McKeown-Eyssen, G, C Baines, DEC Cole, N Riley, RF Tyndale, L Marshall, V Jazmaji. 2004. Case-control study of genotypes in multiple chemical sensitivity: CYP2D6, NAT1, NAT2, PON1, PON2, and MTHFR. International Journal of Epidemiology 33 (5): 1-8.

Multiple chemical sensitivity (MCS) has been a controversial condition, with widely varying opinions about its characteristics and causes, indeed about its very existence as a disease with physical origins. In this epidemiological study, McKeown-Eyssen et al. report several genetic differences between cases and controls in genes involved in detoxifying contaminants.

Not only do differences exist, but they were discovered in genes that are important to detoxifying toxic compounds, with MCS patients more likely to have forms of the genes that would impair their ability to detoxify them. The most striking result emerged when they looked at two genes simultaneously: Women with MCS were over 18 times more likely than controls to have a specific combination of forms of two separate genes. Their data are the first to demonstrate such genetic differences, providing strong evidence for the physical origin of this syndrome.

What did they do?

Mckeown-Eyssen et al. conducted a case-control study of Toronto women ages 30-64. 223 cases with MCS were recruited from among 439 candidates identified by an initial survey. These were matched with 194 controls who were drawn from family practices and who did not show signs of MCS.

The team used several criteria to identify MCS cases:

  • symptoms linked to low-level exposure, which disappear with removal of the agent;
  • chronic symptoms

and either:

  • a stronger sense of smell than others or any two of the following: feeling dull or groggy; feeling spacey; difficulty concentrating.
 

Background on MCS

Multiple Chemical Sensitivity is a chronic, recurring disease that can occur in response to low levels of exposure to multiple unrelated chemicals, with symptoms in multiple organ systems. People with MCS are generally hypersensitive to low levels of chemicals found in everyday substances, such as household cleaning agents, pesticides, fresh paint, new carpeting, synthetic building materials, newsprint, and perfume. MCS has been a controversial and poorly understood condition. Not all in the medical community even agree that MCS exists. Some consider MCS to be a psychiatric illness or allergic condition.

Persons with MCS report a range of symptoms that often include headaches, rashes, asthma, depression, muscle and joint aches, fatigue, memory loss, and confusion. Their symptoms improve or resolve when they avoid exposure to the offending chemicals/products.

To avoid confounding the analysis with ethnic variations in genotype, only Caucasian women were included in the study.

To find participants, 4,126 were sent a survey asking about symptoms relevant to MCS. Out of these, 493 potential cases and 481 potential controls were identified (i.e., women who were 30-64, not pregnant, indicated their willingness to participate in future research studies, and met the criteria for cases and controls). Ultimately, 203 cases and 162 controls participated in the research.

The research team collected blood samples from all participants for genetic analysis. They isolated genomic DNA from these samples, and identified specific polymorphisms to determine if certain polymorphisms occurred more frequently in cases compared to controls for several different genes that play key roles in how the body responds to chemicals:

  • CYP2D6 (Cytochrome P450 2D6) encodes for enzymes that metabolize many toxic chemicals as well as therapeutic drugs (e.g., antidepressants, antipsychotics, beta blockers). Cytochrome P450 enzymes are a large family of enzymes that comprise an important part of the body’s tools to get rid of potentially harmful substances by making them more water-soluble. This study looked at four alleles of CYP2D6: *3, *4 and *5, all of which make CYP2D6 non-functional, and *1, the active form of CYP 2D6.
  • NATI and NAT2 (N-acetyltransferases) encode for enzymes responsible for metabolizing (by either slow or rapid acetylation) a number of different drugs and toxic chemicals, including aromatic amines, a class of compounds widely used in industry, many of which cause cancer. They are able to bioactivate several known carcinogens .
  • PON (paraoxonase) genes code for protein products that react with toxins such as pesticides and nerve agents and have been linked to Gulf War Syndrome. This study looked at PON1-55, PON1-192, and PON2-148.
  • MTHFR (methylenetetrahyrofolate reductase) encodes for a key enzyme involved in metabolism of some B vitamins (including B12 and folate). Women with the C677T polymorphism of the MTHFR gene, and whose diet is low in folate, have a higher risk of having babies with neural tube malformations. The MTHFR-C677T gene was studied since impaired vitamin B12 metabolism can contribute to neurological symptoms, and because the authors had previously observed that serum vitamin B12 levels were higher in cases than in controls.

Allele frequencies and genotype distributions

To compare the genetic make-up of cases vs. controls, McKeown-Eyssen et al. used two classic measurements developed by scientists studying how groups differ genetically, allele frequency and genotype distribution.

Each gene can have multiple forms, determined by the sequence of nucleotides making up each gene's DNA. Different forms are called alleles. Differences in nucleotide sequence are translated during gene expression to differences in molecular structure of the proteins made by that gene. The protein products of different alleles can behave very differently. Hence allele differences can be extremely important in determining whether the protein can do its job, in this case, detoxification. Alleles examined in the current study contribute to differences among people in their ability to detoxify contaminants.

One of the measurements the research team used was allele frequency. Out of the total number of copies of that gene in the population being studied, how many were one form vs. another. Each allele will have a different label, for example, *1 or *2, which identify for the geneticists the precise nucleotide sequence within that allele.

A second measurement used was genotypic distribution. Each individual has (normally) two copies of each gene, one on each pair of paired chromosomes. Some possess two copies of the same allele, e.g.,*1/*1 , whereas others possess combinations, e.g., *1/*4. The genotypic distribution is total pattern across the population of the proportion of individuals that possesses those different combinations, for example, 20% might be *1/*1, 60% might be *1/*4 and another 20% might be *4/*4

 

The research team then carried out statistical analyses designed to identify genetic differences between cases and controls. Specifically, the team looked for differences in allele frequencies and genotype distributions for the set of genes tested.

What did they find?

Mckeown-Eyssan et al. found several significant genetic differences between women with MCS and those without symptoms.

Testing for differences between allele frequencies in specific genes, they found significant differences for the CYP2D6 gene (p = 0.02) and a marginally significant difference for the NAT2 gene (p =0.07)

Genotype distributions between cases and controls were significantly different for both the CYP2D6 (p=0.02) and NAT2 p=0.03) genes.

  • Women who were homozygous for the active form of the CYP2D6 gene (the *1 allele, which codes for increased rates of CYP2D6 activity) had over three times the risk of having MCS, with an odds ratios of 3.36 and 95% confidence interval running from 1.33 to 8.50. (p=0.01), compared to women who were homozygous for the inactive form of CYP2D6 (e.g., *3/*5, *4/*4, *4/*5, *5/*5).
  • Women who were heterozygous for the active form of the CYP2D6 gene (the *1 allele) also showed an increased risk of MCS, although it was just shy of statistical significance. This pattern--significant odds ratios for the homozygous active, marginally-significant odds ratios for the heterozygous--suggests a “gene-dose effect”, with the risk of having MCS increasing as the number of active CYP2D6 alleles .
  • Women who were homozygous for the rapid form of the NAT2 gene (the *4 allele), known as “rapid acetylators,” had over four times the risk, with an odds ratio of 4.14 (95% confidence interval 1.36-12.64, p=0.01) of having MCS, compared to “slow acetylators” who lack the *4 allele of the NAT2 gene. Women who were intermediate (had only one *4 allele (were heterozygous)) did not show an increased risk.

 

The most striking finding was the gene-gene interaction between CYP2D6 and NAT2. Women with genes encoding for rapid metabolism by both enzymes (i.e., CYP 2D6 homozygous active and NAT2 'rapid acetylator' genotypes) were over 18 times more likely to have MCS compared to women with the slow metabolic forms (odds ratio = 18.7 , confidence interval 2.9 -122.5), p=0.02).

Women who were heterozygous for the PONI-55 and PONI –192 genes also were more likely to be cases than women who were not (for PONI-55, odds ratio = 2.05, confidence interval 1.04-4.05;, p=0.04. For PONI-192, odds ratio = 1.57, confidence interval 1.01-2.45; p=0.04).

There was no relationship between MTHFR-C677T and case-control status.

McKeown-Eyssen et al. found no statistical link between potential cofounders (such as cigarette smoking, birthplace and the use of vitamin supplements) and genotype.

What does it mean?

This groundbreaking study by McKeown-Eyssen et al. is the first to show genetic differences in enzymes important to detoxification of contaminants comparing people with MCS to those who don’t have the condition. It thus provides significant genetic evidence for the existence of MCS as a physically-based phenomenon, not something existing 'only in the mind of the patient,' as skeptics of the condition have asserted.

The enzymes studied are only a few within the many important biochemical pathways used to detoxify chemicals that are known to vary genetically. For example, while this study examined 4 alleles of the 2yP2D6 gene, 46 different alleles have been identified. Their impacts differ, with some increasing, others decreasing and still others having no effect on the body’s ability to detoxify a particular chemical. And the 2YP2D6 gene is but one of many genes within the cytochrome P450 family that are important to metabolizing toxic compounds. It is reasonable to expect that one or more of these genetically variable detoxification pathways could also be important in generating a genetic susceptibility to MCS.

Stepping back from the details of the genetic analysis, these results are important in establishing genetic susceptibilities for MCS, but they fall short of proving that MCS is caused by exposure to contaminants. The pathways affected by the enzymes studied are clearly important to detoxifying external contaminants, but they are also important to regulating the metabolism of compounds that occur naturally in the body. It will require additional layers of research to test these different possibilities.

While McKeown-Eyssen et al. offer no final word on causation, their research should help shift the debate away from "does this condition exist?" to "what are the causal pathways that lead women with a specific genotype to be at much greater risk to MCS?" The answer lies not in the mind, as MCS critics have claimed, but rather in the interaction between genes, the enzymes they produce, and the compounds those enzymes detoxify.