Human oocyte chromosome analyses need a standardized presentation of the results

Abstract

Studies of DNA polymorphisms in human trisomic abortions and liveborn have revealed a chromosome-specific variation in the importance of meiosis I versus meiosis II errors. As a general rule, maternal meiosis I errors predominate among almost all trisomies (Hassold and Hunt 2001). It is evident that a direct analysis of female gametes would be useful in order to confirm these tendencies and elucidate causative mechanisms. This possibility actually exists because programmes of assisted reproduction provide oocytes that failed to become fertilized in vitro. Moreover, a few untreated gametes have been donated for research. During the past 20 years, more than 10,000 oocytes have been analysed cytogenetically by the conventional technique, which consists of cell fixation followed by homogeneous staining or, more rarely, by attempts to obtain banding of the chromosomes (Rosenbusch 2006a). Additional information has been obtained from molecular cytogenetic studies (Pellestor et al. 2005) that primarily employed fluorescence in situ hybridization (FISH). The available oocytes are normally arrested at metaphase II (MII). Though their analysis will only allow conclusions on errors arising during the first meiotic division, the corresponding studies have substantially improved our understanding of the type and frequency of chromosomal abnormalities occurring during this stage of development. Of utmost importance was the demonstration of two aneuploidycausing mechanisms in human oocytes, i.e. nondisjunction (ND) of bivalents and premature centromere division at meiosis I (predivision, PD). Briefly, ND will lead to the loss or gain of whole chromosomes whereas PD can cause the loss or gain of single chromatids (Angell 1991, 1997). Unfortunately, this field of research also implies shortcomings and limitations that impair the interpretation of