Intracellular accumulation of advanced glycation end products induces apoptosis via endoplasmic reticulum stress in chondrocytes

Authors


Correspondence

J. Hirose, Department of Orthopaedic Surgery, Kumamoto University Hospital, 1-1-1 Honjo, Chuoku, Kumamoto 860-8556, Japan

Fax: +81 96 373 5228

Tel: +81 96 373 5226

E-mail: hirojun-mk@umin.ac.jp

Abstract

Mammalian cells attempt to maintain their homeostasis under endoplasmic reticulum (ER) stress. If the stress cannot be alleviated, cells are led to apoptosis through induction of C/EBP homologous protein (CHOP). ER stress is provoked in osteoarthritis chondrocytes, and intracellular accumulation of advanced glycation end products (AGEs) in chondrocytes is a possible cause. To clarify the role of intracellular AGE accumulation in chondrocytes, the present study investigated the effect of intracellular AGE accumulation on ER stress and apoptosis by in vitro and in vivo analysis. Intracellular AGE accumulation induced by AGE precursors caused apoptosis, induced expression of ER stress markers, and led to co-localization of AGEs with glucose-regulated protein 78, leading to formation of high-molecular-weight complexes in cultured chondrocytes. These reactions were inhibited by an AGE formation inhibitor. CHOP deletion inhibited apoptosis induced by intracellular AGE accumulation. In vivo intracellular AGE accumulation induced by intra-articular injection of AGE precursors caused ER stress and apoptosis in chondrocytes and led to degradation of articular cartilage. Additionally, intracellular AGE accumulation increased the degree of cartilage degradation in an osteoarthritis model. These data indicate that intracellular accumulation of AGEs induces modification of unfolded protein response-related protein by AGEs and apoptosis via ER stress in chondrocytes. Moreover, the in vivo study showed that intracellular AGE accumulation in chondrocytes is involved in the occurrence and progression of osteoarthritis through ER stress. Thus, research on mechanisms of apoptosis via ER stress induced by intracellular AGE accumulation in chondrocytes will lead to a new understanding of osteoarthritis pathology.

Abbreviations
AGE

advanced glycation end product

AMG

aminoguanidine

CHOP

C/EBP homologous protein

ER

endoplasmic reticulum

GA

glycolaldehyde

Grp78

glucose-regulated protein 78

MTT

3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide

OA

osteoarthritis

PI

propidium iodide

TUNEL

terminal deoxynucleotidyl transferase dUTP nick end labeling

UPR

unfolded protein response

XBP1

Xbox-binding protein 1

Xbp1s

spliced form of Xbox-binding protein 1 mRNA

Xbp1u

unspliced form of Xbox-binding protein 1 mRNA

Introduction

Osteoarthritis (OA) is a disease characterized by degeneration of joint components, including articular cartilage. The number of apoptotic chondrocytes is significantly correlated with the OA grade [1], and inhibition of the pro-apoptotic caspase signaling pathway suppressed progression of OA [2], suggesting that the pathology of OA is closely related to chondrocyte apoptosis [3]. Several pathways induce apoptosis in chondrocytes, including the mitochondrial pathway, the death receptor pathway, and the endoplasmic reticulum (ER) stress pathway [4]. Among these pathways, the ER stress pathway has attracted attention over recent years.

ER stress is defined as an accumulation of unfolded proteins in the lumen of the ER [5]. The unfolded protein response (UPR), which includes signaling pathways assisted by ER chaperone glucose-regulated protein 78 (GRP78), is activated to relieve stress. However, if stress cannot be resolved, signaling in the cell switches from pro-survival to pro-apoptotic, and apoptosis occurs through induction of transcription factor C/EBP homologous protein (CHOP) [6]. ER stress-induced apoptosis is probably involved in a number of diseases, including neurodegenerative disease, diabetes, atherosclerosis and renal disease [7]. A recent report showed that chondrocytes in human OA cartilage displayed ER stress, and that ER stress and apoptosis were increased in cartilage during progression of OA [8]. These data suggest that apoptosis through ER stress is related to OA pathology.

Aging is an important risk factor for OA [9]. Age-associated accumulation of advanced glycation end products (AGEs) in articular cartilage contributes to the pathology of OA [10-12]. AGEs are compounds formed by the non-enzymatic reaction of reducing sugars with protein, and they accumulate in various tissues within the living body. In articular cartilage, accumulation of AGEs leads to increased susceptibility to OA [13], possibly through an increase in matrix brittleness induced by extracellular matrix AGEs [14], reduction of matrix turnover [15-17], and induction of inflammation or matrix metalloproteinases via the receptor for AGEs [18-20].

Recent research has revealed that AGEs also accumulate intracellularly, specifically in OA chondrocytes [21, 22], but the role of intracellular accumulation of AGEs in these cells remains unclear. Studies have shown the induction of apoptosis with accumulation of intracellular AGEs in mouse podocytes [23] and human synovial cells [24]. Accumulation of intracellular AGEs modified molecular chaperones in liver carcinoma cells through advanced glycation, leading to their dysfunction [25]. Therefore, intracellular accumulation of AGEs may induce ER stress via UPR dysfunction. Thus, we hypothesized that intracellular accumulation of AGEs induces apoptosis through ER stress in chondrocytes.

The purpose of this study was to clarify the role of intracellular AGE accumulation in apoptosis and ER stress in chondrocytes. We administered AGE precursors to Chop+/+ and Chop−/− chondrocytes and studied the relationship between intracellular accumulation of AGEs and apoptosis. In addition, we investigated in vitro mechanisms of induction of apoptosis by accumulation of intracellular AGEs, and the in vivo effect of accumulation of intracellular AGEs on cartilage degradation in chondrocytes.

Results

Evaluation of intracellular AGE accumulation-induced apoptosis in cultured chondrocytes

To assess the effects of incubation period with glycolaldehyde (GA), chondrocytes were incubated with 500 μm GA for 12, 24 and 48 h. Incubation with GA for 24 and 48 h caused intracellular accumulation of AGEs in chondrocytes (Fig. 1A). The 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) assay revealed that viability of chondrocytes was significantly reduced to 91.4% by 12 h incubation with GA (< 0.05), to 76.0% by 24 h incubation (< 0.001), and to 73.9% by 48 h incubation (< 0.001) (Fig. 1B). TUNEL staining (Fig. 1C) and cleaved caspase-3 levels (Fig. 1D) were increased by GA treatment.

Figure 1.

Evaluation of intracellular AGE accumulation, cell death and apoptosis after various incubation periods with GA in chondrocytes. Chondrocytes were incubated with 500 μm GA. (A) AGE accumulation was analyzed by immunocytochemistry. AGEs show green fluorescence (Alexa Fluor® 488), and nuclei appear red (PI). (B) The viability of chondrocytes was determined by the MTT assay. Values are normalized to the control (0 h) and expressed as means ± SD. Asterisks indicate statistically significant differences versus control (0 h) (*< 0.05, **< 0.001, one-way ANOVA with post hoc Bonferroni correction). (C) Fragmentation of DNA was analyzed by TUNEL staining. TUNEL fluorescence is green, and nuclei appear red (PI). (D) Cleaved caspase-3 was analyzed by immunocytochemistry. Fluorescence of cleaved caspase-3 appears green (Alexa Fluor® 488), and nuclei appear red (PI). Scale bars = 100 μm. Representative images are shown.

To estimate the effect of GA concentration, chondrocytes were incubated with 0, 50, 100, 500 or 1000 μm GA for 24 h. AGEs were detected in chondrocytes after incubation with 500 μm GA, with a further increase in AGE staining after incubation with 1000 μm GA (Fig. 2A). The viability of chondrocytes was significantly reduced to 57.0% by incubation with 500 μm GA and to 32.4% by incubation with 1000 μm GA (< 0.001, Fig. 2B). Total chondrocyte number was reduced to 32.7% by incubation with 500 μm GA, and to 14.4% by incubation with 1000 μm GA (< 0.001, Fig. 2C). DNA fragmentation increased to 334.6% by incubation with 500 μm GA (< 0.001) and to 257.1% by incubation with 1000 μm GA (< 0.05, Fig. 2D). TUNEL staining (Fig. 2E) and cleaved caspase-3 levels (Fig. 2F) were increased by treatment with both 500 and 1000 μm GA.

Figure 2.

Evaluation of intracellular AGE accumulation, cell death and apoptosis after incubation with various concentrations of GA in chondrocytes. Chondrocytes were incubated with GA for 24 h. (A) AGE accumulation was analyzed by immunocytochemistry. AGEs show green fluorescence (Alexa Fluor® 488), and nuclei appear red (PI). (B) The viability of chondrocytes was determined by the MTT assay. Values are normalized to the control (GA, 0 μm) and expressed as means ± SD. (C) Total chondrocyte numbers were calculated using a Countess™ automated cell counter. Values are normalized to the control (GA, 0 μm) and expressed as means ± SD. (D) Fragmentation of DNA was assessed quantitatively by ELISA. Values are normalized to the control (GA, 0 μm) and expressed as means ± SD. (E) Fragmentation of DNA was analyzed by TUNEL staining. (F) Cleaved caspase-3 was analyzed by immunocytochemistry. Fluorescence of cleaved caspase-3 shows green (Alexa Fluor® 488), and nuclei appear red (PI). (G) AGE content in culture medium was measured quantitatively by ELISA. Chondrocytes were incubated with 0–3000 mm GA for 24 h. Values are expressed as means ± SD. Scale bars = 100 μm. Representative images are shown. Asterisks indicate statistically significant differences versus control (GA, 0 μm) (*< 0.05, **< 0.001, one-way ANOVA with post hoc Bonferroni correction).

To detect extracellular AGEs, the concentration of AGEs in the culture medium was measured. Chondrocytes were treated with 500, 1000, 2000 or 3000 μm GA for 24 h, and AGEs were detected in culture medium after incubation with 2000 or 3000 μm GA (Fig. 2G).

Evaluation of effects on apoptosis by inhibition of AGE formation

AGE formation was inhibited by aminoguanidine (AMG) and the effect on apoptosis was analyzed. Pre-incubation with AMG at concentrations > 1 mm for 2 h prior to 24 h incubation with GA inhibited intracellular AGE accumulation in chondrocytes (Fig. 3A). Pre-incubation with 1 mm AMG increased chondrocyte viability to 125.0% (< 0.05, Fig. 3B), with incremental increases in chondrocyte viability caused by treatment with AMG at concentrations > 1 mm (< 0.05). Total chondrocyte number was increased by pre-incubation with 1 mm AMG or more (< 0.05, Fig. 3C). DNA fragmentation was inhibited to 10.0% by pre-incubation with 25 mm AMG (< 0.001), and to 36.2% by pre-incubation with 50 mm AMG (< 0.05, Fig. 3D). TUNEL staining (Fig. 3E) and cleaved caspase-3 levels (Fig. 3F) were inhibited by pre-incubation with AMG in a concentration-dependent manner.

Figure 3.

Effect of inhibition of AGE formation by AMG on intracellular AGE accumulation, cell death and apoptosis in chondrocytes. Chondrocytes were pre-incubated with AMG for 2 h prior to incubation with 500 μm GA for 24 h. (A) AGE accumulation was analyzed by immunocytochemistry. Fluorescence of AGEs appears green (Alexa Fluor® 488), and nuclei appear red (PI). (B) Chondrocyte viability was determined by the MTT assay. Values are normalized to the control (AMG, 0 mm) and expressed as means ± SD. (C) Total chondrocyte numbers were calculated using a Countess™ automated cell counter. Values are normalized to the control (AMG, 0 mm) and expressed as means ± SD. (D) Fragmentation of DNA was assessed quantitatively by ELISA. Values are normalized to the control (AMG, 0 mm) and expressed as means ± SD. (E) Fragmentation of DNA was analyzed by TUNEL staining. Fluorescence of TUNEL appears green, and nuclei appear red (PI). (F) Cleaved caspase-3 was analyzed by immunocytochemistry. Cleaved caspase-3 shows green fluorescence (Alexa Fluor® 488), and nuclei appear red (PI). Scale bars = 100 μm. Representative images are shown. Asterisks indicate statistically significant differences versus control (AMG, 0 mm) (*< 0.05, **< 0.001, one-way ANOVA with post hoc Bonferroni correction).

Evaluation of ER stress

ER stress was evaluated on the basis of expression of the spliced form of Xbox-binding protein 1 (Xbp1s) mRNA and Chop mRNA. Incubation with GA for 24 h increased the expression of Xbp1s mRNA (Fig. 4A) and Chop mRNA (Fig. 4B). Expression of Chop mRNA was increased to 224.5% by incubation with 500 μm GA and to 501.5% by incubation with 1000 μm GA (< 0.001). Pre-incubation with 25 or 50 mm AMG prior to incubation with 500 mm GA inhibited the expression of Xbp1s mRNA (Fig. 4C) and Chop mRNA (Fig. 4D). Expression of Chop mRNA was inhibited to 60.6% by pre-incubation with 25 mm AMG (< 0.05) and to 41.6% by pre-incubation with 50 mm AMG (< 0.001).

Figure 4.

Induction of ER stress by intracellular AGE accumulation in chondrocytes. Chondrocytes were incubated with 0–1000 μm GA for 24 h (A, B), or pre-incubated with 0–50 mm AMG for 2 h prior to incubation with 500 μm GA for 24 h (C, D). (A, C) Expression of Xbp1 mRNA was analyzed by RT-PCR using the restriction enzyme PstI. The upper panels show expression of total Xbp1 mRNA, and the lower panels show Xbp1s and Xbp1u mRNA after treatment of the PCR products with the restriction enzyme PstI. Representative image are shown. (B,D) The Chop/Actb mRNA expression ratio was analyzed by real-time RT-PCR. Data are normalized to each control (B: GA, 0 μm; D: AMG, 0 mm) and expressed as means ± SD. Asterisks indicate statistically significant differences versus each control (*< 0.05, **< 0.001, one-way ANOVA with post-hoc Bonferroni correction).

Analysis of Chop−/− chondrocytes

To directly analyze the function of Chop, Chop+/+ chondrocytes and Chop−/− chondrocytes were incubated with 500 μm GA for 24 h. Accumulation of intracellular AGEs (Fig. 5A) and expression of Xbp1s mRNA (Fig. 5B) were increased by incubation with GA in both Chop+/+ chondrocytes and Chop−/− chondrocytes, and there was no difference between Chop+/+ chondrocytes and Chop−/− chondrocytes. Chondrocyte viability decreased to 23.9% in Chop+/+ chondrocytes and to 53.1% in Chop−/− chondrocytes after incubation with GA (< 0.001, Fig. 5C). The viability of Chop−/− chondrocytes after incubation with GA was significantly higher than that of Chop+/+ chondrocytes (< 0.001). DNA fragmentation was increased to 184.3% in Chop+/+ chondrocytes and to 137.3% in Chop−/− chondrocytes by incubation with GA (< 0.001, Fig. 5D). The DNA fragmentation in Chop−/− chondrocytes after incubation with GA was significantly lower than that in Chop+/+ chondrocytes (< 0.05).

Figure 5.

Effect of CHOP deletion on apoptosis resulting from intracellular AGE accumulation via ER stress. Chop+/+ and Chop−/− chondrocytes were incubated with 500 μm GA for 24 h. (A) AGE accumulation was analyzed by immunocytochemistry. Green fluorescence (Alexa Fluor® 488) indicates AGEs, and red fluorescence (PI) indicates nuclei. (B) Expression of Xbp1 mRNA was analyzed by RT-PCR using the restriction enzyme PstI. The upper panel shows total Xbp1 mRNA and the lower panel shows Xbp1s and Xbp1u mRNA. (C) The viability of chondrocytes was determined by the MTT assay. Data are normalized to the control (incubated without GA) and expressed as means ± SD. (D) Fragmentation of DNA was assessed quantitatively by ELISA. Data are normalized to the control (incubated without GA) and expressed as means ± SD. Scale bar = 100 μm. Representative images are shown. Open bars represent Chop+/+ chondrocytes and black bars represent Chop−/− chondrocytes. Asterisks indicate statistically significant differences (*< 0.05, **< 0.001 versus each control; †< 0.05, ††< 0.001 versus Chop+/+ chondrocytes, Student's t test).

Modification of UPR-related molecules by AGEs

AGE modification of Grp78, UPR-related protein, was determined by a double immunostaining method and western blot analysis. Grp78 co-localized with AGEs in response to 500 μm GA , and the formation of AGEs was inhibited by pre-treatment with 25 mm AMG (Fig. 6A). Treatment with 500 μm GA generated complexes of various sizes containing intracellular AGEs, as determined by western blot analysis of chondrocyte lysates, and the immunoreactivity of AGEs was inhibited by pre-treatment of chondrocytes with AMG (Fig. 6B). We also observed that Grp78 moved from a 78 kDa band to a high-molecular-weight complex by treatment with GA, and this change was also inhibited by pre-treatment with AMG.

Figure 6.

AGE modification of Grp78 in chondrocytes. Chondrocytes were incubated with 500 μm GA for 24 h, or pre-incubated with 25 mm AMG for 2 h prior to incubation with 500 μm GA for 24 h. (A) Co-localization of Grp78 with AGEs was analyzed by double immunostaining. Grp78 is visualized by red fluorescence and AGEs are visualized by green fluorescence. Representative images are shown. Scale bar = 20 μm. (B) Identification of AGEs and Grp78. Cell lysates (35 μg per lane) were loaded onto polyacrylamide gels, and expression of AGEs and Grp78 was determined by western blot analysis.

In vivo evaluation of intracellular AGE accumulation in chondrocytes, and during ER stress, apoptosis and degradation of articular cartilage

To establish whether AGEs accumulate in chondrocytes in vivo, GA was injected intra-articularly into the knee joint of mice. Intracellular accumulation of AGEs in chondrocytes was observed 1 week after injection of 500 or 1000 mm GA (Fig. 7). Chop-positive and TUNEL-positive chondrocytes were also observed after injection of 500 or 1000 mm GA.

Figure 7.

Induction of intracellular AGE accumulation, ER stress and apoptosis by intra-articular GA injection in chondrocytes. GA was administrated to mice knee joints intra-articularly, and immunostaining for AGEs, immunostaining for CHOP, and TUNEL staining were performed 1 week after injection. Fluorescence of AGEs, CHOP and TUNEL appears green, and fluorescence of nuclei appears red (PI). Representative images are shown.

To analyze whether intracellular AGE accumulation in chondrocytes is related to the occurrence of OA, cartilage was evaluated after injection (Fig. 8A). Intracellular AGE accumulation in chondrocytes and expression of Chop-positive chondrocytes was observed 1 week after GA injection, but Mankin's score for the joint injected with GA was 0.00 at this time point. Intracellular AGE accumulation in chondrocytes and expression of Chop-positive chondrocytes was also observed 5 weeks after GA injection, but toluidine blue staining was reduced and the score increased to 0.86 at this time point (< 0.05, Fig. 8B).

Figure 8.

Occurrence and progression of cartilage degradation by intra-articular GA injection. Joint instability was surgically induced in mice knee joints by the medial model 1 week after intra-articular injection of 500 mm GA. At 1 and 5 weeks after injection, toluidine blue staining and immunostaining for AGEs and CHOP were performed. (A) Representative images are shown. (B) Cartilage degradation was evaluated quantitatively by modified Mankin's scoring system (= 6). Values are means ± SD. Open bars represent control (NaCl/Pi-injected) knee joints, and black bars represent GA-injected knee joints. Asterisks indicate statistically significant differences (*< 0.05 versus 1 week after injection, †< 0.05 versus control, ‡< 0.001 versus 5 weeks without operation (medial model –), Student's t test).

To examine whether intracellular AGE accumulation in chondrocytes is involved in progression of OA, an OA model was surgically created 1 week after injection, and cartilage was evaluated 4 weeks post-surgery (Fig. 8A). No intracellular AGE accumulation was observed in the operated control joint, but Chop-positive chondrocytes were present. In the GA-injected and operated joint, intracellular AGE accumulation and Chop-positive chondrocytes were observed, but at a lower level than in the GA-injected and unoperated joint. OA changes were observed in both the operated control joint and the GA-injected and operated joint, Mankin's score for the control joint increased from 0.00 to 4.20 after the operation (< 0.001, Fig. 8B), and the score for the joint injected with GA increased from 0.86 to 6.33 after the operation (< 0.001). The score for the GA-injected and operated joint was significantly higher than that of the operated control joint (< 0.05).

To further elucidate the effect of intracellular AGE accumulation in chondrocytes on cartilage degradation via ER stress, we created a surgical OA model in Chop−/− mice after intra-articular GA injection, and evaluated cartilage degradation. As in Chop+/+ mice, there was no cartilage degradation 1 week after NaCl/Pi or GA injection (Fig. 9A,B). However, 4 weeks after the operation, Mankin's score for the control joint increased to 2.8, and the score for the GA-injected joint increased to 3.5 in Chop−/− mice. The scores for operated Chop−/− mice were significantly lower than the scores for operated Chop+/+ mice for each injection (< 0.05, Fig. 9C,D).

Figure 9.

Effect of CHOP deletion on cartilage degradation by intra-articular GA injection. Joint instability was surgically induced in Chop+/+ and Chop−/− mice knee joints by the medial model 1 week after intra-articular injection of 500 mm GA. At 1 week (A, B) and 5 weeks (C, D) after injection, degradation of cartilage was analyzed by toluidine blue staining. (A, C) Representative images are shown. (B, D) Cartilage degradation was evaluated quantitatively by a modified Mankin's scoring system (= 6). Values are means ± SD. Open bars represent control (NaCl/Pi-injected) knee joints, and black bars represent GA-injected knee joints. Asterisks indicate statistically significant differences (*< 0.05 versus control, < 0.001 versus Chop+/+ mice, Student's t test).

Discussion

This study investigated the effect of intracellular AGE accumulation on ER stress and apoptosis in chondrocytes. GA-derived accumulation of intracellular AGEs induced apoptosis and expression of ER stress makers in cultured chondrocytes, and this reaction was inhibited by AMG. CHOP deletion inhibited apoptosis induced by accumulation of intracellular AGEs. The UPR-related protein Grp78 co-localized with AGEs, and formed a high-molecular-weight complex as the intracellular AGEs accumulated. In vivo accumulation of intracellular AGEs induced ER stress and apoptosis in chondrocytes, led to degradation of articular cartilage, and was involved in the development of OA.

In the present study, we used GA as an AGE precursor to accumulate AGE intracellularly and to demonstrate that accumulation of intracellular AGEs induces apoptosis in chondrocytes. It has been shown that GA carbonylate rat liver cell or human breast carcinoma cell, thereby inducing apoptosis [26, 27]. Other studies have shown that other AGE precursors such as methylglyoxal or glyceraldehyde also cause apoptosis [28, 29]. GA is generated by neutrophils through the myeloperoxidase system [30] and forms Nε-(carboxymethyl)lysine and a GA-specific AGE, namely GA-pyridine [31]. Because of the relationship between accumulation of GA-pyridine in chondrocytes and the severity of OA [22], we consider GA as the appropriate AGE precursor for this study. Administration of GA caused intracellular AGE accumulation and induction of apoptosis in chondrocytes, and inhibition of AGE formation by AMG suppressed apoptosis. Based on this observation, we showed, for the first time, that accumulation of intracellular AGEs induces apoptosis in chondrocytes. However, as the mechanisms of apoptosis induction were still unclear, we then focused on ER stress.

The transcription factor Xbox binding protein 1 (XBP1) is a key regulatory factor of ER stress, and is widely used as an ER stress marker [32]. Expression of Xbp1s mRNA increased with the accumulation of intracellular AGEs in chondrocytes, and this increase was inhibited by inhibition of AGE formation. These results indicate that accumulation of intracellular AGEs induces ER stress. The transcription factor CHOP is also an ER stress marker, and expression of Chop mRNA showed an increase similar to that of Xbp1s mRNA in this study. In addition, CHOP is an important apoptosis-inducing factor during ER stress [7]. ER stress-induced apoptosis was provoked by over-expression of Chop in a rat fibroblast cell line [33], and was inhibited by deletion of Chop in a mouse embryonic fibroblast cell line [34]. We demonstrated that accumulation of intracellular AGEs in chondrocytes induced apoptosis via ER stress, because accumulation of intracellular AGEs increased expression of Chop mRNA and apoptosis, and CHOP deletion inhibited apoptosis.

Together with the accumulation of AGEs in chondrocytes, the ER chaperone Grp78 was found to co-localize with AGEs, and, rather than as a 78 kDa band, Grp78 was present in a high-molecular-weight complex induced by cross-linking and aggregation, an effect that is characteristic of AGEs [25]. These data indicate that Grp78 is modified by AGEs. Studies have shown that the function of chaperone proteins in liver carcinoma cells was impaired by AGE modification [25]. Therefore, we surmise that the dysfunction of Grp78 by AGE modification may be an inducer of ER stress.

Intra-articular injection of GA in the mouse knee joint induced AGE accumulation, ER stress and apoptosis in chondrocytes 1 week after injection in in vivo experiments, but no apparent degradation of the cartilage matrix was observed at this time. However, we did observe a slight degradation of the cartilage matrix 5 weeks after GA injection. This result suggests that apoptosis resulting from the accumulation of intracellular AGEs via ER stress may cause degradation of the cartilage matrix. Additionally, injection of GA predisposed the joint to OA in a joint instability model. Intracellular AGE accumulation and Chop expression in GA-injected and operated joints were weaker than that in GA-injected and unoperated joints, possibly due to a decrease in chondrocytes induced by the surgical OA model. DeGroot et al. [13] reported that injection of 350 mm ribose into the canine knee joint twice weekly for 7 weeks predisposed the joint to OA after transection of the anterior cruciate ligament. Our data are consistent with the results of a study by DeGroot et al. in this respect, but the authors did not observe degradation of the cartilage matrix in a joint in which only ribose was injected (without instability). It was suggested that, due to the strong interaction with the receptor for AGEs and the subsequent biological effects, AGEs derived from reactive carbonyl GA play a more important role in pathophysiological processes than AGEs derived from reducing sugars such as ribose [35]. Although mechanical stress is involved in a large part of OA pathology, this study suggests that intracellular AGE accumulation in chondrocytes plays a role in the occurrence and progression of OA. This study has two limitations. First, not all types of AGEs were evaluated. Many researchers have reported that there are many types of AGEs in a living organism. To the best of our knowledge, the AGE with the highest concentration in articular cartilage is Nε-(carboxymethyl)lysine; the ratio of Nε-(carboxymethyl)lysine to lysine residues was approximately 0–8.0 mmol·mol−1 [11]. The concentration of Nε-(carboxyethyl)lysine was approximately 0.1–1.25 mmol·mol−1, and that of pentosidine was approximately 0–0.18 mmol·mol−1 in articular cartilage. The anti-AGE antibody 6D12 used in this study reacted with not only Nε-(carboxymethyl)lysine [36] but also Nε-(carboxyethyl)lysine [37] and GA-pyridine [31]. Hence, 6D12 is useful for the evaluation of major AGEs in articular cartilage. The second limitation of the study is the existence of apoptotic pathways other than the ER stress/CHOP pathway. The induction of apoptosis by AGE accumulation in chondrocytes was not completely inhibited in Chop−/− chondrocytes, indicating the existence of pathways other than the CHOP pathway. Other possible signaling pathways include c-Jun N-terminal kinases or reactive oxygen species, but further investigation is required.

In conclusion, the present study demonstrated that intracellular accumulation of AGEs in chondrocytes induced apoptosis via ER stress. Additionally, an in vivo study showed that intracellular accumulation of AGEs in chondrocytes occurred and caused progressive cartilage degradation, and ER stress-induced apoptosis may play an important role in this mechanism. AGEs accumulate in chondrocytes with aging [10-12], and impairment of UPR with aging induces apoptosis signals via ER stress [38, 39]. Thus, apoptosis via ER stress induced by intracellular AGE accumulation in chondrocytes may be a new mechanism for understanding the age-associated occurrence and progression of OA. The research on prevention and treatment of OA, the most common form of joint disease, is an important issue, the mechanism demonstrated in this study may be a new therapeutic target for OA.

Experimental procedures

Reagents

Dulbecco's modified Eagle medium, fetal bovine serum, phosphate-buffered saline (NaCl/Pi) and propidium iodide (PI) were purchased from Gibco/Life Technologies Corporation (Grand Island, NY, USA). EDTA, ethidium bromide, glycerol, RIPA buffer kit (50 mm Tris/HCl pH 7.6, 150 mm NaCl, 1% Nonidet P-40 (NP-40), 0.5% sodium deoxy cholate, 0.1% SDS, protease inhibitor cocktail), SDS and 2-mercaptoethanol were purchased from Nacalai Tesque Inc. (Kyoto, Japan). AMG and GA were obtained from Sigma-Aldrich (St Louis, MO). Agarose, paraformaldehyde, PstI and toluidine blue were obtained from Wako Pure Chemical Industries Ltd (Osaka, Japan).

Animals

C57BL/6J mice were obtained from Japan SLC Inc. (Shizuoka, Japan). Chop−/− mice in the C57BL/6 background [40] were kindly provided by the Department of Molecular Genetics, Graduate School of Medical Sciences, Kumamoto University (Kumamoto, Japan). Mice were fed a standard diet and housed under a 12 h light/dark cycle. All procedures were approved by the Animal Care and Use Committee of Kumamoto University.

Mice chondrocyte culture

Primary costal chondrocytes were isolated by digestion of rib cartilage from 1-day-old mice and cultured in Dulbecco's modified Eagle medium with 10% fetal bovine serum under standard cell culture conditions (humidified atmosphere, 5% CO2, 37 °C) as described previously [41]. Treatments with the AGE precursor GA and the inhibitor of AGE formation AMG were performed in 2% fetal bovine serum/Dulbecco's modified Eagle medium after chondrocytes reached 70% confluence. The chondrocytes were incubated with 0–3000 μm GA for 12–48 h. To investigate the specific effects of AGEs, chondrocytes were pre-incubated for 2 h with 0–50 mm AMG before incubation with 500 μm GA for 24 h [25]. All experiments were performed in triplicate, and the results were confirmed in at least three independent experiments.

Induction of AGE accumulation and mechanical instability in the mouse knee joint

For induction of intracellular AGE accumulation in chondrocytes, 72 male 8-week-old C57BL/6J mice and 24 Chop−/− mice were anesthetized by intraperitoneal injection of pentobarbital (Somnopentyl; Kyoritsu Shoji, Tokyo, Japan), prior to injection of 20 μL NaCl/Pi containing 0, 100, 500 or 1000 μm GA into the right knee joint space through the patellar tendon using a 29 gauge needle. Mice were killed 1 or 5 weeks after injection, and the right knee joint was harvested. For surgical induction of mechanical instability, a medial model was created on the mouse right knee joint, as previously described [42], 1 week after injection of NaCl/Pi with or without 500 μm GA into the right knee joint. In brief, after mice were anesthetized, a medial parapatellar skin incision was made. The medial compartment of the knee joint was visualized and the medial collateral ligament and medial meniscus were transected. Mice were killed 4 weeks after operation, and the right knee joint was harvested. All the harvested tissues were fixed in 4% paraformaldehyde at 4 °C for 16 h, and decalcified in 20% EDTA at 4 °C for 1 week. After embedding in paraffin, 4 μm sections were cut and used for toluidine blue staining, immunostaining for AGEs or CHOP, and TUNEL staining. Six mice were used in each experimental group.

MTT assay

Viability of cultured chondrocytes was evaluated by an MTT colorimetric assay using an MTT cell count kit (Nacalai Tesque Inc.), according to the manufacturer's instructions. The absorbance of samples was read using an iMark™ microplate reader (Bio-Rad Laboratories, Hercules, CA, USA).

Evaluation of chondrocyte number

Total chondrocyte numbers were calculated using a Countess™ automated cell counter (Invitrogen/Life Technologies Corporation, Carlsbad, CA, USA), according to the manufacturer's manual.

ELISA for the detection of apoptosis

The extent of apoptosis was determined using a Cell Death Detection ELISAPlus kit (Roche Diagnostics GmbH, Mannheim, Germany), which measures DNA fragments. All procedures were performed according to the manufacturers' instructions, and the absorbance of samples was read using an iMark™ microplate reader. The results were normalized by total protein content of the well, measured by the Lowry protein assay [43]. The enrichment factor was calculated by dividing the absorbance of the sample by the absorbance of the controls without treatment.

TUNEL staining

For detection of apoptotic chondrocyte death at the single-cell level, we used an In Situ Cell Death Detection Kit, AP (Roche Diagnostics). Briefly, cultured chondrocytes were fixed with 4% paraformaldehyde for 1 h at 4 °C. Paraffin sections (4 μm thick) of the knee joint were deparaffinized using xylene. Thereafter, samples were permeabilized using 0.1% sodium citrate buffer containing 0.1% Triton X-100 (Nacalai Tesque Inc.) for 2 min on ice, incubated at 37 °C for 1 h with TUNEL reaction buffer, and counterstained with PI. Apoptotic chondrocytes (green signal) and nuclei (red signal) were visualized using an FV300 scanning confocal microscope system (Olympus, Tokyo, Japan). Samples incubated without terminal deoxynucleotidyl transferase served as negative control samples.

Immunochemical analysis of AGEs

To stain for AGEs immunochemically, mouse monoclonal antibody against AGE (6D12; Transgenic Inc., Kumamoto, Japan) was used. Chondrocytes fixed using 4% paraformaldehyde and paraffin sections deparaffinized using xylene were blocked with 10% normal goat serum (Nichirei Bioscience, Tokyo, Japan), and stained with fluorescently labeled 6D12 using a Zenon Alexa Fluor® 488 mouse IgG1 labeling kit (Invitrogen/Life Technologies Corporation) according to the manufacturer's instructions. Cells were counter-stained with PI and were visualized using a scanning confocal microscope system. The samples incubated without 6D12 served as negative control samples.

Measurement of AGE contents in medium

To measure extracellular AGE contents, culture media were centrifuged at 1500 g for 10 min, and AGE contents of supernatants were analyzed using a glucose-derived AGEs ELISA kit (Transgenic Inc.). The assay was performed according to the manufacturer's instructions, and the absorbance of samples was read using an iMark™ microplate reader.

Immunochemical analysis of cleaved caspase-3 and Chop

For detection of cleaved caspase-3 and Chop, the cultured chondrocytes were fixed with 4% paraformaldehyde and 4 μm thick paraffin sections were deparaffinized using xylene. Samples were treated with proteinase (Nichirei Bioscience) for 3 min on ice. Then the samples were incubated first with 10% normal goat serum, and then with rabbit anti-cleaved caspase-3 (product number #9661; Cell Signaling Technology, Beverly, MA, USA) or rabbit anti-CHOP (SC-575; Santa Cruz Biotechnology, Santa Cruz, CA, USA). For visualization, secondary antibodies conjugated to Alexa Fluor® 488 goat anti-rabbit IgG (Invitrogen) were used. Nuclei were counterstained using PI and scanning confocal microscopy was performed. Samples incubated without the first antibody served as negative control samples.

Double immunostaining

To detect co-localization of Grp78 or proteasomes with AGEs, double immunostaining was performed. Cultured chondrocytes were fixed with 4% paraformaldehyde and permeabilized using proteinase K (Dako, Carpinteria, CA, USA) before blocking with 5% skim milk. Cells were then incubated for 1 h at room temperature with antibody against AGE, which was labeled with Zenon Alexa Fluor® 488 mouse IgG1 labeling kit. Samples were incubated for 1 h at room temperature with antibody against Grp78 (SPA-862D; Enzo Life Sciences, Farmingdale, NY, USA), and then incubated with Alexa Fluor® 594 goat anti-rabbit IgG (Invitrogen/Life Technologies Corporation) for 1 h at room temperature and visualized by confocal microscopy.

Western blot analysis

The chondrocytes were washed with ice-cold NaCl/Pi and lysed in RIPA buffer. Protein concentrations were measured using the Lowry protein assay [43]. Chondrocyte lysates (35 μg of protein/lane) dissolved in 2× sample buffer (50 mm Tris/HCl pH 6.8, 4% SDS, 20% glycerol, 0.02% bromophenol blue, 5% 2-mercaptoethanol) were boiled for 5 min at 95 °C and separated by SDS/PAGE, and were then electrotransferred onto Immobilon transfer membrane (Millipore, Billerica, MA, USA). Protein Ladder One (Nacalai Tesque Inc.) was used as a molecular marker. Next, the membranes were blocked for 30 min at room temperature with 5% skim milk. After washing the membranes for 5 min at room temperature three times with Tris-buffered saline containing Tween-20, they were incubated with antibody against AGE (6D12) or antibody against Grp78 (SC-166490; Santa Cruz Biotechnology) at room temperature for 1 h. Subsequently, the membranes were washed three times with Tris-buffered saline containing Tween-20 and incubated with horseradish peroxidase-conjugated goat antibody against mouse IgG (SC-2005; Santa Cruz Biotechnology) for 1 h, and the immunoreactive proteins were detected using Chemi-Lumi One Super (Nacalai Tesque Inc.) using a luminescent image analyzer (EZ-capture2; ATTO, Tokyo, Japan).

Toluidine blue staining

To evaluate histological changes in cartilage, paraffin sections were stained with 0.04% toluidine blue by the method previously described [44]. Histological evaluation of cartilage was performed using a modified Mankin's scoring system as previously described [45]. The grading system comprised three criteria [inter-territorial matrix staining of uncalcified cartilage (two points), pericellular matrix staining (three points) and spatial arrangement of chondrocytes (three points)] with a possible maximum score of eight points; normal cartilage scored 0.

RNA isolation and quantitative real-time PCR analysis

Total RNA was isolated from cultured chondrocytes and reverse-transcribed using an RNeasy Cells-to-CT kit (Qiagen, Valencia, CA, USA) according to the manufacturer's instructions.

Xbp1 splicing assay

To evaluate XBP1 activation, splicing of Xbp1 mRNA was examined by RT-PCR analysis as described previously [46]. PCR amplification of Xbp1 mRNA was performed using TaKaRa LA Taq (Takara, Kyoto, Japan) according to the manufacturer's instructions. The following primer pair was used for Xbp1: 5′-GAGCAGCAAGTGGTGGATTT-3′ (sense) and 5′-AGGGTCCAACTTGTCCAGAA-3′ (antisense). This primer pair was designed so that the PCR products contained both the spliced form of Xbp1 (Xbp1s) and the unspliced form of Xbp1 (Xbp1u), encompassing the 26 bp region excised by Ire1a. Because this 26 bp region contains a PstI restriction site, electrophoresis of PCR products after PstI treatment separates Xbp1s from Xbp1u. Thermal cycling was performed with denaturation at 94 °C, followed by 35 cycles of denaturation at 94 °C for 30 s, annealing at 56 °C for 30 s, and extension at 72 °C for 30 s. PCR products were incubated with the PstI restriction enzyme for 2 h at 37 °C, and were visualized on a 2% agarose gel using ethidium bromide. The amount of Xbp1s and Xbp1u mRNA was analyzed by densitometric measurement of each band using a densitometer (AE-6920-MF; ATTO).

Real-time RT-PCR

Quantitative real-time PCR was performed using TaqMan gene expression assay probes and TaqMan Universal PCR Master Mix (Applied Biosystems, Foster City, CA, USA). The mouse Chop probe (Mm01135937_g1) and the mouse Actb probe (Mm00607939_s1) were purchased from Applied Biosystems. Quantitative real-time PCR assays were performed using an Applied Biosystems 7500 real-time PCR system with the following cycling parameters: 10 min at 95 °C, 40 cycles of 95 °C for 15 s and 40 cycles of 60 °C for 60 s. The Chop mRNA levels for each gene were normalized to the internal control Actb mRNA using the ΔΔCt method [47].

Statistical analysis

All data are expressed as means ± standard deviation (SD). One-way analysis of variance (ANOVA) with Bonferroni post hoc correction was used to evaluate the effect of incubation duration with GA, concentration of GA, or incubation duration with AMG in cultured Chop+/+ chondrocytes. Student's t test was performed to investigate the effect of CHOP deletion and the presence or absence of incubation with GA in in vitro analyses. Differences in modified Mankin's scores from in vivo experiments between mice 1 week after injection and 5 weeks after injection, between Chop+/+ mice and Chop−/− mice, between NaCl/Pi-injected mice and GA-injected mice, or between operated mice and non-operated mice were also analyzed by Student's t test. A P value of < 0.05 was considered statistically significant. All statistical analyses were performed using StatView version 5.0 (SAS Institute Inc., Cary, NC, USA).

Acknowledgements

We thank Yuichi Oike (Professor of the Department of Molecular Genetics, Graduate School of Medical Sciences, Kumamoto University, Kumamoto, Japan) for providing the Chop−/− mice. We also thank Koji Takada for valuable comments. This work was supported in part by a Grant-in-Aid for Scientific Research from the Japanese Ministry of Education, Culture, Sports, Science, and Technology (20591784).

Ancillary

Advertisement