24In current clinical practice care of diseased patients is often restricted to separated disciplines. However, 25 such an organ-centered approach is not always suitable. For example, cognitive dysfunction is a severe 26 burden in heart failure patients. Moreover, these patients have an increased risk for age-associated 27 dementias. The underlying molecular mechanisms are presently unknown and thus corresponding 28 therapeutic strategies to improve cognition in heart failure patients are missing. Using mice as model 29 organisms we show that heart failure leads to specific changes in hippocampal gene-expression, a brain 30 region intimately linked to cognition. These changes reflect increased cellular stress pathways which 31 eventually lead to loss of neuronal euchromatin and reduced expression of a hippocampal gene cluster 32 essential for cognition. Consequently, mice suffering from heart failure exhibit impaired memory 33 function. These pathological changes are ameliorated via the administration of a drug that promotes 34 neuronal euchromatin formation. Our study provides first insight to the molecular processes by which 35 heart failure contributes to neuronal dysfunction and point to novel therapeutic avenues to treat cognitive 36 defects in heart failure patients. 37 38 39 40 41 2 1 2 3 1 Results 2 3 Heart failure in CamkIIδc TG mice leads to hippocampal gene expression changes indicative of 4 dementia 5With the aim to elucidate the molecular processes by which cardiovascular dysfunction leads to memory 6 impairment and an increases the risk for dementia, we decided to employ a well-established mouse 7 model for heart failure in which cardiomyocyte-specific kinase CamkIIδc is overexpressed under the 8 control of the alpha-MHC promoter (CamkIIδc TG mice) (Maier, Zhang et al., 2003). Thus, 9 overexpression of CamkIIδc is specific to cardiomyocytes and is not detected in other organs, including 10 the brain (Maier et al., 2003), making it a bona fide model to study the impact of heart failure on brain 11 function ( Fig 1A). We reasoned that this well-defined genetic heart failure model would be superior to 12 other experimental approaches linked for example to cerebral hypoperfusion such as carotid artery 13 occlusion, since it allowed us to study brain function in response to the very precise and exclusive 14 manipulation of cardiac tissue. In line with previous findings, 3-month-old CamkIIδc TG mice displayed 15 heart failure with left ventricular dilatation, impaired ejection fraction, and increased heart mass ( Fig 1B, 16 C), whereas the overall body weight was not affected (P = 0.863 for CamkIIδc TG vs control mice, n =8, 17 unpaired t-test). As a first approach to study the impact of cardiac dysfunction on brain plasticity we 18 decided to analyze the transcriptome of the hippocampal CA1 region in 3-month-old CamkIIδc TG mice 19 ( Fig 1D). This was based on data showing that (1) gene expression is a sensitive molecular correlate of 20 memory function and is de-regulated in dementia patients and corres...