Soma-to-germline BMP signal is essential for Drosophila spermiogenesis

Abstract

In the Drosophila testis, developing germ cells are encapsulated by somatic support cells throughout development. Soma-germline interactions are essential for successful spermiogenesis. However, it is still not fully understood what signaling events take place between the soma and the germline. In this study, we found that a Bone Morphogenetic Protein (BMP) ligand, Glass bottom boat (Gbb), secreted from somatic cyst cells (CCs), signals to differentiating germ cells to maintain proper spermiogenesis. Knockdown of Gbb in CCs or the type I BMP receptor Saxophone (Sax) in germ cells leads to a defect in sperm head bundling and decreased fertility. Our Transmission Electron Microscopy (TEM) analyses revealed that the mutant germ cells have aberrant morphology of mitochondria throughout the stages of spermiogenesis and exhibit a defect in nebenkern formation. Elongating spermatids show uncoupled nuclei and elongating mitochondrial derivatives, suggesting that improper mitochondrial development may cause sperm bundling defects. Taken together, we propose a new role of soma-derived BMP signaling, which is essential for spermiogenesis.

1. Introduction

Bone Morphogenetic Protein (BMP) signaling is involved in a wide array of biological processes including organogenesis, immune response, and tissue homeostasis (reviewed in (Wang et al., 2014) (Sconocchia et al., 2021)). In the Drosophila testis, BMP signaling is responsible for maintaining the stem cell identity of germline stem cells (GSCs). GSCs contact somatic hub cells at the tip of the testis which secrete the BMP ligands Decapentaplegic (Dpp) and Glass bottom boat (Gbb) (Kawase et al., 2004). These ligands signal through a complex of the type I receptor Thickveins (Tkv) and the type II receptor Punt on GSCs (Kawase et al., 2004; Shivdasani et al., 2003). Binding of Dpp and Gbb to Punt and Tkv activates Tkv to phosphorylate the downstream effector Mothers against Dpp (Mad) intracellularly (Kawase et al., 2004). Phosphorylated Mad, along with its cofactor Medea (Med), then enter the nucleus. In GSCs, Mad represses the differentiation factor Bag of marbles (Bam) to maintain stem cell identity (Kawase et al., 2004).

After the asymmetric division of GSCs, the differentiating daughter gonialblast (GB) is pushed away from the niche, losing contact-dependent BMP signaling from the hub and beginning differentiation (de Cuevas et al., 2011). With each step of spermatogenesis, the differentiating cells move farther away from the GSC niche at the apical tip and closer to the distal end of the testis and seminal vesicle where mature sperm are stored (de Cuevas et al., 2011). Typically, the GB undergoes four rounds of mitotic division with incomplete cytokinesis through the 2-, 4-, 8-, and 16-cell spermatogonia (SG) stages before growing into spermatocytes (SCs) and finally entering meiosis (Siddall et al., 2017). However, when tissue homeostasis becomes unbalanced with too few stem cells in the niche, differentiating cells can be called back to the hub through de-differentiation (Sheng et al., 2009).

Interestingly, de-differentiation is also regulated by BMP signaling. Recent work has demonstrated that Dpp can freely diffuse from the hub, and this diffusible fraction of Dpp signals in a different manner than contact-dependent Dpp signaling at the Hub-GSC interface (Ridwan et al., 2024). Diffusible Dpp signals through the type II receptor Punt in complex with either of the type I receptors Tkv or Saxophone (Sax) on GBs and SGs to upregulate Bam expression and prevent differentiating cells from returning to the niche through de-differentiation (Ridwan et al., 2024). Moreover, the expression of Bam is known to regulate the number of transit-amplifying SG divisions (Insco et al., 2009). Throughout the process of spermatogenesis, each differentiating germline cyst is encapsulated by a pair of somatic cyst cells (CCs) (Schulz et al., 2002). Sax, together with the downstream effector Smad on X (Smox), is also found in these CCs, where it contributes to restricting the over proliferation of SGs (Li et al., 2007). Mutant testes lacking either Sax or Smox in CCs have excessive spermatogonial division (Li et al., 2007). These studies have suggested that BMP signaling is not exclusively occurring between the niche and stem cells in the testis.

Although it is well known that soma-germline interactions are vital throughout spermatogenesis in the Drosophila testes, signaling pathways between CCs and developing germ cells have not been fully identified (Zoller et al., 2012). Direct contact between SGs and CCs is required for germline proliferation since mutations in zero population growth (zpg) and discs large (dlg), involved in gap junctions and septate junctions, respectively, exhibit SG death (Tazuke et al., 2002; Papagiannouli et al., 2009). However, it is unclear what regulatory signals are exchanged at these junctions. SGs are also known to signal to CCs through the epidermal growth factor receptor (EGFR) since loss of the EGF ligand Spitz (Spi), the Spi activating protease Stem cell tumor (Stet), and the EGFR itself lead to buildup of SGs and failure to transition to the SC stage (Schulz et al., 2002; Zoller et al., 2012; Sarkar et al., 2007). Though signaling from SGs to CCs has been described, there are currently no known signaling pathways from CCs to SGs or SCs (Zoller et al., 2012).

2. Results

2.1. Sax is required in the germline for proper spermiogenesis

After meiosis, syncytial cysts of 64 haploid spermatids undergo drastic changes in their nuclei (sperm heads), mitochondria, and flagellar axonemes, followed by plasma membrane remodeling and individualization (Fabian et al., 2012). Nuclear remodeling occurs through chromatin condensation facilitated by the exchange of histones for transition proteins and protamines (Rathke et al., 2007). Stages of spermiogenesis are easily identified and named by their level of chromatin condensation; round, leaf, canoe, and needle (Fabian et al., 2012). In the late canoe stage, spermatids transition to protamine-based chromatin, allowing the nucleus to form a compact needle shape (Rathke et al., 2007; Awe et al., 2010). Throughout the process of spermiogenesis, sperm nuclei are bundled together until they are transferred to the seminal vesicle (Fabian et al., 2012).

Knockdown of Sax in differentiating germline cells under the bamGal4 driver caused a defect in sperm head bundling demonstrated by complete scattering of sperm heads throughout the distal half of the testis (Fig. 1A and B). In the control testes, nuclei of post-meiotic spermatids were detectable as clusters from the beginning of the early elongation stage and seen as multiple bundles at the distal end of the testis (Fig. 1A-C). In Sax RNAi testes, post-meiotic nuclei were still relatively clustered in the round stage of spermatids (Fig. 1D). However, they were found completely scattered in leaf ~ canoe stages (Fig. 1B-D). Although the spermatids were able to continue maturing while unbundled, they were not able to fully complete spermiogenesis, appearing to halt at the canoe stage, and no needle stage nuclei were detected (Fig. 1C and D).

Fig. 1. Sax is required in the germline for proper spermiogenesis.

A, B) Representative images of testis tip without (A) or with (B) expression of shRNA against Sax under the bamGal4 driver. C, D) Representative images of post-meiotic nuclei of spermatids in indicated stages without (C) or with (D) Sax RNAi under the bamGal4 driver. F, G) Representative images of ProtamineB-GFP incorporation in sperm nuclei in the testis without (F) or with (G) Sax RNAi under the bamGal4 driver.

All flies are 0- to 3-days old. Scale bars in A, B are 100 μm. Other scale bars represent 10 μm. Asterisks indicate approximate location of the hub.

Since protamine incorporation occurs in the late canoe stage and is required to form mature spermatids, we used ProtamineB-GFP to determine the timing of spermiogenesis arrest in Sax RNAi spermatids. Assuming that half of all canoe spermatids in each testis are in the early canoe stage and half are in the late canoe stage, we expected that approximately 50% of all canoe spermatids should be ProtamineB-GFP positive. Consistent with this assumption, we found that approximately 51 ± 7% (n = 10 testes) of bundled canoe spermatids in ProtamineB-GFP flies were GFP positive, while 7 ± 7% (n = 10 testes) of scattered canoe spermatids in bamGal4>Sax RNAi, ProtamineB-GFP flies were GFP positive (Fig. 1F and G), indicating that Sax RNAi spermatids likely stop development during the early canoe stage and very rarely progress to the late canoe stage. Due to halted spermiogenesis, seminal vesicles were completely empty in ~95% of Sax RNAi testes (Figure S1A-B, n = 113) with the seminal vesicles of the remaining testes containing only a small number of sperm cells (Fig. S1B’). Consistently, Sax RNAi flies showed significantly reduced fertility (Fig. S1C).

2.2. Soma (CC)-derived Gbb signals to the germline to maintain proper spermiogenesis

To determine which BMP ligand (s) signal to Sax to maintain spermatid bundling, we first investigated Gbb as Sax preferentially binds Gbb, and Gbb is known to be expressed in CCs in the testis (Ridwan et al., 2024; Haerry et al., 1998). We knocked down Gbb in CCs under the c587Gal 4 driver combined with TubGal80^ts. After seven days of temperature shift at 29 °C, Gbb RNAi testes displayed the same sperm head scattering phenotype as seen in bamGal4>Sax RNAi flies, indicating that CC-derived Gbb is responsible for this phenotype (Fig. 2A-D).

Fig. 2. Soma (CC)-derived Gbb signals to the germline to maintain proper spermiogenesis.

A-D) Representative images of post-meiotic nuclei of spermatids in indicated stages without (A, C) or with (B, D) Gbb RNAi under the c587Gal 4 driver combined with TubGal80^ts. Testes were dissected and analyzed for DAPI staining after 7-day temperature shift (TS) at 29° . E) A schematic showing the cell types in the testis and expression pattern of different Gal4 drivers. Scale bars represent 10 μm.

In the testes, both Dpp and Gbb, likely as a heterodimer, are required for stem cell maintenance and prevention of dedifferentiation (Shivdasani et al., 2003; Ridwan et al., 2024; Bauer et al., 2023). However, RNAi of Dpp through either the same c587Gal4^ts driver (n = 46) FasIIIGal4^ts (n = 20), or through DppGal4^ts (n = 20), did not lead to sperm-head scattering, indicating that Dpp is not involved in this pathway, and that BMP signaling from CCs to developing germline cells may be solely through Gbb (Table 1).

Table 1. Summary of sperm-head scattering phenotypes observed in indicated genotypes.

0-to-7 day old flies were used for all experiments except when temperature shift was required. 7- to 10-day old flies were used for temperature shift data. Results come from three independent biological replicates for all genotypes. Normal (bundled) sperm heads: (−), Partial scattering: (+), Complete scattering: (++), Not examined: (NA).

	Driver	bamGal4 on 3		bamGal4 on X	nosGal4	c587Gal4ts	no driver
	BDSC#		qPCR
gbb RNAi	34898	NA		NA	NA	++ (n = 47)	NA	Ridwan et al. (2024)
gbb4/gbb4	98345	NA		NA	NA	NA	+ (n = 24)	Khalsa et al. (1998)
dpp RNAi	33618	NA		NA	NA	− (n = 46)	NA	Ayyaz et al. (2015)
sax RNAi	42546	++ (n = 200)	20.40%	++ (n = 37)	+ (n = 57)	− (n = 59)	NA	Ridwan et al. (2024)
sax RNAi	36131	− (n = 30)	56.70%	+ (n = 25)	NA	NA	NA	Redhai et al. (2016); Chng et al. (2014)
sax RNAi	57319	− (n = 20)	68.80%	− (n = 39)	NA	NA	NA	Ayyaz et al. (2015)
tkv RNAi	40937	− (n = 46)		− (n = 42)	NA	NA	NA	Ridwan et al. (2024)
punt RNAi	39025	− (n = 45)		− (n = 18)	NA	NA	NA	Ridwan et al. (2024)
wit RNAi	41906	− (n = 28)		− (n = 22)	NA	NA	NA	Chng et al. (2014)
mad RNAi	31316	− (n = 39)		+ (n = 27)	NA	NA	NA	Ridwan et al. (2024)
med RNAi	43961	− (n = 35)		+ (n = 30)	NA	NA	NA	Ridwan et al. (2024)

A previous study has demonstrated that BMP signal from the niche oppositely regulates Bam expression levels between GSCs and SGs, and that Sax upregulates Bam expression levels in SGs (Ridwan et al., 2024). Moreover, upregulation of Bam is known to regulate dedifferentiation (Ridwan et al., 2024), as well as the number of SG divisions (Insco et al., 2009). Although we did not observe severe defects in GSC number and niche appearance (Figs. S2A and B), we detected excess SG division (cysts containing more than 16-cells) in about ~10% of bam > Sax RNAi testes (n = 22, Figs. S2C and D). To further determine the cell types responsible for the observed phenotype during spermiogenesis, we analyzed additional tissue-specific knockdown of Gbb and Sax.

Importantly, when knockdown of Sax was driven by nosGal4, an early germline driver including GSCs, we observed only a subtle scattering phenotype, excluding the possibility that this phenotype is caused secondarily by a niche-GSC signaling defect (Table 1). Similar to bamGal4>SaxRNAi, we observed complete scattering of spermatids using the SG.18.1Gal 4 driver (n = 20), which is known to be expressed in SCs (Desai et al., 2009). Because Sax is also reported to function in CCs (Li et al., 2007), we examined c587Gal4 driven knockdown of Sax. These testes showed no scattering (Table 1), indicating that the role of Sax in differentiating germ cells to maintain spermatid bundling is independent of its role in CCs for SG proliferation restriction.

To identify other signaling components, we examined known BMP pathway core components for spermatid scattering. RNAi of the transcription factor Mad, as well as its cofactor Med, showed partial sperm head scattering, suggesting their involvement in this process. However, we could not successfully determine any type II receptors.

2.3. BMP signaling controls morphology of mitochondria

Spermiogenesis involves extreme morphological changes to the spermatids to optimize the cells for fertilization. Spermiogenesis is coupled with the dynamic rearrangement of mitochondria. After the completion of meiosis, mitochondria fuse together, and form two long mitochondrial derivatives. These mitochondrial derivates wrap around each other to form the nebenkern (Fabian et al., 2012).

Our Transmission Electron Microscopy (TEM) analyses showed clear morphological differences between control and Sax RNAi mitochondria. In SCs, the Sax RNAi mitochondria were larger with an average diameter at their widest point of 0.94 ± 0.20 μm (n = 57) compared to 0.61 ± 0.19 μm (n = 42) in the wildtype (p < 0.0001, Fig. 3A and B). The Sax RNAi mitochondria were also much less electron dense than wild type mitochondria (Fig. 3C). Control mitochondria had tightly aligned inner and outer membranes, as well as numerous cristae (Fig. 3A’). Sax RNAi mitochondria often had either a missing or broken outer mitochondrial membrane and very few cristae (Fig. 3B’). When we expressed the mitochondrial redox sensor, roGFP, in sax RNAi germ cells, the localization of the sensor was dramatically affected (Fig. 3D and E). No notable change in mitochondrial oxidation level (i.e., roGFP 405/488 wavelength ratio) was observed. Although the Sax RNAi mitochondria were able to aggregate and form the nebenkern, this structure was still less electron dense than in the control, and the membranes often appeared tangled together rather than concentric (Fig. 3D and E). In the SG stage, some Sax RNAi mitochondria showed normal morphology and electron density, while others displayed the same low density as in SCs, suggesting that the defect begins in SGs, and that the mutant phenotype is unlikely caused by a technical artifact (Fig. 3F and G). The onset of mitochondrial defects in Sax RNAi SGs is consistent with our idea that these defects are not caused by previously characterized niche (Hub)-GSC BMP signaling, but instead CC to SG signaling.

Fig. 3. BMP signaling controls the morphology of mitochondria.

A, B) Representative electron micrographs of spermatocytes in the testes from 3- to 7-day old males of indicated genotypes. A′ and B′ are magnified areas in the cytoplasm showing mitochondria and graphical interpretations of their appearance. Scale bars represent 2 μm. C) Quantification of electron densities of mitochondria. Density ratios (Mitochondria/Cytoplasm) were measured from the indicated number of mitochondria from more than five spermatocytes from each samples. The p-value was calculated by student-t-test. D-E) Representative images of spermatocytes in live testes expressing mitochondrial sensor, Tub-mito-roGFP2-Grx1, show lower intensity in mitochondria in bam > Sax RNAi spermatids. The lower left panels show the magnification of a cell from the same images. Scale bars represent 10 μm. F-I) Representative electron micrographs of indicated cell stages in the testes from 3- to 7-day old males of indicated genotypes. Scale bars represent 2 μm. In I, black arrowheads indicate normal mitochondria and purple arrowheads indicate large mitochondria with abnormal electric density found in Sax RNAi SGs. Similar results were obtained by 2 more independent biological replicates. Scale bars represent 2 μm

2.4. Mitochondria are uncoupled from axonemes and nuclei in sax RNAi flies

In the early elongation stage, the nebenkern separates into the major and minor mitochondrial derivatives, which pair with the axoneme during flagellar elongation (Fig. 4A). In Sax RNAi spermatids, we observed either complete loss of mitochondrial derivatives or failure of the mitochondrial derivatives to pair with the axoneme (Fig. 4B). In the late-elongation stages, individualized sperm were found in Sax RNAi, but they often contained more than two axonemes, and coupled mitochondrial derivatives were rarely found (Fig. 4C and D).

Fig. 4. Mitochondria are uncoupled from axonemes and nuclei in Sax RNAi flies.

A-D) Representative electron micrographs of “early-elongation” or “late-elongation” stages of spermatids in the testes from 3- to 7-day old males of indicated genotypes. Similar results were obtained by 2 more independent biological replicates. Scale bars represent 2 μm. In A and C, cyan and red arrowheads indicate minor and major mitochondria derivatives respectively. Purple arrowheads in B and D indicate abnormal mitochondria found in the bundles of Sax RNAi spermatids. Ax: Axoneme. E-F) Representative images of live testes from 3- to 7-day old males expressing nuclear marker, His2Av-mRFP1, together with mitochondrial marker, Tubmito-roGFP2-Grx1, which show uncoupled nuclei and elongating mitochondria in bam > Sax RNAi spermatids. Scale bars represent 10 μm. G) Model. CC-derived Gbb signals to developing germline cells to maintain proper spermiogenesis,likely through regulation of mitochondria.

What is the relationship between sperm head scattering and mitochondrial defects? In Sax RNAi testes, elongated mitochondria can still be found, but they are often uncoupled from nuclei and disorganized (Fig. 4E and F). Previously, several sterile mutations with similar defects in mitochondrial morphology have been reported. We examined two mutant lines, Emmenthal (Emm) (Dorogova et al., 2013) and fuzzy onions (fzo) (Hales et al., 1997), and found that these both show similar sperm-head scattering phenotypes (Fig. S3). These results suggest a possible link between mitochondrial defects and sperm head scattering.

Taken together, we propose that soma-derived Gbb signals to Sax on SGs and SCs to ensure mitochondrial morphology and function. We suggest the possibility that proper rearrangement of the mitochondria is essential in maintaining bundling of spermatids during spermiogenesis.

3. Discussion

Soma-germline interaction is known to regulate spermiogenesis in a broad range of organisms. In this study, we show that the BMP signaling pathway, a highly conserved pathway across species, is an essential component of soma-germline interaction during spermiogenesis in Drosophila. The ligand, Gbb, secreted from CCs, signals to differentiating germ cells through the receptor Sax. Knockdown of Gbb in CCs or the receptor Sax in germ cells leads to a defect in sperm-head bundling and decreased fertility. TEM analyses of mutant spermatocytes/spermatids revealed a morphological defect in mitochondria. Elongating spermatids showed uncoupled nuclei and elongating mitochondrial derivatives, suggesting that improper mitochondrial development may cause the sperm bundling defect. With these results, we propose a new role of BMP signaling in Drosophila spermiogenesis, distinct from previously characterized roles of BMP signaling in the niche.

Interestingly, a previous study has suggested a possible link between BMP signal and mitochondria function. Kumar et al. identified the mitofusin 2 (Mfn2) as an interactor of Smad2 and showed that TGF-β signal regulates mitochondrial dynamics and metabolic functions through a transcription-independent role of Smad2 (Kumar et al., 2016). We found a similar phenotype in the fzo mutant, which is the fly homolog of Mfn2 that mediates mitochondrial fusion (Hales et al., 1997). It would be an interesting future study to test direct interaction between Mad and Fzo in fly spermiogenesis.

4. Methods

4.1. Fly husbandry and strains

Flies were raised on standard Bloomington medium (Lab express) at 25 °C (unless temperature control was required) and young flies (0- to 7-day-old adults) were used for all experiments. The following fly stocks were obtained from Bloomington Drosophila Stock Center (BDSC); nosGal4 (BDSC64277); tkv RNAi: TRiP.HMS02185 (BDSC40937); medea RNAi:TRiP.GL01313 (BDSC43961); mad RNAi:TRiP.JF01264 (BDSC31316); sax RNAi:TRiP.HMJ02118 (BDSC42546); sax RNAi TRiP. HMS04520 (BDSC57319): saxRNAi TRiP.JF03431 (BDSC36131); punt RNAi: TRiP.HMS01944 (BDSC39025); gbb RNAi:TRiP.HMS01243 (BDSC34898); gbb⁴ (BDSC98345); dpp RNAi:TRiP.HMS00011 (BDSC 33618); Smox RNAi: TRiP.JF02320 (BDSC26756). fzo (Wang et al., 2014)/TM3 (BDSC80071); emm 1/CyO (BDSC11769): ProtamineB-eGFP/CyO (BDSC11769);His2Av-mRFP1 (BDSC23651). Tub-mito-roGFP2-Grx1 (BDSC67669) (Albrecht et al., 2011) was used for visualizing mitochondria and nebenkern. dppGal4 (Matsuda et al., 2021) lines are described elsewhere. FasIIIGal4 was obtained from DGRC, Kyoto Stock Center (A04-1-1 DGRC#103-948). bamGal4 on 3rd was kind gift from Yukiko Yamashita. Temperature shift was performed by culturing flies at room temperature and shifted to 29 ° C upon eclosion for the 7 days before analysis. Combinations of Tub-Gal 80^ts with c587Gal 4 (a gift from Yukiko M. Yamashita) were used.

4.2. Immunofluorescence staining

Testes were dissected in phosphate-buffered saline (PBS) and fixed in 4% formaldehyde in PBS for 30–60 min. Next, testes were washed in PBST (PBS + 0.2% TritonX-100, Thermo Fisher) for at least 60 min, followed by incubation with primary antibody in 3% bovine serum albumin (BSA) in PBST at 4 °C overnight. Samples were washed for 60 min (three times for 20 min each) in PBST, incubated with secondary antibody in 3% BSA in PBST at room temperature for 2 h and then washed for 60 min (three times for 20 min each) in PBST. Samples were then mounted using VECTASHIELD with 4’,6-diamidino-2-phenylindole (DAPI) (Vector Lab).

The primary antibodies used were as follows: rat anti-Vasa (1:20; Developmental Studies Hybridoma Bank, DSHB); mouse anti-Hts (1:20; DSHB); AlexaFluor-conjugated secondary antibodies (Abcam) were used at a dilution of 1:400. Images were taken using Zeiss LSM800 confocal microscope with airyscan module by using 1AU-pinhole with 63× oil immersion objective (NA = 1.4). Images were processed by image J/FIJI.

Primary-secondary antibody steps were skipped for DAPI only imaging.

4.3. Short-term live imaging

We used short term live imaging for static image acquisition to observe/quantify fluorescent-tagged proteins to avoid loss of fluorescent signal or tagged protein itself located in extracellular space by fixation and permeabilization.

Testes from newly eclosed flies were dissected into Schneider’s Drosophila medium containing 10% fetal bovine serum and glutamine–penicillin–streptomycin. These testes were placed onto Gold Seal Rite-On Micro Slides’ 2 etched rings with media, then covered with coverslips. Images were taken using a Zeiss LSM800 airyscan with a 63 × oil immersion objective (NA = 1.4), with 10–20 z-stacks (interval 1 μm). For all short-term live imaging experiments, imaging was performed within 30 min and no time-lapse imaging was performed using this method.

4.4. Fertility assay

Individual 0- to 3-day-old males were crossed with three 0- to 3-day-old control virgin females (yw) in a narrow vial at 25 °C. After 3 days, males were removed. Females were left to produce embryos for 5 days. Eclosed offspring were counted for 10 consecutive days.

4.5. Electron microscopy

Testes were dissected into phosphate buffered saline (PBS) and then fixed in a solution of 2.5% glutaraldehyde and 3% paraformaldehyde in 0.1 M sodium cacodylate buffer on ice for 30 min. Samples were then washed in cacodylate buffer containing 2 mM calcium chloride and incubated in a solution of 1.5% potassium ferrocyanide and 2% osmium tetroxide in cacodylate buffer, followed by washing with water and a subsequent incubation in 2% aqueous osmium tetroxide at room temperature. Samples were then washed with water and placed in 1% aqueous uranyl acetate overnight at 4 °C.

The next day, samples were then dehydrated via a graded series of alcohol dilutions, then washed with propylene oxide in epoxy resin and allowed to polymerize at 60 °C for 48 h.

Ultrathin sections (60 nm) of Lowicryl HM-20-embedded testes were cut on a UC-7 ultramicrotome (Leica Biosystems) with a diamond knife (Diatome, Hatfield, PA) and imaged using Hitachi H-7650 transmission electron microscope. 3 testes were examined from each genotype (Control and bam > Sax RNAi)

4.6. Quantitative RT-PCR to estimate sax RNAi efficiency

Testes from ~30 males progeny, age ~7 day, were collected and RNA was extracted RNeasy Micro Kit (Qiagen) following the manufacturer’s instructions. One microgram of total RNA was reverse transcribed to complementary DNA using SuperScript III First-Strand Synthesis Super Mix (Invitrogen) with Oligo (dT)20 Primer. The qPCR was performed, in duplicate, using SYBR green Applied Biosystems Gene Expression Master Mix on a CFX96 Real-Time PCR Detection System (Bio-Rad). Relative quantification was performed using the comparative threshold cycle method. Following primers were used:

sax 5′- CAAGACCCTTGGGAGCTTATA −3′and 5′-GCTTCTATGCGTGGGTATCTAA-3’.

αTub84 B (control): 5′-TCAGACCTCGAAATCGTAGC-3′ and 5′-AGCAGTAGAGCTCCCAGCAG-3’.

4.7. Statistical analysis and graphing

No statistical methods were used to predetermine sample size. The experiments were not randomized. The investigators were not blinded to allocation during experiments and outcome assessment. Statistical analysis and graphing were performed using GraphPad Prism 10 software. Data are means and standard deviations. The p-values (two-tailed Student’s t-test) are provided.

BioRender was used for the graphical abstract.

Data availability

Data will be made available on request.