piRNA loading triggers MIWI translocation from the intermitochondrial cement to chromatoid body during mouse spermatogenesis

Abstract

The intermitochondrial cement (IMC) and chromatoid body (CB) are posited as central sites for piRNA activity in mice, with MIWI initially assembling in the IMC for piRNA processing before translocating to the CB for functional deployment. The regulatory mechanism underpinning MIWI translocation, however, has remained elusive. We unveil that piRNA loading is the trigger for MIWI translocation from the IMC to CB. Mechanistically, piRNA loading facilitates MIWI release from the IMC by weakening its ties with the mitochondria-anchored TDRKH. This, in turn, enables arginine methylation of MIWI, augmenting its binding affinity for TDRD6 and ensuring its integration within the CB. Notably, loss of piRNA-loading ability causes MIWI entrapment in the IMC and its destabilization in male germ cells, leading to defective spermatogenesis and male infertility in mice. Collectively, our findings establish the critical role of piRNA loading in MIWI translocation during spermatogenesis, offering new insights into piRNA biology in mammals.

Fig. 1. piRNA loading-deficient mutations cause the failure of MIWI releasing from the IMC in late-pachytene and diplotene spermatocytes.

a A schematic model showing the domain composition of MIWI and trajectory of the 5′ and 3′ ends of piRNA anchored with MID and PAZ-domain, respectively. The Y569/K573 and Y346/Y347 conserved in PIWI proteins are required for the 5′ end or 3′ end piRNA loading capacity of MIWI. b RNA co-IP assay of MIWI-associated-piRNAs (top) in wild-type (lane 1), Miwi^YY/YY (lane 2), Miwi^YK/YK (lane 3), and Miwi^−/− testes (lane 4), with anti-MIWI IB as a loading reference (bottom). c, d Immunostaining of MIWI (c, red) and TDRKH (d, red) on testis sections from adult wildtype, Miwi^YY/YY, Miwi^YK/YK, and Miwi^−/− mice using regular microscopy. Left: representative staining images of indicated mouse testis sections, scale bar, 20 μm; right, the enlargement of yellow framed regions, scale bar, 5 μm. The developmental stages of spermatocytes and spermatids were distinguished according to γH2AX (green) and DAPI (greyscale) staining. White arrows indicated MIWI (c) or TDRKH (d) aggregations. PS pachytene spermatocytes, DS diplotene spermatocytes, RS round spermatids. e Co-immunostaining of MIWI (red) and TDRKH (green) on testis sections from adult wildtype, Miwi^YY/YY and Miwi^YK/YK mice using confocal microscopy, with nuclei counterstained by DAPI (greyscale). White arrowheads indicated colocalization sites; yellow and white open arrowheads respectively indicated the unique localization of MIWI and TDRKH at non-colocalization sites. Scale bar, 10 μm. The results shown are representative of three independent experiments. Source data are provided as a Source Data file.

Fig. 2. piRNA loading facilitates MIWI dissociation with TDRKH.

a, b Co-IP assay of the association of MIWI and TDRKH in mouse testes from 18 dpp wildtype, Miwi^YY/YY, and Miwi^YK/YK mice. Quantification of blot intensity of TDRKH protein in anti-MIWI pellets is shown in parentheses [the one from wildtype control mouse (lane 4) is set as 1.0 after normalization with MIWI blotting]. c RNA co-IP assay of MIWI-interacting piRNAs in anti-MIWI (lane 1) and anti-TDRKH IP pellets (lane 2) from adult wildtype mouse testes, with anti-MIWI IB as a loading reference (bottom). d Anti-MIWI^unloaded preferably pulled down piRNA-unloaded MIWI (left) and TDRKH (right) in adult mouse testicular lysate. Left, RNA co-IP assays using anti-MIWI^unloaded and control anti-MIWI antibodies in adult wildtype mouse testicular lysate, with anti-MIWI IB as loading references. Right, co-IP assay of the association of MIWI and TDRKH using anti-MIWI^unloaded (lane 3) and control anti-MIWI antibodies (lane 4) in adult wildtype mouse testicular lysate, with testicular lysate (lane 1) and IgG IP (lane 2) serving as positive and negative controls, respectively. Quantification of blot intensity of TDRKH is shown in parentheses [the one anti-MIWI IP (lane 4) is set as 1.0 after normalization with MIWI blotting]. e RNase A treatment enhanced the MIWI-TDRKH interaction in wild-type mouse testes. Quantification of blot intensity of TDRKH protein in anti-MIWI pellets is shown in parentheses [the one from RNase A-untreated (lane 4) is set as 1.0 after normalization with MIWI blotting]. f Transfection of piRNA attenuated the MIWI-TDRKH interaction in Flag-tagged MIWI-stable-expressed GC-2spd (ts) cells. OE overexpression. Quantification of blot intensity of TDRKH is shown in parentheses [the one with RNase A-untreated and piRNA-free condition in Flag-tagged MIWI-stable-expressed GC-2spd (ts) cell lysates (lane 2) is set as 1.0 after normalization with MIWI blotting]. g Schematic diagram showing that piRNA loading facilitates the dissociation of MIWI from TDRKH. The results shown are representative of three independent experiments. Quantification of western blot analysis are represented as mean ± SD. Source data are provided as a Source Data file.

Fig. 3. piRNA loading is required for MIWI interaction with TDRD6 and transportation to the CB during male germ cell differentiation.

a Co-immunostaining of TDRKH (red) and TDRD6 (green) on testis sections from adult wildtype mice using confocal microscopy, with nuclei counterstained by DAPI (greyscale). PS pachytene spermatocytes, DS diplotene spermatocytes, RS round spermatids. Scale bar, 10 μm. b Co-immunostaining of MIWI (red) and TDRKH (green) on testis sections from adult wildtype and Tdrd6^−/− mice using confocal microscopy, with nuclei counterstained by DAPI (greyscale). White arrowheads indicated colocalization sites; yellow and white open arrowheads respectively indicated the unique localization of MIWI and TDRKH at non-colocalization sites. Scale bar, 10 μm. c, d Co-IP assay of the association of MIWI with TDRD6 and TDRKH in mouse testes from adult wildtype (c and d, lanes 3 and 4), Miwi^YY/YY (c, lanes 5 and 6) and Miwi^YK/YK (d, lanes 5 and 6) mice. Anti-MIWI IP pellets (c and d, lanes 4 and 6) were immunoblotted by the indicated antibodies, with testicular lysate (c and d, lanes 1 and 2) and IgG IP (c and d, lanes 3 and 5) serving as positive and negative controls, respectively. Quantification of blot intensity of indicated proteins in anti-MIWI IP pellets is shown in parentheses [the one from wildtype control mouse (lane 4) is set as 1.0 after normalization with MIWI blotting]. e Co-immunostaining of MIWI (red) and TDRD6 (green) on testis sections from adult wildtype, Miwi^YY/YY and Miwi^YK/YK mice using confocal microscopy, with nuclei counterstained by DAPI (greyscale). White arrowheads indicated colocalization sites. Scale bar, 10 μm. The results shown are representative of three independent experiments. Quantification of western blot analysis are represented as mean ± SD. Source data are provided as a Source Data file.

Fig. 4. piRNA loading enhances MIWI arginine methylation to facilitate its interaction with TDRD6.

a RNase A treatment barely altered MIWI methylation and MIWI-TDRD6 interaction in the adult testicular lysate. Quantification is shown in parentheses [the one from RNase A-untreated (lane 2) is set as 1.0 after normalization with MIWI blotting]. b Co-IP assay of the association of MIWI (lane 2) or arginine methylation-deficient MIWI^R-K mutant (lane 3) with TDRD6 in co-transfected HEK293T cells. Quantification is shown in parentheses [the one with wildtype MIWI (lane 2) is set as 1.0 after normalization with MIWI blotting]. c Co-IP assay of the effect of methyltransferase inhibitor methylthioadenosine (MTA, Sigma, D5011) on the MIWI-TDRD6 interaction in co-transfected HEK293T cells. Quantification is shown in parentheses [the one with DMSO treatment (lane 1) is set as 1.0 after normalization with MIWI blotting]. d, e piRNA loading-deficient mutations impaired arginine methylation of MIWI in mouse testes. Quantification is shown in parentheses [the one from the wildtype control mouse (lane 2) is set as 1.0 after normalization with MIWI blotting]. f Anti-MIWI^unloaded antibody pulled down less methylated MIWI and TDRD6 in adult wildtype mouse testicular lysate (lane 3) compared with control anti-MIWI antibody (lane 4). Quantification is shown in parentheses [the one anti-MIWI IP (lane 4) is set as 1.0 after normalization with MIWI blotting]. g Sequential co-IP showing that TDRKH is mainly associated with unmethylated MIWI. Quantification is shown in parentheses [the first anti-MIWI IP (lane 2) is set as 1.0 after normalization with MIWI blotting]. h TDRKH reduced MIWI methylation in co-transfected HEK293T cells. Quantification is shown in parentheses [the one without TDRKH transfection (lane 1) is set as 1.0 after normalization with MIWI blotting]. i Schematic diagram showing that piRNA loading promotes MIWI dissociation from TDRKH, leading to the exposure of the N-terminal of MIWI for arginine methylation by PRMT5 to enhance the MIWI-TDRD6 interaction. The results shown are representative of three independent experiments. Quantification of western blot analysis are represented as mean ± SD. Source data are provided as a Source Data file.

Fig. 5. piRNA loading-deficient mutations in Miwi lead to spermiogenic arrest in mice.

a All tested Miwi^YY/YY and Miwi^YK/YK males were infertile. b Testes from adult Miwi^YY/YY and Miwi^YK/YK mice were moderately reduced compared with wildtype control. Left, a representative image of testes from indicated mice; right, the average weight of testes from wildtype, Miwi^YY/YY (p = 0.033) and Miwi^YK/YK (p = 0.014) mice (n = 6, data are represented as mean ± SD, P values were calculated using two-tailed Student’s t-test, *p < 0.05). c PAS staining of the testis (top) and H&E staining of the epididymis (bottom) sections from adult wildtype, Miwi^YY/YY, Miwi^YK/YK, and Miwi^−/− mice. Scale bar, 30 μm. d Acrosome staining (ACRV1, red) of testis sections from wildtype, Miwi^YY/YY, Miwi^YK/YK, and Miwi^−/− mice using regular microscopy. Developmental stages of the seminiferous tubules were distinguished according to γH2AX (green) and DAPI (grayscale) staining. Scale bar, 10 μm. e TUNEL assays (red) of testis sections from adult wildtype, Miwi^YY/YY, Miwi^YK/YK, and Miwi^−/− mice using regular microscopy, with nuclei counterstained by DAPI (blue). Scale bar, 30 μm. Results shown in c–e are representative of three independent experiments. Source data are provided as a Source Data file.

Fig. 6. piRNA loading-deficient mutations in Miwi impair piRNA production and MIWI stability in mouse testes.

a Detection of piRNA expression in adult wildtype, Miwi^YY/YY, Miwi^YK/YK, and Miwi^−/− testes. b The length distribution of small RNAs from adult wildtype, Miwi^YY/YY, Miwi^YK/YK, and Miwi^−/− testes. Data were normalized by miRNA reads (21–23 nt). c Nucleotide distributions at the first position in the piRNAs from adult wildtype, Miwi^YY/YY, Miwi^YK/YK, and Miwi^−/− testes. d Genomic annotation of the piRNAs from adult wildtype, Miwi^YY/YY, Miwi^YK/YK, and Miwi^−/− testes. The percentage of mapped reads is shown. e Scatter plot of total piRNA reads mapped to individual piRNA clusters from adult wildtype, Miwi^YY/YY, Miwi^YK/YK, and Miwi^−/− testes. Data were normalized by miRNA reads (21–23 nt). f Western blotting of MIWI and MILI expression in testes from wildtype, Miwi^YY/YY, Miwi^YK/YK, and Miwi^−/− mice with indicated ages. β-actin served as a loading control. Quantification of blot intensity of MIWI is shown in parentheses (the one in wildtype testis is set as 1.0 after normalization with β-actin). Results shown in a and f are representative of three independent experiments, and small RNA-seq experiments shown in b–e with two replicates. Quantification of western blot analysis are represented as mean ± SD. Source data are provided as a Source Data file.

Fig. 7. Schematic model showing the panoramic view for the translocation of MIWI between germ granules during mouse spermatogenesis.

Upon its expression in mid-pachytene spermatocytes, MIWI protein is recruited to the IMC for piRNA processing via interacting with TDRKH through its unmethylated N-terminus, while piRNA loading induces a conformational change of MIWI and, in turn, weakens its interaction with TDRKH, leading to its release from the IMC. Meanwhile, the disassociation of MIWI with TDRKH simultaneously results in the arginine residues in its N-terminus exposed for methylation by PRMT5, thereby enhancing TDRD6 binding to prime its localization in the CB for piRNA function.