Introduction

Mosquitoes are vectors of many diseases, including malaria, dengue, Japanese encephalitis, yellow fever and filariasis1. These vector-borne diseases are major global public health threats2 and pose complex biological problems, such as the emergence of drug-resistant strains of pathogens and insecticide-resistant vector populations1. Therefore, medical research teams have been struggling for many years to develop effective vaccines or drugs for mosquito-borne diseases3.

Pathogens in mosquitoes are always transmitted to the host during probing or blood feeding4. The pump organs positioned in the head create the suction pressure necessary to suck blood through the blood canal of its proboscis5. As a result, liquid-phase fluids are sucked through a tightly sealed proboscis6. Most blood-feeding arthropods have piercing-sucking mouthparts at their prosoma to feed on blood5. However, the mouthparts of blood-feeding arthropods have different morphological structures and functional mechanisms5. Mosquitoes regulate blood feeding by controlling the two pump organs the cibarial dilator pump (CP) and the pharyngeal dilator pump (PP)7. The valve-like organ (Vap) positioned between the two pumps appears to prevent backflow in the conduit connecting the two pumps (C-P)7. The two pump organs assist in the feeding of liquid food and in the effective regulation of the blood-sucking process. The CP located at the curved conduit takes up blood from the host through the food canal. The curved conduit is directly connected to the proboscis and to the digestive organs in the main body. Without the curved conduit in the food passage, it would be difficult for mosquitoes to hunt and pierce the skin of a host to suck its blood8. The morphological and functional features of the pump system of a female mosquito have a strong influence on its feeding behavior, thus affecting host exposure to pathogens9.

Mosquitoes exhibit different postures in the static or resting condition. Interestingly, Anopheles and Aedes show different sitting postures during the resting and feeding phases. Anopheles commonly rests by thrusting its abdomen upward to straighten the proboscis and main body, whereas Aedes lies parallel to its resting surface. These two mosquitoes spread different types of diseases. The transmission of malarial parasites to humans usually occurs through Anopheles, whereas Aedes transmits dengue and yellow fever10. The sitting posture of mosquitoes was suggested to reflect the kinematic characteristics of their structures and behaviors11.

In this study, the 3D morphological features of the pumping organs of An. sinensis and Ae. togoi were investigated using synchrotron X-ray microscopic computed tomography (SR-μCT). To evaluate the blood feeding volume of a female mosquito, the two pump chambers were extracted from its 3D reconstructed image. The volumetric variations and feeding behaviors of the feeding pump chambers of An. sinensis and Ae. togoi were quantitatively compared. In addition, differences in the actin-based structures of the heads of An. sinensis and Ae. togoi were investigated. These experimental results will aid in understanding the biophysical characteristics of the pumping organs of An. sinensis and Ae. togoi, which transmit different tropical diseases.

Results

SEM images of the female mosquito heads

Figure 1a–d show typical SEM images of the proboscis and head parts of Ae. togoi and An. sinensis. The length of the maxillary palp of Ae. togoi is shorter than that of An. sinensis (Fig. 1b,e). Figure 1c,f show images of the typical sitting postures of the two different species of mosquitoes in the static or resting state.

Figure 1
figure 1

SEM images of the heads and proboscises of Ae. togoi (a,b) and An. sinensis (d,e) female mosquitoes. Images of the resting state of Ae. togoi (c) and An. sinensis (f).

3D reconstruction of the mosquito heads

The feeding pump organs are in the head of a mosquito. The CP is located under the clypeus and the PP is located above and behind the cibarial pump. The CP conduit is located between these two pumps12. The CP has a curved shape and is supported against the pull of dilator muscles positioned between the sub-esophageal ganglion and the brain12.

The sagittal, coronal and axial images of the feeding pumps are shown in Fig. 2. Using an image-segmentation tool, the feeding pumps of the mosquitoes were quantitatively analyzed and the pixels in the salient image regions were clustered. Mosquito samples fixed with an ethanol solution were selected to represent the reference cases of the static state. Based on the reference images, the pump systems of these two different species of mosquitoes were found to be clearly different.

Figure 2
figure 2

3D morphological structures of the heads of An. sinensis (a) and Ae. togoi (b) female mosquitoes reconstructed using SR-μCT. The viewing angle is slightly turned from the coronal view. (c) Sagittal view, (d) coronal view, (e) axial view of An. sinensis. (f) Sagittal view, (g) coronal view, (h) axial view of Ae. togoi.

SR-μCT cross-sectional image of the mosquito feeding pump organs

To compare the pump systems of the two mosquitoes, their cross-sectional images are shown in Fig. 3. Figure 3a,e show the dorsal inner walls of the CP. This CP is located near the cibarial dome situated directly above the cibarial armature12. The cross-sectional images of the PP are shown in Fig. 3c,g. The PP has a triangular shape comprising three sclerotized plates. These three plates, one dorsal plate and two ventro-lateral plates, are drawn together during the contraction process of the dilator muscles. These contraction processes induce the formation of the posterior pharyngeal valve12.

Figure 3
figure 3

Cross-sectional images (a–h) of the pump systems reconstructed using SR-μCT. The Cross-sectional shape and area of the pump organs are changed according to the sectional position. SR-μCT cross-sectional images of An. sinensis (a–d) and Ae. togoi (e–h). (a,e) Cross-sectional images of the CP. (b,f) Cross-sectional images of the C-P. (c,g) Cross-sectional images of the PP. (d,h) Cross-sectional images of the proboscis.

The pumps of Ae. togoi and An. sinensis have similar shapes as those obtained in the static state. The size of the sclerotized plates of the PP in An. sinensis is mostly larger than that of Ae. togoi (Fig. 3). For example, the distance between the two points of the sclerotized plates of An. sinensis and Ae. togoi is 319.12 ± 105.70 μm and 235.31 ± 71.63 μm, respectively (Fig. 3c,g).

Volume variation of the pumps in the static and expansion states

The blood-feeding pump organs were segmented from the 3D reconstructed head of the two different types of female mosquitoes to measure variations in volume. To estimate the pump volume, the volumes of the feeding pump organs of 4 to 5 similarly sized mosquito samples were statistically averaged.

The whole volumes of the feeding pumps organs in the static state are compared in Fig. 4a. The volumes of the CP, C-P and PP of both mosquitoes are compared in Fig. 4b. The whole volumes of the feeding pumps of Ae. togoi and An. sinensis are 43.03 ± 25.25 μm3 and 139.93 ± 84.30 μm3, respectively (Fig. 4a). The whole volume of An. sinensis is much larger than that of Ae. togoi in the static state. The volumes of the CP are approximately 4.77 ± 1.29 μm3 and 38.94 ± 9.45 μm3 for Ae. togoi and An. sinensis, respectively. The volumes of the PP are 13.54 ± 2.99 μm3 and 78.74 ± 37.52 μm3 for Ae. togoi and An. sinensis, respectively. The volumes of the CP and PP are larger in An. sinensis compared with those in Ae. togoi. The volumes of the C-P in both mosquito species were comparable (Fig. 4b).

Figure 4
figure 4

Comparison of the whole volumes of the blood-feeding pumps at rest in An. sinensis and Ae. togoi.

(a) The whole volume of the CP, C-P and PP of both mosquitoes. (b) A volumetric comparison of each pump organ for both mosquitoes. Error bars indicate standard deviation; *P < 0.05.

Figure 5 shows a sagittal view of the head of a mosquito sucking a diluted iodine solution. The whole volume of the feeding pumps does not show a large difference in the expansion process. The C-P of An. sinensis is more expanded compared with that of Ae. togoi (Fig. 5).

Figure 5
figure 5

2D X-ray images showing the sagittal view of the head of a mosquito sucking a dilute iodine solution.

The expansion state of the CP, C-P and PP of An. sinensis (a) and of Ae. togoi (b). Scale bar indicates 200 μm.

Systematic operation of the two pumping organs in both mosquitoes

Figure 6 shows temporal variations in the instantaneous volumes of the two pumps, the CP and the PP, of the two mosquitoes. The fluid-sucking process of a female mosquito occurs as follows: The mosquito sucks the diluted iodine solution. The CP begins to expand to suck the liquid food, whereas the PP remains static. The light intensity of the CP region gradually increases as the iodine solution accumulates. Simultaneously, the PP starts to expand with a certain time interval. Following the contraction and expansion processes of the two pumps, the mosquito completes one feeding cycle13.

Figure 6
figure 6

Temporal variations of the two pump organs of An. sinensis (a) and Ae. togoi (b). Each data point represents the mean value. The values are the means ± standard deviations.

The feeding cycles of both mosquitoes are less than one second. To analyze the feeding cycles quantitatively, the intensity values of the test samples were condition-averaged by assigning the instant of the maximum intensity value in each period as the reference time (t = 0). High-intensity values correspond to the expansion state of the pumping organ, in which the penetration depth of the X-ray beam is increased. During the pumping process, the two pump organs exhibit a phase shift in the expansion phase. The presence of the phase shift indicates that the temporal volume variations of the two pumps are dynamically coordinated and synchronized in each feeding cycle8. The light intensity of the CP starts to increase before the PP begins to rise. Thus, the feeding cycles of the two different mosquitoes are clearly distinguished. The feeding frequency of An. sinensis is 2.0 ± 1.2 Hz and that of Ae. togoi is 3.9 ± 1.2 Hz.

Figure 7 represents cyclic volume variations of the two pumps of a typical An. sinensis female mosquito (a) and a typical Ae. togoi female mosquito (b). The experimental data were obtained by phase-averaging five feeding cycles for each species. The expansion periods of the CP and PP of Ae. togoi in a feeding cycle are longer than those of An. sinensis. The expansion ratio of CP/PP for Ae. togoi is 1.6-fold greater than that of An. sinensis. These results demonstrate that the two mosquito species exhibit markedly different pumping behaviors in their pumping frequencies and expansion ratios.

Figure 7
figure 7

Cyclic volume variations of the two pumps, CP and PP, of. An. sinensis (a) and Ae. togoi (b). The values are the means ± standard error of the mean (SEM). Intensity (I) is normalized by the initial intensity I0. t/t0 is the dimensionless time.

Cytochemistry of actin of both mosquitoes

To investigate the local distribution of actin inside the heads, the mosquitoes were dissected and stained with FITC-conjugated phalloidin, which specifically binds to fibrous actin (F-actin). By using the TPM system, the local distribution of actin associated with muscle actuations was captured. As shown in Fig. 8, the two species of mosquitoes exhibit F-actin expression. Near the pump chamber PP, F-actin is clearly more localized in Ae. togoi than in An. sinensis. The intensity of F-actin is 44.3% higher in An. sinensis compared with Ae. togoi. The local distribution of actin inside of midgut, please refer to the supplementary information paper which accompanies this paper.

Figure 8
figure 8

TPM images of the mosquito heads.

Images of Ae. togoi stained with phalloidin-FITC (a). Images of An. sinensis stained with phalloidin-FITC (b).

Discussion

A female mosquito sucks blood to provide protein to her eggs. During this blood-feeding process, mosquitoes can transmit pathogens or parasites to the host. To feed on blood from the host, female mosquitoes utilize the pumping organs located in their heads8. The pumping organs in their heads create a large negative suction pressure to suck the blood through the proboscis into their mouths5. A few theoretical studies have been conducted on the blood-feeding process of mosquitoes. However, these studies have focused only on the pumping characteristics of a single mosquito species. In this study, we compared the blood-feeding pump systems of An. sinensis and Ae. togoi. Understanding the pump system of two different species of mosquitoes may help to understand differences in feeding behavior.

In this study, we compared the sizes of the pharynx pumps in the two different mosquitoes. The cross-sectional area of the An. sinensis PP is larger than that of Ae. togoi at rest. However, the volumes of the C-P in the expansion state are not appreciably different between the two species of female mosquitoes. Interestingly, An. sinensis and Ae. togoi also exhibit different sitting postures during rest or feeding. This feeding posture may affect the expansion or contraction of the feeding pumps. An. sinensis may have more pressure in its pumps compared with Ae. togoi. Therefore, at rest, An. sinensis may not contract its feeding pumps completely.

The differences between the feeding behaviors of the two mosquitoes are not fully understood yet. Therefore, in this study, we explored the distinctive feeding behaviors of the two mosquitoes. The expansion period of the pump organs of Ae. togoi in one feeding period is longer compared with that of An. sinensis. In addition, the pumping frequency of Ae. togoi is faster than that of An. sinensis; although Ae. togoi has a longer expansion period, it has a faster pumping frequency. This result implies that Ae. togoi contracts its feeding pump much faster than An. sinensis. This phenomenon is closely related to the different volumes of the feeding pumps of the two mosquitoes. The volume of the feeding pumps of An. sinensis is much larger than that of Ae. togoi at rest. However, the pump volume in the expansion state is not noticeably different in either mosquito species. This result may explain why Ae. togoi needs a longer expansion period compared with An. sinensis. Ae. togoi has a much smaller pump volume at rest compared with An. sinensis. Therefore, Ae. togoi needs more time to fill its pump chamber with the fluid. These results suggest some explanations for the observed differences in feeding behaviors between the two mosquito species. First, different sitting postures may explain the distinct mechanistic feeding behaviors of the two mosquitoes. The feeding posture appears to be closely related to the pumping pressure, feeding rate, salivary flow and transmission of parasitic diseases. A previous study showed that in mosquitoes that suck blood via capillary feeding, blood is driven under pressure into the mosquito. Therefore, capillary feeding is faster than pool feeding14. In addition, the position of the mosquito positioning, particularly its legs, helps maintain the proboscis at a suitable height before penetration without excess pressure on its tip15. These studies did not directly address the pressure of the feeding pump, but they mentioned the existence of pressure inside the mosquito proboscis and its tip.

Next, the viscosity of the feeding solution may affect feeding behavior. The viscosities of the iodine solution and whole blood were measured using a rheometer with increasing shear rates in our previous study8. The iodine solution exhibits a similar viscosity range with that of blood, indicating a similar non-Newtonian fluid shear-thinning effect8. The feeding response to feeding solutions of different viscosities may be dissimilar for the two species of mosquitoes in this study. These distinctive differences between the two mosquito species may lead to different feeding behaviors. Further detailed studies are required to explain the effects of the distinctive differences on their behaviors.

In conclusion, the micro-3D reconstructed structures of the pumping organs of An. sinensis and Ae. togoi female mosquitoes were observed using SR-μCT. From the 3D reconstructed images of the pump systems, the cyclic variations in the pump volumes of the two mosquitoes were quantitatively compared. The blood-sucking pump organs and the feeding behaviors of the two species were distinct. These results will aid in understanding the mechanistic functions of the blood-feeding pump system as a disease vector.

Methods

Mosquito rearing and sample preparation

Following established rearing procedures8, mosquitoes (Anopheles sinensis s.l. and Aedes togoi Theobald 1907) were fed a 10% sugar solution at 27 °C with 80% humidity and a 16 h: 8 h light/dark cycle. Larvae were fed a slurry of ground fish food and baker’s yeast. After pupation, the mosquitoes were transferred to a cage and provided a 10% sucrose-soaked cotton rod. Among the piercing-sucking proboscises of a mosquito, the labrum, which has a feather-like opaque cuticle, does not pierce the skin of a host.

Synchrotron X-ray microscopic computed tomography

All SR-μCT experiments were performed at the biomedical imaging (BMI) beamline of the Pohang Light Source-II (PLS-II) located in Pohang, Korea. This beamline was recently constructed and opened for general use in September 2013. A white beam emitted by the multi-pole wiggler source of the PLS-II 3 GeV storage ring was filtered by graphite with 1 mm thickness after passing through a double multilayer monochromator, resulting in photon energies of 10 KeV to 50 KeV. The optimal monochromatic X-ray energy for the present tomographic scanning experiments was experimentally determined to be approximately 24 KeV. The detector was positioned 20 downstream from the test sample for capturing phase-contrast images. The monochromatic X-ray image transmitted through the test sample was recorded by an imaging detector (Andor Zyla) with a spatial resolution of 2560 × 2160 pixels, upon which a YAG:Ce (30-μm thick) scintillation crystal was adhered. The X-ray image was converted on the scintillator into a visible image. All mosquito samples immersed in ethanol were placed at the tip of a heat-sealed pipette made of polypropylene that has a comparatively low X-ray absorption coefficient16. Parafilm was used to prevent alcohol evaporation. Each test sample was placed on a rotating sample stage and 2D tomographic slice images of the sample were captured by rotating the stage from 0° to 180° in 0.5° intervals. The 3D morphological structure of each mosquito sample was reconstructed from the captured tomographic images using the Octopus image-processing program. To obtain the 3D volumetric image of the sample, a filtered back-projection algorithm was applied to the projection images. Cross-sectional images were stacked to produce a 3D structural image. The stacked images were rendered as a 3D morphological structure using the Amira® 5.3.3 image-analysis software (Visualization Sciences Group, Burlington, MA, USA).

Scanning electron microscopy

Scanning electron microscopy (SEM) was employed to illustrate the morphological configurations of the test samples. Mosquito specimens were fixed with ethanol and submerged in hexamethyldisilazane or prepared by air-drying. The samples were then Ni-coated using a coater (Quorum Technology, SC7640 mode, East Sussex, United Kingdom) and examined in a Field Emission Scanning Electron Microscope (XL30S FEG, Philips Electron Optics B.V., the Netherlands) connected to an EDXS system at an acceleration voltage of 5 kV.

Two-photon microscopy

The heads of the mosquitoes were dissected in PBS and fixed in 10% buffered formalin in a 0.1-M sodium phosphate buffer (pH 7.4) and 0.25% Triton X-10017. After fixation, the mosquito samples were incubated in phalloidin-FITC (Sigma-Aldrich, St. Louis, MO, USA) overnight at 4 °C. All fluorescent images were obtained using a Leica two-photon microscopy (TPM) system (TCS SP5II MP, Leica Microscopy Systems, GMBH) with a 20× objective lens (Leica Microscopy Systems, GMBH). The snapshot images captured by TPM were analyzed and processed with LAS AF 2.7 (Leica Microscopy Systems, GMBH).

Statistical analysis

All data were statistically analyzed using the SPSS t-test (IBM, Chicago, IL, USA) at 95% coverage.

Additional Information

How to cite this article: Ha, Y.-R. et al. Comparison of the functional features of the pump organs of Anopheles sinensis and Aedes togoi. Sci. Rep. 5, 15148; doi: 10.1038/srep15148 (2015).