FIGURE SUMMARY
Title

Cysteamine-bicalutamide combination therapy corrects proximal tubule phenotype in cystinosis

Authors
Jamalpoor, A., van Gelder, C.A., Yousef Yengej, F.A., Zaal, E.A., Berlingerio, S.P., Veys, K.R., Pou Casellas, C., Voskuil, K., Essa, K., Ammerlaan, C.M., Rega, L.R., van der Welle, R.E., Lilien, M.R., Rookmaaker, M.B., Clevers, H., Klumperman, J., Levtchenko, E., Berkers, C.R., Verhaar, M.C., Altelaar, M., Masereeuw, R., Janssen, M.J.
Source
Full text @ EMBO Mol. Med.

Toxicity profile of dimethyl α‐ketoglutarate (DMKG) and the drugs tested for the treatment of cystinosis

Cell viability curves of dimethyl α‐ketoglutarate (DMKG) in CTNSWT and CTNS−/− cells after 4 h of incubation in fed and starved condition (n = 3).

Cell viability curves of the increasing concentration of DMKG in control (CTNSWT) and CRISPR‐generated cystinotic (CTNS−/−) cells after 24 h of incubation in fed and starved condition, respectively (n = 3).

Cell viability test in CTNSWT, CTNS−/− cells treated with the increasing concentrations of cysteamine, bicalutamide, luteolin, genistein, 8‐bromo‐cAMP, disulfiram, and a combination of cysteamine and bicalutamide (100 and 35 µM, respectively), respectively (n = 3).

Data information: Data are expressed as mean ± SEM.

E

Bicalutamide–cysteamine combination treatment corrects the metabolome and proteome profile of cystinotic proximal tubule cells

Metabolomic analysis of CRISPR‐generated cystinotic cells (CTNS−/−) treated with cysteamine (100 µM), bicalutamide (35 µM) and a combination of cysteamine and bicalutamide (100 and 35 µM, respectively) (n = 3).

Heat map analysis of the measured metabolites in CTNS−/− cells upon treatment with cysteamine or cysteamine–bicalutamide combination treatment (n = 3). Metabolites significantly decreased were displayed in green, while metabolites significantly increased were displayed in red.

Heat map and REACTOME analysis of the altered proteins in CTNS−/− cells upon treatment with cysteamine, bicalutamide and cysteamine–bicalutamide combination treatment (n = 3). Proteins significantly decreased were displayed in green, while metabolites significantly increased were displayed in red.

Proteomic analysis of CTNS−/− cells treated with cysteamine, bicalutamide, and a combination of cysteamine and bicalutamide, respectively (n = 3). AKGDH: Alpha‐ketoglutarate dehydrogenase, GLUD1: GLUD2; Mitochondrial glutamate dehydrogenase 1/2, IGF2R; Cation‐independent mannose‐6‐phosphate receptor, GSTK1: Glutathione S‐transferase kappa 1, COX6B1: Cytochrome c oxidase subunit 6B1, ACACA: Acetyl‐CoA carboxylase 1‐Biotin carboxylase, GLS: Glutaminase kidney isoform, mitochondrial, CASP3: Caspase‐3, NPC1: Niemann‐Pick C1 protein.

Data information: Data are expressed as mean ± SEM. P‐values < 0.05 were considered to be significant. One‐way ANOVA with Dunnett’s correction (A,B, C, D and E). * significantly different from CTNSWT cells (P < 0.05). # significantly different from CTNS−/− cells (P < 0.05). Exact P‐values and statistical tests are listed in Appendix Table S1.

F

Cysteamine–bicalutamide combination treatment shows a synergic effect in treatment of cystinotic ciPTEC

Volcano plot illustrates significantly differentially abundant proteins (n = 3). The ‐log10 (Benjamini–Hochberg corrected P‐value) is plotted against the log2 (fold change: CTNS−/− no drug treatment/CTNS−/− cysteamine treatment), (fold change: CTNS−/− no drug treatment/CTNS−/− bicalutamide treatment) and (fold change: CTNS−/− no drug treatment/CTNS−/− combination treatment), respectively. The non‐axial vertical lines denote ±1.5‐fold change while the non‐axial horizontal line denotes P = 0.05, which is our significance threshold (prior to logarithmic transformation).

Western blotting and densitometric analyses for LC3‐II/LC3‐I ratio in CRISPR‐generated CTNS−/− cells treated with cysteamine (100 µM), bicalutamide (35 µM), and a combination of cysteamine and bicalutamide (100 and 35 µM, respectively). β‐Actin was used as a loading control (n = 3).

Quantification of TFEB‐GFP nuclear translocation in CTNSWT, and CTNS−/− cells upon treatment with bicalutamide (35 µM) (n = 3).

TFEB mRNA expression of the CTNSWT cells upon starvation and treatment with bicalutamide (35 µM) (n = 3).

Data information: Data are expressed as mean ± SEM. *P‐values < 0.05 were considered to be significant. One‐way ANOVA with Dunnett’s correction (D, E and F) or unpaired t‐test (E). Exact P‐values and statistical tests are listed in Appendix Table S1.

Metabolomic and proteomic profiling reveal αKG accumulation in cystinosis

Principal component analysis (PCA) of control (CTNSWT), CRISPR‐generated cystinotic (CTNS−/−) and patient‐derived cystinotic (CTNSPatient) cells based on the metabolites measured. Each dot represents one sample, and the dots of the same colour are biological replicates (n = 3).

Heat map analysis of top 25 metabolites distinctively expressed in control and CTNS−/− cells. Metabolites significantly decreased (P < 0.05) were displayed in green, while metabolites significantly increased (P < 0.05) were displayed in red (n = 3).

Global test pathway enrichment analysis of the intracellular metabolic interactions distinctively affected in CTNS−/− cells compared to healthy control cells (n = 3). Larger circles further from the y‐axis and orange‐red colour show higher impact of pathway.

List of metabolites that were shared and significantly altered in both the CTNS−/− and CTNSPatient cells compared to control cells (n = 6).

Plasma levels of αKG (µM) in healthy individuals (n = 4) and cystinotic patients (n = 6). All cystinotic patients were on cysteamine treatment at the time of blood sampling. For this experiment, non‐parametric Mann–Whitney t‐test was used to demonstrate the significance.

Principal component analysis (PCA) of the measured proteins in CTNSWT, CTNS−/− and CTNSPatient (n = 3).

Volcano plot illustrates differentially abundant proteins (n = 3). The ‐log10 (Benjamini–Hochberg corrected P‐value) is plotted against the log2 (fold change: CTNSWT/CTNS−/−).

Gene ontology (GO) analysis illustrates classes of proteins differing between CTNSWT and CTNS−/− cells (n = 3).

List of proteins that were significantly upregulated and downregulated in CTNS−/− cells compared to control cells (n = 3). AKGDH; Alpha‐ketoglutarate dehydrogenase, LIPA; Lysosomal acid lipase, ACP2; Lysosomal acid phosphatase, CTSS; Cathepsin S, CTSC; Dipeptidyl peptidase, IGF2R; Cation‐independent mannose‐6‐phosphate receptor, SORT1; Sortilin, CAT; Catalase, CTSA; Lysosomal protective protein and SOD; Superoxide dismutase.

Data information: Data are expressed as mean ± SEM. *P‐values < 0.05 were considered to be significant. One‐way ANOVA with Dunnett’s correction (A, B, C, D, E, F, G, H and I). Exact P‐values and statistical tests are listed in Appendix Table S1.

Cysteamine–bicalutamide combination treatment is effective in patient‐derived cystinotic tubuloids and cystinotic zebrafish

Immunocytochemistry of patient‐derived cystinotic tubuloids (CNTSPatient‐1 and CTNSPatient‐2) and healthy kidney tissue‐derived control tubuloids (CNTSWT‐1 and CTNSWT‐2) for PAX8, TP63 and F‐actin. Scale bar = 100 µm.

Quantification of cystine levels (nmol/mg protein) by HPLC‐MS/MS in control and cystinotic tubuloids (n = 3).

Quantification of cystine levels (nmol/mg protein) by HPLC‐MS/MS in two different patient‐derived cystinotic tubuloids in the absence of the drugs (NT) or upon treatment with cysteamine (100 μM), bicalutamide (35 μM) or cysteamine (100 μM)‐bicalutamide (35 μM) combination treatment (n = 3).

αKG levels measured in patient‐derived cystinotic tubuloids (CNTSPatient‐1 and CTNSPatient‐2) in the absence of the drugs (NT) or upon treatment with cysteamine, bicalutamide or cysteamine–bicalutamide combination treatment using metabolomics (n = 3).

Representative images of control and cystinotic zebrafish.

Quantification of cystine levels (nmol/mg protein) by HPLC‐MS/MS in control and cystinotic zebrafish (n = 3).

Quantification of cystine levels (nmol/mg protein) by HPLC‐MS/MS in cystinotic zebrafish after treatment with cysteamine (1,000 µM), bicalutamide (10 µM), and a combination of cysteamine and bicalutamide (1,000 and 10 µM, respectively) (n = 40 embryos per group).

Survival rates in ctns−/− zebrafish upon treatment with cysteamine, bicalutamide, and a combination of cysteamine and bicalutamide (n = 40 embryos per group).

Deformity rates in ctns−/− zebrafish after treatment with cysteamine, bicalutamide, and a combination of cysteamine and bicalutamide (n = 40 embryos per group).

Hatching rates in surviving ctns−/− zebrafish evaluated at 72‐ and 96‐h post‐fertilization (hpf) with cysteamine, bicalutamide, and a combination of cysteamine and bicalutamide (n = 40 embryos per group).

Data information: Data are expressed as mean ± SEM. *P‐values < 0.05 were considered to be significant. One‐way ANOVA with Dunnett’s correction (B, C, D, E, G, H and J) or unpaired t‐test (C, D and H). Exact P‐values and statistical tests are listed in Appendix Table S1.

Working model summarizing the results obtained in this work

The blue arrows indicate wild‐type situation, red arrows/text indicate the changes in cystinotic cells, and green arrows indicate intervention with medication. DMKG: dimethyl α‐ketoglutarate, ROS: Reactive oxygen species.

Generation of <italic>CTNS</italic><sup>−/−</sup> isogenic cell line of ciPTEC using CRISPR

Schematic overview of the CRISPR‐based strategy to knockout the CTNS gene in ciPTEC.

Sanger sequencing chromatogram shows resulting sequence in CRISPR‐generated cystinotic cells (CTNS−/−).

Quantification of cystine levels (nmol/mg protein) by HPLC‐MS/MS in control (CTNSWT), CRISPR‐generated cystinotic cells (CTNS−/−; line 3, 7 and 35, and patient‐derived cystinotic cells (CTNSPatient) (n = 3–6).

Quantification of cystine levels (nmol/mg protein) by HPLC‐MS/MS in CTNS−/− lines (3, 7 and 35), and CTNSPatient cells upon treatment with cysteamine (100 µM) (n = 3–6).

Data information: Data are expressed as mean ± SEM. *P‐values < 0.05 were considered to be significant. One‐way ANOVA with Dunnett’s correction (C and D) or unpaired t‐test (D). Exact P‐values and statistical tests are listed in Appendix Table S1.

Decreased retention of mTOR on the lysosomal membrane of cystinotic cells

Representative immunofluorescent staining of CTNSWT, CTNS−/− and CTNSPatient co‐immunolabelled with lysosomal‐associated membrane protein 1 (LAMP1; green) and mTOR (Red) (n = 3). Merge images with zoomed areas are representative of the localization of mTOR with lysosomes in various experimental conditions. Scale bars are 10 µm.

G

Cysteamine–bicalutamide combination treatment efficiently lowers lysosomal cystine, abolishes αKG‐mediated autophagy distortion and cell death in cystinotic proximal tubule cells

Cell viability test of DMKG (2 mM)‐treated CRISPR‐generated cystinotic cells (CTNS−/−) upon pre‐treatment with cysteamine (100 µM), bicalutamide (35 µM), and a combination of cysteamine and bicalutamide (100 and 35 µM, respectively). Results are shown relative to the starved condition (n = 3).

Western blotting and densitometric analyses for LC3‐II/LC3‐I ratio in DMKG (2 mM)‐treated CTNSWT and CTNS−/− cells upon pre‐treatment with cysteamine, bicalutamide, and a combination of cysteamine and bicalutamide (n = 3).

Relative reactive oxygen species (ROS/mg protein) production in DMKG‐treated CTNSWT and CTNS−/− cells upon pre‐treatment with cysteamine, bicalutamide, and a combination of cysteamine and bicalutamide (n = 3).

Representative confocal micrographs and quantification of DQ‐BSA in CTNSWT, and CTNS−/− cells upon treatment with bicalutamide. Scale bars are 20 µm (n = 8–14 quantified images).

Quantification of cystine levels (nmol/mg protein) by HPLC‐MS/MS in CTNS−/− in the absence of the drug (NT) (n = 4) or upon treatment with cysteamine (n = 6), bicalutamide (n = 6), and a combination of cysteamine and bicalutamide (n = 6).

Quantification of cystine levels (nmol/mg protein) by HPLC‐MS/MS in CTNSPatient in the absence of the drug (NT) (n = 4) or upon treatment with cysteamine (n = 4), bicalutamide (n = 4), and a combination of cysteamine and bicalutamide (n = 6).

Data information: Data are expressed as mean ± SEM. *P‐values < 0.05 were considered to be significant. One‐way ANOVA with Dunnett’s correction (A, B, C, D, E, F and G) or unpaired t‐test (B, C, E, F and G). Exact P‐values and statistical tests are listed in Appendix Table S1.

Source data are available online for this figure.

G

Cysteamine‐bicalutamide combination treatment is safe in patient‐derived cystinotic kidney tubuloids and in wild‐type zebrafish

Patient‐derived cystinotic tubuloids (CTNSPatient‐1 and CTNSPatient‐2) and tubuloids established from healthy kidney tissue (CTNSWT‐1 and CTNSWT‐2) were differentiated for 7 days and analysed by quantitative real‐time PCR for markers of various segments of the nephron (n = 1).

Brightfield images of cystinotic tubuloids and healthy control tubuloids at the start of treatment and after 5 days of cysteamine (100 μM)‐bicalutamide (35 μM) combination treatment or treatment with medium only (negative control). Scale bars are 2,000 µm.

Bicalutamide safety screening in cystinotic tubuloids. Tubuloid viability upon treatment with cysteamine (100 μM) in combination with increasing concentrations of bicalutamide was compared to treatment with cysteamine alone (= 100% viability) (per donor n = 4 replicates for each dose).

Survival rates in wild‐type zebrafish upon treatment with bicalutamide (10 µM), and a combination of cysteamine and bicalutamide (1,000 and 10 µM, respectively) (n = 40 embryos per group).

Deformity rates in wild‐type zebrafish upon treatment with bicalutamide (10 µM), and a combination of cysteamine and bicalutamide (1,000 and 10 µM, respectively) (n = 40 embryos per group).

Hatching rates in surviving wild‐type zebrafish evaluated at 72‐ and 96‐h post‐fertilization (hpf) with bicalutamide (10 µM), and a combination of cysteamine and bicalutamide (1,000 and 10 µM, respectively) (n = 40 embryos per group). Drugs were administered at 48‐h post‐fertilization in all experiments dissolved in the swimming water with the specified concentrations.

Data information: Data are expressed as mean ± SEM.

αKG is a key metabolite responsible for impaired autophagy and proximal tubule dysfunction in cystinotic proximal tubule cells

Relative reactive oxygen species (ROS/mg protein) production in control (CTNSWT), CRISPR‐generated cystinotic cells (CTNS−/−) and patient‐derived cystinotic (CTNSPatient) cells under fed condition (n = 3).

ROS production (ROS/mg protein) in CTNSWT, CTNS−/− and CTNSPatient cells upon starvation for 4 h in the presence and absence of DMKG (4 h), respectively (n = 3).

Viability test in starved (−AA) CTNSWT, CTNS−/− and CTNSPatient cells in the presence or absence of DMKG (2 mM) for 24 h. Results are shown relative to the fed condition (n = 3).

Representative confocal micrographs (scale bars are 20 µm) and immunofluorescence analysis of caspase 3/7 activation in DMKG (2 mM)‐treated CTNSWT, CTNS−/− and CTNSPatient cells for 24 h (n = 3).

Western blotting and densitometric analyses for LC3‐II/LC3‐I ratio in CTNSWT, CTNS−/− and CTNSPatient cells cultured in the presence or in the absence of BafA1 (25 nM) and DMKG (2 mM) for 4 h, respectively (n = 3).

Data information: Data are expressed as mean ± SEM. P‐values < 0.05 were considered to be significant. One‐way ANOVA with Dunnett’s correction (A, B, C, D, E, G and I) or Unpaired t‐test (B, C, D, E and G). Exact P‐values and statistical tests are listed in Appendix Table S1.

Source data are available online for this figure.

CRISPR‐generated <italic>CTNS</italic><sup>−/−</sup> ciPTEC demonstrate cystinosis phenotype

Quantification of cystine levels (nmol/mg protein) in control (CTNSWT) (n = 4), CRISPR‐generated cystinotic cells (CTNS−/−) (n = 6), and patient‐derived cystinotic cells (CTNSPatient) (n = 4).

Quantification of transcription factor EB (TFEB)‐GFP nuclear translocation in CTNSWT, CTNS−/−, and CTNSPatient, respectively. Data are demonstrated as the ratio between number of the cells with nucleus‐TFEB positive over the total number of TFEB‐transfected cells. The ratios were then presented as a fold change compared to control cells (n = 3).

TFEB mRNA expression in CTNS−/− and CTNSPatient cells compared to control cells (n = 3).

Representative confocal micrographs and quantification of LC3‐II accumulation in CTNSWT and CTNS−/− cells in presence and absence of 25 nM bafilomycin (BafA1) for 4 h, respectively (n = 3). Scale bars are 20 µm.

Western blotting and densitometric analyses for LC3‐II/LC3‐I ratio and SQSTM1 protein levels in CTNSWT, CTNS−/− and CTNSPatient cells cultured in the presence or in the absence of 25 nM BafA1 for 4 h, respectively (n = 3).

Representative confocal micrographs and quantification of DQ‐BSA and BSA in CTNSWT, CTNS−/− and CTNSPatient cells, respectively (n = 7–14 quantified images). Scale bars are 20 µm.

Data information: Data are expressed as mean ± SEM. *P‐values < 0.05 were considered to be significant. One‐way ANOVA with Dunnett’s correction (A, B, C, E, G, H, I and J) or Unpaired t‐test (E, G and H). Exact P‐values and statistical tests are listed in Appendix Table S1.

Source data are available online for this figure.

Acknowledgments
This image is the copyrighted work of the attributed author or publisher, and ZFIN has permission only to display this image to its users. Additional permissions should be obtained from the applicable author or publisher of the image. Full text @ EMBO Mol. Med.