System genetics in the rat HXB/BXH family identifies Tti2 as a pleiotropic quantitative trait gene for adult hippocampal neurogenesis and serum glucose

Neurogenesis in the adult hippocampus contributes to learning and memory in the healthy brain but is dysregulated in metabolic and neurodegenerative diseases. The molecular relationships between neural stem cell activity, adult neurogenesis, and global metabolism are largely unknown. Here we applied unbiased systems genetic methods to quantify genetic covariation among adult neurogenesis and metabolic phenotypes in peripheral tissues of a genetically diverse family of rat strains, derived from a cross between the spontaneously hypertensive (SHR/OlaIpcv) strain and Brown Norway (BN-Lx/Cub). The HXB/BXH family is a very well established model to dissect genetic variants that modulate metabolic and cardiovascular disease and we have accumulated deep phenome and transcriptome data in a FAIR-compliant resource for systematic and integrative analyses. Here we measured rates of precursor cell proliferation, survival of new neurons, and gene expression in the hippocampus of the entire HXB/BXH family, including both parents. These data were combined with published metabolic phenotypes to detect a neurometabolic quantitative trait locus (QTL) for serum glucose and neuronal survival. We subsequently fine-mapped a key phenotype to a locus that includes the telo2-interacting protein 2 gene (Tti2)—a chaperone that modulates the activity and stability of PIKK kinases. To validate variants in or near Tti2 as a cause for differences in neurogenesis and glucose levels, we generated a targeted frameshift mutation on the SHR/OlaIpcv background. Heterozygous SHR-Tti2+/- mutants had lower rates of hippocampal neurogenesis and hallmarks of dysglycemia compared to wild-type littermates. Our findings highlight Tti2 as a causal genetic and molecular link between glucose metabolism and structural brain plasticity. In humans, more than 800 genomic variants are linked to TTI2 expression, seven of which have associations to protein and blood stem cell factor concentrations, blood pressure and frontotemporal dementia. Author summary Metabolic and neurological disorders are often comorbid, suggesting that biological pathways which orchestrate peripheral homeostasis and the integrity of the nervous system intersect. The genetic architecture behind these relationships is still poorly described, in part because molecular processes in the human brain are very difficult to study. We thus used a rodent genetic reference population to investigate links between adult hippocampal neurogenesis— a cellular plasticity mechanism important for learning flexibility—and metabolism. We measured adult neurogenesis in the family of 30 HXB/BXH rat recombinant inbred strains, who are characterised by stable differences in metabolism, behaviour, and gene expression levels. Because gene variants affecting distinct traits segregated into different members of the family, we discovered that previously published phenotypes correlated to adult neurogenesis due to shared genomic sequence. We found that expression levels of Tti2—a part of a specialised protein chaperone complex regulating stability of PIKK kinases—were concomitantly influencing adult neurogenesis and serum glucose levels. In human populations hundreds of genomic variants regulate TTI2 expression, potentially affecting brain function and glucose homeostasis.


175
The search for genetic associations between adult neurogenesis and physiological traits 176 was performed in several steps. First, we identified QTL for both neurogenic traits ( Fig. 2A, 177 B). We detected a suggestive QTL for net neurogenesis on chromosome 16 (LOD = 3.12, 178 genome-wide corrected p value = 0.14) and a weak suggestive QTL for proliferation on 179 chromosome 17 (LOD = 2.39, p = 0.52). Second, we identified phenotypes from the 180 GeneNetwork database [34] that significantly correlated to both traits (Table 1). We then 181 used these phenotypes as covariates in conditional QTL scans to screen for potential 182 interactions at a genomic level. A substantial change in the LOD score after using another 183 phenotype as a covariate indicates that genetic variation within a QTL may have pleiotropic 184 effects on these two phenotypes [47]. Among pairs of correlating phenotypes, we detected 185 only one such association: QTL mapping for net neurogenesis revealed a LOD drop below 186 suggestive level to 0.43 after conditioning on serum glucose levels. Accordingly, an 187 overlapping significant QTL for serum glucose was found on chromosome 16 (LOD = 5.13,    242 was associated with higher levels of Tti2 mRNA, which we verified using quantitative RT-243 PCR in RNA isolated from the hippocampus, liver, muscle, kidney and pancreas (Table S1).

244
The allelic variation underlying a QTL can either change the expression level of a gene, or 245 the function of its product by altering its structure. We inspected the genes located within the 246 eigenphenotype QTL confidence interval for non-synonymous amino acid substitutions.
247 Interestingly, only Tti2 carried several missense mutations, including one at a highly 248 conserved position, although none of the substitutions were predicted as damaging by 249 Polyphen or SIFT (Table S2). Together, these data support Tti2 as a causal candidate gene 250 for the combined serum glucose and net neurogenesis QTL.  Table 3. Statistical analysis of response to insulin and adrenaline stimulation in 298 tissues isolated from SHR-Tti2 +/rats and SHR wild type littermates.

364
We first checked whether observed gene expression changes could be connected to 365 altered activity of any of the PIKK. Among canonical pathways, IPA indicated enrichment for 366 genes associated with ATM signalling in the liver, muscle, and fat, and with mTOR signalling 367 in the muscle (Fig. S2B-D). Additionally, IPA predicted SMG1 as one of potential upstream 368 regulators in the liver and hippocampus (File S1). Nevertheless, the majority of changes in 369 gene expression were not directly connected to PIKK activity.

425
In this study we used unbiased systems genetics methods to i) discover a number of 426 metabolic and endocrine genetic correlates of the two critical parameters of adult 427 hippocampal neurogenesis-precursor cell proliferation and net neurogenesis; and ii) identify 428 Tti2 as a molecular link between brain cellular plasticity and peripheral metabolism.

499
Transcriptome profiling of SHR-Tti2 +/rats revealed extensive gene expression changes in 500 the liver, and to a lesser degree in the skeletal muscle, perirenal fat and hippocampus 501 compared to SHR wild type littermates. Liver, together with skeletal muscle and adipose 502 tissue are decisive organs in maintenance of glucose homeostasis and hence development 503 of insulin resistance [66]. Functional analysis of differentially expressed genes in the liver 504 identified networks of genes and potential regulators whose activation and inhibition could 505 explain insulin resistance and dysglycemia in the heterozygous animals. We also recorded 506 significant upregulation of Insr in the muscle, which IPA interpreted as consistent with 507 hypoglycaemia and insulin resistant diabetes (File S3).

508
We also used IPA to predict which upstream regulators could be activated or inhibited in a 509 manner consistent with observed gene expression changes. The vast majority of differentially 525 Interestingly, in SHR-Tti2 +/rats IPA detected enrichment of differentially expressed genes 526 related to ATM signalling in the liver, muscle and fat and mTOR signalling in the muscle.
527 While these results do not imply that other PIKK were not affected, ATM might be the most 528 sensitive to Tti2 downregulation also in rats used in our study.

529
Pleiotropy occurs when a single genomic variation, or more broadly a change in a function 530 of a single gene, has multiple consequences at the phenotypic level [9]. Because metabolic 531 diseases can, to some extent, be modified by lifestyle interventions in order to prevent or 532 dampen cognitive decline, but genes cannot, it is clinically crucial to understand which 534 due to pleiotropic genes. Because we used targeted mutagenesis at the Tti2 locus rather 535 than tissue specific approaches to confirm association with target phenotypes, we cannot 536 exclude that Tti2 affects neurogenesis through circulating metabolites or hormones.

555
The support in favour of direct effects of Tti2 reduction on neurogenesis comes from the   827 Glucose utilization in isolated epididymal adipose tissue and brown adipose tissue.

828
Distal parts of epididymal adipose tissue or interscapular brown adipose tissue were 829 rapidly dissected and incubated for 2 hours in Krebs-Ringer bicarbonate buffer with 5 mmol/L 830 glucose, 0.1 Ci 14 C-U-glucose/mL (UVVR, Prague, Czech Republic) and 2% bovine serum 831 albumin, gaseous phase 95% O 2 and 5% CO 2 in the presence (250 U/mL) or absence of 832 insulin in incubation media. All incubations were performed at 37 °C in sealed vials in a 833 shaking water bath. Then we estimated incorporation of 14 C-glucose into neutral lipids.
834 Briefly, adipose tissue was removed from incubation medium, rinsed in saline, and 835 immediately put into chloroform. The pieces of tissue were dissolved using a Teflon pestle 836 homogenizer, methanol was added (chloroform:methanol 2:1), and lipids were extracted at 837 4 °C overnight. The remaining tissue was removed, KH 2 PO 4 was added and a clear extract

849
For determination of triglycerides in liver, gastrocnemius muscle, kidney, and heart, 850 tissues were powdered under liquid N 2 and extracted for 16 hours in chloroform:methanol, 851 after which 2 % KH 2 PO 4 was added and the solution was centrifuged. The organic phase 852 was removed and evaporated under N 2 . The resulting pellet was dissolved in isopropyl 1315 Supporting Tables   1316 Table S1. Quantitative RT-PCR analysis of Tti2 mRNA expression in BN and SHR rats.