375 UNEXPECTED SEVERE ABNORMALITIES IN MOUSE ROSI OFFSPRING

Abstract

Live offspring resulting from round spermatid injection (ROSI) was first accomplished in the mouse, but similar success has been obtained in rat, hamster, rabbit, mastomys, pig, monkey, and human. ROSI has received clinical attention because some infertile men have no spermatozoa or just a very few in their testes, and these are difficult to harvest and are frequently dead and deformed. Although some clinicians were able to generate healthy children by ROSI, others could not (reviewed in Yanagimachi 2004 Reprod. Biomed. Online 9). The clinical value of ROSI has been widely debated. It remains unclear if post-meiotic and pre-fertilization modifications of sperm cells are necessary to ensure normal development. In order to answer this question, we decided to study and compare mouse offspring generated by ROSI and intracytoplasmic sperm injection (ICSI). ROSI and ICSI with fresh sperm cells were carried out in the B6D2 mouse strain as described (Marh et al. 2003 Biol. Reprod. 69, 169–176; Moreira et al. 2005 Hum. Reprod. 20, 3313–3317). In vitro-produced embryos were transferred at the 2-cell stage into Day 1 pseudopregnant females. As shown in Table 1 oocyte survival after injection was significantly higher (z-test, P < 0.05) with ICSI (91%) than with ROSI (68%). The proportion of live offspring obtained by ICSI was also significantly higher (26% vs. 6%; z-test, P < 0.05). Moreover, fertilization with spermatozoa produced healthy offspring more efficiently than with round spermatids. Out of 30 live offspring generated by ROSI, 6 (20%) presented severe abnormalities during their first 6–8 weeks of age. One ROSI animal presented an abnormally swollen skull (hydroencephaly) with a very thin and soft cranial wall. Another developed a subcutaneous engrossment of the forehead, producing a crest-like appearance. Three others presented deviations in their vertebral columns (hyperkyphosis and scoliosis), and recently a testicular tumor was detected in another animal. These types of malformations were not observed in the control offspring. In our experience, very rarely are they observed after ICSI or in naturally mated animals. To our knowledge, and although the risks of the ROSI procedure have been extensively highlighted in human and other species, the phenotypic abnormalities observed in this study have never been reported. Presently, we keep monitoring these animals as they age, as part of an ambitious plan that is also intended to characterize and understand the origin of the possible phenotypic consequences of the ROSI procedure.

Table 1.  In vitro development and development to term of B6D2 mouse embryos generated by ROSI or ICSI