Bionomics and vectorial role of anophelines in wetlands along the volcanic chain of Cameroon.

BACKGROUND: The epidemiological profiles of vector-borne diseases, such as malaria, are strongly associated with landscape components. The reduction of malaria burden in endemic and epidemic regions mainly depends on knowledge of the malaria-transmitting mosquito species, populations and behavioural characteristics, as well as malaria exposure risks. This work aimed at carrying out a holistic study in order to characterise Anopheles species in relation to human malaria in seven wetlands along the lower section of the volcanic chain of Cameroon.
RESULTS: Eight malaria vectors: Anopheles arabiensis, An. coluzzii, An. funestus (s.s.), An. gambiae, An. hancocki, An. melas, An. nili and An. ziemanni, were found biting humans. Anopheles gambiae was widespread; however, it played a secondary role in the Ndop plain where An. ziemmani was the primary vector species (79.2%). Anophelines were more exophagic (73.6%) than endophagic (26.4%), showing a marked nocturnal activity (22:00-4:00 h) for An. coluzzii and An. gambiae while An. funestus (s.s.) was mostly caught between 1:00 and 6:00 h and An. ziemanni having an early evening biting behaviour (18:00-00:00 h). Female Anopheles were mostly observed to have relative high parity rates (≥ 70%), with the exception of the Meanja site where species parity varies from 46 to 55%. Overall, the transmission level was low with entomological inoculation rates estimated to 0.7 infected bites per person per month (ib/p/mth) in Tiko and Ndop, 1.4 ib/p/mth in Mamfe and 2.24 ib/p/mth in Santchou.
CONCLUSIONS: The present study represents detailed Anopheles vector characterisation from an understudied area along the volcanic chain of Cameroon with endemic malaria transmission. The significant differences in bionomics and Anopheles species distribution within the studied wetlands, highlights the importance of providing baseline data and an opportunity to assess the outcome of ongoing malaria control interventions in the country.

Background

Malaria, transmitted by the bite of infected female Anopheles mosquitoes, is a major health burden in the Cameroon hill tracts [1-5]. The area has an estimated Plasmodium falciparum prevalence rate (PfPR) ranging from 50 to 70%, thus the disease remains the leading cause of consultation in health services [6, 7]. Although there are 48 Anopheles species recorded throughout Cameroon, only 18 are responsible for malaria transmission [8-11]. These include: Anopheles arabiensis, An. coluzzii, An. gambiae and An. funestus (s.s.) as dominant vectors species, while An. carnevalei, An. coustani, An. hancocki, An. leesoni, An. marshallii, An. melas, An. moucheti, An. nili, An. paludis, An. pharoensis, An. ovengensis, An. rivulorum-like, An. wellcomei and An. ziemanni play a secondary role.

Various factors contribute to differing malaria epidemiological profiles including altitude, topography, hydrology and land use/land cover types [12, 13]. Specifically, changes of environmental factors can impact malaria transmission by altering the microclimate of the immature stages and adult mosquitoes [14, 15], as it has been observed that the shortening of Plasmodium sporogony and vector gonotrophic cycle lead to an increase of the malaria transmission risk across highlands [16]. Furthermore, malaria prevalence is related to the existence of wetlands or swampy areas, because mosquitoes require stagnant water as their larval habitat [17, 18]. In fact, along the volcanic chain of Cameroon (VCC), there exist a considerable number of wetlands in valley bottoms of mountain massifs that are maintained by high annual precipitation and favourable surface hydrology [19]. Of further note, these volcanic mountains possess highly fertile and productive soils resulting in a concentration of human populations [20, 21].

The level of malaria transmission is determined by the interactions between Plasmodium parasites, the Anopheles vectors and the human host. Understanding adult vector population dynamics by identifying the different species, their abundance, biting behaviour and entomological inoculation rates are important steps towards effective control of malaria, with vector abundance being a key determinant of malaria transmission force [22]. Potential reductions of the malaria burden in endemic and epidemic regions depends on knowledge of the malaria-transmitting mosquito species, populations and behavioural characteristics, and malaria exposure risks. While vector populations have been extensively studied in urban [23-25], forested [26-28] and savannah [9, 29–31] areas of Cameroon, only a few studies on malaria vector populations have been conducted in some isolated wetlands across the VCC [1, 2, 4, 5]. This disparity in the level of knowledge of the malaria vectorial system (species composition, feeding behaviour, biting cycle, entomological indices of transmission) between the Cameroon hill tract and the rest of the country is mainly due to the difficult access towards those areas. In this study, sampling was carried out in seven wetlands along the lower section of the VCC in order to assess the bionomics and vectorial role of Anopheles species in relation to human malaria.

Methods

Study sites

The volcanic chain of Cameroon is a 1600 km long massif of cenozoic volcanic and subvolcanic complexes that straddles the continent-ocean boundary. It extends from the Gulf of Guinea to the interior of the African continent. In Cameroon, it starts with Mount Cameroon, extends to the Western High Plateau and northeast towards Lake Chad (see Additional file 1: Figure S1). The line strikes almost south-west to north-east, forming a swell and basin structure where the six islands and the four central volcanoes of Mount Cameroon, Kupe Manengouba, Bamboutous, Bamenda and Oku represent swells. The presence of these large land masses and several volcanic cones facilitate the occurrence of orographic uplift, which accounts for the intensive precipitation resulting on a major input water into the wetlands [32].

Many wetlands exist in Cameroon, with some located inland and across the national frontier, and others located within the coastal lowlands of mountain massifs. Some of these wetlands result from lava filled depressions, topographic channels which widen into flood plains and braided profiles, tidal activities in estuarine environments and shallow depressions beside the coast [33, 34]. For the current study, seven wetlands surrounded by five volcanic massifs along the lower section of the Cameroon volcanic line were selected (Fig. 1, Additional file 2: Table S1). These were: (i) Mount Cameroon massif: Tiko [4°07'N, 9°36'E, 64 m above sea level (masl)], Kumba (4°38'N, 9°27'E, 240 masl) and Meanja (4°43'N, 9°38'E, 305 masl); (ii) Western Highlands: Mamfe (5°46'N, 9°17'E, 126 masl); (iii) Mount Kupe Manengouba: Santchou (5°15'N, 9°50'E, 750 masl); (iv) Bamboutos: Ndop plain (5°56'N, 10°32'E, 1152 masl); and (v) Mount Oku/Ndu massif: Mbaw plain (5°55'N, 10°10'E, 400 masl).

Fig. 1

3D map of the study area

Study design and adult mosquito sampling

Repeated cross-sectional surveys were carried out to conduct this study (see Additional file 3: Figure S2). Mosquito collections were undertaken using landing catches on human volunteers (human-landing catch: HLC) during the rainy season months of April and May 2010 (beginning of rainy season for Tiko, Meanja and Santchou), September and October 2013, and October 2014 (end of rainy season for Kumba, Mamfe, Ndop and Mbaw plain). Depending on size of the volcanic massifs and/or wetlands, one to six communities were chosen as sampled stations. During five consecutive nights (from 18:00 to 6:00 h), HLCs were performed in two houses per community with one collector indoors and one outdoors (approximately 5 m from the main entrance) selected households (see Additional file 4: Table S2). Householders were asked to assist designated mosquito collectors (four men per night per sampled station) in each community for 5 days per study site. Using this strategy, the Anopheles species man biting rate (MBR) or Anopheles density per person per time unit (month) was estimated and used for the calculation of the entomological inoculation rates (EIR). All mosquitoes collected were placed in holding cups labelled by hour until they were processed for morphological identification.

Morphological identification of Anopheles species and age-grading methods

After being differentiated at the level of genus as Culex, Anopheles, Aedes or Mansonia, Anopheles mosquitoes were identified to species level using the taxonomic and identification key to female Afrotropical anophelines [35, 36]. Age-grading was based on the mosquitoes’ physiological age. More precisely, the physiological age was determined from inspecting its ovaries, based on the ovary tracheation method of Detinova [37] and as described by the simplified technique of Warrel & Gilles [38]. This dissection technique of the ovaries can differentiate parous female from nulliparous. Female Anopheles that have taken a blood meal and have laid their eggs at least once are parous, while those that have never laid eggs are nulliparous.

DNA isolation, species identification and molecular detection of Plasmodium spp. infection in Anopheles mosquitoes

Genomic DNA was individually isolated from most recurrent Anopheles specimens using a DNA extraction buffer [38]. Species were molecularly identified using sequences of the ribosomal DNA intergenic spacer (rDNA IGS) and internal transcribed spacer region 2 (ITS2) of An. gambiae (s.l.) and An. funestus (s.l.), respectively. These regions were isolated and amplified using PCR with IGS and ITS2 specific primers as previously documented [39, 40].

A restricted fragment length polymorphism (RFLP) was further performed on all the specimens diagnosed as An. gambiae (s.s.) to distinguish between molecular forms M and S [41]. The M molecular form of An. gambiae (s.s.) is presently known as An. coluzzii while the S molecular form retained the designation An. gambiae [42].

The infection status of female Anopheles was determined using a multiplex PCR for P. falciparum, P. malariae, P. vivax and P. ovale. Oligonucleotides were designed by Snounou et al. [43] and Padley et al. [44].

Data processing and statistical analysis

Data were compiled in EpiInfoTM v.3.5.3 (Centre for Disease Control and Prevention, Atlanta, USA) and imported to SPSS v.20 (IBM, Chicago, USA) and GraphPad Prism v.7.0.0 for Windows (La Jolla, California, USA) for analysis. The data of malaria vector density (including diversity, biting density, biting behaviour and biting cycle) were summarized using descriptive statistics. Chi-square test was used to compare mosquito density distributions between volcanic massifs and wetlands. T-test and two-way ANOVA (on normalized data using the logarithm procedure) followed by Bonferroni’s multiple comparison test were used to compare the mean indoor and outdoor exposure to main Anopheles vectors within and between wetlands.

Mosquito life expectancy (longevity, life span) was estimated using Davidson formula [45]:where “e” is the natural logarithm of the constant 2.7183 and P is the probability of the vector surviving for 1 day. P is calculated from the following formula:

where gc is the duration of the Anopheles gonotrophic cycle, the time (days) needed by a female mosquito to complete the processing of egg development in the ovaries from the blood meal to the time when the fully developed eggs are laid [46]. The parity rate (PR) is the rate of parous female Anopheles from the total number of ovaries dissected and observed. In this study, the gc anthropophagous species was 3 days while zoophagic species had a gc of 2 days [47]. The parity rate between Anopheles species and surveyed sites was compared using the Chi-square test.

The entomological inoculation rate (EIR) was calculated using the formula:where IR is the rate of Anopheles tested positive for the molecular detection of Plasmodium spp. infection, and MBR the man biting rate expressed as the ratio of the number of Anopheles mosquitoes captured by HLC by the total number of man-nights.

Results

Survey summary

A total of 5918 mosquitoes were collected during the entire survey period. Four genera were recorded based on the morphological identification of the specimens (Fig. 2): Aedes (5.4%), Culex (12.9%), Anopheles (38.6%) and Mansonia (43.1%). The Mount Cameroon massif had the highest number of Anopheles specimens caught (53.5%) while the Santchou valley in the Kupe Manengouba massif had the lowest contribution (6.2%). Except in Mount Cameroon (Tiko, Kumba, Meanja) and Western Highlands (Mamfe) that were dominated by Anopheles mosquitoes (75.9 and 80.8%, respectively), mosquitoes of the genus Mansonia was dominant in the Kupe Manengouba, Bamboutous and Oku volcanic massifs. For these last study sites, catches of Anopheles mosquitoes represented 10.1, 18.2 and 34.4%, respectively.

Fig. 2

Main genera of mosquitoes identified morphologically from wetlands of the different massifs of the volcanic chain of Cameroon. A significant difference based on a Chi-square test (P < 0.0001) was noted in the mosquito distribution for all sites

Anopheline species diversity and biting density

Five Anopheles morphological species leading to eight molecular species were identified (Fig. 3): An. hancocki (n = 14 specimens caught), An. nili (n = 19 specimens caught), An. ziemanni (n = 213 specimens caught), within the group of sister species An. funestus (n = 52 specimens caught), the sole species identified molecularly was An. funestus (s.s.) and within the complex An. gambiae (s.l.) (n = 1990 specimens caught), four species were identified throughout this study: An. arabiensis (1.0%), An. melas (6.7%), An. coluzzii (34.5%) and An. gambiae (56.4%). Additionally, some hybrids An. coluzzii × An. gambiae (1.3%) were also molecularly identified within the Gambiae complex in Kumba, Mamfe and Ndop sites.

Fig. 3

Diversity of Anopheles species (identified morphologically and using molecular techniques) from wetlands of major volcanic massifs across the study area. A significant difference based on a Chi-square test (P < 0.0001) was noted in anopheline abundance in all sites

Anopheles coluzzii was dominant in the western part of the study area, especially towards the sea at Tiko and Meanja, whereas An. gambiae was dominant in other sites except for the Ndop site where An. ziemanni was widely represented among night HLCs. The Mount Cameroon area had the highest species diversity [six Anopheles species out of the eight identified: An. funestus (s.s.), An. hancocki, An. nili, An. melas, An. coluzzii and An. gambiae] while only two Anopheles species (An. coluzzii and An. gambiae) were found in Mamfe towards the Western Highlands. Overall, it was noted a significant difference in the diversity of the different anopheline vectors across the study area (χ2 = 4520.1, df = 48, P < 0.0001).

Biting behaviour

The HLC collections indicated indoor and outdoor exposure of residents to host-seeking anophelines. In this study 1683 (73.6%) of the total female Anopheles specimens collected were captured outdoors, showing a strongly exophagic biting behaviour. Two-way ANOVA test showed some significant differences in the mean number of Anopheles species caught outdoors within wetlands as well as between wetlands (see Additional file 5: Table S3), whereas only 605 (26.4%) anophelines were captured indoors with no significant difference observed in the average number of Anopheles species caught within and between wetlands. Moreover, no statistical difference between indoor and outdoor exposure to Anopheles species was usually noted in the surveyed wetlands, except in the cities of Kumba and Mamfe where all specimens of the Gambiae complex were collected outside the houses (Fig. 4).

Fig. 4

Indoor and outdoor exposure to major anophelines per wetland. Anophelines were mostly collected outdoors; note that there were no specimens collected indoors in Kumba and Mamfe. t-test statistical significance was determined using the Boferonni-Dunn method for each Anopheles species biting behaviour per wetland. #P = 0.0986, *P < 0.05, **P < 0.001. Abbreviation: ns, not significant

Comparison of anophelines biting cycle showed that dominant vectors (An. gambiae and An. coluzzii) were collected both indoors and outdoors with a significant number of bites registered late at night between 22:00 h and 4:00 h (An. coluzzii: χ2 = 417.6, df = 2, P < 0.0001; An. gambiae: χ2 = 588, df = 2, P < 0.0001) as compared to An. funestus (s.s.) which was mostly (χ2 = 19.69, df = 1, P < 0.0001) captured between 1:00 h and 6:00 h (Fig. 5). Indoor species were captured between 21:00 h and midnight. Conversely, An. ziemanni had an early evening biting behaviour (18:00–00:00 h) both indoors and outdoors (χ2 = 59.95, df = 1, P < 0.0001).

Fig. 5

Biting cycles of most predominant anophelines collected indoors (upper panels) and outdoors (lower panels) per wetland. An. coluzzii and An. gambiae were the most abundant specimens sampled both indoors and outdoors; An. ziemanni bites early in the evening

Anopheles parity rate and longevity

The parity rate (parous ratio: PR) and life expectancy or life span were analysed for Tiko (An. melas and An. coluzzii), Meanja (An. funestus and An. coluzzii), Santchou (An. gambiae) and Ndop (An. ziemanni and An. gambiae). PR was used to estimate mosquito longevity based on the formula proposed by Davidson [45]. Since there was no direct observation of the gonotrophic cycle (gc) value, the mosquito life span calculation used a gc value of 2 or 3 days as previously explained.

Overall, parity ratios varied from 0.55 in An. funestus (s.s.) to 0.85 in An. ziemanni, whereas An. coluzzii, An. gambiae and An. melas had an estimated parity of 0.67, 0.73 and 0.79, respectively (Table 1). Likewise, throughout the study area, species life expectancies estimates were: 5.02 days in An. funestus (s.s.), 7.49 days in An. coluzzii, 9.53 days in An. gambiae, 12.31 days in An. ziemanni and 12.73 days in An. melas. No statistical difference was observed in the distribution of parity rate between species; however, only An. coluzzii exhibited significant different parous rates per site (χ2 = 10.25, df = 1, P = 0.0014) while comparing the Tiko and Meanja surveys.

Table 1

Parous rate and duration of life expectancy of Anopheles species from the different study sites

Volcanic massif	Wetlanda	Species	Dissected	Parous	PR	P b	Life span (days)
Mount Cameroon	Tiko	An. melas	99	78	0.79	0.92	12.73
		An. coluzzii	337	261	0.77	0.92	11.48
	Meanja	An. funestus (s.s.)	38	21	0.55	0.82	5.02
		An. coluzzii	157	72	0.46	0.77	3.86
Kupe Manengouba	Santchou	An. gambiae	116	80	0.69	0.88	8.08
Bamboutos	Ndop	An. ziemanni	181	154	0.85	0.92	12.31
		An. gambiae	31	28	0.90	0.97	28.47

aNo Anopheles dissected in Kumba, Mamfe and Mbaw plain

bNo statistical difference was observed in the distribution of parity rate between species; however, a significant difference (Chi-square test: P < 0.05) was noted in the An. coluzzii parous rate while comparing Tiko and Meanja sites

Abbreviations: PR parous ratio, P probability of daily survival.

Plasmodium infection and entomological inoculation rates

A total of 2209 Anopheles mosquitoes [20 An. arabiensis, 51 An. funestus (s.s.), 131 An. melas, 213 An. ziemanni, 685 An. coluzzii and 1109 An. gambiae] were tested for Plasmodium infection using a TaqMan assay. Only three Anopheles species were found positive during the molecular detection of Plasmodium infection, namely An. coluzzii (0.15%), An. gambiae (1.08%) and An. ziemanni (1.41%) (Table 2). Infected Anopheles were found only in four wetlands out of the seven surveyed: Tiko (An. coluzzii), Mamfe (An. gambiae), Santchou (An. gambiae) and Ndop (An. gambiae and An. ziemanni). Infection rates (IRs) varied between Anopheles species and wetlands and ranged between 0–6.9% with no statistical difference (based on Chi-square tests) found between anophelines and wetlands. Anopheles coluzzii was the species most commonly found biting humans with a man biting rate (MBR) of 328.3 bites per person per month (b/p/mth) while for An. gambiae it ranged from 7.2 b/p/mth in Ndop to 121.3 b/p/mth in Mamfe, and An. ziemanni had 49.7 b/p/mth. Thus, the level of overall malaria transmission varies between Anopheles species and wetlands. An. gambiae appears to be the most frequently competent malaria vector with entomological inoculation rates (EIRs) estimated at 0.23 infected bites per person per month (ib/p/mth) in Ndop, 1.4 ib/p/mth in Mamfe and 2.24 ib/p/mth in Santchou. However, similar EIRs (0.7 ib/p/mth) were observed for An. coluzzii in Tiko and An. ziemanni in Ndop.

Table 2

Infection rate and weekly entomological inoculation rate of Anopheles vectors in the study site

Volcanic massif	Wetland	Species	Tested	Positive	IR (%)	MBR (b/p/mth)	EIR (ib/p/mth)
Mount Cameroon	Tiko	An. coluzzii	469	1	0.21	328.3	0.70
		An. melas	131	0	0	91.7	0
		An. funestus (s.s.)	10	0	0	7.0	0
	Meanja	An. coluzzii	167	0	0	116.9	0
		An. funestus	41	0	0	28.7	0
	Kumba	An. coluzzii	26	0	0	18.2	0
		An. gambiae	357	0	0	249.9	0
Western Highlands	Mamfe	An. coluzzii	23	0	0	16.1	0
		An. gambiae	260	3	1.15	121.3	1.40
Kupe Manengouba	Santchou	An. gambiae	116	8	6.90	32.5	2.24
Bamboutos	Ndop	An. gambiae	31	1	3.22	7.2	0.23
		An. ziemanni	213	3	1.41	49.7	0.70
Oku/Ndu	Mbaw	An. arabiensis	20	0	0	4.7	0
		An. gambiae	345	0	0	80.5	0

Abbreviations: IR infection rate, MBR man biting rate as estimated by the number of bites per person per month (b/p/mth), EIR entomological inoculation rate as the number of infected bites per person per month (ib/p/mth)

Discussion

Malaria remains a significant public health problem in Cameroon. Using standard procedures of collecting and processing samples, this study demonstrates the wide range of vector composition and distribution, exposure in sampled sites, and entomological indices of transmission in seven wetlands along the downer section of the VCC.

During this study, Anopheles and Mansonia were the most predominant culicid genera found throughout the study area. The Mount Cameroon massif contributed up to 75.9% of the entire Anopheles fauna collected while Mansonia species were mostly associated with rice flood plains (Santchou, Ndop and Mbaw). These observations were contrary to other studies which associated Anopheles mosquitoes with a wide range of habitats including rice field ecosystems [48-50], whereas Mansonia species showed a preference to ponds infested with Pistia spp. which are also common in such areas [51]. In addition to the species of these genera, Aedes spp. and Culex spp. were also encountered in wetlands of the VCC. In Africa, these mosquitoes are known to transmit Wuchereria bancrofti microfilariae, the causative agents of lymphatic filariasis in Africa (mainly Anopheles spp. and Culex spp.), and viruses causing dengue fever and yellow fever (Aedes spp.) as well as Rift Valley fever (Culex spp. and Aedes spp., but also Anopheles spp. and Mansonia spp. in the enzootic cycle in East Africa).

All five morphological malaria vector taxa (leading to eight molecular species) recorded in Cameroon [8-11] were found biting humans along the volcanic chain of Cameroon. High species diversity was noted in the Mount Cameroon massif (6 out of 8 species), Kupe Manengouba and Oku massifs (4 out of 8 species each), whereas only two Anopheles species were found in the Western Highlands and Oku massifs, with observations being in accordance with previous reports [1, 4, 5].

The most widespread and competent malaria vector in sub-Saharan Africa, An. gambiae, was found to be present throughout the studied wetlands. Its distribution and predominance are consistent with previous studies [1, 9, 52, 53]. The adaptability of this species to different topographic settings has also been widely reported [4, 10]. In Ndop, although An. gambiae and An. ziemanni were both involved in malaria transmission, the dominant vector was An. ziemanni. Here, the high density of An. ziemanni was probably a consequence of the ecosystem consisting of paddy fields irrigated by water (vegetative swamps) sourcing from numerous natural lakes, which enables the proliferation of breeding sites that favour its development [5, 54]. The localised predominance of An. ziemanni has been previously observed in the same region [5], as well as in western Kenya [55] and south-central Ethiopia [54] where the swamps contain a lot of aquatic vegetation as with Ndop plain.

The two sibling species of Anopheles coluzzii and An. gambiae were spatially segregated across the study sites. This is in-line with observations made by other studies [42, 56] which report that these molecular forms are cohesive and constitute exclusive taxonomic groups across their shared range. An. coluzzii is more adapted to urbanized and polluted environments (such as Tiko and Meanja) [57], and environments having high salinity rates [58], while An. gambiae seems predominant in semi-urban, rural and arid environments (such as Kumba, Mamfe, Santchou and Mbaw plain) [59]. Hybridization between forms occurs rarely (~1%) in Central Africa [60, 61] as confirmed in our study where a low rate of An. coluzzii × An. gambiae hybrids was noted. Furthermore, 5.6% of An. coluzzii × An. gambiae hybrids in the Ndop area was noted, whereas only An. gambiae molecularly identified, opposing observations made by Tabue et al. [5] who reported An. coluzzii in the same region. Nevertheless, this difference could be justified by the fact that during this study investigators surveyed different sampled stations than the previous team. Thus, more studies should be conducted in the Ndop area in order to understand the distribution of An. gambiae (s.s.) (formerly grouping An. gambiae M and S molecular forms) species.

Other malaria vectors such as An. funestus (s.s.), An. nili and An. hancocki were recorded in the Mount Cameroon, Kupe Manengouba and Ndu massifs. These are secondary malaria vectors in forested areas [1, 2, 62, 63] even if An. funestus (s.s.) has been reported as primary malaria vector in Savannah areas [29, 30, 64]. The presence of An. melas in the coastal plain (Tiko) confirms its association with saltwater where its larvae develop [2, 61]. Anopheles arabiensis occurred exclusively in the Sudan-Savannah domain (Mbaw plain), in agreement with literature data [52, 61, 65].

Overall, the anopheline fauna were more frequently caught outdoors (73.6%) than indoors (26.4%). However, the biting behaviour should be considered for each Anopheles spp. in each wetland as there was considerable variation. For example, in Kumba and Mamfe, both An. gambiae and An. coluzzii were exclusively exophagic, while in Meanja there was both indoor and outdoor biting of these sibling species. To calculate the extent of transmission that occurs outdoors, it is important to consider the location of the human population during peak biting times, because if biting is primarily nocturnal then the majority of the humans may be indoors. This was demonstrated in south Cameroon where An. gambiae (s.l.) was found both outdoors and indoors [11], although 52–85% of An. gambiae (s.l.) bites occurred indoors as observed by Antonio-Nkondjio et al. [26] when studying malaria transmission dynamics. On the other hand, An. funestus (s.l.) has a preference for indoor-feeding but a significant percentage of bites occur outdoors [66] as well as An. ziemanni [5]. Killeen et al. [67] had recently shown that although most outdoor malaria transmission is due to a behavioural resistance mechanism, each individual mosquito does have opportunities to enter houses during the extrinsic incubation period and could be potentially exposed to indoor control tools. However, further studies are urgently needed in Kumba and Mamfe in order to assess the extent of outdoor transmission and potential vector control measures that could be implemented in those areas.

Anopheles gambiae and An. coluzzii showed a marked nocturnal activity with intense human bites taking place between 22:00 h and 4:00 h while An. funestus (s.s.) was mostly caught from 1:00 to 6:00 h. These observations are consistent with others studies [1, 11, 66] which reported that members of the An. gambiae complex and An. funestus group mostly bite late at night, between midnight and the early hours of the morning (00:00–5:00 h), contrasting with An. ziemanni whose biting activity declined throughout the night with a peak period early in the evening until midnight (18:00–00:00 h). This later species when in sympatry with An. gambiae could ensure an increased risk of malaria transmission as previously observed in Ethiopia by Kenea et al. [54]. However, there is no evidence of that the early biting behaviour observed in An. ziemanni in this study or in others is a result of behavioural modifications or shifts.

This study revealed relatively high parity rates of female Anopheles (≥ 70%) in the study area, except the Meanja site (≤ 55%) in the Mount Cameroon massif. These variations underline the great variability of parity rates observed sometimes between study sites, and within the same site depending on the species density dynamics. High parous rates could be influenced by temporal fluctuations such as the beginning of rainfall coupled to the emergence of nulliparous individuals (for example, in Tiko, Meanja and Santchou, mosquitoes were collected after the dry period) or associated with the non-application of control measures, thus leading to an accumulation of older populations with time [68], especially in the Ndop plain where An. gambiae had the highest life expectancy (28.47 days). The parous ratio and longevity of An. gambiae in Santchou was comparable to data obtained by Tchuinkam et al. [4] when studying the bionomics of Anopheles species and malaria transmission in the same area. Alternatively, the low parous rate and consequently the decrease in longevity of An. coluzzii in Meanja, could be further investigated and associated to the application of control measures and malaria endemicity level in this area.

Overall, the transmission of malaria in the wetlands of the volcanic chain of Cameroon seemed to be low as shown by an overall EIR ranging from 0.23 to 2.24 infected bites per person per month. Although An. coluzzii was the most commonly found biting humans and its sibling An. gambiae was the most frequently tested positive to Plasmodium infection, infection rates showed no significant difference between anophelines tested positive for Plasmodium infection. These observations are in-line with that of Gnémé et al. [69] who showed that there was no difference between An. coluzzii and An. gambiae in their infection susceptibility to Plasmodium falciparum. Moreover, An. gambiae, generally known to be the main malaria vector in most areas of the southern forested parts of the country [2, 70], played a secondary role in Ndop. The contribution of An. ziemanni in the Ndop plain (0.7 infected bites per month) is comparable with values obtained by Tabue et al. [5] in the same area, and comparable with its low anthropophagy as previously shown [71, 72]. However, this species has been previously reported to play a secondary role in the transmission of malaria in several eco-epidemiological settings of Cameroon [2, 9].

Conclusions

To our knowledge, this study represents the most formal and extensive work available to date of Anopheles species in wetlands along the lower section of the volcanic chain of Cameroon. Eight Anopheles species were present across the study area, all known to be potential malaria vectors in Cameroon. Anopheles gambiae was widespread and probably the main vector in the land, whereas in coastal areas An. coluzzii dominated. These dominant vector species were relayed by other species of Anopheles, sometimes of considerable local importance such as An. ziemanni in the Ndop plain. The unequal distribution of the Anopheles species within the studied wetlands, may further confirm that the occurrence of these mosquitoes truly varies according to macro- and microenvironmental differences exhibited by different bio-ecological areas. Most of the host-seeking vectors showed a preference for feeding outdoors rather than indoors which could be a challenge to the malaria control strategies which uses mainly predominantly indoor control intervention tools such as long-lasting insecticidal nets (LLINs) and indoor residual spraying (IRS). Of further challenge is the observation that these mosquitoes were found biting early in the evening as well as late in the night depending on the species when people might not be protected by LLINs. A regular update of vector species is thus necessary for effective malaria control.

Figure S1. Cameroon volcanic line. (PDF 405 kb)

Table S1. Summary information of the surveyed wetlands and environmental factors. (PDF 194 kb)

Figure S2. Flow chart of activities. (PDF 47 kb)

Table S2. Sampling parameters for mosquito collections. (PDF 185 kb)

Table S3. Two-way ANOVA statistical significance for each outdoor-biting Anopheles species within and between wetlands. (PDF 212 kb)