Seroprevalence of five arboviruses in Zebu cattle in the Central African Republic

31

TRANSACTIONSOFTHEROYALSOCIETYOFTROPICALMEDICINEANDHYGIENE(1996)90,31-33

Seroprevalence

of five arboviruses

in Zebu cattle in the Central African

Republic

J. M. Guilhermel, Christine Gonella-Legall*, F. Legalll, E. Nakoume2 and J. Vincent ** ‘Association Nationale pour le D6veloppement de 1’Elevage (ANDE), Bangui, R6publique Centrafiz’caine; 21nstitut Pasteur, B. I’. 923, Bangui R6publique Centraficaine Abstract

Sequential blood samples taken from 237 adult and 147 young Zebu cattle in the Central African Republic (CAR) were examined by enzyme-linked immunosorbent assayfor the presenceof antibodies to Rift Valley Fever (RVF), Dugbe (DUG), Crimean-Congo haemorrhagic fever (CCHF), Chikungunya (CHIK) and Wesselsbron (WSL) viruses. Antibodies to DUG and CCHF viruses were detected in 70% and 61% respectively, of the sera from adult cattle, probably due to common antigens. This hypothesis is supported by the fact that CCHF virus was isolated only 3times from ticks, whereas DUG virus was isolated from almost all ticks. Ticks of the genus Hyalorama, the most important vectors of CCHF, are rare on Zebus in CAR. CCHF virus has a limited role in local human pathology, probably becauseits distribution is limited by the scarcity of its main vector and the immunological barrier resulting from the presenceof antibodies to DUG virus. Antibodies to RVF virus were found in about 8% of adult cattle. Zebus do not play an important role in the spread of CHIK and WSL viruses among humans; these viruses were found in only one and 5, respectively, of the 183cattle examined. Keywords: Zebu cattle, Rift Valley fever virus., Dugbe virus, Crimean-Congo haemorrhagic fever virus, Chikungunya virus, Wesselbron virus, Central African Republic Introduction

Rift Valley Fever (RVF) virus was first isolated in the Central African Republic (CAR), under the name of Zinga, from mosquito pools by DIGOUTTE in 1969. Following this isolation, 836 human sera from different regions of CAR and 261 sera from Zebu cattle were examined by indirect immunofluorescence; only 3 human sera gave a positive reaction and all the Zebu sera were negative. A limited number of individual clinical human cases,confirmed by virus isolation, have been reported by GONZALEZ (1985), with no reported infection in livestock. Crimean-Congo haemorrhagic fever (CCHF) virus is circulating

in Africa with variable freauencies. but hu-

man cases-areinfrequent, despite indications ofhigh levels of virus transmission in livestock, in certain parts of the continent (FISHER-H• CH et al., 1992). In CAR, CCHF virus was occasionally isolated from the ticks Amblyommavariegatum (2 strains) and Hyalomma nitidum (1 strain) and once from a small peridomestic rodent, Mastomys (1 strain). Only one human casewas reported in 1976(strain HB760493). More than 3000 sera of Zebu, sampled from 1978 to 1985,were examined by GEORGES & GONZALES (1986), and a low percentageof seropositivity was revealed by indirect immunofluorescence. Unlike CCHF virus, Dugbe (DUG) virus was often detected in bovine ticks, but the seroprevalence in Zebus of this virus, which does not seemto be pathogenic, was not examined.

Considering that the very large population of Zebus in CAR can act as a sentinel for human infections, we used enzyme-linked immunosorbent assay (ELISA), which is considered to be more sensitive than immunofluorescence, to investigate the seroprevalence of RVF, CCHF and DUG.viruses in Zebu cattle in the Bambari region (South-Central CAR), from which we could obtain sequential individual blood samples. Wesselsbron (WSL) and Chikungunya (CHIK) viruses, frequently isolated from mosquitoes and sometimes from human cases,were included in this study to determine whether bovines play a part in their transmission. Materials and Methods Blood samples

From July 1992 to June 1993, blood samples were taken from each animal of 7 different herds of Zebu cattle in the region of Bambari, CAR (Table 1). Four herds received anti-tick treatment (Flumethrinm) from July to *Author for correspondence

Table 1. Number of cattle sampled

No. of samplings 1

2 3 4-7 Total

Age of cattle (years) a2 <2 71 48 61 39 55 28 11 76 240 149

November 1992 and treatment against flying insects (DeltametrhinB) form December 1992to June 1993. Ticks

As it was not possible at the time of blood sampling to collect the ticks present on each animal, we independently checked for the presence of CCHF virus in 1514 different batches of ticks, collected between May 1993 and March 1994 from 1005Zebus (also blood sampled) at the slaughter-house in Bangui. Each batch, consisting of 2-8 ticks, was homogenized in 1 mL of phosphate buffered saline, pH 7.4 (PBS), supplemented with 1% bovine serum albumin and antibiotics, and inoculated to a litter of suckling mice by the intracerebral route. The isolated strains were passaged in suckling mice until a stable averagesurvival time was attained. Antigens Cell culture antigens. Antigens derived from cell cultures were produced by ultrasonic disruption, in the presence of 1% Triton X1008 (v/v), of infected Vero E6 cells, maintained in serum-free medium and presenting an almost complete cytopathic effect. This type of antigen was prepared with RFV (wild strain kindly supplied by Laboratoire des Fievres Hemorragiques, Institut Pasteur, Paris), with a locally isoalted WSL strain (ARBM 4177), and with an indigenous strain of CHIK (ARBM 27675). Uninoculated cells processed similarly were used as the control preparation.

Mouse brain antigens For DUG virus (strain ARBT 5247) and CCHF virus (strain 10200/Nigeria), antigens were derived from suckling mouse brain homogenized in sterile borate saline, pH = 9 (lmL/brain) and submitted to brief ultrasonic disruption. The viral brain suspension was inactivated with J3-propiolactone, l/2000 final concentration, over-

32 Table 2. Cumulative seroprevalence the Central African Republic

J.M.GUILHERME ETAL. and, seroconversion

(incidence)

rates to three arboviruses

in Zebu cattle in

Age of Virusesa cattle DUG CCHF DUG+CCHF RVF Prevalence Incidence Prevalence Incidence Prevalence Prevalence Incidence (years) 22 70.4%(1691240) 47% (31/66) 61% (146/240) 36% (35/97) 58% (140/240) 8.3% (20/240) 2.6% (61224) <2 4.0% (6/149) 18%(10156) 7.3% (11/149) 5%c(3/58) 4.7% (7/149) 4.7% (7/149) 2% (3/45) aDUG=Dugbe, CCHF=Crimean-Congo haemorrhagic fever, RVF=Rift Valley fever. Results are expressedas percentagesof positive cattle, with number positive/number examined in parentheses,between June 1992and June 1993.‘Incidence’ refers to animals which were initially seronegative but had becomeseropositive at a second test. bOnlv one animal seroconverted to CCHF alone: all the others were either initially seropositive for, or seroconverted to, DUG. ~A113 seroconverted to both CCHF and DUG. night at 4’C, and centrifuged at 12 000 g for 10 min at 4°C. The final supernatant was used as antigen. Uninoculated suckling mouse brains were harvested and processedunder the sameconditions for use as control. Serological reactions

Antibodies were detected by ELBA. The washing solution was PBS plus 0.1% Tween 208. The antigens derived from cell cultures were coated directly on to the wells of microtitre plates (Dynatech, catalogue no. 001-010-2101)in PBS for 12-18 h at 4°C. Positive and negative antigens were dispensed in alternating rows. The brain antigens were indirectly attached by immunocapture. Hyperimmune mouse ascitic fluid (neutralization index in mice 23), diluted as predetermined in carbonate-bicarbonate buffer, pH 9.5, was coated overnight at 4°C on to all the wells of a microtitre plate. After 3 washes in PBS plus 0.1% TweenB, positive and control antigens diluted in Blotto@ (washing solution plus 5% skim milk (Difco, catalogue no. 0032-17-3)were dispensed into alternate wells of successivecolumns and reacted for 1 h at 37°C with the fixed antibodies. After coating with the antigens, the plates were washed 3 times, and the bovine serum samples (diluted l/100 in BlottoB) were added to 4 successive wells in each horizontal row. Unfixed antibodies were washed off (3 washes) after 1 h incubation at 37°C and peroxidaselabelled antibovine serum (Kirkegaard & Perry Laboratories, catalogue no. 14-12-06),or peroxidase-labelled antisheep rabbit immunoglobulins (Kirkegaard & Perry Laboratories, catalogue no. 14-23-06),diluted in BlottoB according to previous titrations, were added and allowed to react for 1 h at 37°C. Positive reactions were revealed with either o-phenylenediamine or 2,2-azinobis(3-ethylbenzthiazoline-sulphonic acid) (ABTS@; Kierkegaard & Perry Laboratories, catalogue no. 50-62-00). Optical densities (OD) were visually evaluated and scored from + to + + + +. They were also recorded with the aid of an LP 300@ spectrophotometer (Diagnostic Pasteur) fitted with the appropriate filters. Visual evaluation was concordant with the spectrophotometric records. Based on the ODs obtained with control antigens and with 5 reference positive sera,we considered as Fositive a serum for which the mean OD obtained with the antigen was at least double that obtained in the control wells. Results and Discussion Serology CCHF and DUG viruses. Prevalence rates were high in

adult cattle (22 vears old), and most of the uositive sera reacted with’both antigens (Table 2). In animals under 2 years of age the prevalence rates were much lower. Incidence rates, indicated by seroconversions, were also significantly higher in the adult cattle (Table 2). Perhaps tick infestation is reduced in calves, either due to mater-

nal care or becauseticks are less attracted by young animals. The double reactivities can be explained by the existence of common antigens, and the monosoecific DUG reactions may represent early infections wifh antibodies directed mainly to specific surface proteins. The few reactions annarentlv monosoecific for CCHF virus are difficult to explain, unless this virus can be transmitted in some other way than by tick bites. R VF virus. In adult cattle the prevalence rate ranged from 2.5% to 15% in different herds (Table 2). An average prevalence rate of 4% was determined in 1991-1992 with 540 sera from young adult female Zebus (3 years old) in the sameregion (unpublished observations). The apparent increase in prevalence rates in some herds could be related to the extended agerange of the animals sampled in the present survey (from 2 to over 12 years of age). The incidence rates, in adults (seroconversions) ranged from 1.5%to 4.3%. 1; 4 herds no young animal gave a positive reaction, but in 2 herds the nercentaae of oositive samnles (5% and lo%, including-3 serocoiversibns) was equal to; or greater than, that in the adults in the sameherds. These 2 herds (and one of those in the first group) were among those treated with FlumethrinB. WSL and CHIK viruses. Only one of 183 sera examined reacted with CHIK antigen and only 5 (2.7%) with WSL antigen. It appears that cattle do not play an important part in the transmission cycle of these viruses, which are both currently being isolated from mosquito pools and occasionally from human cases. Virus isolation from ticks

Most of the ticks collected belonged to the generaAmblyomma or BoophiZus; less than 1% of the 1514 batches were identified as Hyalomma. Only DUG virus strains were isolated, from A. variegatum (55 strains) and B. decoloratus (4 strains). No correlation was observed be-

tween the presenceor absenceof antibodies to DUG antigen in the cattle from which the ticks were collected and the isolation of viral strains from the corresponding ticks. Conclusions

The very high prevalence of antibodies to DUG virus correlates with its freauent isolation from ticks. In adult Zebus this high Prevalencewas urobablv due to cumulative infections by tick bites; almost all the animals over 7 Yearsof aaehad antibodies to CCHF/DUG antigens *Antigen; relationships between DUG and CCHF viruses could explain the large number of samplesreacting to both antigens. The monospecific reactions to DUG, that we have considered as possible primary infections, could also have been due to the presence of variant strains of DUG virus possessingfewer antigens in common with CCHF virus. The use of recombinant moteins specific for each virus will perhaps help to explain the double positives (ANTONIADIS et al., 1992).

ARBOVIRUS

PREVALENCE

IN AFRICAN

33

CATTLE

The very high seroprevalence rate of DUG virus in adult animals may represent an immunological barrier to CCHF virus, reinforced ecologically by the rare occurrence of Hyalomma ticks, which are considered to be the main vector of CCHF in West Africa. The detection of antibodies to RVF virus, particularly in young cattle, implies that this virus is still active in CAR. Nevertheless,no abnormal rate of abortions or other specific symptom has been reported in Zebus. Contrary to the situation in Sahelian Africa, herds in CAR consist mainly of cattle with very few sheep,the main hosts of the virus in other regions. This could be a limiting factor in the amplification necessaryfor epizootics to develop in domesticanimals. Another possibility is that at least some of the local strains of RVF virus are less pathogenic for livestock, but we have no strain isolated from local animals with which to investigate this hypothesis. Apart from the very few human clinical casesreported in the past, a recent unpublished serosurvey in humans in CAR revealed that approximately 7% of 700 sera were positive for RVF virus, with no indication of specific symptomatology in the past. Most of these positive sera were collected from Pygmies living in forested zones, having very limited contact with domestic animals. A sylvatic cycle for RVF virus is a possibility.

vironmental Microbiology, Oxford, UK) for comments on the manuscrint, Dr H. Zeller (Institut Pasteur, Dakar) for suoolv7 ing a set bf bovine CCHF‘reference sera? Dr J. I?.’ Cornei (okSTOM, Dakar) for his assistance in tick tdentification, and all our colleagues who helped with the blood sampling.

Acknowledgements

Received

We thank Dr Patricia Nuttall (Institute

of Virology

and En-

References Antoniadis, A., Polizoni, T., Tomori, O., Williams, O., Marriott. A. C:& Nuttall. I’. A. (1992). Evidence of nreviouslv unrecognized nairovirus infections of humans ih Nigeria and Greece.Journal of Viral Disease, 1,34-39. Digoutte, J. l’. (1969). Rapport Annuel de Unstitut Pasteur de Bang@ 1969. Bangui, Republique Centrafricaine: Institut Pasteur, p. 46. Fischer-Hoch, S. I’., McCormick, J. B., Swanepoel, R. S., Van Middlekoop, A., Harvey, S. & Kustner, H. G. V. (1992). Risk of human infections with Crimean-Congo hemorrhagic fever virus in a South African rural community. American Journal of Tropical Medicine and Hygiene, 47,337-345. Georges, A. J. & Gonzalez, J. P. (1986). Rapport Annuel de Z’Institut Pasteur de Bang@ 1984-1985. Bangui, Rtpublique Centrafricaine: Institut Pasteur, pp. 40-41. Gonzalez, J. I’. (1985). Surveillance epidemiologique de l’endtmie Rift Valley fever en RCA. Rapport Bisannuel d’Activite!, Institut Pasteur de Ban@, 1984-1985. Bangui, Republique Centrafricaine: 1’Institut Pasteur, pp. 49-51. 29 March 199s; revised 4 September 1995; acceptedfor publication 5 September 1995

EMOP VII VII European Multicolloquium of Parasitology Parma, Italy 2-6 September I996

Last day for early registration and receipt of abstracts: 20 April 1996; last day for hotel reservation: 30 June 1996; last day for late registration: 15July 1996. Further information can be obtained from EMOP VII Organizing Secretariat, New Team, Via C. Ghiretti 2, 43100 Parma, Italy; phone +39 521 293913,fax +39 521 294036; or from EMOP VII Scientific Secretariat, Istitut di Patologia Generale Veterinaria, Via Celoria lo,20133 Milano, Italy; phone +39 2 2666079,fax +39 2 2364470.

Annual International Course on Identification of Insects and other Arthropods of Medical and Veterinary Importance I-19July 1996 The Natural History Museum, London, UK

Further information can be obtained from Dr M. J. R. Hall, Medical and Veterinary Division, Department of Entomology, The Natural History Museum, Cromwell Road, London, SW7 5BD, UK; telephone +44 (0)171 938 9451, fax +44(0)171 938 9395 or 8937, e-mail [email protected]