behaviour and life history to increase its transmission and that these
changes would be dependent on the developmental stage of the parasite.Landing and probing on a parasitized host
An important factor influencing host selection is host defensive
behaviour, which occurs during the landing and probing stage (Burkot,
1988). Day and Edman (1983) and Dayet al. (1983) showed that animals
experimentally infected with rodent malaria,P. berghei,P chabaudiand
P. yoelii, exhibited reduced defensive behaviour and were consequently
preferentially fed upon by mosquitoes. This reduced host activity
coincided with peaks of gametocytaemia and would result in increased
transmission.
Parasite infection of the host may also lower the quality of the blood
meal by lowering the haematocrit. However, the consequence is more
freely flowing blood, so the mosquito spends less time on the host probing
for a blood meal (Daniel and Kingsolver, 1983; Rossignolet al., 1985;
Shieh and Rossignol, 1992). Recent work by Taylor and Hurd (2001)
showed that, as a result of the slight anaemia early in the infection of mice
withP. yoelii nigeriensis,A. stephensitook a larger blood meal than from
uninfected mice and hence a greater number of erythrocytes infected with
gametocytes. A further consideration was that a blood meal taken at this
early anaemic/gametocyte stage of infection would provide sufficient
protein to mature a batch of eggs; therefore the mosquito would not
require a further blood meal in the current gonotrophic cycle. This in turn
would have a beneficial outcome in terms of mosquito fitness by reducing
risky host contact and hence improving the chances of the parasite
surviving to produce sporozoites. It is likely, however, that different
transmission profiles will occur in other parasite–host–vector complexes.
A field study by Baylis and Mbwabi (1995) reported that the success of
the tsetse-flyG. pallidipesin obtaining a blood meal from oxen infected
withT. congolensewas 60% greater than when feeding on T. vivax-
infected oxen. They suggested that vasodilatation, caused by T.
congolense attaching with their flagella to the walls of the host
microvasculature, might account for this difference, rather than levels of
parasitaemia, anaemia or skin temperature. Similar results were obtained
in the laboratory by Molooet al. (2000), working with combinations of
T. congolense,T. vivax,G. pallidipesandGlossina morsitans centralis.
The results were more convincing forT. congolensethan forT. vivax
infections. These results taken together would suggest that parasites do
indeed modify their host to improve the feeding success of their vector
and thus improve their chances of transmission.
The sandflyL. longipalpispreferentially probes on the cutaneous
lesions produced when mice are infected withLeishmania mexicana
mexicana. These lesions are rich in amastigotes, the parasite stage that
infects the vector, and selective feeding on them could facilitate
Leishmaniatransmission (Coleman and Edman, 1988). In the same study,270 J.G.C. Hamilton and H. Hurd