Zbl. Bakt. Hyg. A 259, 268-274 (1985)

Cross Protection of Mice Against Different Rabies Virus Isolates

O. P. LARGHI and A. M. O . DfAZ

PAHOIWHO, Pan American Zoonoses Center, P. O. Box 3092, 1000 Buenos Aires,Argentina

With 1 Figure· Received May 17, 1984 · Accepted October 17, 1984

Abstract

In an attempt to identify "atypical" strains which could account for vaccination failures,10 street, one intermediate (DR19) and 4 fixed rabies virus isolates from men, cattle, dogs,cats, mongoose and vampire bats in five countries (Argentina, Brazil, Chile, Cuba, andFrance) were studied by cross-protection tests in mice.

For the purpose of this study, any virus that killed more than 20 % of the vaccinated micechallenged with that virus was considered "atypical". When the suckling mouse brain rabiesvaccine was used, two " atypical" isolate s were found : one, from a human case in Chile (91,125 mouse passages) and the other, from a vampire bat in Brazil (DRI9, 22 mouse passa­ges). However, when mice immunized with a cell culture vaccine (PV-BHK) of a higherantigenic value than the brain vaccine, were challenged with those same isolates, mortalitywas below 20 %.

The fact that these two isolates killed enough vaccina ted mice to be conside red "atypical"could be related to antigenic differences between these viruses and tho se included in thevaccine. However, since this mortality was observed only in the mice immunized with thevaccine with a lower antigenic value, reasons are given why it could be attributed to otherbiological characteristics of those strains than antigenic differences . Causes for vaccinesfailures other than immunological differences of rabies virus strains are also analyzed anddiscussed.

Zusammenfassung

Bei einem Versuch zur Identifizierung "atypischer" Starnme, die moglicherweise fiir Fallevon Impfversagen verantwortlich waren, wurden 10 Isolate von Stralsenvirus, eine Zwi ­schenform (DR 19) und 4 Isolate von Virus fixe von Menschen, Rindern, Hunden, Katzen ,Mungos und Vampir-Fledermausen aus 5 Landern (Argentinien, Brasilien, Chile, Kuba undFrankreich) mittels Priifung der Kreuzirnmunitat an Mausen untersucht.

Fiir die Zwecke der Untersuchung galt jedes Virus, das mehr als 20 % der geimpftenMause bei der Immunitatsprufung totete, als "atypisch". Bei Verwendung von Babymaus­hirn-Tollwutimpfstoff wurden zwei "atypische" Isolate festgestellt. Eines stammte von einermenschlichen Tollwuterkrankung in Chile (91, 125 Mausepassagen) und das andere voneiner Vampirfledermaus in Brasilien (DR 19, 22 Mausepassagen). Bei der Imrnunitatsprii­fung mit gleichen Isolaten an Mausen, die mit einem Zellkulturimpfstoff (PV-BHK) von

Cross Protection of Mice Against Different Rabies Virus Isolates 269

hoherern Antigenwert als der Hirnimpfstoff immunisiert worden waren, lag die Morralitatunter 20 %.

Daf diese zwei Isolate so viele geimpfte Manse toteten, daf sic als "atypisch" anzusehenwaren, konnte auf Unterschiede bei den Antigenen dieser Viren und der Impfviren zuruckge­fuhrt werden. Da jedoch diese hohe Mortalitat nur bei den Mausen beobachtet wurde, diemit dem Impfstoff von geringerem Antigenwert immunisiert worden waren, werden Griindeangefiihrt, warum diese anderen biologischen Eigenschaften dieser Stamme als ihrenAntigenunterschieden zugeschrieben werden kann. Die Griinde fur Impfversager, die nichtin den immunologischen Unterschieden der Tollwutvirusstarnme liegen, werden ebenfallsanalysiert und diskutiert.

Introduction

The causes of rabies vaccination failures are of primary interest to public healthworkers. After the development of rabies-specific monoclonal antibodies (19), it waspossible to better differentiate the four serotypes of the Lyssavirus genus, as well asmany isolates of rabies serotype 1 (7, 18). These findings suggested the possibility thatvaccine failures could be due to antigenic differences between the fixed virus(es) in thevaccines and the street viruses prevailing in a given area (7, 20).

To evaluate this hypothesis, a project was developed in seven Latin American count­ries to study laboratory and street rabies virus strains (11,13). Our participation in theproject consisted in studying 15 rabies virus isolates from four Latin American count­ries and France with which we challenged mice immunized with two different rabiesvaccines. The results obtained in cross-protection tests in mice are presented in thereport.

Material and Methods

Rabies isolates. Ten street, one intermediate (DR19) (5) and four fixed isolates were used.The CV5 strain was also included in each test. The species from which the viruses wereisolated, their geographic origin and the number of mouse passages are presented in Table 1.

Vaccines. Two types of vaccines were used: one was prepared in suckling mouse brain(5MB) at a 1 % tissue concentration using strains CV5, 51 and 91 (6), and the other wasproduced in cell cultures (PV-BHK)as described elsewhere (9) using the Pasteur strain (PV).Antigenic values (AV) of 1.76 for 5MB and 3.68 for PV-BHK vaccines were obtainedaccording to the NIH test, as described before (15).

Mice. Three to four week-old mice were used for the cross-protection tests.Cross-protection tests. Two groups of 30 mice each were immmunized on two consecu­

tive weeks with 5MB and PV-BHK vaccines by intraperitoneal inoculation of 0.5 ml ofvaccine. A third group of 30 mice was not vaccinated . One week after the second dose, 10animals from each vaccinated group and 10 control mice were challenged intracerebrallywith each virus isolate at dilutions prepared to contain 1, 10 and 100 LDlOo per 0.3 m!.

Three cross-protection tests were performed with the isolates listed on Table 1 (except PVadapted to BHK cells). Firstly, mice were immunized with 5MB diluted 1 : 5 and PV-BHKdiluted 1: 25. The isolates which killed more than 20 % of the vaccinated mice were usedsubsequently to challenge mice vaccinated with undiluted 5MB, and with PV-BHK diluted1: 5. Following this, other groups of mice were vaccinated with undiluted 5MB and PV­BHK, and challenged with the isolates that had killed more than 20 % of the vaccinatedmice in the preceding test, as well as with PV adapted to BHK cells.

Any isolate killing more than 20 % of the vaccinated mice at any challenge virus dilutionwas arbitrarily considered "atypical".

270 O.P.Larghiand A.M.O.Diaz

Dog Goat Dog CotMendoza Catamarca Moron Buenos AiresArgentino Argentino Argentino Argentino 5MB: 1'5

roo pv- BHK: 1'25

~ ~~80~ :Unvaccinated controls

60 - : 5MB Vaccinees (AV: 1.73)40

0----0 :PV- BHK Voccinees (AV:3.68)

20

aYom. Bot Dog Dog Dog Mongoose Catt'eCorrientes Moreno Merlo Tucumcin (PV - rabbit)Argent ino Argentino Argentino Argentino Cuba Fronce

*-

\ \ \ <, • • •>~ 80-J« 60 5MB : " ,I-0: PV- BHK: 1'50 4 0~

20

• .. ..a

Cott le Vam.Bot Human Dog Catt'e Cottle(DRI9) (9') (5 1) CPV-BHK) (CVS)

Argentino Brazil Chile Chile France France

100~ t---+-.-.t • • • • • •~

BO

~60

~5MB: 1'1

PV-BHK:I "

40

20

a0 ~ 0 ~ 0 ~ 0 ~ 0 ~ - 0 ~~ ~ 0

~ ~ ~

CHALLENGE DOSES LD ,OO

Fig. 1. Mortality of vaccinated mice challenged with different rabies virus isolates.

Results

Fig. 1 shows the percentages of mortality in mice immun ized with the highest vacci­ne concentration used for each of the isolates included in Table 1.

The following isolates killed more than 20 % of the mice vaccinated with 5MB 1 : 5or PV-BHK 1 : 25: Cattl e (CV5) France; Vampire Bat (DRI 9) from Brazil; Human (91)and Dog (51) from Chile; Mongoose from Cuba; Cattle from France, and Cattle(Balcarce), Vampire Bat (Corrientes), Dog (Merlo), Dog (Moreno), and Dog (Tucu­man) all from Argentina .

Of the above isolates, Cattle (CV5) France, Vampire Bat (DRI 9) Hum an (91), Dog(51) and Cattle (Balcarce), killed more than 20 % of the mice vaccinated with und ilu­ted 5MB or PV-BHK 1: 5.

Cross Protection of Mice Against Different Rabies Virus Isolates 271

Table 1. Fixed, intermediate, and animal street rabies virus

Country Province Locality Species Provided N° mouseby brain

passages

Chile Dog (51) 126Chile Human (91) 125France Cattle (Pv-rabbit) 0France Cattle (PV-BHK) 0France Cattle (CVS) 33Brazil Vamp.bat (DRI9) 22Argentina Buenos Aires Moreno Cat 1Argentina Buenos Aires Balcarce Cattle 1Argentina Buenos Aires Merlo Dog 1Argentina Buenos Aires Moreno Dog 1Argentina Buenos Aires Moron Dog 1Argentina Mendoza Dog Jose C. Vargas 1Argentina Catamarca Cat Clelia Riera 1Argentina Tucuman Dog C. Ch. de Alonso 1Argentina Corrientes Vampire bat 1Cuba Mongoose 2

Only two isolates, Vampire Bat (DRI9) and Human (91) acted as " typical" in miceimmunized with undil uted 5MB vaccine (AV = 1.76), but not in those immunized withundi luted PV-BHK(AV = 3.68).

Discussion

The mortality in vaccinated mice challenged with the rabies virus isolates suggests itcould be associa ted with the antigenic potency of the vaccines used. When immunizedwith the PV-BHK vaccine (AV = 3.68), mice resisted challenge with two virus isolatesthat killed over 20 % of those immunized with a 5MB vaccine (AV = 1.76) . Also, whenbot h vaccines were used at dilutions of 1 : 5 or more, expos ure of mice to the same andother virus isolates from Argentina, Brazil, Chile, Cuba and France resulted in anincreased mortality. Our results would agree with studies (17,20) in which three out ofseven and two out of six street isolates, respectively, killed more than 20 % of miceimmunized with vaccines produced with the Pasteur or Pitman-Moore strains of rabiesvirus, the potency of which was not reported. Antigenic differences between the streetvirus isolates obtained from different sources, and the fixed virus immunogens (7),would not therefore explain the vaccina tion failures observed in this expe rimentalmodel.

It is of interest to note that the two viral isolates (DR 19 and 91) that causedmortality in mice immunized with the undil uted 5MB vaccine were not street viruses,and were at the 22nd and 125th brain passages . The possibility that failures of thevaccine to confer protection agains t these viruses resulted from their ada ptation toreplicate more readily in mice does not seem compatible with the observation thatCattle CV5 and Dog 51 isolates, which did not kill immunized mice, were at the 33rdand 126th mouse passages, respectively.

272 O.P.Larghi and A.M .O .Diaz

Evidence obtained in previous studies (2) suggests that variations in the number andlocation of the antibody binding sites exposed in the outer envelope of street virusesoccur during the course of their adaptation and following transfer to a new host. Onthis basis, the present observation that immunized mice were not killed by the CattleCVS and Dog 51 isolates may be interpreted in terms of their outer envelopes contai­ning a larger proportion of the antigens common to the fixed viruses in the vaccinesthan of the antigens of DR19 and 91 isolates. Therefore, more of the latter virionswould survive and continue to replicate in the absence of an adequate immune respon­se to the vaccine, while those of the former isolates would be more readily susceptibleto neutralization.

The possibility that the surface antigens in the DR19 isolate may differ quantitative­ly or qualitatively from those in the fixed viruses in the vaccines cannot therefore bedisregarded solely on the basis of the present findings. This is supported by the fact thatthe Mongoose isolate from Cuba, which did not kill enough mice to be considered"atypical", had antigenic differences with CVS and another street rabies virus strain, asdetected by counterimmunoelectrophoresis tests with antisera absorbed with the hete­rologous strain and by crossed neutralization tests (1). Although Mongoose virusantigens shared by CVSgradually increase in concentration during adaptation to a newhost, quantitative differences in the common antigen(s) persist between both viruses,even after the street virus has been fixed (2). In any case, the fact that mice survivedchallenge with DR19 when immunized with the vaccine with higher potency confirmsthe importance of high antigenic values in immunogens .

The observation that the Cattle CVS and Dog 51 did not kill immunized micesuggests that these isolates do not differ significantly in their surface antigen composi­tion and distribution from the fixed viruses in the vaccines. This would be in agreementwith previous studies in which no antigenic differences were detected between the twofixed viruses CVS and 91 (1). Accordingly, the possibility that the mortality observedin mice immunized with the undiluted lower-potency vaccine in the present study maybe partly related to the biological characteristics of isolate 91 (virulence, invasiveness)cannot be disregarded. The biological characteristics of isolates 91 and CVS and theirimmunogenic differences are reasons for inclusion of isolate 91 in 5MB vaccines (6).

Vaccine potency and stability, conditions of storage, individual variations in hostsusceptibility, size of the inoculum, site of exposure, invasiveness, virulence of theinfecting strain, and/or antigenic differences may be involved in vaccine failures.

The results obtained in the present study suggest that differences in the antigenicpotency of the vaccines were associated with mortality in mice when challenged withthe Vampire Bat (DR19) and Human (91) isolates. This possibility is in agreement withthe fact that vaccines of adequate potency, prepared with a strain isolated a centuryago (12), have been effective in rabies control programs throughout the world (3,6,14). Vaccines of high potency are most likely to protect against strains with antigenicdifferences, and would also provide longer duration-of-immunity in dogs (10, 16),cattle (4, 8) and other species.

In general, the quality control of vaccines should be performed on samples obtainedbefore licensing. This control should also be performed on vaccine samples collectedfrom those being used in the field (8). Cases of vaccine failures in humans and animalsshould therefore be investigated through epidemiological studies. The isolation of theinfecting virus for laboratory studies, including monoclonal antibodies, should beattempted.

Cross Protection of Mice Against Different Rabies Virus Isolates 273

Acknowledgements. We appreciate the review of the manuscript by V. Varela-Diaz andH. Koprowski, the help in its writing by N. Bonomini and the technical assistance of j. C.Areitio, A . E. Nebel and G. Perdomo .

References

1. Diaz, A. M. and V. M. Varela-Diaz: Detection of antigenic differences among street andfixed rabies virus strains by the counter immunoelectrophoresis test. Zbl. Bakt. Hyg., 1.Abt. Orig. A 236 (1976) 185-190

2. Diaz, A . M. and V. M. Yareia-Diaz : Persistence and variation of mangosta street rabiesvirus antigens during adaptation into mice. Zbl. Bakt. Hyg., 1. Abr. Orig . B 171 (1980)73-78

3. Fermi, c.: Uber die Immunisierung gegen Wutkrankheit. Z. Hyg. Infektionskr. 58(1908) 233

4. Fuenzalida, E., P. N. Acha, P. Atanasiu, O. Largbi, and B. Szyfres: Rabies immunity invaccinated cattle. Ann. Meet. U.S. Anim Hlth Ass. 73 (1969) 307-322

5. Fuenzalida, E. and O. P. Larghi: Caracterfsticas de una cepa de virus rabico aislada delcerebro de Desmodus rotundus. BioI. Ofic, Sanit. Panamer. 73 (1972) 93-99

6. Fuenzalida, E. and P. Palacios: Un metodo mejorado para la preparacion de vacunaantirabica, Inst, Bact. Chile 8 (1955) 3-10

7. Koprowski, H. and T. Wiktor: Monoclonal antibodies against rabies virus. In: Mono­clonal Antibodies, pp. 335-351, Eds. R. H. Kennett, T . j. McKearn, and K. B. Bechtol.Plenum Press, London (1980)

8. Largbi, O. P.: Tendencias recientes y desarrollo de vacunas contra la rabia en AmericaLatina. In: Primeira Conferencia Internacional sobre Impactos das Doen..as Virais, noDesenvolvimento dos Paises Latino-Americanos a da regiao do Caribe, Rio de Janeiro,1982, Trabalhos, pp. 354-367. Instituto Oswaldo Cruz, Rio de Janeiro (1983)

9. Largbi, O. P., V. L. Savy, E. A. Nebel, and A. Rodriguez: Ethylenimine inactivatedrabies vaccines of tissue culture origin. J . Clin. Microbiol. 3 (1976) 26-33

10. Largbi, O. P., V. L. Savy, A. E. Nebel, and A. Rodriguez: Vacuna anrirrabica inact ivadacon etilenimina: Duracion de immunidad en perros . Rev. Arg. Microbiol. 11 (1979)102-107

11. Meeting of the study group on rabies vaccine and potency test. Pan American HealthOrganization, Washington DC (1980)

12. Pasteur, L.: Methode pour prevenir la rage apres morsure. C. R. Acad. Sci. (Paris) 101(1885) 765

13. Second meeting of the working group on immunological differences between street virusstrains and production of rabies vaccines. Pan American Zoonoses Center (PAHO/WHO), Ramos Mejia/Argentina (1980)

14. Semple, D.: On the nature of rabies and antirabic treatment. Brit. med. J. 2 (1919) 33315. Sikes, R. K. and O. P. Larghi: Purified rabies vaccine. Development and comparison of

potency and safety with two human rabies vaccines. J. Immunol. 99 (1967) 545-55316. Sikes, R. K., G. V. Peacock, P. Acha, R. Arko, and R. Dierks : Rabies Vaccines: Dura­

tion of immunity study in dogs. J. Amer. vet. Med. Ass. 159 (1971) 1491-149917. Sureau, P., P. Rollin, and T.]. Wiktor: Epidemiologic analysis of antigenic variations of

street rabies virus: Detection by monoclonal antibodies. Amer. J. Epidem. 117 (1983)605-609

18. Wiktor, T.]., A. Plamand, and H. Koprowski: Use of monoclonal antibodies in diagno­sis of rabies virus infection and differentiation of rabies and rabies-related viruses. J.Virol. Meth . 1 (1980) 33-46

19. Wiktor, T.]. and H. Koprowski: Monoclonal antibodies against rabies virus produced

274 O.P.Larghi and A.M.O.Diaz

by somatic cell hybridation: Detection of antigenic variants . Proc. nat . Acad. Sci.(Wash.) 75 (1978) 3938-3942

20. Wiktor, T. J. and H. Koprowski: Antigenic variants of rabies virus. Do present vaccinesprotect all exposed persons? Rabies Inf. Exch. 2 (1980) 23-25

Dr. O. P. Larghi, PAHOIWHO, Pan American Zoonoses Center, P.O. Box 3092,1000Buenos Aires, Argentina