Viewpoint: High susceptibility to Chikungunya virus of Aedes aegypti from the French West Indies and French Guiana

Authors


Corresponding Author Romain Girod, Institut Pasteur de la Guyane, Unité d’entomologie médicale, BP 6010, 97300 Cayenne, Guyane française. E-mail: rgirod@pasteur-cayenne.fr

Summary

Objectives  To estimate the vector competence of Aedes aegypti populations sampled from distinct anthropogenic environments in French Guiana, Guadeloupe and Martinique for the strain CHIKV 06.21.

Methods  F1/F2 females were orally infected at titres of 106 and 107.5 pfu/ml in blood-meals. Disseminated infection rates (DIR) of mosquitoes were estimated using indirect fluorescent antibody assay on heads’ squashes, 7 or 14 days post-infection (pi).

Results  At a titre of 107.5 pfu/ml, DIR ranged from 88.9% to 100.0% and were not significantly different whether assessed at day 7 or 14 pi. At a titre of 106 pfu/ml, DIR observed 7 days pi ranged from 37.6 to 62.0%.

Conclusions Ae. aegypti from French Guiana and French West Indies are highly competent to transmit CHIKV. An evaluation of DIR 7 days rather than 14 days pi is adequate to estimate vector competence. The titre of 106 pfu/ml allows us to distinguish Ae. aegypti populations originating from distinct environments (dense or diffuse housing) by their vector competence. This assessment is a prerequisite to better evaluate the potential risk of Chikungunya outbreaks once the virus is introduced from endemic regions.

Abstract

Objectifs:  Estimer la compétence vectorielle de populations d’Ae. aegyptiéchantillonnées dans différents environnements anthropogéniques en Guyane française, en Guadeloupe et en Martinique, pour la souche CHIKV 06.21.

Méthodes:  Des femelles F1/F2 ont été infectées par voie orale à des titres de 106 et 107,5 pfu/mL dans le repas sanguin. Les taux d’infection disséminée (TID) de moustiques ont été estimés à l’aide d’un test indirect fluorescent de dosage des anticorps sur des têtes broyées, 7 ou 14 jours post-infection (pi).

Résultats:  Au titre du 107,5 pfu/mL, les TID variaient de 88,9 à 100,0% et n’étaient pas significativement différents lorsqu’évalués au jour 7 ou 14 pi. Au titre de 106 pfu/mL, les TID observés 7 jours pi variaient de 37,6 à 62,0%.

Conclusions:  Les populations d’Ae. aegypti en Guyane et dans les Antilles françaises sont très compétentes pour transmettre CHIKV. Une évaluation des TID à 7 jours au lieu de 14 jours pi est suffisante pour estimer la compétence vectorielle. Le titre de 106 pfu/mL nous permet de distinguer des populations d’Ae. Aegypti provenant d’environnements distincts (logement dense ou diffus) par leur compétence vectorielle. Cette évaluation est une condition pré requise pour mieux évaluer le potentiel risque d’épidémie de chikungunya une fois que le virus est introduit dans des régions endémiques.

Abstract

Objetivos:  Estimar la competencia vectorial de las poblaciones de Ae. Aegypti, muestreadas en ambientes antropogénicos de la Guyana Francesa, Guadalupe y Martinica, frente a la cepa CHIKV 06.21.

Métodos:  Se infectaron oralmente hembras F1/F2 con titulaciones de 106 y 107.5 pfu/mL en sangre. Las tasas de diseminación de la infección (TDI) en los mosquitos se estimaron utilizando una prueba de inmunofluorescencia indirecta sobre las cabezas de los mosquitos aplastadas, 7 o 14 días post-infección (pi).

Resultados:  Con una titulación de 107.5 pfu/mL, la TDI tenía un rango de entre 88.9 y 100.0% y no eran significativamente diferentes si se medían el día 7 o 14 pi. Con una titulación de 106 pfu/mL, la TDI observada 7 días pi tenía un rango de 37.6 a 62.0%.

Conclusiones: Ae. aegypti de las Guyanas Francesas y las Indias Francesas Occidentales eran altamente competentes para la transmisión CHIKV. Una evaluación de TDI a los 7 días, en vez de a los 14 días pi es adecuada para estimar la competencia vectorial. Una titulación de 106 pfu/mL nos permite distinguir, según su competencia vectorial, las poblaciones de Ae. aegypti originarias de ambientes distintivos (con una densidad de construcciones alta o difusa). Esta evaluación es un prerrequisito para una mejor evaluación del riesgo potencial de brotes de Chikungunya una vez este haya sido introducido de regiones endémicas.

Introduction

Chikungunya virus (CHIKV) is an arthropod-borne virus of the Alphavirus genus (Togaviridae family) transmitted to humans by mosquitoes. A CHIKV infection induces fever, rash, myalgia, headache but the more specific symptom is arthralgia, often persistent, sometimes severe, which may result in long-lasting disability (Das et al. 2010).

Chikungunya virus is endemic to tropical Africa, South and Southeast Asia. In tropical Africa, CHIKV is maintained within a sylvatic cycle that involves primates and tree-hole inhabiting mosquitoes of the genus Aedes (Jupp & McIntosh 1990; Diallo et al. 1999). Transmission to humans is common (Jupp & Kemp 1996; Thonnon et al. 1999). In South and Southeast Asia, CHIKV is primarily transmitted within an urban cycle by Aedes aegypti and Aedes albopictus and human cases occur (Thaikruea et al. 1997; Lam et al. 2001).

The latest large-scale Chikungunya outbreak started in Kenya in 2004 (Chretien et al. 2007) and spread to the Indian Ocean islands, where it caused more than 250 000 cases in La Réunion in 2005–2006 (Renault et al. 2007). Then, the epidemic reached South and more recently Southeast Asian countries (Arankalle et al. 2007; Ng et al. 2009). In 2007, local transmission of CHIKV was observed for the first time in Europe in Ravenna Province, Northern Italy (Rezza et al. 2007), and outbreaks also occurred recently in Africa (Leroy et al. 2009; Peyrefitte et al. 2007).

During the La Réunion outbreak, a new variant of CHIKV emerged with a point mutation in the E1 gene sequence changing the amino acid at position 226 in the envelope protein from an alanine to a valine (Schuffenecker et al. 2006). This minor genetic change clearly enhances the transmissibility of CHIKV by Ae. albopictus (Tsetsarkin et al. 2007; Vazeille et al. 2007) and may be responsible for the expansion of CHIKV into new regions where Ae. albopictus is predominant or the only potential vector (De Lamballerie et al. 2008).

Aedes albopictus has a wide distribution including the Americas (Forattini 1986; Hawley et al. 1987) and some islands of the Caribbean (Chadee et al. 2003, Wheeler et al. 2009). Fortunately, it has not yet been seen in French Guiana and the French West Indies, but the risk of introduction remains high. The main potential mosquito vector of CHIKV in the French territories of America is Ae. aegypti. Despite intensive vector control in the context of dengue management, the species is very well established in the region where it essentially colonizes urban areas. Depending of the density and structure of housing, larval stages are developing in different types of containers, permanent or not, with variable food resources. This could favour the emergence of genetic variability in adult populations more or less susceptible to CHIKV.

In spite of several imported cases from endemic/epidemic countries, no local CHIKV transmission has been reported yet in French Guiana and the French West Indies. Indeed, between 2006 and 2009, nine travellers entering the French West Indies and French Guiana were diagnosed with confirmed CHIKV infection (Ardillon, personal communication). In the aim to assess the risk of local CHIKV transmission, we decided to evaluate the vector competence to CHIKV of different Ae. aegypti populations collected from French Guiana, Martinique and Guadeloupe, by measuring disseminated infection rates (DIR) in the laboratory.

Materials and methods

Mosquito sampling and breeding

Six localities were sampled in May–June 2008 and April–May 2009 in Guadeloupe, Martinique and French Guiana. In each of the three territories, two collections were performed, one in an environment characterized by dense housing and another one by diffuse housing (Table 1). Eggs were collected in ovitraps that consist of a 500-ml black plastic jar filled with hay infusion and lined with rough absorbent paper (Reiter et al. 1991). Papers with eggs (field parental stock) were brought back to insectaries and after hatching, larvae were reared until the adult stage. Ae. aegypti adults were sorted, and females were blood-fed on mice to produce first filial generation eggs (F1 generation). Egg batches were then sent to the Institut Pasteur in Paris where, after hatching, larvae were reared to adult stage under standardized conditions (temperature: 26 °C ± 2 °C, relative humidity: 80% ± 10%, photoperiod: 12 h:12 h). Females of the F1 generation were used for experimental oral infections with CHIKV. For the two samples collected in 2009 in Guadeloupe, because the number of F1 eggs was too small, F1 females were blood-fed on mice to obtain a second filial generation (F2 generation) eggs and F2 females were used for oral experiments.

Table 1.   Geographic location and environmental characteristics of Aedes aegypti collected in French Guiana and the French West Indies in 2008 and 2009
SampleTerritoryLocalityEnvironment
GDL-CARGuadeloupePointe à Pitre – CarénageDense housing
GDL-PEGuadeloupePetit bourg – Prise d’eauDiffuse housing
MAR-ERMMartiniqueFort de France – ErmitageDense housing
MAR-CAFMartiniqueRobert – CaféDiffuse housing
GUY-CVFrench GuianaCayenne – centre villeDense housing
GUY-MADFrench GuianaCayenne – MadeleineDiffuse housing

Virus strain

Chikungunya virus 06.21 isolated in November 2005 from a newborn male from La Réunion presenting meningo-encephalitis symptoms was provided by the French National Reference Center for Arboviruses at the Institut Pasteur. This strain harboured the A->V mutation at the position 226 in the E1 glycoprotein (E1-226V) (Schuffenecker et al. 2006). Viral stock used was a third passage on Ae. albopictus C6/36 (Igarashi 1978) stored at −80 °C in aliquots. Procedure for C6/36 cell infections and passages are described elsewhere (Vazeille et al. 2007).

Oral infection of mosquitoes

Infection assays were performed with 7-day-old females that were allowed to feed for 15 min through a chicken skin membrane covering the base of a glass feeder containing the blood-virus mixture maintained at 37 °C. The infectious blood-meal was composed of a virus suspension diluted (1:3) in washed rabbit erythrocytes isolated from arterial blood collected 24 h before the infectious blood-meal (Vazeille-Falcoz et al. 1999). A phagostimulant, ATP, was added at a final concentration of 5 × 10−3 m. Fully engorged females were transferred to small cardboard containers and maintained with 10% sucrose solution at 28 ± 1 °C for 7 or 14 days. The titre of the blood-meal was 106 or 107.5 plaque-forming unit (pfu)/ml.

To evaluate DIR and thus vector competence, surviving females were killed at −80 °C and tested for the presence of CHIKV antigens in head squashes by immunofluorescence assay (IFA) (Kuberski & Rosen 1977). Mosquito samples were compared according to the DIR that corresponds to the proportion of females positive by IFA on head squashes among surviving females 7 or 14 days post-infection (pi).

Statistical analysis

Disseminated infection rates (DIR) were compared using the χ2 test or the Fischer’s exact test when sample sizes were too small, resorting to the Stata® software (StataCorp LP, TX, USA). When the P value was <0.05, the DIR were considered as significantly different.

Results

Twelve Ae. aegypti samples were collected in French Guiana and the French West Indies, six in May–June 2008 and six in April–May 2009 (Table 1). From the six samples collected in 2008, 918 F1 females were successfully orally infected with CHIKV 06.21 at titre 107.5 pfu/ml in the blood-meals. From the six samples collected in 2009, 1031 and 634 F1 females (or F2 females for samples collected in Guadeloupe) were successfully orally infected with CHIKV 06.21 at titres of 107.5 and 106 pfu/ml in the blood-meals, respectively. Table 2 shows DIR observed 7 or 14 post-infection (pi).

Table 2.   Percentages of infected females (number of tested females) of Aedes aegypti collected in French Guiana and the French West Indies in 2008 and 2009, 7 or 14 days after oral infection with CHIKV 06.21 (106 or 107.5 pfu/mL in the blood-meal)
Titre of the blood-meal107.5 pfu/mL107.5 pfu/mL107.5 pfu/mL106 pfu/mL
Incubation periodDay 14 piDay 7 piDay 14 piDay 7 pi
Year2008200920092009
  1. Pi, post-infection; nt, not tested.

GDL-CAR98.0 (145)96.6 (148)100.0 (145)54.7 (148)
GDL-PE95.8 (213)nt97.9 (95)nt
MAR-ERM98.9 (173)100.0 (48)96.8 (31)54.8 (93)
MAR-CAF97.4 (117)88.9 (90)93.4 (151)38.6 (184)
GUY-CV100.0 (188)97.5 (80)95.5 (88)62.0 (92)
GUY-MAD98.8 (82)94.7 (75)97.5 (80)37.6 (117)
All dense housing samples together99.0 (506)97.5 (276)98.5 (264)56.8 (333)
All diffuse housing samples together96.8 (412)91.5 (165)95.7 (326)38.2 (301)

Disseminated infection rates, observed at day 14 pi, for a titre of 107.5 pfu/ml, ranged from 95.8 to 100.0% for the 2008 samples and from 93.4 to 100.0% for the 2009 samples. Of all six comparisons between samples from the same location, all DIR obtained in 2008 and 2009 were similar (Fisher exact tests, P > 0.05) except for the GUY-CV samples (Fischer exact test, P = 0.01). For the same titre, DIR observed at day 7 pi ranged from 88.9 to 100.0% for the 2009 samples. Of all five comparisons between samples from the same location, all DIR obtained at day 7 or day 14 pi were similar (Fisher exact tests, P > 0.05).

Disseminated infection rates observed at day 7 pi, but for a titre of 106 pfu/ml, ranged from 37.6 to 62.0% and were highly heterogeneous among samples (χ2 test, P < 0.001). Of all five comparisons between samples from the same location, all DIR estimated at day 7 pi for a titre of 106 pfu/ml in the blood-meal were significantly lower than those obtained for a titre of 107.5 pfu/ml (χ2 tests, P < 0.001).

The examination of samples according to their environment (dense vs. diffuse housing) highlighted a significant difference among DIR at day 7 pi for a titre of 106 pfu/ml. Samples collected in dense housing environments showed higher DIR than samples collected in diffuse housing environments (χ2 test, P < 0.001).

Discussion

Aedes aegypti collected in 2008 and 2009 in Martinique, Guadeloupe and French Guiana were highly competent for CHIKV 06.21. Our results show that (i) disseminated infection rates can be assessed as early as day 7 pi, as rates observed at day 7 or day 14 pi were not significantly different, (ii) most mosquito samples exhibited DIR that did not vary from 1 year to the other, (iii) the DIR should be estimated using a titre of 106 pfu/ml in the blood-meal to emphasize significant variations among samples, and (iv) mosquito samples collected in dense housing environments exhibited the highest DIR.

Aedes aegypti vector competence to CHIKV has already been widely estimated (Mourya et al. 1987; Banerjee et al. 1988), demonstrating that the species has a lower competence in the laboratory than Ae. albopictus (Mangiafico 1971; Turell et al. 1992). More recently, Tsetsarkin et al. (2007) demonstrated that the E1-226V mutation was increasing vector competence for CHIKV in Ae. albopictus but not in Ae. aegypti laboratory strains. This was confirmed by Reiskind et al. (2008) and Pesko et al. (2009) with Ae. aegypti and Ae. albopictus from Palm Beach (Florida, USA): if both species were indeed susceptible to high doses of CHIKV, only Ae. albopictus developed DIR after exposure to low doses. However, based on our results, Ae. aegypti populations from French Guiana and the French West Indies displayed DIR similar to those of Ae. albopictus from La Réunion, Mayotte (Indian Ocean) and Hanoï (Vietnam) and Ae. aegypti from Ho Chi Minh City (Vietnam) when assessed at day 14 pi with similar viral titres (Vazeille et al. 2007). Under the same laboratory conditions, our Ae. aegypti populations were even more competent than Ae. albopictus and Ae. aegypti from Libreville (Gabon) (Vazeille et al. 2008) or Yaoundé (Cameroun) (Paupy et al. 2010).

The opportunity to measure vector competence by assessing DIR at day 7 pi is in accordance with the short extrinsic incubation period of CHIKV in Aedes mosquitoes (Dubrulle et al. 2009). Indeed, the virus reaches the salivary glands and is delivered in the saliva as soon as 2 days pi with a maximum of viral particles excreted at day 7 pi.

Estimating the DIR using a titre of 106 pfu/ml with a view to evaluate vector competence is supported by the observation of Lanciotti et al. (2007) who showed that CHIKV was circulating at titres from 103.9 to 106.8 pfu/ml in the blood of travellers returning to the USA from endemic countries.

Furthermore, the use of such a titre of CHIKV in the infectious blood-meal allows us to detect variations in vector competence between mosquito samples. Mosquitoes from dense housing environments presented higher DIR than samples collected in a diffuse housing environment. In French Guiana and the French West Indies, development of immature stages mainly occurs in man-made containers: (i) permanent containers productive throughout the year such as water storage containers (jars, barrels, drums, cisterns etc.) and flower-holding containers (vases, pots etc.) and (ii) temporal containers only productive at the rainy season such as trash recipients (tin cans, discarded bottles, tires, junked cars etc.). Differences in number and types of containers as well as their organic resources could lead to different patterns of population dynamics across seasons therefore leading to variations observed in DIR and then vector competence. Similar variations have already been shown in Ae. aegypti collected from different districts in Phnom Penh (Cambodia) (Paupy et al. 2003) and in Ho Chi Minh City (Vietnam) (Huber et al. 2002) when infected with a dengue virus.

Although Ae. aegypti from French West Indies and French Guiana are efficient laboratory vectors, their potential role in a CHIKV transmission cycle will also depend on other factors in relation to mosquito bio-ecology (densities, longevity, anthropophily, duration of gonotrophic cycle). Nevertheless, attention should be maintained considering the regular introductions of CHIKV in those territories through viremic travellers and the presence of dense populations of Ae. aegypti in the human environment. Thus, an efficient sanitary surveillance must be kept and a swift response including efficient vector control measures must be prepared.

Acknowledgements

These studies were funded by the Institut Pasteur international network through an ACIP ‘Action concertée interpasteurienne’ grant and benefit support from the Direction de la santé et du développement social of French Guiana, Martinique and Guadeloupe and from the Conseil général from Martinique. We thank Franck Berger for help with the statistical analysis.

Ancillary