Mlkl-ko mice on a C57BL/6J background (45) and WT littermate controls were generated by heterozygous matings and screened by PCR of tail tip genomic DNA by the Walter and Eliza Hall Institute for Medical Research (WEHI) (45). Mice were housed in microisolator cages in a pathogen-free facility with a 12-hour light-dark cycle under standard conditions of temperature and humidity and fed commercial mouse chow diet with free access to drinking water. All animal experiments were conducted in compliance with the Australian Code of Practice for the Care and Use of Laboratory Animals for Scientific Purposes (Eighth edition, 2013) and the Prevention of Cruelty to Animals Act 1986, Victoria, Australia. All experiments were carried out with the approval of the Animal Ethics Committee of St. Vincent’s Hospital Melbourne.

10–12-week-old male mice were anaesthetised by intraperitoneal injection of ketamine (100 mg/kg) and xylazine (15 mg/kg), followed by right nephrectomy prior to left renal pedicle clamping for 18 min using a microvascular clamp (Roboz, Rockville, MD). After removal of the clamp, the kidney was assessed for even reperfusion prior to abdominal wound closure. Mice received 200 µL warm normal saline (37°C) i.p. post-operatively, and core body temperature was maintained at 35.5 - 36.5°C throughout. Sham mice had a right nephrectomy but the left renal pedicle was not clamped. Separate cohorts of mice were euthanised by anesthetic overdose (ketamine/xylazine) and exsanguination immediately following injury (baseline), at 3, 12, 24, 48, 72 hours or 4 weeks after reperfusion. Blood and kidney samples were obtained to assess kidney function, kidney injury, inflammation, and necroptosis. Group sizes varied depending on availability of homozygous Mlkl-ko and WT littermates during experimental procedures. Specific group sizes are specified in all figure legends.

Serum creatinine was measured using a kinetic colorimetric assay on a COBAS Integra 400 Plus analyser (Roche, Castle Hill, NSW, Australia).

Kidney tissue blocks were fixed overnight in 10% formalin and embedded in paraffin wax. 3 µm sections were cut and de-waxed prior to Periodic Acid Schiff (PAS) staining (57). Each section was divided into 12 regions. A representative area of each region was viewed under 400X magnification, with a focus on the corticomedullary junction. A score was derived by calculating the number of damaged proximal tubules as a percentage of total proximal tubules manually counted in each area (58). Markers of a damaged tubule included tubular atrophy, tubular dilatation, tubular cast formation, vacuolization, and tubular cell degeneration with loss of brush border or thickening of tubular basement membranes. An average of the 12 scores obtained per section was calculated. Scoring was performed under blinded conditions by personnel trained by an experienced veterinarian pathologist.

3 µm paraffin sections were de-waxed prior to antigen retrieval in a citrate buffer (pH 6.0) using a pressure cooker at 125°C for 90 s (pMLKL staining) or 180 s (cC3 staining), and left to cool to room temperature. Slides were then washed on a shaker prior to incubation in 10 μL 3% H₂O₂ v/v in double-distilled water for 5 min and blocked with 50 μL 10% swine serum for 1 hr at room temperature. Sections were then incubated with rabbit monoclonal anti-phospho-MLKL (pSer345) (ab196436, 1:100, Abcam, Melbourne, Australia) or rabbit monoclonal anti-human/mouse cleaved Caspase-3 (cC3) (Asp175) (IC835G, 1:200, RD Systems, Noble Park, Australia) in 2% swine serum at 4°C overnight. After incubation with DAKO anti-rabbit IgG HRP secondary antibody (Envision+ System, K4003, 1:1, Dako, Santa Clara, USA) for 1 hr, sections were developed using DAB and counterstained with hematoxylin. pMLKL- or cC3-stained sections were scanned using an Aperio ScanScope (Leica Biosystems) to generate a digitized image of the whole section. 12 representative areas of the cortex were analysed (4 upper pole, 4 mid pole and 4 lower pole) at X400 magnification. A score was manually calculated by counting the number of tubules with pMLKL or cC3 staining as a percentage of the total number of tubules in the section.

Harvested kidneys were stored in RNA Later^® at 4°C for 24-48 h, followed by storage at -80°C until processing. Total RNA was extracted using the ReliaPrep ™ RNA Tissue Miniprep system (Promega Australia, Alexandria, NSW, Australia) according to manufacturer’s instructions. RNA concentration and quality were measured using a Fluorostar Omega multimode microplate reader (BMG Labtech, Mornington, Australia). First-strand complementary DNA (cDNA) was generated in a reaction volume of 22 µL containing 1 µg oligo (dT), 1 µg random hexamers (Invitrogen, Carlsbad, CA), 12 µg of RNA and sterile Milli-Q H₂O. The reaction was incubated for 10 min at 70°C. Following this, a 28 µL mix comprising 2.5 µL 10 mM dNTPs, 1 µL SuperScript III recombinant reverse transcriptase, 1 µL RNaseOUT recombinant ribonuclease inhibitors, 2.5 µL 0.1 M DTT, 10 µL 5 x first strand buffer (Invitrogen, Carlsbad, USA) and 11 µL Milli-Q H₂O was added to the first reaction. Reverse transcription was performed at 42°C for 1 hr and 70°C for 10 min. The cDNA was stored at -20°C. Quantitative real-time PCR was performed using the TaqMan Universal PCR Master Mix system and TaqMan Gene Expression Assays for Kidney injury molecule 1 (Kim1) (Mm00506686_m1), Neutrophil gelatinase-associated lipocalin (Ngal) (Mm01324470_m1), Tumour necrosis factor alpha (Tnfa) (Mm00443258_m1), Interleukin 1 beta (Il1b) (Mm00434228_m1), Interleukin 33 (Il33) (Mm00505403_m1), Interleukin 6 (Il6) (Mm00446190_m1), Macrophage inflammatory protein-2 (Mip2) (Mm00436450_m1), Receptor interacting protein kinase 1 (Ripk1) (Mm00436354_m1), Receptor interacting protein kinase 3 (Ripk3) (Mm00444947_m1), Mixed lineage kinase domain-like protein (Mlkl) (Mm01244222_m1), Toll-like receptor 2 (Tlr2) (Mm00442346_m1), Toll-like receptor 4 (Tlt4) (Mm00445273_m1), and Glyceraldehyde 3-phosphate dehydrogenase (Gapdh) (Mm99999915_g1) (Applied Biosystems, Carlsbad, CA), using a 7400Fast Real-Time PCR System (Applied Biosystems).

Descriptive statistics were used to summarise the kinetic changes at different time points after reperfusion. Results were expressed as mean ± SEM unless otherwise indicated. A Kruskal-Wallis test and Dunn multiple comparisons test were used when comparing > 2 groups. Comparisons between 2 groups were performed using a Mann-Whitney U-test. Statistical analyses were performed using GraphPad Prism version 9.3 (GraphPad, San Diego, CA) with p<0.05 considered significant.

C57BL/6J WT mice were subjected to right nephrectomy and clamping of the left renal pedicle for 18 min, and cohorts were analyzed at 7 timepoints post-reperfusion (0, 3, 12, 24, 48, 72 hrs, and 4 weeks). Kidney injury was assessed by measuring kidney function (serum creatinine), tubular injury (semi-quantitative morphological analysis), and mRNA expression of the genes for the kidney injury markers KIM-1 and NGAL. Serum creatinine ( Figure 1A ) and tubular injury ( Figure 1B ; Supplementary Figure 2 ) were elevated at 3 hrs, although these increases did not reach statistical significance and there was no change in expression of Kim1 ( Figure 1C ) or Ngal ( Figure 1D ) at this early timepoint. By 12 hrs, creatinine levels and tubular injury were significantly increased, and expression of both Kim1 and Ngal was massively upregulated. All parameters remained elevated at 24, 48 and 72 hrs before returning to baseline by 4 weeks.

We then examined the expression of several pro-inflammatory genes relevant to necroptosis and/or kidney IRI. TNFα is the best characterized activator of necroptosis (53) and has been described as a necroptosis-associated alarmin (59). Tnfa expression was significantly increased at 12, 48 and 72 hrs, and remained high at 4 weeks ( Figure 2A ). Interleukin-6 (IL-6) is an acute inflammatory cytokine that is upregulated in necrotic cells but not in apoptotic cells (60). Il6 expression was elevated earlier than Tnfa (3 hrs) and returned to baseline levels by 72 hrs ( Figure 2B ). IL-1β and IL-33 are released by necroptotic cells (61, 62). Il1b expression was upregulated at 12, 24 and 48 hrs ( Figure 2C ), whereas Il33 was increased later, from 24 hrs ( Figure 2D ). Signalling via Toll-like receptor 4 (TLR4) can activate necroptosis, whereas TLR2 has not been shown to play a role in necroptotic cell death (53). Both Tlr4 ( Figure 2E ) and Tlr2 ( Figure 2F ) were upregulated relatively late (48 and 72 hrs, respectively), with Tlr2 expression remaining high at 4 weeks. Macrophage inflammatory protein-2 (MIP-2) is a neutrophil chemoattractant/activator that is upregulated in kidney IRI (63). Mip2 expression was significantly increased at all timepoints except 48 hrs, peaking at 12 hrs ( Figure 2G ).

Although elevated expression of the necroptotic pathway genes Ripk1, Ripk3 and Mlkl does not necessarily equate with increased necroptosis in vivo (53), it may indicate an increased propensity for this form of cell death. We therefore measured IRI-associated changes in the expression of these genes in the kidney at different times post-reperfusion. All 3 genes were significantly upregulated at most timepoints, with the highest expression observed at 48 hrs ( Figure 3 ).

Sections were cut from formalin-fixed left kidney tissue of WT mice from each cohort and stained for pMLKL using a rabbit monoclonal antibody [ab196436, EPR9515 (2)] This antibody has previously been used to detect pMLKL in methanol-fixed mouse dermal fibroblast cell lines (64) and in formalin-fixed mouse cecal tissue (65). Reproducible staining was achieved after optimization of the antigen retrieval step based initially on a method described by He et al., 2021 (65). Mlkl-ko mice subjected to IRI were used as a control to validate the antibody; Mlkl-ko kidneys showing substantial morphological injury were negative for pMLKL staining at each timepoint ( Supplementary Figure 3 ). pMLKL was not detected in WT/IRI kidneys at 0 or 3 hrs post-reperfusion ( Figures 4A, B ). At 12 hrs, numerous tubules containing cells displaying dense punctate cytoplasmic staining were observed ( Figure 4C ). Interestingly, the tubules at this timepoint were either completely positive or completely negative. At 24 hrs, the pattern of staining had changed from cytoplasmic to predominantly membrane/intraluminal i.e., on the apical surface of tubular epithelial cells ( Figure 4D ). At 48 hrs, strong staining of proteinaceous casts within the tubules was observed, along with patchy intraluminal staining ( Figure 4E ). At 72 hrs ( Figure 4F ) and 4 weeks ( Figure 4G ), staining was weaker and mainly restricted to casts.

We performed a semi-quantitative analysis by counting the number of pMLKL-positive tubules as a percentage of the total number of tubules. The degree of positivity was highest at 12 hrs and remained significantly elevated at 24 and 48 hrs ( Figure 4H ). The data were further stratified into cytoplasmic and intraluminal staining, likely representing the activation and execution phases of necroptosis, respectively. On this basis, maximum activation of necroptosis was observed at 12 hrs, while the terminal stage of necroptotic cell death was relatively constant at 12, 24 and 48 hrs, and returned to close to baseline by 72 hrs ( Figure 4H ).

Compared to WT mice, Mlkl-ko mice showed reduced kidney injury at 24 hrs post-reperfusion, evidenced by significant decreases in serum creatinine ( Figure 5A ) and tubular injury score ( Figure 5B ). Kidney function and injury were not significantly different at the other timepoints. Expression of Kim1 ( Figure 5C ) and Ngal ( Figure 5D ) were reduced at the earlier timepoint of 12 hrs, although only the former was statistically significant. Expression of the pro-inflammatory cytokine genes Tnfa, Il1b and Mip2 was also significantly lower at 12 hrs in Mlkl-ko mice compared to WT mice, whereas expression of Il6, Il33, Tlr2 and Tlr4, and of the necroptosis pathway genes Ripk1 and Ripk3, was generally similar in Mlkl-ko and WT mice at all timepoints ( Supplementary Figure 1 ). As expected, Mlkl expression and pMLKL staining were undetectable in kidneys from Mlkl-ko mice (data not shown).

To investigate the mode of IRI-associated cell death in a necroptosis-deficient setting, kidney sections from the 24 hr timepoint were stained for cleaved Caspase-3 (cC3) as a marker of apoptosis. While WT kidneys showed a significant but modest increase in apoptotic tubules compared to sham kidneys, Mlkl-ko kidneys showed a major increase ( Figure 6 ), suggesting a greater role for apoptosis in this model when necroptosis is inactivated. The pattern of cC3 staining in Mlkl-ko tubules at 24 hrs was like that of pMLKL staining in WT tubules at 12 hrs ( Figure 4C ) i.e., tubules were either completely positive or negative.