Inhibition of retinoic acid signaling in proximal tubular epithelial cells protects against acute kidney injury

Abstract

Retinoic acid receptor (RAR) signaling is essential for mammalian kidney development but, in the adult kidney, is restricted to occasional collecting duct epithelial cells. We now show that there is widespread reactivation of RAR signaling in proximal tubular epithelial cells (PTECs) in human sepsis-associated acute kidney injury (AKI) and in mouse models of AKI. Genetic inhibition of RAR signaling in PTECs protected against experimental AKI but was unexpectedly associated with increased expression of the PTEC injury marker Kim1. However, the protective effects of inhibiting PTEC RAR signaling were associated with increased Kim1-dependent apoptotic cell clearance, or efferocytosis, and this was associated with dedifferentiation, proliferation, and metabolic reprogramming of PTECs. These data demonstrate the functional role that reactivation of RAR signaling plays in regulating PTEC differentiation and function in human and experimental AKI.

Keywords: Molecular pathology; Nephrology.

Figure 1. Activation of retinoic acid receptor (RAR) signaling in patients with sepsis-associated AKI.

Gene set enrichment analysis (GSEA) of bulk RNA as well as cell-specific snRNA-Seq kidney data sets from patients with SA-AKI compared with age-matched controls from Hinze et al. (11). (A and B) Volcano plots indicating fold change in expression of core enrichment genes (AKI versus controls) from bulk RNA and PTEC snRNA-Seq data sets. Dotted line indicates P < 0.05.

Figure 2. Widespread activation of RAR signaling after Rhabdo-AKI.

(A) BUN time course after Rhabdo-AKI in BALB/c mice: 15 mice before injury (day 0) and day 1 after injury, 9 at day 3, and 3 at day 7. (B) Expression of injury marker mRNAs. (C) RAR target gene mRNAs in 5 uninjured and day 1 mice, 4 at day 3, and 3 at day 7. (D) Spatial distribution and kinetics of RAR signaling after Rhabdo-AKI in RARE-LacZ mice. Kidneys stained for LacZ activity, 3 before injury (day 0); 4 at days 1 and 3; 3 at day 7; and 5 at day 14 after injury. (E) Percentage of area staining for LacZ in the cortex (C); OSOM; inner stripe of the outer medulla (ISOM); and inner medulla (IM). LacZ staining at different time points after injury. LacZ staining is pseudocolored in white, and kidney regions are demarcated by dotted lines in the first panel. Scale bars: 500 mM. B and C used 1-way ANOVA; if P < 0.05, q values are shown for between-group comparisons corrected for repeat testing.

Figure 3. RAR signaling is extensively activated in Kim1^– PTECs after Rhabdo-AKI.

RARE-LacZ reporter mice were used to evaluate the cellular localization of RAR signaling after Rhabdo-AKI. (A and C) Percentage of LacZ⁺ cells in LTL⁺ and Kim1^– PTECs (A) and in Kim1⁺ PTECs (C) at the different time points after injury in the cortex and OSOM: 2 mice before injury (day 0) and 4 at days 1 and 3, 3 at day 7, and 5 at day 14 after injury. (B and D) Images showing LacZ, LTL, and Kim1 staining in the cortex and OSOM. (B) Low-power images of the kidneys; regions of the kidney are demarcated by dotted lines, as indicated. Scale bars: 500 mM. (D) High-power images of the OSOM. Scale bars: 100 mM.

Figure 4. Inhibition of RAR signaling in PTECs protects against AKI but also increases Kim1 expression in PTECs.

PEPCK Cre⁺DN-RAR (PTEC DN RAR) mice and Cre^– controls mice underwent bilateral IRI-AKI, or Rhabdo-AKI, and kidneys were harvested after 3 days. (A–F) Bilateral IRI-AKI. (A) Survival curves. Mouse numbers indicated after each event. (B) BUN time course. Mouse numbers are indicated in A. (C) Tubular injury scores in the cortex and OSOM. (D) PAS-stained kidney images from the OSOM. Red arrows indicate necrosis; green arrows indicate detached epithelial cells. (E) Quantification of Kim1 staining. (F) Kim1 and LTL staining in OSOM. (G–L) Rhabdo-AKI. (G) Survival curves. (H) BUN time course. (I) Tubular injury scores. (J) PAS-stained kidney images. Red and green arrows, as described above; yellow arrows indicate tubular casts. (K) Quantification of Kim1 staining in the OSOM. (L) Kim1 and LTL staining in OSOM. Scale bars: 20 mM (D and J), 100 mM (F and K). B and H used 2-way ANOVA; P values are indicated, and if P < 0.05, q values are shown for between-group comparisons corrected for repeated testing at the indicated time points. C, E, I, and K used 1-way ANOVA, and if P < 0.05, q values are shown for between-group comparisons corrected for repeat testing.

Figure 5. PTEC DN RAR mice have increased expression of PTEC dedifferentiation and proliferation markers after AKI.

PTEC DN RAR mice underwent Rhabdo-AKI or bilateral IRI-AKI, and kidneys were harvested after 3 days. (A–D) Rhabdo-AKI. (A) The percentage of LTL⁺ PTECs that are Sox9⁺ in the OSOM. (B) Sox9, Kim1, and LTL staining. Left panels are low magnification, showing OSOM staining extending into the cortex. Right panels show higher magnification of the OSOM. (C) The percentage of LTL⁺ PTECs that are Ki67⁺ in the OSOM. (D) Ki67, Sox9, and LTL staining. (E–G) Bilateral IRI-AKI. (E and F) The percentage of LTL⁺ cells that are Sox9⁺ and Ki67⁺ in the OSOM 3 days after IRI-AKI. (G) Sox9 and Ki67 staining in the OSOM. Scale bars: 100 mM. (A, C, E, F) 1-way ANOVA; if P < 0.05, then q values are shown for between group comparisons corrected for repeated testing.

Figure 6. Increased expression of Kim1 and dedifferentiation markers in uninjured PTEC DN RAR mouse kidneys.

(A) The percentage of the area of Kim1 staining, and the percentage of LTL⁺ PTECs that are Sox9⁺ or Ki67⁺ in the cortex and OSOM of uninjured PEPCK Cre⁺ and Cre^– mice. Ki67, Sox9, and Kim1 staining in uninjured Cre⁺ and Cre^– kidneys shown in Figure 5, B and D. (B) Renal Kim1, Sox9, and FoxM1 mRNA expression. (C) Tubular injury scores. (D) PAS-stained images showing areas with flattened epithelia (green arrow) and peritubular inflammatory cells (red arrow) in the OSOM of uninjured Cre⁺ and Cre^– kidneys. Scale bars: 50 mM. The OSOM Sox9 and Ki67 data in A are the same as the uninjured controls in Figure 5, A and C. A–C used t tests, with P values comparing PEPCK Cre⁺ and Cre^– mice.

Figure 7. Inhibition of RAR signaling increases proliferation, glycolysis, and oxidative phosphorylation in cultured PTECs.

Data are shown for replicates performed on separate preparations of primary PTECs isolated from different mice. (A) RAR target genes. (B) Expression of Kim1 mRNA. (C) Cell growth. Cell numbers/10× microscopy field at the indicated time points in PTECs from 4 PEPCK Cre⁺ and 4 Cre^– mice. (D) Glycolytic stress tests on PTECs from 6 Cre⁺ (red) and 6 Cre^– (black) mice. (E) Glycolytic rate, maximal glycolytic capacity, and glycolytic reserve. (F) Mitochondrial stress test. (G) Basal mitochondrial respiration, maximal mitochondrial respiration, and spare respiratory capacity. A, E, and G used t tests, with P values comparing PEPCK Cre⁺ and Cre^– PTECs. D and F used 2-way ANOVA, with P values comparing PEPCK Cre⁺ and Cre^– PTECs over time.

Figure 8. Increased efferocytosis in PTEC DN RAR mice after AKI.

(A–E) Apoptosis in PTEC DN RAR mice 3 days after Rhabdo-AKI. (A) The percentage of TUNEL⁺LTL⁺ cells in the cortex and OSOM. (B) TUNEL and LTL staining in the OSOM. (C) Percentage of MLKL staining. (D) MLKL with LTL and Kim1 staining in the OSOM. (E–G) Kim1 expression and endocytic activity in cultured PTECs. (E) Kim1 fluorescence intensity. (F) PTECs uptake of fluorescently conjugated oxidized LDL (Ox-LDL). Fluorescence intensity and numbers of PTECs taking up Ox-LDL. (G) Kim1 staining and Ox-LDL uptake. (H and I) Lysosomal clearance of apoptotic cells in PTEC DN RAR mice. (H) PTEC DN RAR mice after bilateral IRI-AKI pretreated with bafilomycin A1. Kidneys harvested at 24 hours. The percentage of TUNEL⁺LTL⁺ cells in the OSOM. (I) TUNEL and LTL staining. Scale bars: 100 mM (B, D, E, and J), 20 mM (H). A, C, E, and F used t tests, with P values comparing PEPCK Cre⁺ and Cre^– mice. H used 1-way ANOVA, with P < 0.05 considered significant; q values are shown for between-group comparisons corrected for repeated testing.

Figure 9. Increased renal macrophages with a reduced proinflammatory signature in PTEC DN RAR mice after AKI.

PTEC DN RAR mice underwent Rhabdo- or bilateral IRI-AKI, and kidneys were harvested after 3 days. (A and B) Rhabdo-AKI. The percentage of F4/80 area staining in the OSOM, and images showing F4/80, Sox9, and LTL staining. Left panels show F4/80 staining largely restricted to the OSOM. Right panels show higher magnification of the OSOM. (C and D) Bilateral IRI-AKI. Quantification, and F4/80 and LTL staining in the OSOM. (E and F) Expression of inflammatory markers in CD11B⁺ cells after IRI-AKI. PTEC DN RAR mice underwent bilateral IRI-AKI, and bulk RNA-Seq was performed on renal CD11B⁺ cells 3 days after injury. (E) GSEA of downregulated genes using GO data sets. (F and G) GSEA for proinflammatory (“M1”) and antiinflammatory (“M2”) gene sets in the RNA-Seq data. Set size=no. of genes from each gene set that are represented. NES=normalized expression score for the gene set sizes. Volcano plots showing fold change in expression of core enrichment gene from the CD11B⁺ RNA-Seq data set. Dotted line indicates P < 0.05. A and C used 1-way ANOVA, with P < 0.05 considered significant; q values shown for between-group comparisons corrected for repeated testing. Scale bars: 100 μM.

Figure 10. Flow cytometric analysis of renal monocyte/macrophages in PTEC DN RAR mice.

Kidneys were harvested from PTEC DN RAR mice 3 days after bilateral IRI-AKI. (A and B) F4/80^hi cells (kidney-resident macrophages). (A) F4/80^hi cells as the percentage of gated CD11B⁺CD45⁺Ly6G^– cells. (B) CD11B and F4/80 expression charts in PEPCK Cre⁺ and Cre^– mice after IRI-AKI (percentage in gated area indicated). (C–F) Ly6C^hi cells (inflammatory monocyte/macrophages). (C and E) F4/80^– (infiltrating monocytes) Ly6C^hi cells and F4/80^int (BM-derived macrophages) Ly6C^hi cells, as the percentage of gated F4/80^– and F4/80^int cells, respectively. (D and F) CD11B and Ly6C expression charts. (G and H) CD206/mannose receptor expression (an M2 activated macrophage marker). (G) CD206⁺ cells as the percentage of gated F4/80^hi cells. (H) CD11B and CD206 expression charts. A, C, E, and G used 1-way ANOVA, with P < 0.05 considered significant; q values shown for between-group comparisons corrected for repeated testing.