Autism risk assessment on children by quantification of male sperm mosaicism

Scientists from the University of California (USA) propose a method for assessing the risk of autism by quantification of male sperm mosaicism. De novo mutations arising on the paternal chromosome make the largest known contribution to autism risk, and correlate with paternal age at the time of conception but the potential impact of assessing parental gonadal mosaicism has not been considered.

“Gonadal mosaicism” means that a mutation affects only those cells that subsequently need to differentiate into gonads that is the mutation will be contained in gametal cells and the somatic cells of the individual will be absolutely normal. Clinicians are facing an ever-increasing incidence of autism spectrum disorders (ASD) in the population, without effective strategies available to prevent disease or counsel families. Recent studies have identified gene-damaging de  novo mutations (DNMs) in at least 10–30% of simplex ASD cases1–4 , along with the realization that the number of DNMs increases as a function of paternal age at the time of conception, doubling in DNM number in an offspring every 16.5 years of the father’s age at the time of conception5,6 . A DNM, defined as a genetic variant present in an offspring but not detectable in either parent, can have any of several different origins. While classically considered as occurring in the fertilized egg at the one-cell stage, most probably occur either postzygotically in the offspring or in a parent, either in the gonads or broadly in a mosaic pattern. DNMs that occur during embryogenesis of a parent cause mosaicism in the soma, the gonads or both, and remain throughout life yet may be undetectable or barely detectable in blood. However, the balance of gonadal-specific compared to broadly distributed DNMs in the father has not been carefully assessed, and thus the role of gonadal mosaicism in DNM recurrence risk remains uncertain. Knowledge of the rates and mechanisms by which gonadal mutations arise has been advanced through assessment of multiple transmissions of DNMs within families, where approximately 1.3% of DNMs are shared by siblings.

Authors recruited eight families from our ASD cohort, where each father agreed to submit a sperm sample for sequencing. Employing 30×whole-genome sequencing (WGS) from blood they defined 912 de novo single-nucleotide variants (dSNVs) in the 14 offspring. They then isolated sperm from the ejaculates and performed 200×WGS on paternal blood and sperm cells to determine which dSNVs were detectable in sperm based on three or more mutant reads. We found 23 dSNVs that were also detected in paternal blood or sperm.

Using the ratio of mutant to reference reads in blood and sperm, they defined four dSNV classes: sperm-detectable only (SDO); sperm-detectable enriched (SDE)—for which the AF was >3-fold higher in sperm than in blood (i.e., α >3); sperm–blood equal (SBE, enrichment) and blood-detectable only (BDO). Of the 23 variants, 34.8, 30.4, 26.1 and 8.7% were SDO, SDE, SBE and BDO, respectively. Of the 23 mosaic variants, 20 resided on the paternal chromosome (40% SDO, 35.0% SDE, 25.0% SBE and 0% BDO). Thus, assessment of blood or using population risk underestimates paternal gonadal mosaicism (PGM) for most mosaic dSNVs.

“Thus, the patterns of sperm mosaicism and the resulting framework for its detection that we present have the potential to impact clinical testing in two ways. First, direct assessment of previously transmitted pathogenic variants in paternal sperm allows for the stratification of fathers with low and high recurrence risk through TAS or ddPCR analysis. Second, even without any previous risk or family history, prospective fathers who may want to know their risk of transmitting a high-impact variant to their child could undergo deep sequencing of their sperm, followed by mosaic analysis of these data”.

The results of the study are published in the Nature Medicine