Comparison of the haemagglutinin production of influenza A2 Hong Kong variant and recombinant strains*

Bull. Org. mond. Sant 1970 43 275-279Bull. Wld Hitha Org. j190432727 Comparison of the Haemagglutinin Production of Influenza A2 Hong Kong Variant and Recombinant Strains* NICOLA M. TAURASO 1 & THOMAS C. O'BRIEN 2 Seven Hong Kong variant strains were comparedfor their capacity to produce haemag- glutinin (HA). From the results of 9 serial egg passages made at 33°C and 36°C, the A2/Aichi/2/68 strain was found suitable as the vaccine production strain for the 1968-69 monovalent influenza vaccine manufactured in the United States of America. Results of initial experiments comparing the HA production by 2 Hong Kong variant recombinants and by the A2/Aichi/2/68 strain showed no essential difference among these strains. When these experiments were repeated starting with viruses that had undergone the same number ofegg passages from isolation, the mean HA and chicken cell agglutina- tion titres ofthe 3 recombinant strains used in these studies were only 1.5 times greater than those ofthe A2/Aichi/2/68 strain. By comparison, the HA producing capacity ofAO/PRI8/34 was almost 6 times that of the A2/Aichif2/68 strain. The basic principle for enhancing the growth capacity of influenza viruses through recombination was established shortly after the 1957 pandemic by Kilbourne & Murphy (1960). The possible usefulness of this procedure, suggested by Kilbourne& Murphy, for influenza vaccine production prompted us to compare the growth capacities of three influenza A2 Hong Kong variant recombinant strains and their parental strains. These comparisons were made under conditions that simulated, as closely as possible, those encountered in the large-scale manufacture of vaccine. MATERIALS AND METHODS Virus strains Two Hong Kong variant strains (A2/Aichi/2/68 and A2/HK/16/68) were recombined with AO/PR/ 8/34 by Dr Edwin D. Kilbourne, who kindly * A preliminary report of these data was presented in brief discussion at the International Working Conference on Hong Kong Influenza, Atlanta, Georgia, 14-16 October 1969. 1 Acting Chief, Laboratory ofVirology and Rickettsiology, Division of Biologics Standards, National Institutes of Health, Bethesda, Maryland 20014, USA. 'Staff Associate, Laboratory of Virology and Rickett- siology, Division of Biologics Standards, National Institutes of Health, Bethesda, Maryland 20014, USA. supplied the recombinant strains. Recombination was accomplished by previously described methods (Kilbourne & Schulman, 1965; Sigiura & Kilbourne, 1965; Kilbourne et al., 1967; Kilbourne, 1968). According to Kilbourne (1969), the recombinant strains had the antigenicity of the Hong Kong variant strains and the growth capacity in the chick embryo allantoic sac, as measured by haemagglutina- tion (HA) titres, of the PR/8 parent. We received the Aichi/2 Hong Kong variant strain from Dr Fred Davenport and in turn supplied it to Dr Kilbourne. The PR/8 strain used was part of the reagent collec- tion maintained in our laboratory and was not the actual strain used for recombination. All other virus strains were obtained either from Dr Marion Coleman, WHO International Influenza Center for the Americas, National Communicable Disease Center, Atlanta, Georgia, or from Dr Edward Buescher, Walter Reed Army Institute of Research, Washington, D.C. Virus passage and harvesting Each influenza virus strain (AO, A2, and A2 recombinants) was passaged in 10-day-old em- bryonated White Rock chicken eggs according to the following procedure: (1) 0.1 ml of an appropriate virus dilution, which 2553 -275- 276 N. M. TAURASO & T. C. O'BRIEN depended upon the HA titre, was inoculated into the allantoic sac of 24 eggs; 1 (2) half of these were incubated at 33°C and half at 36°C; (3) 6 eggs were harvested from each temperature group at 48 hours and the remaining 6 at 72 hours after inoculation; (4) eggs were candled at the time of harvesting and the allantoic fluids from live embryos only were harvested and pooled; (5) the allantoic fluid pools were centrifuged at 1000 rev/min (180 g) for 10 minutes; (6) the supernatant fluids were tested for HA and chicken cell agglutination (CCA) titre and were used as inoculum for the subsequent passage. At each passage there were thus 4 pools of virus, obtained from eggs incubated at different tempera- tures and for different times (temperature-harvest- time pools). In each temperature group, the pool having the higher HA titre was used for passage. All strains being compared were passaged simul- taneously. After HA tests had been performed, the virus pools were stored at -70°C for future reference. Haemagglutination tests These tests were performed according to pre- viously described methods (Tauraso et al., 1969) at the time of each passage, using infectious allantoic fluids that had not been subjected to freezing and cockerel red blood cells (RBC). At the termination of a series of passages, samples of all frozen pools of all strains were tested simultaneously using cockerel, human type 0, and guinea-pig RBC. The HA value for each pool was the average of the end- points of the serial 2-fold dilutions (starting at 1: 8 and 1: 12). The highest average of the 4 tempera- ture-harvest-time pools was the HA value for that passage of a particular virus strain. Chicken cell agglutination tests The CCA tests were performed according to the method of Miller & Stanley (1944) as described in a National Institutes of Health (NIH) memorandum.2 1 The inoculum dilution depended upon the HA titre according to the following schedule, expressed as reciprocal of HA titre/log,, of inoculum dilution: = 16-48/-i; 64-96/-2; 128-256/-3; 256-512/-4; 1024-1536/-5; and 92048/-6. ' Titration of chicken red cell agglutination (CCA) value (NIH memorandum of 16 September 1946, available from the Division of Biologics Standards, National Institutes of Health, Bethesda, Md. 20014, USA). Samples of all pools of all strains were tested simultaneously at the termination of a series of passages, as was done for the HA tests described above. RESULTS Selection of a strain for vaccine production Soon after it became evident that the Hong Kong variant was sufficiently different antigenically from previously occurring influenza virus strains, con- sideration was given to selecting a strain suitable for vaccine production. The seven strains available to us at that time underwent 9 serial passages. Table 1 compares the HA production of these strains. The 2 strains that produced the most haemagglu- tinin were both egg-derived. One of these, the Aichi/2 strain, was selected for the production of 1968-69 monovalent vaccine. Evaluation of recombinant strains When the first Hong Kong variant recombinant, HK/16-PR/8 (LVR 1086), became available, we compared its HA production with that of the Aichi/2 vaccine strain. The mean HA titres of these two strains over a period of 10 serial egg passages were approximately equal (Experiment 1, Table 2). The next recombinant we obtained was the Aichi/ 2-PR/8 (LVR 1119) strain. We compared the HA and CCA productions of this strain with those of the two previous strains passaged simultaneously. The mean HA and CCA values of both the Aichi/2 parent and the Aichi/2-PR/8 recombinant were approximately 1.5 times higher than those of the HK/16-PR/8 recombinant (Experiment 2, Table 2). These results indicate that there was little differ- ence between the Aichi/2 and the recombinant strains. The discrepancy between these results and those reported by Kilbourne (1969) may be due to a bias produced by the higher passage level in favour of the Aichi/2 strain. In addition, Kilbourne (per- sonal communication) used human type 0 RBC, whereas we routinely use cockerel RBC for influenza virus haemagglutination tests. To resolve this discrepancy, an experiment was performed starting with viruses that had undergone the same number of egg passages from isolation, with the exception of the PR/8 strain. Table 3 compares the HA production of the strains tested using cockerel, human type 0, and guinea-pig RBC and the CCA titres using cockerel RBC. The mean HA and CCA values of the recombinants were only 1.5 times greater than those of the Aichi/2 k trent strain. HAEMAGGLUTININ PRODUCTION OF HONG KONG INFLUENZA STRAINS TABLE I COMPARISON OF INFLUENZA A2 HONG KONG VARIANT STRAINS FOR HAEMAGGLUTININ PRODUCTION Virussrain a Isolation] Passage Study [Mean HA~ HA dVirus train system b r on passages c HAd standard rating6 ___________ _____ ~~receipt, _ __ titre deviation rtn A2/Aichil/2/68 Egg E/5 Eg-18 730 751 1.7(LVR 1067) A2/Philippines/303/68 Egg E/1 E3-12 749 708 1.7(LVR 1070) A2/HK/50/68 MK MK/?,f E/i E5-14 688 752 1.6(LVR 1061) A2/HK/1/68 MK MK/?,f E/i E2-11 504 296 1.2(LVR 1066) A2/HK/8/68 MK MK/?,f E/i E5-14 518 594 1.2 (LVR 1060) A2/HK/8138562/68 Egg E/? g E3-12 461 575 1.1(LVR 1065-13) A2/Tha(land/303/68 Egg E/i E3-12 435 371 1.0(LVR 1069) a Serial number assigned to virus strains received by our laboratory. b Virus was inoculated into the allantoic sac of embryonated eggs. MK= monkey kidney cell cultures. c Actual egg passages made during this comparative study. d HA= haemagglutination. e Rating determined from mean HA titre. A value of 1.0 was arbitrarily given to the lowest value, i.e., for the A2/Thailand/303/68 strain. The other ratings were obtained by dividing the mean HA value for a particular strain by the mean HA value for the A2/Thailand/303/68 strain. f Number of MK passages not known. g Number of egg passages not known; first passage In our laboratory was arbitrarily called passage E/1 for the purposes of this analysis. Using cockerel RBC, the mean HA titre for PR/8 was almost 6 times greater than that for the re- combinant strains. The low HA titres for PR/8 using human type 0 RBC were surprising and unexplainable. DISCUSSION When a new influenza virus variant appears, a suitable strain for vaccine production must, as a rule, be selected within a very short time. The situation during the early weeks of the 1968 Hong Kong influenza epidemic was no exception. We were fortunate in being able to select a strain for vaccine production much sooner than in the 1957 Asian influenza epidemic (Murray, 1969). The A2/ Aichi/2/68 strain, used for manufacturing the 1968-69 monovalent vaccine in the United States of America, proved to be a satisfactory production strain. Comparison of the HA and CCA production of three Hong Kong recombinant strains and of the Aichi/2 and PR/8 parent strains showed that, under certain conditions, the recombinant strains could produce approximately 1.5 times more haemag- glutinin than the Aichi/2 strain. However, contrary to the observations of Kilbourne (1969), the HA and CCA activities of the recombinant strains did not closely approximate to those of the PR/8 strain. The selection of an Aichi/2-PR/8 recombinant yielding HA and CCA titres equivalent to those obtained with PR/8 could possibly be achieved by recombining an Aichi/2-PR/8 recombinant with PR/8. It seems appropriate to comment on the practi- 277 N. M. TAURASO & T. C. O'BRIEN TABLE 2 COMPARISON OF HA AND CCA TITRES FOR A2/Aichi/2/68 AND RECOMBINANT STRAINS a V Study IMean HA HA Mean CCA CCAVirus strain passages b titre rating titrec rating Experiment 1: A2/Aichi/2/68 E20-30 463 1.05 ND _(LVR 1067) A2/HK/16/68-AO/PR/8/34 El1-21 442 1.00 ND - (LVR 1086) d Experiment 2: A2/Aichi/2168 E30-32 368 1.3 217 1.6(LVR 1067) A2/Aichi/2/68-AO/PR/8134 E14-16 395 1.4 198 1.5(LVR 1119) c A2/HK/1 6/68-AO/PR/8/34 E22-24 275 1.0 135 1.0 (LVR 1086) d a See Table I footnotes. b Cumulative egg passages since isolation. c ND= not done. d The A2/HK/16/68 recombinant strain (LVR 1086) had a passage history of MK/?, E/3 prior to recombination, during which it underwent 6 additional egg passages. e The A2/Aichi/2/68 (X-31) recombinant strain (LVR 1119) had 7 egg passages prior to recom- bination, during which it underwent 4 additional egg passages. TABLE 3 COMPARISON OF HA AND CCA TITRES FOR A2/Aichil2/68, A0/PR/8134, AND THREE RECOMBINANT STRAINS a RBC species used d Virus strain b Study Cockerel Human Guinea-pig F Cockerelpassages c Mean HA Mean HA Mean HA Mean CAHA HAtin HA HAt HHA HAt CCA CaCtAtitre rating titre rating titre rating titre rating A2/Aichi/2168 (LVR 1067) E13-24 680 1.0 2371 1.0 1226 1.0 105 1.0 A2/Aichi/2/68-A0/PR/8/34 (LVR 1119A) e E13-24 903 1.3 2642 1.1 1774 1.4 161 1.5 A2/Aichi/2/68-AO/PR/8/34(LVR 1119) E13-24 952 1.4 2990 1.3 1815 1.5 156 1.5 A2/H K16/68-AO/PR/8/34 E13-24 1136 1.7 3438 1.5 2183 1.8 160 1.5(LVR 1086) AO/PR/8/34 (LVR 548A) 4025 5.9 5120 2.2 5013 4.1 270 2.6 a See Table I footnotes. b Strains are listed in order of increasing mean HA titres. c Cumulative egg passages since isolation. d1 HA and CCA tests were performed simultaneously with samples of all pools of all virus strains at the termination of the final passage. e The A2/Aichi/2/68 (X-31) recombinant strain (LVR 1119A) had 7 egg passages prior to recombination, during which it underwent 5 additional egg passages. The origin and passage history of all other strains are described in the text or in the foot- notes to Table 2. 278 HAEMAGGLUTININ PRODUCTION OF HONG KONG INFLUENZA STRAINS 279 cability of the recombination procedure as a means of selecting strains with improved growth capacity. If the same increase in growth potential could be achieved through rapid passage in eggs, this would obviate the need for the laborious recombination procedure. The results of our experiments have not supported the suggestion that recombinant strains would solve the vaccine production problems en- countered in the past (Kilbourne & Murphy, 1960; Kilbourne, 1969). ACKNOWLEDGEMENTS The authors wish to thank Dr Edward Buescher, Dr Marion Coleman, Dr Fred Davenport, and Dr Edwin Kilbourne for the virus strains, and Rachel Yahwak and Michael Klutch for their excellent technical assistance. RESUME' iTUDE COMPARATIVE DE LA PRODUCTION D'HEMAGGLUTININE PAR LE VARIANT HONG KONG DU VIRUS GRIPPAL A2 ET PAR DES SOUCHES OBTENUES PAR RECOMBINAISON On a compar6 la capacite de production d'h6magglu- tinine (HA) de sept souches du variant Hong Kong: A2/Aichi/2/68; A2/Philippines/303/68; A2/HK/1/68; A2/HK/8/68; A2/HK/50/68; A2/HK/8138562/68; et A2/Thailand/303/68. Apres 9 passages en serie sur l'ceuf a 33°C et 36°C, la souche A2/Aichi/2/68, l'une des deux souches produisant le plus d'h6magglutinine, a ete choisie en vue de la preparation de vaccin antigrippal monovalent pour la periode 1968-1969 aux Etats-Unis d'Amerique. Lors de premieres recherches portant sur deux recom- binants Hong Kong (obtenus par recombinaison de la souche AO/PR/8/34 avec respectivement les souches A2/HK/16/18 et A2/Aichi/2/68) et la souche A2/Aichi/ 2/68, on n'a constat6 aucune difference sensible entre ces diverses souches sous le rapport de la production d'he- magglutinine. Lorsque les experiences ont et6 repetees en utilisant des virus ayant subi un meme nombre de passages sur l'aeuf apres l'isolement, les titres moyens HA et CCA (agglutination des hematies de poulet) de trois recombinants n'ont ete que 1,5 fois superieurs 'a ceux de la souche A2/Aichi/2/68. Leur capacite de production d'hemagglutinine n'etait en rien comparable a celle de la souche AO/PR/8/34 qui etait pres de 6 fois plus elev&e que celle de la souche A2/Aichi/2/68. Ces observations ne plaident pas en faveur du recours aux recombinants pour resoudre les problemes rencontr6s dans le passe en matiere de production de vaccins anti- grippaux. Si l'on peut obtenir un accroissement equivalent du potentiel de croissance d'une souche par passages rapides sur l'ceuf, on dispose d'un procede qui permet d'eviter les manipulations delicates que requierent les recombinaisons de virus. REFERENCES Kilbourne, E. D. (1968) Science, 160, 74-76 Kilboume, E. D. (1969) Bull. Wld Hlth Org., 41, 643-645 Kilbourne, E. D., Lief, F. S., Schulman, J. L., Jahiel, R. I. & Laver, W. G. (1967) Antigenic hybrids of influenza virus and their implications. In: Pollard, M., ed., Perspectives in virology, vol. 5, New York, Academic Press, pp. 87-106 Kilbourne, E. D. & Murphy, J. S. (1960) J. exp. Med., 111 387406 Kilbourne, E. D. & Schulman, J. L. (1965) Trans. Ass. Amer. Phycns, 78, 323-333 Miller, G. L. & Stanley, W. M. (1944) J. exp. Med., 79, 185-195 Murray, R. (1969) Bull. Wid Hlth Org., 41, 495-496 Sugiura, A. & Kilbourne, E. D. (1965) Virology, 26, 478-488 Tauraso, N. M., O'Brien, T. C. & Seligmann, E. B., Jr (1969) Bull. Wld Hlth Org., 41, 497-506