Arbovirus studies in Luanda, Angola

Bull. Org. mond. Santi 93 9 74 Bull. Wld Hith Org. 1973, 49, 37 40 Arbovirus studies in Luanda, Angola 2. Virological and serological studies during an outbreak of dengue-like disease caused by the Chikungunya virus ARMINDO R. FILIPE1 & MANUEL R. PINTO 2 An outbreak ofdengue-like disease was observed in Luanda, Angola, at the end of 1970 and beginning of 1971. Chikunganya virus was isolated from the blood of a patient with typical symptoms of dengue and from a pool of Aedes aegypti mosquitos. A survey for antibodies to arboviruses in the sera ofpersons living in Luanda showed that the Chikun- gunya virus was indeed responsiblefor the outbreak. Thefact that this outbreak immediately preceded and continued concurrently with a yellow fever epidemic in Luanda shows that two arboviruses from different antigenic groups may circulate simultaneously in the same ecological area. The Chikungunya virus, which Ross isolated in Tanzania in 1956, has since been found in Zaire, Zambia, and South Africa (Taylor, 1967). In Angola, the occurrence of a dengue-like disease has been observed, but the etiology of the disease has never been demonstrated. However, serological studies car- ried out in 1960 by Kokernot et al. (1965) in Angola have shown that sera collected in the northern region of the territory had neutralizing antibodies to the Chikungunya virus. Osterrieth et al. (1961) also reported an epidemic of dengue-like disease caused by this virus. The epidemic coincided with an outbreak of yellow fever in Doruma, Zaire, near the borders of Sudan and the Central African Republic. During the last months of 1970 an outbreak of an infectious disease characterized by high fever, headache, and severe pains in the joints and muscles was observed in Luanda. The disease, known locally as Katolu T6lu (Kimbundu dialect for " break-bone disease") was observed before the yellow fever epi- demic that started in January 1971. Cases continued to be seen until the end of the yellow fever epidemic at the beginning of April, when the vector was eradi- cated as a result of the intensive antimosquito mea- sures taken by the public health services. * From the Institute of Hygiene and Tropical Medicine, Lisbon, Portugal. 1 Research Virologist. Professor of Bacteriology and Virology. Two strains of arbovirus were isolated-one from the blood of a person with a typical dengue-like syndrome, obtained at the beginning of the disease, and the other from A. aegypti mosquitos captured in Luanda. A study of their properties showed both strains to be of the Chikungunya virus. MATERIALS AND METHODS Blood from several acute cases of the dengue-like disease was collected and the serum was inoculated by the intracerebral and intraperitoneal routes into 1-2-day-old Swiss mice of the Charles River strain. The usual methods (Hammon & Work, 1964; Work, 1964) were used for recovering the virus from mosquitos. Fifteen pools of the mosquito species A. aegypti, Mansonia africana, M. uniformis, and Culex fatigans (Ribeiro, 1971) were captured in Luanda and its suburbs and were processed for virus isolation. A suspension in Hanks's solution with 0.4% of bovine serum albumin (Armour, Fraction V) was prepared from the triturated frozen mosquitos from each pool. The supernatant fluid of this suspen- sion was administered to baby mice in the form of intracerebral and intraperitoneal inoculations. When the mice showed signs of disease, they were killed and a 10% suspension of their brains was made in the same diluent for further passages. Immune sera to the isolated viruses were prepared by inoculating this suspension by the intraperitoneal route into adult mice at weekly intervals for 5-6 weeks. An attempt was made to extract haemagglutinins from 3084 - 37- A. R. FILIPE & M. R. PINTO the isolated strains by means of the sucrose-acetone method of Clarke & Casals (1958). The human sera used for the survey for antibodies to arboviruses of group A were those collected in Luanda for the serological studies carried out during the yellow fever epidemic of 1971 (Pinto & Filipe, 1973). In all, 589 sera were studied by the haemag- glutination (HI) test, following the techniques de- scribed by Clarke & Casals (1958), using the micro- test with 4-8 haemagglutinating units and the fol- lowing antigens from group A: Sindbis (Ar 339), CChikungunya (Ross), and O'Nyong Nyong. In addition, 80 sera from persons suspected of having had the dengue-like disease were studied by the HI test. The 50 sera positive to the Chikungunya virus were further studied by the complement- fixation (CF) test with the Chikungunya antigen, also using the microtest. RESULTS Virological studies Serum from a patient with typical " Katolu TOlu ", collected 24 hours after the onset of the disease, yielded a virus (9-CT) pathogenic for baby mice. Some of the inoculated animals became sick and paralysed on the 4th or 5th day after inoculation of the serum; after the 4th passage, paralysis gen- erally set in on the 4th day. The supernatant fluid from a pool of 33 A. aegypti, inoculated by the intracerebral route into two litters of baby mice, yielded another virus (A-4) with patho- genic properties similar to those of the virus strain isolated from the blood of the above-mentioned patient. Both strains were highly neurotropic for suckling mice; they rarely killed adult mice, although some of these developed alopecia, as has occasionally been observed in adult mice inoculated with Chikun- gunya virus or O'Nyong Nyong virus. Both strains were sensitive to sodium deoxicholate, with a sensi- tivity index of 4.4 log for strain 9-CT and 5.5 log for strain A-4. It proved difficult to extract haemagglutinins from these strains; however, strain 9-CT yielded a haem- agglutinin that could be used for the HI test. The optimum pH was 6.0. The similarities between these two strains and the Chikungunya and O'Nyong Nyong reference viruses, as revealed by serological tests, are shown in Tables 1, 2, and 3. As a result of these similarities, it is dif- ficult to differentiate the four viruses. Table 1. HI titre a of the immune sera against the isolated strains and reference viruses Immune mouse Antigen serum against: ONN Chik. 9-CT A-4 strain 0 160 9-CT strain 0 80 160 O'Nyong Nyong 160 40 0 Chikungunya 0 160 80 a Reciprocal of serum titre. Table 2. Mouse neutralization test with the isolated virus Antiserum Virus Chik. ONN A-4 2.6 a 3.4 9-CT 2.6 3.5 a Log neutralization index. Table 3. Complement fixation tests with isolated and reference viruses a Antiserum Antigen A-4 9-CT Chik. ONN b A-4 16/16 16/32 16/16 8/16 9-CT 8/16 16/16 16/32 4/4 Chik. 32/8 16/32 16/32 16/16 ONN 16/8 8/8 8/8 32/128 a Reciprocals of serum dilution (numerator) and antigen dilution(denominator). b Immune reference serum. The neutralization indices of the acute and con- valescent serum samples obtained from the patient from whom the 9-CT strain was isolated are given in Table 4. This shows a significant rise in the homo- logous neutralizing antibody, which neutralized also the Chikungunya and O'Nyong Nyong viruses. Serological studies The results of the investigation of sera for anti- bodies to arboviruses of group A (Table 5) show that only the Chikungunya antigen elicited a serolo- gical response. The results of HI testing of 80 sera from patients suspected of having had the dengue- 38 ARBOVIRUS STUDIES IN ANGOLA: 2 Table 4. Mouse neutralization test with sera from a patient Virus Serum Chik. ONN 9-CT Acute 0 0.5 0.2 Convalescent 1.7 a 1.2 2.1 a Log neutralization index. Table 5. Number of sera reacting with group A viruses (HI test) a Positive to Serum No. Chikungunya No. % Luanda residents 310 20 6.4 Family contacts of yellow fever cases 113 20 17.6 Yellow fever patients and sus- pected cases 166 41 24.6 Total 589 81 13.7 a All results for the Sindbis virus were negative. like disease are shown in Tables 6 and 7. Of the 50 Chikungunya-positive sera subjected also to the CF test with Chikungunya antigen, only one was negative; another reacted at a low titre, and the others fixed complement at high titres. The results obtained with 20 of those 50 sera demonstrated that the infection by Chikungunya or an antigenically closely related virus was recent. Table 6. Reaction with group A viruses (H I test) of sera from 80 patients suspected of having had " KAtolu T6lu Serum Antigen Positive % Sindbis 0 0 Chikungunya 52 65.0 O'Nyong Nyong 1 1.3 Chik. + ONN 27 33.7 Total 80 100 Table 7. Titres observed against Chikungunya and O'Nyong Nyong viruses Positive for: Titres Chikungunya O'Nyong Nyong No. % No. % 1/20 22 42 21 40 1/40 15 29 7 14 1/80 11 21 0 0 1/160 4 8 0 0 Total 52 100 28 54 DISCUSSION It was not easy to identify the arbovirus strains isolated during the outbreak of dengue-like disease. The neutralization indices for these strains and the Chikungunya and O'Nyong Nyong viruses were similar, and CF testing with the homologous antisera did not allow the Chikungunya and O'Nyong Nyong viruses to be differentiated. However, HI tests of the homologous antisera showed higher antibody titres to the Chikungunya virus. These results are similar to those obtained by Williams et al. (1962) in their studies of the antigenic differentiation of the two viruses. The results of the serological survey carried out in Luanda helped to elucidate the etiology of the disease. It is usually difficult to differentiate anti- bodies to the Chikungunya and O'Nyong Nyong viruses, as Henderson et al. (1970) pointed out when discussing the results of a survey for antibodies to arboviruses, carried out in Africa. However, the interpretation of the serological data obtained dur- ing a study of an epidemic of dengue-like disease does not seem to present the same difficulties. The results presented in Table 7 show that HI antibodies to the Chikungunya virus were always found in higher proportions and titres than those against the O'Nyong Nyong virus. A study of sera of Luanda residents with the disease showed that 65% reacted only with the Chikungunya antigen, 33.7% with the Chikungunya and O'Nyong Nyong antigens, and only 1.3% with the O'Nyong Nyong antigen alone. This reactivity was to be expected because of the antigenic overlap between the two antigens. The results obtained in the study of sera positive to the Chikungunya antigen by the CF test indicated recent infection by this virus. 39 40 A. R. FILIPE & M. R. PINTO These data are supported by the epidemiological observations on the activity of the Chikungunya and O'Nyong Nyong viruses in Africa, which show that the Chikungunya virus has been endemic mainly in West Africa, whereas the O'Nyong Nyong virus has been endemic mainly in East Africa (Bres, 1970). The simultaneous or consecutive appearance of two antigenically unrelated arboviruses having the capacity to survive and circulate in the same ecolo- gical area has been already observed in Africa with the Chikungunya and yellow fever viruses (Osterrieth et al., 1961) and in Asia with the Chikungunya and dengue viruses (Myers & Carey, 1967). The same phenomenon has now been reported in Luanda in 1971. This simultaneous presence of two arboviruses from different antigenic groups may add new dimen- rions to the epidemiology of arbovirus diseases in stopical regions. Such epidemiological situations should be further investigated to elucidate their pos- sible implications for public health in tropical countries. ACKNOWLEDGEMENTS We are grateful to the authorities of the Public Health Service of Angola for their cooperation; the Director and staff of the Institute of Scientific Investigation, Angola (IICA), Luanda, for laboratory facilities; Mr J. Teixeira dos Santos, Seca e Cruz and Mrs F. Dias Flora (IICA) and Mrs C. R. Bettencourt and Mr C. Gongalves, Institute of Hygiene and Tropical Medicine, Lisbon, for their valuable technical assistance; Dr. H. Ribeiro, for classifying the mosquitos used in this study; and Dr N. Karabatsos, Yale Arbovirus Research Unit, Yale University School of Medicine, New Haven, Conn., USA, through whose courtesy the immune reference serum to the O'Nyong Nyong virus was supplied. These studies were supported in part by the Insti- tute of Advanced Culture (Instituto de Alta Cultura), Lisbon. RtSUMt tTUDES SUR LES ARBOVIRUS A LUANDA (ANGOLA): 2. ETUDES VIROLOGIQUES ET SEROLOGIQUES PENDANT UNE EPIDEMIE D'UNE MALADIE DE TYPE DENGUE CAUSEE PAR LE VIRUS CHIKUNGUNYA A la fin de 1970 et pendant le ler trimestre de 1971, on a observ6 a Luanda (Angola) plusieurs cas d'une maladie infectieuse benigne, caracterisee par une fievre elevee et de violentes arthralgies, qui a fait penser a une affection de type dengue. Cette epidemie a pr&ede et accompagn6 l'epidemie de fievre jaune qui a s6vi pendant les premiers mois de 1971. On a isole une souche de virus chikungunya a partir du sang d'un malade presentant la symptomatologie clinique classique de I'affection - design6e par la popu- lation locale par le nom de K2tolu T6lu (ROS bris6s ))). Une autre souche du meme virus a ete isolee & partir d'un lot d'Aedes aegypti captures A differents endroits de la ville de Luanda. Simultanement, on a proc6de a une enquete serolo- gique en utilisant les 589 serums preleves parmi la population de Luanda et de sa banlieue dans le cadre des investigations sur 1'epidemie de fievre jaune. Parmi ces serums, 81 (13,7%) renfermaient des anticorps anti- chikungunya. La presence simultan6e de deux arbovirus appartenant A des groupes antig6niques differents, capables de sur- vivre et de circuler dans le meme contexte ecologique, a de'jA e observee en Afrique et en Asie. Ce phenomene peut ouvrir de nouvelles perspectives pour l'etude epidemiologique des affections causees par les arbovirus dans les regions tropicales. REFERENCES Bres, P. (1970) Bull. Wld Hlth Org., 43, 233 Clarke, D. H. & Casals, J. (1958) Amer. J. Trop. Med. Hyg., 7, 561 Hammon, W. McD. & Work, T. D. (1964) In: Lennette, E. H. & Schmidt, N. J., ed., Diagnostic procedures for viral and rickettsial diseases, 3rd ed., New York, American Public Health Association, p. 268 Henderson, B. E. et al. (1970) Bull. WldHlth Org., 42,797 Kokernot, R. H. et al. (1965) Trans. roy. Soc. trop. Med. Hyg., 59, 563 Myers, R. M. & Carey, D. E. (1967) Science, 157, 1307 Osterrieth, P. et al. (1961) Ann. Soc. belge Med. trop., 41, 207 Pinto, M. R. & Filipe, A. R. (1973) Bull. Wld Hlth Org., 49, 31 Ribeiro, H. (1971) Rev. med. Angola, 13, 67 Taylor, R. M. (1967) Catalogue of arthropod-borne viruses of the world, Washington, D.C., US Govern- ment Printing Office (Public Health Service Publica- tion No. 1760) Williams, M. C. et al. (1962) Trans. roy. Soc. trop. Med. Hyg., 56, 166 Work, T. H. (1964) In: Lennette, E. H. & Schmidt, N. J., ed., Diagnostic procedures for viral and rickettsial diseases, 3rd ed., New York, American Public Health Association, p. 312