The expression patterns of zebrafish penka and human PENK in the kidneys.

a scRNA-seq analysis revealed that penka was specifically expressed in zebrafish PTECs. t-SNE plots showing zebrafish kidney cell clusters and the expression of penka. VECs, vascular endothelial cells; DTECs, distal tubular epithelial cells; Mφ, macrophages; HSCs, hematopoietic stem cells; MSCs, mucin-secreting cells; and RICs, renal interstitial cells. b Confocal images showing double labeling of FISH-penka and anti-Pax2a in un-injured (Un-Inj) adult zebrafish kidney sections (n  =  3). c Confocal images showing triple labeling of FISH-slc20a1a, FISH-trpm7, and anti-Met-ENK in Un-Inj adult zebrafish kidney sections. The Met-ENK signal co-localized with the signals of slc20a1a and trmp7, which are markers of PCT and PST, respectively (n  =  3). d Confocal images of combined FISH-PENK and LTL staining in kidney sections of patients with AKI and patients with no detectable lesions (Un-Inj). Human PENK was expressed in PTs and downregulated after AKI (n  =  3). e, f RT-PCR (e) and qRT-PCR (f) analyses of penka in zebrafish kidneys during Gent-induced AKI (n  =  3). penka expression was decreased by 1 dpi and reached its lowest level at 7 dpi and returned to its un-injured level at 15 dpi. The data in (f) were analyzed by two-sided t-test and are presented as mean values  ±  SD. Scale bars in (b), (c), and (d), 50 μm. Source data are provided as a Source data file.

PENK-A deficiency accelerates kidney regeneration.

a RT-PCR analysis of lhx1a expression in WT and penka^−/− zebrafish kidneys during Gent-induced AKI. b The band intensities in a are normalized to that of the loading control, β-actin, and the relative expression levels of lhx1a were quantified (n  =  3 biological replications per group). The data are presented as the fold change relative to the 0 dpi WT groups. c WISH analysis of lhx1a was performed in WT and penka^−/− zebrafish kidneys without injury (Un-Inj) and at 5, 7, and 9 dpi. d Quantitation of lhx1a⁺ RPCAs (blue points) per kidney (n = 5) for each condition in (c). e, f RT-PCR (e) and WISH (f) analyses of lhx1a in Un-Inj, 5, 7, and 9 dpi kidneys after administration of NAL-M or DMSO (n = 3 in e, n = 4 in f). g The lhx1a⁺ RPCAs per kidney (n = 4) were quantified for each condition in (f). The data in (b), (d), and (g) were analyzed by two-sided t-test and are presented as mean values  ±  SD. Scale bars in (c) and (f), 600 μm. Source data are provided as a Source data file.

PENK-A acts as a brake in kidney regeneration.

a‒e RT-PCR (a, c), qRT-PCR (b, d), and WISH (e) analyses of lhx1a were performed on kidneys administered with Met-ENK (a, b) or TRAM (c, d) at various doses at 7 dpi after AKI (n  =  3 in b and d). Met-ENK 50, 50 μM Met-ENK; Met-ENK 100, 100 μM Met-ENK; TRAM 10, 10 μM TRAM; and TRAM 16, 16 μM TRAM. f The lhx1a⁺ RPCAs per kidney (n = 5) were quantified for each condition in (e). The data in (b) and (d) are presented as the fold change relative to the Un-Inj group. g RT-PCR analysis of penka and lhx1a in WT and Tg(hsp70l:penka) zebrafish kidneys with heat shock (HS) or without HS (Un-HS) in the Un-Inj group or at 7 dpi (n = 3). h WISH analysis of lhx1a in WT and Tg(hsp70l:penka) zebrafish kidneys with HS (n = 6 in WT group, n = 10 in Tg(hsp70l:penka) group) or Un-HS (n = 9 in WT group, n = 6 in Tg(hsp70l:penka) group) in the Un-Inj group or at 7 dpi. i Quantitation of lhx1a⁺ RPCAs per kidney for each condition in (h). j‒l RT-PCR (j), qRT-PCR (k), and WISH (l) analyses of lhx1a at 7 dpi after administration (at 4 and 6 dpi) of Met-ENK 100 (100 μM Met-ENK, 10 μL per fish) or TRAM 16 (16 μM TRAM, 10 μL per fish) following AKI. The data in (k) (n = 3) are presented as the fold change relative to the Un-Inj group. m The lhx1a⁺ RPCAs per kidney (n = 5) were quantified for each condition in (l). n Confocal images showing adult Tg(lhx1a:DsRed) kidneys at 5 dpi after administration (at 2 and 4 dpi) of Met-ENK or DMSO following AKI (n = 9). o Quantitation of the individual RPCs (iRPCs, arrowhead) and RPCAs in (n). Data in (b), (d), (f), (i), (k), (m), and (o) were analyzed by two-sided t-test and are presented as mean values  ±  SD. Scale bars in (e), (h), and (l), 600 μm; (n) 50 μm. Source data are provided as a Source data file.

PENK-A regulates H₂O₂ production.

a Gene expression of PENK-A receptors (ogfr, ogfrl1, and ogfrl2) in kidney cells. VECs (vascular endothelial cells), DTECs (distal tubular epithelial cells), Mφ (macrophages), HSCs (hematopoietic stem cells), MSCs (mucin-secreting cells), and RICs (renal interstitial cells). b Confocal images of H₂O₂ signal in adult Tg(cdh17:DsRed) kidneys after Gent-induced AKI, detected using the PBSF fluorescence probe. Scale bar, 50 μm. c Relative H₂O₂ concentration in kidneys after NAL-M or DMSO administration following AKI, presented as fold change relative to the DMSO-treated Un-Inj group (n = 3). d Relative H₂O₂ concentrations in the kidneys of WT, penka^+/−, and penka^−/− zebrafish following AKI, presented as fold change relative to the Un-Inj WT group (n = 3). e H₂O₂ concentration in kidneys after TRAM 10 (10 μM TRAM, 10 μL per fish), TRAM 16 (16 μM TRAM, 10 μL per fish), or DMSO administration following AKI, presented as fold change relative to the DMSO-treated Un-Inj groups (n = 3). f H₂O₂ concentration in kidneys after Met-ENK 50 (50 μM Met-ENK, 10 μL per fish), Met-ENK 100 (100 μM Met-ENK, 10 μL per fish), or DMSO administration following AKI, presented as fold change relative to the DMSO-treated Un-Inj groups (n = 3). g H₂O₂ concentrations in the kidneys of WT and Tg(hsp70l:penka) with HS or without HS (Un-HS) following AKI, presented as fold change relative to the Un-HS and Un-Inj WT groups (n = 3). No significant differences between Un-HS WT, HS WT, and Un-HS Tg(hsp70l:penka) zebrafish were found using two-tailed t-test. h, i RT-PCR (h) and WISH (i) analyses of lhx1a in 5 dpi penka^−/− kidneys after administration (at 2 and 4 dpi) of VAS2870, duox1 vivo-MO, Con vivo-MO, or DMSO after AKI, Scale bar, 600 μm. j Quantification of lhx1a⁺ RPCAs per kidney (n = 5) for each condition in (i). Data in (c), (d), (e), (f), (g), and (i) were analyzed by two-sided t-test and are presented as mean values  ±  SD. p values are listed. Source data are provided as a Source data file.

The PENK-A–H₂O₂ pathway affects the remodeling of global H3K4me3 in kidney cells.

a Western blot analysis of H3K4me3 levels in WT and penka^−/− zebrafish kidneys following AKI. b The protein band intensities in a were normalized to the loading control, Histone 3 (H3), and the relative expression levels of H3K4me3 were quantified (n = 3 biological replications per group). The data are presented as the fold change relative to the Un-Inj WT group. c Western blot analysis of H3K4me3 levels in WT and penka^−/− zebrafish kidneys at 3 dpi after administration (at 2 dpi) of duox1 vivo-MO, VAS2870, or DMSO following AKI. d The protein band intensities in (c) were normalized to the loading control, H3, and the relative expression levels of H3K4me3 were quantified (n = 3 biological replications per group). The data are presented as the fold change relative to the Un-Inj WT group. e Western blot analysis of H3K4me3 levels in the kidneys at 3 dpi after administration (at 2 dpi) of CPI-455 or DMSO following AKI. f The protein band intensities in e were normalized to the loading control, H3, and the relative expression levels of H3K4me3 were quantified (n = 3). The data are presented as the fold change relative to the Un-Inj groups. g, h RT-PCR (g) and WISH (h) analyses of lhx1a at 7 dpi after administration (at 2, 4, and 6 dpi) of CPI-455 or DMSO following AKI (n = 3). i The lhx1a⁺ RPCAs per kidney (n = 5) were quantified for each condition in (h). j Confocal images showing adult Tg(lhx1a:DsRed) kidneys at 5 dpi after administration (at 2 and 4 dpi) of duox1 vivo-MO (n = 4), CPI-455 (n = 7), or DMSO (n = 8) following AKI. Scale bar, 100 μm. k Quantitation of individual RPCs (iRPCs, arrowheads) and RPCAs in (j). The data in (b), (d), (f), (i), and (k) were analyzed by two-sided t-test and are presented as mean values  ±  SD. Source data are provided as a Source data file.

The PENK-A–H₂O₂ pathway regulates tcf21 expression through promoter H3K4me3 remodeling.

a ChIP-seq analysis of the H3K4me3 pattern in the promoter region of tcf21. The H3K4me3 level upstream of the ATG start codon (red box) was decreased significantly at 3 dpi and increased at 5 dpi. b qRT-PCR analysis of tcf21 in zebrafish kidneys during AKI. The data were presented as the fold change relative to the 0 dpi group (n = 3 biological replications per group). c FACS coupled with RT-PCR analysis of tcf21 expression in lhx1a:DsRed-labeled RPCs at 5 dpi after AKI. Water was used as the RT-PCR negative control. Mk, Marker. d Confocal images revealed that the combination of tcf21 FISH with Pax2a immunofluorescence showed high expression of tcf21 in Pax2a⁺ RPCAs at 5 dpi after AKI. Scale bar, 50 μm. e qRT-PCR analysis of tcf21 in WT and penka^−/− kidneys following AKI (n = 3 biological replications per group). The data were presented as the fold change relative to the 0 dpi WT group. f qRT-PCR analysis of tcf21 in 7 dpi WT kidneys after administration (at 2, 4, and 6 dpi) of Met-ENK, Con vivo-MO, duox1 vivo-MO, VAS2870, CPI-455, and DMSO following AKI (n = 3 biological replications per group). The data were presented as the fold change relative to the 7 dpi DMSO-treated group. The data in (b), (e), and (f) were analyzed by two-sided t-test and are presented as mean values  ±  SD. Source data are provided as a Source data file.

The PENK-A–H₂O₂ pathway regulates kidney regeneration through tcf21.

a Confocal images showing adult Tg(lhx1a:DsRed) kidneys at 5 dpi afer administration (at 2 and 4 dpi) of tcf21 vivo-MO, or Con vivo-MO following AKI (n = 5 biological replications per group). b Quantitation of individual RPCs (iRPCs, arrowhead) and RPCAs in a. c, d RT-PCR (n = 3) (c) and WISH (d) analyses of lhx1a in 7 dpi WT and penka^−/− kidneys with administration (at 2, 4, and 6 dpi) of tcf21 vivo-MO (n = 5 in WT group, n = 8 in penka^−/− group) or Con vivo-MO ((n = 5 in WT group, n = 4 in penka^−/− group) following AKI. e The lhx1a⁺ RPCAs per kidney were quantified for each condition in d. f Confocal images showing 5 dpi Tg(lhx1a:DsRed;hsp70l:tcf21) kidneys with HS (at 2 and 4 dpi) or Un-HS after administration (at 2 and 4 dpi) of Met-ENK (n = 9 in Un-HS group, n = 5 in HS group), CPI-455 (n = 9 in Un-HS group, n = 6 in HS group), or DMSO (n = 8) following AKI. g Quantitation of individual RPCs (iRPCs, arrowheads) and RPCAs in (f). h RT-PCR analysis of tcf21 and lhx1a in 7 dpi WT and Tg(hsp70l:tcf21) kidneys with HS (at 2, 4, and 6 dpi) or Un-HS after administration (at 2, 4, and 6 dpi) of Met-ENK, CPI-455 or DMSO following AKI (n = 3). i WISH analysis of lhx1a in 7 dpi WT and Tg(hsp70l:tcf21) kidneys with HS (at 2 and 4 dpi) or Un-HS after administration (at 2, 4, and 6 dpi) of Met-ENK (n = 5), CPI-455 (n = 5) or DMSO (n = 5 in Un-HS group, n = 4 in HS group) following AKI. j Quantitation of lhx1a⁺ RPCAs per kidney was performed for each condition in (i). Data in (b), (e), (g), and (j) were analyzed by two-sided t-test and are presented as mean values  ±  SD. Scale bars in (a) and (f), 50 μm; (d) and (i), 600 μm. Source data are provided as a Source data file.

Graphical abstract summarizing the research findings.

PTEC-expressed PENK-A levels decreased with the loss of PTs after AKI. This decrease triggered an increase in H₂O₂ production, resulting in the expansion of RPC aggregation. This process occurred by upregulating tcf21 expression through remodeling of H3K4me3 in the tcf21 promoter. In the later stage of regeneration, as PTECs recovered, elevated penka expression suppressed the production of H₂O₂, facilitating the termination of kidney regeneration.