Mesangial cell-derived factors alter monocyte activation and function ...

Am J Physiol Renal Physiol 297: F1229–F1237, 2009. First published September 9, 2009; doi:10.1152/ajprenal.00074.2009.

Mesangial cell-derived factors alter monocyte activation and function through inflammatory pathways: possible pathogenic role in diabetic nephropathy Danqing Min,1 J. Guy Lyons,2,3 James Bonner,1 Stephen M. Twigg,1,2 Dennis K. Yue,1,2 and Susan V. McLennan1,2 1

Department of Endocrinology, Royal Prince Alfred Hospital, 2Discipline of Medicine, University of Sydney, and 3Sydney Head & Neck Cancer Institute, Sydney Cancer Centre, Royal Prince Alfred Hospital, Sydney, New South Wales, Australia

Submitted 10 February 2009; accepted in final form 4 September 2009

Min D, Lyons JG, Bonner J, Twigg SM, Yue DK, McLennan SV. Mesangial cell-derived factors alter monocyte activation and function through inflammatory pathways: possible pathogenic role in diabetic nephropathy. Am J Physiol Renal Physiol 297: F1229 –F1237, 2009. First published September 9, 2009; doi:10.1152/ajprenal.00074.2009.— Infiltration of macrophages to the kidney is a feature of early diabetic nephropathy. For this to happen monocytes must become activated, migrate from the circulation, and infiltrate the mesangium. This process involves degradation of extracellular matrix, a process mediated by matrix metalloproteinases (MMPs). In the present study we investigate the expression of proinflammatory cytokines TNF-␣, IL-6, and MMP-9 in glomeruli of control and diabetic rodents and use an in vitro coculture system to examine whether factors secreted by mesangial cells in response to a diabetic milieu can induce monocyte MMP-9 expression and infiltration. After 8 wk of diabetes, the glomerular level of TNF-␣, IL-6, and macrophage number and colocalization of MMP-9 with macrophage were increased (P ⬍ 0.01). Coculture of THP1 monocytes and glomerular mesangial cells in 5 or 25 mM glucose increased MMP-9 (5 mM: 65% and 25 mM: 112%; P ⬍ 0.05) and conditioned media degradative activity (5 mM: 30.0% and 25 mM: 33.5%: P ⬍ 0.05). These effects were reproduced by addition of mesangial cell conditioned medium to THP1 cells. High glucose (25 mM) increased TNF-␣, IL-6, and monocyte chemoattractant protein-1 in mesangial cell conditioned medium. These cytokines all increased adhesion and differentiation of THP1 cells (P ⬍ 0.05), but only TNF-␣ and IL-6 increased MMP-9 expression (50- and 60-fold, respectively; P ⬍ 0.05). Our results show that mesangial cellsecreted factors increase monocyte adhesion, differentiation, MMP expression, and degradative capacity. High glucose could augment these effects by increasing mesangial cell proinflammatory cytokine secretion. This mesangial cell-monocyte interaction may be important in activating monocytes to migrate from the circulation to the kidney in the early stages of diabetic nephropathy. matrix metalloproteinases; matrix degradation; cell adhesion and infiltration; inflammation response DIABETIC NEPHROPATHY is the most common cause of renal failure worldwide (21). Its development is a complex process that involves many cell types, including mesangial cells and monocytes/macrophages. It is well accepted that mesangium matrix accumulation with fibrosis is a prominent feature of the late stage of diabetic nephropathy (7). However, there also is evidence of an early phase of inflammatory cell infiltration by monocytes/macrophages, associated with focal mesangiolysis (8, 18, 23). Although the mesangium matrix accumulation phase of diabetic nephropathy has been studied in detail, the early inflammatory and mesangiolytic phase is much less well characterized.

Address for reprint requests and other correspondence: D. Min, Dept. of Endocrinology, Bld. 96, Royal Prince Alfred Hospital, Missenden Road, Camperdown, NSW 2050, Australia (e-mail: [email protected]). http://www.ajprenal.org

Increased infiltration as well as increased activation of macrophages has been demonstrated in renal biopsies in both experimental diabetes and patients with diabetic nephropathy (3, 4, 6, 10, 18), and kidney macrophage accumulation and activation correlate with the development of diabetic nephropathy (3, 6, 8). Although the exact mechanism of macrophage recruitment to the glomerulus is unknown, increased kidney expression of monocyte chemoattractant protein-1 (MCP-1) is considered to be important in the initiation of this process (5, 9, 11). Once resident in the kidney, increased monocyte activation and differentiation to activated macrophages can further induce inflammatory cytokines, resulting in increased activation and expression of matrix metalloproteinase (MMPs), particularly MMP-9 (2, 26). For monocytes/macrophages to infiltrate the kidney, they must first traverse the blood vessel wall. This involves the MMPs, a family of at least 24 structurally related proteolytic enzymes important in the regulation of many normal and pathological processes (1, 22, 27–29). MMPs are secreted as proenzymes that are activated by proteolytic cleavage and inactivated by specific inhibitors called tissue inhibitors of metalloproteinases (TIMPs) (19, 28). Most MMP genes are not constitutively expressed but can be induced by a number of biologically active agents, which include growth factors such as transforming growth factor-␤ (TGF-␤) and proinflammatory cytokines, e.g., TNF-␣ and ILs (14, 19). With regard to monocytes/ macrophages, proinflammatory cytokines are known to induce the expression of MMPs. Recent data suggest that the expression of gelatinase MMP-9 is associated with activated macrophages (12). Whether macrophages found in the kidney in diabetes express MMP-9 is not known. The chemotaxis of monocytes to the glomerulus in conjunction with subsequent cellular and cytokine changes may be pivotal in the genesis of diabetic nephropathy. The effect of diabetes and cellular interactions on this process in mesangial cells is not well known. Therefore, this study was designed to investigate whether MMP-9 is expressed by macrophages resident in kidney tissue and whether the interaction between mesangial cells and monocytes/macrophages can affect the expression and activation of MMPs and monocyte/macrophage infiltration. MATERIALS AND METHODS

Reagents. Recombinant human MCP-1 protein, antibodies for TNF-␣, IL-6, IL-8, MCP-1, mouse and goat IgG isotype controls, and mouse IgG isotype control [phycoerythrin (PE) conjugated], and ELISA assays for TNF-␣, IL-6, IL-8, MCP-1, and macrophage inflammatory protein-1␤ (MIP-1␤) were purchased from R&D Systems (Minneapolis, MN). Recombinant human TNF-␣ and Cell Titer

0363-6127/09 $8.00 Copyright © 2009 the American Physiological Society

F1229

F1230

KIDNEY-SECRETED FACTORS ALTER MONOCYTE ACTIVATION

96 AQueous assay reagent (MTS assay) were obtained from Promega (Madison, WI); recombinant human IL-6, IL-8, and TRI reagent were obtained from Sigma-Aldrich (St. Louis, MO). Electrophoresis reagents and protein assay reagents were obtained from Bio-Rad (Hercules, CA). The molecular weight marker, SuperScript III RNase Hreverse transcriptase, and Platinum quantitative PCR SuperMix-UDG were obtained from Invitrogen (Carlsbad, CA). Oligo(dT)18 was obtained from Bioline (London, UK). The BioCoat growth factor reduced Matrigel invasion chamber (pore size 8.0 ␮m) and mouse anti-CD68 antibody were obtained from BD Biosciences (San Jose, CA). Mouse anti-human CD11b monoclonal antibody (PE conjugated) was obtained from Abcam (Cambridge, UK). FcR blocking reagent was obtained from Miltenyi Biotec (Bergisch Gladbach, Germany). Mouse anti-MMP-9 antibody was obtained from Chemicon (Millipore, Billerica, MA). Envision 2/G and ARK kits were acquired from Dako (Carpinteria, CA). Proteinase K was obtained from Roche (Mannheim, Germany). The albumin blue fluorescent assay kit was acquired from Active Motif (Carlsbad, CA) and the creatinine assay kit from Cayman Chemical (Ann Arbor, MI). Streptozotocin was obtained from Calbiochem (San Diego, CA). Longacting insulin (Ultratard) was obtained from Novo-Nordisk (Bagsvaerd, Denmark). Rodent studies and renal pathology. Diabetes was induced in 6-wk-old male Sprague-Dawley rats (ARC, Perth, Australia) (n ⫽ 6) using streptozotocin (65 mg/kg ip). A further six age-matched animals, injected with citrate buffer, served as nondiabetic controls. All diabetic rats (fasting blood glucose ⬎15 mM) were treated with 2 units of long-acting insulin (Ultratard) every second day to maintain body weight and prevent ketoacidosis. Fasting blood glucose levels were measured weekly throughout the study as well as on the day before termination and were in the range of 19 –28 mM. Spot urine specimens were collected at termination for measurement of urinary albumin-to-creatinine ratio (ACR) using commercially available kits. This study was approved by the Animal Ethics Committee of Sydney South West Area Health Service and was carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and national laws. After 8 wk of diabetes, the animals were euthanized, and the kidneys were removed, formalin-fixed, and embedded in paraffin for staining with periodic acid-Schiff’s base (PAS) and hematoxylin for severity of glomerular sclerosis and immunohistochemical analysis of macrophage number as well as colocalization of macrophages with MMP-9. The effect of diabetes on the expression of glomerular TNF-␣ and IL-6 was also studied. The severity of glomerular sclerosis was graded on a number scale from 0 to 4 as follows: 0 score, no change; 1⫹, the tuft of ⬍25% of the glomerulus is sclerotic; 2⫹, ⬍50% is sclerotic; 3⫹, ⬍75% is sclerotic; and 4⫹, 75–100% is sclerotic. Scoring was performed in a blinded fashion, and the values of the glomerular sclerosis index were averaged in 75 glomeruli observed in each sample. For immunohistochemical analysis of TNF-␣ and IL-6, the paraffin sections (4 ␮m) were cut, deparaffinized, and rehydrated. To retrieve antigenicity, the sections were incubated for 10 min in 10 mM sodium citrate buffer using a microwave oven. Endogenous peroxidase activity was blocked by further pretreatment with 1% H2O2-methanol followed by incubation in blocking solution. The sections were then incubated with either TNF-␣ or IL-6 antibody at a concentration of 1:200. The primary antibodies, which were diluted in PBS with 2% BSA, were applied to tissue sections and incubated at 4°C overnight in a humidified atmosphere. The slides were then washed with PBS and incubated with the appropriate horseradish peroxidase-conjugated secondary antibody followed by avidin-biotin peroxidase complex (Vector, Burlingame, CA). Immunoreactivity was visualized with the chromagen diaminobenzidine tetrachloride (DAB, 0.025%; Sigma) in PBS. Staining for the negative controls proceeded in the same fashion as for the primary antibodies. All sections were counterstained with hematoxylin. The TNF-␣- and IL-6-positive cells in the glomerulus AJP-Renal Physiol • VOL

were counted at high magnification (⫻400) in 20 glomeruli per animal. To evaluate the colocalization of MMP-9 and CD68 (a macrophage marker), tissue sections were microwaved and pretreated with proteinase K (0.5 units, 3 min) to retrieve antigens. Immunostaining was then performed sequentially using the Envision 2/G and ARK kits as described by the manufacturer and according to the method of van der Loos et al. (25). Sections were incubated first in mouse anti-CD68 antibody (1:100) followed by DAB for secondary antibody. Sections were then incubated with mouse anti-MMP-9 antibody (1:50) followed by permanent red chromagen for secondary antibody. CD68stained cells and their colocalization (seen as a brown color) were counted at high magnification (⫻400) in 20 glomeruli per animal. Immunostained interstitial cells were counted in 25 consecutive highpower (⫻400) interstitial fields. Cell culture. The human promonocytic leukemic cell line (THP1 cells; ATCC, Manassas, VA) were routinely cultured under serumfree conditions containing 0.1% BSA (16). Primary cultures of human fetal mesangial cells were obtained from isolated glomeruli, maintained in RPMI medium containing 10% fetal calf serum, and used between the third and fourth passage (13). All experiments were performed under serum-free conditions using cells either in coculture (Co: THP1 ⫹ mesangial cells) or individually in media containing either normal (5 mM; NG) or high-glucose concentration (25 mM; HG) and 0.1% BSA in the presence or absence of conditioned medium (CM; 20%) from the other cell type. In relevant experiments, THP1 cells or mesangial cells were also cultured in the presence of 20 mM mannitol and 5 mM glucose as an osmotic control. Ethical approval for these studies was obtained from the Human Ethics Committee of the University of Sydney. For the coculture experiments, undifferentiated THP1 cells (0.5 ⫻ 106/ml) were layered onto 50 – 60% confluent monolayers of mesangial cells and cultured for 2–5 days before removal of CM. To investigate the effect of secreted factors, CM was collected from mesangial cells cultured under serum-free conditions in NG or HG medium containing 0.1% BSA for 24 or 48 h. To test the effect of addition of proinflammatory cytokines, THP1 cells were treated with different concentrations of TNF-␣, IL-6, MCP-1, or IL-8 for 48 h or cultured for 24 h in the presence of mesangial cell CM (20%) that had been preincubated with anticytokine antibody (10 ␮g/ml) at 37°C for 1 h. All cultures were maintained at 37°C in a humidified atmosphere of 5% CO2 in air. Measurement of MMP and TIMP expression and activity by 2-methoxy-2,4-diphenyl-3(2H)-furanone-labeled gelatin zymography and reverse zymography. For all samples, the protein concentration in medium was determined using the Bio-Rad DC assay, and aliquots containing 10 ␮g of total protein were analyzed by 2-methoxy-2,4diphenyl-3(2H)-furanone (MDPF)-labeled gelatin zymography and reverse zymography according to the method of Min et al. (15). Measurement of MMP-9 gene expression by real-time PCR. To examine the effect of CM on MMP-9 gene expression, cells were cultured in the presence of CM as described above. After 48 h, the cells were washed with PBS, and RNA was isolated using TRI reagent. A total of 2 ␮g of RNA from each sample were transcribed to cDNA using oligo(dT)18 (25 pmol) and SuperScript III RNase Hreverse transcriptase. The samples were analyzed in duplicate by quantitative real-time RT-PCR (Rotor-Gene-3000; Corbett Research, Sydney, Australia) using SYBR green flurophore. Briefly, all amplicons were amplified from 1 ␮l of cDNA using quantitative PCR SuperMix-UDG, 5 pmol of each forward and reverse primer, and the following PCR conditions: 50°C for 2 min, 95°C for 5 min, followed by 45 cycles of 95°C for 10 s, 63°C for 15 s, and 72°C for 20 s. The expression of the housekeeping gene ␤-actin was measured as control. The primers used generated a single band of 336 bp for MMP-9 and 164 bp for ␤-actin as follows: for MMP-9: forward 5⬘-CTG CCC CAG CGA GAG ACT CTA C-3⬘, reverse 5⬘-GCT GTC AAA GTT CGA GGT GGT A-3⬘; for ␤-actin: forward 5⬘-GAA TTC TGG CCA

297 • NOVEMBER 2009 •

www.ajprenal.org


Table 1. Animal characteristics Body Weight, g

Day 4

Day 19

Day 55

Urinary ACR at Day 55, mg/ mmol

517⫾15 387⫾25*

6.5⫾3.4 21.5⫾3.6*

8.4⫾1.9 23.5⫾2.6*

8.5⫾2.6 22.5⫾5.6*

14.0⫾10.2 52.9⫾46.8†

Blood Glucose Level, mM

Control Diabetic

Values are means ⫾ SD. ACR, albumin-to-creatinine ratio. *P ⬍ 0.01; † P ⬍ 0.05, significantly different from control (by t-test).

CGG CTG CTT CCA GCT-3⬘, reverse 5⬘-AAG CTT TTT CGT GGA TGC CAC AGG ACT-3⬘. The identify of the amplicons was confirmed by sequence analysis. Results were calculated using the “⌬⌬” method and expressed as a ratio of control values corrected for abundance of the housekeeping gene ␤-actin (20). Data were generated from three independent experiments. Measurement of degradative activity. To examine the effect of coculture of THP1 cells and mesangial cells on degradative activity, the CM was assayed using a solution-phase fluorescent assay as previously described (16). Determination of monocyte viability and adhesion. The effect of the various treatments on THP1 cell adhesion was determined by MTS assay of both nonadherent cells and adherent cells. The adhesion rate was calculated from the number of adherent cells expressed as a percentage of the sum of nonadherent and adherent cells. Monocyte viability was assessed by trypan blue exclusion. Measurement of monocyte invasion. In vitro cell invasion was measured as the ability of cells to invade through a chemotaxis

F1231

chamber containing a PET filter (pore size 8.0 ␮m) precoated with Matrigel. Briefly, THP1 cells were placed in the upper well of the chamber, and media with or without treatment were added to the lower well in serum-free conditions. Cells were then cultured at 37°C. After a further 48 h, the inserts were removed and fixed with 100% methanol for 10 min, and the cells were stained by incubation in 1% toluidine blue in 1% sodium tetraborate for 2 min. Excess toluidine blue was then removed by washing in water, and Matrigel and noninvading cells were removed from the inside of the chamber using a cotton swab. The filter was then placed on a hemocytometer, and the invading cells on the underside of the filter were viewed microscopically, photographed, and counted. The cell invasion rate was expressed as the percentage of treatment over control. Flow cytometric analysis for CD11b expression. The expression of the macrophage marker CD11b by THP1 monocytes was determined by flow cytometry. Cells (at least 1 ⫻ 106 cells/group) either from suspension culture or after removal from the plate by incubation in EDTA (0.02% for 15–30 min) were collected by centrifugation (300 g for 10 min). The pelleted cells were washed in fluorescenceactivated cell sorting (FACS) buffer [3⫻; PBS containing 0.5% (wt/vol) BSA, 0.1% (wt/vol) NaN3, and 2 mM EDTA] and incubated with mixing in the presence of FcR blocking buffer at room temperature (5 min). Cells were incubated in the dark (1 h at 4°C) with 1 ␮g of anti-human CD11b conjugated to PE, and excess antibody was removed by washing in ice-cold FACS buffer (3⫻). The labeled cells were then resuspended in 400 ␮l of FACS buffer before analysis. All analyses were performed on the BD FACSAria (Becton Dickinson, San Jose, CA). A minimum of 30,000 events was counted for

Fig. 1. Glomerular immunohistochemistry: immunohistochemical staining of kidney sections. Shown are representative images of colocalization of CD68 and matrix metalloproteinase-9 (MMP-9) in control [at ⫻400 magnification (A)] and diabetic glomeruli [at ⫻400 (B) or ⫻1,000 magnification (C)]. Representative images of staining for TNF-␣ (D–F) and IL-6 (G–I) are also shown. Small arrows indicate the brown color of CD68/MMP-9 colocalization. AJP-Renal Physiol • VOL

297 • NOVEMBER 2009 •

www.ajprenal.org

F1232


Table 2. Glomerular immunohistochemical staining of proinflammatory cytokines and MMP-9

Control Diabetic

TNF-␣⫹ve Cells

IL-6⫹ve Cells

CD68⫹ve Cells

CD68/MMP-9⫹ve Cells

12.2⫾10.4 43.5⫾29.4†

6.6⫾8.48 39.0⫾26.4‡

0.6⫾0.4 2.1⫾1.5*

0.0 1.1⫾0.5*

Statistical analysis. Results from at least three independent experiments were pooled and expressed as means ⫾ SD. Results were then compared using either Student’s t-test or analysis of variance (ANOVA) with post hoc analysis using Bonferroni’s multiple comparison test. Significance was accepted at P ⬍ 0.05. RESULTS

Values are means ⫾ SD, expressed as the average number of positive (⫹ve) cells per glomerular cross section. MMP-9, matrix metalloproteinase-9. *P ⬍ 0.01; †P ⬍ 0.05; ‡P ⬍ 0.005, significantly different from control (by t-test).

each cell sample, and all cell populations were gated on high forward scatter for further assurance of monocytic characteristics. All treatments were compared with an isotype control, and the data were analyzed using FlowJo version 8.1.1 (Ashland, OR). Enzyme-linked immunosorbent assays. The concentration of inflammatory factors TNF-␣, IL-6, IL-8, MCP-1, and MIP-1␤ in mesangial cell CM cultured in either NG or HG or in the presence or absence of recombinant human TNF-␣ (5 nM) were quantified using commercially available ELISA kits.

Diabetes increased macrophage numbers and macrophage/ MMP-9 colocalization. The diabetic animals weighed less than and their average fasting blood glucose levels were consistently higher than those of nondiabetic controls. Measurement of urinary ACR at termination was fourfold increased in the diabetic animals compared with controls (Table 1). After 8 wk of diabetes, despite evidence of only mild glomerulosclerosis (5/6 animals with sclerosis score 1⫹), the staining of TNF-␣ and IL-6 was significantly increased in the glomeruli of diabetic animals (Fig. 1 and Table 2). The number of CD68⫹ve cells in the interstitium (control: 2.6 ⫾ 1.4 and diabetic: 7.1 ⫾ 4.5 cells/25 fields) and the glomeruli (Table 2) of diabetic rats

Fig. 2. Effect of coculture of mesangial cells and THP1 cells on MMP and tissue inhibitors of metalloproteinase (TIMP) expression and matrix degradation. Human THP1 cells and human mesangial cells (Mes) were cultured together (Co) or alone in normal (NG) or high glucose concentrations (HG). After 5 days, the culture media were collected and analyzed for MMP-2 and MMP-9 activities by gelatin zymography and for TIMPs by reverse gelatin zymography. A representative zymography gel is shown in A. The levels of proMMP-9 (B), proMMP-2 (C), TIMP-1 (D), and TIMP-2 (E) were quantified from the band area, expressed as area under the curve (AUC). Matrix degrading activity in conditioned medium (CM) was determined by degradation of a 2-methoxy-2,4-diphenyl3(2H)-furanone (MDPF)-labeled gelatin substrate and expressed as a percentage of the trypsin-degraded fraction (F). Results are from 3 individual experiments and are means ⫾ SD. MW, protein molecular marker. *P ⬍ 0.05, significantly different from either cell type alone. AJP-Renal Physiol • VOL

297 • NOVEMBER 2009 •

www.ajprenal.org


F1233

Fig. 3. Effect of cell CM on the expression of MMPs. THP1 or Mes were cultured either individually or in the presence of CM (20%) obtained from the other cell type and grown in serum-free media containing either NG or HG. MMP-9 and MMP-2 levels were analyzed by zymography gels. MCM, Mes cell conditioned medium; TCM, THP1 cell conditioned medium.

was also increased. We next examined by dual-color staining whether the MMP-9 in the glomeruli colocalized with macrophages (Fig. 1, A–C). Adjacent kidney sections from a diabetic rat stained for CD68 with DAB secondary antibody and for MMP-9 with permanent red for secondary antibody are shown in Supplemental Fig. 1. (Supplemental data for this article are available online at the American Journal of Physiology-Renal Physiology website.) The number of CD68⫹ve cells in the interstitium and glomeruli of the diabetic animals that also expressed MMP-9 was increased (30 and 50%, respectively). By contrast, in the control animals, none of the CD68⫹ve cells in the glomeruli expressed MMP-9. A representative image for each stain is shown in Fig. 1, and the data are presented quantitatively in Table 2. Coculture of THP1 and mesangial cells increase MMP-9 and degradative activity. Zymographic analysis of media from THP1 and mesangial cells showed that THP1 cells secreted mainly proMMP-9 and mesangial cells mainly proMMP-2,

which is another gelatinase (Fig. 2A). We next investigated the effect of coculture of THP1 and mesangial cells. In either NG (5 mM) or HG (25 mM), coculture of these two cell types significantly increased the media concentration of proMMP-9 (65 and 112%, respectively; Fig. 2B) but had no effect on MMP-2, TIMP-1, and TIMP-2 concentration (Fig. 2, C, D, and E, respectively). Coculture of THP1 and mesangial cells in NG or HG also significantly increased media degradative activity to a level greater than either cell type alone (Fig. 2F). Mesangial cell CM increases THP1 secretion of MMP-9, cell adhesion, differentiation, degradative activity, and cell invasion. To determine whether the increased MMP-9 expression and activity required cell-cell contact, we examined the effect of addition of CM from one cell type to the other cells on expression of MMP-9. As shown in Fig. 3, addition of CM from mesangial cells to THP1 cells increased proMMP-9 expression and was significantly greater than the sum of the individual cell types. The magnitude of the increase was the

Fig. 4. Effect of MCM on THP1 MMP expression. A: THP1 cells were cultured for 48 h in MCM obtained from mesangial cells grown for either 24 or 48 h in serum-free medium containing either NG or HG. The expression of proMMP-9 was quantified from the band area and expressed as AUC. B: THP1 cells were cultured with MCM under NG or HG conditions for 6, 17, 24, and 48 h, and the expression of proMMP-9 was analyzed from the gel image and quantified. C and D: the level of MMP-9 mRNA was detected from THP1 control cells (C) and THP1 cells cultured in MCM (D) under NG, HG, and mannitol conditions for 48 h. MMP-9 mRNA level was expressed as fold change relative to THP1 control under NG after adjustment for the housekeeping gene ␤-actin. Results are from 3 independent experiments and are means ⫾ SD. *P ⬍ 0.05, significantly different from THP1 cells alone and MCM. †P ⬍ 0.05, significantly different from NG and mannitol.

AJP-Renal Physiol • VOL

297 • NOVEMBER 2009 •

www.ajprenal.org

F1234


same when the CM was collected from mesangial cells grown in NG or HG medium. In contrast, addition of THP1 CM to mesangial cells had no effect on MMP concentration (Fig. 3). Furthermore, mesangial cell CM increased THP1 proMMP-9 expression three- to fivefold (P ⬍ 0.05) in a dose (Fig. 4A)- and time (Fig. 4B)-dependent manner. As shown in Fig. 4C, culture of THP1 cells in medium containing HG caused a nonsignificant increase in the gene expression of MMP-9. However, when THP1 cells were cultured in the presence of CM obtained from mesangial cells grown in NG or HG, the expression of MMP-9 mRNA was increased (245- and 428-fold, respectively). These results are shown graphically in Fig. 4D. Also, as shown in Fig. 4, C and D, mannitol as an osmotic control had no effect on MMP-9 expression. As with the coculture experiments, addition of mesangial cell CM had no effect on the expression of MMP-2 or TIMPs by THP1 cells (data not shown). We next examined the effect of mesangial cell CM on THP1 cell adhesion and invasion. The results of these studies are

Table 3. Effect of mesangial cell conditioned medium on THP1 cell adhesion, invasion, and degradative activity

THP1 THP1 ⫹ MCM MCM

Cell Adhesion Rate, %

Cell Invasion Rate, %

Degradative Activity, %

0 51.5⫾18.7* N/A

100 121.8⫾0.8* N/A

29.4⫾6.5 39.3⫾7.3 31.3⫾3.8

Values are means ⫾ SD; cell adhesion rate and cell invasion rate are expressed as percent changes from THP1 control; degradative activity in mesangial cell conditioned medium (MCM) is expressed as a percentage of the trypsin degraded fraction. *P ⬍ 0.05, significantly different from THP1 cells alone. N/A, not applicable.

shown in Table 3. Culture of THP1 cells with mesangial cell CM significantly increased THP1 cell adhesion and invasion (P ⬍ 0.05), and there was a trend toward increased matrix degradative activity. Incubation of THP1 cells with mesangial cell CM changed THP1 rounded morphology to a morphologically heteroge-

Fig. 5. Effect of MCM on THP1 cell differentiation. A and B: photomicrographs of THP1 cells cultured in the absence (A) or presence (B) of MCM (20%) obtained from cells grown in NG for 24 h (magnification, ⫻100). Cells displaying morphological change are indicated by the arrows. C–H: representative flow cytometry analysis using anti-CD11b antibody for THP1 cells grown in the absence (C, E, and G) or presence of MCM (D, F, and H). Representative dot plots of forward scatter (FSC; indicating size) against side scatter (SSC; indicating granularity) are shown in C and D. Cell debris was excluded by electronic means (dots outside main population). Also shown are dot plots representative of target antigen (CD11b) on THP1 surface (E and F) and isotype controls (G and H). Data were obtained by gating fluorescent cells, and results are expressed as single-color dot plots with the SSC on the y-axis and the log10 fluorescence intensity on the x-axis. Phycoerythrin-positive (PE⫹VE) and -negative (PE⫺VE) cells are shown. I: graphical representation of results of 3 individual experiments showing the percentage of PE⫹VE cells for CD11b surface antigens in THP1 cultured with or without MCM. *P ⬍ 0.05, significantly different from THP1 cells alone in the absence of MCM. AJP-Renal Physiol • VOL

297 • NOVEMBER 2009 •

www.ajprenal.org


F1235

Fig. 6. Effect of HG and TNF-␣ on mesangial cell proinflammatory cytokines. Mesangial cells were cultured in the presence of either NG or HG for 48 h (A) or in NG with 0 or 5 nM TNF-␣ for 24 h (B). The concentrations of proinflammatory cytokines IL-6, monocyte chemoattractant protein-1 (MCP-1), TNF-␣, IL-8, and macrophage inflammatory protein-1␤ (MIP-1␤) in CM were determined by ELISA and are shown as pg/ml. *P ⬍ 0.05, significantly different from NG control.

neous population of more fusiform-shaped macrophage-like cells (Fig. 5, A and B). Furthermore, FACScan demonstrated that mesangial cell CM-treated THP1 cells became larger in size and more granular (Fig. 5, C and D), and the percentage of cells stained with the macrophage marker CD11b was significantly increased (Fig. 5, E–I). Proinflammatory cytokines from mesangial cells increase THP1 secretion of MMP-9 and cell adhesion. Mesangial cell CM contained IL-6, MCP-1, TNF-␣, and IL-8 but no detectable

MIP-1␤. Culture of the mesangial cells in HG or TNF-␣ significantly increased mesangial cell secretion of IL-6, MCP-1, TNF-␣, and IL-8 but had no detectable effect on MIP-1␤ (Fig. 6). Addition of TNF-␣ or IL-6 to THP1 cells significantly increased the THP1 secretion of MMP-9 protein in a dosedependent manner (Fig. 7A). By contrast, addition of MCP-1 or IL-8 had no effect. None of these cytokines affected the protein expression of MMP-2 (data not shown). These cytokines also significantly increased THP1 cell adhesion (Fig. 7B).

Fig. 7. Effect of TNF-␣, IL-6, MCP-1, and IL-8 on THP1 expression of MMPs and cell adhesion. THP1 cells were treated with TNF-␣, IL-6, MCP-1, or IL-8 for 48 h. A: MMP-9 and MMP-2 were measured by zymography, and the expression of MMP-9 was quantitated by the band area. Results are fold changes from THP1 control. B: THP1 cell adhesion rate was detected by MTS assay, and results are percent changes from THP1 control. *P ⬍ 0.05, significantly different from THP1 control. AJP-Renal Physiol • VOL

297 • NOVEMBER 2009 •

www.ajprenal.org

F1236


As shown in Fig. 8, addition of anti-TNF-␣ or anti-IL-6 antibodies significantly attenuated the MMP-9 response of THP1 cells to mesangial cell CM (by 45 ⫾ 4.8 and 55 ⫾ 5.2%, respectively; P ⬍ 0.05). In contrast, the anti-IL-8 and antiMCP-1 antibodies had no effect (Fig. 8A). Parallel to their effect on MMP-9 expression, the anti-TNF-␣ and anti-IL-6 antibodies, but not the anti-MCP-1 antibody, reduced the ability of mesangial cell CM to induce monocyte adhesion (Fig. 8B). DISCUSSION

Macrophage accumulation and activation in the kidney have been shown to correlate with the onset of diabetic nephropathy (3, 6, 8). Despite this known relationship, the mechanism of monocyte activation, recruitment to the kidney, and subsequent differentiation to macrophages is incompletely understood. In our study and similar to others we have shown increased macrophage accumulation in the kidneys of diabetic animals (3, 6). Using expression of MMP-9 as a marker of macrophage activation, we have extended this observation and shown increased activation of macrophages in the interstitium and glomeruli of kidney tissue from diabetic rodents. Using an in vitro coculture system, we have demonstrated that mesangial cells grown in a diabetic milieu can activate monocytes to increase their adhesion and migration, MMP expression, and matrix degradative activity. These changes were shown to be induced by proinflammatory cytokines, in particular, TNF-␣ and IL-6, which can induce both monocyte MMP-9 expression and adhesion, whereas others only affect monocyte adhesion. Furthermore, direct cellular contact is not necessary for these phenomena, since the interactions can be mediated by factors released from mesangial cells. Together, these results suggest that interaction with mesangial cells increases monocyte adhesion, differentiation, MMP-9 expression, and degradative capacity, which may be important in the early phase of diabetic nephropathy. Our in vitro studies also demonstrated that high glucose can have effects on mesangial cell-secreted factors, which can alter monocyte activation. Among the proinflammatory cytokines,

TNF-␣, IL-6, and MCP-1 were shown to mediate some of these effects. This is evidenced by our observations that these cytokines are present in the CM obtained from mesangial cells and that their effects on THP1 cells can be blocked by specific antibodies. The concentrations of these cytokines in the mesangial cell CM are augmented by high glucose concentration, providing a possible mechanism for monocyte recruitment and activation that may be important in the early stages of development of diabetic nephropathy. There are subtle differences in the actions of the various cytokines. For example, MCP-1 affected only monocyte adhesion and differentiation, not MMP-9 expression. By contrast, TNF-␣ and IL-6 exhibited effects on all three parameters. This suggests that increased expression of MCP-1 is a primary event in the initiation of monocyte attraction to the kidney in diabetic nephropathy (5, 9, 11). The increased MMP-9 expression to facilitate subsequent monocyte infiltration appears to be mediated more by the other cytokines. Although our study demonstrated that the presence of factors secreted from mesangial cells has great impact on the monocyte-to-macrophage differentiation cascade, it is only part of a complex interaction that may include direct cell-cell interaction at the tissue level or systemic action of proinflammatory cytokines. Our in vivo results show a diabetes-induced increase in glomerular expression of TNF-␣ and IL-6, findings that support our in vitro observation of increased mesangial cell TNF-␣ and IL-6 production in response to high glucose. It is likely that increased expression of these proinflammatory cytokines can act in both an autocrine and paracrine fashion to increase MMP-9. This would further augment mesangial cell secretion of cytokines and subsequent increased activation of monocytes. It is also well recognized that diabetes is an inflammatory condition, and the circulating concentrations of MCP-1, TNF-␣, and IL-6 are increased in diabetes (6, 17) and are reported to correlate with the degree of proteinuria, glomerular macrophage number, and monocyte expression of CD11b (24). It is therefore possible that these cytokines, whether from mesangial cells or other sources, may exert systemically similar actions on monocytes. Whether inhibition of any one step in this complex interaction could reduce

Fig. 8. Effect of inhibition of cytokines in MCM on THP1 expression of MMPs and cell adhesion. THP1 cells were cultured with 20% MCM preincubated with anti-cytokine antibody (10 ␮g/ml) or an isotype control (10 ␮g/ml) at 37°C for 1 h. After 24 h, the THP1 culture medium was analyzed for MMPs by zymography. A: expression of MMP-9 was quantitated, and results are fold changes from THP1 control. B: THP1 cell adhesion rate was determined using the MTS assay, and the results are percent changes of THP1 treated with MCM. *P ⬍ 0.05, significantly different from THP1 only. †P ⬍ 0.05, significantly different from THP1⫹MCM.


297 • NOVEMBER 2009 •

www.ajprenal.org


monocyte infiltration and differentiation and prevent development of diabetic nephropathy must be investigated in further studies. ACKNOWLEDGMENTS We thank members of the Department of Pathology, University of Sydney, for assistance with FACS analysis. GRANTS This project was supported by a grant from the Diabetes Australia Research Trust and the Endocrinology and Diabetes Research Foundation, University of Sydney. REFERENCES 1. Birkedal-Hansen H, Moore WGI, Bodden MK, Windsor LJ, BirkedalHansen B, DeCarlo A, Engler JA. Matrix metalloproteinases: a review. Crit Rev Oral Biol Med 4: 197–250, 1993. 2. Chana RS, Martin J, Rahman EU, Wheeler DC. Monocyte adhesion to mesangial matrix modulates cytokine and metalloproteinase production. Kidney Int 63: 889 – 898, 2003. 3. Chow F, Ozols E, Nikolic-Paterson DJ, Atkins RC, Tesch GH. Macrophages in mouse type 2 diabetic nephropathy: correlation with diabetic state and progressive renal injury. Kidney Int 65: 116 –128, 2004. 4. Chow FY, Nikolic-Paterson DJ, Atkins RC, Tesch GH. Macrophages in streptozotocin-induced diabetic nephropathy: potential role in renal fibrosis. Nephrol Dial Transplant 19: 2987–2996, 2004. 5. Chow FY, Nikolic-Paterson DJ, Ma FY, Ozols E, Rollins BJ, Tesch GH. Monocyte chemoattractant protein-1-induced tissue inflammation is critical for the development of renal injury but not type 2 diabetes in obese db/db mice. Diabetologia 50: 471– 480, 2007. 6. Chow FY, Nikolic-Paterson DJ, Ozols E, Atkins RC, Rollin BJ, Tesch GH. Monocyte chemoattractant protein-1 promotes the development of diabetic renal injury in streptozotocin-treated mice. Kidney Int 69: 73– 80, 2006. 7. Fogo AB. Mesangial matrix modulation and glomerulosclerosis. Exp Nephrol 7: 147–159, 1999. 8. Galkina E, Ley K. Leukocyte recruitment and vascular injury in diabetic nephropathy. J Am Soc Nephrol 17: 368 –377, 2006. 9. Kanamori H, Matsubara T, Mima A, Sumi E, Nagai K, Takahashi T, Abe H, Iehara N, Fukatsu A, Okamoto H, Kita T, Doi T, Arai H. Inhibition of MCP-1/CCR2 pathway ameliorates the development of diabetic nephropathy. Biochem Biophys Res Commun 360: 772–777, 2007. 10. Kelly DJ, Chanty A, Gow RM, Zhang Y, Gilbert RE. Protein kinase C beta inhibition attenuates osteopontin expression, macrophage recruitment, and tubulointerstitial injury in advanced experimental diabetic nephropathy. J Am Soc Nephrol 16: 1654 –1660, 2005. 11. Kiyici S, Erturk E, Budak F, Ersoy C, Tuncel E, Duran C, Oral B, Sigirci D, Imamoglu S. Serum monocyte chemoattractant protein-1 and monocyte adhesion molecules in type 1 diabetic patients with nephropathy. Arch Med Res 37: 998 –1003, 2006.


F1237

12. Loftus IM, Naylor AR, Goodall S, Crowther M, Jones L, Bell PR, Thompson MM. Increased matrix metalloproteinase-9 activity in unstable carotid plaques. A potential role in acute plaque disruption. Stroke 31: 40 – 47, 2000. 13. McLennan S, Yue D, Turtle J. Effect of glucose on matrix metalloproteinase activity in mesangial cells. Nephron 79: 293–298, 1998. 14. McLennan SV, Death AK, Fisher EJ, Williams PF, Yue DK, Turtle JR. The role of the mesangial cell and its matrix in the pathogenesis of diabetic nephropathy. Cell Mol Biol 45: 123–135, 1999. 15. Min D, Lyons JG, Jia J, Lo L, McLennan SV. 2-Methoxy-2,4-diphenyl3(2H)-furanone-labeled gelatin zymography and reverse zymography: a rapid real-time method for quantification of matrix metalloproteinases-2 and -9 and tissue inhibitors of metalloproteinases. Electrophoresis 27: 357–364, 2006. 16. Min D, Moore AG, Bain MA, Breit SN, Lyons JG. Activation of macrophage promatrix metalloproteinase-9 by lipopolysaccharide-associated proteinases. J Immunol 168: 2449 –2455, 2002. 17. Mora C, Navarro JF. The role of inflammation as a pathogenic factor in the development of renal disease in diabetes. Curr Diab Rep 5: 399 – 401, 2005. 18. Nguyen D, Ping F, Mu W, Hill P, Atkins RC, Chadban SJ. Macrophage accumulation in human progressive diabetic nephropathy. Nephrology (Carlton) 11: 226 –231, 2006. 19. Parks WC, Mecham RP. Matrix Metalloproteinases. San Diego: Academic, 1998. 20. Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29: e45, 2001. 21. Ritz E, Rychlik I, Locatelli F, Halimi S. End-stage renal failure in type 2 diabetes: a medical catastrophe of worldwide dimensions. Am J Kidney Dis 34: 795– 808, 1999. 22. Shapiro SD, Senior Matrix metalloproteinases RM. Matrix degradation and more. Am J Respir Cell Mol Biol 20: 1100 –1102, 1999. 23. Stout LC, Kumar S, Whorton EB. Focal mesangiolysis and the pathogenesis of the Kimmelstiel-Wilson nodule. Hum Pathol 24: 77– 89, 1993. 24. Takahashi K, Mizuarai S, Araki H, Mashiko S, Ishihara A, Kanatani A, Itadani H, Kotani H. Adiposity elevates plasma MCP-1 levels leading to the increased CD11b-positive monocytes in mice. J Biol Chem 278: 46654 – 46660, 2003. 25. van der Loos CM, Gobel H. The animal research kit (ARK) can be used in a multistep double staining method for human tissue specimens. J Histochem Cytochem 48: 1431–1438, 2000. 26. Vos C, Gartner S, Ransohoff R, McArthur J, Wahl L, Sjulson L, Hunter E, Conant K. Matrix metalloprotease-9 release from monocytes increases as a function of differentiation: implications for neuroinflammation and neurodegeneration. J Neuroimmunol 109: 221–227, 2000. 27. Wall SJ, Sampson MJ, Levell N, Murphy G. Elevated matrix metalloproteinase-2 and -3 production from human diabetic dermal fibroblasts. Br J Dermatol 149: 13–16, 2003. 28. Woessner JF, Nagase H. Matrix Metalloproteinases and TIMPs. Oxford: Oxford University Press, 2000, p. xiii. 29. Woessner JJ. The family of matrix metalloproteinases. Ann NY Acad Sci 732: 11–21, 1994.

297 • NOVEMBER 2009 •

www.ajprenal.org