Novel Variants in HFM1 Lead to Male Infertility Due to Non-obstructive Azoospermia

Dongdong Tang First A liated Hospital of Anhui Medical University https://orcid.org/0000-0002-9966-3642 Mingrong Lv First A liated Hospital of Anhui Medical University Yang Gao First A liated Hospital of Anhui Medical University Huiru Cheng First A liated Hospital of Anhui Medical University Kuokuo Li First A liated Hospital of Anhui Medical University Chuan Xu First A liated Hospital of Anhui Medical University Hao Geng First A liated Hospital of Anhui Medical University Guanjian Li First A liated Hospital of Anhui Medical University Qunshan Shan First A liated Hospital of Anhui Medical University Chao Wang First A liated Hospital of Anhui Medical University Xiaojin He First A liated Hospital of Anhui Medical University Yunxia Cao (  caoyunxia6@126.com ) First A liated Hospital of Anhui Medical University


Introduction
Azoospermia is a medical condition characterized by an absence spermatozoa in the ejaculated semen. It is the most serious form of male infertility, and approximately 1% of the male population and 10% of the infertile males are affected by this condition worldwide [1,2]. According to spermatogenic function, azoospermia is classi ed into obstructive azoospermia (OA) and non-obstructive azoospermia (NOA).
The development of assisted reproductive technology (ART), such as intracytoplasmic sperm injection (ICSI), has allowed most patients with OA to conceive their own offspring [3,4]. However, sperm retrieval in NOA patients has always been di cult even by microsurgical testicular sperm extraction (microTESE) [5,6]. Therefore, it is important to investigate the etiology and pathogenesis of NOA, which will facilitate the development of a targeted therapy.
Helicase family member 1 (HFM1) plays an important role in crossover formation and complete synapsis during meiosis [14]. Previous studies have reported that variants in HFM1 were related to female primary ovarian failure (POF) and male NOA or severe oligozoospermia in human [15,16]. And phenotypes of POF and NOA were also found in HFM1 −/− female and male mice, respectively [17]. In the present study, we reported two novel homozygous variants in HFM1 leading to NOA, which upgrade the clinical signi cance of HFM1 as a pathogenic gene for NOA. Additionally, we also investigated results of microTESE in cases with HFM1 variants for the rst time, which may provide a clinical reference that microTESE is not bene cial for these patients.

Subjects
An azoospermic patient from a Chinese consanguineous family (F1 II-1), who was admitted to the First A liated Hospital of Anhui Medical University for further diagnosis and treatment, was included in the study. Additionally, other 50 NOA patients were also recruited in the study. The ejaculated semen and urine were centrifuged, and some other analyses, including the determination of somatic karyotypes, screening of Y chromosome microdeletions, sex hormone testing, subsequent testicular biopsy, and testicular pathological analysis, were performed to identify the NOA phenotype in patients.

Ethical approval
All participants, including patients, a control individual with normal fertility, and two controls with OA, signed written informed consent to participate in the study. Hematoxylin and eosin (H&E) staining and immuno uorescence (IF) Fluorescence staining was performed as previous described. The following antibodies were used: mouse monoclonal antibody γH2AX (1:1000, Millipore #05-636, to detect meiotic DSBs and the XY body in pachytene nuclei, and apoptotic metaphases in case of pan-chromosomal signal). Histopathological assessment of spermatogenesis was performed by H&E staining. The location for candidate gene expression in seminiferous tubules was determined by IF. H&E staining and IF assay were performed as previously described [18,19]. The Probable ATP-dependent DNA helicase HFM1 antibody was used as the primary anti-HFM1 antibody (Abbexa, 1:50).
Quantitative real-time PCR (RT-PCR) and western blotting The mRNA and protein expression levels of the candidate gene were detected by RT-PCR and western blotting, respectively, according to previously described procedures [19,20]. The same antibody used for IF was applied for western blotting (1:1000), and the PCR primers used are listed in Supplementary table 2.

MicroTESE
As no sperm was detected in routine testicular biopsy, microTESE was performed to retrieve testicular sperms, according to a previous study [21].

Results
Homozygous mutations in HFM1 were identi ed in two of the selected patients Normal somatic karyotypes and Y chromosome microdeletions were observed in all 51 patients. H&E staining of the testicular tissue veri ed NOA in all selected patients. A rare homozygous HFM1 mutation (c.3490C>T: p.Q1164X) was observed and veri ed by WES and Sanger sequencing in F1 II-1, and his parents were classi ed as heterozygous carriers. Subsequently, sequencing of the coding region of HFM1 was performed in the 50 idiopathic NOA patients to detect HFM1 mutations. Another novel mutation of the HFM1 gene (c.3470G>A: p.C1157Y) was observed in four of the fty NOA patients. Three of the four cases were heterozygous and one was homozygous from a non-consanguineous family (F2 II-1). The detailed clinical features and genetic information of the mutants are presented in Table 1 and Figure 1.
Expression of HFM1 was signi cantly decreased in the testicular tissues of F1 II-1 and F2 II-1 Testicular tissues of F1 II-1 and F2 II-1 were collected and examined to determine the potential effects of HFM1 mutations. IF of the testicular tissues of OA patients revealed that HFM1 was concentrated in cytoplasm of spermatogonia and spermatocytes in seminiferous tubules. The IF signal of HFM1 were signi cantly decreased in both F1 II-1 and F2 II-1. Detailed information is present in Figure 2. Similarly, the expression levels of HFM1 mRNA and protein in the testicular tissues of F1 II-1 and F2 II-1 were signi cantly low compared with those of the control group ( Figure 3).
Outcome of microTESE in F1-II-1 After an exhaustive understanding of microTESE, both of F1 II-1 and F2 II-1 provided consent to undergo the procedure. However, no sperm were found in these two cases, although many spermatocytes were observed.

Discussion
In this present study, two homozygous variants of HFM1 were identi ed as genetic causes of spermatogenesis impairment in two patients with NOA. Additionally, the expression levels of HFM1 mRNA and HFM1 protein were signi cantly decreased in the two patients with homozygous HFM1 mutations. Moreover, no sperm were found in mutants despite the application of microTESE. These results indicate that HFM1 variants are novel causative mutations of NOA in humans.
The NOA phenotype in HFM1 -/adult male mice has been reported previously [17]. HFM1 variants related to NOA in human was rst reported in a study conducted in 2017, and compound heterozygous variants of HFM1 were observed in only one patient with NOA [16]. HFM1, comprising 39 exons mapped to the human chromosome 1q22, is a meiosis-speci c gene that plays important roles during synapsis and meiotic recombination [14][15][16][17]. HFM1 variants are not only associated with male NOA, but also female POF [15,16,22]. Apart from HFM1, a series of meiosis-related genes, such as STAG3 (OMIM: 608489) and DMC1 (OMIM: 602721), are also associated with fertility, and their variants can lead to male spermatogenic failure or female POI due to germ cell defects [12,23,24]. STAG3 encodes a meiosisspeci c protein, which plays an important role in chromosomal axis formation and sister chromatid recombination. Additionally, DMC1 is also a meiosis-speci c gene that plays an essential role in homologous recombination during meiosis and DNA double-strand break repair [22]. In our study, we observed two novel homozygous HFM1 variants in two unrelated NOA patients. F1 II-1 carried a homozygous stop-gain variant in HFM1 (c.3490C>T: p.Q1164X), which led to a truncated protein. This variant was classi ed "likely pathogenic" according to the American College of Medical Genetics and Genomics (ACMG) and the Association for Molecular Pathology (AMP) guidelines [25]. The functions of the truncated protein may be different from those of the normal one, and it can cause defects in chromosome synapsis and recombination, leading to spermatogenesis failure. F2 II-1 carried a homozygous missense variant in HFM1 (c.3470G>A), which changed the amino acid at position 1157 from cysteine to tyrosine. The replacement of a highly conserved hydrophilic amino acid with a hydrophobic amino acid may lead to a change in the secondary structure of the HFM1 protein and in uence its function. Potential high pathogenicity of the variant was predicted from Sorting Intolerant from Tolerant (SIFT), PolyPhen-2, and MutationTaster. The RT-PCR and western blot analyses also reported that the expression levels of HFM1 mRNA and protein were signi cantly decreased in the two patients with homozygous HFM1 mutations compared with the normal individuals.
Routine testicular biopsy of these two patients revealed no sperm. Considering the high sperm retrieval rate by microTESE in NOA patients [26,27], F1 II-1 and F2 II-1consented to undergo the procedure. However, no sperms were retrieved, either. It is the rst time to report outcomes of microTESE in NOA patients with variants in HFM1. Although only two patients were enrolled, it may provide a clinical reference that microTESE is not bene cial for these patients.
Some limitations should be taken into consideration in this study. Firstly, since sequencing of HFM1 was not performed in healthy fertile males, we cannot comment on the variant frequency of the gene in normal fertile males. Secondly, the chromosome spread experiments, which can detect possible defects during synapsis and recombination, was not performed as no enough testicular samples. Thirdly, only two cases with homozygous variants in HFM1 underwent microTESE, further research is necessary to verify the probability of successful sperm retrieval by microTESE in NOA patients with bialletic HFM1 mutations.

Conclusions
Our results provide further evidences for HFM1 is a candidate gene of NOA in human, and suggest that homozygous variants in HFM1 can cause autosomal recessive male infertility due to NOA. In addition, it is likely that microTESE cannot contribute to retrieve sperm in these patients.

Consent for publication
All the individuals and their family members, as well as two controls signed written informed consents after having received complete information about the research.
Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.