Ashutosh K. Pandey

Joint mouse-human phenome-wide association to test gene function and disease risk

Phenome-wide association is a novel reverse genetic strategy to analyze genome-to-phenome relations in human clinical cohorts. Here we test this approach using a large murine population segregating for ∼5 million sequence variants, and we compare our results to those extracted from a matched analysis of gene variants in a large human cohort. For the mouse cohort, we amassed a deep and broad open-access phenome consisting of ∼4,500 metabolic, physiological, pharmacological and behavioural traits, and more than 90 independent expression quantitative trait locus (QTL), transcriptome, proteome, metagenome and metabolome data sets—by far the largest coherent phenome for any experimental cohort (www.genenetwork.org). We tested downstream effects of subsets of variants and discovered several novel associations, including a missense mutation in fumarate hydratase that controls variation in the mitochondrial unfolded protein response in both mouse and Caenorhabditis elegans, and missense mutations in Col6a5 that underlies variation in bone mineral density in both mouse and human.

Functionally enigmatic genes: a case study of the brain ignorome.

What proportion of genes with intense and selective expression in specific tissues, cells, or systems are still almost completely uncharacterized with respect to biological function? In what ways do these functionally enigmatic genes differ from well-studied genes? To address these two questions, we devised a computational approach that defines so-called ignoromes. As proof of principle, we extracted and analyzed a large subset of genes with intense and selective expression in brain. We find that publications associated with this set are highly skewed—the top 5% of genes absorb 70% of the relevant literature. In contrast, approximately 20% of genes have essentially no neuroscience literature. Analysis of the ignorome over the past decade demonstrates that it is stubbornly persistent, and the rapid expansion of the neuroscience literature has not had the expected effect on numbers of these genes. Surprisingly, ignorome genes do not differ from well-studied genes in terms of connectivity in coexpression networks. Nor do they differ with respect to numbers of orthologs, paralogs, or protein domains. The major distinguishing characteristic between these sets of genes is date of discovery, early discovery being associated with greater research momentum—a genomic bandwagon effect. Finally we ask to what extent massive genomic, imaging, and phenotype data sets can be used to provide high-throughput functional annotation for an entire ignorome. In a majority of cases we have been able to extract and add significant information for these neglected genes. In several cases—ELMOD1, TMEM88B, and DZANK1—we have exploited sequence polymorphisms, large phenome data sets, and reverse genetic methods to evaluate the function of ignorome genes.

An evolutionarily conserved role for the aryl hydrocarbon receptor in the regulation of movement

The BXD genetic reference population is a recombinant inbred panel descended from crosses between the C57BL/6 (B6) and DBA/2 (D2) strains of mice, which segregate for about 5 million sequence variants. Recently, some of these variants have been established with effects on general metabolic phenotypes such as glucose response and bone strength. Here we phenotype 43 BXD strains and observe they have large variation (∼5-fold) in their spontaneous activity during waking hours. QTL analyses indicate that ∼40% of this variance is attributable to a narrow locus containing the aryl hydrocarbon receptor (Ahr), a basic helix-loop-helix transcription factor with well-established roles in development and xenobiotic metabolism. Strains with the D2 allele of Ahr have reduced gene expression compared to those with the B6 allele, and have significantly higher spontaneous activity. This effect was also observed in B6 mice with a congenic D2 Ahr interval, and in B6 mice with a humanized AHR allele which, like the D2 allele, is expressed much less and has less enzymatic activity than the B6 allele. Ahr is highly conserved in invertebrates, and strikingly inhibition of its orthologs in D. melanogaster and C. elegans (spineless and ahr-1) leads to marked increases in basal activity. In mammals, Ahr has numerous ligands, but most are either non-selective (e.g. resveratrol) or highly toxic (e.g., 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)). Thus, we chose to examine a major environmental influence—long term feeding with high fat diet (HFD)—to see if the effects of Ahr are dependent on major metabolic differences. Interestingly, while HFD robustly halved movement across all strains, the QTL position and effects of Ahr remained unchanged, indicating that the effects are independent. The highly consistent effects of Ahr on movement indicate that changes in its constitutive activity have a role on spontaneous movement and may influence human behavior.

RNAseq expression analysis of resistant and susceptible mice after influenza A virus infection identifies novel genes associated with virus replication and important for host resistance to infection.

BackgroundThe host response to influenza A infections is strongly influenced by host genetic factors. Animal models of genetically diverse mouse strains are well suited to identify host genes involved in severe pathology, viral replication and immune responses. Here, we have utilized a dual RNAseq approach that allowed us to investigate both viral and host gene expression in the same individual mouse after H1N1 infection.ResultsWe performed a detailed expression analysis to identify (i) correlations between changes in expression of host and virus genes, (ii) host genes involved in viral replication, and (iii) genes showing differential expression between two mouse strains that strongly differ in resistance to influenza infections. These genes may be key players involved in regulating the differences in pathogenesis and host defense mechanisms after influenza A infections. Expression levels of influenza segments correlated well with the viral load and may thus be used as surrogates for conventional viral load measurements. Furthermore, we investigated the functional role of two genes, Reg3g and Irf7, in knock-out mice and found that deletion of the Irf7 gene renders the host highly susceptible to H1N1 infection.ConclusionsUsing RNAseq analysis we identified novel genes important for viral replication or the host defense. This study adds further important knowledge to host-pathogen-interactions and suggests additional candidates that are crucial for host susceptibility or survival during influenza A infections.

Genetics of gene expression in CNS.

Transcriptome studies have revealed a surprisingly high level of variation among individuals in expression of key genes in the CNS under both normal and experimental conditions. Ten-fold variation is common, yet the specific causes and consequences of this variation are largely unknown. By combining classic gene mapping methods-family linkage studies and genomewide association-with high-throughput genomics, it is now possible to define quantitative trait loci (QTLs), single-gene variants, and even single SNPs and indels that control gene expression in different brain regions and cells. This review considers some of the major technical and conceptual challenges in analyzing variation in expression in the CNS with a focus on mRNAs, rather than noncoding RNAs or proteins. At one level of analysis, this work has been highly successful, and we finally have techniques that can be used to track down small numbers of loci that control expression in the CNS. But at a higher level of analysis, we still do not understand the genetic architecture of gene expression in brain, the consequences of expression QTLs on protein levels or on cell function, or the combined impact of expression differences on behavior and disease risk. These important gaps are likely to be bridged over the next several decades using (1) much larger sample sizes, (2) more powerful RNA sequencing and proteomic methods, and (3) novel statistical and computational models to predict genome-to-phenome relations.

Evaluation of Methyl-Binding Domain Based Enrichment Approaches Revisited

Methyl-binding domain (MBD) enrichment followed by deep sequencing (MBD-seq), is a robust and cost efficient approach for methylome-wide association studies (MWAS). MBD-seq has been demonstrated to be capable of identifying differentially methylated regions, detecting previously reported robust associations and producing findings that replicate with other technologies such as targeted pyrosequencing of bisulfite converted DNA. There are several kits commercially available that can be used for MBD enrichment. Our previous work has involved MethylMiner (Life Technologies, Foster City, CA, USA) that we chose after careful investigation of its properties. However, in a recent evaluation of five commercially available MBD-enrichment kits the performance of the MethylMiner was deemed poor. Given our positive experience with MethylMiner, we were surprised by this report. In an attempt to reproduce these findings we here have performed a direct comparison of MethylMiner with MethylCap (Diagenode Inc, Denville, NJ, USA), the best performing kit in that study. We find that both MethylMiner and MethylCap are two well performing MBD-enrichment kits. However, MethylMiner shows somewhat better enrichment efficiency and lower levels of background “noise”. In addition, for the purpose of MWAS where we want to investigate the majority of CpGs, we find MethylMiner to be superior as it allows tailoring the enrichment to the regions where most CpGs are located. Using targeted bisulfite sequencing we confirmed that sites where methylation was detected by either MethylMiner or by MethylCap indeed were methylated.

In Inflamed Intestinal Tissues and Epithelial Cells, Interleukin 22 Signaling Increases Expression of H19 Long Noncoding RNA, Which Promotes Mucosal Regeneration

BACKGROUND & AIMS Inflammation affects regeneration of the intestinal epithelia; long noncoding RNAs (lncRNAs) regulate cell functions, such as proliferation, differentiation, and migration. We investigated the mechanisms by which the lncRNA H19, imprinted maternally expressed transcript (H19) regulates regeneration of intestinal epithelium using cell cultures and mouse models of inflammation. METHODS We performed RNA-sequencing transcriptome analyses of intestinal tissues from mice with lipopolysaccharide (LPS)-induced sepsis to identify lncRNAs associated with inflammation; findings were confirmed by quantitative real-time polymerase chain reaction and in situ hybridization analyses of intestinal tissues from mice with sepsis or dextran sulfate sodium (DSS)-induced mucosal wound healing and patients with ulcerative colitis compared to healthy individuals (controls). We screened cytokines for their ability to induce expression of H19 in HT-29 cells and intestinal epithelial cells (IECs), and confirmed findings in crypt epithelial organoids derived from mouse small intestine. IECs were incubated with different signal transduction inhibitors and effects on H19 lncRNA levels were measured. We assessed intestinal epithelial proliferation or regeneration in H19ΔEx1/+ mice given LPS or DSS vs wild-type littermates (control mice). H19 was overexpressed in IECs using lentiviral vectors and cell proliferation was measured. We performed RNA antisense purification, RNA immunoprecipitation, and luciferase reporter assays to study functions of H19 in IECs. RESULTS In RNA-sequencing transcriptome analysis of lncRNA expression in intestinal tissues from mice, we found that levels of H19 lncRNA changed significantly with LPS exposure. Levels of H19 lncRNA increased in intestinal tissues of patients with ulcerative colitis, mice with LPS-induced and polymicrobial sepsis, or mice with DSS-induced colitis, compared with controls. Increased H19 lncRNA localized to epithelial cells in the intestine, regardless of Lgr5 messenger RNA expression. Exposure of IECs to interleukin 22 (IL22) increased levels of H19 lncRNA with time and dose, which required STAT3 and protein kinase A activity. IL22 induced expression of H19 in mouse intestinal epithelial organoids within 6 hours. Exposure to IL22 increased growth of intestinal epithelial organoids derived from control mice, but not H19ΔEx1/+ mice. Overexpression of H19 in HT-29 cells increased their proliferation. Intestinal mucosa healed more slowly after withdrawal of DSS from H19ΔEx1/+ mice vs control mice. Crypt epithelial cells from H19ΔEx1/+ mice proliferated more slowly than those from control mice after exposure to LPS. H19 lncRNA bound to p53 and microRNAs that inhibit cell proliferation, including microRNA 34a and let-7; H19 lncRNA binding blocked their function, leading to increased expression of genes that promote regeneration of the epithelium. CONCLUSIONS The level of lncRNA H19 is increased in inflamed intestinal tissues from mice and patients. The inflammatory cytokine IL22 induces expression of H19 in IECs, which is required for intestinal epithelial proliferation and mucosal healing. H19 lncRNA appears to inhibit p53 protein and microRNA 34a and let-7 to promote proliferation of IECs and epithelial regeneration.

Genetic control of weight loss during pneumonic Burkholderia pseudomallei infection.

Burkholderia pseudomallei (Bp) is the causal agent of a high-morbidity/mortality disease syndrome known as melioidosis. This syndrome can range from acute fulminate disease to chronic, local, and disseminated infections that are often difficult to treat because Bp exhibits resistance to many antibiotics. Bp is a prime candidate for use in biologic warfare/terrorism and is classified as a Tier-1 select agent by HHS and APHIS. It is known that inbred mouse strains display a range of susceptibility to Bp and that the murine infection model is ideal for studying acute melioidosis. Here, we exploit a powerful mouse genetics resource that consists of a large family of BXD-type recombinant inbred strains, to perform genome-wide linkage analysis of the weight loss phenotype following pneumonic infection with Bp. We infected parental mice and 32 BXD strains with 50-100 CFU of Bp (strain 1026b) and monitored weight retention each day over an eleven-day time course. Using the computational tools in GeneNetwork, we performed genome-wide linkage analysis to identify an interval on chromosome 12 that appears to control the weight retention trait. We then analyzed and ranked positional candidate genes in this interval, several of which have intriguing connections with innate immunity, calcium homeostasis, lipid transport, host cell growth and development, and autophagy.

Xylosylation of the Notch receptor preserves the balance between its activation by trans-Delta and inhibition by cis-ligands in Drosophila

The Drosophila glucoside xylosyltransferase Shams xylosylates Notch and inhibits Notch signaling in specific contexts including wing vein development. However, the molecular mechanisms underlying context-specificity of the shams phenotype is not known. Considering the role of Delta-Notch signaling in wing vein formation, we hypothesized that Shams might affect Delta-mediated Notch signaling in Drosophila. Using genetic interaction studies, we find that altering the gene dosage of Delta affects the wing vein and head bristle phenotypes caused by loss of Shams or by mutations in the Notch xylosylation sites. Clonal analysis suggests that loss of shams promotes Delta-mediated Notch activation. Further, Notch trans-activation by ectopically overexpressed Delta shows a dramatic increase upon loss of shams. In agreement with the above in vivo observations, cell aggregation and ligand-receptor binding assays show that shams knock-down in Notch-expressing cells enhances the binding between Notch and trans-Delta without affecting the binding between Notch and trans-Serrate and cell surface levels of Notch. Loss of Shams does not impair the cis-inhibition of Notch by ectopic overexpression of ligands in vivo or the interaction of Notch and cis-ligands in S2 cells. Nevertheless, removing one copy of endogenous ligands mimics the effects of loss shams on Notch trans-activation by ectopic Delta. This favors the notion that trans-activation of Notch by Delta overcomes the cis-inhibition of Notch by endogenous ligands upon loss of shams. Taken together, our data suggest that xylosylation selectively impedes the binding of Notch with trans-Delta without affecting its binding with cis-ligands and thereby assists in determining the balance of Notch receptor’s response to cis-ligands vs. trans-Delta during Drosophila development.

The genetic architecture of murine glutathione transferases

Glutathione S-transferase (GST) genes play a protective role against oxidative stress and may influence disease risk and drug pharmacokinetics. In this study, massive multiscalar trait profiling across a large population of mice derived from a cross between C57BL/6J (B6) and DBA2/J (D2)—the BXD family—was combined with linkage and bioinformatic analyses to characterize mechanisms controlling GST expression and to identify downstream consequences of this variation. Similar to humans, mice show a wide range in expression of GST family members. Variation in the expression of Gsta4, Gstt2, Gstz1, Gsto1, and Mgst3 is modulated by local expression QTLs (eQTLs) in several tissues. Higher expression of Gsto1 in brain and liver of BXD strains is strongly associated (P < 0.01) with inheritance of the B6 parental allele whereas higher expression of Gsta4 and Mgst3 in brain and liver, and Gstt2 and Gstz1 in brain is strongly associated with inheritance of the D2 parental allele. Allele-specific assays confirmed that expression of Gsto1, Gsta4, and Mgst3 are modulated by sequence variants within or near each gene locus. We exploited this endogenous variation to identify coexpression networks and downstream targets in mouse and human. Through a combined systems genetics approach, we provide new insight into the biological role of naturally occurring variants in GST genes.