A pathogenic AKAP4 variant, p.R429H, causes male in/subfertility in humans and mice

Dear Editor,

Azoospermia, oligozoospermia and asthenozoospermia are well‐established causes of male infertility. Next‐generation sequencing has contributed to understanding Mendelian forms of male sterility. ¹ We identified a pathogenic hemizygous AKAP4 variant (c.1286G > A/p.R429H) shared by two siblings suffering from non‐obstructive azoospermia (NOA), while a different missense change involving the same amino acid residue, p.R429C, caused multiple morphological abnormalities of the sperm flagellum (MMAF) and severe oligozoospermia in a prior study. ² An equivalent Akap4 ^R428H mutation knock‐in mouse model was generated using CRISPR/Cas9 technology, which exhibited pronounced male subfertility characterized by diminished sperm count and motility, as well as fibrous sheath (FS) abnormalities in the flagella.

A‐kinase anchor protein 4 (AKAP4), an X chromosome‐linked gene, is exclusively expressed in spermatids and mature spermatozoa in previous studies. ³ AKAP4 participates in tethering Cyclic‐AMP dependent protein kinase A (PKA) to substrates for protein phosphorylation and constructing FS skeleton structure. ⁴ Recently, testicular single‐cell transcriptomic studies ⁵ , ⁶ have indicated that AKAP4, which escapes meiotic sex chromosome inactivation, is also expressed in spermatogonia and spermatocytes, despite lower expression than in spermatids. This knowledge suggests that AKAP4 might function beyond flagella development during spermatogenesis. Notably, variants in the AKAP4 gene have been identified in infertile males exhibiting distinct phenotypes, such as asthenozoospermia, ⁷ MMAF, ² and azoospermia/NOA, ⁸ , ⁹ which shed light on the divergence of male infertility phenotypes of AKAP4 mutations.

We recruited a family including two brothers with histopathologically confirmed NOA (Figure 1A,B). The characteristics of the two probands (II‐1, II‐2) are shown in Table 1. After precluding common etiological factors, whole‐exome sequencing was applied to search for genetic causes (Figure 1C). After strictly filtering, six genes remained: three showed biallelic missense mutations (OBSCN, SYNE1 and ZNF282), and three had X‐linked missense mutations (AKAP4, PLXNB3, and SRPK3) (Table S1, Figure S1). The contributions of mutations in OBSCN, SYNE1, PLXNB3 and SRPK3 to NOA phenomena were excluded by previous knockout (KO) mouse studies through a search of the Mouse Genome Informatics database (https://www.informatics.jax.org/). Segregation analysis was applied to identify hemizygous AKAP4 variation (Figure 1D, Table S2) and compound heterozygous ZNF282 variations (Figure S2). AKAP4 c.1286G > A/p.R429H is located in exon 5, and in silico information is detailed in Table 1 and Figure 1E–G. As yet, there is no literature report of Akap4 homologous mutant knock‐in and Zfp282‐KO mouse models (mouse ZFP282 is orthologous to human ZNF282), so we constructed corresponding mice to explore the effect of these variants on male fertility.

FIGURE 1.

Pedigree of the NOA family and identification of candidate variants by whole‐exome sequencing. (A) Pedigree of the non‐obstructive azoospermia family which included two probands. WES, whole‐exome sequencing. (B) Representative images of testicular tissue stained with hematoxylin‐eosin from proband II‐1 showed comprehensive meiotic arrest. Scale bars, 50 μm. (C) Flow chart of bioinformatics analysis of whole exome sequencing data. NOA, non‐obstructive azoospermia; KO, knockout. (D) Sanger sequencing of pedigree members to validate the AKAP4 candidate variant genotype. (E) Conservation analysis of the AKAP4 R429 amino acid. (F) AKAP4 candidate mutation located in exon 5 of the AKAP4 gene. (G) The missense mutation of AKAP4 p.R429H located in the AKAP4 protein FISP1 binding region.

TABLE 1.

Clinical information of the two non‐obstructive azoospermia (NOA) siblings and the in silico information of the hemizygous deleterious AKAP4 variant identified in the two probands.

	II‐1	II‐2
Age	35 years	28 years
Type of sterility	Primary
Medical history	None
Physical examination	Normal
Semen analysis	No sperm
Testicular biopsies	Maturation arrest
Karotype	46, XY
Y micro‐deletion detection	Normal
Hormone levels
E2 (20–47)	28.00 pg/mL	25.99 pg/mL
FSH (1.3–19.3)	3.76 mIU/mL	5.52 mIU/mL
LH (1.2–8.6)	6.96 mIU/mL	3.46 mIU/mL
PRL (2.3–13.1)	7.62 ng/mL	9.78 ng/mL
T (1.75–7.81)	6.53 ng/mL	4.73 ng/mL
AKAP4 variant
cDNA alteration	c.1286G > A
Variant allele	Hemizygous
Protein alternation	p.R429H
Variant type	Missense
Allele frequency in human population
All individuals in gnomAD	.0001
East Asians in gnomAD	0
1000 Genomes Projects	0
Function prediction
SIFT	Damaging
PolyPhen‐2	Damaging
MutationTaster	Disease_causing
CADD	24.3
Conservation analysis
GERP++	Conserved
PhyloP	Conserved
SiPhy	Conserved

Note: NCBI reference sequence number of AKAP4 is NM_003886. Variants with CADD value >15 are considered to be deleterious.

Zfp282‐KO mice (Figure S3A–C, Table S3 and S4) were viable and exhibited no overt abnormalities. A series of experiments were applied to test the fertility of Zfp282‐KO mice, including fertility test (Figure S3D), sperm counts and motility (Figure S3E), H&E staining of testis/epididymis sections and Papanicolaou staining of sperm (Figure S3F), which collectively showed that there is no impact of ZNF282 on male fertility. We generated Akap4 p.R428H mice (Figure 2A,B, Table S5 and S6), which is equivalent to the human AKAP4 ^R429H mutation. The AKAP4 protein was significantly reduced in the testes of Akap4 ^R428H mice (Figure 2C), indicating that the R428H variant might affect AKAP4 protein stability. The fertility test indicated severe male subfertility in which only 8/24 female mice mated with Akap4 ^R428H males were pregnant and produced 37 offspring; in contrast, 24/25 female mice mated with wild‐type (WT) males became pregnant and gave rise to 203 offspring (Figure 2D). The male reproductive system, testis/body weight ratio and histological examination of testis sections of Akap4 ^R428H mice were not obviously different from those of WT mice (Figure 2E–F). Transmission electron microscopy analysis showed no obvious abnormalities in manchette structure of spermatids in Akap4 ^R428H mice (Figure S4).

FIGURE 2.

The equivalent mutation in the mouse Akap4 gene leads to male subfertility and the Akap4 ^R428H mice show reduced sperm count/motility and a defective fibrous sheath. (A) Schematic illustration of the targeting strategy for generating Akap4 ^R428H (CGT to CAT) mice by using CRISPR/Cas9 technology. (B) Representative results of Sanger sequencing‐based genotyping using tail DNA. (C) Immunoblotting of AKAP4 was performed in wild‐type (WT) and Akap4 ^R428H testes. β‐actin served as a loading control. Bar graphs represent band intensities of blots, and data represent the mean ± SEM of three biological replicates. Student's t‐test, ***p 〈 .001. (D) A fertility test was performed in three Akap4 ^R428H male mice and three littermate WT male mice for 2 months. Each male mouse was mated with two C57BL/6J female mice. Vaginal plugs were observed in all mated females, and the number of pups per litter was recorded. Data are presented as the mean ± SEM. Student's t test, ***p 〈 .001. (E) The morphology of the male reproductive system and the testis/body weight ratio of Akap4R428H mice and WT mice. t, testis; epi, epididymis; vd, vas deferens; sv, seminal vesicle. Data represent the mean ± SEM of three biological replicates. Student's t test, ns, not significant. (F) Periodic acid‐Schiff staining of testis sections from Akap4R428H mice and WT mice. According to the components of the spermatids, the seminiferous tubules can be divided into stages I–XII. The differentiating spermatids within each stage are enlarged and shown in the upper right corners of the figures. Scale bars, 50 μm. (G) Sperm from the cauda epididymis of Akap4 ^R428H mice and WT mice. The sperm number was counted with a fertility counting chamber under a light microscope, and sperm motility was assessed by a computer‐assisted sperm analysis (CASA) system. Data are presented as the mean ± SEM (n = 8 each group). Student's t test was performed, **p 〈 .001. (H) Morphological analyses of epididymal sperm in Akap4 ^R428H mice and WT mice by Papanicolaou staining. Mp, mid‐piece; Pp, principal piece. Scale bars, 10 μm. (I) Percentage of sperm with short tails or bent tails from the cauda epididymis of Akap4R428H mice and WT mice. At least 100 sperm were counted for each mouse. Data are presented as the mean ± SEM (n = 3 each group). Student's t test was performed, *p 〈 .05. (J) Epididymal sperm from WT mice and Akap4 ^R428H mice were fixed and sectioned for transmission electron microscopy. Cross‐sections of the principal piece of flagella are shown. The vacant positions of the fibrous sheath (FS) are indicated by arrows. LC, longitudinal column; CR, circumferential ribs. Scale bars, 100 nm. (K) Percentage of sperm with FS deficiency in Akap4 ^R428H mice and WT mice. At least 50 principal pieces of flagella were counted for each mouse. Data are presented as the mean ± SEM (n = 3 each group). Student's t test was performed, *p 〈 .05, ***p 〈 .001. (L and M) Immunoblotting of AKAP3 and QRICH2 was performed in WT and Akap4 ^R428H testes. β‐actin served as a loading control. Bar graphs represent band intensities of blots. Mean ± SEM of three biological replicates. Student's t test, ***p 〈 .001.

To illustrate the cause of reduced fertility in Akap4 ^R428H mice, we performed sperm analysis using mature sperm from the cauda epididymis. Significantly reduced sperm count and progressive motility were identified in Akap4 ^R428H mice (Figure 2G). Papanicolaou staining further indicated that Akap4 ^R428H mice produced short‐tailed and bent‐tailed sperm with dramatically attenuated principal piece (Figure 2H). Short‐tailed and bent‐tailed sperm accounted for approximately 33% and 25% of the total sperm in Akap4 ^R428H, respectively (Figure 2I). TEM showed that FS was either unrecognizable or partially lost (Figure 2J). The ratio of FS deficiency in sperm from Akap4 ^R428H was significantly higher than that in WT sperm (54.00% vs. 5.90%) (Figure 2K). We further found that the protein expression of AKAP3 (a FS protein) and QRICH2 (a known target of AKAP4) was significantly lower in the testis lysates of Akap4 ^R428H mice (Figure 2L,M). Collectively, these data confirmed that the R429H variant of AKAP4 is a pathogenic mutation to cause male in/subfertility.

To exploit the regulatory mechanism of AKAP4 in mouse spermatogenesis, we reanalyzed the single‐cell transcriptome data of Akap4‐KO and WT testes from the Sequence Read Archive database (access number: SRR9107534), ¹⁰ which were detailed in the supplementary Material and Methods (Figure S5A,B). Through differential expression analysis without distinguishing cell types we identified thirteen main differentially expressed genes (DEGs) (|log2FC|≥.5, adjusted p < .05) including Ccdc38 and Haspin, which were reported in previous literature ¹⁰ (Figure S5C). Trajectory analysis provided novel findings compared with the previous results. ¹⁰ The shorter velocity vectors indicated a decreased accumulation of mRNA in Akap4‐KO mice (Figure S6A). Round spermatids (RSs) were positioned at the starting point of the pseudotime trajectory in KO mice, whereas spermatocytes served as the early state of cell differentiation in WT mice (Figure S6B,C). Enrichment of RS DEGs did not show a significant pathway for spermatogenesis (Figure S7A–C). Protein‐protein interaction network (PPI) analysis of these DEGs showed potential AKAP4 interacting partners (Figure S7D,E). Both up/downregulated DEGs in elongating spermatids were significantly enriched in spermatogenesis (Figure S6D,E), and PPI revealed a central node containing AKAP4, H1FNT (Figure S6F,G), suggesting that AKAP4 may act through its interactors to mainly affect the late stages of spermiogenesis.

In conclusion, the functional alterations in AKAP4 have been demonstrated to exert significant contributions to male infertility, encompassing asthenozoospermia (including MMAF), severe oligozoospermia and even a complete failure of spermatogenesis (Figure 3). The underlying mechanisms of AKAP4 mutations leading to NOA and the phenotypic difference between the AKAP4 p.R429H variant in humans and Akap4 ^R428H mice remain to be studied in the future.

FIGURE 3.

A pathogenic hemizygous variant in AKAP4 causes azoospermia in humans and severe male subfertility in mice. A pathogenic hemizygous AKAP4 variant (c.1286G〉A/p.R429H) was identified in two siblings suffering from non‐obstructive azoospermia, while a different missense change involving the same amino acid residue, p.R429C, has been associated with multiple morphological abnormalities of the sperm flagellum (MMAF) and severe oligozoospermia in a prior study. To further validate the pathogenicity of this mutation on AKAP4 function and male fertility, we constructed corresponding mutant (knock‐in) mice using CRISPR/Cas9 technology. Akap4 ^R428H mice displayed severe male subfertility, showing reduced sperm count/motility and fibrous sheath abnormalities in flagella. Together, our study provides clinical and animal evidence to reveal a hemizygous variant in AKAP4 (c.1286G〉A/p.R429H) causes male in/subfertility in both humans and mice.

AUTHOR CONTRIBUTIONS

Wei H and Zhang XH performed the major experiments and wrote the manuscript. Wang CY. and Wang J undertook the bioinformatics analysis. Li TY completed Sanger sequencing. Chen SR, Li HJ and Wang BB designed the study and revised the manuscript. All authors approved the final version for submission.

CONFLICT OF INTEREST STATEMENT

The authors declare no competing interests in relation to publication of this study.

FUNDING INFORMATION

This work was supported by the Beijing Municipal Natural Science Foundation (7232112), the National Key Research and Development Project (2019YFA0802101) and the Open Fund of Key Laboratory of Cell Proliferation and Regulation Biology, Ministry of Education.

ETHICS STATEMENT

This study was approved by the ethics committee from the Peking Union Medical College Hospital and National Research Institute for Family Planning. Animal experiments were approved by the Animal Care and Use Committee of the College of Life Sciences, Beijing Normal University.

Supporting information

Supporting information

DATA AVAILABILITY STATEMENT

The datasets generated during the current study are available from the corresponding author upon reasonable request.