Urinary Excretion of Monocyte Chemoattractant Protein-1 in Autosomal Dominant Polycystic Kidney Disease

ADPKD Patients

Aliquots of discarded urine from 55 ambulatory ADPKD patients (22 men, 33 women; age 22 to 79 yr) in the Polycystic Kidney Clinic at the Kansas University Medical Center during 1998 to 2001 were evaluated in the study. Fresh samples were collected and frozen at −20°C. ADPKD was diagnosed by characteristic ultrasound or computed tomography presentation of the kidneys and confirmed by the presence of liver cysts or positive family history. None of the patients had evidence of clinical renal infection when urine samples were obtained. None had been treated with dialysis or renal transplantation. Serum creatinine levels ranged between 0.5 and 11.9 mg/dl (median, 1.2). In 17 patients, additional urine samples (two to seven) were obtained over a maximum interval of 21 mo. The initial urine sample was used to date entry of an individual patient into the study. Aliquots of serum were obtained from 15 patients after informed, written consent. Unfortunately, in this retrospective study, serum samples were not available in all patients from whom urine was obtained.

Samples of cyst fluid were obtained from 73 different individuals with ADPKD (7 to 73 yr old) who were nephrectomized in preparation for renal transplantation. Sealed kidneys were placed in wet ice and transported to the laboratory, usually by overnight, expedited delivery. Samples of clear cyst fluid of widely varying amounts were obtained from one to approximately 50 cysts on the surface of each pair of kidneys. Fluids were generally pooled together, although a few individual cyst fluid samples were also obtained from two patients.

Normal Subjects

Nineteen healthy volunteers (9 men, 10 women; 25 to 65 yr old) had serum creatinine values ranging from 0.6 to 1.1 mg/dl. Early morning midstream urine samples were obtained, in women in the middle of the menstrual cycle. The protocol was approved by the Kansas University Medical Center Institutional Review Board, and each volunteer gave informed consent.

Enzyme-Linked Immunosorbent Assay

Samples of urine, serum, and cyst fluid were thawed, and aliquots were taken for the determination of MCP-1, MCP-3, IL-8, and TNF-α by enzyme-linked immunosorbent assay (ELISA) using methods recommended by the manufacturer (Biosource International, Amarillo, CA). These assays used a multiple sandwich, solid-phase enzyme immunoassay that included monoclonal antibodies raised against the respective cytokines. Specifically, the MCP-1 kit used a hamster monoclonal MCP-1 capture antibody and a rabbit polyclonal MCP-1 antibody for detection. Standard buffer in the kit was used to dilute samples with high levels of MCP-1. The enzymatic reactions were quantified in an automatic microplate photometer (Dynatech Laboratories, Chantilly, VA). Cytokine concentrations of unknown samples were determined by interpolation using a standard curve based on recombinant MCP-1 supplied by the manufacturer. Recovery of MCP-1 added to urine was 110 ± 7% (mean ± SEM, n = 5). Urine samples were generally diluted 1:2 to 1:20, cyst fluids 1:30 to 1:50, and serum 1:2. The sensitivity of the ELISA for MCP-1 was 20 pg/ml. Samples could be thawed and frozen at least three times without causing a change in MCP-1 concentration (data not shown). Urine cytokine samples (pg/ml) were related to creatinine (mg/ml) concentration and expressed as pg/mg. Urine creatinine was determined with a kit method based on an alkaline picrate colorimetric assay (Sigma Diagnostics, St. Louis, MO) and protein by a kit method (Quantimetrix Corporation, Redondo Beach, CA).

Immunoprecipitation

Samples of urine and cyst fluid were immunoprecipitated to increase sensitivity of Western blots of MCP-1. Protein A Sepharose beads (Amersham Pharmacia Biotech AB, Uppsala, Sweden) were washed with Triton lysis buffer (TLB; 20 mM Tris [pH 7.4], 137 mM NaCl, 25 mM β-glycerophosphate, 2 mM Pyrophosphate, 1 mM phenymethylsulphonyl fluoride, 5 μg/ml leupeptin, 5 μg/ml aprotinin, 2 mM benzamidine, and 0.5 mM dithiothreitol) three times, then incubated with polyclonal rabbit anti-human MCP-1 (Research Diagnostics Inc., Flanders, NJ) for 2 h. The antibody-laden beads were washed with TLB three times, and aliquots were incubated with urine (2.5 ml) or cyst fluid (150 to 500 μl) overnight at 4°C, each aliquot sufficient for a single PAGE lane (vida infra). MCP-1 bound to the beads was washed with TLB three times then released by heating to 95°C for 5 min in TLB and 2× sample buffer (132 mM Tris-HCl [pH 6.8], 4.2% SDS, 21% glycerol, 0.72 mM β-mercaptoethanol, 20 mM dithiothreitol, and 0.005% bromophenol blue).

Western Blotting of MCP-1

Samples of urine or cyst fluid concentrated by immunoprecipitation were resolved by 15% SDS-PAGE. MCP-1 standards and polyclonal rabbit anti–MCP-1 were obtained from Research Diagnostics Inc., and anti-rabbit IgG–horseradish peroxidase was obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The transfer of proteins from gels to nitrocellulose membranes (Midwest Scientific, St. Louis, MO) was carried out at 4°C. Kaleidoscope prestained standards or prestained polypeptide markers (Bio-Rad Laboratories, Inc., Hercules, CA) were transferred from gel to membranes, and their position in relation to the experimental samples was aligned carefully. Blocking was carried out with 5% dry milk in Tris-buffered saline Tween (TBS-T; 20 mmol/L Tris-HCl [pH 8.0], 137 mmol/L NaCl, and 0.05% Tween 20). Blocked membranes were cut into left and right portions through the lane of visible markers, then incubated in separate systems. One portion of the divided membrane was incubated with 1:1000 dilution of rabbit polyclonal anti-human MCP-1 and the other with anti–MCP-1 antibody that had been incubated for 8 h in a 5:1 excess (wt:wt) of immunizing peptide (MCP-1). Both portions were incubated in 5% dry milk in TBS-T at 4°C for 20 h, followed by washing three times with TBS-T and incubation with a 1:2500 dilution of anti-rabbit IgG–horseradish peroxidase (Santa Cruz biotechnology) in 5% dry milk in TBS-T for 2 h. The membranes were washed three times in TB-T and visualized with an enhanced chemiluminescence system (Amersham Life Sciences, Arlington Heights, IL).

Culture of ADPKD Cyst Epithelial Cells and Human Kidney Cortex Cells

Epithelial cells from the renal cysts of ADPKD patients and cells from humans with normal renal cortex were harvested and grown as primary cultures using methods described in detail previously (8,9⇓). Briefly, primary epithelial cell cultures from renal cysts of ADPKD patients or normal renal cortex were used for up to four passages. Cells were maintained in DMEM and Ham’s F12 (DMEM/F12; JRH Biosciences, Lenexa, KS) supplemented with 5% FBS (Hyclone, UT), 5 μg/ml insulin, 5 μg/ml transferrin, and 5 ng/ml sodium selenite (ITS; Collaborative Biomedical Products, Bedford, MA). Monolayers of ADPKD and human kidney cortex (HKC) cells were grown on Snapwell supports (0.4-μm pore size; Corning Costar Corp., Cambridge, MA) until they were confluent. The FBS content was reduced to 1% FBS DME/F12 supplemented medium for 48 h, then the FBS and ITS were removed for 24 h to reduce cytokine stimulation. Fresh DMEM/F12 lacking FBS and ITS was added to both apical (0.4 ml) and basolateral (3 or 5 ml) compartments for 24 h. Samples of apical and basolateral media were taken for the determination of MCP-1 by ELISA. For recovering sufficient MCP-1 for Western blotting, DMEM/F12 supplemented with 1% FBS and ITS was added to the chambers for 72 h after which apical samples were taken.

Statistical Analyses

Data are presented as means ± SEM, and unpaired two-tailed t test was used for comparisons between two groups; P < 0.05 was considered significant.

Determination of MCP-1 in Urine and Cyst Fluid by Western Blot

The amounts of MCP-1 in urine and cyst fluid were too low to detect in unmodified samples of urine or cyst fluid by conventional Western blotting. Consequently, urine and cyst fluid samples were immunoprecipitated using anti–MCP-1–coated Protein-A Sepharose beads. Western blots were performed on three urine samples from normal and ADPKD patients and on three samples of cyst fluid obtained from surgically removed ADPKD specimens. MCP-1 standards run as a control for the Western blots showed a distinct band at approximately 9 kD (Figure 1A). In two urine samples (70 and 79) selected on the basis of representative levels of MCP-1 detected by ELISA (vida infra), two adjacent bands were detected, the lower of which matched the mobility of the monomeric MCP-1 standard. The heavier band is consistent with a posttranslational glycosylated MCP-1 product (10). Both of these bands as well as the standard band were eliminated after preincubation of the antibody with an excess of immunizing MCP-1 peptide (Figure 1A).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Monocyte chemoattractant protein-1 (MCP-1) protein in autosomal dominant polycystic kidney disease (ADPKD) urine and cyst fluid. Western blots of immunoprecipitated MCP-1 in urine (A) and cyst fluid (B). (A) MCP-1 alone (10 ng) and urine samples from two ADPKD patients (70 and 79) examined with and without an excess of competing peptide (CP). MCP-1 standard in lanes 1 and 6 were not subjected to immunoprecipitation. Molecular weight of monomeric MCP-1 (indicated by arrow at 8.2 kD noting kaleidoscope marker) was approximately 9 kD. The faint band at approximately 18 kD in lane 1 is dimeric MCP-1. The band immediately above MCP-1 in urine (lanes 2 and 3) is most likely glycosylated MCP-1. Heavy staining of the upper portions of the gels, unaffected by incubating anti–MCP-1 with an excess of CP (lanes 4 and 5) was due to secondary antibody binding to nonspecific proteins added during the immunoprecipitation procedure. This staining was absent in the first and last lanes containing MCP-1 standards in medium. (B) Cyst fluid 123 examined with and without anti–MCP-1 incubated with an excess of CP. As in urine, MCP-1 of molecular weight approximately 9 kD was detected as well as a heavier band, both of which were eliminated by incubating anti–MCP-1 with an excess of CP. Lane 5 contains cyst fluid to which secondary antibody was added alone.

In immunoprecipitated samples of cyst fluid, two bands were found, one aligned with the MCP-1 standard and the other at a slightly higher molecular weight (Figure 1B). Both of these bands were eliminated by an excess of competing antigen. These findings indicate that proteins with a molecular weight and immunoreactivity characteristic of MCP-1 are present in ADPKD urine and cyst fluids.

MCP-1 in Urine, Serum, and Cyst Fluid

MCP-1 levels in urine and serum are summarized in Table 1. Urinary MCP-1 was referenced to creatinine, thereby providing an indication of cytokine excretion in relation to GFR (Tables 1 and 2⇓, Figure 2). The concentration of urine MCP-1 referenced to creatinine of 55 ADPKD patients ranged from 43 to 2835 pg/mg. By contrast, the range of values in 19 normal subjects, 36 to 262 pg/mg, was much narrower. The greater urinary excretion of MCP-1 in ADPKD was reflected in a higher mean level (623 ± 73 versus 151 ± 16 pg/mg; P < 0.001). There did not seem to be a gender difference in the urinary excretion of MCP-1 in either ADPKD patients or normal subjects (Tables 1 and 2⇓). The range and mean values for serum concentrations of MCP-1 did not seem to be different between ADPKD patients and normal subjects, and there was no a gender effect (Table 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. Relation between urine MCP-1 excretion and serum creatinine concentration. Normal values for urine MCP-1 fell below 263 mg/mg (vertical dashed line) and serum MCP-1 below 1.5 mg/dl (horizontal dashed line). ADPKD patients were divided into groups 1 to 3 determined by urine MCP-1 excretion and serum creatinine levels. The ADPKD data could be fit best to a two-parameter exponential equation that inscribed a parabolic curve (r² = 0.605).

If renal MCP-1 has either an active or a passive role in the pathogenesis of disease progression as suggested by the earlier study of the Han:SPRD rat (7), then one would expect to see an inverse relationship between the urinary excretion of MCP-1 and the decrease in renal function reflected in the GFR or increased levels of proteinuria, traditional markers of disease progression. Because of the nature of this study, we could evaluate only two classical markers of renal dysfunction, serum creatinine concentration and urine protein excretion. There seemed to be a direct relation between urine MCP-1 excretion and serum creatinine concentration, but it was not linear (Figure 2). Urine MCP-1 excretion and serum creatinine concentration were best fit to a parabolic curve inscribed by a two-parameter equation (curve and equation not shown, r² = 0.605). The relation between urine MCP-1 and serum creatinine concentration suggests but does not prove a possible linkage between the magnitude of chemokine excretion and the degree to which creatinine clearance may be decreased.

When the data for normal subjects were added to this plot, the narrow ranges compared with that of the ADPKD cohort suggested another way to examine the data. In this snapshot study of ADPKD including patients with the disease ranging from relatively mild to severe, more than likely some of the 40 patients with serum creatinine levels in the normal range would ultimately progress to the point at which the serum creatinine level would become clearly abnormal. Furthermore, it would be interesting to know whether urine MCP-1 or protein excretion might indicate which of those patients might be at higher risk of progressing to ESRD.

To facilitate the evaluation, we selected upper limits for urine MCP-1 (263 pg/mg) and serum creatinine (1.5 mg/dl) on the basis of the normal subjects in the study. These limits are represented by dashed lines in Figure 2. Examined in this light, the ADPKD patients were sorted into three groups (Table 2). Thirteen patients fell within the normal limits (group 1), and in these the urine protein excretion rate was normal as well.

Twenty-seven of the ADPKD patients had urinary MCP-1 excretion levels higher than normal, yet their serum creatinine concentrations were within normal limits (group 2). In these group 2 patients with urine MCP-1 excretion levels that were clearly elevated, there was no detectable increase in urine protein excretion (Table 2). Fifteen of the ADPKD patients had elevated urine MCP-1 and serum creatinine levels (group 3); in these patients, urine protein excretion was also significantly increased.

In 17 ADPKD patients, urine MCP-1 measurements were made on more than one occasion (Figure 3). In most of the patients, MCP-1 excretion was relatively stable for as long as 21 mo.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3. Time course of urine MCP-1 excretion in ADPKD patients. Seventeen ADPKD patients had sequential measurements of MCP-1 for as long as 21 mo.

In cyst fluids removed from ADPKD kidneys in a different cohort of patients, MCP-1 concentrations ranged from 767 to 40,860 pg/ml (mean, 6434 ± 841 pg/ml; n = 73). This mean level greatly exceeded the average MCP-1 concentrations in urine or serum from both ADPKD patients and normal subjects (Tables 1 and 2⇑).

Secretion of MCP-1 by ADPKD Cyst Epithelial Cells in Culture

We cultured ADPKD cyst cells on polarized supports to determine whether the mural epithelium might contribute to the high levels of MCP-1 found in cyst fluid. After the cultures reached confluence, they were washed thoroughly with fresh medium and placed in a nutrient medium containing minimal growth factors and no known cytokine agonists. Primary cultures from normal human renal cortex were treated identically. After 24 h, the conditioned media bathing the apical and basolateral surfaces of the cultured cells were collected and MCP-1 was determined by ELISA. The concentrations of MCP-1 achieved in the fluids bathing the apical surfaces of both ADPKD and HKC cells were seven to eightfold higher than in the basolateral fluids (Figure 4). This indicates that the MCP-1 was released by the cells across the apical plasma membranes. Taking into account the different volumes of the apical (0.4 ml) and basolateral compartments (3 or 5 ml), the total amounts of MCP-1 produced in vitro were not different between ADPKD and HKC cells (Figure 4).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4. Appearance of MCP-1 in medium used to culture renal epithelial cells. (A) MCP-1 concentrations determined in cyst epithelial cultures from four ADPKD specimens and four human kidneys without ADPKD (human kidney cortex [HKC]). Vertical bars indicate MCP-1 concentrations (mean ± SEM) in apical and basolateral fluids. The total amounts of MCP-1 accumulated in both apical and basolateral fluids of the cultures are shown by the two vertical bars on the right. (B) Western blot of conditioned medium (72 h) from ADPKD and HKC cultures with and without incubation with anti–MCP-1 antibody containing an excess of CP. In the first lane, MCP-1 (10 ng) was added to medium alone. The bands represent monomeric MCP-1 of approximately 9 kD and dimeric MCP-1 of approximately 18 kD. Two bands were detected in the conditioned medium of PKD and HKC cells, one consistent with MCP-1 and the other a glycosylated derivative.

To determine the nature of the proteins secreted by ADPKD and HKC cells in culture, we performed Western blots on unmodified conditioned medium collected over a 72-h interval. As was found in ADPKD urine and cyst fluid (Figure 1), two adjacent bands were detected in conditioned medium obtained from both ADPKD and HKC cells. Both bands were eliminated by incubation of the antibody used in the Western blot with an excess of immunizing peptide (Figure 4).

We could not quantify MCP-1 in urine and cyst fluid by Western blotting because the samples were concentrated by immunoprecipitation. The concentrations of MCP-1 in conditioned medium were high enough and that of other proteins sufficiently low that Western blots could be obtained after direct application of sample to the gels. The levels of MCP-1 in conditioned media determined by ELISA and those determined by quantitative Western blot analysis fell within the same order of magnitude (data not shown).

Other Cytokines Excreted in Urine and Cyst Fluid

MCP-1, MCP-3, IL-8, and TNF-α levels were determined in the urine selected from five ADPKD and three normal individuals and in cyst fluids from seven polycystic kidneys (Table 3). Because the intent of this experiment was to determine the extent to which the different cytokines were expressed in concert with MCP-1, the ADPKD urine and cyst fluids were selected to give a broad range of MCP-1 concentrations. Because we did not anticipate any correlation among the cytokines in the normal urine, we selected three samples from those with the higher levels of MCP-1 to gain a small glimpse of the level of cytokine excretion in normal individuals. In ADPKD patients, the excretion rates of MCP-3, IL-8, and TNF-α were significantly less than MCP-1, as were the levels of these cytokines in cyst fluid (Table 3). The urinary excretion of MCP-1 correlated directly with the excretion of IL-8, although the urine levels of IL-8 did not seem to rise to the same extent as MCP-1. There was no correlation between urinary MCP-1 and MCP-3 or TNF-α. In cyst fluid, MCP-1 levels seemed to correlate with IL-8 and to a lesser extent with MCP-3 but not with TNF-α.