Identification of bi-allelic DNAH12 variants in infertile men with asthenoteratozoospermia

(A) Pedigrees of family 1 with 3 infertile males, P1 (IV:1), P2 (IV:2), and P3 (IV:4), family 2 with 3 infertile males, P4 (IV:3), P5 (IV:4) and P6 (IV:5), family 3 with 2 infertile males, P7 (IV:1) and P8 (IV:3), and three sporadic infertile cases P9, P10 and P11. Red arrows point to the individuals whom whole-exome sequencing was performed. Candidate mutations were validated by Sanger sequencing and the chromatograms were shown on the right. Red arrowheads point to mutant sites. Double horizontal lines represent consanguineous marriages. WT, wild-type allele; MT, the mutant allele. (B) The positions of the DNAH12 variants at the transcript (ENST00000351747.2) and protein levels (Q6ZR08-1). MT-like, Microtubule-binding-like domain. (C) Conservation analyses of affected amino acids across different species, the arrowheads indicate the mutation sites.

Detailed information of bi-allelic DNAH12 variants identified in infertile men

Abnormal sperm morphology and undetectable DNAH12 expression in men with bi-allelic DNAH12 variants

(A) Representative micrographs show normal sperm morphology from a fertile control and abnormal spermatozoa from P10 with absent, short, coiled, bent, irregular flagella or abnormal head under light microscopy. Scale bars, 5μm. (B) Representative SEM micrographs show details about the sperm morphology of a fertile control and P10. The normal sperm morphology can be observed in the fertile control (left) and abnormal types like absent, short, coiled, bent, irregular tails, and head abnormality in P10 (right). Scale bars, 5μm. (C) Immunoblotting assay revealed that DNAH12 was absent in the spermatozoa from P9, P10, and P11 harboring DNAH12 variants. β-actin was used as a loading control. (D) Sperm cells were co-stained with α-Tubulin (green) and DNAH12 (red) antibodies while no DNAH12 signals were observed in sperm flagella of patients. DNA was counterstained with Hoechst 33342. Scale bars, 10μm.

Clinical characteristics of the infertile patients with bi-allelic DNAH12 mutations

Flagellar axoneme defects were detected in the patients with DNAH12 variants

(A-B) Representative TEM micrographs showing cross-sections of the midpiece, principal piece, and end piece of sperm flagella from a fertile control and P11. The axoneme structure in control presents a “9+2” microtubules arrangement, including mitochondrial sheath (MS, indicated with red arrows), central pair of microtubules (CP, indicated with yellow arrows), outer dense fibers (ODFs, indicated with cerulean arrows), peripheral doublet microtubules (DMTs, indicated with orange arrows), inner dynein arms (IDAs, indicated with green arrows) and outer dynein arms (ODAs, indicated with purple arrows), while in P11, axonemal defects like missing CP, disorganized axonemal structures or missing DMTs were observed, the red asterisks mark CP loss (A). Impaired IDAs in the sperm axoneme of patient P11 (B). The red triangle marks impaired IDAs while adjacent ODAs are identifiable. Scale bars, 200 nm. (C-D) The proportion of different categories of observed cross-sections in the control and patients. Cross-sections were classified into four categories: Intact “9+2”, missing CP, disorganization, and missing DMTs (C), and the proportion of intact IDAs or impaired IDAs in the cross-sections of flagellar axoneme of which ODAs were identifiable (D). n, the total number of cross-sections for quantification. (E-G) Representative images of spermatozoa from fertile controls and patients carrying bi-allelic DNAH12 variants co-stained α-Tubulin with SPAG6 (E), DNAH1 (F), DNALI1 (G), or DNAH17 (H), and Hoechst 33342 for DNA (blue). Scale bars, 10 µm.

Disruption of Dnah12 results in impaired spermatogenesis and abnormal sperm morphology

(A) Fertility test of adult Dnah12+/+ and Dnah12−/− male mice. (B) The testes to body weight ratios of Dnah12+/+ and Dnah12−/− mice. (C) The number of sperm per epididymis of Dnah12+/+ and Dnah12−/− male mice. (D) Percentages of motile sperms in Dnah12+/+ and Dnah12−/−male mice. (E) Histological sections of testis and epididymis from Dnah12+/+and Dnah12−/− mice after H&E staining. Scale bars, 50 μm. (F) Morphology of the sperm from caudal epididymis of Dnah12+/+and Dnah12−/− mice. Scale bars,10 μm. (G) Quantitative analysis of sperm morphology of Dnah12+/+ and Dnah12−/−male mice. The experiments were repeated three times with at least 200 sperms counted every time. The above data were obtained from 3 adult mice for each genotype. (H) Representative images of testicular cells from Dnah12+/+ and Dnah12−/− mice co-stained DNAH12 and α-Tubulin antibodies. Scale bars, 10 μm. The red asterisk indicates abnormal manchette structure and the red arrow indicates abnormal sperm flagellum at the elongated spermatid stage of Dnah12−/− mice. (I) Representative immunofluorescence images of seminiferous tubule squash from 8-week-old Dnah12+/+ and Dnah12−/− mice, co-stained by PNA (red) and α-Tubulin (green) antibody. DNA was stained with Hoechst 33342. Scale bars, 10 μm. The asterisks indicate the manchette defects and the red triangles mark the abnormal sperm heads. Scale bars,10 μm. (J) Representative TEM micrographs of testicular elongated spermatids from Dnah12+/+ and Dnah12−/− mice. Scale bars, 1μm. (K) TUNEL assay on the testicular sections from Dnah12+/+and Dnah12−/− mice. The green triangles mark the apoptotic elongating or elongated spermatids. Scale bars, 50 μm. (L) The proportion of tubules with apoptotic cells in Dnah12+/+ and Dnah12−/− mice. (M) Average numbers of TUNEL-positive cells per tubule in Dnah12+/+ and Dnah12−/− mice. Data are obtained from three mice with at least 50 tubules scored in each repeated experiment. **p < 0.01; ***p <0.001; ****p < 0.0001, Student’s t test.

Deletion of Dnah12 causes axoneme defects in mice

(A) Representative TEM micrographs showing cross-sections of sperm flagella from Dnah12+/+, Dnah12+/-, and Dnah12−/− mice. In Dnah12+/+, Dnah12+/- groups, the normal axoneme “9+2” microtubule arrangement mainly consisted of the MS, (indicated with red arrows), CP (yellow arrows), ODFs (cerulean arrows), DMTs (orange arrows), IDAs (green arrows) and ODAs (purple arrows). However, the CP and DMTs structures were missing or disarranged in the Dnah12−/− group. Red asterisks mark the loss of CP. Scale bars, 200 nm. (B) Immunoblotting of sperm lysate from Dnah12+/+, DNAH12+/- and Dnah12−/−mice using DNAH1, DNALI1 or SPAG6 antibodies. DNAI2 and α-Tubulin were used as the loading controls. (C-H) Representative images of caudal epididymal sperm from Dnah12+/- and Dnah12−/− mice co-stained α-Tubulin and SPAG6 (C), SPEF2 (a component of CP complex) (D), DNAH1 (E), DNALI1 (F), DNAH17 (G) or DNAI2 (H) antibodies. Scale bars, 10 μm.

DNAH12 interacts with IDA components DNALI1 and DNAH1

(A) Phylogenetic and protein domain analyses of the DNAH family. The dashed box indicates the shortest-length member, DNAH12, while the arrow points to the missing microtubule-binding domain(MTBD) region. (B) Schematic diagram showing the procedure of the immunoprecipitation/mass spectrometry (IP/MS) proteomics experiment. (C) GO term enrichment analysis of DNAH12 potential interactors. (D) Protein-protein interaction network of DNAH12 among the motor protein term. The blue circle marks the proteins involved in IDA assembly. (E-F) Co-IP assays using DNAH12 antibody (E) or DNALI1 antibody (F) showed strong and specific interactions among DNAH12, DNALI1, and DNAH1. (G) Predicted DNAH12-DNALI1 interaction details by the HDOCK server. The top three of DNALI1 possible binding conformations were shown in types 1-3. (H-J) Construction of mouse-mCherry-DNAH12-Stem (H) and mouse-GFP-DNALI1-full length (FL) (I) plasmids, and the validation of the interaction between the Stem domain of DNAH12 and full-length of DNALI1 (J).

DNAH12 facilitates the recruitment of DNALI1 and DNAH1 to flagella, but not cilia

(A) Immunoblotting of testis lysate from Dnah12+/+, DNAH12+/- and Dnah12−/− mice using DNAH12, DNAH1, and DNALI1 antibodies. DNAI1 (a component of ODAs) and α-Tubulin were used as the loading controls. (B-C) Immunofluorescence assays of α-Tubulin and DNAH1 (B), or DNALI1 (C) antibodies on round spermatids from adult Dnah12+/−and Dnah12−/− mice. Scale bars, 10 μm. (D-E) Immunofluorescence assays of α-Tubulin and DNAH1 (D), or DNALI1 (E) antibodies on elongated spermatids from adult Dnah12+/− and Dnah12−/− mice. Scale bars, 10 μm. (F-G) Immunoblotting of oviduct (F) or trachea (G) lysate from Dnah12+/+, DNAH12+/- and Dnah12−/− mice using DNAH12, DNAH1, and DNALI1 antibodies. GAPDH was used as the loading control. (H-I) Immunofluorescence assays of the oviduct (H) or trachea (I) sections with α-Tubulin and DNAH1 antibodies. Scale bars, 50 μm. (J-K) Immunofluorescence assays of the oviduct (J) or trachea (K) sections with α-Tubulin and DNALI1 antibodies. Scale bars, 50 μm. (L) Pedigree of a family with 3 infertile males, P12 (III-2), P13 (III:6), and P14 (III:8); MT: the DNAH1 mutation c.11275C>T; p.Arg3759Cys. (M-O) Representative images of spermatozoa from a fertile control and P13 co-stained by α-Tubulin antibody and DNAH1 (M), DNALI1(N), or DNAH12 (O) antibodies, respectively. Scale bars,10 μm.

DNAH12 may interact with RS head associated proteins to regulate CP stability

(A) Co-IP assays of the potential interactions of DNAH12 and RS head proteins RSPH1, RPSH9, or DNAJB13. DNAI2 antibody was used as an negative control. (B) Interaction analyses of DNAH12 and RS head proteins RSPH1, RPSH9, or DNAJB13 by AlphaFold3. (C) Immunoblotting of testis lysate from Dnah12+/+ and Dnah12−/− mice using RSPH1, RPSH9, DNAJB13 antibodies. DNAI2 (a component of ODAs), α-Tubulin and β-Tubulin were used as the loading controls. (D) Immunofluorescence assays of α-Tubulin and RSPH1 antibodies on sperm collected from Dnah12+/− and Dnah12−/− mice caudal epididymis. Scale bars, 10 μm. (E) Representative two-cell embryos and blastocysts of Dnah12+/+, and Dnah12-/ male mice after intracytoplasmic sperm injection. Scale bar, 100 μm. (F) Percentages of 2-cell-stage embryos of Dnah12+/+ and Dnah12−/− groups. (G) Percentages of blastocyst-stage embryos of Dnah12+/+ and Dnah12−/− groups. (H) Schematic diagram showing the proposed function of DNAH12 in sperm flagellar development in humans and mice. DNAH12 is essential for recruiting DNALI1 and DNAH1 to sperm flagella and maintaining the proper axonemal arrangement, especially IDAs and CP structures. Loss of DNAH12 causes the failure of DNALI1 and DNAH1 to be recruited to sperm flagella and results in abnormal sperm morphology with compromised axoneme organization, causing male infertility. ODA, outer dynein arm; IDA, inner dynein arm; N-DRC, nexin-dynein regulatory complex; CP, central pair of microtubules; MTBD, microtubule-binding domain; RS, radial spoke.