Detoxification genes and toxic metals: what gene-environment research can and cannot show
Two children can face the same environmental exposure and carry very different internal doses. Detoxification genes are part of why, and a careful reason not to overclaim.
By Manouchehr Hessabi, MD, MPH
Two children can grow up in the same neighborhood, drink from the same tap, and eat from the same garden, and still end up with measurably different amounts of a toxic metal in their blood. Part of that difference is exposure. Part of it is chemistry the children were born with. Understanding the second part, without overstating it, is one of the harder and more interesting problems in environmental-health research.
This is an explainer about how genes that handle toxicants interact with the exposures themselves. It is educational and not a substitute for personal medical advice, and it does not recommend any test, supplement, or course of action.
The same exposure is not the same internal dose
Environmental-health research often starts with a tidy mental model: more exposure, more harm. That model is useful, but it skips a step. Between the contaminant in the water and the effect in the body sits the body's own processing system, and that system is not identical from one person to the next.
The National Institute of Environmental Health Sciences frames this as gene-environment interaction: subtle differences in a person's genes can cause them to respond differently to the same exposure, so some people are affected while others, similarly exposed, are not (NIEHS). For a toxic metal, that often means two people with the same external exposure carry different internal doses, because their bodies clear the substance at different rates.
When we measure a metal in a child's blood, we are measuring the result of that whole process, not just the exposure. Which is exactly why the genes that govern clearance are worth understanding.
What glutathione S-transferases actually do
Among the body's clearance systems is a family of enzymes called glutathione S-transferases, usually abbreviated GST. They are phase II detoxification enzymes, meaning they act in the second stage of how the body neutralizes foreign or reactive compounds. Their job is conjugation: they attach a small molecule called glutathione to a reactive, potentially harmful compound, which makes it more water-soluble and easier to excrete in urine or bile.
Several members of this family come up repeatedly in environmental-health studies, including GSTM1, GSTT1, and GSTP1. Two of them, GSTM1 and GSTT1, are notable for a particular quirk of human genetics. A large fraction of people carry a "null" genotype, meaning they inherited deletions of the gene from both parents and produce none of that particular enzyme. A genotype is simply the specific genetic variant a person carries, and a polymorphism is a spot in the genome that commonly differs between people. The GSTM1 and GSTT1 null genotypes are among the most studied polymorphisms in all of toxicology, because they offer a natural experiment: some people have the enzyme, some do not.
The intuition is straightforward. If an enzyme helps clear a toxicant, a person who makes none of it might clear that toxicant less efficiently and could carry a higher internal dose for the same exposure. That intuition is the starting point for a great deal of research. It is also where the science gets more careful than the intuition.
Why "null" does not mean defenseless
It would be convenient if a missing gene meant a missing function, but biology is rarely that clean. The GST family is large and overlapping, and other members can partly cover for one that is absent.
One illustrative study found that GSTM2, a sibling gene, can functionally compensate for the GSTM1 null genotype, taking on part of the same detoxification work (Bhattacharjee and colleagues, 2013). The practical lesson is that carrying a null genotype shifts the odds rather than flipping a switch. A person without GSTM1 is not chemically defenseless, and a person with it is not fully protected.
Bhattacharjee P, Paul S, Banerjee M, Patra D, Banerjee P, Ghoshal N, Bandyopadhyay A, Giri AK (2013). Functional compensation of glutathione S-transferase M1 (GSTM1) null by another GST superfamily member, GSTM2. Scientific Reports.DOI: 10.1038/srep02704
This is the first reason to be cautious about single-gene stories. The body has redundancy, and any one polymorphism explains only a slice of the variation between people.
Gene meets environment: a concrete example
The more honest framing is not "this gene determines risk" but "this gene may modify how an exposure relates to an outcome." For fifteen years, our team at UTHealth Houston has worked with Jamaican colleagues on the Epidemiological Research on Autism in Jamaica (ERAJ) program, a case-control study that recruits young children with and without autism spectrum disorder, matches them by age and sex, and measures both blood metal concentrations and GST genotypes.
One ERAJ analysis examined blood manganese, an essential nutrient that becomes neurotoxic in excess, together with the GSTP1 genotype. Among children who carried the GSTP1 Ile/Ile variant, those with higher blood manganese had roughly four times the odds of an autism diagnosis compared with those who had lower manganese, with a matched odds ratio of 3.9 and a wide confidence interval of 1.15 to 13.26 (Rahbar and colleagues, 2015). An odds ratio is a way of comparing the odds of an outcome between two groups, and a confidence interval that wide signals real uncertainty in the estimate.
Rahbar MH, Samms-Vaughan M, Ma J, Bressler J, Dickerson AS, Hessabi M, et al. (2015). Synergic effect of GSTP1 and blood manganese concentrations in Autism Spectrum Disorder. Research in Autism Spectrum Disorders.DOI: 10.1016/j.rasd.2015.08.001
A separate ERAJ analysis of blood mercury showed something equally instructive: the relationship between the same GSTP1 variant and the metal ran in different directions in children with and without autism (Rahbar and colleagues, 2021). That kind of result is a reminder that a genotype is not a simple dial turning risk up or down. Its effect can depend on the outcome, the specific metal, and the population.
Rahbar MH, Samms-Vaughan M, Saroukhani S, Bressler J, Hessabi M, et al. (2021). Associations of metabolic genes (GSTT1, GSTP1, GSTM1) and blood mercury concentrations differ in Jamaican children with and without autism spectrum disorder. International Journal of Environmental Research and Public Health.DOI: 10.3390/ijerph18041377
Notice the careful language. These are associations measured in one population, not demonstrations that manganese or mercury causes autism, and not evidence that any single genotype determines a child's future.
Why this is so easy to get wrong
Gene-environment findings are valuable and unusually fragile at the same time. A few of the reasons are worth naming, because they apply far beyond these particular studies.
The first is multiple comparisons. When researchers test many genes against many exposures and many outcomes, some pairings will look significant by chance alone. The 2015 manganese analysis explicitly noted that its p-values were not adjusted for multiple comparisons and that replication in other populations is warranted. Honest papers flag this; careful readers look for it.
The second is sample size. The manganese study rested on 100 matched pairs. That is real, hard-won data, and it is still a small foundation for a strong claim. Small studies produce wide confidence intervals and unstable estimates, which is part of why a single result, however striking, is a starting point rather than a conclusion.
The third is timing. Blood concentrations usually reflect recent exposure, not the prenatal or early-life window that often matters most for development. A measurement taken at age five is not necessarily a record of the exposure that mattered at age one.
The fourth is confounding, the situation where some third factor is tied to both the exposure and the outcome and quietly distorts the apparent link. Diet, water source, and socioeconomic conditions all travel with metal exposure, and untangling them is the daily work of environmental epidemiology rather than a detail to wave away.
What this means, and what it does not
The defensible takeaway from this body of work is modest and genuinely useful: how a person responds to an environmental exposure depends partly on inherited differences in the systems that clear toxicants, and studying that interaction sharpens our understanding of who is most vulnerable and why.
What the research does not support is the leap from a genotype to a destiny, or from an association in one matched case-control sample to a cause. It also does not point to any action a family should take. Detoxification genotypes are not a clinical test for everyday decisions, and nothing here should be read as a recommendation to seek one. The value of this science is in the population-level understanding it builds, paper by careful paper, not in a verdict about any individual child.
That restraint is not a weakness of the field. It is the discipline that lets the findings hold up. Readers who want to see how these questions are pursued in practice can review the peer-reviewed work on the publications page.