Adenosine monophosphate-activated protein kinase activator inhibits activation of fibroblast-like synoviocytes but promotes hyaluronan and proteoglycan link protein 1 secretion

Objectives: To determine whether any correlation exists between disease activity and AMPK levels in rheumatoid arthritis (RA) patients and investigate the effects of AMPK activator treatment on RA fibroblast-like synoviocytes (RA-FLS). Methods: Serum AMPK-α1, p-AMPK-α1, TNF-α and IL-17 levels between osteoarthritis (OA) and RA patients having different disease activities were compared by ELISA. Differentially expressed genes (DEGs) between RA and OA synovium from NCBI GEO Profiles (accession numbers: GSE1202112, GSE55235, GSE5545713) were identified and the genes intersecting in all the three datasets were selected for enrichment analysis. Immunohistochemical staining was done with synovium obtained from OA and RA patients for p-AMPK-α1. AMPK gene expression in synovium was semi-quantified by RT-qPCR. RNA sequencing of FLS was performed and DEGs were selected for KEGG enrichment analysis. AMPK activator, metformin, treated RA-FLS were tested for proliferation and migration by MTT and scratch test, respectively. Expression of IL-6, AMPK-α1, PKA-α, RAPTOR, mTOR, HAPLN1, RUNX1 and RUNX2 genes were determined by qPCR. Phosphorylated AMPK-α1 and HAPLN1 levels were determined by an automated electrophoresis-western blot analysis method. Results: In RA sera, a positive correlation between p-AMPK-α1 levels and DAS28 (r = 0.270, 95%CI: 0.142-0.492, p < 0.0001) as well as CRP levels (r = 0.259, 95%CI: 0.009-0.478, p < 0.05) was found. Similarly, a positive correlation was observed between AMPKα1 and TNF-α levels (r = 0.460, 95% CI: 0.241-0.640, p = 0.0002). DEGs between OA and RA synovium from NCBI GEO profiles and our RNA sequencing data suggested activation of metabolic pathways specific to RA-FLS. AMPK-α1 in the synovium of RA but not OA patients. Metformin at higher concentrations inhibited RA-FLS proliferation in a dose dependant manner, at lower concentrations it has an opposite effect. On other hand, AMPK inhibitor, dorsmophin, promoted the RA-FLS significantly. Interestingly, both metformin and dorsmophin substantially inhibited the migration of RA-FLS. In FLS, relative expression level of IL-6 mRNA was significantly decreased after metformin treatment, while the expression of AMPK-α1, PKA-α and HAPLN1 genes were significantly increased. Western blot analysis confirmed increased expression of p-AMPK-α1 and HAPLN1 genes in the metformin treated FLS. Conclusions: Inflammatory stress in RA synovium leads to an increase in AMPK levels, possibly as a protective mechanism. AMPK activator but not metformin per se could be a potential therapeutic for RA by promoting HAPLN1 secretion to protect the joints.


Introduction
Growing evidences show the importance of metabolic variations in several autoimmune diseases. For example, inflammation associated macrophages and T-helper 17 cells display a shift towards enhanced glucose uptake, glycolysis and an increased activity of the pentose phosphate pathway. In contrast, anti-inflammatory cells like M2 macrophages, regulatory T cells and quiescent memory T cells exhibit lower glycolytic rates and higher levels of oxidative metabolism (1). The role of fibroblast-like synoviocytes (FLS) in the pathogenesis of rheumatoid arthritis (RA) is increasingly appreciated and recognized as key effector cells (2). The synovium in RA transforms from a quiescent state with relatively acellular structure to a hyperplastic, invasive tissue infiltrated with many inflammatory cells. While the activated FLS are hyper-proliferative to form pannus and acquire resistance to apoptosis, their migration and mobility capacities are also increased.
These properties contribute to the invasion potential of FLS, which by producing inflammatory cytokines like tumor necrosis factor-ɑ (TNF-ɑ) and interlukin-17 (IL-17), chemokines, and matrix-degrading molecules (3) contribute to the destruction of joint cartilage. Recently, choline metabolism was reported to be activated in RA-FLS. Hence, the metabolic variations in RA-FLS could potentially lead to identification of novel signaling pathways and therapeutic agents (4).
Adenosine monophosphate-activated protein kinase (AMPK) is a highly conserved metabolic fuel gauge in eukaryotes that senses changes in the intracellular AMP/ATP ratio in response to energy deprivation by regulating mitochondrial biogenesis (5). Studies have reported that AMPK plays an important role in several inflammatory pathways. For example, AMPK controls the transcriptional regulation of autophagy and lysosomal genes (6); promotes autophagy through phosphorylation of Unc-51 like autophagy activating serine/threonine protein kinase Ulk1 (7); attenuates CD40-mediated inflammatory activities; toll-like receptor (TLR)-induced inflammatory functions and decreases the capacity of antigen presenting cells (8). Some anti-inflammatory agents may function by activating AMPK, a state akin to pseudo-starvation, which alter the metabolism and may participate in the signal-directed programs either by promoting or inhibiting inflammation AMPK-α1 was performed using formalin fixed, paraffin embedded specimens (For detailed methods, see supplementary file 1). Image J was used for analysis.

Analysis of differentially expressed genes between RA and OA synovium
We analyzed the data of synovial tissues from RA and OA patients obtained from NCBI GEO Profiles (accession numbers: GSE12021(11), GSE55235 and GSE55457 (12)). The common differentially expressed genes (DEGs) between RA and OA patients intersecting in all the three datasets were selected for enrichment analysis by submitting to KOBAS 3.0 web tool (http://kobas.cbi.pku.edu.cn/) for KEGG pathway enrichment analysis. Results with corrected p values < 0.05 were selected to plot the bubble diagram.

Quantitative real-time polymerase chain reaction (qPCR)
Synovium specimens from 10 RA and 9 OA patients obtained through knee arthroscopy (for patients' details, see supplementary file 1) were soaked in TRIzol® Reagent (Thermo Scientific, USA) after removing the adipose tissues under aseptic conditions. All the samples were stored in -20℃ freezer until used. Expressions of AMPK-α1, AMPK-α2, AMPK-γ1 and AMPK-γ3 genes were analyzed by qPCR (for methods, see supplementary file 1). To evaluate the effects of metformin on FLS, relative mRNA expression levels of IL-6, AMPK-α1, PKA-α, RAPTOR, mTOR, HAPLN1, RUNX1 and RUNX2 genes were measured by qPCR (Primers list is given in supplementary file1).

Isolation and culture of RA-FLS
FLS were derived from synovial tissue specimens harvested from patients by needle arthroscopy. FLS were isolated by enzyme digestion and subsequently cultured in Dulbecco's modified essential medium (DMEM) containing 10% fetal bovine serum (FBS, Invitrogen) containing antibiotics (penicillin and streptomycin) at 37°C with 5% CO 2 . Cells cultured between passages 4 and 9 were used for this study. Cells were frozen with cell freezing medium and stored in -80℃ freezer until used.

High-throughput mRNA sequencing
High-throughput RNA sequencing was performed with FLS from RA and OA patients (n = 3/group) (for patients' information, see supplementary file 1). Gene expression differences between RA and OA-FLS were investigated using KEGG pathway enrichment analysis.

Methyl thiazolyl terazolium (MTT) assay
MTT assay was used to determine the effects of metformin and dorsomorphin on FLS viability at different concentrations (For detailed method, see supplementary file 1).

Scratch test
Scratch test was performed to evaluate the effects of metformin and dorsomorphin on FLS migration viability (for detailed method, see supplementary file 1).

Automated electrophoresis western blot analysis
FLS treated with metformin (5 mM) or saline for 36 h were examined for relative changes in p-AMPK-α1 and hyaluronan and proteoglycan link protein 1 (HAPLN1) levels by automated electrophoresis followed by western blot analysis (for detailed methods, see supplementary file 1).

Statistical analysis
Statistical analysis was performed using GraphPad Prism 7.0 software. All the data were given as mean ± SD. Differences between two groups were evaluated for statistical significance using Student's t-test. One-way ANOVA with Tukey's multiple comparisons test was used to evaluate the differences among three or more groups. Correlations were evaluated using Liner regression and correlation test. p < 0.05 was considered as statistically significant.

Serum p-AMPK-α1 levels positively correlated with disease activity in RA
To investigate the difference in AMPK levels between RA and OA patients and to detect the presence of correlation between AMPK levels and RA disease activity, we determined AMPK-α1 and p-AMPK-α1 levels using 20 OA and 61 RA patients having different disease activities. No significant differences exist in AMPK-α1 levels between OA and RA patients ( Figure 1A). However, p-AMPK-α1 levels were higher in OA compared to RA patients, who had lower disease activity (p < 0.01) (figure 1B). This significance level was increased when the values for p-AMPK-α1 levels were log transformed (p < 0.0001). Interestingly, after log transformation, RA patients having higher disease activity was found to have significantly higher levels of p-AMPK-α1 compared to patients having low disease activity (p < 0.01) ( Figure 1C). In addition, p-AMPK-α1 levels were positively correlating with DAS28 scores (r = 0.270, 95%CI: 0.142 -0.492, p < 0.0001) and CRP levels (r = 0.259, 95%CI: 0.009 -0.478, p < 0.05) ( Figure 1D and E). However, such a correlation did not exist with ESR levels ( Figure 1F).
IL-17 and TNF-α levels were reported to have correlation with RA activity and thus were currently selected as therapy targets (13,14). We analyzed the presence of possible correlation between AMPK levels with IL-17 and TNF-α in the serum samples. The results demonstrated an increased expression of both the inflammatory cytokines in RA than OA patients, and AMPK-α1 levels were moderately correlated with TNF-α levels (r = 0.46, 95% CI: 0.241-0.640, p = 0.0002) ( Figure 1G, H & I). However, no statistical correlation exists between log transformed p-AMPK-α1 and IL-17 or TNF-α levels (supplementary file 2).
Although AMPK is a key regulator of glucose metabolism (15), and RA and other autoimmune diseases appear to be associated with an increased risk of diabetes mellitus (16), no differences in glucose levels between groups were observed in our study (supplementary file 2).

AMPK levels were more significantly present in RA synovial samples
Since AMPK has an anti-inflammatory effect in many inflammation related diseases (17,18), and our data showed a mild positive correlation with disease activity, we next evaluated the data available in NCBI GEO Profiles [accession numbers: GSE12021(11), GSE55235, and GSE55457 (12)] for AMPK levels from RA and OA patients' synovial tissue samples. In total, 119 DEGs intersected in all the three datasets (supplementary file 3), which demonstrated the high consensus existing between results from different experiments. Although AMPK was not directly detected in the analysis, one of the most obviously changed pathway was the metabolic pathway consisting of 9 genes (Figure 2A).
Since AMPK has a controlling function in metabolism (6,19), we focused on its expression at both protein and mRNA levels in the synovium samples from OA and RA patients. The immunohistochemical staining revealed higher level of p-AMPK-α1 expression in RA than OA synovium ( Figure 2B). Similarly, relative expression levels of AMPK-α1, AMPK-α2 and AMPK-γ3 genes were higher in synovium of RA than OA patients ( Figure 2C).

Metabolism variations in FLS
Phosophorylated AMPK-α1 expressed in the proliferating FLS identified using IHC staining was confirmed by immunofluorescence studies (see supplementary file 2). FLS are the most common cell types present at the pannus-cartilage junction, which contribute to joint destruction through their production of cytokines, chemokines and matrix-degrading

Metformin inhibited FLS migration
Based on the concentration gradient of metformin for its inhibitory effects on FLS proliferation, we selected metformin at 5 mM as well as dorsmorphin at 5 μM to study their effect on FLS migration ability by testing wound repair rate in a scratch test experiment. Interestingly, both metformin and dorsmorphin inhibited FLS migration significantly ( Figure 5). Earlier, dorsmorphin was reported to inhibit the migration of certain cancer cells and this phenomenon was explained by AMPK-independent mechanisms (23).

Metformin increased AMPK-α1 and HAPLN1 expression
Results from semi-quantification of mRNA levels in FLS after metformin treatment by RT-qPCR showed a significant decrease in IL-6 gene expression, while the expressions of AMPK-α1, PKA-α and HAPLN1 genes were significantly increased ( Figure 6A). Automated electrophoresis western blot analysis confirmed the up regulation of p-AMPK-α1 and HAPLN1 at the protein level ( Figure 6H).
The role of IL-6 in the pathogenesis of joint and systemic inflammation in RA has been clearly demonstrated (24), and IL-6 inhibitor has been used for the treatment of RA with favorable outcomes (25). Our results confirmed AMPK-dependent effects of metformin on IL-6 gene expression (26, 27) as we noticed a significant negative correlation (r = -0.422, 95%CI: -0.672 to -0.0865, p = 0.016) between AMPK-α1 and IL-6 gene expressions ( Figure   6B). PKA-α is a regulatory subunit of the cAMP-dependent protein kinases involved in cAMP mediated signaling events in the cells and a mutual promotion effect between AMPK and PKA-α had been reported earlier (28, 29). In this study, we found a significant increase in PKA-α gene expression in FLS after metformin treatment. AMPK also phosphorylates the mammalian target of rapamycin complex 1 (mTORC1) subunit, regulatory associated protein of mTOR (RAPTOR), which is essential for AMPK function as a metabolic checkpoint (30). Although RAPTOR and mTOR did not have any significant changes after metformin treatment, a negative correlation between RAPTOR and AMPK-α1 expression was detected (r = -0.470, 95%CI: -0.682 to -0.185, p = 0.002) ( Figure 6C) confirming an earlier report (31). This pathway was reported to regulate cell growth in response to nutrient and insulin levels.
Interestingly, after treating FLS with metformin, an up-regulation of HAPLN1 expression was observed, which was significantly positive correlated with AMPK-α1 gene expression (r = 0.560, 95%CI: 0.308 to 0.738, p < 0.0001, Figure 6D) as well as at protein level (r = 0.785, 95%CI: 0.3869 to 1.238, p = 0.0015, Figure 6I). HAPLN1 was reported as one of the distinctive genes expressed in RA FLS correlating with the disease activity (32). However, Though metformin as an AMPK activator had inhibitory potential at higher concentrations, it promoted FLS proliferation at lower concentrations in our hands, which is in disagreement with the study of Chen et al (44). They have reported inhibitory capacity of metformin even at very low concentrations (5-60 μM) on FLS proliferation. Since the dose differences used in both the studies were very high, difference in metformin preparations obtained from different manufacturers may not be the major cause. It is plausible that differences in the source of FLS, patient characteristics, genetic variations and treatment conditions could have contributed to these contradictory observations. Most publications reported inhibitory effect of metformin mainly using cancer cells (45,46). An analogy to our observation is the use of methotrexate and radiation, which can be used to treat cancer while being carcinogenic. In fact, pharmacological concentrations of metformin at low or high doses affect cells by different mechanisms (47). At a higher concentration (5 mM), it was proposed to inhibit the respiratory chain complex 1 in intact hepatocytes causing an increase in the AMP/ATP ratio (48). However, more studies are needed to clarify this issue.
Our results are in accordance with metformin effects on the activation of AMPK and in attenuating inflammation (49). We have observed down regulation of IL-6 gene expression, while expression of PKA-α, which exerts a negative effect on AMPK activation (50) was found to be up regulated by metformin treatment. Interestingly, we have identified up-regulation of HAPLN1 expression, which showed a positive correlation with AMPK-α1. In mesenchymoma tissues, HAPLN1 is significantly elevated at both RNA and protein levels (51). HAPLN1 reappears in aggressive hepatocytes that express cytoplasmic β-catenin and stem cell markers, and it is associated with poor disease outcomes (52).
HAPLN1 is also a susceptibility gene for lung cancer (53).

Conclusion
In summary, we tested metformin as an AMPK activator and found its inhibitory role on RA-FLS proliferation and migration. Metformin also affected the expression of many inflammatory and metabolic molecules. Surprisingly we found that at lower concentrations metformin increased the proliferation of RA-FLS. Considering the dosage used in the in vitro studies, it is most unlikely that metformin will be used as an effective drug for the treatment of RA patients. However, it is clear that AMPK pathway is a valuable target for treatment and developing new drugs targeting AMPK activation might be more beneficial for protecting the joints in RA patients.        Metfomin treatment changed the relative mRNA expression levels in FLS.

Consent for publication
Compared to controls, expression of IL-6 gene was significantly decreased in