Toll-Like Receptors—Intrarenal Mechanisms beyond Immune Function

The unraveling of the Toll receptor and the Toll-like receptor (TLR) family is a fascinating story full of surprises. It all started with the observation of the Nobel laureate Nu[Combining Diaeresis]sslein-Volhard (related to the family of the nestor of nephrology Franz Volhard [1]) that the dorsal-ventral polarity in Drosophila resulted from the action of a specific gene product; because the resulting shape of the disorganized body looked “crazy” (in German, “toll”) the gene was called Toll (2). It was later found that the Toll gene encodes an interleukin1 receptor-like protein triggering synthesis of bactericidal and fungicidal peptides in blood cells of Drosophila (3,4), linking the gene to the innate immune response—obviously leading to the question of whether analogues existed in mammalian species. In the search for human homologues of Drosophila Toll, the group of Medzhitov et al. (5) identified the human Toll-like receptor 4 (TLR4), which was shown to activate NFκB-controlled genes such as IL-1, IL-6, and IL-8, and to cause the induction of members of the B7 family required for the activation of nai[Combining Diaeresis]ve T cells. This observation pointed to a potentially important link between pathogen detection and the induction of the adaptive immune response suggesting that TLR functioned as a link between innate and acquired immunity. It was soon recognized that the underlying common principle of the rapidly expanding family of TLR, currently comprising 10 human isoforms, was the recognition of phagocyte-related chemical patterns, e.g., mannans, lipopolysaccharide (LPS), teichoic acids, etc. Such recognition is obviously important to distinguish potential pathogens from self to preserve tolerance to self and defend against foreign (6). The specificity for the different ligands is provided by heterodimerization of a given TLR with cytoplasmic adaptor molecules (7). What has also become clear in recent years, however, is that not only exogenous microbial products but also endogenous ligands released after cell injury or inflammation may activate the TLR (8)—in other words, TLR sense danger signals of exogenous or endogenous origin. It has further been recognized that TLR are not only expressed by antigen-presenting cells, as originally thought, but also by cells that are not normally involved in host defense, but are involved in tissue damage. Thus TLR have a Janus-like aspect: They are beneficial by defending against microbes, but they may also be also deleterious by promoting tissue damage. Against this background it may not come as a complete surprise that TLR are also expressed in the kidney: Constitutive expression of TLR2 mRNA has been documented by Wolfs et al. (9) in tubular epithelial cells as well as in epithelial cells of Bowman’s capsule, and it had also been shown that TLR2 mRNA is upregulated by ischemia, although the endogenous cellular signals had remained undefined. A new twist is now provided by the experiments of Leemans et al., which document a causal role in acute renal dysfunction after ischemia reperfusion injury—and this may have ramifications far beyond acute renal failure. The investigators studied renal ischemia reperfusion injury comparing wild-type (TLR2+/+) and TLR2 knockout mice (TLR2−/−). To distinguish whether the beneficial effect seen in TLR2−/− mice was the result of deficient TLR2 expression by circulating blood cells or by intrinsic renal cells, they created and studied chimeric mice with deficient expression of TLR2 either by blood cells or by renal cells. Finally, to confirm the results by an independent methodological approach and to investigate a potential modality of intervention, they studied knockdown of TLR2 in cultured tubular epithelial cells as well as in vivo using antisense oligonucleotides. In a first step, the authors showed in primary cultures of tubular epithelial cells of TLR−/− mice that ischemia simulated by immersion into mineral oil caused less production of cytokines and chemokines, such as KC (granulocyte chemotactic keratinocyte chemoattractant), MIP-2 (macrophage inflammatory protein2), MCP1 (monocyte chemotactic monocyte chemoattractant protein-1), and IL-6. The same pattern was seen when the cultures were stimulated by homogenates of kidneys subjected to ischemia-reperfusion. This finding suggests that ischemia produces some stimulatory molecule(s). In a second step, the concentrations of these cytokines and chemokines in kidney homogenates were assessed after ischemia-reperfusion injury in wild-type and TLR2 knockout animals. In agreement with the results in the tubular cell cultures, the cytokine and chemokine concentrations were lower in the knockout mice. In addition, the influx of granulocytes and macrophages into the kidney was studied and again the transient infiltration was less pronounced in the TLR2 knockout mice. In a third step, the authors tried to provide evidence that deletion of TLR2 protected the function and the morphology of the kidney after ischemia-reperfusion injury. Indeed, lower early increase of serum creatinine or urea and less morphologic damage (tubular cell necrosis, tubular dilation, brush border loss, cast formation) were noted in TLR2 knockout mice. In addition, less caspase 3 as an index of apoptosis was detected by immunohistochemistry in TLR2 knockout mice. This finding is in line with recent observations that TLR2 activates apoptotic signaling pathways (10). More apoptosis in the wild-type mice was also associated with more cell regeneration as measured by incorporation of BrdU (brom-deoxy-uridine). In a fourth step, the authors compared mice in which the expression of TLR2 was selectively deficient in either blood cells only or in kidney cells only. Protection against injury, assessed as increased serum creatinine concentration, neutrophil infiltration, and apoptosis, was provided by absence of TLR2 in renal tubular epithelial cells, but not by absence of TLR2 in blood cells. In a final step, the potential protection against ischemia reperfusion injury was studied by administering antisense oligonucleotides in vivo to invalidate the TLR2 gene. These oligonucleotides were taken up by tubular epithelial cells and ameliorated the ischemia reperfusion injury as assessed by blood chemistry, apoptotic cells, neutrophil infiltration, and recovery of renal function. The results indicate that the proinflammatory cascade triggered by stimulation of TLR2 and presumably NFκB plays an important role in the genesis of acute renal failure after ischemia reperfusion injury. This role is all the more plausible as TLR2-mediated activation of NFκB and oxidative stress has been shown before in other organs, e.g., cardiomyocytes (11). What remains unclear, however, is which endogenous signal activates TLR2 in kidney cells. It has been shown that TLR2-stimulating endogenous ligands are released by necrotic cells, but not by apoptotic cells (12). Among the substantial number of molecules known to stimulate TLR2 (13), the authors discuss as candidates heat-shock protein 70 (14) and debris including cellular matrix components (15). It is also clear that the pathogenetic role of TLR2 (and of other members of the TLR family) in the genesis of renal disease may well extend beyond renal failure and comprise infection- and injury-associated renal damage and kidney diseases ranging from immune complex glomerulonephritis, e.g., infection-associated glomerulonephritis or lupus nephritis (15), to urinary tract infection (16,17) and malfunction of renal transplants (18–20). What is particularly intriguing with respect to the latter possibility is the observation that, in TLR2 knockout mice, skin allograft rejection was delayed, pointing to a role of TLR2 in transplantation immunology (20).