Bacteria- and IMD pathway-independent immune defenses against Plasmodium falciparum in Anopheles gambiae

Abstract

The mosquito Anopheles gambiae uses its innate immune system to control bacterial and Plasmodium infection of its midgut tissue. The activation of potent IMD pathway-mediated anti-Plasmodium falciparum defenses is dependent on the presence of the midgut microbiota, which activate this defense system upon parasite infection through a peptidoglycan recognition protein, PGRPLC. We employed transcriptomic and reverse genetic analyses to compare the P. falciparum infection-responsive transcriptomes of septic and aseptic mosquitoes and to determine whether bacteria-independent anti-Plasmodium defenses exist. Antibiotic treated aseptic mosquitoes mounted molecular immune responses representing a variety of immune functions upon P. falciparum infection. Among other immune factors, our analysis uncovered a serine protease inhibitor (SRPN7) and Clip-domain serine protease (CLIPC2) that were transcriptionally induced in the midgut upon P. falciparum infection, independent of bacteria. We also showed that SRPN7 negatively and CLIPC2 positively regulate the anti-Plasmodium defense, independently of the midgut-associated bacteria. Co-silencing assays suggested that these two genes may function together in a signaling cascade. Neither gene was regulated, nor modulated, by infection with the rodent malaria parasite Plasmodium berghei, suggesting that SRPN7 and CLIPC2 are components of a defense system with preferential activity towards P. falciparum. Further analysis using RNA interference determined that these genes do not regulate the anti-Plasmodium defense mediated by the IMD pathway, and both factors act as agonists of the endogenous midgut microbiota, further demonstrating the lack of functional relatedness between these genes and the bacteria-dependent activation of the IMD pathway. This is the first study confirming the existence of a bacteria-independent, anti-P. falciparum defense. Further exploration of this anti-Plasmodium defense will help clarify determinants of immune specificity in the mosquito, and expose potential gene and/or protein targets for malaria intervention strategies based on targeting the parasite in the mosquito vector.

Figure 1. Removal of bacteria from the midgut by antibiotic treatment of adult female mosquitoes.

Culture-dependent methods of bacterial cultivation (aerobic vs. anaerobic conditions) were unsuccessful at growing any bacteria from the midguts of aseptic (antibiotic-treated) mosquitoes after feeding on either (A) sugar or (B) 24 h post-blood meal. In contrast, bacteria from the midguts of septic (untreated) mosquitoes fed on (A) sugar or (B) 24 h post-blood meal could be cultured under aerobic and anaerobic conditions. (C) and (D) Culture-independent analysis of bacterial 16s rRNA by qRT-PCR measured almost no 16s rRNA in aseptic mosquitoes (sugar or blood-fed). For (A) and (B), colony forming units (CFU) from three biological replicates were pooled. For (C) and (D), 10 midguts from each treatment were assessed individually by qRT-PCR, and the relative amount of 16s rRNA from aseptic midguts (sugar or blood-fed) was compared to that of the septic (sugar or blood-fed) midgut groups, respectively. Black bars represent the mean CFU or mean -fold change, and error bars represent the standard error of the mean.

Figure 2. Global gene regulation of mosquitoes at 24 h post-P. falciparum infection under septic and aseptic conditions.

(A) Numbers of up- or down-regulated genes in distinct functional groups according to tissue (midgut/carcass) and treatment (septic/aseptic) at 24 h post-P. falciparum infection (not including DIV/UKN). (B) Same as in (A) but including DIV/UKN. (C) Venn diagrams comparing the total numbers of regulated genes between tissues and treatments. Red arrows correspond to the tissues/treatments in the left circles, and green arrows correspond to the tissues/treatments in the right circles. The arrow direction indicates up- or down-regulation. I/A: putative immunity and apoptosis; R/S/M: oxidoreductive, stress-related and mitochondrial; C/S: cytoskeletal, structural; MET: metabolism; R/T/T: replication, transcription, translation; P/D: proteolysis, digestion; TRP: transport; DIV: diverse; UKN: unknown functions.

Figure 3. Tissue-specific expression of SRPN7 and CLIPC2 after Plasmodium infection.

Fold change in transcript abundance of (A) SRPN7 and (B) CLIPC2 at 24 h post-P. falciparum infection. (C) Fold change in expression of SRPN7 and (D) CLIPC2 at 24 h post-P. berghei infection. Bars represent the mean -fold change in transcript abundance of SRPN7 and CLIPC2 between tissues (Midgut/Abdomen) and treatments (Septic/Aseptic) when compared to naïve blood-fed controls of the same tissue/treatment. Data are from three independent biological replicates, and error bars represent the standard error of the mean. Statistical analysis performed by Mann-Whitney test comparing the dCT values of infected to uninfected samples of the same tissue/treatment type resulted in no significant difference between any of the comparisons. There was also no significant difference between tissues when comparing transcript abundance of aseptic to septic samples of the same tissue compartment. These data were processed according to Livak and Schmittgen 2001 .

Figure 4. Plasmodium infection intensity in aseptic mosquitoes after depleting SRPN7 or CLIPC2 through RNAi gene silencing.

(A) P. falciparum infection intensity following RNAi-mediated depletion of SRPN7 (Dunn's post test, p<0.05) and CLIPC2 (Dunn's post test, p>0.05). (B) P. falciparum infection intensity following double RNAi-mediated depletion of SRPN7 and CLIPC2 (p = 0.87). (C) P. berghei infection intensity following RNAi-mediated depletion of SRPN7 and CLIPC2 (Kruskal-Wallis test p = 0.42). Circles represent the number of oocysts from a single midgut; horizontal black bars represent the median oocysts in each RNAi treatment. Three independent biological replicates were pooled, and significance was determined by a Kruskal-Wallis test followed by Dunn's post-test in the case of multiple comparisons. Statistical analysis of the double RNAi knockdown was performed using a Mann-Whitney test. RNAi treatments were compared to dsGFP-injected control mosquitoes.

Figure 5. Influence of SRPN7 and CLIPC2 silencing on mosquito resistance to bacterial challenge and midgut microbiota proliferation.

Adult female mosquitoes were subjected to RNAi-mediated depletion of SRPN7 or CLIPC2 transcripts and then challenged with (A) either Gram-positive Staphylococcus aureus or (B) Gram-negative Escherichia coli bacteria. Depletion of SRPN7 (p = 0.56) or CLIPC2 (p = 0.028) had no effect on the survival of mosquitoes challenged with (A) S. aureus, whereas there was a significant increase (p<0.01) in the survival of CLIPC2-depleted mosquitoes challenged with (B) E. coli but not SRPN7-depleted mosquitoes (p = 0.18). For both (A) and (B), data were pooled from three independent biological replicates (for A, n = 145; for B, n = 111), and a control group injected with dsGFP RNA was included in each replicate. Statistical significance was determined using Kaplan-Meier survival analysis with a log-rank test using Bonferonni's correction for multiple comparisons (significance  = p<0.025. (C) RNAi-mediated gene silencing of SRPN7 or CLIP2 resulted in a significant decrease (p<0.05) in the colony forming units (CFU) of cultivable midgut bacteria when compared to dsGFP-injected control mosquito midguts. Data were pooled from three independent biological replicates (n = 27 for each dsRNA group), and statistical significance was determined by one-way ANOVA followed by Dunnett's multiple comparison test. Error bars represent the standard error of the mean.

Figure 6. SRPN7 or CLIPC2 depletion has no effect on the expression of IMD pathway-regulated anti-P. falciparum genes.

(A) Silencing of SRPN7 and CLIPC2 was measured over a period of 4 days by qRT-PCR. Fifteen midguts, from aseptic mosquitoes, were pooled on each day post-injection, and the results represent the mean silencing for two independent biological replicates. Error bars represent the standard error of the mean. Expression of TEP1, FBN9, and LRRD7 genes following single knockdown of (C) SRPN7 or (D) CLIPC2. Bars represent the -fold change in expression of the listed genes on days 1–4 post-dsRNA injection, as compared to dsGFP-injected controls. qRT-PCR was used to assess changes in expression of the genes indicated above each graph. Error bars represent the standard error of the mean for three biological replicates Statistical analysis was performed at each time point by one-way analysis of variance (ANOVA) followed by Dunnett's post-test to account for multiple comparisons; all genes showed no significant difference in expression when compared to dsGFP-injected controls (not depicted).