Molecular aspects of transferrin expression in the tsetse fly (Glossina morsitans morsitans)

Abstract

Iron is an essential element for metabolic processes intrinsic to life, and yet the properties that make iron a necessity also make it potentially deleterious. To avoid harm, iron homeostasis is achieved via proteins involved in transport and storage of iron, one of which is transferrin. We describe the temporal and spatial aspects of transferrin (GmmTsf) expression and its transcriptional regulation in tsetse where both the male and female are strictly hematophagous. Using Northern, Western and immunohistochemical analysis, we show that GmmTsf is abundant in the hemolymph and is expressed in the adult developmental stages of male and female insects. It is preferentially expressed in the female milk gland tubules and its expression appears to be cyclical and possibly regulated in synchrony with the oogenic and/or larvigenic cycle. Although no mRNA is detected, GmmTsf protein is present in the immature stages of development, apparently being transported into the intrauterine larva from the mother via the milk gland ducts. Transferrin is also detected in the vitellogenic ovary and the adult male testes, further supporting its classification as a vitellogenic protein. Similar to reports in other insects, transferrin mRNA levels increase upon bacterial challenge in tsetse suggesting that transferrin may play an additional role in immunity. Although transferrin expression is induced following bacterial challenge, it is significantly reduced in tsetse carrying midgut trypanosome infections. Analysis of tsetse that have cured the parasite challenge shows normal levels of GmmTsf. This observation suggests that the parasite in competing for the availability of limited dietary iron may manipulate host gene expression.

Fig. 1. Tissue specificity of transferrin mRNA and protein expression

A. Northern blot analysis of tissue specific expression. Lanes: 1: midgut, 2: fat body and milk gland, 3: reproductive tract, 4: carcass, 5: larvae. Five μg of total RNA from each tissue sample was loaded per lane and the blot was hybridized to transferrin and control tubulin cDNA sequences B. Tissue specific expression of transferrin protein. Lanes: 1: female carcass, 2: female fat body and milk gland, 3: female digestive tract, 4: female reproductive tract, 5: larvae, 6: male carcass, 7: male fat body, 8: male digestive tract, 9: male reproductive tract. Fly tissues were homogenized in equal volumes of buffer and homogenates from three flies were pooled for each sample. Each lane was loaded with the equivalent of 0.625% of the total protein from each sample. C. Reproductive tract specific western blot of transferrin protein. Lanes: 1: female hemolymph, 2: uterus, 3: spermatheca, 4: inactive ovary, 5: active ovary, 6: male hemolymph, 7: testes. Fly tissues were homogenized in equal volumes of cracking buffer and homogenates from five flies were pooled for each sample. Each lane was loaded with a protein amount equivalent to that of the tissue of one fly.

Fig. 2. Time course analysis of GmmTsf mRNA and protein expression

A. Northern blot analysis of GmmTsf levels over time in days post eclosion. Each time point in the course is equal to 10 μg of pooled total RNA collected from 3 individual flies for that time point. GmmTub was used as a loading control. Data from the Northern blots was quantitated by phosphoimager and normalized against GmmTub levels. Data points on the bar graphs represent the average level of expression over a 3-day period. Data points used to generate the graph are shown in the Northern blot image below the graph. B. Western blot analysis of GmmTsf levels over time in days post eclosion. Each time point in the course is equal to 10 μg of total protein collected from three individual flies for that time point. Blots were probed with polyclonal antisera specific to GmmTsf. Antibodies against Drosophila Actin were used as a loading control.

Fig. 3. Transferrin transfer dynamics between mother and intrauterine offspring

A. Northern blot analysis of GmmTsf expression in mothers versus offspring. Lanes: 1: mother of 1^st instar larvae, 2: 1^st instar larvae, 3: mother of 2^nd instar larvae, 4: 2^nd instar larvae, 5: mother of 3^rd instar larvae, 6: 3^rd instar larvae. Five μg of total RNA from each tissue sample was loaded per lane and the blot was hybridized with DIG labeled probes generated from transferrin and tubulin cDNA sequences B. Western blot analysis of GmmTsf in mothers versus offspring. Lanes: 1: embryo, 2: mother of embryo, 3: 1^st instar larvae, 4: mother of 1^st instar larvae, 5: 2^nd instar larvae, 6: mother of 2^nd instar larvae, 7: 3^rd instar larvae, 8: mother of 3^rd instar larvae. Either individual whole flies or their larvae were homogenized in equal volumes of cracking buffer. Each lane was loaded with the equivalent of 0.625% of the total protein from each sample

Fig. 4. Localization of GmmTsf by immunohistochemistry

Milk gland/fat body and reproductive tissues were dissected from pregnant flies carrying 2^nd or 3^rd instar larvae. Tissues were fixed in paraformaldehyde and stained with either preimmune serum (negative controls A + C) or antiserum generated against recombinant transferrin as primary antibodies (B + D). Tissues were stained with vector red dye and photographed under a dissecting scope. A. View of milk gland and fat body tissue (negative control). B. View of milk gland and fat body tissue (stained with transferrin antibody) C. Close up view of milk gland and fat body tissue (negative control). D. Close up view of milk gland and fat body tissue (stained with transferrin antibody)

Fig. 5. Expression of GmmTsf following immune challenge

A. qRT-PCR expression analysis of GmmTsf 24 h post immune challenge with Ytat1.1 trypanosomes and E. coli normalized against the housekeeping gene tsetse tubulin (GmmTub). (The * represents the difference between the normal sample and the E. coli sample as statistically significant with a t-test score of <0.05.) B. qRT-PCR expression of GmmTsf from Ytat1.1 infected and uninfected flies normalized against GmmTub. C. Northern analysis of GmmTsf from normal flies (lane 1), Ytat1.1 infected (lane 2) and self-cured flies (lane 3).