Direct ionic stress sensing and mitigation by the transcription factor NFAT5.

Homeostatic control of intracellular ionic strength is essential for protein, organelle and genome function, yet mechanisms that sense and enable adaptation to ionic stress remain poorly understood in animals. We find that the transcription factor NFAT5 directly senses solution ionic strength using a C-terminal intrinsically disordered region. Both in intact cells and in a purified system, NFAT5 forms dynamic, reversible biomolecular condensates in response to increasing ionic strength. This self-associative property, conserved from insects to mammals, allows NFAT5 to accumulate in the nucleus and activate genes that restore cellular ion content. Mutations that reduce condensation or those that promote aggregation both reduce NFAT5 activity, highlighting the importance of optimally tuned associative interactions. Remarkably, human NFAT5 alone is sufficient to reconstitute a mammalian transcriptional response to ionic or hypertonic stress in yeast. Thus NFAT5 is both the sensor and effector of a cell-autonomous ionic stress response pathway in animal cells.

Gene-teratogen interactions influence the penetrance of birth defects by altering Hedgehog signaling strength.

Birth defects result from interactions between genetic and environmental factors, but the mechanisms remain poorly understood. We find that mutations and teratogens interact in predictable ways to cause birth defects by changing target cell sensitivity to Hedgehog (Hh) ligands. These interactions converge on a membrane protein complex, the MMM complex, that promotes degradation of the Hh transducer Smoothened (SMO). Deficiency of the MMM component MOSMO results in elevated SMO and increased Hh signaling, causing multiple birth defects. In utero exposure to a teratogen that directly inhibits SMO reduces the penetrance and expressivity of birth defects in Mosmo-/- embryos. Additionally, tissues that develop normally in Mosmo-/- embryos are refractory to the teratogen. Thus, changes in the abundance of the protein target of a teratogen can change birth defect outcomes by quantitative shifts in Hh signaling. Consequently, small molecules that re-calibrate signaling strength could be harnessed to rescue structural birth defects.

Avoidable Serum Potassium Testing in the Cardiac ICU: Development and Testing of a Machine-Learning Model.

OBJECTIVES: To create a machine-learning model identifying potentially avoidable blood draws for serum potassium among pediatric patients following cardiac surgery. DESIGN: Retrospective cohort study. SETTING: Tertiary-care center. PATIENTS: All patients admitted to the cardiac ICU at Boston Children's Hospital between January 2010 and December 2018 with a length of stay greater than or equal to 4 days and greater than or equal to two recorded serum potassium measurements. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We collected variables related to potassium homeostasis, including serum chemistry, hourly potassium intake, diuretics, and urine output. Using established machine-learning techniques, including random forest classifiers, and hyperparameter tuning, we created models predicting whether a patient's potassium would be normal or abnormal based on the most recent potassium level, medications administered, urine output, and markers of renal function. We developed multiple models based on different age-categories and temporal proximity of the most recent potassium measurement. We assessed the predictive performance of the models using an independent test set. Of the 7,269 admissions (6,196 patients) included, serum potassium was measured on average of 1 (interquartile range, 0-1) time per day. Approximately 96% of patients received at least one dose of IV diuretic and 83% received a form of potassium supplementation. Our models predicted a normal potassium value with a median positive predictive value of 0.900. A median percentage of 2.1% measurements (mean 2.5%; interquartile range, 1.3-3.7%) was incorrectly predicted as normal when they were abnormal. A median percentage of 0.0% (interquartile range, 0.0-0.4%) critically low or high measurements was incorrectly predicted as normal. A median of 27.2% (interquartile range, 7.8-32.4%) of samples was correctly predicted to be normal and could have been potentially avoided. CONCLUSIONS: Machine-learning methods can be used to predict avoidable blood tests accurately for serum potassium in critically ill pediatric patients. A median of 27.2% of samples could have been saved, with decreased costs and risk of infection or anemia.

Cholesterol accessibility at the ciliary membrane controls hedgehog signaling.

Previously we proposed that transmission of the hedgehog signal across the plasma membrane by Smoothened is triggered by its interaction with cholesterol (Luchetti et al., 2016). But how is cholesterol, an abundant lipid, regulated tightly enough to control a signaling system that can cause birth defects and cancer? Using toxin-based sensors that distinguish between distinct pools of cholesterol, we find that Smoothened activation and Hedgehog signaling are driven by a biochemically-defined, small fraction of membrane cholesterol, termed accessible cholesterol. Increasing cholesterol accessibility by depletion of sphingomyelin, which sequesters cholesterol in complexes, amplifies Hedgehog signaling. Hedgehog ligands increase cholesterol accessibility in the membrane of the primary cilium by inactivating the transporter-like protein Patched 1. Trapping this accessible cholesterol blocks Hedgehog signal transmission across the membrane. Our work shows that the organization of cholesterol in the ciliary membrane can be modified by extracellular ligands to control the activity of cilia-localized signaling proteins.

Discovery of gene regulatory elements through a new bioinformatics analysis of haploid genetic screens.

The systematic identification of regulatory elements that control gene expression remains a challenge. Genetic screens that use untargeted mutagenesis have the potential to identify protein-coding genes, non-coding RNAs and regulatory elements, but their analysis has mainly focused on identifying the former two. To identify regulatory elements, we conducted a new bioinformatics analysis of insertional mutagenesis screens interrogating WNT signaling in haploid human cells. We searched for specific patterns of retroviral gene trap integrations (used as mutagens in haploid screens) in short genomic intervals overlapping with introns and regions upstream of genes. We uncovered atypical patterns of gene trap insertions that were not predicted to disrupt coding sequences, but caused changes in the expression of two key regulators of WNT signaling, suggesting the presence of cis-regulatory elements. Our methodology extends the scope of haploid genetic screens by enabling the identification of regulatory elements that control gene expression.

G protein-coupled receptors control the sensitivity of cells to the morphogen Sonic Hedgehog.

The morphogen Sonic Hedgehog (SHH) patterns tissues during development by directing cell fates in a concentration-dependent manner. The SHH signal is transmitted across the membrane of target cells by the heptahelical transmembrane protein Smoothened (SMO), which activates the GLI family of transcription factors through a mechanism that is undefined in vertebrates. Using CRISPR-edited null alleles and small-molecule inhibitors, we systematically analyzed the epistatic interactions between SMO and three proteins implicated in SMO signaling: the heterotrimeric G protein subunit GαS, the G protein-coupled receptor kinase 2 (GRK2), and the GαS-coupled receptor GPR161. Our experiments uncovered a signaling mechanism that modifies the sensitivity of target cells to SHH and consequently changes the shape of the SHH dose-response curve. In both fibroblasts and spinal neural progenitors, the loss of GPR161, previously implicated as an inhibitor of basal SHH signaling, increased the sensitivity of target cells across the entire spectrum of SHH concentrations. Even in cells lacking GPR161, GRK2 was required for SHH signaling, and Gαs, which promotes the activation of protein Kinase A (PKA), antagonized SHH signaling. We propose that the sensitivity of target cells to Hedgehog morphogens, and the consequent effects on gene expression and differentiation outcomes, can be controlled by signals from G protein-coupled receptors that converge on Gαs and PKA.

CRISPR Screens Uncover Genes that Regulate Target Cell Sensitivity to the Morphogen Sonic Hedgehog.

To uncover regulatory mechanisms in Hedgehog (Hh) signaling, we conducted genome-wide screens to identify positive and negative pathway components and validated top hits using multiple signaling and differentiation assays in two different cell types. Most positive regulators identified in our screens, including Rab34, Pdcl, and Tubd1, were involved in ciliary functions, confirming the central role for primary cilia in Hh signaling. Negative regulators identified included Megf8, Mgrn1, and an unannotated gene encoding a tetraspan protein we named Atthog. The function of these negative regulators converged on Smoothened (SMO), an oncoprotein that transduces the Hh signal across the membrane. In the absence of Atthog, SMO was stabilized at the cell surface and concentrated in the ciliary membrane, boosting cell sensitivity to the ligand Sonic Hedgehog (SHH) and consequently altering SHH-guided neural cell-fate decisions. Thus, we uncovered genes that modify the interpretation of morphogen signals by regulating protein-trafficking events in target cells.

Comparative genetic screens in human cells reveal new regulatory mechanisms in WNT signaling.

The comprehensive understanding of cellular signaling pathways remains a challenge due to multiple layers of regulation that may become evident only when the pathway is probed at different levels or critical nodes are eliminated. To discover regulatory mechanisms in canonical WNT signaling, we conducted a systematic forward genetic analysis through reporter-based screens in haploid human cells. Comparison of screens for negative, attenuating and positive regulators of WNT signaling, mediators of R-spondin-dependent signaling and suppressors of constitutive signaling induced by loss of the tumor suppressor adenomatous polyposis coli or casein kinase 1α uncovered new regulatory features at most levels of the pathway. These include a requirement for the transcription factor AP-4, a role for the DAX domain of AXIN2 in controlling ß-catenin transcriptional activity, a contribution of glycophosphatidylinositol anchor biosynthesis and glypicans to R-spondin-potentiated WNT signaling, and two different mechanisms that regulate signaling when distinct components of the ß-catenin destruction complex are lost. The conceptual and methodological framework we describe should enable the comprehensive understanding of other signaling systems.