TSSKL is needed for sperm energy structure development and male fertility in moths

Given the testis's crucial role in sperm morphogenesis and sperm maturation [27], we compared the transcriptomes of the testes, ovaries, whole body of 2 days old virgin adult excluding the testis or ovary (Fig 1A). Comparative analysis revealed 1293 uniquely expressed genes in testes compared to other three organs (Fig 1B). Remarkably, testicle-specific serine/threonine protein kinase-like (TSSKL, accession number: PX116274) was significantly enriched in testes (S1 Table). We additionally performed qRT-PCR analysis of 10 different tissues of B. mori adults and found that BmTSSKL was predominantly expressed in the testis than other tissues analyzed (i.e., head, epidermis, fat body, malpighian tubules, midgut, middle silk gland, posterior silk gland, anterior silk gland, and ovary) (Fig 1C). Furthermore, we quantified BmTSSKL expression pattern in the testis from the fifth-instar larvae stage through the adult stage and observed that it gradually increased during development, reaching a higher level at the virgin adult stage (Fig 1D). These results suggested that the BmTSSKL played a potential role in the regulation of sperm morphogenesis.

Based on the B. mori genome references [28], we found that BmTSSKL is located on chromosome 25 and only has one exon (Fig 1E). To explore the physiological function of BmTSSKL, we used the binary transgenic CRISPR/Cas9 system to engineer loss-of-function mutants. To do this, we created a transgenic silkworm line expressing the two sgRNA targeting BmTSSKL driven by the silkworm U6 promoter (Fig 1F and 1G). Meanwhile, the insertion of the U6-sgRNAs and Nos-Cas9 transgenes was validated by inverse-PCR performed on genomic DNA isolated from the whole bodies of fluorescent G1 moths (DsRed or GFP positive) (Fig 1G). Analysis of the resulting sequences against the silkworm genome database (http://kaikoblast.dna.affrc.go.jp↗) thereby localized the transgenes to distinct chromosomal loci (S1 Fig). Subsequently, the U6-sgRNA (TSSKL) and Nos-Cas9 lines were crossed to generate the TSSKL mutant line (Figs 1G and S2). Genomic DNA was then extracted from randomly selected dual fluorescent (dsRed and GFP) individuals, and the target regions were PCR-amplified and sequenced. Sequencing analysis revealed the occurrence of various large deletions in the BmTSSKL transgenic line (Fig 1H). These results demonstrated that we had successfully obtained the BmTSSKL deficiency mutants. As predicted, the BmTSSKL mutants had viable and normal surface phenotypes compared with WT during all development stages (S3 Fig).

Next, to explore whether the disruption of BmTSSKL affects the fertility of B. mori, we performed fecundity assays. BmTSSKL mutant males mated with either WT or BmTSSKL mutant females produced normal egg yields, but these eggs failed to hatch (0%). Whereas WT males mated with either WT or BmTSSKL mutant females produced normal egg yields with successful hatching (94–96%) (Fig 2A). We also counted the number of eggs in broods produced by different cross combinations. The controls produced an average of ~370 eggs (WT: 370, U6-sgRNAs: 370, Nos-Cas9: 369); WT males mated with BmTSSKL mutant females produced average 368 eggs; BmTSSKL mutant males mated with WT females produced average 348 eggs; BmTSSKL mutant males mated with BmTSSKL mutant females produced average 339 eggs (Fig 2B). After free hybridization of Nos-Cas9 and U6-sgRNAs, four types of offspring were produced, including dual fluorescent BmTSSKL mutant line, Nos-Cas9 line (GFP), U6-sgRNAs line (DsRed), and non-fluorescent line (S2 Fig). Subsequently, random hybridization among these four lines only yields offspring of these same four types. Female mutants carrying the male sterility factor could mate non-mutant adult individuals and effectively reproduced the next generation. In this next generation, male mutants could mate but were sterile, while female mutants could both mate and remain fertile. In every generation of silkworms, BmTSSKL mutant male sterility was observed and inherited (Fig 2C). Meanwhile, we performed immunofluorescence analysis on the testes of virgin adults and found that there was almost no TSSKL protein in the mutants (Fig 2D). We furthermore evaluated the mating behaviors of BmTSSKL mutant moths. Behavior was considered normal if females and male moths could attract each other and copulate for at least 30 min [29]. The response index was determined as the percentage of successful mating relative to the total trials in a group (Fig 2E and 2F). There were no significant differences in female attractiveness between BmTSSKL mutant and control individuals (WT female: 52.07%, mutant female: 47.93%, P = 0.262; U6-sgRNAs female: 50.83%, mutant female: 49.17%, P = 0.448; Nos-Cas9 female: 51.00%, mutant female: 49.00%, P = 0.429) (Fig 2G). And there was no significant difference in male competitiveness between BmTSSKL mutant and control individuals (WT male: 52.05%, mutant male: 47.95%, P = 0.104; U6-sgRNAs male: 49.62%, mutant male: 50.38%, P = 0.532; Nos-Cas9 male: 50.45%, mutant male: 49.55%, P = 0.771) (Fig 2H). In addition, we amplified and sequenced the BmTSSKL gene from the testis of injected males with abnormal fertility and the ovary of injected females with normal fertility. Heterozygous chromatograms and various deletions or insertions were observed in the BmTSSKL mutant individuals of both sexes (S4 Fig). Taken together, these results demonstrate that BmTSSKL is essential for male fertility, but does not affect mating behavior.

We performed further analyses to assess the cause of BmTSSKL mutant male sterility. We first investigated whether the BmTSSKL mutant males exhibited any gross defects in external genitalia or internal genitalia system. However, we confirmed that no significant defects were detected. In lepidopteran insects, the bursa copulatrix consists of proteins from the seminal fluid that are transferred from male moths during copulation [30]. We further explored whether the BmTSSKL mutant influences bursa copulatrix formation. For this purpose, we evaluated the bursa copulatrix 1 h post copulation (hpc) in WT females mated with WT or BmTSSKL mutant males. To our surprise, the results showed that the bursa copulatrix was full in WT females mated with WT or BmTSSKL males (Fig 3A). We then investigated whether BmTSSKL mutant affects sperm transfer into spermatheca. We found that at 3 hpc the spermathecae of females mated with WT males contained sperm, whereas the spermathecae of WT females mated with BmTSSKL mutant males were empty, like those of virgin females (Fig 3B). To elucidate BmTSSKL is involved in sperm migration in the bursa copulatrix, we compared mRNA expression levels of sperm motility-related genes (BmMlc [31], BmMlc2 [32], BmTry [33], and BmFln [34]) at different developmental stages between wild-type and BmTSSKL mutant individuals using qRT-PCR analysis. Notably, the mRNA levels of BmMlc, BmMlc2, BmTry and BmFln were significantly lower in BmTSSKL mutant than those of WT (Fig 3C).

In addition, to gain insights into whether BmTSSKL is involved in sperm morphogenesis, we performed fluorescence staining to examine the morphology of eupyrene and apyrene sperm bundles in WT and BmTSSKL mutant newly emerged male moths. The results showed that the sperm nuclei were assembled regularly at the head of the eupyrene sperm bundles in the WT, whereas in BmTSSKL mutants the eupyrene sperm bundles nuclei were abnormal positioning and exhibited abnormally squeezed eupyrene sperm bundles (Fig 3D). The apyrene sperm of WT had no polarity, and micronuclei were distributed in the middle region of bundles. In contrast, the apyrene sperm bundles in BmTSSKL mutants had defects in sperm nucleus shape and localization (Fig 3E). Thus, we deduced that a defect in BmTSSKL could influence both apyrene and eupyrene sperm morphogenesis.

To explore further how BmTSSKL deficiency hinders sperm morphogenesis and sperm migration, we used transmission electron microscopy (TEM) to examine ultrastructural changes (mitochondria and microtubules) of apyrene and eupyrene sperm in the testes. We observed that the transverse sections of eupyrene sperm in WT showed that the elliptical mitochondrial derivatives were approximately equal in size and regularly arranged by side below the centrosome. Conversely, the mutant's mitochondria were not closely juxtaposed, and the sperm cell membrane ruptured, resulting in the extracellular dissociation of microtubule structure centrosome (Fig 4A). Importantly, we detected the elliptical mitochondrial derivatives of BmTSSKL mutants were markedly atrophied and disappeared, and surrounded by membrane-limited autophagosome vesicles (Fig 4A, BmTSSKL-1 and BmTSSKL-2). The degraded vesicular material and mitochondrial remnants around disintegrated sperm structures were engulfed by autophagosomes (Fig 4A, BmTSSKL-3). In contrast, no autophagosomes were observed in eupyrene sperm WT testes. Moreover, we found the transverse sections of apyrene sperm showed that a pair of mitochondria of apyrene sperm present "V" shaped in WT, whereas, there were one, three or four mitochondrial derivatives in BmTSSKL mutant (Fig 4B). Collectively, TEM data indicated that mutation of BmTSSKL induced the production of defective mitochondrial derivatives and the activation of autophagy in dimorphic sperms.

To gain insights into changes in global gene expression associated with BmTSSKL deficiency, we performed RNA-Seq on sperm bundles isolated from testes of WT and BmTSSKL mutant at 2 days old virgin male adults. In a comparison between sperm in testes of BmTSSKL and WT, we identified 224 significantly upregulated and 59 downregulated genes (S2 Table). Remarkably, Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis showed that the differentially-expressed genes (DEGs) were mainly in amino acid metabolism associated with synthesis and release of ATP (Fig 4C). We further verified the expression changes of genes involved in mitochondrial metabolism by qRT-PCR. The mRNA levels of BmDlp1, BmFis1, BmMfpp1, BmGdap1, BmE3, BmMch, BmMib1, and BmSlp2 were significantly different in BmTSSKL mutants than those of WT (Fig 4D). The deletion of BmTSSKL gene led to the disturbance of sperm mitochondria fusion and fission, that is, the synthesis and release of ATP. This further affects genes related to sperm morphogenesis and motility, leading to male sterility.

In order to further verify the evolutionary conservation of TSSKL in Lepidoptera, the B. mori TSSKL gene was used as a query to identify the putative coding sequence of PxTSSKL from the Plutella xylostella genome (S5 Fig). As expected, we observed that PxTSSKL is expressed at a higher level in the testis than other tissues of adult moths, and the PxTSSKL expression levels are higher in the testes of the 3-day-old pupal and virgin adult moths compared with other life stages (Fig 5A and 5B). Subsequently, we generated PxTSSKL mutants through CRISPR/Cas9 system in P. xylostella (Fig 5C and 5D). We found that PxTSSKL mutations don't affect the mating ability and mating behavior of females and males (S6 Fig). There was no significant difference in egg number among different cross combinations (Fig 5E and 5F). Nevertheless, we also obtained a mutant phenotype consistent with the results in silkworms where males are sterile and females are fertile (Fig 5E and 5G), suggesting that the PxTSSKL mutations cause male sterility in P. xylostella. Similarly, the deletion of PxTSSKL gene function resulted in abnormal sperm development in P. xylostella. Compared to the WT, the nucleus of eupyrene sperm could not be arranged in a needle pattern and became loose in mutant individuals compared to the WT (Fig 6A). The nucleus of apyrene sperm was excreted in WT individuals when it matured, while in PxTSSKL mutants, a large number of the nuclei were retained in apyrene sperm (Fig 6B). Meanwhile, we also observed bursa copulatrix was full in WT females mated with WT or PxTSSKL males (Fig 6C). Whereas, the spermathecae of WT females mated with PxTSSKL mutant males were empty (Fig 6D). Taken together, the resulted phenotypes in PxTSSKL mutants support the role of TSSKL in controlling sperm morphogenesis and motility in additional lepidopteran species.

Silkworm, Nistari, a multivoltine and non-diapausing strain, was used for all experiments. Larvae were reared on fresh mulberry leaves under standard conditions [51,52]. A laboratory strain of diamondback moth, P. xylostella, was obtained from Min-Sheng You (Fujian Agriculture and Forestry University, Fuzhou, China). Larvae were reared in an incubator at 26 ± 1°C, 60 ± 5% relative humidity, a photoperiod of 16:8 h (light: dark) and fed with an artificial diet as described previously [53]. Adult moths were fed with a 10% honey solution.

Total RNAs of testis and other tissues were isolated from several tissues using TRIzol reagent (Invitrogen, USA). 1 μg of total RNA was used with the First Strand cDNA Synthesis Kit for complementary DNA (cDNA) synthesis (Promega, USA). Quantitative real-time PCR (qRT-PCR) analyses were performed using a SYBR Green Realtime PCR Master Mix (TOYOBO, Japan). The PCR conditions were as follows: initial incubation at 95°C for 5 min, 35 cycles at 95°C for 15 s, and 60°C for 1 min. The B. mori gene encoding ribosomal protein 49 (Bmrp49) and P. xylostella Pxactin were used as internal control. A relative quantitative method (^△△Ct) was used to evaluate quantitative variation. The gene-specific primers used for qRT-PCR are listed in S3 Table.

According to the construction principle of the CRISPR/Cas9 system, the open reading frames (ORFs) of BmTSSKL were screened to identify sgRNA target sites using the GGN19GG rule [54 –56]. Two 23-base-pair (bp) sgRNAs that target sites in the exon of BmTSSKL were designed. The activator was the plasmid pBac[IE1-GFP-Nos-Cas9] (Nos-Cas9) with the enhanced green fluorescent protein (GFP) marker, which results in Cas9 expression that is driven by the Nos promoter [57]; The effector plasmid was pBac[IE1-DsRed2-U6-sgRNAs] (U6-sgRNAs) with the discosoma red fluorescent protein (dsRed) fluorescent marker, which results in expression of the sgRNAs under control of the silkworm U6 promoter. Both GFP and dsRed fluorescence markers were controlled by the IE1 promoter. The primers used for plasmid construction are listed in S3 Table.

To generate the TSSKL mutant, we carried out silkworm germline transformation by microinjecting Nos-Cas9 and U6-sgRNA respectively into the pre-blastoderm Nistari embryos along with a mixture of piggyBac helper vectors. Then, embryos (generation 0: G0) were incubated in a humidified chamber at 25°C until hatching. The resulting larvae were reared to adulthood and backcrossed with WT moths to produce progeny. Through fluorescence microscopy screening, we selected G1 individuals displaying single fluorescent markers (GFP or DsRed) and intercrossed them to obtain G2 progeny. This crossing strategy yielded four distinct germlines: (1) dual fluorescent mutants (GFP/DsRed) expressing both Cas9 and sgRNA, (2) Nos-Cas9 line (GFP/+), (3) U6-sgRNA (DsRed/+), and (4) non-transgenic controls (fluorescence-negative) (S2 Fig). Meanwhile. the double-positive individuals (GFP/DsRed) were identified as putative mutants and selected for subsequent genotypic validation.

Inverse PCR was used to determine the genomic insertion sites of the transgene as described previously. In brief, Genomic DNA was extracted from G1 transgenic moths using the CTAB method. Subsequently, 5 μg of DNA was digested with DpnII, purified and self-circularized with T4 DNA ligase. Flanking genomic regions were amplified by nested PCR using outward-facing primers specific to the 5' transposon terminal repeat (). The resulting PCR products were sequenced to identify the insertion sites. S3 Table

Two 23-bp sgRNA target sites were selected in exon 1 of PxTSSKL following GGN19GG principle. The sgRNAs were synthesized with a MegaScript T7 Kit (Thermo Fisher Scientific, USA). The primers used for plasmid construction are listed in S3 Table. A mixture of sgRNA (300ng/μL) and Cas9 proteins (300ng/μL) in 10 μL was injected into embryos within 2 hours using microinjectors. Subsequentlye, eggs were incubated in at 25 ± 1°C and 65 ± 5% relative humidity until hatching, Individuals with PxTSSKL deletion were used in subsequent experiments.

Genomic DNA of individuals was extracted to identify deleted regions. First instar larvae were incubated in DNA extraction buffer (1:1:2:2.5 ratio of 10% SDS to 5 M NaCl to 100 mM EDTA to 500 mM Tris-HCl, pH = 8) with proteinase K (Thermo Fisher Scientific, USA), then purified with a standard phenol: chloroform extraction and isopropanol precipitation, followed by RNaseA (Takara, China) treatment. The genomic PCR conditions were as follows: 94°C for 2 min, 35 cycles of 94°C for 15 s, 55°C for 30 s, and 72°C for 1 min, followed by a final extension period at 72°C for 10 min. The PCR products were sub-cloned into pJET-1.2 vectors, and vectors were sequenced. The primers used for sequencing are listed in. S3 Table

Silkworm female attraction and male competitiveness assay were performed in a box (30 × 18 × 4.5 cm³). Newly emerged control and BmTSSKL mutant female moths were placed on either side of box, and WT male moth was placed in the center. Males reaching one side and mating with females were considered responsive, and the number of responding females was recorded. The response index was calculated as the number of reaction moths divided by the total number of test moths multiplied by 100. Using the same method of female attraction, statistical analysis of male competitiveness was carried out. The cross between Nos-Cas9 line and U6-sgRNA line produced four lines, including BmTSSKL mutant line with double-fluorescent, Nos-Cas9 line with green fluorescence, the U6-sgRNA line with red fluorescence, and a nonmutant line without fluorescence. Random hybridization of these four lines produced only these four types of individuals.

Wild-type virgin females were allowed to mate with wild-type and PxTSSKL mutant males into a mesh cage (25 × 25 × 25 cm³). Similarly, wild-type virgin males were allowed to mate with wild-type and PxTSSKL mutant females. All females were allowed to lay eggs for 24 h. The number of eggs laid by different groups of females and the hatching rate were analyzed. Each pair of mating males and females was placed separately in a clear tube to observe mating behavior and calculate offspring hatching rate. Each group had five pairs of moths, and the experiments were performed three times.

Testes from individuals at the virgin adult stage were dissected and fixed in 4% paraformaldehyde overnight. They were dehydrated with different concentrations of ethanol and then embedded in paraffin. Samples were cut into ultrathin sections (4 μm), and dewaxed with dewaxing agent and ethanol. Subsequently, samples were immersed in a citric acid buffer for antigen repair. The rabbit anti-TSSKL primary antibody was used for TSSKL detection (1:500 dilution). Horseradish peroxidase-conjugated anti-rabbit IgG (1:1000 dilution) was used as a secondary antibody. The cell nucleus was re-stained with DAPI staining solution, and the film was sealed with anti-fluorescence attenuator. Image scanning through the scanner, and then observed by SlideViewer software.

Testes from virgin male adults were dissected on an anatomical plate and placed in phosphate buffered saline (PBS). Each testis was torn in a dye fixing solution and fixed for 10 min, then washed three times with PBS, 5 min/ time. Phalloidin was used to stain actin in a 30 min incubation followed by three washes with PBS for 5 min each. Subsequently, nuclei were stained with Hoechst dye for 15 min, followed by washing 3 times with PBS for 5 min each, followed by transfer of the tissue in a droplet to a slide and covering with a cover glass. Images were captured using Olympus BX51 fluorescence microscope (Olympus, Japan).

Testes of virgin male adults were dissected and fixed with 2.5% glutaraldehyde in 0.1 M PBS overnight at 4°C. Samples were washed with PBS and fixed in the phosphate buffer with 1% osmium tetroxide for 2h. They were dehydrated with different concentrations of ethanol and then embedded in Embed812 resin. Samples were cut into ultrathin sections (60 nm), stained with 2% uranyl acetate (pH = 5) and then stained with 10 mM lead citrate (pH = 12). The stained sections were observed by Hitachi H-7650 transmission electron microscopy (Hitachi, Japan).

Testes, ovary, and whole body of excluding the testis or ovary were dissected from 2 days old virgin adults. Then total RNA was extracted by using the RNAiso plus (Takara, China) following the manufacturer's protocol. Ten individuals were used for each sample, and three independent replicates were performed. Subsequently, total RNA was first enriched and then fragmented for cDNA synthesis and library construction. The DNBSEQ-T7 platform was used to sequence the library by BGI (https://www.genomics.cn/↗), and Fast QC was used to qualify the raw data, then the data were filtered by Trimmomatic. The filtered data were mapped and quantified to the reference B. mori database (https://www.ncbi.nlm.nih.gov/datasets/genome/GCF_030269925.1/↗). In addition, differentially expressed genes (DEGs) between WT and BmTSSKL mutant were normalized and analyzed by the DEGSeq R package (fold change >1 and FDR < 0.05 were used as a cutoff) [58]. Enrichment analyses of DEGs were conducted using the gene ontology (GO) analysis with all the genes which were expressed in our study as a background set [59]. All sequencing data have been deposited in GenBank under an accession code PRJNA1199782.

All the samples have at least three independent replications in this study. Statistical analysis was performed using SPSS 23.0 software with an independent Student's t-test. The data were presented as means ± S.E.M, and statistically significant differences were represented by asterisks as *P < 0.05, **P < 0.01, ***P < 0.001.