Lyophilized combination pools of enterovirus equine antisera: preparation and test procedures for the identification of field strains of 42 enteroviruses

Bull. Org. mond. Sant6s 1 1973, 48, 263-268Bull. Wld Hlth Org. Lyophilized combination pools of enterovirus equine antisera: preparation and test procedures for the identification of field strains of 42 enteroviruses JOSEPH L. MELNICK,' VERLE RENNICK,' BETTYLEE HAMPIL,1 NATHALIE J. SCHMIDT,2 & HELEN H. HO 2 This paper describes the preparation of 8 dried pools (designated A to H) of sera. Each pool is composed of 10 or 11 of 42 individual enterovirus equine sera and contains 500 antibody units of each serum component per 0.1 ml. Procedures for using the anti- serum pools are given, andguidance isprovidedfor interpreting the results ofserum neutrali- zation tests in identifying field isolates. Collaborative studies were carried out by WHO Regional reference centres, WHO virus collaborating laboratories, and other laboratories (Melnick & Hampil, 1965, 1970, 1973; Hampil & Melnick, 1968) on the homologous and heterotypic properties of 42 enterovirus equine sera. These studies provided the basic information required for the preparation of combinations of these sera suitable for use in identi- fying field strains of the enteroviruses that are usually isolated in monkey kidney tissue cultures. The practicability of combining equine enterovirus sera for identifying 42 enterovirus types was evalua- ted by Schmidt et al. (1971). Pools of 42 such sera (P1-3, CA7, 9, and 16, CB1-6, and E1-7, 9, 11-27, and 29-33) were prepared by two methods-that described by Lim & Benyesh-Melnick (1960) and the " inter- secting serum " scheme of Schmidt et al. (1961). The pools were designed to contain 50 antibody units per 0.1 ml of each immune serum present in them. Both sets of pools were tested not only against the proto- type viruses but also against field strains of entero- 1 WHO International Reference Centre for Enteroviruses, Department of Virology and Epidemiology, Baylor College of Medicine, Houston, Tex., USA. 2 Viral and Rickettsial Disease Laboratory, California State Department of Public Health, Berkeley, Calif., USA. Note: Information on how to obtain these reagents is available from the Research Resources Branch, National Institute of Allergy and Infectious Diseases (NIAID), National Institutes of Health (NIH), Bethesda, Md. 20014, USA; from Virus Diseases, World Health Organization, 1211 Geneva 27, Switzerland; or from any WHO Entero- virus Reference Centre. The 8 dried serum pools and mono- valent coxsackievirus B type 3, echovirus 11, and echovirus 12 antisera are supplied as a unit package. A circular giving instructions for use and for the interpretation of results is included in each package. viruses. With test virus doses of 320-3 200 TCD50 approximately equal numbers of the isolates were identified by both methods. Since the Lim-Benyesh-Melnick method requires 8 pools, compared with 13 for the "intersecting serum " scheme, the NIAID Subcommittee on Ente- roviruses recommended, for reasons ofeconomy, that the former be used for preparing dried serum pools suitable for global distribution and long-term storage. The Subcommittee also recommended that the pools be prepared for drying at the level of 500 antibody units. For use in tests, such pools could be rehydrated and further diluted 1 : 10 so as to contain 50 anti- body units per 0.1 ml. This paper describes the preparation of the serum pools and the procedure for their use, and provides guidance for interpreting the results of serum neu- tralization tests for the identification of field isolates. MATERIALS AND METHODS Equine immune sera against 42 enteroviruses (po- lioviruses 1-3, coxsackievirus A types 7, 9, and 16, coxsackievirus B types 1-6, and echoviruses 1-7, 9, 11-27, and 29-33) were used for preparing the pools. Bulk stocks of the sera that had been stored frozen were used, with the exception of 3 sera (P3, E4, and El 1). The latter, which were in dried form, were rehydrated and pooled for use. The frozen sera were shipped to a commercial laboratory for preparation of the pools and further processing. The sera were then thawed and appropri- ate quantities of each were measured and distributed into 8 serum pool bottles (labelled A-H) by two of us (JLM and VR). 3006 -263- J. L. MELNICK ET AL. Table 1. Composition of serum pools A-H and titres of the sera used. Each pool con- tained 500 antibody units per 0.1 ml of each of its serum components. Assigned titre Pools containing the antiserum Antiserum per 0.1 ml(I antibodyunit) A B c D E F G H P1 P2 P3 CA7 CA9 CAl 6 CB1 CB2 CB3 a CB4 CB5 CB6 El E2 E3 E4 E5 E6 E7 E9 Eli E12 El 3 E14 El 5 E16 E17 E18 El 9 E20 E21 E22 E23 E24 E25 E26 E27 E29 E30 E31 E32 E33 8 000 19 000 16 000 9 200 8000 1 500 38 000 22 000 4 000 20 000 22 000 32 000 11 000 35 000 4700 6 000 13 000 8 000 15 000 7 500 2000 26000 17 000 7300 2000 12500 5 000 15 000 35 000 13 000 2 000 13 000 16 000 3 400 2000 19 000 2 000 8000 5 500 9 600 16 000 8 000 x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x xx x x x x x a This serum, erroneously labelled as " CB3 , was subsequently found to be monotypic CA9 serum. Table 1 shows the composition of the pools, each of which consisted of 10 or 11 sera. The table also shows the assigned titres of the immunotypes. These titres were used for calculating the amounts of undi- luted sera needed to prepare 7.5-litre pools containing 500 antibody units per 0.1 ml of each serum compo- nent. For example, using the assigned titre of 1: 8 000 for the poliovirus type 1 serum, it was calculated that 469 ml ofundiluted serum were required for a 7.5-litre pool. The required volumes were measured to the nearest millilitre, using graduated cylinders and pi- pettes. Samples of each serum were taken and stored frozen in case verification of the immunotype should be necessary. This proved to be a wise precaution, as will be shown later. When the pools had been prepared, each was fil- tered and the undiluted filtered serum was dispensed ')6A x x x x x x x x x x x x ENTEROVIRUS EQUINE SERA into 5-ml vials for drying so that, when the contents of each vial were rehydrated with 5 ml of distilled water, the rehydrated material contained 500 anti- body units of each serum component per 0.1 ml of the pool. Since the pools are used at the level of 50 antibody units per 0.1 ml, the content of one rehy- drated vial, when diluted 10-fold, provides 50 ml of the pool for use in neutralization tests. The total serum dilutions of the 8 pools range from 1: 13 to 1: 16 at this level. Test procedures for verifying the composition of the serum pools Prototype viruses were employed to verify the composition of the pools. The Du Toit strain of echo- virus type 4 was used rather than the prototype Pesascek strain. The pools were also tested against echovirus 8. The dried serum pools (A-H) were rehydrated by adding 5 ml of distilled water to each vial to make the undiluted stock. Each pool was further diluted 1: 10 with Melnick's medium B (Melnick & Wenner, 1969) containing antibiotics, to furnish 50-antibody- unit material for the tests. The diluted pools were then distributed into test tubes in 0.3-ml quantities and the sera were inactivated by heating in a water bath at 56°C for 30 minutes; 0.3 ml of the virus dilu- tion was then added to each tube. Each set of pools was tested at 2 dilutions of the prototype virus calcu- lated to contain 100 and 1 000 TCD50 per 0.1 ml. The serum-virus mixtures were then incubated at 37°C for 2 hours. A sample of the virus dilutions used was similarly incubated and subsequently titrated to determine the virus dose employed in the test. After incubation, 2 tissue culture tubes were ino- culated with each serum-virus mixture (0.2 ml per tube). Samples of the challenge virus dilutions were inoculated into each of 4 tissue culture tubes (0.1 ml per tube) and subsequently titrated using 4 tubes per dilution. The inoculated tubes were incubated at 37°C for 7 days. Microscopic readings for cytopathogenic effect (CPE) were made from the second day of incu- bation. As soon as the virus control tubes showed CPE to a degree of 3+ to 4+ (i.e., CPE involving 75% or more of the culture), daily readings were made for the entire test. RESULTS Results of the tests with prototype viruses showed that the serum pools correctly neutralized 37 of the 42 enteroviruses-i.e., polioviruses 1-3; coxsackie- virus A types 7, 9, and 16; coxsackievirus B types 1, 2, and 4-6; and echoviruses 1-7, 9, 13-21, 24-27, and 29-33. None of the pools neutralized CB3 virus, and echoviruses 11 and 12 were not correctly identi- fied by pools containing only the corresponding im- munotype serum. Also, as expected, E22 and E23 viruses could not be distinguished since each of the two antisera had a heterotypic titre of approximately 1 :128. Subsequent investigation of a number of frozen reserve samples taken from each serum bottle when the pools were made showed that incorrectly labelled serum had been used in two instances. The serum labelled " CB3 " by the commercial processor proved to be CA9 antiserum, and the batch ofE5 serum used in the pools was one containing antibody of signifi- cant titres against Ell and E12 viruses (Hampil & Melnick, 1968). On the basis of these results, it is recommended that the serum pools, together with CB3, Ell, and E12 monovalent sera, be used as described below. Identification of enterovirus field isolates by means of pools A-H The use of pools A-H supplemented with mono- valent sera against CB3, Ell, and E12 viruses offers a practical and economical means of identifying field isolates of the above-mentioned 42 entero- viruses. A convenient scheme for identifying an unknown virus neutralized by a given pool or pools is pre- sented in Table 2 with an indication of the instances in which special tests with monotypic antisera are required. These tests are described below. Neutralization tests are performed according to the procedures used with the prototype viruses, except that a single dose of the unknown virus, in a dilution containing 103-010-5 TCD50 per 0.1 ml, is recom- mended. The results of neutralization are checked against the scheme (Table 2) for identity. Often, a tentative identification can be made on the 4th or 5th day, particularly when the dose of virus is high or the virus grows rapidly. Viruses that grow more slowly may require the full 7 days of incubation. Complete neutralization on the 7th day with a dose of virus between 102.5 and 108*5 constitutes a firm identification. There are some instances, however, in which there may be complete neutralization on the 4th or 5th day and evidence of a slow breakthrough on the 6th or 7th day, probably owing to the presence of small 265 J. L. MELNICK ET AL. Table 2. Identification of neutralization test results viruses on the basis of Neutralization Identity Neutralization Identity by pool(s): a of virus by pool(s): of virus A E15 CE CB5 AB CA7 CF P1 AC CB1 CH E12d AD E33 CEG E30 AE CB4 D E25 AF E7 DE E13 AG E4 DF E14 AH El DG E16 ACF E29 DH P3 AEG E5 b DEH E32 B E21 E Ell b BC E2 EF E18 BD CB2 EG E17 BE P2 EH E22e BF E19 F E27 BG(C) CA9 c FG E20 BH E3 FH CB6 BDF E26 G E31 BFH E9 GH E236 C E24 H CA16 CD E6 a Because of a labelling error, CB3 antiserum does not appear in any of the pools. No neutralization of an enterovirus isolate by pools A-H strongly suggests that the isolate may be CB3 virus. For the identification of CB3, see under Special tests. b Pools AEG contain E5 antiserum, which may give a heterotypic neutralization of El 1 virus. To distinguish between E5 and El 1 viruses, see under Special tests. c Strong neutralization by pools B and G and weak neutralization(or no neutralization) by pool C identifies the virus as CA9. Because of a labelling error, pool C contains 10 antibody units against CA9 virus. d Heterotypic neutralization of El 2 by pools A, E, F, and G may occur. For the positive identification of El 2, see under Special tests. e Neutralization by pools E, G, and H indicates that the isolate may be either E22 or E23 (see under Special tests). aggregates of the virus in the culture (Wallis & Mel- nick, 1967). This is most likely to occur with field isolates of echoviruses 9, 17, and 27 tested by the pools. In this situation, identification is considered to be adequate, although in rare instances the investi- gator may deem it advisable to confirm the identifica- tion by the use of type-specific antiserum. Special tests Dried equine sera in 2-ml vials are used for the identification of coxsackievirus B3 and echoviruses 11 and 12. The 50-antibody-unit materials and any unused portion of the rehydrated serum should be stored frozen. Coxsackievirus B3. CB3 virus is not neutralized by any of the pools A-H since specific antiserum does not appear in the pools because of the mis- labelling that occurred (see page 265). If the field isolate is not neutralized by any of the pools, 103-0 TCD5O of the virus should be tested against 50 units of CB3 antiserum. The contents of the 2-ml vial should be reconstituted in 2 ml of distilled water. An appropriate quantity of the rehydrated serum should then be diluted 1: 80 in Melnick's medium B to obtain 50 antibody units per 0.1 ml. Echovirus 11. Eli antiserum appears only in pool E, but unless large doses of the virus are employed it may be neutralized by the heterotypic antibody of pools A and G, which is mainly associated with the E5 serum present in these two pools and in pool E. Therefore, if neutralization bypoolsAEG occurs, spe- cial tests are required to determine whether the isolate is ElI virus or ES virus. Thus, 103-0 TCD50 of the virus should be tested against 50 units of E1 serum. The contents of the 2-ml vial should be reconstituted in 2 ml of distilled water. An appropriate quantity of the rehydrated serum should then be diluted 1: 40 in Melnick's medium B to obtain 50 antibody units per 0.1 ml. If neutralization occurs, the virus is EII; if not, it is E5. Echovirus 12. The specific E12 antiserum appears in poolsC and H; however, there is also strong hetero- typic neutralization of E12 virus by pools A, E, and G and weak neutralization by pool F. Therefore, test results on an isolate neutralized by pools C and H and one or more of pools A, E, F, and G indicate the need for a special test to identify the virus un- equivocally as E12. Thus, 103.0 TCD5O should be tested against 50 units of E12 serum. The contents of the 2-ml vial should be reconstituted in 2 ml of distilled water. An appropriate quantity of the rehydrated serum should then be diluted 1: 520 in Melnick's medium B to obtain 50 antibody units per 0.1 ml. Neutralization identifies the virus as E12. Echovirus 22/23 complex. E22 antiserum appears in pools E and H; E23 serum, in pools G and H. Unless large doses of virus are employed, the two viruses are not clearly distinguished, but may be 266 ENTEROVIRUS EQUINE SERA 267 neutralized by pools E, G, and H. Titration of the isolate against both antisera is required for positive identification of the serotype. Also, heterotypic neu- tralization of small doses of E22 virus may occur with pool B. However, this need not confuse identi- fication, since none of the other enteroviruses can be identified by this combination (BEH). DISCUSSION The functions of Regional enterovirus laboratories are twofold, since these laboratories carry out the diagnostic testing of etiologic agents of enteroviral disease and also act as centres for epidemiological studies on the incidence and distribution of entero- viruses in the local area. Characterization of the many isolates is tedious and burdensome unless combina- tion pools of standard reference monovalent sera are employed. The feasibility of preparing satisfactory combinations ofequine antisera as standard reference reagents was demonstrated by Schmidt et al. (1971), and the large reserves of frozen sera made it possible to prepare a sufficient quantity of these standard reagent pools to last for many years. The 8 dried serum pools (A-H) are similar in composition to those recently reported (Schmidt et al., 1971), although they differ from them slightly in some respects. The earlier pools were prepared in liquid form at the level of 50 antibody units, using the serum titres shown in Table 1 for calculation purposes. Dried sera were rehydrated for use. The pools described in this paper were prepared with fro- zen serum from the reserves available, except for 3 dried sera (P1, E4, and Eli). The pools were pre- pared as 500-antibody-unit material and dried, and tests were performed on the rehydrated sera diluted so as to contain 50 antibody units. As shown in the results of tests for the correctness of the serum mix- tures, it was found that, of the 42 enteroviruses tested, 3 (CB3, Ell, and E12) could not be satis- factorily identified owing to the inadvertent use of 2 frozen sera that had been mislabelled. This error, though unfortunate, does not seriously affect the use- fulness of the materials, since 3 monovalent sera are available for supplementary tests. Whereas no single reagent or set of reagents can be expected to resolve unequivocally all the identi- fication problems inherent in the sometimes complex field of enterovirus infections, the availability of these combination pools of reference enterovirus sera and the standardization of the procedures for using them should be helpful. In addition, their use should prove economically advantageous in terms of the materials and personnel required, thereby increasing the productivity of enterovirus laboratories. ACKNOWLEDGEMENTS The preparation and testing of the serum pools were supported by contracts PH 43-68-1044 and NIH 69-89, from the Research Resources Branch, NIAID. The Subcommittee on Enteroviruses 1 (Research Re- sources Branch, NIAID, NIH, Bethesda) critically ap- praised the results and recommendations. RIESUMJ MELANGES LYOPHILISES DE SERUMS EQUINS ANTI-ENTEROVIRUS: PREPARATION ET TECHNIQUES D'EPREUVE POUR L'IDENTIFICATION DES SOUCHES SAUVAGES DE 42 ENTEROVIRUS On a prepare 8 melanges Iyophilises renfermant, diversement associes, 42 serums 6quins anti-ent6rovirus (poliovirus 1-3, coxsackievirus A des types 7,9 et 16, coxsackievirus B des types 1-6, et echovirus des types 1-7,9,11-27 et 29-33). Chaque melange, compos6 de 10 'a 11 antiserums, contenait 500 unit6s d'anticorps de chaque constituant par 0,1 millilitre. Au moment de 1'emploi, les melanges ont 6te r6hydrat6s puis dilu6s 10 fois afin d'obtenir une activite de 50 unites d'anti- corps par 0,1 millilitre. On a eu recours a des epreuves de neutralisation por- tant sur la serie complte des ent6rovirus, representes par des souches prototypes, pour verifier la presence des constituants prevus dans chacun des melanges. Les r6sultats ont montr6 que 37 des enterovirus 6taient correctement identifi6s par le ou les m6langes appropri6s: poliovirus 1-3, coxsackievirus A 7,9 et 16, coxsackievirus B 1,2,4-6 et echovirus 1-7,9,13-21,24-27 et 29-33. Les echovirus 22 et 23 ont 6te neutralises par les melanges renfermant le serum anti-6chovirus 22 ou le serum anti- 6chovirus 23 ou ces deux antiserums. En raison d'une erreur d'etiquetage de certains lots de serums, il est n6cessaire d'utiliser des antiserums monovalents pour ' The members of the Subcommittee are: M. H. Hatch, J. L. Melnick, L. Rosen, N. J. Schmidt, and H. A. Wenner. 268 J. L. MELNICK ET AL. identifier le coxsackievirus B3 et les 6chovirus 11 et 12. L'article d6crit des techniques normalisees pour les epreuves de neutralisation et expose les modalit6s d'inter- pr6tation des r6sultats. L'emploi des 8 m6langes d'anti- serums et, le cas 6cheant, des 3 antis6rums monovalents (coxsackievirus B3, 6chovirus 11 et 12) represente une m6thode peu cofuteuse d'identification des souches sau- vages d'ent6rovirus. REFERENCES Hampil, B. & Melnick, J. L. (1968) Bull. Wld Hlth Org., 38, 577-593 Lim, K. A. & Benyesh-Melnick, M. (1960) J. Immunol., 84, 309-317 Melnick, J. L. & Hampil, B. (1965) Bull. Wld Hlth Org., 33, 761-772 Melnick, J. L. & Hampil, B. (1970) Bull. Wld Hlth Org., 42, 847-863 Melnick, J. L. & Hampil, B. (1973) Bull. Wld Hlth Org., in press Melnick, J. L. & Wenner, H. A. (1969) In: E. H. Lennette & N. J. Schmidt, ed., Diagnostic procedures for viral and rickettsial infections, 4th ed., New York, American Public Health Association, pp. 529-602 Schmidt, N. J. et al. (1961) J. Immunol., 87, 623-626 Schmidt, N. J. et al. (1971) Bull. Wid Hlth Org., 45, 317-330 Wallis, C. & Melnick, J. L. (1967) J. Virology, 1, 478-488