| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Medicine, Saint Louis University and St. Louis Veterans Affairs Medical Center, St. Louis; 2Departments of Urology and Pathology, New York University School of Medicine, New York, New York; 3Department of Biochemistry, Saint Louis University, St. Louis, Missouri; and 4Department of Medicine, Indiana University, Indianapolis, Indiana
Submitted 17 February 2008 ; accepted in final form 16 May 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
acute kidney injury; Toll-like receptors; uromodulin; ischemia-reperfusion
More recently, it was proposed that THP can activate Toll-like receptor 4 (TLR4) on myeloid dendritic cells (27). TLR4 is a member of the large Toll-like family of innate immune receptors (2). It was first recognized as the endotoxin receptor that mediates the inflammatory response in gram-negative sepsis. It was later shown to be activated by a myriad of endogenous ligands (1, 26). We and others have documented its presence in various segments of the tubular epithelium and its upregulation during sepsis (9, 10). We also recently demonstrated that a TLR4-dependent cyclooxygenase (Cox)-2 upregulation was specific to THP-expressing thick ascending limbs (TAL) during sepsis (11). Furthermore, renal TLR4 was shown to play a detrimental role in renal ischemia-reperfusion injury (IRI) by promoting inflammation and apoptosis (19, 35). Circumstantial evidence was provided to support a role for hyaluronan, biglycan, and high-mobility group box-1 as the TLR4-activating ligands in IRI (35). A role for THP in promoting TLR4-mediated inflammation and injury was not investigated in these studies.
In this study, we tested the hypothesis that THP plays a detrimental role in renal IRI by promoting inflammation and injury. To this end, we measured morphological and functional markers of injury in wild-type and THP-deficient mice. To our surprise, we found that THP in effect protects the kidney from IRI and its absence resulted in more severe inflammation, increased cast formation, and worse renal function. The absence of THP increased the expression of TLR4, and shifted the proximal tubular distribution of this receptor from the apical side to the cellular and basolateral compartments. We hypothesize that a basolateral distribution of TLR4 enhances its interaction with putative pro inflammatory ligands present in the interstitium during renal IRI. By shaping the nature of local innate immune responses in the kidney, THP potentially assumes a prominent role in determining the renal pathophysiology of infectious, inflammatory and other immunologic conditions.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
All animal experimentation was approved by the St. Louis VA Animal Care Committee and conducted in conformity with the Guiding Principles for Research Involving Animals and Human Beings. THP knockout animals (129/SvEv THP–/–) have been characterized in detail in previous publications (22, 23). Wild-type mice (129/SvEv) were purchased from Taconic. Animals were 8 wk old at the time of surgery. Ischemia-reperfusion surgery was as previously described (16). In brief, mice were anesthetized with isoflurane and placed on a homeothermic table to maintain core body temperature at 37°C. Both renal pedicles were exposed via a midline incision and occluded with microaneurysm clamps for clamp times depending on the animal injury group: mild (15 min, n = 6 for THP–/– and 6 for THP+/+), intermediate (22 min, n = 9 for THP–/– and 7 for THP+/+), and severe (30 min, n = 8 for THP–/– and 10 for THP+/+). Sham surgery (n = 3 for THP–/– and 3 for THP+/+) consisted of an identical procedure without application of the microaneurysm clamps. All animals received a subcutaneous analgesic (bruprenorphine 0.1 mg/kg) at the end of surgery. Creatinine was determined by standard picric acid reaction in serum obtained from maxillary vein bleeds immediately before (baseline) and 24 h after surgery.
Tissue Harvesting and Staining
Twenty-four hours after surgery, kidneys were perfusion-fixed in situ with 30 ml 4% paraformaldehyde in phosphate-buffered saline, pH 7.4; 100 µm sections for THP and TLR4 staining were obtained 24 h later with a vibratome. Alternatively, kidneys were dehydrated with increasing concentrations of ethanol and paraffin embedded for histological studies using light microscopy. Protocols for immunostaining vibratome sections were previously reported in detail (10).
We used the following antibodies for immunostaining: a goat polyclonal IgG primary for TLR4 (M-16, cat. no. sc-12511, Santa Cruz Biotechnology). For THP, we used a sheep polyclonal IgG (cat. no. 8595-0054, AbD Serotec). Macrophages were stained with a rat monoclonal antibody (F4/80 AbD Serotec). The following conjugated secondary antibodies were used: Alexa Fluor 555 donkey anti-sheep and goat anti-rat IgG, Alexa Fluor 647 chicken anti-goat IgG (cat. nos. A21436 [GenBank] , A21434 [GenBank] , and A21469, respectively, Molecular Probes). Brush borders were stained with Oregon green-488 phalloidin (cat. no. O7466, Molecular Probes), and nuclei were stained with 4,6-dimadino-2-phenylindoyl (DAPI; cat. no. D21490 [GenBank] , Molecular Probes). Negative controls were obtained by incubating tissues from sham and ischemia animals with secondary antibodies in the absence of primary antibodies.
Light Microscopy
Paraffin-embedded kidneys were sectioned at 4 µm and stained using hematoxylin and eosin (H&E) or periodic acid-Schiff (PAS). Injury scoring was based on three commonly used features of injury: cast formation, tubular dilation, and tubular necrosis. Quantitation was done as described previously (16). Tubules were counted in 10–12 fields each for cortex and outer medulla per section; fields were given an injury score from 0 to 5 for each injury feature based on the percentage of tubules within that field displaying that particular feature: 0–4% (0), 5–24% (1), 25–49% (2), 50–74% (3), 75–99% (4), and 100% (5). Average scores were calculated for each category (cast formation, tubular dilation, and tubular necrosis), and scores were reported for the mild injury group (IRI 15 min, n = 6 each for THP+/+ and THP–/–) and the severe injury group (IRI 30 min, n = 10 for THP+/+ and n = 8 for THP–/–). Estimates were performed on coded sections.
Modified esterase staining specific for neutrophil detection using the naphthol AS-D chloroacetate esterase kit (cat. no. 91C, Sigma) was performed according to the manufacturer's instructions. Neutrophils were counted per high-power field (x40 magnification), and a general average was computed for each animal group. Statistical analysis was done using a two tailed unpaired t-test at the 0.05 significance level.
Fluorescence Microscopy
A fluorescent terminal transferase-mediated dUTP-biotin nick end labeling reaction was performed on paraffin sections to detect apoptosis. We used an ApopTag Red In Situ Apoptosis Detection Kit (with DAPI) according to the manufacturer's instructions with the appropriate positive and negative controls (ca. no. S7165, Chemicon International). Sections were viewed with a Leica epifluorescent microscope using a x40 objective. Quantitation was done by counting positive apoptotic nuclei per high-power field (8–10 fields/section).
Vibratome sections (100-µm thickness) stained for THP and TLR4 were viewed under a Bio-Rad MRC1024 laser-scanning confocal microscope available at Saint Louis University. Images were collected using x60 magnification and analyzed with Zeiss LSM software and MetaMorph (Universal Imaging). Quantitation of TLR4 fluorescence was done as described in previous publications using total tubular fluorescence (10, 11). Previously described fluorescence-based operational definitions of kidney layers and structures were used to guide localizing proteins of interest within each kidney layer (11).
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We first examined histological kidney sections from sham mice of THP–/– and THP+/+ strains. The sections examined extended from the outer cortex to the medulla. Under light microscopy, there were no detectable differences in baseline morphology by PAS (Fig. 1) or H&E stains (not shown) between the two strains. Similarly, there was no significant difference in baseline serum creatinine between the two strains (0.25 mg/dl for THP+/+ vs. 0.28 mg/dl for THP–/–). This average value for each strain was obtained from the pooled baseline creatinine values of the sham, mild, intermediate, and severe injury groups, resulting in a n = 26 for each strain. All mice were 
8 wk old, and the average weight was 22 g for THP+/+ mice and 23 g for THP–/–.
|
We next studied the changes in THP expression with IRI in wild-type mice. Under sham conditions, THP was expressed predominantly at the apical side of medullary and cortical thick ascending limbs (Fig. 2). Twenty-four hours after mild IRI (15-min clamp), THP expression was significantly increased, especially in the outer and inner stripes of the outer medulla. The staining was very strong at the luminal border of TAL with frequent extension into the lumen Furthermore, there was significant cellular and basolateral staining, both not apparent under sham conditions. Finally, THP staining was also evident in the interstitial space close to S3 segments of the outer stripe (Fig. 2D) and in the interstitium of the inner stripe (Fig. 2F). Severe IRI (30-min clamp) caused significant necrosis in the outer medulla (discussed below). THP expression was also increased in this setting in the relatively few regions not affected by the necrotic process (data not shown).
|
We next examined overall renal injury after IRI in wild-type and THP–/– mice. Specifically, we quantified indices of tubular dilation, tubular necrosis, as well as cast formation. We studied IRI with different clamp times, because we believe that mild to moderate injury models (such as IRI with 15-min clamp time) can uncover important differences that would be otherwise hard to discern in the context of more severe damage. Figure 3 shows detailed high-power-field PAS sections of the outer medulla, where the injury is more pronounced. Quantitation of tubular dilation, necrosis, and cast formation is shown in Fig. 4.
|
|
Severe injury. IRI with 30-min clamp time produced significant injury in kidneys from both strains. Figure 3 shows that THP+/+ kidneys sustained significant injury, with marked tubular dilation, necrosis, and cast formation, especially in the outer medulla. Again, THP–/– kidneys had even more damage, with diffuse tubular necrosis in the outer medulla (outer and inner stripes) becoming a cardinal feature.
Serum creatinine. Serum creatinine values 24 h after IRI (15-, 22-, or 30-min clamp) were measured and compared with baseline in wild-type and THP–/– mice (Fig. 5). Creatinine in both types of mice significantly increased as clamp time was augmented from 15 to 22 and 30 min. THP-deficient mice had worse renal function, especially after a 22-min clamp. This is not surprising since serum creatinine is not an ideal marker for kidney damage and extremes of injury (mild or severe) may not be reflected by appropriate increments in creatinine. However, as seen in Fig. 5, the injury-creatinine dose-response curve of THP–/– mice is clearly shifted to the left compared with THP+/+.
|
A role for THP in promoting cast formation has been described extensively in the literature pertaining to acute tubular necrosis, but this is supported by limited experimental data. We therefore examined cast formation in our models of mild and severe injury in wild-type and THP–/– mice. As shown in Fig. 6, cast formation was equally prominent in both mice strains after severe IRI (Fig. 6, C–F). However, after mild IRI, cast formation was only significant in THP–/– mice (Fig. 6, A and B). Therefore, contrary to expectations, cast formation does not depend on the presence of THP. Rather, the presence of casts correlates better with the degree of injury, even in the absence of THP. Quantitation of cast formation was shown in Fig. 4.
|
We next examined whether the absence of THP affected the frequency of apoptosis, another cardinal feature of renal IRI (Fig. 7). Twenty-four hours after a 15-min clamp, there was no significant increase in the rate of apoptosis in either strain. In contrast, a 30-min clamp significantly increased renal apoptosis in both strains compared with sham conditions. Nevertheless, the increase in renal cell apoptosis in THP–/– mice was significantly less than that seen in wild-type mice. Thus the absence of THP is associated with a more necrotic (and less apoptotic) phenotype of cell death. Macrophages are known to phagocytose apoptotic cells (30). Preliminary studies showed that macrophage infiltration was not different between THP–/– and THP+/+ mice after IRI with a 30-min clamp, although there was a trend for an increased macrophage infiltration in THP–/– kidneys after 15-min clamp time (data not shown). Therefore, the discrepancy in detecting apoptosis between the two strains is not entirely explained by a difference in macrophage infiltration.
|
Inflammation plays an important role in the response of kidneys to ischemia-reperfusion. Therefore, we examined neutrophil infiltration in the medulla of kidneys harvested 24 h after sham or IRI from both mice strains (Fig. 8). In kidneys from wild-type and THP–/– mice, no neutrophils were detected under sham conditions. After 15-min clamp IRI, there was still no significant neutrophil infiltration in the medulla of wild-type mice (Fig. 8C). However, in THP-deficient kidneys, neutrophil infiltration was significantly increased (average 3.2/high-power field) (Fig. 8, D and H). This increase was seen predominantly in the outer stripe, in a patchy distribution between S3 segments and TAL (Fig. 8G). Very frequently, the tubules in those areas infiltrated with inflammatory cells were necrotic, especially S3 segments of proximal tubules (Fig. 8G, inset). Interestingly, we did not find significant neutrophil infiltration in the vasa recta areas of the outer medulla. With 30-min clamp time, both wild-type and THP–/– kidneys had a significant increase in neutrophil infiltration (Fig. 8, E and F). However, there was still a trend for a higher number of inflammatory cells in kidneys from THP–/– mice compared with wild-type (Fig. 8H).
|
The sensitivity of THP-deficient mice to ischemic injury is associated with increased inflammation, as discussed above. Inflammation can be regulated by innate immunity through Toll-like receptors. In fact, innate immunity has been shown to play a significant role in modulating the response of the renal epithelium to injury. Indeed, TLR2 and more recently TLR4 on renal cells have been shown to mediate ischemia-reperfusion injury in mice (20, 35). Furthermore, THP itself has been shown to modulate TLR4 signaling in dendritic cells (27). Therefore, we investigated whether the expression and localization of TLR4 are affected by IRI and the absence of THP (Fig. 9). In the outer medulla of THP /+ sham kidneys, TLR4 expression was mostly localized to the apical brush border of proximal tubules (S3 segments in the outer stripe). In contrast, TLR4 expression in sham kidneys of THP–/– mice was strikingly different: TLR4 expression was significantly increased in S3 segments, and was predominantly cellular and basal, with only occasional apical staining (compare Fig. 9, B and C, vs. Fig. 9, E and F). Only minimal TLR4 staining was seen in TAL. Figure 9, M and N, shows the quantitation of TLR4 expression.
|
Figure 10 shows costaining for THP and TLR4 in kidney sections from THP+/+ mice. TAL and S3 segments are contiguous to each other in the outer medulla. With ischemia, the expression of both proteins increases significantly. Although there is no significant colocalization within one tubule type, THP in the TAL and interstitium is present in very close proximity to TLR4 is S3 segments (arrowhead in Fig. 10H). With more severe ischemia, there was positive TLR4 staining in the lumen of S3 tubules, likely representing shed brush border and epithelial cells (Fig. 10, I–L).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We first showed that there were no histological or functional differences between wild-type and our strain of THP knockout mice under sham conditions. In a different THP knockout strain, other investigators showed that creatinine clearances normalized to kidney weight were different between wild-type and knockout (5). However, the number of animals used in these studies in each group was small and the differences were only apparent after normalization to kidney weights. In the current study, we used a sample size large enough to detect minor differences, if present, in baseline creatinine values. There was also no difference in body weight between our two groups, which is a better index of muscle mass.
THP expression in sham mice was similar to what was described by Bachmann and colleagues (3, 4) in several landmark publications, with localization restricted to the apical surface of thick limbs. However, only limited studies examined the effect of ischemia on THP expression. Using Northern blot and gene array techniques, two groups reported a decrease in renal THP mRNA after clamp times from 30 to 50 min (28, 36). However, we believe such clamp times result in profound necrosis, especially in the THP-expressing TAL segments. This was the rationale behind including more moderate clamp times in our studies. In our hands, a more relevant clamp time of 15 min resulted in a robust increase in THP expression by immunofluorescence microscopy. With 30-min clamps, THP expression was also increased but could be detected only in nonnecrotic regions. The increased expression of THP after ischemia likely has a protective role. This is supported by the finding that THP–/– mice had significantly worse injury for comparable clamp times. The localization of THP at the apical border as well as the cellular compartments and interstitium suggests a complex role for this protein.
While wild-type mice showed the typical increase in apoptosis after ischemia, it was surprising to see that ischemic kidneys from THP–/– mice showed less tubular apoptosis. However, these mice had significantly more severe morphological and functional injury, and it is possible that the extensive necrosis is in part secondary in nature. That is, in the setting of severe ischemia, many cells cannot complete the energy-requiring apoptotic program and thus undergo secondary necrosis (18). However, we failed to detect increased apoptosis in the knockout strain even with milder injury. Therefore, it is possible that THP possesses or conveys properties that reduce ischemia-induced cell necrosis but not apoptosis. This underscores the intrinsic differences between necrotic and apoptotic cell death as two discrete phenotypes of ischemic cell death that can be modulated separately (16, 17).
Another major finding in this study concerns the role of THP in cast formation. We show for the first time that the presence of THP is in fact not necessary for cast formation in ischemic renal injury. Rather, cast formation correlates with the degree of injury. Our findings thus question the relative importance that has been historically attributed to THP in this process (21, 25, 34). It is important to note, however, that our conclusions regarding THP and cast formation apply specifically to ischemic injury. Indeed, Sanders and colleagues (15, 29) have established a role for THP in promoting myeloma cast formation through precipitation with Bence Jones proteins.
Inflammation is an important contributor to the pathogenesis of kidney injury (7, 8). We measured neutrophil infiltration in this study as a marker of the inflammatory process. We demonstrated that in the absence of THP there is a marked neutrophil infiltration, especially in the interstitial space of the outer stripe between injured S3 segments and TAL. This leads us to hypothesize that the adjacent thick limbs and S3 segments, along with the surrounding interstitial space, could in fact represent a functional unit that controls the inflammatory response in ischemic injury. We also propose that THP is the mediator of cross talk between TAL and the S3 segments.
TLR4 is highly expressed in the kidney (9). Furthermore, renal TLR4 can mediate ischemic injury as was recently demonstrated (35). Therefore, an increased expression of TLR4 in THP–/– kidneys, as shown in our results, could explain the greater susceptibility of these kidneys to inflammation and ischemia. The cellular/basal localization of this receptor in THP–/– tubules could also indicate an increased activation level. Indeed, previous investigation suggested that stimulated TLRs accumulate in lipid rafts and are targeted to the Golgi apparatus (12, 13).
We have previously shown that TLR4 expression is dramatically increased in sepsis, specifically at the apical border of proximal tubules (10). In the current study, we show that ischemic injury, like sepsis, increases the expression of TLR4 predominantly at the apical side of S3 segments in THP+/+ kidneys. In sepsis, an apical location for TLR4 allows it to interact with filtered endotoxin, which is taken up by the proximal tubule (10). In ischemia, the significance of an apical location for TLR4 in THP+/+ mice is not obvious a priori. It is possible that apical TLR4 interacts with luminal endogenous ligands and thus contributes to injury. However, the magnitude of this contribution cannot be determined with certainty because of the extensive brush border shedding noted in S3 segments. Conversely, the apical expression of TLR4 in THP+/+ mice could represent a means of inactivation by priming this receptor for shedding during ischemia. The phenomenon of TLR4 urinary shedding, shown in Fig. 10, was also described recently by others (38).
In THP–/– mice, the basolateral distribution of TLR4 can also contribute to injury. For example, Wu and colleagues (35) demonstrated that endogenous ligands of TLR4 like hyaluronan are increased in the interstitial space following ischemic injury. These endogenous ligands can potentially interact with TLR4 expressed at the basolateral side of S3 cells and thus promote an inflammatory response (33). In fact, mice deficient in TLR4 had significantly less injury and inflammation after ischemia compared with a wild-type strain (35). Our studies support such a mechanism of injury. We show that the absence of THP in mutant mice increases basolateral expression of TLR4 after ischemia. This allows for a greater interaction with interstitial endogenous ligands and therefore results in more inflammation and injury. Even under sham conditions, the absence of THP seems to promote a cellular and basolateral localization of TLR4 as opposed to an apical one. This may imply a positive role for THP in determining the targeting of TLR4 in polarized epithelial cells. The mechanism by which THP affects TLR4 targeting remains unresolved. The presence of THP in TAL segments and the interstitium after ischemic injury suggests that both direct and indirect mechanisms could be involved.
In summary, we show that THP has an overall protective role in ischemic renal injury. Animals lacking THP have more tubular casts, increased inflammation and necrosis, and worse renal function. The mechanisms by which THP exerts its protective function are likely very complex but potentially involve enhanced expression of TLR4 and its targeting to the apical membrane of S3 segments. This apical location of TLR4 could minimize injury by decreasing the interaction of TLR4 with proinflammatory interstitial ligands released after ischemia.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. W. Good, T. George, and B. A. Watts III Lipopolysaccharide directly alters renal tubule transport through distinct TLR4-dependent pathways in basolateral and apical membranes Am J Physiol Renal Physiol, October 1, 2009; 297(4): F866 - F874. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. A. Ortiz Toll-like receptor 4 (TLR-4) regulates renal ion transport Am J Physiol Renal Physiol, October 1, 2009; 297(4): F864 - F865. [Full Text] [PDF]
|
|
|
|
 |
|
 B. Kruger, S. Krick, N. Dhillon, S. M. Lerner, S. Ames, J. S. Bromberg, M. Lin, L. Walsh, J. Vella, M. Fischereder, et al. Donor Toll-like receptor 4 contributes to ischemia and reperfusion injury following human kidney transplantation PNAS, March 3, 2009; 106(9): 3390 - 3395. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
