Long Noncoding RNAs in Single Cells

Long noncoding RNAs (lncRNAs) refer to transcripts that do not encode proteins and are more than 200 nucleotides long. A length threshold of 500 nucleotides has also been suggested.¹ The human genome includes tens of thousands of genes that encode lncRNAs. Several lncRNAs have been shown to have potent effects on transcriptional and post-transcriptional regulation as well as on other cellular functions.² For instance, the lncRNA XIST (X-inactive specific transcript) is pivotal in silencing genes on the inactive X chromosome by coating the chromosome and recruiting protein complexes. HOTTIP (HOXA distal transcript antisense RNA) interacts with the HOXA (homeobox A) gene cluster through chromatin folding and recruits chromatin-modifying complexes to the locus. Dysregulation of this process is implicated in leukemogenesis. On activation of the estrogen receptor, the lncRNA encoded by the enhancer of NRIP1 (nuclear receptor interacting protein 1) recruits cohesin to promote the formation of short-range and long-range chromatin loops, thereby activating the transcription of NRIP1 and TFF1 (trefoil factor 1).

Numerous studies have explored the role of lncRNAs in kidney function, with a select few providing robust evidence.³ For instance, ablation of the lncRNA Hoxb3os (HOXB cluster antisense RNA 1) worsened cyst growth in two orthologous Pkd1 mouse models.⁴ Overexpression of lncRNA H19 (H19 imprinted maternally expressed transcript) in mouse kidneys was shown to attenuate ischemic AKI.⁵ Knockout of H19 did not have a significant effect, which the author attributed to the low basal abundance of H19 in the kidney. Human genetic studies further hint at the involvement of lncRNAs in kidney-related traits, with around a quarter of the single nucleotide polymorphisms associated with human BP located within lncRNA genes.⁶

LncRNAs can affect multiple target molecules or pathways, and it is often difficult to predict the function of a specific lncRNA. The apparent promiscuity, i.e., a regulatory mechanism potentially affecting multiple targets, is not unique to lncRNAs. Most signal transduction pathways, transcriptional factors, microRNAs, and other regulatory mechanisms are promiscuous. However, these regulatory mechanisms, including lncRNAs, may achieve various degrees of regulatory specificity in a defined cellular context.^7,8 Similar to microRNAs, a lncRNA might target a single or a few genes or pathways in a specific cellular context, depending on the abundance of the lncRNA, the presence or abundance of its potential interacting partners, and the presence of parallel or orthogonal regulatory mechanisms. Understanding the tissue-specific roles of lncRNAs and the underlying mechanisms is vital for addressing misperceptions about noncoding RNA promiscuity and advancing noncoding RNA research.

Identifying cell type–specific expression patterns of lncRNAs is crucial for understanding tissue-specific functions of lncRNAs. The abundance of a lncRNA in a cell type and its coexpression with mRNAs and proteins, or the lack of, provide important clues to the functional role of the lncRNA. Single-cell RNA sequencing (RNA-seq) analysis detects RNAs in individual cells and enables the construction of gene expression profiles in more cell types, subtypes, and states than can be physically purified. However, lncRNAs are often underrepresented by commonly used single-cell RNA-seq methods because of their generally lower abundance compared with mRNAs in many tissues including the kidney.⁹

In this issue of JASN, Kim and colleagues presented a study in which capture probes were used to enrich single-cell RNA-seq libraries for lncRNAs before sequencing.¹⁰ These probes were designed to target over 10,000 annotated mouse lncRNAs, enhancing the detection of lncRNAs per cell by up to 30%, increasing the read counts of lncRNAs, and boosting the identification of differentially expressed lncRNAs between cell types by several fold. The investigators applied the method to examine several mouse tissues with a focus on the kidney. By integrating these data with single-cell mRNA expression profiles and existing single-cell chromatin accessibility data, they constructed potential gene regulatory networks for lncRNAs that were differentially expressed between cell types or experimental conditions. Notably, changes of lncRNA expression in kidney cell types, including glomerular cells, were observed between mice aged 6–8 weeks and 21 months. The genes with greater expression in older mice were primarily associated with immune response, while those with lower expression were associated with energy metabolism.

The study has created a valuable resource of cell type–specific lncRNA expression profiles in the kidney and several other organs in mice and highlighted the potential physiological importance of lncRNAs in the kidney. This resource offers numerous opportunities for further research into lncRNAs, potentially unveiling new aspects of kidney physiology and disease regulation. The kidney is composed of a remarkably large number of physiologically distinct cell types that work together to perform the essential functions of the organ. While most kidney cells are terminally differentiated, some can revert to less differentiated states under stress, playing a role in disease progression. It would be important to understand if, and which, lncRNAs influence the development, maintenance, alteration, or loss of specific cellular characteristics in the kidney (Figure 1).

Figure 1.

Single-cell lncRNA profiling may help to reveal the diverse functions of lncRNAs in the kidney. GRN, gene regulatory network; lncRNA, long noncoding RNA; TF, transcription factor.

Many lncRNAs are polyadenylated and can be detected by the single-cell RNA-seq method used in this study, which utilizes the poly(A) tail of a transcript for sequence capture. However, not all lncRNAs have poly(A) tails, depending on how the lncRNAs are generated. Indeed, over a quarter of lncRNAs expressed in the kidney outer medulla of rats lack a poly(A) tail.⁹ Emerging techniques that facilitate whole transcriptome detection at the single-cell level promise to offer deeper insights into all transcript classes within the kidney.

Most lncRNAs lack sequence conservation between human and rodents. Although some lncRNAs may exhibit structural or functional conservation, demonstrating the relevance of rodent lncRNAs to human conditions is challenging when sequence conservation is absent. Identifying whether lncRNAs maintain sequence or functional similarity across humans, rodents, and other model organisms is crucial for advancing lncRNA research.

Finally, common DNA sequence variants associated with human traits are often found within the noncoding regions of the genome. For example, over 20% of haplotypes in linkage disequilibrium with single nucleotide polymorphisms associated with human BP are located more than 10 kbp from any protein-coding gene.⁶ Understanding how these noncoding sequence variants influence biological processes remains one of the most pressing challenges in human genetic research. lncRNAs are among the epigenetic mechanisms potentially mediating the effects of these noncoding variants on associated traits, including those related to kidney function.

Disclosures

Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/JSN/E698.

Funding

M. Liang: National Heart, Lung, and Blood Institute (HL121233 and HL149620) and National Institute of Diabetes and Digestive and Kidney Diseases (DK129964).

Author Contributions

Conceptualization: Mingyu Liang.

Visualization: Qiongzi Qiu.

Writing – original draft: Mingyu Liang, Qiongzi Qiu.

Writing – review & editing: Mingyu Liang, Qiongzi Qiu.