Human osteoarthritic cartilage shows reduced in vivo expression of IL-4, a chondroprotective cytokine that differentially modulates IL-1β-stimulated production of chemokines and matrix-degrading enzymes in vitro

Abstract

Background: In osteoarthritis (OA), an inflammatory environment is responsible for the imbalance between the anabolic and catabolic activity of chondrocytes and, thus, for articular cartilage derangement. This study was aimed at providing further insight into the impairment of the anabolic cytokine IL-4 and its receptors in human OA cartilage, as well as the potential ability of IL-4 to antagonize the catabolic phenotype induced by IL-1β.

Methodology/principal findings: The in vivo expression of IL-4 and IL-4 receptor subunits (IL-4R, IL-2Rγ, IL-13Rα1) was investigated on full thickness OA or normal knee cartilage. IL-4 expression was found to be significantly lower in OA, both in terms of the percentage of positive cells and the amount of signal per cell. IL-4 receptor type I and II were mostly expressed in mid-deep cartilage layers. No significant difference for each IL-4 receptor subunit was noted. IL-4 anti-inflammatory and anti-catabolic activity was assessed in vitro in the presence of IL-1β and/or IL-4 for 24 hours using differentiated high density primary OA chondrocyte also exhibiting the three IL-4 R subunits found in vivo. Chemokines, extracellular matrix degrading enzymes and their inhibitors were evaluated at mRNA (real time PCR) and protein (ELISA or western blot) levels. IL-4 did not affect IL-1β-induced mRNA expression of GRO-α/CXCL1, IL-8/CXCL8, ADAMTS-5, TIMP-1 or TIMP-3. Conversely, IL-4 significantly inhibited RANTES/CCL5, MIP-1α/CCL3, MIP-1β/CCL4, MMP-13 and ADAMTS-4. These results were confirmed at protein level for RANTES/CCL5 and MMP-13.

Conclusions/significance: Our results indicate for the first time that OA cartilage has a significantly lower expression of IL-4. Furthermore, we found differences in the spectrum of biological effects of IL-4. The findings that IL-4 has the ability to hamper the IL-1β-induced release of both MMP-13 and CCL5/RANTES, both markers of OA chondrocytes, strongly indicates IL-4 as a pivotal anabolic cytokine in cartilage whose impairment impacts on OA pathogenesis.

Figure 1. Analysis of IL-4 expression in normal and osteoarthritic cartilage.

A: representative examples of IL-4 staining in controls (NC), and OA cartilage as detected by confocal microscopy. Panels show IL-4 staining (Dy Light 647, rendered in red), nuclear counterstaining (sybr green, rendered in green) and merged signals (bar  = 10 µm). B: percentage of IL-4 positive cells in mid-deep layers of control (NC, n = 3), and OA cartilage (n = 6). Bars and whiskers indicate mean and S.E.M., respectively. Data were compared with unpaired data analysis using the Mann-Whitney U test (p = 0.0476). C: level of IL-4 expression (shown as mean intensity per positive cell area) in controls (NC), and OA cartilage following confocal analysis. Comparison was performed by the Mann-Whitney U test (p = 0.0357).

Figure 2. Analysis of IL-4R subunits in normal and osteoarthritic cartilage.

A and B: representative examples of IL-4R subunits staining in control (NC, A) and OA (B) cartilage (original magnification ×100) with details of superficial and mid-deep layer. Insets: higher magnification views (original magnification ×200). C: Percentage of positive cells expressing the three IL-4R subunits in controls (NC, white pattern, n = 5), and OA cartilage (dark pattern, n = 6), in the mid-deep layers. Comparison of NC and OA data was performed by the Mann-Whitney U test (p = ns). Boxes indicate the 25% and 75% percentiles, whiskers indicate the minimum to maximum values, and bars indicate the median.

Figure 3. Co-localization of IL-4R subunits expressed in normal and OA cartilage.

A: a representative example of IL-4Rα and IL-2Rγ colocalization (IL-4 Receptor type I) in control and OA cartilage. B: a representative example of IL-4Rα and IL-13Rα1 (IL-4 Receptor type II) in control and OA cartilage (bar  = 10 µm). Panels show IL-4R staining (Dy Light 647, rendered in red), either IL-2Rγ or IL-13Rα (Alexa 555, rendered in white), nuclear counterstaining (sybr green, rendered in green) and merged signals.

Figure 4. Assessment of the maintenance of a differentiated phenotype and evaluation of IL-4R subunits repertoire by high density cultures of primary chondrocytes.

A: SOX-9 protein expression in high density cultures compared to proliferating chondrocytes by western blot analysis. The housekeeping protein GAPDH is used as a loading control. A representative example out of the four examined. B: Cumulative evaluation of IL-4 receptor subunit gene expression in high-density chondrocytes in 5 different patients. Values represent the number of molecules per 100,000 GAPDH as assessed by real time PCR. Boxes indicate the 25% and 75% percentiles, whiskers indicate the minimum to maximum values, and bars indicate the median. C: flow cytometric analysis of the level of expression of each IL-4R subunit in high density culture OA chondrocytes. Dark pattern: isotype control; white pattern: Dy Light 647 signal referring to IL-4Rα (left), IL-2Rγ (middle) and IL-13Rα (right). Each curve refers to 5,000 cells. D representative example of IL-4Rα and IL-2Rγ colocalization (IL-4 Receptor type I) in high-density cultures of OA chondrocytes (bar  = 25 µm). On the right, higher magnifications showing IL-4R staining (Dy Light 647, rendered in red in d4), IL-2Rγ (Alexa 555, rendered in white in d3), nuclear counterstaining (sybr green, rendered in green in d2) and merged signals (d1).

Figure 5. Evaluation of IL-4 anti-inflammatory activity on CXC chemokine expression in high-density OA chondrocytes.

A: mRNA expression; B: protein production. Chemokines were detected in chondrocytes (cells and culture supernatants) after 24 hours of incubation in unstimulated condition (UNS) and in the presence of IL-1β (2 ng/ml), IL-4 (10 ng/ml), or a combination of the two cytokines. Boxes indicate the 25% and 75% percentiles, whiskers indicate the minimum to maximum values, bars indicate the median; solid circles indicate outliers. After Bonferroni's correction, p values below 0.0165 were considered significant.

Figure 6. Evaluation of IL-4 anti-inflammatory activity on CC chemokine expression in high-density OA chondrocytes.

A: mRNA expression; B: protein production. Chemokines were detected in chondrocytes (cells and culture supernatants) after 24 hours' incubation in unstimulated conditions (UNS) and in the presence of IL-1β (2 ng/ml), IL-4 (10 ng/ml), or a combination of the two cytokines. Boxes indicate the 25% and 75% percentiles, whiskers indicate the minimum to maximum values, bars indicate the median; solid circles indicate outliers. After Bonferroni's correction, p values below 0.0165 were considered significant.

Figure 7. Evaluation of IL-4 anti-inflammatory activity on extracellular matrix-degrading enzymes (MMP-13 and ADAMTS4) in high-density OA chondrocytes.

A: mRNA expression; B: protein production. MMP-13 was detected in chondrocytes (cells and culture supernatants) after 24 hours of incubation in unstimulated conditions (UNS) and in the presence of IL-1β (2 ng/ml), IL-4 (10 ng/ml), or a combination of the two cytokines. Whereas ADAMTS4 mRNA was detected as for the other transcripts, ADAMTS4 protein was assessed in lysates of cells scraped with their extracellular matrix. WB show the band at 64 kDA along with β-tubulin as a loading control. Boxes indicate the 25% and 75% percentiles, whiskers indicate the minimum to maximum values, bars indicate the median; and solid circles indicate outliers. After Bonferroni's correction, p values below 0.0165 were considered significant.