Host-Seeking Behavior in Hematophagous Mosquitoes (Diptera: Culicidae)

Abstract

Despite their small size, mosquitoes (Diptera: Culicidae) are pests of vast economic and medical 
importance.  They are the vectors for life-threatening human and animal diseases, in addition to being a 
buzzing, biting annoyance.  Through an understanding of mosquito host-seeking behavior, it may be 
possible to better control the spread of dangerous pathogens, which lead to such diseases as malaria and 
encephalitis.  Blood sucking (or hematophagous) mosquitoes are always females seeking to nourish 
developing eggs: a number of internal and external cues cause mosquitoes to carry out this behavior.    
Circadian rhythms, amongst other internal cues, dictate the time of day within which a specific mosquito 
species will feed.  Furthermore, an individual mosquito's host-seeking behavior is either inhibited or 
enhanced by her state in the gonotrophic cycle.  A mosquito receives external environmental information 
from sensory receptors, the physiology of which will be further reviewed in this paper.  In particular, 
anthropophilic, or mosquitoes more likely to feed on human hosts, respond to varying concentrations of 
attractants such as Carbon dioxide and L-lactic acid in the air.  Infact, mosquitoes may orient to odor 
plumes of these substances using upwind optomotor anemotaxis--much like the well studied response of 
gypsy moths to pheremones.  A mosquito, upon finding a host, then chooses a specific body region for 
feeding upon according to skin temperature and humidity.  Many factors, such as the presence of a 
pathogen within the mosquito, can alter host-seeking behavior (in this case for the increased dispersal of 
disease).  We are currently challenged to find a newer, safer means of repelling mosquitoes as they gain 
resistance to our chemical means of control.  This review will explore the evolution and physiology of 
mosquito host-seeking behavior as well as the challenge to inhibit it.

Introduction

A female mosquito seeking human  blood is faced with a very hostile quest.  She may never receive cues 
to human presence if the cues are masked by a chemical repellant.  If she is successful in sensing a host 
cue, she senses a very small change in the concentrations of carbon dioxide in the air and orients towards 
it. Upon arrival at a host, the mosquito could face such physical barriers as clothing, or bed netting to 
impede her ability to blood feed.  Other animals have their means of defense as well.  Elephants, for 
example, may avoid mosquito attack by covering themselves with mud (Edman 1984).  Even in the 
absence of physical barriers, a mosquito may be susceptible to the defensive action we may employ (ie. 
smashing, shooing, trapping, or zapping).   Yet vast numbers of mosquitoes do succeed and, in addition 
to an irritating bite, leave us with diseases like malaria.  In 1992, an estimated 365 million people were 
infected with the Plasmodium parasite, transmitted by Anopheline mosquitoes (Lehane 1996).  It is 
because of their ability to transmit disease that host-seeking behavior in mosquitoes has been and continues 
to be such an important and fascinating area of study.  J.D. Gillette said of this behavior " Most 
entomologists must go out into the world to seek their chosen quarry, that is, the particular group of 
insects that especially interests them.  We, however, have only to wait for our chosen insects to seek us 
out; for to the mosquito, or at least those that are of special concern to us, we ourselves are the quarry 
(1979)."  

Natural History

It has been proposed that hematophagy evolved as insects equipped with piercing mouthparts (for plant 
feeding)  shifted their feeding preference from other insects to vertebrate hosts--a switch which provided 
better nourishment.  It is also possible that progress toward hematophagy occurred  in small steps--from 
organic matter and feces of vertebrates to skin and blood (Waage 1979).  In the family Culicidae, only 
females have the necessary mouthparts to feed on vertebrate blood.  

The mouthparts of a female mosquito are highly modified for the purpose of blood-feeding.  The maxillae, 
for example, are sharply tapered piercing mechanisms, while the labrum forms a food channel, and a 
complex network of muscles aids in blood sucking.   Mouthparts are highly innervated with sensory 
receptors.   Receptors in the head region are responsible for mechanoreception, proprioception, and 
chemoreception.   Information from such receptors is sent to the central nervous system of the mosquito, 
where it is integrated and a behavioral response, such as the fixed action pattern for host-seeking behavior, 
is triggered.   The mosquito, however, must be in the correct physiological state in order to receive and 
respond to stimuli (Klowden 1996).  Female mosquitoes use sensory receptors to exploit both vertebrate 
host information and nectar resources for energy in between blood feedings.   Male mosquitoes feed only 
on nectar; their mouthparts are, generally, shorter and weaker than those of the female.  Strong female 
mouthparts are vital in reproduction of mosquito species.

Mosquitoes exhibit holometabolous metamorphosis and, therefore, have a life cycle consisting of egg, 
larvae, pupae, and adult.  Adult female culicids feed according to their state in the gonotrophic cycle.  
Following oviposition, the mosquito will  seek a blood meal.  Blood ingestion begins and continues until 
stretch receptors in the abdomen signal that she can't hold anymore.  This is referred to as Distention-
induced Host-seeking Inhibition.  Blood is then digested and egg maturation begins.  Host-seeking is then 
further inhibited in Oocyte-induced Host-seeking Inhibition.  Following the oviposition of this batch of 
eggs, the cycle begins again (Claudine 1996).  Variations in the cycle occur in mosquito species which 
blood feed more than once during egg maturation.  Vertebrate blood contains essential proteins necessary 
for eggs to mature.  Egg production is dependent upon amino acids and does not require the presence of 
other constituents in the blood.  Kogan found that female Aedes aegypti fed on an artificial diet high in 
protein developed equal amounts of eggs to those fed on bovine blood (1990).  Feeding is also dependent 
upon the circadian rhythms of the specific mosquito species.

Circadian rhythms determine the time of day or night that a mosquito will feed.  This is of great importance 
in determining a means of mosquito control; for example, bed netting can be used to prevent blood feeding 
while prospective hosts sleep. Mosquito behavior is also dictated by a species' tendency to blood feed on 
humans (anthropophagic) or animals (zoophilic), and whether the mosquito prefers to feed indoors 
(endophagic) or outdoors (exophagic).  Anopheles Gambiae, a vector of malaria,  to feed on sleeping 
hosts inside their homes (Gibson 1996).  Host-seeking behavior is dependent upon chemical and visual 
cues specific to a desirable host, as well as these internal species specific cues.

Carbon dioxide as a mosquito attractant

Carbon dioxide has long been recognized as a cue for mosquito host location.  Odor bursts of human 
breath contain approximately 4.5% carbon dioxide, while normal atmospheric levels of carbon dioxide are 
between .03-.04%(Gillies 1980).  As with other wind dispersed odor, a plume of carbon dioxide is 
distributed by turbulence in the atmosphere.  An insect detects the odor as bursts (Carde 1996).  Research 
done by Kennedy showed Ae. Aegypti to orient to a plume of odor in a wind tunnel using upwind 
optomotor anemotaxis (1939).  This implies that a mosquito will increase its turning rate to stay in the 
range of a carbon dioxide plume.  In contrast to gypsy moths orienting to pheremone, mosquitoes seek a 
potentially hostile host.  The  model of upwind optomotor anemotaxis by gypsy moths may not precisely 
fit the behavioral pattern of mosquitoes, but gives researchers a good idea of insect orientation 
mechanisms.  Gillies argued that a mosquito uses this technique (in the absence of visual cues) to orient to 
a host in darkness (1979). 

Sensory receptors for carbon dioxide are located in the palps, each containing 3 neurons to detect 
fluctuations in  carbon dioxide levels.  These receptors are sensitive enough to detect changes of 0.01%, 
and can't detect changes above 4.0%--a percentage corresponding to that of human breath (Bowen 1991).  
An overabundance of carbon dioxide will have a narcotic effect. This effect was used in the Bruce 
Christensen laboratory, University of Wisconsin to anesthetize mosquitoes (personal experience).
Perhaps the most interesting property of carbon dioxide in this sense is as an enhancer of other cues to 
host location.  L-lactic acid, 1-octen-3-ol, warmth, and other cues in the presence of carbon dioxide are 
considered stimulants in host-seeking behavior.	

L-Lactic acid as a mosquito attractant

In 1968, using an olfactometer, Acree et. al. recognized L-lactic acid as another chemical cue used by 
mosquitoes in host-seeking.  The group found a positive correlation between the amount of L-lactic acid in 
test subjects and the attractiveness of the subjects to Aedes Aegypti, a vector of yellow fever.  L-lactic 
acid, however, is ineffective in the absence of carbon dioxide.  In a bioassay done by Eiras et al, take-off 
rate, flight activity, landing behavior, and probing behavior were measured in terms of L-lactic acid and 
carbon dioxide concentrations (1991).  All of these behaviors, excluding landing, were generally enhanced 
by the presence of both substances.  The presence of carbon dioxide, however, was proven to be the most 
important stimulant in eliciting behaviors associated with host-seeking, with L-lactic acid enhancing the 
behaviors.

Receptors for L-lactic acid are located on the antennae in the grooved peg sensilla, and are physically 
separated from receptors of carbon dioxide.  In addition, receptors for L-lactic acid are not stimulated by  
concentrations of carbon dioxide.  The relation of these two stimulants must therefore occur as a higher 
function of the mosquito central nervous system (Eiras 1991).  Similar results were obtained in studies of 
1-octen-3-ol.  Combined with carbon dioxide, this compound may be an attractant to mosquitoes; 
however,  the validity of this compound as a mosquito attractant is currently debatable (Cork 1996).

Biting site selection

 In my review of the literature on host-seeking behavior, the most interesting studies were those on a 
mosquitoes selection of a biting site.  For zoophilic mosquitoes, host-accessibility is dependent upon 
physical barriers to blood feeding.  Many mosquito species prefer to feed on relatively hairless, or 
featherless areas of a host.  These areas are, conveniently, more vacularized to keep the host warm.  
Chipmunks and squirrels are more likely to be bitten on the ear (versus the back or feet) according to 
Edman et al (1984).  As relatively hairless beings, it seems unlikely that mosquitoes would prefer any one 
biting site over another.  Research by deJong et al proved otherwise.

DeJong used a test subject inside a bed net, wearing only close-fitting underwear, to determine body 
regions most likely to be bitten by specific mosquito species.  Mosquitoes were released into the bed net, 
and were recorded if they bit within three minutes.   E. Aegypti were found to prefer the shoulder and head 
region, while An gambiae preferred the feet (deJong 1996)..  An. Gambiae have also been proven to be 
more attracted to Limburger cheese than human breath in wind tunnel experiments (deJong 1995).  This 
suggests that bacterial stimuli may also stimulate host-seeking behavior, and is an interesting possible clue 
to better understanding host-seeking behavior in mosquitoes.  Selection of a biting site can also be further 
dictated by the presence of a parasite. 

Parasite enhancement of host-seeking

 Host-seeking behavior may also be enhanced by the presence of a parasite in the vector's body.  To 
increase their dispersal, pathogens within a mosquito vector may alter the host's behavior.  It is clear that 
by altering vector behavior to increase hematophagy, a pathogen will gain increased dispersal. Pathogens 
can affect host attractiveness, increase probing behavior,  and reduce engorgement time (Edman 1984), as 
well as alter the vector's circadian rhythms. Filarial worms, for example, develop according to the activity 
of the vector and are successfully dispersed. 

Terminating host-seeking behavior

After orienting toward a host according to chemical cues and using upwind optomotor anemotaxis, a 
mosquito has to have cues to land.  Arresting stimuli, such as sweat, humidity,  heat, color, and 
movement are possible factors which lead to the termination of  host-seeking, and the beginning of 
probing behavior.  As was mentioned earlier, host-seeking behavior is prevented by certain stages of the 
gonotrophic cycle. 

Claudine determined that the host-seeking inhibitory responses of E aegypti results form the neuropeptide 
known  as Head Peptide 1.  Increased amounts of this neuropeptide then act to desensitize sensory 
receptors to inhibit host-seeking behavior (Claudine 1993).  Claudine further noted that nutritional and 
mating states, and other hormones released during  egg maturation control the inhibition of host-seeking 
behavior (1996b).  There is a great deal of research yet to be done on hormonal control of host-seeking 
behavior as understanding of these mechanisms could lead to genetic manipulation for vector control.

Vector Control

The African Malaria mosquito, An. Gambiae is probably the most dangerous of all animals (Curtis 1996).  
This mosquito is a particularly dangerous vector as it is anthropophilic and has a long life-span.  
Mosquitoes are also the vectors for such diseases as yellow fever (E. aegypti), LaCrosse encephalitis (E. 
triseriatus), and dengue (E. albopictus).  Control of these vectors by employing our knowledge of host-
seeking behavior is an important means of improving public heath.

Effective vector control is dependent upon knowledge of a species' specific ecology.  Five elements of a 
program for vector control include 1) determining the vector species, 2) knowledge of the mosquito's 
behavior and ecology, 3)surveillance, 4) education for the people affected, and 5) control measures 
(Mitchell 1996).

Control measures relating to host-seeking behavior include trapping, repellants, and genetic manipulation.  
Exploiting a mosquito's attraction to carbon dioxide using traps is a good means of capturing mosquitoes 
for study, and could be modified to protect potential vector hosts.  Light traps, for example, used by the 
Centers for Disease Control, attract mosquitoes using dry ice and light according to Dr. Roger Nasci (pers. 
comm.).  The most commonly manufactured mosquito repellant is currently N,N-diethyl-m-toluamide 
(DEET).  Though this repellant is a step above DDT in terms of toxicity, it can be dangerous to young 
children, tends to eat away at plastics, and does not last overnight (Curtis 1996).  My father suggested 
that, if I wanted to make a million dollars, this was the industry to get into (Beaty, pers. comm.).  We are 
faced with a challenge to invent a repellant that lasts, is non-toxic, and economical.    Genetic manipulation 
of vectors is another possible means of preventing pathogen transmission.

It is already possible to make a vector non-susceptible to pathogens, but it may also be possible to switch a 
mosquitoes host preference from anthropophilic to zoophilic.  With better understanding of host-seeking 
behavior it may soon be possible to regulate genetic expression of behavioral characteristics and insect 
morphology involved in host-seeking.  As mosquitoes become resistant to chemical means of control, 
research in the area of host-seeking behavior is critical for  public health.

Acknowledgments

Mosquitoes have played an important part in my family's existence for many years, thanks to my father.  It 
is because these animals have potential to cause so much pain and suffering that my life has been so 
interesting and comfortable (ironically enough).  My thanks goes out to Dr. Barry Beaty for these things, 
but especially for never giving me a straight answer.   His response to my questions, even those on his 
favorite topic of mosquitoes, is always "LOOK IT UP."  This review served to answer my questions on a 
mosquitoes means of seeking out a "victim"--little did I know how complex the subject is, or how much is 
left to be researched.

References

Kogan PH (1990) Substitute blood meal for investigating and maintaining E. Aegypti (Diptera: Culicidae).  
J Med Entomol 27: 709-712

LehaneMJ (1996) Vector insects and their control.  In: Olfaction in mosquito-host interactions.  Wiley, 
Chichester (Ciba Found Symp 200) p8-16

Kennedy JS (1939) The visual responses of flying mosquitoes.  Proc Zool Soc Lond 109: 221a-242a

Gibson G (1996) Genetics, ecology, and behaviour of anophelines.  In: Olfaction in mosquito-	host 
interactions.  Wiley, Chichester (Ciba Found Symp200) p22-37

Curtis CF (1996) Introduction 1: an overview of mosquito biology, behaviour and importance.  In: 
Olfaction in mosquito-host interactions.  Wiley, Chichester (Ciba Found Symp200) p3-7

De Jong R, Knols BGJ (1996) Selection of biting sites by mosquitoes.   In: Olfaction in mosquito-host 
interactions.  Wiley, Chichester (Ciba Found Symp 200) p89-103

De Jong R, Knols BGJ (1995) Olfactory responses of host-seeking An gambiae ss Giles (Diptera: 
Culicidae).  Acta Trop 59: 333-335

Carde R (1996) Odour plumes and odour-mediated flight in insects.  In: Olfaction in mosquito-host 
interactions.  Wiley, Chichester (Ciba Found Symp 200) p54-70

Cork A (1996) Olfactory basis of host location by mosquitoes and other haematophagous Diptera.  In: 
Olfaction in mosquito-host interactions.  Wiley, Chichester (Ciba Found Symp 200) p71-	88

Klowden MJ (1996) Vector behavior.  In: The Biology of Disease Vectors.  University Press of Colorado, 
Niwot p34-50

Klowden MJ (1996b) Endogenous factors regulating mosquito host-seeking behaviour.  In: Olfaction in 
mosquito-host interactions.  Wiley, Chichester (Ciba Found Symp 200) p 212-225

Klowden MJ, Brown MR, Crim JW, Young L, Shrouder LA, Lea AO (1993) Endogenous regulation of 
mosquito host-seeking behavior by a neuropeptide.   J Insect Physiol 40:399-406

Gillett JD (1979) Out for blood; flight orientation upwind in the absence of visual cues.  Mosq News 
39:221-229

Gillies MT (1980) The role of carbon dioxide in host-finding by mosquitoes (Diptera: Culicidae): a review.  
Bull Entomol Res 70:525-532

Mitchell CJ (1996) Environmental management for vector control.  In: The Biology of Disease Vectors.  
University Press of Colorado, Niwot p492-501 

Bowen MF (1991) The sensory physiology of host-seeking behavior in mosquitoes.  Annu Rev Entomol 
36:139-158

Acree F, Turner RB, Gouck HK, Beroza M, Smith N (1968) L-lactic acid: a mosquito attractant isolated 
from humans.  Science161: 1346-1347

Eiras AE, Jepson PC (1991) Host-location by Ae. aegypti (Diptera: Culicidae): a wind tunnel study of 
chemical cues.  Bull Entomol Res 81:151-160.

Edman J, Day J, Walker E (1984) Vector-host interplay-factors affecting disease transmission.  In: 
Ecology of Mosquitoes:  Proceedings of a Workshop.  Florida Medical Entomology Laboratory, Vero 
Beach p273-285

Waage JK (1979) The evolution of insect/vertebrate associations.  Biol J Linn Soc 12:187-224