Molecular biological and monoclonal antibody techniques: their application to the diagnosis, epidemiological study and control of viral infections of man: report on a WHO meeting, Berne, 30 August–1 September 1982

World Health. Organization .,/;. 1tJ Regional Office for Europe ~ Copenhagen EURO Reports and Studies 88 11 Molecular biological and monoclonal antibody techniques Their application to the diagnosis, epidemiological study and control of viral infections of man Report on a WHO meeting

World Health Organization - Regional Office for Europe ~ I Copenhagen ~ EURO Reports and Studies 88 Molecular biological and monoclonal antibody techniques Their application to the diagnosis, epidemiological study and control of viral infections of man Report on a WHO meeting Berne 30 August - 1 September 1982 ICP/BVM 002 ISBN 92 890 1254 4 © World Health Organization 1983 Publications of the World Health Organization enjoy copyright protection in accordance with the provisions of Protocol 2 of the Universal Copyright Conven- tion . For rights of reproduction or translation, in part or in toto, of publications issued by the WHO Regional Office for Europe application should be made to the Regional Office for Europe, Scherfigsvej 8, DK-2100 Copenhagen 0, Denmark. The Regional Office welcomes such applications. The designations employed and the presentation of the material in this publica- tion do not imply the expression of any opinion whatsoever on the part of the Secretariat of the World Health Organization concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. The mention of specific companies or of certain manufacturers' products does not imply that they are endorsed or recommended by the World Health Organiz- ation in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. The views expressed in this publication are those of the participants in the meeting and do not necessarily represent the decisions or the stated policy of the World Health Organization. PRINTED IN DENMARK ISSN 0250-8710 CONTENTS Page Introduction Molecular biological methods relevant to the epidemiological study, diagnosis and control of viral disease . . . . . . . . . . . . . . . . . . 2 Methods of characterizing viral RNA Methods of characterizing viral DNA 3 7 Viral protein characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Some applications of modern molecular biological methods to medical virology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Migration rate analysis of RNA gene segments in viruses with segmented genomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 RNA oligonucleotide mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 DNA endonuclease restriction mapping of human herpesviruses 19 Molecular studies of poxviruses and adenoviruses . . . . . . . . . . . . . . . . . . . . . 21 Molecular hybridization of viral genes - studies of influenza viruses . . . . . . . . 21 Molecular biological studies of papilloma viruses . . . . . . . . . . . . . . . . . . . . . . 23 Applications of DNA recombinant technology to the epidemiology and diagnosis of HBV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Production and properties of monoclonal antibodies . . . . . . . . . . . . . 25 Rationale for production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Properties of monoclonal antibodies 26 Applications of monoclonal antibody in viral diagnosis and epidemiological studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Use in epidemiological and diagnostic studies . . . . . . . . . . . . . . . . . . . . . . . . 29 Investigations of virus structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Some limitations of monoclonal antibodies . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Monoclonal antibodies and molecular mimicry . . . . . . . . . . . . . . . . . . . . . . . 32 Applications of molecular biology and cell hybridization towards the eventual control of viral infection and disease 33 Genetic engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Potential applications of DNA recombinant technology . . . . . . . . . . . . . . . . . 34 Influenza viruses: DNA cloning and expression . . . . . . . . . . . . . . . . . . . . . . . 34 Hepatitis B vaccines: perspectives of new methods of vaccine preparation . . . . . 36 Synthetic peptides as viral antigens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Developments in the immunology of infectious agents . . . . . . . . . . . . . . . . . . 37 Summary of some major practical applications of the new biotechnologies relevant to the prevention and control of viral disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Recombinant DNA technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Hybridoma technology 38 Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Annex 1. List of participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 INTRODUCTION A WHO working group on molecular biological methods applicable to the epidemiology, diagnosis and control of viral infections was held in Berne, Switzerland from 30 August to I September 1982. The participants were welcomed by Dr B. Velimirovic, Regional Officer for Communicable Diseases, who opened the meeting on behalf of the WHO Regional Direc- tor for Europe. From the beginnings of medical virology early in the twentieth century up to the 1970s, the characterization of medically important viruses for diagnostic and epidemiological purposes has relied almost entirely on serological and biological tests. Indeed, the rapid developments in medical virology that occurred from the middle of the 1950s were made possible because of the development and wide availability of tissue culture tech- niques for virus cultivation, an essential prerequisite for much biological and serological work. These approaches continue to serve an essential role in many areas of routine and research work in virology. They are now complemented, however, by developments in molecular biological methods and new possibilities for serological investigation. The past decade has seen the development of many molecular biological techniques applicable to the detailed characterization of viral nucleic acids and proteins, which might be regarded as a second phase in the develop- ment of virology. In addition, new methods for the antigenic character- ization of viruses based on hybridoma technology are now available. Among the specific techniques now available for the characterization of RNA virus genomes are migration rate analysis by polyacrylamide gel electrophoresis (PAGE) and oligonucleotide mapping (RNA fingerprint- ing), which provides for detailed comparisons of RN A from related viruses. The genomes of DNA viruses can be characterized and compared by mapping specific restriction endonuclease sites. For both RNA and DNA viruses, molecular hybridization of nucleic acids and the detailed com- parison of strains provide a method for detecting virus-specific nucleic acid sequences in cell materials, an approach of value for diagnostic and epidemiological investigators. Modern methods, including in particular re- combinant DNA technology and gene cloning, have allowed particular com- plete sequences of the genomes of RNA or DNA and viruses to be obtained, thus leading to their definitive characterization and the precise comparison of related viruses. Viral proteins can be characterized by gel migration methods, by peptide mapping procedures, and ultimately by amino acid sequencing. Prediction of the amino acid sequences from the primary nucleotide sequences is of considerable value. The development of cell fusion technology, which has enabled the preparation of specific monoclonal antibodies directed against defined viral antigens, has provided a powerful tool for the precise antigenic characterization and comparison of viruses, and for the precise analysis of the role that individual virus components contribute to the protective immune response of the host or to immunopathology. These approaches are directly applicable to improving the diagnosis and epidemiology of viral infection and to controlling viral disease through the development of new, cheap, more effective, or safer vaccines and through chemotherapy. The meeting was convened to provide for a broad review of molecular biological approaches in the field of medical virology, but was not in- tended to provide a complete coverage of the field of viral disease. The reader's attention is therefore drawn to several specialized reports that may serve to complement the contents of the present report (1-5). MOLECULAR BIOLOGICAL METHODS RELEVANT TO THE EPIDEMIOLOGICAL STUDY, DIAGNOSIS AND CONTROL OF VIRAL DISEASE The past decade has seen the wide application of several molecular bio- logical approaches to the diagnostic and epidemiological study of viruses of medical importance. These methods and their potential value to studies on RNA and DNA viruses are outlined below. It is not the intention of this report to provide a detailed technical description of the methods, which can be obtained from the many excellent and specialized scientific pub- lications available, but rather to review the main scientific concepts and the relative advantages and disadvantages of the techniques available, to- gether with some of their important applications to medical virology. Table I provides a summary of the methods available, some of the major applications, and a comparison of their technical complexities. 2 Methods of characterizing viral RNA During RN A virus replication, mutations very frequently arise in the viral genome such that alterations at 0.1-3% of the genomic nucleotides are detected after only a few cycles of replication in man or in animal hosts ( 6) . Mutations in viral RNA are useful markers for strain identification, and methods for the rapid characterization of RNA molecules are of value in a wide range of epidemiological work for this group of viruses. Migration rate analysis in polyacrylamide gels For viruses with segmented RNA genomes, such as influenza viruses, rotaviruses and reoviruses , the individual gene segments can be partially characterized and compared by analysis of their electrophoretic mobilities in polyacrylamide gels. The method is simple and practicable, and permits the analysis of many samples at one time. For single-stranded RNA molecules, one or a few mutations can significantly alter the electrophoretic mobility of the segment. However, identity of mobility of gene segments for different virus strains does not necessarily imply high nucleotide se- quence homology. The method is proving a useful tool and is becoming widely employed in epidemiological work with rotaviruses and influenza viruses. A useful application is not only the characterization of field isolates for epidemiological purposes but also the analysis of genetically reassorted viruses, either produced in the laboratory or occurring naturally, so that the parental origin of individual gene segments of the reassortment can be established (7). The latter can be done with certainty only by using the additional methods set out below. O/igonuc/eotide mapping of viral RNA genomes Any viral RNA can be analysed by oligonucleotide mapping, a method first used by Szekely & Sanger ( 8) and subsequently developed ( 7, 9). This is a powerful and highly reproducible method for separating longer oligo- nucleotides, and allows the detailed comparison of closely related viral RNA molecules. Purified RNA (0.2-0.4µg) is digested with a base-specific RNAse; in general RNAse T 1 is used which cleaves after guanosine residues. The T 1-resistant oligonucleotides are 5' -terminus labelled with ATP-y-32P and polynucleotide kinase, and the labelled oligonucleotides separated by two-dimensional PAGE. In the first dimension, a pH of 3.5 is chosen to maximize the separation of oligonucleotides according to their nucleotide composition and size. In the second dimension at pH 8.0, the separation is based on molecular weight differences only. The oligonucleo- tide pattern is visualized by autoradiography. Base substitutions at the unique T 1 oligonucleotides are readily detected. 3 ~ Table 1. Molecular techniques for characterization of viruses, and their applications Method Main fields of application Spec ial requirements Technical Usesb complexity8 RNA virus migration rate analysis in influenza virus si lver staining ++ DER genomes polyacrylamide gels rotavirus radioisotopes not essential T -oligonucleotide influenza virus. po liovi rus rad ioisotopes +++ DER 1 . flavivirus. rotavirus mapping molecular hybridization influenza vi rus. rotavirus rad ioisotopes and specialized +++ ER enzymes base sequenc ing wide potential implications radioisotopes and specialized ++++ R enzymes DNA vi rus restrict ion endonuclease adenovirus . herpesvirus. cyto- silver staining ++ DER genomes mapp ing. agarose or megalovirus, varicella virus. rad ioisotopes. polyacrylamide gel papilloma virus. poxvirus special ized enzymes migration analysis molecular hybridization herpesvirus radioisotopes +++ DER hepatitis B virus specialized enzymes base sequenc ing wide potential applications radioisotopes ++++ R enzymes V, Table 1 (contd) Viral proteins ant1gen1c analysis migration rate analysis in polyacrylamide gels 1soelectr1c focusing peptide mapping amino acid sequencing widely applicable influenza virus pol1ov1rus poliov1rus influenza virus generally applicable 8 Degrees of technical complexity. from simple(+) to very co mplex(++++) b O = d1agnost1c use E = use in ep1dem1olog1cal studies. A = use 1n research work none radioisotopes usually required rad101sotopes usually rad101sotopes + + ++ +++ ++++ DER ER ER EF R The method detects any mutations located in the large, unique oligo- nucleotides, which generally make up 7-10% of the total viral RNA sequences. It is useful to compare closely related viral RN As, such as those from viruses isolated during a single outbreak of infection. If the base sequence homology of the RN A of two viruses is below 90% the probability of detecting a common unique oligonucleotide is low. This method has been used extensively for the characterization and comparison of polio- virus, influenza virus and, recently, rotavirus genomes, as outlined later (see p. 17). Molecular hybridization of RNA for gene analysis The genetic relatedness of viral subtypes can be estimated by the technique of molecular hybridization , using riboendonucleases that specifically at- tack single-stranded but not double-stranded RNA molecules. This tech- nique takes advantage of the fact that double-stranded RNA is resistant to T 1 or pancreatic RN Ase, while single-stranded RN A is digested to acid- soluble oligonucleotides. One strand of RN A is labelled and annealed to a surplus ofunlabelled complementary RN A ( cRN A). If cRN A of a genetic- ally related strain is used for hybridization, the regions of mismatching (indicating genetic diversity) will be digested, leaving intact the domains with a perfect base sequence homology. Recently cRNA has been made by in vitro transcription from rotavirus cores, and has been used for liquid phase and filter dot hybridization assays to determine relationships of human and animal rotavirus RNA genomes (JO). Alternatively, cDNA probes can be used to analyse rota- virus genomes by Northern blot hybridization (11). Nucleotide sequencing of RNA virus genomes RNA is terminally labelled and then partially cleaved by chemicals (12) or by dose-specific enzymes (J 3) and the products are separated by PAGE. Sequences are read from the band patterns. Alternatively, RNA may be reverse transcribed into DNA and the dideoxynucleotide chain termin- ation method (14) applied, either using a terminally labelled primer or incorporating a labelled triphosphate. These primer-extension techniques are powerful tools for the rapid determination of partial gene sequences (I 5). The chain termination method has often been used after cloning the viral ON A into bacteriophage M 13 containing single-stranded ON A. Alter- natively, a double-stranded DNA copy of the viral RNA may be cloned in a suitable vector and sequenced by the Maxam & Gilbert chemical degrad- ation method (16). For example, the sequences of certain picornavirus, influenza virus and parts of rotavirus genomes have been determined by these recombinant DNA procedures. 6 Whilst complete sequence analysis provides a definitive description of a genome, it can be a time-consuming procedure and unsuitable when a large number of genomes are to be compared. Partial sequences of viral genomes can more readily be obtained by either end sequencing or primer- extension methods, and have been of considerable value for studies of influenza viruses ( 17) and polioviruses ( 18). Methods of characterizing viral DNA Many DNA viruses of medical importance, such as human (alpha) herpes- viruses I and 2 (herpes simplex viruses I and 2), human (alpha) herpesvirus 3 (varicella-zoster virus), human (beta) herpesvirus 5 (cytomegalovirus), pox- viruses and adenoviruses, can be propagated in vitro. In contrast other DNA viruses, for example human papilloma viruses and hepatitis B virus (HBV), have not been cultivated in vitro, while others, such as human (gamma) herpesvirus 4 (Epstein-Barr virus) and Creutzfeldt-Jakob disease agent, grow only poorly. This has hampered the detailed molecular study of these viruses. However, the problem has now been largely overcome by molecular cloning of their genomic DNA. It is now possible to obtain large quantities of purified viral DNA fragments, and to prepare specific radio- labelled probes for use in diagnostic and epidemiological procedures. Analysis of viral DNA may be performed with DNA from infected tissues, or with radiolabelled or non-labelled DNA isolated from virus particles or infected cells. The general techniques available for the analysis of viral DNAs for diagnosis or epidemiological studies are outlined below. Restriction endonuclease mapping of viral DNAs Restriction endonucleases are bacterial enzymes that cleave DNA molecules at specific sites, and this produces a characteristic set of fragments. The choice of enzyme depends on the size of the viral DNA (from 3200 base pairs for HBV to 2.5 X l05 base pairs for poxviruses). The fragments resulting from treatment of viral DNA with restriction endonuclease are resolved by electrophoresis in agarose or polyacrylamide gels, and are visualized by staining with ethidium bromide or silver nitrate or by auto- radiography. DNA cleavage patterns have allowed the identification of different papilloma viruses, the typing of human (alpha) herpesviruses I and 2, and the identification of genetic variants of types 1, 2 and 3. The characterization of bacterial plasmid DNAs by endonuclease map- ping is of considerable importance as plasmids are used as vectors for cloning viral genomes and other DNAs. Also, plasmids are important as components affecting pathogenicity and drug resistance of bacteria. In 7 outbreaks of nosocomial bacterial infections, plasmid characterization by incompatibility testing is of particular epidemiological interest. DNA hybridization procedures Blot hybridization. DNA fragments, generated by restriction endo- nucleases and separated by gel electrophoresis, are denatured and trans- ferred by "blotting" on to a nitrocellulose membrane to which they are firmly bound. The immobilized DNA is identified by hybridization with virus-specific DNA probes, and radiolabelled with 32P by nick translation; the labelled bands are then detected by autoradiography. This approach has been of value, for example, in detecting HBV DNA in liver samples in connection with studies on the relationship between chronic HBV infection and primary hepatocellular carcinoma. Dot hybridization. The DNA is directly immobilized on nitrocellulose filters and is detected with a radiolabelled ON A probe. This procedure has been used to detect HBV in serum samples and is suffi ciently sensitive to detect 104 - I 05 virus particles per millilitre of serum. It has also been used to identify human (gamma) herpes virus 4 ON A in cells from patients with nasopharyngeal carcinoma. DNA hybridization in situ. This method allows the detection and local- ization in tissue sections or cells of virus-specific DNA or RNA molecules transcribed from the viral genome. The radiolabelled DNA probes are hybridized in situ and the hybrids identified by autoradiography and microscopy. The procedure has been used to detect RNA complementary to the ON A of human (alpha) herpesviruses I and 2 in biopsy samples from patients with cervical carcinoma, and to detect human (gamma) herpes- virus 5 DNA in biopsy samples from nasopharyngeal, tonsillar and supra- glotic laryngeal carcinomas. Nucleotide sequencing of DNA virus genomes Two alternative methods of ON A sequencing can be applied: the Maxam & Gilbert chemical degradation method (16) and the dideoxy chain termin- ation method (14). These can be used for sequence analysis of DNA viral genomes after cloning all or parts of the genome as described below (pages 15, 19-25). Viral protein characterization Antigenic analysis has been widely exploited for the characterization of virus-specific structural or non-structural proteins for diagnostic and epi- demiological work. This approach is extensively covered in the virological literature and will not be further discussed here. 8 A number of biochemical procedures involving analyses of virus-specific proteins are of potential value for discriminating between related viruses, and usefully complement the results of antigenic analyses. The simplest are comparisons of electrophoretic mobility in polyacrylamide gels, or of isoelectric points of identifiable virus-coded proteins extracted from puri- fied viruses or from virus-infected cells. More detailed comparisons of structurally equivalent proteins, by chromatographic and electrophoretic analysis of the peptides obtained following their proteolytic digestion , have been of considerable value, for example, in research on antigenic variation. The most definitive comparisons of viral proteins a re achieved by partial or complete amino acid sequence determinations; these, however , require the most sophisticated equipment and are unlikely to be under- taken routinely. Predicted amino acid sequences of viral proteins are commonly obtained from the base sequence of the appropriate coding region of DNA, and sequencing of the proteins often provides essential confirmatory data on the N- and C-terminal regions of the proteins. For ease of reference a number of the basic techniques described above are shown in diagrammatic form in Fig. 1-6. Descriptions of the methods are provided in the notes to the figures . SOME APPLICATIONS OF MODERN MOLECULAR BIOLOGICAL METHODS TO MEDICAL VIROLOGY Migration rate analysis of RNA gene segments in viruses with segmented genomes Two families of viruses of medical importance have segmented RNA genomes, the Reoviridae and the Orthomyxoviridae. RNA segment analy- sis has been extensively employed in studies of rota viruses and reoviruses, both members of the Reoviridae , and also of influenza viruses. Rotaviruses are now recognized as major etiological agents of acute enteritis in man and several animal species. These agents contribute sig- nificantly to the enormous impact of diarrhoeal disease on infant mortality in many countries. Epidemiological surveys of these important agents in man have been complicated by the difficulties of culturing the human rota virus strains, and consequently of defining virus serotypes. Because of these difficulties , studies on the direct genetic characterization of rota- viruses in human clinical specimens , not requiring cultivation of the agents, play a useful role in epidemiological studies. The genome of rota viruses consists of 11 segments of double-stranded RNA classified according to size in four groups: segments I, 2, 3 and 4 9 -0 Fig. 1. Chara cterizat ion of fragm ented RNA vi rus genomes by mig ration rate analysis in polyacrylam ide gel s Fluorocarbon extraction Phenol extraction Centrifugation / \ l / . . .. . . Free RNA in Faeces resuspended in EDTA the aqueous phase r r r 1 \ \ / Ethanol -precipitation and resuspension Aqueous phase ~1 - ) Purified RNA PAGE This technique 1s especially useful 1n rotav, rus characte rization. Vi ral RNA 1s extracted di rectly from faec es by f luorocarbon and phenol extraction and centr1fugat1on. RNA present in the aqueous phase is purified by ethanol precipi tat ion and ana lysed by PAGE . By comparison of fragment m1grat1on patterns of different strains of rotavirus. an index of genetic relat1onsh1p between them can be established . ...... ...... Fig. 2. Oligonucleotide mapping of viral RNA "In vivo" labelling "P?!~•·M ®--~) M ~ disruption (:) • disruption 1-(. ! Cell ~ Virion ~ --- ~ :. ..~::on sr Infection in radioactive~abelled medium Replication "I n vitro" labelling @ ~ e) Cell 0 d~onO O - 00 o @~ Oo 0 0 Infection in cold medium Free virions Free Free labelled RNA labelled virions in the aqueous phase Virion ~n Phenol extraction ~ . . Free ANA in the aqueous phase ~i ~ I C ~ 0 G ·;;a .2'~ ::.E w Purified RNA o~~ 0 00 0 ct> 0 o:S,~ O O 0 o .; ., ; :;, ',:•t.,•-,..-,,;, , :: ~ RNAseT1 ~ ,,,,., ....... _, ' ~ ~::~ .. ,.~ Purified labelled RNA Uniformly labelled oligonucleotides RNAse T1 WPolyn_ucleotidu ~ " P-ATP 1.'{ _; Unlabelled oligonudeotides Oligonucleotides 5'- 31 p Migration_!_ense {elec!!_ic charge) ~o~ ~•M ~•U Autoradiography App(ication Application ~ point : .;;_, '-' 1,, = r,; 1~, point Oligonucleotide map of the RNA analysed Vira l o l1 gonucleot ides may be obtained d ire ctl y 1n a labelled form by culturing vi rus in the presence of ATP-y-3 2P. followed by extract io n of labelled vRNA by pheno l and RNAse T digestion (" in vivo" labelling) . On the other hand. ol igonucleotides may be prepared fr om col d vRNA by RNAse T d igestion an1d subsequently labelled 1n the 5' terminus using polynucleotide kinase and ATP-y- 32 P as phosphate donor ("i n vi trci" labelling) . The ol igonucleotide mixture obtained by either of these methods is processed in a two-d imensional gel electrophoresis system. f i rst at pH 3.5 (migration dependent on electric charge) and then at pH 8 .0 (migration dependent on molecular weight). The gel is processed by autoradiography to obtain an oligonucleotide map character- istic of the RNA analysed . Sy comparing maps obta ined with different strains of v irus . different levels of genetic relationship may be established between them . -N Fig. 3. RNA-RNA hybridization with virion RNA ·::~-~ (CT\ ~ distupti2f1 0 0 Cell ~ Infection in radioactive-labelled medium Replication Radioactive•labelled virus 0 Free labelled virions Virion ~ disruption /J~ ~ Phenol V:-.-:". ~ ex traction •. ~· •. -~ :_ .. . . ·:-.· . Free labelled RN:e in the aqueous ph 6j 3 • 1 §· 2s 1 • ~ ., ;.Single-<trand-<pecific ~ 2 • nuclease • Hybridization PAGE or Measurement of ~ • • .- acid-insoluble 3 §. radioactivity tJ by filtration Partial double-stranded hybrid cRNAof reference strain + UNFAAGMENTED GENOMES ~ Elution of labelled fragment or complete genome l'urified labelled RNA I GMENTEO ENOMES ~ Labelled vRN A is obtained by cultu ring vi ru s in the presence of ATP-y- 32 P. followed by virion extraction and disruption and vRNA phenol extraction . Whole vRNA or vRNA fragments previously separated by PAGE (viruses wi th a fragmented genome. 1.e. influenza viruses) hybridize with unlabelled cRNA of a reference strain . Genome sequences that do not show complementarity rema in as single -stranded zones in the hybrid and can be totally digested by incubat ion with a single-strand-specific nu clease. Double- stranded RNA fragments obtained in this way are precipitated by add ition of ethanol and collected on filte r paper. and ra dioact ivi ty is measured. The percentage of hybridization inferred from the radioactivity measuremen ts constitu tes an in dex of the genetic relationsh ip between the analysed strains . Cycloheximide Fig. 4. RNA-RNA hybridization with complementary RNA ll P-phosphate -v ~ "\;;;I Infection in radioactive-labelled medium in presence of cycloheximide @ cRNApurilication ~ 1 \°\ L~ Synthesis of radioactive-labelled cRNA Radioactive-labelled cRNA S El· ~ I Single-strand,.peci fic nuclease PA . FRAGMENTED GENOMES PAGE UNFAAGMENTEO GENOMES • Hybridization /1- Partial double-stranded vRNA of reference strain Measurement of acid-insoluble rad ioactivity by filtration hybrid + ~ - :, ii w- Labelled fra~ent or complete genome Labelled cRNA 1s obtained by culturing virus in the presence of ATP-y- 32 P and cyclohex1m1de . This drug inh ibits viral protein synthesis . and therefore does not permit viral repl1 case synthes is without affecting the cRNA synthes is. wh ich is accomplished by the viral transcriptase present in the v1rions of all negative strand RNA viruses. Vira l cRNA extracted from cel ls and purified 1s subjected to hybrid izat ion w ith vRNA of a reference strain of virus . Genome sequences that do not show complementarity (single-stranded zones 1n the hybrid) are totally digested by a s1ngle-strand-spec1f1c nuclease. and the double-stranded RNA fragments obtained are ethanol precip itated and collected on filter paper . Radioact1v1ty is measured on the precipitate and the w percentage hybridization calculated . These data constitute an index of the genetic relationship between the analysed strains . -""" Fig . 5. Restriction endonuclease analysis of viral DNA "In vivo" labelling "P~h~{O\ @y--- v Infection in radioactive~abelled medium Unlabelled virus M ®~u Infection in cold medium ~ Ci) (!} Replication E!J Replication = -- C OM .fl E ·s ~ !l ~ a: 0. Cell disruption '®j lf) Virion ~ Phenol extraction ~ u . ~ r ·- ; i' ... Free labelled virions Free labelled DNA in the aqueous phase Purified labelled DNA Cell disru_etioo = = C C ON o - .fl E .fl E ·~ ~ ·s ~ ~ ~ ~ 0. a: 0. ~ 0 co Free virions Virion [j..._ w ~ /) ------ - Phenol :- ~:: ~ _, .- extraction ·._:._:.: :<· / / '\ Free DNA in the aqueous phase Purified unlabelled DNA I Autoradiography for / 11 - I 11.,,,;c,;°" labelled virus end°" Restriction fragments 1 "Cfea:se 1 / PAGE / ;-,-1._ / Re,triction endonuclease: c\ea¢ Restriction fragments 2 . ef\Oof'"' ' "" ~P-Fluorescence staining for ,\ -j/1 L•~;led unlabelled virus Restriction fragments 3 unlabelled DNA Viral DNA is prepared by pheno l extraction from a crude preparation of cold medium (un labelled v irus) or ATP-y- 32 P labelled med ium ( .. in vivo .. labelled virus) infected cell s. DNA o bta ined in this way is digested with selected restriction endonucleases and the resulting restriction fragments are analysed by PAGE . Aft er autoradiography or fluorescent staining of slabs. different levels of genetic relationship between vi rus strains ca n be established on the basis of restriction patterns . -Vl ... Viral RNA L__ Reverse transcriptase Oligo dT Polyadenylated viral RNA Pst I restriction point 0 Pst I Cloned RNA products Fig . 6. Cloning of viral nucleic acid J"C C C 'f'" RNAse • or a'kali Reverse • i Ol igo dC Viral ONA ~ J " 11 s· s· C , · 1 ~ S~ ~ tailing.with ~__J . . , . ' ) ' A - l ) " ONA-RNA hybrid or or ONA terminal heating polymerase · transferase , ,. , . ... ,., .. ,. ,· ,. ,. C 1 ; :: : : : / Complementary Double-stranded ~ single-stranded DNA , .• . . DNA copy ~- cONA , -~, OligodG r r tl::irnng ,- th ~ j trans ,· ,· - 1inal 'erase ,.I /'fJ,__~,'at~ •r,?t, Lineal ~ plasmid ONA i ,· Expression ~(J~ ~ Selection.by antibiotic. oa resistance 'V ~ {('<,..;,..,o'(I. (6\ot - £. . coli \_t• - ~ C·(; C-G C -c; C • Ct u Recombinant DNA Viral RNA can be copied into cDNA by reverse transcriptase and DNA polymerase I. The DNA 1s treat ed with S -nuclease. tailed with 01190 dC by terminal transferase. and annealed to Pst I -cleaved. 01190 dG -tailed pBR 322 DNA. Alternat1vely. ol1gonucleotide linkers may be used to JO in the two DNAs . Transformation of E. col, and selection of Tc'. Ap' transform ants yields recombinant DNAs with pBR 3 2 2 including inserts of viral cDNA. For DNA viruses the reverse transcription step 1s not needed. and DNA fragments may be cloned into a suitable vector using ol1gonucleot1de tails. linkers or after cleavage with restr1ct1on enzymes. comprise group I, segments 5 and 6 group II , segments 7, 8 and 9 group III, and segments 10 and 11 group IV (19). The electrophoretic mobility of the RNA segments may differ for rotavirus strains obtained from different hosts, or for viruses from the same host. Rotavirus RNA may be obtained from virus particles purified from faeces, where virus particles may be present at high concentrations, by sedimentation in cesium chloride gradients. A simple and rapid technique has been developed for recovery of RNA (20). The RNA is extracted from virus by treatment with phenol , precipitated with ethanol , loaded on to polyacrylamide gels, and the RNA segment separated by electrophoresis (21). This procedure has a number of advantages: only small quantities (about 0.2 g) of faecal material are required; it is possible to treat 20 samples of faeces in one hour; and all the extraction steps can be made in closed containers. This simplified method makes epidemiological surveys of rota- virus infections practical. The technique may be made highly sensitive by silver staining of RNA (22); 32 1 labelling of viral RNA followed by auto- radiography can also be used to demonstrate the RN A bands after PAGE. In attempts to characterize and compare rota virus strains isolated from a variety of sources, the relative electrophoretic mobilities of the genome segments described above have been precisely measured. By this approach the electrophoretype of an isolate can be expressed. Studies in several laboratories have indicated a wide genetic heterogeneity among different human rotaviruses. RNA oligonucleotide mapping This method has been used widely for characterizing RNA viruses since its development in I 972. Applications include the study of influenza viruses , polioviruses , measles virus, rhabdoviruses, corona virus, toga viruses, flavi- viruses, retroviruses , and viruses of the African haemorrhagic fever group. Comparisons of maps from RNA of viral isolates may be of value for epidemiological and diagnostic studies and for the precise identification of vaccine virus strains. Limitations of the method are: (a) only a small proportion of the viral genome _is screened unless segments of RN A of low complexity, such as individual influenza virus genes, are se lected for mapping; and (b) the oligonucleotide pattern may not necessarily correlate with any biological property of the virus , such as pathogenicity. Even though T I mapping involves labelling and purification of viral RN A and a two-dimensional electrophoresis, it is a relatively simple and reproducible method for the characterization of RNA genomes , and it has already established itself as a potent tool for studies of the epidemiology of RN A viruses. 16 Influenza viruses Applications of this method to epidemiological and surveillance work in influenza have recently been reviewed ( 4) and will not be discussed in detail here. A good example of the power of the method for the genetic analysis of influenza viruses is its application to the study of influenza A (HIN I) viruses isolated in 1977. These viruses were shown (23) to be genetically very close to HIN l viruses that had been in circulation some 27 years earlier in 1950, suggesting that the 1977 virus had re-emerged from a genetically well preserved but unidentified repository of this virus in nature. Oligonucleotide maps of the eight genes of the 1977 virus showed their close identity with each of the corresponding genes of the 1950 virus. Oligonucleotide mapping was instrumental in proving that reassort- ment occurs in nature within influenza A virus subtype HIN l (24). Polioviruses Studies with polioviruses serve as an excellent example to illustrate the use ofT 1 mapping in diagnostic and epidemiological work. Oligonucleotide maps of poliovirus RNA may be analysed to different degrees of detail depending on the nature of the information required. At the simplest level, the oligonucleotides of individual strains form readily recognizable pat- terns on visual inspection, and genetically unrelated strains from a single poliovirus type produce entirely distinct patterns (25). Where more de- tailed analysis is required, for example when strains appear to give similar oligonucleotide constellations, it is convenient to number the individual oligonucleotides to make quantitative comparison easier. Usually 60 or more oligonucleotides may be identified in this way. Firm identification of the oligonucleotides, and thus complete analysis of the map, requires co-migration of digests of the pairs of strains to be compared. Individual oligonucleotides may be sequenced to check their identity. Epidemiology. With oligonucleotide mapping it has proved possible to follow individual outbreaks of poliomyelitis caused by wild poliovirus strains. In one such instance (26), closely related type I polioviruses were isolated from communities in Canada, the Netherlands and the United States. Epidemiological evidence indicated that the outbreaks were prob- ably linked, and this conclusion was confirmed by oligonucleotide map- ping of the viruses, which remained clearly recognizable even though they were isolated over a period of 13 months. There was, however, evidence of some alteration in the causative virus during this transmission period, involving an estimated 100 base changes in all from a genome of7500 bases. The maps of all strains were similar, though distinct. On a more general level, it has been shown in some countries where live attenuated vaccines are widely used that wild strains have been displaced 17 by vaccine derived strains. This conclusion has been confirmed by oligo- nucleotide mapping of strains isolated from individuals with no sign of poliomyelitis. There are frequent modifications of the maps in such cases when comparisons are made with vaccine virus, but in all cases at least three quarters of the 55-60 oligonucleotides found in maps of RN A of the parental Sabin vaccine are to be found in the maps of RNA of such isolates. Similarly, strains isolated from poliomyelitis patients who had been given live vaccine have been shown to be closely related to the vaccine strain . In some cases these strains were isolated post mortem from the central nervous system, and they have been shown to cause paralysis in monkeys. This strongly suggests that they were the causative agents of the disease, even though they were derived from the Sabin vaccine. Prolonged excretion of virus has been observed in certain immuno- deficient people fed oral poliomyelitis vaccine. In such cases sequential isolates have been examined to follow the evolution of the virus in the human gut. Oligonucleotide mapping has not yet been extensively applied to the evolution ofpoliovirus strains in the wild; however, a WHO collab- orative study is in progress and the results to date suggest that many poliovirus strains of each serotype may circulate concurrently. Biological properties. Because only 10-25% of the RN A is ultimately found in the large, unique T 1-resistant oligonucleotides characteristic of the oligonucleotide map of the RNA, relationships between changes in biological properties and the oligonucleotide maps of strains are likely to be rare. Two type 3 strains (Leon, the precursor of the Sabin type 3 vaccine strain, and the revertant strain 119, isolated from a fatal case of vaccine- associated poliomyelitis) are both highly neurovirulent for monkeys, but give identical maps to the nonvirulent Sabin type 3 strain itself. A more detailed analysis, such as complete base sequencing, is therefore required to determine the genetic basis of the difference in the virulence of these three viruses. On the other hand, relationships between biological properties and oligonucleotide maps do arise, and have been demonstrated in antigenic variants of poliovirus type 3 Leon. Non-neutralized antigenic variants were selected in the presence of virus neutralizing monoclonal antibodies , and a specific and characteristic alteration in a single oligonucleotide was associated with alterations in the antigenic properties of the mutants . This indicated that the mutations were concentrated in a small portion of the RNA (27) , and led to the identification of a small region of the capsid protein, eight amino acids in length, as a major antigenic site for virus neutralization. The particular portion of the genome detectable in oligonucleotide maps varies from strain to strain, so that different strains give oligo- nucleotides that are not only different in their chemical composition but 18 also ongmate from different parts of the RNA . Clear relationships between biological properties and oligonucleotide maps of strains might therefore be expected only by selecting appropriate groups of related strains for study. Rotaviruses Two-dimensional oligonucleotide mapping has been applied to the analy- sis of rotavirus field isolates. During a single outbreak over a period of 3-4 months, isolates with an identical RNA migration pattern differed one from another by sequential point mutations, whereas isolates with differ- ences in their RN A migration patterns differed exteMsively in the positions of their larger, unique T 1-resistant oligonucleotides (28). DNA endonuclease restriction mapping of human herpesviruses Molecular biological tools have been used to examine the epidemiology of infections by human herpesviruses; these studies have largely concentrated on human (alpha) herpesviruses 1 and 2, although some information is available on human (beta) herpesvirus 5 and human (gamma) herpesvirus 4. Human (alpha) herpesvi rus I is most frequently associated with facial lesions and type 2 with genital lesions. Lesions caused by both virus types may recur in some individuals, and the pathogenesis of recurrent infections involves establishment of a latent virus infection in the sensory ganglia following a primary episode. Reactivation of latent virus results in replication of the virus, which then reaches the epithelial surface where further replication occurs in conjunction with a developing lesion. Recurrent lesions also may arise through exogenous reinfection or by transmission of virus from one region of the body to another. Genital type I and 2 infections are a major public health problem in many countries, particularly in the United States where I in 5 adults is now infected ; in 1982 approximately 500 000 new cases were reported. Severe, ulcerating infections have been associated with immunodeficiency states (acquired immunodeficiency syndrome) in male homosexuals, and with immunosuppression associated with organ transplantation or cancer therapy. Reactivation of the virus occurs in some 50% of bone marrow transplant patients and may have life threatening consequences. An understanding of the epidemiology and natural history of herpes- virus infections requires precise identification of the virus strain(s) involved in the primary episodes of infection, and of the reactivated virus present in recurrent infections, and an assessment of the possibi lity of reinfection by exogenous virus. Recently , these questions have been addressed through the application of restriction endonuclease technology. 19 The ON A genomes of human (alpha) herpes viruses l and 2 are linear, duplex molecules of molecular weight approximately 100 X 106 u. The restriction endonuclease cleavage patterns of type l and type 2 DNAs are quite distinct for a range of restriction endonucleases (29). A rapid, simplified technique based on cleavage of radiolabelled viral DNA, which unambiguously distinguishes between the two types, has been reported (30). This procedure allows 50 or more isolates to be screened in 4 or 5 days. Some experimental conclusions derived from the use of endonuclease map- ping techniques Human (alpha) herpesviruses l and 2 isolated from different clinical and geographical sources exhibit different DNA cleavage patterns; they can be differentiated by the loss or gain of a cleavage site for at least one of the restriction endonucleases used. Epidemiologically related viruses, such as repeated isolates from the same individual , are identical, as are viruses passed numerous times in vitro. The cleavage patterns of clonally related type l virus stocks vary, but this variability is confined to small alterations in the mobility of certain DNA fragments. Mutational events causing the loss or gain of a cleavage site are rare. Type l isolates from the left and right ganglia of the same individual , or multiple isolates from different explants of the same ganglion, are indis- tinguishable. This indicates that a single virus strain initially infects each individual, and becomes latent in ganglion cells. The DNA mapping approach was applied to an outbreak of herpes- viral encephalitis in which eight cases appeared within 3 months. Each isolate was type l but the cleavage patterns of each were distinct , indicating that a single strain was not responsible for the outbreak. Investigation of 17 virus isolates from eight patients with type 2 in- fection identified two individuals with non-identical virus isolates. One individual was infected with two distinguishable virus isolates present at a single site, while the other had genetic variants serially isolated from lesions at different sites. This suggests that reinfection with exogenous virus can occur. Human (beta) herpesvirus 5 is present as a large number of genetic variants. A study on 21 strains from 17 members of nine unrelated families showed genetic identity in six out of seven mother-baby pairs, including two congenitally infected infants born several years apart. Genetic hetero- genicity of virus was observed in one woman in whom viruses were isolated a year apart, and in one instance virus obtained from a congenitally infected child differed from an earlier maternal isolate. This suggests that reinfection may occur, although it is less frequent than virus reactivation . 20 Molecular studies of poxviruses and adenoviruses DNA isolated from representative human and animal strains of poxvirus, including rabbitpox, vaccinia, monkeypox, variola, cowpox and ectro- melia viruses , has been mapped following cleavage with various restriction endonucleases (31). The different poxvirus types could be distinguished by this method of analysis, and within the virus types strain differences were clearly apparent. Vaccinia and rabbitpox appear to be the most closely related orthopox- virus types. Viruses from variola major infection could be distinguished from those causing variola minor. Whitepox variants of red cowpox virus have been identified as deletion mutants that lack 12% of red cowpox DNA sequences from one end of the parental genome. Some 38 adenovirus serotypes have been recognized, and adenovirus type 7 is frequently associated with epidemic outbreaks of severe res- piratory illness. Four genome types of adenovirus type 7 have been identi- fied by cleavage of viral DNA with different restriction endonucleases (32) . In addition, restriction endonucleases have been used to identify a genome type of adenovirus 19 associated with keratoconjunctivitis (33). Molecular hybridization of viral genes - studies of influenza viruses These techniques have been applied most extensively to the genetic analy- sis of influenza viruses (34) but are also potentially applicable to studies on other groups of viruses. The studies with influenza viruses serve to il- lustrate the principles involved in the use of the methods, and these are outlined below. Four main approaches have been used. Direct RNA-RNA hydridization This method involves the isolation of 32P-labelled single-stranded RNA from purified influenza virus particles and separation of the RNA into eight virion RNA (vRNA) segments by PAGE. To each vRNA segment is added an excess of unlabelled complementary RNA (cRNA) isolated from the microsomal fraction of infected cells , and the mixture is treated with ribonucleases. If the cRNA is from the homologous virus, the double- stranded RNA resulting from hybridization is resistant to enzyme activity. If cRNA from a genetically heterogeneous virus is used for hybridization, those regions at which low base sequence homology is found are degraded to acid-soluble oligonucleotides. Thus the degree of resistance to ribo- nuclease treatment of the vRNA-cRNA hybrids, formed from the vRNA of one virus and the cRNA of another, provides a relative measure of genetic relatedness of the viruses. The direct hybridization technique can be rendered very sensitive by heating the hybrid molecules close to the 21 melting point in the presence of I% formaldehyde prior to digestion with RN Ase. This is desirable if the genes differ by only a small number of point mutations (35). The technique has contributed significantly to taxonomic data on influenza viruses and the establishment of a new nomenclature system for influenza A viruses (36). It has also been applied to studies on the origins of the various RN A segments in the new pandemic strains of influenza A virus (7) and in laboratory derived reassortant influenza viruses. The limitations of the method are (a) that a few point mutations will not be recognized, and (b) that in regions of the genome with relatively low base sequence homology the few matching bases form relatively unstable double strands, which are degraded by RN Ase. To this extent the RN Ase protection may underestimate the real base sequence homology. Competitive hybridization For this technique, homologous double-stranded RNA molecules are prepared by annealing 1251-labelled individual influenza vRNA segments to unlabelled cRNA. The labelled, double-stranded RNA is then melted and allowed to reanneal in the presence of graded amounts of an un- labelled vRNA of the influenza virus under test. If the unlabelled vRNA contains an RNA sequence identical or partially homologous to the 1251-labelled vRN A, the unlabelled vRN A will compete for annealing sites on the cRNA and will block the reannealing of the labelled vRNA, which then becomes digestible with RN Ase. If RN Ase-resistant radioactivity is plotted against the amount of competing RN A, plateaux are reached that give a relative measure of genetic relatedness of the various influenza virus genes (37). RNA-DNA hybridization A relatively stable 3H-labelled DNA probe for hybridization can be pre- pared by transcribing individual vRNA segments using reverse transcriptase in the presence of dactinomycin and 3H-dNTP. This single-stranded cDNA can be hybridized with non-labelled vRNA of the homologous or heterol- ogous strains. In the heterologous hybridization single-stranded regions are digested by S1-nuclease. The degrees of protection against RN Ase obtained by this method are somewhat low compared with the direct RNA-RNA hybridization tech- nique. However , an advantage of the RNA-DNA hybridization technique is that, once prepared, a batch of 3H-labelled DNA is relatively stable and enables many test viruses to be studied. 22 Analysis of double-stranded hybrid molecules by PA GE after S 1-nuclease treatment This technique takes advantage of the fact that after infection of cells with influenza virus in the presence of cycloheximide, essentially only viral cRNA is synthesized and thus can be labelled with 3H-uridine. The labelled cRN A can be hybridized to an excess of non-labelled homologous or heterologous vRNA. After treatment with SI-nuclease, the double- stranded molecules are analysed by PAGE. Extensive S1-nuclease incu- bation destroys all single-stranded RNA and leaves only the homologous double-stranded RNA segments intact. More limited SI-nuclease treat- ment leads to a retardation of the migration rate in gels of the heterologous hybrids compared to the homologous molecules. This technique has been used for the analysis of artificially produced and naturally occurring recombinant influenza viruses, and can detect small genetic differences within a subtype of influenza A virus - for example, strains that show minor degrees of antigenic drift (38). The advantage of this method is that it is relatively simple and the 3H-cRNA probe used is stable. A disadvantage is that it does not provide quantitative data. Although sensitive, it may not detect all point muta- tions (38). Molecular biological studies of papilloma viruses Papilloma viruses are characterized by a genome comprising a circular, double-stranded DNA molecule of about 8000 base pairs. These viruses induce benign epithelial or fibroepithelial tumours, warts, papillomas or fibropapillomas of the skin and mucosa in at least a dozen animal species (39,40). In man, papilloma viruses are particularly associated with skin warts of various types, orolaryngeal papillomas, anogenital condylomas, and lesions showing no papillomatous structure, such as condylomas of the uterine cervix and the cutaneous macules that characterize epidermo- dysplasia verruciformis (EV). Lesions associated with papilloma virus usually regress but in rare instances may become malignant. Recently at least five types of bovine and nine types of human papil- loma virus have been recognized on the basis of the degree of nucleotide sequence homology of their genomes. Most of these genomes have been molecularly cloned in £. coli, after insertion in a plasmid vector or in bacteriophage DNA. Information on the relationship of gene structure to the biological properties of papilloma viruses is incomplete. D ata so far available for viruses of human and bovine origin suggest that each ty pe of virus is preferentially associated with a distinct type of lesion . Further- more, it appears that some genotypes of human and bovine virus have a higher oncogenic potential ; for example human papilloma virus type 5 has been identified in carcinomas of patients with EV ( 41) . 23 Currently the basis for classification of papilloma viruses is their degree of nucleotide sequence homology. Virus types do not share more than 50% base homology, as determined in the most stringent hybridization con- ditions, while subtypes share more than 50% homology. Types are labelled by numbers and subtypes by letters (a,b,c, ... ) in the order of their identifi- cation. The existence of type-specific and group-specific antigens has been recognized, but the production of type-specific antibodies is generally difficult owing to a lack of sufficient antigen, which is generally obtained by biopsy. Prototypes of bovine and human viruses yield unique restriction endo- nuclease patterns. In man, in addition to the nine types identified which possess up to 15% DNA sequence homology, at least 20 viruses have been recognized on the basis of completely different endonuclease cleavage patterns. Some correspond to subtypes sharing a substantial homology with one of the prototypes, whilst others may constitute additional types. Moreover, minor variations in the patterns are observed, and these define variants. Currently, typing is performed after selective extraction of viral DNA from single lesions, or from pooled specimens from the same type of lesion on a single patient. Viral DNA is cleaved by enzymes such as Pst I or Hind II , or by a mixture of enzymes such as Hind II + III which generate 3-6 frag- ments . Digestion products are separated by electrophoresis on agarose gels and visualized by staining with ethidium bromide. The size of the frag- ments is determined by their mobility compared with standards such as the Hind III fragments of phage DNA. A limitation of this method is the widely variable virus content of lesions. Moreover, patients may be in- fected by several viruses, as is often observed in butchers ( 42) or in patients with EV, and this results in complex electrophoretic patterns. These prob- lems may be overcome by using the blot hybridization technique, whereby viral DNA fragments are denatured in situ in the gels, transferred to nitrocellulose sheets, hybridized with 32P-labelled probes prepared by nick translation of prototype cloned DNAs, and detected by autoradiography. The degree of nucleotide sequence homology must be determined under the most stringent conditions of hybridization before concluding that any virus isolate is a new type. The characterization of papilloma viruses represents a new field in human and animal virology. These studies are stimulated by the potential role played by these viruses in the development of certain cancers ( 43). Applications of DNA recombinant technology to the epidemiology and diag- nosis of HBV An important application of recombinant DNA techniques has been in studies of the epidemiology and diagnosis of HBV using cloned 24 HBV-specific DNA as a probe. In addition, the new methods of genetic engineering provide hope for the development of a new generation of hepatitis B vaccines. Hepatitis B is a major public health problem of wide international distribution . The number of chronic HBV carriers in the world has been estimated at approximately two hundred million. Several liver diseases are associated with HBV infection, including acute or fulminant hepatitis , chronic hepatitis and cirrhosis. Moreover, epidemiological studies (J) have clearly demonstrated an association between chronic HBV infection and the development of primary hepatocellular carcinoma. HBV DNA as an epidemiological probe Blot hybridization techniques allow the clear detection of HBV-specific sequences integrated in the chromosomal DNA of hepatocytes from cases of primary hepatocellular carcinoma. Tests on livers of patients with alcoholic cirrhosis revealed no detectable HBV. Detection ofHBV DNA is therefore a valuable marker of HBV infection for epidemiological pur- poses, and for studies of the association between chronic HBV infection and the incidence of primary hepatocellular carcinoma. Detection of HBV particles in serum Transmission of HBV infection is related to the presence of infectious HBV particles in the serum. During both the acute and chronic stages of hepatitis a titre as high as 109 physical HBV particles per millilitre can be detected. Molecular hybridization techniques using HBV DNA as a probe permit the direct detection of HBV particles in serum at concentrations as low as 104-105 particles per millilitre. PRODUCTION AND PROPERTIES OF MONOCLONAL ANTIBODIES Rationale for production Antiviral immune sera have been used extensively for virus characteriz- ation and diagnosis since the beginning of medical virology, but for experimental work they have some shortcomings. First , they are inherently heterogeneous; even the simplest antigen will raise a large family of anti- bodies directed against different epitopes of the antigen , and comprising a wide range of molecular species of immunoglobulin. Second, since antigen preparations are rarely pure in the immunological sense, the antibody 25 response may be equally impure and may give rise to unexpected and confusing cross-reactions. Third, antisera are not reproducible; each im- munized animal produces a unique assortment of antibodies , and even sequential samples of serum from the same animal are unlikely to be uniform. A great deal of effort has gone into purifying and standardizing antibodies, but this is technically cumbersome. A solution to these problems was realized by Kohler & Milstein ( 44). Antibody-producing cells are terminally differentiated and normally live for only a short period. However, when fused with myeloma cells (tumours of the immune system), B lymphocytes are effectively immortalized and can be cloned and grown in continuous culture. Fusion of cells and their nuclei is a rare event. Isolating rare hybrids from the majority of vigorously multiplying myeloma cells may be achieved by using for fusion a mutant myeloma line that can use only one of the two pathways through which purines are made. Such cells die in the presence of a folic acid analogue while normal cells, including hybrids that have inherited the gene missing from the myeloma, can survive. Thus, in the selective medium, unfused myeloma cells die but some of the hybrids will survive. To be useful, hybrids need to carry the two chromosomes coding for the two chains of the specific immunoglobulin required. The probability of all these events occurring is about I in I 06 , but that still leaves between I 00 and 200 specific hybrids since a single mouse spleen contains over 108 cells. Fig. 7 illustrates schematically a procedure for producing monoclonal antibody as it was devised originally in 1975, and a modern adaptation involving the use of a cytostat ( 45,46). Current practice is to distribute fused cells in a large number of culture wells, thus partially eliminating the need for the time consuming cloning step. In addition the original fusing agent, a paramyxovirus (Sendai virus), has been replaced by polyethylene glycol. Clones of hybrid cells appear between days 4 and 20, and after a week or two the supernatants can be tested for antibody and the cells preserved in liquid nitrogen. For the production of monoclonal antibodies in large quantities cell suspension cultures can be grown using, for example, a cytostat (Fig. 7), or hybridomas may be inoculated intraperitoneally into mice and ascitic fluids collected. Properties of monoclonal antibodies Cross-reactivity A monoclonal antibody may be regarded as a single antibody species, selected from a large family of potential antibodies reacting with the same antigen. Thus, the monoclonal antibodies may be narrowly specific for the 26 N --.J Fig. 7. Production and uses of monoclonal antibodies Produce 1'1 K lg HAT selection Cloning ✓ n'i. 10 ' Screening for specificity 24-48 independent cultures Mass culture Titres ~ 102 -103 -fold C~ higher than culture supernatants Screening for X~ mouse myeloma lines HAT selection on feeder layer 500-1000 independent cultures ----------------.1 / o o o o'J Screening 0 0 0 0 .. . o o o o for spec1f1c1ty l i Cytostat in serum-free medium ~ 50 TC bottles/day . X63 . FO HGPRT HAT hypoxanthine-guanosine phosphoribosyl transferase. the key enzyme in the rescue pathway of purine synthesis hypoxanthine-aminopterine-thymidine medium: forces the cells to use the rescue pathway TC t issue culture lg ,mmunog lobul in PEG polyethylen e glycol Feeder layer culture of peritoneal macrophages. to supply intermediate metabolites for the arising clones homologous antigen or may be broadly reactive, within a group of related antigens. This broad distribution of reactivities has advantages. In the case of monoclonal antibodies to influenza virus haemagglutinin, broad re- activity within a subtype may be observed or there may be highly specific reaction with the homologous virus strain only ( 47). Affinity The firmness of binding to antigen varies greatly among components of a polyclonal antibody population, with poorly binding antibodies making up the majority. Such a spectrum of affinities has its practical uses, but it may cause considerable confusion. Monoclonal antibodies with defined binding affinity may be selected for some purposes. For many uses , such as affinity chromatography for the purification of antigens, o nly antibodies of high affinity are of value; other uses may require antibodies oflower but known affinity. Jsotypes Monoclonal antibodies of identical specificity and affinity may belong to different isotypes, that is, the conserved regions of their globulin chains are chemically different and associated with differing biological and immun- ological properties. Consequently, antibodies may be selected that are, for example, suitable for complement fixation or agglutination tests or for plaque-inhibition assays, while others may be easily purified or made into univalent (Fab) fragments having the antigenic specificity. While monoclonal antibody techniques hold great promise for im- proving diagnostic reagents , for purifying and concentrating macromole- cules and for carrying special reagents to particular markers, the scope for non-critical and improper use is equally large. Since monoclonal anti- bodies are random samples of a large and heterogeneous population, they should not be applied without first evaluating their cross-reactivity and affinity, and defining as closely as possible the epitope0 or epitopes they recognize. a An antigen is any macromolecule, usually of molecular weight above 10000 u, with which antibodies can react. The actual contact area during reaction is called epitope on the antigen and paratope on the antibody. Antigens, even if they are single molecules, may carry several different epitopes. Anti- bodies have also several different paratopes, but these are always identical on any one molecule and in any one population of a monoclonal antibody. Antibody molecules themselves are antigenic; their epitopes are called idiotopes. ldio- topes are of two kinds, either paratope-related (i.e. inaccessible when the paratope is com- bined with the corresponding epitope) or paratope-independent. 28 APPLICATIONS OF MONOCLONAL ANTIBODY IN VIRAL DIAGNOSIS AND EPIDEMIOLOGICAL STUDIES Conventional polyclonal antisera have been used for the last five decades in the laboratory diagnosis of a wide range of viral diseases . Serological tests based on these conventional reagents continue to be valuable in diagnosis and epidemiological studies. The availability of monoclonal antibodies, however, provides the possibility of improving the degree of reliability , standardization, specificity and precision of the tests . In addi- tion , the use of hybridoma techniques may enable specific antibodies to be made against the antigens of viruses that are uncultivable or grow poorly, such that immune sera cannot easily be made by conventional methods. Use in epidemiological and diagnostic studies Monoclonal antibodies can be used to search for unambiguous evidence of antigenic variation among closely related viruses. For example, antigenic variants of measles virus ( 48) have been detected by immunoselection techniques with monoclonal antibodies. Cloned measles virus was grown in the presence of an excess of neutralizing monoclonal antibody and non-neutralized variants were selected. The frequency at which antigenic variants occurred was approximately 10- 4 _ Variants differed from parental virus in the antigenic structure of their haemagglutinins and were gen- etically stable. It seems almost certain that antigenic variation plays no role in the epidemiology of measles; nevertheless , precise antigenic characterization of strains by monoclonal antibodies may be of va lue in distinguishing vaccine strains from "wild" measles virus . Furthermore, it should be possible to determine whether antigenic variants have any role in the persistence of measles-like viruses as observed in subacute sclerosing panencephalitis. Antigenic drift among influenza B viruses ( 49) has been studied using a collection of monoclonal antibodies to analyse the haemagglutinin anti- gens. The antigenic determinants could be subdivided into three partially overlapping groups, and antigenic drift occurred in each of the epitopes defined by the monoclonal antibodies. The study clearly showed that several antigenically distinct influenza B variants co-circulated in a com- munity at one time. Furthermore, similar studies have recently shown that viruses isolated during circumscribed outbreaks of influenza A (H3N2) and of influenza Bin residential schools were antigenically heterogeneous in respect of their haemagglutinin antigens. A study of antigenic variation in rabies virus has been reported by Wiktor & Koprowski (50). CVS-I I , a standard challenge rabies virus, was grown in the presence of a concentration of monoclonal antibody able to 29 neutralize approximately 105 infectious units. Antigenic variants in single glycoprotein epitopes were present in the cloned rabies seed at a frequency of 10- 4•3 to 10- 5 and the variants probably represent single point glyco- protein mutants of the CVS virus. Moreover, the study showed that viruses isolated from seven fatal cases of human rabies represented antigenic variants. The use of monoclonal antibodies reacting with rabies virus has per- mitted the identification of numerous antigenic variants of rabies virus isolated from various animal and human sources in different parts of the world. Plaque-purified variants of rabies virus resistant to individual mono- clonal antibodies do not immunize against challenge with parental virus, even though antibodies from mice immunized with the variant virus neu- tralize equally well the variant and the parental virus. There is no indication at present that vaccination failures in man are due to antigenic differences between the vaccine and the field virus, but it might be desirable for licensing authorities in countries where rabies variants are prevalent to introduce experimental potency tests in which vac- cinated mice will be challenged with local field strains (street viruses) and not only by a fixed Pasteur virus derived strain. Also, rabies virus variants isolated as being resistant to an individual monoclonal antibody have been shown to be apathogenic on intracerebral injection into adult mice. Poly- peptide analysis of the glycoprotein of these attenuated variants showed substitution of arginine by isoleucine or glutamine. When the attenuated variant is manipulated so that it reverts to a more virulent form, the isoleucine or glutamine residues are substituted by arginine. Investigations of virus structure Antigenic variants of influenza virus selected in vitro using monoclonal antibodies are valuable in detailed investigations of the molecular basis for antigenic variation in the most important antigen of the virus, the haemag- glutinin. The complete amino acid sequences of haemagglutinins for several influenza viruses have been determined, and the tertiary structure of the haemagglutinin protein has also been determined by X-ray crystal- lography (51). Studies with mutants selected in the presence of monoclonal antibodies have revealed the existence offi.ve independent antigenic sites. Knowledge of the precise three-dimensional structure of the haemagglutinin molecule has enabled the positions of the antigenic sites to be exactly located in the globular distal region of the molecule (52). The amino acids specifying the antigenic sites have been found to be located non-contiguously in the amino acid sequence of the molecule. The study of changes in the indi- vidual antigenic sites of the haemagglutinin are likely to contribute to an understanding of antigenic variation and epidemiology of the virus . 30 Antigenic changes at individual sites on the molecule may differ in their epidemiological significance. It is not clear at presen t if amino acid changes occurred in the actual antigenic site recognized by the monoclonal anti- body, or if they were located elsewhere and induced conformational changes that altered the antigenic site. Monoclonal antibodies against the major structural and nonstructural components of human adenovirus have been developed (53) . At least five different epitopes were found to be associated with the hexon antigen, and preliminary data suggest that complex interrelationships between the hexons of different serotypes may exist. Further subdivision of the various adenovirus types can probably now be achieved with these antibodies. The anti-hexon monoclonal antibodies could prove to be valuable tools in diagnostic virology, enab ling rapid typing of virus and perhaps finer discrimination of virus types in association with particular disease syndromes. Monoclonal antibodies to poliovirus type 3 secreted by mouse and rat hybridoma cells have been used to study the antigenic characteristics of infectious poliovirus (D antigen) and noninfectious, empty particles (C antigen) (54). Antigen blocking tests were used to separate a collection of 51 monoclonal antibodies into three groups exhibiting exclusive re- activity with D antigen, exclusive reactivity with C antigen, or common reactivity with both antigens. Thus, in addition to possessing unique determinants, the D and C antigens of poliovirus type 3 appear to share one or more common antigenic determinants . All of the monoclonal antibodies that bound to the determinant(s) common to D and C antigen neutralized a wide range of type 3 poliovirus strains. However , only I 2 of the 19 D antigen-specific antibodies had virus neutralizing properties, and some of these neutralized or reacted in blocking tests with only a narrow range of type 3 strains. Some antibodies were diagnostically useful in that they neutralized only Sabin vaccine strains and not wild strains of poliovirus. The antibodies were a lso tested for their ability to react with isolated poliovirus capsid proteins (VPI, VP2, VP3 and VP4) separated on poly- acrylamide gels in "immunoblot" techniques. Six of the 51 antibodies tested bound to the denatured proteins, five with VP I and one with VP3 . All six antibodies that reacted in this test were non-neutralizing and C antigen-specific. None of the 20 antibodies of D + C or D specificity that possessed virus neutralizing properties reacted in the immunoblot test. Nevertheless, evidence is available that the majority of these anti- bodies are directed against VP I . These findings suggest that the antigenic determinants of the poliovirus capsid involved in neutralization are specified largely by the three-dimensional tertiary or quaternary con- formation of the capsid proteins, rather than by primary amino acid sequence alone. 31 Some limitations of monoclonal antibodies Monoclonal antibodies cannot always form the complex lattices necessary for precipitation in immunodiffusion tests, immunoelectrophoresis or routine agar diffusion, and some classes or subclasses of monoclonal antibodies do not fix complement. In addition, monocolonal antibodies may often be too specific for routine virological diagnosis, recognizing only a narrowly defined virus variant when an antibody reacting widely with all viruses of the group is desirable. This problem may be overcome by using for diagnostic purposes a mixture of several monoclonal antibodies reacting with different virus epitopes. A biased repertoire of antibody specificities is another potential complication in the interpretation of results on the antigenic analysis of viruses by means of hybridomas. For example, the Balb/c mice used widely for work with monoclonal antibodies may possess a limited antibody repertoire for the particular virus antigen. Finally, a single monoclonal antibody prepared, for example, against a viral antigen may unexpectedly cross-react with a variety of tissue antigens. Monoclonal antibodies and molecular mimicry Monoclonal antibodies produced after immunization of mice with large T antigen of simian virus 40 (SV40) have been found cross-reactive with normal constituents present in many mammalian cells. This antibody immunoprecipitates an antigen of molecular weight 168 000 u, which is bound to nuclear membrane and not expressed on the surface of the cells. Subsequently, several monoclonal antibodies obtained after immun- ization of mice with viral agents such as human (alpha) herpes virus type I, rabies virus and measles virus have been found to cross-react with normal cell constituents, being at the same time quite specific for a given viral component. For instance, an antibody that immunoprecipitates an (al- pha) herpesvirus phosphorylated protein component of molecular weight 146 000 u reacts with 52 000 u vimentin protein of normal mammalian cells. The vimentin protein is also immunoprecipitated by monoclonal antibodies, which precipitate a 70 000 u phosphorylated protein of measles virus. The herpes and measles monoclonal antibodies do not cross-react, however (55). A striking example of virus-induced antibodies reacting with normal components of the organism has been presented by Notkins et al. (personal communication) in the case of reovirus . Spleens of mice infected with reovirus were used as a source of monoclonal antibody after fusion with myeloma cells. Antibodies secreted by these hybridomas reacted with insulin and growth hormone, and with antigens released by cells of normal pancreas, normal hypophysis and other organs of the body. The biological significance of these observations is as yet unclear. 32 APPLICATIONS OF MOLECULAR BIOLOGY AND CELL HYBRIDIZATION TOWARDS THE EVENTUAL CONTROL OF VIRAL INFECTION AND DISEASE A wide variety of new biological medicinal products is capable of being manufactured as a consequence of recent developments in biotechnology. Two techniques of particular significance are: the manipulation of defined coding sequences of DNA and their controlled expression in eukaryotic and prokaryotic host cells; and cell fusion technology leading to the development ofhybridoma cell lines that secrete monoclonal antibodies of potential value as thera- peutic and diagnostic reagents, and also in preparative and analyti- cal procedures for biologicals. New approaches are also being made to the manufacture of biological products that exploit a range of recently developed techniques, including purification methods such as affinity chromatography using monoclonal antibodies. These methods have stimulated the exploitation of aneuploid cell lines of primate or human origin as substrates in the manufacture of inactivated virus vaccines and other products such as interferon. Genetic engineering Fundamental studies in molecular genetics, together with a precise under- standing of nucleic acid chemistry, has allowed the genes coding for biologi- cally active substances to be identified, analysed in fine detail , transferred within and between organisms , and expressed under controlled conditions so as to obtain efficient synthesis of the product for which they code . Gene manipulation , based on DNA recombinant techniques, may be defined as the formation of new combinations of heritable material by the insertion of nucleic acid molecules into any vector system - commonly used vectors are bacterial plasmids or viruses - so as to allow thei r incorporation into a host organism in which they do not naturally occur, but in which they are capable of continued propagation . Genes coding for a specific product can be isolated and propagated by the insertion of naturally occurring or synthetic genetic source material into a suitable vector, followed by the selection of individual clones of the vector that carry the required gene - the process of gene cloning. Much, but not all, gene cloning work has been carried out in plasmids of E. coli. Key steps in the process involve the insertion of the gene into the vector with the aid of highly specific restriction endonuclease enzymes to cleave the vector DNA at predetermined sites, and ligases to recombine the gene insert into the vector. 33 A gene is characterized by a specific nucleotide sequence in one strand of the double-stranded DNA molecule. When the strands are separated each forms a template for the synthesis of a complementary copy, thus providing a mechanism for the faithful reproduction of genes with con- servation of the linear sequence of the four mononucleotides. The process of decoding this information and the synthesis of the gene product occurs in two phases: first, transcription of the DNA coding strand in the form of a messenger RNA (mRNA) and second, translation of the information carried by the mRNA molecule into an amino acid sequence. The factors affecting the expression of foreign genes introduced into a bacterium by a plasmid are complex; indeed, the efficient and controlled expression of stable, cloned DNA sequences is an important field of current research . Potential applications of DNA recombinant technology The potential application of DNA recombinant technology for the present- ation and control of viral disease covers a very wide field . An important application of the new biotechnologies is the development of vaccines against agents that so far have not been cultivated or which grow poorly . Examples of this are the potential development of vaccines against hepa- titis B virus and rotavirus . The latter agent is a major cause of severe diarrhoeal disease in infancy, but grows too poorly to enable preparation of conventional vaccines. Production of vaccines against dangerous path- ogens, such as African haemorrhagic fever viruses, can potentially be achieved without the risks arising from the use of the infectious virus . Influenza viruses present special problems for immunoprophylaxis because of their high degree of antigenic variability. Modern technologies can be used analytically to study the problem of variation and to seek for alternative approaches to vaccine design or chemotherapy. A worthwhile goal in this respect might be to seek antigenic determinants of the influenza virus, which might be harnessed as vaccines capable of protecting against a wide range of virus variants, i.e. of inducing cross-subtype protection. The methods are also applicable to the production of virus-specific proteins and DNA molecules of value in epidemiological studies and diagnosis. Examples of recombinant DNA techniques relevant to the design and preparation of future vaccines _are given below. Influenza viruses: DNA cloning and expression Mainly as a result of nucleotide sequence analyses of bacterially cloned influenza virus cDNA molecules , the nucleotide sequences of all eight vRNAs of at least one virus are now known. Together with other nucleo- tide sequence analyses involving primer extension and direct RNA se- quencing, these studies have been of value in defining details of genome 34 transcription and replication and in determining the ongms of small defective RN As often produced during infection. Their main value to date, however, has been in allowing deductions of the sequences of virus pro- teins, particularly the haemagglutinins and neuraminidases of antigen- ically distinct viruses. Cloning procedures The various procedures used for the production and cloning of double- stranded DNAs are standard and may be summarized as follows. Prep- aration of cDN As is by reverse transcription, either ofvRNAs polyadenyl- ated at their 3' termini or ofvRNAs in the presence of synthetic oligodeoxy- nucleotide primers complementary to the conserved sequences at their 3' termini. Second-strand synthesis is catalysed either by DNA polymerase I (Kienow fragment) or reverse transcriptase, utilizing either the ability of cDN A to form 3' -terminal hairpin loops or synthetic primers complemen- tary to the conserved 3' termini of cDNA. The double-stranded DNAs are either tailed using terminal transferase and inserted into appropriately complementary tailed linearized plasmids, or complete double-stranded DNAs are blunt-end ligated into linearized plasmids. Recombinant DNA molecules so obtained are used for transformation of£. coli cells such as 1776 or HB IO 1. Transformed cells are selected for the antibiotic resistance conferred by the recombinant plasmid, and the particular virus sequences in the cloned recombinants are usually determined by in situ hybridization. In addition to these procedures, restriction fragments derived from double- stranded DNAs synthesized in vitro have been cloned into the bacterio- phage M 13 for extensive sequence analyses using the dideoxy technique. Expression Reports of the expression of cloned influenza genes are at present restricted to a number ofhaemagglutinin genes from different viruses that have been studied in prokaryotic systems (56,57). The prokaryotic vectors used in- clude plasmids containing the transcriptional control and ribosome bind- ing site regions of the £. coli tryptophan and lac operons. Reports of expression in eukaryotic cells (58) have involved inserting haemagglutinin DNA into the late region of SV40 DNA, and utilizing early detection mutants of SV40 as helper viruses in the generation of appropriate virus stocks for the infection of monkey cells. In addition, recombinant DNAs with haemagglutinin DNA inserts in the early region of SV40 have been used for expression of haemagglutinin in SV40 trans- formed cells. The expression ofhaemagglutinin genes in£. coli was rather low, much lower than theoretically attainable. The product was non-glycosylated and 35 little is known about its stability and immunogenicity. The production of haemagglutinin achieved in simian cells with the use of SV40-derived vector was much more efficient and the product was glycosylated and appeared normal in all respects. Hepatitis B vaccines: perspectives of new methods of vaccine preparation Cloning and sequencing of the complete genome of HBV has provided important information on the genetic organization of the virus (59). The gene coding for the viral surface antigen (HBsAg) has been identified and characterized. This gene has been inserted into suitable vectors and can be expressed in a variety of host cell systems (£. coli, yeast, mammalian cell lines). It is well established that the 22-nm HBsAg particles present in the serum of chronic HBV carriers induce protective antibodies when injected in man. In the absence ofa cell culture system able to propagate the virus in vitro, 22-nm HBsAg particles for use in vaccine are purified from human serum. An alternative to this first-generation method of vaccine preparation is to produce HBsAg in a suitable host cell transfected with cloned DNA of HBV containing the gene coding for the HBsAg protein. For example, a transfected cell line has been shown to synthesize and excrete 22-nm HBsAg particles in the cell culture supernatant. These particles contain both the group and the subtype HBV determinants. Injected into mice they induced antibodies against HBsAg of high binding affinity to human HBsAg. Expression of the same gene in E. coli leads to the biosynthesis of the major polypeptide of the viral envelope. Injected into animals this polypeptide induces anti-HBs, but its immunogenicity is reported to be low compared to that of complete 22-nm particles. A further approach consists in chemically synthesizing oligopeptides containing HBsAg determinants . Such peptides have been known to in- duce anti-HBs when injected into animals, but their protective value has yet to be determined. Synthetic peptides as viral antigens The development of small, synthetic peptides mimicking the antigenic structures of certain viruses has already provided encouraging results, for example with HBV, rabies virus and foot-and-mouth disease virus (a picornavirus with many of the structural features of human poliovirus). For this last agent, small peptides representing amino acid sequences of the major capsid protein of the virus VPl efficiently stimulate virus neutral- izing antibody. This field of work is worthy of careful exploration and development to determine the value of the peptides as vaccines. 36 Following cloning of rabies virus and description of its genetic code, a number of polypeptides representing amino acid sequences of viral struc- tural proteins have been synthesized. Immunization of animals with these polypeptides has resulted in the production of serum antibodies that bind to native rabies virus antigens, including the surface. So far, however, attempts to induce antipeptide antibodies capable of neutralizing rabies virus have been unsuccessful. Developments in the immunology of infectious agents The past decade has seen a rapid growth in knowledge of the various components of the immune response to microbial antigens in man, and this was reviewed in 1982 by a WHO Scientific Group (2). Both antibody and cell-mediated responses play a role in immunity and recovery from micro- bial infection. There is now much new information on the cells involved in the immune response, which depends on a complex network of inter- dependent components. A full understanding of the functioning of the system, and consequently of the mechanisms for artificial induction of immunity, depends on analysing interactions among its components as well as the function of each component. It is well recognized that the parameters of the immune response and protection differ widely for dif- ferent types of microorganism. An important aspect of immunology is its relevance to the understanding of immunopathological responses to microbial infection or vaccination . Detailed analysis of the responses to individual infectious agents is required to provide a scientific basis for immunoprophylaxis and immunotherapy. This field of study is thus highly relevant to the harnessing of antigens developed by the new biotechnol- ogies as effective vaccines . Synthetic antigens, usually small peptides, may not by themselves stimulate a satisfactory immune response. The reason, apart from con- formational instability, is that they may be incapable of stimulating helper T cells, i.e. they behave like haptens. This shortcoming might be corrected by attaching them to larger carrier molecules known to be effective in inducing the required macrophage T-cell interaction. If synthetic antigens are made in bacteria, a possible solution is to construct a plasmid in which the gene sequence coding for the required antigen is inserted directly after the ON A coding for , say ,,8-galactosidase, which may act as a carrier protein . If the antigenic peptide is produced separately, it can be covalently coupled to knoW'h good carriers. For experimental purposes, su itable carriers might be tetanus toxoids, immuno- globulins or haemocyanin. In either case the immune response can be greatly enhanced by priming the recipient with the carrier only, 2-3 weeks before injecting the carrier-antigen complex, so as to induce a primary response. Much development work is required to identify carriers and 37 adjuvants that might safely be used in man to harness the immunogenicity of small synthetic antigens. Monoclonal antibodies are now available for use in the evaluation of different functional subsets of T and B lymphocytes. In studies using monoclonal antibodies as analytical tools, perturbation of the balance between certain T-cell subsets has been shown to occur during certain viral infections, such as acute cytomegalovirus infection and hepatitis B. Such studies may be of value in monitoring the cellular basis of immune response and immunopathology in viral diseases. SUMMARY OF SOME MAJOR PRACTICAL APPLICATIONS OF THE NEW BIOTECHNOLOGIES RELEVANT TO THE PREVENTION AND CONTROL OF VIRAL DISEASE Recombinant DNA technology The production of nucleic acids of defined microbial specificity for use as diagnostic reagents and tools for epidemiological research. The definite modification of microbial genomes ( e.g. by site-specific mutagenesis) for the production of stable, safe, attenuated mutants for development as live vaccines. The prediction of amino acid sequences of antigenic epitopes of microbes leading to the development of synthetic peptide vaccines. The production, by controlled gene expression in suitable vectors, of defined microbial proteins or oligopeptides for use as vaccines, in diagnosis, or in epidemiological research. - The production of "synthetic" antimicrobial antibodies. This ap- proach is feasible but not yet explored. Hybridoma technology 38 The preparation of monoclonal antibodies for the analysis of mi- crobial antigens, so that antigenic structures relevant to immuno- genesis may be identified and harnessed for vaccine development purposes. Large-scale production of defined antimicrobial monoclonal im- munoglobulins for use in passive immunoprophylaxis or therapy or as diagnostic reagents. Production of immunoglobulins linked to antimicrobial or anti- cellular toxins for use in therapy (targeted drugs). Preparation of clones of immunocompetent cells (B and T cells) that have a role in immunological research and potentially in disease control. RECOMMEND A TIO NS I. The use of established molecular biological methods appropriate to studies on the epidemiology, diagnosis and control of viral infections is strongly encouraged. Collaboration among research laboratories in- volved in molecular virology and those concerned with epidemiological studies will enhance developments in this field and should be fostered by WHO. 2. Analysis by gel electrophoresis and other appropriate methods of the nucleic acid components of noncultivable or poorly cultivable viruses with segmented RNA genomes is a simple, practical and inexpensive method. It is recommended for the diagnosis, epidemiological study and partial char- acterization of rotaviruses and influenza viruses. 3. Oligonucleotide mapping is a suitable method for the differentiation of viruses with closely related RNA genomes. It is a powerful tool for the differentiation of wild and vaccine-derived poliovirus strains, for the study of influenza subtype variants, and for the study of rota virus genomes. The technique is demanding in the use of radioisotopes and requires consider- able technical expertise. 4. Restriction endonuclease analysis of viral DNA provides a va luable method for appropriately equipped laboratories. It has been of va lue in epidemiological studies of human (alpha) herpesviruses I and 2 and adeno- viruses, in the study of mother-child transmission involving congenital infections by human (beta) herpesvirus 5, and in the characterization of human papilloma viruses. 5. Techniques based on molecular hybridization are highly discriminating for closely related viruses and may enable the analysis of a large number of isolates. They have been shown to be of value for the epidemiological study and characterization of influenza, papilloma and hepatitis B viruses, rota- viruses and retroviruses. These techniques are indispensable in studying the 39 role of viruses in human cancer. Like oligonucleotide mapping, these methods require the use of radioisotopes and are technically demanding. 6. Nucleotide sequencing provides definitive analysis of the genomes of both DNA and RNA viruses, together with information on the gene products. However, the methods are complex and are available in only a few specialized laboratories. Recombinant DNA technology is being in- creasingly used for the cloning and sequencing of viral genomes. 7. The identification and cloning of specific viral DNA sequences, and their controlled expression in eukaryotic or prokaryotic host cells, are potent tools for the preparation of diagnostic reagents through the pro- duction of specific nucleic acid probes or amino acid sequences. It is also of great potential importance for the development of novel or improved vaccines and their production. In this respect, the detailed characterization of bacterial plasmids of epidemiological and medical importance, such as antibiotic resistance factors for bacteria and those used as gene vectors in experimental work, is very important. 8. Cell fusion (hybridoma) technology and the production of mono- clonal antibodies against defined viral antigens is a potentially important tool for diagnosing viral infection for epidemiological studies, and in purifying viral proteins for the development of new vaccines . Monoclonal antibodies are particularly valuable for use in studies on antigenic struc- ture and variation in viruses, and on mechanisms of immunity to virus infection. 9. Recombinant DNA and monoclonal antibody techniques have great potential, and should be vigorously exploited in attempts to improve the diagnosis and control of virus disease. These approaches are of particular value for viruses that cannot readily be cultivated, and for use with highly pathogenic viruses requiring special facilities for their handling. 10. The development of new diagnostic reagents and vaccines based on recombinant ON A technology should go in parallel with quality and safety control techniques. New sets of analytical tests should be developed to assay for the purity of each product , its identity with natural proteins, and its consistency. The criteria of identity should rely on both immunological and physicochemical analyses. Although vaccines for human use will not be available in the near future, the methodology of examining such products in selected groups of individuals and in the field should be carefully elab- orated well in advance. Guidelines for these activities should be developed. 11 . Future developments in the application of molecular methods to the diagnosis, epidemiology and control of viral disease will benefit from a 40 combined immunological, microbiological, molecular biological and gen- etic approach. WHO has a role in encouraging and coordinating collab- oration among appropriate laboratories. 12. The wider use of the new biotechnological techniques to study viruses or reagents, based on their use for the epidemiology and control of viral disease, is strongly encouraged. 13. Research work on the targeting of drugs should be encouraged. Speci- fic monoclonal antibodies directed against critical viral antigens and Jinked to potent biological reagents has important potential applications in the control of viral infection. 14. Although the specific considerations and recommendations encom- passed in this report deal with the field of virology, it is emphasized that most of the considerations have a general relevance to the diagnosis and control of other infectious diseases, including those caused by chlamydia, rickettsiae, bacteria and parasites. REFERENCES l. Deinhardt, F. & Gust, I.D. Bulletin of the World Health Organization, 60: 661-691 ( 1982). 2. WHO Technical Report Series, No. 693, 1983 (Viral vaccines and antiviral drugs). 3. WHO Technical Report Series, No. 676, 1982 (Interferon therapy). 4. The role of genetic and molecular characterization of viruses in relation to influenza surveillance and epidemiology: a WHO Memorandum. Bulletin of the World Health Organization, 59: 875-879 (1981). 5. WHO Scientific Working Group. Bulletin of the World Health Organ- ization, 58: 183-198 (1980). 6. Holland, J. et al. Science, 215: 1577-1585 (1982). 7. Webster, R.G. et al. Nature, 296: 115-121 (1982). 8. Szekely, M. & Sanger, F. Journal of molecular biology, 43: 607-617 (1969). 9. Pedersen, F.S. & Haseltine, W.A. Methods in enzymology, 65: 680- 687 (1980). 10. Flores, J. et al. Lancet, 1: 555-559 (1983). 11. Schroeder, B.A. et al. Journal of virology, 43: 379-385 (1982). 12. Peattie, D.A. Proceedings of the National Academy of Sciences of the USA, 76: 1760-1764 ( 1979). 13. Donis-Keller, H. et al. Nucleic acid research, 4: 2527-2538 (1977). 41 14. Sanger, F. et al. Proceedings oft he National Academy of Sciences oft he USA, 74: 5463-5467 (1977). 15. Caton, A.J. et al. Cell, 31: 417-427 (1982). 16. Maxam, A.M. & Gilbert, W. Proceedings of the National Academy of Sciences of the USA. 74: 560-564 ( 1977). 17. Air, G.M. Proceedings of the National Academy of Sciences of the USA, 78: 7639-7643 (1981). 18. Minor, P.D. Journal of virology, 34: 73-76 (1980). 19. Lourenco, M.H. et al. Anna/es de viro/ogie, 132E: 161-173 (1981). 20. Nicolas, J.-C. et al. Anna/es de viro/ogie, 134E: 135-139 (1983). 21 . Rodger, S.M. et al. Journal of clinical microbiology, 13: 272-278 (1981). 22. Follett, E.A.C. & Desselberger, U. Journal of medical virology, ll: 39-52 ( 1983). 23. Nakajima, K. et al. Nature, 274: 334-339 (1978). 24. Young, J.F. & Palese, P. Proceedings of the National Academy of Sciences of the USA , 76: 6547-655 l (l 979). 25. Minor, P.D. Journal of general virology, 59: 307-317 (1982). 26. Nottay, B.K. et al. Virology, 108: 405-423 (1981). 27. Minor, P.D. et al. Nature, 301: 674-679 (1983). 28. Follett, E.A.C. & Desselberger, U. In: Bishop, D.H. & Com pans, R.W., ed. The replication of negative strand viruses. Amsterdam, Elsevier, 1981 (Developments in Cell Biology Series, Vol. 7). 29. Wilkie, N.M. et al. Journal of antimicrobial chemotherapy, 3(Suppl. A): 47-62 (1977). 30. Lonsdale, D.M. Lancet, I: 849-851 ( 1979). 31. Esposito, J.J. et al. Virology, 89: 53-66 (1978). 32. Wadell, G. et al. Infection and immunity, 34: 368-372 (1981). 33. Wadell, G. & de Jong, J.C. Infection and immunity, 27: 292-296 (1980). 34. Scholtissek, C. In: Palese, P. & Kingsbury, D.W., ed. Genetics of influenza viruses. Vienna & New York, Springer, 1983. 35. Scholtissek, C. et al. Virology, 74: 332-344 (1976). 36. A revision of the system of nomenclature for influenza viruses: a WHO Memorandum. Bulletin of the World Health Organization, 58: 585- 591 (1980). 37. Bean, W.J. et al. Nature, 284: 638-640 (1980). 38. Hay, A.J. et al. Developments in biological standardization, 39: 15- 24 (1977). 39. zur Hausen, H. & Gissman, L. In: Klein, G ., ed. Viral oncology. New York, Raven Press, 1980. 40. Gissman, L. et al. International journal of cancer, 29: 143-146 (1982). 41. Kremsdorf, D. et al. Journal of virology, 43: 436-447 (1982). 42. Orth, G. et al. Journal of Danish dermatology, 76: 97-122 (1981). 43. Orth, G. et al. Cold Spring Harbor conferences on cell proliferation, 7: 259-282 (1980). 42 44. Kohler, G. & Milstein, C. European journal of immunology, 6: 511- 519 (1976). 45. Fazekas de St Groth, S. & Scheidegger, D. Journal of immunological methods, 35: 1-21 (1980). 46. Fazekas de St Groth, S. Journal of immunological methods, 51: 121- 136 (1983). 47. Gerhard, W. & Webster, R.G. Journal of experimental medicine, 148: 383-392 ( 1978). 48. Birrer, M.J. et al. Nature, 293: 67-69 (1981). 49. Webster, R.G. & Berton, M.T. Journal of general virology, 54: 243- 251 (1981). 50. Wiktor, T.J. & Koprowski, H. Journal of experimental medicine, 152: 99-112 (1980). 51. Wilson, I.A. et al. Nature, 289: 366-373 (1981 ). 52. Wiley, D.C. et al. Nature, 289: 373-378 (1981). 53. Russell, W.C. et al. Journal of general virology, 56: 393-408 (1981). 54. Ferguson, M. et al. Journal of general virology, 65: 197-201 (1984). 55. Fujinami, R.S. et al. Proceedings oft he National Academy of Sciences of the USA, 80: 2346-2350 (1983). 56. Emtage, J.S. et al. Nature, 283: 171-174 (1980). 57. Davis, A.R. et al. Proceedings oft he National Academy of Sciences oft he USA, 78: 5376-5380 ( 1981 ). 58. Gething, M.-J. & Sambrook, J. Nature, 293: 620-625 (1981). 59. Tiollais, P. et al. Science, 213: 406-411 (1981). 43 Annex 1 LIST OF PARTICIPANTS Temporary advisers 44 Dr F. Bricout, Laboratory of Virology, Hopital Trousseau, Paris, France Dr J.B. Clements, Institute of Virology, University of Glasgow, United Kingdom Dr E. Domingo, Centre for Molecular Biology, Autonomous Univer- sity of Madrid, Spain Professor S. Fazekas de St Groth, Basie Institute for Immunology, Switzerland Dr H . Koprowski, Wistar Institute of Anatomy and Biology, Phila- delphia, USA Professor G. Lebek, Institute of Hygiene and Medical Microbiology, University of Bern, Switzerland Dr R. Najera, Director, National Centre for Microbiology, Virology and Immunology applied to Health, Madrid, Spain Dr J.-C. Nicolas, Laboratory of Virology, Hopital Trousseau, Paris, France Dr G.C.J. Orth, Head, Papilloma Virus Unit, Pasteur Institute, Paris, France Dr G .C. Schild, National Institute for Biological Standards and Con- trol, London, United Kingdom (Rapporteur) Professor C. Scholtissek, Institute of Virology, Justus-Liebig Univer- sity, Giessen, Federal Republic of Germany Professor B.F. Semenov, Director, Mechnikov Research Institute for Vaccines and Sera, Moscow, USSR Dr J .J . Skehel, Co-Director, World Influenza Centre, Division of Virology, National Institute of Medical Research, London, United Kingdom Professor P. Tiollais, Faculty of Medicine Lariboisiere St Louis, Pas- teur Institute, Paris , France Dr V. Vonka, Chief, Department of Experimental Virology, Institute of Sera and Vaccines, Prague, Czechoslovakia Professor P. Wildy, Department of Pathology, University of Cam- bridge, United Kingdom (Chairman) World Health Organization Headquarters Dr F. Assaad, Director, Division of Communicable Diseases Dr T.A. Bektimirov, Medical Officer, Virus Disease Unit Regional Office for Europe Dr B. Velimirovic, Regional Officer for Communicable Diseases (Secretary) 45 WHO publications may be obtained, direct or through booksellers, from: ALGERIA ARGHITINA AUSTRALIA AIJ!>"TRI ,\ BANGLADESH BELGI UM BH IJTA N BOTSWANA BRAZIL BURMA CANADA C HINA CYPRUS CZECHO- SLOVAKIA DEMOCRATIC PEOPLE'S REPUBLIC OF KOREA DENMARK ECUADOR EGYPT FIJI Fl LAND FRANCE GABON GERMAN DEMOCRATIC REPUBLIC GERMANY. FEDERAL REPUBLIC OF GHANA GREECE HAITI HONG KONG HUNGARY ICELAND INDIA INDONESIA IRAN (ISLAMIC REPUBLIC OF) IRAQ IRELAND ISRAEL ITALY JAPAN JORDAN. THE HASHEMITE KINGDOM OF KUWAIT LAO PEOPLE'S DEMOCRATIC REPUBLIC LEBANON LUXEMBOURG '.\,IALAWI S<x:1c1c :Sa11onalc d 'Edi11on ct de Diffusion. 3 bd Zirout Youcef. ALGIEltS Carlo, Hirsch SRL. Florida 165. Galcrias Giicmes. Escritorio 453/465. BUENOS AIRES Hun1cr Publica11ons. 58A Gipps Street. COUJNGWOOO, VIC 3066 - Australian Government Publishing Service /Mu,/ urdn salts). P.O. Box 84. CANBERRA A.C.T. 2600: or o,.·~r tht coumu from . Australian Gov• crnmcnt Publi,hing Service Bookshops at: 70 Alinga Street. CANIERRA CITY A.C.T 2600 ; 294 Adelaide Sareet , BRISBANE. Queensland 4000 : 347 Swanston Street, MEUOURNE, VIC 3000 : 309 Piu Street. SYDNEY, NS W 2000 : Mt Newman House. 200 St . George's Terrace. l'EUH, WA 6000 : Industry House, 12 Pirie Street. ADELAIDE. SA 5000 : 156-162 Macquarie Street, HOBART, TAS 7000 - R. Hill & Son Ltd., 608 St.. K1lda Road. MEUIOLRNE, VIC 3004; Lawson House, 10-12 Clark Street, CRow·s NEST, NSW 2065 Gerold & Co .. Graben 31 . 1011 VIENNA I The WHO Progr•mmc Coordinator. G.P.0 . Box 250. DHAKA 5 - The A..sociauon of Voluntary Agencies, P O Box 5045. DHAKA 5 For hooJ..J·. Office lntcrnauonal de L1brainc s.a .. avenue Marn1x 30. 1050 BRliSSELS. For ptnodicals and wh,·, nptuuu: Orficc International des PCnodiques. avenue Marntx 30, 1050 BRUSSELS - Subscnpnoru to World J-lrulth 1m/r: Jean de Lannoy. 202 avenue du Roi . 1060 BRUSSELS " ' I nd1a, WHO Regio nal Office Botsalo Book, (Ply) Ltd .. P.O. Box 1532. GABO•ONE Bibli0tcca Regional de Medicina OMS/OPS. Unidade de Venda de Publica¢cs. Caixa Postal 20.381 , Vila Clcmcnuno. 04023 S.-o PAULO, S.P. , ,, India. WHO Regional Office Canadian Public Health Assoc1a11on, 1335 Carling Avenue, Suite 210, OTTAWA, Ont. KIZ 8N8. SubscnJ>- tum ordtn. accompamtd bi• chtqut madt out to tht Royal Bank of Canada. Ottawa. Account World Health Organiza11on. ma,· also~ s,nr ro rh, World Health Organizauon, PO Box 1800. Postal Stauon B, OTTAWA, Ont. KIP 5R5 China Na11onal Publicallons lmpon & fapon Corporauon, P.O. Box 88, BEUING (PEKING) "MAM". P 0 . Box 1722. NICOSIA Ania. Ve Smcckach 30, 111 27 PRAGUE I ,,, India. WHO Regional Office '.\,lunksgaard fapon and Subscription Service, Nerre Segadc 35. 1370 COPENHAGEN K (Tel : +45 I 12 85 70) Librcria Cicntifica S.A .. P 0 . Box 362. Luque 223. GLAYAOUIL Osiris Office (or Books and Reviews, 50 Kasr El Nil Street. CAIRO The WHO Programme Coordinator. P.O. Box 113. Su vA Akatccmincn KirJakauppa, Kcskuskatu 2. 00101 HELSINKI 10 Libramc Arnette, 2 rue Cas1mir•Oclav1gnc, 75006 PARIS Librainc Un1vcrs1tairc du Gabon. B.P. 3881 , L11R£VILI.E Buchhaus Leipzig, Postfach 140, 701 LEIPZIG Govi-Vcrlag GmbH . Ginnhcimcrstrassc 20. Postfach 5360. 6236 EscHBORN - W. E. Saarbach. Post(ach 101 610, Follcrstrassc 2. 5000 COLOGNE I - Alex. Horn. Spiegclgassc 9, Postfach 3340, 6200 WIESBADEN Fidcs Enterprises, P.O. Box 1628. ACCRA G .C. Elcfthcroudakis S.A., Libra1nc intcrnauonale. rue Nikis 4, ATHE s CT. 126) '.\,lax Bouchcrcau. Libraanc "A la Caravclle", Boitc postalc 111-B. PoRT-AU-l'RINCE Hong Kong Government lnformauon Services. Beaconsfield House. 6th Floor. Queen's Road, Central. VICTORIA Kultura. P.0 .B. 149, BUDAPEST 62 - Akadcm1ai Konyvcsbolt. Vaci utca 22, BUDAPEST V Snacbjorn Jonsson & Co .. P.O. Box I 131. Hafnarstraeu 9, REYKJAVIK WHO Regional Office for South-East Asia. World Health House. Indraprastha Estate. '.\,lahatma Gandhi Road, NEW DELHI 110002 - Oxford Book & Stauoncry Co., Scindia House, :-IEw DELHI I 10001 : 17 Park Street. CALCL'TTA 700016 (Sub-ag,nr) P. T. Kalman Media Pusaka. Pusat Pcrdagangan Sencn, Block I. 4th Floor. P.O. Box 3433/Jkt . JAKARTA Iran University Press. 85 Park Avenue. P.O. Box 54/551 , TEHRAN Ministry of lnformauon, National House for Publishing. Distributing and' Advcrusing. BAGHDAD TDC Publishers. 12 Nonh Fredenck Street. DuauN I (Tel : 744835-749677) Heiliger & Co .. 3 Nathan Strauss Street. JERUSAUM 94227 Ediziona Minerva Medica, Cor,o Bramante 83-85. 10126 TURIN : Via Lamarmora 3. 20100 MILAN Maruzcn Co. Ltd .. P.O. Box 5050. TOKYO lnterna11onal. 100-31 Jordan Book Centre Co. Ltd .. Univcrsuy S1ree1. P.O. Box 301 (AI-Jubc1ha/. AMMAN The Kuwait Bookshops Co. Ltd .. Thunayan AI-Ghancm Bldg, P.O. Box 2942. Ki;WAIT The WHO Programme Coordinator. P.O. Box 343. VIENTIANE The Levant Distnbu1ors Co. S.A.R.L .. Box 1181 . Makdass1 Street. Hanna Bldg, BEIRUT Librairic du Centre. 49 bd Royal. LUXEMBOURG '.\,lalaw1 Book Service. P.O. Box 30044. Chichiti. BLANTYRE 3 A/1 /84 WHO pubHonon, m,v be obtaJ IIIIIII IIIII IIIII IIIII IIIII IIIIIIIIIIIIIIIIIIIIIII MALAYSIA MALDIVES MEXICO MONGOLIA MOROCCO MOZAMBIQUE NEPAL NETHERLANDS NEW ZEAi.AND NIGERIA NORWAY PAKISTAN PAPUA NEW GUINEA PHILIPPINES POLAND PORTUGAL REPUBLIC OF KOREA SIERRA LEONE SINGAPORE SOUTH AFRICA SPAIN SRI LANKA SWEDEN SWITZERLAND THAILAND TUNISIA TURKEY UNITED KINGDOM UNITED STATES OF AMERICA URUGUAY USSR VENEZUELA YUGOSLAVIA ZAIRE The WHO Propammc trick's Buildina). Jalan I Book Center, le.. L Hill - India, WHO Reaio~ Oflioe .. 0 0 0 0 0 5 9 6 .. Libreria Internacional, S.A. de C.V., Av. Sonona 206, 06l~Mtxico, D.F. - India, WHO Rep,nal Oflioe Editions La Porte, 281 avenue Mohammed V, RA&.\T INLD, Caixa l'oltal ~JO. MAP\1TO - India, WHO Reponal Ofrioe Medical Books Europe BY, Noorderwal 38, 7241 BL LocHIN Government Printina Ofrl<ll. Publicatio111 Section, Mu)arave St,-. Privaie Bae, WIIJJNOTON I ; Walter St,-, WIIWNOTON ; World Trade BuiJdina, Cubacadc, Cuba St,-, WIIJJNOTON. Govem-nt Boolcsloop, 01: Hannaford Bunon Buildina, Rutland Si,-, Privaie Bae, Aucxu.No; 159 Hereford Street, Privaie Bae, Ciwnoru.a<; Aleundra St,-, P.O. Box 857, Ii.um.TON ; T A. G Buildina, Princes St,-. P.O. Box 1104, DuNIDIN - R. Hill A. Son Ltd, Ideal Howe, Cnr Gillies A venue A. Eden Street, Newmarket, AUCXUNOI University Bookshop Ni,eria Ltd, University of Ibadan, •- J. G. Tanum A/S, P.O. Box 1177 Sentrum, OsLO I Mirza Book Aaenc:Y, 65 Shahrah-E-Quaid-E--Azam, P.O. Box 729, Lutoall 3; Sui Limited, Sui Centre, G.P.O. Box 779, I.i. Chundripr Road, K.uAatt The WHO Propammc Coordinaior, P.O. Box 646, KoNIIDOeU World Health Orpnization, Reaiooal Oflioe for the Wesiem Pacific, P.O. Box 2932, MANIU - The Modem Book Company Inc., P.O. Box 632, 922 Rizal Avenue, MANIU 2800 Skladnica Ksqanlta, uJ Mazowiecl<a 9, 00052 WAU>.w (txctpt {Hriodiaw) - BKWZ Ruch, ul Wronia 23, 00840 WAUAW ({Hriodicau 011ly) Livnaria RDdriaues, 186 Rua do Ouro, lJsaoN 2 The WHO Propamme Coordinator, Central P.O. Box 540, SaouL Njala Univemty Collqo Bookshop (Univenity or Sioma Leone). Privaie Mail Bae, Flllffl>WN The WHO Propammc Coordinator, 144 Moulmein Road, SIN<W'OU 1130; Newton P.O. Box 31, SINGA- POU 9122 - Selec1 Books (Pte) Ltd, 215 Tan1Jlin Shoppina Centre, 2/F, 19 Tan1Jlin Road, SINOAIORI! 10 Van Schailt's Bookstore (Pty) Ltd, P.O. Box 724, 268 Chun:h Street, PuroaiA 0001 Comercial Atheneum S.A., Consejo de Ciento 130-136, IIAacnoNA IS; General Moscard6 29, MADRJO 20 - Libreria Diaz de SanlOS, Lapsca 95 y Maldonado 6, MADRJO 6; Balmes 417 y 419, IIAacnoNA 22 - India, WHO Reaional orrioe For boo/cs: Al<tiebol.lpt C.E. Friues Kun1Jl. Hovbokhandel, Rea,erinpptan 12, 103 27 Sl'OCICHOLM. For {Hriodlcals: Wennerpen-Williams AB, Box 30004, 104 2S STOCIOtOI.M Mediziniscber Verla& Hans Huber, Una,us Sll'1IS50 76, 3012 IIDNI 9 ~- India, WHO Reaional orrioe Societe Tunisienne de Diffusion, S avenue de Catthqe, TuNIS Huct IGtapeV~ 469 lslil<lal Caddes~ Ileyoslu, ISTANIIUL H.M. Stationery Ofrioe : 49 Hiah Holborn, l..oNOON WCIV 6HB; 13a Castle St,-. EocNIUROH EH2 3AR; 80 Chichester Street, BluAST BTI 4JY; Bruennooe St,-. MANamTu. M60 BAS; 258 Broad Street, Illll- MtNOHAM 81 2HE; Southey House, Wine Street, 8aJSTOL !ISi 28Q. All moil orMn 1/oould b, ~nt to: HMSO PublicatioM' Centre, SI Nine Elms Lane, l..oNOON SW8 SDR Sinik ONi bulk copiu of indi•idMal publication, (nor .n,bxriprlom): WHO Publica1ions Centre USA, 49 Sheridan Avenue, Al.&ANY, NY 12210. Sub,cription,: Sub,criprlon ord,n, occomponi,d by ch«k mDlk our to tht Chemical Bank, New York, Account World Health Orpnization, 1/wuld b, Mnt to 1/w World Health Orpnization, PO Box 5284, Chun:h Street Station, NEW You, NY 10249; Corrupo,uknct co,._ ctmi"I .n,b,cription, ,hou/d b, oddraud 10 tht World Health Orpnization, Diaribution and Sales, 1211 Gl!NEVA 27, Switzerland. Pwblicatlon, ore obo owJi/Qbk from rlw United Nations Bookshop, N!W You, NY 10017 (retoil only) Libreria Ap,:,pecu&ria S. R. L , Casilla de Correo 17SS, Alzaibar 1328, M0NTEV1080 For rtaun in 1/w USSR requirini Rwsion ,dition,: KoffllOmolsl<ij proapeln 18, Medicinsl<aja Knip, Moscow - For ~" our.ruk rht USSR fW1Mirin1 Ru.uion tdirlon,: Kuzneclcij moo1 18, Metdunarocl- naja Knip, Moscow G-200 Libreria del Esle, Aputado 60.337, C.....cu 106 - Libreria Medic& Paris, Aputado 60.681, CuACAS 106 Juplovensl<a Knjip, Teruije 27/11, 11000 Ilaolw:ll! Librairie universilaire, avenue de la Paix NO 167, B.P. 1682, KJNSHASA I Spo,:ial term, for developing countrin are obtainable on application to the WHO Programme Coordina- tors or WHO Regional Officn lilted above or to the World HNlth Organization, Diltribution end Sain Service, 1211 Geneva 27, Switzerland. Order■ from countriea where aaln 11ge/lt1 have not yet been appointed may also be sent to the GeneYI eddr .... but muat be paid for in pound• sterling, US dollars, or Swill franc■. Price: Sw. fr. 5 .- Pricea 1111 1ubject to change without notice.