The liver is an orchestra in which every cell plays its part from a shared score. Males and females have different scores: in males, the genes responsible for detoxifying foreign substances and steroid hormones sound louder, while in females, the genes for fat metabolism and bile acids do. Sex hormones and growth hormone conduct this orchestra, setting the rhythm from the first days of life.
Seven common toxicants, arsenic, lead, bisphenol A, tributyltin, the phthalate DEHP, the dioxin TCDD, and fine particulate matter PM2.5, were given to mice from two weeks before the parents mated until three weeks after birth. The animals were then raised without exposure, and at 3 weeks, 5 months, and 10 months of age the researchers looked into the liver to see how the score had changed.
Each toxicant made its own edits, and those edits in males and females hardly ever matched. In males, gene activity changed more often; in females, chemical marks on DNA did. Tributyltin, for example, produced almost 33,000 altered marks in females, and 95% of them had been erased. Arsenic, by contrast, acted consistently in both sexes: shared DNA regions became more open, while the marks became denser. But the most alarming finding was that the toxicants scrambled natural sex differences: in males, many male genes fell silent, while female genes woke up. With tributyltin exposure, the activity of 38 male genes dropped in males, and the activity of 10 female genes rose.
This means that early exposures do not merely cause temporary glitches: they rebuild the liver's sex program, and males are more vulnerable to this feminization.
It was already known that toxicants harm the liver, but no one had systematically compared different substances at the molecular level while accounting for sex and age. Now it is clear that each toxicant leaves a unique trace in gene activity, DNA packaging, and chemical marks, and that these traces in males and females barely overlap. The findings explain why men and women have different predispositions to liver disease after identical environmental exposures, and they point to DNA methylation as a potential biomarker of early exposure.
The false conductors leave, but the orchestra keeps playing from a corrupted score. And if we learn to read those edits, we will be able to predict who among us is threatened by disease long before it ever shows up.
Only two genes kept altered expression across all three life stages: Tmem200b in females exposed to PM2.5-CHI and Ces1d in males exposed to lead. The other thousands of changes came and went with age.
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.