Inflammatory chemokine (C‐C motif) ligand 8 inhibition ameliorates peritoneal fibrosis

Abstract

Peritoneal fibrosis (PF) is an irreversible complication of peritoneal dialysis (PD) that leads to loss of peritoneal membrane function. We investigated PD effluent and serum levels and the tissue expression of chemokine (C‐C motif) ligand 8 (CCL8) in patients with PD. Additionally, we investigated their association with PF in a mouse model. Eighty‐two end‐stage renal disease (ESRD) patients with PD were examined. CCL8 levels were measured via enzyme‐linked immunosorbent assays in PD effluents and serum and analyzed with peritoneal transport parameters. Human peritoneal mesothelial cells (hPMCs) were obtained from the PD effluents of 20 patients. Primary cultured hPMCs were treated with recombinant (r) transforming growth factor (TGF)‐β, and CCL8 expression was assessed via western blotting. As the duration of PD increased, the concentration of CCL8 in PD effluents significantly increased. Correlations between peritoneal transport parameters and dialysate CCL8 levels were observed. Western blotting analysis showed that CCL8 was upregulated via rTGF‐β treatment, accompanied by increases in markers of inflammation, fibrosis, senescence, and apoptosis in hPMCs after induction of fibrosis with rTGF‐β. Anti‐CCL8 monoclonal antibody (mAb) treatment suppressed the rTGF‐β‐induced increase in all analyzed markers. Immunohistochemical analysis revealed that CCL8 along with fibrosis‐ and inflammation‐related markers were significantly increased in the PF mouse model. Functional blockade of CCL8 using a CCR8 inhibitor (R243) abrogated peritoneal inflammation and fibrosis in vivo. In conclusion, high CCL8 levels in PD effluents may be associated with an increased risk of PD failure, and the CCL8 pathway is associated with PF. CCL8 blockade can ameliorate peritoneal inflammation and fibrosis.

Abbreviations

CCL8

chemokine (C‐C motif) ligand 8

CG

chlorhexidine gluconate

EMT

epithelial‐to‐mesenchymal transition

ESRD

end‐stage renal disease

hPMCs

human peritoneal mesothelial cells

mAb

monoclonal antibody

PD

peritoneal dialysis

PF

peritoneal fibrosis

qRT‐PCR

quantitative real‐time polymerase chain reaction

TGF‐β

transforming growth factor‐β

1. INTRODUCTION

Peritoneal fibrosis (PF) is a severe complication of peritoneal dialysis (PD) that leads to failure of peritoneal membrane function. Peritoneal membrane function failure is characterized by acute and chronic peritoneal inflammation, angiogenesis, and epithelial‐to‐mesenchymal transition (EMT). ¹ , ² During this transition, mesothelial cells migrate from the superficial mesothelial layer toward the sub mesothelium where they produce extracellular matrix, thereby contributing to fibrosis. ³ Moreover, bioincompatible PD solutions (high glucose, glucose degradation products, and advanced glycation end‐products) alter mesothelial cell function and proliferation. ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹

Transforming growth factor‐β (TGF‐β) functions as a key factor in the regulation of PF. ² The TGF‐β superfamily is broadly expressed in almost every cell type and has multiple biological functions, such as cell proliferation, apoptosis, differentiation, autophagy, and organ fibrosis. ¹⁰ Activation of TGF‐β is a crucial event that mediates fibrogenesis via exposure to PD solutions containing high concentrations of glucose. ¹¹ TGF‐β mediates fibrosis, inflammation, and senescence through an intracellular signaling cascade involving the Smad family of proteins, which are transduced via its downstream effectors such as p65 (nuclear factor [NF]‐κB) and β‐galactosidase. ¹² Additionally, connective tissue growth factor signaling, inflammatory pathways involving nucleotide oligomerization domain (NOD)‐like receptor protein 3/interleukin (IL)‐1β signaling, and other proinflammatory cytokines modulate peritoneal inflammation for fibrosis development. ¹³

Presently, reliable diagnostic biomarkers for PF remain scarce. Therefore, we performed a prospective study in a mouse model of chlorhexidine gluconate (CG)‐induced PF and analyzed microarrays of peritoneal tissues to identify new biomarkers. Among many differentially expressed genes, chemokine (C‐C motif) ligand 8 (CCL8) was the most upregulated chemokine in PF model mice compared to that in control mice. This result suggests that chemokines and their receptors play a crucial role in PF. Therefore, we next investigated whether CCL8 can be used as a marker for PD failure.

CCL8, also known as monocyte chemoattractant protein‐2, was first identified in human osteosarcoma cells and functions in various inflammatory cells as a chemotactic factor by binding chemokine receptors, mainly CCR2. ¹⁴ , ¹⁵ Until recently, human CCL8 and its increased levels have been detected in limited diseases, such as graft‐versus‐host disease (GVHD), idiopathic pulmonary fibrosis (IPF), and chronic kidney disease. The serum CCL8 level in patients with GVHD correlates closely with GVHD severity and is a promising serum marker for early diagnosis. ¹⁴ Moreover, a recent study suggested that CCL8 concentrations in bronchoalveolar lavage fluids and CCL protein expression in lung tissues are significantly high in patients with IPF. Increased CCL8 levels are also associated with mortality. ¹⁶ Recently, Lee et al. have proposed the role of CCL8 in kidney disease for the first time. ¹⁷ The CCL8 pathway is associated with an increased risk of kidney fibrosis and CCL8 blockade can relieve kidney fibrosis and apoptosis. ¹⁷ However, the role of CCL8 in PF has not been previously described. The objective of this study was to investigate the association between PD effluent levels of CCL8 and PD failure and the function of CCL8 in PF using cell‐based and in vivo models.

2. MATERIALS AND METHODS

2.1. Study population and data collection

A total of 82 end‐stage renal disease (ESRD) patients with PD were enrolled at the Seoul National University Hospital, Seoul, South Korea, between January 2015 and October 2020. Demographic and clinical features as well as biochemical parameters were collected from the electronic medical records of the patients. The demographic and clinical features included age, sex, primary renal disease, and follow‐up duration. Biochemical parameters included serum creatinine (Cr) levels, estimated glomerular filtration rate (eGFR), and other laboratory findings (serum hemoglobin, albumin, and high‐sensitivity C‐reactive protein levels). Serum samples were obtained at the same time point as PD effluent. Peritoneal transport was assessed by calculating small solute transport, and peritoneal Kt/V (pKt/V) was estimated simultaneously as the PD effluent was collected.

The present study was approved by the Institutional Review Board of the Seoul National University Hospital (IRB No. H‐2004‐148‐1118). Informed consent was obtained from all patients before serum and PD effluent collection, which was performed and supervised by the Seoul National University Hospital Human Biobank (IRB No. H‐1506‐097‐681). All clinical investigations were performed in accordance with the guidelines of the 2013 Declaration of Helsinki.

2.2. Measurement of serum CCL8 levels using enzyme‐linked immunosorbent assay (ELISA)

Serum CCL8 levels were quantified using an ELISA kit (Lifespan Biosciences, Seattle, WA, USA) according to the manufacturer's protocol. All measurements were performed in duplicate in a blinded manner. The lower limit of detection for CCL8 was 31.25 pg/ml.

2.3. Isolation and primary culture of human peritoneal mesothelial cells (hPMCs)

Effluent from clinically stable patients was obtained and immediately processed. ¹⁸ Bags were suspended for 3–4 h in an incubator at 37°C to allow the cells to settle at the bottom. The supernatant was removed via vacuum using a sterile pipette, leaving approximately 200 ml of sediment. The cells were transferred to four 50 ml tubes, centrifuged at 1500 rpm for 20 min, and washed twice with phosphate‐buffered saline (PBS). The cell pellets were resuspended in a 5–7 ml culture medium, counted in a Neubauer chamber, seeded in 25‐cm² tissue culture flasks, and incubated at 37°C in a humidified atmosphere with 5% CO₂. The culture medium was M199 (Gibco, Grand Island, NY, USA) supplemented with 20% fetal bovine serum, 100 IU/ml penicillin, 100 mg/ml streptomycin, and 2% Biogro‐2 (Gibco), which contains insulin, transferrin, ethanolamine, and putrescine. The medium was replaced every 2–3 days. After 48 h, mesothelial cells (attachment cells) were stably attached to the plate, and peritoneal leukocytes could be sufficiently removed in subsequent washes as they were in suspension. hPMCs from the second passage were used for the experiments. Primary cultured hPMCs were seeded in six‐well plates or 10‐cm² plastic culture dishes containing Dulbecco's modified Eagle's medium/F12. When the cells reached 70%–80% confluence, they were serum‐starved for 24 h to synchronize cell growth.

2.4. Confocal microscopic examination

hPMCs stimulated with rTGF‐β (R&D Systems, Wiesbaden, Germany) and treated with or without an anti‐CCL8 monoclonal antibody (mAb) (Invitrogen, Carlsbad, CA, USA) were washed with PBS and fixed in 4% paraformaldehyde for 20 min. Following fixation, the cells were permeabilized using 0.3% Triton X and stained with antibodies against CCL8 and alpha‐smooth muscle actin (α‐SMA) (Abcam, Cambridge, United Kingdom) in a blocking agent overnight at 4°C. Alexa 488/555‐conjugated probes (Invitrogen) were used as secondary antibodies, and 4′, 6‐diamidino‐2‐phenylindole (Invitrogen) was used to counterstain the nuclei. The primary antibodies were excluded from the negative controls. Immunofluorescence images were acquired using a confocal microscope (Leica TCS SP8; Leica Microsystem GmbH, Wetzlar, Germany). The quantification of the fluorescence was analyzed using the Leica Las X software (Leica Microsystem GmbH, Wetzlar, Germany).

2.5. Western blot analysis

After the primary cultured cells were removed from culture dishes, the proteins were retrieved using a radioimmunoprecipitation assay buffer containing Halt protease inhibitor (Pierce, Rockford, IL, USA). Western blotting was performed using primary antibodies against α‐SMA, periostin, CCR2, p53, p65, phosphorylated (p‐)p65, tumor necrosis factor receptor 1 (TNFr1), IL‐1β (all from Abcam), CCL8 (Novus Biologicals, Centennial, CO, USA), fibronectin and β‐galactosidase (Santa Cruz Biotechnology, Dallas, TX, USA), intercellular adhesion molecule 1 (ICAM‐1) (Thermo Fisher Scientific, Waltham, MA, USA), and β‐actin (Sigma–Aldrich, St. Louis, MO, USA). Extracted proteins (20–80 μg) were separated on 10% sodium dodecyl sulfate‐polyacrylamide gels and transferred onto Immobilon‐FL 0.4 μm polyvinylidene difluoride membranes (Millipore, Burlington, MA, USA). Anti‐rabbit IgG (Cell Signaling Technology, Danvers, MA, USA) and anti‐mouse IgG (Cell Signaling Technology) were used as horseradish peroxidase‐conjugated secondary antibodies. The immunoblot bands were visualized, and images were captured using an ImageQuant LAS 4000 Mini instrument (GE Healthcare, Princeton, NJ, USA). Western blotting results were quantified using ImageJ (National Institutes of Health, Bethesda, MD, USA).

2.6. Flow cytometry

Apoptosis and necrosis were quantified via flow cytometry using an annexin V/propidium iodide (PI) assay. The fibrotic response was induced using 1 ng/ml rTGF‐β (R&D Systems) for 24 h. rTGF‐β‐stimulated hPMCs were simultaneously treated with an anti‐CCL8 monoclonal antibody (mAb) (1 μg/ml; Invitrogen). Cells stained with PI and fluorescein isothiocyanate‐conjugated annexin V (BD Biosciences, Franklin Lakes, NJ, USA) were incubated for 30 min in the dark, followed by analysis using a BD FACS Diva instrument (version 8.0; BD Biosciences).

2.7. Quantitative real‐time polymerase chain reaction (qRT‐PCR)

Total RNA was isolated from hPMCs and peritoneal tissues using TRIzol reagent (Thermo Fisher Scientific). cDNA was synthesized using a reverse transcription kit (Promega, Madison, WI, USA) and a C1000 thermal cycler (Bio‐Rad, Hercules, CA, USA). Subsequently, qRT‐PCR was performed using a LightCycler‐480 instrument II (Roche Molecular Systems, Basel, Switzerland). Fibronectin, collagen 1a (Col1a), protein S100‐A8 (S100a8), and CCL8 mRNA levels were analyzed using the comparative Ct method (ΔΔC _t ) after normalization to the glyceraldehyde 3‐phosphate dehydrogenase levels. PCR primers used for qRT‐PCR are listed in Table 1.

TABLE 1.

Primer sets were used for qRT‐PCR

Genes	Forward (5′‐>3′)	Reverse (5′‐>3′)
CCL8	CTCAGCCAGATTCAGTTTCCA	GGTGATTCTTGTGTAGCTCTC
Fibronectin	CCACCCCCATAAGGCATAGG	GTAGGGGTCAAAGCACGAGTCATC
Collagen 1a (Col1a)	GATTCCCTGGACCTAAAGGTGC	AGCCTCTCCATCTTTGCCAGCA
S100a8	AGACCGAGTGTCCTCAGTATATC	TGCCACGCCCATCTTTATC
GAPDH	TCGACAGTCAGCCGCATCT	CCGTTGACTCCGACCTTCA

Abbreviations: CCL8, chemokine (C‐C motif) ligand 8; GAPDH, glyceraldehyde 3‐phosphate dehydrogenase.

2.8. Animal model establishment

Animal experiments were performed with the approval of the Institutional Animal Care and Use Committee of Seoul National University Hospital (Approval No. 15‐0271‐S1A0). C57BL/6 (WT) mice (The Jackson Laboratory, Bar Harbor, ME, USA) weighing 20 g were randomly assigned to two groups (n = 6/group), which consisted of sham and CG‐PD groups. PF was induced via daily intraperitoneal injection of 0.2 ml chlorhexidine solution composed of 0.1% CG and 15% ethanol for 4 weeks. After 4 weeks, the mice were anesthetized, and the peritoneal tissue was dissected. The peritoneum at the pole was embedded in paraffin for immunostaining or snap‐frozen in liquid nitrogen for immunoblotting. For assessing peritoneal inflammation, fibrosis, and apoptosis, staining was performed using Masson's trichome and antibodies against ICAM‐1, F4/80, and CCL8. Three to five fields (×200) were randomly chosen for each sample and quantified using computer‐based morphometric analysis (Leica Qwin 3). Scoring was conducted under blind conditions using the mean values of the positive areas (%).

2.9. In vivo inhibition of CCL8‐CCR8 with R243 treatment

Mouse CCL8 is an agonist for the chemokine receptor CCR8, ¹⁹ and R243 is a small molecule that functionally inhibits CCR8 in vivo and in vitro. ²⁰ , ²¹ We, therefore, decided to treat with R243 (MedKoo Biosciences, Morrisville, NC, USA) to block the CCL8 function indirectly. Mice were treated with R243 (1.0 mg/kg per intraperitoneal injection thrice a week) at day 0 and continued until day 28. On day 28, the effects of treatment on peritoneal inflammation, fibrosis, and apoptosis were assessed.

2.10. Statistical analysis

The results were compared using either the t‐test or a one‐way analysis of variance. Pearson correlation coefficients were calculated to explore the linear relationship between dialysate CCL8 levels and various clinical parameters. All statistical analyses were performed using the STATA software (version 16.1; StataCorp LP, College Station, TX, USA) and GraphPad Prism (version 9.0.2; GraphPad Software, La Jolla, CA, USA). p < .05 was considered significant.

3. RESULTS

3.1. Association of dialysate CCL8 with peritoneal membrane function and transport parameters

The baseline characteristics of the 82 patients with ESRD and PD are shown in Table 2. The mean age of the patients was 46.4 ± 11.9 years, and 40 (48.8%) were male. The median duration of PD was 35.5 ± 21.6 months. Dialysate and serum CCL8 levels were 356.6 ± 212.0 and 258.8 ± 118.1 pg/ml, respectively. As the duration of PD increased, the concentration of CCL8 in PD effluents increased significantly (p = .036, Figure 1A). When the peritoneal equilibration test and adequacy of PD (weekly peritoneal Kt/V; pKt/V) were evaluated, there were positive associations between dialysate CCL8 levels and pKt/V, D/P Cr, and D/P60 sodium (Na). Additionally, a negative association was noted between the CCL8 level and D/D0 glucose (All p < .05, Figure 1B). In particular, D/P Cr and D/P60 Na had a strong positive correlation with the CCL8 concentration in PD effluents (Spearman r = 0.52 and 0.53, respectively).

TABLE 2.

Clinical baseline characteristics of patients undergoing peritoneal dialysis

Clinical characteristics	n (%)
Patients	82
Age (years)	46.4 ± 11.9
Men	40 (48.8)
Follow‐up duration (months)	35.5 ± 21.6
Primary renal disease
Diabetic nephropathy	19 (23.2)
Hypertension	17 (20.7)
Glomerulonephritis	24 (29.3)
Others	22 (26.8)
Laboratory findings
Hemoglobin (g/dl)	10.4 ± 1.3
Albumin (g/dl)	3.7 ± 0.3
eGFR (ml/min/1.73 m2)	4.3 ± 2.0
Creatinine (mg/dl)	12.7 ± 4.2
C‐reactive protein	0.3 ± 1.0
Peritoneal membrane function
D/P creatinine	0.75 ± 0.11
D/P urea	0.99 ± 0.05
D/D0 glucose	0.27 ± 0.07
D/P60 sodium	0.86 ± 0.04
Weekly peritoneal Kt/V	1.45 ± 0.45
Weekly renal Kt/V	0.48 ± 0.57
Total weekly Kt/V	1.93 ± 0.42
Dialysate CCL8 (pg/ml)	356.6 ± 212.0
Serum CCL8 (pg/ml)	258.8 ± 118.1

Note: Values are expressed as the mean ± standard deviation or percentages.

Abbreviations: CCL8, chemokine (C‐C motif) ligand 8; eGFR, estimated glomerular filtration rate.

FIGURE 1.

Association of dialysate CCL8 and peritoneal membrane function and transport parameters (n = 82). (A) There was a positive association between dialysate CCL8 level and PD duration (p = .036). (B) Association of dialysate CCL8 with peritoneal equilibration test and adequacy of PD. CCL8, chemokine (C‐C motif) ligand 8; PD, peritoneal dialysis.

3.2. rTGF‐β induced fibrotic response in primary cultured hPMCs

To determine whether CCL8 is expressed during PF in hPMCs, the cells were treated with 1 ng/ml rTGF‐β for 24 h, and CCL8 and fibronectin expression levels were evaluated via western blotting. CCL8 was upregulated after rTGF‐β treatment (Figure 2A,B). CCL8 and fibronectin, which is a marker of fibrosis, were analyzed via qRT‐PCR. The mRNA expression levels of fibronectin and CCL8 showed a similar pattern of a significant increase in a dose‐dependent manner after treatment with 1 and 2 ng/ml rTGF‐β (Figure 2C). These results indicate that rTGF‐β induces a fibrotic response in hPMCs with upregulation of CCL8.

FIGURE 2.

rTGF‐β‐induced fibrosis in primary cultured hPMCs. (A) Western blot analysis of hPMCs after induction of fibrosis with rTGF‐β (1 ng/ml). rTGF‐β increased CCL8 protein expression. (B) Quantification of (A). (C) Quantitative real‐time PCR analysis of fibronectin and CCL8 with rTGF‐β (1 and 2 ng/ml) treatment (n = 6). Analyses were performed at least three times with duplicate samples. *p < .05, **p < .01, and ***p < .001. CCL8, chemokine (C‐C motif) ligand 8; hPMC, human peritoneal mesothelial cells; PCR, polymerase chain reaction; rTGF, recombinant transforming growth factor.

3.3. Anti‐inflammatory effect of anti‐CCL8 therapy

To evaluate the role of CCL8 in inflammation, hPMCs were simultaneously treated with rTGF‐β (1 ng/ml) and various concentrations of anti‐CCL8 mAb (0.25, 0.5, and 1.0 μg/ml). rTGF‐β treatment induced the expression of markers of inflammation and proinflammatory cytokines (TNFr1, IL‐1b, IL‐6, ICAM‐1, p65, and pp65) as determined via western blotting (Figure 3A,B). These increases were significantly reduced after blocking CCL8 function, particularly at anti‐CCL8 mAb concentrations above 0.5 μg/ml. However, there was no significant decrease upon treatment with 0.25 μg/ml of anti‐CCL8 mAb. These results indicated that the upregulation of CCL8 contributed to peritoneal inflammation.

FIGURE 3.

Anti‐inflammatory effect of CCL8 blockade on hPMCs. (A) Western blot analysis of rTGF‐β‐stimulated hPMCs after treatment with various concentrations of anti‐CCL8 mAb (0.25, 0.5, and 1.0 μg/ml) (n = 6). (B) Quantification of (A). *p < .05, **p < .01, and ***p < .001. hPMC, human peritoneal mesothelial cells; mAb, monoclonal antibody; rTGF, recombinant transforming growth factor.

3.4. Anti‐CCL8 treatment modulated fibrosis and apoptosis in hPMCs

We further investigated whether CCL8 plays a similar role in fibrosis and apoptosis in hPMCs. Western blotting was performed to analyze fibrosis markers such as fibronectin, α‐SMA, and periostin; p53, an apoptosis marker; β‐galactosidase, a marker of senescence; and CCR2, a receptor for CCL8. The simultaneous treatment with various concentrations of anti‐CCL8 mAb (0.25, 0.5, and 1.0 μg/ml) suppressed the rTGF‐β‐induced increase in all analyzed markers, particularly at concentrations above 0.5 μg/ml (Figure 4A,B).

FIGURE 4.

(A) Fibrosis‐ and apoptosis‐related markers are highly induced after induction of fibrosis with rTGF‐β (1 ng/ml) and attenuated by CCL8 blockade (n = 6). (B) Quantification of (A). *p < .05, **p < .01, and ***p < .001. CCL8, chemokine (C‐C motif) ligand 8; rTGF, recombinant transforming growth factor.

Confocal images with immunohistochemical staining for α‐SMA and CCL8 showed that the cell morphology was prominently altered via the administration of rTGF‐β. The rTGF‐β treated cells showed a more elongated shape, considerable cytoplasmic projection, and enriched fibers. Additionally, the simultaneous treatment with anti‐CCL8 mAb alleviated the morphological changes (Figure 5A). Quantitative analysis of fluorescence following FACS showed significant decreases in fluorescence intensities of α‐SMA and CCL8 after treatment with anti‐CCL8 mAb. The anti‐apoptotic effect of anti‐CCL8 mAb was also demonstrated via Annexin V/PI staining in hPMCs (Figure 5B). When treated with rTGF‐β, the number of apoptotic cells increased significantly compared with that after control IgG treatment and significantly decreased after treatment with anti‐CCL8 mAb. Thus, TGF‐β may induce EMT in hPMCs, and this effect is attenuated by blocking the CCL8 function.

FIGURE 5.

CCL8 Mediates rTGFβ‐induced fibrosis and apoptosis in hPMCs. (A) Representative confocal microscopic images of hPMCs co‐stained for CCL8 (red), α‐SMA (green), and DAPI (blue) (left panel). Original magnification: ×400. Bar graphs show fluorescence intensities of α‐SMA and CCL8 after rTGF‐β treatment in the presence or absence of anti‐CCL8 mAb (right panel). (n = 5) Bar, mean ± SEM, **p < .01. (B) The anti‐apoptotic effect of the anti‐CCL8 mAb was demonstrated by Annexin V/propidium iodide staining using hPMCs (n = 3). The histogram on the right of B is described as the percentage (sum) of annexin V‐positive or PI‐positive cells in the dot plots. CCL8, chemokine (C‐C motif) ligand 8; DAPI, 4′,6‐diamidino‐2‐phenylindole; hPMC, human peritoneal mesothelial cells; SEM, standard error of the mean; α‐SMA, alpha‐smooth muscle actin.

3.5. Expression of CCL8 in a mouse model

We performed an immunohistochemical analysis of a mouse model of CG‐induced PF. In control mice, peritoneal tissues showed no thickening of the submesothelial area. Masson's trichome (MT) staining and immunohistochemical detection of ICAM‐1 revealed an increase in peritoneum thickness accompanying the upregulation of CCL8 protein in the submesothelial layer in PF model mice (Figure 6A,B). The expression of F4/80, which is used to identify macrophages, and CCL8 expression in peritoneal tissues also demonstrates the contribution of CCL8 to peritoneal inflammation. Moreover, the mRNA expression of Col1a, S100a8, and CCL8 was increased in the peritoneal tissues of CG‐treated mice (Figure 6B).

FIGURE 6.

CCL8 expression in a mouse model of PF (CG‐PD). (A) Examination of PF model mice by microscopy (original magnification: ×200; scale bar: 100 μm). (B) Quantification of Masson's trichome (MT) staining and three indicated marker‐positive areas (ICAM‐1, F4/80, and CCL8). (C) The mRNA expression of in the peritoneum of PF model mice. (n = 5–6 mice per group) *p < .05, **p < .01, and ***p < .001. CCL8, chemokine (C‐C motif) ligand 8; CG, chlorhexidine gluconate; hPMC, human peritoneal mesothelial cells; ICAM, intercellular adhesion molecule 1; PF, peritoneal fibrosis.

3.6. Amelioration of PF in CCL8‐CCR8 blockade by R243 in vivo

In mice, CCR8 is a receptor for CCL8. ¹⁹ Since R243 is a small molecule that functionally inhibits CCR8 both in vivo and in vitro, ²⁰ , ²¹ we hypothesized that blocking CCL8‐CCR8 by R243 would indirectly inhibit CCL8 function. PF model mice were treated with R243 (1.0 mg/kg per dose; intraperitoneal) thrice a week from day 0 to 28. We finally examined whether CCL8 blockade could ameliorate PF in CG‐treated mice. As shown in Figure 7A, at day 28, the peritoneum of CG‐treated mice revealed marked thickening and significant fibrosis on MT and Sirius Red staining. Moreover, CG‐ and R243‐treated mice showed a significant improvement in peritoneal thickening (300.8 ± 31.34 μm in CG‐treated mice versus 151.4 ± 25.13 μm in CG‐ and R243‐treated mice; p = .003). We also observed a significant reduction in peritoneal fibrosis by Sirius Red staining (23.1 ± 1.15% in CG‐treated mice versus 10.8 ± 1.88% in CG‐ and R243‐treated mice; p < .001) and peritoneal inflammation by F4/80 expression (3.7 ± 0.59% in CG‐treated mice versus 1.8 ± 0.35% in CG‐ and R243‐treated mice; p = .009) (Figure 7B). These results suggest that blocking CCL8 function ameliorates peritoneal fibrosis by modulating anti‐fibrotic pathways and anti‐inflammatory signals.

FIGURE 7.

CCL8 blockade with R243 showed anti‐fibrotic and anti‐inflammatory effects in PF mice. (A) Representative microscopic images of the peritoneum on day 28 after treatment with R243 in the CG‐induced PF mouse model (original magnification: ×200; scale bar: 100 μm). (B) Bar graphs demonstrating peritoneal thickness and quantification of Masson's trichome (MT), Sirius Red staining, and F4/80‐positive areas from indicated groups. (n = 7 mice per group) Bar, mean ± SEM, **p < .01 and ***p < .001. CCL8, chemokine (C‐C motif) ligand 8; CG, chlorhexidine gluconate; PAS, periodic acid‐Schiff; PF, peritoneal fibrosis; SEM, standard error of the mean.

4. DISCUSSION

The role of CCL8 in PF has not been previously investigated; therefore, we investigated the association of CCL8 and PD parameters with PF. To the best of our knowledge, this is the first study to suggest the role of CCL8 in the process of PF. Correlations of peritoneal transport parameters with CCL8 levels in PD effluents were shown. The inhibition of CCL8 signaling using anti‐CCL8 mAb attenuated the rTGF‐β‐induced peritoneal inflammation, fibrosis, and apoptosis in primary cultured hPMCs. Using in vivo CG‐treated PF models, we demonstrated that functional blockade of CCL8 using a small molecule CCR8 inhibitor (R243) abrogated peritoneal inflammation and fibrosis.

CCL8 is a C‐C chemokine that belongs to a family of small cytokines. The CCL8 protein starts as a 109‐amino‐acid precursor, which is then cleaved to create the mature CCL8 containing 75 amino acids. ¹⁵ , ²² CCL8 chemotactically attracts and activates diverse immune cells, including mast cells, eosinophils, basophils, monocytes, T cells, and natural killer cells, all of which are involved in the inflammatory reaction. ²³ , ²⁴ Fibrosis in various organs and tissues is considered a pathological consequence of a chronically altered damage recovery response by recruiting immune cells. CCL8 is produced via toll‐like receptors when fibroblasts and endothelial cells are co‐stimulated with IL‐1β and interferon‐γ. ²⁵ Although several chemokines and chemokine receptors, such as CXCL, CCL, CX3CL, and XCL groups, are associated with the fibrosis of diverse tissues, ²⁶ the role of CCL8 in PF has not been investigated.

The most meaningful finding of the present study is the validation of the anti‐inflammatory, anti‐fibrotic, and anti‐apoptotic effects of CCL8 blockade in PF. Research over the prior two decades has demonstrated that PF is closely associated with inflammation, angiogenesis, and EMT and that these processes influence each other. ² EMT is one of the processes receiving considerable attention in recent years. However, EMT is not the only source of myofibroblast‐like submesothelial cells promoting fibrosis during PD. Recent studies have shown that vascular endothelial growth factor‐mediated signaling in angiogenesis is associated with ultrafiltration failure and closely related to EMT. ¹³ , ²⁷ , ²⁸ , ²⁹ , ³⁰ Further, IL‐6 trans‐signaling appears to be associated with TGF‐β/Smad3 signaling because the blockade of TGF‐β or Smad3 using siRNA inhibits IL‐6/sIL‐6R‐mediated EMT in hPMCs. ³¹ In the present study, treatment of cultured hPMCs with rTGF‐β induced EMT and increased the expression of fibronectin, α‐SMA, and periostin; p53, an apoptosis marker; and β‐galactosidase, a marker of senescence. Anti‐CCL8 treatment markedly alleviated fibrosis, apoptosis, and inflammation in rTGF‐β‐induced hPMCs.

Our results showed correlations between dialysate CCL8 levels and peritoneal transport parameters, with a weak correlation with serum CCL8. Based on previous studies, ³² , ³³ , ³⁴ , ³⁵ we propose that tissue‐resident macrophage‐derived CCL8 and CCL8 secreted from peritoneal mesothelial cells may play a major role in peritoneal fibrosis. The expression of F4/80, which is used to identify macrophages, and CCL8 in peritoneal tissues also supports the role of resident macrophages. Circulating monocyte‐derived CCL8 may play a role; however, tissue‐derived CCL8 may represent a greater proportion. Further investigation is required to determine the role of various types of resident macrophages and interactions between these cells and CCL8‐mediated fibrosis.

This study has several limitations. First, the effects of CCL8 blockade were investigated with R243, a small molecule inhibitor of CCR8, but not with an anti‐CCL8 mAb in vivo. Whether R243 blocks the CCL8‐CCR8 by inhibiting ligand binding to CCR8 on the cell surface or modulates CCR8 function by acting on downstream signaling molecules remains undetermined. However, the CG‐ and R243‐treated PF mouse model and TGF‐β‐induced in vitro fibrosis model clearly demonstrated the effects of CCL8 blockade on PF. Second, we did not examine the difference in CCL8 changes over time in our study, and it was impossible to confirm the long‐term association with the CCL8 expression level. Third, the detailed mechanisms underlying the association between CCL8 and inflammation, fibrosis, and apoptosis pathways were not explored during the progression of PF.

Our results demonstrate that high CCL8 levels in the PD effluent may be associated with an increased risk of PD failure and that the CCL8 pathway is associated with PF. CCL8 blockade can relieve peritoneal inflammation and fibrosis, and CCL8 represents a potential therapeutic target for preventing PF development.

AUTHOR CONTRIBUTIONS

Seung Hee Yang and Kwon Wook Joo designed the study. Dong Ki Kim, Kook‐Hwan Oh, Yon Su Kim, and Seung Hee Yang provided resources. Yeonhee Lee, Jangwook Lee, Minkyoung Park, Areum Seo, Kyu Hyeon Kim, and Seung Hee Yang were involved in the investigation and methodology. Yeonhee Lee and Seung Hee Yang performed data analysis and visualization. Seonmi Kim, Minjung Kang, Eunjeong Kang, Kyung Don Yoo, and Sunhwa Lee supervised and assisted in the analyses. Yeonhee Lee drafted the manuscript. Seung Hee Yang and Kwon Wook Joo reviewed the manuscript. All authors read and approved the final manuscript.

DISCLOSURES

The authors declare no competing interests.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available in the main article.