Tubuloids derived from human adult kidney and urine for personalized disease modeling

Adult stem cell-derived organoids are three-dimensional epithelial structures that recapitulate fundamental aspects of their organ of origin. We describe conditions for the long-term growth

of primary kidney tubular epithelial organoids, or ‘tubuloids’. The cultures are established from human and mouse kidney tissue and can be expanded for at least 20 passages (>6 months) while

retaining a normal number of chromosomes. In addition, cultures can be established from human urine. Human tubuloids represent proximal as well as distal nephron segments, as evidenced by

gene expression, immunofluorescence and tubular functional analyses. We apply tubuloids to model infectious, malignant and hereditary kidney diseases in a personalized fashion. BK virus

infection of tubuloids recapitulates in vivo phenomena. Tubuloids are established from Wilms tumors. Kidney tubuloids derived from the urine of a subject with cystic fibrosis allow ex vivo

assessment of treatment efficacy. Finally, tubuloids cultured on microfluidic organ-on-a-chip plates adopt a tubular conformation and display active (trans-)epithelial transport function.

The data that support the findings of this study are available from the corresponding author upon reasonable request. WGS data have been deposited at the EGA, which is hosted by the EBI,

under accession code EGAS00001002729. Single-cell and bulk sequencing data can be found in the Supplementary Information and have been deposited at GEO under accession code GSE107795.

We thank the Hubrecht Imaging Center for assistance with (confocal) microscopy. We thank H. Begthel and J. Korving (both Hubrecht Institute) for preparation of histological and

immunohistochemical specimens. We thank T. Nguyen (UMC Utrecht, the Netherlands) and R. de Krijger (Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands) for help with the

analysis of histological specimens. We thank the Hubrecht FACS facility for help with the sort of single (EPCAM+) cells and G. Posthuma (Department of Cell Biology and Institute of

Biomembranes, UMC Utrecht, the Netherlands) for excellent support with transmission electron microscopy. We thank E. Driehuis for the photography of the VP-1 staining and for help with

finalizing the manuscript. We thank J. Hoenderop (Radboud UMC, the Netherlands) for kindly providing UMOD and SLC12A1 antibodies, and NC3Rs for development of the kidney-on-a-chip assays

(Nephrotube, CRACK-IT challenge). This work was supported by a grant from the Dutch Kidney Foundation (grant no. DKF14OP04), and Zwaartekracht (NWO). This work was supported by the partners

of Regenerative Medicine Crossing Borders (www.regmedxb.com). Powered by Health~Holland, Top Sector Life Sciences & Health.

Hubrecht Institute—Royal Netherlands Academy of Arts and Sciences, Utrecht, the Netherlands

Frans Schutgens, Carola Ammerlaan, Fjodor Yousef Yengej, Benedetta Artegiani & Hans Clevers

Department of Nephrology and Hypertension, University Medical Centre Utrecht, Utrecht, the Netherlands

Frans Schutgens, Maarten B Rookmaaker, Carola Ammerlaan, Fjodor Yousef Yengej & Marianne C Verhaar

Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands

Thanasis Margaritis, Anne Rios, Sepide Derakhshan, Ruben van Boxtel, Marry M. van den Heuvel-Eibrink, Frank Holstege, Jarno Drost & Hans Clevers

Division of Pharmacology, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Utrecht, the Netherlands

Regenerative Medicine Center Utrecht, Utrecht, the Netherlands

Department of Medical Microbiology, University Medical Center Utrecht, Utrecht, the Netherlands

Wilhelmina Children’s Hospital, University Medical Center Utrecht, Utrecht, the Netherlands

Center for Molecular Medicine, Cancer Genomics Netherlands, Department of Genetics, University Medical Center Utrecht, Utrecht, the Netherlands

Institute of Human Genetics, Medical University of Graz, Graz, Austria

Laboratory of Medical Microbiology and Immunology, St. Elisabeth TweeSteden Ziekenhuis, Tilburg, the Netherlands

F.S., M.B.R., M.C.V. and H.C. designed, performed, analyzed experiments and wrote the manuscript. C.A. established and maintained tubuloid cultures and performed and analyzed karyotyping

experiments. A.R. performed, imaged and analyzed immunofluorescent stainings. T.M. and F.H. analyzed the single-cell sequencing data. F.S., J.J. and R.M. designed, analyzed and performed the

P-gp transporter assays in tubuloids. A.P.A.H. provided support with scanning electron microscopy images. F.S., M.Vi., M.B.R. and J.M. designed, performed and analyzed the BK virus

experiments. R.v.B analyzed and interpreted the WGS analysis and E.C. gave input on the interpretation. B.A. helped with the analysis of the the bulk RNA-seq. M.M.H.E. established the

logistics of obtaining clinical samples of nephroblastoma tissue. F.S. and J.D. established nephroblastoma cultures and designed experiments. E.H. performed and analyzed the nephroblastoma

CNV analysis. S.D. established the clonal tubuloid line and F.Y.Y. analyzed the clonal tubuloid line. K.M.W.G. obtained CF urine samples. F.S., A.V. and J.B. designed FSK-induced swelling

experiments and analyzed data. F.S., L.G., M.Vo. and H.L. designed, performed and analyzed the ‘tubuloid-on-a-chip’ experiments. All authors commented on the manuscript.

H.C. is a holder of several patents related to organoid technology. L.G., M.Vo. and H.L. are employees of MIMETAS BV, the Netherlands, that is marketing the OrganoPlate. OrganoPlate is a

trademark of MIMETAS.

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

> 40 spreads were quantified, from tubuloids in P11, P14 and P18 in 3 independent experiments, from 3 independent tubuloid cultures.

A tubuloid line was established from cortical kidney tissue after nephrectomy and tubuloids in passage 2 (P2) and passage 8 (P8) were, along with the matched tissue sample, harvested for

WGS. A pie chart with types of mutations and a list of the observed missense mutations in coding regions are included. None of the reported mutations is reported in the COSMIC database or

associated with kidney (dys)function. Green boxes indicate the part of the tubuloids that are used for establishing the next passage. Red boxes indicate the samples that were harvested for

WGS.

Scheme of the experimental protocol (a). The development of mouse kidney epithelial cells into folded / branching structures after seeding; image of passage 6 (representative image of at

least n = 3 lines) (b). H&E stain at passage 7 (representative image of an H&E stain that was performed for at least n = 3 lines) (c). An example of a typical metaphase spread, from a

tubuloid culture of passage number > 11 (d). Quantification of > 40 spreads in 3 independent experiments, from tubuloids at P11, P15 and P16 (e). Scale bars 100 µm (b, c) and 10 µm (d).

EPCAM+ kidney epithelial cells and tubuloid cells were analysed. 67% of the primary kidney epithelial cells were in GO/G1, compared to 40% of the tubuloid cells (see Methods).

Normalized log2 transcript counts show expression of the pan-kidney epithelium marker PAX8 across different clusters (192 kidney and 192 tubuloid cells were sequenced in one run and after

quality checks (see Methods), 51 kidney cells and 149 tubuloid cells were used for clustering analysis. Cluster 1: n = 8; Cluster 2: n = 45; Cluster 3: n = 33; Cluster 4: n = 19; Cluster 5:

n = 39; Cluster 6: n = 28; Cluster 7: n = 28. Tukey box plots are used to visualize the distributions. All points of the populations are shown as jittered dots.) (a). The expression of PAX8

is confirmed by staining (n = 4 tubuloid lines) (b). Scale bar: 100 µm.

Cluster 3–7 contains only tubuloid-derived cells; cluster 2 contains only primary kidney epithelial cells and cluster 1 is a mix between tubuloid and primary kidney epithelial cells.

Normalized log2 transcript counts show expression of (kidney) epithelium markers PAX8, EPCAM, KRT18 and KRT19 across different clusters, showing that all cells are epithelial in nature (a).

Normalized log2 transcript counts of typical intercalated genes SLC26A7, SLC4A1, ATP6V1B1, that are expressed in cluster 1, a combination of primary kidney epithelial cells and

tubuloid-derived cells (b). Normalized log2 transcript counts show expression of collecting duct principal cell genes SLC14A1 and CLDN8 in cluster 4 (c). Normalized log2 transcript counts

show expression of glucose handling genes SLC2A1, ALDOC as well as VEGFA, suggestive of proximal tubule cells in cluster 5 (d). Cluster 6 does not express many segment-specific genes, except

SLC22A5, which may suggest a (proximal tubule) progenitor cell phenotype (e) Normalized log2 transcript counts how higher expression of COL4A3 and COL4A4 and lower expression of KRT18 and

KRT19 in cluster 7, suggesting a pro-fibrotic phenotype (f). In a-f, 192 kidney and 192 tubuloid cells were sequenced in one run and after quality checks (see methods), 51 kidney cells and

149 tubuloid cells were used for clustering analysis. Cluster 1: n = 8; Cluster 2: n = 45; Cluster 3: n = 33; Cluster 4: n = 19; Cluster 5: n = 39; Cluster 6: n = 28; Cluster 7: n = 28.

Tukey box plots are used to visualize the distributions. All points of the populations are shown as jittered dots.

CALB1 expression is absent in kidney tubuloids on the protein level (a). By withdrawal of growth factors from the culture medium, CALB1 can be induced, as visualized with immunofluorescence

(b). UMOD expression is absent in kidney tubuloids on the protein level (c). By withdrawal of growth factors from the culture medium, UMOD can be induced, as visualized with

immunofluorescence (d). Representative images of n = 2 independent experiments. Scale bars: 75 µm.

When tubuloids are exposed to calcein-AM, a substrate of P-gp (the xenobiotics efflux pump, located at the apical membrane) that diffuses freely into cells and that becomes fluorescent

inside cells after cleaving the acetomethoxy group resulting in calcein, in the presence of an inhibitor (PSC-833) of P-gp, calcein accumulates (a). In absence of the inhibitor, P-gp pumps

calcein from the cells, thereby preventing accumulation of fluorescent signal (b).

In the presence of specific P-gp-inhibitor PSC-833, calcein accumulates in tubuloids, as measured by fluorescent plate reader. Normalized quantification of n = 3 independent experiments,

with each experiment n ≥ 5 plate reader measurements. Lines indicate means per experiment. * P = 0.0016 with an unpaired two-tailed t test, after pooling the individual measurements from n =

 3 experiments.

Gene expression of marker genes of the proximal tubule (ANPEP, ABCC4, SLC4A4), Loop of Henle (SLC12A1, CLDN10), distal tubule (SLC12A3, SLC41A3, PCBD1) and collecting duct (AQP3, NR3C2) was

determined in the clonal tubuloid line and compared with the bulk tubuloid line that was used for establishing the clonal line. The clonal line preserves expression of markers of multiple

segments: the proximal tubule genes ABCC4 and SLC4A4 are similar to the bulk tubuloid line, whereas expression of the typical Loop of Henle gene SLC12A1 is increased and the typical distal

tubule marker SLC12A3 is decreased. These data indicate multi-lineage potential of a single tubuloid cell (proximal tubule and Loop of Henle). Expression levels were normalized to RPLP0, and

expressed as fold change to expression levels of a colon organoid line.

Sanger sequencing shows a heterozygous 8 base pair deletion in the healthy kidney tissue (H tissue); tubuloids derived from the healthy tissue (H tubuloid); tumor tissue (T tissue) and the

tubuloids derived from the tumor tissue (T tumoroid) (a). Sanger sequencing of WT1 shows a frameshift mutation in exon 10: it is a heterozygous 4 base pair insertion in the tumor tissue (T

tissue) and the tumoroids derived from the tumor tissue (T tumoroid). This insertion is absent in the healthy kidney tissue (H tissue) and tubuloids derived from the healthy tissue (H

tubuloid) (b). In both a and b: representative sequences of at least n = 3 independent experiments.

Assessed with a PAX8 staining (PAX8 staining performed once on this tubuloid line and on n = 3 other lines). Scale bar 100 µm.

In urine-derived tubuloids and intestinal organoids from the same patient, VX-770 increased FSK-induced swelling. Average is plotted of n = 3 independent experiments that were performed in

duplicate, error bars represent standard deviation.

Rhodamine 123 is transported into cells on the basal side by OCT-transporters and removed on the apical side by efflux pump P-gp (a). In the presence of P-gp inhibitor PSC-833, rhodamine 123

fluorescence in the lumen on the apical side, is expected to be reduced (b).

Dissociation-induced stress was quantified using a scoring strategy (see methods) based on the expression of heat shock genes. A score over 0.4 was used to filter out cells that were

severely stressed. 192 kidney and 192 tubuloid cells were sequenced in one run and after quality checks (see methods), cells were analyzed for dissociation-induced stress (Kidney 1: n = 41;

Kidney 2: n = 43; Tubloid 1: n = 76; Tubloid 2: n = 73). Tukey box plots are used to visualize the distributions. All points of the populations are shown as jittered dots.

Supplementary Figures 1–15 and Supplementary Tables 1 and 2.

Wilms tumor–derived tumoroids (from patient 1) have a different morphology than matched normal tubuloids. This 6-day time-lapse video shows the matched normal tubuloids.

Wilms tumor–derived tumoroids (from patient 1) have a different morphology than matched normal tubuloids. This 6-day time-lapse video shows the tumoroids.

Wilms tumor–derived tumoroids (from patient 1) have a different morphology than matched normal tubuloids. This 6-day time-lapse video shows a detail of the tumoroids.

Forskolin-induced swelling assay of CF urine-derived tubuloids without VX-770.

Forskolin-induced swelling assay of CF urine-derived tubuloids with VX-770.

Forskolin-induced swelling assay of CF intestinal organoids without VX-770.

Forskolin-induced swelling assay of CF intestinal organoids with VX-770.

Bulk RNA sequencing data. CSV file of n = 3 (1, 2, 3) tissue samples (FS1) and matched tubuloid lines (FS2).

DNA sequencing data at day 1 of infection and day 30 of infection.

Anyone you share the following link with will be able to read this content: