deletions | additions       diff --git a/section_Discussion__.tex b/section_Discussion__.tex  index 445f8db..e258d79 100644  --- a/section_Discussion__.tex  +++ b/section_Discussion__.tex 
...
\section{Discussion}  Francisella tularensis is a Gram negative mammalian pathogen that has the potential to be used as a bioterrorist agent. The need for an improved insect model to study this pathogen is ever increasing as researchers search for alternatives to mammalian models. Here we show the tropical cockroach Blaptica dubia as a surrogate host for the in vivo study of F. tularensis LVS.  In the first demonstration of a roach model for studying F. tularensis, we show as the number of bacterial cells injected into the roaches increases, the mortality of B. dubia also increased. This indicates B. dubia has an intact immune system that is capable of defending itself from a lower concentration of F. tularensis LVS cells (104 CFU). However, at higher concentrations, the roaches are susceptible to fatality via this pathogen. Similar results have been found in recent studies using the Galleria mellonella insect model (9).   In addition, we studied the differences between a small panel of F. tularensis LVS deletion mutants in the B. dubia roach model. The lethal dose fifty-percent (LD50) values were determined for F. tularensis LVS, ΔdsbA, ΔdipA, ΔiglC, and ΔdeoB. The mutants ΔdsbA and ΔdipA are known essential genes for virulence (12). Our results express this essential gene because the killing rate of F. tularensis LVS wild type was greater than that of the ΔdsbA and ΔdipA isolates. It is hypothesized that ΔiglC, and ΔdeoB mutants are nonessential virulence genes, and our results support this claim through a decreased killing rate of B. dubia by these mutant strains and a higher LD50. These finding verify the B. dubia roach as a surrogate model host to better understand certain genes because a difference in survival of roaches can be observed between virulent and non-virulent mutants. The B. dubia roach model can therefore be used with a larger panel of mutants for screening of genomic libraries to identify more necessary virulence factors of F. tularensis.   An in vivo growth curve of F. tularensis LVS was determined with B. dubia roaches using a gentamicin protection assay. Both intracellular and extracellular growth was observed as shown in figure X. The bacterial burden of both intracellular and extracellular F. tularensis LVS increased throughout each of the selected time points post infection, with the majority of viable F. tularensis LVS being extracellular. This correlates with data that has been observed with F. novicidia in Drosophila melanogaster (13). Because only free-flowing hemolymph was extracted from the roaches, further work is needed to characterize the activity of F. tularensis LVS in other tissues of B. dubia roaches. Additional research into the antimicrobial activity of B. dubia hemolymph will also provide insight for understanding how F. tularensis LVS is able to grow extracellularly, or if intracellular growth is being inhibited.   Further applications of the B. dubia roach model include drug testing and variable incubation temperatures. Both intrahemoceol and oral administration of antibiotics are possible routes of administration. Our results indicate that select antibiotics can successful rescue B. dubia roaches from a lethal dose of F. tularensis LVS (fig X). An oral administration of streptomycin is ineffective due to inability of crossing the gastrointestinal track to reach the site of infection, while an intrahemoceol injection of the same concentration is successful at rescuing roaches to 80\% survival after seven days post-infection. This provides evidence for the potential use of the B. dubia roach model as a preliminary in vitro screening of novel drug treatments. Additionally, B. dubia roaches can be incubated at various temperatures without detrimental effects of host survival. Therefore virulence activity of F. tularensis LVS can be observed under fever-like conditions by incubating roaches at 40°C. Similarly, incubation at 30°C and 22°C is possible. A decrease in virulence of F. tularensis LVS was observed at these two lower temperatures. A decrease in virulence is also reflected in the increase in LD50 values of F. tularensis LVS at the lower temperatures compared to the groups incubated at 37°C and 40°C (table X). These data could relate to ineffective activation of temperature-dependent survival or virulence factors.   The optimization of the B. dubia insect model has aided in the investigation of F. tularensis LVS, and can be applied to other pathogens. This invertebrate model can provide insight of these microbes through high throughput investigation of virulence factors. In addition, the B. dubia model can be extended to explore both oral and systematic delivery of novel drug therapies. In conclusion, the B. dubia roach is a suitable alternative to mammalian models.