Kidneys derived from brain death organ donors show an inferior survival when compared to kidneys derived from living donors. Brain death is known to induce organ injury by evoking an inflammatory response in the donor. Neuronal injury triggers an inflammatory response in the brain, leading to endothelial dysfunction and the release of cytokines in the circulation. Serum levels of interleukin-6, -8, -10, and monocyte chemoattractant protein-1 (MCP-1) are increased after brain death. Binding with cytokine-receptors in kidneys stimulates activation of nuclear factor-kappa B (NF-κB), selectins, adhesion molecules and production of chemokines leading to cellular influx. Mitogen-activated protein kinases (MAP-kinases) mediate inflammatory responses and together with NF-κB they seem to play an important role in brain death induced renal injury. Altering the activation state of MAP-kinases could be a promising drug target for early intervention to reduce cerebral injury related donor kidney damage and improve outcome after transplantation.
To date organs derived from brain-dead donors are the main source used in transplantation. However, when compared to organs retrieved from living donors, deceased donor kidneys have a significantly increased risk of delayed graft function and inferior long-term graft survival (1–3). The unphysiological state of brain death followed by preservation and ischemia/reperfusion-induced injury (4,5) will damage the kidney graft-to-be. After transplantation, the functioning nephron mass is reduced, which leads to exhaustion of the remaining nephrons due to hyperfiltration trying to meet the metabolic needs (6) and result in accelerated dysfunction (7). Recent microarray studies have demonstrated upregulated expression of genes related to inflammatory responses and reparative mechanisms in kidneys after brain death (4,8). Transcription can be induced via activation of signal transduction pathways, which can be activated by circulating cytokines after binding to cytokine receptors. Several kinases involved in intracellular signal transduction are activated to induce activation of specific signal transduction pathways that will lead to the induction of a local inflammatory response. This review will focus on some important molecular aspects of brain death-induced renal injury and describe a number of promising drug targets. The signal transduction pathways described in this review for the kidney are probably also relevant for decreased viability due to brain death in other organs.
Induction of brain death leads to hemodynamic changes
Brain death is most often due to either cerebral hemorrhage, trauma or an anoxic event. The subsequent formation of edema will suppress the blood flow and diminish cerebral perfusion leading to an ischemic state. Ischemia of the hypothalamus induces the Cushing reflex that results in hemodynamic instability with severe hypotension and poor organ perfusion. This significant ischemic state is marked by elevated serum lactate levels (8–10). Appropriate donor management with hemodynamic stabilization after brain death toward normotensive levels limits organ dysfunction (8,11) and decreases the severity of the inflammatory response after brain death, which is demonstrated by a decrease in the number of CD8-positive cells in glomeruli and interstitium (9). The induction of this inflammatory response in kidneys could be due to hypoperfusion and ischemia of the peripheral organs. However, brain death itself causes release of cytokines in the circulation, which are able to trigger an inflammatory response in peripheral organs. Since hemodynamic stabilization decreases the severity but does not prevent the inflammatory response in kidneys, this response is probably due to the combination of peripheral hypoperfusion and brain death.
Neuronal injury triggers an inflammatory reaction in the brain
Besides hemodynamic changes influencing the inflammatory response, neuronal injury itself induces an inflammatory response in the brain leading to endothelial injury and leakiness of the blood-brain-barrier. Dysfunction of the blood-brain-barrier allows cytokines produced in the brain to leak away into the serum. Patients with traumatic brain injury show elevated serum levels of S100A9 (4), S100β (12,13), interleukin-6 (IL-6) (14,15), complement activation and coagulation disorders (16). S100β appears to be an important peptide as it is not only involved in protein phosphorylation, enzyme activation, cell proliferation and differentiation, but also in cytoskeletal dynamics, intracellular Ca2+-homeostasis and protection against oxidative injury (12,13). The timeprofile of release suggests a role for S100β in delayed reparative processes. Incubating neuronal cell cultures with anti-S100 reduces the release of S100β and increases neuronal injury in traumatized cells, which confirms its neuroprotective role (17). It has been suggested that serum levels of S100β can be used to monitor neuronal injury, unless brain-blood barrier has remained intact, which prevents S100β from leaking into the systemic circulation (18). S100A9 can induce DNA fragmentation, activation of caspases and inflammatory responses in endothelial cells, ultimately leading to loss of endothelial cell contacts (19) and cell death (20). Activation of S100A9 will diminish the function of the blood-brain-barrier, allowing substances produced in the brain now to enter the systemic circulation. Activation of C5b-9 affects brain function negatively and correlates with S100β activity (21). Activation occurs mostly on the blood side of the brain-blood barrier, which is demonstrated by the fact that no difference between jugular (venous) and arterial levels (transcranial gradient) can be found and that activation of C5b-9 in blood is higher than in cerebrospinal fluid (16).
Several studies reported elevated serum levels of IL-6 after neuronal injury (14,15,22). Synthesis is probably caused by interleukin-1β (IL-1β) in the brain. Activation of microglial cells induces transcription of pro-IL-1 β (23,24), which accumulates in the cell in the absence of cell death, but is released after apoptosis (25). Intracellularly, it can be cleaved by caspase-1 to its active form IL-1β and then released after apoptosis, cell lysis (necrosis), or due to secretory lysosomes or microvesicle shedding (apoptosis) (26). Binding with the IL-R1R receptor induces transcription of nuclear factor-kappa B (NF-κB) and IL-6 (27) via activation of intracellular signal transduction pathways. Levels of IL-6 in brain parenchyma correlate with survival after brain injury, which suggests a neuroprotective role of IL-6 in the brain after injury (22). In patients with brain injury progressing toward brain death, jugular IL-6 levels are higher than arterial IL-6 levels, which suggests that IL-6 originates (‘leaking away’) from the brain (14). Release of IL-6 in the circulation leads to induction of a wide range of systemic effects (28,29) and IL-6 is considered the most important mediator in the acute phase response (30). In this respect it is worthwhile mentioning that an inverse correlation has been demonstrated between serum levels of IL-6 and kidney graft survival (31,32) (Figure 1).
Besides elevated levels of IL-6 (5,11,33), also, elevated levels of interleukin-8 (IL-8) (11), interleukin-10 (IL-10) (5) and MCP-1 (11) can be found in the serum of brain-dead donors. Probably these cytokines originate as well from the brain. In the brain, transcription of CD-40 and MCP-1 is induced in macrophages and microglia after activation of NF-κB and interferon-β (IFN-β) (34). CD40-CD40 interaction induces secretion of IL-8 (35). IL-10 in the brain serves a regulatory role by inhibiting IFN-β (36). Despite elevated serum levels of IL-6, IL-8, IL-10 and MCP-1, it is remarkable that only minor increases in the level of IL-1β and that no increases in the level of tumor necrosis factor-alpha (TNF-α) are found in serum (5,11,33). The question remains whether this is due to a short half life of these cytokines. As a result of cerebral injury and brain death the elevated levels of circulating cytokines will induce a local inflammatory response in the kidney and affect renal function already prior to retrieval of the graft-to-be.
Mitogen-Activated Protein Kinases (MAP-Kinases)
Although it is known that brain death leads to elevated levels of cytokines in serum, it remains questionable which signal transduction pathways are involved in inducing the inflammatory response in the kidney. An important family of protein kinases involved in signal transduction is the family of MAP-kinases. Four MAP-kinases are known: c-Jun N-terminal activated kinase (JNK) (37–43); p38-kinase (44,45); p42/44 or extracellular signal regulated kinase (ERK1/2) (46–48) and big MAP-kinase (BMK or ERK5) (49). MAP-kinases can be activated by a variety of stimuli including cytokines, growth factors, peptide hormones, but also by pathological stressors like ischemia/reperfusion (50–55), ultravioletirradiation (40,41,56) and neuronal injury or osmotic shock (57,58). Although each MAP-kinase exerts specific effects, redundancy is demonstrated between the pathways. Stimuli known to activate JNK can also lead to marginal activation of ERK1/2 and vice versa (40). TNF-α-induced IL-6-production in mesangial cells can be reduced by incubating the cells by using an ERK1/2- and a JNK-inhibitor at the same time, while the inhibitors are separately not effective (59). Cross-talk between pathways is also known and is demonstrated by the fact that sustained activation of JNK will lead to inhibition of ERK, ultimately leading to apoptosis of the cell (60). The activity of MAP-kinases is based on the balance between the presence of stimuli acting via receptor stimulation and upstream MAPK-kinases and the presence of MAPK-phosphatases intracellularly (61,62).
MAP-kinases and NF-κB as key players in the inflammatory response in the kidney
MAP-kinases could play a major role in initiating the inflammatory response in the kidney after brain death. As described above, induction of brain death leads to an elevated level of IL-6 in serum. IL-6 can bind plasma membrane receptors containing glycoprotein 130 (gp130) leading to dimerization of the receptor and activation of Janus kinase) and STAT. This activation induces phosphorylation of JNK, p38 and ERK1/2 (63,64), but also activation of NF-κB (65–68). Activation of NF-κB occurs rapidly after induction of brain death (11), possibly due to phosphorylation of p38. Phosphorylated p38 can induce activation of protein kinase casein kinase II, which in turn can phosphorylate p53 (known to induce cell cycle arrest or apoptosis (69)) or the NF-κB-inhibitor (IκBα). Phosphorylation of IκBα will lead to ubiquination and proteolysis of IκBα resulting in an inhibitor-free p50/RelA-complex, NF-κB. This NF-κB-complex translocates to the nucleus, where it exerts its effects in immunity, inflammation, and cell survival, including cell growth and proliferation or cell death (65–68). Besides, translocation of NF-κB inhibits JNK (66,70–72) (Figure 1).
Activation of JNK can lead to apoptosis of the cell via activation of caspase-3 and -9 (66,70,71,73). Since the fate of the cell in terms of survival or apoptosis is determined by the balance between ERK1/2 and JNK, we accomplished a pilot experiment to assess the function of this ‘rheostat’ after brain death. To examine activation of ERK1/2 and JNK in kidneys of brain-dead rats (experiment described in (11)), we used immunohistochemistry and Western Blotting techniques with phosphorylation-specific antibodies. Phosphorylation of ERK1/2 decreased while phosphorylation of JNK increased time-dependently from 30 min to 4 h after brain death (p < 0.05).
Signal transduction involved in transcription of genes related to inflammatory responses and protective mechanisms
Activation of MAP-kinases ultimately leads to activation of down-stream transcription factors and expression of genes related to inflammatory responses. c-Myc is a down-stream substrate of ERK1/2 and p38, while c-Jun and c-Fos are well-known down-stream substrates of JNK and ERK1/2. In kidneys derived from brain-dead rats gene expression of c-Myc is increased (4,8) and protein production of c-Jun and c-Fos is increased as well (74). Especially, the activation of c-Myc, c-Jun and c-Fos after brain death suggests a role for MAP-kinases in mediating gene expression involved in the inflammatory response after brain death. The most profound changes in the kidney are found in expression of E-selectin, MCP-1 (75), interferon-γ (IFN- γ), interleukin-2 (76), IL-6 (8,11,32,76,77), IL-8 (8), IL-10, tumor growth factor-β1 (TGF-β1) and platelet-derived growth factor- β (PDGF- β) (76).
The role of MAP-kinases in transcription of these inflammatory genes has been addressed by several in vitro studies. Transcription of IL-6 and IL-8 is dependent on activation of p38 (78–89) and ERK (59,79–82), which is confirmed after blocking these MAP-kinases. Most potent reduction of IL-6-expression was observed when both p38 and ERK were inhibited (59). Expression of IL-8 cannot be inhibited with a JNK-inhibitor (79,85). Transcription of MCP-1 is not dependent on ERK-activation (90) or JNK-activation (91), but is dependent on activation of p38 (90–92). In regulating gene expression of E-selectin, JNK plays a pivotal role in combination with p38, while no role is put aside for ERK. Activation of p38 and JNK leads to phosphorylation of activating transcription factor 2 (ATF-2) and c-Jun. These transcription factors are together with NF-κB needed to induce expression of E-selectin (93,94). Both a JNK-inhibitor (93,95–99) as well as a p38-inhibitor (93,99–102) are able to reduce gene expression of E-selectin (Figure 1).
MAP-kinases may be regarded as key players in the inflammatory response, by initiating expression of genes related to inflammation in the kidney after brain death. NF-κB is also regarded as a key player in regulating the inflammatory response by inducing gene expression of IL-6 (102–104), IL-8 (102,105,106), E-selectin (102,106) and MCP-1 (107) (Reviewed by Pahl et al. (108)). The effects of p38 are possibly exerted via NF-κB since inhibiting one of these two will lead to a reduction in transcription of IL-6, IL-8, E-selectin and MCP-1 (109). Since p38-activation will also cause activation of NF-κB (65–69), activation of p38 will be sufficient to induce this p38/NF-κB-cascade leading to transcription of IL-6, IL-8, E-selectin and MCP-1. (66,70–72).
Besides these changes inducing organ injury, signal transduction and gene expression involved in protective mechanisms are also upregulated in expression after brain death. After brain death in organ donors, expression of heme-oxygenase-1 (HO-1) and heat-shock protein 70 (Hsp70) increases 3- and 2.5-fold, respectively (8,110). Expression of the suppressor of cytokine signaling-3 (SOCS-3) is increased almost 3-fold after induction of brain death (8). SOCS-3 plays an inhibiting role in inflammatory responses by inhibiting signal transduction of the IL-6-receptor (gp130). Expression of SOCS-3 is dependent on activation of MAP-kinases and IL-10 (64,111). Also, activation of p38 correlates with the activation of SOCS-3 (112). Activation of MAP-kinases induces a negative feedback loop via SOCS-3, which inhibits activation of STAT1, STAT3 and gp130 (63,64,112) (Figure 1).
Expression and production of chemokines, selectins and adhesion molecules stimulates cellular influx
The MAP-kinase-mediated inflammatory response in the kidney stimulates production and activation of chemokines, selectins and adhesion molecules, which is followed by cellular influx. Activation of ICAM-1 and VCAM-1 is increased in interstitium and glomeruli after brain death in rats (9). Expression of E-selectin (11), P-selectin (8) and the chemokines MCP-1 (8,11,75), macrophage inflammatory protein 2 (MIP2) (8), IL-8 (8) and Chemokine (C-X-C motif) ligand-10 (CXCL10 or IP-10) (4,8) are all upregulated in the kidney after brain death. Activation of E-/P-selectin (110,113), ICAM-1 and VCAM-1 (mainly on proximal tubular epithelial cells) (113) was confirmed in human kidney samples using immunohistochemistry. An increased number of CD4-positive and CD8-positive cells, polymorphonuclear cells, and circulating (ED-1) and resident macrophages (ED-2) are found in glomeruli and interstitium after brain death (9,114). Activation of adhesion molecules and selectins, and production of chemokines after brain death leads to subsequent leukocyte infiltration in the kidney.
MAP-kinases as targets for early intervention
The inflammatory response induced by brain death contributes to organ injury, which affects graft survival negatively. Elevated levels of cytokines in serum are found after induction of brain death, followed by NF-κB-activation in kidneys and subsequent expression of genes involved in inflammation. The induced inflammatory response leads to influx of leukocytes in the kidney, which further deteriorates the quality of the potential donor organ. MAP-kinases probably play a major role in the transition from the systemic inflammatory signal to the local inflammatory reaction in the kidney. These signal transduction pathways can be activated by the hormonal, hemodynamic and inflammatory systemic alterations after brain death. The already preexisting organ injury is then aggravated by cold ischemia (preservation) and during the transplantation procedure (ischemia/reperfusion injury).
Inhibitors of signal transduction pathways are very promising therapeutics to reduce the inflammatory response. Preventing activation of p38 and JNK could be beneficial for organ transplantation, since these two kinases are important regulators in the inflammatory response. Studies in healthy human volunteers showed a decreased leukocyte activation (115), E-/L-selectin (115,116) and ICAM-1 activation (115) and a reduction of TNF-α (117–119) (120), IL-1β (120), IL-6 (117–119), IL-8 (118,119), IL-10 (117), IL-12 (120) and C-reactive-protein (CRP) (117) after treatment with RWJ-67657 or BIRB 796 BS (inhibitors of p38) on stimulation with endotoxin in vivo or on isolated leukocytes ex vivo. Other types of MAP-kinase inhibitors have been tested for their safety in phase-I clinical trials in patients with different types of cancer (121,122). Randomized-controlled trials in patients with Crohn's disease demonstrated clinical improvements with the use of CNI-1493 (inhibitor of JNK and p38) (123), but not with BIRB 796 (inhibitor of p38) (124). Treatment of the brain-dead organ donor with specific MAP-kinase inhibitors could influence the inflammatory response and prevent further organ injury. In addition, inhibiting MAP-kinases will also limit the damage induced by ischemia/reperfusion (125,126).
A different approach to limit or delay inflammatory reactions may be found by observing hibernating rodents. Hibernating rodents are able to ‘delay’ the inflammatory response to injection of endotoxin during their cold, sleeping phase until the next arousal when their immune system is reactivated (127). Hibernating animals lower their metabolism until their body temperature almost equals the environmental temperature and are able to survive many cycles of cooling and rewarming (128–130). D-Ala2-Leu5-enkephalin (DADLE, a delta opioid agonist) is synthesized according to ‘hibernating-inducing trigger’ (HIT), which can be found in blood of hibernating animals when they are about to enter their sleeping phase. In a rabbit model of cardiac transplantation, DADLE activated Akt and ERK and protected against reperfusion injury (measured in terms of functional recovery) (131). Furthermore, DADLE is able to inhibit the mRNA expressions of p53 and c-Fos and it promotes cell survival (132,133). DADLE causes a ‘hibernating’ state with inhibition of transcription by reversibly binding to perichromatin fibrils and dense fibrillar components, where transcription and early splicing of pre-mRNAs and pre-rRNAs occur (134). Another study used a Langendorff's setup for the rat heart demonstrated a definite role for delta-opioid receptors in improving functional recovery after ischemia. Measured effects were similar to that conferred by classic ischemic preconditioning and could be reversed by naloxone, which suggests that these effects involve signaling through opioid receptors (135). Thus, addition of DADLE to the preservation solution could induce a hypometabolic state in the donor organ with minimal gene expression and further protection against ischemia/reperfusion injury.
Further research is required to clarify the role of MAP-kinases in brain death-induced organ injury. Besides studying gene expression and protein production, studying kinomics could provide more insights in signal transduction patterns involved in the induction of organ injury after brain death. Inhibiting transcription of genes involved in inflammatory responses by interfering with signal transduction pathways like MAP-kinases may be very promising drug targets to improve organ quality from deceased brain-dead donors in organ transplantation.