Annexin-A5 monomer as a membrane repair agent for the treatment of renal ischemia-reperfusion injury

Background: Renal ischemia-reperfusion injury (IRI) is one of the causes of acute kidney injury. Annexin A5 (AnxA5) as a calcium-dependent cell membrane binding protein has shown good protective effects in various organ IRI models. This study explored the therapeutic effect of exogenous AnxA5 monomer protein on renal IRI and its potential mechanism. Methods: In vivo, different doses of AnxA5 were injected intravenously to treat bilateral renal IRI models in SD rats. This model con�rms the protective effect of AnxA5 on the structure and function of the kidneys. In vitro, HK-2, the renal tubular epithelial cells, was subjected to hypoxia for 12 hours, followed by restoration of normal oxygen supply to simulate IRI. In vitro experiments have demonstrated the mechanism of action of AnxA5 by measuring cell activity and permeability. The comparison of the mutant AnxA5 protein M23 and the application of calcium-free culture medium further validated the protective effect of AnxA5 by forming a network structure. Results: In vivo, AnxA5 monomer improved kidney function damage caused by IRI and prevented the deterioration of the pathological structure. In vitro, AnxA5 protected cell viability, reduced cell membrane permeability, and inhibited cell apoptosis induced by IRI treatment. However, AnxA5 lost its protective effect in the calcium-free culture medium and the M23 that lack the ability of forming a two-dimensional network structure could not reduce cell membrane permeability and lost its protective effect on cell viability, indicating that the formation of a two-dimensional network structure of AnxA5 is critical to the repair of the cell membrane. Conclusions: Exogenous AnxA5 monomer could prevent renal IRI by binding to the damaged renal tubular epithelial cell membrane, forming a two-dimensional network structure to maintain cell membrane integrity, and ultimately prevent cell death.


Introduction
Ischemia-reperfusion injury (IRI) refers to a situation where tissues or organs experience further deterioration, or even irreversible damage, when blood ow is restored after a period of ischemia [1].The kidney has a rich blood supply and is one of the organs most likely to cause IRI.In clinical practice, the causes of renal IRI are common in the treatment of renal diseases such as kidney transplantation, nephron sparing surgery and shock [2].Moreover, renal IRI can easily lead to acute kidney injury, one of the most common critical conditions in clinical practice with a high mortality rate [3].Renal IRI can also lead to long-term adverse events, including chronic kidney disease and renal brosis [4].At the pathological level, the most common form of renal IRI is acute tubular necrosis, characterized by the death of renal tubular epithelial cells and dysfunction of one or more tubular segments [5][6].
Currently, the prevention of renal IRI is partly achieved by reducing the ischemic time as much as possible to lower the risk of renal IRI [7].On the other hand, changing the conditions of reperfusion can also affect the occurrence of IRI, which is mainly related to the pressure, temperature, pH value, and electrolyte concentration of the perfusion uid during the reperfusion process [8][9].In terms of the pathogenesis of renal IRI, anti-in ammation treatment plays a therapeutic role in the renal IRI.Some active ingredients extracted from plants, such as ganoderic acid, allicin, astragaloside, and asiaticoside, have been proven to have good anti-in ammation activities and achieve good therapeutic effects in animal models [10][11][12][13].
During IRI, ischemic conditions lead to ATP depletion, subsequently causing the translocation of phosphatidylserine (PS) to the surface of endothelial cells [14].Once exposed to the extracellular environment, PS serves as a binding site for leukocytes and platelets, leading to thrombus formation and an in ammatory state.Endogenous annexin A5 (AnxA5) binds to PS and forms a two-dimensional crystal structure through a calcium-dependent pathway, blocking PS exposure on the surface of endothelial cell membranes.In animal models, AnxA5 has been shown to have antithrombotic effects and prevents the recruitment of monocytes and macrophages to allograft endothelial cells [15].Recently, AnxA5 has being widely studied as an anti-in ammatory agent.It has been also used in various IRI animal models.Studies have shown that AnxA5 homodimers can prevent acute lung injury caused by ischemia-reperfusion in lung transplantation, and likely works through regulating in ammation [16].Furthermore, AnxA5 has also been proven to reduce infarct size and improve cardiac function after myocardial ischemia-reperfusion injury by inhibiting cardiac in ammatory responses [17].In terms of renal IRI treatment, AnxA5 homodimers DA5 have been proven to protect kidney function and inhibit renal in ammation [15].However, whether exogenous AnxA5 monomers have a protective effect on renal IRI and the mechanisms behind their protective effects have not been fully explained.
Current approaches are mainly focused on anti-in ammatory and antioxidant mechanisms.However, the occurrence and development of IRI is a complex process involving multiple mechanisms.Cellular membrane injury, as an intermediate process of IRI, plays a bridging role in the entire IRI development process.Previous studies have found that AnxA5 monomers have the ability to self-assemble into a twodimensional (2D) network structure at the bio-membrane level [18].Experiments have con rmed that by laser irradiation to rupture the membrane, AnxA5 plays a core role in the membrane repair by forming 2D bandages at the edge of the torn membrane, thereby preventing the expansion of the membrane wound and promoting membrane resealing [19].AnxA5 also plays an important role in the repair process of cellular membrane permeability changes caused by strong electromagnetic pulses [20].Based on the inherent characteristics of AnxA5, we hypothesized that AnxA5 might promote the cellular membrane repair by forming 2D network structure on the surface of the damaged cellular membrane, thereby playing a therapeutic role in renal IRI.

Animal model and AnxA5 treatment
Male Sprague Dawley (SD) rat (8 weeks, 280-300 g) were purchased from Shanghai Lab Animal Research Center (Shanghai, China).Establish a bilateral renal ischemia-reperfusion injury model in rats.The bilateral renal pedicles of rats were exposed and clamped for 60 min to induce ischemia [21].Then, the clamps were released.The animals were randomly divided into 6 groups, which consisted of the Control group, Sham group, IRI group, IRI + low AnxA5 (40 μg/kg) group, IRI + medium AnxA5 (200 μg/kg) group, and IRI + high AnxA5 (1000 μg/kg) group.Each group was assigned ve rats.AnxA5 was purchased from Novoprotein (China).The AnxA5 treatment group injected AnxA5 through the tail vein before releasing the vascular clamp.Fresh blood and urine from all rats 1 day before IRI and 1, 3, 7 days after reperfusion was collected (Fig. 1 A).The animals were sacri ced 7 days later.Animals without clamping the renal pedicle were used as controls.The rats during the surgery were placed on a heating mat at 32℃.All surgical procedures are performed in a constant temperature operating room at 28℃.

Assessment of Renal Function
Blood was centrifuged by 20 min of centrifugation at 2000 ´g.Blood urea nitrogen (BUN), creatinine (Scr) and uric acid (UA) were measured with an automated analyzer (Shenzhen, China).Urine protein was detected using the Urinary protein quanti cation kit (Jiancheng, China).

Histopathological Analysis
Rat right kidney specimens were xed in 4% paraformaldehyde for 24 h, graded with alcohol dehydration, and embedded in para n.Para n embedded tissue was cut into 4 μm thickness sections and performed hematoxylin and eosin (HE), periodic acid-silver metheramine (PASM), and Masson staining.Collagen deposition was scored semi-quantitatively by Image-Pro Plus version 6.0 (Medium Cybernetics, Bethesda, MD, USA).

Immunohistochemistry (IHC)
The 4 μm-thick sections of rat kidney tissues were depara nized and rehydrated with xylene and deescalated ethanol.Antigen retrieval was conducted in sodium citrate buffer.To block endogenous peroxidase activity, the slices were incubated in 3% hydrogen peroxide in methanol for 20 min.After being washed with phosphate buffered saline (PBS), the sections were blocked in PBS containing 2% normal goat serum, 5% BSA, and 0.1% triton-X.The slices were then incubated with primary antibodies overnight at 4 ℃, followed by incubation with the secondary antibody (goat anti-rabbit, Abcam, UK) for 1 h at room temperature.DAB Kit was used to develop colors.The slices were counterstained with hematoxylin.The primary antibodies used in this study are listed below, including kidney injury molecule 1 (KIM-1) (NBP1-76701, 1:400, Novus), neutrophil gelatinase-associated lipocalin (NGAL) (ab216462, 1:1000, Abcam).

Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining
Renal tissue sections were xed in xylene, hydrated with a gradient of ethanol, and permeabilized with proteinase K for 30 min.Subsequently, samples were subjected to stain with TUNEL reagent (Roche Switzerland) for 1 h at 37°C.The brown stained cell nuclei were identi ed as TUNEL-positive.

Cell culture and IRI treatment
Human proximal tubular cells (HK2) were purchased from American Type Culture Collection (ATCC, USA).HK2 cells were cultured in Dulbecco's modi ed eagle medium (DMEM, Gibco, USA) mixed with 10% fetal bovine serum (FBS, Gibco), and 1% penicillin-streptomycin (Sigma-Aldrich, USA) in an incubator of 5% CO 2 at 37℃.All culture media should be changed every 2 d.For the IRI treatment, HK2 cells were cultured for 12 h under hypoxic conditions (1% O 2 , 94% N 2 , and 5% CO 2 ) in medium without nutrients (glucose-free, serum-free) to induce hypoxic injury.Then, replace with fresh normal culture medium containing nutrients during reoxygenation, and the plates were moved to a cell incubator (5% CO 2 and 95% air) for 24 h.The control group cells were cultured in complete medium in a conventional incubator (5% CO2 and 95% air), and the frequency of medium replacement was consistent with that of the experimental group cells [22].

Cell viability assay
HK2 cells were inoculated into 96-well plates at a density of 5×10 3 cells/well and cultured for 24 h in DMEM (supplemented with 10% FBS, 1% penicillin-streptomycin).Add AnxA5 according to grouping while changing the culture medium after reoxygenation.On the one hand, different concentrations of AnxA5 (0, 5, 50, 100 μg/ml) were used to treat normal HK2 cells for 24 h to determine the cytotoxicity of AnxA5.On the other hand, different doses of AnxA5 (0, 5, 50, 100 μg/ml) were used to treat HK2 cells after IRI treatment for 24 h to verify the protective effect of AnxA5.To verify whether the mutant M23 of AnxA5 has a protective effect, 50 μg/ml of normal AnxA5 and M23 were respectively used to treat HK2 cells after IRI for 24 h, followed by cell viability testing.A CCK-8 Assay Kit (Dojindo, Japan) was utilized to test cell viability.CCK-8 absorbance was tested at 450 nm in each well on a microplate reader.

Lactate dehydrogenase (LDH) release assay
Membrane permeability was evaluated using the LDH Cytotoxicity Assay Kit (Beyotime, China).HK2 cells were inoculated into 96 well plates and cultured normally for 24 h.After IRI treatment, the cells were treated with different doses of AnxA5 (0, 5, 50, 100 μg/ml) for 24 h.When verifying the effect of Ca 2+ on AnxA5, AnxA5 was used to treat cells under reoxygenation conditions for 6 h, and the supernatant of different treatment groups was taken for LDH detection.When comparing the effects of normal AnxA5 and M23 on LDH release rate, corresponding tests were performed on HK2 cells after 24 h of treatment.
The LDH level in the supernatant was measured according to the instructions of the kit.Then measure the absorbance at 490nm using a microplate reader.

Flow cytometry analysis
Cell apoptosis was assessed by ow cytometry using the Apoptosis and Necrosis Detection Kit with YO-PRO-1 and PI (Beyotime, China) according to the manufacturer's instructions.In short, HK2 cells were reoxygenated for 24 h after hypoxia, during which the experimental group was treated with different concentrations of AnxA5 (0, 5, 50, 100 μg/ml).After reaching the experimental endpoint, cells from different intervention groups were collected and washed twice with PBS.Then, cells were stained with 499 μL binding buffer containing 0.5 μL YO-PRO-1 and 0.5 μL PI under dark and room temperature conditions for 20 min.The apoptotic cells were detected with a FACS ow cytometer (BD, Germany).YO-PRO-1 single positive cells are apoptotic cells, while PI positive cells are necrotic cells.
Further validate the effect of AnxA5 on early cell membrane permeability and its binding form with damaged cells, perform uorescence staining on cells treated with different concentrations of AnxA5 (0, 5, 50, 100 μg/ml) for 2 h.Annexin V-PE (Beyotime, China) was used for detecting the binding of AnxA5 to cell membranes, and cells were stained with 194.5μl binding buffers containing 5 μL Annexin V-PE and 0.5 μL YO-PRO-1.After incubating at room temperature in the dark for 20 min, the FACS ow cytometer was used for detection.

AnxA5 binding assay
The control group cells were cultured normally, while the experimental group HK2 cells were cultured for 24 h after hypoxia and reoxygenation.After washing the cells twice with PBS, incubate cells with 195 μL binding buffers containing 5 μL Annexin V-FITC (Beyotime, China) under dark room temperature conditions for 20 min.After incubation, clean the cells again with PBS and reserve 0.5 mL of PBS in the well plate to prevent the cells from drying out.Observe the sample under an inverted uorescence microscope (ZEISS, Germany).

Statistical Analysis
In vitro experiments were repeated at least three times.Data were expressed as the mean ± SEM.One-way ANOVA was used for multiple comparisons using SPSS version 22.0 (SPSS, Inc., IL, USA).Single groups were compared by an unpaired two-tailed t-test.P < 0.05 was considered signi cant.

Results
3.1 AnxA5 alleviated renal function damage caused by IRI.
As shown in Fig. 1, on the rst day after IRI, the levels of BUN, Cr, and UA were signi cantly elevated in all IRI groups compared to the control and sham surgery groups.The levels of BUN, Cr, and UA in all IRI groups gradually decreased over time.On the 7th day, the BUN, Cr, and UA levels in the IRI group without AnxA5 treatment were still signi cantly higher than those in the AnxA5-treated groups.The BUN, Cr, and UA levels in the AnxA5-treated groups were close to those of the Sham group (Fig. 1B-D).The random midstream urine protein tests also showed that different doses of AnxA5 all improved renal function, with the medium dose group having the most signi cant effect (Fig. 1E).There was no animal death in each group of experimental animals before reaching the observation endpoint.
From H&E staining, it was clearly observed that renal tubules in the IRI group were dilated, with epithelial atrophy and many necrotic epithelial cells in the tubular lumen.The AnxA5-treated groups showed less renal tubular damage compared to the IRI group (Fig. 2A and B).In Masson staining, collagen deposition was clearly observed in the IRI group.The low-dose AnxA5 group showed a low level of collagen deposition, while collagen deposition in the medium and high dose AnxA5 groups were close to the Sham group (Fig. 2A and C).PASM staining basically shows the continuity and integrity of the glomerular basement membrane as well as the collagen extravasation.Part of the glomerular basement membrane was damaged in the IRI group, but this damage was not observed in the medium-and high-dose AnxA5 groups (Fig. 2A).

AnxA5 prevents renal injury by protecting renal tubular epithelial cells.
The results of IHC staining showed that after IRI, the expression levels of early renal injury markers KIM-1 (Fig. 3A) and NGAL (Fig. 3B) increased and concentrated in the renal tubules.After applying AnxA5 to IRI rats, the expression level of early renal injury markers was signi cantly reduced and showed a dosedependent effect (Fig. 3C and D).Indicating that AnxA5 can prevent early renal tubular injury caused by IRI.To further detect the protective role of AnxA5 on renal tubular epithelial cells, HK2 cells were cultured in the presence of AnxA5 with or without IRI treatment.As shown in Fig. 4A, no effect of AnxA5 on cell viability was observed under the non-IRI condition, while a signi cant protective effect of AnxA5 on cell viability was observed under the IRI treatment (Fig. 4B).Expressions of KIM-1 and NGAL, the important indicators of early renal tubular injury, con rmed the protective effect of AnxA5 on renal tubular epithelial cells (Fig. 4C and D).

AnxA5 inhibit cell apoptosis in vivo and vitro.
To further analyze the protective role of AnxA5, the apoptotic cells were measured by TUNEL staining in the tissue samples.About 15% of apoptotic cells were observed in the IRI group, while less apoptotic cells were observed in the AnxA5-treated groups (Fig. 5A and B).The inhibitory role of AnxA5 on cell apoptosis was further con rmed in cultured renal tubular epithelial cells.IRI treatment could signi cantly increase the number of apoptotic cells, while the number of apoptotic cells was decreased in the presence of AnxA5 (Fig. 6A and B).And the early protective effect of AnxA5 was given, which prevented the progression of injury, reduced the proportion of cell necrosis, and changed the outcome of IRI induced cell necrosis (Fig. 6A and C).
3.5 AnxA5 binds to cell membrane to improve cell membrane integrity.
To investigate how AnxA5 protects cell apoptosis, HK2 cells were stained by FITC-labeled AnxA5 with or without IRI treatment.No AnxA5 binding was observed in normally cultured cells, while a punctate aggregation form of AnxA5 positive staining on the cell membrane was observed after IRI treatment (Fig. 7A).Flow cytometry analyses of YO-PRO-1 staining show that the permeability of HK2 cells signi cantly increases after IRI treatment, while AnxA5 reduced cell membrane permeability in a dose-dependent manner (Fig. 7B and C).And it can be found that there is competitive inhibition between AnxA5 and the surface of damaged cell membranes (Fig. 7B and D).LDH exists inside the cell membrane.Normally, LDH is released from the cell when cell membrane permeability is increased.The LDH release assay showed that AnxA5 maintained cell membrane integrity and reduced LDH release caused by IRI treatment (Fig. 7E).

3.6
The function of AnxA5 depends on the formation of 2D network structure.
AnxA5 has been proven to bind to PS on the surface of the cell membrane and form a 2D network structure in the injured cells [26][27].The formation of a 2D network structure relies on Ca 2+ .After reoxygenation, HK2 cells were treated with normal culture medium for 2 hours and then incubated with AnxA5 for 0.5 h in the presence or absence of Ca 2+ .LDH release assay showed that the cell membrane integrity of damaged cells was signi cantly improved by AnxA5 in the presence of Ca 2+ (Fig. 8A).However, no improvement was observed in a calcium-free buffer solution (Fig. 8B).To further con rm the results, M23, a mutant of AnxA5 that lacks the ability to form a 2D network structure [28-29], was tested on HK2 cells after IRI treatment.As shown in Fig. 8C, no protective effect on cell viability was observed in the M23-treated group.LDH release assay con rmed that cell membrane integrity was improved by AnxA5 but not M23 (Fig. 8D).These results suggest that the protective role of AnxA5 depends on the formation of a 2D network structure.

Discussion
In renal IRI, the pathological manifestations in kidney tissue include swelling and necrosis of renal tubular epithelial cells, renal tubular atrophy, formation of a large number of cellular casts in renal tubules, and partial damage to the continuity of the glomerular basement membrane.The pathological damage is related to the large production of reactive oxygen species, release of in ammatory factors, Ca 2+ overload, and disorder of cell energy metabolism in the kidney after reperfusion, ultimately leading to necrosis of renal tubular epithelial cells [30][31][32].Once the structure of the renal unit is destroyed, the metabolic function of the kidney is greatly affected, speci cally manifested as increased levels of urea, creatinine, and uric acid in the blood, accumulation of metabolic waste products, and impaired renal barrier function, leading to increased urinary protein content.In the current study, IRI-induced kidney damage could be alleviated by intravenous injection of the AnxA5 monomer during reperfusion.The BUN, Cr, and UA levels in blood and the protein level in random midstream urine were signi cantly decreased after AnxA5 treatment (Fig. 1).In addition, less pathological damages were observed in the AnxA5-treated groups (Fig. 2), indicating that AnxA5 plays a protective role in the renal IRI.
Previous attention to AnxA5 function has mostly focused on its anti-in ammatory ability [33][34][35][36].The ipping of PS from the inner to the outer cell membrane signi cantly occurs when cells are damaged, which subsequently recruits platelets and leukocytes, leading to thrombus formation and in ammation.By binding AnxA5 to PS, the exposure of PS to platelets and leukocytes is reduced, and the overactivation of in ammation is avoided.This is the major explanation of the protective effect of AnxA5 in the lung, heart and renal IRI that have been reported [15].However, it is known that AnxA5 itself has the characteristic of binding to the damaged cell membrane and forming a 2D network structure to repair the cell membrane and maintain membrane integrity [18].In the process of IRI, the damage to the cell membrane comes from lipid peroxidation.The damaged cell membrane not only expose PS to the in ammatory cells, but also cause an increase in cell membrane permeability, which is the main cause of cell death [37].Clearly, cell membrane damage is central to the damage transmission chain caused by IRI.If IRI-induced cell membrane damage could be prevented, the cascade damage response and cell death could be inhibited.In this study, we found that damage to renal tubular epithelial cells is the primary pathological change during renal IRI (Fig. 2, 3 and 5).In vitro experiments con rmed that HK2 cells were protected by AnxA5 from IRI (Fig. 4 and 6), and the protective effect is likely through direct binding of AnxA5 on cell membrane (Fig. 7).Interestingly, this process did not involve in ammation.KIM-1 and NGAL, as markers of early kidney injury, could be produced by renal tubular epithelial cells and correlate with the degree of injury [38].As observed in this study, AnxA5 could effectively reduce the expression levels of KIM-1 and NGAL in kidney (Fig. 3A and B) and HK2 cells (Fig. 4C and D) after IRI, indicating that AnxA5 works in the early stage of IRI process.
By comparing perivascular cells from wild-type and AnxA5-de cient mice, it was found that wild-type cells could spontaneously and quickly repair membrane rupture.In contrast, cells lacking AnxA5 showed defects in their membrane repair mechanism [39].Previous studies also con rmed that AnxA5 could assemble into 2D network structures at the edge of the damaged cell membrane, prevent the expansion of the membrane wound and promote resealing of the membrane [19].In the current study, AnxA5 reduced the YO-PRO-1 staining in HK2 cells and inhibited the LDH release caused by IRI, proving that AnxA5 could maintain cell membrane integrity and decrease cell membrane permeability (Fig. 7).As we expected, AnxA5 lost its protective effect in a calcium-free medium.In addition, M23, a mutant of AnxA5 which can bind to PS but cannot form a 2D network structure, does not have the protective function on HK2 cells from IRI damage (Fig. 8).These results can verify that the AnxA5 monomer mainly binds to the damaged cell membrane, assemblies a 2D network structure to repair the cell membrane, maintain the membrane integrity, and prevent cell death.
Wever et al. demonstrate that the AnxA5 homodimer DA5 ameliorates renal function after IRI, reduces leukocyte in ux and tubular damage and reduces renal injury marker expression [15].The protective function of DA5 is explained by its anti-in ammation ability.DA5 is thought to shield externalized PS on vascular endothelial cells, thereby reducing leukocyte adherence and transmigration, as well as preventing the formation of the pro-thrombinase complex and diminishing the in ammatory response [15].In the current study, we proved the protective effects of AnxA5 monomer on renal tubular epithelial cells, the direct target cell type in IRI.Although DA5 was claimed to have an increased half-life in the circulation in vivo as well as high a nity for exposed PS in vitro [40], the protective effect was observed on the 3rd day after medication and was not enhanced by prolonging the observation time [15].In addition, about 5 times higher accumulation of AnxA5 in kidney than DA5 was observed in the previous study [15], indicating that in the treatment of renal IRI, AnxA5 might be better than DA5.Further experiments are needed to con rm this prediction.

Conclusions
In summary, the current study demonstrated that exogenous AnxA5 monomer could prevent renal IRI by binding to the damaged renal tubular epithelial cell membrane, forming a 2D network structure to maintain cell membrane integrity, and ultimately prevent cell death.
Annexin A5 inhibits atherogenic and pro-in ammatory effects of lysophosphatidylcholine.

Figures
Figures