Juvenile hormone suppresses resistance to infection in mated female Drosophila melanogaster

Summary

Hormonal signaling provides metazoans with the ability to regulate development, growth, metabolism, immune defense, and reproduction in response to internal and external stimuli. The use of hormones as central regulators of physiology makes them prime candidates for mediating allocation of resources to competing biological functions (i.e. hormonal pleiotropy) [1]. In animals, reproductive effort often results in weaker immune responses (e.g. [2–4]) and this reduction is sometimes linked to hormone signaling (see [5–7]). In the fruit fly, Drosophila melanogaster, mating and the receipt of male seminal fluid proteins results in reduced resistance to a systemic bacterial infection [8, 9]. Here, we evaluate whether the immunosuppressive effect of reproduction in female D. melanogaster is attributable to the endocrine signal juvenile hormone (JH), which promotes the development of oocytes and the synthesis and deposition of yolk protein [10, 11]. Previous work has implicated JH as immunosuppressive [12, 13], and the male seminal fluid protein Sex Peptide (SP) activates JH biosynthesis in female D. melanogaster after mating [14]. We find that transfer of SP activates synthesis of JH in the mated female, which in turn suppresses resistance to infection through the receptor germ cell-expressed (gce). We find that mated females are more likely to die from infection, suffer higher pathogen burdens, and are less able to induce their immune responses. All of these deficiencies are rescued when JH signaling is blocked. We argue that hormonal signaling is important for regulating immune system activity and, more generally, for governing trade-offs between physiological processes.

Graphical abstract

eTOC Blurb

Schwenke and Lazzaro identify the hormonal basis for post-mating immunosuppression in Drosophila melanogaster females. They find that transfer of Sex Peptide (Acp70a) during mating activates Juvenile Hormone (JH) synthesis, which suppresses resistance to bacterial infection through the receptor Germ Cell Expressed (GCE).

Results and Discussion

Under a model of antagonistic hormonal pleiotropy, reproduction and immunity are hypothesized to be interlinked by molecular cues that promote reproduction at the expense of immunity [1, 7]. Thus, we tested whether the endocrine signaling molecule JH is responsible for post-mating immune suppression in females. We first tested whether application of a synthetic JH analogue, methoprene, blocks immune system activation. Methoprene exposure (10⁻² μg) suppressed the induction of antimicrobial peptides (AMPs) by unmated females after an inoculation with heat-killed Gram-negative bacteria, Providencia rettgeri, (Tukey’s HSD, p < 0.0001), rendering them similar to untreated, mated females (Tukey’s test, p = n.s.; Figure 1A). Exposure to methoprene also significantly reduced the ability of virgin females to restrict the growth of live P. rettgeri (acetone vs methoprene across doses; t-test, t₃₇ = 6.54, p < 0.0001; Figure 1B) and survive the infection (Log-rank, $X_1^2$ = 13.9, p < 0.0001; Figure 1C). Application of methoprene to mated females also promoted fecundity, increasing the average number of eggs laid over the course of 5 days from 75.6 ± 24.3 to 95.4 ± 32.4 (t-test, t₄₅ = 2.35, p = 0.0230), again consistent with the general understanding of hormonal control of reproduction in D. melanogaster [11]. We conclude that JH facilitates reproduction but is immunosuppressive, and that its application can phenocopy the immunosuppression observed in mated females (Figure 1A).

Figure 1. Juvenile hormone is immunosuppressive.

(A) mRNA expression of antimicrobial peptide genes 8 hours after an injection with heat-killed P. rettgeri relative to CO₂ controls. Unmated females were exposed to methoprene, acetone, or CO₂ and mated females were exposed to CO₂ (Repeated-measures ANOVA, p < 0.0001, Treatment group: F_3,52 = 27.24, p < 0.0001). Tukey’s HSD was performed within each gene. Means with the same letter are not significantly different (p > 0.05) and error bars represent standard errors of the mean (SEM).

(B) Bacterial load of individual unmated females that received acetone or methoprene (ANOVA, Treatment: F_{3, 72} = 13.92, p < 0.0001). Means with the same letter are not significantly different (p > 0.05) and error bars represent one SEM. None of the sterile wound treatments (dashed lines) experienced mortality events. n = 19 ± 1; two replicates.

(C) Survivorship of unmated females subsequent to methoprene exposure and injection with sterile medium (PBS, dotted lines) or P. rettgeri (solid lines) (Cox, Chemical (infected only): $X_3^2$ = 35.98, p < 0.0001). n = 50 ± 6; two replicates.

Next, we tested whether the immunosuppressive effects of JH stem from the receipt of the male seminal fluid protein Sex Peptide (SP). SP drives a large number of physiological changes in mated D. melanogaster females [15] and is important in post-mating immunosuppression [8]. We used mRNA expression levels of JH acid methyltransferase (jhamt), which encodes a key regulatory enzyme in the JH biosynthesis cascade [16], as an indirect indicator of JH activation. jhamt expression has been previously established as a proxy for JH titres in D. melanogaster [e.g. 38]. We evaluated jhamt expression levels in females mated to: wildtype males (SP^WT), males lacking SP entirely (SP^null), or males lacking the N-terminus of SP that has been previously shown to promote JH synthesis ex vivo (SP^Δ2-7) [14, 17]. We found that females mated to SP^WT males expressed significantly higher levels of jhamt than females mated to SP^null or SP^Δ2-7 males (Tukey’s HSD, WT-null, p = 0.00977; WT-Δ2-7, p = 0.0158; Figure 2A). Thus, we conclude that transfer and receipt of the N-terminus of SP is required for JH production in mated females.

Figure 2. Transgenic males lacking the N-terminus of Sex Peptide do not elicit immunosuppression in recipient females.

(A) jhamt mRNA expression in females 10 hours after mating to SP genotypes relative to unmated females (ANOVA, Status: F_{2, 6} = 24.87, p = 0.00124). Bars represent the mean ± SEM. Means with the same letter are not significantly different (p > 0.05); three replicates.

(B) Amino acid sequences of Sex Peptide; SP^QQ: the R₇K₈ trypsin cleavage site has been changed to Q₇Q₈; SP^Δ2-7: N-terminal amino acids (E₂-R₇) deleted; SP^WT: wildtype.

(C) Bacterial load of individual females that had been infected with P. rettgeri subsequent to mating with males of different SP genotypes (ANOVA, mating status: F_{4, 189} = 13.91, p < 0.0001). Means with the same letter are not significantly different (p > 0.05) and error bars represent one SEM. n = 40 ± 6; three replicates.

(D) Infection survivorship subsequent to mating with males of different SP genotypes (Cox, mating status: $X_4^2$ = 23.37, p = 0.00011). Letters indicate levels of significance. n =110 ± 15; three replicates.

To test whether the inferred induction of JH leads to female immunosuppression, we mated females to several male genotypes expressing variant versions of SP (Figure 2B). SP is transferred to the female in the seminal fluid and binds to sperm tails by the N-terminus after reaching the female reproductive tract [18]. Bound SP is cleaved from sperm tails at a trypsin cleavage site, providing females with a continued source of the C-terminus [18]. Males with SP mutated at the trypsin cleavage site (SP^QQ) provide an intact N-terminus during mating but deprive females of long-term access to the C-terminus. Females mated to SP^null or SP^Δ2-7 males exhibited virgin-levels of bacterial load and survivorship after mating. However, females mated to SP^QQ and SP^WT males exhibited significantly higher bacterial load and lower survivorship than virgins (Figures 2C and 2D). Therefore, the immunosuppressive effect of SP can be specifically attributed to the N-terminus, which promotes JH synthesis.

To further substantiate the role of JH in post-mating immunosuppression, we tested whether blocking JH synthesis or receptor binding within the female would prevent the reduction in immunity after mating. JH is synthesized in the corpus allatum (CA). We used an inducible driver to overexpress either Diphtheria toxin (DTI) [19] or NIPP1 [20] in the CA, partially ablating the tissue in late-stage pupae to avoid any early developmental defects caused by JH removal [21]. We found that females whose CA were partially ablated exhibited bacterial loads and infection survivorship that were no different than those of virgins (Figure 3A-D). Thus, reduction of the CA was sufficient to prevent post-mating immunosuppression.

Figure 3. Genetic ablation of JH biosynthesis rescues virgin-levels of resistance.

(A)(B) Bacterial load (CFU) of individual CA+ and CA-ablated females subsequent to mating and infection. Mean bacterial load of unmated and mated females within a genotype were compared with a Wilcoxon test. (A) DTI, n = 60 ± 10; four replicates. (B) NIPP1, n = 30 ± 5; three replicates. Error bars represent one SEM.

(C)(D) Survivorship of CA+ and CA− females subsequent to mating and infection. Survivorship was compared within a genotype using a log-rank test. (C) DTI, n > 120; five replicates. (D) NIPP1, n = 48 ± 15; three replicates.

* p<0.05; ** p<0.01; *** p<0.001; **** p<0.0001

Because the CA was not fully ablated in our experiments, we performed a separate validation experiment to confirm that the observed partial ablation resulted in lower levels of JH activation after mating. First, we measured the expression of jhamt and two downstream targets of JH (mnd and JHI-21 [22]) 10 hours after mating. Expression levels for all three genes were significantly reduced by 50% or more in both CA-ablation genotypes relative to their controls (t-tests: t₄ = 5.96–15.9, p < 0.01). This is consistent with previous work showing that CA-ablated^NIPP1 females have significant reductions in JH titre [20]. Additionally, CA-ablated^DTI and CA-ablated^NIPP1 laid significantly fewer eggs (33.3 ± 29.5 and 51.6 ± 35.0) than control genotypes (130.9 ± 53.0 and 167.5 ± 82.5) (Tukey’s HSD comparisons, p < 0.05), also in concordance with prior findings [20]. Based on the full set of data, we conclude that CA-ablated females are deficient in JH synthesis and therefore exhibit reduced fecundity and virgin-levels of resistance to bacterial infection.

Finally, we sought to identify the receptor through which JH suppresses immunity in reproductively active females. Two recently-duplicated paralogs are thought to be responsible for mediating the JH signal during development [23, 24]. While Methoprene-tolerant (Met) and germ cell-expressed (gce) are partially redundant during development, it is unknown whether either or both of these receptors are required in post-mating immune suppression in adult females. We ubiquitously expressed an RNAi knockdown construct targeted against each gene in adult females. The respective knockdowns resulted in a 62% reduction in gce expression relative to the control genotype (t₄ = 6.27, p = 0.00330) and a 73.4% reduction in Met expression (t₄ = 4.21, p = 0.0136) relative to the control genotype as measured by RT-qPCR.

RNAi knockdown of gce significantly improved resistance to infection and eliminated post-mating immunosuppression (Figures 4B, D), whereas knockdown of Met had no effect on immune defense (Figures 4A, C). Specifically, bacterial loads within mated versus unmated females were not significantly different in the absence of gce (Wilcoxon, W = 494, p = n.s.; Figure 4B). In contrast, Met knockdown females continued to suffer from significantly higher bacterial loads as a consequence of mating (Wilcoxon, W = 610.5, p = 0.00796; Figure 4A). RNAi knockdown of gce improved female survivorship after mating and infection, with mated and unmated females experiencing similar rates of mortality (Log-rank, $X_1^2$ = 0.5, p = n.s.; Figure 4D). On the contrary, Met knockdown females remained immunologically sensitive to mating and experienced higher levels of infection-induced mortality as a consequence of mating (Log-rank, $X_1^2$ = 23.7, p < 0.0001; Figure 4C). Interestingly, a reduction in gce expression significantly improved survivorship relative to background controls as well (Tukey’s HSD, p < 0.05), suggesting that even basal levels of JH in unmated females may negatively influence immune defense. We predicted that if gce expression mediates resistance to infection via JH signaling, then gce knockdown females should be resistant to the immunological effects of methoprene. We tested this, and found that methoprene application increased infection-induced mortality in all genotypes except for gce-RNAi (Log-rank, $X_1^2$ = 11.3–37.6, p < 0.0001; Figures 4E, F). Thus, we conclude that GCE is the receptor that mediates the post-mating reduction in resistance driven by JH and SP, and we have solidified a role for JH as a central mediator of the physiological trade-off between reproduction and immunity in D. melanogaster.

Figure 4. RNAi-mediated knockdown of gce, a JH receptor mediates the effect of JH on immunity.

(A)(B) Bacterial load (CFU) within Met-RNAi or gce-RNAi females and their controls, respectively. Mean bacterial load of unmated and mated females within a genotype were compared with a Wilcoxon test. Error bars represent one SEM. n = 29 ± 2; three replicates.

(C)(D) Infection survivorship of Met-RNAi or gce-RNAi females and their controls, respectively. Survivorship was compared within a genotype using a log-rank test. n = 105 ± 15; four replicates.

(E)(F) Infection survivorship of unmated Met-RNAi or gce-RNAi females exposed to methoprene (MP+) or acetone (MP−). Survivorship was compared within a genotype using a log-rank test. In the absence of infection, methoprene did not impact survivorship. (E) Met, n = 80 ± 15; three replicates. (F) gce, n =120 ± 20; three replicates.

* p<0.05; ** p<0.01; *** p<0.001; **** p<0.0001

Our finding that gce alone regulates post-mating immune suppression highlights the intricate nature of the molecular action of JH. While MET and GCE have apparent redundancies [23], new evidence posits a divergence in the functionality of the two bHLH-PAS transcription factors [25–27]. For example, Reiff et al 2015 [25] demonstrated that the effect of JH on enterocyte growth and concomitant increases in reproduction are mediated largely by GCE. It is worth noting that while the duplication of the JH receptor is specific to Dipterans, gce is the ancestral gene [27, 28], suggesting JH-mediated immunosuppression might occur via a similar mechanism in other taxa.

Why has JH evolved an immunosuppressive function, and is post-mating immunosuppression adaptive? Under the immunopathology-avoidance hypothesis [29], the risk of damage from autoimmunity is potentially greater than the risk associated with of immune system functionality (i.e. being immunocompromised). If immune activation disrupts reproductive tissues and output [30, 31], such processes would be strongly selected against due to their fitness consequences. Under this hypothesis, JH may act to suppress immune signaling to prevent instances of autoimmunity, especially in cases where reproductive tissues may be targeted. Thus, immunosuppression could occur to support reproductive output.

A perhaps more likely explanation is that the trade-off stems from a simple competition for resources. Both immune function and reproduction are resource-intensive [32–34]. Under the resource-limitation hypothesis [29], JH may operate as the molecular cue for the investment in reproduction rather than immunity. While evidence for reallocation of a specific nutrient to antibacterial immunity has not been demonstrated in Drosophila, protein and specific amino acids strongly influence both reproduction and immunity in insects [e.g. 35, 36]. Recently, JH was shown to increase reproductive output through enhanced lipid metabolism, with sterile females storing more triacylglycerides [25]. Sterile females have also been shown to be resistant to the effects on mating on immunity [8]. The fat body is a tissue that drives systemic immunity, regulates central metabolism and allocation to egg provisioning, and stores lipid [37], and thus may be the organ regulating the trade-off. Whether this trade-off operates as a simple function of resource availability or whether there is a more direct antagonism remains to be conclusively demonstrated. Altogether, our work demonstrates an unambiguous role for JH in suppressing immunity in mated females, thus providing a mechanism for a classic life history trade-off and supporting the hormonal theory of pleiotropy.

Highlights.

• Male Sex Peptide activates female JH synthesis and reduces resistance to infection.

• Removal of the corpus allatum suppresses the post-mating reduction in resistance.

• JH inhibits resistance through the receptor germ cell-expressed (gce).

• Hormonal pleiotropy could mediate the evolution of life-history traits.