Functional integration of the circulatory, immune, and respiratory systems in mosquito larvae: pathogen killing in the hemocyte-rich tracheal tufts
© League and Hillyer. 2016
Received: 25 July 2016
Accepted: 5 September 2016
Published: 19 September 2016
As both larvae and adults, mosquitoes encounter a barrage of immune insults, ranging from microbe-rich communities in larval habitats to ingested blood-borne pathogens in adult blood meals. Given that mosquito adults have evolved an efficient means of eliminating infections in their hemocoel (body cavity) via the coordinated action of their immune and circulatory systems, the goal of the present study was to determine whether such functional integration is also present in larvae.
By fluorescently labeling hemocytes (immune cells), pericardial cells, and the heart, we discovered that fourth instar larvae, unlike adults, contain segmental hemocytes but lack the periostial hemocytes that surround the ostia (heart valves) in abdominal segments 2–7. Instead, larvae contain an abundance of sessile hemocytes at the tracheal tufts, which are respiratory structures that are unique to larvae, are located in the posterior-most abdominal segment, and surround what in larvae are the sole incurrent openings for hemolymph entry into the heart. Injection of fluorescent immune elicitors and bacteria into the larval hemocoel then showed that tracheal tuft hemocytes mount rapid and robust immune responses against foreign insults. Indeed, green fluorescent protein-labeled Escherichia coli flowing with the hemolymph rapidly aggregate exclusively at the tracheal tufts, where they are killed within 24 h post-infection via both phagocytosis and melanization.
Together, these findings show that the functional integration of the circulatory, respiratory, and immune systems of mosquitoes varies drastically across life stages.
Insects have evolved powerful innate immune responses to neutralize infectious agents [1, 2]. Pathogens, including bacteria, viruses, malaria parasites, and fungi, invade the insect hemocoel (body cavity) after ingestion and penetration of the midgut, through breaches in the cuticle, or through the tracheal system. Upon entering the hemocoel, cellular and humoral immune factors kill pathogens by phagocytosis, melanization, lysis, and other mechanisms [2, 3]. The primary immune cells that drive these processes are the hemocytes, which are found both circulating with the hemolymph and attached to tissues (sessile) [4–6].
Immune responses in the hemocoel occur in a dynamic space, where hemocytes, humoral immune factors, and pathogens are propelled through the body by the action of an open circulatory system [2, 7]. This circulation is primarily mediated by the contractile action of a muscular dorsal vessel that comprises two distinct regions: the abdominal heart and the thoracic aorta [1, 8]. Because circulatory currents move both pathogens and immune components, functional integration is expected to exist between the immune and circulatory systems. Such integration was recently documented in the adult stage of the malaria mosquito Anopheles gambiae, where cellular immune responses are rapid and inducible on the surface of the heart . Specifically, in adult mosquitoes, hemolymph enters the lumen of the dorsal vessel via valves, called ostia, that are located in the anterior portion of abdominal segments 2–7 [10, 11]. A population of sessile hemocytes, called periostial hemocytes, is always present on the surface of the heart in the regions that flank the ostia, where they phagocytose pathogens as they are swept with the hemolymph toward the heart [9, 12, 13]. This initial capture of circulating pathogens induces the migration of additional hemocytes to each periostial region, where they aggregate with the initial population of first responders and amplify the phagocytosis response. Although heart-associated immune aggregations have been observed in multiple mosquito studies [9, 12–19], they are not unique to this group of insects. Such aggregations have also been directly or indirectly observed in Drosophila melanogaster adults [20–27], though these studies did not seek to link immune responses with the functional mechanics of the heart or with hemolymph flow.
In the present study we assessed whether the circulatory and immune systems of mosquito larvae interact during the course of an infection, and describe a pattern that is remarkably different from what is seen in adults. Specifically, we show that, unlike in adults, infection of a mosquito larva does not induce the aggregation of hemocytes and pathogens in the periostial regions of the heart. Instead, infection induces the rapid aggregation and killing of pathogens on the surface of the tracheal tufts, which are respiratory structures that are unique to larvae and associate with the posterior terminus of the heart in the eighth abdominal segment. In a process that is functionally analogous to what is seen in the periostial regions of adults, the tracheal tuft hemocytes of larvae phagocytose pathogens as they near entry into the dorsal vessel, and thus sequester and kill pathogens in the area of the body with the highest hemolymph flow. Taken together, these findings show that the functional integration of the circulatory and immune systems of mosquitoes displays drastic stage-specific differences that reflect the broad changes in body plan that occur during metamorphosis.
Larval sessile hemocytes are segmentally arranged and enriched in the eighth abdominal segment
The larval heart lacks periostial hemocytes
When the pericardial cells and hemocytes were co-stained, the periostial regions of the larval heart were devoid of hemocytes (Fig. 3e). Though at first glance some of the nearby segmentally arranged hemocytes appeared to occupy the periostial space, 3D deconvolved Z-stacks clearly showed that these segmental hemocytes are attached to the integument that is immediately underneath the cuticle, and thus, these cells lie far dorsal to the ostia, where the periostial hemocytes would be located (Fig. 3f). Staining of heart muscle with Alexa Fluor-conjugated phalloidin, which binds F-actin, in conjunction with hemocyte staining, confirmed this observation by revealing a distinct lack of periostial hemocytes in larvae (Fig. 3g, h). Instead, segmental hemocytes are present on the dorsal integument in a location that is dorsal to both the heart (Fig. 3g, h) and the surrounding swim muscles (Fig. 3i, j).
Larval hemocytes densely populate the eighth abdominal segment tracheal tufts
Additional file 2 Movie S1. The tracheal tufts move as a consequence of the wave-like contractions of the heart. Intravital video through the dorsal cuticle of the eighth abdominal segment of a larva showing that the tracheal tuft tracheoles move in concert with each contraction of the heart. Frame rate: real time.
The tracheal tuft hemocytes comprise a population of cells that is distinct from the segmental hemocytes, as the former are located ventral to the tracheal trunks, whereas the latter are attached to the integument (Fig. 4b, c). In addition to binding the tracheal tuft tracheoles, larval hemocytes often bind to other thin trachea but not to larger trachea, as is evident in the comparative scarcity of hemocytes bound to the dorsal longitudinal tracheal trunks (Fig. 4a, b). A similar phenomenon also occurs in adult mosquitoes; though lacking tracheal tufts and their free-floating tracheoles, hemocytes in adults often bind to thin trachea, and only occasionally bind to the thick trachea, such as those that replace the tracheal tufts in the eighth abdominal segment (Additional file 3: Figure S2). In summary, larvae contain a dense population of hemocytes, called the tracheal tuft hemocytes, in a region of the body that is functionally analogous to the periostial regions of adults.
Larval tracheal tuft hemocytes are phagocytically active
Additional file 4 Movie S2. Phagocytosis of GFP-E. coli in the eighth segment tracheal tufts begins within seconds of infection. Four-minute time-lapse intravital video through the dorsal cuticle of a larva immediately following injection with GFP-E. coli, showing the rapid accumulation of fluorescent bacteria (white) in the eighth abdominal segment. GFP-E. coli begins to accumulate at the tracheal tufts in less than 15 s after treatment. Frame rate: 10 s of real time equals 1 s of movie.
E. coli in the larval hemocoel aggregate in the tracheal tufts, where they are phagocytosed and killed
Injection of both fluorescent microspheres and GFP-E. coli into the hemocoel of larvae and adults resulted in patterns of aggregation that mirrored the respective hemocyte distributions in these two life stages: microspheres and E. coli aggregated at the periostial regions of the heart in adults and at the eighth abdominal segment tracheal tufts of larvae (Additional file 5: Figure S3 and Additional file 6: Figure S4). To quantitatively determine whether pathogen aggregation in larvae was indeed higher at the tracheal tufts when compared to the periostial regions of the heart, larvae were injected with GFP-E. coli, and the fluorescence emitted by these bacteria within the periostial regions and the tracheal tufts was measured at 4 and 24 h post-treatment (Additional file 7: Figure S5).
At 24 h post-treatment, little or no GFP-E. coli was detected in any specimen, regardless of treatment (Fig. 7d–f, h). Interestingly, the fluorescence signal in infected larvae was significantly lower than that of the injured larvae (RM 2 W ANOVA P < 0.0001 for treatment), and a Šidák test that compared the different body regions of infected larvae to the corresponding body regions of injured larvae showed that, with the exception of the first periostial region (P = 0.2990), this was the case for all regions of the body (P ≤ 0.0005 for all other comparisons). Furthermore, there was an interaction between the type of treatment and the segment of the body being assayed (RM 2 W ANOVA P < 0.0001), indicating that the differences in the fluorescence signal emitted in different segments changed with treatment. Visual analysis of the data and a Šidák post hoc test showed that this is due to a more pronounced decrease in the fluorescence signal at the tracheal tufts relative to the periostial regions in infected larvae (P ≤ 0.0005 for all comparisons). Collectively, these findings indicate that larvae mount vigorous and effective antibacterial immune responses that are largely completed within 24 h of infection. These immune responses are strongest at the tracheal tufts, which surround the only incurrent openings of the larval dorsal vessel.
Infection induces a robust melanization immune response at the tracheal tufts
The fluorescence signal emitted by GFP-E. coli at 4 h post-treatment is indicative of a vigorous phagocytosis response at the tracheal tufts, and the absence of this fluorescence at 24 h post-treatment is indicative of bacterial clearance. However, the fluorescence signal at 24 h is significantly lower than the fluorescence signal in injury controls, and visual examination of the specimens showed that antibacterial melanization deposits are minor at 4 h post-infection but far more prevalent 20 h later, with these deposits being most abundant at the tracheal tufts. We hypothesized that the negative fluorescence signal seen at 24 h after infection is due to the dampening of fluorescence by dark melanin deposits, and this hypothesis was supported by high magnification fluorescence imaging of the tracheal tufts of infected larvae (Additional file 9: Figure S7).
To quantitatively confirm that a melanization response is indeed taking place, we measured melanin deposition by calculating the mean pixel optical density (OD) in all the periostial regions as well as in the tracheal tufts of the same larvae used to calculate the phagocytosis response (see the preceding section; Additional file 7: Figure S5B). In this type of analysis, a higher OD value indicates darker pixels and therefore a stronger melanization response.
Because the periostial regions of larvae are devoid of hemocytes, we scrutinized the location of melanin deposits in these regions using a combination of fluorescence and bright-field microscopy, and found that melanin accumulates at the periostial regions of infected larvae because of the pinocytic activity of pericardial cells (Fig. 8b; Additional file 7: Figure S5B). At the tracheal tufts, however, melanin accumulation is much more pronounced, and this is primarily due to the accumulation and phagocytosis of melanized bacteria by the tracheal tuft hemocytes (Fig. 7; Fig. 8b, d; Additional file 9: Figure S7). Together, these data demonstrate a robust antibacterial melanization response in larvae, with much of the resultant melanin being deposited at the tracheal tufts, near the sole incurrent openings of the dorsal vessel.
Hemocytes are more prevalent in the tracheal tufts but do not aggregate upon infection
At 4 h post-treatment, there was no difference in the fluorescence signal emitted by hemocytes in naïve, injured, and infected mosquitoes (Fig. 9a; RM 2 W ANOVA P = 0.1123 for treatment). However, there was a significant difference between the regions of the body being measured (RM 2 W ANOVA P < 0.0001). Specifically, for naïve, injured, and infected mosquitoes, the fluorescence intensities of the two tracheal tufts were significantly higher than the fluorescence intensities in the corresponding periostial regions (Šidák’s P ≤ 0.0312 for 39 out of 42 comparisons). This indicates that hemocytes are absent from the periostial regions of infected larvae (any signal emanates from segmental hemocytes) but are abundant at the tracheal tufts. This also indicates that at 4 h post-infection the relative number of tracheal tuft hemocytes in infected larvae is similar to that found in naïve and injured larvae.
At 24 h post-treatment, hemocyte intensity was significantly different between naïve, injured, and infected larvae (Fig. 9b; RM 2 W ANOVA P = 0.0012). There was also a significant difference between the regions of the body being measured (RM 2 W ANOVA P < 0.0001), as the fluorescence signal in the tracheal tufts remained elevated relative to the periostial regions (Šidák’s P ≤ 0.0003 for all comparisons). Furthermore, there was a significant interaction between the region of the body and the treatment (RM 2 W ANOVA P < 0.0001 for interaction), indicating that the distribution of fluorescence signal in the different body regions changes depending on the infection status of the larva. Specifically, the fluorescence signal emitted by segmental hemocytes was similar between the periostial regions of interest (ROIs) of naïve, injured, and infected larvae (Šidák’s P ≥ 0.1528 for all comparisons), but the signal in the tracheal tufts was lower in infected larvae when compared to naïve and injured larvae (Šidák’s P < 0.0001 for all comparisons).
Taken together, these results suggest that although hemocytes are more abundant at the tracheal tufts, infection does not recruit circulating hemocytes to the periostial regions or to the tracheal tufts. At 24 h post-treatment, hemocyte fluorescence in the tracheal tufts of infected larvae was lower than that of naïve and injured larvae, but visual analysis of these specimens strongly suggests that this is not due to a reduction in the hemocyte population but rather due to melanization-induced dampening of the fluorescence signal (Fig. 7h; Fig. 8d; Additional file 7: Figure S5B and Additional file 9: Figure S7).
A major finding in this study is that bacteria aggregate and are killed by hemocytes that are attached to the tracheal tufts of mosquito larvae, which are structures that surround the only incurrent openings of the heart. This represents an immune strategy where pathogen sequestration and killing occurs in the areas of the body that receive the highest amount of hemolymph flow. These locations are ideal for mounting immune responses, as they increase the probability of hemocytes encountering circulating pathogens and facilitate the rapid dissemination of hemocyte-produced humoral immune factors. Immune responses at the tracheal tufts are functionally, though not spatially, analogous to the immune responses that occur in the periostial regions of adult mosquitoes [9, 12, 13], as in both life stages pathogens and hemocytes associate with the entry points for hemolymph into the heart. Similar aggregations of pathogens at the periostial regions have also been observed in adult Drosophila, though these aggregations have not been linked to hemolymph flow or the structural mechanics of the heart, and have not been scrutinized at the cellular level [23, 28]. Furthermore, although melanotic nodules and melanized bacteria accumulate on the surface of the heart of lepidopteran larvae [29, 30], these melanin deposits are diffused along the length of the heart and do not associate specifically with the periostial regions (the tracheal tufts were not examined). Thus, it remains unclear whether these melanization responses in larval Lepidoptera are modulated by hemolymph flow.
A difference between the immune response at the periostial regions of adult mosquitoes and the immune response at the tracheal tufts of larvae is that infection induces the migration of additional hemocytes to the former but not the latter. The reason for this is unclear, but it may be due to the kinetics of the antibacterial immune response in these two life stages. In adults, we have observed bacteria at the periostial regions as far as 12 days post-infection (the last time point that was studied) , and other studies have documented the presence of bacteria in the hemocoel days or weeks after infection [31, 32]. In larvae, however, our method measuring fluorescence pixel intensity could not detect any bacteria at 24 h post-infection, and higher magnification observation of larval tissues detected few bacteria. Another possibility for the lack of an infection-induced increase in the number of hemocytes in the tracheal tufts is that, unlike in adults, larval hemocytes do not increase in number following immune stimulation or alter their adherence properties upon infection, as occurs in adults after blood feeding or immune stimulation [9, 12, 13, 33–36]. However, as these properties have yet to be tested in larval hemocytes, we presently hypothesize that the basal number of hemocytes at the tracheal tufts, in combination with the activity of other sessile and circulating hemocytes, is sufficient to rapidly clear systemic infections. This immune proficiency may in part be facilitated by the streamlined circulatory system of larvae, which contains a single pair of incurrent heart openings , as opposed to that of adults, which contains seven pairs of incurrent ostia .
In adult mosquitoes, the primary cause of pathogen death at the periostial regions is phagocytosis [9, 13], and our data show that this immune response is also prominent at the tracheal tufts of mosquito larvae. This response begins within seconds of infection, which is similar to the speed of the phagocytosis response by the periostial, non-cardiac sessile, and circulating hemocytes of adult mosquitoes [4, 9, 13, 37–40]. Furthermore, much like in adult mosquitoes [2, 9, 38, 39], melanization of bacteria is prevalent in larvae. Our observation that melanin deposits are predominant at the tracheal tufts suggests that hemolymph flow drives the accumulation of melanized bacteria at these respiratory structures, and some of the melanized bacteria are phagocytosed by hemocytes. These observations also provide a functional explanation for the spatial distribution of the melanization response against vertically transmitted Elizabethkingia meningoseptica that has been observed in A. gambiae larvae , which also occurs in the eighth abdominal segment. Furthermore, our findings provide a new host-immunity perspective on a previous report linking parasitoid wasp development to the tracheal tufts of Heliothis virescens caterpillars, where the authors suggest that the trapping of Toxoneuron nigriceps eggs in the tracheal tufts is advantageous for the development of the parasite because it places them in an environment that is rich in oxygen . Although the authors note that the propensity of the eggs to aggregate near the tracheal tufts is due to the natural flow of hemolymph, it is possible that the host immune response at the tracheal tufts outweighs any respiration-related benefits gained by the wasp eggs.
Our discovery that a high concentration of immunologically active hemocytes is associated with the eighth abdominal segment tracheal tufts of A. gambiae larvae represents the first description of this unique population of sessile hemocytes in Diptera. Populations of hemocytes in the posterior abdominal segments have been observed in other dipteran larvae, such as Drosophila [43–52], Calliphora sp. [53–55], and Musca sp. [53, 56]. However, unlike the tracheal tuft-bound sessile hemocytes we describe in mosquito larvae, the posterior hemocytes described in other Diptera have not been documented as a hot spot of phagocytic activity during systemic infections or described as being associated with tracheal tufts.
Mosquito larvae are metapneustic, meaning they breathe through a single pair of posterior spiracles . Air enters these spiracles and is transported the length of the body by a pair of large dorsal longitudinal tracheal trunks. As they traverse the body, these tracheal trunks branch into a series of smaller segmental branches, which branch into increasingly smaller trachea, finally terminating as thin-walled tracheoles that directly oxygenate the tissues to which they are permanently attached. However, the tracheoles that comprise the eighth abdominal segment tracheal tufts differ from other tracheoles in that, aside from some attachments to the posterior terminus of the heart, they float freely in the hemolymph instead of being attached to tissues .
Tracheal tufts, which have also been referred to as the “tracheole plexus” or “tracheal lungs,” have been observed in the larvae of the mosquitoes Aedes aegypti , Culex sp. [59, 60], Anopheles maculipennis [59, 61], and Anopheles quadrimaculatus , as well as in the larvae of other dipterans [59, 63, 64], but not in Drosophila, as our own experiments have confirmed. However, the tracheal tufts have been most comprehensively described in the lepidopteran Calpodes ethlius . Because of their obvious role in the oxygenation of hemolymph and their close association with hemocytes, the tracheal tufts were originally hypothesized to act as a “lung” for hemocytes. However, given that circulating and sessile hemocytes persist following eclosion and the tracheal tufts do not, we hypothesize that the tracheal tufts are not necessary for the direct oxygenation of hemocytes and instead function in the dissolution of oxygen into the hemolymph. This is supported by our observation that hemolymph flow is highest at the tracheal tufts , and thus, the contractile action of the heart disseminates newly dissolved oxygen to other parts of the body. More relevant to immune defense, we hypothesize that hemocytes associate with the tracheal tufts because, in a process analogous to the function of periostial hemocytes [9, 12, 13], the attachment of immune cells to structures that surround the only incurrent opening of the heart places them in an ideal location to survey the hemolymph for foreign invaders. Additionally, we observed that hemocytes, both in larvae and adults, have a propensity for binding to thin trachea, and we hypothesize that this also confers an immunological advantage, because the sites for gas exchange are also locations where pathogens often invade the hemocoel [12, 66–73].
To our knowledge, this study contains the first description of segmentally arranged sessile hemocytes in larval mosquitoes. These segmentally arranged cells are phagocytic, but infection does not induce the aggregation of hemocytes or the dense aggregation of pathogens within these locations. Because of their immune competence and their sheer abundance in the larval stage, this newly discovered population of sessile hemocytes represents a formidable immune reservoir. A similar arrangement of hemocytes, referred to as the segmentally arranged sessile hemocyte islets or patches, has been documented in Drosophila [44, 46, 47, 50, 52, 74, 75], as well as in other Diptera, such as the larger Calliphora flies . In Drosophila, larval segmental hemocytes detach at the onset of pupation to engage in tissue remodeling [46, 50, 76] and in order to encapsulate parasitoid wasp eggs [48, 52]. In adult mosquitoes, segmentally arranged hemocytes participate in the consumption of autolysing larval swim muscles during the first couple of days post-eclosion, and then dissipate . Our finding that these cells also contribute to bacterial clearance points to an even broader role for these sessile hemocytes in larval insect immune responses.
A. gambiae is a major vector of malaria in sub-Saharan Africa, and as such, understanding the innate immune defenses of this mosquito against invading pathogens is crucial to the development of novel pest and disease control programs. In spite of this, the host immune response in larvae has received scant attention compared to the adult stage, a significant oversight considering that larvae are often a major target in pest control efforts [77, 78], and many common bacterial and fungal pesticides are seeded into mosquito breeding sites where they exert their effects in the hemocoel of larvae [79–82]. The present study uncovers an immune strategy that mosquito larvae use in combating microbial infections, which relies on the coordinated action of their circulatory, immune, and respiratory systems to sequester and kill pathogens present in the hemocoel.
Mosquito rearing and maintenance
Anopheles gambiae Giles sensu stricto (G3 strain; Diptera: Culicidae) were reared as previously described . Briefly, larvae were reared in deionized water and fed a mixture of koi food and baker’s yeast. Upon pupation, mosquitoes were transferred to plastic containers with a marquisette net top, and adults were fed a solution of 10 % sucrose ad libitum. All life stages were maintained in an environmental chamber set to 27 °C and 75 % relative humidity under a 12 h:12 h, light:dark cycle. Experiments were conducted on fourth instar larvae and adult female mosquitoes at 5 days post-eclosion.
Mosquito injection, inoculation of immune elicitors, and bacterial infection
Mosquito larvae were immobilized by removing excess water and then injected at the lateral center of the mesothorax. Mosquito adults were cold anesthetized and then injected at the thoracic anepisternal cleft. Finely pulled glass capillary tubes were used as needles, and 0.2 μl of the following substances, alone or in combination, were injected: 0.08 % solids 1-μm diameter green fluorescent (505/515) carboxylate-modified microspheres (Invitrogen, Carlsbad, CA, USA) in phosphate-buffered saline (PBS) (pH 7.0), GFP-expressing E. coli (modified DH5α) in Luria-Bertani’s rich nutrient medium (LB), LB medium, 1 mg/ml pHrodo Red E. coli Bio Particles (560/585 nm; Invitrogen) in PBS, or cellular staining solutions (see the next section). For bacterial infections, E. coli were grown in LB overnight at 37 °C in a shaking incubator until cultures reached approximately OD600 = 5, as measured using a BioPhotometer Plus spectrophotometer (Eppendorf AG, Hamburg, Germany).
Staining of hemocytes, pericardial cells, and heart muscle
To fluorescently label hemocytes, live larvae or adults were injected with 67 μM Vybrant CM-DiI or DiO Cell-Labeling Solutions (Invitrogen) in PBS and were incubated at 27 °C and 75 % relative humidity for 20 min or 1 h, respectively. Larvae were then fixed in 16 % paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA, USA) for 1 h and either rinsed briefly in PBS and the whole body mounted directly on a glass slide using Aqua-Poly/Mount (Polysciences Inc., Warrington, PA, USA), or dissected along a coronal plane, rinsed briefly in PBS, and the dorsal abdomen sans the internal organs mounted. Adults were fixed by injecting 16 % paraformaldehyde, and 10 min later the specimens were dissected along a coronal plane, placed in 0.5 % Tween in PBS to reduce surface tension, rinsed briefly in PBS, and the dorsal abdomen sans the internal organs mounted on a glass slide using Aqua-Poly/Mount.
To fluorescently co-stain phagocytic events and hemocytes, larvae were injected with green fluorescent microspheres in PBS, GFP-E. coli in LB, or pHrodo Red E. coli Bio Particles in PBS and allowed to incubate for 4 h at 27 °C and 75 % relative humidity prior to staining with either CM-DiI or DiO. To co-stain hemocytes and heart muscle, following staining with CM-DiI the larvae were incubated in 16 % formaldehyde for 1 h, dissected, and incubated in 0.7 μM phalloidin-Alexa Fluor 488 (Invitrogen) in PBS on a Pelco R1 Rotary Mixer (Pelco Instruments, Redding, CA, USA) for 1 h. To fluorescently stain pericardial cells, larvae and adults were injected with 0.2 mg/ml Alexa Fluor conjugated IgG 568 nm (Invitrogen) and 0.8 mM Hoechst 33342 nuclear stain (350/461; Invitrogen) in PBS and allowed to incubate for 2 h. To co-stain pericardial cells and hemocytes, larvae were injected with 0.2 mg/ml IgG 488 nm and allowed to incubate for 2 h prior to staining hemocytes with CM-DiI. After staining, specimens were dissected and mounted as described above.
Light and fluorescence microscopy of aldehyde-fixed mosquito specimens
Larval and adult specimens were imaged under bright-field and fluorescence illumination using a Nikon 90i compound microscope connected to a Nikon Digital Sight DS-Qi1Mc monochrome digital camera (Nikon, Tokyo, Japan). For the rendering of detailed fluorescence images with extended focal depth, Z-stacks of whole mounts were acquired using a linear encoded Z-motor, and all images in a stack were combined to form a single focused image using the Extended Depth of Focus (EDF; for image viewing) or Maximum Intensity Projection (for image quantification) modules of Nikon’s NIS Elements software. For three-dimensional rendering, Z-stacks were quantitatively deconvolved using the AQ 3D Blind Deconvolution module of NIS Elements and rendered using the volume view feature.
Quantification of fluorescently labeled E. coli and hemocyte fluorescence intensity
GFP-E. coli and CM-DiI-labeled hemocyte fluorescence intensity were quantified in abdominal segments 1–8 of larvae using maximum intensity projections of Z-stacks created in NIS Elements software. For abdominal segments 1–7, custom polygonal regions of interest (ROIs) were drawn over the anterior portion of each abdominal segment, encompassing the periostial regions of the heart (Additional file 7: Figure S5). Specifically, these ROIs contained the area in each abdominal segment that lay between the dorsal longitudinal tracheal trunks, and stretched from the abdominal suture to the dorsal abdominal tracheal commissure. For segment 8, two custom ROIs circumscribing each tracheal tuft were constructed. The mean pixel intensity, defined as the average intensity of pixels within a given ROI, was then calculated to compare fluorescence emission across ROIs quantitatively.
For GFP-E. coli quantification, intensity measurements for any given ROI in the GFP-E. coli and injury (injected with LB alone) groups were normalized by subtracting the intensity values from the corresponding ROI from the naïve group, as naïve mosquitoes represent background fluorescence. Three independent trials were conducted at both 4 and 24 h post-treatment, and each trial consisted of at least 10 mosquitoes per group. For CM-DiI-labeled hemocyte quantification, intensity measurements for any given ROI in the CM-DiI-stained naïve, injured, and E. coli groups were normalized by subtracting the intensity values from the corresponding ROI of the unstained naïve group. Four and five independent trials consisting of on average 8 and 6 mosquitoes per group were conducted at 4 and 24 h post-treatment, respectively. Because for each mosquito measurements were obtained for all periostial regions and the tracheal tufts (left and right), data (Additional file 10: Data file S1) were analyzed by repeated measures two-way ANOVA (RM 2 W ANOVA; repeated measures being the nine regions of the body of an individual mosquito), using treatment and ROI as the variables, followed by a Šidák post hoc test.
Quantification of melanization in response to E. coli infection
Melanization was quantified in abdominal segments 1–8 of larvae using maximum intensity projections created from the bright-field images of the Z-stacks of the same specimens used to measure GFP-E. coli fluorescence intensity. These data (Additional file 10: Data file S1) were analyzed using the same ROI scheme and statistical tests described above. For the melanization analysis, the mean optical pixel density (OD), defined as the average OD of pixels within a given ROI, was calculated to compare melanization across ROIs quantitatively. The OD measurements for any given ROI in the E. coli and injured groups were normalized by subtracting the OD values from the corresponding ROI from the naïve group.
We thank Dr. Patrick Abbot, Dr. Lisa Brown, Leah T. Sigle, and Tania Estévez-Lao for fruitful discussions and for commenting on this manuscript.
This research was funded by US National Science Foundation (NSF) grant IOS-1456844 and US National Institutes of Health (NIH) grant 1R21AI119596 to JFH. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of this manuscript.
Availability of data and materials
All relevant data are within the paper and its supporting Additional files.
GPL and JFH designed the study; GPL performed the experiments; GPL and JFH analyzed the data and wrote the manuscript. Both authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Chapman RF, Douglas AE, Siva-Jothy MT. Circulatory system, blood, and the immune system. In: Simpson SJ, Douglas AE, editors. The insects: structure and function. Cambridge: Cambridge University Press; 2013. p. 107–31.Google Scholar
- Hillyer JF. Insect immunology and hematopoiesis. Dev Comp Immunol. 2016;58:102–18.PubMedView ArticleGoogle Scholar
- Hillyer JF. Mosquito immunity. Adv Exp Med Biol. 2010;708:218–38.PubMedView ArticleGoogle Scholar
- Hillyer JF, Strand MR. Mosquito hemocyte-mediated immune responses. Curr Opin Insect Sci. 2014;3:14–21.PubMedPubMed CentralView ArticleGoogle Scholar
- Lavine MD, Strand MR. Insect hemocytes and their role in immunity. Insect Biochem Mol Biol. 2002;32(10):1295–309.PubMedView ArticleGoogle Scholar
- Strand MR. The insect cellular immune response. Insect Sci. 2008;15(1):1–14.View ArticleGoogle Scholar
- Hillyer JF. Integrated immune and cardiovascular function in Pancrustacea: lessons from the insects. Integr Comp Biol. 2015;55(5):843–55.PubMedView ArticleGoogle Scholar
- Klowden AJ. Circulatory systems. In: Physiological systems in insects. 3rd ed. Boston: Academic Press; 2013. p. 365–413.View ArticleGoogle Scholar
- King JG, Hillyer JF. Infection-induced interaction between the mosquito circulatory and immune systems. PLoS Pathog. 2012;8(11):e1003058.PubMedPubMed CentralView ArticleGoogle Scholar
- Glenn JD, King JG, Hillyer JF. Structural mechanics of the mosquito heart and its function in bidirectional hemolymph transport. J Exp Biol. 2010;213(4):541–50.PubMedView ArticleGoogle Scholar
- League GP, Onuh OC, Hillyer JF. Comparative structural and functional analysis of the larval and adult dorsal vessel and its role in hemolymph circulation in the mosquito Anopheles gambiae. J Exp Biol. 2015;218(Pt 3):370–80.PubMedPubMed CentralView ArticleGoogle Scholar
- King JG, Hillyer JF. Spatial and temporal in vivo analysis of circulating and sessile immune cells in mosquitoes: hemocyte mitosis following infection. BMC Biol. 2013;11:55.PubMedPubMed CentralView ArticleGoogle Scholar
- Sigle LT, Hillyer JF. Mosquito hemocytes preferentially aggregate and phagocytose pathogens in the periostial regions of the heart that experience the most hemolymph flow. Dev Comp Immunol. 2016;55:90–101.PubMedView ArticleGoogle Scholar
- Hernández-Martínez S, Lanz-Mendoza H, Martínez-Barnetche J, Rodríguez MH. Antimicrobial properties of Anopheles albimanus pericardial cells. Cell Tissue Res. 2013;351(1):127–37.PubMedView ArticleGoogle Scholar
- Hillyer JF, Barreau C, Vernick KD. Efficiency of salivary gland invasion by malaria sporozoites is controlled by rapid sporozoite destruction in the mosquito haemocoel. Int J Parasitol. 2007;37(6):673–81.PubMedView ArticleGoogle Scholar
- Michel K, Budd A, Pinto S, Gibson TJ, Kafatos FC. Anopheles gambiae SRPN2 facilitates midgut invasion by the malaria parasite Plasmodium berghei. EMBO Rep. 2005;6(9):891–7.PubMedPubMed CentralView ArticleGoogle Scholar
- Schnitger AK, Kafatos FC, Osta MA. The melanization reaction is not required for survival of Anopheles gambiae mosquitoes after bacterial infections. J Biol Chem. 2007;282(30):21884–8.PubMedView ArticleGoogle Scholar
- Yassine H, Kamareddine L, Chamat S, Christophides GK, Osta MA. A serine protease homolog negatively regulates TEP1 consumption in systemic infections of the malaria vector Anopheles gambiae. J Innate Immun. 2014;6(6):806–18.PubMedPubMed CentralView ArticleGoogle Scholar
- Yassine H, Kamareddine L, Osta MA. The mosquito melanization response is implicated in defense against the entomopathogenic fungus Beauveria bassiana. PLoS Pathog. 2012;8(11):e1003029.PubMedPubMed CentralView ArticleGoogle Scholar
- Akbar MA, Tracy C, Kahr WH, Krämer H. The full-of-bacteria gene is required for phagosome maturation during immune defense in Drosophila. J Cell Biol. 2011;192(3):383–90.PubMedPubMed CentralView ArticleGoogle Scholar
- Chambers MC, Lightfield KL, Schneider DS. How the fly balances its ability to combat different pathogens. PLoS Pathog. 2012;8(12):e1002970.PubMedPubMed CentralView ArticleGoogle Scholar
- Cuttell L, Vaughan A, Silva E, Escaron CJ, Lavine M, Van Goethem E, Eid JP, Quirin M, Franc NC. Undertaker, a Drosophila Junctophilin, links Draper-mediated phagocytosis and calcium homeostasis. Cell. 2008;135(3):524–34.PubMedView ArticleGoogle Scholar
- Elrod-Erickson M, Mishra S, Schneider D. Interactions between the cellular and humoral immune responses in Drosophila. Curr Biol. 2000;10(13):781–4.PubMedView ArticleGoogle Scholar
- Ghosh S, Singh A, Mandal S, Mandal L. Active hematopoietic hubs in Drosophila adults generate hemocytes and contribute to immune response. Dev Cell. 2015;33(4):478–88.PubMedPubMed CentralView ArticleGoogle Scholar
- Kocks C, Cho JH, Nehme N, Ulvila J, Pearson AM, Meister M, Strom C, Conto SL, Hetru C, Stuart LM, et al. Eater, a transmembrane protein mediating phagocytosis of bacterial pathogens in Drosophila. Cell. 2005;123(2):335–46.PubMedView ArticleGoogle Scholar
- Shiratsuchi A, Mori T, Sakurai K, Nagaosa K, Sekimizu K, Lee BL, Nakanishi Y. Independent recognition of Staphylococcus aureus by two receptors for phagocytosis in Drosophila. J Biol Chem. 2012;287(26):21663–72.PubMedPubMed CentralView ArticleGoogle Scholar
- Stone EF, Fulton BO, Ayres JS, Pham LN, Ziauddin J, Shirasu-Hiza MM. The circadian clock protein timeless regulates phagocytosis of bacteria in Drosophila. PLoS Pathog. 2012;8(1):e1002445.PubMedPubMed CentralView ArticleGoogle Scholar
- Horn L, Leips J, Starz-Gaiano M. Phagocytic ability declines with age in adult Drosophila hemocytes. Aging Cell. 2014;13(4):719–28.Google Scholar
- Bao YY, Xue J, Wu WJ, Wang Y, Lv ZY, Zhang CX. An immune-induced reeler protein is involved in the Bombyx mori melanization cascade. Insect Biochem Mol Biol. 2011;41(9):696–706.PubMedView ArticleGoogle Scholar
- Pereira MF, Rossi CC, de Queiroz MV, Martins GF, Isaac C, Bossé JT, Li Y, Wren BW, Terra VS, Cuccui J, et al. Galleria mellonella is an effective model to study Actinobacillus pleuropneumoniae infection. Microbiology. 2015;161(Pt 2):387–400.PubMedView ArticleGoogle Scholar
- Gorman MJ, Paskewitz SM. Persistence of infection in mosquitoes injected with bacteria. J Invertebr Pathol. 2000;75(4):296–7.PubMedView ArticleGoogle Scholar
- Hillyer JF, Schmidt SL, Fuchs JF, Boyle JP, Christensen BM. Age-associated mortality in immune challenged mosquitoes (Aedes aegypti) correlates with a decrease in haemocyte numbers. Cell Microbiol. 2005;7(1):39–51.PubMedView ArticleGoogle Scholar
- Bryant WB, Michel K. Blood feeding induces hemocyte proliferation and activation in the African malaria mosquito, Anopheles gambiae Giles. J Exp Biol. 2014;217(Pt 8):1238–45.PubMedPubMed CentralView ArticleGoogle Scholar
- Bryant WB, Michel K. Anopheles gambiae hemocytes exhibit transient states of activation. Dev Comp Immunol. 2016;55:119–29.PubMedView ArticleGoogle Scholar
- Castillo J, Brown MR, Strand MR. Blood feeding and insulin-like peptide 3 stimulate proliferation of hemocytes in the mosquito Aedes aegypti. PLoS Pathog. 2011;7(10):e1002274.PubMedPubMed CentralView ArticleGoogle Scholar
- Christensen BM, Huff BM, Miranpuri GS, Harris KL, Christensen LA. Hemocyte population changes during the immune response of Aedes aegypti to inoculated microfilariae of Dirofilaria immitis. J Parasitol. 1989;75(1):119–23.PubMedView ArticleGoogle Scholar
- Hillyer JF, Schmidt SL, Christensen BM. The antibacterial innate immune response by the mosquito Aedes aegypti is mediated by hemocytes and independent of Gram type and pathogenicity. Microbes Infect. 2004;6(5):448–59.PubMedView ArticleGoogle Scholar
- Hillyer JF, Schmidt SL, Christensen BM. Hemocyte-mediated phagocytosis and melanization in the mosquito Armigeres subalbatus following immune challenge by bacteria. Cell Tissue Res. 2003;313(1):117–27.PubMedView ArticleGoogle Scholar
- Hillyer JF, Schmidt SL, Christensen BM. Rapid phagocytosis and melanization of bacteria and Plasmodium sporozoites by hemocytes of the mosquito Aedes aegypti. J Parasitol. 2003;89(1):62–9.PubMedView ArticleGoogle Scholar
- Hillyer JF, Christensen BM. Characterization of hemocytes from the yellow fever mosquito, Aedes aegypti. Histochem Cell Biol. 2002;117(5):431–40.PubMedView ArticleGoogle Scholar
- Akhouayri IG, Habtewold T, Christophides GK. Melanotic pathology and vertical transmission of the gut commensal Elizabethkingia meningoseptica in the major malaria vector Anopheles gambiae. PLoS One. 2013;8(10):e77619.PubMedPubMed CentralView ArticleGoogle Scholar
- Rao A, Henderson RE, Bradleigh Vinson S. The probable significance of tracheal tufts in the 8th abdominal segment of Heliothis virescens (F.) on the development of its parasitoid, Toxoneuron nigriceps (Viereck). J Insect Physiol. 2009;55(9):769–73.PubMedView ArticleGoogle Scholar
- Babcock DT, Brock AR, Fish GS, Wang Y, Perrin L, Krasnow MA, Galko MJ. Circulating blood cells function as a surveillance system for damaged tissue in Drosophila larvae. Proc Natl Acad Sci U S A. 2008;105(29):10017–22.PubMedPubMed CentralView ArticleGoogle Scholar
- Honti V, Csordás G, Kurucz É, Márkus R, Andó I. The cell-mediated immunity of Drosophila melanogaster: hemocyte lineages, immune compartments, microanatomy and regulation. Dev Comp Immunol. 2014;42(1):47–56.PubMedView ArticleGoogle Scholar
- Kurucz E, Váczi B, Márkus R, Laurinyecz B, Vilmos P, Zsámboki J, Csorba K, Gateff E, Hultmark D, Andó I. Definition of Drosophila hemocyte subsets by cell-type specific antigens. Acta Biol Hung. 2007;58(Suppl):95–111.PubMedView ArticleGoogle Scholar
- Lanot R, Zachary D, Holder F, Meister M. Postembryonic hematopoiesis in Drosophila. Dev Biol. 2001;230(2):243–57.PubMedView ArticleGoogle Scholar
- Makhijani K, Alexander B, Tanaka T, Rulifson E, Brückner K. The peripheral nervous system supports blood cell homing and survival in the Drosophila larva. Development. 2011;138(24):5379–91.PubMedPubMed CentralView ArticleGoogle Scholar
- Márkus R, Laurinyecz B, Kurucz E, Honti V, Bajusz I, Sipos B, Somogyi K, Kronhamn J, Hultmark D, Andó I. Sessile hemocytes as a hematopoietic compartment in Drosophila melanogaster. Proc Natl Acad Sci U S A. 2009;106(12):4805–9.PubMedPubMed CentralView ArticleGoogle Scholar
- Shrestha R, Gateff E. Ultrastructure and cytochemistry of the cell types in the larval hematopoietic organs and hemolymph of Drosophila melanogaster. Dev Growth Differ. 1982;24(1):65–82.View ArticleGoogle Scholar
- Stofanko M, Kwon SY, Badenhorst P. A misexpression screen to identify regulators of Drosophila larval hemocyte development. Genetics. 2008;180(1):253–67.PubMedPubMed CentralView ArticleGoogle Scholar
- Welman A, Serrels A, Brunton VG, Ditzel M, Frame MC. Two-color photoactivatable probe for selective tracking of proteins and cells. J Biol Chem. 2010;285(15):11607–16.PubMedPubMed CentralView ArticleGoogle Scholar
- Zettervall CJ, Anderl I, Williams MJ, Palmer R, Kurucz E, Ando I, Hultmark D. A directed screen for genes involved in Drosophila blood cell activation. Proc Natl Acad Sci U S A. 2004;101(39):14192–7.PubMedPubMed CentralView ArticleGoogle Scholar
- Arvy L. Contribution à l’étude de la leucopoïèse chez quelques Diptères. Bull Soc Zool Fr. 1953;78:158–69.Google Scholar
- Crossley ACS. An experimental analysis of the origins and physiology of haemocytes in the blue blow-fly Calliphora erythrocephala, (Meig). J Exp Zool. 1964;157(3):375–97.PubMedView ArticleGoogle Scholar
- Zachary D, Hoffmann JA. The haemocytes of Calliphora erythrocephala (Meig.) (Diptera). Z Zellforsch Mikrosk Anat. 1973;141(1):55–73.PubMedView ArticleGoogle Scholar
- Arvy L. Données sur la leucopoièse chez Musca domestica, L. Proc R Entomol Soc London. 1954;29:39–41.Google Scholar
- Clements AN. The biology of mosquitoes: development, nutrition and reproduction, vol. 1. London: Chapman & Hall; 1992. p. 100–23.Google Scholar
- Christophers SR. Aedes aegypti (L.) The yellow fever mosquito: its life history, bionomics and structure. New York: Cambridge University Press; 1960.Google Scholar
- Imms AD. On the larval and pupal stages of Anopheles maculipennis, Meigen. J Hyg (Lond). 1907;7(2):291–318.View ArticleGoogle Scholar
- Raschke EW. Die Larve von Culex nemorosus. Zool Anz. 1887;10:18.Google Scholar
- Wigglesworth VB. The physiology of mosquitoes, 1. Philadelphia and London: Saunders; 1949.Google Scholar
- Jones JC. The heart and associated tissues of Anopheles quadrimaculatus Say (Diptera: Culicidae). J Morphol. 1954;94(1):71–123.View ArticleGoogle Scholar
- Dell JA. On the structure and life-history of Psychoda sexpunctata. Trans Ent Soc Lond. 1905;293.Google Scholar
- Vaney C. Contributions à l’étude des larves et des metamorphoses des Diptères. Ann de l'Univ de Lyon. 1902;1(9):138.Google Scholar
- Locke M. Caterpillars have evolved lungs for hemocyte gas exchange. J Insect Physiol. 1997;44(1):1–20.PubMedView ArticleGoogle Scholar
- Bukhari T, Middelman A, Koenraadt CJ, Takken W, Knols BG. Factors affecting fungus-induced larval mortality in Anopheles gambiae and Anopheles stephensi. Malar J. 2010;9:22.PubMedPubMed CentralView ArticleGoogle Scholar
- Clark TB, Kellen WR, Fukuda T, Lindegren JE. Field and laboratory studies on the pathogenicity of the fungus Beauveria bassiana to three genera of mosquitoes. J Invertebr Pathol. 1968;11(1):1–7.PubMedView ArticleGoogle Scholar
- Engelhard EK, Kam-Morgan LN, Washburn JO, Volkman LE. The insect tracheal system: a conduit for the systemic spread of Autographa californica M nuclear polyhedrosis virus. Proc Natl Acad Sci U S A. 1994;91(8):3224–7.PubMedPubMed CentralView ArticleGoogle Scholar
- Franz AW, Kantor AM, Passarelli AL, Clem RJ. Tissue barriers to arbovirus infection in mosquitoes. Viruses. 2015;7(7):3741–67.PubMedPubMed CentralView ArticleGoogle Scholar
- Lacey CM, Lacey LA, Roberts DR. Route of invasion and histopathology of Metarhizium anisopliae in Culex quinquefasciatus. J Invertebr Pathol. 1988;52(1):108–18.PubMedView ArticleGoogle Scholar
- Miranpuri GS, Khachatourians GG. Infection sites of the entomopathogenic fungus Beauveria bassiana in the larvae of the mosquito Aedes aegypti. Entomol Exp Appl. 1991;59(1):19–27.View ArticleGoogle Scholar
- Romoser WS, Wasieloski LP, Pushko P, Kondig JP, Lerdthusnee K, Neira M, Ludwig GV. Evidence for arbovirus dissemination conduits from the mosquito (Diptera: Culicidae) midgut. J Med Entomol. 2004;41(3):467–75.PubMedView ArticleGoogle Scholar
- Scholte E-J, Knols BGJ, Samson RA, Takken W. Entomopathogenic fungi for mosquito control: A review. J Insect Sci. 2004;4(19). doi: 10.1673/031.004.1901.
- Moreira CG, Jacinto A, Prag S. Drosophila integrin adhesion complexes are essential for hemocyte migration in vivo. Biol Open. 2013;2(8):795–801.PubMedPubMed CentralView ArticleGoogle Scholar
- Narita R, Yamashita H, Goto A, Imai H, Ichihara S, Mori H, Kitagawa Y. Syndecan-dependent binding of Drosophila hemocytes to laminin alpha3/5 chain LG4-5 modules: potential role in sessile hemocyte islets formation. FEBS Lett. 2004;576(1-2):127–32.PubMedView ArticleGoogle Scholar
- Tepass U, Fessler LI, Aziz A, Hartenstein V. Embryonic origin of hemocytes and their relationship to cell death in Drosophila. Development. 1994;120(7):1829–37.PubMedGoogle Scholar
- Fillinger U, Lindsay SW. Larval source management for malaria control in Africa: myths and reality. Malar J. 2011;10:353.PubMedPubMed CentralView ArticleGoogle Scholar
- Tusting LS, Thwing J, Sinclair D, Fillinger U, Gimnig J, Bonner KE, Bottomley C, Lindsay SW. Mosquito larval source management for controlling malaria. Cochrane Database Syst Rev. 2013;8:CD008923.PubMedPubMed CentralGoogle Scholar
- Bukhari T, Takken W, Koenraadt CJ. Development of Metarhizium anisopliae and Beauveria bassiana formulations for control of malaria mosquito larvae. Parasit Vectors. 2011;4:23.PubMedPubMed CentralView ArticleGoogle Scholar
- Capone A, Ricci I, Damiani C, Mosca M, Rossi P, Scuppa P, Crotti E, Epis S, Angeletti M, Valzano M, et al. Interactions between Asaia, Plasmodium and Anopheles: new insights into mosquito symbiosis and implications in malaria symbiotic control. Parasit Vectors. 2013;6(1):182.PubMedPubMed CentralView ArticleGoogle Scholar
- Federici BA, Park HW, Bideshi DK, Wirth MC, Johnson JJ. Recombinant bacteria for mosquito control. J Exp Biol. 2003;206(Pt 21):3877–85.PubMedView ArticleGoogle Scholar
- Otieno-Ayayo ZN, Zaritsky A, Wirth MC, Manasherob R, Khasdan V, Cahan R, Ben-Dov E. Variations in the mosquito larvicidal activities of toxins from Bacillus thuringiensis ssp. israelensis. Environ Microbiol. 2008;10(9):2191–9.PubMedView ArticleGoogle Scholar
- Estévez-Lao TY, Boyce DS, Honegger HW, Hillyer JF. Cardioacceleratory function of the neurohormone CCAP in the mosquito Anopheles gambiae. J Exp Biol. 2013;216(Pt 4):601–13.PubMedView ArticleGoogle Scholar