It is clear that mosquitos possess a sophisticated olfactory system, the outputs of which can drive orientation behavior . Attraction to the volatiles emanating from a vertebrate host is a potent driver of behavior. But, obtaining blood from an active host exposes the insect to risk [30, 31], so it follows that host-seeking behavior has evolved to balance the benefits and risks associated with the blood-feeding behavior. In other words, it is reasonable to expect that the sensitivity and specificity of CO2 receptor neurons would be optimized. The age of the mosquito and the physiological state of the insect can modulate host-seeking behavior [14, 32]. If Plasmodium-infected mosquitoes have a more sensitive peripheral sensory apparatus, this could lead to more efficient host-location and blood-feeding. This change in sensory sensitivity could, in turn, result in a more efficient transfer of the parasite between vector and host. In most cases, we do not know the precise mechanisms that underlie these modulations. Due to the long co-evolutionary history between insect vectors and their hosts, complex co-adaptive interactions between these species may have occurred, especially with regard to semiochemicals produced by hosts and detected by insects.
Mosquitoes may be infected by a variety of pathogens, some of which can be transmitted to alternative hosts. These pathogens include viruses, bacteria, fungi and protozoans such as Plasmodium. Several reports indicate that various aspects of mosquito behavior can be modulated by infection status [21, 22]. Relatively simple behaviors, such as overall activity level or basal metabolic rates, and more complex behaviors, such as host-seeking or blood-feeding, can be altered by mosquito infection status. Although it is clear that most behaviors are multistep processes, many of the mechanisms involved in eliciting particular vector behaviors are unknown or poorly defined. However, since much of an insect’s behavior can be mediated by olfactory signals, it seemed reasonable to postulate that one way to mediate behaviors would be to alter the peripheral sensory capabilities of the insect. Thus, behavior could be enhanced or suppressed as the sensitivity of the responsible receptor is regulated up or down.
Previous work has demonstrated that blood feeding alone can modulate host-seeking responses [33–35]. Mosquito midgut distention produced by blood-feeding leads to the release of peptides, which in turn reduces the sensitivity of lactic acid-sensitive receptor neurons on the antenna [34, 36]. Although these changes would provide a reasonable sensory mechanism to account for inhibition of host seeking following blood-feeding, data presented here do not reveal a reduction in the sensitivity of CO2 detection by the maxillary palp sensilla after blood-feeding. The lack of effect of blood-feeding or infection on maxillary palp sensilla CO2 sensitivity may simply reflect a fundamental difference between the lactic acid-sensitive neurons on the antenna and the carbon dioxide sensors on the maxillary palps. Recent work with Ae. aegypti has indicated that specific peptides transferred from the male to the female during mating can also influence host-seeking behavior in the female . We should note that the relatively long post-blood feeding intervals were chosen so that we could directly compare these data with the post-infection responses (10 and 20 days). It is possible that blood-feeding alone may indeed have effects on sensory sensitivity at intervals closer to the blood feeding event.
Our data indicate that infection of An. stephensi with P. berghei does not affect the sensitivity of maxillary palp sensilla of female An. stephensi with regard to the detection of carbon dioxide or octenol. This is the case for receptor neurons in mosquitoes 10 d post-infection, when oocysts are present in the midgut, and at 20 d post-infection, when sporozoites are present in the salivary glands. This lack of effect from infection on the peripheral sensory system suggests that the behavioral changes that do occur in conjunction with infection must reflect specific modifications through other sensory inputs or through central nervous system processes. It should be noted that some of the behavioral changes observed in infected mosquitoes are associated with high-intensity infections , which could lead to a general decline in the overall health of the insect rather than the modulation of a specific sensory mechanism by the pathogen. However, other studies suggest more specific modes of action .
Although the data presented here indicate that infection status does not affect the CO2-sensitive peripheral sensory neurons in maxillary palp sensilla, analysis of animals at different ages does suggest that the age of the mosquito can modulate peripheral sensitivity. There is a slight but significant shift in the threshold level responses to CO2 stimulation with increasing age. Age-related differences in peripheral sensitivity have been reported previously [9, 38]. In these earlier studies, very young Ae. aegypti (less than 5 days old) exhibited a reduced sensitivity to CO2-stimulation. In this current study with An. stephensi, we note an age-related increase in CO2 sensitivity in older insects (21–30 d post-emergence). These data suggest that older insects are more sensitive to lower concentrations of CO2, with an increasing percentage responding to the 150 ppm CO2 concentration as the insect ages. The significance of this change in sensitivity is unclear; however, its implications for blood-feeding and pathogen transmission could be important if it affects activation or host orientation. Since development of infectious sporozoites requires a prolonged extrinsic incubation period in the mosquito, older mosquitoes are more infectious than younger insects. Having a more sensitive peripheral sensory receptor system might increase the mosquito’s ability to efficiently locate a host, and could consequently promote pathogen transmission.
Host-seeking and blood-feeding behaviors are absolutely crucial steps in the transmission of the malaria parasite; and as such, they represent potential control points for reducing the spread of vector-transmitted diseases. Clearly, CO2 and octenol are two of the important signals driving host location behavior. However, the fact that we do not observe any modification of the sensitivity of peripheral sensors to carbon dioxide or octenol as a consequence of infection does not mean that other sensory inputs are similarly unaffected. The sensitivity change associated with age [ and the present study] suggests that this peripheral arm of the sensory system is more dynamic than generally assumed. These findings may have implications for control methodologies in which systems utilizing the attractive nature of carbon dioxide could be tuned to attract, repel or disorient older, potentially-infected mosquitoes. Recent work by Turner et al.[39, 40] suggests that prolonged activation of the CO2 sensors can lead to behavioral disorientation.