O-090 Correcting a PLCζ mutation in the human germ line to overcome hereditary infertility

Study question Can clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 gene editing result in the correction of a single base pair substitution that causes male infertility? Summary answer CRISPR/Cas9 administration during intracytoplasmic sperm injection (ICSI) leads to correction attempts of mutant phospholipase C zeta (PLCζ), howeverc loss-of-heterozygosity (LOH). What is known already Failed fertilization after ICSI can be caused by mutations in the sperm-related oocyte factor PLCζ which can be overcome by assisted oocyte activation (AOA). In this way, children may inherit the infertility-causing mutation. Mutation transmission can be overcome through CRISPR/Cas9 delivery during ICSI. In previous studies using CRISPR/Cas9 in the human germline for mutation correction, loss-of-heterozygosity (LOH, loss of the allele of one of the parents) was observed. Two different explanations were given, namely partial or complete paternal chromosomal loss or the correction of the mutation by using the maternal wild-type allele instead of the exogeneous supplied repair template. Study design, size, duration We injected a gRNA-Cas9 protein complex to target the PLCζ mutant allele, a repair template harboring the desired nucleotide substitution and an additional synonymous variant to track template usage, together with patient’s sperm. To overcome fertilization failure, AOA was applied during ICSI. After a culture period of maximal 6 days the embryos were collected. At day 3, some embryos were dissociated in individual blastomeres. The extracted DNA was analyzed through different genetic sequencing techniques. Participants/materials, setting, methods Donated sperm of a patient experiencing complete fertilization failure after routine ICSI, harboring a heterozygous base pair substitution in PLCZ1 (c.136-1G>C), was utilized. Sperm was injected in donated in vitro matured oocytes or in vivo matured oocytes containing clusters of smooth endoplasmic reticulum. Next-generation sequencing was used to assess correction potential. Short tandem repeat (STR) and single nucleotide polymorphism (SNP) assays were used to determine whether the sperm contained the mutation and to evaluate LOH. Main results and the role of chance CRISPR/Cas9 injections had no significant impact (p > 0.05) on embryonic development. Due to the heterozygous nature of the mutation, 47% (27/58) of the embryos originated from mutated sperm injection. The CRISPR components showed a high specificity with absence of insertions/deletions in 97% of the embryos originating from wild-type sperm (n = 31). Embryos originating from mutant sperm (n = 27) fall into three categories:(1) 22% showed the untargeted mutant allele, (2) 52% showed additional mutagenesis and (3) 26% showed the wild-type allele, which could be explained by correction. Mosaicism, defined as various editing events, was present in 17% (1), 21% (2) and 71% (3) of the embryos. The low occurrence of the synonymous variant, incorporated in the repair template, suggests that the template is not used during correction attempts. In only 29% (2/7) and 14% (1/7) of the ‘corrected embryos’, respectively long (>18Mb) or medium width LOH (4Mb) was observed through STR analysis. SNP analysis in closer proximity showed in 71% (5/7) of the embryos LOH, even in the absence of LOH through STR, suggesting also the occurrence of short width LOH. These results will be studied in more detail before definitive conclusions can be made. Chromosomal LOH will be studied by ddRADseq. Limitations, reasons for caution The occurrence of mosaicism and LOH might complicate the use of traditional CRISPR/Cas9 in human embryos and should be studied in detail to draw definite conclusions on its potential future use. To this end, genomic data have been produced from both individual blastomeres and whole-embryos which will be further analyzed. Wider implications of the findings Our findings demonstrate caution to use CRISPR/Cas9 to correct mutations in the germ line. They seem to contradict other reports that show predominant lack of mosaicism and presence of long width LOH. A deeper evaluation will be undertaken to define the length and type of LOH in this study. Trial registration number Not Applicable