Team:EPF Lausanne/Project droso
From 2010.igem.org
(→Conclusion) |
(→Conclusion) |
||
Line 64: | Line 64: | ||
As mentioned before, bacteria that live in insects’ gut are not common and it is possible that because it cannot survive in Drosophila, it is actually very specific to mosquitoes. This specificity of Asaia for Culicidae is reinforced by two recent studies that have been unable to find Asaia in other insect genus [2,3]. Therefore we could modify asaia with little risk of the engineered bacteria to spread to other insect in the wildlife. | As mentioned before, bacteria that live in insects’ gut are not common and it is possible that because it cannot survive in Drosophila, it is actually very specific to mosquitoes. This specificity of Asaia for Culicidae is reinforced by two recent studies that have been unable to find Asaia in other insect genus [2,3]. Therefore we could modify asaia with little risk of the engineered bacteria to spread to other insect in the wildlife. | ||
- | |||
+ | <b>References </b> | ||
<br>[1] Liehl et al Plos Pathogen 2006 Prevalence of local immune response against oral infectionin a Drosophila/Pseudomonas infection model | <br>[1] Liehl et al Plos Pathogen 2006 Prevalence of local immune response against oral infectionin a Drosophila/Pseudomonas infection model | ||
<br>[2] Crotti et al 2010 AEM Acetic acid bacteria, new emerging symbionts of insects | <br>[2] Crotti et al 2010 AEM Acetic acid bacteria, new emerging symbionts of insects |
Revision as of 23:54, 24 October 2010
Contents |
Experiments on Drosophilia
The final goal of our project is for our modified Asaia to survive and produce proteins in the mosquito's gut. Working with mosquitoes however requires special equipment that we do not have at EPFL, and we wondered if we could work on another insect which is less demanding. We therefore turned towards Drosophila, commonly known as the fruit fly, which is much easier to work with.
Considering the fact that bacteria that live in the guts of insects are not very common, we assumed that there was a fair chance that Asaia could persist in Drosophila and that we could use the it as an alternative to mosquitos for our basic experiments.
Using Drosophila melanogaster we aimed to address two questions:
i. Is Asaia toxic for Drosophila?
ii.Is Asaia able to colonize the Drosophila gut and persist?
(See Materials and Methods for details on how the experiments were conducted.)
Our main results
1) Asaia is not toxic for Drosophila
We infected Drosophila with different bacterial strains, a pathogenic control starin (P. entomophila), a non-pathogenic control strain (Ecc 15) and our Asaia bacteria. For these experiments we used two different fly strains (Oregon and Relish).
We found that Asaia does not cause significantly more deaths than the non-pathogenic bacteria or in the uninfected control (Figure 2).
2) Asaia is not persistant in Drosophila
With this last experiment we wanted to ascertain wether asaia persisted in drosophila and quantize how many asaia were present in the drosophila’s gut after 3h, 24h and 48h.
To be able to “count” the number of asaia at these three different periods of time, we had to retrieve the bacteria inside the flies, plate them and after incubation count the number of colonies present.
To do this we disinfected the exterior of the flies by washing them with ethanol (for less than five seconds) and then rinsing them with water. The flies were then crushed in the medium corresponding to the bacteria we were interested in (i.e: Gly+LB for asaia, LB for Pe, etc). We then did a serial dilution eleven times with a factor of ten with the crushed flies, and plated each dilution with antibiotics to specifically select the bacteria we were interested in.
Conclusion
The survival assay reveales that Asaia is not lethal for Drosophila. Asaia is likely to be rapidly cleared by Drosophila innate defense mechanisms such as reactive oxygen species, gut pH.
The persistence assay showed the inability of Asaia to establish itself in the fruit fly gut. This enables us to use Drosophila as a substitute to mosquitoes for our characterization assays.
As mentioned before, bacteria that live in insects’ gut are not common and it is possible that because it cannot survive in Drosophila, it is actually very specific to mosquitoes. This specificity of Asaia for Culicidae is reinforced by two recent studies that have been unable to find Asaia in other insect genus [2,3]. Therefore we could modify asaia with little risk of the engineered bacteria to spread to other insect in the wildlife.
References
[1] Liehl et al Plos Pathogen 2006 Prevalence of local immune response against oral infectionin a Drosophila/Pseudomonas infection model
[2] Crotti et al 2010 AEM Acetic acid bacteria, new emerging symbionts of insects
[3] Bessem et al 2010 AEM Typing of Asaia spp. bacterial symbionts in four mosquito species molecular evidence for multiple infections