sign in sign up
<<
Home , Collecting duct system, Loop of Henle, Medullary ray (anatomy), Podocyte

REVIEW www.jasn.org

Renal-Tubule Epithelial Nomenclature for Single-Cell RNA-Sequencing Studies

Lihe Chen,1 Jevin Z. Clark,1 Jonathan W. Nelson,2 Brigitte Kaissling,3 David H. Ellison ,2 and Mark A. Knepper1

1Epithelial Systems Biology Laboratory, Systems Biology Center, National Heart, Lung, and Institute, National Institutes of Health, Bethesda, Maryland; 2Division of Nephrology and Hypertension, Oregon Health & Science University, Portland, Oregon; and 3Institute of , University of Zurich, Zurich, Switzerland

JASN 30: 1358–1364, 2019. doi: https://doi.org/10.1681/ASN.2019040415

The assignment of cell type names to also include information on the per- the International Union of Physiologic individual cell clusters in single-cell centage of each epithelial cell type in Sciences and published in in RNA-sequencing (RNA-seq; including mouse kidney and on widely accepted International in 1988.3 The 1988 nomen- single-nucleus RNA-seq) studies is a mRNA/ markers that are consid- clature was focused on what to call each critical bioinformatic step because re- ered prerequisites for identification of renal segment, and not neces- searchers will need to be able to reliably individual cell types. sarily what to call individual cell types. map the pertinent transcriptomes to The mammalian renal tubule is made Accordingly, we refocused the recom- other types of information in the litera- up of at least 14 segments, containing at mended terminology to the individual ture. Ambiguous or redundant cell-type least 16 distinct epithelial cell types.4–6 cell types present in the various tubule names would impede such mapping Each cell type has its own characteristic segments. tasks. The solution to this problem is to set of cellular functions (mostly related Before presenting the list of recom- create “controlled vocabularies”1,2 that to transport and metabolism), which mended cell-type terminology, we sum- can be preferentially used in reports of have been elucidated largely over the marize, in a greatly simplified manner, single-cell RNA-seq data and other types past 50 years since the development of the current procedures for mapping of -omic data at a single-cell level. In single-tubule microdissection approaches gene expression lists from single-cell this review, we provide such a controlled by Burg et al.7 and the expansion of the RNA-seq to the various known cell types. vocabulary for the epithelial cells that micropuncture technique to mamma- A single-cell RNA-seq dataset contains make up the mammalian renal tubule. lian physiology.8,9 Furthermore, each (for each cell) a list of expressed genes The nomenclature list was not produced cell type has its own unique transcrip- together with some measure of their de novo in this review, but instead repre- tome as elucidated by recent RNA-seq mRNA expression levels. The initial sents an adaptation of a prior, widely ac- studies carried out at the level of single task, carried out independently of cell cepted nomenclature list for renal tubule microdissected tubules6 and single type assignment, is to cluster the cells segments published in 1988 in Kidney cells.10–26 The new transcriptomic data into groups that are similar on the basis International.3 This was done because promises to expand and sharpen our un- of the lists of their expressed mRNAs historical consistency is important to as- derstanding of renal tubule function, as along with their expression levels. These “ ” sure that names assigned to clusters in the data are mapped to cellular processes clusters are then mapped to cell-type single-cell RNA-seq studies mesh well responsible for the macroscopic function names that are assigned on the basis of with the “pregenomic” 20th century lit- of each renal tubule segment. erature, which provides robust and well To allow this mapping to be done ac- validated information on kidney cell curately, it is imperative that the cell-type Published online ahead of print. Publication date structure and function. The 1988 article terminology used for single-tubule and available at www.jasn.org. offered several alternative names for single-cell gene expression studies be Correspondence: Dr. Mark A. Knepper, Epithelial some of the segments and cell types. consistent with terminology used in the Systems Biology Laboratory, Systems Biology Our strategy here is to propose nomen- functional and biochemical studies done Center, National Heart, Lung, and Blood Institute, fl National Institutes of Health, MSC-1603, 10 Center claturethatre ects the most common in the past. Here, we summarize recom- Drive, Bethesda, MD 20892-1603. Email: knep@ usage from 1988 to the present in each mended terminology for renal tubule helix.nih.gov case. In this report, we not only describe epithelial cells derived from prior recom- Copyright © 2019 by the American Society of preferred names for each cell type, but mendations by the Renal Commission of Nephrology

1358 ISSN : 1046-6673/3008-1358 JASN 30: 1358–1364, 2019 www.jasn.org REVIEW representation of cell-type selective Ontology (RECON) webpage (https:// the latter two junctions are easily recog- marker genes. The problem we are ad- hpcwebapps.cit.nih.gov/ESBL/Database/ nizable macroscopically, but the corti- dressing in this review is that there is RECOn/). Figure 1 shows both a short-loop comedullary junction is obscure, at least presently no generally accepted “con- and a long-loop superimposed on in rodents, without benefitofastereo- trolled vocabulary” for renal tubule epi- regional boundaries of the kidney. The struc- microscope. Here, we present the rationale thelial cell types, i.e., there is no standard tures that define the regional boundaries are for the terms presented, some alternative list of acceptable terms. Consequently, the arcuate arteries (corticomedullary junc- terminology, and some alternative terms to terminology has been inconsistent among tion), the transition from proximal straight be avoided. individual single-cell RNA-seq articles,10–26 (PSTs) to descending thin limbs and the correct mappings to functional and (DTLs) of the loops of Henle (inner/outer biochemical studies done in the pregenomic stripe of outer medulla junction) and the era are uncertain. transition from the ascending thin limbs NOMENCLATURE Recommended terms for renal tubule of Henle of the long-loop to epithelial cell types in kidneys of mature the medullary thick ascending limbs There are two sets of terms commonly mice, rats, or humans are given in Figure 1 (MTALs) of the loops of Henle (inner- used to describe proximal tubule cells. and are listed on the Renal Epithelial Cell outer medullary junction).4 Generally, One divides the proximal tubule into three

Figure 1. Renal tubule cell nomenclature. The scheme shows the connection of both a short-looped nephron and a long-looped nephron to the . The numbers point to different renal tubule segments and the cell types that they contain are listed on the rightof thescheme.Dashedlines indicateregional boundaries.The corticomedullary junction isdefinedbythearcuate arteries(not showninthisdiagram).Thedivisionbetweentheouter stripe and inner stripe of the outer medulla is defined by the transitions from PSTs to DTLs of the loops of Henle. The junction between the outer medulla and inner medulla is defined by the transitions from the ascending thin limbs (ATL) of Henle of the long-loop nephrons to the MTALs of the loops of Henle. The division between the cortical labyrinth and cortical medullary rays is indicated but not labeled. Drawing is on the basis of the original Renal Commission of the In- ternational Union of Physiologic Sciences publication.3

JASN 30: 1358–1364, 2019 Renal Epithelial Cell Nomenclature 1359 REVIEW www.jasn.org segments called S1, S2, and S3 on the basis topologically distinct from the other three considered to include the -2 of differences in morphology, topology, thin limb segments.29 All of the thin limb (AQP2) expressing cells of the collecting and location in the kidney.27,28 Function- segments are difficult to microdissect from duct. Hence, the term distal tubule is am- ally, the S2 is recognized as the segment the kidney and difficult to study by mi- biguous and should be avoided in RNA-seq responsible for para-aminohippurate se- cropuncture, limiting the amount of studies. Another similar term is “distal cretion and, by implication, responsible functional information we have about nephron,” commonly used to designate for general organic anion and cation se- them. the portion of the renal tubule from the cretion.28 A limitation of this classifica- In contrast, the thick ascending limb through the end of the inner tion is the lack of a discrete transition (TAL) is relatively easy to study in iso- medullary collecting duct (IMCD), thereby from one to the other morphologically. lation and has well defined functional comprised of multiple cell types. This lack of clear transitions suggests roles.30–32 The TAL is normally divided Rabbits exhibit a distinct transition that there are probably intermediate cell into the MTAL and the CTAL on the point at the end of the DCT that allow types, which would preclude clear-cut basis of a difference in morphology, lo- unambiguous dissection of a uniform clustering into three distinct groups of cation, and function.30 The transition DCT (Figure 2).39,40 Rodents, cells. The S3 segment is found chieflyin between these cell types occurs in the however, exhibit a transition region41 that the outer stripe of the outer medulla, al- outer stripe of the outer medulla. Near includes cells expressing markers nor- lowing it to be separated from S1 and the end of the TAL, the CTAL comes into mally attributed to downstream seg- S2 simply by selecting the outer stripe contact with its own . At this ments, such as the three subunits of as starting material for single-cell isola- point, special epithelial cells called mac- the epithelial channel (ENaC) tion. An alternative classification divides ula densa cells transduce feedback sig- (Figure 2). The DCT in human kidney the proximal tubule into the proximal nals to the , regulating also has such a transition region (Figure 2), convoluted tubule (PCT) and PST on GFR and renin production.33 The mac- leading investigators to divide the DCT the basis of the course of the tubule (con- ula densa consists of only 14–24 cells per into two subsegments, referred to as voluted versus straight) and the location nephron in rodent species,34,35 and so has DCT1 and DCT2.41,42 The rodent and in the cortex (the PCT is found in the been elusive in single-cell RNA-seq stud- human DCT1 appears equivalent to the cortical labyrinth whereas the PST is ies. In several recent studies using single- rabbit DCTwith a single unique cell type. found in the of the cortex cell RNA-seq, clusters were labeled “loop In addition, the DCT2 region contains and the outer stripe of the outer medulla). of Henle cells.” This is an ambiguous label interspersed intercalated cells (ICs; see Thus, in tissue sections and microdissec- because it could apply to any of seven cell below). Whether there are uniquely de- tion experiments, the two can be readily types discussed in this section. Thus, the fined DCT2 cells remains an unresolved discriminated. The PCT includes the S1 term cells should generally issue and is certainly addressable in fu- (early PCT) and S2 (late PCT), whereas be avoided in cluster labeling. ture single-cell RNA-seq studies. This the PST includes the S2 (cortical PST) scientific question is complicated by the and S3 (medullary PST). marked plasticity of the DCT segment, undergoing changes in cell number and VERSUS DISTAL TUBULE tubule length under different physiologic LOOP OF HENLE circumstances.43,44 It seems possible, NOMENCLATURE The distal convoluted tubule (DCT) is a therefore, that the cell types present in short segment that follows the CTAL a theDCTcanchangewithphysiologic The loop of Henle spans from the DTL to few cells beyond the macula densa. It state. An ancillary question is: “What cells the cortical thick ascending limb (CTAL) transports NaCl at a high rate in a regu- in this region are derived from the meta- (Figure 1). It is made up of six distinct lated manner, and is the chief target for nephric mesenchyme and what cells are segments corresponding to seven mor- .36–38 Morphologically, derived from the ?” Careful phologically distinct cell types. Three of DCT cells are quite distinct from any studies are needed to address this question. these are the cells of the three DTL seg- other cell type in the kidney, on the basis ments, viz. DTL1, DTL2, and DTL3.3,29 of the unique subapical location of its The DTL1 segment makes up the entire nucleus. The term “distal convoluted THE COLLECTING DUCT SYSTEM DTL in short-loop nephrons, whereas the tubule” should be distinguished from DTL of long-loop nephrons is made up of the nonspecificterm“distal tubule.” The collecting duct system consists of DTL2 in the outer medulla and DTL3 in the The latter is defined differently in differ- the connecting tubule (CNT), cortical inner medulla. The transition between the ent contexts, sometimes referring to all collecting duct (CCD), outer medul- two is indistinct, occurring various distances tubule segments beyond the proximal lary collecting duct (OMCD), and into the inner medulla. The fourth thin limb tubule, sometimes denoting the region IMCD. These collecting duct segments segment, the ascending thin limb of the between the macula densa and the initial contain different cell types that are in- loop of Henle is morphologically and collecting tubule38 and sometimes terspersed in the same epithelium.

1360 JASN JASN 30: 1358–1364, 2019 www.jasn.org REVIEW

Figure 2. Distributions of transporters in the renal tubule segments from CTAL to CCD in rabbit, mouse, rat, and human kidney. A uniform DCT epithelium expressing NCC (Slc12a3) is present in rabbit, whereas the DCT in mouse, rat, and human has a transi- tional region (DCT2) expressing NCC and ENaC subunits (Scnn1a, Scnn1b,andScnn1g). Thus, in rodents and humans, but not rabbit, the DCT is divided into DCT1 and DCT2. The distribution of ICs (mostly type B ICs expressing [Slc26a4]) is in- dicated by large dots. AQP2 (Aqp2), -dependent channel ; ENaC (Scnn1a, Scnn1b,andScnn1g), -sensitive epithelial Na channel; G, glomerulus; NKCC2 (Slc12a1), -sensitive Na-K-2Cl cotransporter; NCC (Slc12a3), thiazide-sensitive Na-Cl cotransporter; TRPM6 (Trpm6), magnesium channel of distal convoluted tubule, TRPV5 (Trpv5), calcium channel of CNT cells.

These cell types are usually classified as PCs uniquely express the water chan- amounts. These data are taken from principal cells (PCs) and ICs. The latter are nel AQP2. AQP2-expressing cells can be Clark et al.,58 who compiled morphom- of two general types, the type A (or a)ICs divided into four morphologically and etric data from multiple sources including that secrete and the type B (or b)ICs functionally distinct cell types: CNT cells, the classic comparative morphology that secrete base () and reab- CCD PCs, OMCD PCs, and IMCD cells. study of Sperber59 (available at https:// sorb chloride. The collecting duct seg- Many kidney morphologists consider the esbl.nhlbi.nih.gov/Databases/Sperber/). ments appear to be quite plastic, and CNT cell a unique cell type different from As expected, proximal tubule cells (com- even in adult kidneys PCs and ICs were the PC on the basis of morphologic dif- bination of S1, S2, and S3) are the most reported to be interconvertible.13,45–54 In ferences.57 Likewise, some kidney mor- abundant cell type amounting to 43% of accordance with that observation, hybrid phologists do not consider the IMCD all renal tubule epithelial cells and 65% of PC/IC cells have been reported.13,14 cell a PC, largely because it is the only the total renal tubule protein mass. The Type A ICs predominate in the cell type throughout most of the IMCD second most abundant cell type is the OMCD, whereas type B ICs predominate length. TAL cell (combination of MTAL and in the CCD and CNT.55 There have been CTAL), amounting to 21% cell count reports of noncanonical ICs (so called and 15% protein mass. DCT cells ac- “non-A, non-B cells”)thataredistin- ABSOLUTE AND RELATIVE count for 7% of tubule epithelial cells guished on the basis of the distribution NUMBERS OF EACH RENAL and AQP2-expressing cells (CNT, PC, of acid-base transporters between the TUBULE CELL TYPE and IMCD) account for 12%. In the pre- apical and basolateral plasma mem- sent context, can be consid- brane,56 but specificproteinmarkers Assignment of names to individual cell ered a “‘minority” cell type accounting for these cells that could be used to sort clusters in single-cell RNA-seq studies for only 3% of cells. If one makes this cells within RNA-seq datasets were not can benefit fromknowledge of the relative calculation for cortex rather than whole well established. It is possible that these number of each cell type in a kidney. Fig- kidney, the percent of proximal tubule represent hybrid type A IC/type B IC cells, ure 3 shows the absolute and relative cells increases to 57% and the percent although existing single-cell RNA-seq numbers of each renal tubule cell type of podocytes increases to 4% (see Supple- studies failed to identify cells with markers (including podocytes) in a mouse kidney mental Figure 1). Thus, despite their for both.14 as well the absolute and relative protein importance to overall kidney function,

JASN 30: 1358–1364, 2019 Renal Epithelial Cell Nomenclature 1361 REVIEW www.jasn.org

Figure 3. Cell type and protein mass distribution among mouse renal epithelial cells. (A) The contributions of individual mouse renal epithelial cell types to whole kidney epithelial cell count (given at bottom). (B) The contributions of individual mouse renal epithelial cell types to whole kidney tubule protein mass (given at bottom). A-IC, type A intercalated cell; B-IC, type B intercalated cell; Thin Limbs, thin limbs of the loop of Henle (sum of DTL1, DTL2, DTL3, and ascending thin limbs). The data are from Clark et al.58 transcriptomes may be expec- to mediate the major function of the SUMMARY ted to be rare in whole-kidney shotgun DCT cells, namely reabsorption of single-cell RNA-seq studies and it may be NaCl, and has been found to be ex- In this review, we have provided a set of necessary to isolate glomeruli before cell pressed exclusively in DCT cells.36 terms for each renal tubule epithelial cell dissociation to get large numbers of po- Thus, this single marker is considered type, recommended for use in cluster nam- docyte transcriptomes.24 In general, iso- necessary and sufficient to identify a ing in single-cell RNA-seq studies. A con- lation of cells from solid tissue can distort DCT cell. Table 1 lists these sine-qua- sistent nomenclature is critical to ensure the proportions of cell types because non identifiers for the cell types de- accurate mapping of the pertinent tran- some cells are more fragile than others, scribed in this review, known before scriptomes to other types of information less easily dissociable than others, or less the invention of the RNA-seq tech- in the literature accrued over manydecades. metabolically stable than others. Conse- nique. An important characteristic of We have pointed out examples of am- quently, the distribution of cell types that such identifiers is that their mRNAs biguouscell-typenames(i.e.,loopof can be successfully analyzed by single-cell must be very abundant because (de- Henle cells and distal tubule cells) that RNA-seq are unlikely to match the actual pending on the platform) single-cell impede such mapping tasks. The con- distribution displayed in Figure 3. RNA-seq does not always capture trolled vocabulary introduced here is the full transcriptome of each cell likely to be useful for naming tasks (discussed by Clark et al.58) and it is that extend beyond single cell RNA- PROTEIN AND MRNA MARKERS important for discriminating markers seq studies as measurements in single FOR INDIVIDUAL CELL TYPES to be detectible at their native expres- cells expand to other modalities, e.g., sion levels. In Supplemental Figure 2 chromatin accessibility measurements Identification of the cell types corre- and Supplemental Appendix 1, we using ATAC-seq.60 In this report, we sponding to individual clusters in sin- show a set of additional markers found not only described preferred names gle-cell RNA-seq data depends in part in RNA-seq studies of microdissected for each cell type, but also included in- on matching the transcriptomes of rat renal tubule segments6 (panel A) formation on widely accepted mRNA/ the cells included in each cluster to prior in combination with mouse single-cell protein markers that are considered informationaboutwhatproteinsare RNA-seq14 (panel B). In theory, a com- pre-requisites for identification of indi- functionally important in the cells. For bination of these markers can be used vidual cell types and on the percentage example, the thiazide sensitive NaCl co- to specifically identify any renal tubule of each epithelial cell type in mouse transporter (Slc12a3) is well established cell type. kidney.

1362 JASN JASN 30: 1358–1364, 2019 www.jasn.org REVIEW

Table 1. Classic cell type selective markers mechanism: Evidence for the countercurrent hypothesis. Am J Physiol 196: 927–936, 1959 Cell Typea Gold Standard Marker(s) 10. Brunskill EW, Park JS, Chung E, Chen F, Podocyte Nphs1 (), Nphs2 () Magella B, Potter SS: Single cell dissection of Proximal S1 cell Slc5a2 (SGLT2) early : Multilineage prim- Proximal S2 cell Slc22a6 (OAT1) ing. Development 141: 3093–3101, 2014 Proximal S3 cell Agt (angiotensinogen) 11. Lu Y, Ye Y, Yang Q, Shi S: Single-cell RNA- All proximal tubule Slc34a1 (Na-Pi2), Lrp2 (megalin) sequence analysis of mouse glomerular DTL type 1 cell Slc14a2 (UT-A2) mesangial cells uncovers mesangial cell es- sential genes. Kidney Int 92: 504–513, 2017 DTL type 2 cell Slc14a2 (UT-A2) 12. Lu Y, Ye Y, Bao W, Yang Q, Wang J, Liu Z, DTL type 3 cell Slc14a2 (UT-A2) et al.: Genome-wide identification of genes Ascending thin limb cell Clcnka (chloride channel Ka) essential for podocyte cytoskeletons based – MTAL limb cell Slc12a1 (NKCC2), Umod (Tamm Horsfall protein) on single-cell RNA sequencing. Kidney Int CTAL cell Slc12a1, Umod 92: 1119–1129, 2017 Macula densa cell Slc12a1, Oxtr (oxytocin receptor), Nos1 (nitric oxide synthase 1) 13. Park J, Shrestha R, Qiu C, Kondo A, Huang S, Distal convoluted tubule cell Slc12a3 (NCC) Werth M, et al.: Single-cell transcriptomics CNT cell Aqp2 (aquaporin-2), Scnn1g (g-ENaC), Calb1 (calbindin) of the mouse kidney reveals potential cellu- PC (CCD) Aqp2, Scnn1g lar targets of kidney disease. Science 360: – PC (OMCD) Aqp2, Scnn1g 758 763, 2018 14. Chen L, Lee JW, Chou CL, Nair AV, Type A IC Slc4a1 (anion exchanger 1) Battistone MA, Paunescu TG, et al.: Tran- Type B IC Slc26a4 (pendrin) scriptomes of major renal collecting duct IMCD cell Aqp2, Slc14a2 (UT-A1) cell types in mouse identified by single-cell a Cell type markers identified in the pregenomic era. RNA-seq. Proc Natl Acad Sci U S A 114: E9989–E9998, 2017 ACKNOWLEDGMENTS Supplemental Appendix 1. Additional 15. Young MD, Mitchell TJ, Vieira Braga FA, Tran markers found in RNA-seq studies. MGB, Stewart BJ, Ferdinand JR, et al.: Single- cell transcriptomes from human kidneys reveal Dr. Chen, Dr. Clark, and Dr. Knepper the cellular identity of renal tumors. Science outlined the paper. Dr. Chen, Dr. Clark, 361: 594–599, 2018 and Dr. Knepper wrote the paper. Dr. Clark REFERENCES 16. Wang, P, Chen, Y, Yong, J, Cui, Y, Wang, R, made the webpage. Dr. Chen, Dr. Clark, and Wen, L, et al.: Dissecting the global dynamic molecular profiles of human fetal kidney de- Dr. Kaissling made the figures. All authors 1. Jupe, S, Jassal, B, Williams, M, Wu, G: A velopment by single-cell RNA sequencing. fi controlled vocabulary for pathway entities edited the text and approved the nal version. Cell Rep 24: 3554–3567.e3, 2018 and events. Database (Oxford) 2014: bau060, 17. Wu, H, Uchimura, K, Donnelly, EL, Kirita, Y, 2014 Morris, SA, Humphreys, BD: Comparative 2. Mayer G, Jones AR, Binz PA, Deutsch EW, analysis and refinement of human PSC- DISCLOSURES Orchard S, Montecchi-Palazzi L, et al.: Con- derived kidney organoid differentiation with trolled vocabularies and ontologies in pro- single-cell transcriptomics. Cell Stem Cell None. teomics: Overview, principles and practice. 23: 869–881.e8, 2018 Biochim Biophys Acta 1844[1 Pt A]: 98–107, 18. Wu H, Kirita Y, Donnelly EL, Humphreys BD: 2014 Advantages of single-nucleus over single- 3. Kriz W, Bankir L; The Renal Commission of the FUNDING cell RNA sequencing of adult kidney: Rare International Union of Physiological Sciences cell types and novel cell states revealed in Thework was primarily fundedbytheDivisionof (IUPS): A standard nomenclature for structures fibrosis. JAmSocNephrol30: 23–32, 2019 of the kidney. Kidney Int 33: 1–7, 1988 Intramural Research, National Heart, Lung, 19. Hochane M, van den Berg PR, Fan X, and Blood Institute (projects ZIA-HL001285 and 4. Kaissling B, Kriz W: Structural analysis of the Bérenger-Currias N, Adegeest E, Bialecka M, ZIA-HL006129, to Dr. Knepper). rabbit kidney. Adv Anat Embryol Cell Biol 56: et al.: Single-cell transcriptomics reveals gene – 1 123, 1979 expression dynamics of human fetal kidney 5.KnepperM,BurgM:Organizationofnephron development. PLoS Biol 17: e3000152, 2019 – SUPPLEMENTAL MATERIAL function. Am J Physiol 244: F579 F589, 1983 20. Combes AN, Zappia L, Er PX, Oshlack A, 6. Lee JW, Chou CL, Knepper MA: Deep se- Little MH: Single-cell analysis reveals con- Thisarticle containsthe following supplemental quencing in microdissected renal tubules iden- gruence between kidney organoids and hu- tifies nephron segment-specific transcriptomes. material online at http://jasn.asnjournals.org/ man fetal kidney. Genome Med 11: 3, 2019 J Am Soc Nephrol 26: 2669–2677, 2015 21. Harder JL, Menon R, Otto EA, Zhou J, Eddy lookup/suppl/doi:10.1681/ASN.2019040415/-/ 7. Burg M, Grantham J, Abramow M, Orloff J: S, Wys NL, et al.: European Renal cDNA Bank DCSupplemental. Preparation and study of fragments of sin- (ERCB); Nephrotic Syndrome Study Network Supplemental Figure 1. Cell type and gle rabbit nephrons. Am J Physiol 210: (NEPTUNE): Organoid single cell profiling – protein mass distribution among mouse re- 1293 1298, 1966 identifies a transcriptional signature of glo- 8. Windhager EE, Giebisch G: Micropuncture nal epithelial cells. merular disease. JCI Insight 4: e122697, study of renal tubular transfer of sodium chloride 2019 Supplemental Figure 2. Selected marker in the rat. Am J Physiol 200: 581–590, 1961 22. Cao J, Cusanovich DA, Ramani V, Aghamirzaie transcript expression in different renal tubule/ 9. Gottschalk CW, Mylle M: Micropuncture study D, Pliner HA, Hill AJ, et al.: Joint profiling of cell types. of the mammalian urinary concentrating chromatin accessibility and gene expression

JASN 30: 1358–1364, 2019 Renal Epithelial Cell Nomenclature 1363 REVIEW www.jasn.org

in thousands of single cells. Science 361: 36. McCormick JA, Ellison DH: Distal convoluted TFCP2L1 patterns cells in the mouse kidney 1380–1385, 2018 tubule. Compr Physiol 5: 45–98, 2015 collecting ducts. eLife 6: e24265, 2017 23. Wu H, Malone AF, Donnelly EL, Kirita Y, 37. Bachmann S, Bostanjoglo M, Schmitt R, Ellison 49. Kim WY, Nam SA, Choi A, Kim YM, Park SH, Uchimura K, Ramakrishnan SM, et al.: Single-cell DH: Sodium transport-related in the Kim YK, et al.: Aquaporin 2-labeled cells transcriptomics of a human kidney allograft bi- mammalian distal nephron - distribution, on- differentiate to intercalated cells in response opsy specimen defines a diverse inflammatory togeny and functional aspects. Anat Embryol to depletion. Histochem Cell Biol response. J Am Soc Nephrol 29: 2069–2080, (Berl) 200: 447–468, 1999 145: 17–24, 2016 2018 38. Biner HL, Arpin-Bott MP, Loffing J, Wang X, 50. Mukherjee M, deRiso J, Otterpohl K, 24. Karaiskos N, Rahmatollahi M, Boltengagen Knepper M, Hebert SC, et al.: Human cortical Ratnayake I, Kota D, Ahrenkiel P, et al.: En- A, Liu H, Hoehne M, Rinschen M, et al.: A distal nephron: Distribution of dogenous notch signaling in adult kidneys single-cell transcriptome atlas of the mouse and water transport pathways. JAmSoc maintains segment-specific epithelial cell glomerulus. J Am Soc Nephrol 29: 2060–2068, Nephrol 13: 836–847, 2002 types of the distal tubules and collecting 2018 39. Morel F, Chabardès D, Imbert M: Functional ducts to ensure water . JAmSoc 25. Han, X, Wang, R, Zhou, Y, Fei, L, Sun, H, Lai, S, segmentation of the rabbit distal tubule by Nephrol 30: 110–126, 2019 et al.: Mapping the mouse cell atlas by mi- microdetermination of -dependent 51. Blomqvist SR, Vidarsson H, Fitzgerald S, crowell-seq. Cell 172: 1091–1107.e17, 2018 adenylate cyclase activity. Kidney Int 9: Johansson BR, Ollerstam A, Brown R, et al.: 26. Tabula Muris Consortium; Overall coordina- 264–277, 1976 Distal renal tubular in mice that lack tion; Logistical coordination; collec- 40. Loffing J, Loffing-Cueni D, Macher A, Hebert the forkhead transcription factor Foxi1. JClin tion and processing; Library preparation and SC, Olson B, Knepper MA, et al.: Localization Invest 113: 1560–1570, 2004 sequencing; Computational data analysis; of epithelial and aquaporin- 52. Christensen BM, Marples D, Kim YH, Wang Cell type annotation; Writing group; Sup- 2inrabbitkidneycortex.Am J Physiol Renal W, Frøkiaer J, Nielsen S: Changes in cellular plemental text writing group; Principal in- Physiol 278: F530–F539, 2000 composition of kidney collecting duct cells in vestigators: Single-cell transcriptomics of 20 41. Loffing J, Pietri L, Aregger F, Bloch-Faure M, rats with lithium-induced NDI. Am J Physiol mouse organs creates a Tabula Muris. Nature Ziegler U, Meneton P, et al.: Differential Cell Physiol 286: C952–C964, 2004 562: 367–372, 2018 subcellular localization of ENaC subunits in 53. Jeong HW, Jeon US, Koo BK, Kim WY, Im SK, 27. Maunsbach AB: Observations on the seg- mouse kidney in response to high- and low- Shin J, et al.: Inactivation of Notch signaling mentation of the proximal tubule in the rat Na diets. Am J Physiol Renal Physiol 279: in the renal collecting duct causes nephro- kidney. Comparison of results from phase F252–F258, 2000 genic in mice. JClinInvest contrast, fluorescence and electron micros- 42. Obermüller N, Bernstein P, Velázquez H, Reilly 119: 3290–3300, 2009 copy. J Ultrastruct Res 16: 239–258, 1966 R, Moser D, Ellison DH, et al.: Expression of 54. Wu H, Chen L, Zhou Q, Zhang X, Berger S, Bi 28. Woodhall PB, Tisher CC, Simonton CA, the thiazide-sensitive Na-Cl cotransporter in J, et al.: Aqp2-expressing cells give rise to Robinson RR: Relationship between para- rat and human kidney. Am J Physiol 269: renal intercalated cells. J Am Soc Nephrol aminohippurate secretion and cellular mor- F900–F910, 1995 24: 243–252, 2013 phology in rabbit proximal tubules. JClin 43. Kaissling B, Bachmann S, Kriz W: Structural 55. Alper SL, Natale J, Gluck S, Lodish HF, Brown Invest 61: 1320–1329, 1978 adaptation of the distal convoluted tubule D: Subtypes of intercalated cells in rat kidney 29. Kriz W: Structural organization of the renal to prolonged treatment. Am J collecting duct defined by antibodies medulla: Comparative and functional aspects. Physiol 248: F374–F381, 1985 against erythroid band 3 and renal vacu- Am J Physiol 241: R3–R16, 1981 44. Saritas, T, Puelles, VG, Su, XT, McCormick, olar H1-ATPase. Proc Natl Acad Sci U S A 30. Burg MB: Thick ascending limb of Henle’s JA, Welling, PA, Ellison, DH: Optical clearing 86: 5429–5433, 1989 loop. Kidney Int 22: 454–464, 1982 in the kidney reveals potassium-mediated tu- 56. Kim J, Kim YH, Cha JH, Tisher CC, Madsen 31. Knepper MA, Kim GH, Fernández-Llama P, bule remodeling. Cell Rep 25: 2668–2675.e3, KM: Intercalated cell subtypes in connecting Ecelbarger CA: Regulation of thick ascend- 2018 tubule and cortical collecting duct of rat and ing limb transport by vasopressin. JAmSoc 45.WangF,LuX,PengK,FangH,ZhouL,SuJ, mouse. J Am Soc Nephrol 10: 1–12, 1999 Nephrol 10: 628–634, 1999 et al.: Antidiuretic action of collecting duct 57. Madsen KM, Tisher CC: Structural-functional 32. Dimke H, Schnermann J: Axial and cellular (Pro)Renin receptor downstream of vasopres- relationships along the distal nephron. Am J heterogeneity in electrolyte transport path- sin and PGE2 receptor EP4. J Am Soc Nephrol Physiol 250: F1–F15, 1986 ways along the thick ascending limb. Acta 27: 3022–3034, 2016 58. Clark JZ, Chen L, Chou CL, Jung HJ, Lee JW, Physiol (Oxf) 223: e13057, 2018 46. Trepiccione F, Capasso G, Nielsen S, Knepper MA: Representation and relative 33. Briggs JP, Schnermann JB: Whys and Christensen BM: Evaluation of cellular plas- abundance of cell-type selective markers in wherefores of juxtaglomerular apparatus ticity in the collecting duct during recovery whole-kidney RNA-Seq data. Kidney Int 95: function. Kidney Int 49: 1724–1726, 1996 from lithium-induced nephrogenic diabetes 787–796, 2019 34. Peti-Peterdi J, Morishima S, Bell PD, Okada Y: insipidus. Am J Physiol Renal Physiol 305: 59. Sperber I: Zoologiska Bidrag fran Uppsala. In Two-photon excitation fluorescence imaging F919–F929, 2013 Studies on the Mammalian kidney,Vol22, of the living juxtaglomerular apparatus. Am 47. Guo Q, Wang Y, Tripathi P, Manda KR, 1944, pp. 249–432 J Physiol Renal Physiol 283: F197–F201, Mukherjee M, Chaklader M, et al.: Adam10 60. Jung HJ, Raghuram V, Lee JW, Knepper 2002 mediates the choice between principal cells MA: Genome-Wide mapping of DNA 35. Razga Z, Nyengaard JR: The effect of an- and intercalated cells in the kidney. JAmSoc accessibility and binding sites for CREB giotensin II on the number of macula densa Nephrol 26: 149–159, 2015 and C/EBPb in vasopressin-sensitive col- cells through the AT1 receptor. Nephron 48. Werth M, Schmidt-Ott KM, Leete T, Qiu A, lecting duct cells. JAmSocNephrol29: Physiol 112: 37–43, 2009 Hinze C, Viltard M, et al.: Transcription factor 1490–1500, 2018

1364 JASN JASN 30: 1358–1364, 2019 Erratum

CORRECTION Figure 1 of our article in the August issue of JASN lacked a numerical label indicating the position Lihe Chen, Jevin Z. Clark, Jonathan W. Nelson, Brigitte of the thick ascending limb of Henle in short-looped Kaissling, David H. Ellison, and Mark A. Knepper: Renal-Tubule nephrons. The corrected figure is provided below and Epithelial Cell Nomenclature for Single-Cell RNA-Sequencing at https://hpcwebapps.cit.nih.gov/ESBL/Database/ Studies. J Am Soc Nephrol 30: 1358–1364, 2019. RECOn/.

Figure 1. Renal tubule cell nomenclature. The scheme shows the connection of both a short-looped nephron and a long-looped nephron to the collecting duct system. The numbers point to different renal tubule segments and the cell types that they contain are listed on the right of the scheme. Dashed lines indicate regional boundaries. The corticomedullary junction is defined by the arcuate arteries (not shown in this diagram). The division between the outer stripe and inner stripe of the outer medulla is defined by the transitions from PSTs to DTLs of the loops of Henle. The junction between the outer medulla and inner medulla is defined by the transitions from the ascending thin limbs (ATL) of Henle of the long-loop nephrons to the MTALs of the loops of Henle. The division between the cortical labyrinth and cortical medullary rays is indicated but not labeled. Drawing is on the basis of the original Renal Commission of the International Union of Physiologic Sciences publication.

JASN 30: 2475, 2019 ISSN : 1046-6673/3012-2475 2475 Supplemental Materials Table of Contents

Supplemental Figure 1. Cell type and protein mass distribution among mouse renal epithelial cells in cortex.

Supplemental Figure 2. Selected marker transcript expression in different renal tubule/cell types. Supplemental Figure Legends

Supplemental Figure 1. Cell type and protein mass distribution among mouse renal epithelial cells in cortex. (A) The contributions of individual mouse renal epithelial cell types to epithelial cell count

(given at bottom). (B) The contributions of individual mouse renal epithelial cell types to renal cortex tubule protein mass (given at bottom). A-IC, type A intercalated cell; B-IC, type B intercalated cell; CCD, cortical collecting duct; CNT, connecting tubule; CTAL, cortical thick ascending limb of the loop of Henle; DCT, distal convoluted tubule; PC, principal cell; PCT, proximal convoluted tubule; PST, proximal straight tubule. The data are from Clark and colleagues.58

Supplemental Figure 2. Selected marker transcript expression in different renal tubule/cell types.

(A) Representative marker expression in different renal tubule segments of rat determined by single- tubule RNA-seq from Lee et al.6 (B) Distribution of selected markers from (A) in collecting duct cell types

(A-IC, B-IC and PC) determined by single-cell RNA-seq (median TPM values) from Chen et al.14 Not all

the transcripts from (A) are detected by single-cell RNA-seq in (B). Color gradient indicates relative

abundance.

A B Cell Type Distribution Protein Mass Distribution

A-IC (2%) PC-CCD (4%) B-IC (5%) Proximal Tubule (PCT + PST) (77%) Podocyte (4%)

CNT (7%)

CTAL DCT (8%) (10%)

CTAL (5%) Proximal Tubule DCT (PCT + PST) (11%) CNT (57%) (4%) Podocyte (2%) B-IC (2%) PC-CCD (1%) Total Cell Count per Cortex Total Protein Mass per Cortex A-IC (1%) 3.6 X 107 21.3 mg A Rat single-tubule RNA-Seq B Mouse single-cell RNA-Seq

Gene Gene S1 S2 S3 DTL1 DTL2 DTL3 AT L M T AL C T AL DCT CNT CCD OMCD IMCD A-IC B-IC PC Symbol Symbol Umod 19 0 17 2 1 0 1 1597 12485 329 1 1 6 0 Umod 0 0 0 Aqp2 2 9 4 10 5 4 1 12 6 15 4240 6776 4608 6087 Aqp2 3 0 1821 Aqp1 95 297 151 1317 1831 3748 8 0 0 1 1 1 0 0 Calb1 1 0 675 Calb1 7 2 0 0 0 0 0 4 1 222 3623 145 1 0 Hsd11b2 1 0 770 Hsd11b2 20 32 50 15 9 5 5 8 25 179 1368 1151 952 11 Lgals3 3 267 1357 Ggt1 97 743 1394 136 111 56 0 5 39 11 3 0 2 1 Slc34a1 1 0 0 Lgals3 0 2 0 2 0 17 4 2 15 15 92 544 1077 881 Atp6v1g3 2331 3806 4 Slc34a1 333 1242 215 0 0 0 0 0 5 2 1 0 0 0 Defb1 1941 7 1661 Atp6v1g3 1 1 0 0 0 0 0 0 1 9 754 559 886 1 Gclc 7 26 3 Defb1 8 3 0 0 0 0 0 7 28 345 785 503 877 6 Avpr2 0 0 455 Fst 0 0 0 365 1061 56 1 0 0 0 0 0 0 0 Clcnkb 507 660 123 Pfn3 324 898 270 0 0 0 0 0 3 1 1 0 0 0 Muc20 1 0 0 Slc7a13 0 137 780 0 1 23 0 1 0 0 0 0 1 0 Gss 0 0 0 Gclc 1 408 635 1 12 32 10 0 4 4 5 12 2 15 Aqp3 46 137 1303 Slc3a1 50 114 710 2 2 4 3 0 3 4 2 4 9 7 Rhbg 226 38 56 Vim 0 0 3 431 495 380 38 0 14 8 0 2 0 1 Avpr2 0 0 0 0 0 0 2 15 34 10 177 399 472 371 Pck1 0 0 0 Clcnkb 3 1 0 0 0 0 6 41 309 228 318 550 135 1 Aqp6 55 0 0 Muc20 0 0 0 0 0 0 0 0 0 6 124 256 328 511 Ass1 0 0 0 Gss 33 74 593 11 31 0 25 21 10 13 8 9 18 11 Osgin1 0 0 0 Gpha2 0 0 0 0 0 31 0 0 0 0 0 0 0 575 Slc12a1 0 0 0 Acss2 31 98 560 8 26 0 10 21 8 10 6 4 8 9 Pdzk1 0 0 0 Aqp3 1 1 0 1 0 0 0 0 0 0 174 234 312 348 Klk1 207 914 5 Rhbg 0 0 0 0 0 0 0 0 1 20 218 203 418 17 Aqp4 0 0 0 Pck1 280 375 186 0 0 0 0 0 2 0 0 1 0 0 Atp6v1b1 277 201 31 Aqp6 0 0 0 0 0 0 0 0 0 4 88 95 461 0 Casr 0 1 0 Ass1 398 250 105 1 0 1 0 1 3 10 1 0 0 0 Slc26a4 0 1135 0 Osgin1 0 105 448 0 1 1 2 0 0 0 0 2 0 2 Hmx2 0 108 0 Slc12a1 1 0 1 0 0 0 2 346 255 2 0 0 0 0 Scnn1b 3 0 308 Pdzk1 299 257 203 0 0 4 0 0 1 0 2 0 4 1 Pth1r 0 0 0 Klk1 1 0 0 0 0 0 0 0 34 220 327 3 0 0 Acaa1b 0 0 0 Aqp4 0 0 0 0 0 0 0 0 0 0 44 18 73 376 Hepacam2 24 561 0 Atp6v1b1 1 0 0 0 0 0 0 0 47 83 236 196 202 0 Slc4a1 94 0 0 Casr 0 0 0 0 0 0 0 7 332 30 3 14 0 0 Tsc22d3 0 0 0 Slc26a4 0 0 0 0 0 0 0 0 0 0 241 203 17 0 Gls 20 23 62 Slc12a3 1 0 0 0 0 0 0 0 2 287 0 0 0 0 Kl 0 0 3 Hmx2 0 0 0 0 0 0 0 0 0 2 160 211 1 0 Wnk1 228 56 3 Scnn1b 0 0 0 0 0 0 0 0 3 32 160 191 110 0 Foxi1 26 396 0 Pth1r 144 179 60 0 0 0 0 1 132 48 11 1 0 0 Anpep 0 0 0 Dak 94 197 114 0 0 0 0 0 1 1 1 1 0 1 Cdh1 39 28 97 Acaa1b 190 136 112 1 3 7 2 1 20 19 6 11 13 4 Scnn1a 1 20 8 Hepacam2 0 0 0 0 0 0 0 0 0 0 149 184 48 0 Scnn1g 0 0 102 Slc4a1 0 0 0 0 0 0 0 0 0 6 83 103 182 0 Khk 0 0 0 Tsc22d3 10 8 3 131 35 160 121 14 3 5 6 10 0 71 Ptger3 0 0 0 Gls 0 0 0 10 22 203 86 0 3 2 13 13 1 23 Kcne1 1 0 554 Kl 8 48 4 0 0 0 0 28 57 157 121 7 7 0 L1cam 0 0 9 Mx2 7 1 0 0 0 4 185 0 2 0 0 1 16 27 Gata2 1 9 345 Wnk1 4 1 0 7 4 4 7 1 45 182 41 10 1 4 Slc4a9 76 424 0 Foxi1 0 0 0 0 0 0 0 0 0 1 79 112 94 0 Anpep 2 43 141 2 1 5 0 0 0 0 1 2 0 0 Hk1 4 8 4 Cdh1 1 0 2 11 17 20 19 1 25 15 59 76 65 128 Scnn1a 0 0 0 0 0 0 0 11 39 69 88 104 47 0 Slc8a1 0 0 0 0 0 0 0 0 0 41 130 16 0 0 Slc22a6 83 105 20 0 0 0 0 0 1 0 0 0 0 0 Scnn1g 0 0 0 0 0 0 0 0 0 0 53 89 61 0 Khk 24 77 76 0 0 0 0 0 1 1 0 0 0 1 Ptger3 4 0 0 0 0 0 2 91 33 50 0 1 0 0 Kcne1 0 0 0 0 0 0 0 0 0 0 66 71 0 0 L1cam 0 0 0 3 3 39 16 0 4 2 16 31 4 89 Slc14a2 0 0 0 18 0 22 0 0 0 0 0 0 0 90 Gata2 0 0 0 0 0 0 0 0 0 0 56 56 0 58 Slc5a2 91 8 0 0 0 0 0 0 2 5 4 5 0 1 Slc4a9 0 0 0 0 0 0 0 0 0 1 29 73 6 0 Hk1 0 0 1 18 36 26 36 23 29 42 33 51 42 50