many patients with vasculitis associated with anti-neutrophil cytoplasmic autoantibodies (ANCA) develop pauci-immune necrotizing crescentic glomerulonephritis, which is frequently associated with rapidly progressive deterioration of renal function (5). Although the pathogenesis of this glomerulonephritis is not precisely known, leukocyte infiltration, matrix degradation, and necrosis are essential steps in disease development. Following institution of immunosuppressive treatment, some glomeruli with active crescentic lesions resolve with improvement of glomerular function, whereas others become acellular and fibrotic leading to permanent loss of glomerular function (8).

The role of matrix metalloproteinases (MMPs) in glomerular disease is increasingly appreciated. MMPs mediate both degradation of extracellular matrix components and cell proliferation and facilitate leukocyte function (9, 15). The balance between MMPs and their specific inhibitors, tissue inhibitors of MMPs (TIMPs), might direct the long-term disease course. This balance could lead to either glomerular recovery or, alternatively, to development of fibrosis.

Recently, we have found that renal immunohistochemical expression of MMP-2, -3, and -9 and TIMP-1 correlates with histological renal disease activity in ANCA-associated vasculitis (12). To assess whether urinary and plasma levels of MMP-2, -9, and TIMP-1 reflect local renal MMP production and disease activity in ANCA-associated vasculitis, we measured urinary MMP activity and urinary and plasma levels of MMP-2, MMP-9, and TIMP-1 and compared these to renal disease activity, damage, and immunohistochemical expression of MMP-2 and MMP-9.

METHODS

Patients.

Patients with ANCA-associated vasculitis (n = 29) who underwent a renal biopsy at the University Medical Center Groningen between 1995 and 2004 were included in the study.

One patient underwent two biopsies, and another patient underwent three renal biopsies. In total, 32 renal biopsies were evaluated. Ten biopsies were performed in patients newly presenting with ANCA-associated vasculitis, 13 biopsies were performed at the time of relapse, and 9 biopsies were performed some weeks after immunosuppressive therapy had been started to evaluate remaining renal inflammation as recovery of renal function was considered unsatisfactory.

Age, ANCA status, serum creatinine, proteinuria, and C-reactive protein (CRP) were recorded at biopsy, and disease activity was assessed by the Birmingham Vasculitis Activity Score (BVAS). For the patients who underwent a biopsy after starting immunosuppressive therapy, BVAS was recorded at the start of the disease episode. Additionally, serum creatinine was recorded 1 yr after diagnosis. Thirty-one healthy volunteers (13 women, 18 men; median age 51 yr, range: 44–67 yr) provided plasma and urine specimens for comparison. The study was approved by the local medical ethical committee.

Histopathology.

Sections of formalin-fixed, paraffin-embedded renal biopsy tissue (5 μm) were stained with hematoxylin and eosin, periodic acid-Schiff, and methenamine silver. Biopsies were evaluated by light microscopy blinded to clinical status and renal function. Normal, crescentic, sclerosed glomeruli, and glomeruli with fibrinoid necrosis were counted in all glomerular cross sections (GCS) per biopsy (median 12.5 GCS; range 6–50). Results were expressed as percentage of the total number of GCS. Interstitial fibrosis, tubulointerstitial inflammation, and atrophy were scored semiquantitatively from 0 to 4 (0 = no interstitial fibrosis/inflammation/atrophy; 1 = <10%; 2 = 10–25%; 3 = 25–50%; 4 > 50%). The interstitial scores were calculated as the mean score of at least 10 adjacent cortical fields individually scored at ×400 magnification.

Urine and serum samples.

Before the biopsy, freshly voided urine and EDTA blood samples were collected from each patient. Urine samples were spun at 3,000 rpm for 10 min to remove debris; supernatants were stored at −80°C until use. In aliquots of urine specimens, creatinine and protein were measured. EDTA blood samples were spun at 3,000 rpm for 10 min, and supernatants were stored at −80°C until use. All specimens underwent a minimum of freeze-thaw cycles.

MMP activity assays.

Endogenous urinary activity of MMP-2 and MMP-9 was determined by specific activity assays (Amersham Biosciences, Piscataway, NJ) (7). In brief, MMP-2 and MMP-9 were captured by MMP-specific antibodies. Endogenous MMP activity was measured by adding MMP substrate (modified prourokinase) and a chromogenic urokinase peptide substrate, S-2444, and color development was recorded. Results were expressed in units per micromole creatinine (a unit is defined as 1,000·ΔA405/t2).

ELISA.

Urinary and plasma levels of MMP-2, MMP-9, and TIMP-1 were determined by ELISA. In brief, for MMP-9 and TIMP-1 ELISA, 96-well plates (M129A, Greiner) were precoated with F(ab)₂ fragments of goat anti-mouse IgG-Fc (Jackson, West Grove, PA) in 0.1 M carbonated buffer (pH = 9.6) for at least 48 h. After a washing, plates were coated with 100 ng/ml mouse anti-human MMP-9 (clone 36020.111, R&D Systems, Oxon, UK) or 500 ng/ml mouse anti-human TIMP-1 (clone 63515.111, R&D Systems), respectively. Human urine samples and recombinant MMP-9 and TIMP-1 were diluted in 1% BSA, 0.05% Tween 20 in 50 mM Tris·HCl (pH = 8.0), and 300 mM NaCl, and plasma samples were diluted in HPE buffer (Sanquin, Amsterdam, The Netherlands). For MMP-2 ELISA, 96-well plates (MaxiSorp, Nunc) were coated overnight with 2 μg/ml mouse anti-human MMP-2 (clone 36006.211, R&D Systems), and, after being washed, blocked with 1% BSA in PBS for 60 min. Human urine samples, plasma samples, and recombinant MMP-2 were diluted in 0.1% BSA, 0.05% Tween 20 in PBS. For MMP-2, MMP-9, and TIMP-1, an ELISA of both urine and plasma samples was performed in duplicate, and the samples were incubated at room temperature for 1 h. After washing, bound MMP-2, MMP-9, or TIMP-1 was detected by a 1-h incubation with, respectively, biotinylated rabbit anti-human MMP-2, rabbit anti-human MMP-9, or rabbit anti-human TIMP-1 (all R&D Systems). After being washed, samples were incubated with 0.125 μg/ml peroxidase-conjugated streptavidin (Sanquin) for 30 min, and the color reaction was performed with tetramethylbenzidine (Roth, Karlsruhe, Germany). The colorimetric reaction was stopped by the addition of 100 μl/well 0.5 M 2N H₂SO₄, and adsorption at 450/575 nm was measured with a microplate reader.

Sensitivity of the ELISA was 3.9 ng/ml for MMP-2, 0.039 ng/ml for MMP-9, and 0.39 ng/ml for TIMP-1. Urine and plasma concentrations were determined by interpolation from a standard curve. Urinary concentrations were adjusted for the creatinine concentration and expressed in picograms per millimole creatinine.

Immunohistochemistry.

Paraffin sections of 26 renal biopsies were available for immunohistochemistry. For the remaining six biopsies, insufficient material was available for immunohistochemistry. Samples of normal kidney tissue from patients with renal cell carcinoma (n = 8) were used as control specimens. For immunohistochemistry, paraffin-embedded sections (5-μm thickness) were deparaffinized and then microwaved for 10 min in 10 mM citrate, pH 6.0. Sections were then (and after all subsequent steps) washed in Tris-buffered saline (TBS). After blocking of endogenous peroxidase, sections were incubated for 1 h with the following primary antibodies diluted in TBS: monoclonal mouse anti-human MMP-2 (Calbiochem, La Jolla, CA), polyclonal goat anti-human MMP-9 (R&D Systems), and monoclonal mouse anti-human CD68 (DAKO, Glostrup, Denmark), respectively. Peroxidase-conjugated secondary and, subsequently, tertiary antibodies were added, and peroxidase activity was developed in a freshly prepared solution of 3-amino-9-ethylcarbazol containing 0.03% hydrogen peroxide for 10 min. Sections were counterstained using hematoxylin. Control slides, which were incubated in PBS in the absence of the primary antibody, were consistently negative. Histology slides were analyzed and scored by two observers. For each biopsy, positively staining cells in glomeruli were manually counted and expressed as cells per glomerular cross section (GCS). Interstitial staining was scored semiquantitatively from 0 to 4 (0 = no staining; 1 = occasional positive cells in <10% of interstitium; 2 = positive cells in 10–25% of interstitium; 3 = positive cells in 25–50% of interstitium; 4 = positive cells in >50% of interstitium).

Statistics.

The nonparametric Mann-Whitney test was used to compare data between various patient groups. Correlations were tested using Spearman's correlation coefficient. Statistical analysis was performed with GraphPad Prism Software, version 4.0 for Windows (GraphPad Software, San Diego, CA). Statistical significance was defined as a two-sided P < 0.05.

RESULTS

Clinical and histological characteristics.

Clinical characteristics of 29 patients at renal biopsy are shown in Table 1; the median age in the patient group was 67 yr (range 23–86 yr), median BVAS was 19 (range 0–36), median serum creatinine was 233 μmol/l (range 99–1033), and median proteinuria was 1.5 g/l (range 0.1–14.1). In total, 18 patients were PR3-ANCA positive (including 1 patient who underwent 2 biopsies), and 11 were MPO-ANCA positive (including 1 patient who underwent 3 biopsies). Ten renal biopsies were performed at diagnosis. Thirteen renal biopsies were taken because renal relapse was suspected. In the renal biopsy of one of the latter patients (patient 19), no active lesions were found. Nine renal biopsies were performed in patients with an unsatisfactory response to treatment to evaluate remaining renal disease activity several weeks (median 5, range 3–8) after they started immunosuppressive treatment. In six of these renal biopsies, persistent activity with cellular crescents was found. Table 2 shows histological damage in biopsies of all patients and PR3- and MPO-ANCA-positive patients, respectively. Histological parameters did not differ between biopsies of PR3- and MPO-ANCA-positive patients.

Urinary MMP and TIMP-1 levels.

In urine samples of healthy controls, both endogenous MMP-2 (median 0.002 U/μmol creatinine; range 0.00–0.032) and MMP-9 activity (median 0.010 U/μmol creatinine; range 0.00–0.964) could be detected (Fig. 1A). However, urinary MMP-2 activity was significantly higher in patients with ANCA-associated vasculitis (median 0.005 U/μmol creatinine; range 0.00–0.052) (P = 0.02) (Fig. 1A). By ELISA, urinary MMP-2 was detected in none of the controls and in only two urine samples from patients (6.6 and 7.5 ng/mmol creatinine, respectively). Urinary MMP-9 activity was significantly elevated in patients (median 0.088 U/μmol creatinine; range 0.002–1.791) compared with controls (P = 0.0002) (Fig. 1A). Urinary MMP-9 levels, as measured by ELISA, were not significantly higher in patients (median 0.07 pg/mmol creatinine; range 0.00–6.77) than in controls (median 0.00 pg/mmol creatinine; range 0.00–3.77) (P = 0.15) (Fig. 1B). Urinary MMP-9 activity (activity assay) and MMP-9 levels (ELISA) correlated significantly (r = 0.73, P < 0.001). TIMP-1 levels in urine samples of patients (median 2.86 pg/mmol creatinine; range 0.00–23.50) were significantly higher than in urine samples of healthy controls (median 1.10 pg/mmol creatinine; range 0.30–5.20) (P = 0.001) (Fig. 1B).

Plasma MMP and TIMP-1 levels.

Plasma MMP-2 levels (median 85 ng/ml; range 0–1340) were not significantly elevated in patients with ANCA-associated vasculitis compared with healthy controls (median 93 ng/ml; range 0–723) (P = 0.31) (Fig. 2A). Also, plasma MMP-9 levels in patients (median 53.1 ng/ml; range 4.8–410.0) were not higher than in healthy controls (median 65.2 ng/ml; range 30.1–376.3) (Fig. 2B). Plasma TIMP-1 levels were significantly higher in patients with ANCA-associated vasculitis (median 649 ng/ml, range 86–1253) than in healthy controls (median 380 ng/ml; range 50–620) (P < 0.0001) (Fig. 2C). Plasma levels of MMP-2, -9, and TIMP-1 did not correlate significantly with urinary MMP activity or levels and with TIMP-1 levels, respectively. Plasma levels of MMP-2, -9, and TIMP-1 did not correlate with CRP or BVAS.

MMP-2 and MMP-9 expression in renal biopsies.

MMP-2-expressing cells were present in many glomeruli, particularly in active lesions, and to a lesser extent in normal appearing glomeruli (median 15.6 cells/GCS; range 0–52.5) (Figs. 3A and 4). In the interstitium MMP-2-expressing cells were present as well (median score 2; range 0–4); a major part of the tubulointerstitial MMP-2 expression was located in tubuloepithelial cells (Fig. 5). In control sections, significantly fewer MMP-2-expressing cells were present, both per glomerular cross section (2.5 cells/GCS; range 1.3–16.7) (P = 0.003) (Figs. 3A and 4) and in the interstitium (median score 1; range 0–1) (P = 0.01) (Fig. 5). Compared with MMP-2, fewer cells expressed MMP-9 in glomeruli from patients with ANCA-associated vasculitis [median 0.8 cells/GCS; range (0.0–12.2)] (Figs. 3 and 4). In sections from controls, significantly more glomerular cells expressed MMP-9 (median 2.3 cells/GCS; range 0.8–3.8) than in patients with crescentic glomerulonephritis (P = 0.04) (Fig. 4). However, in some severely inflamed glomeruli many cells expressed MMP-9 (Fig. 3D). Both in controls and patients, hardly any interstitial expression of MMP-9 was present (Fig. 5). In glomeruli from patients, significantly more CD68-positive cells were present than in control renal biopsy sections (P = 0.0017) (Fig. 4). Similarly, CD68 was expressed by significantly more cells in the interstitium in patients than in controls (P = 0.0016).

Glomerular expression of both MMP-2 (r = 0.044, P = 0.83) and MMP-9 (r = 0.33, P = 0.11) did not correlate significantly with the number of CD68-positive cells per glomerulus. Neither correlated glomerular MMP-2 and MMP-9 expression significantly (r = 0.32, P = 0.11). Interstitial expression of MMP-2 correlated significantly with interstitial expression of CD68 (r = 0.59, P = 0.002). Neither glomerular nor interstitial MMP-2 and MMP-9 expression correlated with urinary and plasma levels of these MMPs. Also, glomerular and interstitial expression of CD68 did not correlate with urinary and plasma levels of MMPs and TIMP-1.

Correlation of urinary MMP and TIMP-1 levels with histological and clinical characteristics.

Urinary MMP-2 activity correlated negatively with the percentage of glomeruli with fibrinoid necrosis (r = −0.39, P = 0.029) and cellular crescents (r = −0.42, P = 0.017), and tended to correlate positively with the percentage of fibrous crescents (r = 0.35, P = 0.051) (Table 3). Urinary MMP-9 activity also correlated negatively with the percentage of glomeruli with fibrinoid necrosis (r = −0.46, P = 0.009) and cellular crescents (r = −0.41, P = 0.02) and positively with the percentage of fibrous crescents (r = 0.42, P = 0.017) (Table 3). Urinary levels of TIMP-1 did not correlate with glomerular damage (Table 3). However, urinary TIMP-1 correlated significantly with tubulointerstitial atrophy (r = 0.41, P = 0.022) and tubulointerstitial fibrosis (r = 0.35, P = 0.050) (Table 3). Urinary MMP-2 activity also correlated with tubulointerstitial atrophy (r = 0.49, P = 0.005) and tubulointerstitial fibrosis (r = 0.41, P = 0.019) (Table 3). Urinary MMP-9 activity did not correlate with tubulointerstitial damage (Table 3). Plasma levels of MMPs and TIMP-1 did not correlate with glomerular and tubulointerstitial damage.

Urinary MMP-2 activity (activity assay) correlated negatively with creatinine clearance at biopsy (r = −0.41, P = 0.019) and 1 yr after diagnosis (r = −0.40, P = 0.042). Urinary MMP-9 activity (activity assay, r = −0.30, P = 0.098) and urinary MMP-9 levels (ELISA, r = −0.30, P = 0.10) tended to correlate negatively with creatinine clearance at diagnosis. At 1 yr after diagnosis, no correlation was found between creatinine clearance and urinary MMP-9 activity (r = −0.21, P = 0.30). Urine levels of TIMP-1 correlated negatively with creatinine clearance at biopsy (r = −0.49, P = 0.006), and urinary TIMP-1 levels tended to correlate with creatinine clearance 1 yr after diagnosis (r = −0.35, P = 0.08). Plasma levels of MMP-2, -9, and TIMP-1 (ELISA) did not correlate with creatinine clearance at biopsy or 1 yr after diagnosis.

Correlation of renal MMP-2 and MMP-9 expression with histological and clinical characteristics.

Glomerular MMP-2 expression correlated positively with the percentage of glomeruli with fibrinoid necrosis (r = 0.55, P = 0.004) and negatively with the percentage of glomeruli with fibrous crescents (r = −0.42, P = 0.035) (Table 3). Glomerular MMP-9 expression correlated significantly with the percentage of glomeruli with fibrinoid necrosis (r = 0.68, P < 0.001) and with cellular crescents (r = 0.66, P < 0.001) (Table 3). The glomerular number of CD68-positive cells correlated negatively with the percentage of normal glomeruli (r = −0.42, P = 0.032) but not with glomerular damage (Table 3).

Interstitial expression of MMP-2 and CD68 correlated significantly with tubulointerstitial inflammation (r = 0.51, P = 0.009; r = 0.59, P = 0.002) but not with tubulointerstitial atrophy and fibrosis.

Glomerular MMP-2 and MMP-9 expression did not correlate with creatinine clearance at biopsy. In contrast, glomerular expression of CD68 correlated significantly with creatinine clearance at biopsy (r = −0.41, P = 0.036). Interstitial expression of MMP-2 did not correlate significantly with creatinine clearance at biopsy and 1 yr after diagnosis. Interstitial expression of CD68 correlated negatively with creatinine clearance at biopsy (r = −0.43, P = 0.028).

DISCUSSION

Renal lesions in ANCA-associated vasculitis are characterized by a wide variation in activity and chronicity. As histological characteristics are significantly related to clinical outcome, clinical assessment in many cases warrants a renal biopsy (8). However, a renal biopsy is not without risk, and repeat biopsies to evaluate treatment response are usually not performed. Therefore, other means of monitoring renal disease activity would facilitate and improve patient care. Urinary measurement of MMPs offers a potential alternative, as the presence and function of MMPs have been associated with disease activity and severity in both human and animal models of glomerulonephritis. Recently, we found that renal expression of MMP-2, MMP-9, and TIMP-1 correlates with renal histological disease activity in ANCA-associated vasculitis, a finding that was confirmed in the current study (12). In the current study, we measured urinary levels and activity of MMP-2, MMP-9, and TIMP-1 and correlated these with renal protein expression, renal damage, plasma levels of MMPs, and clinical disease activity.

Both urinary MMP-2 and MMP-9 protein levels and/or activity was significantly elevated in patients with active crescentic glomerulonephritis. In contrast, plasma MMP levels were not elevated in patients compared with healthy controls and did not correlate with urinary MMPs or with measures of disease activity. Thus urinary MMP excretion is not a reflection of increased systemic inflammation and MMP expression. Previously, in systemic lupus erythematosus, giant cell arteritis, and Takayasu arteritis, elevated protein levels and activity of MMP-9 were found, whereas MMP-2 was elevated in Takayasu arteritis (4, 11, 13). In Wegener's granulomatosis, Bjerkeli (1) found increased plasma MMP levels, including MMP-2. Additionally, Bjerkeli found TIMP-1 levels were increased and correlated with CRP. In line with these latter findings, we found both elevated urinary and plasma levels of TIMP-1 in patients with ANCA-associated vasculitis. However, we did not find a correlation between plasma TIMP-1 levels and disease activity as measured by BVAS and CRP. Thus plasma levels of MMP-2, MMP-9, and TIMP-1 do not reflect disease activity in ANCA-associated vasculitis.

Additionally, we studied whether urinary MMP-9 levels reflected histological MMP-9 expression and consequently glomerular inflammation. We observed that in patients with active crescentic glomerulonephritis, a smaller number of glomerular cells expressed MMP-9 than in normal-appearing control renal sections. Although glomeruli were extensively damaged in most patients, only a few glomerular lesions contained an accumulation of MMP-9-expressing cells. Nevertheless, glomerular MMP-9 expression correlated strongly with active glomerular inflammation. In contrast to local renal expression, urine MMP-9 levels and activity were higher in patients than in healthy controls. Apparently, MMP-9 present in glomeruli from control sections is not excreted. MMP-9 excretion in patients might be the consequence of inflammation and damage. However, neither urine MMP-9 levels nor activity reflected renal damage. Thus from our study we cannot advocate the clinical use of MMP-9 in urine as a disease marker in ANCA-associated renal vasculitis.

In contrast to MMP-9 expression, significantly more glomerular cells expressed MMP-2 in patients with crescentic glomerulonephritis than in control sections. In our previous study, in a smaller number of patients, we also found only a small number of MMP-9- but more MMP-2-positive cells in glomeruli of patients with crescentic glomerulonephritis (12). In contrast to the previous study, we now found more MMP-9 in control sections. This might be due to technical differences as we used formalin-fixed, paraffin-embedded sections instead of frozen biopsy sections. Additionally, in the current study we used a different polyclonal MMP-9 antibody compared with the previous study. In concordance with our current findings Lods et al. (10) previously found lower glomerular MMP-9 expression and increased MMP-2 expression in kidney biopsy specimens of patients with glomerulonephritis compared with normal-appearing glomeruli from tumor nephrectomy specimens. However, in both the study of Lods et al. and our study renal malignancy and related immunological responses might explain the increased glomerular MMP-9 expression in control sections. Animal and in vitro data suggest that MMP-2 is not only proteolytic but also involved in remodeling during fibrosis. MMP-2 was able to induce transformation of normal rat tubular epithelium into a myofibroblastic phenotype (3). Our data could to some extent be interpreted in support of epithelial-mesenchymal transition. We detected strong MMP-2 expression in tubuloepithelial cells of severely affected kidneys. Additionally, urinary MMP-2 activity correlated with tubulointerstitial damage and creatinine clearance at biopsy and 1 yr after diagnosis. However, urinary MMP-2 activity did not reflect renal MMP-2 expression. Also, regarding immunohistochemical findings, no correlation was found between tubulointerstitial MMP-2 expression and fibrosis.

From our study, we cannot conclude from what cellular sources the urinary MMPs were derived. However, the absence of a correlation with glomerular expression and the negative correlation with glomerular inflammatory activity, when immunohistochemical expression is most pronounced, preclude a glomerular origin. Meanwhile, urinary MMP activity and TIMP-1 reflected tubulointerstitial damage. Thus, although MMP-9 expression was only present in small amounts in the interstitium, urine MMPs might originate from the major renal compartment, the tubulointerstitium. For MMP-2, this is supported by the strong expression we detected in tubuloepithelial cells.

In our study, we measured both MMP activity and protein levels by activity assays and ELISA. Activity assays have been suggested to be superior, as results reflect proteolytic amounts. We did not find major differences between the two techniques, and MMP activity and protein levels correlated significantly Because sensitivity of the MMP-2 ELISA was insufficient regarding MMP-2, we cannot compare activity assay and ELISA results.

In ANCA-associated vasculitis, several serum markers are clinically used as indicators of disease activity, in particular, ANCA and CRP. However, these are not specific regarding renal disease activity. Urinary levels of inflammatory mediators could be more sensitive and specific markers of renal disease. We are the first to report on urinary MMPs as a marker of glomerular and tubulointerstitial disease in ANCA-associated glomerulonephritis. We did not find a relationship among urinary levels or activity of MMPs, glomerular damage, and renal expression. No difference in renal expression of MMPs and levels in urine and plasma was found in patients who used immunosuppressive agents compared with patients who did not. We found elevated urinary levels of MMP-2, MMP-9, and TIMP-1. Moreover, urinary MMP-2 correlated negatively with creatinine clearance at biopsy and 1 yr after diagnosis, whereas urinary TIMP-1 levels correlated negatively with creatinine clearance at biopsy. Other studies have found urinary markers reflecting renal disease severity in ANCA-associated glomerulonephritis, in particular macrophage migration inhibitory factor (MIF) and monocyte chemoattractant protein-1 (MCP-1) (2, 14). These studies stress the role of macrophages in glomerular disease induction and development. Macrophages play different roles in both acute and chronic disease, resulting in different phenotypes. MMP expression is a defining feature of these different macrophage phenotypes. In vitro, the IFN-γ and TNF-α induced so-called classic activation of macrophages results in a proinflammatory state (6). These macrophages predominantly express MMP-9 (16), corresponding to the high MMP-9 expression that we found in what might be acute stages of glomerular injury, whereas high expression of MMP-2 as detected in chronic lesions corresponds to a deactivated, dampened macrophage state induced by TGF-β (6, 16). Ultimately, more insight in this heterogeneity could lead to better disease monitoring and eventually manipulation of macrophage function.

In conclusion, we found that urinary MMP-2 and MMP-9 activity and TIMP-1 levels were elevated in ANCA-associated glomerulonephritis. Urinary levels of MMP did not reflect renal MMP expression and inflammation. However, urinary MMP-2 activity and TIMP-1 levels correlated with tubulointerstitial damage and creatinine clearance at biopsy.

GRANTS

J.-S. F. Sanders is an MD-clinical research trainee sponsored by ZonMw. This study was supported by a grant from the Dutch Kidney Foundation (C02.2033).

FOOTNOTES

• The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.