Research Roundup: December 6, 2019
Faster bacterial diagnostics, massive-scale analysis methodologies, a new view into ADHD and autism genetics, and more.
Faster bacterial diagnostics, massive-scale analysis methodologies, a new view into ADHD and autism genetics, and more.
Welcome to the December 6, 2019 installment of Research Roundup, a recurring snapshot of recent studies published by scientists at the Broad Institute and their collaborators.
While every cell in our body contains the same genetic sequence, it is the enhancers that control how genes are expressed in different cell types, ensuring, for example, that a liver cell does not go rogue and start switching on kidney genes. However, the ability to determine and predict which enhancers regulate which genes remains elusive. Charlie Fulco, Joseph Nasser, Jesse Engreitz, and institute president and founding director Eric Lander, along with colleagues from the Broad and elsewhere, describe in Nature Genetics both an experimental technique that can determine which enhancers regulate which genes, and a model that predicts enhancer-gene connections across the genome. Since prior studies have connected enhancer mutations with disease, these new tools will be important for understanding human health.
A quicker way to diagnose bacterial infections could help patients recover sooner and prevent the spread of drug-resistant microbes, which kill 35,000 Americans each year. Roby Bhattacharyya, core institute member and Infectious Disease and Microbiome Program (IDMP) co-director Deborah Hung, and colleagues developed a new diagnostic approach, called GoPhAST-R, which combines genotypic and phenotypic testing to determine a bacterium’s antibiotic susceptibility. GoPhAST-R looks for patterns in antibiotic-induced gene expression and also identifies key resistance genes to distinguish susceptible from resistant strains. Described in Nature Medicine, the approach can provide results with 94 to 99 percent accuracy in less than four hours, compared to 28 to 40 hours using standard clinical laboratory methods.
The rules governing how transcription factors (TFs) work with gene promoters to control gene expression, cell phenotype, and cell state remain murky, in part due to issues of scale. In Nature Biotechnology, Carl de Boer, core institute member and Klarman Cell Observatory director Aviv Regev, and colleagues unveiled the Gigantic Parallel Reporter Assay (GPRA): a method that merges machine learning with a laboratory system that measures how TFs interact with more than 100 million randomly-synthesized gene promoter sequences in yeast to influence gene expression. GPRA reveals key features of TF-promoter binding, and provides an opportunity to create complex, comprehensive models for studying how genetic variation affects gene expression and disease risk.
To make the most of all the single-cell RNA-sequencing (scRNA-seq) studies out there, researchers need to be able to bring together data from a variety of tissues, data sources, sequencing platforms, and more. Ilya Korsunsky, institute member Soumya Raychaudhuri in the Program in Medical and Population Genetics (MPG), and colleagues have developed Harmony, an algorithm that allows scientists to integrate scRNAseq data from multiple datasets. In Nature Methods, they demonstrate Harmony's ability to 1) work with large datasets, 2) identify both broad and fine-grained cell populations in integrated data, 3) process data generated in complex experiments, and 4) work with data from multiple experimental platforms. Harmony's R package is available on GitHub.
Potential prion disease therapeutics aim to lower prion protein (PrP) in the brain, but current methods to measure PrP in the cerebrospinal fluid (CSF) don’t capture protein fragments or different conformations. Eric Vallabh Minikel, Eric Kuhn, Sonia Vallabh, institute scientist and Proteomics Platform director Steven Carr, and colleagues developed a mass spectrometry-based method based on multiple reaction monitoring that precisely measures PrP peptide concentrations in humans and other model species. Reporting in Molecular and Cellular Proteomics, they found that CSF PrP is reduced as the disease progresses, so dose-finding studies for PrP-lowering drugs should focus on presymptomatic at-risk individuals. Read more in a press release from the American Society for Biochemistry and Molecular Biology.
Regulatory T cells (Tregs) can weaken anti-tumor immune responses and therefore are associated with poor prognosis of several cancers. In order to better understand the role of Tregs in tumor development, researchers used single-cell RNA sequencing to map the diversity of these cells throughout tumor development in a genetically engineered mouse model of lung adenocarcinoma. In Cell Reports, the team, led by Amy Li, Rebecca Herbst, David Canner, Aviv Regev, Cancer Program senior associate member Tyler Jacks, and colleagues, provides a high resolution view of Tregs diversity in the tumor microenvironment, thus highlighting potential pathways for therapeutic interventions.
Human kidney organoids grown from patient-derived induced pluripotent stem (iPS) cells are a promising new tool for the development of much-needed precision therapies. To learn how reproducible these organoids are across iPS cells lines, Ayshwarya Subramanian, Eriene-Heidi Sidhom, Maheswarareddy Emani, institute member and Kidney Disease Initiative director Anna Greka, and colleagues analyzed roughly 450,000 single cells in kidney organoids derived from four iPS cell lines and compared them to single cell profiles from human adult and fetal kidneys. The team found that organoid composition and development were faithful surrogates of human kidney tissue, and that transplanting organoids into mice further improved the organoids’ quality. Learn more in Nature Communications.
Little is known about the role of sphingolipid metabolism in regulating inflammation of the central nervous system (CNS) and in diseases such as multiple sclerosis. Julian Avila-Pacheco, associate member Francisco Quintana, institute scientist and Metabolomics Platform senior director Clary Clish, and colleagues used a combination of proteomic, metabolomic, transcriptomic, and in vivo genetic perturbation studies to find out the effects of sphingolipid metabolism in CNS cells called astrocytes. Their findings, reported in Cell, define a novel mechanism that drives pro-inflammatory astrocyte activities, outline a new role for mitochondrial antiviral signaling protein in CNS inflammation, and identify sphingolipid metabolism as a promising therapeutic target in CNS inflammation.
Plasmodium falciparum has rapidly developed resistance to every malaria drug in clinical use. Figuring out the parasite’s molecular escape routes early on during drug development could help researchers come up with better drugs. To address this issue, institute member Dyann Wirth in the IDMP and her team have devised a method to predict Plasmodium resistance mechanisms, which they describe in Science Translational Medicine. The researchers exposed the parasite, both in vitro and in infected mice, to compounds that block malarial dihydroorotate dehydrogenase (DHODH). They then selected the resistant organisms and sequenced their genomes. The team found similar rapid emergence of resistance and common mutations shared by resistant parasites in vitro and those in mice. The researchers conclude that selecting for resistance in vitro can predict resistance in vivo and suggest that this approach could be used to triage potential new drugs.
Proteins participate in functional pathways and form a complex set of interactions to drive a cell’s behavior. Deciphering this “interactome” is crucial for understanding the mechanisms that drive biology. While scientists have created a valuable “reference interactome” that generalizes these interactions into a single resource, such a tool cannot provide information specific to distinct cell types. Shahin Mohammadi, Jose Davila-Velderrain, and Epigenomics Program associate member Manolis Kellis describe in Cell Systems a computational framework (SCINET) that can analyze this interactome at single-cell resolution. Using scRNA-seq, SCINET reconstructs interactomes in individual cells, allowing researchers to identify perturbed single-cell interactions under a variety of conditions.
Amyotrophic lateral sclerosis (ALS) is a late-onset neurodegenerative disease, for which genetics is known to be a contributing risk factor. To discover novel genes related to ALS, a team led by Sali Farhan and institute member Benjamin Neale in the MPG analyzed exomes from 3,864 patients and 7,839 healthy individuals, the largest ALS exome case–control study to date. The team observed rare protein-truncating genetic variants in ALS cases, as well as associations with known ALS genes and with a novel gene, DNAJC7. The summary statistics of the ALS genetic data have been made available via the ALS Knowledge Portal. Check out the full story in Nature Neuroscience.
Autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD) have substantial genetic components, but gathering large cohorts for genetic analyses has been a challenge for both. A team led by Kyle Satterstrom, institute member and MPG co-director Mark Daly, and colleagues in Denmark's iPSYCH research initiative used DNA from archived bloodspots to analyze the exomes of approximately 8,000 children with ASD and/or ADHD and 5,000 controls to better understand the genetic architecture of these disorders. The researchers found that ASD and ADHD have a similar burden of truncating variants in constrained genes, and identified truncating variants in the MAP1A gene as associated with risk for the disorders. Learn more in Nature Neuroscience and a press release from iPSYCH.
To learn more about research conducted at the Broad, visit broadinstitute.org/publications, and keep an eye on broadinstitute.org/news.
Infectious Disease and Microbiome Program
Program in Medical and Population Genetics
New diagnostic approach rapidly identifies the right antibiotics