Biallelic mutations in RNA-binding protein ADAD2 cause spermiogenic failure and non-obstructive azoospermia in humans

STUDY QUESTION

What are some pathogenic mutations for non-obstructive azoospermia (NOA) and their effects on spermatogenesis?

SUMMARY ANSWER

Biallelic missense and frameshift mutations in ADAD2 disrupt the differentiation of round spermatids to spermatozoa causing azoospermia in humans and mice.

WHAT IS KNOWN ALREADY

NOA is the most severe cause of male infertility characterized by an absence of sperm in the ejaculate due to impairment of spermatogenesis. In mice, the lack of the RNA-binding protein ADAD2 leads to a complete absence of sperm in epididymides due to failure of spemiogenesis, but the spermatogenic effects of ADAD2 mutations in human NOA-associated infertility require functional verification.

STUDY DESIGN SIZE DURATION

Six infertile male patients from three unrelated families were diagnosed with NOA at local hospitals in Pakistan based on infertility history, sex hormone levels, two semen analyses and scrotal ultrasound. Testicular biopsies were performed in two of the six patients. Adad2 mutant mice (Adad2^Mut/Mut) carrying mutations similar to those found in NOA patients were generated using the CRISPR/Cas9 genome editing tool. Reproductive phenotypes of Adad2^Mut/Mut mice were verified at 2 months of age. Round spermatids from the littermates of wild-type (WT) and Adad2^Mut/Mut mice were randomly selected and injected into stimulated WT oocytes. This round spermatid injection (ROSI) procedure was conducted with three biological replicates and >400 ROSI-derived zygotes were evaluated. The fertility of the ROSI-derived progeny was evaluated for three months in four Adad2^WT/Mut male mice and six Adad2^WT/Mut female mice. A total of 120 Adad2^Mut/Mut, Adad2^WT/Mut, and WT mice were used in this study. The entire study was conducted over 3 years.

PARTICIPANTS/MATERIALS SETTING METHODS

Whole-exome sequencing was performed to detect potentially pathogenic mutations in the six NOA-affected patients. The pathogenicity of the identified ADAD2 mutations was assessed and validated in human testicular tissues and in mouse models recapitulating the mutations in the NOA patients using quantitative PCR, western blotting, hematoxylin-eosin staining, Periodic acid-Schiff staining, and immunofluorescence. Round spermatids of WT and Adad2^Mut/Mut mice were collected by fluorescence-activated cell sorting and injected into stimulated WT oocytes. The development of ROSI-derived offspring was evaluated in the embryonic and postnatal stages.

MAIN RESULTS AND THE ROLE OF CHANCE

Three recessive mutations were identified in ADAD2 (MT1: c.G829T, p.G277C; MT2: c.G1192A, p.D398N; MT3: c.917_918del, p.Q306Rfs*43) in patients from three unrelated Pakistani families. MT1 and MT2 dramatically reduced the testicular expression of ADAD2, likely causing spermiogenesis failure in the NOA patients. Immunofluorescence analysis of the Adad2^Mut/Mut male mice with the corresponding MT3 mutation showed instability and premature degradation of the ADAD2 protein, resulting in the spermiogenesis deficiency phenotype. Through ROSI, the Adad2^Mut/Mut mice could produce pups with comparable embryonic development (46.7% in Adad2^Mut/Mut versus 50% in WT) and birth rates (21.45 ± 10.43% in Adad2^Mut/Mut versus 27.5 ± 3.536% in WT, P = 0.5044) to WT mice. The Adad2^WT/Mut progeny from ROSI (17 pups in total via three ROSI replicates) did not show overt developmental defects and had normal fertility.

LARGE SCALE DATA

N/A.

LIMITATIONS REASONS FOR CAUTION

This is a preliminary report suggesting that ROSI can be an effective treatment for infertile Adad2^Mut/Mut mice. Further assisted reproductive attempts need to be carefully examined in humans during clinical trials.

WIDER IMPLICATIONS OF THE FINDINGS

Our work provides functional evidence that mutations in the ADAD2 gene are deleterious and cause consistent spermiogenic defects in both humans and mice. In addition, preliminary results show that ROSI can help Adad2^Mut/Mut to produce biological progeny. These findings provide valuable clues for genetic counselling on the ADAD2 mutants-associated infertility in human males.

STUDY FUNDING/COMPETING INTERESTS

This work was supported by the National Natural Science Foundation of China (32000587, U21A20204, and 32061143006), and the National Key Research and Developmental Program of China (2019YFA0802600 and 2021YFC2700202). This work was also supported by Institute of Health and Medicine, Hefei Comprehensive National Science Center, Hefei, China. The authors declare no competing interests.

Transcriptome and DNA Methylation Profiles of Mouse Fetus and Placenta Generated by Round Spermatid Injection

Low birth efficiency and developmental abnormalities in embryos derived using round spermatid injection (ROSI) limit the clinical application of this method. Further, the underlying molecular mechanisms remain elusive and warrant further in-depth study. In this study, the embryonic day (E) 11.5 mouse fetuses and corresponding placentas derived upon using ROSI, intracytoplasmic sperm injection (ICSI), and natural in vivo fertilized (control) embryos were collected. Transcriptome and DNA methylation profiles were analyzed and compared using RNA-sequencing (RNA-seq) and whole-genome bisulfite sequencing, respectively. RNA-seq results revealed similar gene expression profiles in the ROSI, ICSI, and control fetuses and placentas. Compared with the other two groups, seven differentially expressed genes (DEGs) were identified in ROSI fetuses, and ten DEGs were identified in the corresponding placentas. However, no differences in CpG methylation were observed in fetuses and placentas from the three groups. Imprinting control region methylation and imprinted gene expression were the same between the three fetus and placenta groups. Although 49 repetitive DNA sequences (RS) were abnormally activated in ROSI fetuses, RS DNA methylation did not differ between the three groups. Interestingly, abnormal hypermethylation in promoter regions and low expression of Fggy and Rec8 were correlated with a crown-rump length less than 6 mm in one ROSI fetus. Our study demonstrates that the transcriptome and DNA methylation in ROSI-derived E11.5 mouse fetuses and placentas were comparable with those in the other two groups. However, some abnormally expressed genes in the ROSI fetus and placenta warrant further investigation to elucidate their effect on the development of ROSI-derived embryos.

Ninety babies born after round spermatid injection into oocytes: survey of their development from fertilization to 2 years of age

OBJECTIVE

To compare physical and cognitive development of babies born after round spermatid injection (ROSI) with those born after natural conception.

DESIGN

Comparison of efficiencies of ROSI and ICSI using testicular spermatozoa, performed in the St. Mother Clinic. Physical and cognitive development of ROSI babies recorded by parents in the government-issued Mother-Child Handbook was checked and verified by attending pediatricians. Data included baby's weight gain and response to parents' voice/gesture.

SETTING

Assisted reproduction technology practice.

PATIENT(S)

A total of 721 men participated in ROSI; 90 ROSI babies were followed for 2 years for their physical and cognitive development. Control subjects were 1,818 naturally born babies.

INTERVENTION(S)

Surgical retrieval of spermatogenic cells from testes; selection and injection of round spermatids into oocytes; oocyte activation, in vitro culture of fertilized eggs, and embryo transfer to mothers.

MAIN OUTCOME MEASURE(S)

Physical and cognitive development of ROSI babies (e.g., body weight increase, response to parents, and understanding and speaking simple language) compared with naturally born babies.

RESULT(S)

Of 90 ROSI babies, three had congenital aberrations at birth, which corrected spontaneously (ventricular septa) or after surgery (cleft lip and omphalocele). Physical and cognitive development of ROSI babies was similar to those of naturally born babies.

CONCLUSION(S)

There were no significant differences between ROSI and naturally conceived babies in either physical or cognitive development during the first 2 years after birth.

CLINICAL TRIAL REGISTRATION NUMBER

UMIN Clinical Trials Registry UMIN000006117.

Development of blastocyst-stage embryos after round spermatid injection in patients with complete spermiogenesis failure

PURPOSE

Our purpose was to evaluate the progression of embryos derived from round spermatid injection to the blastocyst stage and compare the results with those obtained by the use of testicular or epididymal spermatozoa.

METHODS

Thirty-eight patients with azoospermia enrolled in this study. In 29 patients with obstructive or nonobstructive azoospermia, spermatozoa were recovered from epididymis or testis. In the remaining nine cases with nonobstructive azoospermia, only round spermatids were found in seven, whereas in two of the patients, there were no elongated or round spermatids. Six of these cases underwent round spermatid injection.

RESULTS

Twenty-one of 29 patients with injection of spermatozoa underwent embryo transfer on day 3, and 10 pregnancies (47.6%) were obtained. In eight cycles, embryos were further cultured for delayed transfer. In six cases undergoing round spermatid injection, no transfer was performed on day 3 and extended culture with delayed embryo transfer was applied. The mean number of fertilized oocytes and mean number of embryos on day 3 and also the fertilization rate and mean number of good-quality embryos on day 3, mainly grade 1 or 2, were statistically significantly higher in the spermatozoa group than the round spermatid injection group. Compared to the spermatozoa group, the number of arrested embryos was significantly higher and the number of blastocyst-stage embryos and number of good-quality blastocysts were significantly lower in the spermatid injection group. No blastocysts developed in two spermatid cycles and embryo transfer was not possible, and in the remaining four cycles, after at least one blastocyst transfer, no pregnancies were achieved. However, in eight cycles with extended culture in the spermatozoa group, embryo transfers were achieved in all and three pregnancies, for a pregnancy rate of 37.5%, were obtained after blastocyst transfer.

CONCLUSIONS

Our preliminary results showed that round spermatid injection was associated with a significantly lower fertilization and embryo development rate and a significantly higher developmental arrest rate compared with the injection of spermatozoa. Extended culture and delayed embryo transfer did not improve the clinical outcome after round spermatid injection, and these results suggested a developmental failure in embryos preventing successful implantation after round spermatid injection.

Bypassing spermiogenesis for several generations does not have detrimental consequences on the fertility and neurobehavior of offspring: a study using the mouse

PURPOSE

This study was conducted to determine whether the omission of spermiogenesis and all prefertilization events for five generations in mice affects the fertility or behavior of offspring.

METHODS

Fifth-generation hybrid (C57BL/6 x DBA/2) mice were produced using round spermatid injection (ROSI). Control groups consisted of mice born after natural mating with and without sham operation. The growth, fertility, and behavior of offspring were compared. Behavior tests conducted assessed elementary reasoning (Krushinsky test), emotionality (Mouse Defense Test Battery), and spatial learning and memory (Morris water maze).

RESULTS

There were no significant differences in the growth and fertility of fifth-generation ROSI mice compared to natural fertilization mice. We also found no evidence of significant learning or behavioral deficits of the fifth-generation ROSI mice.

CONCLUSIONS

In this study, we found no evidence that bypassing the natural biological processes involved in spermiogenesis produces adverse effects on the growth, fertility, or behavior of mouse offspring.