistr. 52, 2000 Hamburg Albert-Ludwigs-Universitiit, Hugstetter Str. 20, FRG 55, 7800 Freiburg, FRG

Haematology and Blood Transfusion Vol. 35 CXXVII Modern Trends in Human Leukemia IX R. Neth et al. (Eds.) © Springer-Verlag Berlin Heidelberg 1992 Dr. Hans Martin Dr. Felicia Mechtild Rosenthal Klinikum d. Johann Wolfgang Goethe• Klinikum der Albert Ludwigs-Universitiit Universitiit, Zentrum Innere Medizin, Freiburg, Abt. Innere Medizin I, Hugstetter Theodor Stern Kai 7, 6000 Frankfurt am Str. 55, 7800 Freiburg, FRG Main 70, FRG Dr. A. P. Schwarer Dr. Peter Moller LRF Centre for Adult Leukaemia, Pathologisches Institut, 1m Neuenheimer MRCjLRF Leukaemia Unit, Royal Feld 220, 6900 Heidelberg, FRG Postgraduate Medical School, Du Cane Dr. Gareth Morgan Road, London W12 ONN, UK LRF Centre of Cellular & Molecular Biology, Institute of Cancer Research, Josef Vormoor Chester Beatty Laboratories, Fulham Road, Steinfurter Str. 63 B, 4400 Munster, FRG London SW3 6JB, UK Dr. Stephan Wagner Dr. Bruce Morland Klosestr. 37, 7500 Karlsruhe, FRG Department of Child Health, Southampton General Hospital, Shirley, Southampton Dr. Ursula Wargalla S09 4XY, UK Westfiilische Wilhelms-U niversitiit Munster, Dr. Michael Neumaier Klinik u. Poliklinik f. Kinderheilkunde, Abt. f. Klin. Chemie, II. Med. Klinik, Piidiatr. Hiimatologie, Onkologie u. Allgem. U niversitiits-Krankenhaus Eppendorf, Poliklinik, Albert Schweitzer Str. 33, 4400 Martinistr. 52, 2000 Hamburg 20, FRG Munster, FRG Dr. Torsten Pietsch Dr. Jeremy Whelan Abt. f. Piidiatrische HiimatologiejOnkologie, Department of Medical Oncology, St. Kinderklinik d. Med. Hochschule, Bartholomew's Hospital, West Smithfield, Konstanty-Gutschow-Str. 8, 3000 Hannover London EC1A 7BE, UK 61, FRG Dr. Chris Price Dr. Christina Zechel Department of Medical Oncology, St. Genetisches Institut der Justus-Liebig Bartholomew's Hospital, West Smithfield, Universitiit, Heinrich Buff Ring 58-62, 6300 London EC1A 7BE, UK Giessen, FRG Dr. Susan Ridge Cornelia Zeidler LRF Leukaemia Unit, Department of Medizinische Hochschule Hannover, Haematology, University of Wales College Kinderklinik· MHH, Piidiatrische of Medicine, Heath Park, Cardiff CF4 4XN, Hiimatologie und Onkologie, Konstanty• UK Gutschow-Str. 8, 3000 Hannover 61, FRG

CXXVIII Acknowledgements

Bundesministerium fUr Jugend, Familie, Leukemia Research Fund, Great Britain Frauen und Gesundheit Leukemia Society of America Deutsche F orschungsgemeinschaft Elisabeth J annsen-Stiftung Deutsche Krebshilfe For their generous hospitality we thank Erich und Gertrud Roggenbuck-Stiftung the Stiftung F.V.S~ zu Hamburg, Verein zur Krebshilfe, Hamburg Naturschutzpark e. v., Hamburg, the Amerikahaus in Hamburg and the Freie Freie und Hansestadt Hamburg und Hansestadt Hamburg. Niedersachsisches Ministerium fUr Wissenschaft und Kunst I would like to thank Dr. Jurgen Wiec• zorek, Ms. Anne Clauss of Springer• Hamburger Landesverband fur Krebs• Verlag for their assistance in the produc• bekampfung und Krebsforschung e. V., tion of the book. On behalf of the authors Hamburg and editors: Rolf Neth.

Haematology and Blood Transfusion Vol. 35 CXXIX Modern Trends in Human Leukemia IX R. N eth et al. (Eds.) © Springer-Verlag Berlin Heidelberg 1992