Arbovirus studies in Luanda, Angola

Bull. Org. mond. Sant) 1973, 49, 31-35 Bull. Wid Hith Org. Arbovirus studies in Luanda, Angola 1. Virological and serological studies during a yellow fever epidemic MANUEL R. PINTO1 & ARMINDO R. FILIPE2 A yellow fever epidemic broke out in Luanda, Angola, in January 1971 and was halted in less than 3 months by prompt mass vaccination and intensive antimosquito measures. Nine strains of yellow fever virus were isolated from hospitalized cases, but attempts to isolate the virus from Aedes aegypti mosquitos failed. A serological survey for antibodies to arboviruses in 589 sera from unvaccinated persons showed that two arboviruses were circulating in the epidemic area: theyellowfever virus and the Chikungunya virus responsible for an outbreak ofdengue-like disease that occurred shortly before the yellowfever epidemic. Yellow fever epidemics have not been reported in Angola since 1872 (Mora, 1933). Serological surveys for antibodies to arboviruses carried out in Angola in 1952 (Cambournac et al., 1955) and in 1960 (Kokernot et al., 1965) showed that the yellow fever virus was probably active in some northern areas of the country where the ecological conditions are similar to those found in the forest zone of the Guinea jungle biome. However, if any cases oc- curred, they went unnoticed in the absence of facili- ties for diagnosis. During the Luanda epidemic of 1971, 65 cases of yellow fever with 42 deaths were officially notified (Wkly epidem. Rec., 1972), but these numbers prob- ably do not reflect the extension of the epidemic as shown by the serological survey. Prompt mass vaccination and vector control by ULV spraying of malathion from aircraft (Ribeiro, 1971) inter- rupted transmission of the virus in the first week of April. The results of the virological and serological studies carried out during the epidemic are reported in this paper. MATERIALS AND METHODS Virological studies Samples of blood were taken from patients on the day of their admission to hospital. The serum was separated and kept frozen until it was inoculated * From the National School of Hygiene and Tropical Medicine, Lisbon, Portugal. Professor of Bacteriology and Virology. Research Virologist. into Swiss mice of the Charles River strain. Two litters of mice 1-2 days old were inoculated by the intracerebral and intraperitoneal routes with whole serum and with the serum diluted 1 : 10 in Hanks's solution with bovine serum albumin (Armour, Fraction V) at 0.4 %. A 10% suspension of the brains of diseased mice in the same diluent and a suspension of a liver fragment from a fatal case were used for the passages. Attempts were made to isolate virus from 15 pools of mosquitos of the following species: A. aegypti, Mansonia africana, M. uniformis, and Culex fatigans. The viruses isolated were studied by means of the techniques recommended by Hammon & Work (1964) and by Work (1964) for the study of arbo- viruses. The strains were characterized serologically with reference ascitic fluid anti-yellow fever virus, Asibi strain,3 and reference serum anti-yellow fever virus.4 The immune serum against the isolated viruses was prepared in adult mice that had received 5 or 6 weekly intraperitoneal injections of a 10% suspen- sion of mouse brain infected with the virus studied. The neutralization tests were carried out by inocu- lating infant mice by the intracerebral route with a mixture of the reference serum or ascitic fluid and an equal volume of each tenfold dilution of the virus preparations. The neutralization index expresses the difference between the titres of the virus in ' Supplied by Dr N. Karabatsos, Yale Arbovirus Research Unit, New Haven, Conn., USA. ' Supplied by Dr J. H. S. Gear, South African Institute for Medical Research, Johannesburg, South Africa. 3083 -31 M. R. PINTO & A. R. FILIPE Fig. 1. Map showing the localities in Luanda and its suburbs where Chikungunya virus (indicated by stars) and yellow fever virus (hatched area) were found. diluent and in the reference ascitic fluid or serum, calculated by Kiirber's method (Karber, 1931). Serological studies A total of 589 blood specimens were collected from groups of nonvaccinated residents of Luanda and its suburbs (Fig. 1) during the period February- May 1971; 113 of the specimens were obtained from family contacts of yellow fever cases and 166 from persons suspected of having had yellow fever. These sera were studied by the haemagglutination- inhibition test (HI) performed by the method of Clarke & Casals (1958) with 4-8 units of antigen and by the microtest. The following antigens were used: Sindbis (Ar 339), Chikungunya (Ross), O'Nyong Nyong, yellow fever (Asibi), West Nile (Eg 101), Zika, Ntaya, Wesselsbron, Banzi, dengue 1, dengue 2, and Bunyamwera. RESULTS Virological studies Nine strains of an encephalitogenic virus were isolated: 8 from the blood of yellow fever patients and 1 from a liver suspension from a fatal case. One strain (13 CT) was isolated from the blood of a patient with a dengue-like disease and another (52 FA) from a patient hospitalized with infectious hepatitis. Serological characterization showed all the strains to be yellow fever virus, with marked neurotropism and similar biological properties (Table 1), and antigenic analysis indicated that they were antigenically identical. All the sera prepared against the strains reacted by HI with yellow fever antigen, and a few of them with Ntaya antigen; no reaction to the other antigens was observed. All the strains cross-reacted among themselves with titres almost equivalent to those for the homologous strain. The baby mice inoculated with the serum from patients usually became sick after an incuba- tion period of 9-11 days and, after the first or second passage, all the strains had adapted to the mice, causing paralysis by the 5th day. The strain obtained from the liver fragment killed 2 of the 10 mice after 9 and 13 days, respectively. However, on the first passage the mice were paralysed by the 5th day. All the strains gave high infectious titres in infant mice and produced haemagglutinin that was easily extracted by the sucrose-acetone method. 32 ARBOVIRUS STUDIES IN ANGOLA: I Table 1. Biological properties of the virus strains isolated Isolated and reference virus strains Tests YFAsibi 13FA 14FA 15FA 27FA 43FA 47FA 48FA 52FA 13CT pas. 197 index of sensitivity to s.d.c. pH of haemagglutination + 4.7 log 4 log 5.1 log 5.6 log 4.5 log 6.5 log 4 log 4.5 log 5.2 log 6.2/6.4 6.2/6.4 6.2/6.4 6.2/6.4 6.2/6.4 6.2/6.4 6.2/6.4 6.2/6.4 6.2/6.4 6.2/6.4 neutralization test with reference serum a 7.8/4.3 5.8/1.6 5.4/1.0 7.1/1.5 7.2/4.1 4.8/1.0 6.0/1.5 6.4/4.1 5.5/1.5 5.6/3.5 neutralization index 3.5 log 4.2 log 4.4 log 5.6 log 3.1 log 3.8 log 4.5 log 2.3 log 4.0 log 2.1 log a LD 50 titre of the virus/LD 50 titre of the virus and reference serum mixtures. Table 2. Results of HI tests performed with sera collected in Luanda and suburbs Number of sera reacting with the antigens Antigen total positive % Group A 589 81 13.7 Group B 589 131 22.2 Bunyamwera 589 0 0 Inoculation into baby mice of the supernatant fluid of the macerate from several pools of mosquitos did not yield yellow fever virus. Serological studies Of the 589 sera studied, 22.2% had antibodies to the group B antigens used in the HI test and 13.7% had antibodies to the group A antigens (Table 2); 59.4% of those reacting with the group B antigens were positive to the yellow fever antigen and 40.6% were of the " multiple-infection " type (Table 3). Table 3. Number of sera reacting with group B viruses Antigen No. of sera % positive yellow fever 78 59.4 - multiple infections - 53 40.6 positive 131 22.2 negative 458 77.8 Total 589 100 Table 4. HI titres observed against yellow fever virus in primary infections Titre No. of positive sera 1/20 20 1/40 25 1/80 19 1/160 8 1/320 3 1/640 1 1/1280 2 Total 78 HI testing of sera from 113 family contacts of yellow fever patients showed that 45.1 % had anti- bodies to yellow fever antigen and 17.6% to the Chikungunya antigen; 166 sera from persons sus- pected of having had yellow fever showed corre- sponding percentages of 44.5 and 24.6. Of the 131 sera positive to the group B viruses, 78 showed reactions characteristic of primary infec- tion with yellow fever antigen (Table 4). DISCUSSION The occurrence of a yellow fever epidemic in Angola in 1971 shows how the virus may suddenly attack a territory that has been free from its activity for nearly a century, and calls attention to the need for more studies on the epidemiology of this disease in Africa, as was pointed out by BrMs (1970) and by the WHO Expert Committee on Yellow Fever (1971). 33 34 M. R. PINTO & A. R. FILIPE In 1970, the virus was active in countries as far away from Angola as Cameroon, Equatorial Guinea, Ghana, Nigeria, and Togo (Wkly epidem. Rec., 1972). Its appearance in Luanda points to the possible existence of endemic jungle foci where the virus may survive and from which it may eventually spread. The virological studies reported here showed that the 9 strains of yellow fever virus isolated in Luanda were antigenically similar to the Asibi strain. The study of the sera collected during the epidemic illustrates the frequent difficulty met by physicians in recognizing yellow fever and shows the extensive spread of the epidemic. It is noteworthy that the finding of antibodies to the Chikungunya virus in 13 % of the sera studied coincided with an outbreak of dengue-like disease caused by the Chikungunya virus, which immediately preceded and continued concurrently with the yellow fever epidemic (Filipe & Pinto, 1973). The Luanda epidemic was soon brought under control by prompt mass vaccination and intensive antimosquito measures. However, the existence of mosquito vectors in the territory (Ribeiro, 1971) and the possibility that the yellow fever virus may again be introduced into an urban area show that it is necessary to organize a regular vaccination pro- gramme as proposed by BrMs (1971) and to ensure permanent surveillance. 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), 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. Gonralves, National School of Public Health and Tropical Medicine, Lisbon, for their valuable technical assistance; and Dr M. Ribeiro, for having classified the mosquitos used in the virological studies. This investigation was supported in part by the Instituto de Alta Cultura, Lisbon. RtSUMt ETUDE SUR LES ARBOVIRUS A LUANDA (ANGOLA): 1. ETUDES VIROLOGIQUES ET SEROLOGIQUES PENDANT UNE EPIDEMIE DE FItVRE JAUNE Au debut de 1971, la fievre jaune a fait son apparition a Luanda (Angola). L'epid6mie qui en a r6sult6 s'est developpee pendant les mois de f6vrier et mars, mais a et6 enrayee dans la premiere semaine d'avril sous l'effet des mesures anti-moustiques et de la campagne de vacci- nation. On a enregistr6 officiellement 65 cas et 42 dc&s. Les etudes virologiques qui ont ete effectu6es ont permis d'isoler 9 souches de virus, 8 6tant obtenues a partir du s6rum de malades et une a partir d'un fragment de foie pr6lev6 sur un patient d6cedd. La caracterisation des isolats a montre qu'il s'agissait dans tous les cas du virus amaril. Tous les essais d'isolement de virus A partir de lots de moustiques sont rest6s infructueux. On a d'autre part examin6 en epreuve d'inhibition de l'hemagglutination 589 serums pr6lev6s chez des per- sonnes habitant a Luanda ou dans sa banlieue; 113 d'entre elles etaient des contacts familiaux de malades atteints de fievre jaune et 166 des personnes suspectees d'avoir Wt6 infect6es par le virus amaril. Dans le premier groupe, on a d6nombre 45,1% de sujets poss6dant des anticorps antiamarils et dans le second groupe 44,5%. L'enquete serologique a 6galement montr6 qu'un autre virus avait ete present parmi la population en m8me temps que le virus amaril, en l'occurrence le virus chikungunya. REFERENCES BrMs, P. (1970) Bull. Wld Hlth Org., 43, 223 BrMs, P. (1971) In: Proceedings of the Conference on the Application of Vaccines Against Viral, RPckettsial, and Bacterial Diseases of Man, December 1970, Washing- ton, D.C. (Pan American Health Organization Scien- tific Publication No. 226), p. 15 Cambournac, F. J. C. et al. (1955) An. Inst. Med. trop. (Lisboa), 12, 101 Clarke, D. H. & Casals, J. (1958) Amer. J. trop. Med. Hyg., 7, 561 Filipe, A. R. & Pinto, M. R. (1973) Bull, Wld Hlth Org., 49, 37 ARBOVIRUS STUDIES IN ANGOLA: 1 35 Filipe, A. R. et al. (1972) An. Esc. nac. Sa&de publ. Med. trop., 6, 101 Hammon, W. McD. & 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. 268 Karber, G. (1931) Naunyn-Schmiedeberg's Arch. exp. Path. Pharmak., 162, 480 Kokernot, R. H. et al. (1965) Trans. roy. Soc. trop. Med. Hyg., 59, 563 Mora, A. D. (1933) Bull. trim. Org. Hyg., 2, 56 Ribeiro, H. (1971) Rev. mid. Angola, 13, 67 WHO Expert Committee on Yellow Fever (1971) Wi Hith Org. techn. rep. Ser., No. 479 Wkly epidem. Rec., 1972, 47, 229 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