Anti-Inflammatory Effect of Hepatocyte Growth Factor in Chronic Kidney Disease: Targeting the Inflamed Vascular Endothelium

Cell Culture Studies

Human umbilical vein endothelial cells (HUVEC) were purchased from VEC Technologies (Rensselaer, NY) and maintained in MCDB-131 complete medium. HUVEC (two to eight passages) were seeded on gelatin-coated (1.5%) cultures at approximately 80% confluence. After growth for 24 h in complete medium, cells underwent serum starvation for 6 h in Medium 199. Human recombinant HGF (Genentech, South San Francisco, CA) and human recombinant TNF-α (R&D Systems, Minneapolis, MN) were added to the culture with fresh serum-free medium at a final concentration of 100 and 0.1 ng/ml, respectively, or as otherwise indicated.

Human monocytes (THP-1) and rat alveolar macrophages (RAM) were purchased from American Type Culture Collection (ATCC; Manassas, VA) and were cultured, respectively, in RPMI supplemented with 10% FBS and Ham’s F12K that contained 15% FBS. For fluorescence-viable labeling, THP-1 and RAM cells (1 × 10⁷) were incubated in medium that contained 5 μg/ml Calcein-AM (Invitrogen, Carlsbad, CA) at 37°C for 30 min. Excess dye was removed by washing three times with PBS.

Animal Studies

Study 1 was designed to determine how manipulation of HGF activity affects endothelial inflammation in an experimental chronic renal disease. Male Sprague-Dawley rats (Harlan Sprague-Dawley, Indianapolis, IN) that initially weighed 180 to 210 g were housed in an approved animal care facility and fed a standard diet. Five-sixths renal ablation was performed as described previously (13–15). Two weeks after surgery, rats were randomly assigned to one of three groups (n = 7; Figure 1A): (1) HGF-treated group: Rats received recombinant human HGF as a continuous intravenous infusion via an implanted osmotic minipump into the jugular vein at a dose of 100 μg/kg per d for 14 d; (2) anti-HGF antibody (α-HGF)–treated group: Rats received daily injections of a high-titer, rabbit polyclonal anti–HGF-neutralizing antibody intravenously through the tail vein for 14 d; or (3) preimmune IgG-treated group: Rats received a continuous intravenous infusion of saline via minipump as well as daily tail-vein injections of rabbit preimmune IgG. Sham-operated rats with no renal ablation served as normal controls (n = 7). Two weeks after treatment, rats were killed.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Schematic diagrams illustrate experimental design for animal studies. (A) In Study 1, 2 wk after renal ablation or sham operation (Sham), rats were treated for another 2 wk with minipump continuous intravenous (iv) infusion of hepatocyte growth factor (HGF), daily (qd) tail-vein injection of anti-HGF antibody (α-HGF), or minipump vehicle infusion plus daily tail-vein injection of preimmune IgG (Ctrl-IgG). (B) In Study 2, 2 wk after renal ablation or sham operation (Sham), rats were treated for another 2 wk with minipump continuous intravenous infusion of HGF or vehicle before a bolus injection of saline, anti–E-selectin antibody (αE-sel), or preimmune IgG through the carotid artery. Four hours later, all rats were infused with Calcein-AM–labeled rat macrophages and killed after 30 min.

Study 2 was performed to test how HGF or E-selectin blockade affects macrophage adhesion to the endothelium in vivo in the remnant kidney. Figure 1B provides a summary of the treatments that were given to each of the five groups (n = 5). For examination of macrophage adhesion in vivo, rats were anesthetized, a tracheostomy was performed, and catheters were inserted into the jugular vein and carotid artery. All rats received an infusion of plasma as described previously (13–15) to maintain euvolemic conditions. Rats also received a bolus injection of a rabbit anti–E-selectin–blocking antibody (Santa Cruz Biotechnology, Santa Cruz, CA), preimmune IgG, or saline into the left carotid artery. Four hours later, fluorescence-labeled RAM cells (1 × 10⁴) suspended in 1 ml of normal saline were infused into the carotid artery as a bolus injection. Rats were killed 30 min after macrophage administration.

In all animal experiments, kidneys were excised for analysis. One portion was fixed in 10% phosphate-buffered formalin for morphology and immunohistochemistry analyses. Another part was frozen immediately for cryostat sectioning. The remainder was snap-frozen in liquid nitrogen and stored at −80°C for RNA and protein extraction. For quantification of the fluorescence RAM cells that were sequestrated in the kidney, kidney homogenates with equal amount of protein (100 μg) were subjected to fluorometric analysis in a fluorescence plate reader as described below.

Morphologic Studies

Formalin-fixed kidneys were embedded in paraffin and prepared in 3-μm-thick sections. Immunohistochemical staining for ED-1 was performed as described previously (18). The antibody for ED-1 was purchased from Serotec (Oxford, UK). Indirect immunofluorescent staining of E-selectin was carried out on methanol/acetone-fixed (1:1) frozen cryostat sections using rabbit polyclonal anti–E-selectin antibody (Santa Cruz Biotechnology). Frozen sections were double stained for aminopeptidase P, a marker for vascular endothelial cells, using a specific murine mAb, JG-12 (Bender MedSystems GmbH, Vienna, Austria), which has been shown to bind specifically to endothelial cells of blood vessels in rat kidney (32). The Alexa Fluor 488 goat anti-rabbit and Alexa Fluor 594 goat anti-mouse antibody (Invitrogen) were as used as secondary antibody. As a negative control, the primary antibody was replaced by nonimmune serum from the same species; no staining occurred. Finally, all sections were counterstained with 4′,6-diamidino-2-phenylindole or Evans blue and mounted with Vectashield mounting medium (Vector Laboratories, Burlingame, CA). For visualization of fluorescence monocytes that were sequestrated in the tissue, cryostat sections were fixed with methanol/acetone and then subjected to counterstaining. Morphologic results were interpreted with a Nikon Microphot-FX fluorescence microscope equipped with a Spot II digital camera.

Static Monocytic Adhesion Assay

Adhesion studies were performed with the human monocytic cell line THP-1 under static conditions (33). Fluorescence-labeled THP-1 cells were resuspended in Medium 199 (1 × 10⁶ cells/ml). For static adhesion assays, HUVEC monolayers with equal cell numbers in 12-well plates were treated with HGF and/or TNF-α for the stated intervals. HUVEC cultures were washed twice with PBS before addition of 1 ml (1 × 10⁶) of labeled THP-1 cells per well. The plates were incubated for 30 min at 37°C. After incubation, the monolayers were washed gently three times with PBS and photomicrographs were taken using a phase-contrast microscope. Adherent monocytes were lysed with RIPA buffer (1% Nonidet P-40, 0.1% SDS, 100 μg/ml PMSF, 0.5% sodium deoxycholate, 1 mM sodium orthovanadate, 2 μg/ml aprotin, 2 μg/ml leupeptin, and 5 mM EDTA in PBS), and fluorescence intensity was measured by a Spectramax GEMINI EM fluorescence plate reader (Molecular Devices, Sunnyvale, CA) at an excitation wavelength of 485 nm and emission at 530 nm. HUVEC monolayers that were adhered with nonlabeled THP-1 cells served as negative controls.

Transient Transfection and Reporter Gene Assay

The reporter construct pGL-3κB-Luc, a firefly luciferase reporter gene driven by three copies of an NF-κB consensus sequence, was provided by Dr. Fu (Division of Gastroenterology, Brown Medical School). HUVEC were transfected by electroporation using the Amaxa HUVEC Nucleofector kit (Amaxa GmbH, Koeln, Germany). A fixed amount (50 ng) of internal control reporter Renilla reniformis luciferase driven under thymidine kinase (TK) promoter (pRL-TK; Promega, Madison, WI) was co-transfected for normalizing the transfection efficiency. After transfection with equal amounts of reporter constructs, HUVEC were seeded and treated as indicated. Total cell lysates were collected, and luciferase activity was determined using the Dual-Glo luciferase assay kit (Promega). The relative NF-κB transactivation activity was assessed as the fold change of firefly luciferase activity after normalization for Renilla reniformis luciferase activity.

DNA Affinity Precipitation Assay

The interaction between NF-κB and DNA was examined by DNA affinity precipitation assay as described elsewhere (34) with minor modifications. After various treatments, HUVEC were collected and washed with PBS. The nuclear extracts were prepared with the NE-PER nuclear extraction kit (Pierce, Rockford, IL). A 50-μg sample of the nuclear lysate was mixed with 2× Laemmli’s sample buffer, boiled, and used for Western blotting as an input fraction. The residual lysate was used for the DNA precipitation assay. The 5′-biotinylated, double-stranded oligonucleotides (0.2 nM) that contained two putative NF-κB cis elements at the nucleotide position −129 to −80 bp of human E-selectin promoter (5′-CGTGGATATTCCCGGGAAAGTTTTTGGATGCCATTGGGGATTTCCTCTTT-3′; Genbank accession no. M64485) were mixed with 200 μg of nuclear extracts. DNA affiliated protein then was pulled down by streptavidin-agarose and washed with ice-cold PBS that contained 0.5% Triton X-100 and 5 mM EDTA. Proteins were eluted from the beads by the addition of 2× Laemmli’s sample buffer, followed by boiling for 10 min and immunoblot analysis with specific anti–NF-κB p65 antibody (Santa Cruz Biotechnology).

Chromatin Immunoprecipitation Assay

The in situ interaction between NF-κB and E-selectin promoter in HUVEC was examined by chromatin immunoprecipitation assay (35) by using a commercially available kit (Upstate Biotechnology, Charlottesville, VA) according to the manufacturer’s instructions. Briefly, HUVEC were fixed and cross-linked with 4% formaldehyde. After collection of cell pellets and sonication, aliquots of the samples were set aside as input fraction, and the rest were subjected to immunoprecipitation using an anti–NF-κB p65 antibody or preimmune IgG and Protein A agarose that were used to pull down the immune complexes. After elution, precipitated chromatin as well as the input fraction was heated at 65°C to reverse cross-links, and the DNA was extracted with Qiagen PCR purification kit (Qiagen, Valencia, CA). DNA sequences that spanned the κB responsive elements in the promoter region of E-selectin gene were amplified by PCR using specific primers (forward 5′-GGCATGGACAAAGGTGAAGT-3′; reverse 5′-CAGCCAAGAACAGCTGAACA-3′; Genbank accession no. M64485) for a number of cycles in the exponential phase as estimated in pilot experiments. DNA samples that were extracted from input fractions were amplified in parallel for normalization. PCR products that were resolved in approximately 1.5 to 2% agarose gels were photographed under ultraviolet light.

Semiquantitative Reverse Transcription–PCR

Total RNA was prepared and reverse transcription–PCR was performed as before (18) using specific primers for human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and E-selectin (forward 5′-TTATGATGAGGCCAGTGCTTATTG-3′; reverse, 5′-CTTTTTGCCTATTGTTGGGTTCAC-3′; Genbank accession no. NM_000450) for a number of cycles in the exponential phase as estimated in pilot experiments. PCR products that were resolved in approximately 1.5 to 2% agarose gels were photographed under ultraviolet light.

Western Immunoblot Analysis

Rat kidneys were homogenized and HUVEC monolayers were lysed in RIPA buffer supplemented with protease inhibitors. Protein concentrations were determined using a bicinchoninic acid protein assay kit (Sigma, St. Louis, MO). Samples with equal amounts of total protein (40 to 80 μg/ml) were fractionated by 7.5 to 10% SDS–polyacrylamide gels under reducing condition and analyzed by Western immunoblot as described previously (16). The antibodies against E-selection, NF-κB p65, IκBα, and actin were purchased from Santa Cruz Biotechnology, and those for phosphorylated NF-κB p65 and phosphorylated IκBα were purchased from Cell Signaling Technology (Beverly, MA).

Statistical Analyses

Immunoblot bands were scanned and the integrated pixel density was determined using a densitometer and the National Institutes of Health image analysis program. All data are expressed as mean ± SD. Statistical analysis of the data from multiple groups was performed by ANOVA followed by Student-Newman-Kuels tests. Data from two groups were compared by t test. P < 0.05 was considered significant.

HGF Attenuates Endothelial Inflammation in Rat Remnant Kidney

Endothelial cell activation and leukocyte-to-vascular endothelial adhesion are the initial steps in the genesis of an interstitial inflammatory infiltrate (6,7). As shown in Figure 2, immunohistochemistry staining for ED-1, a rat macrophage marker, demonstrated that adhesion of inflammatory cells to vascular endothelium was evident in rat remnant kidney 4 wk after renal ablation (Figure 2C). Adhesion of ED-1–positive cells was markedly reduced by HGF treatment (Figure 2D) and significantly increased after HGF neutralization by a specific antibody (Figure 2E). Absolute counting of ED-1–positive macrophages in the renal sections revealed that the extent of macrophage-to-endothelial adhesion correlated with the magnitude of interstitial inflammation and was modulated by HGF (Figure 2F). This is consistent with the important role of endothelial activation in the renal inflammatory response (6).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

HGF supplement suppresses whereas HGF blockade enhances renal inflammation, accompanied by parallel alternations in endothelial inflammation, marked by adhesion of immune competent cells to renal vascular endothelium. (A through E) Representative micrographs of ED-1 immunohistochemistry staining depicting macrophages adherent to renal vascular endothelium (arrowheads) in kidney sections from sham-operated rats (B) and remnant kidney rats treated with preimmune IgG (Ctrl-IgG; A and C), HGF (D), or an anti-HGF antibody (α-HGF; E). Sections processed with nonimmune serum from the same species as the primary antibody were examined as a negative control (A) for immunohistochemistry. (E) Counting of ED-1–positive cells in glomeruli (□), tubulointerstitium (□), and those adherent to vascular endothelium (▪) in kidney sections from each group. *P < 0.05, ^#P < 0.05 versus other groups with remnant kidney disease (n = 7). Magnifications: ×100 in A through E; × 400 for inserts.

Adhesion of Monocytes to TNF-α–Activated Endothelial Cells Is Attenuated by HGF

The direct effects of HGF on leukocyte adhesion to activated endothelial cells were investigated. HUVEC monolayers were pretreated with HGF or vehicle and then stimulated by the proinflammatory cytokine, TNF-α, for different intervals. Calcein-AM-labeled human monocytes (THP-1 cells) were applied for static adhesion assay. Microphotographs showed that very few cells were adherent to vehicle-treated (Figure 3A) or HGF-treated (Figure 3B) endothelial monolayers. TNF-α increased THP-1 cell adherence (Figure 3C), and this effect was significantly blunted by HGF (Figure 3D). Fluorometric analyses of cell lysates were consistent with the microscopic observations and revealed time-dependent suppression of monocytic adhesion to activated HUVEC by HGF (Figure 3E).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

HGF abrogates monocytic adhesion to endothelial cell monolayers that is activated by TNF-α, mimicking the action of E-selectin–blocking antibody. Human umbilical vein endothelial cells (HUVEC) were pretreated with vehicle (A and C) or 100 ng/ml HGF (B and D) for 30 min before addition of 0.1 ng/ml TNF-α (C and D) or vehicle (A and B). After various intervals (4 h for A and B, and 2, 4, and 6 h for E), HUVEC monolayers were washed and Calcein-AM–labeled THP-1 cells were applied. Before TNF-α stimulation, a rabbit anti–E-selectin antibody (2 μg/ml) or preimmune IgG (2 μg/ml) was given to demonstrate the role of E-selectin in monocyte adhesion. (A and B) Representative microphotographs demonstrate THP-1 adhesion to HUVEC monolayers after different treatments. (E) Aliquots of cell lysates were subjected to fluorometric analysis to quantify the inhibitory effect of various maneuvers on THP-1 adhesion (n = 3).

After endothelial activation, E-selectin is induced and captures circulating leukocytes (10). To determine whether E-selectin was responsible for the static adhesion that we observed in endothelial cell cultures in vitro, we used a rabbit polyclonal anti–E-selectin antibody to block the adhesion molecule. Shown in Figure 3E, preimmune IgG from the same species as the anti–E-selectin antibody had little effect on THP-1 adhesion to TNF-α–activated HUVEC at all time points. In contrast, the E-selectin blocking antibody significantly reduced monocytic adhesion in a pattern that was similar but more potent than HGF. This suggests that the effect of HGF on monocyte-to-endothelial adhesion is consistent with E-selectin blocking.

De Novo Expression of E-Selectin in Cultured Endothelial Cells Is Suppressed by HGF

The direct effect of HGF on endothelial expression of E-selectin in HUVEC was examined. E-selectin expression was minimal under basal conditions but was induced beginning 1 to 2 h after exposure to TNF-α. HGF suppressed TNF-α–induced expression of E-selectin in a dosage- and time-dependent manner; the maximal inhibitory effect was observed at 500 ng/ml and at 4 h, respectively (Figure 4, A and B). Reverse transcription–PCR demonstrated that HGF also reduced E-selectin mRNA levels, suggesting that HGF suppresses E-selectin primarily at the transcriptional level (Figure 4C). Of note, TNF-α–induced E-selectin expression was markedly suppressed by HGF regardless of whether HGF was added before, simultaneously with, or after TNF-α stimulation (Figure 4C).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

HGF suppresses TNF-α–induced expression of E-selectin in cultured endothelial cells. (A and B) HUVEC were pretreated with HGF (100 ng/ml or otherwise indicated) or vehicle for 30 min before stimulation by TNF-α (0.1 ng/ml). Cell lysates were harvested at different time points after TNF-α and analyzed for E-selectin by Western immunoblot. Cytoplasmic β-actin was used as a standard molecule for normalization. (C) HGF (100 ng/ml) treatment was initiated at different time points relative to TNF-α (0.1 ng/ml). E-selectin message was examined by reverse transcription–PCR 4 h later.

HGF Supplement Decreases Vascular Endothelial E-Selectin Expression in Rat Remnant Kidney

For examination of whether HGF also regulates endothelial E-selectin expression in the diseased kidney in vivo, remnant kidneys from rats in which HGF action was unaltered, augmented, or blocked were stained fluorescently for E-selectin. Shown in Figure 5A, there was minimal expression in sham-operated kidney; however, prominent focal E-selectin expression was evident in remnant kidneys primarily in the interstitial area (Figure 5, B and E) and, to a lesser extent, in glomerular tufts (Figure 5F). E-selectin staining co-localized with the vascular endothelial marker JG-12 (Figure 5, E and F), suggesting that E-selectin was expressed predominantly by renal vascular endothelium. Continuous HGF infusion for 2 wk markedly reduced E-selectin staining (Figure 5C) and decreased its absolute abundance as estimated by immunoblot (Figure 5G) and densitometry (Figure 5H). Conversely, neutralization of endogenous HGF by a specific antibody significantly enhanced E-selectin staining (Figure 5D) and substantially increased its expression level (Figure 5, G and H).

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

HGF infusion inhibits and HGF neutralization induces renal vascular endothelial expression of E-selectin. (A through E) Representative micrographs showing immunofluorescent staining for E-selectin (arrowheads) in kidney sections from sham-operated rats (A) and remnant kidney rats treated with preimmune IgG (Ctrl-IgG; B, E, and F), HGF (C), or anti-HGF antibody (α-HGF; D). (E and F) Co-localization of E-selectin (green) with the rat vascular endothelial maker JG-12 (red) in the interstitium (E) and glomerulus (F). Tissue sections were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). (G) Representative immunoblots for E-selectin in kidney homogenates. (H) Arbitrary units of E-selectin abundance in immunoblot analysis expressed as densitometric ratios of E-selectin/actin as folds above that found in the normal control group. *P < 0.05, ^#P < 0.05 versus other disease groups (n = 7). Magnifications: ×100 in A through D; ×400 in E and F.

HGF Affects NF-κB–Mediated E-Selectin Expression in Endothelial Cells

NF-κB activation and subsequent transactivation of downstream proinflammatory genes is a fundamental mechanism that initiates and promotes inflammation in chronic renal disease (1–4,36). Two specific inhibitors for NF-κB, n-tosyl-l-phenylalanine chloromethylketone and pyrrolidine dithiocarbamate, reduced de novo expression of E-selectin in endothelial cells (Figure 6), consistent with the hypothesis that NF-κB regulates E-selectin expression. To explore further whether HGF suppression of E-selectin is mediated by NF-κB inhibition, we examined the effect of HGF on NF-κB–mediated transcriptional activity in HUVEC. HUVEC were transiently transfected with a reporter gene construct driven by three copies of an NF-κB consensus sequence. Shown in Figure 7A, 3κB promoter activity essentially was unchanged by HGF treatment but increased significantly after TNF-α stimulation. HGF co-treatment significantly blunted TNF-α’s effect. The transcriptional activity of NF-κB is fundamentally affected by its interaction with κB cis elements. To determine whether HGF regulates NF-κB by altering DNA binding, we performed DNA affinity precipitation assay on cell lysates by using an oligonucleotide that contained two tandem copies of a κB cis element in the human E-selectin promoter region to pull down NF-κB with binding activity. Immunoblot analysis demonstrated that HGF alone had only a minor effect on constitutive binding of NF-κB p65 to the exogenous oligo (Figure 7B). TNF-α promoted total input as well as oligo-affiliated NF-κB p65, whereas HGF reduced NF-κB p65 binding that was elicited by TNF-α without altering total NF-κB abundance. Chromatin immunoprecipitation assay was used to examine further the inhibitory effect of HGF on NF-κB p65-DNA interaction in situ at the target gene (E-selectin) level. Untreated or HGF-treated HUVEC showed no NF-κB to E-selectin gene binding. In contrast, TNF-α treatment induced recruitment of NF-κB to the E-selectin promoter, and this effect was abrogated by HGF (Figure 7C). Collectively, these findings suggest that HGF suppresses NF-κB-mediated TNF-α–induced E-selectin expression.

• Download figure
• Open in new tab
• Download powerpoint

Figure 6.

De novo expression of endothelial E-selectin is regulated by NF-κB. HUVEC were pretreated with n-tosyl-l-phenylalanine chloromethylketone (TPCK) or pyrrolidine dithiocarbamate (PDTC), two selective NF-κB inhibitors, for 30 min before TNF-α activation. Cell lysates were harvested 4 h later for immunoblot analysis of E-selectin or actin.

• Download figure
• Open in new tab
• Download powerpoint

Figure 7.

HGF suppresses the transcription activity of NF-κB and alters NF-κB–E-selectin gene interactions. (A) HUVEC were transiently transfected with a reporter gene driven by three copies of an NF-κB consensus sequence. After 24 h, cells were treated with HGF (100 ng/ml) and/or TNF-α (0.1 ng/ml) for another 24 h before cell lysates were prepared for luciferase assay. Relative luciferase activity (fold induction) is presented. *P < 0.05 versus TNF-α stimulation alone (n = 3). (B) HUVEC were treated with HGF and/or TNF-α as described in A. After 4 h, cell lysates were subjected to DNA affinity precipitation assay using synthesized oligonucleotides detailed in Materials and Methods. Precipitated or input fractions were analyzed by immunoblot for NF-κB p65. (C) After 4 h of treatment as described in B, HUVEC were fixed and subjected to chromatin immunoprecipitation assay. DNA was extracted from precipitated or input fractions and underwent PCR amplification in the exponential phase using primers that were specific for E-selectin gene.

HGF Intercepts NF-κB Signal Transduction in Endothelial Cells

To further unravel the mechanism by which HGF suppresses NF-κB–meditated E-selectin expression, we examined NF-κB signal transduction (Figure 8A). As documented previously (37), TNF-α induced a rapid activation of the NF-κB pathway in HUVEC, as evidenced by phosphorylation of IκBα and NF-κB p65 as well as degradation of IκB. HGF had little effect at early time points; however, beginning at 45 min, HGF inhibited NF-κB p65 phosphorylation and, to a lesser extent, IκBα phosphorylation so that the restoration of IκB clearly was accelerated at 90 min. To quantify more accurately NF-κB signaling activity, we calculated densitometric ratios of phosphorylated to total IκBα (Figure 8B) and NF-κB (Figure 8C) and confirmed that HGF intercepts TNF-α–induced IκB and NF-κB activation at a late phase.

• Download figure
• Open in new tab
• Download powerpoint

Figure 8.

HGF intercepts NF-κB signal transduction in endothelial cells. (A) HUVEC were treated with HGF (100 ng/ml) and/or TNF-α (0.1 ng/ml) for various intervals. Cell lysates were subjected to immunoblot analysis for phosphorylated or total IκBα or NF-κB p65. (B and C) Arbitrary units of relative IκB or NF-κB activity expressed as the ratio of phosphorylated/total IκBα or NF-κB p65 as fold of nontreated cells at the same time points. *P < 0.05 versus HGF treatment at the same time point (n = 3).

HGF Suppresses E-Selectin–Mediated Macrophage Sequestration in Rat Remnant Kidney

For examination of the functional consequences of HGF suppression of E-selectin in vivo, fluorescence-labeled RAM cells were infused into nephrectomized rats after 2 wk of HGF or vehicle administration. Shown in Figure 9A, a significant number of fluorescent RAM cells were observed in the tubulointerstitial area in vehicle-treated rats; most were adherent to the microvasculature endothelium with some migrating into the renal parenchyma (Figure 9C). In contrast, very few fluorescence cells were noted in the HGF-treated group (Figure 9B). Blocking E-selectin with a specific antibody (Figure 9E) also reduced RAM sequestration as compared with the group that was given preimmune IgG (Figure 9D). Fluorometric analysis of kidney homogenates corroborated the morphologic findings (Figure 9F). These data suggest that suppression of E-selectin contributes to HGF’s inhibitory effect on inflammatory sequestration in CKD.

• Download figure
• Open in new tab
• Download powerpoint

Figure 9.

HGF suppresses E-selectin–mediated inflammatory cell sequestration in rat remnant kidney. (A through E) Fluorescence micrographs show overall sequestration (A, B, D, and E) and detailed adhesion and transmigration (C; arrowheads) of infused macrophages in rat remnant kidney. Two weeks after renal ablation, rats received vehicle (A and C through E) or HGF via a continuous minipump infusion (B) for another 2 wk. Then rats were infused with Calcein-AM–labeled rat macrophages. After 30 min, remnant kidneys were harvested. For estimation of the functional role of E-selectin, a specific anti–E-selectin blocking antibody (αE-sel; E) or preimmune IgG (D) was injected 4 h before macrophage infusion. Tissues were counterstained with DAPI and Evans blue. Sham-operated rats (Sham) were treated with vehicle and injected with preimmune IgG before macrophage infusion. (E) Aliquots of kidney homogenates from differently treated groups with an equal amount of total protein were subjected to fluorometry to quantify sequestered rat alveolar macrophages cells. *P < 0.05 versus vehicle- or IgG-treated rats with remnant kidney disease (n = 5). Magnifications: ×100 in A, B, D, and E; ×400 in C.