Clinical Role of ASCT2 (SLC1A5) in KRAS-Mutated Colorectal Cancer

Mutation in the KRAS gene induces prominent metabolic changes. We have recently reported that KRAS mutations in colorectal cancer (CRC) cause alterations in amino acid metabolism. However, it remains to be investigated which amino acid transporter can be regulated by mutated KRAS in CRC. Here, we performed a screening of amino acid transporters using quantitative reverse-transcription polymerase chain reaction (RT-PCR) and then identified that ASCT2 (SLC1A5) was up-regulated through KRAS signaling. Next, immunohistochemical analysis of 93 primary CRC specimens revealed that there was a significant correlation between KRAS mutational status and ASCT2 expression. In addition, the expression level of ASCT2 was significantly associated with tumor depth and vascular invasion in KRAS-mutant CRC. Notably, significant growth suppression and elevated apoptosis were observed in KRAS-mutant CRC cells upon SLC1A5-knockdown. ASCT2 is generally known to be a glutamine transporter. Interestingly, SLC1A5-knockdown exhibited a more suppressive effect on cell growth than glutamine depletion. Furthermore, SLC1A5-knockdown also resulted in the suppression of cell migration. These results indicated that ASCT2 (SLC1A5) could be a novel therapeutic target against KRAS-mutant CRC.


Introduction
Colorectal cancer (CRC) is one of the most common cancers worldwide; therefore, development of novel diagnostic measures and treatment is very important [1]. KRAS mutations are found in approximately 40% of CRC cases [2][3][4]. A number of clinical trials have shown that KRAS mutations in CRC can predict a lack of responses towards anti-epidermal growth factor receptor (EGFR)-based therapy [2][3][4]. Therefore, development of new therapy for CRC with mutated KRAS has been desired clinically. Some studies have investigated the correlation between KRAS mutations and metabolic alterations in pancreatic and lung cancers [5][6][7][8][9] as well as in CRC [10][11][12][13][14][15]. We have recently reported that, using metabolome analysis, concentration of amino acids is elevated in CRC cells with mutated KRAS compared to CRC cells with wild-type KRAS [12]. The increase in glucose transporter 1 (GLUT1) expression and glucose uptake was critically dependent on mutated KRAS [16][17][18]. However, it remains to be investigated which amino acid transporter is specifically regulated by mutated KRAS in CRC. In the present study, we performed a screening of amino acid transporters in KRAS-mutant CRC cells transfected by siKRAS and found that ASCT2 (SLC1A5) was particularly up-regulated through KRAS signaling.
However, the role of ASCT2 in CRC has not yet been reported. In addition to glucose, glutamine is an essential source of cellular building blocks to fuel cell proliferation. Recent studies have established a better understanding about the importance of glutamine as a critical nutrient in fast growing cancer cells [32][33][34]. In the present study, we investigated the significance of ASCT2 expression in CRC using in vitro cultures and clinical samples.

SLC1A5 (ASCT2) Is Regulated through KRAS Signaling in KRAS-Mutant CRC Cells
We have recently reported that mutated KRAS induces metabolic alterations in many amino acids [12]. Therefore, we hypothesized that the expression of amino acid transporters might be regulated by mutated KRAS. Several amino acid transporters (SLC1A5, SLC7A5, SLC7A11, SLC3A2, and SLC43A1) have been reported to be up-regulated in different cancers [32]. To determine the specific transporter that could be regulated by mutated KRAS, we introduced two different small interfering RNAs (siRNAs) targeting KRAS in KRAS-mutant CRC cell lines (HCT116 and DLD-1). We confirmed that siKRAS significantly reduced the mRNA levels of KRAS in both cell lines ( Figure S1a). Interestingly, KRAS-knockdown significantly reduced SLC1A5 expression in both KRAS-mutant cell lines ( Figure 1a). SLC1A5 (ASCT2) is a known glutamine transporter. Next, we investigated whether glutamine transporters other than SLC1A5 (i.e., SLC1A4, SLC38A1, SLC38A2, SLC38A3, and SLC38A5) could be regulated by KRAS signaling [33]. Expression levels of SLC1A4 and SLC38A1 were decreased after KRAS-knockdown in HCT116; however, their expression was not decreased in DLD-1 (Figure 1b). We also found that KRAS-knockdown significantly reduced protein expression of ASCT2 in both KRAS-mutant cell lines (Figure 1c). The mutated KRAS continuously activates both Raf/MEK/ERK and PI3K/Akt/mTOR pathways. To investigate which pathway regulates ASCT2 expression, we used specific inhibitors of each pathway. Western blot analysis revealed that ASCT2 expression was dramatically reduced in KRAS-mutant CRC cell lines by addition of LY 294002 (PI3K inhibitor) or rapamycin (mTOR inhibitor), which suggested that KRAS signaling may regulate ASCT2 expression in CRC mainly via the PI3-Akt-mTOR pathway (Figure 1d).

Knockdown of SLC1A5 (ASCT2) Results in Suppression of Cell Growth
To investigate the role of SLC1A5 (ASCT2) in CRC cell lines with mutated KRAS, we introduced non-silencing siRNA and two different siRNAs targeting SLC1A5 (referred as siSLC1A5#1 and siSLC1A5#2) into CRC cell lines ( Figure S2). Knockdown of SLC1A5 (ASCT2) significantly suppressed the cell growth in all the 3 cell lines with mutated KRAS (HCT116, DLD-1, and SW480), whereas in 1 out of 3 cell lines with wild-type KRAS (RKO) (Figure 3a). Furthermore, we investigated the knockdown effect of SLC1A5 on cell apoptosis. Knockdown of SLC1A5 induced a significant increase in caspase 3/7 activities in all the 3 cell lines with mutated KRAS (HCT116, DLD-1, and SW480), whereas in 2 out of 3 cell lines with wild-type KRAS (HT29 and RKO) (Figure 3b). Oncogenic PIK3CA mutations were reported to reprogram glutamine metabolism in CRC [35]. PIK3CA mutations are observed in HCT116 (a H1047R mutation), DLD-1 (E545K; D549N mutations), HT29 (a P449T mutation), WiDR (a P449T mutation), and RKO (a H1047R mutation), whereas the PIK3CA status is wild-type in SW480. The PIK3CA status might be related to the differences in the knockdown effect of SLC1A5 between cell lines.

Knockdown of SLC1A5 (ASCT2) Results in Suppression of Cell Growth
To investigate the role of SLC1A5 (ASCT2) in CRC cell lines with mutated KRAS, we introduced non-silencing siRNA and two different siRNAs targeting SLC1A5 (referred as siSLC1A5#1 and siSLC1A5#2) into CRC cell lines ( Figure S2). Knockdown of SLC1A5 (ASCT2) significantly suppressed the cell growth in all the 3 cell lines with mutated KRAS (HCT116, DLD-1, and SW480), whereas in 1 out of 3 cell lines with wild-type KRAS (RKO) (Figure 3a). Furthermore, we investigated the knockdown effect of SLC1A5 on cell apoptosis. Knockdown of SLC1A5 induced a significant increase in caspase 3/7 activities in all the 3 cell lines with mutated KRAS (HCT116, DLD-1, and SW480), whereas in 2 out of 3 cell lines with wild-type KRAS (HT29 and RKO) (Figure 3b). Oncogenic PIK3CA mutations were reported to reprogram glutamine metabolism in CRC [35]. PIK3CA mutations are observed in HCT116 (a H1047R mutation), DLD-1 (E545K; D549N mutations), HT29 (a P449T mutation), WiDR (a P449T mutation), and RKO (a H1047R mutation), whereas the PIK3CA status is wild-type in SW480. The PIK3CA status might be related to the differences in the knockdown effect of SLC1A5 between cell lines. Cell proliferation measured by CCK-8 assay. CRC cells were transfected with negative control (NC) or two independent siSLC1A5 and cultured for 72 h. Viability in each siSLC1A5 was normalized to that in NC. Student's t-test; * p < 0.05; (b) caspase 3/7 activities measured by Caspase-Glo assay. CRC cells transfected with negative control (NC) or two independent siSLC1A5 were cultured for 72 h. Caspase 3/7 activity was normalized to the cell viability measured by CCK-8 assay under the same density and conditions. Mean; bars, ± SD, n = 3 (Student's t-test; * p < 0.05).

Role of SLC1A5 (ASCT2) in KRAS-Mutant CRC Cells
SLC1A5 (ASCT2) is generally regarded as a glutamine transporter. In a KRAS-mutant CRC cell line (HCT116), glutamine depletion resulted in decreased cell proliferation and enhanced caspase 3/7 activities. Importantly, even in the presence of glutamine, siSLC1A5 dramatically suppressed cell proliferation and up-regulated caspase 3/7 activities (Figure 4a,b), which indicated that the effect of SLC1A5-knockdown was more prominent on cell growth and apoptosis than glutamine depletion. To further investigate the functional role of SLC1A5, we established stable HCT116 transfectant cell lines in which SLC1A5 was knocked down by shRNA constructs targeting SLC1A5 (referred as shSLC1A5#1 and shSLC1A5#2) ( Figure S3). In the clonogenic assay, SLC1A5-knockdown significantly suppressed colony number as compared to the control (Figure 4c). Moreover, in the wound healing assay, SLC1A5-knockdown significantly inhibited wound closure as compared to the control ( Figure   Figure 3. SLC1A5 knockdown inhibits cell proliferation and induces cell apoptosis of CRC cells. (a) Cell proliferation measured by CCK-8 assay. CRC cells were transfected with negative control (NC) or two independent siSLC1A5 and cultured for 72 h. Viability in each siSLC1A5 was normalized to that in NC. Student's t-test; * p < 0.05; (b) caspase 3/7 activities measured by Caspase-Glo assay. CRC cells transfected with negative control (NC) or two independent siSLC1A5 were cultured for 72 h. Caspase 3/7 activity was normalized to the cell viability measured by CCK-8 assay under the same density and conditions. Mean; bars, ± SD, n = 3 (Student's t-test; * p < 0.05).

Role of SLC1A5 (ASCT2) in KRAS-Mutant CRC Cells
SLC1A5 (ASCT2) is generally regarded as a glutamine transporter. In a KRAS-mutant CRC cell line (HCT116), glutamine depletion resulted in decreased cell proliferation and enhanced caspase 3/7 activities. Importantly, even in the presence of glutamine, siSLC1A5 dramatically suppressed cell proliferation and up-regulated caspase 3/7 activities (Figure 4a,b), which indicated that the effect of SLC1A5-knockdown was more prominent on cell growth and apoptosis than glutamine depletion. To further investigate the functional role of SLC1A5, we established stable HCT116 transfectant cell lines in which SLC1A5 was knocked down by shRNA constructs targeting SLC1A5 (referred as shSLC1A5#1 and shSLC1A5#2) ( Figure S3). In the clonogenic assay, SLC1A5-knockdown significantly suppressed colony number as compared to the control (Figure 4c). Moreover, in the wound healing assay, SLC1A5-knockdown significantly inhibited wound closure as compared to the control (Figure 4d). Taken together, these results indicate that the inhibition of SLC1A5 (ASCT2) could be a therapeutic target in KRAS-mutant CRC.   Table 1 shows the relationship between ASCT2 expression and clinicopathologic variables. ASCT2 expression was significantly correlated with tumor location, but not with age, sex, tumor size, stage, T-/N-/M-category, lymphatic invasion, or vascular invasion. We further investigated the clinical significance of ASCT2, based on the KRAS mutational status. Interestingly, we found that high ASCT2 expression was significantly associated with tumor depth and vascular invasion in KRAS-mutant CRC, which was not observed in wild-type KRAS CRC. Mean; bars, ± SD, n = 3 (Student's t-test; * p < 0.05); (c) clonogenic assay with HCT116 transfected with control or two independent shSLC1A5 vectors. Cells were maintained under 4 mM glutamine condition containing 10% fetal bovine serum (FBS) for 10 days. Mean; bars, ± SD, n = 3 (Student's t-test; * p < 0.05); (d) wound healing assay with HCT116 transfected with control or shSLC1A5 vector. Cells were photographed at 50× magnification at 0, 24, and 48 h. Wound closure (%) was evaluated. Mean; bars, ± SD, n = 3 (Student's t-test; * p < 0.05). Table 1 shows the relationship between ASCT2 expression and clinicopathologic variables. ASCT2 expression was significantly correlated with tumor location, but not with age, sex, tumor size, stage, T-/N-/M-category, lymphatic invasion, or vascular invasion. We further investigated the clinical significance of ASCT2, based on the KRAS mutational status. Interestingly, we found that high ASCT2 expression was significantly associated with tumor depth and vascular invasion in KRAS-mutant CRC, which was not observed in wild-type KRAS CRC.

Patients' Prognosis
To evaluate the relationship between ASCT2 expression and patients' prognosis, we performed the log-rank test analysis with CRC patients who underwent curative resection of primary CRC (n = 90). Kaplan-Meier survival curves indicated that ASCT2 expression was not significantly correlated with recurrence-free survival (RFS) in all cases (Figure 5a). However, in KRAS-mutant CRC cases (n = 38), the estimated RFS rate at 5-year tended to be lower in the high ASCT2 group than in the low ASCT2 group (52.9% vs. 70.2%; p = 0.251) (Figure 5b, right). On the other hand, in wild-type KRAS CRC cases (n = 52), the estimated RFS rate at 5-year was almost similar between the high and low ASCT2 groups (84.6% vs. 75.8%; p = 0.513) (Figure 5b, left). Taken together, high ASCT2 expression can be one of the crucial prognostic factors in KRAS-mutant CRC.

Patients' Prognosis
To evaluate the relationship between ASCT2 expression and patients' prognosis, we performed the log-rank test analysis with CRC patients who underwent curative resection of primary CRC (n = 90). Kaplan-Meier survival curves indicated that ASCT2 expression was not significantly correlated with recurrence-free survival (RFS) in all cases (Figure 5a). However, in KRAS-mutant CRC cases (n = 38), the estimated RFS rate at 5-year tended to be lower in the high ASCT2 group than in the low ASCT2 group (52.9% vs. 70.2%; p = 0.251) (Figure 5b, right). On the other hand, in wild-type KRAS CRC cases (n = 52), the estimated RFS rate at 5-year was almost similar between the high and low ASCT2 groups (84.6% vs. 75.8%; p = 0.513) (Figure 5b, left). Taken together, high ASCT2 expression can be one of the crucial prognostic factors in KRAS-mutant CRC.

Discussion
KRAS mutations are found in a variety of human cancers, including pancreatic cancer, non-small cell lung cancer, and CRC. Recent studies have shown that mutated KRAS promotes metabolic reprogramming through nutrients uptake, glycolysis, glutaminolysis, and synthesis of nucleotides

Discussion
KRAS mutations are found in a variety of human cancers, including pancreatic cancer, non-small cell lung cancer, and CRC. Recent studies have shown that mutated KRAS promotes metabolic reprogramming through nutrients uptake, glycolysis, glutaminolysis, and synthesis of nucleotides and fatty acids. The mechanism by which mutated KRAS coordinates the metabolic reprogramming to promote tumor growth remains to be investigated. The International CRC Subtyping Consortium has suggested that CRC can be divided into four subtypes with distinguished features: CMS1, CMS2, CMS3, and CMS4 [15]. Notably, CMS3 is characterized by metabolic dysregulation and is strongly associated with KRAS mutations. Using a comprehensive metabolomics analysis with isogenic CRC cell lines harboring mutated or wild-type KRAS, we have recently reported that mutated KRAS induces some metabolic alterations in glycolysis, the pentose phosphate pathway (PPP), the tricarboxylic acid (TCA) cycle, and most significantly in the amino acid pathway [12]. We identified that mutated KRAS regulated asparagine synthetase (ASNS), an enzyme that is involved in de novo synthesis of asparagine from aspartate, and that KRAS-mutant CRC cells could become adaptive to glutamine depletion through ASNS-dependent asparagine biosynthesis. There is also some evidence from other groups that KRAS mutations in CRC are associated with glutamine metabolism. Wong et al. reported that SLC25A22 (a mitochondrial glutamine transporter) was a synthetic lethal metabolic gene in KRAS-mutant CRC cells and that expression of SLC25A22 was correlated with poor prognosis in patients harboring KRAS mutations [11]. Miyo et al. reported that glutamine dehydrogenase 1 (GLUD1) and SLC25A13 (a mitochondrial aspartate-glutamate carrier) played an essential role in cell survival of CRC cells under glucose-deprived conditions, and that combined expression of GLUD1 and SLC25A13 was significantly associated with tumor aggressiveness and poorer prognosis in CRC patients [13]. These results indicate that the amino acid metabolism including glutaminolysis is more essential for cell survival in KRAS-mutant CRC than in wild-type KRAS CRC.
In this study, we focused on the amino acid transporter which was exclusively regulated by mutated KRAS, although several amino acid transporters have been reported to be up-regulated in cancer [32]. Herein, we identified SLC1A5 as a novel target gene regulated by mutated KRAS in CRC. Expressions of SLC25A22 and SLC25A13 were not affected by KRAS-knockdown in our experiments ( Figure S1b). Up-regulation of SLC1A5 (ASCT2) and its clinical significance has been reported in a variety of human cancers [19][20][21][22][23][24][25][26][27][28][29][30][31]. In the present study, we demonstrated that SLC1A5 (ASCT2) expression was regulated through KRAS signaling, and that SLC1A5-knockdown resulted in reduced cell growth and increased cell apoptosis in KRAS-mutant CRC cells (Figures 1 and 3). Importantly, the effect of SLC1A5-knockdown was more prominent on cell growth and apoptosis than that of glutamine depletion (Figure 4a,b), which indicates that SLC1A5 (ASCT2) plays a critical role in the malignant progression of KRAS-mutant CRC. Furthermore, SLC1A5-knockdown resulted in the suppression of cell migration (Figure 4d). In primary CRC clinical specimens, we found that ASCT2 expression was significantly associated with tumor depth and vascular invasion in KRAS-mutant CRC, but not in wild-type KRAS CRC (Table 1). In conclusion, our data indicates that SLC1A5 (ASCT2) could be a novel biomarker as well as a potential therapeutic target in KRAS-mutant CRC.

Quantitative Reverse Transcription Polymerase Chain Reaction (RT-PCR) Analysis
Total RNAs were extracted from cells with High Pure RNA Isolation Kit (Roche, Mannheim, Germany) according to the manufacturer's instructions. RNA was reverse transcribed to cDNA with Transcriptor First Strand cDNA Synthesis Kit (Roche) according to the manufacturer's instructions. The relative levels of respective genes were quantified using StepOnePlusTM Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). The respective mRNA levels were normalized to that for ACTB. Primer sequences were found in Table S1.

Western Blot Analysis
Cells were washed with ice-cold phosphate-buffered saline and lysed in sodium dodecyl sulfate lysis buffer supplemented with inhibitor cocktails of protease and phosphatase. Primary antibodies can be found in Table S2.

Cell Proliferation Assay
A cell proliferation assay was measured by Cell Counting Kit-8 (Dojindo, Kumamoto, Japan) according to the manufacturer's instruction. Cells transfected with siRNA were cultured in 96-well plates at a density of 5000 cells/well for 72 h.

Clonogenic Assay
Cells were seeded in 6-well plates at a density of 100 cells per well in complete media. At the end point, colonies were fixed in 1% glutaraldehyde and stained with 0.2% crystal violet for 30 min, and number of colonies was counted. A colony was defined as a cluster of at least 50 cells.

Apoptosis Assay
The activity of Caspase-3 and -7 was measured by using Caspase-Glo 3/7 assay (Promega, Madison, WI, USA) according to the manufacturer's protocol. Caspase activity was normalized to the cell number counted by CCK-8 cell proliferation assay under the same density and conditions.

Wound Healing Assay
Cell lines were seeded into 12-well plates and grew until 80-90% confluence. Confluent cultures were scratched with sterile tips, washed with PBS, and cultured in DMEM containing 5% FBS. Cells were photographed by a 50× magnification at 0, 24, and 48 h. Wound closure (%) was evaluated using the ImageJ software.
Two researchers (Kosuke Toda and Gen Nishikawa) independently evaluated all immunohistochemistry samples without prior knowledge of other data. The slides with different evaluations among them were reinterpreted at a conference to reach the consensus. No patients received chemotherapy and/or radiation therapy. KRAS mutational status in all patients was analyzed by using an ABI 3130 Genetic Analyzer (Applied Biosystems, foster City, CA, USA), as described previously. Pathologic staging was categorized in accordance with the 7th edition of Union for International Cancer Control (UICC) classification of malignant tumors.

Statistical Analysis
All values were expressed as mean ± standard deviation (SD). Statistical analyses were conducted with the JMP Pro 12 (SAS Institute, Inc., Cary, NC, USA). Student's t-test was used for comparing means between two groups. In clinical data, the statistical significance of differences between variables of two groups was determined by student's t-test, chi-squared test, or Fisher's exact test. Relapse-free survival (RFS) rates were evaluated by the Kaplan-Meier survival curve and log-rank test. All analyses were two-sided, and differences with a p value of less than 0.05 were considered statistically significant in all analyses.