Frauke Degenhardt

Genome-wide association analysis identifies variation in vitamin D receptor and other host factors influencing the gut microbiota

Human gut microbiota is an important determinant for health and disease, and recent studies emphasize the numerous factors shaping its diversity. Here we performed a genome-wide association study (GWAS) of the gut microbiota using two cohorts from northern Germany totaling 1,812 individuals. Comprehensively controlling for diet and non-genetic parameters, we identify genome-wide significant associations for overall microbial variation and individual taxa at multiple genetic loci, including the VDR gene (encoding vitamin D receptor). We observe significant shifts in the microbiota of Vdr−/− mice relative to control mice and correlations between the microbiota and serum measurements of selected bile and fatty acids in humans, including known ligands and downstream metabolites of VDR. Genome-wide significant (P < 5 × 10−8) associations at multiple additional loci identify other important points of host–microbe intersection, notably several disease susceptibility genes and sterol metabolism pathway components. Non-genetic and genetic factors each account for approximately 10% of the variation in gut microbiota, whereby individual effects are relatively small.

Genome-wide analysis of rare copy number variations reveals PARK2 as a candidate gene for attention-deficit/hyperactivity disorder

Attention-deficit/hyperactivity disorder (ADHD) is a common, highly heritable neurodevelopmental disorder. Genetic loci have not yet been identified by genome-wide association studies. Rare copy number variations (CNVs), such as chromosomal deletions or duplications, have been implicated in ADHD and other neurodevelopmental disorders. To identify rare (frequency ⩽1%) CNVs that increase the risk of ADHD, we performed a whole-genome CNV analysis based on 489 young ADHD patients and 1285 adult population-based controls and identified one significantly associated CNV region. In tests for a global burden of large (>500 kb) rare CNVs, we observed a nonsignificant (P=0.271) 1.126-fold enriched rate of subjects carrying at least one such CNV in the group of ADHD cases. Locus-specific tests of association were used to assess if there were more rare CNVs in cases compared with controls. Detected CNVs, which were significantly enriched in the ADHD group, were validated by quantitative (q)PCR. Findings were replicated in an independent sample of 386 young patients with ADHD and 781 young population-based healthy controls. We identified rare CNVs within the parkinson protein 2 gene (PARK2) with a significantly higher prevalence in ADHD patients than in controls (P=2.8 × 10−4 after empirical correction for genome-wide testing). In total, the PARK2 locus (chr 6: 162 659 756–162 767 019) harboured three deletions and nine duplications in the ADHD patients and two deletions and two duplications in the controls. By qPCR analysis, we validated 11 of the 12 CNVs in ADHD patients (P=1.2 × 10−3 after empirical correction for genome-wide testing). In the replication sample, CNVs at the PARK2 locus were found in four additional ADHD patients and one additional control (P=4.3 × 10−2). Our results suggest that copy number variants at the PARK2 locus contribute to the genetic susceptibility of ADHD. Mutations and CNVs in PARK2 are known to be associated with Parkinson disease.

A protein-truncating R179X variant in RNF186 confers protection against ulcerative colitis

Protein-truncating variants protective against human disease provide in vivo validation of therapeutic targets. Here we used targeted sequencing to conduct a search for protein-truncating variants conferring protection against inflammatory bowel disease exploiting knowledge of common variants associated with the same disease. Through replication genotyping and imputation we found that a predicted protein-truncating variant (rs36095412, p.R179X, genotyped in 11,148 ulcerative colitis patients and 295,446 controls, MAF=up to 0.78%) in RNF186, a single-exon ring finger E3 ligase with strong colonic expression, protects against ulcerative colitis (overall P=6.89 × 10−7, odds ratio=0.30). We further demonstrate that the truncated protein exhibits reduced expression and altered subcellular localization, suggesting the protective mechanism may reside in the loss of an interaction or function via mislocalization and/or loss of an essential transmembrane domain.

Sparse Modeling Reveals miRNA Signatures for Diagnostics of Inflammatory Bowel Disease

The diagnosis of inflammatory bowel disease (IBD) still remains a clinical challenge and the most accurate diagnostic procedure is a combination of clinical tests including invasive endoscopy. In this study we evaluated whether systematic miRNA expression profiling, in conjunction with machine learning techniques, is suitable as a non-invasive test for the major IBD phenotypes (Crohns disease (CD) and ulcerative colitis (UC)). Based on microarray technology, expression levels of 863 miRNAs were determined for whole blood samples from 40 CD and 36 UC patients and compared to data from 38 healthy controls (HC). To further discriminate between disease-specific and general inflammation we included miRNA expression data from other inflammatory diseases (inflammation controls (IC): 24 chronic obstructive pulmonary disease (COPD), 23 multiple sclerosis, 38 pancreatitis and 45 sarcoidosis cases) as well as 70 healthy controls from previous studies. Classification problems considering 2, 3 or 4 groups were solved using different types of penalized support vector machines (SVMs). The resulting models were assessed regarding sparsity and performance and a subset was selected for further investigation. Measured by the area under the ROC curve (AUC) the corresponding median holdout-validated accuracy was estimated as ranging from 0.75 to 1.00 (including IC) and 0.89 to 0.98 (excluding IC), respectively. In combination, the corresponding models provide tools for the distinction of CD and UC as well as CD, UC and HC with expected classification error rates of 3.1 and 3.3%, respectively. These results were obtained by incorporating not more than 16 distinct miRNAs. Validated target genes of these miRNAs have been previously described as being related to IBD. For others we observed significant enrichment for IBD susceptibility loci identified in earlier GWAS. These results suggest that the proposed miRNA signature is of relevance for the etiology of IBD. Its diagnostic value, however, should be further evaluated in large, independent, clinically well characterized cohorts.

Hypothalamic Inflammation in Human Obesity is Mediated by Environmental and Genetic Factors

Obesity is associated with hypothalamic inflammation (HI) in animal models. In the current study, we examined the mediobasal hypothalamus (MBH) of 57 obese human subjects and 54 age- and sex- matched nonobese control subjects by MRI and analyzed the T2 hyperintensity as a measure of HI. Obese subjects exhibited T2 hyperintensity in the left but not the right MBH, which was strongly associated with systemic low-grade inflammation. MRS revealed the number of neurons in the left hypothalamic region to be similar in obese versus control subjects, suggesting functional but not structural impairment due to the inflammatory process. To gain mechanistic insights, we performed nutritional analysis and 16S rDNA microbiome sequencing, which showed that high-fat diet induces reduction of Parasutterella sp. in the gut, which is significantly correlated with MBH T2 hyperintensity. In addition to these environmental factors, we found subjects carrying common polymorphisms in the JNK or the MC4R gene to be more susceptible to HI. Finally, in a subgroup analysis, bariatric surgery had no effect on MBH T2 hyperintensity despite inducing significant weight loss and improvement of peripheral insulin sensitivity. In conclusion, obesity in humans is associated with HI and disturbances in the gut-brain axis, which are influenced by both environmental and genetic factors.

Serologic Anti-GP2 Antibodies Are Associated with Genetic Polymorphisms, Fibrostenosis, and Need for Surgical Resection in Crohn's Disease.

Background:The presentation of Crohns disease (CD) is heterogeneous and often leads to serious complications and need for surgery. We tested serum anti-zymogen granule glycoprotein 2 (GP2) antibodies, including its novel isoform alpha, for association with genetic variants, diagnosis, disease stratification, and prediction of CD courses in a combined cross-sectional and cohort study. Methods:Serum samples of 303 CD, 108 ulcerative colitis, 72 other inflammatory gastrointestinal diseases, and 206 controls without predominant gastrointestinal diseases controls (HC) were tested for the presence of Anti-GP2 and Anti-Saccharomyces cervisiae (ASCA) by enzyme-linked immunosorbent assay. Genetic analysis was performed using the Illumina Immunochip. Results:GP2 IgA and IgG had the highest discriminatory capability for CD versus ulcerative colitis and CD versus inflammatory gastrointestinal diseases. We identified an association of GP2 IgA and IgG each with 5 distinct single-nucleotide polymorphisms. Levels of anti-GP2 IgG were moderately associated with ileal disease location. Interestingly, both, anti-GP2 IgA and IgG were exclusively associated with the occurrence of stenosis and need for surgery, independently of disease location, but not with fistulizing CD, early disease onset or disease activity. ASCA IgG and IgA were qualitatively and quantitatively linked to CD, CD complications, and need for surgery. Increased levels of ASCA IgG and IgA and positivity for ASCA IgG, but neither levels nor positivity for GP2 IgG or IgA were predictive of the earlier occurrence of complications or surgery. Conclusions:Anti-GP2 antibodies may aid as a tool for diagnosis and differentiation of CD and could indicate a more complicated CD course.

Impact of high risk thrombophilia status on recurrence among children and adults with VTE: An observational multicenter cohort study.

BACKGROUND Antithrombin [AT]-, protein C [PC]- or protein S [PS]-deficiency [D] constitutes a major risk factor for venous thromboembolism [VTE]. Primary study objective was to evaluate if the clinical presentation at first VTE onset differs between children and adults and to compare the individual recurrence risk among patients with respect to age at onset and their thrombophilia status ATD, PCD or PSD. METHODS/PATIENTS/RESULTS In 137 of 688 consecutively enrolled pediatric and adult VTE patients we calculated the absolute risk of VTE recurrence and event-free-survival adjusted for thrombophilia and positive family VTE history. At first VTE children manifested i) with a lower rate of pulmonary embolism, ii) a higher rate of cerebral vascular events or multiple VTEs, and iii) showed a higher proportion of unprovoked VTE compared to adolescents and adults. Adult patients reported more often a positive VTE history compared to younger study participants. The adjusted odds of recurrence in adults was 2.05 compared to children. CONCLUSION At disease manifestation children and adults differ with respect to i) thrombotic locations, ii) percentage of unprovoked versus provoked VTE, and iii) different rates of positive VTE family histories. Furthermore, adults showed a two-fold increase risk of VTE recurrence compared to children.

Low-Frequency Blood Group Antigens in Switzerland

Background: High-frequency blood group antigens (HFA) are present in >90% of the human population, according to some reports even in >99% of individuals. Therefore, patients lacking HFA may become challenging for transfusion support because compatible blood is hardly found, and if the patient carries alloantibodies, the cross-match will be positive with virtual every red cell unit tested. Methods: In this study, we applied high-throughput blood group SNP genotyping on >37,000 Swiss blood donors, intending to identify homozygous carriers of low-frequency blood group antigens (LFA). Results: 326 such individuals were identified and made available to transfusion specialists for future support of patients in need of rare blood products. Conclusion: Thorough comparison of minor allele frequencies using population genetics revealed heterogeneity of allele distributions among Swiss blood donors which may be explained by the topographical and cultural peculiarities of Switzerland. Moreover, geographically localized donor subpopulations are described which contain above-average numbers of individuals carrying rare blood group genotypes.

Serum anti-glycan-antibodies in relatives of patients with inflammatory bowel disease

Background Serum anti-glycan antibodies are a promising tool for differential diagnosis, disease stratification and prediction of Crohn’s disease (CD). To investigate possible heritability of the markers we assessed the presence of serum anti-glycan antibodies in affected and unaffected relatives of patients with CD. Methods Serum samples of 169 IBD patients of the German inflammatory bowel disease (IBD) network (140 CD & 29 Ulcerative colitis (UC)), 349 relatives of CD patients, 63 relatives of UC patients and 46 healthy controls were tested for the presence of anti-glycan antibodies by ELISA in a blinded fashion. Clinical data of the IBD patients and controls were available. Results A higher proportion of non-affected CD relatives was positive for anti-glycan antibodies compared to healthy subjects. No inheritance of a specific pattern of anti-glycan antibodies could be detected. No difference in marker expression depending on the degree of relationship in the non-affected relatives was noted and the presence of family history did not lead to a difference in marker levels in the affected CD subjects. Conclusions Non-affected CD relatives had a higher frequency of anti-glycan antibodies compared to healthy subjects. This difference was mild and was found to be true for the overall reactivity to glycan antigens, but not for specific patterns. This may indicate an inherited mechanism resulting in a non-specific increased reactivity to microbial antigens in IBD.

Evaluation of variable selection methods for random forests and omics data sets

Abstract Machine learning methods and in particular random forests are promising approaches for prediction based on high dimensional omics data sets. They provide variable importance measures to rank predictors according to their predictive power. If building a prediction model is the main goal of a study, often a minimal set of variables with good prediction performance is selected. However, if the objective is the identification of involved variables to find active networks and pathways, approaches that aim to select all relevant variables should be preferred. We evaluated several variable selection procedures based on simulated data as well as publicly available experimental methylation and gene expression data. Our comparison included the Boruta algorithm, the Vita method, recurrent relative variable importance, a permutation approach and its parametric variant (Altmann) as well as recursive feature elimination (RFE).  In our simulation studies, Boruta was the most powerful approach, followed closely by the Vita method. Both approaches demonstrated similar stability in variable selection, while Vita was the most robust approach under a pure null model without any predictor variables related to the outcome. In the analysis of the different experimental data sets, Vita demonstrated slightly better stability in variable selection and was less computationally intensive than Boruta. In conclusion, we recommend the Boruta and Vita approaches for the analysis of high-dimensional data sets. Vita is considerably faster than Boruta and thus more suitable for large data sets, but only Boruta can also be applied in low-dimensional settings.