点击显示 收起
【摘要】 Kidney transplantation is the preferred and definitive treatment for end-stage renal disease (ESRD), and kidneys from deceased donors are a major source for it. These kidneys are routinely cold stored to prolong viability, which, however, when prolonged can cause injury, resulting in reduced graft function and survival. Recent experimental studies have identified the release of iron and free radicals, activation of calpain, and formation of F 2 -isoprostanes as important components of cold ischemic injury, as are the swelling of mitochondria and activation of mitochondrial apoptotic pathways. Moreover, studies have also suggested that fortifying the storage solution with deferoxamine or preconditioning the donor kidneys with hemeoxygenase-1 may prove viable clinical strategies to limit cold ischemic injury. This review will summarize these and other new experimental data that have implications for reducing cold ischemic transplant injury, a step necessary to improve deceased-donor allograft survival.
【关键词】 ischemiareperfusion free radical injury cold preservation cold storage delayed graft function reduced allograft survival
THE KIDNEYS FROM DECEASED donors are subjected to a period of cold ischemia while waiting for tissue matching and transplantation. Unlike these kidneys, kidneys from live donors do not undergo cold ischemia. As a result, renal allografts from live donors, regardless of human leukocyte antigen matching, have better survival rates ( 22, 74 ). Prolonged cold ischemia of deceased donor kidneys leads to higher graft dysfunction, higher recipient mortality, and higher health care costs. Furthermore, based on the "response-to-injury" hypothesis ( 21, 34, 67 ), injury from cold ischemia may set the stage for chronic injury. The number of patients waiting for transplantation is steadily increasing, and currently this number in the United States is close to 88,000, a threefold increase over the last decade. The higher graft failure due to cold ischemia adds to kidney scarcity by requiring the recipients to return to the waiting list. According to the United Network of Organ Sharing (UNOS) registry, the average cold storage time has remained at 21 h over the last decade ( 61 ). Thus prolonged cold ischemia remains a persistent problem with costly and serious consequences that merit scientific consideration.
This article will review briefly 1 ) the reasons deceased donor kidneys are routinely cold stored, 2 ) the history of cold storage, 3 ) cold ischemia as a persistent clinical problem, 4 ) recognized mechanisms of cold injury, 5 ) the inadequacies of existing cold storage solutions, 6 ) new mechanisms of cold injury, and 7 ) new potential strategies to prevent cold storage injury.
WHY DECEASED DONOR KIDNEYS ARE COLD STORED
After the procurement of kidneys, the immediate cause of renal injury is ischemia, which initiates complex injury processes such as ATP loss, hypoxanthine accumulation, loss of the Na + -K + pump, cell swelling, and cytosolic calcium increase, among others ( 77 ). Currently, a rapid in situ flushing of the kidneys with a preservation solution and rapid cooling to 4°C remain the main strategy to minimize ischemic injury and increase viability. Thus storage of organs at a lower temperature plays a very important role in organ transplantation. However, cold storage ironically allows prolonged ischemia, and this coupled with direct injury from hypothermia seriously damages organs, even if they are viable at the end.
HISTORY OF COLD PRESERVATION
The first recorded attempt at perfusion of an isolated organ occurred in 1849 by Loebel ( 24 ). In 1895, Langendorf devised a simple organ-perfusion technique. In 1937, Lindbergh and Carrel, fascinated with hypothermic preservation, created a perfusion apparatus. In 1953, Lapchinsky from the Soviet Union started successfully transplanting limbs and kidneys preserved at 4°C. In 1964, Belzer, while working with Najarian to develop a cadaver kidney transplant program at the University of California (San Francisco, CA), started to work on hypothermic perfusion techniques for the preservation of kidneys ( 1 ).
COLLINS' SOLUTION: THE SOLUTION THAT CHANGED TRANSPLANTATION PRACTICE
However, it was Collins et al. ( 9 ) who first developed a simple yet effective cold storage solution in 1969. The solution contained a high concentration of glucose and electrolytes, mostly of intracellular composition. The seminal observation of Collins et al. that storage at 4°C after a simple perfusion can extend the viability of cadaver kidneys changed the practice of transplantation from an emergency procedure to a semielective one. It also resulted in better tissue matching and effective utilization of retrieved kidneys, such that cadavers remain the significant source for transplant organs in the United States. While Collins' solution had extended the viability of kidneys, it soon became clear that it was less suitable for prolonged periods and for other organs.
RECOGNIZED MECHANISMS OF COLD INJURY
There are at least four components to cold ischemic transplant injury: the coupled effect of ischemia and hypothermia during cold storage and the coupled effect of reperfusion and rewarming after transplantation. The effects of ischemia and reperfusion are widely studied, but the contribution of hypothermia and rewarming to them is difficult to separate and rarely studied. The known mechanisms of cold ischemic injury are related to perturbations in osmoregulation, energetics, and aerobic metabolism ( 2 ). Reduced Na-K-ATPase activity and ATP levels permit intracellular accumulation of sodium and water, resulting in cell swelling. During cold ischemia, lactic acid generated from the anaerobic metabolism of glucose contributes to intracellular acidosis, resulting in lysosomal instability and altered mitochondrial function. Cold storage, due to a reduction in interstitial oncotic pressure, allows interstitial expansion and edema, leading to capillary compression, tissue injury, and poor organ function after transplantation.
UNIVERSITY OF WISCONSIN SOLUTION: A BETTER SOLUTION WITH PROBLEMS
Based on a series of attempts, Belzer and Southard ( 2 ) encountered some of the above-mentioned problems limiting Collins' simple storage solution, and in the early 1980s developed University of Wisconsin (UW) solution. A number of large-molecular-weight cell impermeants were added to reduce tissue swelling ( 69 ). Glucose was replaced with phosphate buffers. On theoretical grounds, adenosine for rapid ATP repletion and glutathione (GSH) and allopurinol for antioxidant property during the reperfusion phase were added ( 69 ). UW solution, proven to be superior to many other solutions, has provided a significant step toward an ideal solution. However, although it extends organ viability, severe organ injury is quite common with extended cold storage. Furthermore, recent work has shown that GSH added to UW solution at the time of manufacture is oxidized with time and that UW solution is contaminated with reactive iron ( 15, 26 ). The limitation of UW solution is also evident in clinical practice. Data analysis from the UNOS shows high failure rates for kidneys subjected to cold storage over 24 h ( 47, 61 ). With the accrual of new information on the cellular and molecular mechanisms of cold ischemic injury, it is also clear that the protection strategy employed for UW solution is too simplistic.
REPERFUSION-REWARMING INJURY: THE UNAVOIDABLE SECOND HIT
After transplantation, additional injury accrues with rewarming and reperfusion of the cold ischemic organs. An increase in reactive oxygen species and cytosolic calcium and activation of several calcium-dependent and -independent proteolytic enzymes, including activation of caspases, are implicated in the process ( 41, 51, 60 ). The process and mechanism of reperfusion injury of cold ischemic organs are likely similar to the reperfusion injury of warm ischemic organs, although no direct comparison studies are available. Furthermore, although hypothermia may modify the organs' reactions to ischemia and subsequent reperfusion, separating the hypothermic effect from the ischemic effect in cold ischemic organs remains practically impossible.
In the warm ischemia-reperfusion model, neutrophils, adhesion molecules, and lately T cells have been shown to play some role in the injury process ( 38, 53, 54 ). Endothelial cell dysfunction and activation of leukocytes contribute to the inflammatory process with the coordinated release of several cytokines and chemokines ( 21 ). It is widely held that cold ischemic injury increases allograft immunogenicity, provoking acute and chronic rejections ( 21, 34 ). Whether innate and adaptive immune responses play any key role in the process and whether the immune system within the allograft contributes to the injury process are prevailing questions of considerable interest ( 35 ).
ENDOTHELIAL DYSFUNCTION AFTER COLD ISCHEMIA-REPERFUSION
Renal allograft vascular integrity is critical for early and longer term kidney transplant function. As in native kidney warm ischemia-reperfusion injury, intense vasoconstriction and endothelial injury are prominent features of cold storage injury ( 42 ). In the transplant setting, recipient-derived repair mechanisms including the homing in of stem cells have been proposed to influence allograft acceptance and recovery of function ( 12, 43 ). However, few precise indicators exist to predict the recovery and repair of transplanted organs.
CLINICAL SIGNIFICANCE: COLD ISCHEMIA REMAINS A DEVASTATING CLINICAL PROBLEM
Sharing deceased donor kidneys at the national level may improve tissue matching, which in practice, unfortunately, increases cold ischemia time. With better immunosuppression, it is debatable whether cold ischemia for better tissue matching is justifiable. However, for logistical reasons cold storage and hence cold ischemic injury are likely to continue. A recent analysis of recent UNOS clinical data reported that 1 ) cadaveric kidneys on average still continue to undergo long periods of cold ischemia, 2 ) kidneys with longer cold ischemia time have the worse posttransplant kidney function, and 3 ) cold ischemia continues to negatively impact renal allograft survival ( 61 ).
NEWER INSIGHTS INTO COLD ISCHEMIC INJURY MECHANISMS
Mode of Cell Death During Cold Storage and Reperfusion: Targeting Apoptosis
Although acute warm ischemic injury causes massive cell necrosis, apoptosis does contribute to injury, particulalry to the accrual of postreperfusion injury and is determined by several factors including the loss of ATP and, as recently described, GTP ( 31, 49 ). Although cold ischemic injury, as in warm ischemic injury, is considered to be due to massive cell necrosis, several recent studies suggest that the apoptotic form of cell death does occur, but only after the reperfusion of transplanted organs ( 6, 46, 62, 65 ). Our current view is that cold ischemia causes cell necrosis, but during reperfusion a significant proportion of sublethally injured cells undergo apoptosis ( 65 ). Data from Burns et al. ( 6 ) in cold-stored human kidney transplants and our own in vitro study in renal tubular cells ( 64 ) are consistent with this view. Currently, suppressing apoptosis in the acute injury setting is considered beneficial. Limited early reports from a liver reperfusion model suggest that antiapoptotic treatments might reduce dysfunction of cold-stored organs ( 46 ). The events during cold storage that lead to the apoptotic form of cell death have not been studied in any detail. During an electron microscopic analysis, we unexpectedly came across the presence of previously undescribed massive mitochondrial swelling of cold-stored cells ( Fig. 1 ) ( 65 ). This prompted us to study the role of the mitochondrial apoptotic pathway in the setting of cold ischemic renal injury ( 62, 65 ).
Fig. 1. Electron microscopy of cold-stored human renal proximal tubular cells grown on Aclar embedding films as reported in Salahudeen et al. ( 65 ). A : cells at 37°C. B : 4 h of cold in University of Wisconsin (UW) solution: many mitochondria show signs of swelling (arrow). C and D : mitochondrial swelling is progressive at 12 ( C ) and 24 h ( D ) of cold. E : at 48 h of cold, plasma membranes are disrupted, mitochondria are undiscernible, and vacuoles appear in cytoplasm. F : well-preserved mitochondria in cells cold stored with 2.5 mM deferoxamine (DFO) for 48 h.
Cold-Induced Mitochondrial Swelling and Activation of the Mitochondrial Apoptotic Pathway
Subjecting hearts to warm ischemia-reperfusion was shown to be attended by cyclosporine A (CsA)-suppressible mitochondrial permeability transition pore (PTP) opening ( 20 ). The PTP, formed from mitochondrial membrane-bound adenine nucleotide translocase and cyclophilin D, opens with calcium and free radicals and closes with CsA ( 10, 17, 20 ). In our study, cold storage of human renal tubular cells was attended by mitochondrial PTP opening and mitochondrial swelling ( 62 ). Moreover, cold-induced mitochondrial swelling was attended by a number of key proapoptotic events such as cytochrome c translocation and Bax-to-Bcl-2 ratio increase ( Fig. 2 ) ( 62 ). The final event of caspase-3 activation and the actual occurrence of apoptosis were observed only during rewarming. These experiments were conducted using cultured cells in UW solution and exposed to cold, which, unlike cells in a retrieved kidney, do not endure disruption or ischemia. These limitations notwithstanding, the cell culture system has offered a platform for testing basic mechanisms in cold storage injury, which cannot be readily conducted in whole kidney experiments. Supporting our cell culture work was the recent clinical study by Castaneda et al. ( 7 ) demonstrating a role for mitochondrial events in the apoptosis of human renal allografts transplanted after cold ischemia.
Fig. 2. A likely scenario for cold-reperfusion-associated apoptosis based on data presented by Salahudeen et al. ( 62 ). Cold ( top ) via calcium and free radicals opens permeability transition pore (PTP), causing marked mitochondrial swelling, which, in turn, triggers key apoptotic events and sets the stage for apoptosis during rewarming ( bottom ). Bcl-2 family of proteins, Bax, Bcl-X L, and Bcl-2, are particularly abundant at the junction between inner and outer mitochondrial membranes, which is the site where PTPs are formed and the site of membrane disruption during cold storage-rewarming ( 62 ). Bcl-2 counteracts the proapoptotic activity of the pore-forming Bax protein. Normally, the ratio of Bcl-2 to Bax is maintained in favor of Bcl-2, but during cold storage it is shifted toward Bax ( 62 ). Mitochondrial leakage of cytochrome c and other proapoptotic proteins such as Apaf-1 leads to the formation of apoptosome complexes composed of cytochrome c (cyt c), Apaf-l, ATP, and procaspase-9. Formation of the complex makes Apaf-I more competent at binding procaspase-9 and recruiting other caspases-1, -2, -3, and -4 through its recruitment domain (CARD), triggering the caspase cascade, the latter occurring during the rewarming phase after cold storage ( 62 ).
Any Role for Receptor-Mediated Apoptosis in Cold Ischemic Transplant Injury?
Fas is a widely expressed cell-surface receptor that can initiate apoptosis when activated by its ligand (FasL). In a mouse myocardial infarction model, Fas-deficient lpr mice, compared with wild-type mice, exhibited smaller infarct size and reduced myocyte apoptosis, suggesting indirectly that Fas might be important in this model ( 37 ). Others have found a role for CD95/Fas/Apo1 and TNF- in the induction of postreperfusion apoptosis in isolated rat hearts ( 28 ). Again, it is to be noted that all these studies had been conducted in warm ischemia-reperfusion models. Whether Fas-FasL plays a role in cold reperfusion-induced apoptosis and inhibition of downstream events, such as caspase-3 activation, would protect against cold reperfusion-associated renal transplant apoptosis is presently not clear.
Calcium, Calpain, Calpastatin, and PKC in Cold Ischemia-Reperfusion Injury: Potential Opportunity for Additional Intervention
Although cytosolic calcium levels are tightly regulated, injuries can often lead to a rise in free calcium that, in turn, can activate calcium-dependent injury pathways ( 14 ). UW solution does not contain calcium. Yet, in our preliminary study cold storage of renal tubular cells in UW solution was found to increase cytosolic calcium ( 29 ). Addition of calcium-chelating EGTA in this setting reduced cold storage-induced cell injury ( 29 ). Because cytosolic calcium is increased in the cold, it is reasonable to examine the role of calcium-dependent process such as calpain activation and PKC signaling in the injury process. Calpains are calcium-dependent proteases: two ubiquitous isoforms are µ-calpain (I) and m-calpain (II). The endogenous calpain inhibitor calpastatin also regulates calpain activity ( 17a ). Germane to cold ischemic injury is the finding during warm ischemia studies that calpain activation leads to the breakdown of the cytoskeletal protein spectrin, partly accounting for the loss of cell structure, polarity, and cell-cell contact after reperfusion ( 17a ). Additionally, Kohli et al. ( 33 ) reported in a cold storage model of ex vivo liver transplantation that hepatic calpain activity steadily increased in the cold and intensified with rewarming and that a calpain inhibitor reduced liver injury. Although other studies on warm ischemia-reperfusion injury of native organs have suggested a role for calpain in the ischemic injury process ( 8, 75 ), the study by Kohli et al. ( 33 ) is the only one in cold-stored transplanted organs, and this remains to be confirmed. Furthermore, whether the calpain inhibitor or calpastatin reduces cold ischemic transplant injury is yet to be determined.
Besides its regulatory role in the cell cycle, PKC is also shown to participate in several pathophysiological conditions, including in the recovery process of warm ischemic organs ( 48 ). PKC is composed of at least 10 isoenzymes, which, once activated, undergo translocation across cellular compartments ( 48 ). Activated PKCs use receptors for activated C kinase as anchoring proteins that enable compartmentalization. Prior pharmacological activation of PKC has been shown to protect against ischemia and, consistent with this, PKC activation, specifically that of the -isoform ( 19 ), is linked to the preconditioning phenomenon in myocardium ( 70 ). In cardiomyocyte cell culture work, hypoxia induced both PKC- and -, but it was the -isoform antagonist, not the -isoform antagonist, that completely abrogated hypoxic preconditioning ( 70 ). Moreover, the latest data suggest that PKC- can be a proapoptotic molecule by functioning as a lamin kinase, disassembling lamina during apoptosis. That PKC- can be protective, whereas PKC- can be destructive, is also suggested by other studies ( 27, 39 ). Very few studies have been carried out examining the role of PKC during cold storage, and in one, the use of nonspecific PKC inhibitors has been shown to improve rat heart preservation ( 16 ). The presence of many PKC isoforms and their seemingly contradictory roles can make PKC studies difficult to interpret. Nonetheless, the recent availability of specific antibodies and inhibitors against PKC isoforms ( 71 ) will help clarify the role of PKC in cold ischemia- and warm ischemia-reperfusion-associated organ injury during transplantation.
Free Radical Injury in the Cold: A True But Counterintuitive Phenomenon
Free radicals are important mediators of reperfusion injury of transplanted organs, but their involvement in cold storage-induced injury is only beginning to be recognized ( 55, 56, 63 ). The notion that cold might be associated with free radical injury is somewhat counterintuitive, because cold temperatures, in themselves, are associated with reduced cellular metabolism and enzymic activity. Indeed, the objective of Belzer and Southard ( 2 ) to add antioxidants to UW storage solution was more to minimize reperfusion injury than to prevent radical-mediated cold storage injury. A number of recent studies ( 4, 23, 52 ), including those from our laboratory ( 63, 66 ), suggest a role for free radicals in cold storage injury. In our study, the genetic expression of mitochondrial Mn SOD, but not of cytosolic Cu-Zn-SOD or of glutathione peroxidase, increased with cold exposure, suggesting mitochondria as a cellular source of free radicals ( 63 ). Indeed, our recent work demonstrating striking mitochondrial derangement with cold storage supports this view ( 62 ). As alluded to earlier, UW solution has minimal antioxidant efficacy, and this is confirmed by our finding that the addition of deferoxamine to UW solution affords further protection against cold storage injury ( 26, 63 ).
Free Iron Release During Cold Storage: A Clinically Addressable Problem
In normal cells, free iron is nearly undetectable. Free iron at higher levels can catalyze the Haber-Weiss reaction, resulting in the generation of toxic hydroxyl radicals. Earlier studies have shown that iron can be toxic in many forms of renal injury ( 50, 68 ), including cold storage injury ( 18, 32, 57, 78 ). Consistent with a role for iron toxicity in cold storage injury, we have recently found the copious release of free iron during cold storage, mainly of microsomal origin ( 26 ). In the clinical setting, sequestering free iron released during cold storage is possible, and such a strategy may prove to be useful as suggested by our work in the rat transplant model ( 25 ). Although iron chelators including deferoxamine have been used in a number of clinical studies ( 13, 40, 58, 73 ), no published studies to our knowledge have addressed the efficacy of similar agents in cold ischemic injury of human organs removed for transplantation. While deferoxamine's protective effect that we found in the transplant model against cold is considered mainly due to its antioxidant properties ( 25 ), other effects of deferoxamine such as upregulation of cytoprotective genes, HO-1, EPO, and VGEF via HIF-I transcription factor are to be considered ( 3, 76 ). Consistent with the release of iron during cold ( 26 ) is our unpublished but novel observation in human tubular cells by microarray that cold storage causes specific and severe suppression of the ferritin-H gene. Thus the increase in the injurious free iron noted by us ( 26 ) and Rauen et al. ( 57 ) during cold storage could also be due to a reduction in the cold of iron-sequestrant ferritin.
Cold-Induced F 2 -Isoprostanes: A New Player in Cold Storage-Associated Vasoconstriction?
Cold storage of human renal tubular cells or rat kidneys was associated with a striking time-in-the-cold-dependent increase in F 2 -isoprostanes, which were significantly suppressed with the inclusion of the antioxidant 2-methy aminochroman or deferoxamine ( 66 ). Similarly, inclusion of deferoxamine in the cold storage solution reduced renal isoprostanes levels and improved the cold ischemia-associated reduction in renal blood flow and transplant dysfunction in a syngeneic kidney transplant model ( 25 ). Isoprostanes are vasoconstrictive prostaglandins formed during free radical-catalyzed lipid peroxidation. The discovery of isoprostanes was prompted, in part, by the mass-spectrometric observation that storage of plasma at -20°C was associated with the production of PGF 2 -like compounds ( 44 ). These compounds in subsequent studies were found to be formed by lipid peroxidation of arachidonyl-containing lipids in plasma and tissue and, because of their structural similarity to PGF 2, these compounds are collectively referred to as F 2 -isoprostanes ( 59 ). The biological effects of isoprostane have been explored with one of the synthetically available isoprostanes, 8-epi-PGF 2. Infusion of 8-epi-PGF 2 at nanomolar concentrations into rat renal artery has been shown to cause a dramatic dose-related reduction in the glomerular filtration rate and renal blood flow mainly due to afferent arteriolar vasoconstriction ( 72 ). Data from recent experimental and clinical studies support the suggestion that isoprostanes may play a role in renal vasoconstriction, demonstrable in certain experimental models and clinical settings of renal injury including cold ischemic injury of transplanted organs ( 45, 66 ).
Deferoxamine to Reduce Cold Ischemic Renal Transplant Injury: A Simple Strategy for Clinical Testing
The idea of adding antioxidants to the storage solution to reduce transplant injury is not new, neither is the potential use of deferoxamine during cold storage ( 18, 78 ). However, ours was the first study to test deferoxamine in a kidney transplant model ( 25 ). The transplanted kidneys' function assessed on day 10 of transplantation was 75% higher in the deferoxamine-treated kidneys compared with untreated kidneys. Moreover, deferoxamine treatment was attended by a significant reduction of renal apoptotic and necrotic scores and F 2 -isoprostane content. Our findings in the transplant model suggest that a simple strategy of including deferoxamine in the cold storage solution may reduce cold ischemic renal transplant damage in the clinical setting. In support of a role for iron in ischemia-reperfusion injury is the finding of DeBoer and Clark ( 11 ) in a myocardial ischemia-reperfusion model. Deferoxamine added to the storage solution was cardioprotective but not when administered on reperfusion. Deferoxamine is widely available and water soluble. Moreover, by adding it to the storage solution, the effect is confined to the donor kidneys as the storage solution is flushed out before transplantation. This attribute, along with a strong scientific rationale and a great deal of experimental evidence, suggests that deferoxamine is ready for testing in the clinical transplant setting.
Inducing HO-1 Before Cold Storage: Another Strategy for Clinical Testing
Stress protein HO-1 is a 32-kDa microsomal enzyme that degrades heme proteins and, in the process, releases bilirubin, CO, and free iron. We demonstrated that overexpression of HO-1 in renal tubular cells, either through gene transfer or induction with hemin, can protect these cells against cold storage injury ( 64 ). Our in vitro finding corroborates in vivo findings of others in heart, kidney, and liver transplantation models ( 30 ). HO-1 through the release of free iron stimulates synthesis of ferritin, the main cellular repository for free iron ( 64 ). The finding that iron is released in copious amount in the cold ( 26 ) and that HO-1, which induces ferritin, protects cells against cold strongly suggests ferritin as an important mechanism for HO-1 protection. HO-1 also produces CO and bilirubin, both of which are suggested to contribute to the cytoprotective properties of HO-1 ( 5, 30 ). In essence, there is a strong scientific basis and much experimental backing to put the HO-1 concept to the test in a clinical trial.
In summary, cold ischemic injury remains a persistent problem with serious clinical consequences. Recent experimental studies identify a number of novel mechanisms, but a role for iron toxicity and mitochondrial apoptosis appears to be compelling. Future experimental studies will define the utility of antiapototic, anticalpain, anti-inflammatory, and anti-immune strategies to limit cold ischemic injury. Meanwhile, studies are required to test the utility of HO-1 preconditioning or adding deferoxamine to the storage solution in the clinical transplant setting.
GRANTS
This work was funded by National Institute of Diabetes and Digestive and Kidney Diseases Grant RO-1 DK-56835-01.
Address for reprint requests and other correspondence: A. K. Salahudeen, Dept. of Medicine, Univ. of Mississippi Medical Ctr., 2500 North State St., Jackson, MS 39216 (E-mail: asalahudeen{at}medicine.umsmed.edu