A pervasive characteristic of parasite infections is their tendency to be overdispersed. Understanding the mechanisms underlying this overdispersed distribution is of key importance as it may impact the transmission dynamics of the pathogen. Although multiple factors ranging from environmental stochasticity to inter-individual heterogeneity may explain parasite overdispersion, parasite infection is also observed to be overdispersed in inbred host population maintained under laboratory conditions, suggesting that other mechanisms at play. Here, we show that the aggregated distribution of malaria parasite within mosquito vectors is partially explained by a temporal heterogeneity in parasite infectivity triggered by the bites of mosquitoes. The transmission of the parasite was tripled between the first and the last blood fed mosquito in a period of only three hours. Surprisingly the increase in transmission is not associated with an increase in parasite investment in the production of the transmissible stage. Overall, we highlight that Plasmodium is capable of responding to the bites of mosquitoes to increase its own transmission at a much faster pace than initially thought and that this is partly responsible for overdispersed distribution of infection. We discuss the underlying mechanisms as well as the broader implications of this plastic response for the epidemiology of malaria.

To investigate the impact of mosquito bite-driven plasticity on Plasmodium transmission, three experiments were carried out in which infected birds were exposed to mosquitoes for 3 hours (6 – 9 p.m.) and mosquitoes were sampled at regular intervals thereafter (different protocols for the three experiments, see below). To investigate the impact of vector bites on parasite population growth, the parasitaemia (number of parasites in the blood) and gametocytaemia (number of gametocytes in the blood) of vertebrate hosts exposed or not (control) to mosquitoes were measured just before and just after the mosquito exposure period using blood smears.

All experiments were carried out using domestic canaries (Serinus canaria).

Experiment 1_ Oocyst burden and mosquito biting order: batch experiment

Two experimental blocks were carried out with 14 and 5 infected birds respectively with two different isolates of Plasmodium relictum. Day 11-13 post-infection, corresponding to the acute phase of infection, blood was sampled from each bird at 5:45 p.m. Straight afterwards blood sampling, birds were placed individually in an experimental cage (L40 x W40 x H40 cm).

At 6:00 p.m., 8 and 3 haphazardly chosen birds, for block 1 and 2 respectively, were exposed to four successive batches of 25 ± 3 uninfected females’ mosquitoes. Each mosquito batch was left in the cage for 45 minutes before being taken out and replaced by a new batch (i.e. batch 1 (T0min), batch 2 (T45min), batch 3 (T90min) and batch 4 (T135min)). Blood fed mosquitoes in each batch were counted and individually placed in numbered plastic tubes (30 ml) covered with a mesh with a cotton pad soaked in a 10% glucose solution. At the end of the last mosquito exposure session (9:00 p. m.) a second blood sample was taken from each bird. A red lamp was used to capture blood fed mosquitoes without disturbing the birds and the mosquitoes. Unexposed birds (control) were placed in the same experimental condition but without mosquitoes.

Tubes containing the blood fed mosquitoes were kept in standard insectary conditions to obtain an estimate of the blood meal size and the success of the infection (infection prevalence and oocyst burden). For this purpose, 7-8 days post blood meal, the females were taken out of the tubes and the amount of haematin excreted at the bottom of each tube was quantified as an estimate of the blood meal size. Females were then dissected and the number of Plasmodium oocysts in their midgut counted with the aid of a binocular microscope.

Experiment 2_ Oocyst burden and mosquito biting order: individual monitoring

To obtain a finer measurement of the impact of mosquito biting order on the oocyst burden, a second experiment was carried out with the same protocol as described above, except that exposed birds (4 of the 8 infected birds) were exposed to a single batch of 100 uninfected mosquitoes for 3h (6:00-9:00 p.m.). Female mosquitoes were continuously observed and individually removed from the cages immediately after blood feeding in order to record the order of biting of each female.

Experiment 3_ Mosquito biting order and density of parasites ingested

The third experiment was carried out to investigate whether the total number of parasites in the blood meal, immediately after the blood feeding, fluctuated during the feeding bout. Two infected birds were exposed to a single batch of 100 mosquitoes for 3h (6.00 pm – 9.00 pm) and mosquitoes were removed from the cages immediately after blood feeding. The order of biting of each female was recorded and every second mosquito collected was either immersed immediately in liquid nitrogen to quantify the number of parasites ingested by qPCR or stored in plastic tubes and dissected one week later to count the number of oocysts in the midgut.

NA = missing values

Batch_ID = 1: batch 1 (T0_min), 2: batch 2 (T45_min), 3: batch 3 (T90_min) and 4: batch 4 (T135_min)

Haematin = ng

Bird status = Exposed (Exp) or not (Unexp) to mosquito bites