posted 07-30-2005 08:52 PM
The first genomwide scan for intelligence trait loci by Posthuma et al. has revealed two regions with putative evidence of linkage: 2q24.1-2q31.1 and 6.p25.3-6p22.3 (see https://genepi.qimr.edu.au/staff/nick_pdf/CV411.pdf ).
Independently, a novel phenotyping strategy has allowed Hallmayer et al. to identify a familial subtype of schizophrenia, chacterized by pervasive cognitive deficit. This subtype, which accounts for up to 50% of the sample, has a distinct genetic basis and its linkage to 6p24 has been confirmed
(see http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db= pubmed&dopt=Abstract&list_uids=16012923&query_hl=1 ). Among the gene loci in the neighborhood of 6p24 only NQO2 (6p25),
guanine monophosphate reductase (6p23, where rs1042391 should be investigated),
DTNBP1 (6p22.3) and ALDH5A1 (6p22) have common polymorphisms, which could account for contributions to interindividual differences in general cognitive ability. Plomin et al. have already been confirmed ALDH5A1 as a minor gene, responsible for differences of about 1.5 IQ points. Already in 2003 Harada et al. (see http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db= pubmed&dopt=Abstract&list_uids=14639047&query_hl=6 )
had published a study about a possible association between a common
insertion/deletion polymorphism of the NQO2 gene and schizophrenia. They investigated NQO2 because: "Glutathione S-tranferases and NADP(H):quinone oxidoreductases provide important cellular defences against the neurotoxity induced by catecholamine-derived o-quinones and oxidative stress during redox cycling." In other words, NQOs and GSTs are parts of a pool of metabolites with overlapping functions and regulations, of which about the specifity of NQO2 is little known until now. The NQO2 gene shows extensive polymorphism with two nonsynonymous coding SNPs (rs1143684 and rs2756081), where the common allele in both cases has a frequency of about 0.80.
Since 23 years I am accumulating evidence that the glutathione status (Chiueh and Rauhala 1999, Janaky et al. 1999, Dringen 2000, De Maria et al. 2003) of an individual is related to its general cognitive ability and hence its IQ, also leading to correlations of cognitive decline and biochemicalparameters during normal aging and diseases as alzheimer and parkinson. After publication of my editorial "The advent of a molecular genetics of general intelligence" (Weiss 1996, see http://www.volkmar-weiss.de/intellig.html ), in 1996 all known polymorphisms of glutathione S-transferases (GST) and glutathione peroxidase (GSHPx) were investigated (and newly discovered polymorphisms again in 2000 and 2003) for any relationship with IQ by six independent research groups and nothing was found at all. (These negative findings were not published.). Therefore, in order to explain the well-established correlations between glutathione status and IQ, we had to draw the conclusion that glutathione and still unknown proteines share a regulation where an unknown gene is the cause behind the correlation. On the basis of Mendelian segregation within families and frequency of highly gifted among relatives, I had estimated the allele frequency of a hypothetical major gene of general intelligence in socially representative samples of populations of Eurasian descent to be about 0.80, where 0.20 is the frequency of the allele underlying higher IQ performance (Weiss 1992, see http://www.volkmar-weiss.de/majgenes.html and http://www.volkmar-weiss.de/table.html ). In most cases heterozygotes of this rare allele should have an IQ above 105, homozygotes above 124.
From our present state of knowledge the hypothesis that NQO2 is the most probable candidate gene for general cognitive ability IQ cannot be avoided.
What is lacking, is any direct and undeniable proof that NQO2 is the major
gene locus of general intelligence. In the first world high IQ is highly correlated with high social status. If any research group uses data on the social status (occupation, education) of their subjects, the hypothesis can be checked in a rough way (without IQ testing) with moderate effort. The association between this polymorphism and IQ should be a strong one, for social reasons at least among healthy male individuals.
Only a limited number of polymorphisms have an allele frequency of about 0.20 for the rare allele in the general population. On chromosome 2 in the range 2q24.1-2q31.1 NOSTRIN, MAP1D and GAD1 have such coding polymorphisms.
Proteins with polymorphisms with a frequency of the rare allele below 0.05 can be excluded theoretically to contribute much to the variation of IQ in the normal range.
A complete understanding of alzheimer, dyslexia, attention deficit, autism, schizophrenia and other disorders will be impossible without considering the IQ of an individual before the onset of the cognitive decline and its genetic underpinning. For example, a person with an original IQ of 130 will have an IQ of 100 some time after the onset of the disease and will be diagnosed far later than a person with an original IQ of 90 reduced to an IQ of 60 by alzheimer, even when the time passed after the onset of the disease is similar in both cases. Therefore, a major gene locus of IQ has to be expected to turn out as a minor gene of alzheimer, dyslexia, attention deficit, autism, and schizophrenia and vice versa. What we need, therefore, is a better understanding of the genetics of normal IQ, too.
[This message has been edited by Volkmar Weiss (edited 07-30-2005).]