LAUSR

Abstract Background Impaired memory is among the most profound cognitive deficits observed in patients with schizophrenia. Evidence from twin studies suggests that memory is mainly influenced by genetics. Moreover, a few twin studies have demonstrated genetic overlap between specific memory functions and schizophrenia. Memory deficits in schizophrenia seem to involve abnormalities in frontal cortical areas and the middle temporal lobe, particularly the hippocampus. In the general population, twin studies have consistently demonstrated genetic influences on brain volumes, however, evidence from twin pairs discordant for schizophrenia suggests that hippocampus volumes may be more susceptible to environmental effects in patients. Methods Twin pairs concordant or discordant for a diagnosis in the schizophrenia spectrum were recruited nation-wide by linking The Danish Twin Register and The Danish Psychiatric Central Research Register. Both monozygotic (MZ) and dizygotic (DZ) proband pairs as well as healthy control (HC) pairs were identified. A total of 216 twins participated in this study consisting of 32 complete MZ and 24 complete DZ proband pairs, 29 complete MZ and 20 complete DZ HC pairs, and six twins from proband pairs were included without their sibling. Verbal memory was assessed using the list learning task from the Brief Assessment of Cognition in Schizophrenia (BACS), visual memory using the Rey Complex Figure Test (RCFT) and associative memory using 15 word pairs. Structural brain scans were acquired with T1-weighted sequence on a Philips 3.0 T Achieva MRI scanner with a 32-channel SENSE head coil. Images were processed using FreeSurfer (version 5.3) and the Desikan-Killiany atlas was used to extract the volumes of bilateral hippocampi, superior frontal, rostral and caudal middle frontal cortices as well as the whole brain volume. Structural equation modelling was applied to examine the genetic and environmental contributions to the variability in memory and brain measures and to quantify associations with schizophrenia spectrum liability. Results Significant heritability estimates were observed for verbal memory (h2=0.53), visual memory (h2=0.58) and associative memory (immediate h2=0.33, delayed h2=0.54), whereas the copy and recognition task from RCFT were only explained by unique environmental factors. Except for verbal memory, all memory measures were significantly associated with schizophrenia spectrum liability, and these were mainly due to overlapping genetic factors. Genetic factors also significantly contributed to whole brain (h2=0.36), right superior frontal (h2=0.48), left rostral middle frontal (h2=0.40) and hippocampus volumes (right h2=0.29, left h2=0.50). Common environmental factors significantly influenced whole brain (c2=0.51), right hippocampus (c2=0.51) and right rostral middle frontal (c2=0.47) volumes. Hippocampus volumes were significantly associated with schizophrenia spectrum liability, and for the left hippocampus this association was due to overlapping genetic factors. Discussion Specific memory measures and related brain areas were heritable, providing further evidence of the importance of genetics in memory functioning. Furthermore, the majority of the applied memory measures and left hippocampal volume were (genetically) associated with schizophrenia spectrum liability, suggesting a partially shared etiology. The heritable memory measures and related brain areas showing associations with disease may represent endophenotypes for schizophrenia spectrum disorders. In future analyses, we plan to examine the covariance between memory, brain volumes and schizophrenia.