DNA sequencing of the sperm genome retrieved at different levels of the male genital tract

Abstract

To test whether sampling epididymal and testicular tissues yields spermatozoa with higher genomic integrity than those found in the ejaculate. Nine chromosome FISH was performed on 2 donor controls, ejaculates of 67 men, and surgical specimens from 9 azoospermic men. DNA sequencing was used to analyze ejaculates and surgical samples from 20 men. A combined assessment was performed on cryptozoospermic men with high DNA fragmentation in their ejaculates. ICSI pregnancy outcomes were recorded and compared. An analysis was also carried out to compare obstructive (OA) and non-obstructive (NOA) men. Consenting men undergoing treatment for infertility provided specimens. FISH was performed on at least 1,000 spermatozoa, with a threshold of >1.6% with 2-3% FISH error. DNA was extracted and amplified from a comparable number of spermatozoa by PCR-based random hexamer amplification (DNA concentration, 610±102 ng/ul; quality, 1.7±0.1nm). Through NGS, duplications and deletions by copy number variants (CNVs) were calculated using CASAVA and VarScan2 software programs. A total of 76 couples were included in our study (maternal age, 35.8±4 yrs; paternal age, 39.4±8 yrs). Aneuploidy rates as determined by FISH in the study group were 3.6% for the ejaculated, 1.2% for the epididymal, and 1.1% for the testicular spermatozoa, while it was only 0.9% for the donor control specimens. NGS performed in the study group yielded aneuploidy rates of 11.1% for ejaculated, 1.8% for epididymal, and 1.9% for testicular spermatozoa, whereas it was only 1.2% for the control (P<0.0001). ICSI pregnancy rates were 47.2% and 55.5% for the ejaculated and surgically retrieved specimens, respectively. Aneuploidy assessment to characterize azoospermic patients did not show any difference between OA and NOA men. However, when we analyzed the mutations from those who sustained a pregnancy, the infertile group had a higher incidence of mutations (P<0.05) for both OA and NOA men. In a sub-analysis, specimens were concurrently collected from the same cryptozoospermic individuals. Paired assessment showed a sperm chromatin fragmentation rate of 20% for ejaculated specimens but only 8% for testicular samples. In addition, aneuploidy rates were 2.8% for ejaculated and 1.2% for testicular specimens, while NGS showed aneuploidy rates of 8.4% for ejaculated and 1.3% for testicular spermatozoa (P=0.02). Interestingly, in these individuals, the clinical pregnancy rate was 0% with the ejaculated specimens but 100% with the testicular spermatozoa. DNA sequencing of the male genome proved superior to standard FISH by showing higher aneuploidy rates in the ejaculated than in the surgically retrieved specimens. This was confirmed in the concurrent analysis of the same individuals. In azoospermic men, whether OA or NOA, a higher incidence of gene mutations, rather than aneuploidy, was responsible for the impaired reproductive competence of the male gamete.