Control of Disease Vectors using Fungi - Biological and Environmental Control of Disease Vectors - page 74

resistance among fi eld populations (N'Guessan
et al. , 2007; Kabula et al. , 2012).
Insecticides, which do not kill immediately
but do so after several gonotrophic cycles are
completed, have been termed as late-life-acting
(LLA) insecticides (Read et al. , 2009). Fungi such
as B. bassiana and M. anisopliae take from 3 to 15
days to kill their host, during which time
mosquitoes may still feed and reproduce (albeit
at a lower propensity; Scholte et al. , 2006;
Blanford et al. , 2011; Darbro et al. , 2012). The
reduction in the overall reproductive output of
mosquitoes killed by LLA insecticides is less than
25% of that of rapidly killing chemical products
(Read et al. , 2009). Models indicate that the
fi tness of mosquitoes resistant to the commonly
used chemical insecticides would be 6.5 times
that of wild insects, whereas LLA-resistant
mutants would be less than 1.5 times fi tter (Read
et al. , 2009). Therefore, selective pressure to
develop resistance to LLA insecticides would be
considerably less than that of conventional
insecticides (Koella et al. , 2009). Furthermore,
mutations conferring resistance may carry some
fi tness cost (Rowland, 1991a,b). Studies have
shown that in the absence of insecticide pressure,
the frequency of these mutations is reduced
(Casimiro et al. , 2007). If resistant mutants were
to arise, only those mosquitoes that survived up
to the point that the LLA insecticide killed the
insect would benefi t. Older mosquitoes comprise
a small proportion of the total population, yet the
whole population would have to carry the cost of
the gene for the benefi t of a few. Models have
demonstrated that, in such cases, a LLA
insecticide could be 'evolution-proof ' if the costs
of resistance outweigh the fi tness gains of
possessing the mutation (Read et al. , 2009). Due
to the complex nature of fungal invasion, it is
assumed that multiple mechanisms of resistance
would be required, resulting in the slowing of the
emergence and spread of resistance and high
fi tness costs (Blanford et al. , 2005).
Natural physiological mechanisms in
insects following fungal invasion include the
production of antifungal toxins and enzyme
inhibitors (Gillespie et al. , 2000; Dean et al. ,
2002; Rosengaus et al. , 2007). In An . gambiae ,
the injection of B . bassiana spores into the
haemolymph of adults resulted in the up-
regulation of 96% of genes, of which 11 have
been identifi ed as potentially infl uencing

immune responses (Aguilar et al. , 2005). It may
be imagined that some metabolic mechanisms
of resistance seen in pyrethroid- and DDT-
resistant mosquitoes, such as elevated
expression of detoxifying enzymes, may increase
an insect's ability to reduce fungal infection.
However, studies to date do not support this
theory. Laboratory-reared lines and wild caught
An . gambiae s.s., An . arabiensis and An . funestus
with target-site or metabolic mechanisms of
resistance to pyrethroids and DDT have been
shown to be equally susceptible to B . bassiana
and M . anisopliae formulations as fully insecticide
susceptible insects (Farenhorst et al. , 2009,
2010; Kikankie et al. , 2010; Blanford et al. ,
2011). Moreover, An . gambiae mosquitoes with
both target-site mutations and elevated
expression of detoxifying enzymes were shown
to be signifi cantly more susceptible to B . bassiana
than mosquitoes with only target-site resistance
or which were fully susceptible to pyrethroids
(Stevenson, 2008), supporting the notion that
resistance mutations may carry a fi tness cost. In
addition to demonstrating that insecticide-
resistant mosquitoes are equally or more
susceptible to fungi than insecticide-susceptible
mosquitoes, studies have also demonstrated that
prior exposure to B . bassiana and M . anisopliae
has the potential to increase mortality rates of
fi eld-caught pyrethroid-resistant An . gambiae s.s.
exposed to permethrin. Reciprocally, the same
ef ect was seen: prior exposure to permethrin
increased fungal susceptibility (Farenhorst et al. ,
2010).

Isolates with sub-lethal effects
The selective pressure for insecticide resistance is
reduced if its mode of action does not directly
result in mortality. Following research
demonstrating reduced fl ight capabilities of
desert locusts infected with M . anisopliae var.
acridum (Seyoum et al. , 1994, 2002), Blanford et
al . (2011) also demonstrated that An . stephensi
exposed to B . bassiana took longer to initiate
fl ight and experienced reduced fl ight stamina.
Elevated resting metabolic rates were recorded
in fungal-infected mosquitoes, possibly reducing
energy reserves available for fl ight. Four days
after fungal exposure, no response to feeding
cues was observed. Combined with impacts on
mosquito survival, it was estimated that malaria