Discovery of a non-canonical GRHL1 binding site using deep convolutional and recurrent neural networks.

BACKGROUND: Transcription factors regulate gene expression by binding to transcription factor binding sites (TFBSs). Most models for predicting TFBSs are based on position weight matrices (PWMs), which require a specific motif to be present in the DNA sequence and do not consider interdependencies of nucleotides. Novel approaches such as Transcription Factor Flexible Models or recurrent neural networks consequently provide higher accuracies. However, it is unclear whether such approaches can uncover novel non-canonical, hitherto unexpected TFBSs relevant to human transcriptional regulation. RESULTS: In this study, we trained a convolutional recurrent neural network with HT-SELEX data for GRHL1 binding and applied it to a set of GRHL1 binding sites obtained from ChIP-Seq experiments from human cells. We identified 46 non-canonical GRHL1 binding sites, which were not found by a conventional PWM approach. Unexpectedly, some of the newly predicted binding sequences lacked the CNNG core motif, so far considered obligatory for GRHL1 binding. Using isothermal titration calorimetry, we experimentally confirmed binding between the GRHL1-DNA binding domain and predicted GRHL1 binding sites, including a non-canonical GRHL1 binding site. Mutagenesis of individual nucleotides revealed a correlation between predicted binding strength and experimentally validated binding affinity across representative sequences. This correlation was neither observed with a PWM-based nor another deep learning approach. CONCLUSIONS: Our results show that convolutional recurrent neural networks may uncover unanticipated binding sites and facilitate quantitative transcription factor binding predictions.

Single-cell transcriptomics reveals common epithelial response patterns in human acute kidney injury.

BACKGROUND: Acute kidney injury (AKI) occurs frequently in critically ill patients and is associated with adverse outcomes. Cellular mechanisms underlying AKI and kidney cell responses to injury remain incompletely understood. METHODS: We performed single-nuclei transcriptomics, bulk transcriptomics, molecular imaging studies, and conventional histology on kidney tissues from 8 individuals with severe AKI (stage 2 or 3 according to Kidney Disease: Improving Global Outcomes (KDIGO) criteria). Specimens were obtained within 1-2 h after individuals had succumbed to critical illness associated with respiratory infections, with 4 of 8 individuals diagnosed with COVID-19. Control kidney tissues were obtained post-mortem or after nephrectomy from individuals without AKI. RESULTS: High-depth single cell-resolved gene expression data of human kidneys affected by AKI revealed enrichment of novel injury-associated cell states within the major cell types of the tubular epithelium, in particular in proximal tubules, thick ascending limbs, and distal convoluted tubules. Four distinct, hierarchically interconnected injured cell states were distinguishable and characterized by transcriptome patterns associated with oxidative stress, hypoxia, interferon response, and epithelial-to-mesenchymal transition, respectively. Transcriptome differences between individuals with AKI were driven primarily by the cell type-specific abundance of these four injury subtypes rather than by private molecular responses. AKI-associated changes in gene expression between individuals with and without COVID-19 were similar. CONCLUSIONS: The study provides an extensive resource of the cell type-specific transcriptomic responses associated with critical illness-associated AKI in humans, highlighting recurrent disease-associated signatures and inter-individual heterogeneity. Personalized molecular disease assessment in human AKI may foster the development of tailored therapies.

Nuclei Isolation from Adult Mouse Kidney for Single-Nucleus RNA-Sequencing.

The kidneys regulate diverse biological processes such as water, electrolyte, and acid-base homeostasis. Physiological functions of the kidney are executed by multiple cell types arranged in a complex architecture across the corticomedullary axis of the organ. Recent advances in single-cell transcriptomics have accelerated the understanding of cell type-specific gene expression in renal physiology and disease. However, enzyme-based tissue dissociation protocols, which are frequently utilized for single-cell RNA-sequencing (scRNA-seq), require mostly fresh (non-archived) tissue, introduce transcriptional stress responses, and favor the selection of abundant cell types of the kidney cortex resulting in an underrepresentation of cells of the medulla. Here, we present a protocol that avoids these problems. The protocol is based on nuclei isolation at 4 °C from frozen kidney tissue. Nuclei are isolated from a central piece of the mouse kidney comprised of the cortex, outer medulla, and inner medulla. This reduces the overrepresentation of cortical cells typical for whole-kidney samples for the benefit of medullary cells such that data will represent the entire corticomedullary axis at sufficient abundance. The protocol is simple, rapid, and adaptable and provides a step towards the standardization of single-nuclei transcriptomics in kidney research.

Technologies for profiling the impact of genomic variants on transcription factor binding.

Transcription factors (TFs) bind DNA in a sequence-specific manner and thereby regulate target gene expression. TF binding and its regulatory activity is highly context dependent, and is not only determined by specific cell types or differentiation stages but also relies on other regulatory mechanisms, such as DNA and chromatin modifications. Interactions between TFs and their DNA binding sites are critical mediators of phenotypic variation and play important roles in the onset of disease. A continuously growing number of studies therefore attempts to elucidate TF:DNA interactions to gain knowledge about regulatory mechanisms and disease-causing variants. Here we summarize how TF-binding characteristics and the impact of variants can be investigated, how bioinformatic tools can be used to analyze and predict TF:DNA binding, and what additional information can be obtained from the TF protein structure.

Hypertonicity-Affected Genes Are Differentially Expressed in Clear Cell Renal Cell Carcinoma and Correlate with Cancer-Specific Survival.

The heterogeneity of renal cell carcinoma (RCC) subtypes reflects the cell type of origin in the nephron, with consequences for therapy and prognosis. The transcriptional cues that determine segment-specific gene expression patterns are poorly understood. We recently showed that hypertonicity in the renal medulla regulates nephron-specific gene expression. Here, we analyzed a set of 223 genes, which were identified in the present study by RNA-Seq to be differentially expressed by hypertonicity, for the prediction of cancer-specific survival (CSS). Cluster analyses of these genes showed discrimination between tumor and non-tumor samples of clear cell RCC (ccRCC). Refinement of this gene signature to a four-gene score (OSM score) through statistical analyses enabled prediction of CSS in ccRCC patients of The Cancer Genome Atlas (TCGA) (n = 436) in univariate (HR = 4.1; 95% CI: 2.78-6.07; p = 4.39 × 10-13), and multivariate analyses including primary tumor (T); regional lymph node (N); distant metastasis (M); grading (G)(p = 2.3 × 10-5). The OSM score could be validated in an independent ccRCC study (n = 52) in univariate (HR = 1.29; 95% CI = 1.05-1.59; p = 0.011) and multivariate analyses (p = 0.016). Cell culture experiments using RCC cell lines demonstrated that the expression of the tumor suppressor ELF5 could be restored by hypertonicity. The innovation of our novel gene signature is that these genes are physiologically regulated only by hypertonicity, thereby providing the possibility to be targeted for therapy.

Claudins in the Renal Collecting Duct.

The renal collecting duct fine-tunes urinary composition, and thereby, coordinates key physiological processes, such as volume/blood pressure regulation, electrolyte-free water reabsorption, and acid-base homeostasis. The collecting duct epithelium is comprised of a tight epithelial barrier resulting in a strict separation of intraluminal urine and the interstitium. Tight junctions are key players in enforcing this barrier and in regulating paracellular transport of solutes across the epithelium. The features of tight junctions across different epithelia are strongly determined by their molecular composition. Claudins are particularly important structural components of tight junctions because they confer barrier and transport properties. In the collecting duct, a specific set of claudins (Cldn-3, Cldn-4, Cldn-7, Cldn-8) is expressed, and each of these claudins has been implicated in mediating aspects of the specific properties of its tight junction. The functional disruption of individual claudins or of the overall barrier function results in defects of blood pressure and water homeostasis. In this concise review, we provide an overview of the current knowledge on the role of the collecting duct epithelial barrier and of claudins in collecting duct function and pathophysiology.