Introduction

Close to half of the world's population is at risk of malaria infection [1], [2]. Attempts to curtail disease transmission have focused on the development of drugs to treat infected individuals, insecticide spraying to kill the mosquito vectors, and the use of physical barriers to prevent vector-human contact. Understanding the biology of mosquito-malaria interactions may aid in the development of a new generation of vector-based measures for malaria control.

In sub-Saharan Africa, where greater than 85% of malaria-associated mortality occurs [1], the major Plasmodium vectors are A. gambiae, A. arabiensis, A. funestus, A. nili, and A. moucheti, with heterogeneities in populations occurring both geographically and seasonally [3], [4]. To date, due to practical reasons such as extensive genome information and relative ease of colonization, most studies on mosquito-Plasmodium interactions have focused on the African mosquito, A. gambiae, and the Asian mosquito, A. stephensi. However, A. funestus is a major malaria vector in certain regions of Africa [3], plays a more prominent role than A. gambiae during the dry season [5], and in areas where it co-exists with A. gambiae it has been observed to have higher infection rates [6]. Thus, understanding the vector biology of A. funestus needs to be part of successful malaria control, but currently little is known about this species. Towards this end, we and colleagues recently determined and reported the complete transcriptome sequence of A. funestus using RNA-seq next-generation sequencing technology [7].

When studying mosquito-malaria interactions, model parasite systems are often used because of the ease of manipulation and the laboratory safety afforded by using parasites incapable of infecting humans. One model parasite commonly used is P. berghei, a rodent malaria species originally isolated from the salivary glands of A. dureni and whose vertebrate host in nature is the Central African tree rat, Thamnomys surdaster [8], [9]. P. berghei can be genetically manipulated, and marked transgenic parasites are easily visualized in mosquito and mammalian tissues [10]–[14]. To date, experimental infections have shown that P. berghei is capable of completing its life cycle in A. quadrimaculatus, A. freeborni, A. stephensi, A. annupiles, A. atroparvus, A. maculipennis, and A. gambiae [15]–[21]. A. albimanus, a major malaria vector in South America, is not an efficient laboratory vector of P. berghei [19], [22], and the African mosquito A. quadriannulatus Species A, a member of the A. gambiae species complex, also is not permissive for P. berghei development [23]. Ten other anopheline species have been described as being resistant to P. berghei, including the only published report using A. funestus, which states that P. berghei is unable to infect A. funestus [18], [20]. Here, we show for the first time that P. berghei can indeed infect and complete its life cycle in random natural genotypes of the important African malaria vector A. funestus, and that infection progression proceeds with similar kinetics and efficiency to that observed in A. gambiae.

Methods

Results and Discussion

We used the F1 progeny of field-collected A. funestus females to assess whether P. berghei can complete its life cycle in this mosquito species, and we did not attempt to obtain further generations. Thus, the mosquitoes tested should represent an unbiased sample of random wild mosquito genotypes.

Following blood feeding on infected mice, numerous P. berghei oocysts were observed on A. funestus midguts as early as 7 days post-feeding, the earliest time point tested (Figure 1A). By 18–20 days post-infection, both midgut oocysts and salivary gland sporozoites were observed in dissected tissues (Figure 1B, D–E), and through the cuticle of live mosquitoes (Figure 1C). By this time several oocysts had ruptured, and 57% (9/17) of infected mosquitoes carried salivary gland sporozoites. In the hemocoel, migrating sporozoites were observed to flow with the hemolymph through the dorsal vessel in a manner similar to that recently reported in the A. gambiae-P. berghei system [12], [14]. Furthermore, sporozoites were also occasionally observed attached to the cuticle and in the appendages. When the salivary glands were closely examined, sporozoites were observed to preferentially invade the median and distal lateral lobes (Figure 1D–E). These observations are in agreement with reports in other mosquito species [25], [26], including the A. gambiae-P. berghei system [27].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Light and GFP epi-fluorescence microscopy show fluorescent Plasmodium berghei parasites developing in Anopheles funestus mosquitoes and murine erythrocytes.

A. A. funestus midgut with greater than 300 P. berghei oocysts (e.g., arrows) at 7 days post-infection. B. A. funestus midgut showing normal oocysts (e.g., arrow) and oocysts that have recently undergone rupture (e.g., arrowhead) at 20 days post-infection. C. Imaging of fluorescent parasites through the cuticle of a live A. funestus showing parasite development in the midgut (MG) and salivary glands (SG). Note that tissues presented in panels B, D, and E originated from this mosquito. D–E. A. funestus salivary glands showing that sporozoites preferentially invade the median (M) and distal lateral (DL) lobes. F–G. Blood smear from a mouse exposed to P. berghei via mosquito bite showing infected erythrocytes, indicating that A. funestus salivary gland sporozoites are infective to the vertebrate host and that the parasite can complete its life cycle inside the insect vector. Bars: A–C = 500 µm; D–E = 200 µm; F = 100 µm; G = 20 µm.

https://doi.org/10.1371/journal.pone.0061181.g001

To test the infectivity of A. funestus salivary gland sporozoites to mice, 15 infected mosquitoes (19 days post-P. berghei infection) were fed on two normal mice. One mouse became infected as determined by the visualization of blood stage parasites at 7 days post-mosquito bite (Figure 1F–G). These experiments indicate that P. berghei is capable of completing its life cycle in A. funestus.

To compare parasite development in A. funestus versus A. gambiae, parallel infections were tracked for the first 7 days post-infection (Figure 2). In the first trial, A. funestus prevalence of infection, as determined by the presence of fluorescent P. berghei oocysts on the midguts of individual mosquitoes, was 62.5% (10/16) and median oocyst intensity was 28 (mean = 47; range = 1–141). In a parallel infection using A. gambiae, prevalence of infection was 56% (20/36) and median oocyst intensity was 4 (mean = 23; range = 1–110). The second trial yielded a similar comparison: A. funestus prevalence of infection was 95% (19/20) as compared to 97% (29/30) for A. gambiae, and median oocyst intensity was 69 (mean = 120; range = 1–347) for A. funestus and 35 (mean = 56; range = 1–285) for A. gambiae. Overall, P. berghei infection prevalence was equivalent in the two mosquito species. Although both trials showed a trend for higher oocyst intensities in A. funestus, the difference was not statistically significant (Mann-Whitney P>0.05). However, due to the negative binomial distribution of oocysts in infected midguts [28], [29], large sample sizes would be necessary to yield statistically significant estimates of mean infection intensity.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. Anopheles funestus and Anopheles gambiae mosquitoes display equivalent susceptibility to Plasmodium berghei infection.

A. gambiae and A. funestus mosquitoes were fed on the same infected P. berghei-infected mouse in two replicate experiments, and midgut oocyst infections were quantified at 7 days post-infection. Each circle represents the number of midgut oocysts in an individual mosquito.

https://doi.org/10.1371/journal.pone.0061181.g002

Conclusions

This study demonstrates that P. berghei can complete its life cycle in A. funestus. Several hurdles remain to be overcome before unraveling the molecular interactions between Plasmodium parasites and A. funestus, including the lack of available diverse and robust laboratory colonies. Nevertheless, the data presented in this study, together with the understanding that A. funestus is a major vector of human malaria, illustrate that, once the reference genome sequence of A. funestus is available, studying A. funestus-Plasmodium interactions is feasible and warranted in a laboratory setting.