Heterogeneity of Distal Convoluted Tubule Cells

A key objective in systems biology is to comprehensively map gene expression in every cell type in the human body to gain better insights into their functions. Recent advancements in single-cell genomics and computational techniques have facilitated the analysis of RNA transcripts expressed in individual cells, enabling a deeper understanding of cell type composition within complex organs. The kidney is a highly complex organ comprising over 43 cell types,¹ including distinct epithelial cells across 14 renal tubule segments, interstitial cells, immune cells, and more. Each cell type carries out unique functions often associated with various kidney diseases. Therefore, classifying these cell types and identifying their molecular signatures becomes crucial for understanding kidney function and disease progression. Despite conventional bulk RNA-seq and single-cell RNA-seq conducted at the whole kidney level providing aggregated expression or a rough estimation of the cell type composition, certain cell types essential for kidney functions remain notably understudied. This gap is partially attributed to their minority status and variation in enzymatic dissociation properties during the single-cell preparation procedure, affecting less-explored cell types, such as glomerular cell types (podocytes, mesangial cells, and parental epithelial cells), macula densa cells, Gli-positive mesenchymal stem cells, distal convoluted tubule cells, and others. To overcome this limitation, a targeted approach focusing on specific cell types before single-cell RNA-seq is an alternative solution.^2,3

In this issue of JASN, Su and colleagues⁴ employed a new targeted single-nucleus RNA-seq (snRNA-seq) method to profile the gene expression in the cells of distal convoluted tubule (DCT), a short renal tubule segment located between the postglomerular part of the cortical thick ascending limb (CTAL) and the connecting tubule (CNT). Despite being a short segment, major transport pathways are associated with the DCT, establishing its involvement in maintaining sodium, potassium, calcium, and magnesium homeostasis. Genetic variants altering the transport activity in DCT can result in inherited diseases such as Gitelman syndrome (GS) and familial hyperkalemic hypertension (FHHt), both linked to altered activity of the NaCl cotransporter (NCC, encoded by Slc12a3). NCC, exclusively expressed in DCT cells, is the target of thiazide diuretics used in the treatment of many forms of hypertension. In addition to NCC, inherited diseases can also arise from other genetic variants affecting transport processes in DCT cells, underscoring its central role in human physiology and disease. Although NCC is confined to DCT cells, the existence of heterogeneity within these cells is increasingly recognized. For instance, parvalbumin (Pvalb) is found in the early part of DCT, while calbindin (Calb1), ENaC-β (Scnn1b), Trpv5, and others are located more distally, separating the DCT into DCT1 and DCT2. Data from Su et al.⁴ coupled with an earlier article in JASN³ also using targeted RNA-seq show that the DCT has cellular heterogeneity that extends beyond the DCT1/DCT2 division.

The developmental origin of DCT cells also remains a topic of controversy. This arises from their location in the conjugation zone, where there is fusion of the nephron originating from mesenchymal mesenchyme and the collecting duct system derived from the ureteric bud.⁵ In addition, the DCT exhibits plasticity and undergoes structural remodeling in physiological perturbations. To address these issues and others, Su and colleagues conducted a high-resolution transcriptomic analysis of DCT cells.⁴

The authors introduced a clever approach to DCT cell enrichment. They generated a NCC-Cre-INTACT^flx/flx mouse line, which involves breeding a DCT-specific inducible NCC-Cre recombinase mouse line⁶ previously developed by the authors with the INTACT^flx/flx mouse line⁷ to produce DCT-INTACT mice. The INTACT (Isolation of Nuclei TAgged in Specific Cell Types) system relies on the expression of the SUN1-sfGFP-Myc fusion protein localized to nuclei only in cells expressing Cre recombinase. On tamoxifen administration, these mice specifically express GFP in the nuclear membrane of DCT cells. This approach is novel, allowing for the identification and isolation of DCT nuclei using fluorescence-activated nuclei sorting (FANS) for subsequent snRNA-seq analysis. Notably, snRNA-seq demonstrates comparable sensitivity with single-cell RNA-seq (scRNA-seq) while mitigating dissociation bias.⁸

Following the standard analysis procedure, the author observed several cell clusters, among which two major distinct cell types, namely DCT1 and DCT2, were revealed. In agreement with morphological studies,⁹ the proportion of DCT1 and DCT2 is close to 3:1, confirming the precision of the DCT-INTACT approach in faithfully representing all NCC-expressing DCT cells. The central question then emerged: Beyond the standard classification (DCT1=NCC⁺/Pvalb⁺/ENaC⁻; DCT2=NCC⁺/Pvalb⁻/ENaC⁺), what distinguishes DCT1 from DCT2 cells and to what extent do these cells differ from each other? To address these questions, the investigators identified the top 10 differential expressed genes in DCT1 (Erbb4, Egf, Trpm7, Fgf13, Col5a2, Umod, Ptgfr, Stk32b, Rtl4, Abca13) and DCT2 (Slc8a1, Arl15, Calb1, Slc2a9, Phactr1, Gls, S100g, Kl, Klk1, Egfem1) and generated a score for each other. DCT1 and DCT2 are apparently different based on the expression of these genes. This is also supported by pseudotime analysis, showing that DCT1 and DCT2 cells are positioned differently on the trajectory. Despite their distinct expression profiles, the transition between these cells is gradual. Indeed, the authors immunostained the microdissected DCT and confirmed this gradual transition, as there was no clear border between DCT1 and DCT2, with parvalbumin and calbindin expression intermingling in the transitional zone. These observations are consistent with previous studies, reconfirming that DCT cells are heterogeneous.

DCT expresses numerous transport proteins crucial for calcium and magnesium excretion, establishing its role as the primary site for active handling of these ions. The authors employed a scoring approach to dissect the varying abilities of calcium and magnesium handling between DCT1 and DCT2 cells. They curated a magnesium cassette with genes such as Trpm6, Trpm7, and Egf and a calcium cassette with genes such as Slc8a1, Calb1, and Vdr. The resulting magnesium and calcium scores highlight the enrichment of the magnesium cassette in DCT1, while DCT2 exhibits enrichment in the calcium cassette. Remarkably, these DCT1/DCT2 and magnesium/calcium scores exhibit similar patterns in previously published datasets, including the mouse-targeted scRNA-seq³ and human snRNA-seq data from the Kidney Precision Medicine Project.¹⁰ This not only underscores consistency across diverse datasets but also implies a comparable organization of the DCT between mouse and human. Notably, the cells within DCT1 also show some degree of heterogeneity. Although DCT1 cells share a fundamental set of genes, they are further separated into several subclusters exhibiting gradient expression of these genes. Similar subclusters are also seen within DCT2 cells. These findings introduce additional complexity to the overall understanding of the DCT. The authors propose that the heterogeneity observed could be attributed to mixed origins and reciprocal induction between mesenchymal mesenchyme and ureteric bud, resulting in a continuum of cell types in this region.

Aside from the artificial cell clusters found in the study, a particularly interesting group of cells, similar to the previous study,³ exhibiting proliferating potential were found. These cells, constituting <1% of the total DCT cell population, express markers associated with proliferation, such as Mki67, Top2a, and Cenpp. An intriguing question arises regarding the potential role of these cells in physiological perturbation and their contribution to the plasticity of the DCT. Future investigations, particularly involving dietary perturbation or loop diuretics treatment, could offer valuable insights and generate significant interest within the research community.

The results from this study, together with earlier studies, provide insights into the heterogeneity of DCT cells and their functional characteristics. The findings have implications for understanding the development and plasticity of the DCT. Future directions may involve exploring the functional consequences of the identified heterogeneity, both within and between sexes, across various stages of kidney development and disease progression. Moreover, these discoveries have implications for potential applications in the field of kidney organoid/tubuloid development. Notably, DCT is located in a region known for frequent alternative splicing events, exemplified by genes such as Wnk1, Stk39, Tsc22d1, and others.¹¹ It is important to note that the snRNA-seq and scRNA-seq methods, while offering expression profiles, lack information on alternative splicing. As a result, the distribution of these isoforms within DCT1 and DCT2 remains unknown. These isoforms may result in different protein structures, adding more complexity to DCT functions. Ultimately, integrating multi-omics, alongside transcriptomics, offers the potential for a more comprehensive understanding of DCT function.

Disclosures

All authors have nothing to disclose.

Funding

None.

Author Contributions

Conceptualization: Lihe Chen.

Writing – original draft: Lihe Chen, Adrián R. Murillo-De-Ozores.