F. Anthony San Lucas

Tumor microenvironment derived exosomes pleiotropically modulate cancer cell metabolism

Cancer-associated fibroblasts (CAFs) are a major cellular component of tumor microenvironment in most solid cancers. Altered cellular metabolism is a hallmark of cancer, and much of the published literature has focused on neoplastic cell-autonomous processes for these adaptations. We demonstrate that exosomes secreted by patient-derived CAFs can strikingly reprogram the metabolic machinery following their uptake by cancer cells. We find that CAF-derived exosomes (CDEs) inhibit mitochondrial oxidative phosphorylation, thereby increasing glycolysis and glutamine-dependent reductive carboxylation in cancer cells. Through 13C-labeled isotope labeling experiments we elucidate that exosomes supply amino acids to nutrient-deprived cancer cells in a mechanism similar to macropinocytosis, albeit without the previously described dependence on oncogenic-Kras signaling. Using intra-exosomal metabolomics, we provide compelling evidence that CDEs contain intact metabolites, including amino acids, lipids, and TCA-cycle intermediates that are avidly utilized by cancer cells for central carbon metabolism and promoting tumor growth under nutrient deprivation or nutrient stressed conditions. DOI: http://dx.doi.org/10.7554/eLife.10250.001

Integrated annotation and analysis of genetic variants from next-generation sequencing studies with variant tools

MOTIVATION Storing, annotating and analyzing variants from next-generation sequencing projects can be difficult due to the availability of a wide array of data formats, tools and annotation sources, as well as the sheer size of the data files. Useful tools, including the GATK, ANNOVAR and BEDTools can be integrated into custom pipelines for annotating and analyzing sequence variants. However, building flexible pipelines that support the tracking of variants alongside their samples, while enabling updated annotation and reanalyses, is not a simple task. RESULTS We have developed variant tools, a flexible annotation and analysis toolset that greatly simplifies the storage, annotation and filtering of variants and the analysis of the underlying samples. variant tools can be used to manage and analyze genetic variants obtained from sequence alignments, and the command-line driven toolset could be used as a foundation for building more sophisticated analytical methods. AVAILABILITY AND IMPLEMENTATION variant tools consists of two command-line driven programs vtools and vtools_report. It is freely available at http://varianttools.sourceforge.net, distributed under a GPL license. CONTACT bpeng@mdanderson.org.

Incorporating prior knowledge into Gene Network Study

MOTIVATION A major goal in genomic research is to identify genes that may jointly influence a biological response. From many years of intensive biomedical research, a large body of biological knowledge, or pathway information, has accumulated in available databases. There is a strong interest in leveraging these pathways to improve the statistical power and interpretability in studying gene networks associated with complex phenotypes. This prior information is a valuable complement to large-scale genomic data such as gene expression data generated from microarrays. However, it is a non-trivial task to effectively integrate available biological knowledge into gene expression data when reconstructing gene networks. RESULTS In this article, we developed and applied a Lasso method from a Bayesian perspective, a method we call prior Lasso (pLasso), for the reconstruction of gene networks. In this method, we partition edges between genes into two subsets: one subset of edges is present in known pathways, whereas the other has no prior information associated. Our method assigns different prior distributions to each subset according to a modified Bayesian information criterion that incorporates prior knowledge on both the network structure and the pathway information. Simulation studies have indicated that the method is more effective in recovering the underlying network than a traditional Lasso method that does not use the prior information. We applied pLasso to microarray gene expression datasets, where we used information from the Pathway Commons (PC) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) as prior information for the network reconstruction, and successfully identified network hub genes associated with clinical outcome in cancer patients. AVAILABILITY The source code is available at http://nba.uth.tmc.edu/homepage/liu/pLasso.

Genomic Landscape of Colorectal Mucosa and Adenomas

The molecular basis of the adenoma-to-carcinoma transition has been deduced using comparative analysis of genetic alterations observed through the sequential steps of intestinal carcinogenesis. However, comprehensive genomic analyses of adenomas and at-risk mucosa are still lacking. Therefore, our aim was to characterize the genomic landscape of colonic at-risk mucosa and adenomas. We analyzed the mutation profile and copy number changes of 25 adenomas and adjacent mucosa from 12 familial adenomatous polyposis patients using whole-exome sequencing and validated allelic imbalances (AI) in 37 adenomas using SNP arrays. We assessed for evidence of clonality and performed estimations on the proportions of driver and passenger mutations using a systems biology approach. Adenomas had lower mutational rates than did colorectal cancers and showed recurrent alterations in known cancer driver genes (APC, KRAS, FBXW7, TCF7L2) and AIs in chromosomes 5, 7, and 13. Moreover, 80% of adenomas had somatic alterations in WNT pathway genes. Adenomas displayed evidence of multiclonality similar to stage I carcinomas. Strong correlations between mutational rate and patient age were observed in at-risk mucosa and adenomas. Our data indicate that at least 23% of somatic mutations are present in at-risk mucosa prior to adenoma initiation. The genomic profiles of at-risk mucosa and adenomas illustrate the evolution from normal tissue to carcinoma via greater resolution of molecular changes at the inflection point of premalignant lesions. Furthermore, substantial genomic variation exists in at-risk mucosa before adenoma formation, and deregulation of the WNT pathway is required to foster carcinogenesis. Cancer Prev Res; 9(6); 417–27. ©2016 AACR.

Genetic risk factors for the development of osteonecrosis in children under age 10 treated for acute lymphoblastic leukemia

Osteonecrosis is a dose-limiting toxicity in the treatment of pediatric acute lymphoblastic leukemia (ALL). Prior studies on the genetics of osteonecrosis have focused on patients ≥10 years of age, leaving the genetic risk factors for the larger group of children <10 years incompletely understood. Here, we perform the first evaluation of genetic risk factors for osteonecrosis in children <10 years. The discovery cohort comprised 82 cases of osteonecrosis and 287 controls treated on Childrens Oncology Group (COG) standard-risk ALL protocol AALL0331 (NCT00103285, https://clinicaltrials.gov/ct2/show/NCT00103285), with results tested for replication in 817 children <10 years treated on COG protocol AALL0232 (NCT00075725, https://clinicaltrials.gov/ct2/show/NCT00075725). The top replicated single nucleotide polymorphisms (SNPs) were near bone morphogenic protein 7 [BMP7: rs75161997, P = 5.34 × 10(-8) (odds ratio [OR] 15.0) and P = .0498 (OR 8.44) in the discovery and replication cohorts, respectively] and PROX1-antisense RNA1 (PROX1-AS1: rs1891059, P = 2.28 × 10(-7) [OR 6.48] and P = .0077 [OR 3.78] for the discovery and replication cohorts, respectively). The top replicated nonsynonymous SNP, rs34144324, was in a glutamate receptor gene (GRID2, P = 8.65 × 10(-6) [OR 3.46] and P = .0136 [OR 10.8] in the discovery and replication cohorts, respectively). In a meta-analysis, the BMP7 and PROX1-AS1 variants (rs75161997 and rs1891059, respectively) met the significance threshold of <5 × 10(-8). Top replicated SNPs were enriched in enhancers active in mesenchymal stem cells, and analysis of annotated genes demonstrated enrichment in glutamate receptor and adipogenesis pathways. These data may provide new insights into the pathophysiology of osteonecrosis.

Improving the sensitivity of sample clustering by leveraging gene co-expression networks in variable selection

BackgroundMany variable selection techniques have been proposed for the clustering of gene expression data. While these methods tend to filter out irrelevant genes and identify informative genes that contribute to a clustering solution, they are based on criteria that do not consider the potential interactive influence among individual genes. Motivated by ensemble clustering, there is a strong interest in leveraging the structure of gene networks for gene selection, so that the relationship information between genes can be effectively utilized, while the selected genes are expected to preserve all the possible clustering structures in the data.ResultsWe present a new filter method that uses the gene connectivity in the gene co-expression network as the evaluation criteria for variable selection. The gene connectivity measures the importance of the genes in term of their expression similarity with others in the co-expression network. The hard threshold and soft threshold transformations are employed to construct the gene co-expression networks. Both simulation studies and real data analysis have shown that the network based on soft thresholding is more effective in selecting relevant variables and provides better clustering results compared to the hard thresholding transformation and two other canonical filter methods for variable selection. Furthermore, a new module analysis approach is proposed to reveal the higher order organization of the gene space, where the genes of a module share significant topological similarity and are associated with a consensus partition of the sample space. We demonstrate that the identified modules can lead to biologically meaningful sample partitions that might be missed by other methods.ConclusionsBy leveraging the structure of gene co-expression network, first we propose a variable selection method that selects individual genes with top connectivity. Both simulation studies and real data application have demonstrated that our method has better performance in terms of the reliability of the selected genes and sample clustering results. In addition, we propose a module recovery method that can help discover novel sample partitions that might be hidden when performing clustering analyses using all available genes. The source code of our program is available at http://nba.uth.tmc.edu/homepage/liu/netVar/.

An S/T-Q cluster domain census unveils new putative targets under Tel1/Mec1 control.

BackgroundThe cellular response to DNA damage is immediate and highly coordinated in order to maintain genome integrity and proper cell division. During the DNA damage response (DDR), the sensor kinases Tel1 and Mec1 in Saccharomyces cerevisiae and ATM and ATR in human, phosphorylate multiple mediators which activate effector proteins to initiate cell cycle checkpoints and DNA repair. A subset of kinase substrates are recognized by the S/T-Q cluster domain (SCD), which contains motifs of serine (S) or threonine (T) followed by a glutamine (Q). However, the full repertoire of proteins and pathways controlled by Tel1 and Mec1 is unknown.ResultsTo identify all putative SCD-containing proteins, we analyzed the distribution of S/T-Q motifs within verified Tel1/Mec1 targets and arrived at a unifying SCD definition of at least 3 S/T-Q within a stretch of 50 residues. This new SCD definition was used in a custom bioinformatics pipeline to generate a census of SCD-containing proteins in both yeast and human. In yeast, 436 proteins were identified, a significantly larger number of hits than were expected by chance. These SCD-containing proteins did not distribute equally across GO-ontology terms, but were significantly enriched for those involved in processes related to the DDR. We also found a significant enrichment of proteins involved in telophase and cytokinesis, protein transport and endocytosis suggesting possible novel Tel1/Mec1 targets in these pathways. In the human proteome, a wide range of similar proteins were identified, including homologs of some SCD-containing proteins found in yeast. This list also included high concentrations of proteins in the Mediator, spindle pole body/centrosome and actin cytoskeleton complexes.ConclusionsUsing a bioinformatic approach, we have generated a census of SCD-containing proteins that are involved not only in known DDR pathways but several other pathways under Tel1/Mec1 control suggesting new putative targets for these kinases.

Genomic Landscape of Atypical Adenomatous Hyperplasia Reveals Divergent Modes to Lung Adenocarcinoma

There is a dearth of knowledge about the pathogenesis of premalignant lung lesions, especially for atypical adenomatous hyperplasia (AAH), the only known precursor for the major lung cancer subtype adenocarcinoma (LUAD). In this study, we performed deep DNA and RNA sequencing analyses of a set of AAH, LUAD, and normal tissues. Somatic BRAF variants were found in AAHs from 5 of 22 (23%) patients, 4 of 5 of whom had matched LUAD with driver EGFR mutations. KRAS mutations were present in AAHs from 4 of 22 (18%) of patients. KRAS mutations in AAH were only found in ever-smokers and were exclusive to BRAF-mutant cases. Integrative analysis revealed profiles expressed in KRAS-mutant cases (UBE2C, REL) and BRAF-mutant cases (MAX) of AAH, or common to both sets of cases (suppressed AXL). Gene sets associated with suppressed antitumor (Th1; IL12A, GZMB) and elevated protumor (CCR2, CTLA-4) immune signaling were enriched in AAH development and progression. Our results reveal potentially divergent BRAF or KRAS pathways in AAH as well as immune dysregulation in the pathogenesis of this premalignant lung lesion. Cancer Res; 77(22); 6119-30. ©2017 AACR.

Mutational Heterogeneity in APC and KRAS Arises at the Crypt Level and Leads to Polyclonality in Early Colorectal Tumorigenesis

Purpose: The majority of genomic alterations causing intratumoral heterogeneity (ITH) in colorectal cancer are thought to arise during early stages of carcinogenesis as a burst but only after truncal mutations in APC have expanded a single founder clone. We have investigated if the initial source of ITH is consequent to multiple independent lineages derived from different crypts harboring distinct truncal APC and driver KRAS mutations, thus challenging the prevailing monoclonal monocryptal model. Experimental Design: High-depth next-generation sequencing and SNP arrays were performed in whole-lesion extracts of 37 familial adenomatous polyposis colorectal adenomas. Also, ultrasensitive genotyping of hotspot mutations of APC and KRAS was performed using nanofluidic PCRs in matched bulk biopsies (n = 59) and crypts (n = 591) from 18 adenomas and seven carcinomas and adjacent normal tissues. Results: Multiple co-occurring truncal APC and driver KRAS alterations were uncovered in whole-lesion extracts from adenomas and subsequently confirmed to belong to multiple clones. Ultrasensitive genotyping of bulk biopsies and crypts revealed novel undetected APC mutations that were prominent among carcinomas, whereas abundant wild-type APC crypts were detected in adenomas. KRAS mutational heterogeneity within crypts was evident in both adenomas and carcinomas with a higher degree of concordance between biopsy and crypt genotyping in carcinomas. Nonrandom heterogeneity among crypts was also observed. Conclusions: The striking degree of nonrandom intercrypt heterogeneity in truncal and driver gene mutations observed in adenomas and carcinomas is consistent with a polycryptal model derived from multiple independent initiation linages as the source of early ITH in colorectal carcinogenesis. Clin Cancer Res; 23(19); 5936–47. ©2017 AACR.

Haploscope: A tool for the graphical display of haplotype: Structure in populations

Patterns of linkage disequilibrium are often depicted pictorially by using tools that rely on visualizations of raw data or pairwise correlations among individual markers. Such approaches can fail to highlight some of the more interesting and complex features of haplotype structure. To enable natural visual comparisons of haplotype structure across subgroups of a population (e.g. isolated subpopulations or cases and controls), we propose an alternative visualization that provides a novel graphical representation of haplotype frequencies. We introduce Haploscope, a tool for visualizing the haplotype cluster frequencies that are produced by statistical models for population haplotype variation. We demonstrate the utility of our technique by examining haplotypes around the LCT gene, an example of recent positive selection, in samples from the Human Genome Diversity Panel. Haploscope, which has flexible options for annotation and inspection of haplotypes, is available for download at http://scheet.org/software.