Aberrant activation of the complement system in renal grafts is mediated by cold storage

Abstract

Aberrant complement activation leads to tissue damage during kidney transplantation, and it is recognized as an important target for therapeutic intervention. However, it is not clear whether cold storage (CS) triggers the complement pathway in transplanted kidneys. The goal of the present study was to determine the impact of CS on complement activation in renal transplants. Male Lewis and Fischer rats were used, and donor rat kidneys were exposed to 4 h or 18 h of CS followed by transplantation (CS + transplant). To study CS-induced effects, a group with no CS was included in which the kidney was removed and transplanted back to the same rat [autotransplantation (ATx)]. Complement proteins (C3 and C5b-9) were evaluated with Western blot analysis (reducing and nonreducing conditions) and immunostaining. Western blot analysis of renal extracts or serum indicated that the levels of C3 and C5b-9 increased after CS + transplant compared with ATx. Quite strikingly, intracellular C3 was profoundly elevated within renal tubules after CS + transplant but was absent in sham or ATx groups, which showed only extratubular C3. Similarly, C5b-9 immunofluorescence staining of renal sections showed an increase in C5b-9 deposits in kidneys after CS + transplant. Real-time PCR (SYBR green) showed increased expression of CD11b and CD11c, components of complement receptors 3 and 4, respectively, as well as inflammatory markers such as TNF-α. In addition, recombinant TNF-α significantly increased C3 levels in renal cells. Collectively, these results demonstrate that CS mediates aberrant activation of the complement system in renal grafts following transplantation.

NEW & NOTEWORTHY This study highlights cold storage-mediated aberrant activation of complement components in renal allografts following transplantation. Specifically, the results demonstrate, for the first time, that cold storage functions in exacerbation of C5b-9, a terminal cytolytic membrane attack complex, in renal grafts following transplantation. In addition, the results indicated that cold storage induces local C3 biogenesis in renal proximal cells/tubules and that TNF-α promotes C3 biogenesis and activation in renal proximal tubular cells.

INTRODUCTION

Optimizing long-term graft function after transplantation continues to be a challenge, especially for people who receive kidneys from deceased donors (1, 2). Although the 5-yr median graft and patient survival rates are higher when kidneys come from living donors, kidneys from deceased donors are more readily available (2) and constitute up to 70% of total transplants (1). These kidneys are flushed with and stored in cold storage (CS) solutions, typically Viaspan or University of Wisconsin (UW) solution (3–7), to increase their viability until a recipient is identified. These donor grafts may be particularly vulnerable to damage during CS because of prior injuries inflicted by the vigorous inflammatory response to brain death. Decreasing the injury associated with CS could help organs be used more effectively, reduce immune activation, delay graft failure, improve long-term graft survival, and lower the mortality rates for patients with end-stage kidney disease, thereby improving health. The main goal of this study was to characterize CS-related immune mechanisms of renal damage, specifically complement activation, to decrease the incidence of transplant-associated graft failure.

The pathophysiology of organ damage and dysfunction after transplantation is linked to an abnormal immune response. An increasing body of evidence suggests that CS-related injuries to the donor kidney trigger an innate and/or adaptive immune response within the kidney after transplantation (8–13). There is evidence that complement is activated during kidney transplant (14). Complement proteins are an integral part of the innate immune surveillance system (15–17), which participates in eliminating cellular debris and foreign intruders from host cells. The complement system consists of ∼25 proteins that are synthesized in the liver and transported in the blood, mostly as inactive precursors. The complement system comprises three mechanistic pathways: classical, lectin, and alternative. All of these pathways converge at C3 convertase, which cleaves the complement component C3 into C3a and C3b. C3a is an anaphylatoxin and C3b participates in forming multimeric complexes with other complement molecules, yielding C5 convertase. C5 convertase cleaves the complement component C5 to C5a (an anaphylatoxin) and C5b. C5b binds to complement components C6–C9, and, eventually, the complement cascade leads to the formation of the terminal, cytolytic activation product C5b-9, referred to as the membrane attack complex (MAC) (15–18). C3a and C5a are effector molecules that bind to C3a and C5a receptors (C3aR and C5a receptors, respectively), transmitting proinflammatory signals that induce biological processes such as vasodilation and cytokine and chemokine release (19, 20). It is known that warm ischemia-reperfusion injury triggers the complement system. During organ transplantation (including the kidney), complement activation can lead to tissue damage (14). Therefore, it is recognized as an important target for therapeutic intervention (21–24). However, the mechanisms that trigger complement activation during solid-organ transplantation are poorly understood. Our hypothesis is that cold ischemia during CS is a critical mediator of complement activation in renal transplants. This is the first study, if any, that has studied the effects of cold ischemia on renal transplant outcome in the context of complement activation.

Our recent work with rat transplant models indicated that CS exacerbates renal damage and dysfunction in transplanted kidneys (compared with no CS) (25–27). We further demonstrated that proteasome function is decreased, protein homeostasis is altered, and albumin is modified within kidneys after CS plus transplantation (25, 28). Here, we show that CS augments complement activation and promotes inflammation in renal transplants.

METHODS

Animals

Male Lewis or Fischer rats (200–250 g) were used as transplant donors and male Lewis rats were used as recipients. All animal protocols were approved by the Institutional Animal Care and Use Committee of the University of Arkansas for Medical Sciences, and all animal experiments were performed in compliance with institutional and National Institutes of Health guidelines.

Rat Surgery

Surgeries were performed as previously described (25–28).

CS plus transplant surgery.

For donor surgeries, rats were anesthetized with isoflurane and the left and right kidneys were removed and flushed with and stored in CS solution (UW solution, also referred to as Viaspan) at 4°C for 4 or 18 h. The right kidney was saved immediately as the CS group, and this group was compared with untreated control (to evaluate the effects due to CS) or CS plus transplant surgery (CS + transplant) groups (to compare how transplant affects the CS condition). For recipient surgeries, rats were anesthetized with isoflurane, the native left kidney was removed, and the donor left kidney (exposed to CS) was transplanted by end-to-end anastomosis. The surgical ischemia time was <45 min. The ureter was anastomosed end to end over a 5-mm PE-50 polyethylene stent. For the 1-day postsurgery time point, the native right kidney was removed at the time of transplant so that renal function was entirely dependent on the transplanted left kidney. For the 9-day postsurgery time point, the native right kidney was removed on day 7 posttransplant to increase survival. Postoperatively, animals were given 0.9% (wt/vol) NaCl (saline solution) subcutaneously in the dorsal region and placed on a heating pad to recover from anesthesia; they were given buprenorphine (2 mg/kg sc) for pain. After 1 or 9 days of reperfusion, the transplanted left kidney and blood were collected under anesthesia and saved as the CS + transplant group (n = 6). The survival rate at 1 or 9 days after surgery was >95% for the CS + transplant groups.

ATx surgery.

ATx surgery was included in these experiments so that the effect of CS could be isolated from the effect of transplant surgery alone. ATx (n = 6) was performed as described above in CS plus transplant surgery except that the left kidney was removed, flushed with saline, and immediately transplanted back into the same rat without CS exposure; right nephrectomy followed immediately afterward. After 1 day, the transplanted kidney was harvested under anesthesia. This group served as a transplant control for the CS + transplant group.

Sham surgery.

Rats underwent the same procedure for right nephrectomy but without renal transplantation (sham operation); the right kidney was saved as a control kidney (n = 6). The right kidney was saved immediately as the untreated control kidney. The left kidney and blood were harvested 1 or 9 days later and saved as the sham group (n = 6).

For all groups, kidneys were immediately flash frozen for genetic/biochemistry assays and Western blots.

Sample Collection

The kidneys and blood were collected under anesthesia 1 or 9 days after surgery, and rats were euthanized by exsanguination as described in the approved animal used protocol. Kidneys were immediately flash frozen and saved at −80°C until further use. For serum collection, blood was allowed to clot on ice for 1 h. Blood was then centrifuged (5,000 g) at 4°C for 10 min to remove the clot, and the serum (supernatant) was aliquoted and saved at −80°C until further use.

Cell Culture and Treatment

Normal rat kidney proximal tubular cells [NRK-52E (NRK), no. CRL-1571, American Type Culture Collection] and human kidney proximal tubular cells (HK-2, no. CRL-2190, American Type Culture Collection) were maintained in warm growth medium (DMEM plus 5% FCS and 1% penicillin-streptavidin, 5% CO₂).

CS plus rewarming treatment.

NRK cells at 70% confluence or HK-2 cells at 60% confluence were exposed to CS solution (UW solution, also referred to as Viaspan) as previously reported (25). Briefly, cells were washed two times with cold PBS (4°C) and incubated with cold UW solution (4°C) for 18 h (CS condition). Rewarming/reperfusion was initiated by washing the cells three times with cold PBS and then adding cold growth medium (4°C); finally, cells were incubated in growth medium at 37°C for 6 h.

Cytokine treatment.

HK-2 cells at 60% confluence were treated with warm growth medium containing interferon (IFN)-γ (10 ng/mL, R&D Systems) or TNF-α (10 ng/mL, R&D) at 37°C for 24 h.

SDS-PAGE and Western Blot Analysis

Renal extracts from whole kidney homogenates were prepared with radioimmunoprecipitation assay (RIPA) lysis buffer (Pierce) containing 1 mM PMSF, 1.2 mM Na₃VO₄, 2.5 mM NaF, 1 mM DTT (Sigma-Aldrich), and protease inhibitor cocktail (Pierce). After lysis, the extracts were centrifuged (16,000 g for 20 min at 4°C), and the supernatant was saved as the renal extract. Protein concentrations were determined with the BCA Protein Assay kit (Pierce). Renal extracts (20 µg cells and 30 µg tissues) were separated by SDS-PAGE and transferred to a PVDF membrane. For serum analysis, 2 μL of rat serum were boiled in sample loading buffer, separated by SDS-PAGE, and transferred to a PVDF membrane. After transfer, the membrane was incubated with 1× RedAlert Western Blot Stain (Millipore) for 10 min at room temperature to visualize the transferred protein bands. The membrane was quickly washed with double-distilled H₂O, and an image was taken using Fluorchem 8900. Membranes were then incubated with antibodies to C3 (1:1,000, ab181147, Abcam), C5b-9 (1:1,000, HM3033-IA, Hycult Biotech), β-actin (loading control for 1 day posttransplant kidney samples, 1:1,000, A5441, Sigma-Aldrich), or GAPDH (loading control for 9 day posttransplant kidney samples, 1:1,000, no. 37985, Signalway Antibody). Probed membranes were washed three times, incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (1:30,000, Seracare KPL), and assayed for enhanced chemiluminescence (Thermo Fisher Scientific). Densitometry was performed with AlphaEase FC software (Alpha Innotech). For all SDS-PAGE Western blots of kidney or renal cell extracts, the densitometry ratio of C3 or C5b-9 bands to β-actin (1 day) or GAPDH (9 day) was considered for statistical evaluation. For all SDS-PAGE Western blots of serum samples, a densitometry ratio of C3 or C5b-9 bands to the corresponding whole lane density of RedAlert stain was considered for statistical evaluation. Densitometry evaluation on scanned membranes was performed using AlphaEase FC software.

Native Gel Western Blot Analysis

Renal extracts from whole kidney homogenates were prepared with 0.9% digitonin lysis buffer as previously described (25). The homogenate was mixed with lysis buffer [50 mM Tris·HCl (pH 7.5), 250 mM sucrose, 1 mM EDTA, 5 mM MgCl₂, 0.9% digitonin, 1.2 mM sodium Na₃VO₄, 2.5 mM NaF, and Halt Proteasome Inhibitor Cocktail, no. 78430, Thermo Fisher Scientific] at a 1:1 ratio and then incubated on ice for 20 min with occasional mixing. The lysate was centrifuged (16,000 g for 20 min at 4°C), and the supernatant was collected and saved as the digitonin extract. Protein concentrations were determined with Coomassie Plus Protein Assay Reagent (Pierce). Renal extracts (20 µg) were resolved with a bis-Tris (4%–12%) gel and transferred to a PVDF membrane. After transfer, the membrane was incubated with 1× RedAlert Western Blot Stain (Millipore Sigma) for 10 min at room temperature. The membrane was quickly washed with double-distilled H₂O to visualize protein bands, and images were taken using Fluorchem 8900. Western blot analysis was performed with antibodies for C3 (1:1,000, Abcam) and C5b-9 (1:1,000, Hycult Biotech). Chemiluminescence-based detection and densitometry were performed as described above in SDS-PAGE and Western Blot Analysis. For all native gel Western blot analyses (renal or serum), a densitometry ratio of C3 or C5b-9 to the corresponding whole lane density of RedAlert stain was considered for statistical evaluation.

LC-MS/MS and Bioinformatics Identification of Proteins

The SDS-PAGE gel lane for each sample (n = 3) was cut between ∼55 and 90 kDa, and excised slices were subjected to in-gel trypsin digestion and processed further as previously described (28). Briefly, gel slices were destained in 50% methanol (Thermo Fisher Scientific) and 100 mM ammonium bicarbonate (Sigma), followed by a reduction in 10 mM Tris 2-carboxyethyl]phosphine (Pierce) and alkylation in 50 mM iodoacetamide (Sigma). Gel slices were then dehydrated in acetonitrile (Thermo Fisher Scientific), mixed with 100 ng porcine sequencing-grade modified trypsin (Promega) in 100 mM ammonium bicarbonate (Sigma), and incubated at 37°C for 12–16 h.

Peptide products were acidified in 0.1% formic acid (Pierce). Tryptic peptides were separated with reversed-phase Jupiter Proteo resin (Phenomenex) on a 150 × 0.075-mm column with a nanoAcquity UPLC system (Waters). Peptides were eluted over a 30-min gradient from a 97:3 to 67:33 ratio of buffer A to buffer B (buffer A: 0.1% formic acid and 0.5% acetonitrile; buffer B: 0.1% formic acid and 99.9% acetonitrile). Eluted peptides were ionized by electrospray (2.25 kV), followed by MS/MS analysis with higher-energy collisional dissociation on an Orbitrap Fusion Tribrid mass spectrometer (Thermo Fisher Scientific) in top-speed data-dependent mode.

MS data were acquired with the Fourier Transform Mass Spectrometry analyzer in profile mode at a resolution of 240,000 over a range of 375–1,500 m/z. Following higher-energy collisional dissociation activation, MS/MS data were acquired with the ion trap analyzer in centroid mode and normal mass range with precursor mass-dependent normalized collision energy between 28.0 and 31.0. Proteins were identified through a database search with Mascot (Matrix Science, v. 2.5.1) against the UniprotKB database restricted to Rattus norvegicus (36,170 entries) with a fixed modification of carbamidomethyl on C, variable modification of oxidation on M, a parent ion tolerance of 3 ppm, and a fragment ion tolerance of 0.5 Da. Scaffold (Proteome Software) was used to verify MS/MS-based peptide and protein identifications. Peptide identifications were accepted if they could be established with <1.0% false discovery (Scaffold Local FDR algorithm). Protein identifications were accepted if they could be established with <1.0% false discovery and contained at least two identified peptides. Protein probabilities were assigned by the Protein Prophet algorithm (29). Total spectral counts were exported from Scaffold and transformed to log2-normalized spectral abundance factor values (30) for analysis by ANOVA and Student’s t test.

Immunohistochemical Analysis of Tissue Sections

Two cross sections (4–5-µm thickness) from each paraffin block of kidneys were mounted on a glass slide (Thermo Fisher Scientific) and deparaffinized with xylene and a series of graded ethanol washes. Sections were further processed as follows.

For immunohistochemical analysis, antigens were retrieved by heating sections in 10 mM sodium citrate buffer (pH 6.0) for 20 min. Endogenous peroxidase was quenched by incubating the sections with Peroxidase Suppressor (Thermo Fisher Scientific) for 15 min at room temperature, and sections were blocked with nonserum protein block (Dako) for 20 min at room temperature. Primary antibody (anti-C3, Abcam) was diluted 1:500 in antibody diluent solution (1% BSA and 0.5% nonfat dry milk in Tris-buffered saline) and incubated overnight at 4°C. Immunoreactivity was detected with Envision+ System-HRP (Dako). Counterstaining was performed with Mayer’s hematoxylin (Electron Microscopy Science), and bluing was carried out by dipping in 0.125% ammonia blue solution. Finally, slide-mounted sections were dehydrated, covered with Cytoseal-60 (Electron Microscopy Science), and mounted with a coverslip.

For immunofluorescence analysis, antigens were retrieved by heating sections at 98°C in 10 mM sodium citrate buffer (pH 6.0) for 20 min. Slides were blocked with blocking solution (3% BSA and 0.5% nonfat dry milk in Tris-buffered saline) for 20 min at room temperature and then incubated overnight at 4°C with primary antibody (anti-C5b-9, Hycult Biotech) diluted 1:150 in antibody diluent solution (1% BSA and 0.5% nonfat dry milk in Tris-buffered saline). Immunoreactivity was detected with secondary antibody conjugated with Alexa Fluor 594 (Invitrogen). Counterstaining was performed with DAPI nuclear stain (Invitrogen).

All images were taken on a Nikon Eclipse E800 microscope with Nikon Elements software. C3 staining was semiquantitatively evaluated based on the percentage of positive tubules in 10 high-power (×400) fields from the cortex and medulla with the following scores: 0, null/negative; 1, <10% positivity; 2, 10%–50% positivity; or 3, >50% positivity. Fluorescence intensity of C5b-9 staining was evaluated using ImageJ software (National Institutes of Health).

Quantitative Real-Time PCR

Total RNA from cells or tissue was isolated using an RNeasy kit (Qiagen), and RT-PCR was carried out with Superscript III (Invitrogen) reverse transcribed mRNA and a SYBR green PCR Kit (Qiagen). The PCR involved 45 cycles, and the conditions were as follows: 95°C for 5 s and 60°C for 10 s. Amplification of the target genes was normalized to amplification of TATA box-binding protein and to the levels of an appropriate control using the ΔΔC_t method (where C_t is threshold cycle) (31). The specific primer sequences (RealTimePrimers.com) used were as follows: C3, forward 5′- GTACTTGGGAGACGTGGATG-3′ and reverse 5′- ATGAGGGTGTTCTTGTTGGA-3′; chemokine (C-C motif) ligand 2 (Ccl2), forward 5′- TTGTCACCAAGCTCAAGAGA-3′ and reverse 5′- GGTTGTGGAAAAGAGAGTGG-3′; CD11b, forward 5′- AGCACCATCTGGGACATAAA-3′ and reverse 5′- CTTCACAGGCAACTCCAACT-3′; CD11c, forward 5′- AAATACAAGCCACCAACCAA-3′ and reverse 5′- GAAGTTCTGGTTCTGCCTGA-3′; TNF-α, forward 5′- CCCATTACTCTGACCCCTTT-3′ and reverse 5′- TGAGCATCGTAGTTGTTGGA-3′; and IL-1β, forward 5′- AGAGTGTGGATCCCAAACAA-3′ and reverse 5′- AGTCAACTATGTCCCGACCA-3′; and TATA box-binding protein, forward 5′- CGATAACCCAGAAAGTCGAA-3′ and reverse 5′- AGATGGGAATTCCAGGAGTC-3′.

Statistical Analysis

Results are presented as means ± SE (GraphPad Prism software). Data were analyzed with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons, and an unpaired Student's t test was used when comparing differences between the means of two groups (control vs. CS) at a 95% level of confidence. Differences with P < 0.05 were considered statistically significant. Control kidneys were compared with CS kidneys because both groups were harvested from healthy rats. Sham kidneys served as controls for both transplant models (ATx and CS + transplant) because all underwent a nephrectomy (removal of the right kidney).

RESULTS

CS + Transplant Increases the Level of Complement Components

We have previously reported that Coomassie-stained SDS-PAGE gels of renal extracts from the 18-h CS + transplant group (syngeneic rats) at 1 day postsurgery showed profound increases in higher-molecular-weight proteins (range: ∼55–90 kDa) compared with the Sham and ATx groups (1 day postsurgery) (28). Proteins within this region (∼55–90 kDa) were trypsin digested and analyzed with quantitative MS (28). Here, we evaluated the spectral counts for complement components from the same analysis and identified an increase in complement components C3, C4, and C9 after CS + transplant (fold changes of ∼3: C3, 8: C4, and 2: C9 compared with ATx; Fig. 1).

Figure 1.

Complement proteins increase in kidneys after cold storage plus transplantation. The following three experimental groups were considered: sham (right nephrectomy), 18-h cold storage followed by transplantation (CS + Tx), and auto-transplantation (ATx, transplantation with no cold storage). In all groups, surgical procedures were followed by 1 day of reperfusion; sham rats were used as controls. A − C: proteins from renal extracts (30 µg) prepared from kidney homogenates were resolved with SDS-PAGE and visualized with Coomassie staining. Proteins within the 55- to 65-kDa region of the gel were trypsin digested and evaluated with LC-MS/MS, followed by spectral counting [normalized spectral abundance factor (NSAF)]. C3 (A), C4 (B), and C9 (C) values are expressed as means ± SE (bars, n = 3). Differences between the group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. *P < 0.05, CS + Tx vs. sham; $P < 0.05, CS + Tx vs. ATx.

CS + Transplant Increases the Level of Complement Component C3

To explore the impact of CS on complement C3 in renal grafts, renal extracts from syngeneic (Lewis donor to Lewis recipient) and allogeneic (Fischer donor to Lewis recipient) transplant models were evaluated for C3 levels within kidneys (1 day postsurgery). Sham and ATx kidneys were used as controls. SDS-PAGE and Western blot analyses showed increased C3 in kidneys after ATx, and this effect was further enhanced after 18-h CS + transplant (syngeneic and allogeneic; Fig. 2A, left, downstream products of C3, C3a, iC3b, and C3dg, as indicated). Because C3 was induced, we sought to evaluate the native forms of C3 in renal grafts. Total protein (nonreducing conditions) was extracted from renal homogenates of sham, ATx, and 18-h CS + transplant groups and evaluated using native gel Western blots (see methods). Two distinct bands of ∼400 and 500 kDa were observed in sham kidneys (Fig. 2B). Consistent with SDS-PAGE Western blot data, nonreducing (native) Western blots revealed a modest increase in C3 after ATx (∼400-kDa band), but CS + transplant significantly increased C3 levels at ∼110-, 400-, and 500-kDa bands within kidneys (syngeneic and allogeneic groups) 1 day postsurgery (Fig. 2B). As a targeted verification, we assessed the deposition of complement C3 in kidneys using anti-C3 antibody. Immunohistochemical staining for C3 on paraffin-embedded renal sections and found the expected increase in C3 in ATx kidneys compared with basal levels in sham kidneys. C3 appeared to localize outside the renal cells or within the glomerulus (Fig. 2C, ATx; arrow indicates extratubular C3 localization). Quite strikingly, intracellular C3 levels were profoundly elevated within renal tubules (glomeruli, proximal tubules, loops of Henle, distal tubules, and collecting ducts) after CS + transplant (Fig. 2C, CS + transplant; arrow indicates intracellular C3 localization).

Figure 2.

Complement component C3 increases in renal allografts after cold storage plus transplantation. The following four experimental groups were considered at 1 day postsurgery: sham, auto-transplantation (ATx), cold storage + transplantation (CS + Tx or CS/Tx) syngeneic (syn; Lewis to Lewis), and cold storage + transplantation (allo; Fischer to Lewis). A: proteins from RIPA renal extracts (30 µg) prepared from kidney homogenates were subjected to SDS-PAGE and Western blot analysis of C3 after sham, ATx, or CS + Tx (syngeneic or allogeneic). β-Actin was used as a loading control. A representative blot from three independent experiments is shown. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. B: native Western blot of C3 in digitonin extracts of kidney homogenates from sham, ATx, and CS + Tx (syngeneic and allogeneic) groups. Twenty micrograms of protein per well were loaded in a bis-Tris gel under nonreducing conditions followed by Western blot with C3 antibody. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. P< 0.05 was considered significant. C: kidney tissues (paraffin-embedded section) from sham, ATx, and CS + Tx (syngeneic) groups were subjected to immunohistochemistry with C3 antibody. Brown staining indicates reactivity to C3. Arrows indicate C3 localization (extratubular in ATx and intratubular in CS + Tx). a.u., arbitrary units.

Because 18-h CS increased C3 activation in renal transplants, we evaluated the effects of CS (4 or 18 h) alone on C3 levels in rat kidneys. Surprisingly, SDS-PAGE and Western blot analysis showed reduced C3 levels after 18-h CS (Fig. 3A). However, nonreducing Western blots did not show changes in native forms of C3 after CS (Fig. 3B).

Figure 3.

Cold storage (CS) decreases complement component C3 in kidneys. The following three experimental groups were considered: untreated control (Con), 4-h CS, and 18-h CS. A: proteins from RIPA renal extracts (30 µg) prepared from kidney homogenates were subjected to SDS-PAGE and Western blot analysis of C3 after control, 4-h CS, or 18-h CS. β-Actin was used as a loading control. A representative blot from three independent experiments is shown. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. B: native Western blot of C3 in digitonin extracts of kidney homogenates from control, 4-h CS, or 18-h CS groups. Fifteen micrograms of protein per well were loaded in a bis-Tris gel under nonreducing conditions followed by Western blot with C3 antibody. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. a.u., arbitrary units.

We then sought to determine C3 protein levels in rat serum. Western blots showed that C3 levels in serum from all groups (sham, ATx, and CS + transplant) remained unchanged 1 day after surgery (Fig. 4A, reducing; Fig. 4B, nonreducing). On nonreducing Western blots, C3-reactive ∼1,236- and 800-kDa bands appeared to increase in the CS + transplant group (both syngeneic and allogeneic), but this was not statistically significant (Fig. 4B).

Figure 4.

Complement component C3 is unaltered in rat serum after cold storage plus transplantation (CS + Tx or CS/Tx). The following four experimental groups were considered at 1 day postsurgery as in Fig. 2. sham (right nephrectomy), auto-transplantation (Atx; transplantation with no cold storage), and CS/Tx [syngeneic (syn) Lewis to Lewis or allogeneic (allo) Fischer to Lewis]. A: serum (2 μL) from each rat was subjected to SDS-PAGE and Western blot analysis of C3 after sham, ATx, or CS + Tx (syngeneic or allogeneic) treatment. A representative blot from three independent experiments is shown. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. B: native Western blot of C3 in rat serum from sham, ATx, and CS/Tx (syngeneic and allogeneic) groups. Two microliters of serum per well were loaded in a bis-Tris gel under nonreducing conditions followed by Western blot with C3 antibody. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown-Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. a.u., arbitrary units.

CS + Transplant Exacerbates Complement Activation

The C3b fragment of C3 participates in the downstream convertase-mediated complement cascade to form the terminal activation product C5b-9, a multimeric enzyme complex termed MAC (18), which is cytolytic (Fig. 5A). To determine if CS + transplant induced MAC formation, we performed Western blot analysis (reducing and nonreducing) of renal extracts using C5b-9 antibody; we found comparable basal C5b-9-reactive bands in sham kidneys and ATx renal grafts at 1 day postsurgery (Fig. 5B). However, there was a dramatic increase in C5b-9-reactive bands in renal grafts exposed to 18-h CS (syngeneic and allogeneic) at 1 day postsurgery (Fig. 5B). Because C5b-9 is cytolytic, its targeted deposition in tissues results in cell death (15–17). We assessed C5b-9 deposits in kidneys using C5b-9 antibody. First, kidney extracts from sham, ATx, and CS + transplant groups (syngeneic and allogeneic) were evaluated under nonreducing conditions. Native gel Western blot showed basal reactivity with C5b-9 antibody (∼400–1,236 kDa) in renal extracts of the sham group 1 day postsurgery (Fig. 5C), whereas no increase in C5b-9 was seen in the ATx group (Fig. 5C). However, there was a profound increase in native C5b-9 in CS + transplant groups (syngeneic and allogeneic), with four distinct C5b-9-reactive bands ranging from 800 to 2,000 kDa (Fig. 5C). Interestingly, there were no differences in C5b-9 levels between syngeneic and allogeneic groups at 1 day postsurgery. As anticipated, untreated control kidneys showed basal reactivity (weak) to C5b-9 antibody in reducing or nonreducing gel Western blots, and the CS (4 or 18 h) condition did not show a change in C5b-9 reactivity within kidneys compared with the untreated control group (see Supplemental Fig. S1; all Supplemental material is available at https://doi.org/10.6084/m9.figshare.14388581.v1). As a targeted verification, we assessed the deposition of complement C5b-9 in kidneys using anti-C5b-9 antibody. Immunofluorescence staining of renal sections (paraffin embedded) showed increased C5b-9 deposition after CS + transplant (syngeneic) compared with basal levels in sham kidneys (Fig. 5D; red fluorescence indicates C5b-9). In addition, we determined if C5b-9 levels increased in rat serum at 1 day postsurgery. SDS-PAGE Western blot detected several reactive bands with C5b-9 antibody in rat serum (reducing condition) after 1 day postsurgery (Fig. 6A; arrows indicate C5b-9-reactive bands). Surprisingly, no prominent changes were observed in C5b-9 reactivity in rat serum from any transplant groups [ATx, CS + transplant (syngeneic), and CS + transplant (allogeneic)] compared with sham at day 1 postsurgery (Fig. 6A). Because the C5b-9 complex is cytolytic, we sought to evaluate native levels of circulating C5b-9 in rat serum from all four groups. Native gel Western blot using C5b-9 antibody showed multiple reactive bands for C5b-9 in serum from the sham group, and the band size ranged from 242 to 2,000 kDa (Fig. 6B). Interestingly, the native Western blots for the ATx group showed a decrease in intensity for bands of ∼242–300 kDa in serum (Fig. 6B). However, nonreducing Western blot showed a more intense band in between ∼1,048 and 1,236 kDa and a lower intensity band between ∼242 and 300 kDa after CS + transplant (syngeneic and allogeneic). Collectively, these data suggest that the complement system is activated after CS + transplant and that C5b-9, the terminal cytolytic activation product (MAC), is deposited in kidneys and circulates in rat serum.

Figure 5.

Complement terminal activation product C5b-9 increases in renal allografts after cold storage plus transplantation (CS + Tx or CS/Tx). A: schematic of complement pathway activation downstream of C3 convertase leading to formation of the terminal activation product C5b-9 [membrane attack complex (MAC)]. B and C: the following four experimental groups were considered at 1 day postsurgery as in Fig. 2. sham (right nephrectomy), auto-transplantation (Atx; transplantation with no cold storage), and CS + Tx [syngeneic (syn) Lewis to Lewis or allogeneic (allo) Fischer to Lewis]. B: proteins from RIPA renal extracts (30 µg) prepared from kidney homogenates were subjected to SDS-PAGE and Western blot analysis of C5b-9 after sham, ATx, or CS + Tx (syngeneic or allogeneic) treatment. β-Actin was used as a loading control. A representative blot from three independent experiments is shown. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. C: native Western blot of C5b-9 in digitonin extracts of kidney homogenates from sham, ATx, and CS + Tx (syngeneic and allogeneic) groups. Twenty micrograms of protein per well were loaded in a bis-Tris gel under nonreducing conditions followed by Western blot with C5b-9 antibody. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. D: kidney tissues (paraffin-embedded section) from sham and CS + Tx (syngeneic) groups were subjected to immunofluorescence staining with C5b-9 antibody. Red fluorescence indicates reactivity to C5b-9. DAPI staining from the same region of C5b-9 is shown. a.u., arbitrary units.

Figure 6.

Serum levels of C5b-9 are unchanged in rat serum after cold storage plus transplantation (CS + Tx or CS/Tx). Sham (right nephrectomy), autotransplantation (Atx; transplantation with no cold storage), and CS + Tx [syngeneic (syn) Lewis to Lewis or allogeneic (allo) Fischer to Lewis] groups were considered. A: serum (2 µL) from rats was resolved with SDS-PAGE, and proteins were transferred to a PVDF membrane and analyzed by Western blot with C5b-9 antibody (bands detected with chemiluminescence). A representative blot from three independent experiments is shown. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. B: native Western blot of C5b-9 in rat serum from sham, ATx, and CS + Tx (syngeneic and allogeneic) groups. Two microliters of serum per well were loaded in a bis-Tris gel under nonreducing conditions followed by Western blot with C5b-9 antibody. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown-Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. a.u., arbitrary units.

Complement Activation Persists in Renal Allografts at Day 9 After CS + Transplant

Renal extracts and serum from allogeneic transplant groups with 4- or 18-h CS were evaluated for C3 and C5b-9 levels at day 9 postsurgery; the sham group was used as a control. Western blot analyses using C3 or C5b-9 antibodies were performed on renal extracts or serum under reducing or nonreducing conditions. Interestingly, both C3 and C5b-9 levels remained elevated in kidneys at day 9 postsurgery (reducing and nonreducing conditions; Fig. 7, A–D); likewise, the levels of these proteins remained elevated in the serum at day 9 (Fig. 8, A–D).

Figure 7.

C3 and C5b-9 proteins remain elevated in renal allografts after cold storage plus transplantation (CS/Tx) at 9 days postsurgery. The following three experimental groups were considered: sham (right nephrectomy), 4-h CS/Tx (Fischer to Lewis), and 18-h CS/Tx (Fischer to Lewis). In all groups, surgical procedures were followed by 7 days of reperfusion with the right native kidney intact; the right native kidney was removed at day 7, and organs were harvested at day 9 postsurgery. Sham rats were used as controls. Proteins from RIPA renal extracts (30 µg) prepared from kidney homogenates were subjected to SDS-PAGE and Western blot analysis of C3 (A) and C5b-9 (B) after sham, 4-h CS/Tx, or 18-h CS/Tx. A representative blot from three independent experiments is shown. GAPDH was used as a loading control. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. Native Western blot of C3 (C) and C5b-9 (D) in digitonin extracts of kidney homogenates from sham, 4-h CS/Tx, or 18-h CS/Tx groups. Twenty micrograms of protein per well were loaded in a bis-Tris gel under nonreducing conditions followed by Western blot with C3 and C5b-9 antibodies. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. a.u., arbitrary units.

Figure 8.

Cold storage plus transplantation (CS/Tx) activates complement in rat serum at 9 days postsurgery. The following three experimental groups were considered as in Fig. 6. sham (right nephrectomy), 4-h CS/Tx (Fischer to Lewis), and 18-h CS/Tx (Fischer to Lewis). Serum proteins (2 µL) were resolved with SDS-PAGE and analyzed by Western blot analysis for C3 (A) and C5b-9 (B) after sham, 4-h CS/Tx, or 18-h CS/Tx treatment. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. A representative blot from three independent experiments is shown. Native Western blot of C3 (C) and C5b-9 (D). Digitonin extracts (2 µL serum protein) from sham, 4-h CS/Tx, or 18-h CS/Tx groups were loaded in a bis-Tris gel under nonreducing conditions followed by Western blot with C3 and C5b-9 antibodies. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown-Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. a.u., arbitrary units.

CS + Transplant Induces C3 and Inflammatory Genes Within Kidneys

Most circulating complement proteins, including C3, are synthesized by hepatocytes, but smaller amounts are produced by other sources, including kidneys (32) and macrophages (33, 34). Thus, we assayed the induction of the C3 gene within kidneys during CS + transplant. Quantitative real-time PCR (SYBR green) showed a significant increase in C3 mRNA in rat renal extracts after CS + transplant (Fig. 9A).

Figure 9.

Cold storage (CS) plus transplantation (CS + Tx) induces complement-related inflammatory genes. The following three experimental groups were considered: sham (Sh; right nephrectomy), 18-h CS, and 18-h CS + Tx (Fischer to Lewis). Real-time PCR (SYBR green) was performed with rat renal mRNA extracts to assay expression of C3 (A), chemokine (C-C motif) ligand 2 (CCL2 or MCP-1 (B), CD11b (C), and CD11c (D). Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. *P < 0.05, sham vs. CS + Tx and CS vs. CS + Tx groups.

During the complement cascade, cleaved fragments of complement components, including C3a, induce phagocyte chemotaxis and promote inflammation (35). In fact, CS + transplant increased CD68⁺ macrophage infiltration within rat kidneys (28). In this study, we evaluated whether complement activation triggered by CS contributed to CD68⁺ macrophage infiltration (28) and/or inflammation in transplanted kidneys. Quantitative real-time PCR showed an induction of inflammatory markers, namely Ccl2, a chemoattractant for monocytes (Fig. 9B), and CD11b/CD11c α-integrins (also known as components of complement receptors 3 and 4, expressed by leukocytes and phagocytes, including macrophages) (Fig. 9, C and D) after CS + transplant. These results suggest that C3 and inflammatory gene expression related to phagocytes/macrophages is induced in transplanted kidneys exposed to CS.

Renal Proximal Tubular Cells Produce Basal C3, Which Increases After CS + Rewarming

While the liver is the primary source of C3 in the circulation, extrahepatic synthesis of C3 is observed in various cell types, including renal proximal tubule cells (36–47). In the present study, we exploited the in vitro system to gain mechanistic insights into C3 biogenesis in renal proximal tubular cells, complementing our in vivo transplant models (see methods). NRK cells and human kidney proximal tubular cells (HK-2) were exposed to CS (18 h) followed by rewarming to simulate transplantation. After the treatments, NRK cell extracts were evaluated for C3 gene expression. Quantitative real-time PCR showed induction of C3 following CS + rewarming (Fig. 10A). RIPA extracts of NRK or HK-2 cells following CS or CS + rewarming treatments were evaluated for C3 using Western blots and C3 antibody. Western blots showed a basal level of C3a (∼98 kDa) and C3 cleavage products (∼35 kDa) in NRK cells (Fig. 10B, control lane; C3-reactive bands indicated by arrow) and HK-2 cells (Fig. 10C, control lane; C3-reactive bands indicated by arrow). CS or CS + rewarming increased the levels of C3a and C3 cleavage products in NRK and HK-2 cells (Fig. 10, B and C; C3-reactive bands indicated by arrow).

Figure 10.

In vitro cold storage (CS) plus rewarming (RW) induces expression of complement C3. Normal rat kidney (NRK) cells or HK-2 cells were exposed to 18-h CS or 18-h CS + 6-h RW (CS + RW or CS/RW). A: real-time PCR (SYBR green) was performed with renal mRNA extracts (NRK cells) to assay expression of C3 after CS or CS + RW. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. *$P < 0.05, control vs. CS + RW and CS vs. CS + Tx groups. NRK (B) or HK-2 (C) cell extracts were prepared using RIPA buffer (see methods). Proteins of renal extracts (30 µg) prepared from NRK or HK-2 cells were resolved with SDS-PAGE and analyzed by Western blot for C3 after CS or CS + RW. Untreated cell extracts were used as controls. A representative blot from three independent experiments is shown. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant.

CS + Rewarming Induces Chemokines and Inflammatory Cytokines in Rat Renal Proximal Tubular Cells

Next, to determine if CS induced inflammatory gene expression in renal tubules following transplantation, NRK cells were exposed to CS or CS + rewarming and evaluated for inflammatory gene expression. To assess if CS or CS + rewarming affects chemokine production in renal tubular cells, CCL2 expression was evaluated. Quantitative real-time PCR showed induction of CCL2 only after CS + rewarming (Fig. 11A). Inflammatory genes such as TNF-α and IL-1β have been shown to be induced in human kidneys after transplantation (48). Consistent with this, CS + rewarming induced expression of TNF-α and IL-1β in NRK cells (Fig. 11, B and C).

Figure 11.

In vitro cold storage (CS) plus rewarming (RW)-mediated induction of TNF-α promotes biosynthesis of complement C3 in renal cells. A−C: normal rat kidney (NRK) cells were exposed to 18-h CS or 18-h CS plus 6-h RW (CS + RW). Real-time PCR (SYBR green) was performed with renal mRNA extracts to assay expression of inflammatory genes, namely, chemokine (C-C motif) ligand 2 (CCL2), TNF-α, and IL1β after CS or CS + RW. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. *$P < 0.05, control vs. CS + RW and CS vs. CS + RW groups. B: HK-2 cells were exposed to 18-h CS or 18-h CS plus 6-h RW, and renal extracts were prepared using RIPA buffer (see methods). Proteins of renal extracts (30 µg) were resolved with SDS-PAGE and analyzed by Western blot for C3 after CS or CS + RW. Untreated cell extracts were used as controls. A representative blot from three independent experiments is shown. Data are means ± SE (bars, n = 3). Differences between group means were compared with one-way ANOVA followed by Brown–Forsythe and Welch’s correction for multiple group comparisons. P < 0.05 was considered significant. IFN, interferon.

TNF-α Induces C3 in Renal Cells

C3 synthesis (donor specific) has been detected in rejected transplanted human kidneys (49), and cytokines are known to induce C3 biosynthesis in various cell types, including renal (29, 36–41, 43, 50–54). To investigate whether inflammatory cytokines produced during CS + transplant could trigger C3 induction in renal tubular cells, HK-2 cells were treated with recombinant human IFN-γ or TNF-α. HK-2 cell lysates were then evaluated for C3 levels using SDS-PAGE Western blot and C3 antibody. Interestingly, TNF-α treatment (but not IFN-γ) significantly increased C3 biosynthesis in HK-2 cells (Fig. 11D).

DISCUSSION

Transplantation procedures subject kidneys to episodes of ischemia (cold and warm) and reperfusion injury. Clinically, when the donor is deceased, the injury starts at brain death and continues with cold ischemia during CS and warm reperfusion after transplantation. A number of studies in animal models have suggested that complement activation is an important mediator of renal injury during renal warm ischemia-reperfusion (55–57). Indeed, aberrant complement activation is detected during acute and chronic allograft rejection in humans, but the direct connection between cold ischemia (during CS) and complement activation in renal graft is not clear. Here, we demonstrate, for the first time, that CS induces abnormal complement activity in rat renal transplants compared with ATx (no CS exposure), which correlates with graft dysfunction (26, 27). This is the first study in an animal model, if any, to perform head-to-head comparisons in transplants with or without CS (but similar warm ischemia time during the transplant process), and these results highlight why prolonged CS produces unfavorable outcomes in the grafts after transplantation.

The current study demonstrated that renal C3 cleavage occurs during CS, but C3 gene induction and protein activation within kidneys occurs only after transplantation. Because cold temperature lowers the metabolic rate, molecular changes, including C3 biogenesis or C5b-9 production, occur more slowly within donor kidneys during CS. CS (∼18 h) alone produces subtle injury in donor kidneys (58). Taken together, these data indicate that the subtle injury during CS primes the induction and activation of C3 after blood reperfusion (transplantation), leading to pathological activation of the complement system (i.e., C5b-9) after transplantation. A study of human kidney transplantation by de Vries et al. (59) showed a significant but transient (>3 min) release of soluble C5b-9 from the reperfused kidney graft in brain-death and cardiac-death donor kidney transplantation. This study also demonstrated peritubular and tubular C5b-9 deposition during acute rejection. Additionally, they demonstrated that C5b-9 was not present in renal tissue before transplantation. Taken together, these results and ours suggest that cold ischemia triggers early C3 activation and the subsequent reperfusion triggers complement activation and deposition. Similarly, comparable levels of complement activation within kidneys of syngeneic and allogeneic animals at day 1 after CS + transplant suggests that the allogeneity does not affect complement activation at this early stage. Although our study provided evidence that the complement activation persists up to 9 days after transplant, future studies are warranted to address the effect of CS on the long-term status of complement activation in renal transplants.

All complement pathways converge at C3, which plays an important role in the complement cascade. C3, including donor-specific C3, has been detected in rejected human kidneys (49, 60–66), but it has not been clear whether C3 or CS, or both, have a role in kidney injury following transplantation. Our study showed an increase in C3 (and cleaved products) in rat renal transplants (syngeneic and allogeneic), and local C3 synthesis was shown to participate in renal allograft rejection in a murine model (without CS exposure) (67). Our observation of a significant amount of intracellular C3 in rat renal tubules following CS + transplant, but not after ATx, suggests that CS injury amplifies the local synthesis of C3 in renal transplants. As in our in vivo CS + transplant model, our in vitro renal proximal tubular cell CS + rewarming model demonstrated the induction of local expression/synthesis of C3, suggesting that CS induces C3 biogenesis in the renal proximal tubular cell, a compartment that is vulnerable and particularly damaged during ischemia-reperfusion injury.

Cytokines including IFN-γ, TNF-α, IL-1β, and IL-6 can induce C3 biosynthesis in immune and nonimmune cells (29, 36–41, 43, 50–54). Brain death stimulates the synthesis of proinflammatory cytokines in donor organs, and induction of brain death also increases C3 in rat kidneys (29). Indeed, our results demonstrated that CS + rewarming triggers cytokine production (TNF-α and IL-1β) along with C3 in rat proximal tubular cells, and we showed that recombinant TNF-α induced C3 synthesis in those same cells. These findings provide novel mechanistic insights into the CS-mediated induction of intracellular C3 within renal tubules. Future studies are warranted to explore other factors that potentially mediate local C3 increase or biosynthesis after renal CS + transplant.

Cleaved complement peptides bind to and activate complement receptors, which produce downstream signals. CD11b and CD11c integrins are components of complement receptors CR3 (macrophage-1 antigen or α_Mβ₂-integrin) and CR4 (α_Xβ₂-integrin), respectively, and both integrins are expressed by leukocytes and phagocytes, including macrophages (17, 19). Similarly, anaphylatoxins C3a and C5a are produced during the complement cascade and mediate proinflammatory signals through the G protein-coupled cell surface receptors C3AR1 (C3a receptor) and C5AR1 (C5a receptor), respectively (17). We have previously reported that the numbers of CD68⁺ macrophages increased in kidneys after transplantation (1 day postsurgery), and this infiltration was exacerbated by CS in a time-dependent manner (0 h < 4 h < 18 h) (28). Taken together, these results suggest that CS + transplant induces complement activation and macrophage infiltration. It is worth noting that the monocyte chemoattractant CCL2 increased after CS + transplant in our rat model; CCL2 is produced during tissue injury and recruits monocytes (macrophage precursors) to the injured tissue (68). Furthermore, our data demonstrated that renal cells exposed to CS + rewarming produce CCL2. Collectively, these results suggest that anaphylatoxins (C3a and C5a; Fig. 8) and CCL2 promote monocyte-derived macrophage infiltration in renal grafts after CS + transplant. Interestingly, immunoglobulin-driven complement activation was shown to induce CCL2 in a model of pulmonary hypertension (69), supporting the idea that complement can induce CCL2. Future studies are needed to address the relationship between anaphylatoxins and CCL2 induction in the context of renal CS + transplant.

The complement system is made up of a large number of distinct plasma proteins that react with one another to make a complex to mark pathogens or tissue debris to mark them for destruction by phagocytes. Two proteases, C3 and C5 convertases, cleave C3 and C5, leading to formation of the terminal cytolytic product C5b-9 (MAC). The present study demonstrated, for the first time, the impact of CS on the native (nonreducing) forms of C3 and C5b-9. Although the molecular weight of mature C3 in its native form is suggested to be 190 kDa, we detected two distinct novel C3-reactive bands of ∼500 and 350 kDa under nonreducing conditions within sham kidney homogenates (Fig. 2B), suggesting that the renal tissue harbors C3 as a multimeric protein complex under basal conditions. The intensity of these bands increased after CS + transplant. In addition, we observed an ∼110-kDa band of C3 (possibly α-chain) under nonreducing conditions that remained undetected in sham kidney homogenates. While C3 and other complement components are constitutively expressed, C5b-9 is produced after complement activation; this large complex is made up of monomeric C5b, C6, C7, C8, and six to nine molecules of C9 protein, yielding various sizes of MAC (18). As expected, very weak bands of C5b-9 (∼350−1,236 kDa) were detected in Western blots (nonreducing) from sham or ATx kidneys. CS + transplant significantly increased the C5b-9-reactive bands, with the most prominent observed between ∼720 and 2,000 kDa. Future studies should explore the composition of these multimeric complexes of C3 and C5b-9. Although the levels of C3 and C5b-9 protein increased in renal tissues after CS + transplant at 1 day postsurgery, serum levels remained unchanged at that time point; however, interesting changes were evident in serum at 9 days postsurgery. These data suggest that the injury sustained by the donor kidneys activates and induces the complement system systemically after CS + transplant, and the induction is potentially related to changes in the native conformation of the complement components.

Calcineurin inhibitors have greatly improved the short-term survival of organ transplants; however, long-term exposure to calcineurin inhibitors produces nephrotoxicity and compounds allograft failure (70, 71). Various factors contribute to this nephrotoxicity, and studies have demonstrated that prolonged ischemia time for cadaveric kidneys could be an important factor. A randomized prospective study demonstrated an additive effect of prolonged ischemia time for cadaveric kidneys and calcineurin inhibitor exposure that produced nephrotoxicity, and the study recommended renewed emphasis on minimizing ischemic renal damage (72). In addition, the complement system has been proposed to be involved in calcineurin inhibitor-mediated nephrotoxicity. An interesting study by Renner et al. (73) demonstrated that endothelial cells exposed to cyclosporine (a calcineurin inhibitor), both in vitro and in vivo, released microparticles (vesicles that are released during the endothelial cell activation and/or apoptosis) that activated the complement pathway. Furthermore, calcineurin inhibitor-induced renal injury has been associated with activation of the intrarenal complement system (74). These studies further suggest that the duration of CS for a donor kidney could complicate numerous pathways, including complement activation and/or calcineurin inhibitor-mediated nephroxicity. Therefore, strategies should be developed to reduce ischemia-mediated injury during CS to blunt complement activation as well as to reduce calcineurin inhibitor-mediated nephrotoxicity.

Four different complement-interfering strategies have entered into clinical evaluation in the context of kidney transplantation (75, 76). These strategies are used via early complement blockade through C1 interference to prevent the formation of C3 convertase or via late complement blockade through C5 interference to prevent the formation of MAC. Clinically, C1 interference includes three therapeutic approaches. The first involves treating patients with a humanized monoclonal antibody, BIVV009, which selectively blocks the enzymatic C1 subcomponent C1s (77–80). The second approach uses purified (plasma-derived) C1 inhibitor, which blocks activation of C1 subcomponents C1r and C1s and dissociates the C1 complex (81–83). The third approach uses modified extracorporeal treatment including a porous membrane filter for depletion of the C1 subcomponent, C1q (84). Regardless of the approach, C1 appears to be an attractive target for therapeutic intervention in complement-mediated graft rejection. In recent years, the approach of terminal complement blockade using the humanized monoclonal antibody eculizumab has produced promising early results in preventing acute humoral rejection of kidney transplants (85). Eculizumab binds with high affinity to C5 and prevents C5 convertase-mediated cleavage to C5a and C5b, which results in disruption of C5b-9 assembly, while proximal functions of complement are maintained (please refer to Fig. 4A for the complement pathway) (86). Studies performed at the Mayo Clinic have suggested that eculizumab treatment (post-transplant) of sensitized patients with donor cross-match leads to a dramatically decreased incidence of acute humoral rejection and glomerulopathy at 1 yr posttransplant (87, 88). Eculizumab treatment has also been proposed as an effective agent in the prevention or amelioration of ischemia-reperfusion injury. A prospective randomized, controlled trial of pediatric kidney transplantation showed that eculizumab was associated with better early graft function and improved graft morphology, but there was an unacceptably high number of early graft losses among eculizumab-treated children (89). Although early results show promising results of eculizumab in reducing acute or early graft rejection, many questions remain to be answered to define the best indications with regard to the length of beneficial outcome of the therapy and the use of eculizumab in other forms of transplant rejection, such as chronic antibody-mediated rejection and T cell-mediated rejection. Future studies should consider the strategy of initial (C1) or late complement (C5) blockade early during donor kidney preservation and/or posttransplant and assess if this produces beneficial long-term graft/patient outcomes.

Perspectives and Significance

During organ transplantation, including that of the kidney, aberrant activation of the complement system leads to tissue damage. Therefore, complement is recognized as an important target for therapeutic intervention (21–24). Importantly, the cold ischemia-mediated molecular mechanisms that trigger activation of the complement pathway in renal transplants are largely unknown. The present study adds an important new mechanism to our understanding of complement activation in renal grafts. Critically, we have evaluated the effects of CS on donor kidneys using syngeneic and allogeneic rat kidney transplant models. We demonstrated that CS exacerbates complement activation/induction in renal allografts after transplantation and promotes C3 biogenesis in proximal tubular cells and renal tubules, and TNF-α appears to induce C3 biogenesis in kidney tubules. This type of head-to-head comparison in transplants with or without CS using animal model has never been done before, and these results highlight why prolonged CS produces unfavorable outcomes in the grafts after transplantation. Future studies should explore the mechanisms by which CS activates complement and induces renal graft damage (syngeneic and allogeneic) in recipients and should delineate how allogeneity further compounds these mechanisms. Clinically, intervention related to complement activation is performed posttransplantation. Our findings suggest that therapeutics targeting complement inhibition during CS could provide beneficial outcomes following transplantation. It would be interesting to evaluate early (C1) or late (C5) complement pathway inhibition during CS and assess if this intervention produces a beneficial renal outcome, short term or long term. Another important prospect is to define whether CS-mediated systemic C3 activation (global) and local biogenesis (kidney specific) are critical determinants of poor renal outcome after transplantation. These studies may provide a scientific basis for targeting specific complement components during CS to improve graft function, thereby reducing complications and mortality after transplantation.

SUPPLEMENTAL DATA

All Supplemental Material: https://doi.org/10.6084/m9.figshare.14388581.v1.

GRANTS

This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant R01DK123264 (to N.P.), American Heart Association awards 16SDG276000026 (to N.P.) and 19TPA34850057 (to N.P.), a Barton Pilot award (to N.P.), a Medical Research Endowment award (to N.P.), and a Marion B. Lyon New Scientist Development award (to N.P.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

N.P. conceived and designed research; S.L., L.J., S.S., and H.L. performed experiments; S.L., H.L., and N.P. analyzed data; N.P. interpreted results of experiments; L.J. and N.P. prepared figures; N.P. drafted manuscript; S.L. and N.P. edited and revised manuscript; S.L., S.S., H.L., and N.P. approved final version of manuscript.