Conditioning attenuates kidney and heart injury in rats following transient suprarenal occlusion of the abdominal aorta

Suprarenal aortic clamping during abdominal aortic aneurysm (AAA) repair results in ischemia-reperfusion injury (IRI) in local (i.e. kidney) and distant (i.e. heart) tissue. To investigate perioperative approaches that mitigate IRI-induced tissue damage, Wistar rats underwent suprarenal aortic clamping either alone or in combination with short cycles of ischemic conditioning before and/or after clamping. Serum analysis revealed significant reduction in key biochemical parameters reflecting decreased tissue damage at systemic level and improved renal function in conditioned groups compared to controls (p < 0.05), which was corroborated by histolopathological evaluation. Importantly, the levels of DNA damage, as reflected by the biomarkers 8-oxo-G, γH2AX and pATM were reduced in conditioned versus non-conditioned cases. In this setting, NADPH oxidase, a source of free radicals, decreased in the myocardium of conditioned cases. Of note, administration of 5-HD and 8-SPT blocking key protective signaling routes abrogated the salutary effect of conditioning. To further understand the non-targeted effect of IRI on the heart, it was noted that serum TGF-β1 levels decreased in conditioned groups, whereas this difference was eliminated after 5-HD and 8-SPT administration. Collectively, conditioning strategies reduced both renal and myocardial injury. Additionally, the present study highlights TGF-β1 as an attractive target for manipulation in this context.

The nonselective adenosine receptor antagonist 8-SPT and the ATP-sensitive potassium channel blocker 5-HD were administered intraperitoneally separately in another series of animals exactly as above, 30 min prior to any procedure. The resulting experimental groups of the study will be henceforth identified as A, B, C, D; A 5-HD , B 5-HD , C 5-HD , D 5-HD and A 8-SPT , B 8-SPT , C 8-SPT , D 8-SPT .
Blood was collected via cardiac puncture in vacutainer tubes (Becton Dickinson, BD), centrifuged at RT for 15 min at 3000 g, and serum stored in cryotubes at −80 °C. After sacrifice, tissues from different organs were quickly removed and divided in two parts: one part was fixed immediately in 10% neutral-buffered solution with 4% formaldehyde for 24 h before being embedded in paraffin, and the other part was immediately stored at −80 °C for further processing.
Histopathological evaluation. The histopathological evaluation of rat kidney and heart was done by two experienced histopathologists. For the quantification of the tissue damage, a scoring system was employed based on previously published criteria [14][15][16][17] . Collective histological scoring ranged from 0 (within normal range) to 5 (severe organ injury).
Previously characterized cases served as positive controls 23 .

Statistical analysis.
The results are presented as mean ± (SEM). Comparisons of numeric variables among groups were performed using ANOVA and Tukey's multiple comparisons test. A p value of less than 0.05 was considered statistically significant.

Improved profile in key biochemical markers in conditioned cases.
To examine the effect of pre-, post-and combined pre-and post-conditioning on tissue integrity, we initially assessed the status of key biochemical parameters. Biochemical analysis in the serum of all groups sacrificed in 24 h and 48 h revealed a significant reduction in the acute phase protein C-Reactive Protein (CRP) in the conditioned groups (namely B, C, D; p < 0.05), indicating lower inflammation (Figs. 1 and 2ai). A significant decrease was also observed in urea and creatinine levels in the conditioned cases (p < 0.05) suggesting a protective effect of perioperative conditioning on kidney function (Fig. 2aii). Similarly, compared to non-conditioned group A, serum CPK, LDH, SGPT and K+ levels were lower in groups B, C and D reflecting reduced cellular lysis and tissue damage systematically in conditioned cases (Fig. 2aiii). Our results indicate that all types of ischemic conditioning lead to decreased inflammation and tissue injury.
conditioning lessens the histopathological lesions in kidney and heart. To further investigate the effect of each perioperative strategy in an organ-specific manner, we proceeded with histopathological examination of the kidney and heart. In non-conditioned cases, renal damage was prominent from the earliest time point (6 h), with loss of brush border, cytoplasmic vacuolization, necrotic cells and sloughing of the latter into the tubular lumen, and the formation of protein casts admixed with necrotic debris (Fig. 2b). Similarly, in the myocardium from non-conditioned rats, contraction band necrosis, a form or irreversible myocardial injury was evident from 6 h, accompanied by cellular vacuolization, focal interstitial edema and hemorrhagic infiltration (Fig. 2b). Histopathological examination of kidney and heart at additional time points further verified the protective effect of conditioning ( Fig. 2b) although the differences in rats sacrificed 10d after ischemia were minimal among the various groups. Overall, our results show that ischemic conditioning decreases the histopathological lesions observed in non-conditioned cases both in kidney and heart corroborating the results from the biochemical analyses.
Conditioning reduces the accumulation of DNA damage in kidney and heart. A critical factor in reperfusion injury following blood flow restoration is the excess production of free radicals 6 . To assess oxidative DNA damage, we examined the status of 8-oxo-G, a major product of oxidative damage reflecting cellular oxidative stress 24 . The levels of 8-oxo-G were significantly reduced in all conditioned versus non-conditioned cases from 24 h and 48 h subgroups (Fig. 3ai, ii). Within this frame, assessment of NADPH oxidase (NOX4), an important enzymatic source of ROS in cardiomyocytes 25 , demonstrated elevated NOX4 levels in group A versus conditioned cases ( Supplementary Fig. 1ai, ii). The latter finding comes in line with the increased levels of 8-oxo-G in unconditioned Group A.
Given that free radicals can in turn cause strand breaks in DNA 26 , we examined the status of γH2AX, a prototypic biomarker that is accumulated upon DSBs 27 , along with the levels of phosphorylated ATM kinase at Serine 1981 (designated as pATM), a protein kinase that is activated upon DSB formation 28 (Fig. 3ai, ii). Conditioning was associated with reduced levels of γH2AX both in kidney and heart in 24 h and 48 h subgroups, further supporting the protective effect of this strategy. Western blot analysis for γH2AX and pATM supported our immunohistochemical findings, verifying the protective effect of perioperative conditioning on DNA integrity ( Fig. 3aiii and Supplementary Fig. 1b). In addition, we observed reduced apoptosis in the conditioned cases in 48 h subgroup that comes in line: a) with the decreased activation of DNA damage response and repair machinery (DDR/R) reflected by the status of γH2AX and pATM (further analyzed in the Discussion) and b) with the reduced histopathological lesions in these cases ( Supplementary Fig. 1c). Additionally, given that activation of DDR/R may also trigger senescence, we utilized a novel hybrid histo-/immuno chemical assay employing GL13, a lipophilic biotin-linked Sudan Black-B (SBB) analogue (commercially available as SetTraGor ® ), which reacts Scientific RepoRtS | (2020) 10:5040 | https://doi.org/10.1038/s41598-020-61268-9 www.nature.com/scientificreports www.nature.com/scientificreports/ with lipofuscin, a non-degradable material that accumulates in senescent cells 23 . GL13 staining did not reveal significant differences in our setting (Supplementary Fig. 1d). Collectively, our results demonstrate the protective effect of ischemic conditioning on DNA integrity by assessing the accumulation of 8-oxo-G as well as two sensitive markers of DSBs both in the kidney, a directly affected organ by abdominal aortic occlusion, and in the heart, distant to the occlusion.
Conditioning decreases the levels of serum TGF-β1. The non-targeted effect of IRI on the heart is indicative of the existence of mediators released from the directly affected site and induce DNA damage at distant sites. We focused our analysis on the effect of conditioning on TGF-β1 serum levels for the following reasons: a) TGF-β1 mRNA levels are elevated following ischemia injury in rat kidney 29 , b) TGF-β1 has a central role in myocardial homeostasis after ischemia and reperfusion 30 , c) TGF-β1 induces ROS production mainly through NOX4, and ROS in turn favors TGF-β1 activity forming a vicious cycle 31 and d) TGF-β1 can induce DSBs in bystander cells in a ROS-dependent manner 32 . As expected, group B, C and to a greater degree group D cases exhibited lower levels of serum TGF-β1 ( Supplementary Fig. 2a), which is in accordance with the decreased status of 8-oxo-G, γH2AX, pATM and NOX4 in kidney and heart of conditioned cases. The diminished nuclear immunostaining of p65, a downstream target of TGF-β1 31,33 , in the myocardium of representative conditioned cases further comes in line with TGF-β1 levels in these cases (Supplementary Fig. 2b). Additionally, prompted by previous findings linking Monocyte Chemoattractant Protein 1 (MCP-1) [alternatively knows as chemokine (C-C) ligand 2 (CCL2)] with DNA damage in out-of-field tissues following irradiation 34 , we examined the status of MCP-1/CCL2 in our setting but found no differences among the conditioned and unconditioned groups ( Supplementary Fig. 2c). Collectively, our findings indicate that TGF-β1 can play a role in myocardial IRI after abdominal aortic occlusion. www.nature.com/scientificreports www.nature.com/scientificreports/ Treatment with 5-HD or 8-SPT abolishes the protective effect of conditioning. Earlier studies have shown that administering the KATP channel blocker 5-HD or the non-selective adenosine inhibitor 8-SPT, eliminates the protective effect of IPre and IPost 35,36 . To decipher the role of mitochondrial KATP channel activity and adenosine signaling in this context, we administered 5-HD or 8-SPT before the application of suprarenal abdominal artery occlusion and reperfusion (Fig. 4a). Treatment with 5-HD and 8-SPT resulted in increased levels of urea, creatinine as well as K +in conditioned cases (Fig. 4b). Within this frame, histopathological evaluation revealed significant tissue damage in kidney and heart in 5-HD and 8-SPT treated rats irrespective of the perioperative strategy followed (Fig. 4c). To this end, there was no statistically significant difference among the different groups (A 8-SPT , B 8-SPT , C 8-SPT , D 8-SPT ) after treatment with 8-SPT with respect to 8-oxo-G and γH2AX status ( Supplementary Fig. 3) in 48 h subgroup, which comes in line with the histopathological evaluation. Similar findings were observed in cases with 5-HD treatment (data not shown). Therefore, administration of 5-HD or 8-SPT abrogates the protective effect of conditioning both in kidney and heart. The examination of serum TGF-β1 levels showed similar levels in the different time points among unconditioned (A 5-HD , and A 8-SPT ) and conditioned (B 5-HD , C 5-HD , D 5-HD, B 8-SPT , C 8-SPT , and D 8-SPT ) groups ( Supplementary Fig. 4) further supporting a potential role of TGF-β1 in IRI context.

Discussion
ΙRΙ is associated with increased morbidity, incidence of myocardial infarction, stroke and acute renal injury 2 . Although, renal and heart IRI has been extensively studied, the underlying molecular events are still unclear. The present work contributes to our understanding of the impact of various conditioning strategies on directly (i.e. kidney) and remotely (i.e. heart) affected organs. Of note, the current study revealed that IPre, IPost and combined (IPre + IPost) conditioning resulted in improved biochemical profile as reflected by improved kidney function and reduced tissue destruction at a systemic level. Of note, consistent to reported elevated CRP levels in acute kidney injury 37 , CRP levels in all conditioned groups were decreased compared to the control group (i.e. group A) indicating an attenuated inflammatory response in the former cases. The histopathological analysis in www.nature.com/scientificreports www.nature.com/scientificreports/ kidney and heart further corroborated our findings demonstrating that all different conditioning procedures conveyed a protective effect. Our findings come in line with a review and meta-analysis by Wever et al. in 2012 38 on 58 experimental animal studies showing that IPre reduced the levels of creatinine, blood urea nitrogen and histological damage compared to untreated animals which was more pronounced in the "second window of protection" (SWOP). SWOP refers to a delayed form of protection appearing after 24 h to 48 h upon IPre 39 .
Previous reports have demonstrated the accumulation of the oxidative DNA damage marker 8-oxo-G in kidney tissues following ischemia reperfusion 40 . Herein, we showed -for the first time-that different conditioning strategies result in reduced 8-oxo-G levels both in kidney and heart. Within this context, we further provided evidence that NOX4 is down-regulated in the myocardium of conditioned cases, which is in line with the role of NOX4 as a significant source of free radicals during myocardial ischemia-reperfusion 41 .
To this point, we showed that conditioning reduces oxidative DNA modifications. Taking into consideration a previous study that demonstrated an elevation of γH2AX and pATM in kidney tissue during renal IRI 42 , one would expect a decreased DDR/R activation following IC. Congruent to our hypothesis, γH2AX and pATM levels were decreased across all conditioned groups in comparison to group A both in kidney and heart. γH2AX and pATM are sensitive and robust biomarkers accumulated upon DSBs, indicating for the first time that IC exerts a DNA-protective effect. Previous evidence suggests that inhibition of Poly(ADP ribose) synthase (PARS) [also known as Poly(ADP ribose) polymerase, PARP], a critical component of DDR/R 28 , protects against myocardial IRI 43 . As such, our data provide additional evidence for the significance of DDR/R in IRI and IC. In addition, based on the findings of this study, apoptosis was also reduced in conditioned cases further highlighting the protective effect of this strategy which is in line with previous studies 44 .
Another interesting finding of our study is that TGF-β1 was reduced in all IC groups compared with the IRI group (i.e. group A). Notably, this difference was more pronounced in group D where a combined conditioning (IPre + IPost) was applied. TGF-β1 is a pleiotropic cytokine synthesized by a variety of cells and www.nature.com/scientificreports www.nature.com/scientificreports/ is implicated in a number of cellular functions 45 . Accumulating evidence has demonstrated that TGF-β1 may be up-regulated following acute ischemia having a salutary effect in kidney and heart tissue after IRI 29,30,[46][47][48] . These observations emphasize the beneficial role of TGF-β1. This may at first glance seem counterintuitive with our findings; yet, although up-regulation of TGF-β1 may have a protective effect to some extent, it might exert an adverse effect on cellular homeostasis -when rising above a certain threshold-by inducing redox imbalance. This can be justified by the fact that TGF-β1 induces ROS production in certain contexts. In particular, TGF-β1 signaling has been implicated in the production of mitochondrial ROS through NOX4 induction 31,49 . Hence, the observed reduced TGF-β1 levels after conditioning in our setting, which was pronounced in the combined preand post-conditioning, may favor the protective effect of this cytokine on tissue integrity. In a similar manner, free radicals have a bivalent nature in IC; excess levels of ROS cause tissue damage, whereas small amounts protect cellular functions 50 . To this end, the IRI biology is highly reminiscent of the complexities in radiation-induced bystander/distant effect biology 34 . Within this frame, a systemic response has been depicted in out-of-field tissues upon irradiation verified by γH2AX foci suggesting DNA damage. A hint for the underlying mechanism comes from the elevated circulating levels of certain cytokines including TGF-β1, which may play a role as mediators of non-targeted effects by promoting oxidative DNA damage at distant sites 34 .
Previous studies have demonstrated that mitochondrial KATP channel activity and adenosine signaling exert a protective effect during IRI 9,35,36,51 . Induction of these molecular routes exert an immunomodulatory and anti-oxidant effect in different models of IRI [52][53][54][55] . Therefore, to gain a further insight in the pathogenesis of tissue damage in our context, we separately abrogated these two signaling pathways. Administration of the nonselective adenosine receptor antagonist 8-SPT or the mitochondrial KATP channel blocker 5-HD abrogated the protective effect of IC. Destruction of renal architecture accompanied by impairment of kidney function along with myocardium damage was clearly evident irrespective of the conditioning strategy. In view of the fact that NOX4 and mitochondrial KATP channel form a feed-forward loop in cardiac biology 56 , our findings pinpoint the involvement of this axis in the pathophysiological processes of IRI. Notably the serum levels of TGF-β1 were similar in all groups irrespective of the perioperative approach after treatment with 8-SPT or 5-HD, suggesting the involvement of both signaling pathways in the release of TGF-β1. Previous studies have linked the adenosine signaling with TGF-β1 status in different contexts 57,58 ; however, further research is needed in order to clarify the relative contribution of mitochondrial KATP channel activity and the different types of adenosine-adenosine receptor in the induction of TGF-β1 in the particular setting.
Collectively, these data demonstrate the adverse effect of suprarenal aortic occlusion at organismal, cellular and molecular level and highlight the beneficial role of different conditioning procedures on both kidney and heart, a directly and remotely affected organ, respectively. In addition, this work indicates that TGF-β1 can be a potential mediator of the remote effects in this context. In light of the recent evidence demonstrating a beneficial systemic immune response and higher overall survival of metastatic breast cancer patients receiving TGFβ blockade during radiotherapy 59 , TGFβ1 becomes an attractive target with promising prognostic and therapeutic applications in IRI context. Hence, knowledge of IRI biology could be applied not only to understand the etiology of tissue damage during IRI but also to overcome current challenges in translating the experimental data into the clinical setting. To this end, the promising results yielded from clinical trials that examine transient perioperative episodes of ischemia as well as the evidence from preclinical studies suggest a protective role of the conditioning strategies 60 and highlight the opportunity for clinical application in the setting of AAA repair.