Scientific Correspondence

Role of the cholinergic muscarinic 2 receptor (CHRM2) gene in cognition

SIR – Heritability estimates for intelligence quotient (IQ) range from 0.50 to 0.80.1 This makes IQ a suitable target for attempts to identify the specific genes involved. This first such gene to be identified was by Chorney et al.2 They reported an association between the insulin-like growth factor 2 receptor (IGF2R) gene and IQ. Cholinergic pathways have been widely implicated in cognition and memory,3,4 making the respective genes excellent candidate markers for cognitive abilities. Identification of a possible role of cholinergic receptor genes in humans has been hampered by the lack of reported polymorphisms. We identified a common A→T 1890 polymorphism in the 3′UTR of the CHRM2 gene. To determine if it was associated with IQ, we examined 358 adult males and 470 adult females for a total of 828 adults.

The subjects were the parents of twins from the Minnesota Twin and Family Study, a long-term study of the genetics and environmental factors in substance abuse.5 All subjects in the CHRM2 study were of Caucasian ancestry. All were given the Wechsler Adult Intelligence Scale-Revised (Vocabulary, Information, Block Design, and Picture Arrangement) test. The study was approved by the internal review boards of both the University of Minnesota and the City of Hope Medical Center and all subjects gave written informed consent.

Using the SSCP technique,6 we identified a common single nucleotide polymorphism, A 1890→T in the 3′UTR of the CHRM2 gene based on accession No. M16404. The upstream primer was 5′-ACAAAA-CGTGCAATTCAGGAG-3′. The downstream primer was 5′-CAGAGACTGATAAAATTGTAG-3′. Details of the PCR reaction can be obtained from the DEC. This produced a 208 bp product that was digested with Dpn II restriction endonuclease.

The IQ and years of education by genotype for the fathers, mothers and both, and number of subjects, F-ratio, P and linear P values are shown in Figure 1. To assess which variable was more closely associated with the CHRM2 gene, we performed a MONOVA using both total IQ and years of education as the dependent variables and the CHRM2 gene as the independent variable for the total set. The total MANOVA (Wilks) was significant at P≤0.009. The F-ratio for IQ was 4.12, P≤0.017, and for years of education the F-ratio was 5.86, P≤0.003.

Figure 1
Figure 1

Association of the CHRM2 genotypes with Total IQ (top) and years education (bottom), for fathers (left panel), mothers (middle panel) and both (right panel).

Long-term potentiation has been implicated in memory and cognition and the cholinergic muscarinic M2 receptor plays a role in long-term potentiation in the striatum.7 This suggests that genetic variants at the CHRM2 gene might be associated with different aspects of cognition. In humans, two variables that correlate with cognitive skills are IQ and years of education. The present studies showed a significant linear increase in IQ across the three genotypes of the CHRM2 gene. In males the spread was 4.17 IQ units and for females it was 3.31 IQ units, and 4.06 for the total group. There was also a significant progressive increase in years of education across the three genotypes. This would be expected since there is a high correlation between IQ and years of education. By regression analysis the CHRM2 gene accounted for approximately 1% of the variance of the IQ scores and years of education. While the percent of the variance of IQ attributable to the CHRM2 gene of 1% may seem small, it is very likely that IQ is a true polygenic trait due to the additive effect of many genes, each with a small effect. In our studies of over 200 phenotype–genotype correlations with behovioral traits we have found that even when the association is significant, in the majority of cases the percent of the variance attributable to each gene is less than 3% and averages 1.2%.8

We have replicated these findings using a quantitative TDT method developed by Abecasis et al9 in 230 parent-child trios from the MTFS. While a marginally significant association was found between CHRM2 and total IQ, after stratifying parental origin of transmission, there was a highly significant association for paternal transmission (P=0.007).

Although in need of replication, we believe these preliminary results are consistent with a role of the CHRM2 gene in cognitive processes in humans, as assessed by both total IQ and years of education.



  1. 1.

    , . Science 1981; 212: 1055–1059.

  2. 2.

    et al. Psychol Sci 1998; 9: 159–166.

  3. 3.

    , . Curr Opin Neurobiol 1999; 9: 178–183.

  4. 4.

    , . Annu Rev Psychol 1997; 48: 649–684.

  5. 5.

    et al. Development and Psychopathology 1999; 11: 869–900.

  6. 6.

    et al. Proc Natl Acad Sci USA 1989; 86: 2766–2770.

  7. 7.

    et al. Eur J Neurosci 1998; 10: 3020–3023.

  8. 8.

    et al. Clin Genet 2000; 57: 178–196.

  9. 9.

    et al. Am J Hum Genet 2000; 66: 279–292.

Download references

Author information


  1. Department of Medical Genetics, City of Hope Medical Center, Duarte, CA 91010, USA

    • D E Comings
    • , S Wu
    •  & M Rostamkhani
  2. Department of Psychology, University of Minnesóta, Minneapolis, MN, USA

    • M McGue
    •  & W G lacono
  3. Department of Genomics, MGI Applied Genomics, Long Beach, CA, USA

    • J P MacMurray
  4. Department of Biostatistics, City of Hope Medical Center, Duarte, CA 91010, USA

    • L S-C Cheng


  1. Search for D E Comings in:

  2. Search for S Wu in:

  3. Search for M Rostamkhani in:

  4. Search for M McGue in:

  5. Search for W G lacono in:

  6. Search for L S-C Cheng in:

  7. Search for J P MacMurray in:

Corresponding author

Correspondence to D E Comings.