Fibroblast‑like synoviocyte migration is enhanced by IL‑17‑mediated overexpression of L‑type amino acid transporter 1 (LAT1)via the mTOR/4E‑BP1 pathway

Abstract
In rheumatoid arthritis (RA), activated synovial fibroblasts have the ability to invade joint cartilage, actively contributing to joint destruction in RA. The mechanisms underlying this cell migration and invasion remain unclear. Our previous results and data from the GEO profile indicate that the L-type amino acid transporter gene, LAT1, is overexpressed in the synovium of RA. To identify its potential role in RA, fibroblast-like synoviocytes (FLS) from patients with RA were used to determine the effects of suppressing the LAT1 genes using RNA interference and the LAT inhibitor, BCH. We found that BCH exposure reduced the phosphorylation of mTOR and its downstream target 4EBP1, radiolabeled leucine uptake, and migration of RA FLS. LAT1 silencing by siRNA presented effects similar to BCH inhibition. Treatment of cells with IL-17 stimulated the expression of LAT1. In contrast, applying an inhibitor of mTOR pathway, temsirolimus, or silencing eIF4E neutralized the stimulation of IL-17 on LAT1. BCH and siLAT1 also resulted in lower IL-17-stimulated leucine uptake and cell migration. These results suggest that the migration of RA FLS is aggravated by IL-17-mediated overexpression of LAT1 via mTOR/4E- BP1 pathway. In conclusion, further investigation is warranted into LAT1 as a potential target for drug therapies aimed at attenuating migration of transformed-appearing fibroblasts and subsequently preventing further erosion of bone and cartilage.

Introduction
Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by synovial hyperplasia and progressive joint destruction. Accumulating evidence suggests that activated rheumatoid arthritis synovial fibroblasts (RA FLS), the main cell type in the terminal layer of the hyperplastic synovial tissue, are key players in the pathogenesis of RA (Pap et al. 2005). Recently, RA FLS have been shown to be able to migrate to non-affected areas and joints (Neumann et al. 2010a). Thus, the effectors that enable cell migration may be potential targets in preventing RA progression.
Our previous microarray results (data not published) indicate that LAT1 is overexpressed in RA patients. LAT1 belongs to the L-type amino acid transporters family, which are part of a major nutrient transporter system that supplies cells with large neutral amino acids such as leu- cine, isoleucine, and valine (Neumann et al. 2010a). These amino acids are not only required for protein synthesis, but also contribute to amino acids disorders (Rosario et al. 2013); LAT1 may therefore be related to cell activity. In line with this, LAT1 was reported to correlate with inva- sion and migration of cholangiocarcinoma cells (Janpipat- kul et al. 2014). This evidence is intriguing in relation to RA, since RA FLS share many characteristics with tumor cells, such as reduced apoptosis, increased proliferation, and invasion (Neumann et al. 2010b). Thus, LAT1 could play an important role in affecting migration of RA FLS. Further, evidence indicates that activated RA FLS exhibit more active amino acid metabolism (Ahn et al. 2015). The intricate cellular balance of amino acid influx and efflux is maintained by the L-system of protein transporters via the mTORC1 pathway (Zhenyukh et al. 2017). Sinclair et al. (2008) reported that the mTOR pathway controls T lym- phocyte migration. Taken together, this evidence prompted us to hypothesize that LATs play a role in the activation of RA FLS. Here, we investigate the role of LAT1 in the development of RA to uncover mechanisms behind the invasion of FLS into the synovium.

Results
In a preliminary microarray analysis, we observed that LAT1 was overexpressed in RA ST (data not shown). To confirm this result, expression of LAT1 in ST obtained from RA patients (N = 10) and OA patients (N = 10) was measured using real-time PCR. The expression levels of LAT1 were found to be higher in RA ST than in OA ST (p < 0.001, Fig. 1).LAT1 transports large neutral amino acids, such as leu- cine, isoleucine, and valine (Yanagida et al. 2001). These findings prompted us to hypothesize that LAT1 may affect amino acid influx and efflux in RA FLS, and thereby affect proliferation, migration, and invasion of RA FLS. Based on our finding that LAT1 is overexpressed in RA, we investigated whether its overexpression might play a role in RA FLS proliferation and migration. RNA interference was used to silence LAT1 expression and assess the effect on RA FLS. LAT1 expression was analyzed by RT-qPCR and western blotting 24 h after transfection with either control siRNA or LAT1 siRNA. The expression of LAT1 decreased significantly at both the mRNA (**p < 0.01) and protein level (Fig. 2a). As CD98 is the heavy chain region of LAT1, we also tested whether siLAT1 can impact CD98; CD98 was unchanged in our experiments (data not shown).We next examined cell growth using the MTT assay. We found that siLAT1 had no effect on RA FLS proliferation compared to control (p > 0.05, Fig. 2b).To evaluate whether LAT1 is involved in the migration of RA FLS, transwell migration assays were performed after the impairment of LAT1 expression. The transwell migration assay showed less RA FLS migration after LAT1 expression was impaired with siRNA (Fig. 2c). We then used a leucine analogue and universal inhibitor of LATs to investigate whether decreased amino acid uptake can cause less RA FLS migration. We evaluated the migra- tion capabilities of the cells in the presence or absence of BCH and found that cell migration decreased by 30–40% in the presence of BCH (***p < 0.001, Fig. 2c). This sug- gests that LAT1 may impact RA FLS migration through an effect on amino acid uptake.

Leucine is an essential amino acid for protein synthesis. To determine whether LAT1, an amino acid transporter, is responsible for leucine uptake in RA FLS, we treated cells with the leucine analogue BCH, a competitive inhibi- tor of LATs, and measured leucine levels inside the cells. [3H]-L-Leucine uptake was reduced after LAT1 expression was inhibited by siRNA (***p < 0.001, Fig. 2d). This find- ing was confirmed using BCH treatment, which decreased [3H]-L-leucine uptake by 20–30% in RA FLS (*p < 0.05, Fig. 2d). Together, these results suggest that LATs do indeed play a role in determining leucine uptake in RA FLS.cine-free media with [3H]-L-leucine for 3 min to assess amino acid uptake. BCH and siLAT1 can significantly reduce leucine uptake in RA FLS (***p < 0.001). e Treatment with BCH and downregulated LAT1 using siRNA can decreased the phosphorylation of mTOR and its downstream target 4EBP1 in RA FLS cells. f The expression of MMP-2, 3, 9, and 13 in RA FLS has no significant difference in the presence of BCH or siRNA (p > 0.05). All data are presented as the mean ± SEM of three independent experiments performed in tripli- cateSince amino acids are activators of the mammalian target of rapamycin (mTOR), which is a major player in nutri- ent signaling and can regulate the invasive properties of synovial fibroblasts in rheumatoid arthritis (Laragione and Gulko 2010), we next investigated whether LAT1 mediates mTOR activity. Using western blotting, we found that BCH inhibition and siRNA knockdown of LAT1 both resulted in decreased phosphorylation of mTOR and its downstream target 4EBP1 (Fig. 2e).

Matrix metalloproteases (MMP) degrade extracellular matrix in both physiological and pathological processes, including RA (Itoh 2015). To investigate the mechanism underlying the relationship between LAT1 and RA FLS migration, we assessed the impact of LAT1 inhibition by BCH and siRNA on the expression of MMP in RA FLS. Using RT-qPCR, we analyzed MMP-2, 3, 9, and 13 expres- sion in RA FLS with and without inhibition, and found no significant differences (p > 0.05, Fig. 2f).Because LAT1 is elevated in RA synovium and because syn- ovial fluid is known to contain proinflammatory cytokines (Du et al. 2008), we evaluated whether the expression of LAT1 is stimulated by these cytokines. We stimulated RA FLS with three well-known inflammatory effectors found in RA synovial fluid: IL-17, LPS, and TNF-α. IL-17 consist- ently and significantly stimulated the expression of LAT1 (Fig. 3a, ***p < 0.001) in a dose- and time-dependent man- ner (see Supplementary Fig. 1). The expression reached a maximum level with 50 ng/mL of IL-17, so this concentra- tion was used in subsequent experiments. To confirm this result, we used western blotting to detect the expression of LAT1 in the presence of IL-17 (50 ng/mL) for 12 h. Here, LAT1 was significantly upregulated (Fig. 3b). LPS and TNF-α had no effect on expression of LAT1 (Fig. 3a).To investigate which signaling pathways are responsible for IL-17-induced expression of LAT1 in RA FLS, we used well-established inhibitors of several signaling pathways to determine which might play a role in the IL-17-related sign- aling.

The compounds we used including PDTC, an inhibi- tor of NF-κB activation; PD98059, an inhibitor of ERK1/2 12 h. RA FLS were treated with IL-17 (50 ng/mL) and temsirolimus for 24 h. IL-17-stimulated expression of LAT1 in RA FLS was mark- edly decreased in the presence of mTOR inhibitors at both the mRNA (c) and protein (d) level (***p < 0.001). e si-eIF4E can inhibit the induction of IL-17 on LAT1 in RA FLS. All data are presented as the mean ± SEM of three independent experiments performed in tripli- cate activation; Stattic, an inhibitor of STAT3; and temsirolimus, an inhibitor of mTOR. None of these compounds exhibited cytotoxicity at the concentrations used in our experiments (data not shown). IL-17-stimulated expression of LAT1 mRNA (Fig. 3c) and protein (Fig. 3d) in RA FLS was mark- edly decreased in the presence of the mTOR inhibitor. Other signaling pathway inhibitors had little effect on the expres- sion of LAT1 (data not shown). These results imply that IL-17 may regulate the expression of LAT1 in RA FLS via the mTOR pathways. IL-17-mediated expression of LAT1 via mTOR pathway was further confirmed by the results of eIF4E (downstream target of mTOR) siRNA transfection analysis (Fig. 3e).

Since IL-17 was able to up-regulate LAT1 expression in RA FLS, we continued to explore the role of IL-17 in LAT1-mediated amino acids uptake. Stimulation of RA FLS with IL-17 resulted in enhanced leucine uptake, and this effect was neutralized by BCH or siRNA targeting LAT1 (***p < 0.001, Fig. 4a).Since leucine uptake was enhanced by IL-17-medi- ated overexpression of LAT1, we continued to explore whether enhanced expression of LAT1 might be involved stimulation with IL-17 (50 ng/mL) (***p < 0.001) and this effect was neutralized by BCH and siRNA targeting LAT1 (**p < 0.01,
***p < 0.001). All data are presented as the mean ± SEM of three independent experiments performed in triplicate in IL-17-stimulated RA FLS migration. As expected, we observed that RA FLS migration was significantly increased by stimulation with IL-17 (50 ng/mL), and this effect was attenuated by BCH or siRNA targeting LAT1 (***p < 0.001, Fig. 4b). These results suggest that IL-17 positively regu- lates the expression of LAT1 in RA FLS, which in turn, plays a role in mediating the stimulatory effect of this pro- inflammatory cytokine on RA FLS migration.

Discussion
In our present study, we found that the L-type amino acid transporter, LAT1, was overexpressed in surgically resected synovium from RA patients. This result is consistent with data from the GEO profile (http://www.ncbi.nlm.nih.gov/ geoprofiles/44325905). To investigate the role of LAT1 in RA FLS, siRNA and an inhibitor of LAT1, BCH, was used to inhibit expression of LAT1. Proliferation and migration were tested using MTT and transwell assays, respectively. To verify that the siRNA process has no effect on LAT1 expres- sion or cell proliferation, we used a scrambled RNA con- trol (NC group) that shared transfection reagents, medium, and experiment conditions. siRNA knockdown of LAT1 for 24 h slowed the velocity in proliferation comparing with NC group, but the difference was not statistically significant. The proliferation rates of two groups at 48 h were almost the same. Thus, transfection conditions did not affect the cells, but knockdown of LAT1 did.Though inhibition of LAT1 had no effect on proliferation, it resulted in reduced migration of RA FLS. BCH-induced inhibition of LAT1 produced similar results to siLAT1. These findings are consistent with those reported in pros- tate cancer LNCap cells (Wang et al. 2011) and in human ovarian cancer cells (Fan et al. 2010), in which proliferation of LNCap and SKOV3 (S1) or IGROV1 (I10) cells were all unaffected by LAT1 knockdown. However, others reported that cell proliferation of prostate PC-3 (Wang et al. 2011) and breast cancer MDA-MB-231 cells (Liang et al. 2011) was indeed decreased after LAT1 knockdown. Given these discordant findings, we suggest that LAT1 may play different roles in different cell types; that is, LAT1 may affect both proliferation and migration in some cell types, but may not affect proliferation in other cell types.

A high concentration of amino acids can promote migra- tion of human peripheral blood mononuclear cells (Zheny- ukh et al. 2017). We hypothesized that overexpressed LAT1 may induce more amino acids influx into RA FLS, thereby promoting their migration. In fact, we found that inhibition of LAT1 expression decreased radiolabelled [3H]-L-leucine uptake by 45%, while treatment with the leucine analogue BCH, which is a specific, competitive inhibitor of LATs, presented a relatively low effect to 25%. These findings are consistent with prior work (Wang et al. 2011) in LNCap prostate cancer cell [about 22%, see Wang et al. (2011), Fig. 3d]. Our uptake assay methodology was similar to that study, and we repeated the experiments with good data reproducibility. Thus, we believe these data are reliable.RA FLS secrete various proteases, including MMPs, that degrade extracellular matrix (ECM) components, mainly proteoglycans and collagens, of articular cartilage in the affected joints (Frisenda et al. 2013). MMPs play a critical role in the progressive joint destruction pathogenesis of RA (Itoh 2015). Here, despite LAT1’s role in cell migration, this protein appeared to have no effect on MMP secretion from RA FLS. mTOR signaling is reported to be hyperactive in RA, and amino acids are known to be involved in the mTOR pathway (Babu et al. 2003; Bodoy et al. 2005; Fukasawa et al. 2000). mTOR is also involved in regulation of invasive properties of synovial fibroblasts in RA. Here, we showed that BCH and si-LAT1 can inhibit mTOR activation by attenuating the phosphorylation of mTOR and its downstream target, 4EBP1. Thus, LAT1 may affect RA FLS through the mTOR pathway.

As RA is a chronic inflammatory disease (Siebert et al. 2015), and LPS, TNF-α, and IL-17 are well-known effectors in RA that can up-regulate the expression of many genes (Zhang et al. 2009). IL-17 is an important pro-inflammatory cytokine involved in RA pathogenesis. In vitro, IL-17 can up-regulate and/or synergize with local mediators of inflam- mation, such as TNF-α and IL-1β (Katz et al. 2001). LPS is a component of gram-negative bacteria cell walls and causes inflammatory reactions (Brandenburg et al. 2016). Our observation that LAT1 plays a role in RA FLS migration led us to explore the possibility that LAT1 might be func- tionally linked to IL-17, TNF-α, or LPS. While IL-17 stimu- lated the expression of LAT1 in a dose- and time-dependent manner, LPS and TNF-α did not. RA FLS migration was promoted by stimulation with IL-17, and neutralized by BCH or siRNA targeting LAT1. Further, siLAT1 reduced migration ability of RA FLS by 48%, while in the presence of IL-17, siLAT1 reduced the IL-17-stimulated migration by 80%. Thus, under the inflammatory status, that is, in the presence of IL-17, knock down of LAT1 attenuates the migration of RA FLS at a higher level than in the absence of IL-17 stimulation. Inhibition with BCH acts differently than siRNA knockdown, in that BCH, 2-aminobicyclo-(2,2,1) heptane-2-carboxylic acid, a non-metabolizable analogue of L-leucine, can competitively inhibit amino acid transport of the L-amino acids transport system (Malaisse-Lagae et al. 1982) rather than affecting LAT1 expression. The inhibition rate of BCH on RA FLS migration with and without IL-17 stimulation was nearly the same. We hypothesized that, as a competitive inhibitor, BCH resembles the substrate and binds to the active site of the LATs; the substrate leucine is thereby prevented from binding to the same active site.

At any given inhibitor concentration, competitive inhibition can be relieved by increasing the substrate concentration (Berg et al. 2002); such competitive inhibition is a reversible reaction. Further, an increase in normal substrate concentra- tion can antagonize uptake of the analogue. In our study, IL-17 can increase expression of LAT1, and increased LAT1 should result in enhanced amino acid uptake. Increased leu- cine concentration will antagonize uptake of BCH by RA FLS. We think this phenomenon prevents BCH effects from being as high as those of siLAT1. These results indicate that IL-17 may positively regulate the expression of LAT1 in RA FLS, which in turn, play a role in mediating the stimulatory effect of this pro-inflammatory cytokine on RA FLS migra- tion. Further, using inhibitors of several pathways, includ- ing NF-KB, ERK1/2, STAT3, and mTOR, in combination with IL-17 stimulation, only the mTOR pathway mediated the effects on LAT1. IL-17 can activate mTOR by phos- phorylation (Saxena et al. 2011), and in line with this we found that knockdown of a protein downstream of mTOR, eIF4E (Brunn et al. 1996; Qin et al. 2016), resulted in less stimulation of IL-17 on LAT1. This finding confirms that the mTOR pathway was involved in mediating stimulation of LAT1 by IL-17.

In summary, our results suggest that the expression of
LAT1 in RA FLS is upregulated by IL-17 via the mTOR signaling pathway. Moreover, there appears to be an IL-17–mTOR–LAT1–mTOR loop in RA FLS. We propose that LAT1 may be a new and promising strategy targeting cellular amino acid pools as a way to combat erosion activity of synovium in RA.
Synovial tissues were collected during knee joint replace- ment surgery from patients with RA (n = 15, 9 female, age 29–72 years, mean 51 years) and patients with OA (n = 15, 7 female, age 39–77 years, mean 62 years). All of the patients fulfilled the American College of Rheuma- tology (ACR) diagnosis criteria for RA and OA. The RA patients had disease duration of 3–9 years and were clas- sified as having erosive RA (Larsen class IV–V). These patients presented high levels of C-reactive protein (CRP) (32–298 mg/L, mean 73 mg/L), anti-cyclic citrullinated peptide (anti-CCP) (28–458 U/mL, mean 279.6 U/mL) and rheumatoid factor (RF) (38–316 U/mL, mean 183.2 U/ mL). Participants provided written informed consent to participate in the study and to allow their biological sam- ples to be genetically analyzed. The Ethical Committee of the Shandong Academy of Medicinal Sciences approved this study and the approval number is 2014-19. The study was conducted in compliance with the Temsirolimus Helsinki Agree- ment or with the research ethics standards of the country of origin of the report.