Transmission from humans to mosquitoes is a vital step in the malaria parasite life cycle. Transmission through the mosquito vector represents a bottleneck where parasite populations shrink from millions in the human body to as few as one in the mosquito [1]. Thus, malaria transmission is vulnerable to interruption when transiting from the human to the mosquito host [2]. This transition can be studied by direct membrane feeding assays (DMFA). During DMFA, mosquitoes feed through a membrane on blood kept warm via water-jacketed glass feeders [3] including blood harvested from humans with circulating malaria parasites, [4] to either study the parasites development in the mosquito [4] or to test interventions that disrupt their development hence, interrupting transmission. [5]
The mosquito blood feeding rate, i.e., the proportion of mosquitoes that successfully ingest blood is an important determinant of overall infection success. The success rate of ingesting blood from a membrane feeder can vary depending on the mosquito species, whether the mosquitoes were collected in the wild [6] or reared in a colony [7], as well as the level of adaptation of the colony.
Blood feeding rates also depend on the experimental conditions under which the DMFAs are conducted including i) the duration of starvation before exposure, ii) the starving conditions (access to water or no access to water), iii) the type of membrane used, iv) the amount of time mosquitoes are allowed to feed v) the mosquito age, vi) illumination level, vii) the blood volume in the feeder, viii) the density of mosquitoes attempting to feed and ix) water bath temperature during DMFA (Fig. 1).
However, membrane feeding studies have been conducted with a range of conditions and with varied feeding success [8, 9]. Thus, there is a need to optimize DMFA conditions for each colony mosquito species.
Starving conditions is a key component that greatly impacts mosquito feeding rates and a balance needs to be established between starving the mosquitoes for too long, thereby increasing mosquito mortality or affecting their fitness [10], and not starving for long enough so mosquitoes only partly feed or not at all.
Most studies describe dry starving for durations from 5–36 h [11–14] while other studies performed starving where the mosquitoes had access to water for 12 h [15, 16]. A study conducted by Coulibaly and colleagues compared the feeding rates of mosquitoes dry starved 8 h, 14 h and 20 h and concluded that mosquitoes starved 8–14 h yielded significantly higher feeding rates than mosquitoes starved 20 h. [10] However, most studies did not directly report the impact of starving on the feeding rate.
Membranes take the role of an artificial skin in the feeding experiments. An ideal membrane will yield the highest feeding rates in the shortest period of time. Parafilm and natural membranes such as Baudruche, sausage casing, chicken skin or rat skin have been used [6, 17]. Natural membranes which closely mimic the skin resulted in the highest feeding rates followed by Baudruche membrane which is derived from bovine cecum and finally Parafilm, a wax synthetic membrane [3]. Most studies reported using Baudruche membrane [8, 18, 19] while others used Parafilm membrane [14, 20]. Interestingly, a study done by Coulibaly and colleagues showed that there was no significant difference between the feeding rates, survival and infection rates from feeding experiments with either Baudruche or Parafilm membranes, for Anopheles coluzzii mosquitoes [10].
Mosquitoes 2–8 days post-emergence have been used in different studies [6, 7, 10, 12, 14, 15, 21–23]. The main consideration in this is that mosquitoes are fed at an early age so that they survive for the required duration for either oocysts [8, 9, 14, 24, 25] or sporozoites [8, 24, 25] to develop. Coulibaly and colleagues compared the feeding rate for An. coluzzii mosquitoes between 3 days and 9 days post emergence and found that 3 day old mosquitoes had a significantly higher feeding rate compared to 6 and 9 day old mosquitoes [10].
Mosquito density is another factor that may influence the mosquito feeding rate. Rutledge and colleagues observed that having more mosquitoes per cage can result in lower feeding rates [3] and crowding, making handling, especially removing of unfed mosquitoes, difficult. Vallejo and colleagues observed that 100 An. coluzzii mosquitoes per cage (or 1 mosquito per 5 cm2) resulted in the highest P. vivax infection prevalence after DMFA [25]. However, the study did not report on the feeding rate of the different mosquito densities in relation to infection success.
Not much has been reported also with respect to the impact of the other parameters listed above on the feeding rates. Much of the focus is on the infection rates. As such the focus of this study was to determine the optimal feeding conditions for Anopheles farauti s.s colony mosquitoes in order to maximise the feeding rate during the DMFA.