Animal experimentation was approved by the Institutional Animal Care and Use Committee of the Graduate School of Veterinary Medicine, Hokkaido University (approval No. 16-0124). The investigators adhered to the Guide for the Care and Use of Laboratory Animals of Hokkaido University, Graduate School of Veterinary Medicine (approved by the Association for the Assessment and Accreditation of Laboratory Animal Care International).

Two single guide RNA (sgRNA) designed for the target sequences (5′-CCTAGGGTCAATCTGCAGAT-3′ and 5′-AGGGGGGATC CCACGTACAT-3′), and Cas9 mRNA were prepared as described previously (13). Briefly, an sgRNA expression vector with a T7 promoter was synthesized for the target sequence and transcribed in vitro using a MEGAshortscript kit (Life Technologies, Carlsbad, CA, USA). hCas9 mRNA was synthesized using mMESSAGE mMACHINE T7 kit (Life Technologies) and was polyadenylated with polyA tailing kit (Life Technologies). The purified hCas9 mRNAs at 100 ng/mL and each sgRNAs at 50 ng/mL targeting exon 2 and exon 3 for Ilf6 (coding gene of IL-36α) were co-injected into the cytoplasm of the pronuclear stage eggs from C57BL/6N (B6; Japan SLC; Shizuoka, Japan), and the eggs were transferred into the oviducts of pseudopregnant ICR female mice. The founders were genotyped with primer pairs spanning the CRISPR/Cas9 cleaved site (Table S1 in Supplementary Material). The mutation of Il1f6 gene was confirmed by sequencing analysis. To minimize the risk of off-target effects, the resulting founder (13) mouse was backcrossed to B6 mice for two generations, and the heterozygous IL-36α-KO mice were then intercrossed to produce the homozygous IL-36α-KO mice and wild-type (WT) mice that were used for this study. The genotype of the obtained littermates was confirmed by genotyping PCR analysis using specific primers for Il1f6 mutation (Table S1 in Supplementary Material). The potential off-target site was predicted by CRISPR Design Tool (14) and listed in Table S2 in Supplementary Material.

B6 mice were used in all in vivo study. UUO operation was performed to male mice at 10 weeks of age or IL-36α-KO mice at 4 weeks of ages. At 0–11 days after UUO, both kidneys and serum were collected. BrdU (100 mg/kg, i.p.) was injected to some mice 2 h before sampling. A part of each kidney was fixed in paraformaldehyde (PFA) or glutaraldehyde (GTA), and the remaining parts were stored at −80°C.

For the FA-induced AKI model, FA (800 mg/kg, i.p., Sigma-Aldrich; St. Louis, MO, USA) was injected to male B6 mice at 10 weeks of age. At 8 h after injection, the serum, urine, and kidneys were collected. A part of each kidney was fixed in 4% PFA, and the remaining parts were stored at −80°C. Urinary Cr was determined in FA models by the Creatinine Companion (Exocell; Philadelphia, PA, USA). Soluble proteins were extracted by using RIPA lysis buffer (Santa Cruz; Dallas, TX, USA) from urinary sediments and urinary exosomes. Urine (400 µL) was centrifuged at 21,000 × g for 10 min, and its sediments and supernatants were used. Lithium dodecyl sulfate-sample buffer and sample reducing reagent (Thermo Fisher Scientific; Waltham, MA, USA) were added to the samples or directly added to urinary supernatant. The serum levels of BUN and Cr were measured with Fuji Drichem (Fujifilm Medical Co. Ltd.; Tokyo, Japan).

M-1, MES13, NIH3T3, and AI cell lines [kindly provided from Dr. J.B. Kopp, National Institutes of Health (NIH)] were maintained as previously described (15, 16). Immortalized mouse PT cells (Figure S1 in Supplementary Material) and 209/MDCT (American Type Culture Collection; Manassas, VA, USA) cells were maintained in REGM BulletKit (Lonza; Basel, Switzerland) and DMEM/F12 medium containing 10% fetal bovine serum (FBS) and 1× penicillin/streptomycin (PS; Thermo Fisher Scientific), respectively. For 209/MDCT, after stimulation by NaCl, KCl, H₂O₂, or LPS (Wako; Osaka, Japan) under FBS-free condition, cells were collected for RNA and protein analysis, or were fixed by using 4% PFA and stained by immunofluorescence method using IL-36α antibody (Table S3 in Supplementary Material), phalloidin (Life Technologies), and Hoechst 33342 (Dojindo; Kumamoto, Japan). NIH3T3 cells were stimulated by recombinant mouse IL-36α (7059-ML-010, R&D Systems; Minneapolis, MN, USA), and the culture medium and cells were collected for RNA and protein analysis. IL-6 in the culture medium was measured by ELISA (IBL; Gunma, Japan).

Total RNA was extracted from frozen UUO and collateral control (Cont) kidneys of B6 mice at 7 days or WT and IL-36α-KO mice at 48 h. One pooled sample collected from four kidneys was analyzed in each group. Gene expression was analyzed using a GeneChip Mouse Gene 2.0 ST Array (Affymetrix; Santa Clara, CA, USA). Microarray signals were normalized by using the RMA algorithm. Generic gene ontology (GO) term finder (http://go.princeton.edu/cgi-bin/GOTermFinder) was used for GO analysis.

Total RNA from frozen kidneys was isolated and used as a template to synthesize cDNA using ReverTra Ace qPCR RT Master Mix (Toyobo; Osaka, Japan). Quantitative PCR analysis was performed by using Brilliant III SYBR Master Mixes for kidneys (Agilent; Santa Clara, CA, USA) or TaqMan Real-Time PCR Master Mixes for cells (Thermo Fisher Scientific) and specific primers or probes (Table S1 in Supplementary Material) with an MX3000P system (Agilent). The specificity of each PCR reaction was confirmed by melting curve analysis. The expression data were normalized to the expression levels of Actb or Gapdh.

From collected kidneys, cells, and urines, soluble proteins were extracted by using RIPA lysis buffer (Santa Cruz). Lithium dodecyl sulfate-sample buffer and sample reducing reagent (Thermo Fisher Scientific) were added to the samples. Immunoblotting was performed by using the NuPAGE electrophoresis system (Thermo Fisher Scientific) with the antibodies listed in Table S3 in Supplementary Material. Immunocomplexes were detected by using Typhoon Variable-Mode Imagers (GE Healthcare; Little Chalfont, UK). The intensity of band was quantified using Image J (NIH).

Paraffin-embedded kidney sections were stained with periodic acid Schiff (PAS) or Masson’s trichrome for histopathological analysis, and the area of renal pelvis (RP) was measured using ImageJ (NIH) or BZ-H3M (Keyence; Osaka, Japan). Sections were analyzed by immunohistochemistry or immunofluorescence using primary antibodies listed in Table S3 in Supplementary Material as described previously (9, 16) and evaluated for the number of B220⁺ cells, Gr1⁺ cells, Iba1⁺ cells, single strand DNA (ssDNA)⁺ cells, Caspase 3⁺ cells, and IL-36α⁺ tubules. The signals detected in areas positive for CD3, BrdU, alpha-smooth muscle actin (αSMA), and IL-36α were measured by using ImageJ (NIH) or BZ-H3M (Keyence).

Four percent PFA/0.1 M phosphate buffer (PB) containing 2.5% GTA or 4% PFA/0.1 M containing 0.5% GTA was used for the fixation of transmission electron microscopy (TEM) and immunoelectron microscopy (IEM) specimens, respectively. For IEM, kidneys were fixed for 4 h, and PFA was replaced with 0.1 M PB containing 10–30% sucrose. The kidneys were incubated overnight, and then embedded by using O.C.T. Compound (Sakura Finetech; Tokyo, Japan). Immunohistochemistry using 5-μm thick cryosections and streptavidin-biotinylated horseradish peroxidase complex (Nichirei; Tokyo, Japan) was performed to detect IL-36α. TEM specimens and kidney sections for IEM were post-fixed with 1% OsO₄ and embedded in Quetol 812 (Nisshin EM; Tokyo, Japan), and the ultrathin sections were stained with or without uranyl acetate and lead citrate for TEM and IEM, respectively.

For scanning electron microscopy (SEM), specimens were fixed with 2.5% GTA/0.1 M PB and cut into 200-µm sections by vibratome, and then post-fixed with OsO₄ and dehydrated in critical point dyer (HCP-2; Hitachi, Tokyo, Japan). For IEM using SEM, specimens were fixed with 0.5% PFA and 0.25% GTA. Immunohistochemistry was performed to detect IL-36α using 5-μm thick paraffin sections, IL-36α primary antibody, biotinylated secondary antibody, gold conjugated streptavidin-biotin, and gold enhance kit (Nanoprobe; Yaphank, NY, USA). After post-fixation by 0.5% OsO₄ and dehydration, ion-spattered sections were observed under SEM.

The results are displayed as the mean ± SE and were statistically analyzed by using a Mann–Whitney U test (P < 0.05). The correlation between two parameters was analyzed by using Spearman’s rank correlation test (P < 0.05).

First, microarray analysis was performed in mouse kidneys at 7 days after UUO. Among the IL-1 family, Il1f6 was the most upregulated gene in UUO kidneys; 2.67-fold vs. Cont kidneys (Figure 1A). Similar to Kim-1, Il1f6 mRNA expression in the kidneys increased from 12 h after UUO, and its elevation was prominent at 24 h compared to that of other members (Figure 1B). At 12 h, Il1rn, Il1f6, Il1f9, and Kim-1 mRNA expression was significantly higher in UUO kidneys than that in Control kidneys, but that of Il18 and Il1f5 was decreased (Figure 1C). At 24 h, Il1rn, Il1f6, and Kim-1 mRNA expression was more significantly increased in UUO kidneys. IL-36α was detected in the kidneys from 24 h after UUO (Figure 1D).

In UUO kidneys, the dilated tubules were observed at 0.5–24 h after UUO (Figure 2A) and were positive for calbindin D28k, a DT marker. IL-36α⁺ immunostaining appeared in DTs, first in the macula densa (MD), from 12 h after UUO, and the immunostaining spread to surrounding DT segments at 24 h (Figures 2A,B). In DTs, IL-36α⁺ immunostaining was observed in the cytoplasm and in some nuclei (Figure 2B). Increased IL-36α protein expression remained until 11 days after UUO without sex-related differences (Figures 2C,D). The number of IL-36α⁺ tubules increased from 12 h after UUO and was significantly higher in UUO kidneys than in Cont kidneys at 12 and 24 h (Figures 2E,F). In some Cont kidneys, a small number of cells were IL-36α⁺ in DTs, especially close to the MD (Figure 2G). The probability of IL-36α⁺ DT segments was significantly higher in DT segments attached to the renal corpuscle compared to the other DT segments in Cont kidneys; however, the latter DT segments showed higher values than the former DT segments in UUO kidneys (Figure 2H). These data indicated that DT segments attached to the renal corpuscle, including the MD, were initiation sites of IL-36α expression, which spread to other DT segments as kidney injury progressed. At 24 h after UUO, IL-36α expression was scarce in Cont kidneys, but clearly co-localized with calbindin D28k expression in UUO kidneys (Figure 2I).

In TEM, the structures of PT epithelial cells remained normal after 24 h of UUO, but that of DT epithelial cells presented a dilated tubular lumen and squamous features (Figure 3A). IEM clarified that DT epithelial cells strongly expressing IL-36α in the cytoplasm and nucleus showed loose intercellular digitations and randomized mitochondria arrangements (Figure 3B). IEM using SEM showed that IL-36α⁺ nanogold particles were observed in dilated DTs; These particles localized to the cytoplasm and nucleus as well as in extracellular regions beside the basement membrane and peritubular capillary (Figure 3C). Furthermore, long primary cilia were observed at the apical portion of DT epithelial cells in UUO kidneys rather than in Cont kidney at 48 h by SEM, and some of these cilia presented bulges and were interconnected (Figure 3D).

After 8 h of FA injection, serum levels of BUN and Cr were significantly increased in the FA groups compared to that in the vehicle groups, and urinary Cr was significantly decreased in the former groups (Figure 4A). At 8 h after FA injection, Il1rn and Il1f6 mRNA expression was significantly increased, but that of Il1f5 and Il1f8 was decreased in the kidneys of FA-AKI groups compared to that in the vehicle groups (Figure 4B). Kim-1 mRNA expression also tended to increase after FA injection without statistical significance (P = 0.062). IL-36α protein expression in the kidneys was increased at 8 h after FA injection (Figures 4C–E), and IL-36α⁺ immunostaining was observed in DT epithelial cells as well as in luminal cell debris (Figure 4D). In the urine, IL-36α protein was detected in the urinary sediment of FA-AKI individuals showing increased urinary cells suggested by increased β-actin expression, but not in their supernatant (Figure 4F, see #3, 6, 8). Table 1 summarizes the correlation between IL-36α expression and renal dysfunction of the FA-AKI model. Il1f6 mRNA expression significantly correlated with the serum levels of BUN and Cr and urinary Cr. IL-36α protein expression also significantly correlated with serum level of BUN and urinary Cr.

The correlation between IL-36α expression and kidney injury parameters was examined at 7 days after UUO because UUO kidneys at 7 days show severe interstitial inflammation. The ratio of kidney weight (KW)/body weight (BW) and the area of RP were significantly increased in UUO kidneys compared to those in Cont kidneys (Figures 5A,B). In UUO kidneys, infiltration of B220⁺ B-cells, CD3⁺ T-cells, Gr1⁺ granulocytes, and Iba1⁺ macrophages was observed at TILs (Figure 5C). Furthermore, BrdU⁺ proliferating cells and ssDNA⁺ or Caspase 3⁺ dead cells were detected in the tubular epithelium, and αSMA⁺ myofibroblasts increased. The quantified parameters, except for BrdU⁺ cells, and KW/BW, or the area of the RP, were significantly higher in UUO kidneys than in Cont kidneys (Figure 5D) and positively correlated with the number of IL-36α⁺ tubules (Table 2).

IL-36α induction was examined in 209/MDCT, a mouse DT epithelial cell line, and LPS stimulation for 24 h strongly induced IL-36α mRNA and protein (Figures 6A–C).

Next, the expression of IL-1RL2, a receptor of IL-36α, was examined in UUO kidneys at 48 h. IL-1RL2⁺ immunostaining was detected in PTs, podocytes, platelets, and interstitial cells (Figures 7A,B). Interestingly, dotted or rod-like IL-1RL2⁺ immunostaining was observed at the apical portion of the epithelial cells of DT, including the MD in UUO and Cont kidneys at 48 h (Figure 7A), indicating IL-1RL2 expression in primary cilia. At 12 and 24 h after UUO, Il1rl2 mRNA expression tended to increase in UUO kidneys compared to that in Cont kidneys without statistical significance (Figure 7C). Among the examined mouse cell lines, Il1rl2 mRNA expression was detected only in NIH3T3, a fibroblast line (Figure 7D). Therefore, IL-36α stimulation was performed in NIH3T3 cells (Figure 7E). Il1rl2 mRNA expression was not affected, but Il6 mRNA expression in cell lysates and IL-6 protein level in the culture medium significantly increased from 6 h after IL-36α stimulation.

Parts of exon 2 and exon 3 were deleted by the CRISPR/Cas9 system to generate IL-36α-KO mice (Figure 8A). At 48 h after UUO, KO mice expressed no IL-36α (Figure 8B). KW/BW, KW ratio (UUO/Cont), BUN, and serum Cr, were significantly higher at 48 h in the UUO group than in the Sham group in both WT and KO mice (Figure 8C). However, serum potassium concentration was significantly increased in the UUO group of WT mice, but not in that of KO mice compared with each Sham group, and a significant difference was observed between WT and KO mice in the UUO group. CD3⁺ T-cells and Gr1⁺ granulocytes were detected in 48 h UUO kidney in both WT and KO mice (Figure 8D), but infiltration of these cells and renal tubule dilation were significantly milder in UUO kidneys of KO mice than in WT mice (Figure 8E).

Table 3 summarized the GO analysis of genes differently expressed in the kidneys of WT and KO mice at 48 h after UUO. Genes involved in inflammatory response and acute inflammatory response were significantly downregulated (under 1.5-fold, P < 0.01) in the kidneys of KO mice compared to that in the kidneys of WT mice. Further, genes involved in sensory perception or detection of chemicals were significantly upregulated or downregulated (over or under 1.5-fold, P < 0.01). In particular, genes belonging to the olfactory receptor (Olfr) and vomeronasal receptor (Vmnr) family were frequently downregulated or upregulated (Figure S2 in Supplementary Material).

Figure 9A summarizes the genes differentially expressed in the kidneys of WT and IL-36α-KO mice at 48 h after UUO in microarray analysis. In particular, protease serine 35 (Prss35) mRNA expression was decreased in KO mice, suggesting Prss35 as a candidate downstream molecule of IL-36α signaling. NIH3T3 stimulated by IL-36α showed increased Prss35 mRNA and protein expression (Figures 9B–F), and Prss35⁺ immunostaining was detected in interstitial fibroblasts and DTs of UUO kidneys (Figure 9G).