HSD17B12 Dosage Insufficiency Induced Premature Ovarian Insufficiency in Humans and Mice
Qiqi Wang,Qing Chen,Yixin Zhang,Xue Zhang,Chunyu Liu,Daqi Wang,Yanhua Wu,Yixi Sun,Ling Zhang,Chengcheng Song,Yongming Wang,Yanpeng An,Huiru Tang,Congjian Xu,Yanting Wu,Li Jin,Hefeng Huang,Feng Zhang
DOI: https://doi.org/10.1002/ctm2.737
IF: 8.554
2022-01-01
Clinical and Translational Medicine
Abstract:Premature ovarian insufficiency (POI) is a severe female reproductive disorder that affects 1%−2% of women in general populations.1 The analysis of familial POI implies that genetic aberrations strongly influence the onset of POI.2 However, a large proportion of POI cases remain idiopathic, suggesting that novel causative or susceptible factors are yet to be discovered.3 Here, we identify HSD17B12 dosage insufficiency as a novel mechanism underlying human POI. A nonconsanguineous Han Chinese pedigree with two daughters exhibiting POI was investigated (Table S1). The proband (II-1, Figure 1A) presented primary amenorrhea. Ultrasound examination revealed the infantile uterus (size: 19 × 14 × 18 mm) and invisible ovaries. She was treated with hormone replacement therapy at age 19 to induce menarche and maintain sexual development and cyclical bleeding. Her younger sister (II-2) experienced normal first menarche at 12 years old. But she was diagnosed with early-onset POI at 13 years old when occurring amenorrhea. Both sisters have normal 46,XX karyotypes. Their FMR1 CGG repeat lengths are in the regular polymorphic range. Both parents were reported as healthy with normal high-resolution karyotypes. The mother had normal pubertal development and continued to have regular menstrual periods at her age of 49. No history of associated endocrinopathies or autoimmune disorders was found in this family. We conducted whole-exome sequencing and subsequent array-based comparative genomic hybridization (aCGH) and found that both index POI sisters harbored compound heterozygous variants in HSD17B12 (Figure 1A), including a novel missense variant (M1: c.610G > C, p.A204P) and a novel deletion (M2). The maternally inherited missense variant (Figure 1B) was confirmed by Sanger sequencing (Table S2) and predicated to be pathogenic by bioinformatic tools (Table 1). Additionally, long-range polymerase chain reaction (PCR) confirmed a paternally inherited deletion affecting HSD17B12 exons 5 and 6 (Figure 1C). HSD17B12 (hydroxysteroid 17-beta dehydrogenase 12) encodes a major enzyme responsible for the estrone to estradiol (E2) conversion.4 Importantly, the A204 position of HSD17B12 is conserved according to phyloP and phastCons programs (Table 1), and the p.A204P variant could disturb the structural stability of HSD17B12 by missing hydrogenbonds formed between A204 and other amino acids (Figure S1). Additionally, both HSD17B12 missense and deletion variants carried by both index sisters are located in the crucial short-chain dehydrogenase/reductase domain (Figure 1D) that is conserved during evolution.5 Dramatically, no obvious HSD17B12 was detected in the blood samples of index sisters (Figure S2), indicating deleterious effects of HSD17B12 variants. To further investigate the pathogenesis of bi-allelic HSD17B12 variants in POI, we knocked these variants into human ovarian KGN cells using CRISPR-Cas9 system (Table S2). Firstly, we sequenced the low-abundance complementary DNA of HSD17B12M2/M2 KGN cells and revealed that the M2 deletion of HSD17B12 exons 5 and 6 (c.392_501del) created a frameshift in canonical HSD17B12 (NP_057226.1: p.V131Dfs*51) (Figure S3). Secondly, quantitative reverse transcription PCR (RT-PCR) demonstrated that HSD17B12+/M2 KGN cells expressed only a half abundance of HSD17B12 when compared with wild-type (WT) controls (Figure 1E), suggesting the possibility of nonsense-mediated mRNA decay.6 Thirdly, we observed a further reduced expression of HSD17B12 in HSD17B12M1/M2 KGN cells (Figure 1E,F). The above experimental evidence suggested a gene dosage effect caused by the combination of paternally and maternally inherited deleterious variants in HSD17B12. Furthermore, we generated gene-edited mouse models to investigate in vivo function of Hsd17b12 (Figure S4). We crossed Hsd17b12+/− and Hsd17b12+/A204P mice of 8 weeks old. However, no Hsd17b12−/A204P mice were found at birth (Table S3), which was due to the embryogenic lethality at E11.5. Coincidentally, we failed to obtain Hsd17b12−/− or Hsd17b12A204P/A204P live mice by intercrossing heterozygous mutants, since they died at E8.5 and P0, respectively (Table S3). These experimental observations indicate that mice are more vulnerable than human subjects to HSD17B12/Hsd17b12 dosage insufficiency. These divergent phenotypes between bi-allelic HSD17B12/Hsd17b12-mutated humans and mice might be attributed by that the 17b-hydroxysteroid dehydrogenases are largely multifunctional enzymes, and their functions depend on the substrates in different species or the isoenzymes.7 Since bi-allelic Hsd17b12-mutated mice were embryonically lethal, heterozygous Hsd17b12-mutated adult mice were recruited to determine the effects of Hsd17b12 variants on female fertility. Although there was no significant difference in ovarian index between heterozygous Hsd17b12-mutated and WT groups (Table S4), significantly decreased mRNA abundances of Hsd17b12 were observed in the ovaries of Hsd17b12+/− and Hsd17b12+/A204P female mice when compared with their WT littermates (Figure 2A). Subsequently, Hsd17b12+/− and Hsd17b12+/A204P female mice of 8 weeks old and their WT control littermates were mated with WT adult males for up to 6 months. Surprisingly, although the offspring fit a conventional pattern of Mendelian inheritance (Table S5), both Hsd17b12+/− and Hsd17b12+/A204P female mice showed significantly longer times between pregnancies and smaller litter sizes than WT controls (Table 2). Additionally, after 2 months daily examination of estrus cycle, Hsd17b12+/− female mice showed inordinate prolongations in the metestrus and diestrus and so that the entire estrous cycle when compared to WT controls (Figure 2B,C). To further visualize the potential female subfertility, the morphologies of follicles at different developmental stages were examined by hematoxylin and eosin staining. Both Hsd17b12+/− and Hsd17b12+/A204P female mice showed significantly reduced numbers of developing-follicles in their ovaries when compared to WT controls (Figure 2D), indicating the correlation between Hsd17b12 dosage insufficiency and female subfertility in mice. As for Hsd17b12 homozygous mutants, Hsd17b12A204P/A204P mice lived till P0; therefore, the P0 female mice were used for three-dimensional structure reconstruction of cleared-ovaries, which showed smaller and thinner ovaries, and consistently, fewer primordial follicles in Hsd17b12A204P/A204P female mice than those in WT controls (Figure S5), further suggesting that Hsd17b12 is essential for ovarian morphology and reserve. HSD17B12 was previously reported in the regulation of female reproduction potentially via the metabolism of arachidonic acid, a precursor of prostaglandins.8-10 Coincidently, the metabolic processes of arachidonic acid and related products are affected in the ovaries of heterozygous Hsd17b12-mutated female mice (Figure S6). In conclusion, our observations based on human POI subjects, gene-edited KGN cell, and mouse models supported the pathogenic roles of HSD17B12 deleterious variants and their associated HSD17B12 dosage insufficiency in POI. HSD17B12 is genetically involved in human early-onset POI and even primary amenorrhea via the autosomal recessive inheritance. The authors would like to thank the family for participating and supporting this study. This work was supported by National Natural Science Foundation of China (grant numbers: 31625015 and 31521003), Shanghai Municipal Science and Technology Major Project (grant number: 2017SHZDZX01), State Key Laboratory of Reproductive Medicine (grant number: SKLRM-K202002), Science and Technology Major Project of Inner Mongolia Autonomous Region of China (grant number: zdzx2018065), and the 111 Project (grant number: B13016). The authors declare that they do not have any commercial or associative interest that represents a conflict of interest in connection with the work submitted. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.