A novel therapeutic approach for anaplastic thyroid cancer through inhibition of LAT1

A novel therapeutic approach is urgently needed for patients with anaplastic thyroid cancer (ATC) due to its fatal and rapid progress. We recently reported that ATC highly expressed MYC protein and blocking of MYC through its selective inhibitor, JQ1, decreased ATC growth and improved survival in preclinical models. One of the important roles of MYC is regulation of L-neutral amino acid transporter 1 (LAT1) protein and inhibition of LAT1 would provide similar anti-tumor effect. We first identified that while the human ATC expresses LAT1 protein, it is little or not detected in non-cancerous thyroidal tissue, further supporting LAT1 as a good target. Then we evaluated the efficacy of JPH203, a LAT1 inhibitor, against ATC by using the in vitro cell-based studies and in vivo xenograft model bearing human ATC cells. JPH203 markedly inhibited proliferation of three ATC cell lines through suppression of mTOR signals and blocked cell cycle progression from the G0/G1 phase to the S phase. The tumor growth inhibition and decrease in size by JPH203 via inhibition of mTOR signaling and G0/G1 cell cycle associated proteins were further confirmed in xenograft models. These preclinical findings suggest that LAT1 inhibitors are strong candidates to control ATC, for which current treatment options are highly limited.

Xtt assay. The Cell Proliferation Kit II (XTT) assay (Sigma Aldrich, cat. 11465015001; St. Louis, MO) was used according to manufacturer's protocol. Briefly, cells were grown in 96-well microplates in a final volume of 100 μl culture medium per well. The cells were incubated with JPH203 for 24 hours and then 50 μl of the XTT labeling mixture (final XTT concentration 0.3 mg/ml) was added to each well. After incubation of the microplate for 4 hours in 5% CO 2 at 37 °C in humidified incubator, the formazan dye formed was quantitated using a scanning multi-well spectrophotometer (MTP-450 microplate reader, Corona, Hitachinaka, Japan). flow cytometric analysis of cell cycle. Cells were seeded into six-well plates at a density of 3 × 10 5 cells.
After 24 hours, JPH203 at 100 µM concentration was applied for 24 hours at 37 °C. Control cells were treated with DMSO. After treatment, all harvested cells were washed with ice-cold PBS and then fixed with cold 70% (vol/vol) ethanol and stored at −20 °C for at least 2 hours. After this, cells were washed in PBS buffer and then incubated (2019) 9:14616 | https://doi.org/10.1038/s41598-019-51144-6 www.nature.com/scientificreports www.nature.com/scientificreports/ with 1 mg/mL DAPI solution for 10 minutes before being analyzed. The cell cycle profiles were determined using flow cytometry (LSRFortessa II, BD Bioscience, San Jose, CA) and analyzed with FlowJo (FlowJo LLC, Ashland, OR). All experiments were performed at least triplicate.
Knockdown of LAT1 by RNA interference. Short interference RNA (siRNA) against LAT1 was purchased from QIAGEN (Cat. No. 1027416; Hilden, Germany). The target sequences of LAT1 were siRNA_#1; TGGGCTTGTGACATTCGTGAA and siRNA_#2; GAACATTGTGCTGGCATTATA, respectively. The negative control siRNA was also obtained from QIAGEN (Cat. No. 1022076; Hilden, Germany). 8505C, OCUT-2, and OCUT-6 cells at a density of 1 × 10 5 cells in each well were cultured for 16 hours in 6-well plates. The siRNAs were transfected into the cells using the Lipofectamine RNAiMAX reagent (Thermo Fisher Scientific, Cambridge, MA). After transfection, the cells were incubated for 24 hours at 37 °C and subjected to further analyses.
In vivo xenograft tumor assays. All animal experiments were performed under protocols approved by the Animal Care and Use Committee of Wakayama Medical University (No. 877), and all methods involving animals were performed in accordance with the relevant guidelines and regulations. Female athymic nude mice with ages of 6 to 8 weeks old (BALB/c-nu, CAnN.Cg-Foxn1 nu /CrlCrlj, Charles River Laboratories Japan, inc., Yokohama, Japan) were used for the xenograft assays. 8505C cells (1 × 10 6 cells) in 200 μl suspension mixed with Matrigel basement membrane matrix (BD Biosciences, cat. 354234; San Jose, CA) were inoculated subcutaneously into the right flank of mice. After 3 days, the mice were randomly divided into two groups: a control group (n = 5), and JPH203 treated group at 12.5 mg/kg/day concentration (n = 5). JPH203 was first dissolved in DMSO and diluted to 3 volumes of β-cyclodextrin (Sigma Aldrich, cat. H107-5G, St. Louis, MO) in pure water and administered through intraperitoneal injection. The tumor volume was calculated as L x W x H x 0.5236 (mm). JPH203 or vehicle treatment was applied when the median tumor size reached at approximately 100 mm 3 . Tumor growth rate was calculated using the following equation: The tumor growth rate = (V2-V1)/(t2-t1) in which, V2: tumor volume at euthanasia; V1: tumor volume when JPH203 injection was started; t2: the day at euthanasia; t1: the day when JPH203 injection was started.
Treatment was continued until the first mouse reached humane endpoint criteria, upon which all mice were euthanized using isoflurane inhalation. Then, the tumor tissues were collected for further analysis. Formalin-fixed paraffin-embedded sections from the internal organs were analyzed by hematoxylin and eosin (HE) staining according to standard methods. All slides were reviewed by the pathologists, and were photographed using an NIKON Microscope (NIKON, Tokyo, Japan) with standard software.
Statistical analysis. Data are expressed as mean ± standard error (SE). All tests were two-sided and p < 0.05 was considered to be statistically significant. GraphPad Prism version 5.0 for Mac OS X (GraphPad Software, La Jolla, CA) was used to perform analyses of variances.

Results
Overexpression of LAT1 and 4F2hc in human ATC tissues. We first investigated the expression of LAT1, 4F2hc and Ki67 in human ATC tissues by IHC. The distribution of each molecules was summarized in Table 1. Significant LAT1 immunoreactivity, at least score 1, was detected in the cancer tissues of ATC (78%: 11/14 cases) as compared with the adjacent non-cancerous tissues and it was predominantly localized in the cell membrane of cancer cells (Fig. 1). On the other hand, no LAT1 immunoreactivity was observed in all adenomatous goiter (AG) tissues (0%: 15/15 cases). 4F2hc immunoreactivity over the score 2 was detected in the all ATC tissues (100%: 14/14 cases), and it was highly detected in the AG tissues as well (67%: 10/15 cases). Ki67 immunoreactivity over the score 2 was frequently shown in the ATC tissues (85%: 12/14 cases), whereas it was barely detected in AG tissues (6%: 1/15 cases).
To know whether overexpressed LAT1 is associated with survival outcome or not, we further analyzed the survival data and the score of LAT1 expression (Fig. 1m). The patients with strong LAT1 expression of score 2 or 3, were associated with poor prognosis than those with weak or no LAT1 expression of score 1 or 0, but there was no significance (p = 0.2291). Cause-Specific Survival (CSS) at 1 year were 0 and 45.5%, respectively.

Human ATC cells express LAT1and 4F2hc proteins. To confirm the expression of LAT1 and 4F2hc
in human ATC cells, we first performed western blot analysis for 8505C, OCUT-2, and OCUT-6 cells. All three cells were confirmed for the expressions of LAT1 and 4F2hc proteins ( Fig. 2A). We also noticed that the level of LAT1 expression rate was in order in OCUT-2 > 8505C > OCUT-6 ( Fig. 2B) and of 4F2hc expression rate was in order in OCUT-2 > 8505C = OCUT-6 ( Fig. 2C) according to western blot analysis. Additionally, we confirmed the localization of the LAT1 and 4F2hc proteins by immunocytochemistry. Both LAT1 and 4F2hc antibodies accumulated at the cell membrane (Fig. 2D). These data indicate that these three human ATC cells absolutely express the LAT1 together with 4F2hc.

JPH203 inhibits cell proliferation in human ATC cells.
We further investigated whether a selective LAT1 inhibitor, JPH203, inhibits ATC progression or not. To examine the effect on cell viability, 8505C, OCUT-2 and OCUT-6 cells were treated with JPH203 at various concentrations using an XTT assay. As shown in Fig. 3A, when cells were treated with 0.1 to 100 µM JPH203 for 2 days, it potently inhibited cell proliferation in all human ATC cells in a dose dependent manner. To test whether JPH203 was effective in inhibiting LAT1 in human ATC, we evaluated the effect of JPH203 on the proliferation of these three human ATC cells: 8505C, OCUT-2 and OCUT-6 cells. After 96-hour exposure at the concentration of 100 µM, JPH203 showed inhibition of 87.0%, 78.6%, and 75.0% in proliferation of 8505C, OCUT-2 and OCUT-6 cells, respectively (Fig. 3B). Of note, JPH203 was similarly effective in inhibiting cell proliferation irrespective of the genetic background of these three cell lines. We performed further immunofluorescence staining analysis to assess the extent of JPH203-induced www.nature.com/scientificreports www.nature.com/scientificreports/ inhibition of tumor cell proliferation using the cell proliferation marker Ki67. Figure  Previous reports showed that JPH203 inhibited amino acid transport and suppressed the mTOR signals in other cancer types [30][31][32][33][34] . Thus, the three ATC cells prompted us to examine whether the phosphorylated p70S6K, S6 and 4EBP1 proteins' abundance, which are the downstream of mTOR, were affected by JPH203 or not. In accordance with the previous reports in other carcinomas, western blot revealed that JPH203 inhibited the mTOR signaling pathway in the human ATC cells as shown in Fig. 3E. The counting data of western blotting showed statistically significant differences in JPH203 group as compared to the control group in all human ATC cells (Fig. 3F-H).
We therefore examined cell cycle distribution in human ATC cells with or without JPH203 inhibitors. The FACS assay confirmed that treatment with JPH203 increased percentage of G0-G1 phase cells in all human ATC cells (Fig. 4A). Quantified results confirmed the statistically significant increase of G0-G1 phase cells by JPH203 inhibitors (Fig. 4B,C). Consistent with cell cycle distribution findings, western blot assay results also confirmed that cyclin D1, CDK4 and E2F1, the protein associated with G1 restriction checkpoint of cell cycle, were also reduced (Fig. 4D). The counting data of western blot also showed statistically significant differences in JPH203 group as compared to control group in all human ATC cells (Fig. 4E-G).

LAT1 knockdown inhibits cell proliferation in human ATC cells.
To reveal effects of JPH203 as LAT1 knockdown, we further examined whether LAT1 knockdown affected mTOR signaling and subsequent cell proliferation in ATC cells or not. Both siRNA_#1 and _#2 for LAT1 knockdown significantly decreased endogenous LAT1 protein expression in 8505C, OCUT-2, and OCUT-6 (Supplemental Fig. S1). Expressions of 4F2hc and phosphorylated p70S6K were decreased by LAT1 knockdown using siRNA_#1 and _#2. In addition, the cyclin D1 expression was significantly decreased by LAT1 knockdown using each of siRNA_#1 and _#2.
JPH203 suppressed ATC cell growth in mouse xenograft models. Inhibition of cell proliferation in the cultured ATC cells after JPH203 treatment prompted us to assess the effects of JPH203 in vivo using mouse www.nature.com/scientificreports www.nature.com/scientificreports/ xenograft models. We studied the induction of tumor growth through 8505C cell injection in athymic mice because it is the most commonly used ATC cell line. JPH203 administration intraperitoneally decreased the growth ratio of xenograft tumors (Fig. 5A, p < 0.05) and also reduced the tumor size (Fig. 5B). And there was no difference between body weight of the mice treated with JPH203 and body weight of the mice treated with control DMSO (p = 0.1449, Supplemental Fig. S2). Histology of the tumor between control and JPH203 treatment was similar (Fig. 5C). To confirm whether our xenograft model tumors expressed the LAT1 and 4F2hc, we performed immunohistochemistry. There were no differences in terms of the expression of LAT1 and 4F2hc proteins between JPH203-treated mice and control mice. To clarify cell growth suppression mechanism, we performed immunohistochemistry for the expression of a cell proliferation marker Ki67 in mouse tumor implanted models. Reduction of Ki67 positive cells were observed in JPH203 treatment group as compared with control group (Fig. 5C,D, p < 0.001). In addition, we found that cyclin D1 and phosphorylated S6 protein levels were lower in JPH203 treatment group than in control group (Fig. 5E,F; p < 0.05 and p < 0.05, respectively). However, LAT1 and 4F2hc expressions were not changed by JPH203 treatment. These results indicate that JPH203 suppresses tumor growth via suppression of mTOR signals as similar to our in vitro observations.

Discussion
LAT1 expression seems to be one of the promising therapeutical target in various cancers. LAT1 inhibitors, JPH203, BCH, and RNAi, lead to mainly attenuate cell proliferation. Of them, several reports have provided the molecular analysis and showed the anti-tumor proliferation mechanism by western blot assay, through downregulating the phosphorylation of mTOR pathway proteins such as p70S6K, and S6, and upregulating the phosphorylation of 4EBP1 [31][32][33] . Moreover, JPH203 activated the mitochondria-dependent apoptotic signaling pathway by www.nature.com/scientificreports www.nature.com/scientificreports/ upregulating pro-apoptotic factors, such as Bad, Bax, and Bak, and the active form of caspase-9, and downregulating anti-apoptotic factors, such as Bcl-2 and Bcl-xL in human osteosarcoma Saos2 cells 35 . The other groups also showed induction of the apoptosis in oral cancer YD-38 cells 36 . We demonstrated that the tumor proliferation was suppressed by inhibiting LAT1 in ATC preclinical model both in vitro and in vivo. Inhibition of LAT1 causes down regulation of mTOR signaling pathway through suppressing amino acid uptake into tumor cells. In ATC cells of 8505C, OCUT-2, and OCUT-6, the mTOR downstream signals also dramatically influenced in vitro model in this study (Fig. 3E-H). The suppressed mTOR signals led to the G1 cell cycle arrest by decreasing cyclin D1, CDK4, and E2F1 expressions (Fig. 4).
So far two reports provided the preclinical cancer xenograft mouse models of JPH203 administration 37,38 . JPH203 showed anti-tumor efficacy in nude mice bearing human colon cancer and cholangiocarcinoma cell xenografts with doses of 12.5 and 25 mg/kg/day. JPH203 significantly inhibited tumor growth in HT-29 and KKU-213 cell xenografts in the nude mice model in a dose-dependent manner with no toxicity. In our ATC xenograft model, JPH203 administration with a dose of 12.5 mg/day also suppressed the tumor growth through blocking downstream mTOR signaling pathway.
To the best of our knowledge, only two studies exist for targeting LAT1 in thyroid cancer 39,40 . Barollo S et al. reported the overexpression of the LAT1 in patients with human medullary thyroid cancer (MTC) as a part of the neuroendocrine tumors 39 . They showed that the expression of glucose transporter 1 correlated with LAT1 expression in MTC and pheochromocytoma by using RT-PCR and IHC but no prognostic analysis. They also tested the overexpression of LAT1 in MTC cell line, TT, by western blot assay. Their data imply that MTC also overexpresses LAT1, and LAT1 may be therapeutic target in MTC.  www.nature.com/scientificreports www.nature.com/scientificreports/ The other study, which was published very recently, focused on the role of LAT1 in ATC and papillary thyroid cancer (PTC) 40 . Although they reported similar findings to our study, the methodologies are completely different. First of all, Häfliger et al. analyzed LAT1 at mRNA expression level alone. On the other hand, we showed the relationship between LAT1 and histopathological differences in the protein levels by using immuohistochemical (IHC) procedures. Moreover, we evaluated the co-expressed 4F2hc protein, which is the chaperone protein of LAT1. Secondly, Häfliger et al. showed inactivation of mTOR pathway in vitro. However, they did not provide www.nature.com/scientificreports www.nature.com/scientificreports/ any data about cell proliferations. Our current data clearly showed the G1 cell cycle arrest through cyclin D1 and CDK4 reduction and Ki67 expression was also decreased after JPH203 treatment (Figs 3 and 4). We also showed that LAT1 knockdown by siRNA treatment attenuated mTOR signaling and cell proliferation as similar to JPH203 treatment (Supplemental Fig. S1). Finally, Häfliger et al. used BRAF V600E /PIK3CA H1047R double KO mice for establishing in vivo ATC model. This is quite important and there are radical differences. This mouse model was well known as spontaneous ATC model based on the activated MAPK and PI3K pathway. However, it is also well recognized that the pathogenesis of human ATC involved p53 mutation with activated MAPK and PI3K pathway 41 . Our xenograft mouse model using ATC cell line 8505C that consists of BRAF, PI3K3R1/2 and p53 mutations are much popular to investigate pathogenesis of ATC. Based on these facts, our xenograft model is much appropriate for the preclinical evaluation of the effectiveness of JPH203 against ATC. Nevertheless, the difference of experimental design at the same period, their findings strongly support our conclusion. We can conclude that LAT1 inhibitors would be effective therapeutic candidates toward to ATC with strong reliability.
Recently, the novel Boron Neutron Capture Therapy (BNCT) is developed for malignant brain cancer and salivary gland carcinoma 42,43 . It is a binary radiotherapeutic modality based on the nuclear capture and fission reactions that occur when the stable isotope, boron-10, is irradiated with neutrons to produce high energy alpha particles. To deliver boron-10 into cancer cells, the low molecular weight boron-containing drugs, boronophenylalanine (BPA), currently are being used clinically. LAT1 is a well-known key protein for BNCT because LAT1 transport boronophenylalanine into cancer cells 44,45 . In present study, we found that ATC overexpresses LAT1 protein, and then our findings imply that BNCT may also be effective for ATC.
Our study also has some limitations. Firstly, we analyzed 14 ATC patients and 15 AG patients as control for histological analysis. Sample number is relatively small, and cannot show the statistical difference in survival data as shown in Fig. 1m. Secondly, we did not know the effectiveness of JPH203 for other ATC cell lines such as THJ-16, 8305C, ARO, and KTC-2, which have different gene mutations than the cells used in the current study.
A phase I clinical trial using JPH203 was completed (UMIN000016546) and a phase II clinical trial for advanced cholangiocarcinoma has just started in Japan (UMIN000034080). These trials will demonstrate promising results.
In conclusion, our study suggests that inhibition of LAT1 by JPH203 would be a novel molecular targeted therapy for one of the most fatal cancer types, ATC.