CYP1A1: Ethnic and Population Differences in Allelic Frequencies and Interpretation of Bone Biology Studies

To the Editor: We read with great interest the report by Napoli et al.(1 published in JBMR. In this report, the authors noted that women carrying an A allele for the C4887A polymorphism of the CYP1A1 gene had increased estrogen metabolism and that this was associated with increased bone resorption and a lower femoral BMD. They concluded that the C4887A polymorphism of the CYP1A1 gene might represent a genetic risk factor for osteoporosis. In our opinion, this finding may have been confounded by erroneous classification of women based on their genotype or the result of a strong bias in the sampled women. Indeed, the high allelic frequency of the rare allele observed in their small sample of women (n =156) is intriguing. They report a frequency close to 10% in white women. This is much higher than what was previously reported; Cascorbi et al.(2) reported a frequency of 2.95% in a group of 880 unrelated German individuals, and Zhang et al.(3 reported a frequency of 4.5% among 406 controls in a case/controls breast cancer study among white women in Connecticut. We also analyzed the CYP1A1 gene in our large sample of white pre- and postmenopausal women for two polymorphisms, C4887A and A4889G, and no association with BMD measures was observed. Furthermore, among 1791 unrelated French-Canadian women genotyped for C4887A, we found a 3.5% allelic frequency in agreement with the expected Hardy-Weinberg distribution (1668 CC, 121 CA, and 2 AA). The 95% confidence limits of this frequency are 0.0289–0.0411. In our sample, genotyping was performed by allele-specific PCR followed by SybrGreen detection of amplicons. To validate the genotype, a second reaction was performed among 540 samples with primers designed in the other direction (other DNA strand) taking into account polymorphisms other than the one analyzed (three known polymorphisms are very close to C4887A). In this analysis, we obtained 99.7% concordance between the two reactions; the two discordant samples were analyzed a third time in duplicate to obtain a final result. To obtain a population sample with characteristics similar to that of Napoli et al.,(1) we extracted from our 1791 women those at menopause, not taking hormone therapy, and not currently smoking. This generated a subsample of 191 women. No difference was observed in BMD at the spine or femoral neck between women grouped as CC and women carrying an A allele; no AA individual was observed (Table 1). In this subgroup, the allelic frequency was 3.4%, with 95% confidence limits of 0.0157–0.0524. Power calculations could not be performed because the authors did not provide data on the distribution of dependent variables. The authors mentioned that 26 women in this study were participants in a previous related study in which they measured estrogen metabolite and BMD. No description of these women and how they were chosen was provided, so we cannot eliminate the possibility of a bias introduced at this level. Although a higher rate of estrogen catabolism and consequently lower free estradiol might indeed correlate with lower BMD, we doubt that this is caused by this single CYP1A1 gene polymorphism in the general population. Furthermore, to our knowledge, no functional effect has been reported for this polymorphism in CYP1A1 gene.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.