A novel AMH variant at the prehelix loop impairs the binding to AMHR2 and causes persistent Müllerian duct syndrome

Dear Editor,

Persistent Müllerian duct syndrome (PMDS; Online Mendelian Inheritance in Man [OMIM]: #261550) is a rare autosomal recessive inherent difference in sex development in males with Müllerian duct derivatives, including the uterus, fallopian tubes, and upper part of the vagina.1 Approximately 88% of PMDS patients were found to have biallelic variants in the anti-Müllerian hormone (AMH; OMIM: *600957) or AMH type 2 receptor (AMHR2; OMIM: *600956) genes, and the rest were referred as idiopathic.1 The serum AMH levels are usually very low or undetectable in PMDS patients with AMH gene mutations but normal or elevated in those with AMHR2 gene mutations. Here, an adult Chinese patient with PMDS carrying biallelic variants of the AMH gene was reported to have unexpectedly normal serum AMH level. Please refer to Supplementary Information of Patient and Methods for methodologies.

A 32-year-old male who was referred to the Center of Reproductive Medicine of Shengjing Hospital of China Medical University (Shenyang, China) on March 18^th, 2021, was enrolled in this study. Written informed consent was obtained and all protocols were approved by the Ethics Committee of Shengjing Hospital of China Medical University (Approval No. 2021PS523K). He was born to nonconsanguineous parents, with a typical sister who delivered a healthy boy (Figure 1a). He had experienced 8 years of infertility, with occasional hematospermia. He had transverse testicular ectopia and underwent cryptorchidectomy at 1 year old after a failed attempt to return the left testis. Physical examination revealed an empty right scrotum and inguinal hernia on the left side. No gynecomastia or other abnormalities were observed. Semen analysis revealed oligoasthenozoospermia with low sperm volume, low normal morphology, low survival rate, high sperm nucleoprotein immaturity, low DNA fractionation index, and no progressive motility. Hormone analyses revealed slightly elevated levels of prolactin and estradiol. Serum AMH (3.18 ng ml⁻¹) was within the normal range (Supplementary Table 1). Ultrasonography showed a single testis of 6.3 cm × 2.1 cm × 2.5 cm in the left scrotum. Magnetic resonance scanning showed 10.0 cm × 2.1 cm × 2.5 cm of tissue with a fluid-filled cavity (Figure 1b).

Figure 1.

(a) Pedigree of the patient with persistent Müllerian duct syndrome. The affected proband is indicated by a black arrow. (b) Pelvic magnetic resonance scanning. The Müllerian structure is indicated by red arrows. The left testis is indicated by blue arrows. (c) Sanger sequencing of the heterozygous c.1165G>T (p.E389X) and c.1507T>C (p.Y503H) variants (indicated by red arrows) of the AMH gene. (d) Evaluation of the amino acid conservation by Ugene. Each residue box in alignment is assigned a color according to ClustalX color scheme. The Y503 of AMH is indicated in a black box. (e) Three-dimension molecular docking of AMH (green molecule) with AMHR2 (red molecule). The cluster was constructed by Haddock2.410 and visualized by PyMOL. The interface of wild-type (left) and Y503H mutant (right) AMH with the extracellular domain of AMHR2 showed differences in both bonds and surface. (f) ELISA of the secreted AMH in the transfected HEK293 cell culture media. The kit was specific to the C-terminal bioactive domain. ^*P ˂ 0.01 in the post hoc Tukey analysis. (g) The binding of AMH with AMHR2 evaluated by IP and IB. The Myc-tagged AMHR2-transfected cells were incubating with AMH-transfected culture media. The interactive bands (IP: Myc; IB: AMH) were normalized to that of internal control (IP: Myc; IB: Myc), with the wild-type taken as “1”. ^*P ˂ 0.01 in the Student’s t-test. (h) H&E staining of the Müllerian remnant. Immunohistochemistry for (i) ER and (j) PR. Scale bars = 20 μm. AMH: anti-Müllerian hormone; ELISA: enzyme-linked immunosorbent assay; AMHR2: AMH type 2 receptor; IP: immunoprecipitation; IB: immunoblotting; H&E: hematoxylin and eosin; ER: estrogen receptor; PR: progesterone receptor.

Supplementary Table 1.

Sperm and hormone parameters in the patient with persistent Müllerian duct syndrome

Parameter	Patient	Reference range
Sperm volume (ml)	1.0*	≥1.5
Sperm pH	7.4	7.2–8.0
Sperm concentration (×106 ml−1)	27.49	≥15
Progressively motile sperm (%)	0*	≥32
Normal morphology sperm (%)	3.0*	≥4
Sperm nucleoprotein immaturity (%)	43.0*	≤30
Hypo-osmotic swelling test (%)	3.0*	≥58
DNA fractionation index (%)	81.7*	0–30
Follicle-stimulating hormone (mIU ml−1)	3.50	1.5–12.4
Luteinizing hormone (mIU ml−1)	3.62	1.7–8.6
Prolactin (ng ml−1)	17.01*	2.64–13.13
Testosterone (ng dl−1)	4.41	2.49–8.36
Estradiol (pg ml−1)	58.35*	14–55
AMH (ng ml−1)	3.18	0.73–16.05

^*Exceeding reference range values. AMH: anti-Müllerian hormone

Cytogenetic analysis revealed a typical 46,XY karyotype. Compound heterozygous variants of c.1165G>T (p.E389X) and c.1507T>C (p.Y503H) in the AMH gene were identified by whole-exome sequencing and confirmed by Sanger sequencing (Figure 1c). The nonsense variant has been reported in Chinese patients.2,3,4 The novel missense variant occurs at a highly conserved amino acid across various species (Figure 1d). In silico analyses by Sorting Intolerant From Tolerant (SIFT; http://sift.jcvi.org), ProVean (http://provean.jcvi.org/index.php), Polyphen2 (http://genetics.bwh.harvard.edu/pph2), and MutationTaster (http://www.mutationtaster.org), all predicted the variant as “deleterious”. Considering the unexpectedly normal serum AMH level, we evaluated three-dimensional (3D) molecular docking to the receptor AMHR2 using Haddock2.4 and visualized the top cluster using PyMOL (https://pymol.org/; Figure 1e). The Y503 at the prehelix loop of the AMH palm generated a hydrogen bond (3.7 Å) to AMHR2 C559 at the terminus of finger 4, creating a connective molecular surface. The substitution by H503 disrupted the H-bond and the connective surface, prolonging the distance to T47 of AMHR2 from 4.9 Å to 5.6 Å.

Secretion and receptor binding were investigated in transfected HEK293 cells. In the enzyme-linked immunosorbent assay (ELISA) specific for C-terminal mature AMH (Figure 1f), the E389X mutant AMH was almost undetectable in the culture media. In contrast, the Y503H mutant AMH mildly decreased to 72.8% of the wild-type control. Receptor binding was assayed by incubating AMHR2-transfected cells in the AMH-transfected culture media. After immunoprecipitation with Myc-tagged AMHR2, the proteins were immunoblotted to the Y503H mutant AMH with a significantly decreased level (36.3% of the wild-type) but not to the E389X mutant AMH (Figure 1g). These data indicate that the pathogenic mechanism of E389X AMH is impaired biosynthesis, and Y503H AMH is impaired receptor binding.

The patient underwent testicular sperm extraction and intracytoplasmic sperm injection (ICSI) and got a healthy child. An informed choice for subtotal hysterectomy was made. The resected Müllerian remnants included a seemingly well-developed uterus filled with 30 ml of bloody liquid and a fallopian tube; the cervix was preserved. Pathological analysis confirmed the uterine structure (Figure 1h). Immunohistochemistry revealed positive staining in estrogen receptor (Figure 1i) and progesterone receptor (Figure 1j). The left inguinal hernia was repaired without complications.

Most cases of PMDS were incidentally identified during evaluation or corrective surgery for cryptorchidism or inguinal hernia in children. Overlooking adult patients were rare. The patients in our present and previous studies demonstrated asthenozoospermia with no progressive motility,2 which was supported by other adult patients with azoospermia, oligozoospermia, or asthenozoospermia.4,5,6 Infertility is a major concern in PMDS, and we have added successful experience in testicular sperm retrieval and ICSI. The development of the Müllerian remnant varies, and interventions remain controversial. The risk of Müllerian neoplastic degeneration (4%) is less than that of undescended testes (33%).7,8 Surgical removal may unavoidably damage the vas deferens or the deferential artery to some degree. However, the extended uterus may induce urinary retention by mechanical obstruction, and highly differentiated Müllerian derivatives may cause excess estrogen, inducing gynecomastia and hematospermia. Considering our patient’s large uterus and occasional hematospermia, subtotal hysterectomy was suggested after successful assisted reproduction.

AMH comprises a large N-terminal pro region crucial for proper folding and stability and a smaller C-terminal mature domain responsible for receptor binding.1 The p.E389X variant at the N-terminal domain created a premature termination codon with a long distance to the poly(A) binding protein, which might activate nonsense-induced mRNA decay. Cells transfected with p.E389X expression vectors demonstrated null expression using C-terminal-specific antibodies.2 Indeed, almost all patients with AMH mutations showed very low or undetectable serum AMH levels. Our patient is an exception, with similar observations in two patients with homozygous p.Q496H and p.R502L variants.8,9 Together with the novel p.Y503H variant in our patient, all three resided in the highly conserved prehelix loop of the C-terminal domain, which is important for receptor interaction. The 3D molecular docking analysis of Y503H AMH demonstrated disruption in the H-bond and connective surface to AMHR2, which was confirmed by a co-immunoprecipitation experiment. Therefore, the Y503H variant may be destructive in tethering parallel fingers to maintain a stable interface for receptor binding. The decrease to 72.8% of wild-type AMH secretion might not exceed the wide reference range if set in human serum, indicating a minor pathogenic effect.

In conclusion, we report an adult patient with PMDS carrying biallelic loss-of-function AMH variants with unexpectedly normal serum AMH level. The pathogenic mechanism of the novel p.Y503H variant in the prehelix loop of the palm was found to be defective in receptor binding.

AUTHOR CONTRIBUTIONS

QD evaluated the clinical data and drafted the manuscript. CQ performed the cellular and molecular experiment. YYZ performed the genetic analysis. XJS performed the bioinformatic analysis and reviewed the literature. XLL conceived and supervised the study. All the authors took part in revising the manuscript, and read and approved the final manuscript.

COMPETING INTERESTS

All authors declare no competing interests.

SUPPLEMENTARY PATIENT AND METHODS

PATIENT

A patient at the Center of Reproductive Medicine of Shengjing Hospital of China Medical University (Shenyang, China) was enrolled in this study, with clinical data collected and peripheral blood samples extracted for genetic tests after obtaining written informed consent. We followed the Declaration of Helsinki, and all protocols were approved by the Ethics Committee of Shengjing Hospital of China Medical University (Approval No. 2021PS523K).

CLINICAL DATA COLLECTION

Semen analysis was performed in the regular way. Serum hormone levels including estradiol (E2), testosterone (T), follicle-stimulating hormone (FSH), luteinizing hormone (LH), and prolactin (PRL), were determined by chemiluminescent immunoassay (Beckman Counter, Pasadena, CA, USA). Anti-Müllerian hormone (AMH) levels in the serum and cell culture media were analyzed using by ELISA (Beckman Counter). Radiographic examination of MRI and ultrasound scanning was performed on the pelvic and urogenital system.

IMMUNOHISTOCHEMISTRY

The Müllerian remnant was extracted and fixed in 10% neutral buffered formalin at 4°C overnight. The tissues were embedded in low temperature paraffin wax and sliced at 3 μm thick for regular HE staining. Immunohistochemistry was performed by the avidin-biotin-peroxidase complex method using the VECTASTAIN^® ABC Kit (Vector Laboratories, Newark, CA, USA) according to the manufacturer’s instructions. The slices were incubated with anti-estrogen receptor (ER) and anti-progesterone receptor (PR) antibodies (Yaji Biological, Shanghai, China) at 1:500 dilution (Alomone Labs, Jerusalem, Israel). The photomicrographs were taken using an inverted microscope (Olympus IX51, Olympus, Tokyo, Japan).

GENETIC AND IN SILICO ANALYSIS

Genomic DNA was isolated from peripheral blood samples using the Blood Genomic DNA Miniprep Kit (Axygen, Union City, CA, USA). Whole-exome sequencing (WES) was performed on proband using biotinylated capture probes (MyGenostics, Baltimore, MD, USA), and the Illumina NextSeq 500 platform (Illumina, San Diego, CA, USA). Variant calling was performed with the Illumina NextSeq Reporter Software using NCBI37/hg19 assembly of the human genome as reference sequences. Sanger sequencing was performed to confirm the AMH variants on the proband and his parents. The conservation of the amino acid sequences was assayed by Ugene (https://ugene.net). The harmfulness of the variants was assayed by SIFT (http://sift.jcvi.org), ProVean (http://provean.jcvi.org/index.php), Polyphen2 (http://genetics.bwh.harvard.edu/pph2), and MutationTaster (http://www.mutationtaster.org). The structural models of AMH (UniProt ID P03971) and AMH receptor 2 (AMHR2; UniProt ID Q16671) generated by Swiss-MODEL, and protein-protein docking was performed using Haddock2.4 (https://bianca.science.uu.nl/haddock2.4/). The three-dimensional (3D) structural analysis was performed using PyMOL software.

PLASMID CONSTRUCTION AND CELL TRANSFECTION

Human Myc-tagged AMHR2 (NM_020547.3) as well as wild-type, E389X, and Y503H mutant AMH (NM_000479.5) expression vectors were constructed by GeneChem Co., Ltd (Shanghai, China). HEK293 cells were transfected with 2 µg of AMH expression vectors using the Advanced DNA RNA Transfection Reagent (Zetalife, Menlo Park, CA, USA). The cells were changed into serum free media at 24 h posttransfection and were maintained for additional 24 h to collect the cell media. Another set of HEK293 cells were transfected with 2 µg of AMHR2 expression vectors and changed at 24 h posttransfection into the media of cells transfected with AMH expression vector. Plasmin (1 µg ml-1, Sigma-Aldrich, St. Louis, MO, USA) was added to promote the digestion into active AMH. The cells were harvested after 24 h of incubation for further protein analysis.

IMMUNOPRECIPITATION AND IMMUNOBLOTTING

Total cell protein was isolated using gentle RIPA lysis buffer and incubated with 2 µg of anti-Myc antibodies (Proteintech, Chicago, IL, USA) or non-immune rabbit IgG, conjugated to Protein A/G magnetic beads (Bimake, Houston, TA, USA) by rotation at 4°C overnight. The immunoprecipitated proteins were washed and eluted from the beads using 2× SDS sample buffer. The proteins underwent electrophoresis on 10% sodium dodecyl sulfonate-polyacrylamide gel electrophoresis (SDS-PAGE) and membrane transfer for the subsequent immunoblotting. After membranes blocking in TBST buffer containing 5% non-fat milk, the immunoblotting was performed at 4°C overnight with anti-AMH (1:2000, Zen BioScience, Chengdu, China) and anti-Myc (1:1000, Proteintech, Rosemont, IL, USA) as primary antibodies, followed by HRP-conjugated IgG (1:5000, Proteintech) as secondary antibodies for 2 h at room temperature. The bands were detected using SuperLumia ECL Kit (Abbkine Scientific, Wuhan, China), and the densitometry was quantified using Quantity One software version 5.0 (Bio-Rad Laboratories, Hercules, CA, USA).

STATISTICAL ANALYSES

Quantitative data are presented as mean ± standard error of the mean. Statistical analysis was performed using SPSS version 23.0 (IBM Corp, New York, NY, USA) for Windows. The Student’s t-test or the post hoc Tukey analysis was performed comparisons as appropriate. P < 0.05 was considered statistically significant.