INTRODUCTION
Acute kidney injury (AKI) is a common clinical syndrome which results in substantial short and long-term morbidity and mortality.[1,2] Even with a full recovery of renal function after an episode of AKI, there is an increased risk of developing chronic kidney disease (CKD), or exacerbating pre-existing CKD.[1,3] Indeed, there is now great interest in how AKI transitions into CKD.[4]
There are many potential insults that can lead to AKI such as renal ischaemia, sepsis, nephrotoxic drugs and hypovolaemia. While most cases of AKI cannot be anticipated, renal ischaemia leading to AKI can be anticipated in settings such as cardiopulmonary bypass (CPB) surgery or kidney transplantation. However, a major challenge in the AKI field is the lack of specific therapies to prevent anticipated AKI or to intervene in diagnosed AKI. In particular, clinical trials to prevent AKI following CPB surgery have been very disappointing.[5] Although there can be many reasons why any one individual trial fails, it has been proposed that a common problem underlying these failures is that the pre-clinical animal models used to justify these trials do not adequately represent the clinical situation.[6]
Up to 40% of patients undergoing CPB surgery will develop AKI, and this can be predicted by a number of patient-based risk factors such as advanced age, CKD, hypertension, congestive heart failure and type 1 or type 2 diabetes.[5,7] However, almost all preclinical studies of renal ischaemic injury use healthy animals in which a single ischaemic event induces severe AKI; this does not reflect the clinical reality. Thus, for an animal model of ischaemia injury to be relevant to AKI following CPB surgery it should: (i) cause little or no AKI in the absence of patient-based risk factors, and; (ii) induce clearly demonstrable AKI in the presence of patient-based risk factors. Therefore, the aim of this study was to develop an animal model in which renal ischaemic injury induces robust AKI with maladaptive repair in the setting of a patient-based risk factor (established diabetes), but cause only mild AKI with recovery in healthy animals without the risk-factor.
MATERIALS AND METHODS
Animals
Male inbred C57BL/6J mice were sourced from the Monash Animal Research Platform (Clayton, Victoria, Australia). The Monash Medical Centre Animal Ethics Committee gave approval for these studies (MMCB/2018/22 approved on 29 July 2018), and they conformed with the 8th Edition of the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes.
Induction of type 1 diabetes
Type 1 diabetes was induced in 10 to 12-week-old male C57BL6/J mice using the low dose streptozotocin (STZ) regimen. On each of 5 consecutive days, mice were fasted for 6 h and then given an intraperitoneal injection of 55 mg/kg STZ freshly dissolved in sodium citrate buffer (pH 4.5). Blood glucose was measured after a 3 h fast by tail vein sampling (Stat Strip Xpress, Nova Biomedical, Whaltham, MA, USA) on days 7 and 14 after starting the STZ injections. Mice achieving a fasting blood glucose level of > 16 mmol/L on both days 7 and 14 were considered diabetic. Fasting blood glucose levels were measured weekly thereafter and maintained within the target range (22 to 26 mmol/L), with 0.4U of long-acting insulin (Protophane, Novo Nordisk, Baulkham, Australia) given by subcutaneous injection as required. A plasma sample was taken from the tail vein to measure HbA1c ((DCA Vantage Analyzer, Siemens Healthcare Diagnostics, Tarrytown, NY, USA) in the week prior to undergoing IRI surgery.
Pilot study of IRI in diabetic mice
Pilot studies were performed on mice at 8 weeks after STZ-induced type 1 diabetes in which ischaemia times of 12, 13, 14 and 15 min were tested. An ischaemic time of 12 or 13 min induced moderate to severe AKI at 24 h in diabetic mice, whereas 14 or 15 min of ischaemia induced severe AKI with animals unwell at 24 h and unlikely to survive beyond this time. For long-term studies, an ischaemia time of 12 min was selected which was well tolerated with no deaths due to AKI. This technique is detailed below.
IRI surgical procedure
Mice were anaesthetized with ketamine and xylazine and body temperature was maintained at 37 °C using a rectal thermometer linked to heating blanket (Homeothermic monitor system, Harvard Apparatus, Holliston, MA, USA). After a midline abdominal incision, the right and then the left renal pedicle were clamped using non-traumatic vascular clamps (B-1A microvascular clamp for arteries (00397), S&T Microsurgical Instruments, Birmingham, AL, USA) and the abdomen closed with a temporary suture to minimise fluid loss. Clamps were then removed from the right and then the left renal pedicle giving each kidney exactly 12 min of clamping (ischaemia), with reperfusion of each kidney monitored visually. Next, the abdomen was sutured in 2 layers using 5-0 suture, and analgesia provided by 2 or 3 drops of bupivacaine onto the sutures and a subcutaneous injection of 0.05 mg/kg buprenorphine. Fluid management was provided by subcutaneous saline injection (1 mL spread over several sites) at the time of surgery and at 4 h after surgery. In addition, mice were given mouse chow mashed together with liquid Ensure (Abbot Laboratories, Chicago, IL, USA) in a petri dish on the cage floor for the first 2 days after surgery, and cages were kept on a heating pad overnight after surgery. Sham operated mice underwent the same procedure, but without clamping of the renal pedicles. All mice survived both the STZ-induced diabetes and IRI surgical procedures. Plasma creatinine and blood urea nitrogen levels were measured using a Dupont ARL Analyser by the Department of Biochemistry, Monash Health.
Experimental design
At 8 weeks after starting STZ injections, groups of 6 mice diabetic mice underwent renal IRI surgery with a 12 min ischaemic time and were killed on days 1, 7 or 28 (Figure 1A). A sham operated diabetic group was used as the control for changes due to the diabetic state. Groups of 6 age-matched, non-diabetic mice underwent the same IRI surgical procedure and were killed on days 1, 7 or 28. A non-diabetic sham group was used as the normal control.
;)
Figure 1:: Renal function and kidney damage following renal ischaemia/reperfusion injury (IRI). (A) Schematic diagram showing the experimental design for streptozotocin-induced diabetes followed by renal IRI and animals killed at different times after. Groups of 6 diabetic (red) or non-diabetic mice (blue) were killed on day 1, 7 or 28 after IRI surgery or on day 28 after sham surgery. (B) Plasma creatinine levels. (C) Blood urea nitrogen (BUN) levels. (D) The percentage of damaged tubules in the inner cortex and outer medulla. (E) RT-PCR for Kim1 mRNA levels. (F) Fasting blood glucose levels from before (time 0) and weekly after STZ injection in diabetic groups that underwent sham or IRI surgery. (G) Plasma HbA1c levels for the non-diabetic day 28 sham controls and all of the diabetic groups. # P < 0.05, ### P < 0.01 vs. non-diabetic sham; * P < 0.05, ** P < 0.01, *** P < 0.001 vs. diabetic day 28 sham.
Renal histology
Kidney tissue was fixed in freshly prepared Carnoy's solution (60% methanol, 30% chloroform, 10% glacial acetic acid), washed in graded ethanol and embedded in paraffin. Tissue sections (2 μm) were stained with Periodic acid-Schiff (PAS) reagent and Harris' haematoxylin. Tubular damage was scored in the inner cortex and outer medulla under high power (x400). Tubules were deemed either normal or damaged, with damage characterized as any of the following: loss of the brush border, nuclear loss, sloughing of cells into the lumen, and atrophy. Scoring was performed on blinded slides.
Immunohistochemistry
Sections (4 μm) of Carnoy-fixed kidney tissue were deparaffinised and rehydrated through graded ethanol. After blocking with 5% bovine serum albumin (BSA) at room temperature for 40 min, sections were incubated overnight at 4 °C with the following primary antibodies diluted with 5% mouse serum and 5% rabbit or sheep serum (depending on the species of the primary antibody): rat anti-mouse F4/80 (MCA497GA at 1 in 100 dilution; Bio-Rad, Hercules, CA, USA), rabbit anti-α-SMA (EPR5368, 1 in 8000 dilution; Abcam, Melbourne, Australia), or rabbit anti-collagen type I (E8F4L, 1 in 400 dilution; Cell Signaling Technology, Danvers, MA, USA). After washing (x3) in PBS, endogenous peroxidase was blocked in 3% H2O2 in distilled water for 10 min, followed by washing (x3) in PBS and then an avidin/biotin blocking step performed before incubation with secondary antibodies; biotinylated rabbit anti-rat IgG (BA-400, 1 in 500 dilution; Thermo Fisher, Waltham, MA, USA) or biotinylated goat anti-rabbit IgG (656140, 1 in 500 dilution; Thermo Fisher). Sections were then developed with diaminobenzidine/H2O2 to produce a brown reaction product, dehydrated through graded ethanol into histolene, and mounted with Eukitt mounting medium (Sigma Aldrich, Melbourne, Australia).
Real time RT-PCR
Approximately 100 mg of snap-frozen kidney tissue was added to TRIzol solution (Thermo Fisher), homogenised in a Bullet Blender (Next Advance, Troy, NY, USA), and total RNA was extracted using the RiboPure RNA purification kit (AM1924, Ambion, Texas, USA) according to the manufacturer's protocol. Total RNA (7 μg) per sample was reversed transcribed into cDNA using the Superscript III First-Strand Synthesis kit (Thermo Fisher) according to the manufacturer's protocol. Real time PCR used a StepOne Real-Time PCR System (Applied Biosystems, MA, USA) using a total reaction volume of 20 μL containing cDNA, Taqman Gene Expression Master Mix (Applied Biosystems, Waltham, MA, USA), and Taqman gene expression assays for the gene of interest and the 18S internal control (Applied Biosystems). Thermal cycling conditions were: 37 °C for 10 min, 95 °C for 5 min, followed by 50 cycles of 95 °C for 15 s, 60 °C for 20 s and 68 °C for 20 s. The relative amount of cDNA was calculated using comparative threshold cycle (ΔΔCt) method.
Statistics
GraphPad Prism (GraphPad Prism 9.0 software, San Diego, CA, USA) was used for the statistical analysis. Data are shown as mean ± one standard deviation, and were analysed using one-way ANOVA with Tukey's multiple comparison test.
RESULTS
AKI in non-diabetic mice following renal IRI
A 12 min ischaemic time for bilateral IRI was selected based upon pilot studies (described in section Pilot study of IRI in diabetic mice). Non-diabetic mice undergoing 12 min of bilateral IRI showed AKI on day 1 based on a 2.3-fold increase in plasma creatinine levels, which returned to sham control levels on days 7 and 28 (Figure 1B). A similar pattern was seen with blood urea nitrogen levels (Figure 1C). Kidney histology showed tubular damage ranging from loss of brush border to tubular necrosis with sloughing of cells into the lumen, which affected 25% of tubules in the inner cortex and outer medulla on day 1 after IRI (Figures 1D and 2A), with effective repair of tubular damage evident on days 7 and 28 compared to the sham control (Figures 1D, 2C, 2E, 2G). We also assessed renal repair by the anti-apoptotic protein, survivin, which is expressed by proximal tubular cells in the normal kidney,[8] and plays a functional role in tubular repair after renal IRI.[9] Prominent staining of proximal tubular cells on days 7 and 28 after IRI was comparable to that of the sham control, indicating effective renal repair (Figure 3A, C, E). Consistent with the histological picture, there was a marked increase in mRNA levels of the tubular damage marker, Kim1/Havcr1, on day 1 after IRI, and this returned to sham control levels on days 7 and 28 (Figure 1E).
Figure 2:: Renal histology following renal ischaemia/reperfusion injury (IRI). (A-H) Kidney sections were stained with Periodic acid-Schiff reagent and Harris' haematoxylin. Images for non-diabetic mice are shown for (A) day 1 IRI, (C) day 7 IRI, (E) day 28 IRI, and (G) day 28 sham control. Images for diabetic mice are shown for (B) day 1 IRI, (D) day 7 IRI, (F) day 28 IRI, and (H) day 28 sham control. Bar = 200 μm.
Figure 3:: Tubular expression of survivin following renal ischaemia/reperfusion injury (IRI). Immunostaining for survivin is shown in non-diabetic mice on day 7 IRI (A), day 28 IRI (C) and day 28 sham control (E), and for diabetic mice on day 7 IRI (B), day 28 IRI (D) and day 28 sham control (F). Bar = 700 μm.
AKI in diabetic mice following IRI
Induction of diabetes by 5 low dose STZ injections was confirmed by weekly fasting blood glucose levels and by measuring HbA1c levels in the week before surgery (Figure 1F, G). Sham operated diabetic mice showed normal plasma creatinine levels on day 28. By contrast, diabetic mice undergoing IRI showed a 6.1-fold increase in plasma creatinine on day 1, which improved substantially on days 7 and 28 and was not different to the day 28 diabetic sham control (Figure 1B). A similar pattern was seen with blood urea nitrogen levels, although levels remained elevated on day 7 after IRI and returned to diabetic sham levels on day 28 after IRI (Figure 1C). Kidney histology showed extensive tubular necrosis in the outer medulla and inner cortex on day 1 after IRI, with extensive loss of tubular cell nuclei and cells sloughing into the tubular lumen to give strongly stained cellular casts. A cellular infiltrate was also evident around many of the severely damaged tubules. At least 80% of tubules in the outer cortex and inner medulla were damaged (Figure 1D and Figure 2B). A majority of tubules in this area still showed damage on day 7, with some cellular casts still evident and many tubules showing dilation with only a partial recovery of normal cellularity. An interstitial cellular infiltrate was also evident (Figure 1D and Figure 2D). Significant tubular damage remained evident on day 28: necrotic cells had been removed, leaving a mixed histologic picture with areas of tightly packed, healthy tubules surrounded by areas with irregularly packed tubules showing atrophy and a thickened tubular basement membrane, with substantial interstitial cell infiltration (Figure 1D and Figure 2F). Immunostaining showed a marked loss of survivin expression in proximal tubular cells on day 7 after IRI, and while there was a substantial recovery of survivin expression on day 28, some areas with damaged tubular cells still lacked survivin expression when compared to the sham control (Figure 3B, D, F). Diabetic IRI mice showed strongly elevated Kim1 mRNA levels on day 1, which remained significantly elevated on day 7, but returned to diabetic sham control levels on day 28 (Figure 1F).
Chronic inflammation in diabetic mice following IRI
In non-diabetic mice, kidney levels of the pro-inflammatory cytokine Tnf were significantly elevated on days 1 and 7 after IRI (Figure 4A). This was associated with a significant increase in the mRNA levels of the macrophage marker, CD68, on day 1 which then returned to sham control levels on days 7 and 28 (Figure 4B). Immunostaining for F4/80+ cells showed a pattern consistent with resident renal macrophages in non-diabetic mice at day 28 after IRI, which was comparable to that seen in the non-diabetic sham control group (Figure 4C, E).
Figure 4:: Inflammation following renal ischaemia/reperfusion injury (IRI). Groups of 6 diabetic (red) or non-diabetic mice (blue) were killed on day 1, 7 or 28 after IRI surgery or on day 28 after sham surgery. RT-PCR for (A) Tnf, and (B) Cd68, mRNA levels. ### P < 0.001 vs. non-diabetic sham; * P < 0.05, *** P < 0.001 vs. diabetic day 28 sham. Immunostaining for F4/80+ macrophages is shown in non-diabetic mice on day 28 IRI (C) and day 28 sham control (E), and for diabetic mice on day 28 IRI (D) and day 28 sham control (F). Bar = 200 μm.
Diabetic mice showed a greater increase in Tnf mRNA compared to that seen in non-diabetic mice on days 1, 7 and 28 after IRI (Figure 4A). In addition, Tnf mRNA levels in diabetic mice were significantly elevated on days 1, 7 and 28 after IRI compared to day 28 diabetic sham controls (Figure 4A). Diabetic sham mice showed a significant increase in CD68 mRNA levels compared to non-diabetic sham controls (Figure 4B), consistent with the early macrophage infiltrate into the diabetic kidney.[10] Diabetic mice exhibited a substantial increase in CD68 mRNA levels on days 1 and 7 after IRI compared to diabetic sham controls and to non-diabetic IRI groups (Figure 4B). Immunostaining showed patchy areas of interstitial F4/80+ macrophage infiltration on day 28 after IRI in diabetic mice, which contrasted with the mild and diffuse increase in F4/80+ macrophages seen in the diabetic sham control (Figure 4D, F).
Interstitial fibrosis in diabetic mice following IRI
There was no clear evidence of renal fibrosis in the non-diabetic groups on days 7 or 28 after IRI, with no change in Tgfb1 or Col1a1 mRNA levels (Figure 5A, C), although an increase in α-SMA mRNA was seen on day 28 after IRI (Figure 5B). Immunostaining showed a similar pattern of fine staining of collagen I in some tubular basement membranes, as well as staining in vessel walls, in non-diabetic day 28 IRI and sham controls (Figure 6A, C), with no significant difference in the area of interstitial collagen I staining (Figure 5D).
Figure 5:: Renal fibrosis following renal ischaemia/reperfusion injury (IRI). Groups of 6 diabetic (red) or non-diabetic mice (blue) were killed on day 1, 7 or 28 after IRI surgery or on day 28 after sham surgery. RT-PCR for (A) Tgfb1, (B) α-SMA, and (C) Col1a1, mRNA levels. (D) Area of interstitial collagen I staining. ## P < 0.01 vs. non-diabetic sham; ** P < 0.01, *** P < 0.001 vs. diabetic day 28 sham.
Figure 6:: Collagen I deposition following renal ischaemia/reperfusion injury (IRI). Immunostaining for collagen I is shown in non-diabetic mice on day 28 IRI (A) and day 28 sham control (C), and for diabetic mice on day 28 IRI (B) and day 28 sham control (D). Bar = 200 μm.
Diabetic sham controls showed no evidence of renal fibrosis compared to the non-diabetic sham controls based on Tgfb1, α-SMA and Col1a1 mRNA levels and the area of interstitial collagen I staining (Figure 5A–D). However, diabetic mice showed a substantial fibrotic response on days 7 and 28 after IRI. There was a peak of Tgfb1 and Col1a1 mRNA levels on day 7, whereas α-SMA mRNA levels showed a progressive increase over days 7 and 28 (Figures 5A–C). Patchy areas with dilated tubules showed a significant increase in interstitial collagen I staining on days 7 and 28 after IRI in diabetic mice (Figure 5D, 6B).
DISCUSSION
We have established a new animal model of renal ischaemic injury. A brief period (12 min) of renal ischaemia in otherwise healthy mice induced a transient AKI which underwent effective repair by day 7 without the development of chronic inflammation or fibrosis on day 28 after IRI. By contrast, the same ischaemic insult in mice with established diabetes induced a more severe AKI with a failure of the repair process leading to chronic tubular damage (atrophy and thickened tubular basement membrane), inflammation and progressive fibrosis up to day 28 despite a return to normal plasma creatinine levels. The diabetic state by itself did not cause significant inflammation or fibrosis in the day 28 sham control group; however, the additional insult of IRI markedly enhanced the severity of tubulointerstitial damage in the diabetic kidney.
A small number of studies have investigated renal ischaemia in diabetic animals. A standard unilateral renal ischaemic time of 30 min in rats at one to two weeks after inducing diabetes with STZ results in a virtually complete loss of function in the ischaemic kidney, such that animals rely upon the unaffected kidney for survival.[11,12] Similarly, bilateral renal ischaemia for 20 or 22 min in young Akita mice with type 1 diabetes results in very severe AKI mice with significant animal death within the first 4 days after reperfusion.[13] While these studies were designed to investigate specific mechanistic questions, we consider that the diabetic IRI model described in this paper has a number of advantages over these previous approaches. First, the use of a short ischaemic period enables studies to contrast the acute and longer-term effects in diabetic and non-diabetic mice without animal mortality. Second, waiting for a chronic diabetic state (8 weeks post STZ) before performing IRI provides time for completion of the proximal tubule hypertrophy response to the acute onset of diabetes,[14] and enables careful matching of the diabetic state between animals based on weekly fasting blood glucose levels and elevated HbA1c levels prior to IRI surgery.
We selected diabetes as the patient-based risk factor for this ischaemic model for several reasons. First, diabetes can be induced by low dose STZ in most strains of mice and rats, making this applicable for use in inbred or outbred animals and in genetically modified animals. Second, diabetes is relatively straightforward to monitor via fasting blood glucose and HbA1c. Third, diabetes is a common patient-based risk factor in those undergoing CPB surgery,[15] and diabetic patients who have an AKI episode have a greater risk of developing CKD.[16] Other patient-based risk factors have been investigated in IRI models, such as aging. For example, 12-month old mice show enhanced AKI in response to renal IRI compared to younger mice, despite a baseline of normal renal function.[17] In a different strategy, rats with subclinical CKD induced by administration of adenine or aristolochic acid show increased sensitivity to a low dose of the nephrotoxic drug cisplatin, with enhanced AKI and exacerbation of tubulointerstitial damage.[18]
An important consideration for establishing our model is that the ischaemic time in IRI studies is often quite variable between laboratories. Factors that affect ischaemic time are: (i) bilateral renal artery clamping needs to be of shorter duration to ensure animal survival, while unilateral clamping can be of long duration to induce robust fibrosis and the animal survives via the contralateral kidney; (ii) body temperature such that higher temperature induces more severe AKI; (iii) some anaesthetics can induce resistance or susceptibility to renal ischaemia, and; (iv) post-operative fluid treatment reduces the severity of AKI.[19,20] Therefore, it is recommended than anyone seeking to establish this model of diabetic IRI in their own lab perform pilot studies in diabetic and non-diabetic mice focusing on small changes in the ischaemic time to achieve the desired features of this model.
A limitation of the study is that we have not assessed the long-term outcomes beyond day 28. While diabetic C57BL6/J mice develop relatively mild diabetic kidney disease at 18 weeks after STZ-induced diabetes,[8] the addition of IRI may dramatically accelerate the progression of DKD beyond the day 28 time point examined herein. It will be interested to determine whether the tubulointerstitial damage caused by IRI will affect the longer-term development of diabetic glomerulosclerosis, podocyte loss and the severity of albuminuria.
In conclusion, we have characterised a model of diabetic IRI which has clear relevance to the development of AKI following CPB surgery since healthy animals showed transient AKI with effective recovery, whereas mice with a patient-based risk factor – diabetes – showed a more severe AKI with delayed recovery and transition to CKD. This model can be used to investigate the mechanisms by which diabetes facilitates ischaemia-induced AKI and in testing new therapies to prevent AKI in diabetic patients following CPB surgery.
Author contributions
Conceptualization, Nikolic-Paterson DJ, Ma FY, Grynberg K. and Mulley WR; methodology, Ma FY, Ozols E, Tesch GH, and Grynberg K; formal analysis, Ma FY, Tesch GH and Nikolic-Paterson DJ; resources, Nikolic-Paterson DJ; data curation, Ma FY, Ozols E, Tesch GH and Nikolic-Paterson DJ; writing—original draft preparation, Nikolic-Paterson DJ, Ma FY, Tesch GH, Grynberg K and Mulley WR; writing—review and editing, Nikolic-Paterson DJ and Ma FY; supervision, Nikolic-Paterson DJ and Mulley WR; funding acquisition, Ma FY and Nikolic-Paterson DJ. All authors have read and agreed to the published version of the manuscript.
Ethics approval and consent to participate
The animal study protocol was approved by the Animal Ethics Committee of the Monash Medical Centre (protocol MMCB/2018/22, approved on 29 July 2018).
Financial support and sponsorship
This study was funded by the National Health and Medical Research Council of Australia (APP1156982).
Conflict of interest statement
David J Nikolic-Paterson is an Editorial Board Member of the journal. The article was subject to the journal's standard procedures, with peer review handled independently of this member and his research group. No other authors have conflicts of interest.
REFERENCES
1. Kellum JA, Romagnani P, Ashuntantang G, Ronco C, Zarbock A, Anders HJ. Acute kidney injury. Nat Rev Dis Primers. 2021; 15;7(1): 52.
2. Lewington AJ, Cerdá J, Mehta RL. Raising awareness of acute kidney injury: a global perspective of a silent killer. Kidney Int. 2013; 84(3): 457–467.
3. Chawla LS, Eggers PW, Star RA, Kimmel PL. Acute kidney injury and chronic kidney disease as interconnected syndromes. N Engl J Med. 2014; 371(1): 58–66.
4. Yu SM, Bonventre JV. Acute kidney injury and maladaptive tubular repair leading to renal fibrosis. Curr Opin Nephrol Hypertens. 2020; 29(3): 310–318.
5. O'Neal JB, Shaw AD, Billings FT 4th. Acute kidney injury following cardiac surgery: current understanding and future directions. Crit Care. 2016 4;20(1): 187.
6. de Caestecker M, Humphreys BD, Liu KD, et al. Bridging Translation by Improving Preclinical Study Design in AKI. J Am Soc Nephrol. 2015; 26(12): 2905–2916.
7. Hertzberg D, Sartipy U, Holzmann MJ. Type 1 and type 2 diabetes mellitus and risk of acute kidney injury after coronary artery bypass grafting. Am Heart J. 2015; 170(5): 895–902.
8. Lechler P, Wu X, Bernhardt W, et al. The tumor gene survivin is highly expressed in adult renal tubular cells: implications for a pathophysiological role in the kidney. Am J Pathol. 2007; 171(5): 1483–1498.
9. Chen J, Chen JK, Conway EM, Harris RC. Survivin mediates renal proximal tubule recovery from AKI. J Am Soc Nephrol. 2013; 24(12): 2023–2033.
10. Chow FY, Nikolic-Paterson DJ, Ozols E, Atkins RC, Rollin BJ, Tesch GH. Monocyte chemoattractant protein-1 promotes the development of diabetic renal injury in streptozotocin-treated mice. Kidney Int. 2006; 69(1): 73–80.
11. Abu-Saleh N, Awad H, Khamaisi M, et al. Nephroprotective effects of TVP1022, a non-MAO inhibitor S-isomer of rasagiline, in an experimental model of diabetic renal ischemic injury. Am J Physiol Renal Physiol. 2014; 306(1): F24–F33.
12. Melin J, Hellberg O, Larsson E, Zezina L, Fellström BC. Protective effect of insulin on ischemic renal injury in diabetes mellitus. Kidney Int. 2002; 61(4): 1383–1392.
13. Peng J, Li X, Zhang D, et al. Hyperglycemia, p53, and mitochondrial pathway of apoptosis are involved in the susceptibility of diabetic models to ischemic acute kidney injury. Kidney Int. 2015; 87(1): 137–150.
14. Seyer-Hansen K, Hansen J, Gundersen HJ. Renal hypertrophy in experimental diabetes. A morphometric study. Diabetologia. 1980; 18(6): 501–505.
15. Raza S, Blackstone EH, Sabik JF 3rd. The diabetes epidemic and its effect on cardiac surgery practice. J Thorac Cardiovasc Surg. 2015; 150(4): 783–784.
16. Jiang G, Luk AO, Tam CHT, et al. Clinical Predictors and Long-term Impact of Acute Kidney Injury on Progression of Diabetic Kidney Disease in Chinese Patients with Type 2 Diabetes. Diabetes. 2022; 71(3): 520–529.
17. Ferenbach DA, Nkejabega NC, McKay J, et al. The induction of macrophage hemeoxygenase-1 is protective during acute kidney injury in aging mice. Kidney Int. 2011; 79(9): 966–976.
18. Succar L, Pianta TJ, Davidson T, Pickering JW, Endre ZH. Subclinical chronic kidney disease modifies the diagnosis of experimental acute kidney injury. Kidney Int. 2017; 92(3): 680–692.
19. Kennedy SE, Erlich JH. Murine renal ischaemia-reperfusion injury. Nephrology (Carlton). 2008; 13(5): 390–396.
20. N. Le Clef, A. Verhulst, P. C. D'Haese, B. A. Vervaet: Unilateral Renal Ischemia-Reperfusion as a Robust Model for Acute to Chronic Kidney Injury in Mice. PLoS One. 2016; 11: e0152153.
