- Email: [email protected]
New paradigms in cell death in human diabetic nephropathy
mini review
http://www.kidney-international.org & 2010 International Society of Nephrology
New paradigms in cell death in human diabetic nephropathy Maria-Dolores Sanchez-Nin˜o1, Alberto Benito-Martin1 and Alberto Ortiz1,2,3 1 3
Department of Nefrologia, IIS Fundacion Jimenez Diaz, Madrid, Spain; 2Fundacio´n Renal In˜igo Alvarez de Toledo, Madrid, Spain and Universidad Autonoma de Madrid, Madrid, Spain
Cell death is thought to contribute to progressive renal cell depletion in diabetic nephropathy. Unbiased gene expression profiling identified novel cell death molecules in human diabetic nephropathy. The expression of TNF-related apoptosis-inducing ligand (TRAIL), its decoy receptor osteoprotegerin, and receptors Fas (a Fas ligand receptor) and CD74 (a migration inhibitory factor (MIF) receptor) were induced in human diabetic nephropathy. Cell culture studies supported the functional relevance of this observation and the relationship to a high glucose environment. To define novel proapoptotic proteins upregulated in diabetic nephropathy, functional genomic screens for novel apoptosis mediators were integrated with genome-wide expression profiling and identified candidates for further functional analysis, including brain acid-soluble protein 1 (BASP1). Several lines of evidence point toward induction of endoplasmic reticulum stress response in human diabetic nephropathy. Functional studies defining an unequivocal contribution of endoplasmic reticulum stress to cell death in this setting are still needed. Further comparative studies will be required to define whether there is a specific aspect of apoptosis in progressive human diabetic nephropathy or whether the mechanisms are shared among all patients with chronic kidney disease. The next challenge will be to define the consequence of therapeutic interference of the apoptosis pathways in diabetic nephropathy and chronic kidney disease. Kidney International (2010) 78, 737–744; doi:10.1038/ki.2010.270; published online 11 August 2010 KEYWORDS: apoptosis; BASP1; CD74; diabetic nephropathy; MIF; TRAIL
Correspondence: Alberto Ortiz, Department of Nefrologia, IIS Fundacion Jimenez Diaz, Avenida Reyes Catolicos 2, 28040 Madrid, Spain. E-mail: [email protected] Received 15 February 2010; revised 23 June 2010; accepted 6 July 2010; published online 11 August 2010 Kidney International (2010) 78, 737–744
MECHANISMS FOR CELL LOSS IN DIABETIC NEPHROPATHY
Diabetic nephropathy is characterized by proteinuria and progressive loss of renal function, leading to end-stage renal disease requiring renal replacement therapy. The pathological substrate of diabetic nephropathy is a progressive loss of parenchymal renal cells and extracellular matrix accumulation. Recently, it has become clear that inflammation also contributes to diabetic nephropathy.1 Unbalanced apoptotic cell death leads to renal cell loss and has been observed in podocytes, tubular cells, and endothelial cells in experimental and human diabetic nephropathy.2–8 Diabetic nephropathy is the most frequent cause of end-stage renal disease in developed countries. Only a comprehensive understanding of the molecular basis for renal injury initiation and progression will allow the rational design of novel, successful therapeutic strategies. In recent years, high-throughput techniques exploring gene expression have uncovered novel networked mediators that contribute to renal cell death and may be involved in human diabetic nephropathy.4,9–12 MOLECULAR BASIS FOR APOPTOTIC CELL DEATH
Cell death is usually a response to the cell microenvironment.13 Among the different forms of cell death, apoptosis is molecularly best characterized and may be manipulated therapeutically. Cell survival depends on the interplay between extracellular survival factors that activate intracellular survival pathways and lethal factors that activate cell death pathways leading to apoptosis (Figure 1). A key role for radical oxygen species in cell death and diabetic nephropathy was recently reviewed.14 The two main pathways for apoptosis are ligation of plasma membrane death receptors (‘extrinsic’ pathway) and perturbation of the intracellular homeostasis (‘intrinsic’ pathway). In the extrinsic pathway, ligation of death receptors promotes the assembly of multimolecular complexes that include adaptor proteins and initiator caspases-8 and -10. The intrinsic pathway involves injury to intracellular organelles, the most important being the mitochondria. Sentinel activator ‘BH3-only proteins’ triggers the allosteric activation of Bax and/or Bak, which oligomerize at the mitochondria, inducing permeabilization of the outer mitochondrial membrane and release of proapoptotic factors such as cytochrome c, SMAC/DIABLO and AIF, which 737
mini review
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
MIF Lethal cytokines
Cell stress
TRAIL
CD74 BASP1?
Death receptor activation
Mitochondria
ER stress
TRAILR1/R2 FasL
FAS CD36
HSPA5 HYOU1 XBP1
Caspase activation Mitochondrial injury
Caspase activation
Apoptotic cell death Cell membrane External survival factors OPG DCR3 VEGF EGF
Figure 1 | Summary of apoptotic pathways and location of some key players identified by unbiased gene expression profiling in human diabetic nephropathy (DN). Genes upregulated in human DN include CD74, tumor necrosis factor-related apoptosis-inducing ligand (TRAIL), FAS, brain acid-soluble protein 1 (BASP1), CD36, osteoprotegerin (OPG), heat-shock 70 kDa protein 5 (HSPA5), hypoxia upregulated 1 (HYOU1), and X-box-binding protein 1 (XBP1) and downregulated genes include vascular endothelial growth factor (VEGF), epidermal growth factor (EGF), and DCR3. CD74 activation by migration inhibitory factor (MIF) and other inflammatory stimuli mediate upregulation of the lethal cytokine TRAIL. TRAIL activates the lethal receptors TRAIL receptor 1 (TRAIL-R1) and TRAIL-R2 to induce cell death. Osteoprotegerin is a decoy receptor for TRAIL. The lethal receptor FAS is also increased in human DN. FAS mediates apoptosis on ligation by Fas ligand (FasL). Expression of the FasL decoy receptor DCR3 was downregulated. In human DN the expression of the protective endoplasmic reticulum stress genes HSPA5/GRP78/BiP, HYOU1/ORP150, and XBP1 was upregulated. The precise location of BASP1 action is incompletely characterized, but inhibition of BASP1-induced apoptosis by Bax antagonists and caspase inhibitors suggests a role upstream of mitochondrial injury. CD36 was identified by unbiased gene expression profiling as a downregulated gene in murine DN, but was found increased in human DN. Expression of the extracellular survival factors VEGF and EGF was downregulated in human DN tubular compartment.
promote caspase-dependent and -independent apoptosis. Cytochrome c facilitates the oligomerization of Apaf-1 and caspase-9 in the apoptosome multimolecular complex. Initiator caspase oligomerization activates their enzymatic activity and cleavage of protein substrates, thus activating effector caspases such as caspase-3 and –7, and leading to widespread proteolysis and commitment to cell death. Signaling crosstalk between the intrinsic and extrinsic pathways involves cleavage of proapoptotic BH3-only protein Bid by caspase-8. The active fragment, tBid, translocates to mitochondria and recruits the intrinsic pathway. Bcl2-related antiapoptotic proteins, such as Bcl2 and BclxL, bind and sequester activator BH3-only proteins, Bax and Bak, and may directly inhibit Apaf-1-mediated activation of caspase-9. Sensitizer BH3-only molecules competitively displace activator BH3-only proteins from the Bcl2 pocket. Inhibitors of apoptosis proteins directly inhibit caspases and are, in turn, inhibited by SMAC/DIABLO. Endoplasmic reticulum stress leads to accumulation of unfolded proteins in the endoplasmic reticulum and triggers 738
the unfolded protein response, which strives to reestablish homeostasis.15 Thus, the unfolded protein response is primarily an adaptive response. In resting cells, the endoplasmic reticulum chaperone HSPA5/GRP78/BiP (heat-shock 70 kDa protein 5/glucose-regulated protein 78 kDa) binds to endoplasmic reticulum stress sensors (IRE1, PERK, and ATF6) and keeps them inactive. On endoplasmic reticulum stress, HSPA5/GRP78/BiP is recruited by unfolded proteins, thus releasing and activating PERK, IRE1, and ATF6. PERK activation leads to translation inhibition. IRE1 initiates the nonconventional splicing of Xbp-1 mRNA, encoding the transcription activator XBP1, that collaborates with active ATF6 to promote transcription of endoplasmic reticulum chaperones, including HSPA5/ GRP78/BiP. However, under conditions of severe and prolonged endoplasmic reticulum stress, failure to restore homeostasis is followed by cell death, usually occurring by apoptosis.12,13,15,16 There are three main pathways that may contribute to endoplasmic reticulum stress-induced apoptosis: (a) the CHOP/growth arrest and DNA-damage-inducible Kidney International (2010) 78, 737–744
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
(GADD153) transcription factor, which is mainly induced via PERK, downregulates Bcl2; (b) IRE1-mediated activation of apoptosis signal-regulating kinase 1 (ASK1)/c-Jun NH2terminal kinase (JNK); and (c) activation of the endoplasmic reticulum-localized caspase-12. The three pathways are thought to activate the caspase cascade and may recruit the mitochondrial pathway. However, although caspase-12 processing is widely recognized as a marker of endoplasmic reticulum stress, its catalytic activity is limited to selfprocessing and its role in cell death has been questioned.17 In the context of diabetic nephropathy hyperglycemia may induce or facilitate apoptosis.2,4,10,18 There are no specific pathways for high-glucose-induced cell death. Rather, high glucose promotes the upregulation or activation of several molecules in the different proapoptotic pathways. Early reports disclosed upregulation of proapoptotic molecules such as Bax and downregulation of antiapoptotic molecules (Bcl2 and BclxL) by high glucose in the renal cells.2 However, high glucose regulation of cell death-related genes is more complex than initially thought. In addition, hypoxia, oxidative stress, inflammatory cytokines, the renin–angiotensin–aldosterone system, advanced glycosylation end products, and glucose degradation products may also contribute to cell death in diabetic nephropathy.8–10,19–22 CELL DEATH GENE EXPRESSION IN HUMAN DIABETIC NEPHROPATHY
The European Renal cDNA Bank has characterized the transcriptomic fingerprint of human diabetic nephropathy, that is, the level of expression of multiple mRNAs.11 Cell death-related genes were identified through several approaches4,9–12 (Table 1). First, the gene ontology term ‘Cell death’ identified 66 significantly upregulated and 46 significantly downregulated genes.9 Overall, 25% of apoptosis-related genes were differentially regulated in human diabetic nephropathy.9 Differentially expressed apoptosisrelated genes included those coding for cytokines, receptors, intracellular lethal proteins and caspases. Caspase-1 and caspase-4 were upregulated. These caspases are now widely recognized to have mainly a proinflammatory role. In a second approach, differentially expressed endoplasmic reticulum stress genes were characterized.12 Finally, novel cell death genes upregulated in diabetic nephropathy were identified by combining the database with a functional genomics screen for cDNAs that induce apoptosis when overexpressed in cultured cells: 12 of the 138 full-length cDNAs inducing cell death when overexpressed in human embryonic kidney cells also showed mRNA transcript induction in human diabetic nephropathy biopsies.4 DEATH LIGANDS AND RECEPTORS IN DIABETIC NEPHROPATHY
Cytokines belonging to the tumor necrosis factor (TNF) superfamily (TNFSF) activate a superfamily of receptors (TNFRSF) and are key regulators of cell survival, inflammation, and fibrosis. TNF/TNFSF2 had long been known to contribute to diabetic nephropathy.1 Fas (TNFSFR6), the Kidney International (2010) 78, 737–744
mini review
Table 1 | Cell death genes with differential mRNA expression in human DN4,9,12,a Cell death (gene ontology) genes Upregulated Osteoprotegerin TRAIL: TNF-related apoptosis-inducing ligand TNFAIP8: Tumor necrosis factor-a-induced protein 8 14–3-3, Z STAT1: Signal transducer and activator of transcription 1 Fas APP: Amyloid precursor protein Tubulin-b Osteopontin Caspase 1 CD74 EIF4G2: Eukaryotic translation initiation factor 4 g-2 NRIF3: Neurotrophin receptor interacting factor 3 TIA1: T-cell intracellular antigen 1 LOXL2: Lysyl oxidase-like 2 Clusterin DOCK1: Dedicator of cytokinesis 1 EMP3: Epithelial membrane protein 3 Downregulated GADD45A: Growth arrest and DNA-damage-inducible-45-a Apolipoprotein E Sphingosine kinase 2 Prostaglandin E receptor 3 MAPT: Microtubule-associated protein-t PKC, z: Protein kinase C z Upregulated endoplasmic reticulum stress genes HSPA5/GRP78/BiP: Heat-shock 70 kDa protein 5/glucose-regulated protein 78 kDa HYOU1/ORP150: Hypoxia upregulated 1/oxygen-regulated protein 150 XBP1: X-box-binding protein 1 Genes upregulated in DN that were identified by a functional genomics screening as inducing cell death when overexpressed in cultured cells EMP3: Epithelial membrane protein-3 BASP1: Brain acid-soluble protein 1 PLP2: Proteolipid protein 2 (colonic epithelium enriched) CLDN4: Claudin 4 ITGB2: Integrin, b 2 (complement component 3 receptor 3 and 4 subunit) Abbreviation: DN, diabetic nephropathy. a Only genes with at least a 1.5-fold difference in expression between DN and control biopsies are listed.
receptor for Fas ligand (FasL/TNFSF6), is upregulated in diabetic nephropathy and the proapoptotic role of FasL/Fas in the kidney, including in diabetic nephropathy, has been extensively studied.3,23 Unbiased gene expression profiling confirmed the increase Fas expression in diabetic nephropathy and disclosed decreased expression of the decoy FasL receptor DCR3 (TNFSFR6b) mRNA.9 Increased Fas and FasL protein expression was confirmed by immunohistochemistry in human diabetic nephropathy.6 However, the two most upregulated cell death mRNA transcripts in human diabetic nephropathy were TNF-related apoptosis-inducing ligand (TRAIL/TNFSF10) and its decoy receptor osteoprotegerin (TNFRSF11B).9 TRAIL is a type II transmembrane protein with a molecular mass of 33–35 kDa. Membrane-bound TRAIL can be cleaved from the cell surface to form a soluble trimeric 739
mini review
protein that retains the proapoptotic activity. TRAIL binds to a complex system of four membrane-bound and one soluble receptor.24 TRAIL receptor 1 (TRAIL-R1/TNFRSF10A) and TRAIL-R2 (TNFSFR10B) have a functional death domain that allows caspase activation and apoptosis induction. In contrast, TRAIL-R3 (TNFRSF10C), TRAIL-R4 (TNFRSF10D), and osteoprotegerin do not transduce cell death signals and osteoprotegerin behaves as a decoy receptor. Depending on the relative levels of the different receptors and on the cell type, TRAIL can induce cell death, and also survival, proliferation, and maturation. Exogenous TRAIL is more lethal to cultured cancer cells than to normal cells, and TRAIL-R agonists are in clinical trials as anticancer agents.25 Many normal human tissues including the kidney express low levels of TRAIL, TRAIL-R1, and -R2, which is consistent with the low cytotoxicity of TRAIL to most healthy tissues. However, membrane-bound TRAIL has a higher lethal potential than soluble TRAIL, and data from knockout mice suggest that endogenous TRAIL may induce apoptosis in nontumoral inflamed liver.26 In this regard, hepatotoxicity and nephrotoxicity were observed in preclinical studies with TRAIL agonists.27 TRAIL also has immunoregulatory functions that may be beneficial for the onset of autoimmune diabetes.28 Osteoprotegerin is a secreted protein with no transmembrane domain, which is produced as a monomer (55–62 kDa), undergoes homodimerization, and is secreted as a disulfide-linked homodimeric glycoprotein of 110–120 kDa.29 Osteoprotegerin is a soluble decoy receptor for both TRAIL and receptor activator of nuclear factor-kB ligand (RANKL/TNFSF11) and may have independent functions promoting endothelial cell survival and angiogenesis. Potential cross-regulatory mechanisms involving the balance TRAIL–osteoprotegerin–RANKL may exist. Although the affinity of TRAIL for osteoprotegerin is weaker than for the lethal TRAIL receptors, there is evidence for the biological relevance of the interaction.30 On a tissue level, osteoprotegerin regulates bone mineral density and vascular calcification, and serum osteoprotegerin levels are increased in patients with renal dysfunction, including diabetic nephropathy, and are associated with adverse cardiovascular outcomes.31 In the normal kidney, TRAIL is expressed in tubules but not in glomerular cells.9 In diabetic nephropathy, TRAIL and osteoprotegerin mRNA expression is increased and correlates to clinical and histological severity of renal disease.9 Tubular epithelium is the predominant nephron cell lineage expressing TRAIL protein, but de novo protein expression in podocytes was also noted in diabetic nephropathy.9,10 High glucose did not modulate TRAIL expression in cultured renal cells. However, proinflammatory cytokines such as interferon-g, TNFa, and migration inhibitory factor (MIF) increased TRAIL expression in cultured renal tubular cells and podocytes.9,10 TRAIL promoted a modest increase in tubular cell and podocyte death, which is consistent with a role in the chronic progression of diabetic nephropathy seen 740
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
humans. Nuclear factor-kB activation by TRAIL protected cells from TRAIL lethality.9 In this regard, both TRAIL and TNF activate simultaneous death and survival signals, and this may underlie the low sensitivity of non-neoplastic cells to both cytokines unless the microenvironment is unfavorable for survival. Blockade of nuclear factor-kB-mediated survival signals or additional environmental death signals, such as high glucose or inflammation, increase the rate of cell death.9,10 Osteoprotegerin, even endogenous osteoprotegerin secreted by tubular cells, protected from TRAIL-induced cell death.9 The end result of the increased expression of TRAIL and osteoprotegerin in a particular patient may depend on the relative balance between them. We hypothesize that a misbalance toward TRAIL may result in tissue injury. If osteoprotegerin prevails, the cells would be protected from TRAIL. Podocyte overexpression of the macrophage MIF cytokine results in podocyte injury and proteinuria.32 MIF was increased in experimental diabetic nephropathy, and serum MIF was elevated in patients with diabetic nephropathy (reviewed in Sanchez-Nino et al.10). The protein expression of the MIF receptor CD74 is increased in tubular cells and podocytes in human diabetic nephropathy, and CD74 mRNA expression is increased in diabetic nephropathy.10 MIF did not promote renal cell apoptosis. However, MIF engagement of CD74 activated MAP kinases and increased TRAIL expression.10 CD74 may have MIF-independent functions, as it is a coreceptor for CXCR4 or CD44.33 These functions have not been fully characterized in renal cells. Microarray-based gene expression profiling on whole kidney RNA, together with supervised clustering methods, identified CD36 as a downregulated mRNA in diabetic mice with albuminuria.8 CD36 protein was absent from proximal cells, both in control and diabetic mice. In contrast, CD36 protein was upregulated and colocalized with apoptotic proximal tubular cells in human diabetic nephropathy. This is one of several examples of dissociation in the molecular signature of human and experimental diabetic nephropathy. CD36 is a class B scavenger receptor. Cell culture studies disclosed that high glucose upregulated CD36 mRNA and protein, and engagement by glycated albumin transduced a lethal signal to human proximal tubular cells. Cell death was prevented by caspase-8, but not by caspase-9 inhibition, suggesting a role for death receptors. However, the molecular pathways were not further characterized. MITOCHONDRIAL STRESS IN DIABETIC NEPHROPATHY
Mitochondrial stress has long been known to contribute to high-glucose-induced apoptosis. Increased Bax expression was the first apoptosis-related molecular change observed in kidney cells in response to high glucose, and Bax knockout mice are protected from high-glucose-induced cell death.2,18 More recently, a novel proapoptotic role has been identified for brain acid-soluble protein 1/cortical cytoskeleton-associated protein of 23 kD (BASP1/CAP-23).4 BASP1 is a multifunctional protein involved in cytoskeletal and lipid Kidney International (2010) 78, 737–744
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
raft dynamics, as well as in the nuclear regulation of transcription. BASP1 expression is mainly cytoplasmic in some cell types including tubular cells, and nuclear in others, such as podocytes.4,34 In the adult kidney podocyte, BASP1 is a transcriptional cosuppressor of WT1.34 Kidney BASP1 was upregulated at the mRNA and protein level in human and experimental diabetic nephropathy.4 Tubular cells were the main site of BASP1 upregulation, where it localized to apoptotic cells.4 In cultured human tubular epithelial cells, lethal stimuli such as serum deprivation, high glucose, or proinflammatory cytokines increased BASP1 mRNA and protein.4 Overexpression of BASP1 was sufficient to induce apoptosis in tubular cells. In addition, BASP1 knockdown protected tubular cells from apoptosis induced by serum deprivation or high glucose concentrations, but not from cytokine-induced apoptosis.4 Thus, BASP1 may be critical for cell stress-induced apoptotic cell death. BASP1 is upregulated in other forms of renal injury, such as hypertensive nephropathy, but shows the most robust induction in diabetic nephropathy.4 The precise role of BASP1 in the apoptotic pathway is unclear. Interestingly, BASP1 colocalizes with the apoptotic actin ring and tubulin in the cell cortex, suggesting that the cytoskeleton-binding properties of BASP1 may be associated with its lethal action.4 Cytoskeletal remodeling is a hallmark of apoptosis, and changes in actin dynamics contribute to mitochondrial injury and cell death.35 Prevention of BASP1-induced apoptosis by Bax antagonists and caspase inhibitors suggests that BASP1 has a role upstream of mitochondria or caspase activation, or that BASP1 contributes to a positive feed-back loop, leading to further mitochondrial injury and caspase activation.4 BASP1/ CAP-23, GAP-43 and MARCKS belong to a protein family characterized by the presence of an effector domain, association with the cortical cytoskeleton, and regulation of actin-binding proteins regulated by reversible phosphorylation of the effector domain by protein kinase C.36 Interestingly, different isoforms of protein kinase C have been involved in diabetes-associated apoptosis.37 Deprivation of survival factors is a known inducer of apoptosis that cooperates with the presence of lethal factors such as cytokines and toxins to promote renal cell death.13 In human diabetic nephropathy, the expression of two cytokines with survival activity of renal epithelial cells, vascular endothelial growth factor-A and epidermal growth factor, was reduced at the mRNA and protein level in the tubular compartment.13,38 In glomerular endothelium, receptorbound vascular endothelial growth factor protein was decreased in more advanced lesions.7 The reduced vascular endothelial growth factor-A expression contrasted to previously reported findings in experimental models. Reduced renal epidermal growth factor mRNA and had been previously observed in animal models.39 ENDOPLASMIC RETICULUM STRESS IN DIABETIC NEPHROPATHY
Stimuli that increase the demand on the endoplasmic reticulum to synthesize proteins or degrade improperly Kidney International (2010) 78, 737–744
mini review
folded proteins cause endoplasmic reticulum stress.40 High glucose, free fatty acids, albumin, nephrotoxins, oxidative stress, or inflammation induce endoplasmic reticulum stress in renal cells.12,16,41 In human progressive diabetic nephropathy, expression of mRNA for protective endoplasmic reticulum chaperone proteins HSPA5/GRP78/BiP and HYOU1/ORP150 (hypoxia upregulated 1/oxygen-regulated protein 150) and for the prosurvival transcription factor XBP1 (X-box-binding protein 1) was increased.12 Increased HSPA5/GRP78/BiP and HYOU1/ORP150 protein expression was confirmed at the protein level. However, potentially lethal unfolded protein response mRNAs, such as CHOP, were unchanged or repressed.12 The finding of increased mRNA levels for genes involved in the protective responses of the unfolded protein response, but not for proapoptotic genes, suggests that the endoplasmic reticulum stress in human diabetic nephropathy might be adaptive and, per se, does not support a role of endoplasmic reticulum stress in progression of diabetic nephropathy.40 In this regard, induction of some potentially lethal endoplasmic reticulum stress proteins may not be followed by completion of the endoplasmic reticulum stress pathway.16,41 Thus, cyclosporin A increased CHOP/GADD153 protein expression in tubular cells, but failed to result in caspase-12 processing, indicating that CHOP upregulation may be induced by nonendoplasmic reticulum proapoptotic stressors. However, it is conceivable that interference with adaptive mechanisms, because of the influence of genetic background or additional insults, may result in endoplasmic reticulum-mediated cell death in human diabetic nephropathy. In animal models there is evidence that endoplasmic reticulum stress-induced apoptosis may cause tubular cell injury (reviewed in Sanz et al.13). Thus, the endoplasmic reticulum stressor tunicamycin induced severe histological tubular injury, which was decreased both in CHOP/ GADD153 and caspase-12 knockout mice.42,43 Furthermore, there is evidence that impairing the adaptive endoplasmic reticulum response, as is the case for HSPA5/GRP78/BiPmutant heterozygous mice, results in more severe tubulointerstitial injury in response to albuminuria.44 However, there is no functional evidence yet that endoplasmic reticulum stress, in the presence of an intact adaptive endoplasmic reticulum stress response, contributes to renal cell apoptosis in vivo in animal models of proteinuric kidney disease.45 This is in part due to technical difficulties, given the relative paucity of adequate mouse models. To complicate matters further, significant differences in the endoplasmic reticulum stress response between human and experimental diabetic nephropathy preclude direct extrapolation of animal model results to the human situation. As in humans, increased expression of adaptive endoplasmic reticulum stress proteins, such as HSPA5/GRP78/BiP, has been observed in murine and rat models.46,47 Although early studies did not find increased caspase-12 in db/db mice,48 a recent transcriptomics-based approach identified caspase-12 as an upregulated mRNA in this model.47 Caspase-12 protein was processed and increased 741
mini review
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
expression was noted in apoptotic tubular cells in vivo. Caspase-12 was also increased in rat streptozotocin-induced diabetic nephropathy.46 Furthermore, albumin induced caspase-12-dependent apoptosis in cultured human HK2 tubular cells. However, most humans carry a single-nucleotide premature stop codon polymorphism that inactivates caspase-12.49 In this regard, although HK2 cells carry the inactive caspase-12 allele, they expressed a full-length protein, probably as a result of translational readthrough induced by aminoglycosides in the cell culture media. Thus, the relevance of these findings to human diabetic nephropathy is unclear, with the exception of the 20% of the population of African descent that expresses a functional protein. In contrast to humans, pro-death CHOP/GADD153 mRNA and protein was increased in animal models of diabetic nephropathy.46,47 SPECIFICITY OF THE OBSERVED CHANGES IN GENE EXPRESSION FOR HUMAN DIABETIC NEPHROPATHY
Exposure to a high glucose environment for a few hours will increase CD74 and BASP1, but not TRAIL expression, in cultured tubular cells. At present, it is not possible to differentiate whether these changes are a consequence of diabetes, nephropathy, or diabetic nephropathy in humans. However, there is a progressive gradient for magnitude of changes in the expression of these genes between early and late diabetic nephropathy.9 The fact that inflammatory cytokines can induce the expression of these genes in cultured cells suggests a role beyond diabetic nephropathy and indicates that, once a threshold of kidney injury is reached, these genes might be important for progression of
diabetic nephropathy even when an adequate glycemic control has been achieved. In any case, there appears to be a specific interaction between high glucose concentrations and sensitivity to TRAIL or BASP1-induced death.4,9 There is a limited understanding of the range of human nephropathies that present with changes in the expression of genes such as TRAIL, CD74, or BASP1. Data from our group and a search of the Nephromine database (www.nephromine. org) indicate that some of these changes are present in other proteinuric progressive nephropathies, but not present or milder in nonprogressive proteinuric kidney injury. BASP1 mRNA expression is increased in membranous nephropathy and age-related chronic kidney injury, and a trend was observed for hypertensive nephropathy. BASP1 did not change in IgA nephropathy or minimal change disease.4,50 Upregulation of TRAIL and osteoprotegerin mRNA in age-related chronic kidney injury and minimal change disease was significant but milder than in human diabetic nephropathy.11,50 Fas mRNA was increased in age-related chronic kidney injury, but not in minimal change disease.11,50 CD74 mRNA was increased in hypertensive nephropathy and age-related chronic kidney injury, but not in minimal change disease.4,50 In contrast, evidence for endoplasmic reticulum stress, including increased expression of CHOP/GADD153 (not found upregulated in human diabetic nephropathy) and HSPA5/GRP78/BiP, has been observed in a variety of proteinuric nephropathies.45 Proximal tubular CD36 expression was not observed in focal segmental glomerulosclerosis, but other human diseases were not tested.8 There are additional ligands for CD36, not
Hyperglycemia Increase Increase
Increase Sensitize Basp1
Inflammation Increase TRAIL (MIF) CD74 TNF
Sensitize
Low VEGF, EGF
Intracellular events
Parenchymal renal cell apoptosis
Progression of kidney disease
OPG
Figure 2 | Hypothetical contribution of novel cell death pathways to human diabetic nephropathy (DN): integration of inflammation and high glucose. Renal cell loss by apoptosis is thought to contribute to the progression of DN. Both inflammation and high glucose are known inducers of apoptosis. Renal cell culture studies suggest that inflammation contributes to increased tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) and Fas expression. Thus, migration inhibitory factor (MIF) activation of CD74 or tumor necrosis factor (TNF) increased TRAIL expression and TNF increases Fas expression. In contrast hyperglycemia increased CD74 expression, increased brain acid-soluble protein 1 (BASP1) expression, and also sensitized to the lethal effect of TRAIL on renal cells. Osteoprotegerin (OPG) antagonized the lethal effect of TRAIL in cultured cells. BASP1 had a role in hyperglycemia- and growth factor deprivation-induced apoptosis. Low levels of the survival factors vascular endothelial growth factor (VEGF) and endothelial growth factor (EGF) were observed in human DN and may also contribute to apoptosis sensitization. 742
Kidney International (2010) 78, 737–744
mini review
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
specific for diabetes, such as thrombospondin-1, that also promote tubular cell apoptosis.8 INTEGRATIVE VIEW AND FUTURE PERSPECTIVES
Recent evidence gathered from the direct study of gene expression patterns in human diabetic nephropathy coupled with confirmation at the protein expression level have unraveled a set of new molecules and pathways, and novel information on known players in kidney epithelial cell death. Thus, the interplay among high glucose, inflammation, and lack of survival factors may contribute to activation of death receptor, mitochondrial injury, and endoplasmic reticulum stress pathways that may ultimately lead to renal cell loss (Figure 2). Within each of these pathways, we find a balance of intracellular lethal and adaptive responses. Thus, potentially lethal cytokines such as TNF and TRAIL also activate prosurvival nuclear factor-kB, and current data indicate an adaptive pattern of endoplasmic reticulum stress response genes observed in human diabetic nephropathy. Additionally, increased expression of some decoy receptors has been observed. The net end result observed in vivo is progressive loss of renal parenchymal cells in a significant proportion of patients. The molecular networks deduced from these studies provide a paradigm for understanding cell death regulation in the course of diabetic nephropathy. A high glucose level appears to have a role as sensitizer to cell death. Subsequent features of diabetic nephropathy, such as the presence of albuminuria and renal inflammation, are key additional triggers of cell death and may drive the process even if adequate control of blood glucose levels is reestablished, providing a strong rational for antiapoptotic or antiinflammatory strategies in diabetic nephropathy. Different to previous approaches, recent progress in identification of molecules of potential pathogenic importance has been based on the analysis of human diabetic nephropathy. Changes in the expression of most, but not all, newly identified genes are also observed in animal models of diabetic nephropathy, setting the stage for preclinical intervention on some molecules. These studies will clarify whether any of the reported gene changes is a legitimate therapeutic target. It is conceivable that, in the future, multipronged approaches to the management of diabetic nephropathy may include strategies to prolong renal cell survival. DISCLOSURE
All the authors declared no competing interests. ACKNOWLEDGMENTS
This work was supported by grants from FIS 06/0046, PS09/00447, ISCIII-RETICS REDINREN RD 06/0016, MEC (SAF 03/884), and Sociedad Espanola de Nefrologia. ABM was supported by FIS, MDSN by MEC, and AO by the Programa de Intensificacio´n de la Actividad Investigadora in the Sistema Nacional de Salud of the Instituto de Salud Carlos III, the Agencia ‘Pedro Lain Entralgo’ of the Comunidad de Madrid, and CIFRA S-BIO 0283/2006. We appreciate the critical review, thoughtful comments, and contribution from Matthias Kretzler, University of Michigan. Kidney International (2010) 78, 737–744
REFERENCES 1.
2.
3.
4. 5. 6.
7.
8.
9. 10. 11.
12.
13.
14.
15.
16. 17.
18.
19.
20.
21.
22.
23. 24. 25.
26. 27.
Navarro JF, Mora-Fernandez C. The role of TNF-alpha in diabetic nephropathy: pathogenic and therapeutic implications. Cytokine Growth Factor Rev 2006; 17: 441–450. Ortiz A, Ziyadeh FN, Neilson EG. Expression of apoptosis-regulatory genes in renal proximal tubular epithelial cells exposed to high ambient glucose and in diabetic kidneys. J Investig Med 1997; 45: 50–56. Kelly DJ, Stein-Oakley A, Zhang Y et al. Fas-induced apoptosis is a feature of progressive diabetic nephropathy in transgenic (mRen-2)27 rats: attenuation with renin-angiotensin blockade. Nephrology (Carlton) 2004; 9: 7–13. Sanchez-Nino MD, Sanz AB, Lorz C et al. BASP1 promotes apoptosis in diabetic nephropathy. J Am Soc Nephrol 2010; 21: 610–621. Gilbert RE, Marsden PA. Activated protein C and diabetic nephropathy. N Engl J Med 2008; 358: 1628–1630. Verzola D, Gandolfo MT, Ferrario F et al. Apoptosis in the kidneys of patients with type II diabetic nephropathy. Kidney Int 2007; 72: 1262–1272. Hohenstein B, Hausknecht B, Boehmer K et al. Local VEGF activity but not VEGF expression is tightly regulated during diabetic nephropathy in man. Kidney Int 2006; 69: 1654–1661. Susztak K, Ciccone E, McCue P et al. Multiple metabolic hits converge on CD36 as novel mediator of tubular epithelial apoptosis in diabetic nephropathy. PLoS Med 2005; 2: e45. Lorz C, Benito-Martin A, Boucherot A et al. The death ligand TRAIL in diabetic nephropathy. J Am Soc Nephrol 2008; 19: 904–914. Sanchez-Nino MD, Sanz AB, Ihalmo P et al. The MIF receptor CD74 in diabetic podocyte injury. J Am Soc Nephrol 2009; 20: 353–362. Schmid H, Boucherot A, Yasuda Y et al. Modular activation of nuclear factor-kappaB transcriptional programs in human diabetic nephropathy. Diabetes 2006; 55: 2993–3003. Lindenmeyer MT, Rastaldi MP, Ikehata M et al. Proteinuria and hyperglycemia induce endoplasmic reticulum stress. J Am Soc Nephrol 2008; 19: 2225–2236. Sanz AB, Santamaria B, Ruiz-Ortega M et al. Mechanisms of renal apoptosis in health and disease. J Am Soc Nephrol 2008; 19: 1634–1642. Wagener FA, Dekker D, Berden JH et al. The role of reactive oxygen species in apoptosis of the diabetic kidney. Apoptosis 2009; 14: 1451–1458. van der Kallen CJ, van Greevenbroek MM, Stehouwer CD et al. Endoplasmic reticulum stress-induced apoptosis in the development of diabetes: is there a role for adipose tissue and liver? Apoptosis 2009; 14: 1424–1434. Lorz C, Justo P, Sanz A et al. Paracetamol-induced renal tubular injury: a role for ER stress. J Am Soc Nephrol 2004; 15: 380–389. Roy S, Sharom JR, Houde C et al. Confinement of caspase-12 proteolytic activity to autoprocessing. Proc Natl Acad Sci USA 2008; 105: 4133–4138. Moley KH, Chi MM, Knudson CM et al. Hyperglycemia induces apoptosis in pre-implantation embryos through cell death effector pathways. Nat Med 1998; 4: 1421–1424. Justo P, Sanz AB, Egido J et al. 3,4-Dideoxyglucosone-3-ene induces apoptosis in renal tubular epithelial cells. Diabetes 2005; 54: 2424–2429. Lee SH, Yoo TH, Nam BY et al. Activation of local aldosterone system within podocytes is involved in apoptosis under diabetic conditions. Am J Physiol Renal Physiol 2009; 297: F1381–F1390. Singh DK, Winocour P, Farrington K. Mechanisms of disease: the hypoxic tubular hypothesis of diabetic nephropathy. Nat Clin Pract Nephrol 2008; 4: 216–226. Liu F, Brezniceanu ML, Wei CC et al. Overexpression of angiotensinogen increases tubular apoptosis in diabetes. J Am Soc Nephrol 2008; 19: 269–280. Ortiz A, Lorz C, Egido J. The Fas ligand/Fas system in renal injury. Nephrol Dial Transplant 1999; 14: 1831–1834. Lorz C, Benito A, Ucero AC et al. Trail and kidney disease. Front Biosci 2009; 14: 3740–3749. Falschlehner C, Ganten TM, Koschny R et al. TRAIL and other TRAIL receptor agonists as novel cancer therapeutics. Adv Exp Med Biol 2009; 647: 195–206. Zheng SJ, Wang P, Tsabary G et al. Critical roles of TRAIL in hepatic cell death and hepatic inflammation. J Clin Invest 2004; 113: 58–64. Zou Y, Zhang X, Mao Y et al. Acute toxicity of a single dose DATR, recombinant soluble human TRAIL mutant, in rodents and crab-eating macaques. Hum Exp Toxicol 2010.
743
mini review
28.
29. 30.
31.
32.
33. 34.
35. 36.
37.
38.
744
Mi QS, Ly D, Lamhamedi-Cherradi SE et al. Blockade of tumor necrosis factor-related apoptosis-inducing ligand exacerbates type 1 diabetes in NOD mice. Diabetes 2003; 52: 1967–1975. Zauli G, Melloni E, Capitani S et al. Role of full-length osteoprotegerin in tumor cell biology. Cell Mol Life Sci 2009; 66: 841–851. Pritzker LB, Scatena M, Giachelli CM. The role of osteoprotegerin and tumor necrosis factor-related apoptosis-inducing ligand in human microvascular endothelial cell survival. Mol Biol Cell 2004; 15: 2834–2841. Rasmussen LM, Tarnow L, Hansen TK et al. Plasma osteoprotegerin levels are associated with glycaemic status, systolic blood pressure, kidney function and cardiovascular morbidity in type 1 diabetic patients. Eur J Endocrinol 2006; 154: 75–81. Sasaki S, Nishihira J, Ishibashi T et al. Transgene of MIF induces podocyte injury and progressive mesangial sclerosis in the mouse kidney. Kidney Int 2004; 65: 469–481. Schwartz V, Lue H, Kraemer S et al. A functional heteromeric MIF receptor formed by CD74 and CXCR4. FEBS Lett 2009; 583: 2749–2757. Carpenter B, Hill KJ, Charalambous M et al. BASP1 is a transcriptional cosuppressor for the Wilms’ tumor suppressor protein WT1. Mol Cell Biol 2004; 24: 537–549. Boldogh IR, Pon LA. Interactions of mitochondria with the actin cytoskeleton. Biochim Biophys Acta 2006; 1763: 450–462. Wiederkehr A, Staple J, Caroni P. The motility-associated proteins GAP-43, MARCKS, and CAP-23 share unique targeting and surface activityinducing properties. Exp Cell Res 1997; 236: 103–116. Geraldes P, Hiraoka-Yamamoto J, Matsumoto M et al. Activation of PKCdelta and SHP-1 by hyperglycemia causes vascular cell apoptosis and diabetic retinopathy. Nat Med 2009; 15: 1298–1306. Lindenmeyer MT, Kretzler M, Boucherot A et al. Interstitial vascular rarefaction and reduced VEGF-A expression in human diabetic nephropathy. J Am Soc Nephrol 2007; 18: 1765–1776.
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
39.
40. 41. 42.
43.
44.
45. 46.
47.
48.
49.
50.
Kelly DJ, Cox AJ, Tolcos M et al. Attenuation of tubular apoptosis by blockade of the renin-angiotensin system in diabetic Ren-2 rats. Kidney Int 2002; 61: 31–39. Brosius III FC, Kaufman RJ. Is the ER stressed out in diabetic kidney disease? J Am Soc Nephrol 2008; 19: 2040–2042. Justo P, Lorz C, Sanz A et al. Intracellular mechanisms of cyclosporin Ainduced tubular cell apoptosis. J Am Soc Nephrol 2003; 14: 3072–3080. Nakagawa T, Zhu H, Morishima N et al. Caspase-12 mediates endoplasmic-reticulum-specific apoptosis and cytotoxicity by amyloidbeta. Nature 2000; 403: 98–103. Zinszner H, Kuroda M, Wang X et al. CHOP is implicated in programmed cell death in response to impaired function of the endoplasmic reticulum. Genes Dev 1998; 12: 982–995. Kimura K, Jin H, Ogawa M et al. Dysfunction of the ER chaperone BiP accelerates the renal tubular injury. Biochem Biophys Res Commun 2008; 366: 1048–1053. Cybulsky AV. Endoplasmic reticulum stressin proteinuric kidney disease. Kidney Int 2010; 77: 187–193. Liu G, Sun Y, Li Z et al. Apoptosis induced by endoplasmic reticulum stress involved in diabetic kidney disease. Biochem Biophys Res Commun 2008; 370: 651–656. Brezniceanu ML, Lau CJ, Godin N et al. Reactive oxygen species promote caspase-12 expression and tubular apoptosis in diabetic nephropathy. J Am Soc Nephrol 2010; 21: 943–954. Mishra R, Emancipator SN, Kern T et al. High glucose evokes an intrinsic proapoptotic signaling pathway in mesangial cells. Kidney Int 2005; 67: 82–93. Saleh M, Vaillancourt JP, Graham RK et al. Differential modulation of endotoxin responsiveness by human caspase-12 polymorphisms. Nature 2004; 429: 75–79. Rodwell GE, Sonu R, Zahn JM et al. A transcriptional profile of aging in the human kidney. PLoS Biol 2004; 2: e427.
Kidney International (2010) 78, 737–744
http://www.kidney-international.org & 2010 International Society of Nephrology
New paradigms in cell death in human diabetic nephropathy Maria-Dolores Sanchez-Nin˜o1, Alberto Benito-Martin1 and Alberto Ortiz1,2,3 1 3
Department of Nefrologia, IIS Fundacion Jimenez Diaz, Madrid, Spain; 2Fundacio´n Renal In˜igo Alvarez de Toledo, Madrid, Spain and Universidad Autonoma de Madrid, Madrid, Spain
Cell death is thought to contribute to progressive renal cell depletion in diabetic nephropathy. Unbiased gene expression profiling identified novel cell death molecules in human diabetic nephropathy. The expression of TNF-related apoptosis-inducing ligand (TRAIL), its decoy receptor osteoprotegerin, and receptors Fas (a Fas ligand receptor) and CD74 (a migration inhibitory factor (MIF) receptor) were induced in human diabetic nephropathy. Cell culture studies supported the functional relevance of this observation and the relationship to a high glucose environment. To define novel proapoptotic proteins upregulated in diabetic nephropathy, functional genomic screens for novel apoptosis mediators were integrated with genome-wide expression profiling and identified candidates for further functional analysis, including brain acid-soluble protein 1 (BASP1). Several lines of evidence point toward induction of endoplasmic reticulum stress response in human diabetic nephropathy. Functional studies defining an unequivocal contribution of endoplasmic reticulum stress to cell death in this setting are still needed. Further comparative studies will be required to define whether there is a specific aspect of apoptosis in progressive human diabetic nephropathy or whether the mechanisms are shared among all patients with chronic kidney disease. The next challenge will be to define the consequence of therapeutic interference of the apoptosis pathways in diabetic nephropathy and chronic kidney disease. Kidney International (2010) 78, 737–744; doi:10.1038/ki.2010.270; published online 11 August 2010 KEYWORDS: apoptosis; BASP1; CD74; diabetic nephropathy; MIF; TRAIL
Correspondence: Alberto Ortiz, Department of Nefrologia, IIS Fundacion Jimenez Diaz, Avenida Reyes Catolicos 2, 28040 Madrid, Spain. E-mail: [email protected] Received 15 February 2010; revised 23 June 2010; accepted 6 July 2010; published online 11 August 2010 Kidney International (2010) 78, 737–744
MECHANISMS FOR CELL LOSS IN DIABETIC NEPHROPATHY
Diabetic nephropathy is characterized by proteinuria and progressive loss of renal function, leading to end-stage renal disease requiring renal replacement therapy. The pathological substrate of diabetic nephropathy is a progressive loss of parenchymal renal cells and extracellular matrix accumulation. Recently, it has become clear that inflammation also contributes to diabetic nephropathy.1 Unbalanced apoptotic cell death leads to renal cell loss and has been observed in podocytes, tubular cells, and endothelial cells in experimental and human diabetic nephropathy.2–8 Diabetic nephropathy is the most frequent cause of end-stage renal disease in developed countries. Only a comprehensive understanding of the molecular basis for renal injury initiation and progression will allow the rational design of novel, successful therapeutic strategies. In recent years, high-throughput techniques exploring gene expression have uncovered novel networked mediators that contribute to renal cell death and may be involved in human diabetic nephropathy.4,9–12 MOLECULAR BASIS FOR APOPTOTIC CELL DEATH
Cell death is usually a response to the cell microenvironment.13 Among the different forms of cell death, apoptosis is molecularly best characterized and may be manipulated therapeutically. Cell survival depends on the interplay between extracellular survival factors that activate intracellular survival pathways and lethal factors that activate cell death pathways leading to apoptosis (Figure 1). A key role for radical oxygen species in cell death and diabetic nephropathy was recently reviewed.14 The two main pathways for apoptosis are ligation of plasma membrane death receptors (‘extrinsic’ pathway) and perturbation of the intracellular homeostasis (‘intrinsic’ pathway). In the extrinsic pathway, ligation of death receptors promotes the assembly of multimolecular complexes that include adaptor proteins and initiator caspases-8 and -10. The intrinsic pathway involves injury to intracellular organelles, the most important being the mitochondria. Sentinel activator ‘BH3-only proteins’ triggers the allosteric activation of Bax and/or Bak, which oligomerize at the mitochondria, inducing permeabilization of the outer mitochondrial membrane and release of proapoptotic factors such as cytochrome c, SMAC/DIABLO and AIF, which 737
mini review
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
MIF Lethal cytokines
Cell stress
TRAIL
CD74 BASP1?
Death receptor activation
Mitochondria
ER stress
TRAILR1/R2 FasL
FAS CD36
HSPA5 HYOU1 XBP1
Caspase activation Mitochondrial injury
Caspase activation
Apoptotic cell death Cell membrane External survival factors OPG DCR3 VEGF EGF
Figure 1 | Summary of apoptotic pathways and location of some key players identified by unbiased gene expression profiling in human diabetic nephropathy (DN). Genes upregulated in human DN include CD74, tumor necrosis factor-related apoptosis-inducing ligand (TRAIL), FAS, brain acid-soluble protein 1 (BASP1), CD36, osteoprotegerin (OPG), heat-shock 70 kDa protein 5 (HSPA5), hypoxia upregulated 1 (HYOU1), and X-box-binding protein 1 (XBP1) and downregulated genes include vascular endothelial growth factor (VEGF), epidermal growth factor (EGF), and DCR3. CD74 activation by migration inhibitory factor (MIF) and other inflammatory stimuli mediate upregulation of the lethal cytokine TRAIL. TRAIL activates the lethal receptors TRAIL receptor 1 (TRAIL-R1) and TRAIL-R2 to induce cell death. Osteoprotegerin is a decoy receptor for TRAIL. The lethal receptor FAS is also increased in human DN. FAS mediates apoptosis on ligation by Fas ligand (FasL). Expression of the FasL decoy receptor DCR3 was downregulated. In human DN the expression of the protective endoplasmic reticulum stress genes HSPA5/GRP78/BiP, HYOU1/ORP150, and XBP1 was upregulated. The precise location of BASP1 action is incompletely characterized, but inhibition of BASP1-induced apoptosis by Bax antagonists and caspase inhibitors suggests a role upstream of mitochondrial injury. CD36 was identified by unbiased gene expression profiling as a downregulated gene in murine DN, but was found increased in human DN. Expression of the extracellular survival factors VEGF and EGF was downregulated in human DN tubular compartment.
promote caspase-dependent and -independent apoptosis. Cytochrome c facilitates the oligomerization of Apaf-1 and caspase-9 in the apoptosome multimolecular complex. Initiator caspase oligomerization activates their enzymatic activity and cleavage of protein substrates, thus activating effector caspases such as caspase-3 and –7, and leading to widespread proteolysis and commitment to cell death. Signaling crosstalk between the intrinsic and extrinsic pathways involves cleavage of proapoptotic BH3-only protein Bid by caspase-8. The active fragment, tBid, translocates to mitochondria and recruits the intrinsic pathway. Bcl2-related antiapoptotic proteins, such as Bcl2 and BclxL, bind and sequester activator BH3-only proteins, Bax and Bak, and may directly inhibit Apaf-1-mediated activation of caspase-9. Sensitizer BH3-only molecules competitively displace activator BH3-only proteins from the Bcl2 pocket. Inhibitors of apoptosis proteins directly inhibit caspases and are, in turn, inhibited by SMAC/DIABLO. Endoplasmic reticulum stress leads to accumulation of unfolded proteins in the endoplasmic reticulum and triggers 738
the unfolded protein response, which strives to reestablish homeostasis.15 Thus, the unfolded protein response is primarily an adaptive response. In resting cells, the endoplasmic reticulum chaperone HSPA5/GRP78/BiP (heat-shock 70 kDa protein 5/glucose-regulated protein 78 kDa) binds to endoplasmic reticulum stress sensors (IRE1, PERK, and ATF6) and keeps them inactive. On endoplasmic reticulum stress, HSPA5/GRP78/BiP is recruited by unfolded proteins, thus releasing and activating PERK, IRE1, and ATF6. PERK activation leads to translation inhibition. IRE1 initiates the nonconventional splicing of Xbp-1 mRNA, encoding the transcription activator XBP1, that collaborates with active ATF6 to promote transcription of endoplasmic reticulum chaperones, including HSPA5/ GRP78/BiP. However, under conditions of severe and prolonged endoplasmic reticulum stress, failure to restore homeostasis is followed by cell death, usually occurring by apoptosis.12,13,15,16 There are three main pathways that may contribute to endoplasmic reticulum stress-induced apoptosis: (a) the CHOP/growth arrest and DNA-damage-inducible Kidney International (2010) 78, 737–744
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
(GADD153) transcription factor, which is mainly induced via PERK, downregulates Bcl2; (b) IRE1-mediated activation of apoptosis signal-regulating kinase 1 (ASK1)/c-Jun NH2terminal kinase (JNK); and (c) activation of the endoplasmic reticulum-localized caspase-12. The three pathways are thought to activate the caspase cascade and may recruit the mitochondrial pathway. However, although caspase-12 processing is widely recognized as a marker of endoplasmic reticulum stress, its catalytic activity is limited to selfprocessing and its role in cell death has been questioned.17 In the context of diabetic nephropathy hyperglycemia may induce or facilitate apoptosis.2,4,10,18 There are no specific pathways for high-glucose-induced cell death. Rather, high glucose promotes the upregulation or activation of several molecules in the different proapoptotic pathways. Early reports disclosed upregulation of proapoptotic molecules such as Bax and downregulation of antiapoptotic molecules (Bcl2 and BclxL) by high glucose in the renal cells.2 However, high glucose regulation of cell death-related genes is more complex than initially thought. In addition, hypoxia, oxidative stress, inflammatory cytokines, the renin–angiotensin–aldosterone system, advanced glycosylation end products, and glucose degradation products may also contribute to cell death in diabetic nephropathy.8–10,19–22 CELL DEATH GENE EXPRESSION IN HUMAN DIABETIC NEPHROPATHY
The European Renal cDNA Bank has characterized the transcriptomic fingerprint of human diabetic nephropathy, that is, the level of expression of multiple mRNAs.11 Cell death-related genes were identified through several approaches4,9–12 (Table 1). First, the gene ontology term ‘Cell death’ identified 66 significantly upregulated and 46 significantly downregulated genes.9 Overall, 25% of apoptosis-related genes were differentially regulated in human diabetic nephropathy.9 Differentially expressed apoptosisrelated genes included those coding for cytokines, receptors, intracellular lethal proteins and caspases. Caspase-1 and caspase-4 were upregulated. These caspases are now widely recognized to have mainly a proinflammatory role. In a second approach, differentially expressed endoplasmic reticulum stress genes were characterized.12 Finally, novel cell death genes upregulated in diabetic nephropathy were identified by combining the database with a functional genomics screen for cDNAs that induce apoptosis when overexpressed in cultured cells: 12 of the 138 full-length cDNAs inducing cell death when overexpressed in human embryonic kidney cells also showed mRNA transcript induction in human diabetic nephropathy biopsies.4 DEATH LIGANDS AND RECEPTORS IN DIABETIC NEPHROPATHY
Cytokines belonging to the tumor necrosis factor (TNF) superfamily (TNFSF) activate a superfamily of receptors (TNFRSF) and are key regulators of cell survival, inflammation, and fibrosis. TNF/TNFSF2 had long been known to contribute to diabetic nephropathy.1 Fas (TNFSFR6), the Kidney International (2010) 78, 737–744
mini review
Table 1 | Cell death genes with differential mRNA expression in human DN4,9,12,a Cell death (gene ontology) genes Upregulated Osteoprotegerin TRAIL: TNF-related apoptosis-inducing ligand TNFAIP8: Tumor necrosis factor-a-induced protein 8 14–3-3, Z STAT1: Signal transducer and activator of transcription 1 Fas APP: Amyloid precursor protein Tubulin-b Osteopontin Caspase 1 CD74 EIF4G2: Eukaryotic translation initiation factor 4 g-2 NRIF3: Neurotrophin receptor interacting factor 3 TIA1: T-cell intracellular antigen 1 LOXL2: Lysyl oxidase-like 2 Clusterin DOCK1: Dedicator of cytokinesis 1 EMP3: Epithelial membrane protein 3 Downregulated GADD45A: Growth arrest and DNA-damage-inducible-45-a Apolipoprotein E Sphingosine kinase 2 Prostaglandin E receptor 3 MAPT: Microtubule-associated protein-t PKC, z: Protein kinase C z Upregulated endoplasmic reticulum stress genes HSPA5/GRP78/BiP: Heat-shock 70 kDa protein 5/glucose-regulated protein 78 kDa HYOU1/ORP150: Hypoxia upregulated 1/oxygen-regulated protein 150 XBP1: X-box-binding protein 1 Genes upregulated in DN that were identified by a functional genomics screening as inducing cell death when overexpressed in cultured cells EMP3: Epithelial membrane protein-3 BASP1: Brain acid-soluble protein 1 PLP2: Proteolipid protein 2 (colonic epithelium enriched) CLDN4: Claudin 4 ITGB2: Integrin, b 2 (complement component 3 receptor 3 and 4 subunit) Abbreviation: DN, diabetic nephropathy. a Only genes with at least a 1.5-fold difference in expression between DN and control biopsies are listed.
receptor for Fas ligand (FasL/TNFSF6), is upregulated in diabetic nephropathy and the proapoptotic role of FasL/Fas in the kidney, including in diabetic nephropathy, has been extensively studied.3,23 Unbiased gene expression profiling confirmed the increase Fas expression in diabetic nephropathy and disclosed decreased expression of the decoy FasL receptor DCR3 (TNFSFR6b) mRNA.9 Increased Fas and FasL protein expression was confirmed by immunohistochemistry in human diabetic nephropathy.6 However, the two most upregulated cell death mRNA transcripts in human diabetic nephropathy were TNF-related apoptosis-inducing ligand (TRAIL/TNFSF10) and its decoy receptor osteoprotegerin (TNFRSF11B).9 TRAIL is a type II transmembrane protein with a molecular mass of 33–35 kDa. Membrane-bound TRAIL can be cleaved from the cell surface to form a soluble trimeric 739
mini review
protein that retains the proapoptotic activity. TRAIL binds to a complex system of four membrane-bound and one soluble receptor.24 TRAIL receptor 1 (TRAIL-R1/TNFRSF10A) and TRAIL-R2 (TNFSFR10B) have a functional death domain that allows caspase activation and apoptosis induction. In contrast, TRAIL-R3 (TNFRSF10C), TRAIL-R4 (TNFRSF10D), and osteoprotegerin do not transduce cell death signals and osteoprotegerin behaves as a decoy receptor. Depending on the relative levels of the different receptors and on the cell type, TRAIL can induce cell death, and also survival, proliferation, and maturation. Exogenous TRAIL is more lethal to cultured cancer cells than to normal cells, and TRAIL-R agonists are in clinical trials as anticancer agents.25 Many normal human tissues including the kidney express low levels of TRAIL, TRAIL-R1, and -R2, which is consistent with the low cytotoxicity of TRAIL to most healthy tissues. However, membrane-bound TRAIL has a higher lethal potential than soluble TRAIL, and data from knockout mice suggest that endogenous TRAIL may induce apoptosis in nontumoral inflamed liver.26 In this regard, hepatotoxicity and nephrotoxicity were observed in preclinical studies with TRAIL agonists.27 TRAIL also has immunoregulatory functions that may be beneficial for the onset of autoimmune diabetes.28 Osteoprotegerin is a secreted protein with no transmembrane domain, which is produced as a monomer (55–62 kDa), undergoes homodimerization, and is secreted as a disulfide-linked homodimeric glycoprotein of 110–120 kDa.29 Osteoprotegerin is a soluble decoy receptor for both TRAIL and receptor activator of nuclear factor-kB ligand (RANKL/TNFSF11) and may have independent functions promoting endothelial cell survival and angiogenesis. Potential cross-regulatory mechanisms involving the balance TRAIL–osteoprotegerin–RANKL may exist. Although the affinity of TRAIL for osteoprotegerin is weaker than for the lethal TRAIL receptors, there is evidence for the biological relevance of the interaction.30 On a tissue level, osteoprotegerin regulates bone mineral density and vascular calcification, and serum osteoprotegerin levels are increased in patients with renal dysfunction, including diabetic nephropathy, and are associated with adverse cardiovascular outcomes.31 In the normal kidney, TRAIL is expressed in tubules but not in glomerular cells.9 In diabetic nephropathy, TRAIL and osteoprotegerin mRNA expression is increased and correlates to clinical and histological severity of renal disease.9 Tubular epithelium is the predominant nephron cell lineage expressing TRAIL protein, but de novo protein expression in podocytes was also noted in diabetic nephropathy.9,10 High glucose did not modulate TRAIL expression in cultured renal cells. However, proinflammatory cytokines such as interferon-g, TNFa, and migration inhibitory factor (MIF) increased TRAIL expression in cultured renal tubular cells and podocytes.9,10 TRAIL promoted a modest increase in tubular cell and podocyte death, which is consistent with a role in the chronic progression of diabetic nephropathy seen 740
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
humans. Nuclear factor-kB activation by TRAIL protected cells from TRAIL lethality.9 In this regard, both TRAIL and TNF activate simultaneous death and survival signals, and this may underlie the low sensitivity of non-neoplastic cells to both cytokines unless the microenvironment is unfavorable for survival. Blockade of nuclear factor-kB-mediated survival signals or additional environmental death signals, such as high glucose or inflammation, increase the rate of cell death.9,10 Osteoprotegerin, even endogenous osteoprotegerin secreted by tubular cells, protected from TRAIL-induced cell death.9 The end result of the increased expression of TRAIL and osteoprotegerin in a particular patient may depend on the relative balance between them. We hypothesize that a misbalance toward TRAIL may result in tissue injury. If osteoprotegerin prevails, the cells would be protected from TRAIL. Podocyte overexpression of the macrophage MIF cytokine results in podocyte injury and proteinuria.32 MIF was increased in experimental diabetic nephropathy, and serum MIF was elevated in patients with diabetic nephropathy (reviewed in Sanchez-Nino et al.10). The protein expression of the MIF receptor CD74 is increased in tubular cells and podocytes in human diabetic nephropathy, and CD74 mRNA expression is increased in diabetic nephropathy.10 MIF did not promote renal cell apoptosis. However, MIF engagement of CD74 activated MAP kinases and increased TRAIL expression.10 CD74 may have MIF-independent functions, as it is a coreceptor for CXCR4 or CD44.33 These functions have not been fully characterized in renal cells. Microarray-based gene expression profiling on whole kidney RNA, together with supervised clustering methods, identified CD36 as a downregulated mRNA in diabetic mice with albuminuria.8 CD36 protein was absent from proximal cells, both in control and diabetic mice. In contrast, CD36 protein was upregulated and colocalized with apoptotic proximal tubular cells in human diabetic nephropathy. This is one of several examples of dissociation in the molecular signature of human and experimental diabetic nephropathy. CD36 is a class B scavenger receptor. Cell culture studies disclosed that high glucose upregulated CD36 mRNA and protein, and engagement by glycated albumin transduced a lethal signal to human proximal tubular cells. Cell death was prevented by caspase-8, but not by caspase-9 inhibition, suggesting a role for death receptors. However, the molecular pathways were not further characterized. MITOCHONDRIAL STRESS IN DIABETIC NEPHROPATHY
Mitochondrial stress has long been known to contribute to high-glucose-induced apoptosis. Increased Bax expression was the first apoptosis-related molecular change observed in kidney cells in response to high glucose, and Bax knockout mice are protected from high-glucose-induced cell death.2,18 More recently, a novel proapoptotic role has been identified for brain acid-soluble protein 1/cortical cytoskeleton-associated protein of 23 kD (BASP1/CAP-23).4 BASP1 is a multifunctional protein involved in cytoskeletal and lipid Kidney International (2010) 78, 737–744
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
raft dynamics, as well as in the nuclear regulation of transcription. BASP1 expression is mainly cytoplasmic in some cell types including tubular cells, and nuclear in others, such as podocytes.4,34 In the adult kidney podocyte, BASP1 is a transcriptional cosuppressor of WT1.34 Kidney BASP1 was upregulated at the mRNA and protein level in human and experimental diabetic nephropathy.4 Tubular cells were the main site of BASP1 upregulation, where it localized to apoptotic cells.4 In cultured human tubular epithelial cells, lethal stimuli such as serum deprivation, high glucose, or proinflammatory cytokines increased BASP1 mRNA and protein.4 Overexpression of BASP1 was sufficient to induce apoptosis in tubular cells. In addition, BASP1 knockdown protected tubular cells from apoptosis induced by serum deprivation or high glucose concentrations, but not from cytokine-induced apoptosis.4 Thus, BASP1 may be critical for cell stress-induced apoptotic cell death. BASP1 is upregulated in other forms of renal injury, such as hypertensive nephropathy, but shows the most robust induction in diabetic nephropathy.4 The precise role of BASP1 in the apoptotic pathway is unclear. Interestingly, BASP1 colocalizes with the apoptotic actin ring and tubulin in the cell cortex, suggesting that the cytoskeleton-binding properties of BASP1 may be associated with its lethal action.4 Cytoskeletal remodeling is a hallmark of apoptosis, and changes in actin dynamics contribute to mitochondrial injury and cell death.35 Prevention of BASP1-induced apoptosis by Bax antagonists and caspase inhibitors suggests that BASP1 has a role upstream of mitochondria or caspase activation, or that BASP1 contributes to a positive feed-back loop, leading to further mitochondrial injury and caspase activation.4 BASP1/ CAP-23, GAP-43 and MARCKS belong to a protein family characterized by the presence of an effector domain, association with the cortical cytoskeleton, and regulation of actin-binding proteins regulated by reversible phosphorylation of the effector domain by protein kinase C.36 Interestingly, different isoforms of protein kinase C have been involved in diabetes-associated apoptosis.37 Deprivation of survival factors is a known inducer of apoptosis that cooperates with the presence of lethal factors such as cytokines and toxins to promote renal cell death.13 In human diabetic nephropathy, the expression of two cytokines with survival activity of renal epithelial cells, vascular endothelial growth factor-A and epidermal growth factor, was reduced at the mRNA and protein level in the tubular compartment.13,38 In glomerular endothelium, receptorbound vascular endothelial growth factor protein was decreased in more advanced lesions.7 The reduced vascular endothelial growth factor-A expression contrasted to previously reported findings in experimental models. Reduced renal epidermal growth factor mRNA and had been previously observed in animal models.39 ENDOPLASMIC RETICULUM STRESS IN DIABETIC NEPHROPATHY
Stimuli that increase the demand on the endoplasmic reticulum to synthesize proteins or degrade improperly Kidney International (2010) 78, 737–744
mini review
folded proteins cause endoplasmic reticulum stress.40 High glucose, free fatty acids, albumin, nephrotoxins, oxidative stress, or inflammation induce endoplasmic reticulum stress in renal cells.12,16,41 In human progressive diabetic nephropathy, expression of mRNA for protective endoplasmic reticulum chaperone proteins HSPA5/GRP78/BiP and HYOU1/ORP150 (hypoxia upregulated 1/oxygen-regulated protein 150) and for the prosurvival transcription factor XBP1 (X-box-binding protein 1) was increased.12 Increased HSPA5/GRP78/BiP and HYOU1/ORP150 protein expression was confirmed at the protein level. However, potentially lethal unfolded protein response mRNAs, such as CHOP, were unchanged or repressed.12 The finding of increased mRNA levels for genes involved in the protective responses of the unfolded protein response, but not for proapoptotic genes, suggests that the endoplasmic reticulum stress in human diabetic nephropathy might be adaptive and, per se, does not support a role of endoplasmic reticulum stress in progression of diabetic nephropathy.40 In this regard, induction of some potentially lethal endoplasmic reticulum stress proteins may not be followed by completion of the endoplasmic reticulum stress pathway.16,41 Thus, cyclosporin A increased CHOP/GADD153 protein expression in tubular cells, but failed to result in caspase-12 processing, indicating that CHOP upregulation may be induced by nonendoplasmic reticulum proapoptotic stressors. However, it is conceivable that interference with adaptive mechanisms, because of the influence of genetic background or additional insults, may result in endoplasmic reticulum-mediated cell death in human diabetic nephropathy. In animal models there is evidence that endoplasmic reticulum stress-induced apoptosis may cause tubular cell injury (reviewed in Sanz et al.13). Thus, the endoplasmic reticulum stressor tunicamycin induced severe histological tubular injury, which was decreased both in CHOP/ GADD153 and caspase-12 knockout mice.42,43 Furthermore, there is evidence that impairing the adaptive endoplasmic reticulum response, as is the case for HSPA5/GRP78/BiPmutant heterozygous mice, results in more severe tubulointerstitial injury in response to albuminuria.44 However, there is no functional evidence yet that endoplasmic reticulum stress, in the presence of an intact adaptive endoplasmic reticulum stress response, contributes to renal cell apoptosis in vivo in animal models of proteinuric kidney disease.45 This is in part due to technical difficulties, given the relative paucity of adequate mouse models. To complicate matters further, significant differences in the endoplasmic reticulum stress response between human and experimental diabetic nephropathy preclude direct extrapolation of animal model results to the human situation. As in humans, increased expression of adaptive endoplasmic reticulum stress proteins, such as HSPA5/GRP78/BiP, has been observed in murine and rat models.46,47 Although early studies did not find increased caspase-12 in db/db mice,48 a recent transcriptomics-based approach identified caspase-12 as an upregulated mRNA in this model.47 Caspase-12 protein was processed and increased 741
mini review
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
expression was noted in apoptotic tubular cells in vivo. Caspase-12 was also increased in rat streptozotocin-induced diabetic nephropathy.46 Furthermore, albumin induced caspase-12-dependent apoptosis in cultured human HK2 tubular cells. However, most humans carry a single-nucleotide premature stop codon polymorphism that inactivates caspase-12.49 In this regard, although HK2 cells carry the inactive caspase-12 allele, they expressed a full-length protein, probably as a result of translational readthrough induced by aminoglycosides in the cell culture media. Thus, the relevance of these findings to human diabetic nephropathy is unclear, with the exception of the 20% of the population of African descent that expresses a functional protein. In contrast to humans, pro-death CHOP/GADD153 mRNA and protein was increased in animal models of diabetic nephropathy.46,47 SPECIFICITY OF THE OBSERVED CHANGES IN GENE EXPRESSION FOR HUMAN DIABETIC NEPHROPATHY
Exposure to a high glucose environment for a few hours will increase CD74 and BASP1, but not TRAIL expression, in cultured tubular cells. At present, it is not possible to differentiate whether these changes are a consequence of diabetes, nephropathy, or diabetic nephropathy in humans. However, there is a progressive gradient for magnitude of changes in the expression of these genes between early and late diabetic nephropathy.9 The fact that inflammatory cytokines can induce the expression of these genes in cultured cells suggests a role beyond diabetic nephropathy and indicates that, once a threshold of kidney injury is reached, these genes might be important for progression of
diabetic nephropathy even when an adequate glycemic control has been achieved. In any case, there appears to be a specific interaction between high glucose concentrations and sensitivity to TRAIL or BASP1-induced death.4,9 There is a limited understanding of the range of human nephropathies that present with changes in the expression of genes such as TRAIL, CD74, or BASP1. Data from our group and a search of the Nephromine database (www.nephromine. org) indicate that some of these changes are present in other proteinuric progressive nephropathies, but not present or milder in nonprogressive proteinuric kidney injury. BASP1 mRNA expression is increased in membranous nephropathy and age-related chronic kidney injury, and a trend was observed for hypertensive nephropathy. BASP1 did not change in IgA nephropathy or minimal change disease.4,50 Upregulation of TRAIL and osteoprotegerin mRNA in age-related chronic kidney injury and minimal change disease was significant but milder than in human diabetic nephropathy.11,50 Fas mRNA was increased in age-related chronic kidney injury, but not in minimal change disease.11,50 CD74 mRNA was increased in hypertensive nephropathy and age-related chronic kidney injury, but not in minimal change disease.4,50 In contrast, evidence for endoplasmic reticulum stress, including increased expression of CHOP/GADD153 (not found upregulated in human diabetic nephropathy) and HSPA5/GRP78/BiP, has been observed in a variety of proteinuric nephropathies.45 Proximal tubular CD36 expression was not observed in focal segmental glomerulosclerosis, but other human diseases were not tested.8 There are additional ligands for CD36, not
Hyperglycemia Increase Increase
Increase Sensitize Basp1
Inflammation Increase TRAIL (MIF) CD74 TNF
Sensitize
Low VEGF, EGF
Intracellular events
Parenchymal renal cell apoptosis
Progression of kidney disease
OPG
Figure 2 | Hypothetical contribution of novel cell death pathways to human diabetic nephropathy (DN): integration of inflammation and high glucose. Renal cell loss by apoptosis is thought to contribute to the progression of DN. Both inflammation and high glucose are known inducers of apoptosis. Renal cell culture studies suggest that inflammation contributes to increased tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) and Fas expression. Thus, migration inhibitory factor (MIF) activation of CD74 or tumor necrosis factor (TNF) increased TRAIL expression and TNF increases Fas expression. In contrast hyperglycemia increased CD74 expression, increased brain acid-soluble protein 1 (BASP1) expression, and also sensitized to the lethal effect of TRAIL on renal cells. Osteoprotegerin (OPG) antagonized the lethal effect of TRAIL in cultured cells. BASP1 had a role in hyperglycemia- and growth factor deprivation-induced apoptosis. Low levels of the survival factors vascular endothelial growth factor (VEGF) and endothelial growth factor (EGF) were observed in human DN and may also contribute to apoptosis sensitization. 742
Kidney International (2010) 78, 737–744
mini review
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
specific for diabetes, such as thrombospondin-1, that also promote tubular cell apoptosis.8 INTEGRATIVE VIEW AND FUTURE PERSPECTIVES
Recent evidence gathered from the direct study of gene expression patterns in human diabetic nephropathy coupled with confirmation at the protein expression level have unraveled a set of new molecules and pathways, and novel information on known players in kidney epithelial cell death. Thus, the interplay among high glucose, inflammation, and lack of survival factors may contribute to activation of death receptor, mitochondrial injury, and endoplasmic reticulum stress pathways that may ultimately lead to renal cell loss (Figure 2). Within each of these pathways, we find a balance of intracellular lethal and adaptive responses. Thus, potentially lethal cytokines such as TNF and TRAIL also activate prosurvival nuclear factor-kB, and current data indicate an adaptive pattern of endoplasmic reticulum stress response genes observed in human diabetic nephropathy. Additionally, increased expression of some decoy receptors has been observed. The net end result observed in vivo is progressive loss of renal parenchymal cells in a significant proportion of patients. The molecular networks deduced from these studies provide a paradigm for understanding cell death regulation in the course of diabetic nephropathy. A high glucose level appears to have a role as sensitizer to cell death. Subsequent features of diabetic nephropathy, such as the presence of albuminuria and renal inflammation, are key additional triggers of cell death and may drive the process even if adequate control of blood glucose levels is reestablished, providing a strong rational for antiapoptotic or antiinflammatory strategies in diabetic nephropathy. Different to previous approaches, recent progress in identification of molecules of potential pathogenic importance has been based on the analysis of human diabetic nephropathy. Changes in the expression of most, but not all, newly identified genes are also observed in animal models of diabetic nephropathy, setting the stage for preclinical intervention on some molecules. These studies will clarify whether any of the reported gene changes is a legitimate therapeutic target. It is conceivable that, in the future, multipronged approaches to the management of diabetic nephropathy may include strategies to prolong renal cell survival. DISCLOSURE
All the authors declared no competing interests. ACKNOWLEDGMENTS
This work was supported by grants from FIS 06/0046, PS09/00447, ISCIII-RETICS REDINREN RD 06/0016, MEC (SAF 03/884), and Sociedad Espanola de Nefrologia. ABM was supported by FIS, MDSN by MEC, and AO by the Programa de Intensificacio´n de la Actividad Investigadora in the Sistema Nacional de Salud of the Instituto de Salud Carlos III, the Agencia ‘Pedro Lain Entralgo’ of the Comunidad de Madrid, and CIFRA S-BIO 0283/2006. We appreciate the critical review, thoughtful comments, and contribution from Matthias Kretzler, University of Michigan. Kidney International (2010) 78, 737–744
REFERENCES 1.
2.
3.
4. 5. 6.
7.
8.
9. 10. 11.
12.
13.
14.
15.
16. 17.
18.
19.
20.
21.
22.
23. 24. 25.
26. 27.
Navarro JF, Mora-Fernandez C. The role of TNF-alpha in diabetic nephropathy: pathogenic and therapeutic implications. Cytokine Growth Factor Rev 2006; 17: 441–450. Ortiz A, Ziyadeh FN, Neilson EG. Expression of apoptosis-regulatory genes in renal proximal tubular epithelial cells exposed to high ambient glucose and in diabetic kidneys. J Investig Med 1997; 45: 50–56. Kelly DJ, Stein-Oakley A, Zhang Y et al. Fas-induced apoptosis is a feature of progressive diabetic nephropathy in transgenic (mRen-2)27 rats: attenuation with renin-angiotensin blockade. Nephrology (Carlton) 2004; 9: 7–13. Sanchez-Nino MD, Sanz AB, Lorz C et al. BASP1 promotes apoptosis in diabetic nephropathy. J Am Soc Nephrol 2010; 21: 610–621. Gilbert RE, Marsden PA. Activated protein C and diabetic nephropathy. N Engl J Med 2008; 358: 1628–1630. Verzola D, Gandolfo MT, Ferrario F et al. Apoptosis in the kidneys of patients with type II diabetic nephropathy. Kidney Int 2007; 72: 1262–1272. Hohenstein B, Hausknecht B, Boehmer K et al. Local VEGF activity but not VEGF expression is tightly regulated during diabetic nephropathy in man. Kidney Int 2006; 69: 1654–1661. Susztak K, Ciccone E, McCue P et al. Multiple metabolic hits converge on CD36 as novel mediator of tubular epithelial apoptosis in diabetic nephropathy. PLoS Med 2005; 2: e45. Lorz C, Benito-Martin A, Boucherot A et al. The death ligand TRAIL in diabetic nephropathy. J Am Soc Nephrol 2008; 19: 904–914. Sanchez-Nino MD, Sanz AB, Ihalmo P et al. The MIF receptor CD74 in diabetic podocyte injury. J Am Soc Nephrol 2009; 20: 353–362. Schmid H, Boucherot A, Yasuda Y et al. Modular activation of nuclear factor-kappaB transcriptional programs in human diabetic nephropathy. Diabetes 2006; 55: 2993–3003. Lindenmeyer MT, Rastaldi MP, Ikehata M et al. Proteinuria and hyperglycemia induce endoplasmic reticulum stress. J Am Soc Nephrol 2008; 19: 2225–2236. Sanz AB, Santamaria B, Ruiz-Ortega M et al. Mechanisms of renal apoptosis in health and disease. J Am Soc Nephrol 2008; 19: 1634–1642. Wagener FA, Dekker D, Berden JH et al. The role of reactive oxygen species in apoptosis of the diabetic kidney. Apoptosis 2009; 14: 1451–1458. van der Kallen CJ, van Greevenbroek MM, Stehouwer CD et al. Endoplasmic reticulum stress-induced apoptosis in the development of diabetes: is there a role for adipose tissue and liver? Apoptosis 2009; 14: 1424–1434. Lorz C, Justo P, Sanz A et al. Paracetamol-induced renal tubular injury: a role for ER stress. J Am Soc Nephrol 2004; 15: 380–389. Roy S, Sharom JR, Houde C et al. Confinement of caspase-12 proteolytic activity to autoprocessing. Proc Natl Acad Sci USA 2008; 105: 4133–4138. Moley KH, Chi MM, Knudson CM et al. Hyperglycemia induces apoptosis in pre-implantation embryos through cell death effector pathways. Nat Med 1998; 4: 1421–1424. Justo P, Sanz AB, Egido J et al. 3,4-Dideoxyglucosone-3-ene induces apoptosis in renal tubular epithelial cells. Diabetes 2005; 54: 2424–2429. Lee SH, Yoo TH, Nam BY et al. Activation of local aldosterone system within podocytes is involved in apoptosis under diabetic conditions. Am J Physiol Renal Physiol 2009; 297: F1381–F1390. Singh DK, Winocour P, Farrington K. Mechanisms of disease: the hypoxic tubular hypothesis of diabetic nephropathy. Nat Clin Pract Nephrol 2008; 4: 216–226. Liu F, Brezniceanu ML, Wei CC et al. Overexpression of angiotensinogen increases tubular apoptosis in diabetes. J Am Soc Nephrol 2008; 19: 269–280. Ortiz A, Lorz C, Egido J. The Fas ligand/Fas system in renal injury. Nephrol Dial Transplant 1999; 14: 1831–1834. Lorz C, Benito A, Ucero AC et al. Trail and kidney disease. Front Biosci 2009; 14: 3740–3749. Falschlehner C, Ganten TM, Koschny R et al. TRAIL and other TRAIL receptor agonists as novel cancer therapeutics. Adv Exp Med Biol 2009; 647: 195–206. Zheng SJ, Wang P, Tsabary G et al. Critical roles of TRAIL in hepatic cell death and hepatic inflammation. J Clin Invest 2004; 113: 58–64. Zou Y, Zhang X, Mao Y et al. Acute toxicity of a single dose DATR, recombinant soluble human TRAIL mutant, in rodents and crab-eating macaques. Hum Exp Toxicol 2010.
743
mini review
28.
29. 30.
31.
32.
33. 34.
35. 36.
37.
38.
744
Mi QS, Ly D, Lamhamedi-Cherradi SE et al. Blockade of tumor necrosis factor-related apoptosis-inducing ligand exacerbates type 1 diabetes in NOD mice. Diabetes 2003; 52: 1967–1975. Zauli G, Melloni E, Capitani S et al. Role of full-length osteoprotegerin in tumor cell biology. Cell Mol Life Sci 2009; 66: 841–851. Pritzker LB, Scatena M, Giachelli CM. The role of osteoprotegerin and tumor necrosis factor-related apoptosis-inducing ligand in human microvascular endothelial cell survival. Mol Biol Cell 2004; 15: 2834–2841. Rasmussen LM, Tarnow L, Hansen TK et al. Plasma osteoprotegerin levels are associated with glycaemic status, systolic blood pressure, kidney function and cardiovascular morbidity in type 1 diabetic patients. Eur J Endocrinol 2006; 154: 75–81. Sasaki S, Nishihira J, Ishibashi T et al. Transgene of MIF induces podocyte injury and progressive mesangial sclerosis in the mouse kidney. Kidney Int 2004; 65: 469–481. Schwartz V, Lue H, Kraemer S et al. A functional heteromeric MIF receptor formed by CD74 and CXCR4. FEBS Lett 2009; 583: 2749–2757. Carpenter B, Hill KJ, Charalambous M et al. BASP1 is a transcriptional cosuppressor for the Wilms’ tumor suppressor protein WT1. Mol Cell Biol 2004; 24: 537–549. Boldogh IR, Pon LA. Interactions of mitochondria with the actin cytoskeleton. Biochim Biophys Acta 2006; 1763: 450–462. Wiederkehr A, Staple J, Caroni P. The motility-associated proteins GAP-43, MARCKS, and CAP-23 share unique targeting and surface activityinducing properties. Exp Cell Res 1997; 236: 103–116. Geraldes P, Hiraoka-Yamamoto J, Matsumoto M et al. Activation of PKCdelta and SHP-1 by hyperglycemia causes vascular cell apoptosis and diabetic retinopathy. Nat Med 2009; 15: 1298–1306. Lindenmeyer MT, Kretzler M, Boucherot A et al. Interstitial vascular rarefaction and reduced VEGF-A expression in human diabetic nephropathy. J Am Soc Nephrol 2007; 18: 1765–1776.
M-D Sanchez-Nin˜o et al.: Cell death and diabetic nephropathy
39.
40. 41. 42.
43.
44.
45. 46.
47.
48.
49.
50.
Kelly DJ, Cox AJ, Tolcos M et al. Attenuation of tubular apoptosis by blockade of the renin-angiotensin system in diabetic Ren-2 rats. Kidney Int 2002; 61: 31–39. Brosius III FC, Kaufman RJ. Is the ER stressed out in diabetic kidney disease? J Am Soc Nephrol 2008; 19: 2040–2042. Justo P, Lorz C, Sanz A et al. Intracellular mechanisms of cyclosporin Ainduced tubular cell apoptosis. J Am Soc Nephrol 2003; 14: 3072–3080. Nakagawa T, Zhu H, Morishima N et al. Caspase-12 mediates endoplasmic-reticulum-specific apoptosis and cytotoxicity by amyloidbeta. Nature 2000; 403: 98–103. Zinszner H, Kuroda M, Wang X et al. CHOP is implicated in programmed cell death in response to impaired function of the endoplasmic reticulum. Genes Dev 1998; 12: 982–995. Kimura K, Jin H, Ogawa M et al. Dysfunction of the ER chaperone BiP accelerates the renal tubular injury. Biochem Biophys Res Commun 2008; 366: 1048–1053. Cybulsky AV. Endoplasmic reticulum stressin proteinuric kidney disease. Kidney Int 2010; 77: 187–193. Liu G, Sun Y, Li Z et al. Apoptosis induced by endoplasmic reticulum stress involved in diabetic kidney disease. Biochem Biophys Res Commun 2008; 370: 651–656. Brezniceanu ML, Lau CJ, Godin N et al. Reactive oxygen species promote caspase-12 expression and tubular apoptosis in diabetic nephropathy. J Am Soc Nephrol 2010; 21: 943–954. Mishra R, Emancipator SN, Kern T et al. High glucose evokes an intrinsic proapoptotic signaling pathway in mesangial cells. Kidney Int 2005; 67: 82–93. Saleh M, Vaillancourt JP, Graham RK et al. Differential modulation of endotoxin responsiveness by human caspase-12 polymorphisms. Nature 2004; 429: 75–79. Rodwell GE, Sonu R, Zahn JM et al. A transcriptional profile of aging in the human kidney. PLoS Biol 2004; 2: e427.
Kidney International (2010) 78, 737–744