Medical and Veterinary Entomology (2016) 30, 365–368
doi: 10.1111/mve.12171
S H O R T C O M M U N I C AT I O N
Shift in species composition in the Anopheles gambiae complex after implementation of long-lasting insecticidal nets in Dielmo, Senegal S. S O U G O U F A R A 1,2 , M. H A R R Y 3,4 , S. D O U C O U R É 1 , P. M. S E M BÈ N E 2 and C. S O K H N A 1 1
Unité de Recherche sur les Maladies Infectieuses Tropicales Emergentes (URMITE), Unité Mixte de Recherche (UMR), Centre National de la Recherche Scientifique (CNRS) 6236, Institut de Recherche pour le Développement (IRD) 198, Aix Marseille Université, Campus Universitaire IRD de Hann, Dakar, Senegal, 2 Département de Biologie Animale, Faculté des Sciences et Techniques, Université Cheikh Anta Diop de Dakar, Dakar Fann, Senegal, 3 Unité de Formation et de Recherche (UFR) Sciences, Université Paris-Sud, Orsay, France and 4 UMR Évolution, Génomes, Comportement, Écologie (EGCE), CNRS-IRD Université Paris Sud, Institut Diversité, Écologie et Évolution du Vivant (IDEEV), Université Paris-Saclay, Gif-sur-Yvette, France
Abstract. Long-lasting insecticidal nets (LLINs) and indoor residual spraying (IRS) are the cornerstones of malaria vector control. However, the effectiveness of these control tools depends on vector ecology and behaviour, which also largely determine the efficacy of certain Anopheles mosquitoes (Diptera: Culicidae) as vectors. Malaria vectors in sub-Saharan Africa are primarily species of the Anopheles gambiae complex, which present intraspecific differences in behaviour that affect how they respond to vector control tools. The focus of this study is the change in species composition in the An. gambiae complex after the implementation of LLINs in Dielmo, Senegal. The main findings referred to dramatic decreases in the proportions of Anopheles coluzzii and An. gambiae after the introduction of LLINs, and an increase in the proportion of Anopheles arabiensis. Two years after LLINs were first introduced, An. arabiensis remained the most prevalent species and An. gambiae had begun to rebound. This indicated a need to develop additional vector control tools that can target the full range of malaria vectors. Key words. Anopheles, insecticide-treated bednets, malaria, vector composition.
Over 90% of deaths from malaria have occurred in sub-Saharan Africa [World Health Organization (WHO), 2014]. The last decade has seen impressive progress in malaria control in this area, primarily because of several concurrent control initiatives, particularly antivectorial programmes that have integrated protocols for the use of long-lasting insecticidal nets (LLINs) and indoor residual spraying (IRS) more heavily. Current estimates indicate that 50% of the human population of sub-Saharan Africa is now protected by at least one vector control intervention (WHO, 2014). The use of LLINs and IRS has become markedly more prevalent in sub-Saharan Africa. However, both tools are at risk of becoming less effective if the main malaria
vectors develop widespread physiological and behavioural resistance to insecticide components (Ranson et al., 2011; Okorie et al., 2015). Anopheles gambiae sensu stricto, which has recently been split into Anopheles coluzzii and An. gambiae (Coetzee et al., 2013), and its sibling species Anopheles arabiensis are among the most efficient malaria vectors in sub-Saharan Africa. These three species belong to the An. gambiae complex, which includes other species that are morphologically indistinguishable but present different bioecological features that impact their effectiveness as malaria vectors. Anopheles coluzzii, An. gambiae and An. arabiensis are the most
Correspondence: Dr Seynabou Sougoufara, URMITE, UMR, CNRS 6236, IRD 198, Aix Marseille Université, Campus Universitaire IRD de Hann, Dakar 18524, Senegal. Tel.: + 221 33 849 3523; Fax: + 221 33 832 4307; E-mail:
[email protected] © 2016 The Royal Entomological Society
365
366
S. Sougoufara et al.
Table 1. Numbers of specimens identified using the protocol of Wilkins et al. (2006) for each species of Anopheles in each year. Specimens, n (%∗) (95% confidence interval) Year
An. coluzzii
An. gambiae
Hybrids
An. arabiensis
Total
2006 2008 2010 Total
104 (57.14%) (49.60–64.38) 32 (13.67%) (9.67–18.91) 19 (8.44%) (5.30–13.07) 155 (24.18%) (20.95–27.72)
35 (19.23%) (13.93–25.87) 7 (2.99%) (1.32–6.32) 70 (31.11%) (25.21–37.66) 112 (17.47%) (14.66–20.69)
4 (2.20%) (0.71–5.89) 1 (0.43%) (0.02–2.73) 1 (0.44%) (0.02–2.84) 6 (0.94%) (0.38–2.13)
39 (21.43%) (15.85–28.24) 194 (82.91%) (77.33–87.38) 135 (60.00%) (53.26–66.39) 368 (57.41%) (53.47–61.26)
182 (100%) 234 (100%) 225 (100%) 641 (100%)
∗Percentage of all species sampled.
widespread members of the An. gambiae complex and often occur in sympatry (Fontenille & Simard, 2004). The first two species, An. coluzzii and An. gambiae, are highly anthropophilic, endophagic and endophilic. Thus, indoor interventions work best to decrease the risk for exposure to infectious bites and to lower malaria transmission (Eckhoff, 2013). However, An. arabiensis exhibits greater behavioural plasticity in feeding behaviour, and often feeds on cattle and remains outdoors (Killeen, 2013). Consequently, it is likely to have lower vectorial capacity and is exposed to less insecticide indoors than its sibling species An. coluzzii and An. gambiae. Therefore, in areas in which these species coexist, the implementation of vector control strategies may lead to a change in the composition of the anopheline fauna as high rates of mortality in species that are in contact with insecticides leave their ecological niche open for species that have the behavioral plasticity to fill that niche. The present study focuses on the vector species composition of the An. gambiae complex after the implementation of LLINs (active ingredient: deltamethrin; PermaNet® 2.0; Vestergaard Frandsen Group SA, Lausanne, Switzerland) in July 2008 in Dielmo, a rural Senegalese village located 280 km southeast of Dakar. The rainy season lasts from June to October. A pilot project began in Dielmo in 1990 to investigate the determinants of malaria transmission and has continued since then. Entomological studies are conducted and adult mosquitoes are collected using human landing catches indoors and outdoors at two sentinel sites over two or three consecutive nights per month between 19.00 hours and 07.00 hours (Trape et al., 1994). The households selected in the village have remained unchanged over the years. More details of the pilot project can be found in Sokhna et al. (2013) and Trape et al. (2014). In the present study, 182 Anopheles mosquitoes were captured in September and October 2006 (2 years before the implementation of LLINs) using 24 human landing catches, 234 mosquitoes were collected during July and October 2008 (the time of LLIN implementation) using 16 human landing catches, and 225 were collected in September and October 2010 (2 years after the implementation of LLINs) in 24 human landing catches. Rainfall data recorded during the study periods showed total rainfall of 583.4 mm in 2006, 838.4 mm in 2008 and 745.6 mm in 2010 (Trape et al., 2014). All Anopheles mosquitoes sampled were identified to species by polymerase chain reaction using the methods of Wilkins et al. (2006). To compare bite dynamics and to describe the composition of species in the An. gambiae complex, a generalized linear model (GLM) with the quasi-Poisson family was fitted. The model took into account species (An. coluzzii, An. gambiae, An. arabiensis) and year (2006, 2008, 2010) as
explanatory factors. The response variable was the number of mosquito bites. The number of human-nights was entered in the model as an offset option to allow the model to provide estimates that can be used to compare human biting rates, which refer to the number of mosquitoes captured on human bait divided by the number of person nights during the sampling period. Data analysis was performed using R Version 3.2.1 (R Foundation for Statistical Computing, Vienna, Austria). The proportions of each species collected were calculated using data for each year and the corresponding 95% confidence intervals (CIs) provided by the prop.test function in R. A total of 641 specimens were identified to species, comprising mainly An. coluzzii, An. gambiae and An. arabiensis. Anopheles arabiensis was the most common species (57.41%) of the An. gambiae complex sampled in Dielmo across all years combined (Table 1). Notably, just after the implementation of LLINs in 2008, 82.91% of individuals sampled were An. arabiensis, up from just 21.43% in 2006. By 2010, the proportion of collections represented by An. arabiensis had decreased to 60.00%. Anopheles coluzzii was the most prevalent species in 2006, before the implementation of LLINs (57.14%); however, its density decreased gradually from 2008 (13.67% of collections), when LLINs were first implemented in Dielmo, to 2010 (8.44% of collections). The frequency of An. gambiae decreased between 2006 (19.23% of collections) and 2008 (2.99% of collections) and increased to a much higher density in 2010 (31.11% of collections). Only six An. coluzzii/An. gambiae hybrids (0.94%) were recorded. The number of bites recorded varied significantly according to species (GLM quasi-Poisson family: 𝜒 2 = 9.597, d.f. = 2, P < 0.01) and year (GLM quasi-Poisson family: 𝜒 2 = 37.725, d.f. = 2, P < 0.001). The species × year interaction was also significant (GLM quasi-Poisson family: 𝜒 2 = 45.869, d.f. = 4, P < 0.001). Indeed, before LLIN implementation in Dielmo, in 2006, An. coluzzii was the most prevalent species and averaged 4.33 bites per person per night (b.p.n.). Anopheles gambiae and An. arabiensis were nonetheless present, with means of 1.45 b.p.n. and 1.62 b.p.n., respectively (Fig. 1). After LLIN implementation, in 2008, a shift in vector composition was noted, with human biting rates for An. arabiensis increasing to an average of 12.12 b.p.n. Anopheles gambiae was rarely captured in 2008 (0.44 b.p.n.) and An. coluzzii occurred at very low densities (2.00 b.p.n.). In 2010, at 2 years after the implementation of LLINs, An. gambiae had recolonized in Dielmo. Anopheles arabiensis was less abundant than in 2008, but its human biting rate did not decline to 2006 levels. The aggression of An. coluzzii continued to decline and its human
© 2016 The Royal Entomological Society, Medical and Veterinary Entomology, 30, 365–368
Species composition of An. gambiae complex
Fig. 1. Variations in mean human biting rates per night of Anopheles coluzzii, Anopheles gambiae and Anopheles arabiensis populations in 2006 [before the implementation of long-lasting insecticidal nets (LLINs)], 2008 (at LLIN implementation) and 2010 (2 years after LLIN implementation).
biting rate dropped from 2.00 b.p.n. in 2008 to 0.79 b.p.n. in 2010. Long-lasting insecticidal nets are front-line tools used in malaria vector control programmes in sub-Saharan Africa. An estimated 44% of the at-risk population were sleeping under insecticide-treated mosquito nets by 2013, compared with just 2% in 2004 (WHO, 2014). Since the widespread implementation of these indoor interventions in Africa, many studies have focused on the status of Anopheles susceptibility to insecticides. However, studies on how tools like LLINs are linked to changes in feeding and resting behaviours or to changes in biting time of Anopheles mosquitoes (Yohannes & Boelee, 2012; Sougoufara et al., 2014) shed light on the heterogeneity of Anopheles vectors in Africa and on how readily these species can adapt to control measures. Indeed, the scaling up of intradomiciliary tools can influence the composition of Anopheles vector populations because selective pressure reduces the abundance of sensitive populations. In turn, the abundances of Anopheles populations that are predisposed to feeding outdoors or on non-human hosts may increase as these species fill the ecological niches left open by more sensitive populations. In this study, variable densities of An. coluzzii, An. gambiae and An. arabiensis were observed in 2006, before LLINs were used in Dielmo. This is consistent with previous studies conducted in the same location, which reported that An. gambiae s.s. and An. arabiensis are sympatric in Dielmo and that An. gambiae s.s. is most prevalent during the rainy season, whereas An. arabiensis is present throughout the year and peaks in abundance in the rainy season (Fontenille et al., 1997). In the present study, after LLIN implementation, An. arabiensis was the most prevalent species, probably as a result of its ability to feed outdoors and on cattle. Environmental factors such as rainfall may influence the presence of An. arabiensis during the dry period. Rainfall in 2008 was heavier than in 2006, which should have favoured the presence of An. coluzzii and An. gambiae in Dielmo given that these species are usually more abundant during the wet season. However, only An. arabiensis increased in abundance in 2008. Similar species shifts in the An. gambiae complex were reported in 2009 in a Tanzanian village, in which An. arabiensis became the most prevalent species
367
and was responsible for residual malaria transmission when 47% of the community were using LLINs (Russell et al., 2011). Anopheles gambiae s.s. was the most prevalent species in the pre-LLIN period in that study (Russell et al., 2011). By 2008 in Dielmo, 75.6% of inhabitants declared that they always used LLINs (Trape et al., 2014), which can reduce human–vector contact and increase mortality in species such as An. coluzzii and An. gambiae that are more restricted in their host feeding choice (Fontaine et al., 2015). Just after the implementation of LLINs in 2008, prevalences of malaria in Dielmo fell from 36.8% in 2006 to 12.3% in 2008 in children aged 0–14 years, and from 27.6% in 2006 to 9.0% in 2008 in people aged 15 years or older (Trape et al., 2014). This could be attributed mainly to the prompt treatment of clinical malaria cases with artemisinin-based combination therapy, and to high levels of usage of LLINs that decreased human–vector contact. However, the persistence of An. arabiensis during this period highlights the behavioural plasticity of this species in maintaining its survival and supporting its reproduction by taking blood meals on non-human hosts, thus decreasing the transmission of malaria to people in Dielmo. The less anthropophilic, more zoophilic species An. arabiensis was also reported to be the dominant species in a village located 1.5 km from Dielmo (Niang et al., 2013), probably as a result of the widespread use of mosquito nets and easier access to alternatives to human hosts. In the present study, a resurgence of An. gambiae and a gradual decrease in An. coluzzii were observed 2 years after the introduction of LLINs, at a point at which 58.3% of the inhabitants of Dielmo claimed to always use LLINs. The resurgence of An. gambiae in 2010 may be attributed to insecticide resistance. Indeed, in Dielmo, Trape et al. (2011) reported that WHO bioassay tests performed in Anopheles populations demonstrated low levels of susceptibility to pyrethroids including deltamethrin, with mortality rates ranging from 43 to 63%, compared with high susceptibility to fenitrothion and bendiocarb, with mortality reaching 100%. Furthermore, the frequency of knock-down resistance, which confers resistance to pyrethroids, increased from 8% in 2007 to 48% in 2010 in An. gambiae s.l. populations (Trape et al., 2011). However, Trape et al. (2011) did not specifically show which populations of the An. gambiae complex carry these resistant alleles. Anopheles arabiensis remained the most prevalent species in 2010, although its density was lower than in 2008. One possible explanation may refer to larval competition in Anopheles breeding sites. Indeed, Gimnig et al. (2001) reported that larval habitats of An. gambiae s.s. and An. arabiensis overlapped. Moreover, Mutuku et al. (2011) showed that An. arabiensis became much more prevalent than An. gambiae s.s. in breeding sites after the introduction of insecticide-treated bednets, with higher proportions of An. arabiensis larvae observed in several villages in Kenya. In Dielmo, these three species may share the same breeding sites; therefore, larval competition with An. coluzzii and An. gambiae is probably a limiting factor for An. arabiensis populations. These findings underscore the relevance of investigating the spatial heterogeneity of larval breeding sites of species of the An. gambiae complex in Dielmo and of monitoring the status and evolution of insecticide resistance in these species. These factors could be used to
© 2016 The Royal Entomological Society, Medical and Veterinary Entomology, 30, 365–368
368
S. Sougoufara et al.
improve current control tools and to develop alternative control strategies. The observation of shifts in species composition does not mean that vector control tools have failed because these tools may have dramatically reduced the densities of those vectors that are most often in contact with human hosts. As the resurgence of An. gambiae and the persistence of An. arabiensis observed in 2010 coincided with the decrease in the use of LLINs, higher usage of non-pyrethroid bednets must be promoted in Dielmo, but combined with IRS strategies. Furthermore, in the current context of malaria elimination with the ultimate goal of eradicating this disease, the combination of sustainable and appropriate vector control tools such as IRS and bednet usage must be generally adopted to target the broadest range of Anopheles populations involved in malaria transmission.
Acknowledgements This publication was made possible through support provided by the Institut de Recherche pour le Développement–Direction des Programmes de Recherche et de la Formation au Sud (IDR-DPF). The authors thank Florence Mougel, UMR EGCE (Evolution, Génomes, Comportement, Ecologie) CNRS IRD-Université Paris-Sud, IDEEV, Université Paris-Saclay, Gif-sur-Yvette Cedex, France, for her help with data analysis. References Coetzee, M., Hunt, R.H., Wilkerson, R., della Torre, A., Coulibaly, M.B. & Besansky, N.J. (2013) Anopheles coluzzii and Anopheles amharicus, new members of the Anopheles gambiae complex. Zootaxa, 3619, 246–274. Eckhoff, P. (2013) Mathematical models of within-host and transmission dynamics to determine effects of malaria interventions in a variety of transmission settings. The American Journal of Tropical Medicine and Hygiene, 88, 817–827. Fontaine, M.C., Pease, J.B., Steele, A. et al. (2015) Extensive introgression in a malaria vector species complex revealed by phylogenomics. Science, 347, 1258524. Fontenille, D. & Simard, F. (2004) Unravelling complexities in human malaria transmission dynamics in Africa through a comprehensive knowledge of vector populations. Comparative Immunology, Microbiology and Infectious Diseases, 27, 357–375. Fontenille, D., Lochouarn, L., Diagne, N. et al. (1997) High annual and seasonal variations in malaria transmission by anophelines and vector species composition in Dielmo, a holoendemic area in Senegal. The American Journal of Tropical Medicine and Hygiene, 56, 247–253. Gimnig, J.E., Ombok, M., Kamau, L. & Hawley, W.A. (2001) Characteristics of larval anopheline (Diptera: Culicidae) habitats in Western Kenya. Journal of Medical Entomology, 38, 282–288. Killeen, G.F. (2013) A second chance to tackle African malaria vector mosquitoes that avoid houses and don’t take drugs. The American Journal of Tropical Medicine and Hygiene, 88, 809–816.
Mutuku, F.M., King, C.H., Mungai, P. et al. (2011) Impact of insecticide-treated bed nets on malaria transmission indices on the south coast of Kenya. Malaria Journal, 10, 356. Niang, E.H.A., Touré, A., Ngom, E.H.M. et al. (2013) Malaria transmission pattern in an area selected for clinical trials in the Sudanian area of Senegal (West Africa). Journal of Tropical Medicine, 2013, 907375. Okorie, P.N., Ademowo, G.O., Irving, H., Kelly-Hope, L.A. & Wondji, C.S. (2015) Insecticide susceptibility of Anopheles coluzzii and Anopheles gambiae mosquitoes in Ibadan, south-west Nigeria. Medical and Veterinary Entomology, 29, 44–50. Ranson, H., N’Guessan, R., Lines, J., Moiroux, N., Nkuni, Z. & Corbel, V. (2011) Pyrethroid resistance in African anopheline mosquitoes: what are the implications for malaria control? Trends in Parasitology, 27, 91–98. Russell, T.L., Govella, N.J., Azizi, S., Drakeley, C.J., Kachur, S.P. & Killeen, G.F. (2011) Increased proportions of outdoor feeding among residual malaria vector populations following increased use of insecticide-treated nets in rural Tanzania. Malaria Journal, 10, 80. Sokhna, C., Mediannikov, O., Fenollar, F. et al. (2013) Point-of-care laboratory of pathogen diagnosis in rural Senegal. PLoS Neglected Tropical Diseases, 7, e1999. Sougoufara, S., Diédhiou, S.M., Doucouré, S. et al. (2014) Biting by Anopheles funestus in broad daylight after use of long-lasting insecticidal nets: a new challenge to malaria elimination. Malaria Journal, 13, 125. Trape, J.F., Rogier, C., Konate, L. et al. (1994) The Dielmo project: a longitudinal study of natural malaria infection and the mechanisms of protective immunity in a community living in a holoendemic area of Senegal. The American Journal of Tropical Medicine and Hygiene, 51, 123–137. Trape, J.F., Tall, A., Diagne, N. et al. (2011) Malaria morbidity and pyrethroid resistance after the introduction of insecticide-treated bednets and artemisinin-based combination therapies: a longitudinal study. The Lancet Infectious Diseases, 11, 925–932. Trape, J.F., Tall, A., Sokhna, C. et al. (2014) The rise and fall of malaria in a west African rural community, Dielmo, Senegal, from 1990 to 2012: a 22-year longitudinal study. The Lancet Infectious Diseases, 14, 476–488. Wilkins, E.E., Howell, P.I. & Benedict, M.Q. (2006) IMP PCR primers detect single nucleotide polymorphisms for Anopheles gambiae species identification, Mopti and Savanna rDNA types, and resistance to dieldrin in Anopheles arabiensis. Malaria Journal, 5, 125. World Health Organization (2014) World Malaria Report 2014. WHO, Geneva. Yohannes, M. & Boelee, E. (2012) Early biting rhythm in the Afro-tropical vector of malaria, Anopheles arabiensis, and challenges for its control in Ethiopia. Medical and Veterinary Entomology, 26, 103–105. Accepted 15 January 2016 First published online 5 April 2016
© 2016 The Royal Entomological Society, Medical and Veterinary Entomology, 30, 365–368
