Abstract
ABSTRACT. IFN-inducible protein-10 (IP-10/CXCL10) is a potent chemoattractant for activated T lymphocytes and was recently reported to have several additional biologic activities. In this study, the expression and the function in normal glomeruli and in Thy1.1 glomerulonephritis (GN) were investigated. The expression of IP-10 was detected in normal rat glomeruli mainly in the podocyte. The expression of IP-10 was also detected on the cultured podocyte. The IP-10 expression was elevated at the early phase of Thy1.1 GN. The double staining immunofluorescence study clearly demonstrated that the elevated expression of IP-10 was mostly detected in the podocyte and very partly in mesangial area. A receptor for IP-10, CXCR3, showed similar expression patterns to that of IP-10. Expressions of neither of IP-10 nor of CXCR3 were detected on the inflammatory cells. For elucidating the role of IP-10, the blocking study was carried out with monoclonal anti–IP-10 antibody. The monoclonal anti–IP-10 antibody treatment decreased the expression of IP-10 and podocyte-associated proteins such as nephrin and podocin that are reported to be essential for maintaining the podocyte function (IP-10, 53.0% to control; nephrin, 43.5%; podocin, 60.4%). The findings indicated that the anti–IP-10 treatment disturbed the podocyte function. The anti–IP-10 treatment given to the rats with Thy1.1 nephritis exacerbated proteinuria, mesangiolysis, and matrix expansion. Collectively, the findings indicated that IP-10 plays a role in maintaining the podocyte function. Also, the findings suggested that anti–IP-10 treatment exacerbated the glomerular alterations in Thy1.1 GN by disturbing the podocyte function.
IFN-inducible protein of 10 kD (IP-10/CXCL10) identified as a product of genes induced in response to IFN-γ is a member of the CXC chemokine family (1). IP-10 is a potent chemoattractant for activated T lymphocytes, natural killer (NK) cells, and monocytes (2) and is believed to be a regulator of the type 1 T helper (Th1) inflammatory responses (3). Recent studies indicated that the expression of IP-10 was observed in a variety of cells (4–8⇓⇓⇓⇓), and IP-10 had several additional biologic activities such as the stimulation of monocytes and lymphocytes, modulation of the adhesion molecule expression, and inhibition of angiogenesis (9). It is reported that the expression of IP-10 was elevated in several diseases such as colitis, hepatitis, and multiple sclerosis and that IP-10 was involved in the development of these diseases (6,10–12⇓⇓⇓). Romagnani et al. and other groups (13–16⇓⇓⇓) have reported that IP-10 was expressed in mesangial cells. Gomez-Chiarri et al. (17) reported that mRNA expression was detected in the cultured glomerular epithelial cells treated with adriamycin. Some studies with experimental model suggested that IP-10 contributed to the development of the glomerular diseases (13,17–20⇓⇓⇓⇓). However, the knowledge on the expression and the function of IP-10 in glomeruli is still limited. Mesangial proliferative glomerulonephritis (GN) including IgA nephropathy is one of the most important diseases in the nephrology field. Anti-Thy1.1 antibody–induced GN (Thy1.1 GN) is most commonly used as a model of mesangial proliferative GN (21). However, no precise studies of the role of IP-10 on the pathogenesis of Thy1.1 GN are reported.
In this study, first, an intense investigation was carried out on the expression and the function of IP-10 in normal rat glomeruli. Second, the expression and the role of IP-10 in Thy1.1 GN were studied. This study clearly showed that IP-10 is expressed on the podocyte and plays a role in maintaining the podocyte function. It was also demonstrated that the treatment of blocking anti–IP-10 mAb exacerbated the glomerular alteration in Thy1.1 GN.
Materials and Methods
Animal
All experiments were performed using specific pathogen-free female Wistar rats (6 wk old) that weighed 140 to 180g, purchased from Charles River Japan (Atsugi, Japan). All animal experiments conformed to the National Institutes of Health Guide for the Care and Use of Laboratory Animals.
Culture of Podocyte
Cultivation of conditionally immortalized mouse podocytes was conducted as reported previously (22). In brief, podocytes were maintained in RPMI 1640 medium (Nissui Pharmaceutical, Tokyo, Japan) supplemented with 10% FBS (Life Technologies, Grand Island, NY), 100 U/ml penicillin (Banyu Pharmaceutical, Tokyo, Japan), and 0.1 mg/ml streptomycin (Meiji Seika Kaisha, Tokyo, Japan). To propagate podocytes, we cultivated cells at 33°C (permissive conditions), and the culture medium was supplemented with 10 U/ml mouse rIFN-γ (Pepro Tech EC, England) to enhance expression of a thermosensitive T-antigen. To induce differentiation, we maintained podocytes at 37°C without IFN-γ (nonpermissive conditions) for at least 1 wk before using in the experiment.
Experimental Design
Expression and Localization of IP-10 and CXCR3 in Normal Rat Glomeruli and in Cultured Podocyte.
The expressions of IP-10 and CXCR3 for normal rat glomeruli and cultured podocyte were analyzed by immunofluorescence (IF) study and by Western blot techniques with two anti–IP-10 antibodies. mRNA expression of IP-10 was analyzed by reverse transcriptase–PCR (RT-PCR) with total RNA prepared from isolated glomeruli.
IP-10 Blocking Study in Normal Rat and Cultured Podocyte.
In vivo IP-10 blocking was induced by injections of the anti–IP-10 mAb. This reagent was obtained by immunizing mice with rat CXCL10/Fc fusion protein, and then screened by measuring the binding to the rat CXCL10/Ac2A fusion protein (23). It was already confirmed that this mAb blocked the rat CXCL10-induced chemoattractive effect (24). As a control, mouse IgG1 mAb RVG1 (against rotavirus) was used. The blocking study was carried out according to two different concentrations of the anti–IP-10 mAb, such as 1 mg/100 g body wt as a low-dose injection and 5 mg/100 g body wt as a high-dose injection. Five rats received an injection of the anti–IP-10 mAb of each concentration intravenously daily for 5 d and were killed, and the right kidney was removed, weighed, cut into portions, and used for assessment of IF and light microscopy (LM). The remaining portion of right kidney and the left kidney were used to prepare the total glomerular RNA. mRNA expression of IP-10, nephrin, podocin, and podoplanin was analyzed by the semiquantitative RT-PCR on isolated glomeruli of pooled kidneys from five rats. Twenty-four-hour urine samples were collected just before the rats were killed. Urine protein concentrations were determined by the colorimetric assay (Bio-Rad, Oakland, CA) using BSA as a standard.
For analyzing the effect of anti–IP-10 mAb treatment on the expression of podocyte-associated molecules, immortalized cultured podocyte maintained at 37°C was incubated with anti–IP-10 mAb (1 mg/ml), RVG1 (1 mg/ml), or rabbit anti-podocalyxin antibody (100 mg/ml) for 24 h. mRNA expression of IP-10, nephrin, podocin, and podoplanin was analyzed by semiquantitative RT-PCR.
Expression of IP-10 and CXCR3 in Thy1.1 GN.
Thy1.1 GN was induced in rats by a single injection with 1.0 ml of saline containing 500 μg of anti-Thy1.1 mAb 1-22-3 through tail vein. Preparation of mAb 1-22-3 and induction of Thy1.1 GN have been described previously (25). The rats were killed just before injection of mAb 1-22-3 and on days 1, 5, and 14 after induction of GN (n = 5 per time point). At each time point, a rat was killed and the kidney was treated as described above. The kinetics of the expression of IP-10 and CXCR3 were analyzed by IF and RT-PCR. To investigate the localization of IP-10 and CXCR3, dual-labeling staining was carried out with several cellular markers including leukocyte common antigen, monocyte/macrophage marker, CD5, RECA-1, α-smooth muscle actin (α-SMA), and podocalyxin.
IP-10 Blocking Study in Thy1.1 GN.
Thy1.1 GN was caused in rats by mAb 1-22-3 in the same manner described above. The rats with Thy1.1 GN were treated with anti–IP-10 mAb at 5 h after mAb 1-22-3 injection and treated daily until the day of killing. The rats were killed on day 5 and day 14 (n = 5 per group), and kidneys were removed as described above. Twenty-four-hour urine samples were collected on days 1, 3, 5, 7, 10, and 14 after injection of mAb 1-22-3. Urine protein concentrations were determined as described above. Glomerular injury was assessed by LM and IF. The severity of morphologic alterations was evaluated in a double-blind manner with >30 full-sized glomeruli (80 to 100 μm) from each specimen. The severity of mesangiolysis, mesangial matrix expansion, and the intensity of α-SMA staining were scored as described previously by Ito et al. (21). To analyze whether inflammatory responses are affected by the anti–IP-10 mAb treatment, we counted the number of OX-1–, ED1-, and OX-19–positive inflammatory cells in glomeruli. The expression of nephrin, podocin, and podoplanin was analyzed by immunofluorescence. Glomerular mRNA expression for podocyte-associated molecules was analyzed by RT-PCR. In this study, the expression for nephrin, podocin, and podoplanin to that of podocalyxin was calculated. Considering the mesangial proliferation of Thy1.1 GN, podocalyxin is more proper than GAPDH for internal control of mRNA of podocyte, because podocalyxin is thought to be a stable molecule in podocyte from our experience. Glomerular mRNA expressions for chemokine (IP-10, macrophage-derived chemokine [MDC]), chemokine receptor (CXCR3), and cytokine (IFN-γ, IL-4) were also analyzed by RT-PCR.
Localization of Anti–IP-10 mAb Injected into Normal and Thy1.1 GN Rats
Normal rats or the rats 5 d after induction of Thy1.1. GN received an intravenous injection of 5 mg/100 g body wt anti–IP-10 mAb or RVG1 and were killed 10 min after the injection. To detect the anti–IP-10 mAb injected to rat, we used goat anti-mouse IgG1 as a primary antibody and FITC-conjugated anti-goat IgG as a secondary antibody.
Morphologic and Immunohistochemical Studies
Tissue samples for LM assessment and for the IF studies were prepared as described previously (21). The 3-μm-thick frozen sections were cut with a cryostat and stained with goat anti–IP-10 antibody (Santa Cruz Biotechnology), goat anti-CXCR3 antibody (Santa Cruz Biotechnology), rabbit anti-nephrin antibody (intracellular site) (26), rabbit anti-podocin antibody (N-terminal site) (27), and rabbit anti-podoplanin antibody (28). The double-staining IF for anti–IP-10 and anti-CXCR3 was performed with several cellular markers such as OX-1 (leukocyte marker), α-SMA (as an injured mesangial cell marker), RECA-1 (as endothelial cell marker), podocalyxin (as podocyte marker), CD5 (as a T-cell marker), and ED1 (as a pan monocyte/macrophage marker). Anti-leukocyte common antigen mAb (OX-1), anti–RECA-1 mAb, and anti–OX-19 mAb (anti-CD5) were purchased from Serotec (Oxford, JE, UK); anti–α-SMA mAb was purchased from Sigma (St. Louis, MO); and anti-ED1 mAb was purchased from Chemicon Internation (Temecula, CA). Anti-podocalyxin antibody 4D5 was donated by Dr. Hara (Yoshida Hospital, Niigata, Japan). The clone producing 4D5 was purified from hybridoma cells fused with spleen cells from mice immunized with podocalyxin-rich fraction prepared with WGA-Sepharose 4B (29). The specificity of 4D5 was analyzed with immunohistochemical studies, immunoprecipitation, and Western blot analysis using anti-podocalyxin mAb 5A as positive control (30) (5A was donated by Dr. Miettinen, University of Helsinki, Helsinki, Finland).
FITC-conjugated anti-goat IgG (for anti–IP-10 and anti-CXCR3 antibodies), tetramethyl-rhodamine isothiocyanate (TRITC)-conjugated goat anti-mouse IgG1 (for OX-1, anti–RECA-1, OX-19, and ED1 mAb), and TRITC-conjugated goat anti-mouse IgG2a (for anti–α-SMA mAb and 4D5) were used as secondary antibodies. FITC-conjugated goat anti-mouse IgG1 was used to count the number of OX-1–, OX-19–, and ED1-positive cells. These secondary antibodies were purchased from Southern Biotechnology Associates (Birmingham, AL). FITC-conjugated swine anti-rabbit IgG was used for anti-nephrin, anti-podocin, and anti-podoplanin antibodies. This secondary antibody was purchased from DAKO (Glostrup, Denmark).
RT-PCR
Semiquantitative RT-PCR with glomerular RNA was performed basically according to the method described previously (21). The primers were designed according to the published sequences (Table 1). Negative controls without cDNA and positive controls of cDNA from Con-A–stimulated rat spleen cells were included.
Table 1. PCR primers used in this studya
Western Blot Analysis
Normal rat glomeruli and cultured podocyte were isolated with PBS containing protease inhibitors and solubilized with RIPA buffer (consisting of 0.1% SDS, 1% sodium deoxycholate, 1% Triton X-100, 150 mmol/L NaCl, and 10 mmol/L EDTA in 25 mmol/L Tris-HCl [pH 7.2]) with protease inhibitors. The insoluble material was removed by centrifugation at 15,000 × g for 10 min. The concentration was measured by the bicinchoninic acid method (Pierce Chemical, Rockford, IL), and the solubilized material was subjected to SDS-PAGE with 12% acrylamide gel according to the method of Laemmli et al. (31) and transferred to a nitrocellulose membrane (Bio-Rad, Hercules, CA) by electrophoretic transblotting for 30 min using Trans-Blot SD (Bio-Rad). After blocking with bovine skim milk, strips of the membranes were exposed to two kinds of anti–IP-10 antibodies; one is the goat anti–IP-10 antibody (Santa Cruz Biotechnology), and the other is the anti–IP-10 mAb from the mouse. They both were washed and then incubated with alkaline phosphatase–conjugated rabbit anti-goat IgG (Bio Source International, Tago Immunologicals, Camarillo, CA) or with alkaline phosphatase–conjugated anti-mouse IgG (Bio Source International, Tago Immunologicals). The reaction was developed with an alkaline phosphatase chromogen kit (5-bromo-4-chloro-3-indolil phosphate p-toluidine salt/nitro blue tetrazolium; Biomedica, Foster City, CA).
Statistical Analyses
All values are expressed as means ± SD. The statistical significance (defined as P < 0.05) was evaluated using the unpaired t test or Mann Whitney U test. Data were analyzed using the GraphPad InStat 3.05 (GraphPad Software, San Diego, CA).
Results
IP-10 and CXCR3 Were Expressed in Normal Rat Glomeruli and Cultured Podocyte
IP-10 and CXCR3 expressions were detected in the normal rat glomeruli by IF study with goat anti–IP-10 and anti-CXCR3 antibodies, and no staining with normal goat serum was seen (Figure 1, Panel I-A). Immunostaining of IP-10 was observed as a linear-like pattern along the glomerular capillary wall, and CXCR3 staining was also observed as an epithelial pattern along the glomerular capillary wall. IP-10 and CXCR3 expressions in normal rat glomeruli were also detected by RT-PCR (Figure 1, Panel I-B). The Western blot analysis was performed with the goat anti–IP-10 antibody and the mouse anti–IP-10 mAb that was used for the IP-10 neutralization study. Approximately 10-kD bands of IP-10 were detected by both antibodies (Figure 1, Panel I-C, lanes 1 and 3). No binding bands were detected by either normal goat serum or RVG1 (lanes 2 and 4). On the basis of these observations, we investigated whether IP-10 and CXCR3 are expressed in the cultured podocyte with IF, RT-PCR, and Western blot studies. The expression of IP-10 and CXCR3 were clearly detected in differentiated conditioned podocyte with IF, RT-PCR, and Western blot (Figure 1, Panel II). CXCR3 was not detected in undifferentiated conditioned podocyte with RT-PCR (Figure 1, Panel II-B).
Figure 1. Expression of IFN-inducible protein-10 (IP-10) in normal rat glomeruli and in cultured podocyte. (Panel I) Expression of IP-10 in normal rat glomeruli. (A) Immunofluorescence (IF) finding of IP-10 and CXCR3 in normal rat kidney section. Stainings of IP-10 and CXCR3 were observed as a linear-like pattern along the glomerular capillary wall (arrows) with goat anti–IP-10 and anti-CXCR3 antibodies. No positive signal with normal goat serum was seen. (B) Reverse transcriptase–PCR (RT-PCR) findings of IP-10 and CXCR3 with normal rat glomerular mRNA. The expressions of glomerular mRNA for IP-10 and CXCR3 were detected at 27 and 30 cycles, respectively, of the amplification of PCR. (C) Western blot findings of glomerular extracts. A total of 250 μg of solubilized glomeruli was subjected to each lane. An approximately 10-kD band was detected in lanes 1 and 3. No binding bands were detected in lanes 2 and 4. Lane 1, stained with commercially purchased goat anti-IP-10; lane 3, mouse anti–IP-10 mAb that was used for the IP-10 neutralization study; lane 2, normal goat serum; lane 4, mouse IgG1, RVG1. (Panel II) Expression of IP-10 in cultured podocyte. (A) IF finding of IP-10 and CXCR3 with differentiated immortalized podocytes. The expressions of IP-10 and CXCR3 were clearly detected in differentiated conditioned podocyte with goat anti–IP-10 and anti-CXCR3 antibodies. There were no specific expressions with normal goat serum. (B) RT-PCR findings of IP-10 and CXCR3 with undifferentiated (33°C) and differentiated (37°C) immortalized podocytes. As a positive control, IP-10 and CXCR3 were also amplified from mouse spleen cDNA. The expression of mRNA for IP-10 was seen in both undifferentiated and differentiated conditioned podocytes. The expression of mRNA for CXCR3 was seen in differentiated conditioned podocyte, but no expression was detected in undifferentiated conditioned podocyte. (C) Western
blot findings of differentiated conditioned podocytes. The volume of 0.05, 0.1, and 0.15 mg of the solubilized podocytes was subjected to lanes 1, 2, and 3, respectively. An approximately 10-kD band was detected in lanes 1, 2, and 3 of anti–IP-10. The band intensity is dependent on the charged volume to the lane. No bands were detected in lanes 1, 2, and 3 of normal goat serum. Magnifications: ×200 in Panel I-A, ×600 in Panel II-A.
Anti–IP-10 mAb Treatment Decreased the Expression of IP-10 and Podocyte-Associated Proteins
The effects of anti–IP-10 mAb treatment on the expression of glomerular IP-10 mRNA were shown in Figure 2, Panel I. A high-dose treatment (5 mg/100 g body wt) clearly decreased the IP-10 expression, whereas there was no difference of IP-10 expression between the low-dose treatment (1 mg/100 g body wt) group and the RVG1 (5 mg/100 g body wt) treatment group. Neither abnormal proteinuria nor morphologic abnormality observed by LM was detected in the anti–IP-10 mAb treatment group. mRNA expressions for podocin, nephrin, and podoplanin were clearly decreased by the high-dose anti–IP-10 mAb treatment (Figure 2, Panel II). High-dose anti–IP-10 mAb treatment decreased the IF staining intensity of nephrin (Figure 2, Panel III-B), podocin, and podoplanin. To confirm these in vivo results, we carried out the IP-10 blocking study with cultured podocyte. In this study, we used two kinds of control antibodies, RVG1 and anti-podocalyxin antibody. The mRNA expressions of cultured podocytes for podocin and nephrin were clearly decreased and those of IP-10 and podoplanin were slightly decreased by the anti–IP-10 mAb treatment (Figure 3).
Figure 2. Effect of anti–IP-10 mAb treatment on rat glomerular mRNA and IF expression. (Panel I) Effect of anti–IP-10 mAb treatment on rat glomerular mRNA expression for IP-10. mRNA expression for IP-10 was semiquantified by RT-PCR using cDNA corresponding to 750 ng of RNA. (A) Ratio of the densitometric signals of IP-10 to that of internal control (glyceraldehyde-3-phosphate dehydrogenase [GAPDH]) was analyzed. The data are shown as a ratio (%) relative to the RVG1-injected control group and are expressed as mean ± SD of three independent experiments. (B) Representative agarose gel electrophoretic patterns of PCR product of IP-10 and GAPDH are shown. A high dose of anti–IP-10 mAb treatment clearly decreased the glomerular mRNA expression of IP-10. Anti–IP-10 (L), low dose of anti–IP-10 mAb (1 mg/100 g body wt) injected group; anti–IP-10 (H), high dose of anti–IP-10 mAb (5 mg/100 g body wt) injected group; RVG1 (H), high dose of RVG1 (5 mg/100 g body wt). (Panel II) Effect of anti–IP-10 mAb treatment on rat glomerular mRNA expression for podocyte-associated proteins. (A) mRNA expression for podocyte-associated proteins podocin, nephrin, and podoplanin was semiquantified by RT-PCR. The data are shown as a ratio (%) relative to the RVG1-injected control group and are expressed as mean ± SD of three independent experiments. (B) Representative agarose gel electrophoretic patterns of PCR products are shown. A high dose of anti–IP-10 mAb treatment clearly decreased the glomerular mRNA expression of podocin, nephrin, and podoplanin. (Panel III) Effect of anti–IP-10 mAb treatment on IF staining of nephrin. (A) IF finding of the high-dose RVG1 treatment group. (B) IF finding of the high-dose anti–IP-10 treatment group. A decrease in the staining for nephrin was found in the anti–IP-10 treatment sample.
Figure 3. Effect of anti–IP-10 mAb treatment to cultured podocyte on mRNA expression for IP-10 and podocyte-associated proteins. (A) mRNA expression for IP-10 and podocyte-associated proteins podocin, nephrin, and podoplanin was semiquantified by RT-PCR. The data are shown as a ratio (%) relative to the RVG1-injected control group and are expressed as mean ± SD of three independent experiments. (B) Representative agarose gel electrophoretic patterns of PCR products are shown. Anti–IP-10 mAb treatment clearly decreased the glomerular mRNA expression of podocin, nephrin, and podoplanin, but anti-podocalyxin (anti PCX) antibody treatment did not.
Expression of Glomerular IP-10 and CXCR3 Were Increased after Induction of Thy1.1 GN
Glomerular mRNA expression for IP-10 increased in rats with Thy1.1 GN caused by mAb 1-22-3. There was an almost twofold increase of IP-10 mRNA on day 1 after mAb 1-22-3 injection. The increased expression of IP-10 mRNA peaked on day 5, and then the expression decreased to the normal range on day 14 (Figure 4, Panel I-A and B). A weak immunostaining of IP-10 was detectable in the normal rat glomeruli, but noticeable IP-10 expression with a liner-like pattern was observed in rat glomeruli of day 5 after Thy1.1 GN induction (Figure 4, Panel I-C and D). Dual-labeling IF study with cell markers showed that major parts of IP-10–positive cells were stained with 4D5 (Figure 4, Panel II-D). A small part of IP-10–positive cells was also stained with anti–α-SMA mAb (Figure 4, Panel II-C). IP-10 staining was apart from RECA-1 staining (Figure 4, Panel II-B). No IP-10–positive cells were co-stained with OX-1, OX-19, and ED1 (Figure 4, Panel II-A, E, and F). mRNA expression for CXCR3 increased on day 1 and day 5 similarly to that for IP-10. The constant increase of the expression of CXCR3 was observed on day 14 (Figure 5, Panel I-A and B). The immunostaining of CXCR3 was detected in the normal rat glomeruli. The clear staining of CXCR3 with a linear-like pattern along the glomerular capillary wall was seen in rat with Thy1.1 GN (Figure 5, Panel I-C and D). Dual-labeling IF study with cell markers showed that CXCR3-positive cells were largely co-stained with 4D5 (Figure 5, Panel II-D), and some CXCR3-positive cells were also stained with anti–RECA-1 mAb (Figure 5, Panel II-B) and anti–α-SMA mAb (Figure 5, Panel II-C). No CXCR3-positive cells co-stained with OX-1 were observed at any stages (5 d, 2 h, 24 h) of the disease (Figure 5, Panel II-A, E, and F).
Figure 4. (Panel I) Kinetics of the glomerular mRNA expression and IF findings of IP-10 after induction of Thy1.1 glomerulonephritis (GN). (A) The kinetics of mRNA expression for IP-10 during the development of Thy1.1 GN was analyzed by RT-PCR. Ratio of the densitometric signals of IP-10 to that of the GAPDH was analyzed. The data are shown as a ratio (%) relative to normal rat findings and are expressed as
mean ± SD of three independent experiments. (B) Representative agarose gel electrophoretic patterns of PCR product of IP-10 and GAPDH for each time point are shown. Increased mRNA expression of IP-10 was detected on days 1 and 5 after induction of Thy1.1 GN. (C and D) IF findings of IP-10 of normal rat and on day 5 after induction of Thy1.1 GN, respectively. Intense staining of IP-10 along the capillary wall (arrows) was seen in rats on day 5 (D). Panel II. Dual labeling IF study of IP-10 with cellular markers in rat glomeruli 5 d after induction of Thy1.1 GN. (A to F) Dual-labeling study of IP-10 with OX-1 (leukocyte common marker; A), anti–RECA-1 mAb (endothelial cell marker; B), anti–α-smooth muscle actin (α-SMA) mAb (mesangial cell marker; C), 4D5 (podocyte marker; D), OX-19 (CD5; T-cell marker; E), and ED1 (macrophage marker; F). IP-10 was stained as green, and cellular markers were stained as red. Clear IP-10 staining along the glomerular capillary wall was observed (A, arrow). OX-1–, OX-19–, or ED1-positive cells were not stained with anti–IP-10 antibody (arrowheads in A, arrows in E and F). IP-10 staining was apart from RECA-1 (arrows in B). A part of IP-10 staining was observed in α-SMA–positive mesangial area (arrows, yellow; C). Major parts of IP-10–positive cells were also stained with podocyte marker 4D5 (arrows, yellow; D). Magnifications: ×200 in Panel I-D and C, ×400 in Panel II.
Figure 5. (Panel I) Kinetics of the glomerular mRNA expression and IF findings of CXCR3 after induction of Thy1.1 GN. (A) The kinetics of mRNA expression for CXCR3 during the development of Thy1.1
GN was analyzed by RT-PCR. Ratio of the densitometric signals of CXCR3 to that of the GAPDH was analyzed. The data are shown as a ratio (%) relative to normal rat findings and are expressed as mean ± SD of three independent experiments. (B) Representative agarose gel electrophoretic patterns of PCR product of CXCR3 and GAPDH for each time point are shown. Increased mRNA expression of CXCR3 was detected on days 1, 5, and 14 after induction of Thy1.1 GN. (C and D) IF findings of CXCR3 of normal rat and on day 5 after induction of Thy1.1 GN, respectively. Intense staining of CXCR3 along the capillary wall (arrows) was seen in rats on day 5 (C). (Panel II) Dual-labeling IF study of CXCR3 with cellular markers in rat glomeruli after induction of Thy1.1 GN. (A to F) Dual-labeling study of CXCR3 with OX-1 (leukocyte marker; A, E, and F), anti–RECA-1 mAb (endothelial cell marker; B), anti–α-SMA mAb (mesangial cell marker; C), and 4D5 (podocyte marker; D) in rat glomeruli 5 d (A to D), 2 h (E), and 24 h (F) after the induction of Thy1.1 GN. CXCR3 was stained as green, and cellular markers were stained as red. Clear CXCR3 staining along the glomerular capillary wall was observed (arrows), and no OX-1–positive cells were stained with anti-CXCR3 antibody (arrowheads; A). A part of CXCR3 staining was observed in RECA-1–positive endothelial cells (B) and in α-SMA–positive mesangial area (C). Major parts of CXCR3-positive cells were also stained with podocyte marker 4D5 (arrows, yellow; D). Although many inflammatory cells’ infiltration into glomeruli were observed in the early phase of Thy1.1 GN, no OX-1–positive cells were stained with CXCR3 (E and F). Magnifications: ×200 in Panel I-C, ×400 in Panel II.
Blockade of IP-10 Exacerbated Thy1.1GN
Both the low- and the high-dose treatment of anti–IP-10 markedly enhanced the proteinuria level of Thy1.1 GN (Figure 6, Panel I-A and B). The effect of the anti–IP-10 mAb treatment on the morphologic alterations was evaluated by the number of ED1-, OX-19–, and OX-1–positive inflammatory cells and the mesangiolysis, matrix, and α-SMA staining scores on day 5 and day 14 (Figure 6, Panel II). There were no differences in numbers of ED1- and CD5-positive inflammatory cells between two groups on day 5 (Figure 6, Panel II-A and B). The number of OX-1–positive inflammatory cells in glomeruli of the anti–IP-10 mAb–treated group was slightly higher than that of the RVG1 group; however, there was no statistical difference between the two groups on either day 5 or day 14 (Figure 6, Panel II-C and D). Mesangiolysis observed on day 5 and the extent of the α-SMA staining and mesangial matrix on day 14 were significantly exacerbated by the anti–IP-10 treatment (Figure 6, Panel II-E, F, and H). Figure 7 shows representative findings of OX-1, mesangiolysis, mesangial matrix, and α-SMA staining. We also investigated the IF staining patterns of podocyte-associated proteins in glomeruli on day 5 and day 14. On day 5, more discontinuous staining of nephrin podocin and podoplanin were observed in the anti–IP-10 mAb treatment group than in the RVG1 control group (Figure 8). On day 14, staining intensities of nephrin, podocin, and podoplanin were weaker in the anti–IP-10 mAb treatment group than in the RVG1 control group (Figure 9). The anti–IP-10 treatment clearly decreased the glomerular mRNA expression of nephrin, podocin, and podoplanin on day 5 after induction of Thy1.1 GN (Figure 10). Figure 11 shows the effect of anti–IP-10 mAb treatment on rat glomerular mRNA expression for several chemokines and cytokines on 5 d after Thy1.1 GN induction. Anti–IP-10 mAb treatment did not affect the mRNA expressions of IP-10, CXCR3, MDC, IL-4, and IFN-γ.
Figure 6. (Panel I) Effect of anti–IP-10 mAb treatment on kinetics of proteinuria after induction of Thy1.1 GN. (A and B) Effect of anti–IP-10 mAb treatment of a low dose (1 mg/100 g body wt, daily) and a high dose (5 mg/100 g body wt, daily) on proteinuria was analyzed. The low-dose treatment of anti–IP-10 mAb increased the mount of proteinuria on day 5 (A). Significant increase of proteinuria was observed in the high-dose treatment group on days 1, 3, and 5 (B). •, RVG1-treated control group; □, anti–IP-10 mAb–treated group. Data are expressed as mean ± SD (n = 5; *P < 0.05, **P < 0.01 compared with the control group at the same time point). (Panel II) Effect of anti–IP-10 mAb treatment on rat glomerular changes on day 5 and day 14 after induction of Thy1.1 GN. Glomerular changes were evaluated by the number of ED1-, CD5-, and OX-1–positive inflammatory cells and the mesangiolysis, matrix, and α-SMA staining scores on day 5 and/or day 14. No significant differences of the numbers of inflammatory cells of ED1, CD5, and OX-1 in glomeruli were observed between the anti–IP-10 mAb–treated group and the RVG1 group (A to D). Anti–IP-10 treatment exacerbated mesangiolysis on day 5 (E) and the extension of mesangial matrix and α-SMA staining on day 14 (F and H). Data are expressed as mean ± SD (n = 5; *P < 0.05, **P < 0.01 compared with the control group).
Figure 7. Representative light microscopic (LM) and IF findings of OX-1, mesangiolysis, mesangial matrix, and α-SMA staining. (A, C, E, and G) Findings of anti–IP-10 mAb treatment group. (B, D, F, and H) RVG1 treatment group. IF findings of OX-1–positive cells (A and B) on day 5, LM findings of periodic acid-Schiff staining on day 5 (C and D) and on day 14 (E and F), and IF findings of α-SMA on day 14 (G and H). Severe mesangiolysis finding with ballooning was observed in rats that were treated with anti–IP-10 mAb (C). Anti–IP-10 mAb treatment clearly exacerbated LM and IF findings of α-SMA on day 14. Magnifications: ×400 in A and B, ×100 in C to F, ×200 in G and H.
Figure 8. Effect of anti–IP-10 mAb treatment on IF staining of podocyte functional molecules on day 5 after induction of Thy1.1 GN. The findings of the anti–IP-10 mAb treatment group are shown in left column and of the RVG1 treatment group are shown in right column (A and B, nephrin; C and D, podocin; E and F, podoplanin). More discontinuous coarse granular patterns of nephrin, podocin, and podoplanin were observed in the anti–IP-10 mAb treatment group (arrows) than that in the RVG1 treatment group. Magnification, ×200.
Figure 9. Effect of anti–IP-10 mAb treatment on IF staining of podocyte functional molecules on day 14 after induction of Thy1.1 GN. The findings of the anti–IP-10 mAb treatment group are shown in the left column and of the RVG1 treatment group are shown in the right column (A and B, nephrin; C and D, podocin; E and F, podoplanin). Staining intensity of nephrin, podocin, and podoplanin was weaker in the anti–IP-10 mAb treatment group than that in the RVG1 treatment group. Magnification, ×200.
Figure 10. Effect of anti–IP-10 mAb treatment on mRNA expression of podocyte functional molecules on day 5 after induction of Thy1.1 GN. (A) mRNA expression for podocin, nephrin, podoplanin, and podocalyxin was semiquantified by RT-PCR using cDNA corresponding to 750 ng of RNA. Ratio of the densitometric signals of podocin, nephrin, and podoplanin to that of podocalyxin was calculated. The data are shown as a ratio (%) relative to the RVG1-injected control group and are expressed as mean ± SD of three independent experiments. (B) The mRNA expression of nephrin, podocin, and podoplanin was decreased by anti–IP-10 mAb treatment. Representative agarose gel electrophoretic patterns for PCR product are shown.
Figure 11. Effect of anti–IP-10 mAb treatment on rat glomerular mRNA expression for several chemokines and cytokines on day 5 after induction of Thy1.1 GN. (A) mRNA expression for IP-10, IFN-γ, CXCR3, macrophage-derived chemokine, and IL-4 was semiquantified by RT-PCR using cDNA corresponding to 750 ng of RNA. Ratio of the densitometric signals of chemokines, cytokines, and CXCR3 to that of internal control (GAPDH) was analyzed. The data are shown as a ratio (%) relative to the RVG1-injected control group and are expressed as mean ± SD of three independent experiments. (B) Representative agarose gel electrophoretic patterns of PCR products are shown.
Injected Anti–IP-10 mAb Was Observed along the Capillary Loop
Anti–IP-10 mAb injected into normal rat was observed along the glomerular capillary loop (Figure 12A). Clear staining of anti–IP-10 as capillary loop pattern was observed in rats with Thy1.1 GN (Figure 12B). RVG1 injected into rat with Thy1.1 GN was not detected (Figure 12C).
Figure 12. Localization of anti–IP-10 mAb injected into normal and Thy1.1 GN rats. Normal and Thy1.1 GN rats received an injection of 5 mg/100 g body wt anti–IP-10 or RVG1 and were killed 10 min after injection. (A) Anti–IP-10 mAb injected into normal rat was observed along the capillary loop. (B) Clear staining along the capillary loop was observed in rats 5 d after Thy1.1 GN induction. (C) RVG1 injected into rats 5 d after Thy1.1 GN was not detected.
Discussion
IP-10 and its receptor CXCR3 are reported to be expressed in a variety of cells (4–8⇓⇓⇓⇓). However, the expression and the function of IP-10/CXCR3 in glomeruli are poorly understood. Although the study with IP-10–deficient mouse (32) was reported, there was no description of the disorder of kidney. In this study, we investigated whether and where IP-10 is expressed in normal glomeruli. The expression of IP-10 was detected in normal rat glomeruli by IF, RT-PCR, and Western blot analysis (Figure 1, Panel I). Immunostaining of IP-10 was demonstrated as a linear-like pattern along the glomerular capillary loop, which suggested that IP-10 was expressed on the glomerular visceral epithelial cell (podocyte). We observed that IP-10 was expressed also in cultured podocyte in both undifferentiated and differentiated conditions (Figure 1, Panel II). CXCR3 staining is also observed as an epithelial pattern along the glomerular capillary wall, and CXCR3 expression was detected in cultured podocyte at differentiated condition. Podocyte is a highly differentiated cell characterized by interdigitating foot processes covering glomerular basement membrane and is accepted to play an important role to keep the normal permselectivity of the glomerular capillary wall. Recently, podocyte was reported to have a multi-role for maintaining the glomerular function (33,34⇓). In 1996, Gomez-Chiarri et al. (17) reported that mRNA expression of IP-10 was detected in the cultured glomerular epithelial cells treated with adriamycin. Recently, Huber et al. (35) reported that CXCR3 was expressed on podocyte and that CXCR3 may contribute to podocyte injury. Romagnani et al. (15) reported that IP-10 was expressed in podocyte in cases of human disease. However, few studies analyzing the expression and the function of IP-10/CXCR3 in podocyte with in vivo material are reported. In this study, we clearly demonstrated that IP-10/CXCR3 is expressed on podocyte. In the previous report with a murine acute colitis model, we demonstrated that IP-10 could directly inhibit the proliferation and migration of epithelial cell (36). Chemokines generally have a capability to activate the adhesion molecules on various types of cells (37,38⇓). It is reported that the expressions of integrin and cadherin mediated by IP-10 may contribute to firm cellular interactions with an adjacent cell and with extracellular matrix (36). Although the function of IP-10 on podocyte is uncertain, we hypothesized that IP-10 contributes to maintaining the highly differentiated structure and the function of podocyte.
Next, to test the hypothesis, we analyzed the function of IP-10 expressed on podocyte by the blocking study. A mAb with blocking activity to IP-10 was prepared in mice by immunization with rat IP-10/Fc fusion protein (23). The blocking activity of the anti–IP-10 mAb used in this study was confirmed in the chemotactic assay and in the study with the murine models of acute colitis (36) and encephalomyelitis (39). We confirmed that the anti–IP-10 mAb has no cross-reactivity with other chemokines such as monokine induced by IFN-γ, macrophage inflammatory protein-1α, and MDC (40). To elucidate the effect of the anti–IP-10 mAb treatment on the podocyte, we analyzed the expression of IP-10 and the functional molecules of podocyte such as nephrin, podocin, and podoplanin. Nephrin (26,41⇓) and podocin (27,42⇓) are critical components of slit diaphragm that interacts adjacent foot processes and functions to maintain the barrier function of the glomerular capillary wall (43,44⇓). Podoplanin is believed to be a critical molecule for maintaining the foot process structure of podocyte (45). The glomerular expression of IP-10 and these functional molecules of podocyte were confirmed to be decreased by the treatment of anti–IP-10 mAb (Figure 2). In this study, we showed that the anti–IP-10 mAb injected into normal rat specifically bound to the IP-10 expressed on podocyte (Figure 12A). The injected anti–IP-10 mAb comes across the glomerular basement membrane and acts on its target IP-10 expressed on the podocyte. This can explain why a relatively high dose of the anti–IP-10 treatment was necessary to affect the podocyte (Figure 2). The effects of anti–IP-10 mAb treatment observed in in vivo studies were also confirmed by in vitro study using cultured podocyte (Figure 3). The next question that arose was whether the effect of anti–IP-10 is a specific one. Because our group has been producing some mAbs that recognize glomerular antigens (25,44⇓), we have several mAbs with binding activity to rat podocyte. However, most of these mAbs have no biologic activity. We have experienced that these mAbs did not affect the expression of the functional molecules of podocyte, even when a large amount of antibody was injected into rats (data not shown). In the study with cultured podocyte, we showed that even the very high dose of anti-podocalyxin antibody treatment did not affect the expression for IP-10 or podocyte functional molecules. From these observations, it is conceivable that the specific intervention of IP-10 by anti–IP-10 mAb depressed the expression of the functional molecules of podocyte. The findings suggest that IP-10 plays a role for maintaining the podocyte function.
Next, we investigated the role of IP-10 on the pathogenesis of the glomerular disease. In this study, we adopted Thy1.1 GN. Although Thy1.1 GN is commonly used as a model of mesangial proliferative GN, which is one of the most important diseases in the nephrology field, no studies analyzing the role of IP-10 in Thy1.1. GN are reported. Another reason that we adopted Thy1.1 GN in this study is that we previously reported that podocyte dysfunction was highly concerned with the prognosis of mesangial alterations (28). We analyzed the kinetics of the expression of IP-10 during the development of Thy1.1 GN. Thy1.1 GN is characterized by mesangiolysis followed by inflammatory cell infiltration, mesangial cell proliferation, and the consequent mesangial matrix expansion. The elevated expression of IP-10 mRNA was already observed on day 1, and it peaked on day 5 after the induction of Thy1.1 GN, when the amount of urinary protein excretion peaked (Figure 4, Panel I). The kinetics of the IP-10 expression was basically coincident with that of proteinuria. The clear immunostaining of IP-10 along the glomerular capillary loop was observed in the kidney section on day 5 of Thy1.1 GN. To investigate cellular distribution of glomerular IP-10, we conducted double-staining IF studies with several cellular markers, including OX-1, OX-19, and ED1 as inflammatory cells; RECA-1 as an endothelial cell; α-SMA as a mesangial cell; and 4D5 as a podocyte. The double-staining IF studies elucidated that the elevated expression of IP-10 on day 5 was detected mainly at the podocyte, although some positive staining of IP-10 was also detected at the mesangial area (Figure 4, Panel II). No IP-10–positive cells co-stained with OX-1, OX-19, or ED1 were detected. We also analyzed the expression of CXCR3 during Thy1.1 GN. The expression of CXCR3 mRNA increased with time after the Thy1.1 GN induction (Figure 5, Panel I). Clear immunostaining of CXCR3 was observed on day 5. The marked elevation of CXCR3 expression was still detected on day 14, when severe mesangial cell proliferation and mesangial matrix expansion were observed. The double-staining IF studies of CXCR3 showed that an increased expression of CXCR3 was also detected mainly at the podocyte (Figure 5, Panel II). It was confirmed that OX-1–positive inflammatory cells were not stained with anti-CXCR3 antibody at any stages of Thy1.1 GN investigated in this study. These findings indicate that the inflammatory cells infiltrating into glomeruli are not responsible for the increased expression of IP-10 and CXCR3 in Thy1.1 GN.
Next, we further analyzed the role of IP-10 on the pathogenesis of Thy1.1 GN by blocking study with anti–IP-10 mAb. We demonstrated here that anti–IP-10 treatment increased the amount of proteinuria. It was also shown that the anti–IP-10 mAb treatment exacerbated the mesangial morphologic alterations (Figures 6 and 7⇑). It is thought that the exacerbated mesangiolysis in the early phase resulted in the increased α-SMA staining and the expansion of mesangial matrix in the following phase of Thy1.1 GN. For the explanation of the mechanism of the exacerbation of Thy1.1 GN, first we have to mention that the IP-10 blockade modulated the inflammatory responses, because Th1 cytokines are considered to be involved in the development of Thy1.1 GN (46). However, there were no significant differences in numbers of OX-1–, ED1- or CD5-positive inflammatory cells between both groups (Figure 6, Panel II-A and B). Anti–IP-10 mAb treatment did not affect the expression of IFN-γ, MDC, or IL-4 in Thy1.1 GN (Figure 11). In the previous studies in the experimental models of encephalomyelitis (39) and hepatitis (40), we reported that anti–IP-10 treatment did not affect the Th1/Th2 polarization. From these observations and the previous reports, we consider that the exacerbated glomerular alteration caused by the IP-10 blockade did not result from the modulated inflammatory response. Next, we should mention the possibility that mesangial cell damage caused by the anti–IP-10 treatment directly resulted in the exacerbated mesangial alterations. Romagnani et al. (15,47⇓) reported that IP-10 and CXCR3 are expressed in the mesangial cells. In this study, the increased expression of IP-10 in Thy1.1 GN was observed not only on the podocyte but also in the mesangial area. These findings suggest that the anti–IP-10 mAb treatment may directly exacerbate the mesangial injury in Thy1.1 GN. However, we showed here that IP-10 staining was observed as an epithelial pattern along the glomerular capillary wall in normal rat section and that an increased expression of IP-10 in Thy1.1 GN was mainly detected on the podocyte and that in the mesangial area was partly and weak. These findings suggested that the injected anti–IP-10 mAb mainly acts on podocyte. More discontinuous staining of nephrin, podocin, and podoplanin was observed in the anti–IP-10 treatment group of Thy1.1 GN than in RVG1 control group (Figure 8). We also observed that the anti–IP-10 treatment decreased the glomerular mRNA expression for nephrin, podocin, and podoplanin on day 5 after induction of Thy1.1 GN (Figure 10, Panel I). It should be mentioned that anti–IP-10 treatment did not decrease the mRNA expression for IP-10 on day 5 of Thy1.1 GN, although the treatment in normal rats decreased IP-10 expression. We cannot give a clear explanation for these results. We consider that because mRNA expression for IP-10 clearly increased on day 5 of Thy1.1 GN (Figure 4A), the anti–IP-10 mAb treatment did not affect the mRNA expression for IP-10 at this time point of disease. However, we also observed that anti–IP-10 mAb injected into rat at 5 d of Thy1.1 GN was detected as a podocytic pattern along the capillary loop (Figure 12B). The finding also indicates that the injected anti–IP-10 mAb evidently acts on podocyte.
We have previously demonstrated that the minor podocyte injury evaluated by the decreased expression of the functional molecules of podocyte could exacerbate the mesangial injury of Thy1.1 GN (28). Shih et al. (48) reported that mesangial cell proliferation and the matrix expansion were observed in the knockout mouse of CD2-associated protein, a component of slit diaphragm. It was discussed that the mesangial alterations in this mouse were brought on by the dysfunction of the podocyte. Taken together with these previous reports and all of the observations in this study, it is conceivable that the exacerbated proteinuria and mesangial alterations in Thy1.1 GN resulted from the podocyte dysfunction that was caused by IP-10 blockade.
In conclusion, IP-10 was expressed on normal glomerular podocyte, and these expressions were elevated in Thy1.1 GN. A receptor for IP-10, CXCR3, showed similar expression patterns to that of IP-10. IP-10 plays a role for maintaining the podocyte function. The blockade of IP-10 exacerbated the glomerular alterations in Thy1.1 GN probably by disturbing the podocyte function.
Acknowledgments
This work was supported by Grant-Aids for Scientific Research (B) (13557084 and 14370317 to H.Ka. and 08457286 to F.S.) from the Ministry of Education, Science, Culture and Sports of Japan.
We express our gratitude to Dr. Katsue Kanno, Dr. Yoshio Morioka, and Dr. Koichi Suzuki for helpful discussions. We also thank M. Kayaba and C. Nagasawa for tremendous technical assistance.
- © 2003 American Society of Nephrology
References
Jump to section
More in this TOC Section
Cited By...
- Role of calcineurin (CN) in kidney glomerular podocyte: CN inhibitor ameliorated proteinuria by inhibiting the redistribution of CN at the slit diaphragm
- Tumor Necrosis Factor-{alpha} Produced in the Kidney Contributes to Angiotensin II-dependent Hypertension
- Chemokines in Renal Injury
- Chemokine Receptor CXCR3 Mediates T Cell Recruitment and Tissue Injury in Nephrotoxic Nephritis in Mice
- Compartment-Specific Expression and Function of the Chemokine IP-10/CXCL10 in a Model of Renal Endothelial Microvascular Injury
- IFN-Inducible Protein-10 Plays a Pivotal Role in Maintaining Slit-Diaphragm Function by Regulating Podocyte Cell-Cycle Balance