点击显示 收起
【摘要】 Toll-like receptors (TLRs) are now recognized as the major receptors for microbial pathogens on cells of the innate immune system. Recently, TLRs were also identified in many organs including the kidney. However, the cellular distribution and role of these renal TLRs remain largely unknown. In this paper, we investigated the expression of TLR4 in a cecal ligation and puncture (CLP) model of sepsis in Sprague-Dawley rats utilizing fluorescence microscopy. In sham animals, TLR4 was expressed predominantly in Tamm-Horsfall protein (THP)-positive tubules. In CLP animals, TLR4 expression increased markedly in all tubules (proximal and distal), glomeruli, and the renal vasculature. The staining showed a strong apical distribution in all tubules. A moderately less intense cellular signal colocalized partially with the Golgi apparatus. In addition, kidneys from septic rats showed increased expression of CD14 and THP. They each colocalized strongly with TLR4, albeit in different tubular segments. We also imaged the kidneys of live septic animals with two-photon microscopy after fluorescent lipopolysaccharide (LPS) injection. Within 10 min, LPS was seen at the brush border of some proximal tubules. Within 60 min, LPS was fully cytoplasmic in proximal tubules. Conversely, distal tubules showed no LPS uptake. We conclude that TLR4, CD14, and THP have specific renal cellular and tubular expression patterns that are markedly affected by sepsis. Systemic endotoxin can freely access the tubular and cellular sites where these proteins are present. Therefore, locally expressed TLRs and other interacting proteins could potentially modulate the renal response to systemic sepsis.
【关键词】 TLR CD TammHorsfall protein cecal ligation and puncture lipopolysaccharide
THE RECENT DISCOVERY of receptors for pathogen-associated molecular patterns, known as the Toll-like receptors (TLRs), has increased our understanding of the pathogenesis of sepsis ( 9 ). Indeed, these trans -membrane proteins are the primary mediators of the interaction between pathogens and cells of the innate immune system. In addition, TLRs are now emerging as important modulators of diverse pathophysiological conditions like ischemia, atherogenesis, and various immune-mediated diseases ( 29, 39, 50 ). This is because, in addition to microbial products, TLRs can be activated by endogenous ligands. These include debris from apoptotic and necrotic cells, HSP70, oligosaccharides, and possibly nucleic acid fragments ( 1, 39 ). TLRs are thus poised at a critical position in the pathogenesis of a variety of inflammatory conditions.
Among the 11 known TLRs, TLR4 and TLR2 have received the most scrutiny as they mediate responses to gram-negative and gram-positive bacteria, respectively ( 41 ). These receptors have been well characterized in cells of the immune system like macrophages and monocytes. Their activation by microbial ligands culminates in the secretion of a myriad of cytokines, chemokines, oxygen radicals, and lipid mediators. These, in turn, generate the inflammatory and procoagulant state characteristic of sepsis and possibly contribute directly to organ failure ( 9 ).
More recently, increasing attention has been given to the possible role of local, tissue-specific TLRs in modulating the response of organs to sepsis. Heart, liver, lung, and gut TLRs have been described as potential participants in the response of these organs to sepsis and other forms of injury ( 5, 33, 50 ). However, much less is known about TLRs in the kidney ( 2, 8 ). In particular, our knowledge of the exact cellular localization of TLRs along the nephron remains incomplete despite recent publications suggesting a functional role for these proteins in various models of renal injury ( 23, 25, 27, 28, 43, 46, 48 ). More importantly, while a role for renal TLRs in ascending urinary infections has been suggested ( 3, 14, 15, 44, 51 ), the accessibility of these receptors to systemically generated ligands like endotoxin is unknown ( 10 ).
In this paper, we investigated the hypothesis that the distribution of TLR4 along the nephron is affected by sepsis and that this important receptor is accessible to systemic endotoxin. To this end, we utilized the powerful techniques of confocal and two-photon microscopy to characterize the distribution of renal TLR4 and its changes with sepsis in the rat. We correlated these changes with those of two important proteins in sepsis, namely CD14 and Tamm-Horsfall protein (THP). We also determined the accessibility of renal TLRs to systemically generated endotoxin in live animals. Our findings show a remarkable response of these receptors to systemic sepsis and document for the first time their accessibility to systemic endotoxin. These studies will form the basis of future investigations into the role of TLR4 in the pathophysiology of the kidney.
MATERIALS AND METHODS
Animal surgery. All animal experimentation was approved by our University Animal Care Committee and conducted in conformity with the "Guiding Principles for Research Involving Animals and Human Beings."
Cecal ligation and puncture (CLP) is the preferred model for animal sepsis ( 16, 47 ). It results in a cytokine profile similar to that seen in human sepsis. It lacks the exaggerated TNF- levels observed in the LPS injection models ( 38 ). In brief, under halothane anesthesia, the cecum of Sprague-Dawley rats (200 to 250 mg body wt, Harlan, Indianapolis, IN) is ligated and punctured twice with an 18-gauge needle. The procedure is performed on a homeothermic pad with monitoring of O 2 saturation and blood pressure. The animal is allowed to recover, and blood is collected at baseline and 24-h intervals thereafter. Surviving animals are killed at 24, 48, or 72 h after CLP. Survival rate was 65% at 72 h.
In our hands, this model is characterized by a 50% reduction in glomerular filtration rate (GFR) that occurs between 24 to 48 h, with some recovery in the surviving rats. The histological changes 24 or 48 h after CLP are minimal and consist of vacuolization of tubular cells and very patchy architectural disruption. Very few casts were observed and white blood cell infiltration (esterase reaction) averaged one to two per field. Apoptosis in tubular cells increased up to 5% (TUNEL stain) compared with less than 1% in sham. These changes are similar to those reported by others ( 30 ). Of note is that the changes in protein expression after CLP between 24 and 48 h varied somehow between animals in different experiments. That is, some animals showed maximal changes 24 h after CLP while changes in others peaked more around 48 h after CLP. This represents variability in the response of individual animals to the CLP procedure. We also performed CLP using wild-type and TLR4 knockout mice to validate our antibody staining. In mice, we used a 25-gauge needle to puncture the cecum. The procedure was otherwise identical to that done on rats.
Tissue harvesting. After harvesting, kidneys were cut into small segments and fixed in 4% paraformaldehyde overnight. They were then preserved for 3 to 5 days in 30% sucrose before 10-µm cryosections were obtained. Alternatively, kidneys were perfusion-fixed in situ with 100 ml 4% paraformaldehyde and 100-µm sections were obtained 24 h later with a vibratome. Both methods yielded similar immunofluorescence staining results. The protocols for immunostaining cryosections and vibratome sections were as we reported previously in detail ( 19, 22, 36 ).
We used the following antibodies from Santa Cruz Biotechnology: goat polyclonal IgG primary for TLR4 (L-14. cat. no. sc-16240), rabbit polyclonal IgG primary for TLR4 (H-18. cat. no. sc-10741), rabbit polyclonal IgG for CD14 (M305, cat. no. sc-9150), and rabbit polyclonal IgG for the Golgi marker Rab 6 (C-19, cat. no. sc-310). For THP, we used a rabbit polyclonal IgG (cat. no. 8595-0004, Biogenesis). For improved detection of TLR4 with the primary goat antibody, we used an unconjugated secondary rabbit anti-goat antibody (Zymed labs) and a Cy5-conjugated tertiary donkey anti-rabbit IgG (Jackson Immunoresearch Labs). The blocking peptide for the primary goat TLR4 antibody was also from Santa Cruz Biotechnology (cat. no. sc-16240 P). For the primary rabbit TLR4 antibody, we used an unconjugated goat anti-rabbit followed by a Cy5 tertiary donkey anti-goat antibodies. For CD14, THP and Rab 6 only Cy5 or Texas Red-labeled secondary antibodies were used. During costaining of TLR4 with CD14, THP or Rab 6, only a Cy5-labeled secondary donkey anti-goat IgG was used without a tertiary antibody. Brush borders were stained with FITC-phalloidin and nuclei were stained with DAPI. Negative controls were obtained by incubating tissues from sham and CLP animals with secondary and/or tertiary antibodies, in the absence of primary antibodies.
Validation of immunofluorescence staining for TLR4. To confirm the specificity of our goat TLR4 primary antibody staining, we performed a series of validation procedures. First, we ensured the lack of nonspecific staining by the secondary and tertiary antibodies. This was done by incubating sham and CLP tissues with the secondary and tertiary or the tertiary antibody alone, in the absence of the primary antibody ( Fig. 1, NC1 and NC2). Both negative controls did not show any nonspecific staining (compare with TLR4 goat in which the primary was included). Next, to ensure the lack of nonspecific binding of our primary goat antibody, we stained the tissues in the presence of a specific blocking peptide. Again, no significant staining was seen (BP). Furthermore, to determine whether we can reproduce the same staining with a completely different antibody, we used a rabbit TLR4 antibody raised against a different TLR4 epitope. Both primaries showed a very similar staining pattern (compare TLR4 rabbit and TLR4 goat), although the two antibodies were raised in different animal species and recognize different amino acid epitopes. Finally, we confirmed the specificity of our goat antibody to TLR4 by the absence of staining in tissues obtained from mutant mice with a deletion in the TLR4 region (C57BL/10ScNJ) compared with wild-type background mice C57BL/10ScSnJ (both from Jackson Labs; Fig. 1, TLR4 -/- and TLR4 +/+).
Fig. 1. Validation of immunofluorescence staining for TLR4. Representative kidney sections harvested 24 h after cecal ligation and puncture (CLP; x 40 objective). Nuclei stained with DAPI are shown in blue. Proximal tubules showing typical green autofluorescence are marked P, whereas distal tubules are marked D. NC1 : negative control in which tissue was stained with secondary and tertiary antibodies in the absence of a primary antibody. NC2 : negative control in which tissue was stained with tertiary antibody in the absence of secondary and primary antibodies. BP : tissue stained with primary, secondary, and tertiary antibodies in the presence of a blocking peptide for the primary goat TLR4 antibody. Tertiary TLR4 staining (shown in red) with either goat or rabbit primary antibodies are shown, respectively. TLR4 +/+ and TLR4 -/- represent staining with goat primary antibody of tissues from wild-type and TLR4-deleted mutant mice, respectively.
Western blot analysis. In some experiments, the kidney was removed without fixation and proteins were extracted immediately from the cortex or medulla using standard techniques. Proteins were quantitated, loaded equally (40 µg/lane), and resolved on a 10% Tris·HCl gel. Western blots for TLR4 were performed as previously described ( 20 ). The probing goat antibody and blocking peptide for TLR4 were the same as used for immunofluorescence.
Confocal microscopy. All images from fixed tissues were collected with a Zeiss LSM 510 confocal microscope and analyzed with Zeiss LSM software and MetaMorph (Universal Imaging) ( 21 ). Quantitation of TLR4 and CD14 signal intensities was done using Metamorph version 5.0. Representative fields in sham and CLP-operated rats were analyzed using at least six tubules from each field. Apical areas were sampled from each tubule and their intensities were averaged by the software. Corresponding areas within the cells of each tubule were also sampled and averaged for intensity. Furthermore, the average intensity from negative control image was calculated and subtracted from each reading to adjust for the background. Statistical analysis was done using a two-tailed unpaired t -test at the 0.05 level of significance. Colocalization of signals was done after thresholding the image in each channel, selecting apical and cellular areas, and applying the colocalization function of Metamorph software.
Intravital two-photon microscopy. Imaging of the kidney in the live animal was performed using a Bio-Rad MRC-1024MP Laser-Scanning Confocal/Multiphoton Scanner (Hercules, CA) as previously described in detail ( 11, 21 ). After initial imaging of sham or CLP animals, 500 µg of the nuclear dye Hoechst 33342 were administered intravenously over 2 min to allow identification of all cells and examination of nuclear morphology. Alexa-568 conjugated with LPS (Molecular Probes was injected via tail vein; 300 µg) as a tracer for endogenous LPS. Imaging began few minutes after injection. In some experiments, 1.6 mg FITC-conjugated dextran (molecular weight 3 kDa) were also injected via tail vein.
RESULTS
Effect of CLP on the expression of TLR4 by Western blot analysis. We first studied the presence and changes in TLR4 expression by Western blot analysis ( Fig. 2 ). TLR4 was expressed in the cortex and medulla of sham kidneys at baseline (CLP 0 h). The double bands are composed of the native protein and its higher molecular weight glycated form ( 48 ). TLR4 expression in the cortex and medulla increases after CLP and returns toward baseline in surviving animals at 72 h. Incubation of the blot with a blocking peptide completely abolished the TLR4 bands thus confirming the specificity of our primary antibody.
Fig. 2. Effect of CLP on the expression of TLR4 by immunoblot analysis. Western blots of protein extracts from cortex and medulla of kidneys obtained from rats subjected to sham surgery (0 h) or 24, 48, and 72 h after CLP surgery. Blots are representative of n = 5. In the blot labeled BP, cortical extracts from 24-h CLP were probed with the primary antibody in the presence of a blocking peptide. All lanes were loaded equally with 40 µg protein.
Localization of TLR4 by immunofluorescence microscopy. In kidneys from sham animals, TLR4 showed a strong signal in distal tubules ( Fig. 3 A ). There was a significantly weaker signal from proximal tubules and peritubular capillaries. After CLP, TLR4 staining in proximal tubules increased significantly along with the appearance of peritubular capillary staining ( Fig. 3 B ). In both proximal and distal tubules, high-magnification images show that TLR4 staining had a strong apical component but also extended well into the cytoplasm ( Fig. 3 C ). Kidneys from CLP animals also showed TLR4 staining in glomerular capillary tufts and the vascular pole ( Fig. 3 D ). Low-magnification images show positive TLR4 staining in both cortical ( Fig. 3 E ) and medullary ( Fig. 3 F ) regions of kidneys from CLP rats.
Fig. 3. Localization of TLR4 by immunofluorescence microscopy after sham surgery or CLP. Representative kidney sections obtained from sham-operated rats ( A ) or 24 h after CLP ( B, C, D, E, and F ). Nuclei are shown in blue with DAPI fluorescence. Green fluorescence represents proximal tubular autofluorescence. TLR4 immunostaining was pseudocolored red. Distal tubules (D) lack the green autofluorescence characteristic of proximal tubules (P). B : arrows point to increased TLR4 fluorescence in proximal tubules. Note the lighter green autofluorescence of proximal tubules in B compared with A. It results from the overlap of red cellular TLR4 signal with the green autofluorescence. Arrowheads point to red TLR4 fluorescence in peritubular capillaries. C : high-power view. D : arrowhead points to TLR4 staining at the vascular pole of a glomerulus (G). Note also the positive TLR4 red signal within the glomerular capillary tuft. E and F : x 10 views and represent cortical and medullary sections, respectively. E : arrowheads point to TLR4 staining of glomerular vascular poles.
Costaining of TLR4 and proximal tubular brush border. To further characterize the TLR4 signal, we costained the proximal tubular brush border with FITC-phalloidin. In addition to confirming the identity of proximal tubules (so far identified solely on the basis of their green autofluorescence), the staining delineates the extent of the brush border and thus permits more exact characterization of the apical TLR4 signal. In sham animals ( Fig. 4 A ), TLR4 staining indeed was predominantly localized to distal tubules, whereas the proximal tubules showed a less intense cellular signal. The dark green brush border indicates minimal costaining with TLR4. Figure 4 A tubule marked with an * is shown in high magnification in Fig. 4 C to illustrate this point. In contrast, kidneys from CLP rats showed a significant increase in TLR4 staining that extended throughout the cell into the tips of the brush border ( Fig. 4, B and D ). Figure 5 shows the magnitude of the increase in apical and cellular TLR4 fluorescent signal in CLP rats compared with sham. The increase in both apical and cellular TLR4 fluorescence was significant in CLP rats compared with sham.
Fig. 4. Costaining for TLR4 and the proximal tubular brush border in sham-operated rats and after CLP. Representative kidney cortex sections obtained from sham-operated ( A and C ) or CLP rats ( B and D ). Green fluorescence represents FITC-phalloidin staining of the brush border. Red fluorescence represents TLR4 immunostaining. A : sham surgery x 40, TLR4 staining is seen in distal tubules (D) and to a lesser extent in the cellular compartment of proximal tubules (P). Note the brush border dark green FITC fluorescence indicating little costaining with red TLR4. C : tubule marked with * is shown in high magnification. B : CLP rats x 40, note the yellowish hue of the brush border indicating costaining with red TLR4 which extends well into the tip of the brush border ( D, arrowheads). Note also the increased cellular red TLR4 signal in the cellular compartment of proximal tubules.
Fig. 5. Quantitation of TLR4 fluorescence in sham-operated rats and after CLP. Values represent means ± SE. At least 6 tubules from representative fields per experimental condition were measured. Fluorescence intensity of TLR4 staining was measured separately at the apical border and the cellular compartment. *Statistical significance with P < 0.01 (unpaired t -test) when apical and cellular fluorescence was compared between sham and CLP.
Costaining of TLR4 and THP. THP, an abundant protein in the kidney, has been recently reported to modulate TLR4 signaling ( 42 ). In addition, it is a good marker to characterize the thick ascending loop and early distal tubules. Therefore, we examined the changes in THP staining with sepsis and its colocalization with TLR4. Figure 6 shows the thin cellular and apical staining of THP of sham kidneys ( B ). In kidneys from CLP rats, THP was dramatically increased in quantity and extended well into the lumen of the tubules, occasionally causing complete obstruction ( Fig. 6 E ). The costaining for TLR4 and THP is also shown in this figure. In sham animals, 94% of the TLR4 signal colocalized with THP in THP positive (T+) distal tubules ( Fig. 6, A, B, and C ). In kidneys from CLP animals ( Fig. 6, D, E, and F ), both TLR4 and THP increased and 99% of the TLR4 signal colocalized with THP in T+ tubules. Note also the strong TLR4 signal in THP negative (T-) tubules and in proximal tubules. Thus the increase in TLR4 fluorescence in CLP animals involves all tubular segments (proximal, distal T+, and distal T-) and does colocalize strongly with THP in T+ tubules.
Fig. 6. Immunofluorescence staining of TLR4 and THP in sham and CLP-operated rats. Representative x 40 cortical kidney sections from sham rats ( A, B, and C ) and 48 h after CLP ( D, E, and F ). TLR4 fluorescence is pseudocolored red, THP fluorescence yellow and brush border staining with FITC-phalloidin in green. Colocalization of TLR4 and THP imparts an orange hue to the involved areas in the merged pictures ( C and F ). T- denotes THP-negative tubules. Two THP-positive tubules are labeled with *. Arrowheads point to proximal tubules.
Localization of CD14 with immunofluorescence microscopy. CD14 acts as an adaptor protein for ligands of TLR4 and has also been implicated in TLR4-independent endotoxin uptake by cells ( 12 ). Therefore, we examined the renal distribution of CD14 in sham and CLP rats. In sham kidneys, CD14 was expressed predominantly at the apical border of proximal tubules ( Fig. 7 A ). CLP caused a significant increase in the CD14 signal that was still localized to the apical border of proximal tubules ( Fig. 7 B ). In both sham and CLP kidneys, CD14 showed a faint (but detectable) intracellular signal. We also confirmed the localization of CD14 to proximal tubules by costaining with FITC-phalloidin in sections from sham ( Fig. 7 C ) and CLP ( Fig. 7 D ) rats. High magnification revealed CD14 to be primarily localized basal to the brush border ( Fig. 7 D, inset ). Thus, unlike TLR4, CD14 signal did not extend throughout the brush border (note the discrete red and green fluorescence in Fig. 7 D, inset and compare with the full brush-border extension of TLR4 in Fig. 3 D ). Low-magnification views of sections from CLP kidneys show CD14 expressed predominantly in the cortex ( Fig. 7 E ) with minimal staining in the medulla ( Fig. 7 F ). The quantitative changes in CD14 fluorescence between sham and CLP kidneys are shown in Fig. 8.
Fig. 7. Localization of CD14 by immunofluorescence microscopy after sham surgery or CLP. Representative kidney sections from sham ( A and C ) and 24-h CLP rats ( B, D, E, and F ). A and B : green represents proximal tubule (P) autofluorescence. Arrowheads point to apical red fluorescence of CD14. C and D : nuclei are stained blue with DAPI. Proximal tubular (P) brush border is stained green with FITC-phalloidin. CD14 is shown in red. Distal tubules are labeled D. Tubule labeled with * in D is shown also in inset with distinct green (arrow) and red (arrowhead) fluorescence. E and F : x 10 views of cortex and medulla, respectively, stained for CD14 (red).
Fig. 8. Quantitation of CD14 fluorescence in sham-operated rats and after CLP. Values represent means ± SE. At least 12 tubules from representative fields per experimental condition were measured. Fluorescence intensity of CD14 staining was measured separately at the apical border and the cellular compartment. *Statistical significance with P < 0.01 (unpaired t -test) when apical fluorescence was compared between sham and CLP.
Finally, we determined whether CD14 colocalized with TLR4 in the apical border and cellular compartments of proximal tubular cells. As shown in Fig. 9, there was indeed colocalization between the two signals. At the apical border, 70% of the CD14 signal colocalized with TLR4. Conversely, only 48% of the apical TLR4 signal colocalized with CD14. The remaining 52% likely represent the differential extension of TLR4 staining throughout the brush border (compare Fig. 4 D with Fig. 7 D, inset ). This differential extension of TLR4 is not well resolved under costaining conditions (see MATERIALS AND METHODS ). In the cellular compartment, 30% of the CD14 signal colocalized with TLR4.
Fig. 9. Colocalization of TLR4 with CD14. A representative cortical kidney section is shown from a 24-h CLP rat, costained for TLR4 (yellow) and CD14 (red). In the merged picture ( C ), nuclei are also shown stained with DAPI (blue). C : orange hue represents areas of colocalization of TLR4 and CD14 (arrowheads).
Colocalization of cellular TLR4 and the Golgi apparatus. TLR4 (bound and unbound to endotoxin) has been reported to localize to the Golgi apparatus in various cell types in culture. We examined whether this is also the case in renal tubular cells ( Fig. 10 ). In sham kidneys, 12% of cellular TLR4 fluorescence colocalized with Golgi fluorescence ( Fig. 10, A, B, and C ). In CLP kidneys, 59% of the TLR4 signal colocalized with the Golgi ( Fig. 10, D, E, and F ). Note that Golgi staining itself was altered by CLP, becoming more intense and dispersed ( Fig. 10, B and D ).
Fig. 10. Colocalization of cellular TLR4 with Golgi. Representative kidney sections from sham ( A, B, and C ) and 24-h CLP-operated rats ( D, E, and F ). TLR4 is shown in the yellow channel. Golgi is shown in the red channel. C and F : orange hue indicates colocalization of TLR4 with Golgi.
Intravital imaging of fluorescent endotoxin in kidneys of control and septic rats. Having localized the expression of TLR4 in kidneys of CLP rats, we next examined whether systemic endotoxin has access to this receptor in live septic and control rats. To this end, we imaged the kidneys of live rats with two-photon microscopy after the injection of tracer fluorescent endotoxin. In septic rats, fluorescent endotoxin was seen at the brush border of some proximal tubules within 10 min of its injection ( Fig. 11 A ). It was also observed in peritubular capillaries. Twenty-two minutes after its injection, fluorescent endotoxin was seen in a more subapical compartment in some proximal tubules. A strong signal was seen in the lumen of distal tubules but without any cellular uptake ( Fig. 11 B ). Fifty-seven minutes after its injection, fluorescent endotoxin was totally intracellular ( Fig. 11 C ).
Fig. 11. Intravital imaging of fluorescent LPS in rats after CLP surgery. Representative views of kidney fields from live septic animals obtained with 2-photon microscopy. A - D : time after intravenous injection of fluorescent LPS (red color). Proximal tubules (P) have brown autofluorescence. Distal tubules (D) lack autofluorescence. Nuclei are stained blue with intravenous injection of Hoechst. A : arrows point to red LPS fluorescence in peritubular capillaries. D : yellow/green color represents 3-kDa FITC-dextran.
The differential uptake of endotoxin by different proximal tubular segments raised the possibility of segmental hypoperfusion and/or hypofiltration, as could occur with sepsis. To examine this possibility, we infused fluorescent low-molecular-weight dextran (3 kDa) that is readily filtered. Within minutes, dextran was seen in the lumen of several proximal tubules that showed differential endotoxin uptake ( Fig. 11 D, compare red color in proximal tubule marked with * to that seen in the 2 adjacent proximal tubules. All 3 tubules showed similar luminal dextran fluorescence.). Therefore, the differential uptake of endotoxin by proximal tubules is not related to variations in filtration of the corresponding nephrons.
In nonseptic animals that underwent sham surgery, fluorescent endotoxin was also taken up by some proximal tubules, albeit with a much lower frequency. On average, the number of fields containing endotoxin-positive proximal tubules was 10% of the number seen in CLP rats. In sham rats, fluorescent endotoxin was also noted in the lumen of distal tubules with similar intensity and distribution as in CLP rats.
DISCUSSION
In this paper, we provide the first extensive characterization of TLR4 in kidneys from control rats and after CLP-induced sepsis. We show a remarkable change in the expression level and distribution of these receptors after injury. In addition, they are also accessible to systemic bacterial ligands as we demonstrate in this paper. This suggests a role for these local TLRs that extends beyond interacting with pathogens in the setting of an ascending urinary tract infection.
The potential of TLRs to modulate the renal response to injury is supported by recent data. Indeed, Cunningham et al. ( 10 ) showed that TLR4-positive kidneys transplanted into TLR4 knockout mice do sustain injury after systemic LPS injection. Furthermore, Leemans et al. ( 27 ) convincingly provide data supporting an important role for TLR2 in mediating ischemic injury to the kidney. Finally, Lim et al. and Wolfs et al. ( 23, 28, 48 ) point to possible activation of renal innate immunity through upregulation of TLR4 and TLR2 after ischemia or cyclosporine-induced injury. These data underscore the functional importance of TLRs in various models of renal injury. They also stress the need for a more detailed knowledge of the cellular distribution of these proteins in the kidney and their response to injury. In this paper, we provide the first such characterization of TLR4 distribution and changes in a relevant sepsis model in the rat.
Previously, other investigators have documented the presence of TLR4 in kidney tissues from various species. In humans, TLR4 from renal tissues was detected by two groups using RT-PCR techniques ( 34, 49 ). Conversely, Backhed et al. ( 4 ) failed to detect TLR4 mRNA in isolated human tubular cells. The reasons for these discrepancies are unknown but could be related to the specific cellular origin of the amplified signal. Using immunohistochemical techniques, Samuelsson et al. ( 43 ) did show TLR4 staining in tubules from human renal biopsies. However, no specific markers for various tubular segments were used in this study. In mice, Tsuboi et al. ( 46 ) detected TLR4 mRNA in isolated tubular epithelial cells with Northern blot analysis. Similarly, Wolfs et al. ( 48 ) detected TLR4 mRNA with in situ hybridization in mice renal cortex and to a lesser degree in the medulla. In the rat, Laestadius et al. ( 25 ) demonstrated the presence of TLR4 mRNA with RT-PCR in rat proximal tubular cells. More recently, Lim et al. ( 23, 28 ) used in situ hybridization to detect TLR4 mRNA in distal tubules in the rat kidney. TLR4 expression increased in both distal and proximal tubules after ischemia or cyclosporine toxicity. Our characterization of TLR4 distribution in sham and septic rat kidneys with immunofluorescence microscopy extends these studies to provide more specific localization of this protein at the cellular and subcellular levels.
In sham animals, TLR4 was detected predominantly in THP-positive tubules and to a lesser degree in proximal tubules. The function of TLR4 in normal, nonseptic states is completely unknown. In septic rats, TLR4 was dramatically increased in proximal tubular cells and was present both in the intracellular compartment and throughout the brush border. This apical presence is optimal for interaction with filtered endotoxin and subsequent internalization of the receptor-ligand complex. It is also possible that proximal tubular endotoxin signaling itself upregulates the expression of TLR4 in these segments. The intracellular colocalization with the Golgi apparatus is similar to that described by others in nonrenal tissues ( 17, 26 ). Sepsis also increased TLR4 in THP-positive tubules, where its colocalization with THP exceeded 90%.
Recently, THP was shown to stimulate the maturation of myeloid dendritic cells via interaction with TLR4 ( 42 ). It is very tempting therefore to speculate that THP can also stimulate local tubular TLR4 with which it colocalizes so strongly. This signaling could participate both in the proinflammatory properties of THP and in its role in the antibacterial defense of the urinary tract ( 6, 35, 45 ). Whether THP also prevents the uptake of LPS by distal tubules, as shown in our in vivo imaging studies, is unclear. Finally, we note that our CLP model is characterized by very little cast formation. Nevertheless, THP was dramatically increased in its distal sites. Thus our studies hint to a possible complex functional role for THP that extends well beyond its usual association with tubular casts formation. Such role, in health and disease, remains to be elucidated.
In the live imaging studies, we show that systemic endotoxin has rapid access to all the renal sites where TLRs are expressed. While others have detected endotoxin in renal tissues with radioactive techniques or electron microscopy ( 18, 32 ), ours is the first demonstration of the kinetics of LPS uptake by renal cells in live animals. This uptake of endotoxin could in itself have a detrimental effect on tubular cells, independent of TLR4 signaling. Indeed, in vivo studies have shown that the presence of LPS along with TNF- (itself a product of TLR4 signaling) results in a higher mortality than the administration of TNF- alone ( 40 ). Thus the fate of internalized LPS and its direct effects on cell function deserves further investigation.
The uptake of LPS was strictly localized to proximal tubules. The reason for the varying uptake rate of LPS in adjacent proximal tubules is unclear but does not seem to be related to differential filtration as shown with the dextran studies. One possibility is intrinsic differences in the uptake machinery between proximal tubular subsegments. This was not pursued further in this paper. Of interest is the fact that none of the distal tubules showed any LPS uptake. The presence of THP in some of these tubules might explain this lack of uptake as discussed above. However, it is unclear why THP-negative tubules also failed to take-up LPS. One possibility is the minimal expression of CD14 in all nonproximal tubules as discussed below.
Among the proteins necessary for TLR4 signaling, CD14 plays a central role and was in fact believed to be the endotoxin receptor before the TLRs era ( 13 ). Now, its function as an adaptor protein is well recognized, along with a possible role in TLR-independent LPS uptake ( 9, 24 ). Our results show a strong increase in CD14 staining with sepsis at the apical border of proximal tubules. There was also significant colocalization between TLR4 and CD14 in proximal tubules, a fact that supports the swift uptake of LPS in these tubules. In contrast, there was little if any CD14 staining in all distal tubules. This might explain the lack of LPS in these tubules, including the THP-negative ones. Whether our CD14 signal represents membrane-bound CD14 or soluble CD14, filtered from the circulation, is unknown. The cellular localization of CD14 under the proximal tubular brush border favors the former possibility. In humans, there is evidence that membrane-bound CD14 is not present in renal tubular cells ( 7, 43 ). However, urinary soluble CD14 increases markedly with sepsis and inflammatory conditions ( 7, 9, 37 ). Both membrane-bound and -soluble CD14 are adequate for LPS signaling and uptake. In mice, CD14 was reported to increase in both cortex and medulla in obstruction and ischemia models of acute renal failure ( 31 ).
In conclusion, we characterized the cellular distribution of TLR4 in kidneys from sham and septic rats. We have also investigated the localization of THP and CD14, two potentially important proteins in renal TLR4 signaling. Our data support a role for TLR4 and these related proteins in modulating the renal response to sepsis. This is further underscored by our findings in live septic animals showing significant uptake of systemic endotoxin by renal tubular cells. This study can thus form the basis of future functional and histological investigations into the role of TLR4 in various models of renal injury.
GRANTS
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant 1RO1-DK-60495-01A1 (P. C. Dagher) and a National Kidney Foundation-Indiana grant (T. M. El-Achkar).
ACKNOWLEDGMENTS
We are grateful to B. Molitoris for help and support throughout this project.
【参考文献】
Akira S and Takeda K. Toll-like receptor signalling. Nat Rev Immunol 4: 499-511, 2004.
Anders HJ, Banas B, and Schlondorff D. Signaling danger: toll-like receptors and their potential roles in kidney disease. J Am Soc Nephrol 15: 854-867, 2004.
Backhed F, Meijer L, Normark S, and Richter-Dahlfors A. TLR4-dependent recognition of lipopolysaccharide by epithelial cells requires sCD14. Cell Microbiol 4: 493-501, 2002.
Backhed F, Soderhall M, Ekman P, Normark S, and Richter-Dahlfors A. Induction of innate immune responses by Escherichia coli and purified lipopolysaccharide correlate with organ- and cell-specific expression of Toll-like receptors within the human urinary tract. Cell Microbiol 3: 153-158, 2001.
Basu S and Fenton MJ. Toll-like receptors: function and roles in lung disease. Am J Physiol Lung Cell Mol Physiol 286: L887-L892, 2004.
Bates JM, Raffi HM, Prasadan K, Mascarenhas R, Laszik Z, Maeda N, Hultgren SJ, and Kumar S. Tamm-Horsfall protein knockout mice are more prone to urinary tract infection: rapid communication. Kidney Int 65: 791-797, 2004.
Bussolati B, David S, Cambi V, Tobias PS, and Camussi G. Urinary soluble CD14 mediates human proximal tubular epithelial cell injury induced by LPS. Int J Mol Med 10: 441-449, 2002.
Chowdhury P, Sacks SH, and Sheerin NS. Minireview: functions of the renal tract epithelium in coordinating the innate immune response to infection. Kidney Int 66: 1334-1344, 2004.
Cohen J. The immunopathogenesis of sepsis. Nature 420: 885-891, 2002.
Cunningham PN, Wang Y, Guo R, He G, and Quigg RJ. Role of Toll-like receptor 4 in endotoxin-induced acute renal failure. J Immunol 172: 2629-2635, 2004.
Dunn KW, Sandoval RM, Kelly KJ, Dagher PC, Tanner GA, Atkinson SJ, Bacallao RL, and Molitoris BA. Functional studies of the kidney of living animals using multicolor two-photon microscopy. Am J Physiol Cell Physiol 283: C905-C916, 2002.
Dunzendorfer S, Lee HK, Soldau K, and Tobias PS. TLR4 is the signaling but not the lipopolysaccharide uptake receptor. J Immunol 173: 1166-1170, 2004.
Fearns C, Kravchenko VV, Ulevitch RJ, and Loskutoff DJ. Murine CD14 gene expression in vivo: extramyeloid synthesis and regulation by lipopolysaccharide. J Exp Med 181: 857-866, 1995.
Frendeus B, Wachtler C, Hedlund M, Fischer H, Samuelsson P, Svensson M, and Svanborg C. Escherichia coli P fimbriae utilize the Toll-like receptor 4 pathway for cell activation. Mol Microbiol 40: 37-51, 2001.
Hedlund M, Frendeus B, Wachtler C, Hang L, Fischer H, and Svanborg C. Type 1 fimbriae deliver an LPS- and TLR4-dependent activation signal to CD14-negative cells. Mol Microbiol 39: 542-552, 2001.
Heyman SN, Lieberthal W, Rogiers P, and Bonventre JV. Animal models of acute tubular necrosis. Curr Opin Crit Care 8: 526-534, 2002.
Hornef MW, Frisan T, Vandewalle A, Normark S, and Richter-Dahlfors A. Toll-like receptor 4 resides in the Golgi apparatus and colocalizes with internalized lipopolysaccharide in intestinal epithelial cells. J Exp Med 195: 559-570, 2002.
Kang YH, Falk MC, Bentley TB, and Lee CH. Distribution and role of lipopolysaccharide in the pathogenesis of acute renal proximal tubule injury. Shock 4: 441-449, 1995.
Kelly KJ, Plotkin Z, and Dagher PC. Guanosine supplementation reduces apoptosis and protects renal function in the setting of ischemic injury. J Clin Invest 108: 1291-1298, 2001.
Kelly KJ, Plotkin Z, Vulgamott SL, and Dagher PC. P53 mediates the apoptotic response to GTP depletion after renal ischemia-reperfusion: protective role of a p53 inhibitor. J Am Soc Nephrol 14: 128-138, 2003.
Kelly KJ, Sandoval RM, Dunn KW, Molitoris BA, and Dagher PC. A novel method to determine specificity and sensitivity of the TUNEL reaction in the quantitation of apoptosis. Am J Physiol Cell Physiol 284: C1309-C1318, 2003.
Kelly KJ, Sutton TA, Weathered N, Ray N, Caldwell EJ, Plotkin Z, and Dagher PC. Minocycline inhibits apoptosis and inflammation in a rat model of ischemic renal injury. Am J Physiol Renal Physiol 287: F760-F766, 2004.
Kim BS, Lim SW, Li C, Kim JS, Sun BK, Ahn KO, Han SW, Kim J, and Yang CW. Ischemia-reperfusion injury activates innate immunity in rat kidneys. Transplantation 79: 1370-1377, 2005.
Kitchens RL, Wang P, and Munford RS. Bacterial lipopolysaccharide can enter monocytes via two CD14-dependent pathways. J Immunol 161: 5534-5545, 1998.
Laestadius A, Soderblom T, Aperia A, and Richter-Dahlfors A. Developmental aspects of Escherichia coli-induced innate responses in rat renal epithelial cells. Pediatr Res 54: 536-541, 2003.
Latz E, Visintin A, Lien E, Fitzgerald KA, Monks BG, Kurt-Jones EA, Golenbock DT, and Espevik T. Lipopolysaccharide rapidly traffics to and from the Golgi apparatus with the toll-like receptor 4-MD-2-CD14 complex in a process that is distinct from the initiation of signal transduction. J Biol Chem 277: 47834-47843, 2002.
Leemans JC, Stokman G, Claessen N, Rouschop KM, Teske GJ, Kirschning CJ, Akira S, van der Poll T, Weening JJ, and Florquin S. Renal-associated TLR2 mediates ischemia/reperfusion injury in the kidney. J Clin Invest 115: 2894-2903, 2005.
Lim SW, Li C, Ahn KO, Kim J, Moon IS, Ahn C, Lee JR, and Yang CW. Cyclosporine-induced renal injury induces Toll-like receptor and maturation of dendritic cells. Transplantation 80: 691-699, 2005.
Michelsen KS, Doherty TM, Shah PK, and Arditi M. Role of Toll-like receptors in atherosclerosis. Circ Res 95: e96-e97, 2004.
Miyaji T, Hu X, Yuen PS, Muramatsu Y, Iyer S, Hewitt SM, and Star RA. Ethyl pyruvate decreases sepsis-induced acute renal failure and multiple organ damage in aged mice. Kidney Int 64: 1620-1631, 2003.
Morrissey J, Guo G, McCracken R, Tolley T, and Klahr S. Induction of CD14 in tubular epithelial cells during kidney disease. J Am Soc Nephrol 11: 1681-1690, 2000.
Musson RA, Morrison DC, and Ulevitch RJ. Distribution of endotoxin (lipopolysaccharide) in the tissues of lipopolysaccharide-responsive and -unresponsive mice. Infect Immun 21: 448-457, 1978.
Nemoto S, Vallejo JG, Knuefermann P, Misra A, Defreitas G, Carabello BA, and Mann DL. Escherichia coli LPS-induced LV dysfunction: role of toll-like receptor-4 in the adult heart. Am J Physiol Heart Circ Physiol 282: H2316-H2323, 2002.
Nishimura M and Naito S. Tissue-specific mRNA expression profiles of human toll-like receptors and related genes. Biol Pharm Bull 28: 886-892, 2005.
Pak J, Pu Y, Zhang ZT, Hasty DL, and Wu XR. Tamm-Horsfall protein binds to type 1 fimbriated Escherichia coli and prevents E. coli from binding to uroplakin Ia and Ib receptors. J Biol Chem 276: 9924-9930, 2001.
Phillips CL, Arend LJ, Filson AJ, Kojetin DJ, Clendenon JL, Fang S, and Dunn KW. Three-dimensional imaging of embryonic mouse kidney by two-photon microscopy. Am J Pathol 158: 49-55, 2001.
Pugin J, Schurer-Maly CC, Leturcq D, Moriarty A, Ulevitch RJ, and Tobias PS. Lipopolysaccharide activation of human endothelial and epithelial cells is mediated by lipopolysaccharide-binding protein and soluble CD14. Proc Natl Acad Sci USA 90: 2744-2748, 1993.
Riedemann NC, Guo RF, and Ward PA. The enigma of sepsis. J Clin Invest 112: 460-467, 2003.
Rifkin IR, Leadbetter EA, Busconi L, Viglianti G, and Marshak-Rothstein A. Toll-like receptors, endogenous ligands, and systemic autoimmune disease. Immunol Rev 204: 27-42, 2005.
Rothstein JL and Schreiber H. Synergy between tumor necrosis factor and bacterial products causes hemorrhagic necrosis and lethal shock in normal mice. Proc Natl Acad Sci USA 85: 607-611, 1988.
Sabroe I, Read RC, Whyte MK, Dockrell DH, Vogel SN, and Dower SK. Toll-like receptors in health and disease: complex questions remain. J Immunol 171: 1630-1635, 2003.
Saemann MD, Weichhart T, Zeyda M, Staffler G, Schunn M, Stuhlmeier KM, Sobanov Y, Stulnig TM, Akira S, von Gabain A, von Ahsen U, Horl WH, and Zlabinger GJ. Tamm-Horsfall glycoprotein links innate immune cell activation with adaptive immunity via a Toll-like receptor-4-dependent mechanism. J Clin Invest 115: 468-475, 2005.
Samuelsson P, Hang L, Wullt B, Irjala H, and Svanborg C. Toll-like receptor 4 expression and cytokine responses in the human urinary tract mucosa. Infect Immun 72: 3179-3186, 2004.
Schilling JD, Martin SM, Hung CS, Lorenz RG, and Hultgren SJ. Toll-like receptor 4 on stromal and hematopoietic cells mediates innate resistance to uropathogenic Escherichia coli. Proc Natl Acad Sci USA 100: 4203-4208, 2003.
Serafini-Cessi F, Malagolini N, and Cavallone D. Tamm-Horsfall glycoprotein: biology and clinical relevance. Am J Kidney Dis 42: 658-676, 2003.
Tsuboi N, Yoshikai Y, Matsuo S, Kikuchi T, Iwami K, Nagai Y, Takeuchi O, Akira S, and Matsuguchi T. Roles of toll-like receptors in C-C chemokine production by renal tubular epithelial cells. J Immunol 169: 2026-2033, 2002.
Wichterman KA, Baue AE, and Chaudry IH. Sepsis and septic shock-a review of laboratory models and a proposal. J Surg Res 29: 189-201, 1980.
Wolfs TG, Buurman WA, van Schadewijk A, de Vries B, Daemen MA, Hiemstra PS, and van't Veer C. In vivo expression of Toll-like receptor 2 and 4 by renal epithelial cells: IFN- and TNF- mediated upregulation during inflammation. J Immunol 168: 1286-1293, 2002.
Zarember KA and Godowski PJ. Tissue expression of human Toll-like receptors and differential regulation of Toll-like receptor mRNAs in leukocytes in response to microbes, their products, and cytokines. J Immunol 168: 554-561, 2002.
Zhai Y, Shen XD, O'Connell R, Gao F, Lassman C, Busuttil RW, Cheng G, and Kupiec-Weglinski JW. Cutting edge: TLR4 activation mediates liver ischemia/reperfusion inflammatory response via IFN regulatory factor 3-dependent MyD88-independent pathway. J Immunol 173: 7115-7119, 2004.
Zhang D, Zhang G, Hayden MS, Greenblatt MB, Bussey C, Flavell RA, and Ghosh S. A toll-like receptor that prevents infection by uropathogenic bacteria. Science 303: 1522-1526, 2004.
作者单位：Indiana Center for Biological Microscopy, Department of Medicine, Division of Nephrology, Indiana University, Indianapolis, Indiana