To functionally test the hypothesis that the observed co-morbidity between T2D and psychiatric disorders might be partially due to pleiotropic effects of TCF7L2 alleles, we performed several behavioral studies using mice with null alleles of Tcf7l2 as well as mice over-expressing Tcf7l2. The use of these mouse models allowed for the testing of a range of Tcf7l2 copy numbers that might be similar to gain- or loss-of-function alleles in humans. Our data support a role for Tcf7l2 in behavior, thereby demonstrating that this transcription factor can influence a diverse set of processes.
Using the open field and light dark box tests, we observed anxiety-like phenotypes in Tcf7l2+/− mice; ablation of Tcf7l2 led to decreased time in the center of the arena and in the light compartment. However, these phenotypes were coincident with differences in locomotor activity, namely decreased distance traveled in the open field. While this may confound the interpretation of anxiety phenotypes, we did not observe significant differences in latency to periphery in open field testing. Both latency to the dark compartment and the number of transitions were not significant in light dark box testing, suggesting this test might not have been confounded by differences in activity. Taken together, these observations support a role for Tcf7l2 in anxiety-like behaviors; future studies of the impact of human alleles of TCF7L2 on anxiety are warranted.
Our mouse model demonstrated a dose-dependent role for Tcf7l2
in fear learning. Ablation of Tcf7l2
led to enhanced fear learning while mice overexpressing Tcf7l2
displayed an impairment of fear learning. Interestingly, this gene-dosage effect was only observed for contextual fear, which is believed to be a hippocampus-dependent process 
and therefore may have some relevance for SCZ. Indeed, disruptions of contextual processing are known to occur in SCZ 
and this endophenotype is routinely used in rodent and human studies 
While our study uncovered a role for Tcf7l2
on anxiety and fear learning, we did not identify significant differences in PPI in both heterozygous null mice and BAC transgenic mice. This is significant because PPI is a behavioral test that can be more directly related to the observation of increased risk for SCZ in humans 
. However, we cannot exclude potential differences in PPI in older mice as behavioral studies were conducted on mice 7–8 weeks of age at the start of behavioral testing.
Importantly, none of the mice were diabetic during behavioral testing as we previously reported glucose intolerance in mice overexpressing Tcf7l2
only after a high fat diet stress 
. As a result, this enabled the dissociation of diabetes-related secondary effects in this study. In light of this, our study design is able to detect independent genetic effects on behavior, which is a significant advantage over studies of co-morbidity in humans. Our design allows us to conclude that Tcf7l2 directly regulates behavior.
The idea that seemingly disparate diseases may result from shared susceptibility loci is intriguing and reflects the integrated nature of our genome. Indeed, TCF7L2
functions as a transcriptional regulator of the canonical Wnt signaling pathway that maintains broad roles in development regulating cell fate, survival and proliferation 
. Genes such as TCF7L2
that are involved in pathways that harbor pleiotropic functions represent strong candidates in this respect. Our study demonstrates that TCF7L2
carries pleiotropic effects and provides added evidence of possible common genetic underpinning that may explain the historical comorbidities between T2D and psychiatric disorders.