Background
Cognitive functions are impaired in schizophrenia [
1,
2] and, to a lesser extent, also in bipolar disorder [
3,
4] and major depression [
5]. Among the different cognitive domains, verbal memory, verbal fluency and attention are typically most affected [
2,
3,
5,
6]. Especially in schizophrenia, these impairments are already present in adolescence, long before the onset of psychotic symptoms [
4,
6], in the prodromal state [
7] and can also be found in relatives of patients [
8,
9], suggesting a genetic influence.
In the last years, several susceptibility genes for psychiatric disorders have been identified (for reviews, see [
10‐
13]. Among these,
G72 (recently named D-amino acid oxidase activator, DAOA) is one of the most frequently replicated vulnerability genes [
14]. It shows a genetic overlap across the major psychoses, such as bipolar disorder, major depression and schizophrenia [
15‐
21], questioning the long-held view of a strict nosological separation of psychiatric disorders [
16,
17].
To reveal potential in vivo functions of risk genes several studies have assessed genotype effects in healthy individuals. For several risk alleles, an association with subtle impairments in cognitive functions (e.g., [
22]) or disadvantageous personality traits (e.g., [
21,
23,
24]) have been found, although this does depend on the gene and the respective tagging marker.
In a recent study, we investigated the effect of
G72 genotype on working memory using both neuropsychological tests and functional neuroimaging. Unexpectedly, the putative high-risk haplotype (i.e. homozygote C/C-allele carriers in the single nucleotide polymorphism (SNP) M23 and homozygote T/T-allele carriers in SNP M24) was associated with significant
better performance in verbal working memory. These behavioural differences were accompanied by a stronger deactivation in the right parahippocampus during a working memory 2-back task [
25]. Thus, the high risk variant of
G72 has a beneficial influence on verbal working memory in healthy subjects, although it is known to increase the risk for schizophrenia and affective disorders, diseases that are associated with impairments in this domain [
2,
3,
5,
6].
In the present study we further investigated how genetic alterations in G72 influence cognitive functions in healthy individuals. We examined whether the positive influence of the high risk G72 variant is restricted to verbal working memory or whether this genotype also beneficially influences other cognitive domains.
Discussion
In the present study we investigated the effect of G72 genotype on cognitive functions in a large sample of healthy individuals. Our results show that G72 status influences the performance in a number of cognitive domains (significant differences in verbal working memory and attention, differences on a trend level in spatial working memory and executive functions). Most importantly, the high-risk allele carriers scored significantly better than one or even both other low-risk groups. Thus, healthy individuals with a G72 haplotype that is known to increase the risk for the major psychoses perform better in some cognitive domains than subjects with a low risk status, although these cognitive domains are negatively affected by the psychiatric disorders that are associated with this allele variant.
Only few studies assessed so far the effect of genetic variation in
G72 on cognitive functions. Goldberg et al. investigated the relationship between several SNPs in the
G72 region and select cognitive measures in attention, working memory, and episodic memory in a cohort of over 600 subjects, including patients with schizophrenia, their unaffected siblings, and healthy controls. The authors showed for the markers M23 and M24 a significant genotype by diagnosis interaction with a number of cognitive measures (working memory, attention, verbal learning). The low risk homozygote A/A genotype group scored better than the high risk T/T homozygote group, most notably in the schizophrenia group [
35]. Although the authors also report a main effect of genotype at least for marker M24, this effect seems to be mainly driven by the patient sample. Opgen-Rhein and colleagues investigated the influence of
G72 variation on cognitive performance in a large sample of both patients schizophrenia (n = 178) and healthy controls (n = 144) [
36]. They showed that a certain
G72 haplotype located upstream of the presumed gene borders of
G72 has an impact on semantic fluency. Interestingly, carriers of the risk haplotype showed
better semantic fluency than non-carriers, both in the patients and the control population. Donohue and colleagues report that a functional polymorphism within
G72 (rs 2391191, M15) was associated with poorer verbal memory performance among patients with schizophrenia [
37]. Taken together, these studies show that functional polymorphisms in the
G72 gene region have an impact on cognitive functions. This impact seems to be most notable in psychiatric samples. Our study further extends these previous findings and show that the SNPs in the
G72 gene complex have also an impact on cognitive functions in healthy controls.
Our results suggest that, at least for markers M23 and M24, the high-risk genotype of
G72 has no negative effect on cognitive functions in healthy individuals
per se, but even a positive effect in some cognitive domains (such as verbal working memory and attention). This finding is at first glance counterintuitive, but might be explained by a number of reasons. First, the M23-M24 risk haplotype might influence cognitive functions independent of its role as a risk factor for psychiatric disorders. A similar explanation has been proposed by Opgen-Rhine and colleagues who also report that a risk haplotype in the
G72 region is associated with better performance in semantic processing both in patients with schizophrenia and control subjects [
36]. Second, from a standpoint of evolutionary theory, it might be argued that the risk variant of
G72 is maintained in the population since it has a beneficial influence on cognitive functions which has a positive effect for evolutionary selection [
36]. At last, it cannot be fully excluded that at least some of the results represent false positive findings. It is for instance in particular difficult to understand why the intermediate risk group has a significant worse performance in the d2-test in comparison to both the high- and the low-risk group. A limitation of our study is that we cannot give a stringent neurobiological explanation for these findings. However, all results are based on a large cohort (n = 423), were obtained by stringent statistical analyses and survived Bonferroni corrected thresholds, reducing the likelihood of this interpretation.
The functional mechanisms of
G72 are still not fully understood. Chumakov and colleagues showed that the
G72 protein (which is only known in higher primates) activates a second protein, D-amino acid oxidase (DAAO), that is involved in the mechanisms of D-serine [
38]. D-serine is an agonist at the glycine modulation site of the N-methyl-D-aspartate (NMDA) receptor [
39]. Thus,
G72 might work as an indirect modulator of NMDA neurotransmission, which has been implicated in various cognitive domains. Lower serum level of D-serine has been shown, for instance, in patients with schizophrenia. Furthermore, the administration of D-serine (as add-on medication) has been shown to reduce some of the symptoms in schizophrenia [
40]. This provides a potential link between
G72 and the glutamate hypofunction hypothesis of schizophrenia [
41]. Another study however failed to confirm the interaction between
G72 and DAAO [
42]. Rather, LG72, a splicing isoform of the
G72 gene, encodes for a mitrochondrial protein. It was shown that an overexpression of
G72 led to mitrochondrial fragmentation. The authors proposed that an unknown function of the
G72 in modulating mitochondrial morphology might be responsible for the risk-conferring property of the gene.
Several fMRI studies suggest a modulatory role of
G72 on brain activity in the medial temporal lobe (MTL), in particular the hippocampus and parahippocampus [
25,
35,
43]. Goldberg and colleagues showed that healthy control subjects carrying the homozygous high-risk T/T allele at SNP M24 had decreased brain activity of the right hippocampus and left parahippocampus during an episodic memory encoding task [
35]. Hall et al. investigated subjects with a high familial risk for schizophrenia and report brain activation differences related to the
G72 genotype (as assessed by SNPs M23 and M24) in the left hippocampus and parahippocampus during a verbal sentence completion task [
43]. Jansen and colleagues showed that in healthy control subjects the
G72 genotype (determined by SNPs M23 and M24) is correlated with brain activity of the right parahippocampus during a working memory task [
25].
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
AJ performed the statistical analysis, was involved in the interpretation of data, made substantial contributions to conception and design and drafted the manuscript. SK was involved in the statistical analysis and the interpretation of data and helped to draft the manuscript. AK was involved in the acquisition of data, made substantial contributions to conception and design and was involved in drafting the manuscript. VM was involved in the acquisition of data and was involved in drafting the manuscript. MT was involved in the statistical analysis and the interpretation of data and helped to draft the manuscript. TE was involved in the genetic analyses and was involved in drafting the manuscript. KZ was involved in the genetic analyses and was involved in drafting the manuscript. MN made substantial contributions to conception and design, was involved in the genetic analyses. JT was involved in the genetic analyses and was involved in drafting the manuscript. MR made substantial contributions to conception and design and was involved in the genetic analyses. TK conceived of the study, and participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.