Methods of increasing the sensitivity of the haemagglutination inhibition test for rabies virus antibody

Bulletin ofthe World Health Organization, 62(6): 883 - 892 (1984) (c) World Health Organization 1984 Methods of increasing the sensitivity of the haemagglutination inhibition test for rabies virus antibody K. MANNEN,' K. ISHIKAWA,' J. TACHIBANA,2 & K. MIFUNE' A new method for the removal of non-specific inhibitors of rabies virus haemagglutinin has been developed. Treatment with colloidal silicic acid (Aerosil) or with acetoneplus Aerosil reduced the non-specific inhibitors in human, mouse, anddog sera to a level that was undetectable at the 1:4 starting dilution in the haemagglutination inhibition test. Bromelain-treated goose erythrocytes were much more susceptible to haemagglutination by rabies virus than were untreated erythrocytes, and the sensitivity of the haemagglutin- ation inhibition test was considerably increased by using bromelain-treated erythrocytes. Low levels ofantibodies in serafrom immunized human subjects were detected with higher sensitivity by combining Aerosil treatment of the sera with the use of bromelain-treated goose erythrocytes in the haemagglutination inhibition test. After the establishment of the haemagglutination (HA) test for rabies virus (1), the haemagglutination inhibition (HI) test was expected to be a simple and reliable method of measuring protective antibody against rabies. However, the test has not been widely used in the study of rabies, probably because high titres of non-specific inhibitors are present in sera from humans and various species of animals, and these inhibitors cannot easily be removed. The nature of these inhibitors has been characterized (2, 3) and they have been found to be mainly low-density lipoproteins (2). Kaolin treatment of the serum has been employed to remove these inhibitors (1); however, the serum has to be diluted 1:10 for the treatment, resulting in inability to measure low HI antibody titres, and even at a serum dilution of 1:20, one occasionally encounters some remaining non- specific inhibitors (1, 4). In addition, HI titres have been reported to be 10-100 times lower than neutral- izing antibody titres. In the present study, an attempt was made to increase the sensitivity of the HI test, in two ways. First, we examined the possibility of removing the non-specific inhibitors of rabies virus haemagglutinin by treating the serum with colloidal silicic acid. This ' Department of Microbiology, Medical College of Oita, Hazama-cho, Oita 87956, Japan. Requests for reprints should be addressed to Professor Mifune. 2 Central Laboratory of Kyoritsushoji Co., Inashiki-gun, Ibaraki 30023, Japan. approach was tried because of a previous report that lipoproteins of human sera are completely removed by colloidal silicic acid without any effect on other proteins (5). Secondly, we tried to find erythrocytes that are more susceptible to rabies haemagglutinin. MATERIALS AND METHODS HA antigen The chick-embryo cell-adapted high egg passage (HEP) Flury strain of rabies virus was cultured for 6 days at 36 IC in chick-embryo fibroblast cells in medium 199 without serum. Tissue culture fluid was harvested and subjected to low-speed centrifugation to remove cell debris. The virus was then inactivated by incubation for 60 min at 37 IC in the presence of 0.2 ml/litre 13-propiolactone, as described previously (6), and used as HA antigen. This antigen usually possessed 64 HA units per aliquot of 0.025 ml. In some experiments, HA antigen concentrated by ultra- centrifugation was used. Sera Normal pooled human serum was obtained from the central diagnostic laboratory of the hospital of the Medical College of Oita. Individual normal human sera were obtained from students of the College who 4488 -883 884 K. MANNEN ET AL. had no history of rabies immunization; these sera were stored at - 30 °C until use. Sera from approxi- mately 100 normal mice were used as normal pooled mouse serum. Individual mouse and dog sera were obtained from normal animals that had not been immunized against rabies. Individual sera were obtained from the staff of the Department of Microbiology, who had various histories of immunization with inactivated rabies vaccine (HEP Flury strain) of chick-embryo cell-culture origin.a Antirabies mouse serum and ascitic fluid were obtained by immunizing mice with five weekly intra- peritoneal inoculations of a 10Go brain suspension from suckling mice infected with Challenge Virus Standard (CVS) strain virus. Treatment ofsera to remove non-specific inhibitors Treatment with colloidal silicic acid. Aerosil 200 was supplied by Nippon Aerosil Co., Tokyo, Japan. Aerosil is silica of high purity (more than 99.80/0) and contains no water; it does not produce any adverse effects in human subjects. The average diameter of the first-order particle of Aerosil 200 is approximately 12 nm and the surface area is 200 ± 25 m2 /g. Various dilutions of Aerosil in different buffers were assessed for their efficacy in removing non-specific inhibitors. The effect of different incubation times and temperatures was also investigated using test sera diluted 1:2 with borate saline and an equal volume of 50 g/litre Aerosil in borate saline. The mixture was incubated for various periods at various temperatures with occasional shaking, and centrifuged at 1400 g for 15 min at 4 'C. The supernatant was used for the HI test. Treatment with acetone. Acetone extraction of the serum was performed using essentially the method described by Clarke & Casals (7). Treatment with 250 gllitre kaolin. This was done by the method described by Kuwert (8), except that the sera were tested at a 1:2 dilution in borate saline as well as a 1:10 dilution. Acid-washed kaolinb was used in the study. All treated sera were absorbed with packed goose erythrocytes for 30 min in ice slurry by the usual technique before the HI test. HA test The HA test was carried out in disposable micro- titration plates using 2.5 ml/litre goose erythrocytes U KONDO, A. Preimmunization and postexposure treatment with inactivated rabies vaccine of chick-embryo cell-culture origin. Presented at a WHO/IABS Joint Symposiutn on Standardization of Rabies Vaccines for Human Use Produced in Tissue Cultures; Marburg, 1977. h Fisher Scientific Co., Fairlawn, NJ, USA. Table 1. Hi titres of non-specific inhibitors in pooled normal human and mouse sera after treatment with Aerosil or kaolin Hi titre Treatment Serum dilution Pooled normal Pooled normal human serum mouse serum None 1:2 2560 2560 Aerosil, 20 g/l 1:2 512 NTa Aerosil, 50 g/l 1:2 <4 <4 Aerosil, 100 g/l 1:2 <4 <4 kaolin, 250 g/l 1:10 <20 <20 kaolin, 250 g/l 1:2 32 128 a NT =not tested. at a final pH of 6.2. The diluent of the antigen was borate saline containing 4 ml/litre bovine albumin free of fatty acidc and 0.1 g/litre gelatin. This diluent was also used for dilution of sera. Treatment of goose erythrocytes with bromelain (EC 3.4.22.4) Bromelaind was dissolved in 0.02 mol/litre MES' buffer containing 1.0 ml/litre ethylenediamine- tetraacetic acid (pH 5.8) at various concentrations. An 80 ml/litre suspension of goose erythrocytes in the same buffer was mixed with an equal volume of bromelain and the mixture was incubated for 20 min at 37 IC. The erythrocytes were then washed 3 times with dextrose-gelatin-veronal buffer (pH 7.4) and suspended in virus-adjusting diluent (pH 6.2) immediately before use. HI test The HI test was carried out essentially as described elsewhere (8). A mixture of 4 HA units of antigen and serially diluted test serum was held at room temperature for 1 h, after which 0.05 ml of 2.5 ml/litre goose erythrocytes was added to each well. Neutralizing antibody test The serially diluted human sera, treated or untreated for removal of non-specific inhibitors, were mixed with 200 plaque-forming units of tissue culture- adapted CVS strain of rabies virus (9), and incubated for 1 h at 37 'C. Sigma Chemical Co., St Louis, MT, USA. d Wako Pure Chemical Industries, Osaka, Japan. e 2 (N-morpholino) ethane sulfonate acid. IMPROVED HI TEST FOR RABIES VIRUS ANTIBODY Table 2. Hi titres of non-specific inhibitors of rabies virus in sera treated with 50 g/litre Aerosil at various temperatures HI titre Serum 0 OC Room temperature for: 37 OC 56 OC for for for 1 h 5min 10min 20min 30min 30min 30min Pooled normal 8 4 4 <4 <4 <4 1 6 human serum Pooled normal NT" NT NT NT <4 <4 8 mouse serum a NT =not tested. Aliquots of the mixture were inoculated onto CER cells (10) in a 24-multiwell tissue culture plate in duplicate and incubated for 5 days for plaque formation under the medium supplemented with 10 g/litre methylcellulose. Neutralization antibody titre was expressed as the reciprocal of the serum dilution producing 50% inhibition of plaque formation. RESULTS Preliminary investigations Sera diluted 1:2 with borate saline were treated with an equal amount of various concentrations of Aerosil in borate saline for 30 min at room temperature. As shown in Table 1, untreated human and mouse sera exhibited an HI titre of non-specific inhibitors of 2560. When treated with 50 g/litre or 100 g/litre Aerosil, the inhibitors were no longer detected at the starting serum dilution of 1:4. Aerosil at 20 g/litre did not completely remove the inhibitors. Treatment of sera diluted 1:10 with 250 g/litre kaolin removed the inhibitors below the starting dilution of 1:20. However, significant titres of inhibitors were detected when sera diluted 1:2 were treated with 250 g/litre kaolin. Table 2 shows that non-specific inhibitors can be reduced to undetectable levels when serum at a dilution of 1:2 is treated with an equal amount of 50 g/litre Aerosil for more than 20 min at room temperature or for 30 min at 37 IC. Treatment for less than 10 min at room temperature or for 1 h in ice slurry left low levels of inhibitors. Slightly higher levels of inhibitors were left in the sera treated for 30 min at 56 IC. The reason for this is unclear, but the observation was reproducible. As shown in Table 3, when phosphate-buffered saline or physiological saline was used as diluent, a fairly high degree of naturally occurring haemag- glutinins appeared at higher serum dilutions, resulting in an inability to measure the titres of non- specific inhibitors below the titres of naturally occurring agglutinins. It was also noted that the removal of inhibitors was incomplete if undiluted serum was treated with 50 g/litre Aerosil. The effect of heating the serum at 56 °C for 30 min before treatment was tested using a 50 g/litre suspension of Aerosil in borate saline. The results indicated that treatment of diluted sera with Aerosil reduces the non-specific inhibitors to an undetectable level irrespective of heat treatment. However, after treatment of undiluted serum, higher levels of inhibitors remained in the sera that had been heated than in unheated sera (data not shown). Therefore, sera to be treated with Aerosil were not heat- inactivated in the subsequent experiments. Removal of non-specific inhibitors from individual sera Individual serum samples were then treated with 50 g/litre Aerosil in borate saline for 30 min at room Table 3. Results of treatment of normal pooled human serum with an equal amount of 50 g/litre Aerosil in various buffers Buffer Serum dilution Titre of Titre of for treatment non-specific naturally inhibitors occurring agglutinins Borate saline undiluted 8 < 2 (pH 9.0) 1:2 < 4 < 4 1:4 <8 <8 Phosphate-buffered undiluted 8 2 saline (pH 7.4) 1:2 < 32 16 1:4 <32 16 Physiological undiluted < 2 2 saline (pH 7.0) 1:2 < 16 8 1:4 <32 16 885 K. MANNEN ET AL. Table 4. HI titres of non-specific inhibitors of rabies virus in individual normal sera treated with Aerosil or kaolin No. of Sera showing complete Sera with Serum Treatment Serum dilution sera tested removal of inhibitors' inhibitors remaining No. % No. Titre Human Aerosil 1:2 23 15 65 8 4 acetone and Aerosil 1:2 44 44 100 0 acetone and kaolin 1:2 44 0 44 128-512 none 44 0 44 1280-5120 Mouse Aerosil 1:2 15 15 100 0 none 15 0 15 640-2560 Dog Aerosil 1:2 18 18 100 0 none 18 0 18 1280-10 240 ' At the starting dilution of 1:4. temperature. As shown in Table 4, the inhibitors were and approximately one-third of those in untreated removed completely from mouse and dog sera that sera. had initial titres of 640 to 2560 and 1280 to 10 240, The NT antibody titre of serum treated with respectively. However, in the case of human sera, a acetone before Aerosil was higher than that of serum low level (1:4) of inhibitors remained in 8 (35%) of 23 treated with Aerosil alone. The NT antibody titre of samples. Pretreatment of these 8 samples with the mouse immune serum treated with Aerosil was acetone resulted in removal of inhibitors by Aerosil, higher than that of the serum treated with kaolin. and this was also observed in a further 44 sera treated However these NT antibody titres cannot be in this way. Treatment of sera with acetone and kaolin compared directly with that of untreated serum, since left titres of non-specific inhibitors ranging from 128 animal serum contains a considerable amount of non- to 512 in all the samples. specific inhibitors which neutralize the virus (11, 12). Effectltreatment on NT antibody titre These results indicate that Aerosil and acetone plusAerosil treatments were not significantly more Since it is known that kaolin treatment of serum effective than the currently used kaolin treatment in partially removes NT antibody (2), the NT antibody removing NT antibodies. titres of the sera treated with kaolin or Aerosil were It was also demonstrated that the NT antibody compared with those of untreated sera. Table 5 shows titres in untreated human and mouse immune sera that the NT antibody titres of human immune sera were approximately 40-60 times higher than the HI after treatment with kaolin or Aerosil were similar, antibody titres in treated sera. Table 5. Effect of Aerosil treatment of serum on neutralizing antibody titres against rabies virus Serum Treatment Pretreatment NT antibody HI antibody HI titre/NT titre of serum dilution titre titre untreated serum Human immune none 1902 1280° serum kaolin 1:10 619 40 1/48 Aerosil 1:2 615 32 1/59 acetone + Aerosil 1:2 828 32 1/59 Mouse immune none 9892 2560° serum kaolin 1:10 2447 160 1/62 Aerosil 1:2 4155 256 1/39 ' This figure comprises titres of non-specific inhibitors and of specific HI antibody. 886 Serum dilution (HI) Serum control A A I Human Immune UT serum BR Mouse Immune UT serum RR Mouse Immune ascites UT BR Fig. 1. Result of HI test using untreated goose erythrocytes (UT) and bromelain-treated erythrocytes (BR). Human immune serum was treated with acetone plus Aerosil; mouse immune serum and ascites were treated with Aerosil only. ow a IMPROVED HI TEST FOR RABIES VIRUS ANTIBODY Use of bromelain-treated goose erythrocytes In our previous studies on the haemolysis and fusion activities of rabies virus, it was found that human erythrocytes became more susceptible to haemagglutination after treatment with proteolytic enzymes, and exhibited extensive haemolysis and fusion at acidic pH (13, 14). In the present study, attempts were made to increase the sensitivity of the HI test by applying this observation to goose erythro- cytes. As shown in Table 6, after treatment with bromelain, both human and goose erythrocytes became 4-16 times more susceptible to haemag- glutination. Although the increase in HA titre was greater for the human erythrocytes than for the goose .erythrocytes, the maximum HA titres were almost identical in both. In addition, the sedimentation rate was slower for the human erythrocytes. Autohaemag- glutination was not observed in the bromelain-treated erythrocytes. Therefore, goose erythrocytes treated with 0.05 g/litre bromelain were used in the subsequent experiments. Table 6. Rabies virus HA titres of human and goose erythrocytes after treatment with bromelain Bromelain HA titre concentration (g/litre) Human Goose erythrocytes erythrocytes 0 80 640 0.01 1280 1280 0.05 1280 2560 0.25 2560 5120 0.5 2560 5120 HI titres obtained using bromelain-treated and untreated goose erythrocytes were compared. Human immune serum, mouse immune serum and ascitic fluid treated with Aerosil or kaolin were used as the test samples. In all samples, HI titres obtained using bromelain-treated erythrocytes were 2-4 times greater than those with untreated erythrocytes (Table 7, Fig. 1). Moreover, the titres of the sera treated with Aerosil or acetone plus Aerosil and tested using bromelain-treated erythrocytes were 3-6 times higher than those for kaolin-treated sera tested using untreated erythrocytes, which is the currently used technique. However, one problem found with the system was the occasional appearance of naturally occurring agglutinins in the first well of serum control without HA antigen (Fig. 1). This phenomenon was observed only with the combination of Aerosil-treated human serum and bromelain-treated erythrocytes. Various attempts were made to prevent this haemagglutin- ation, but it was not possible to identify the factor responsible. If naturally occurring haemagglutin- ation is observed in the first control well and no haemagglutination inhibition is observed in experi- mental wells, the HI antibody titre should be judged to be below 1:8. HI test ofserafrom immunized subjects To examine the usefulness of the proposed procedure for the rabies HI test, human sera collected from individuals with different immunizaton histories were tested for their HI antibody and NT antibody titres. Sera were treated with acetone and Aerosil for the HI test, but untreated sera were used for the NT test. Table 7. HI antibody titres measured using bromelain-treated and untreated goose erythrocytes Treatment of serum Increase Serum Goose erythrocytes in titreb 250 g/litre kaolin Aerosil Human immune serum Untreated 40 32 Bromelain-treated 80 128 3.2 Mouse immune serum Untreated 160 256 Bromelain-treated 320 1024 6.4 Mouse immune ascites Untreated 40 64 Bromelain-treated 160 256 6.4 Human immune serum was treated with acetone and Aerosil. b Ratio of HI titre in serum treated with Aerosil and tested with bromelain-treated goose erythrocytes to that of serum treated with kaolin and tested with untreated erythrocytes. 889 K. MANNEN ET AL. Table 8. HI antibody titres of sera from subjects with a history of rabies immunization HI titre Ratio of HI and NT titre Subject With untreated With bromelain- NT titre With untreated With bromelain- goose erythrocytes treated goose erythrocytes treated erythrocytes erythrocytes M.O. 8 32 61 1:7.6 1:1.9 R.0. 8 64 40 1:5 1:0.6 K.l. 8 16 106 1:13.3 1:6.6 Y.K. 8 16 46 1:5.8 1:2.9 A.S. 32 128 1206 1:37.7 1:9.4 T.N. 4 16 102 1:25.5 1:6.4 K.M. 4 8 40 1:10 1:5 K.M. 32 64 370 1:11.6 1:5.8 K.M. 64 128 2015 1:31.5 1:15.7 T.N. <4 <4 < 10 S.T. < 4 < 4 < 10 S.M. 16 32 95 1:5.9 1:3 S.N. <4 <4 15 < 1:3.8 < 1:3.8 T.N. 8 32 69 1:8.6 1:2.2 K.H. 8 32 46 1:5.8 1:1.4 Mean 1:14 1:5 As seen in Table 8, low levels (below 1:20 serum dilution) of HI antibody titres, which would be undetectable in kaolin-treated sera, were detectable in many test sera treated with acetone and Aerosil. The ratio of HI antibody titre measured with untreated goose erythrocytes to NT antibody titre (HI:NT ratio) varied from 1:5 to 1:37.7 (average 1:14); it was generally higher in sera with a low NT antibody titre. HI titres measured using bromelain-treated erythrocytes were 2-4 times higher than those measured with untreated erythrocytes, producing an average HI:NT ratio of 1:5. Non-specific inhibitors were not detectable in any sera at a 1:4 dilution. DISCUSSION The present study demonstrated that non-specific inhibitors of rabies virus haemagglutinin can be efficiently removed from human, mouse and dog sera by treatment with Aerosil (50 g/litre in borate saline). For human sera, acetone extraction should be carried out before the Aerosil treatment. After adsorption with one drop of packed goose erythrocytes in the usual manner, the sera should be stored in a refrigerator until use. When samples are stored at -30 °C, naturally occurring agglutinins are occa- sionally observed in the serum control (Mannen et al., unpublished data, 1983). Aerosil treatment of serum resulted in a decrease in the NT antibody titre to approximately one-third. A similar decrease has been described after kaolin treatment (2). Stephan & Roka (5) reported that lipoproteins could be removed from undiluted human sera by treatment with Aerosil (20 g/litre); serum protein composition other than lipoproteins was not altered and NT antibody titres to herpesvirus and poliovirus and the HI titre to measles virus remained unchanged. The reason for the considerable decrease in antibody titres in the present study is unclear; however, differences in the nature of the inhibitors and in the experimental conditions (e.g., concen- tration of Aerosil suspension, incubation temper- ature, and serum dilution) are possible contributory factors. The mechanisms by which the non-specific inhibitors to rabies virus are removed by Aerosil also remain to be determined. Aerosil particles possess a much greater surface area (200 m2/g) than kaolin and have a weak negative surface charge. These character- istics might play some part in enhancing the adsorption of the inhibitors to the surface. Further 890 IMPROVED HI TEST FOR RABIES VIRUS ANTIBODY 891 characterization of rabies inhibitors is necessary to elucidate the mechanisms and to minimize the loss of NT antibody produced by the treatment. The present study also showed that HI antibody titres obtained with intact goose erythrocytes are approximately 40-60 times lower than NT antibody titres in hyperimmune sera (Table 5) and, on average, 14 times lower in sera from immunized subjects (Table 8). However, the bromelain-treated goose erythrocytes were much more susceptible to haemag- glutination by rabies virus than were untreated erythrocytes; HI titres obtained with bromelain- treated erythrocytes were approximately 3-6 times higher than those obtained using current techniques (Tables 7 and 8). In addition, the removal of non-specific inhibitors from low dilutions of serum permitted the measure- ment of low levels of antibodies below a 1:20 dilution of the serum (Table 8). Since only a few materials were tested in the present study, further studies are being carried out to obtain additional data with a large number of materials. It is also of interest to examine whether antigenic differences between virus strains can be detected by the proposed HI procedures, and how these affect the serum HI antibody titres. The present study has demonstrated that the HI test might be useful in seroepidemiological studies of rabies. The techniques described here might also be used to remove non-specific inhibitors of other viruses that cannot easily be removed by currently available methods. ACKNOWLEDGEMENTS This research was supported in part by a grant from Yakult Honsha Co. Ltd, Tokyo, Japan. RESUME ACCROISSEMENT DE LA SENSIBILITE DE L'EPREUVE D'INHIBITION DE L'HEMAGGLUTINATION EN VUE DE LA RECHERCHE DES ANTICORPS ANTIRABIQUES La presence dans les serums humains et animaux d'inhi- biteurs non specifiques de l'hemagglutinine du virus de la rage et la faible sensibilite de 1'epreuve d'inhibition de l'hemagglutination (IH) pour la recherche des anticorps antirabiques, sont les deux principales raisons qui font que cette epreuve n'est guere utilisee pour l'etude de la rage. Le pr6sent rapport decrit une nouvelle methode permettant d'eliminer ces inhibiteurs non specifiques. Le traitement pendant 30 min, a la temperature ambiante, de serum dilu6 a 1:2 par un volume 6gal d'acide silicique colloidal a 50 g/litre (Aerosil), a permis de ramener a un niveau indecelable les inhibiteurs seriques, chez l'homme, la souris et le chien, a la dilution de 1:4 utilisee au depart de l'epreuve. Dans le cas du s6rum humain, il a fallu faire preceder le traitement a l'Aerosil par un traitement a l'acetone pour eliminer entiere- ment les inhibiteurs non sp6cifiques. Ce traitement entraline certes la perte d'une partie des anticorps neutralisants contenus dans le serum, mais dans une proportion qui reste sensiblement la meme qu'avec le traitement actuel, par le kaolin. La sensibilite de l'6preuve a e renforcee par l'emploi d'erythrocytes d'oie traites par le Bromelain. Apres ce traitement, les erythrocytes se sont reveles beaucoup plus sensibles au pouvoir agglutinant du virus de la rage. Les titres IH obtenus avec ces erythrocytes dans le serums traites a l'Aerosil etaient 3-6 fois plus 6leves que les titres obtenus avec des erythrocytes non traites dans un. serum traite au kaolin. De faibles titres d'anticorps ont pu etre mesures avec une plus grande sensibilite dans le serum de sujets vaccines quand on a associe le traitement du serum par l'Aerosil et l'emploi d'erythrocytes d'oie trait6s par le Bromelain. REFERENCES 1. HALONEN, P. E. ET AL. Hemagglutinin of rabies and some other bullet-shaped viruses. Proceedings of the Society for Experimental Biology and Medicine, 127: 1037-1042 (1968). 2. HALONEN, P. E. ET AL. Non-specific serum inhibitors of activity of hemagglutinins of rabies and vesicular stomatitis viruses. Journal ofgeneral virology, 22: 309- 318 (1974). 892 K. MANNEN ET AL. 3. SUZUKI, M. ET AL. Interaction of non-specific inhibitor and rabies virus hemagglutinin. Journal of general virology, 36: 31-39 (1977). 4. HALONEN, P. Hemagglutinin of rabies virus. In: Baer, G. M., ed., The natural history of rabies, New York, Academic Press, 1975, Vol. 1, pp. 103-113. 5. STEPHAN, V. W. & ROKA, L. Absorption von Lipoproteiden. Zeitschrift fur klinische Chemie und klinische Biochemie, 6: 186-190 (1968). 6. MIFUNE, K. ET AL. A mouse model for the patho- genesis and postexposure prophylaxis of rabies. Microbiology and immunology, 24: 835-845 (1980). 7. CLARKE, D. H. & CASALS, J. Techniques for hemag- glutination and hemagglutination inhibition with arthropod-borne viruses. Americaljournal of tropical medicine and hygiene, 7: 561-573 (1958). 8. KUWERT, E. Haemagglutination and haemagglutin- ation-inhibition tests. In: Kaplan, M. & Koprowski, H. Laboratory techniques in rabies, 3rd ed. Geneva, World Health Organization, 1973, pp. 135-146. 9. BUCKLEY, S. M. & TIGNOR, G. H. Plaque assay for rabies serogroup viruses in Vero cells. Journal of clinical microbiology, 1: 241-242 (1975). 10. SMITH, A. L. ET AL. Isolation and assay of rabies serogroup viruses in CER cells. Intervirology, 8: 92-99 (1977). 11. SEKINE, N. & YOSHINO, K. Inhibitors against rabies virus present in normal rabbit sera. Archiv fur die gesamte Virusforschung, 45: 89-98 (1974). 12. SMITH, J. S. ET AL. A rapid reproducible test for deter- mining rabies neutralizing antibody. Bulletin of the World Health Organization, 48: 535-541 (1973). 13. MANNEN, K. ET AL. pH-dependent hemolysis and cell fusion of rhabdoviruses. Microbiology and immunology, 26: 1035-1043 (1982). 14. MIFUNE, K. ET AL. Hemolysis and cell fusion by rhabdoviruses. FEBS letters, 137: 293-297 (1982).