Using bacteria to express mosquito larvicidal protein or antiPlasmodium molecules

Kampang and collaborators (1999) isolated from Anopheles dirus a strain of Enterobacter amnigenus able to recolonize the gut of larvae from this species This capacity looks somehow specific because the bacteria are not able to recolonize in the gut of larvae of Cx. quinquefasciatus and Ae. aegypti. En. amigenus is able to float in the water for a much longer period than Bacillus thuringiensis and B. sphaericus, two bacteria that for years have been used in the biological control of mosquitoes and other biting flies (Priest, 1992) and that tend to sink when sprayed in the water, thus requiring many applications of the bacteria to effectively control the mosquito larvae. Since the original experiments of recoloni-zation were performed keeping mosquito larvae without food supplement after feeding with bacteria, further experiments were carried out with larvae that after being fed with bacteria were continuously fed a mosquito larval diet These experiments indicated that recolonization by En. amnigenus was not due to fasting (Kampang et al ., 1999) .

When the mosquito larvicidal binary toxin of B. sphaericus 2297 was expressed in En. amnigenus, the effect on An. dirus larvae was dramatic (Kampang et al ., 2001) . Indeed, En. amnigenus carrying a recombinant plasmid containing the toxin genes under the control of the native B. sphaericus promoter not only expressed an amount of protein comparable to that found in B. sphaericus 2297, but also provided around twenty times higher toxicity toward second instar larvae of An. dirus if compared to B. sphaericus 2297. Even though the species-specific pattern of recolonization of E. amnigenus is a limiting factor to its wide application, it still remains a promising candidate for field applications of mosquito control

Riehle and collaborators (2007) genetically engineered Escherichia coli to display anti-Plasmodium effector molecules In particular, two molecules that already proved to be particularly attractive were used in this set of experiments:

1 . SM1, a small dodecapeptide able to interfere with a binding protein of the lumen of the mosquito midgut, needed for Plasmodium invasion, and, as a consequence, interfering with parasite development (Ghosh et al , 2001)

2 . PLA2, a snake venom phospholipase, which also blocks Plasmodium development in the mosquito midgut by the inhibition of the association between ookinetes and midgut surface (Zieler et al ., 2001)

E. coli bacteria expressing both molecules, and supplied with food to mosquitoes 24 hours prior an infective blood meal, inhibited the parasite development quite significantly (41% and 23%, respectively). Nevertheless, E. coli survived poorly in mosquitoes; therefore, Enterobacter agglomerans was isolated from mosquito midguts and selected for midgut survival by the means of serial passages in mosquitoes: after four passages survivorship increased from 2 days to 2 weeks . Due to the fact that En. agglomerans is widespread and nonpathogenic, it is to be regarded as a good candidate for the development of paratrans-genic protocols aimed at the control of mosquito vectors

0 0

Post a comment