Fig 1.
Relative levels of A. stephensi JNK1 and JNK3 phosphorylation varied among mosquito tissues but did not change with age.
Heads, thoraces, abdomens, and midguts from 15 female A. stephensi maintained on 10% sucrose were processed for western blotting at days 7, 14, and 21 post-emergence. Levels of phosphorylated JNK1 and JNK3 were normalized to GAPDH levels in the same samples. This experiment was replicated with four separate cohorts of mosquitoes. Expression levels relative to GAPDH between timepoints were analyzed by Student’s t-test.
Fig 2.
JNK1 was activated in the A. stephensi midgut in response to P. falciparum infection.
(A) Representative western blot. Female A. stephensi mosquitoes (3–5 day old) were provided a meal of P. falciparum-infected red blood cells (Pf) or a meal of uninfected red blood cells (blood). Midguts were dissected at 30 min and 3 h post-feed and processed for western blotting. Levels of GAPDH in the midgut were used to assess loading and for normalization. (B) Levels of phosphorylated JNK1 and JNK3 in midguts of infected A. stephensi were normalized to levels in mosquitoes fed uninfected blood (set at 1.0, dotted line). This experiment was replicated with four separate cohorts of mosquitoes. Fold changes relative to mosquitoes fed uninfected blood at each timepoint were analyzed by Student’s t-test.
Fig 3.
Mosquito MKPs are homologous to fruit fly, human, and mouse MKPs.
(A) A. gambiae (Ag) and A. stephensi (As) MKPs possess the highly conserved extended active site sequences and residues required for MKP catalysis, DX26VLVHCX(A/M)G(V/I)SRSX5AYLM, suggesting that the encoded proteins are enzymatically active. (B) Phylogenetic tree based on MKP catalytic domain sequences. Highly homologous mosquito MKP proteins are grouped with class II, class III, and atypical DUSPs. Class I MKPs are not encoded in the A. gambiae, A. stephensi, D. melanogaster or C. elegans genomes. Circles and squares denote A. gambiae and A. stephensi MKP proteins, respectively.
Table 1.
List of vertebrate, fly, and mosquito MKP genes.
Fig 4.
MKP overexpression in ASE cells differentially altered LPS-induced A. stephensi MAPK phosphorylation.
ERK, p38, and JNK3 phosphorylation was detected in response to LPS stimulation in ASE cells transfected with empty plasmid or plasmids overexpressing A. gambiae MKP3, MKP4, or MKP5. As noted in the text, pJNK1 levels are very low to not detectable in ASE cells. Numbers indicated below the brackets indicate fold changes in MAPK phosphorylation in LPS-treated (+) relative to untreated control (-) cells. Detection of GAPDH was used to assess protein loading and V5 detection confirmed MKP protein overexpression.
Fig 5.
MKP4 transgene transcript and protein expression were midgut-specific and induced by blood feeding in M3 and M4 lines of transgenic female A. stephensi.
(A) Diagram of the transformation construct showing the marker gene EGFP driven by the eye-specific 3XP3 promoter and the A. stephensi MKP4-HA construct driven by the A. gambiae carboxypeptidase (AgCP) promoter and flanked by pBac inverted terminal repeats. (B) EGFP fluorescence in MKP4 transgenic A. stephensi larvae. (C) Transcript levels of the MKP4 transgene and actin in M3 and M4 transgenic lines and non-transgenic (NTG) A. stephensi midgut and carcass (body minus midgut) tissues in non-blood fed (NBF) individuals and at 24 h after blood feeding (BF), and (D) in midgut tissue of M3 and M4 transgenic A. stephensi from 2–72 h post-blood feeding. (E) Transgene protein expression levels in M3 and M4 lines of MKP4 transgenic (TG) and non-transgenic (NTG) A. stephensi midgut and carcass (body minus midgut) tissues before and 24 h after a blood meal. (F) Transgene protein expression patterns in non-blood fed (NBF) transgenic and non-transgenic A. stephensi midgut and carcass and from 6–72 h after blood feeding (BF). Relative MKP4 protein expression levels for all replicates are plotted below. All gene and protein expression assays were replicated with three separate cohorts of mosquitoes.
Fig 6.
Provision of JNK-IN-8 and TCS JNK 6o in artificial blood meals inhibited JNK1 and JNK3 phosphorylation in the A. stephensi midgut.
Three-day old female A. stephensi were allowed to feed for 30 min on an uninfected blood meal containing 1 μM JNK-IN-8 or 1 μM TCS JNK 6o or on a blood meal supplemented with an equivalent volume of diluent (DMSO) as a control. A total of 25 midguts from each group were dissected and pooled at 1–3 h post feeding and processed for western blotting. JNK phosphorylation levels were normalized to control (set at 1.0, dotted line). This experiment was replicated three times with separate cohorts of mosquitoes and fold changes in pJNK levels relative to matched controls were analyzed by Student’s t-test.
Fig 7.
Midgut JNK3 phosphorylation levels were reduced below non-blood fed levels by MKP4 overexpression in P. falciparum-infected M3 line A. stephensi.
Three-day old M3 and M4 line MKP4 transgenic and non-transgenic (NTG) female A. stephensi were allowed to feed on P. falciparum-infected blood meals for 30 min. Total proteins from midguts dissected at 1, 3, 5, and 7 h after feeding were processed for western blotting. pJNK levels were first normalized to GAPDH and then to pJNK in non-blood fed (NBF) mosquitoes within each group (set at 1, black dotted line). These experiments were replicated with 2–5 separate cohorts of mosquitoes. Fold changes relative to NBF controls at each timepoint were analyzed by Student’s t-test (*, P < 0.05).
Fig 8.
MKP4 overexpression in the midgut of A. stephensi increased median lifespan in M3 line transgenic relative to non-transgenic mosquitoes.
Representative survivorship curves for M3 line MKP4 transgenic (A) and M4 line MKP4 transgenic (B) A. stephensi relative to non-transgenic (NTG) controls. Lifespan experiments were replicated four (M3) or six (M4) times with separate cohorts of mosquitoes. (C) Summary of sample sizes, medians, means, and significance relative to NTG mosquitoes as determined by Gehan-Breslow-Wilcoxon analysis within individual experiments. Meta-analysis of the replicates revealed no difference in median lifespans between M4 line and NTG females, but median lifespan of the M3 line was significantly longer than that of NTG females (one-way ANOVA, P = 0.05).
Table 2.
Analysis of replicate lifespan studies of A. stephensi provided weekly blood meals with JNK-IN-8 or TCS JNK 6o relative to controls.
Fig 9.
Provision of JNK-IN-8 and TCS JNK 6o in artificial blood meals reduced eggs laid per female A. stephensi during the first gonotrophic cycle.
Three-day old A. stephensi were provided blood meals supplemented with 1 μM JNK-IN-8, TCS JNK 6o, or blood supplemented with an equivalent volume of diluent as a control. At 24 h post feeding, females were housed individually in modified 50 ml conical tubes and provided water for oviposition for two days. Eggs were washed onto filter paper, photographed, and counted using ImageJ software. The experiment was replicated four times with separate cohorts of mosquitoes. (A) Proportions of females that laid at least one egg were analyzed by Chi-square test. (B) Clutch sizes relative to controls were analyzed by Mann-Whitney test (NS = not significant).
Fig 10.
MKP4 overexpression in the midgut of A. stephensi reduced lifetime fecundity in M3 line transgenic mosquitoes relative to non-transgenic mosquitoes.
Lifetime fecundity was assessed at the individual (A, B) and cage or population (C, D) levels. Bars represent the combined average total number of eggs per individual (A, B) or for the entire cage or population (C, D) through to the end of lifespan. The experiment was replicated four times with separate cohorts of mosquitoes and data were analyzed using Student’s t-test.
Table 3.
Moderate inhibition of JNK reduces P. falciparum oocyst development in A. stephensi.
Fig 11.
MKP4 overexpression in the midgut of A. stephensi reduced P. falciparum infection prevalence and intensity relative to non-transgenic mosquitoes.
Data are represented as percentages of mosquitoes infected (infection prevalence, A) and mean numbers of oocysts in mosquitoes with at least one midgut oocyst (infection intensity, B). Three-day old M3 and M4 line transgenic and non-transgenic (NTG) female A. stephensi were allowed to feed on P. falciparum-infected blood meals. Mosquitoes that did not feed or were not fully engorged were removed from the experiment. Ten days after infection, midguts were dissected and the number of P. falciparum oocysts counted. The experiment was replicated four times with separate cohorts of mosquitoes. Infection intensity data were analyzed using Kruskal-Wallis and Dunn’s post-test. Infection prevalences were analyzed by Chi-square test (NS = not significant).
Fig 12.
Provision of JNK-IN-8 and TCS JNK 6o in artificial blood meals reduced midgut anti-parasite gene expression in P. falciparum-infected A. stephensi.
Three-day old A. stephensi were provided blood meals containing P. falciparum-infected red blood cells supplemented with 1 μM JNK-IN-8 or 1 μM TCS JNK 6o or an identical blood meal supplemented with an equivalent volume of diluent as a control. At 3 and 24 h post-feeding, midguts were dissected for analyses of anti-parasite gene expression. These assays were replicated three times with separate cohorts of mosquitoes. Data for individual cohorts (circles, squares, triangles) are represented as transcript levels normalized to ribosomal protein S7 transcript levels and were analyzed by one-way ANOVA (* P ≤ 0.05, ** P ≤ 0.001).
Fig 13.
Treatment of ASE cells with JNK-IN-8 or TCS JNK 6o did not alter LPS-induced NF-κB signaling.
ASE cells transfected with cecropin, defensin, or gambicin luciferase promoter-reporter plasmid constructs were treated with 1 μM JNK-IN-8, 1 μM TCS JNK 6o, or mock-treated for 1 h prior to stimulation with 100 μg/ml LPS. Data are represented as luciferase activity (relative light units, RLU) normalized to media treated controls (set at 1, dotted line) for each of three replicates (squares, triangles, circles). Treatment with JNK-IN-8 or TCS JNK 6o alone had no effect on NF-κB signaling relative to media control. Further, neither of the JNK SMI + LPS treatments was different from media + LPS by Student’s t-test.
Fig 14.
MKP4 overexpression in the midgut of A. stephensi had no effect on anti-parasite immune gene expression in this tissue.
Transgenic and non-transgenic A. stephensi were allowed to feed on P. falciparum-infected blood meals. Midguts were dissected at 3 h and 24 h and processed as in Fig 15. The experiment was replicated four times with separate cohorts of mosquitoes. Data are represented as transcript levels normalized to the mosquito housekeeping gene ribosomal protein S7. Data were analyzed by one-way ANOVA.
Fig 15.
Midgut permeability was inconsistently altered by treatment with JNK SMIs and unaffected by midgut MKP4 overexpression.
Non-transgenic (control) A. stephensi (3–5 day old) were fed on artificial blood meals containing 2.0–2.4 μm magnetic fluorescent particles (Spherotech) with or without supplementation with 1 μM JNK-IN-8 or 1 μM TCS JNK 6o. Particle numbers were quantified at 72 h post-blood feeding as described in the methods. Dots represent particle numbers per five mosquitoes and bars indicate the means. Data were analyzed by one-way ANOVA.
Fig 16.
Midgut metabolites identified in A. stephensi treated with JNK SMIs.
(A) 3-D bubble plot of the number of midgut metabolites detected in this study by GC-MS/MS. In grey, number of metabolites detected and identified; blue, number of metabolites with a different abundance than controls (LOG2 ratio >0.3 and <-0.3); in red, number of metabolites with higher concentrations than controls; green, number of metabolites with concentrations lower than controls. Bubble area is proportional to the number of metabolites (indicated at the center of each bubble). (B) Linear regression of the LOG2 ratios of metabolites associated with JNK-IN-8 and TCS JNK 6o treatments. A total of 81% of metabolites identified in both treatments showed an overlap (only 11.5% were unique to JNK-IN-8 and 7.3% to TCS JNK 6o). Both treatments resulted in similar metabolomes with relatively similar concentrations as judged by regression (r2 = 0.859; p < 0.001 Pearson’s; XLSAT).
Fig 17.
Identification of midgut metabolites in JNK SMI-treated A. stephensi with fold concentrations and metabolite set enrichment analysis.
(A) Midgut metabolites identified by GC-MS/MS. Controls (n = 3) and treated samples (n = 3 per treatment) were processed to identify midgut metabolites. The relative concentrations of each metabolite were divided by the average of control levels and the resulting ratio was expressed as the LOG2. Cut-off for over-abundance was considered ≥ 0.3 whereas concentrations below that of controls were taken as ≤-0.3. (B) Summary plot for quantitative enrichment analysis (QEA). The QEA was performed using a generalized linear model to estimate a Q-statistic for each metabolite set, which describes the correlation between compound concentration and treatment. The right panel summarizes the average Q statistics for each metabolite in the input set.
Fig 18.
Pathway analysis of midgut metabolites identified in JNK SMI-treated A. stephensi.
For comparison among different pathways, node importance values calculated from centrality measures were further normalized by the sum of the importance of the pathway. Therefore, the total/maximum importance of each pathway is 1. The importance measure of each metabolite node is the percentage of the total pathway importance, and the pathway impact value is the cumulative percentage from the matched metabolite nodes.
Table 4.
List of the pathways identified and their relative impacts.
Fig 19.
Expression of PANK1 and PTEN genes in the M3 and M4 MKP4 transgenic A. stephensi lines.
PANK1 (A, C) and PTEN (B, D) expression levels were assessed by qPCR from pools of five mosquito midguts collected from 2–72 h during a reproductive cycle for the M3 (A, B) and M4 (C, D) MKP4 transgenic A. stephensi lines. All qPCR assays were performed in triplicate and normalized against ribosomal protein S7. The graphs show the fold change in PANK and PTEN expression levels between transgenic and non-transgenic sibling mosquitoes. Significant differences (p>0.05) are indicated with an asterisk. All qPCR experiments were replicated three times with distinct cohorts of mosquitoes.