Elevated Glucose Levels Preserve Glucose Uptake,Hyaluronan Production,and Low Glutamate Release Following Interleukin-1β Stimulation of Differentiated Chondrocytes

Abstract

Objective

Chondrocytes are responsible for remodeling and maintaining the structural and functional integrity of the cartilage extracellular matrix. Because of the absence of a vascular supply, chondrocytes survive in a relatively hypoxic environment and thus have limited regenerative capacity during conditions of cellular stress associated with inflammation and matrix degradation, such as osteoarthritis (OA). Glucose is essential to sustain chondrocyte metabolism and is a precursor for key matrix components. In this study, we investigated the importance of glucose as a fuel source for matrix repair during inflammation as well as the effect of glucose on inflammatory mediators associated with osteoarthritis.

Design

To create an OA model, we used equine chondrocytes from 4 individual horses that were differentiated into cartilage pellets in vitro followed by interleukin-1β (IL-1β) stimulation for 72 hours. The cells were kept at either normoglycemic conditions (5 mM glucose) or supraphysiological glucose concentrations (25 mM glucose) during the stimulation with IL-1β.

Results

We found that elevated glucose levels preserve glucose uptake, hyaluronan synthesis, and matrix integrity, as well as induce anti-inflammatory actions by maintaining low expression of Toll-like receptor-4 and low secretion of glutamate.

Conclusions

Adequate supply of glucose to chondrocytes during conditions of inflammation and matrix degradation interrupts the detrimental inflammatory cycle and induces synthesis of hyaluronan, thereby promoting cartilage repair.

Keywords

chondrocytes osteoarthritis inflammation matrix hyaluronan

Introduction

In spite of its apparent simple structure, articular cartilage is a complex tissue with intricate extracellular matrix (ECM) composition and cellular organization. Cartilage ECM is composed primarily of type II collagen and an interlocking mesh of proteins and proteoglycans, hyaluronan, and chondroitin sulfate.¹ It is becoming increasingly clear that proteoglycans and hyaluronan generate intricate transduction networks comprising signaling pathways involved in immune responses and inflammatory pathways, such as Toll-like receptor (TLR)-2 and TLR-4 signaling.¹

Chondrocytes constitute the only cell type in cartilage and are quiescent cells with limited turnover of the ECM network in the resting steady state.² Chondrocytes respond to mechanical and biochemical changes in ECM through signaling events by various cell surface growth factor receptors and adhesion molecules,³ thereby maintaining the matrix components under normal, low-turnover conditions. The chondrocyte has a low metabolic activity, surviving under relatively hypoxic conditions and with a nutritional supply solely through the synovial fluid. This cell, which is ultimately responsible for remodeling and maintaining the structural and functional integrity of the cartilage matrix, therefore possesses limited regenerative capacity.⁴ During conditions of abnormal cartilage stress, such as osteoarthritis (OA), chondrocytes undergo a phenotypic shift, characterized by increased cell proliferation, cluster formation, and increased production of both matrix proteins and matrix-degrading enzymes, resulting in cartilage degradation.⁵

OA is a low-grade inflammatory progressive disease, frequently associated with joint pain, swelling and stiffness leading to significant functional impairment and disability.⁶ The correlation between the increased levels of catabolic enzymes and inflammatory mediators such as prostaglandins, nitric oxide, and increased levels of interleukin-1β (IL-1β) and tumor necrosis factor (TNF)-α in OA synovial fluids and joint tissue is well documented.⁷ In addition, chondrocytes produce cytokines, chemokines, alarmins, prostanoids, and adipokines and express numerous cell surface receptors for cytokines and chemokines, as well as toll-like receptors (reviewed in Houard et al.²). Synovial inflammation is thus a factor that likely contributes to dysregulation of chondrocyte function in OA, favoring an imbalance between the catabolic and anabolic action of the chondrocyte in remodeling the cartilage extracellular matrix.⁷

Because of the absence of vascular and lymphatic supply in the adult cartilage, distribution of nutrients to the chondrocytes occurs solely by diffusion from synovial fluid. The adequate provision of glucose to articular chondrocytes is essential to sustain their predominantly anaerobic metabolism; glucose is also a precursor for the ECM macromolecules synthesized by these cells.^8-10 Thus, a higher demand for glucose would be expected in OA and other inflammatory conditions but might be limited for chondrocytes due to the avascular nature of the cartilage tissue. Limited glucose supply might compromise cell function and potentially result in an imbalance of matrix synthesis and degradation, leading to osteoarthritis.¹¹ It is therefore plausible that a limited supply of glucose largely contributes to the limited repair capacity of the matrix by the chondrocytes during OA.

Here, we investigate the importance of glucose as a fuel source for matrix production and cartilage repair during conditions of inflammation associated with OA in a 3-dimensional (3D) in vitro model using differentiated equine chondrocyte pellets.

Methods

Tissue Biopsies and Chondrocyte 3-Dimensional Culture

Macroscopically normal cartilage samples were obtained from 4 horses (2-4 years old). Isolation and expansion of chondrocytes were performed as previously described.¹² The study protocol was approved by the Ethical Committee on Animal Experiments, Stockholm, Sweden (N378/12). The protocol for 3D chondrocyte pellet cultures was slightly modified from previously described procedures.¹² Expanded chondrocytes (seeded 300,000 cells/pellet) were cultured for approximately 3 weeks in a 96-well, ultra-low attachment, flat bottom plate (Corning Life Sciences, Lowell, MA, USA) at 37°C in a chondrogenic medium, consisting of Dulbecco’s modified Eagle’s medium (DMEM) with 25 mM glucose concentration (Gibco or Lonza), supplemented with 14 µg/mL ascorbic acid (Sigma-Aldrich), 10⁻⁷ M dexamethasone (Sigma-Aldrich), 1 mg/mL human serum albumin (Equitech Bio, Kerville, TX, USA), 1× insulin–transferrin–selenium (Gibco, Invitrogen), 5 µg/mL linoleic acid (Sigma-Aldrich), 1× penicillin-streptomycin (GE Healthcare Life Sciences HyClone South Logan, Utah, USA), and 10 ng/mL human transforming growth factor (TGF)β-1 (R&D Systems, Abingdon, UK) to form a 3D pellet and generate a defined matrix.

When differentiated 3D chondrocyte pellet were formed, pellets were stimulated with 5 ng/mL IL-1β (R&D Systems Inc., Minneapolis, MN) in DMEM media with 5 mM or 25 mM glucose concentration, supplemented with the additives described above, for 72 hours.

Glucose Uptake

Glucose uptake in 3D equine cartilage pellets were performed essentially as described earlier for monolayer cultured chondrocyte cells.¹³ The 2-deoxy-d-glucose [1,2-3H(N)] (0.2 µCi/pellet) (Perkin Elmer, Waltham, MA, USA) uptake was performed for 35 minutes at 37°C in glucose-free DMEM in the presence or absence of 25 µM cytochalasin B. At the end of incubation, pellets were washed with ice-cold phosphate buffered saline and solubilized in 0.2 M NaOH. Radioactivity was determined in a liquid scintillation counter (Beckman LS 7000 Coulter, Fullerton, CA) and normalized to protein content measured using a bicinchoninic acid (BCA) assay (Thermo Fisher Scientific, Waltham, MA, USA).

Isolation of RNA

Frozen cell pellets were homogenized using 3-mm stainless steel beads and TissueLyser (Qiagen, Valencia, CA, USA). One milliliter QIAzol (Qiagen, Valencia, CA, USA) was added followed by 0.2 mL chloroform and the mixture was centrifuged at 13,000 rpm for 15 minutes at 4°C. Total RNA was extracted using RNeasy Micro Kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions. DNAse I was used to remove contaminating genomic DNA from the isolated RNA. RNA concentrations were measured with NanoDrop 1000 (Thermo Fisher Scientific, Waltham, MA, USA), with a 260/280 nm ratio between 1.7 and 2.1 considered a good purity.

Quantitative Real-Time Polymerase Chain Reaction Analysis

cDNA was prepared from total RNA using High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA). Commercially available equine TaqMan Gene Expression Assays all from Applied Biosystems were used (see table below for details of primers and probes). Samples were analyzed in duplicates. The relative comparative CT method was used to analyze the real-time polymerase chain reaction (PCR) data.

Gene Name	Catalogue No. of Primer/Probe (Applied Biosystems)
HPRT1 (Reference gene)	Ec03470217_m1
SLC2A1/GLUT 1	EC03470363_m1
SGLT1 /SLC5A1	Ec03468144_m1
SLC2A4/GLUT 4	Ec03468109-m1
SLCA2 / GLUT 2	Ec03468195_m1
COL1A1	Ec03469676_m1
COL2A1	Ec03467378_m1
ACAN	Ec03469667_m1
COMP	Ec03468074_m1
MMP13	Ec03467797_m1
HAS2	Ec03467770_m1
TLR4	Ec03468994_m1

Gene Expression of Glucose Metabolism Using PCR Array

Expression of genes involved in glucose metabolism was analyzed by RT2 Profiler PCR Array (Qiagen, Valencia, CA, USA). cDNA for this analysis was prepared by RT2 First Strand Kit (Qiagen, Valencia, CA, USA). Gene expression was analyzed by the real-time-PCR ViiA7 system (Applied Biosystem, Foster City, CA, USA).

Analysis of Protein Interaction Network

To investigate plausible interactions between mediators of glucose metabolism and hyaluronan synthesis mediators, the search tool STRING was used.¹⁴ Analysis settings were set for evidence-based interactions (indicated by lines between nodes) from textmining (yellow), gene neighborhood (green), experimentally determined (pink), gene fusions (red), curated databases (light blue), co-occurrence (dark blue) and co-expression (black). Colored nodes = query proteins and first shell interactions. Small nodes = proteins of unknown 3D structure. Large nodes = some 3D structure is known or predicted.

Glutamate Analysis

Cell culture medium was treated with sulfosalicylic acid and the free amino acids were separated with ion change chromatography in an automatic analysis system. Glutamate was detected with ninhydrin in a photometer at 570 nm at Clinical Chemistry Lab, Sahlgenska University Hospital, Gothenburg, Sweden.

Hyaluronan Analysis

The concentration of hyaluronan in cell culture medium was determined by a Quantikine ELISA according to according to the manufacturer’s instructions (R&D Systems Inc, Minneapolis, MN, USA).

Lactate Analysis

Lactate concentration in cell culture medium was determined by an enzymatic colorimetric method using lactoxidase, peroxidase, and hydrogen peroxide as enzymes and substrate. The color change was detected at 552 nm (Clinical Chemistry Lab, Sahlgenska University Hospital, Gothenburg, Sweden).

Histological Staining

Sections of chondrocyte pellets were stained with safranin O to visualize proteoglycan content and evaluated with light microscopy as described in Ley et al.¹² and scored. The score was based on a scale from 1 to 5 for evaluating intensity of safranin O coloring, pellet size and outer cell layer integrity of the pellet.

Equine cartilage explants were incubated for 24 days in 5 mM glucose or 25 mM glucose with or without 10 ng/ml IL-1β. The explants were sectioned and stained with hematoxylin and eosin, followed by light microscopy analysis.

Statistical Analysis

All data are presented as mean ± standard error of the mean (SEM). One-way analysis of variance with multiple comparisons was used for statistical analysis. A P value <0.05 was considered statistically significant.

Results

To create an OA model, we used equine chondrocytes from four healthy individual horses that we differentiated into cartilage pellets in vitro followed by IL-1β stimulation for 72 hours. Our aim was to investigate the importance of glucose as a fuel source for matrix repair during inflammation as well as the effect of glucose on inflammatory mediators associated with OA. To this end, cells were kept at either normoglycemic conditions (5 mM glucose) or supraphysiological glucose concentrations (25 mM glucose) during the stimulation with IL-1β.

Elevated Levels of Glucose Rescue Glucose Uptake Following IL-1β Stimulation

We found that glucose uptake was decreased by 50% following IL-1β stimulation at 5 mM glucose, whereas glucose uptake remained sustained following IL-1β stimulation at 25 mM glucose ( Fig. 1A ).

Figure 1.

Glucose uptake in response to interleukin-1β (IL-1β) at 5 mM and 25 mM glucose. Equine chondrocytes were differentiated into cartilage pellets in vitro and incubated with or without (basal) 5 ng/ml IL-1β for 72 hours in the presence of 5 mM or 25 mM glucose. (A) Glucose uptake was measured for 35 minutes using 2-deoxy-d-glucose in the presence or absence of 25 µM cytochalasin B and normalized to protein content. (B) mRNA expression of GLUT-1 (Slc2A1), normalized to HPRT1 mRNA. Error bars indicate standard error of the mean (SEM), n = 4 (chondrocytes from 4 different horses). *P < 0.05.

Blocking glucose transporters (GLUT) with cytochalasin B resulted in a marked diminished uptake of glucose for all conditions ( Fig. 1A ), whereas inhibitors of sodium-dependent glucose cotransporters (SGLTs) had no effect (data not shown). Thus, the majority of the glucose taken up in these cells as well as the effect of IL-1β on glucose uptake occurred through GLUT in the chondrocytes. We found the GLUT-1 gene to be robustly expressed ( Fig. 1B ), whereas GLUT-2 or GLUT-4 were not expressed in the differentiated chondrocytes (data not shown). GLUT-1 gene expression was increased following IL-1β stimulation at 25 mM glucose ( Fig. 1B ).

Elevated Glucose Concentration Sustains Matrix Production Following IL-1β Stimulation

We hypothesized that the sustained glucose uptake at 25 mM glucose following IL-1β stimulation was associated with an intact expression and production of cartilage extracellular matrix molecules, which are otherwise known to be subjected to degradation during inflammatory states such as OA. Chondrocyte pellets subjected to IL-1β treatment showed increased staining intensity, larger size, as well as a more intact outer matrix cell layer when incubated at 25mM glucose concentration compared to 5 mM glucose conditions as revealed by histological analysis using safranin O staining ( Fig. 2A and B ). Hematoxylin and eosin staining of equine cartilage explants showed modest fibrillation of the superficial layer of the articular cartilage following IL-1β stimulation at 5 mM glucose, whereas the articular cartilage was intact at 25 mM glucose ( Fig. 2C ), supporting our data for differentiated chondrocytes in cell culture. These data suggest that 25 mM glucose concentration protects the extracellular matrix from degradation during inflammatory conditions.

Figure 2.

(A, B, and D-I) Equine chondrocytes were differentiated into cartilage pellets in vitro and incubated with or without (basal) 5 ng/mL IL-1β for 72 hours in the presence of 5 mM or 25 mM glucose. (A) Cartilage pellets were sectioned and incubated with safranin O to analyze matrix content, followed by (B) scoring of safranin O color intensity, pellet size and morphology. The score was based on a scale of 1 to 5 for evaluating intensity of safranin O coloring, pellet size, and outer cell layer integrity of the pellet. (C) Equine cartilage explants were incubated in vitro with or without (basal) 10 ng/ml IL-1β for 24 days in the presence of 5 mM or 25 mM glucose and analyzed by hematoxylin and eosin staining. Modest fibrillation of the superficial layer of the articular cartilage in IL-1β-stimulated explants at 5 mM glucose is indicated by arrows. Scale bar: 100 μm. (D-I) mRNA expression of COL1A1, COL2A1, ACAN, COMP, MMP13, and HAS2, normalized to HPRT1 mRNA. Error bars indicate standard error of the mean (SEM), n = 4 (chondrocytes from 4 different horses). *P < 0.05.

At 5 mM glucose, IL-1β stimulation resulted in a marked decreased expression of the key matrix genes COL2A1, ACAN, and HAS2, and a pronounced increased gene expression of MMP13, whereas COL1A1 and COMP were not significantly changed ( Fig. 2D-I ). Incubating the cells at 25 mM glucose concentration tended to increase ACAN gene expression both at basal and IL-1β-stimulated states ( Fig. 2F ) compared with 5 mM glucose. However, the most prominent effect was observed for the HAS2 gene, with expression levels maintained at high levels following IL-1β stimulation at 25 mM glucose, in contrast to 5 mM glucose, for which HAS2 gene expression was decreased by >50% in the presence of IL-1β ( Fig. 2I ).

Sustained HAS2 Gene Expression in Response to Elevated Glucose Levels Is Accompanied by Anti-Inflammatory Signals

Based on the HAS2 gene expression data, we hypothesized that the amount of secreted hyaluronan from the cartilage pellets would be similarly regulated with elevated glucose levels following IL-1β stimulation. We found that hyaluronan secretion increased following IL-1β at 25 mM glucose ( Fig. 3A ). There was a trend toward increased hyaluronan secretion following IL-1β at 5 mM glucose as well but did not reach statistical significance ( Fig. 3A ). Gene expression analysis of the pro-inflammatory receptor TLR-4 revealed a large increase in response to IL-1β at 5 mM glucose, whereas this increase was blunted at 25 mM glucose levels, maintaining TLR-4 expression at low levels ( Fig. 3B ). Analysis of glutamate secretion revealed a completely blunted release of glutamate in response to IL-1β stimulation at 25 mM glucose in contrast to 5 mM glucose conditions, for which cartilage pellets from each individual horse showed a marked increased glutamate release ( Fig. 3C ).

Figure 3.

Evaluation of anti-inflammatory activities. Equine chondrocytes were differentiated into cartilage pellets in vitro and incubated with or without (basal) 5 ng/ml interleukin-1β (IL-1β) for 72 hours in the presence of 5 mM or 25 mM glucose. (A) Hyaluronan secretion relative to 5 mM glucose in the basal state. (B) mRNA expression of TLR4, normalized to HPRT1 mRNA. (C) Glutamate release relative to 5 mM glucose in the basal state. Each color indicates chondrocytes from the same individual at the indicated different conditions. Error bars indicate standard error of the mean (SEM), n = 4 (chondrocytes from 4 different horses). *P < 0.05.

Taken together, hyaluronan secretion from differentiated chondrocytes following IL-1β stimulation is increased at 25 mM glucose compared with 5 mM glucose. The sustained HAS2 gene expression was accompanied by low TLR-4 gene expression and inhibition of glutamate release at 25 mM glucose concentrations in response to IL-1β stimulation, indicating enhanced cellular anti-inflammatory actions in response to 25 mM glucose compared with 5 mM glucose.

Elevated Glucose Levels Result in an Increased Glycolytic Rate in Differentiated Chondrocytes

To identify mechanistic mediators of the association between the sustained glucose uptake and HAS2 expression, we performed a PCR array analysis, investigating 84 genes involved in glucose metabolism. The results are summarized in Table 1 . Most enzymes involved in glucose metabolism were similarly affected by IL-1β stimulation, independently of glucose concentration. Changes were generally small, and most did not reach statistical significance. Nevertheless, the most pronounced changes were observed for glycolysis with marked significant differences between 5 mM and 25 mM glucose following IL-1β stimulation for pyruvate dehydrogenase kinase 4 (PDK4) and alpha-enolase (ENO1). PDK4 was downregulated 10-fold with 25 mM glucose and ENO1 was upregulated ~6-fold with 25 mM glucose compared with 5 mM glucose following IL-1β stimulation. Analysis of lactate secretion revealed increased lactate release in response to 25 mM glucose compared with 5 mM glucose following stimulation with IL-1β ( Fig. 4A ), supporting an increased flux of glucose through the glycolysis pathway. STRING protein-protein interaction network analysis revealed a strong association between ENO1 and HAS2 and HAS3 ( Fig. 4B ), suggesting that ENO1 might be involved in HAS2 regulation and/or function.

Table 1.

Equine Chondrocytes were Differentiated into Cartilage Pellets in Vitro and Incubated with or without (basal) 5 ng/mL Interleukin-1β (IL-1β) for 72 Hours in the Presence of 5 mM or 25 mM Glucose.^a

		5 mM Glucose, IL-1β Fold Regulation Relative to 5 mM Glucose Basal	25 mM Glucose, Basal Fold Regulation Relative to 5 mM Glucose Basal	25 mM Glucose, IL-1β Fold Regulation Relative to 5 mM Glucose Basal
Cytosolic acetyl-CoA synthesis
ATP citrate lyase	ACLY	1.0	−1.9	−1.5
Gluconeogenesis
Phosphoenolpyruvate carboxykinase 1, cytosolic	PCK1		Not expressed
Phosphoenolpyruvate carboxykinase 2, mitochondrial	PCK2	−4.1	−2.4	−4.3
Fructose bisphosphatase 1	FBP1		Not expressed
Fructose bisphosphatase 2	FBP2		Not expressed
Glucose-6-phosphatase, catalytic	G6PC		Not expressed
Glucose-6-phosphatase, catalytic 3	G6PC3	−1.2	−1.2	−1.5
Glucose-6-phosphate dehydrogenase	G6PD	2.2	−1.2	1.4
Glycogen metabolism
Amylase	AGL	1.0	−1.6	−1.7
Glycogen synthase kinase 3 alpha	GSK3A	1.0	−1.3	−1.2
Glycogen synthase kinase 3 beta	GSK3B	−0.3	−1.5	−1.2
Glycogen synthase 1	GYS1	−1.7	−2.7	−1.9
Glycogen synthase 2	GYS2		Not expressed
Phosphorylase kinase alpha 1	PHKA1	−1.1	−1.2	−1.3
Phosphorylase kinase beta	PHKB	1.1	−1.5	−1.3
Phosphorylase kinase, gamma 1 (muscle)	PHKG1	−1.7	1.1	−1.1
Phosphorylase kinase, gamma 2 (testis)	PHKG2	1.3	1.0	−1.1
Phosphorylase, glycogen, liver	PYGL	−1.8	−1.8	−1.7
Glucan (1,4-alpha-), branching enzyme 1 (transglycosylase)	GBE1	−1.6	−1.7	−1.8
UTP-glucose-1-phosphate uridylyltransferase-like	UGP2	−1.7	−1.9	−1.9
Glycolysis
Aldolase A, fructose-bisphosphate	ALDOA	1.0	−1.7	1.0
Aldolase B, fructose-bisphosphate	ALDOB	1.0	1.3	−1.4
Aldolase C, fructose-bisphosphate	ALDOC	−1.1	−2.1	−1.1
Bisphosphoglycerate mutase-like	BPGM	−2.2	−1.7	−2.2
Aldose 1-epimerase-like	GALM	1.6	−1.5	1.0
Glucokinase	GCK		Not expressed
Glucose phosphatase isomerase 1 (phosphoglucose isomerase)	GPI1	−1.6	−1.9	−1.9
Hexokinase 2	HK2	−4.2	−3.0	−3.1
Hexokinase 3	HK3	−1.5	−1.2	1.0
Pyruvate dehydrogenase kinase, isoenzyme 1	PDK1	−1.5	−2.4	−1.8
Pyruvate dehydrogenase kinase, isoenzyme 2	PDK2	−1.1	−1.1	1.0
Pyruvate dehydrogenase kinase, isoenzyme 3	PDK3	−1.5	−1.6	−1.3
Pyruvate dehydrogenase kinase, isoenzyme 4	PDK4	1.0	−3.4	−10.3
Phosphofructokinase, liver, B-type	PFKL	−2.0	−1.9	−2.2
Pyruvate dehydrogenase phosphatase catalytic subunit 2	PDP2	1.0	−2.1	−1.7
Pyruvate dehydrogenase phosphatase regulatory subunit	PDPR	1.0	−1.5	−1.2
Phosphoglycerate mutase 2-like	PGAM2	−1.3	−1.1	−1.1
Phosphoglycerate kinase 1	PGK1	−1.6	−2.3	−2.2
Phoshoglycerate kinase 2	PGK2	−1.7	−1.6	−1.6
6-phosphogluconolactonase-like	PGLS	−1.3	−1.5	−1.8
Pyruvate kinase, liver and RBC	PKLR		Not expressed
Triosephosphate isomerase 1	TPI1	−1.4	−1.4	−1.2
Enolase 1, alpha	ENO1	4.2	−2.3	26.2
Enolase 2, gamma (neuronal)	ENO2	1.0	1.1	1.0
Enolase 3, beta (muscle)	ENO3	−1.9	−1.6	−2.4
Phosphoglucomutase, 1	PGM1	1.3	−1.3	1.0
Phosphoglucomutase, 2	PGM2	−1.3	−1.3	−1.9
Phosphoglucomutase, 3	PGM3	−1.1	−1.3	−1.4
Pyruvate dehydrogenase E1 component subunit alpha, somatic form	PDHA1	1.2	−1.5	−1.3
Pyruvate dehydrogenase (lipoamide) beta	PDHB	1.2	−1.3	−1.3
Pentose phosphate pathway
Hexose-6-phosphate dehydrogenase (glucose 1-dehydrogenase)	H6PD	−1.2	−1.2	−1.1
Hypoxanthine phosphoribosyltransferase 1	HPRT1	−1.5	−2.1	−1.9
Phosphoribosyl pyrophosphate synthetase 1	PRPS1	1.7	−1.6	1.0
Phosphoribosyl pyrophosphate synthetase 2	PRPS2	1.7	−1.4	1.0
Phosphoribosyl pyrophosphate synthetase 1-like 1	PRPS1L1		Not expressed
Ribokinase-like	RBKS	−1.4	−1.3	1.3
Ribulose-5-phosphate-3-epimerase	RPE	−1.2	−1.5	−1.3
Ribose 5-phosphate isomerase A	RPIA	−1.1	−1.8	−2.1
Transaldolase 1	TALDO1	2.0	−1.3	1.5
Transketolase	TKT	1.7	−1.7	−1.2
TCA cycle
Aconitase 1, soluble	ACO1	1.2	−1.5	−1.2
Aconitase 2, mitochondrial	ACO2	1.5	−1.4	1.0
Citrate synthase	CS	1.4	−1.5	−1.1
Dihydrolipoamide S-acetyltransferase	DLAT	1.2	−1.6	−1.2
Dihydrolipoamide dehydrogenase	DLD	1.4	−1.5	−1.5
Dihydrolipoamide S-succinyltransferase (E2 component of 2-oxo-glutarate complex)	DLST	1.1	−1.2	−1.1
Fumarate hydratase 1	FH1	1.2	−1.6	−1.5
Isocitrate dehydrogenase 1 (NADP+), soluble	IDH1	2.2	−1.5	1.6
Isocitrate dehydrogenase 2 (NADP+), mitochondrial	IDH2	1.3	−1.5	−1.1
Isocitrate dehydrogenase 3 (NAD+) alpha	IDH3A	1.1	−1.5	−1.3
Isocitrate dehydrogenase 3 (NAD) subunit beta, mitochondrial like	IDH3B	1.3	−1.8	−1.4
Isocitrate dehydrogenase 3 (NAD+), gamma	IDH3G	−1.8	−2.6	−2.2
Malate dehydrogenase, cytoplasmic like	MDH1	1.1	−1.8	−1.7
Malate dehydrogenase 1B, NAD (soluble)	MDH1B	−1.7	−2.1	−3.2
Malate dehydrogenase 2, NAD (mitochondrial)	MDH2	1.4	−1.3	1.0
2-oxoglutarate dehydrogenase, mitochondrial-like	OGHD	1.0	−1.4	−1.3
Pyruvate carboxylase	PC	1.8	−1.5	1.1
Succinate dehydrogenase complex, subunit A, flavoprotein	SDHA	1.0	−1.5	−1.2
Succinate dehydrogenase complex, subunit B, iron sulphur	SDHB	1.0	−1.3	−1.4
Succinate dehydrogenase cytochrome B560 subunit, mitochondrial like	SDHC	1.1	−1.7	−1.1
Succinate dehydrogenase complex, subunit D, integral membrane protein	SDHD	−1.1	−1.3	−1.1
Succinyl-CoA ligase (ADP forming) subunit B, mitochondrial like	SUCLA2	1.1	−1.3	−1.2
Succinate CoA-ligase, GDP-forming, alpha subunit	SUCLG1	1.1	−1.3	1.0
Succinate CoA-ligase, GDP forming, beta subunit	SUCLG2	1.7	−1.4	1.1

Expression of genes involved in glucose metabolism was examined with a polymerase chain reaction array. Fold differences relative to 5 mM glucose in the basal state are shown. Numbers in boldface indicate significant differences relative to 5 mM glucose in the basal state. n = 3.

Figure 4.

Lactate secretion and relation of ENO1 with HAS2. Equine chondrocytes were differentiated into cartilage pellets in vitro and incubated with or without (basal) 5 ng/mL interleukin-1β (IL-1β) for 72 hours in the presence of 5 mM or 25 mM glucose. (A) Lactate secretion relative to 5 mM glucose in the basal state. Error bars indicate standard error of the mean (SEM), n = 4 (chondrocytes from 4 different horses). *P < 0.05. (B) STRING protein-protein interaction network analysis. Analysis settings were set for evidence-based interactions (indicated by lines between nodes) from textmining (yellow), gene neighborhood (green), experimentally determined (pink), gene fusions (red), curated databases (light blue), co-occurrence (dark blue) and co-expression (black). Colored nodes = query proteins and first shell interactions. Small nodes = proteins of unknown 3-dimensional (3D) structure. Large nodes = some 3D structure is known or predicted.

Taken together, when investigating glucometabolic pathways, the most pronounced effect of elevated glucose levels in response to IL-1β stimulation was observed for glycolytic mediators, of which ENO1 might play a role in HAS2 regulation.

Discussion

In this study, we investigated the effect of physiological and supraphysiological levels of glucose on glucose uptake, matrix production and glutamate release in an OA model of differentiated 3D chondrocyte pellets. In contrast to physiological levels of glucose (5 mM), elevated levels (25 mM) resulted in sustained glucose uptake and HAS2 expression in response to IL-1β stimulation as well as increased hyaluronan secretion, maintained low expression of Toll-like receptor 4 and low secretion of glutamate following IL-1β stimulation. These data are all in line with increased functional matrix production and anti-inflammatory processes, supporting cartilage repair.

Previously reported studies on the effect of IL-1β on glucose uptake in isolated primary chondrocytes support our findings in differentiated chondrocyte pellets, showing that, at high glucose concentrations, IL-1β stimulation sustained or increased glucose uptake in chondrocytes,¹⁵ whereas glucose uptake was reduced in response to IL-1β at low glucose concentrations (2.5 mM or lower).¹¹ One report by Shikhman et al.¹⁶ reported results contradictory to our study with increased glucose uptake at 5 mM glucose in response to inflammatory cytokines such as IL-1β, TNFα, and IL-6. However, besides using primary chondrocytes rather than differentiated pellets, the authors also used shorter incubation time (24 hours) with IL-1β, which might explain the discrepancies.

The effect of IL-1β on degradation of matrix molecules is well known and was confirmed in our study with marked effects on COL2A1, ACAN, MMP13, and HAS2 following IL-1β stimulation for 72 hours at 5 mM glucose. Increased glucose levels appeared to sustain matrix production in the presence of IL-1β as indicated by safranin O staining intensity, pellet size, morphology as well as an intact cartilage surface. At gene level, the most marked effect of elevated glucose concentration was found for HAS2, suggesting that elevated glucose levels protected the cells from hyaluronan depletion in the presence of IL-1β, similar to what has been shown for other cytokines, such as TNFα, TGFβ, and interferon-γ. ¹⁷ The importance of glucose for matrix production has been reported by others^18-20; the accumulation of aggrecan at wound sites in the ts5 knockout mouse was a result of marked aggrecan synthesis due to amplified glucose uptake by local reparative cells.¹⁸ Moreover, Mobasheri et al.²⁰ showed that hypoxia and glucose deprivation increase the production of the active form of the matrix metalloproteinase MMP-2. Thus, our data, as well as others’, suggest that an adequate glucose supply is crucial for matrix production and integrity.

The HAS2 gene encodes hyaluronan synthase 2, the main enzyme for full length hyaluronan synthesis.²¹ Hyaluronan (HA) is a crucial matrix component in cartilage. High molecular weight (HMW) HA polymers are associated with anti-inflammatory processes, whereas low molecular weight (LMW) HA polymers, are associated with proinflammatory processes (reviewed in Frey et al.¹). HMW HA has been proposed to promote inhibition of inflammation by forming a protective coating of the chondrocytes, thereby preventing binding of proinflammatory ligands to Toll-like receptors,²² as well as decreasing cellular glutamate secretion.²³ Chondrocytes have been shown to express a self-sufficient machinery for glutamate signaling with increased secretion of glutamate in response to IL-1β stimulation, and glutamate levels have been reported to be increased in the synovial fluid of OA patients.²⁴ Decreased glutamate is known to lead to both decreased inflammation as well as increased glucose uptake.^24-26 Indeed, glutamate release, in response to IL-1β stimulation, was completely blunted at 25 mM glucose levels in our study, in contrast to 5 mM glucose. Moreover, our data on equine explants, show fibrillation of the superficial layer of the articular cartilage following IL-1β stimulation at 5 mM glucose, but not at 25 mM glucose, suggesting an intact protective HMW HA layer at high glucose levels. Based on these data, it is plausible that the composition of the secreted HA might differ, such that the HMW:LMW ratio of HA at 25 mM glucose is larger than that at 5 mM glucose, promoting anti-inflammatory activities and matrix maintenance at elevated glucose levels ( Fig. 5 ). Considering that the maintained HAS2 expression was also accompanied by a maintained low expression of TLR-4, HAS2 might conduct a protective response of chondrocytes to inflammation by exerting anti-inflammatory actions at several different levels.

Figure 5.

Hypothesis of the underlying mechanism for the effect of glucose during osteoarthritis. In conditions with osteoarthritis, inflammation causes degradation of cartilage components, including hyaluronan (HA), due to an insufficient supply of glucose and decreased glucose uptake in the chondrocyte. The decreased glucose uptake results in decreased expression of HAS2 and a lowering of the high molecular weight to low molecular weight (HMW:LMW) HA ratio. Reduced synthesis of HMW HA leads to a breakdown of the HA protective layer, exposing TLR-4 and enables interaction of TLR-4 with proinflammatory cytokines, thereby promoting inflammation. Furthermore, decreased HMW HA results in increased glutamate release, which in turn enhances inflammation. In addition, inflammatory cytokines, such as IL-1β, promote TLR-4 expression. The result is a self-sustained inflammatory cycle. By addition of glucose, the glucose uptake is sustained, resulting in maintained production of HMW HA by sustained HAS2 expression. Maintained HMW HA synthesis further leads to an intact protective HA layer as well as inhibition of glutamate release and maintained low expression of TLR-4. Maintained glucose uptake further enables a high glycolytic rate with increased lactate production. These anti-inflammatory actions interrupt the proinflammatory cycle and induces matrix repair.

To identify molecular players underlying an association between increased glucose uptake and sustained HAS2 expression, we performed a PCR array, investigating 84 genes involved in glucose metabolism. The most pronounced effects of elevated glucose levels were observed for glycolytic mediators, particularly pyruvate dehydrogenase kinase 4 (PDK4) and alpha-enolase (ENO1). Interestingly, performing a STRING protein-protein interaction network analysis,¹⁴ we identified an association between ENO1 and HAS2 and HAS3. Besides functioning as a glycolytic enzyme, alpha-enolase has been shown to be translocated to, and act on, the cell surface of monocytes and tumor cells. On the cell surface, monocyte alpha-enolase induces inflammation and matrix invasion of monocytes,²⁷ and promotes cell growth, migration and metastasis of tumor cells.^28,29 Cell surface expression of alpha-enolase has also been shown to be increased in the synovial fluid monocytes of RA patients and OA patients.^30,31 Whether alpha-enolase is expressed on the chondrocyte cell surface is not known and no direct interaction between alpha-enolase and HAS2 has been reported, but the STRING analysis suggests there might be an expected functional or physical interaction. Future mechanistic in vitro studies will elucidate any such potential association, and whether alpha-enolase has a proinflammatory or anti-inflammatory role in the chondrocyte.

In support of an overall increased flux of glucose through glycolysis, we found that lactate secretion was increased at 25 mM glucose compared to 5mM glucose. Lactate is the principal end-product of glucose metabolism by articular cartilage.³² Interestingly, increased release of lactate is associated with lower pH, and lowering pH has been shown to decrease glutamate release in presynaptic neurons and glial cells.^33,34 Thus, increased lactate release in chondrocytes as a result of increased glucose uptake and glucose flux through the glycolytic pathway, might contribute to anti-inflammatory cellular actions by inhibiting glutamate release ( Fig. 5 ).

Based on the fact that endocrine disorders, such as type 2 diabetes, are associated with osteoarthritis,^35-37 it has been postulated that excess glucose levels associated with diabetes might lead to reactive oxygen species production and advanced glycation end products in cartilage,^38-40 thereby favoring the progression of OA. However, the large majority of diabetes patients are rarely severely hyperglycemic for prolonged periods of time, but have their glucose levels controlled by various treatments, for example insulin injections or metformin. As chondrocytes are insulin insensitive,¹³ lacking the insulin-sensitive glucose transporter GLUT-4, glucose will not be directed to cartilage by diabetes drugs. Instead, glucose is shunted to insulin-sensitive tissues such as muscle and adipose tissue, resulting in predicted glucose concentrations of ~5mM in the synovial fluid.^41,42 Thus, type 2 diabetics are unlikely to have chronically elevated levels of glucose in the synovial fluid. Rather than high glucose levels being the driving force of OA in diabetes patients, it is more plausible that chronically elevated levels of cytokines are a major contributing factor. Cytokine levels are known to be chronically elevated in type 2 diabetics,⁴³ and appears to be a crucial mediator of the OA progression. Higher levels of cytokines in the synovial fluid in combination with insufficient glucose supply would be a very strong factor in driving cartilage degradation and OA.

In summary, in this study we show that elevated glucose levels rescue glucose uptake, matrix production and maintenance and induce anti-inflammatory activities in differentiated chondrocytes during inflammatory conditions associated with OA.

There is an unfulfilled need to identify novel therapeutic targets to halt cartilage degradation and to induce cartilage restoration in OA conditions.⁴⁴ Current clinical treatment strategies have varying success rates, with unsatisfactory long-term results.^45-49 Further research is needed to expand our understanding of glucosensing and the impact of glucose on the ECM in chondrocytes. However, based on our data, intra-articular injections with glucose would be expected to exert beneficial effects in early phases of OA and/or in combination with other treatments to enhance treatment efficacy by rescuing chondrocyte metabolism, inducing HA synthesis and anti-inflammatory processes, thereby interrupting the vicious inflammatory cycle and promoting matrix production and repair ( Fig. 5 ).

Footnotes

Acknowledgments and Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the Swedish Research Council Grant 2015-02959 and Western Region Research Grant ALFGBG-432251.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical Approval

The study protocol was approved by the Ethical Committee on Animal Experiments, Stockholm, Sweden (N378/12).

Animal Welfare

The present study followed international, national, and/or institutional guidelines for humane animal treatment and complied with relevant legislation.

ORCID iD

Victoria Rotter Sopasakis

References

Frey

Schroeder

Manon-Jensen

Iozzo

Schaefer

Biological interplay between proteoglycans and their innate immune receptors in inflammation. FEBS J. 2013;280(10):2165-79.

Houard

Goldring

Berenbaum

Homeostatic mechanisms in articular cartilage and role of inflammation in osteoarthritis. Curr Rheumatol Rep. 2013;15(11):375.

Leong

Hardin

Cobelli

Sun

HB.

Mechanotransduction and cartilage integrity. Ann N Y Acad Sci. 2011;1240:32–7.

Goldring

SR.

Osteoarthritis. J Cell Physiol. 2007;213(3):626-34.

Goldring

Otero

Plumb

Dragomir

Favero

El Hachem

, et al. Roles of inflammatory and anabolic cytokines in cartilage metabolism: signals and multiple effectors converge upon MMP-13 regulation in osteoarthritis. Eur Cell Mater. 2011;21:202-20.

Felson

DT.

Clinical practice. Osteoarthritis of the knee. N Engl J Med. 2006;354(8):841-8.

Loeser

RF.

Molecular mechanisms of cartilage destruction: mechanics, inflammatory mediators, and aging collide. Arthritis Rheum. 2006;54(5):1357-60.

Kim

Conrad

HE.

Kinetics of mucopolysaccharide and glycoprotein synthesis by chick embryo chondrocytes. Effect of D-glucose concentration in the culture medium. J Biol Chem. 1976;251(20):6210-7.

Mobasheri

Vannucci

Bondy

Carter

Innes

Arteaga

, et al. Glucose transport and metabolism in chondrocytes: a key to understanding chondrogenesis, skeletal development and cartilage degradation in osteoarthritis. Histol Histopathol. 2002;17(4):1239-67.

10.

Mobasheri

Glucose: an energy currency and structural precursor in articular cartilage and bone with emerging roles as an extracellular signaling molecule and metabolic regulator. Front Endocrinol (Lausanne). 2012;3:153.

11.

Windhaber

Wilkins

Meredith

Functional characterisation of glucose transport in bovine articular chondrocytes. Pflugers Arch. 2003;446(5):572-7.

12.

Ley

Svala

Nilton

Lindahl

Eloranta

Ekman

, et al. Effects of high mobility group box protein-1, interleukin-1β, and interleukin-6 on cartilage matrix metabolism in three-dimensional equine chondrocyte cultures. Connect Tissue Res. 2011;52(4):290-300.

13.

Richardson

Neama

Phillips

Bell

Carter

Moley

, et al. Molecular characterization and partial cDNA cloning of facilitative glucose transporters expressed in human articular chondrocytes; stimulation of 2-deoxyglucose uptake by IGF-I and elevated MMP-2 secretion by glucose deprivation. Osteoarthritis Cartilage. 2003;11(2):92-101.

14.

Szklarczyk

Morris

Cook

Kuhn

Wyder

Simonovic

, et al. The STRING database in 2017: quality-controlled protein-protein association networks, made broadly accessible. Nucleic Acids Res. 2017;45(D1):D362-D368.

15.

Hernvann

Jaffray

Hilliquin

Cazalet

Menkes

Ekindjian

OG.

Interleukin-1β-mediated glucose uptake by chondrocytes. Inhibition by cortisol. Osteoarthritis Cartilage. 1996;4(2):139-42.

16.

Shikhman

Brinson

Valbracht

Lotz

MK.

Cytokine regulation of facilitated glucose transport in human articular chondrocytes. J Immunol. 2001;167(12):7001-8.

17.

Campo

Avenoso

Campo

D’Ascola

Traina

Calatroni

Effect of cytokines on hyaluronan synthase activity and response to oxidative stress by fibroblasts. Br J Biomed Sci. 2009;66(1):28-36.

18.

Gorski

Xiao

Luo

Lauer

Kisiday

, et al. Deletion of ADAMTS5 does not affect aggrecan or versican degradation but promotes glucose uptake and proteoglycan synthesis in murine adipose derived stromal cells. Matrix Biol. 2015;47:66-84.

19.

Johnson

Stephan

Roberts

The influence of serum, glucose and oxygen on intervertebral disc cell growth in vitro: implications for degenerative disc disease. Arthritis Res Ther. 2008;10(2):R46.

20.

Mobasheri

Platt

Thorpe

Shakibaei

Regulation of 2-deoxy-d-glucose transport, lactate metabolism, and MMP-2 secretion by the hypoxia mimetic cobalt chloride in articular chondrocytes. Ann N Y Acad Sci. 2006;1091:83-93.

21.

Itano

Sawai

Yoshida

Lenas

Yamada

Imagawa

, et al. Three isoforms of mammalian hyaluronan synthases have distinct enzymatic properties. J Biol Chem. 1999;274(35):25085-92.

22.

Ebid

Lichtnekert

Anders

HJ.

Hyaluronan is not a ligand but a regulator of toll-like receptor signaling in mesangial cells: role of extracellular matrix in innate immunity. ISRN Nephrol. 2014;2014:714081.

23.

Jean

Wen

Chang

Lee

Hsieh

, et al. Hyaluronic acid attenuates osteoarthritis development in the anterior cruciate ligament-transected knee: association with excitatory amino acid release in the joint dialysate. J Orthop Res. 2006;24(5):1052-61.

24.

Piepoli

Mennuni

Zerbi

Lanza

Rovati

Caselli

Glutamate signaling in chondrocytes and the potential involvement of NMDA receptors in cell proliferation and inflammatory gene expression. Osteoarthritis Cartilage. 2009;17(8):1076-83.

25.

Azbill

Springer

JE.

Riluzole improves measures of oxidative stress following traumatic spinal cord injury. Brain Res. 2000;870(1-2):66-72.

26.

Nagao

Nishizawa

Bamba

Nakayama

Isozumi

Nagamori

, et al. Increased dynamics of tricarboxylic acid cycle and glutamate synthesis in obese adipose tissue: IN VIVO METABOLIC TURNOVER ANALYSIS. J Biol Chem. 2017;292(11):4469-83.

27.

Wygrecka

Marsh

Morty

Henneke

Guenther

Lohmeyer

, et al. Enolase-1 promotes plasminogen-mediated recruitment of monocytes to the acutely inflamed lung. Blood. 2009;113(22):5588-98.

28.

Song

Luo

Long

Que

Zhang

, et al. Alpha-enolase as a potential cancer prognostic marker promotes cell growth, migration, and invasion in glioma. Mol Cancer. 2014;13:65.

29.

Hsiao

Shih

Fang

Huang

Kuo

Chu

, et al. Surface α-enolase promotes extracellular matrix degradation and tumor metastasis and represents a new therapeutic target. PLoS One. 2013;8(7):e69354.

30.

Zhou

Luo

Jia

Gan

, et al. Analyses of differential proteome of human synovial fibroblasts obtained from arthritis. Clin Rheumatol. 2009;28(2):191-9.

31.

Bae

Kim

Lee

Won

Kim

Hwang

, et al. α-Enolase expressed on the surfaces of monocytes and macrophages induces robust synovial inflammation in rheumatoid arthritis. J Immunol. 2012;189(1):365-72.

32.

Lee

Urban

JP.

Evidence for a negative Pasteur effect in articular cartilage. Biochem J. 1997;321 (Pt 1):95-102.

33.

Fedorovich

Kaler

Konev

SV.

Effect of low pH on glutamate uptake and release in isolated presynaptic endings from rat brain. Neurochem Res. 2003;28(5):715-21.

34.

Billups

Attwell

Modulation of non-vesicular glutamate release by pH. Nature. 1996;379(6561):171-4.

35.

Hart

Doyle

Spector

TD.

Association between metabolic factors and knee osteoarthritis in women: the Chingford study. J Rheumatol. 1995;22(6):1118-23.

36.

Stürmer

Brenner

Günther

KP.

Non-insulin dependent diabetes mellitus (NIDDM) and patterns of osteoarthritis. The Ulm osteoarthritis study. Scand J Rheumatol. 2001;30(3):169-71.

37.

Burner

Rosenthal

AK.

Diabetes and rheumatic diseases. Curr Opin Rheumatol. 2009;21(1):50–4.

38.

Rosenbloom

Silverstein

JH.

Connective tissue and joint disease in diabetes mellitus. Endocrinol Metab Clin North Am. 1996;25(2):473-83.

39.

Rosa

Goncalves

Judas

Mobasheri

Lopes

Mendes

AF.

Impaired glucose transporter-1 degradation and increased glucose transport and oxidative stress in response to high glucose in chondrocytes from osteoarthritic versus normal human cartilage. Arthritis Res Ther. 2009;11(3):R80.

40.

Laiguillon

Courties

Houard

Auclair

Sautet

Capeau

, et al. Characterization of diabetic osteoarthritic cartilage and role of high glucose environment on chondrocyte activation: toward pathophysiological delineation of diabetes mellitus-related osteoarthritis. Osteoarthritis Cartilage. 2015;23(9):1513-22.

41.

Omar

Reichling

Liodakis

Ettinger

Guenther

Decker

, et al. Rapid exclusion of bacterial arthritis using a glucometer. Clin Rheumatol. 2017;36(3):591-8.

42.

Liberg

Magnusson

Schougaard

Studies on the synovia in healthy horses with particular reference to the protein composition. Equine Vet J. 1977;9(2):87-91.

43.

Keane

Calton

Carlessi

Hart

Newsholme

The bioenergetics of inflammation: insights into obesity and type 2 diabetes. Eur J Clin Nutr. 2017;71(7):904-12.

44.

Shen

Abu-Amer

O’Keefe

McAlinden

Inflammation and epigenetic regulation in osteoarthritis. Connect Tissue Res. 2017;58(1):49-63.

45.

Kreuz

Steinwachs

Erggelet

Krause

Konrad

Uhl

, et al. Results after microfracture of full-thickness chondral defects in different compartments in the knee. Osteoarthritis Cartilage. 2006;14(11):1119-25.

46.

Redman

Oldfield

Archer

CW.

Current strategies for articular cartilage repair. Eur Cell Mater. 2005;9:23-32.

47.

Bentley

Biant

Carrington

Akmal

Goldberg

Williams

, et al. A prospective, randomised comparison of autologous chondrocyte implantation versus mosaicplasty for osteochondral defects in the knee. J Bone Joint Surg Br. 2003;85(2):223-30.

48.

Hunziker

EB.

Articular cartilage repair: basic science and clinical progress. A review of the current status and prospects. Osteoarthritis Cartilage. 2002;10(6):432-63.

49.

Kock

van Donkelaar

Ito

Tissue engineering of functional articular cartilage: the current status. Cell Tissue Res. 2012;347(3):613-27.