TRANSACTIONS OF THE ROYAL SOCIETY OF TROPICAL MEDICINE AND HYGIENE (1999) 93,581-586 Aedes aegypti in Ho Chi Minh City (Viet Nam): susceptibility genetic differentiation to dengue virus and Tran Khanh Tien’, Marie Vazeille-Falcoz*, Laurence Mousson 2, Tran Huu Hoang’, Franqois Rodhair?, ‘Institut Pasteur de Ho-Chi-Minh Ville, Laboratoire d’Entomologie Nguyen Thi Huong’ and Anna-Bella Faillow nzgdicale, 167 Duong Pasteur, Ho-Chi-Minh City, Viet Nam; ‘Institut Pasteur, Ecologic des Systenzes Vectoriels, 25 rue du Dr Roux, 75 724 Park, France Abstract vector of dengue viruses, responsible for a viral infection that has become a major public health concern in Asia In Viet Nam, dengue haemorrhagic fever was first detected in the 1960s and is now a leading cause of death in childhood We studied the variability in competence of Ae aegypti as a vector for dengue virus and genetic differentiation in this mosquito species Twenty mosquito samples collected in 1998 in Ho Chi Minh City were subjected to oral infection and isoenzyme polymorphism analysis by starch gel electrophoresis Ae aegypti populations from the centre of Ho Chi Minh City were genetically differentiated and their infection rates differed from those of populations from the commuter belt These results have implications for insecticidal control during dengue outbreaks Aedes aegypti is the principal Keywords: A&s aegypti, dengue, oral susceptibility, population Introduction Dengue epidemics have been reported since 1779 (HIRSCH, 1883) Before the 195Os, epidemics generally occurred at IO-40-year intervals, mainly because both the viruses and their mosquito vectors were displaced only by sea transportation After World War II, the pattern of dengue transmission changed and the disease increasingly affected cities in most tropical countries, particularly those in Asia Increased movement ofhuman populations to urban agglomerations and the resulting uncontrolled urbanization created an ideal situation for dengue transmission At the same time, the increase in air traffic further enhanced the spread of dengue viruses The result was hyperendemicity (simultaneous circulation of the dengue serotypes), which has resulted in an increase in dengue incidence, and the emergence of dengue haemorrhagic fever (DHF) (GUBLER, 1998) In South-East Asia, DHF was first described in Manila et al., 1960) The (Philippines) in 1956 (HAMMON disease was first reported in North Viet Nam in 1958, South Viet Nam in 1960, Singapore in 1960, Penang in 1962, and India in 1963 (HALSTEAD, 1966) Since the 196Os, major dengue epidemics have affected at least million people living around the Red River Delta, with DHF occurring every year in Ho Chi Minh City (HALSTEAD et al., 1965; Vo QUI DAI & NGUYEN THI KIMTHOA, 1967) In this region, the pattern of DHF has developed since 1975 with a gradual rise in the number of fatal cases: 119 429 DHF cases were recorded in 1998 versus 19 16 in 1975 In 1998, dengue virus was isolated from 15% of 1467 serum samples from clinical patients, and in 74% of these cases the virus isolated was denaue3 All the DHF cases observed in Ho Chi Minh C& in 1998 were due to dengue virus (DO QUANG HA et al., 1998) In the dengue-endemic areas of Asia, types of transmission are observed: urban and rural In urban situations, the mosquito vector, Aedes aegypti, is commonly found close to human populations whereas the less-domestic Aedes albopictus lives at low densities in planted open spaces In rural areas, Ae aegypti is less common and Ae albopictus is the principal vector of dengue viruses in rural communities It causes only sporadic cases and local outbreaks are observed There is evidence that Ae albopictus is indigenous to Viet Nam whereas Ae aegypti was introduced, probably around 1915 (STANTON, 1920) In the 192Os, Ae aegyptimade up 4% of the mosquito population in Saigon (Ho Chi Address for correspondence: Dr Anna-Bella Failloux, Unite d’Ecologie des Svstkmes Vectoriels, 25 rue du Dr Roux, 75724 Paris c&Jex 15, -France; phone +j3 40613617, fax’+33 40613089, e-mail afaillou@pasteur.fr genetics, Viet Nam Minh City) (BOREL, 1926) It has been suggested that Ae aegypti invaded Asia during the second half of the 19th Century following the increase in trade and shipping The species was first introduced via the seaports and gradually spread inland along rivers and roads Usually, the peak in dengue incidence occurs during the rainy season, coinciding with the highest densities of Ae aegypti populations In South Viet Nam, dengue cases are observed throughout the year whereas in North Viet Nam, dengue transmission does not occur during the winter In Ho Chi Minh City, DHF cases are mostly reported from June to October A substantial proportion (39.4%) of the mosquitoes collected inside habitations and most of the larvae collected from neridomestic breeding sites are Ae aegypti (data from-the Pasteur Institute in Ho Chi Minh City) Only 1% of the collected adults are Ae albopictus This emphasizes the importance of urbanization in providing breeding places suitable for Ae aegypti The change in the pattern of dengue transmission in Viet Nam from urban epidemics to endemicity is partly correlated with the establishment of Ae aegypti in cities Urban endemic dengue fever was not known in Asia until the end of the 19th Century although epidemics did occur in rural zones The recent arrival of Ae aegypti, which can survive in urban environments, has led to the emergence of urban epidemics Ae aegypti has a limited flight range, which restricts its dispersion around breeding sites (CHRISTOPHERS, 1960) Populations from highly urbanized cities therefore tend to differ greatly in genetic structure (TABACHNICK, 199 1) Genetic differences may be correlated with traits of medical importance such as competence as a vector for dengue viruses In this study, variability in the susceptibility of Ae aegypti to dengue virus was analysed to determine the competence of Ae aeapti as a vector In addition, we estimated genetic variability by investigating isoenzyme polymorphism This indicated the pattern of gene flow and, indirectly, the pattern of dispersion of potentially infected Ae aegypti in Ho Chi Minh City Materials Mosquito and Methods sawtples (Table 1, Figure) Twenty mosquito samples were collected from February to May 1998 in Ho Chi Minh City and its suburbs: from the 2nd district, from the 7th district, from the 8th district, from Binh Chanh, from Cu Chi, from Hoc Mon, from Nha Be, with the remaining samples from the 6th district, Thu Due, 9th district and Binh Thanh The samples collected consisted of larvae or pupae which were reared until the imago stage in an insectarium (temperature 25 & l”C, relative humidity 80 f IO%, 582 TRANKHANHTIEN ETAL Table Location ofAe aegypti February to May 1998 samples collected Sample Location District PHU NHA KIE NBE BIN CHA 7.ABB CUC CHI 10 BCH 11 LOI 12 HA1 13 MON 14 HOC 15 KHA 16 QUA 17 THU 18 BIH 19 THA 20 MYL Phu Thuan Nha Be Tan Kieng Phuoc Kien Binh Chanh Hung Long 5th ward Phuoc Thanh Cu Chi Tan Quy Tay Binh Loi Pham Van Hai Hoc Mon Tan Thoi Nhi An Khanh 8th ward Hiep Binh Phuoc Tan Phu 27th ward Thanh My Loi 7th Nha Be 7th Nha Be Binh Chanh Binh Chanh 8th Cu Chi Cu Chi Binh Chanh Binh Chanh Binh Chanh Hoc Mon Hoc Mon 2nd 6th Thu Due 9th Binh Thanh 2nd and 12-h light/dark cycle) The resulting FO adults were allowed to blood feed on mice Batches of eggs were obtained and hatched and the larvae were reared until the image stage (Fl) FO adults were stored at -80°C until analysis forisoenzyme polymorphism and Fl adults were infected exuerimentallv with denaue virus The Paea strain of Ae: aegy@ was collected i; Paea district (Tahiti, French Polynesia) and has been maintained in our laboratory since 1994 This strain was used as a control for mosquito oral susceptibility to dengue viruses and as a mobility control for isoenzyme polymorphism Experimental infection qf mosquitoes bne to two infection assays were carried out for each Fl mosauito samnle The 5-7-dav-old non-blood-fed females were placed in 0.5-litre plastic-screened containers and starved for 24 h prior to infection They were allowed to feed for 20 on an infected artificial meal maintained at 37°C in glass feeders covered with chicken skin stretched over thi opening at the base (VAZEILLEFALCOZ et aZ., 1999) The infectious meal consisted of parts washed rabbit erythrocytes to part virus suspension and ATP (a phagostimulant) at a final concentration of X 1Om3M The feeding suspension had a final titre of 10”’ MIDso (mosquito infectious dose for 50% for Ae aegypti individuals) per mL of the dengue type strain isolated in 1974 from a serum samnle from Bangkok (Thailand) The preparation of viral stocks has been described elsewhere (VAZEILLE-FALCOZ et al., 1999) Fully engorged females were incubated at 28°C in small cardboard containers and were fed on a 10% sugar solution for 14 days Surviving females were killed and tested for the presence of dengue virus by indirect immunofluorescence assay (IFA) on head squashes The Paea strain was used as an infectivity control (VAZEILLE-FALCOZ et al., 1999) Prior to the infection of field-derived populations of Ae aegypti, we determined the optimal amount of virus necessary to infect an average of 90% of the Paea strain For such purpose, serial dilutions of the viral stock were used to infect mosquitoes The titre of the feeding suspension that gave 90% infected females was estimated and was later used to infect field populations Isoenzyvne po&morp&n Individual mosquitoes were homogenized distilled water The homogenate- was (15 000 g, at +4”C) and half the was subjected to starch gel electrophoresis in 25 L of centrifuged supernatant in the Tris- in Ho Chi Minh City from Collection date 05/02198 05102198 07102198 07102198 10102198 l/02/98 11102198 19102198 19102198 27102198 05/03/98 05lO3198 20105198 20105198 28105198 19/03/98 25103198 31103198 01104198 28105198 maleate-EDTA (TME) 7.4 buffer system (PASTEUR et al., 1988) This made it possible to investigate 10 enzyme svstems oer individual (5 ner eel) We actuallv studied ehzyme systems: esterases (Est, ‘EC 3.1.1.1 )I phosphoglucoisomerase (Pgi, EC 5.3.1.9.), glutamate-oxaloacetate transaminases (Got1 and Got2, EC 2.6.1 l.), giycero-phosphate dehydrogenase (Gpd, EC 1.1.1 S.), hexokinases (Hk, EC 2.7.1.1.), malate dehydrogenase (Mdh, EC 1.1.1.37.), mannose-phosphate isomerase (Mpi, EC 5.3.1.8.), malic enzyme (Me, EC 1.1.1.40.) and phosphoglucomutase (Pgm, EC 2.7.5.1.) A strain developed from an isofemale lineage of Ae aegypti ‘Paea’ (Tahiti, French Polynesia) was used as a control Allelic composition at each locus was evaluated for the Paea strain In field populations, alleles were scored in relation to the most common allele obtained at each locus in the Paea strain Population diSferentiation Differentiation was analysed with GENEPOP (version 3.1) software (RAYMOND & ROUSSET, 1995) I”, and F,, were estimated as described by WEIR & COCKERHAM (1984) Heterozygote deficits were tested using an exact test procedure (ROUSSET & RAYMOND, 1995) Global disequilibrium between pairs of loci was tested for each sample Genotypic differentiation was tested by calculating the P value of an F,, estimate The overall significance of multiple tests was estimated by Fisher’s combined probability test (FISHER, 1970) The number of effective migrants (Nm) was calculated from F,, estimates according to the relationship: Nm = [( 1/F,,) - l] /4 (WRIGHT, 1969) Isolation by distance was assessed by determining the significance of the correlation between F,, estimates and geographical distances (SMOUSE et al., 1986; LEDUC et al., 1992) If necessary, the significance level of each test was adjusted based on the number of tests run, according to the sequential method of Bonferroni (HOLM, 1979) Results Mosquito infection rates For the 15 mosquito samples tested (Table 2), infection rates obtained ranged from 88.8% [QUA (16)] to 100% [BIN (5), BCH (10) and LO1 (ll)] For each sample, if replicates were carried out, the infection rates obtained were compared For PHU (l), NHA (2) and the Paea control, infection rates were homogeneous (P > 0.05) When we compared the rates obtained for the assay with their corresponding controls, only rate ,4EDESAEGYPTI IN HO CHI Mn\rH CITY (WET MM) 583 Figure Geographical location of sites from which Ae aegyptiwas collected in Ho Chi Minh City (Viet Nam) and its commuter belt (for more details, see Table 1) Scale bar represents km remained significantly different [P = 0.01 for THU (17)] In analysis of all the samples, an overall P value calculated in Fisher’s exact test was more highly significant (P = 0.002) than the P value estimated for the corresponding controls (P = 0,038) To better understand geographical differences in infection rates, groups were set up: one comprised sites[PHU(l),KIE(3),ARR(7),KHA(15),QUA(16), THU (17) and THA (19)] corresponding to mosquito populations from tile centre ofHo Chi Mjnh City The second group consisted of mosquito popuiations COIlected in the commuter belt @JHA (2), NBE (4), BIN (5), CHI (9),BCH (lO),LOI (11) andHA (12)] The null hypothesis of homogeneous infection rates was rejecred [P = 0,001) for samples from the city centre whereas the samples collected in the commuter belr were not differentiated (P = 0.35) Genetic polymorphism The enzyme systems studied led to the identification of 10 putative loci Est-1, &t-2, Est-3 and Mpi had uninterpretable profiles and were not considered in the analysis In the hexokinase system (Hkl, Hk2 and Hk3 loci), only Hk2 was taken into account because Hkl and Hk3 displayed only limited polymorphism Got, Gpd and Me were discarded as each had only or alleles Hk2, Pgi, Mdh and Pgm had allozyme profiles with codominant segregation in all samples (Table 3) The following genetic analysis is based on these loci Genotypic association between pairs of loci was ruled out for aI samples by multiple testing (P > 0.05) The loci analysed were considered to segregate independently Analysis of the allele frequencies at these loci showed significant heterozygote deficit (P < 0.05) in of 80 Bonferroni multjple tests: at the Pgm locus [PHU tat (l),CHI(9),QUA(16),TFIA(I9)andMYL(20)1, the Mdh locus [CHA (6)] and at the Pgi locus [CHI WI Genetic differentiation was analysed by comparing genotype frequencies and distributions -The overall differentiation of the 20 samDies evaluated bv estimatinn F,, was highly significant (FE = 1-0.099, P 10e6) (Tgble 4) We investigated the forces generating this pattern of differentiationby grouping samples together accordine to collection site in Ho Chi Minh Citv: samnles from the city centre and samples from the cot&&terbdt The samples collected in the city centre [PHU (I), KIE (3), ARR (7), KHA (15), QUA (16), THU (17), BIH (I@, THA (19), MYL (20)] were significantly and highly differentiated (F,, = +0.07 I, P < 1Om6).Mosquito samples collected from the commuter b&r fNHA (2), NBE (4), BIN (5), CHA (6), CUC (8), CHI (9), BCH (lo), LOI (1 l), HAI (I2), MON (13) and HOC (14)] were also highly (F, = i-0.125 and significantly differentiated (P < 10v6) Therefore, sites were considered according to their district Genetic differentiation was high (high F, values) and significant (P < 0.05) for populations collected in Cu Chi district (2 sites), Hoc Mon (2) and Binh Chanh (5) whereas in Nha Be district, moquito populationswerenotdifferentiated (F,T,= t0.051, P = 0,059) Gene flow (Table 4) estimated based on the number of effective migrants, ~Vtn, calculated from F,, estimates, was low for samples from the city centre (N17z= 3.3) and commuter belt (Nm = 1.7) Gene flow was estimated between samples from the same district (i.e., Cu Chi, Hoc Man, Binh Chati and Nha Be), and the rate of genetic exchange was high only for sites from Cu Chi (iVm = 7.31 and sites from Nha Be lNnr = 4-7) We assessed geographical genetic isolation by examining the slope of the regression between log Nm and log distance 584 TFlAN KHANH Table Experimental dengue virus infection lected from Ho Chi Minh City and its commuter rates for Ae aegypti belt TIEN ETAL col- % infected females (n) Sample PHU Replicate it NHA KIE NBE BIN ARR CHI 10 BCH 11 LO1 12 HA1 15 KHA 16 QUA 17 THU 18 BIH 19 THA All samples i a a a a Assay 99.1 96.5 96.8 96.5 92.2 93.5 100 94.5 98.8 100 100 96 92.6 88.8 98.7 98.5 92.3 (114) (145) (62) (86) (179) (92) (18) (55) (163) (36) (20) (50) (68) (143) (149) (66) (39) 0.002 (0.001) Control 100 100 100 100 96.1 94.7 90.8 90.8 90.8 87.5 93.7 87.5 94.7 93.7 87.5 93.7 93.7 (51) (50) (51) (50) (52) (38) (65) (65) (65j (32) (80) (32) (38) (80) (32) (80) (80) P (SD) (0) 0.335 (0.003) 0.503 (0.003) 0.294 (0.003) 0.533 (0~005) (0) 0.332 (0.003) 0.504 (0.004) 0.009 (0.001 j 0.05 (0.001) 0.579 (0.002) 0.206 (0.003) (0) 0.333 (0.005) 0.010 (0.001) 0.225 (0.003) 0.718 (0.003) 0.038 (0.003) We used a feeding suspensionwith a final tine of 10” MID,,/mL in all experiments to obtain an average of 90% of infected females in the Paea control strain P, probability of homogeneity (Fisher exact test); SD, standard deviation Numbers in bold correspond to significant Pvalues (cO.05) (SLATKIN, 1993) and by calculating the correlation between F,, estimates and geographical distance The correlation-between log Nm andlogdistance was positive (b = +O.OOOl) and F,, was nenativelv (b = -0.0004) but not significantly (P = Infectious Diseases, 4, 442-450 Halstead, S B (1966) Mosquito-borne haemorrhagic fevers of South and South East Asia Bulletin of the World Health Organization, 3.5, 3- 15 Halstead, S B., Voulgaropoulos, E M., Tien, N H & Udomsakdi, S (1965) Dengue hemorrhagic fever in South Vietnam: report of the 1963 outbreak American Journal of Tropical Medicine and Hygiene, 14, 19-830 Hammon, W McD., Rudnick, A., Sather, G., Rogers, I< D & Morse, L J (1960) New hemorrhagic fevers of children in the Philippines and Thailand Transactions ofthe Association of American Physicians, 73, 140-I 55 2.1 1.2 4.7 Pgm, phosphoglu- Hirsch, A (1883) Handbook of Geographical and Historical Pathology, Vol I London: New Sydenham Society Holm, S (1979) A simple sequentially rejective multiple test procedure Scandinavian Journal of Statistics, 6, 65-70 Leduc, A., Drapeau, I’., Bergeron, Y & Legendre, P (1992) Study of partial components of forest cover using partial Mantel tests and path analysis Journal of Vegetal Sciences, 3, 69-78 Mirovsky,J,,Vymola,F.&HoangThucThuy(1965).The1960 summer dengue epidemic in the Democratic Republic of Vietnam I Epidemiology observations Journal of Hygiene, Epidemiology, Microbiology and Immunology, Ix, 356-363 Monath, T P (1994) Dengue: the risk to developed and developing countries Proceedings of the National Academy of Sciences of the United States ofAmerica, 91, 2395-2400 Pasteur, N., Pasteur, G., Bonhomme, F., Catalan, J & BrittonDavidian, J (1988) Practical Isozyme Genetics Chichester, England: John Wiley and Sons/Ellis Horwood Ltd Raymond, M & Rousset, F (1995) GENEPOP (ver 1.2), a population genetics software for exact tests and ecumenicism Journal of Heredity, 86, 248-249 Reiter, P., Amador, M A., Anderson, R A & Clark, G G (1995) Dispersal of Aedes aegyptiin an urban area after blood feeding as demonstrated by rubidium-marked eggs American Journal of Tropical Medicine and Hygiene, 52, 177- 179 Rousset, F & Raymond, M (1995) Testing heterozygote excess and deficiency Genetics, 140, 1413-1419 Slatkin, M (1993) Isolation by distance in equilibrium and non-equilibrium populations Evolution, 47,264-279 Smouse, P E., Long, J L & Sokal, R R (1986) Multiple regression and correlation extensions of the Mantel test of Matrix correspondence Systematic Zoology, 35, 627-632 Stanton, A T (1920) Mosquitoes of far eastern ports with special reference to the prevalence of Stegomyia fasciata Bulletin of Entomological Research, 10, 333-334 Tabachnick, W J (1991) Evolutionary genetics and arthropod-borne disease: the yellow fever mosquito American Entomologist, 37, 14-24 Vazeille-Falcoz, M., Mousson, L., Rodhain, F., Chungue, E & Failloux, A B (1999) Variation in oral susceptibility to dengue type virus of populations of Aedes aegypti from the islands of Tahiti and Moorea, French Polynesia American Journal of TropicalMedicine and Hygiene, 60, 292-299 Vo Qui Dai & Nguyen Thi Kim-Thoa (1967) La fii-ure himorragique au Vietnam en 1964-1965 Etude serologique avec bref aper$u clinique et tpidtmiologique Bulletin de la So&t6 de Pathologic Exotique, 60, l-33 Weir, B S & Cockerham, C C (1984) Estimating F-statistics for the analysis of population structure Evolution, 38, 1358-1370 Wright, S (1969) Evolution and the Genetics of Populations, Vol 2, The Theoy, of Gene Frequencies Chicago: Chicago University Press Received 11 June 1999; revised August 1999; accepted for publication August 1999