Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Yoshimura et al1 reported that although exposure to dioxin (TCDD) (after the explosion at Seveso) was associated with a subsequent significant and substantial drop in offspring sex ratio (proportion male),2 there was no obvious similarity in the offspring sex ratios after accidental ingestion of polychlorinated biphenyls (PCBs) and dibenzofurans (PCDFs) in Yusho, Japan,1 and Yucheng, Taiwan.3 PCBs and PCDFs are toxicologically similar to (although less potent than) TCDD. The authors wrote that in Taiwan, despite exposure to chemicals similar to TCDD sufficient to produce obvious clinical disease, the sex ratio was not altered. They inferred that sex ratio is unlikely to be a sensitive indicator of exposures to chemicals such as PCBs, PCDFs, and dioxins. The argument is not decisive and Yoshimura et al1 called for further research.
I had predicted4 Mocarelli's low paternal offspring sex ratio on the basis of the known endocrine effects of dioxin on exposed men.5 The basis of my prediction was a hormonal hypothesis of mammalian sex determination, the evidence for which was later summarised.6 So I suggest a direction for the further research suggested by Yoshimura et al.1
My hypothesis proposes that the sexes of mammalian (including human) offspring are partially dependent on the hormone concentrations of both parents around the time of conception. It is thought that high levels of estrogens and testosterone are followed disproportionately often by sons; and high levels of gonadotropins and progesterone by daughters. This being so, a contaminant released into the atmosphere or water sources may in principle have opposing tendencies on the offspring sex ratios of exposed mothers and fathers and thus—without further examination—remain undetected. As already suggested, the known effect of dioxin is to lower men's testosterone/gonadotropin ratio, thus predisposing them to sire daughters. By contrast, the effect of this class of chemicals on women is (under some circumstances) estrogensic,7–11 predisposing them to produce sons. If this were correct, the point would be revealed by examining the sex ratios of offspring of exposed men mated to unexposed women, and vice versa, as was done by Mocarelli et al.2 The data of these authors are suggestive but not decisive in this context. The offspring sex ratio of their exposed mothers married to unexposed fathers was higher (but not significantly higher) than the expected overall population sex ratio in Italy at that time. I suggest that workers try to ascertain data of this sort from people exposed in Yucheng and Yusho; and in Vietnam. Lastly it would be useful to have experimental animal data on the point.
We thank James for his comments1 about our paper on sex ratio in offspring of affected parents of Yusho.2 He suggests ascertaining information on exposure of parents to examine his hypothesis that the sexes of offspring partially depend on the hormone concentrations of both parents around the time of conception. Three combinations of parent pairs according to exposure could be possible in the Yusho incident; a pair of an exposed father and a non-exposed mother, a pair of a non-exposed father and an exposed mother, and a pair of both parents exposed. He suggested that the first two types of pair should be informative to his hypothesis, because the first pair would be more likely to have daughters and the second likely to have sons.
In the Yusho incident, affected people were likely to be clustered into families, because the contaminated rice oil was mainly distributed as family cooking oil; therefore, for those who were married at the time of the incident, both husband and wife were likely to be exposed to the dioxin-like compounds. Affected single people, who were living with their family at the time of the incident and were subsequently married, can give some information towards his hypothesis, but cannot be followed up until they have married and had children. This was not done because the follow up time was too short.
We agree with his suggestion that follow up must be extended to ascertain the sex of offspring born to the affected single people at the time of the incident. We are making an effort to conduct such a study; but we are facing difficulties from increasing socioethical concerns of the public—that is, protection of privacy and informed consent for epidemiological studies in recent years in Japan.3