Ammonium transporter expression in sperm of the disease vector Aedes aegypti mosquito influences male fertility

AeAmt1 Is Expressed during All Stages of Spermatogenesis and Spermiogenesis.

Previous characterizations of ammonia transporters (AMT and Rh proteins) in reproductive tissues in the context of male reproductive biology and fertility are scarce and come from vertebrate models (47, 48). In the mosquito Aedes albopictus, Rh protein (AalRh50) transcript is expressed at modest levels in ovary tissue of adult females and is unaltered following blood feeding (15). Given our previous characterizations of AMT proteins (AeAmt1 and AeAmt2) in A. aegypti larvae which are expressed in organs important for ion and water balance and digestion (32, 33, 49), we speculated that AeAmt1 and AeAmt2 would play functionally similar roles in digestive and excretory organs in adults, particularly following ingestion of a protein-rich blood meal by females. To examine this hypothesis, we utilized immunohistological techniques with previously validated AeAmt antibodies (32, 33). To our surprise, our results demonstrate that AeAmt1 protein is immunolocalized almost exclusively to the reproductive organs of both male and mated, sugar-fed female A. aegypti (Fig. 1). In a longitudinal section of a mature testis follicle from male A. aegypti, AeAmt1 is expressed in primary spermatogonia (germ cells), primary spermatocytes, spermatids, and spermatozoa (Fig. 1A). In early stages of spermatogenesis, AeAmt1 localizes to regions consistent with mitochondria found near the centriole which in later stages stretches alongside axial filament formation within the flagella (50). Live cell staining of mitochondrial derivatives that localize to the full length of the flagella of mature spermatozoa (Fig. 1B) (51) shows a localization pattern similar to observed AeAmt1 staining along the length of the flagellum of a mature spermatozoa (Fig. 1 C and D). Interestingly, AeAmt1 immunostaining appears to be absent within the centriole adjunct (Fig. 1D, red arrow), a structure located between the nuclei and flagella of spermatozoa and implicated in proper flagellar development (51).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

AeAmt1 protein localization in male testes and spermatozoa stored within the spermathecae of female Aedes aegypti. (A) AeAmt1 (red) immunolocalization in a longitudinal section of a testis counterstained for nuclei (blue; DAPI). Zones of different stages of spermatogenesis, from primary spermatogonia (germ cells) to mature spermatozoa, are indicated (white labels). (B) High-magnification image of a single spermatozoon showing the nucleus (blue; DAPI) and the mitochondrial derivatives along the length of the flagellum (green; MitoTracker). (C) High-magnification image of AeAmt1 (red) localization counterstained for nuclei (blue; DAPI) in mature spermatozoa isolated from the testis and (D) bright-field image to corresponding C indicating the flagellum (black arrow), nucleus (yellow arrow), and region of the centriole adjunct (red arrow). (E) AeAmt1 (red) localization in a transverse section of spermathecae from mated female A. aegypti counterstained for nuclei (blue; DAPI) and (F) higher-magnification image of AeAmt1 (red) and nuclei (blue; DAPI) immunolocalization in spermatozoa from a single spermatheca of mated female A. aegypti from E. (G) The bright-field image corresponding to E indicating the sperm-filled spermatheca (black arrow). (H) Control slide (anti-AeAmt1 preincubated with 20x molar excess of AeAmt1-specific peptide) and (I) bright-field image corresponding to H. (Scale bars, 50 µm, unless indicated otherwise.)

In mated, sugar-fed females, AeAmt1 protein was immunolocalized to the spermathecae (Fig. 1 E–G), and this immunostaining was absent when the AeAmt1 antibody was preabsorbed with 20× molar excess of AeAmt1 immunogenic peptide prior to application to tissue sections (Fig. 1 H and I). High-magnification images of spermathecae filled with spermatozoa reveal a flagellar-like AeAmt1 staining pattern similar to that observed in the mature spermatozoa within the male testis. The spermathecae are spherical reservoirs for spermatozoa in females, consisting of a cuticular lining surrounded by epithelial cells and glandular cells associated with the entrance of the spermathecal duct (52, 53). Our findings are consistent with those of other Aedine species in that only the largest, median spermatheca and one of the smaller lateral spermatheca are filled with sperm following copulation (46, 53, 54). AeAmt1 immunostaining localized to the nurse cells of the previtellogenic egg chambers within the ovary of sugar-fed females was also observed [SI Appendix, Fig. S1 E and F (55)]. AeAmt1 immunostaining is qualitatively absent in the midgut (MG) and Malpighian tubules (MT) (SI Appendix, Fig. S1 A–D) and the rectum (RM) epithelium and rectal pads (SI Appendix, Fig. S1 E–H), in contrast to AeAmt1 immunostaining of spermatozoa in recently inseminated female A. aegypti (SI Appendix, Fig. S1 E–H). AeAmt1 messenger ribonucleic acid (mRNA) levels corroborate these findings, whereby transcript levels are higher in the reproductive organs (RP) (combined with hindgut [HG] tissue to extract sufficient RNA for qPCR), relative to MG and MT levels (SI Appendix, Fig. S2). AeAmt1 mRNA is highest in fat body (FB) and carcass tissues (SI Appendix, Fig. S2), translating to only moderate AeAmt1 immunostaining in sections of fat body cells in comparison to the spermatozoa (SI Appendix, Fig. S1F, yellow dashed arrow). We suggest that carcass tissues (muscle, nervous, and hypodermis) may also be enriched in AeAmt1 transcript, perhaps explaining the difference in transcript and protein levels but these carcass tissues have yet to be examined. Alternatively, translation efficiency by specific cells can greatly influence the correlation between mRNA and protein levels, in part due to untranslated mRNA that remains in the cell (56). By comparison, AeAmt2 mRNA is enriched in the HG, RP, and FB (SI Appendix, Fig. S2), and we have immunolocalized AeAmt2 within the MG, MT, ovary, HG (including rectal pads), and the fat body cells of sugar-fed, mated adult females (SI Appendix, Fig. S3). From this, we postulate that AeAmt2 likely plays critical roles in ammonia detoxification and excretion during blood meal digestion, as this causes significantly elevated hemolymph [NH₄⁺] within 48 h after blood meal (SI Appendix, Fig. S4) which corresponds to simultaneously elevated ammonia concentrations in the feces (57). Interestingly, AeAmt2 mRNA and protein are also highly enriched in the antennae (SI Appendix, Fig. S3 A–D). In Anopheles gambiae, an ammonium transporter (AgAmt) sharing high sequence homology with AeAmt2 (33) is expressed in chemosensory appendages with evidence that AgAmt functions in critical host-seeking behaviors involving ammonia sensing (22, 31).

AeAmt1 Expression Is Only Detected in Spermathecae from Mated Females Containing Sperm.

In light of the finding that AeAmt1 is expressed in the flagella of mature spermatozoa and this expression persists during storage of spermatozoa within spermathecae, we predicted that spermathecae of nonmated females should lack AeAmt1 expression. Indeed, AeAmt1 immunostaining is absent in the spermathecae of nonmated females (Fig. 2 A and B). Furthermore, Western blotting analysis confirmed these findings as AeAmt1 protein was not detected in spermathecae protein homogenate from nonmated females, in contrast to AeAmt1 protein detected in the spermathecae of mated female A. aegypti (Fig. 2C). A single band at ∼55 kDa was detected, the predicted monomer form (VectorBase), which is antibody specific based on previous peptide-blocking analysis in other A. aegypti tissues (32). In an effort to elucidate a functional role for AeAmt1 in spermatozoa during storage in spermathecae, NH₄⁺-selective microelectrodes were used to measure the [NH₄⁺] within the spermathecae of mated and nonmated female A. aegypti. Mated females exhibit significantly higher NH₄⁺ levels released from spermathecae when punctured in a droplet of saline in comparison to nonmated females (Fig. 2D), and we estimate that nondiluted [NH₄⁺] levels in intact spermathecae are ∼500 mM for mated females and 300 mM for nonmated females, consistent with levels measured along the gastrointestinal tract (58) and within buoyancy organs (7) of other invertebrates. Secretion of proteins by both male accessory glands and glandular cells of spermathecae is required for sperm viability during insemination and sperm storage in insects (59, 60); however, catabolism of these proteins would yield ammonia. These remarkably high levels in contrast to hemolymph [NH₄⁺] (SI Appendix, Fig. S4) suggest that, at least in spermathecae, AeAmt1 in spermatozoa is likely functioning to mitigate ammonia toxicity. This is particularly important considering female A. aegypti undergo a single mating event with one male and sperm are stored in spermathecae for the duration of the female’s life (46, 52, 53). While it is unclear how the spermathecal cells mitigate the effects of high ammonia stored within the spermathecal reservoirs, the morphology of the spermathecae is such that the epithelial cells lie externally to the thick cuticle that lines the reservoirs, and the glandular cells are present on the surface of cuticle-lined spermathecal ducts (52, 53). This suggests that it is the cuticle that separates these cells from the high ammonia levels found within the spermathecae.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.

Absence of AeAmt1 protein expression and localization in nonmated females and ammonium (NH₄⁺) content in spermathecae of nonmated and mated female Aedes aegypti. (A) AeAmt1 (red) immunolocalization is absent in longitudinal sections of spermathecae (white outlines) from nonmated females counterstained for nuclei (blue; DAPI). (B) Bright-field image corresponding to A. (C) AeAmt1 protein expression (Top) in isolated spermathecae from mated (m) and nonmated (n) females and Coomassie total protein staining (Bottom) using Western blotting. (D) Diluted [NH₄⁺] of pooled spermathecae (3 per animal) punctured in a 0.5 µL saline droplet from nonmated and mated female A. aegypti (Mann–Whitney U test, **P = 0.0077). Data are shown as mean ± SEM. (Scale bars, 50 µm.)

During ovulation and oviposition in female A. aegypti, spermatozoa are released from the spermathecae and traverse the spermathecal ducts, and this release is highly coordinated with ovulation of the descending egg for successful egg fertilization (46). At 48 h after blood meal, we found that AeAmt1 shows continued expression in sperm flagella as they exit the spermathecae (Fig. 3 A and B) and migrate through the spermathecal ducts (Fig. 3C). AeAmt1 immunostaining was also present in the follicular epithelium and within nurse cells surrounding the developing oocyte (SI Appendix, Fig. S5), paralleling our observation of AeAmt1 localization in previtellogenic egg chambers of sugar-fed females (SI Appendix, Fig. S1 E and F). The functional significance of AeAmt1 expression in the nurse cells of both previtellogenic and vitellogenic (post blood meal) ovarian follicles is not yet known; however, the high metabolic activities of the nurse cells to meet energy and nutrient demands mirrors that of spermatozoa. In this regard, metabolic processes in sperm cells occur in part through oxidative phosphorylation whereby ammonia (NH₃/NH₄⁺) formation occurs as a by-product and requires regulation and, ultimately, excretion from the cell (61). This putative function for AeAmt1 in facilitating ammonia excretion from these highly metabolically active cells is most plausible considering the similar localizations of AeAmt1 and the mitochondrial derivatives (the site of oxidative phosphorylation) to the sperm flagellum. Findings from a comparative study using a murine model reported the production of potentially inhibitory levels of ammonium resulting from amino acid metabolism during embryonic development (62). Evidently, AeAmt1 likely has a critical function in protecting both spermatozoa and the developing oocyte against ammonia toxicity in reproductive organs in A. aegypti. Similarly, to sugar-fed females, AeAmt1 immunostaining within the midgut epithelium of females was not observed 48 h after blood meal (SI Appendix, Fig. S5B, white arrow).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3.

AeAmt1 protein localization in migrating spermatozoa within spermathecal ducts from blood-fed (48 h post blood meal, pbm) adult female Aedes aegypti. (A) AeAmt1 (red) immunolocalization in a longitudinal section of spermathecae (white outlines) counterstained for nuclei (DAPI, blue), (B) bright-field image corresponding to A, and (C) AeAmt1 (red) immunolocalization within spermathecal ducts containing migrating spermatozoa (white arrow) counterstained for nuclei (DAPI, blue) in a blood-fed female A. aegypti. (Scale bars, 50 µm.)

AeAmt1 Is Essential for Spermatozoa Viability and Male Fertility.

Considering that AeAmt1 is highly expressed in spermatozoa during the entire course of development in males, insemination of females, and storage within spermathecae, we speculated that AeAmt1 plays an essential role in sperm viability. While information from insect models is certainly lacking, a Rh protein (Rhcg) in mice is highly expressed in the epididymis and late stage spermatids in the testes of males in which the epididymal luminal fluid is maintained acidic to immobilize sperm during maturation within this organ (47, 48, 63). Consequently, Rhcg mutant mice exhibit decreased fertility. Here, we employed RNA interference (RNAi) techniques whereby AeAmt1 double-stranded ribonucleic acid (dsRNA) injection of males results in significant decreases in AeAmt1 protein in whole-body tissues within 24 h, which returns to normal levels within 36 h (Fig. 4 A and B). AeAmt1 protein knockdown at 24 h causes significant reductions in mature spermatozoa counts in the seminal vesicle (Fig. 4 C and D), a common storage unit for mature spermatozoa in male A. aegypti (64). We speculated that reductions in the number of spermatozoa are due to viability (i.e., sperm death) rather than delays in spermatogenesis and spermiogenesis, as the latter processes occur over ∼10 d in dipterans (50, 65). To examine if spermatozoa within the seminal vesicle were, indeed, dying or dead prior to when AeAmt1 protein is significantly decreased following dsRNA injection, we utilized a live/dead sperm viability assay. At 16 h after dsRNA injection, a higher proportion of cell death was observed in the AeAmt1 knockdown males, and total sperm counts were significantly reduced within the seminal vesicle (SI Appendix, Fig. S6), indicating that AeAmt1 knockdown begins earlier than 24 h after dsRNA treatment, demonstrated through Western blotting.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4.

Total spermatozoa in the seminal vesicle of male Aedes aegypti following RNAi (dsRNA)-mediated knockdown of AeAmt1. (A) AeAmt1 protein abundance in male A. aegypti at 24 and 36 h postinjection with AeAmt1 (n = 3) or control β-lactamase (β-lac; n = 3) dsRNA (*P = 0.0467 for 24 h, unpaired Student’s t test, two tailed; P = 0.100 for 36 h, Mann–Whitney U test). (B) Representative Western blot images corresponding to A demonstrating AeAmt1 protein abundance (Top) and Coomassie total protein staining (Bottom) at 24 and 36 h after AeAmt1 and β-lac dsRNA injection. (C) Total spermatozoa number within the seminal vesicle of males at 24 h after AeAmt1 and β-lac dsRNA injection (*P = 0.0073; Mann–Whitney U test; each point represents an individual replicate value). (D) Representative images of immunofluorescent nuclei (grayscale, DAPI) from fixed spermatozoa in a 1 µL droplet of PBS from the seminal vesicles of AeAmt1 and β-lac dsRNA-injected males at 24 h postinjection. Data are shown as mean ± SEM.

When nonmated females were inseminated by AeAmt1 knockdown males, significant reductions in the number of eggs laid (Fig. 5A), the number of larvae hatched (Fig. 5B), and, ultimately, the percentage of viable eggs laid per female (Fig. 5C) were observed. Although moderate levels of spermatozoa remain in the seminal vesicle following AeAmt1 protein knockdown at 24 h postinjection, we cannot discount the possibilities that sperm death continues to occur prior to, during, and after mating and insemination of the female or that physiological defects exist in these remaining sperm as a result of AeAmt1 knockdown, which would impact the ability of sperm to fertilize the egg. Mated female A. aegypti injected with AeAmt1 dsRNA produce amounts of eggs and hatched larvae similar to control females, and no significant reduction in the percentage of hatched larvae per female was observed (SI Appendix, Fig. S7 A–C), which may be due to adequate amounts of spermatozoa surviving in the spermathecae following AeAmt1 knockdown. It should be noted that high proportions of male and female A. aegypti survive following AeAmt1 dsRNA injection, with gradual decreases in female survival following blood feeding and decreases in the proportion of females that lay (viable) eggs occurring, to a greater extent in AeAmt1 knockdown groups (SI Appendix, Fig. S8).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 5.

Effects of AeAmt1 protein knockdown in males on egg laying and larval hatching (egg viability) of mated female Aedes aegypti. (A) Number of eggs laid (****P < 0.0001; Mann–Whitney U test), (B) number of larvae hatched (****P < 0.0001; Mann–Whitney U test), and (C) percentage of larval hatching (***P = 0.0006; Mann–Whitney U test) by individual females mated with males that were injected with AeAmt1 and β-lac dsRNA (n = 45 for β-lac; n = 46 for AeAmt1). Data are shown as mean ± SEM (each point represents an individual replicate value).

Taken together, these multiple lines of evidence demonstrate that AeAmt1 expression in spermatozoa is important for sperm survival and successful egg fertilization in A. aegypti. AeAmt1 expression within the developing follicles of the ovaries merits further investigation because there was no effect on fertility when females were injected with AeAmt1 dsRNA, and thus, the functional role of AeAmt1 in this respect is unclear. While our findings provide an initial description of a member of the ammonia transporter (Amt/MEP/Rh) protein family that is localized within reproductive organs in a single mosquito genera, we posit that this feature may be conserved, at the very least across insect taxa, as a method of protecting sperm from ammonia toxicity as they navigate the reproductive tract. The use of spermathecae for sperm storage and maintenance of viability until fertilization is a general characteristic of insect reproductive biology but also extends to other animal phyla, including mollusks, annelids, nematodes, and some vertebrates (e.g., amphibians) (66, 67). In this regard, protection of the spermatozoa against ammonia toxicity is a challenge that likely many animals are tasked with, and our current work highlights the importance of an AMT protein function in this aspect of insect reproduction. These findings also contribute to our understanding of A. aegypti reproductive biology, which is still in its infancy but is of medical importance because of their frequent feeding on human hosts and their high vectorial capacity for arboviral pathogens. In this context, AeAmt1 protein expression during the early stages of sperm development, maturation, and later insemination and fertilization of the egg is necessary for male fertility and, more broadly, the reproductive fitness of A. aegypti, which is significant in terms of the development of novel vector control methods.