Bacterial Infection and Immune Responses in Lutzomyia longipalpis Sand Fly Larvae Midgut

Abstract

The midgut microbial community in insect vectors of disease is crucial for an effective immune response against infection with various human and animal pathogens. Depending on the aspects of their development, insects can acquire microbes present in soil, water, and plants. Sand flies are major vectors of leishmaniasis, and shown to harbor a wide variety of Gram-negative and Gram-positive bacteria. Sand fly larval stages acquire microorganisms from the soil, and the abundance and distribution of these microorganisms may vary depending on the sand fly species or the breeding site. Here, we assess the distribution of two bacteria commonly found within the gut of sand flies, Pantoea agglomerans and Bacillus subtilis. We demonstrate that these bacteria are able to differentially infect the larval digestive tract, and regulate the immune response in sand fly larvae. Moreover, bacterial distribution, and likely the ability to colonize the gut, is driven, at least in part, by a gradient of pH present in the gut.

Fig 1. Infection of sand fly larva midgut by B. subtilis and P. agglomerans.

EGFP- or GFP-expressing Bs and Pa were grown on LB-agar plates with selective media and fed to 3^rd instar sand fly larvae. Larvae guts were dissected and assessed for the distribution of each bacterium. In A, a schematic representation of the sand fly larval gut. Ingested food is moved from right (proventriculus–pv) to left, towards to posterior midgut and hindgut. Confocal images (1024x1024 per tile pixel resolution) of the distribution of EGFP-expressing Bs-infected and GFP-expressing Pa-infected midguts are shown in panels B and C. Posterior (pos) and anterior (ant) portions of midguts are indicated. SG, salivary glands. Inset in 1C: blow up of area of gut delineated by a rectangle (asterisk) showing biofilm formed in Pa infection. Bars = 100 µm.

Fig 2. pH of the sand fly larval gut.

Third instar sand fly larvae were fed with pH indicators Bromothymol blue (A) and Phenol red (B). Shown is the distribution of each indicator within live sand fly larvae guts, with the predicted pH for each area of the gut indicated. Live 3^rd instar sand fly larvae are shown from left (posterior or caudal setae) to right (anterior or head).

Fig 3. Effect of pH on in vitro growth of B. subtilis and P. agglomerans.

Cultured bacteria were grown on LB-agar plates of pH varying from 6–9.5. The colony forming units are measured for Bs (A) and Pa (B). One way ANOVA, with a post hoc Tukey test, was performed to assess significance. pHs 6, 6.5, and 7 were statistically different than pH 9.5 (P<0.01 for pH 6 and 6.6; and P<0.05 for pH 7).

Fig 4. Confocal images of B. subtilis and P. agglomerans infection of sand fly larvae midguts.

Anterior midgut image12h post feeding, EGFP-expressing Bs is able to infect the entire length of the midgut. A) DAPI staining; B) Shows Bs distributed throughout the anterior larval gut; C) Immuno-staining for cleaved caspase3 along the lumen of the midgut (the bright red staining in the bottom corner was due to auto-fluorescence associated with remnants of the colony chow previously fed to the larva). D) Merge. 12h post infection with GFP-expressing Pa localized to the posterior region of the midgut epithelium and induces apoptotic activity. E) DAPI staining depicting the midgut epithelium. F) Shows Pa localized on the apical portion of the lumen of the midgut. G) Immuno-staining for cleaved caspase3 along the lumen of the midgut depicting high levels of caspase3 activity. H) Merge. Bars = 50 µm.

Fig 5. mRNA expression profiles of 3^rd instar L. longipalpis larvae post infection with B. subtilis and P. agglomerans.

B. subtilis (Bs) or P. agglomerans (Pa) bacteria were fed to larvae and the expression of selected transcripts relative to agar fed control larvae was assessed. Expression profiles were obtained for the effector molecules Att and IMPer, in addition to the epithelial growth factor Vein are shown in A. D, and G, Expression profiles for the JAK/STAT receptor Domeless, the transcription factor for immunodeficiency IMD, and the negative regulator of IMD pathway Pirk are shown in are shown in B, E and H. Expression profiles for the E3 Ubiquitin ligase associated with IMD, and the effector molecules dual oxidase DUOX and Def1are shown in C, F, and I. Following bacterial feeding, total RNA was obtained at 12 h (A, B, and C), at 24 h (D, E, and F), and at 36 h (G, H, and I) post infection (PI). All Ct values were normalized to the ribosomal protein S6 (RPS6). Error bars are represented as the standard deviation determined from three biological replicates each with n = 20 midguts. Significance was determined using student t-tests. * denotes P<0.05 and ** denotes P<0.01. Y-axis, fold change.