Some observations on hamster-derived human infection with lymphocytic choriomeningitis virus*

Some observations on hamster-derived human infection with lymphocytic choriomeningitis virus* J. HOTCHIN,1 W. KINCH,2 & E. SIKORA2 The sequence of events leading to the diagnosis of recent outbreaks of hamster-borne lymphocytic choriomeningitis (LCM) in New York State is briefly reviewed. Some relevant experiments on persistent LCM infection of hamsters are described, including measure- ments of complement levels in the affected animals. Evidence is discussed which suggests that complement has an important role in eliminating free virus in the circulation by participating in virus neutralization. During the past several years there have been many cases in man of lymphocytic choriomeningitis (LCM) which originated from hamsters with inap- parent persistent, and sometimes clinical, LCM virus infection. Most of these cases remained undiagnosed for 1 or 2 years; then, as a result of a recent (1974) severe hospital outbreak, definite diagnoses were made in retrospect (1). Until that time LCM was not suspected and leptospirosis was presumed. LCM was only considered when by chance the case his- tories came to the attention of Dr David Axelrod at this laboratory, who suggested their similarity to the laboratory-associated hamster-to-man LCM out- break reported by Baum et al. (2). The scientific and medical publicity given to the 1974 hospital outbreak alerted many physicians and the public to the ham- ster danger, and as a direct result many more cases in the general population of New York State were identified as having been acquired from hamsters bought as pets. Thus in tracing the epidemiological events, two "infections " can be envisaged as spreading through the population. The first was the hamster-to-human spread ofLCM virus, and the second, which occurred after a considerable lag, was the human-to-human spread of the knowledge concerning LCM and ham- sters. Conceivably, if the laboratory outbreak (2) had not been detected and publicized, the New York * From the Division of Laboratories and Research, New York State Department of Health, Albany, NY, USA. 1 Assistant Director, Virus Laboratories. ' Senior Bacteriologist. State outbreak could have been entirely missed. One wonders how long the hamster LCM infection had in fact been present and whether other similar, but undiagnosed, outbreaks of LCM have happened in the past. Interest in the outbreak has been quite intense and several overlapping publications have resulted (3-8). In view of this interest and the recurring problem of the detection of hamster LCM to render animal colonies safe, some observations on hamster LCM made in this laboratory will be presented. These results are not intended as definitive but as prelimin- ary data bearing on the general problem. Many of the experiments were carried out to see whether per- sistently LCM-infected hamsters resulted from neo- natal inoculation; hence in several respects exhaustive tests were not made. The results are summarized in Tables 1 and 2. Hamsters were used from a strain reared in our laboratory (referred to as " Albany ") and from one obtained from a dealer whose stock, referred to as " golden ", had been shown previously (but not currently) to carry LCM. The LM4 variant of LCM virus (9) was used for inoculation, since it readily causes persistent infection of newborn mice. The uninoculated mothers of newborn " Albany " hamsters that received intraperitoneal (IP) inocu- lations of LCM usually became infected and showed specific antibody by fluorescent antibody (FA) as- say (10, 11) within 30 days (Table 1). Only 1 of 6 breeders tested at that time showed viraemia and this animal was also FA-positive. It follows that neo- natally IP inoculated hamsters are readily infectious 3384 - 561 BULL. WORLD HEALTH ORGAN., Vol. 52, 1975 J. HOTCHIN ET AL. Table 1. Virus LCM virus and antibody status of "Albany " strain hamsters inoculated with Type of No. of IV1 virus CYclophos- LCM viraemia hamster aNimals inoculum phamide b (mice + FA post-infectionhamster n route endotoxin) neto Breeder a 2 - - NT NT 180 1 P - NT NT 180 1 lP - - + 30 3 - - - 1/3 + 30 1 lP - + + 30 1 IP - - + 30 Newborn 3 P - - NT 180 5 SC - - NT 120 1 IP - + NT 9 4 - - - - 30 6 IP - + + 30 4 IP - 3/4 + c NT 42 50-day-old 1 SC + + NT 37 1 SC + - NT 58 1 - + - NT 37 a Inoculum was given to infants nursed. b 150 mg/kg of body weight given intraperitoneally on days 0 and 2 post-infection. CAll these sera were negative for virus by plaque assay. Table 2. Virus and antibody status of " golden - hamsters inoculated with LCM virus Age at No. of LM4 virus Cyclophos- LCM viraemia Viraemia test: inoculation animals inoculation phamideb (mice + day(days) a route phmendotoxin) post-infection -2 1 SC (breeder) - - 4 0 4 IP - 3/4 + 28 0 9 IP - - 120 30 1 IP - + 14 30 4 IP + + 14 30 4 IP - + 14 30 2 IP + 1/2 + 90 30 4 IP - 2/4 + 90 Adult breeder 3 via infant - - 120 a Infants only were inoculated after birth. b 150 mg/kg of body weight given intraperitoneally on day 2 post-infection. 562 HAMSTER-DERIVED HUMAN INFECTION to their mothers, possibly via urine; more important, the presence of FA in a hamster does not mean that the animal is " safe ", since viraemia can coexist with FA antibody. This is not surprising, since the same holds true for mice (12). LCM FA possesses no virus-neutralizing power and conveys no immun- ity to the infection. One breeder, whose newborn were not inoculated, was FA-positive. Of 11 neo- natally inoculated hamsters, 10 were viraemic when tested 9-42 days post-inoculation. In the one group of animals tested by plaque assay, the viraemia was not detectable by that method, so either the titre was too low or the virus had lost its ability to give visible plaques. In a litter tested for both virus and FA, all 6 animals had both viraemia and FA 30 days post-inoculation. The presence of LCM-specific FA in a hamster colony must be regarded as a danger sign and not as a sign of immunity to LCM. One of 2 animals inoculated when 50 days old and then given 2 injections of cyclophosphamide showed virus 37 days later. The other (negative) animal was tested only at 58 days post-inoculation. Neither was tested for FA. Similar attempts to induce LCM virus persistence in young " golden" hamsters (Table 2) appeared to indicate that 30-day animals were more susceptible to persistence than " Albany" breeders, since 3 of 6 animals were viraemic on the 90th day post- inoculation whether or not cyclophosphamide was given on day 2 post-inoculation. It is possible but not certain that hamsters of this strain more readily remain LCM carriers, although none of 9 neonatally- infected animals was viraemic at day 120 post- inoculation. The finding that both FA and virus were present at the same time in hamster sera suggested that complement (C)-fixing antibody might also be pre- sent. Tests showed that many of the sera from animals with active or recent persistent infection were anticomplementary. Further tests were there- fore made to determine whether the addition of C (guinea-pig) altered the titre of virus in persistently LCM-infected hamster sera. Sera were selected from the previous experiments and used to make 1/3, 1/6 and 1/9 dilutions in Eagle's tissue culture medium plus 10% fetal bovine serum. A duplicate set of the same dilutions was prepared containing 1/3 volume of 30% fresh guinea-pig serum (for C) in tissue culture medium. All tubes were incubated at 37°C in a water bath for 30 min, then 0.1 ml of each was used for LCM plaque assay (9). The results (Table 3) show that in all cases the addition of Table 3. Effect of 30-min incubation with complement on the LCM titre of persistently infected hamster serum Plaque count Plaque Hamster No. Serum Serum + Serum Titre ratio Titre ratio reduction bySerum Serum + dilution (% complement alone a complement (%) 4 > 200 37 1/3 < 37/200 < 18.5 > 81.5 4 200 30 1/6 30/200 15 85 4 125 15 1/9 15/125 12 88 5 > 200 50 1/3 < 50/200 < 25 > 75 5 > 200 31 1/6 < 31/200 < 15.5 > 85 5 > 200 10 1/9 < 10/200 < 5 > 95 6 > 200 50 1/3 < 50/200 < 25 > 75 6 > 200 20 1/6 < 20/200 < 10 > 90 6 > 200 15 1/9 < 15/200 < 7.5 > 92.5 7 > 200 5 1/3 < 5/200 < 2.5 > 97.5 7 > 200 6 1/6 < 6/200 < 3 > 97 7 > 200 7 1/9 < 7/200 < 3.5 > 96.5 All sera were incubated for 30 min. 563 J. HOTCHIN ET AL. C greatly lowered the titre of circulating virus. The plaque reduction reached 95% and in some cases was clearly much greater, since the end-point titre of the serum without C was reached only in 1 of the 4 cases studied. In a further experiment, persistently infected hamster serum was incubated for 3 h at 37°C with and without added C, and the virus titre assayed by plaque count. The control (no C) showed a titre of 3 x 104/ml, and the serum with C showed no plaques in any dilution. Apparently the persist- ently infected hamsters were C-deficient; this was tested by duplicate C titrations (13) of one persist- ently infected hamster serum and one normal ham- ster serum. The optical density of haemolysis prod- ucts was 4.34 and 2.75 per ml (average= 3.54) for replicate aliquots of the LCM-infected hamster serum, compared to 41 for the control hamster serum. Normal hamsters contain 225 haemolytic units of C per ml (14); thus the infected hamster contained only 3.54/41 of this, or 19.5 units of C. The tendency for hamsters spontaneously to be- come persistently infected if contaminated with LCM virus might be related to their C level (225 units/ml versus 625 for guinea-pigs) (14). It would seem plausible to regard persistence of virus in hamsters as due in part to complement depletion. However, this depletion is probably a consequence of the per- sistence rather than a cause, since it is now well established that the eradication of LCM infection is mainly dependent upon the cellular immune response. The neutralization and removal of virus by complement-fixing antibody can probably be more properly seen as part of a scavenging process for the disease by-products, rather than a major therapeutic event. However, the exact role of C in LCM remains unclear. The very low level of this substance (only 10 units/ml) in mouse blood leads to the tempting hypothesis that the readily induced LCM persistence in mice could be related to this lack of C. However, the 10-unit C level is also shared by rats, which in our experience are much less susceptible to persistent infection with LCM. Our results have not shown murine neutralization of LCM virus to be much enhanced by the addition of C to the incubation mixture, as is the case with early human sera (15). These results must be re- garded as merely preliminary; the role of C in LCM and other chronic virus infections clearly deserves further study. The origin of the hamster LCM infections remains in doubt. While the hospital outbreak may have originated from LCM-infected tumour cells passed in the hamsters, the infection of hamsters from cer- tain pet dealers cannot be explained by this source. The origin of the LCM virus in the tumour cells also needs explaining. It seems likely that LCM could be transmitted without difficulty from wild mice to hamsters in animal farms from time to time, finally resulting in both tumour cell line and human infection. The only preventive available at pre- sent is close monitoring and supervision of animal farms. RtSUM1! QUELQUES OBSERVATIONS SUR LES CAS HUMAINS DE CHORIOMtNINGITE LYMPHOCYTAIRE TRANSMISE PAR LE HAMSTER Les auteurs de l'etude concluent que la detection de flambees recentes de choriomeningite lymphocytaire sur- venues dans l'Etat de New York s'explique en partie par la propagation du virus mais aussi par la prise de conscience que des hamsters pouvaient etre responsables de l'infection. Des virus de la choriomeningite lympho- cytaire ont ete inocules A la naissance A des hamsters dont on a ensuite suivi l'evolution. Chez beaucoup, l'infection a persist6 pendant 30 jours mais non pendant 120 jours. Certains animaux auxquels le virus avait e inocule A l'Age de 30 jours etaient encore viremiques 90 jours plus tard. I1 est apparu que le serum de ces animaux pouvait contenir A la fois des anticorps dece- lables par immunofluorescence et des virus. On a cons- tate que le taux de complement serique, mesure chez un> animal infecte, etait tomb6 A 19,5 unites, la valeur nor- male etant de 225 unites. Dans les cas ou du complement a ete ajout6 A des serums de hamsters viremiques avant incubation A 37°C pendant 30 minutes, le titre de virus a ete en moyenne d'au moins 88% inferieur au titre observe dans des serums temoins ne contenant pas de complement. Apres une incubation de 3 heures, le titre de virus est tombe A 0 dans des serums avec compl6ment alors qu'il etait de 3 x 104/ml dans des s6rums temoins egalement incub6s. Les auteurs pensent que l'epuisement du complement pourrait jouer un role dans la persistance du virus de la choriomeningite lymphocytaire. 564 HAMSTER-DERIVED HUMAN INFECTION 565 REFERENCES 1. HOTCHIN, J. ET AL. Lymphocytic choriomeningitis in a hamster colony causes infection of hospital per- sonnel. Science, 185: 1173-1174 (1974). 2. BAUM, S. G. ET AL. Endemic nonmeningitis lympho- cytic-choriomeningitis-virus infection. An outbreak in a population of laboratory personnel. New England journal of medicine, 247: 934-936 (1966). 3. BiGGAR, R. J. & DOUGLAS, R. G. Lymphocytic choriomeningitis associated with hamsters. Lancet, 1: 856-857 (1975). 4. BIGGAR, R. J. ET AL. Lymphocytic choriomeningitis outbreak associated with pet hamsters. Fifty-seven cases from New York State. Journal ofthe American Medical Association, 232: 494-500 (1975). 5. DEIBEL, R. ET AL. Lymphocytic choriomeningitis virus in man. Serologic evidence of association with pet hamsters. Journal of the American Medical Association, 232: 501-504 (1975). 6. HINMAN, A. R. ET AL. Outbreak of lymphocytic choriomeningitis virus infections in medical center personnel. American journal of epidemiology, 101: 103-110 (1975). 7. HIRSCH, M. S. ET AL. Lymphocytic-choriomeningitis- virus infection traced to a pet hamster. New England journal of medicine, 291: 610-612 (1974). 8. VANZEE, B. E. ET AL. Lymphocytic choriomeningitis in university hospital personnel. Clinical features. American journal of medicine, 58: 803-809 (1975). 9. HOTCHIN, J. ET AL. Lytic and turbid plaque type mutants of lymphocytic choriomeningitis virus as a cause of neurological disease or persistent infection. Infection and immunity, 4: 281-286 (1971). 10. COHEN, S. M. ET AL. Immunofluorescent detection of antibody to lymphocytic choriomeningitis virus in man. Journal ofimmunology, 96: 777-784 (1966). 11. HOTCHIN, J. & SIKORA, E. Laboratory diagnosis of lymphocytic choriomeningitis. Bulletin of the World Health Organization, 52: 555-559 (1975). 12. HOTCHIN, J. E. Persistent and slow virus infections. In: Virology monographs, Basle, Karger, 1971. 13. KENr, J. F. & FIFE, E. H. Precise standardization of reagents for complement fixation. American journal of tropical medicine and hygiene, 12: 103-116 (1963). 14. CARPENTER, P. L. In: Immunology and serology, 2d ed. Philadelphia & London, W. B. Saunders Co., 1965, p. 327. 15. HOTCHIN, J. & KINCH, W. Microplaque reduction: New assay for neutralizing antibody to lymphocytic choriomeningitis virus. Journal of infectious diseases, 131: 186-188 (1975).