Mosquitoes : researchers discovered how they find human to bite


In a paper appearing online February 6 in Science, professor of biology Paul Garrity, Ph.D. student Chloe Greppi, post-doctoral fellow Willem Laursen and several colleagues report that they’ve figured out an important part of how mosquitoes hone in on human warmth to find and bite people.

Mosquitoes are one of the planet’s deadliest animals. Hundreds of thousands of people die each year from such mosquito-borne illnesses as malaria, dengue, West Nile virus and yellow fever, most of them children. Another 200 million are infected and suffer the symptoms.

The discovery holds out the possibility of one day being able to fool or knock-out the insects’ temperature sensors so they don’t spread disease.

“Sensory systems like these are excellent targets for developing new ways to repel or confuse mosquitoes to keep them from biting us or to create new ways to help trap and kill these disease-spreading creatures,” Garrity said.

A quick history lesson

At the beginning of the 20th century, Frank Milburn Howlett, a British scientist serving in India, noticed mosquitoes were always hovering around his teapot at tea-time.

As an experiment, he filled a loose gauze bag with the insects and placed it near a test tube filled with hot water.

When warmth from the tube reached the animals, “the effect was most interesting,” he wrote in a 1910 research paper. The mosquitoes were drawn to the side of the bag closest to the rising hot air.

Howlett also observed that mosquitoes didn’t seem to attack cold-blooded animals, suggesting that it was body heat that drew them to humans.

Other research has since shown that over distances of many feet, mosquitoes rely on the carbon dioxide we exhale, the odors we give off, and visual cues to find us.

But when they get within a few inches, it’s our bodies’ temperature that plays a major role in guiding them.

Only the females of the species behave this way. As was later learned, they use the protein in our blood to nourish their eggs. Males sup only on fruit and plant nectar.

The heat-seeking behavior of Anopheles gambiae mosquitoes activated by a puff of carbon dioxide. Credit: Greppi et al., Science (2019)

Heat-seeking or cool-avoiding?

Last year, Garrity and several colleagues published a paper in the journal Neuron that upended the conventional thinking about the temperature-sensing receptors at the tip of flies’ antennas.

Traditionally, these receptors were thought to act like thermometers, taking the temperature of the surroundings to let the fly know if the environment is hot or cold.

Instead, Garrity and his colleagues found that the receptors only detected whether the temperature was changing, letting the fly know if things were getting hotter or colder.

For this reason, Garrity renamed these temperature sensors the Cooling Cells and Heating Cells. They’re so sensitive they can detect a few hundredths of a degree change in temperature per second.

Mosquitoes, who are close evolutionary relatives of flies, also have Cooling Cells and Heating Cells.

While it would seem to make sense to look at the insects’ heating cells to understand what draws them to human warmth, Garrity’s group considered an alternative – and counterintuitive – hypothesis.

Maybe it wasn’t that the insects were flying toward the heat; maybe they were flying away from the cold. This would mean the Cooling Cells would be the ones to focus on.

The specific Cooling Cells Garrity and his fellow scientists studied for their paper in Science rely on a molecular receptor called IR21a.

IR stands for ionotropic receptor, a group of proteins that help neurons to transmit signals. IR21a facilitates the transmission of a signal that the temperature around the insect is falling.

How they did it

In their experiment, the researchers knocked out the mosquito gene responsible for producing the IR21a receptor.

They then placed about 60 of the mutant insects into a shoebox-sized container with a plate on its back wall heated to near core body temperature, 98.6 degrees, and gave the mosquitoes a puff of carbon dioxide to mimic human breath.

While non-mutant mosquitoes rapidly congregated on the body temperature plate, trying to feed, the mutant mosquitoes largely ignored the plate. Without the IR21a receptor, they could no longer direct themselves to the hottest spot in their vicinity.

In a second experiment, the mosquitoes were placed in a small mesh cage. Above the cage, the researchers placed two vials full of human blood, with one heated to 73 degrees (room temperature) and the other to 88 degrees (the surface temperature of a human hand).

Compared to non-mutant mosquitoes run through the same setup, the mutants showed a reduced preference for the 88-degree blood.

Anopheles gambiae mosquitoes lacking the IR21a receptor show greatly reduced heat-seeking behavior. Credit: Greppi et al., Science (2019)

“Is the world getting better or are things getting worse?”

According to Garrity, the IR21a receptor is activated whenever mosquitoes move toward a cooler temperature. Since humans are usually warmer than their surroundings, this means that as a mosquito is approaching a human, IR21a is silent.

But if the animal should deviate from its course and start to move away from its warm-blooded prey, IR21a becomes activated, only shutting off once the insect course-corrects.

Ultimately tracking temperature change is extremely useful in helping these animals determine precisely where to bite us because blood vessels are the warmest spot on our skin.

Garrity said IR21a seems to act like “an annoying alarm. It goes off whenever the female mosquito heads towards cooler climes. When they are seeking humans, they seem to be driven to do whatever it takes to turn down the sound.”

How it all began

The gene for IR21 originated in a marine creature that lived over 400 million years ago and eventually gave rise to both modern crustaceans (like lobsters and crabs) and insects.

Once the ancestors of the modern insects finally ventured onto land, the gene was passed on to the common ancestor of both flies and mosquitoes.

When the evolutionary trajectories of these insects diverged some 200 million years ago, each species developed different uses for the IR21a receptor.

Flies use it to avoid warmth, mosquitoes to find warmth and feed on human blood.

The other authors of the study are: Gonzalo Budelli, Elaine C. Chang, Abigail M. Daniels ’19 and Lena van Giesen of Brandeis; and Andrea L. Smidler and Flaminia Catteruccia of the Harvard T.H. Chan School of Public Health.

Many of the most deadly human diseases, including malaria, Zika virus, dengue virus and chikungunya virus, are transmitted by mosquitoes (Takken & Knols 1999; Harrington et al., 2001; Lambrechts et al., 2010; Fauci and Morens, 2016).

In addition to transmitting diseases between human hosts, mosquitoes also facilitate the spillover of zoonotic pathogens into human populations (Jones et al., 2008).

As the closest relatives of Homo sapiens, great apes are of interest as potential reservoirs of zoonotic vector‐borne pathogens themselves, and mosquitoes may act as a bridge vector between great ape reservoirs of such pathogens and humans.

There is, however, a fundamental gap in knowledge concerning the transmission dynamics of vector‐borne diseases between great apes and humans, as well as the potential zoonotic threat that they pose to humans.

Studies of Plasmodium parasites in African great apes revealed that the malaria parasites Plasmodium falciparum and Plasmodium vivax, which together account for the majority of global malaria cases (WHO World Malaria Report 2018), both originated from parasites that infect African great apes (Liu et al., 20102014; Loy et al., 2018a).

Indeed, African great apes are infected with at least 13 Plasmodium species, which are further subdivided into the Laverania and Plasmodium subgenera (Liu et al., 201020142016; Loy et al., 2016,2017).

The Laverania parasites tend to be host restricted, whereas Plasmodium parasites appear to have a more promiscuous host tropism.

Although several studies have failed to detect evidence of ape Plasmodium parasites in modern African humans ( Sundararaman et al., 2013; Délicat‐Loembet et al., 2015; Loy et al., 2018b), it is not clear whether this is representive of biological barriers to infection or a lack of exposure to ape parasites. Interestingly, studies of sanctuary apes show that the Laverania host‐species restriction observed in the wild can be broken when chimpanzees and gorillas are housed in the same sanctuary (Ngoubangoye et al., 2016), suggesting that perhaps ecological factors (such as the frequency of exposure to infectious mosquitoes) impact cross‐species transmission.

Although two studies have identified some of the Anopheles species that transmit ape Plasmodium parasites (Paupy et al., 2013; Makanga et al., 2016), little is known about the behaviour of mosquitoes that could serve as bridge vectors between ape and human hosts. One understudied area is the biting behaviour and host species preferences of the mosquitoes that are found in the forest near Plasmodium‐infected apes.

Biting behaviour is largely dependent on the mosquito’s host preference, which in turn is influenced by the body odour profile of the vertebrate hosts (Takken & Verhulst 2017).

Some mosquito species have a specialised attraction towards humans (anthropophilic species), whereas others have a more opportunistic host preference (generalistic species) (Busula et al., 2015).

This preference is mediated by differences in volatile compounds produced by different host species (Busula et al., 2017) and similarities in the odour profile of host species could mediate the transmission of pathogens between these species (Verhulst et al., 2012; Verhulst et al., 2018).

In the present study, experiments were performed at the Tchimpounga Chimpanzee Rehabilitation Centre in the Republic of Congo, where chimpanzees were previously reported to harbour ape Plasmodium parasites (Pacheco et al., 2013), aiming to examine the feeding behaviour and host choice of local mosquito species.

Mosquito traps were baited with different host volatiles to assess the behaviour of mosquitoes towards different host odours. The present study also aimed to characterize the transmission dynamics of ape Plasmodium parasites by screening chimpanzees and mosquitoes for Plasmodium species.


When sampling mosquitoes at a chimpanzee rehabilitation site, situated within a natural chimpanzee habitat where wild mosquitoes would feed on the chimpanzees being rehabilitated, it was found that mosquito species exhibited different host preferences.

Moreover, the location of the trap had a significant effect on the mosquito catches, as seen in other field studies on mosquito host preference (Hiscox et al., 2014; Pombi et al., 2014). Most of the mosquito species caught during the present study, including An. obscurus and M. africana, showed a generalistic host preference and were attracted to all of the host species tested.

Interestingly, a Culex species belonging to the Cx. decens complex showed a much higher attraction towards chimpanzee odours than to human or cow odours. The blood‐feeding behaviour of mosquitoes is highly plastic and the adaptation of mosquitoes to available host species could have implications for pathogen transmission (Chaves et al., 2010; Takken & Verhulst 2013).

Studies have shown that Cx. decens feeds on both birds and bats (Boreham & Snow 1973; Quan et al., 2010). Whether Cx. decens has developed a specialised host preference for chimpanzees remains to be investigated.

The specialisation of mosquitoes could turn a generalist mosquito into a more dangerous anthropophilic vector of human infections, as seen for Aedes aegypti, which is now a dominant vector of Zika virus and dengue virus (Yakob et al., 2010; Scott & Takken 2012; Brown et al., 2014).

Mansonia mosquitoes are aggressive biting mosquitoes and are associated with Rift Valley fever virus, West Nile virus and Wucheria bancrofti (Fontenille et al., 1998; Diallo et al., 2005; Ughasi et al., 2012).

The Mansonia species caught during the present study were all morphologically identified as M. africana and showed a strong preference for cow odour‐baited mosquito traps.

Different studies in the past have observed different host preferences for Mansonia mosquitoes, ranging between anthropophilic, a high preference for cows and a generalistic host preference, respectively (Lefèvre et al., 2009; Busula et al., 2015; Omondi et al., 2015). Although the different results obtained could be a result of the study methodology, Busula et al. (2015) utilized a comparable experimental design to that used in the present study, indicating that, even within a mosquito species, different populations can have different host preferences (Takken & Verhulst 2013).

Anopheles obscurus was the predominant Anopheles species caught during the present study (97.4%).

This typical marsh and swamp breeder is an opportunistic species biting both bovine and primate hosts (Boorman and Service, Boorman & Service 1960). 

Anopheles obscurus has also been associated with feeding on ungulate hosts such as Cephalophus (duikers) species and found to be infected with ungulate Plasmodium spp. (Boundenga et al., 2016; Makanga et al., 2017).

Experimental studies have shown that Anopheles mosquitoes with different host preferences are similarly attracted towards non‐human primates and humans (Verhulst et al., 2018). Moreover, a field study by Makanga et al. (2016) identified mosquitoes biting both humans and great apes.

The Anopheles mosquitoes in the present study were equally attracted towards chimpanzee and human odours. However, although 21 An. obscurus mosquitoes were infected with a variety of Plasmodiidae species, no primate Plasmodium was detected, which suggests that An. obscurus in not a competent vector.

Because of the lack of well‐defined reference sequences, it was impossible to classify the parasites to genus level with any kind of certainty.

Therefore, it was decided to assign them to the Plasmodiidae family, which includes both Plasmodium and Hepatocystis. Other mosquito species such as An. gambiaeAn. ziemanniAn. nili and An. moucheti caught in the present study are known Plasmodium vectors and may play a role as bridge vectors in the circulation of P. falciparum in this area. Although their abundance was very low, this could still be sufficient to sustain transmission (Homan et al., 2016).

Human P. falciparum was found in a chimpanzee blood sample and a human blood sample from a Mansonia mosquito (Fig. 3), which indicated that chimpanzees can be infected with P. falciparum.

Although a chimpanzee to chimpanzee transmission of P. falciparum cannot be ruled out, it is likely that this represents transmission of P. falciparum from humans to chimpanzees, which has previously been observed in apes held in captivity (Krief et al., 2010; Prugnolle et al., 2010; Pacheco et al., 2013; Ngoubangoye et al., 2016).

By contrast, P. falciparum has never been detected in wild chimpanzees, bonobos or gorillas, which suggests that there is something unique about the sanctuary environment that facilitates cross‐species transmission.

It is possible that the rangers working in the sanctuary harbour subclinical densities of P. falciparum (Rayner et al., 2011; Maselli et al., 2014; Rovira‐Vallbona et al., 2017), which, when picked up by anopheline mosquitoes, can be transmitted to the chimpanzees held in the sanctuary.

However, it was not possible to identify bridge vectors for P. falciparum because none of the Anopheles specimens collected were infected with human P. falciparum, although some of the mosquitoes caught at Tchimpounga (An. ziemanniAn. niliAn. moucheti and gambiae s.l.) are known human malaria vectors (Greenwood et al., 2005; Scott & Takken 2012; Sinka et al., 2012).

Additional work is needed to determine the factors that facilitate cross species transmission of P. falciparum in sanctuary apes.

Unravelling the trophic behaviour of mosquitoes in remote locations will provide us with valuable information on potential transmission pathways of vector‐borne diseases between non‐human primates. With increasing human activities in natural environments, the risk of zoonoses increases, as seen in the cases of a non‐human primate derived P. vivax infecting a Caucasian man travelling from Africa and the transmission of Plasmodium knowlesi from monkey to humans in South‐East Asia (Cox‐Singh et al., 2008; Singh et al., 2004; Paupy et al., 2013).

This is further supported by recent zoonotic transmission of P. simium from simians to humans in the Atlantic Forest of Rio de Janeiro (Brasil et al., 2017). Mosquito plasticity in host preference has recently been shown as a major factor for disease outbreaks (Yakob et al., 2018).

In addition, adaptation of arthropod‐borne viruses to new mosquito vectors is a major concern in the spread of vector‐borne diseases and has already caused outbreaks of chikungunya virus and West‐Nile virus (Kilpatrick et al., 2006, Kilpatrick et al., 2008, Tsetsarkin et al., 20072016).

The trophic behaviour of mosquitoes was determined in the present study using a variety of methods, including odour‐baited traps, analyses of blood meal and pathogens of mosquitoes. Each of these methods alone would have led to different conclusions, which showed that a combination of methods is required to be able to fully understand the behaviour of disease vectors.

The present study provides evidence that the majority of mosquito species collected near wild apes are attracted to odours of multiple different species, including chimpanzees and humans, and may thus serve as bridge vectors for ape pathogens.

The identification of P. falciparum in both humans and chimpanzees suggests that there is active circulation of these parasites transmitted by anopheline mosquitoes, although it is unlikely that An. obscurus is part of this transmission cycle given that, regardless of their high abundance, no primate Plasmodium positive specimens were found. Primary human Plasmodium vectors such as An. gambiae s.l., An. nili and An. moucheti, were collected during this study period, which could have played a role as bridge vectors for the circulation of P. falciparum.

More information: C. Greppi el al., “Mosquito heat seeking is driven by an ancestral cooling receptor,” Science (2020). … 1126/science.aay9847

C.R. Lazzari at Institut de Recherche sur la Biologie de l’Insecte in Tours, France el al., “In the heat of the night,” Science (2020). … 1126/science.aba4484


Please enter your comment!
Please enter your name here

Questo sito usa Akismet per ridurre lo spam. Scopri come i tuoi dati vengono elaborati.