Bionomics of Anopheline species and malaria transmission dynamics along an altitudinal transect in Western Cameroon

Highland areas of Africa are known to be malaria hypoendemic, due to climate (low temperatures and relative humidity), which is not appropriate for anophelines development and their reproductive fitness [1]. Nevertheless, the probability that an entomological inoculation is effective (the success probability) in these regions of low transmission is higher than in holoendemic areas [2]. In fact, the highest disease risks are observed among populations exposed to low-to-moderate intensities of transmission, and mean age of disease patients increases with decreasing transmission intensity [3].

For the past decades, there has been an increase in the number of malaria epidemics, and the question is raised of the boosting of malaria transmission in African high altitude areas as a consequence of global warming [4, 5]. Moreover, there is a spread of endemic malaria into the highland fringes [6]. In 2003, it was estimated that 110 million peoples were at risk of malaria epidemics in Africa and 110 000 of these died of the disease [7]. One year later, another survey indicated a higher rate of 155 000-310 000 deaths (out of 12 million malaria episodes) attributable to epidemics [8]. Unlike in lowland environments where the potential for malaria epidemics owing to decreasing levels of natural immunity may be offset by negative impacts of urbanization on anopheline mosquito larvae, and where therefore malaria control may be simpler [9], control strategies in highland areas should be much more based on prevention of epidemics. Tools to predict and forecast malaria epidemics are therefore urgently needed.

Since malaria epidemics appear suddenly and terminate within a few months, detection based solely on increasing incidence and conducted by network of health workers in sentinel stations as applied so far [10, 11], although likely to be cost-effective [12], would not provide sufficient time for an adequate response. In fact, epidemics usually require emergency measures that must be implemented as promptly as possible in order to be effective. Moreover, a declaration of an emergency, generally based on the inability of the medical services to cope with demands of patients following malaria outbreaks must be made. Forecasts of epidemics based on indicators preceding elevated infection and disease rates in human population are therefore required.

Recently, modern tools to help decision makers to predict malaria epidemics were proposed. They rely on the use of: combinations of satellite derived climate-based data [13, 14], weather monitoring combined with disease surveillance [15], the normalized difference vegetation index (NDVI) [16], the El Niño Southern Oscillation phenomenon (ENSO) [17, 18], and soil moisture [19]. Guidelines for implementing some of these models were provided and approved [20, 21]; they defined the steps to be taken while setting up a national malaria early warning system (MEWS). Unfortunately, very few African countries have adequately qualified human resources and appropriate meteorological facilities to be able to implement such measures. Consequently, the challenge of developing new tools for malaria prevention in highlands remains.

It was demonstrated that abundance and increased density of indoor resting Anopheles vectors were positively correlated with incidence of malaria in the human population one month later, indicating that entomological indicators could be used to predict or confirm incipient epidemics as well [18]. Thus, monitoring parameters of anopheline bionomics and vectorial capacity [ie: abundance (indoor density and biting rate), feeding behaviour, gonotrophic cycle and survivorship], may provide good conclusive early warning of malaria outbreaks.

The hypothesis of our study was that there is a relationship between: altitudinal climate variations and vectorial capacity of anopheline populations. Such variations in malaria transmission parameters lead to different risk levels for malaria infections and epidemics, taking into account the little or no functional immunity in human populations of highlands.

Baseline data and detailed knowledge on the biology of each species of the vectorial system are therefore a prerequisite for implementation of any MEWS. Herein, data on anopheline species diversity and abundance, implication in malaria transmission, their survivorship, host-seeking behavior and subsequent malaria stability index are provided for three sites in Western Cameroon, spread out over an altitudinal transect, starting from a lowland plain, across a forest sheer cliff, then a highland plateau till a hilly landscape.

HLC were more efficient than PSC in reflecting the biodiversity and density of the anopheline fauna in all study sites, although the latter was carried out in twice as much houses. Several of these species (Anopheles coustani, An. paludis, An. pharoensis, An. wellcomei, and An. ziemanni) were frequently caught indoor by HLC, but never sampled by PSC. It is likely that they were outdoor resting species. Investigations should be conducted to unveil their resting behaviour which should be included in the strategic planning of malaria control in highlands. In fact, they might be secondary malaria vectors, for they were found carrying Plasmodium antigens in the forest zone of South-Cameroon, although at very low rates [35‐39]. However, they were scarce even outdoor (data not shown), compared to what is usually found in the above other cited areas of the country where microclimatic conditions are favourable, suggesting that the hostile climate in highlands restricts outdoor resting opportunities to very few hidden sites.

There was a qualitative drop in anopheline species diversity with increasing altitude, although two additional species were found on the plateau. The availability of permanent breeding sites in the valleys of Dschang such as lakes and swamps, suitable for the development of An. funestus but also An. moucheti and An. hancocki larvae, could justify the presence of these two latter species rather than altitude. It is likely that up to a certain level, availability of convenient larval habitats attenuates the suppressive effect of altitude and/or climate on the mosquito biodiversity. This antagonistic effects might have limited the number of sibling species of An. gambiae complex along the transect to a single one. Moreover, only the indoor resting species found in lowland could be captured uphill. These same species were those found at the elevated areas of the Mount Cameroon region (another highland zone in south-west of the country), where evidence for implication of Anopheles hancocki in malaria transmission was shown [40, 41]. It is likely that endophilic behaviour is necessary for anophelines to climb above a certain threshold, which in the case of Western-Cameroon is situated beyond the Bamiléké plateau at 1400 m. This observation helps to understand why the idea of using indoor resting anopheles to predict highland malaria epidemics was put forward [42].

Likewise, a quantitative drop in the density of anopheles with altitude was recorded both for HLC and PSC, also probably due to altitudinal climate variations, which empirically was incriminated to reduce their survivorship and reproductive fitness. Apart from An. hancocki, which seemed to be adapting to highland conditions, vector densities from HLC declined across the sheer cliff (650 m height), by 1.7-fold, then rapidly by 30-fold till uphill (465 m height). However, densities of samples from PSC followed a more regular decreasing slope of about 3.5-fold along the transect. This spatial evolution of the densities indicated a moderate decline rate compared to the 50% per 86 m rise in altitude observed in Mount Kilimanjaro [26].

At equal altitude, the maximum density of mosquitoes was always recorded in the SDS and the minimum in the MDS, probably due to both the availability of puddles and the suitability of temperatures. During the MDS, malaria transmission was maintained in Dschang and it was essentially due to An. funestus, whose sporozoite rate reached 4.2%. An. funestus larvae were found to develop in permanent ponds and lake sides, especially widespread around Dschang city. Absence of such permanent breeding habitats in Santchou resulted in a more seasonal pattern of malaria transmission with a steep drop in biting density of all anopheline species during the MDS. Even the river Nkam which meanders around and then across the village of Santchou did not yield enough amount of An. nili as presumed from observations in South-Cameroon [37], probably due to absence of forest ecosystem. A striking increase in biting rate of An. ziemanni was noted at the middle of the MRS both in Santchou and Dschang, indicating a relationship between the rainfall and the density of this species.

As in most localities of sub-Saharan Africa, An. gambiae and An. funestus were the malaria vectors along the transect. An. funestus supplemented malaria transmission and bridged the gap to compensate for lack of malaria transmission which is normally induced by the microclimatic conditions of highlands during the dry season. The vectorial system was not as complex as in the South, where particular effective vectors have adapted to specific niches with population peaks occurring at different times to ensure continuous malaria transmission [35, 36, 38]. There was a global reduction of about 1.5-fold in malaria transmission level, from the holoendemic lowland to the mesoendemic highland plateau, and this EIR rapidly turned to zero in the hypoendemic zone uphill. This confirms the fact that elevated zones are areas of low or absence of malaria transmission [6, 43]. The reduction rate was similar to the one observed in a transect on Mount Cameroon, where the transmission intensity also decreased gradually with increasing altitude, though a third anopheline species was involved in malaria transmission [40, 41]. This reduction was moderate compared to the decreasing rate at similar altitudes in Tanzania [44, 45]. The low reduction rate at which malaria transmission intensity declines with increasing altitude in Cameroon indicates a potential aptitude of An. gambiae and An. hancocki to climb uphill and adapt to highland environment. So far, annulment of malaria transmission occurred beyond 1400 m.

The human blood rate was the highest, scoring 92.5% of the overall total blood meal analyzed, followed in each site by the most available domestic animal, especially those spending their night indoor. Mixed blood meals were recorded only in the lowland, and the overall rate of blood detected to be from more than one host species was very low (0.3%), compared to the 18% obtained in Senegal [46]. There was a significant reduction in the HR from the plain across the plateau until uphill, indicating that anophelines exhibited a preference for humans to a higher extent in lowlands than in highlands. However, an attempt to separate the effect of host availability and attractiveness from the effect of altitudinal climate changes is necessary, since it was suggested that the degree of anthropophagy of An. gambiae has an innate olfactory basis [47].

In the lowlands of tropical forest, the gonotrophic cycle of An. gambiae takes 2 to 3 days. This duration was assessed by mark-release-recapture technique, a method feasible only in localities of high biting rate [48, 49]; which is not the case in highlands. An attempt to estimate this entomological factor was carried out by a different method, successfully used before in Tanzania [26]. The duration of gonotrophic cycle was consistently found to be one day shorter for An. gambiae as compared to An. funestus at equal altitude. Although this period of time increased with altitude in both An. gambiae and An. funestus, there was no effect on the daily survival probability and consequently on life expectancy of both species. In fact, the significant decrease in the parity rate of both species from the lowlands to the highlands was compensated by an increase in the duration of gonotrophic cycle. As a matter of fact, the calculated survivorship and the life expectancies were almost constant with increasing altitude. Thus, the question of how altitude and/or climate influence mosquito population size remains. We argue that the stage-specific effect of altitude and/or climate on the longevity of anopheles is therefore not on the adults, but more likely on aquatic stages whose breeding habitats directly face the altitudinal climate constraints. Adult malaria vectors may be alleviated from the burden of hostile climate in highland sites by their endophilic behaviour, as inside houses had a microclimate warmer than outside. This explains why only indoor resting species were found uphill.

Spatial variations in malaria transmission level indicated lower level on the plateau as compared to the lowland plain, and complete absence of transmission on the hill site. The malaria immune status of the local human populations therefore decreases with increasing altitude. Although there was no epidemic in these highland sites for the two years of survey, data on the indoor resting density and malaria stability, highlight the fact that this highland area of Western-Cameroon is prone to malaria outbreak in case of climate change. In fact, a density of 0.25 An. gambiae/he was said to be associated with epidemic transmission in East-Africa [42], and stability index between 0 and 0.5 to characterize unstable malaria settings [28]. Epidemiological studies based on longitudinal entomological and parasitological surveys, associated to clinical surveillance are ongoing in the same sites in order to establish the threshold values and ranges associated with an eventual epidemic in Western-Cameroon highlands.