1. IGF-1 mediates FASN expression in breast cancer through mTORC1.
Regulation of lipogenic metabolism through FASN has been studied extensively and is often mediated downstream of receptor tyrosine kinases (RTK) such as fibroblast growth factor receptor (FGFR), epidermal growth factor receptor (EGFR), insulin receptor (IR), and the oncogenic human epidermal growth factor receptor 2 (HER2) (24-28). Thus, we sought to evaluate the effect of the IGF-1R signaling axis in FASN regulation. IGF-1 exposure contributed to increases in FASN mRNA and protein expression in both MCF-7 and MDA-MB-231 breast cancer cells (Figure 1A) (Full membranes in Figure S1.). To further elucidate the specific role of IGF-1R, RNAi mediated silencing of IGF-1R was conducted in response to IGF-1 exposure. IGF-1R knockdown significantly decreased (p=.03, MDA-MB-231; p=.009, MCF-7) FASN protein expression in both cell lines (Figure 1B). To further demonstrate IGF-1R induced FASN regulation, we used the mTORC1 inhibitor, rapamycin, to attenuate mTORC1 activity in response IGF-1 exposure. In both cells lines, pre-treatment with rapamycin resulted in significant decreases (p=.027, MDA-MB-231; p=.004, MCF-7) in IGF-1 induced FASN gene expression compared to vehicle treated cells (Figure 1C). Additionally, rapamycin significantly decreased (p=.007, MDA-MB-231; p=.03, MCF-7) FASN protein expression in response to IGF-1 exposure (Figure 1D). This was accompanied by a significant decrease in p-S6K protein expression (p<.001, MDA-MB-231, p<.0001, MCF-7) a direct downstream substrate of mTORC1 signaling (Figure 1D). These results suggest that mTORC1 is involved in IGF-1 induced FASN expression in breast cancer.
2. IGF-1-mTORC1 axis contributes to SRPK2 nuclear localization and phosphorylation.
Recent studies have demonstrated the downstream substrate of mTORC1, ribosomal S6 kinase (S6K) in phosphorylating SRPK2(8, 29). Phosphorylation of SRPKs can alter their conformation and potentially affect nuclear localization through their nuclear localization signals (NLS) (8, 29). Thus, we aimed to determine if the IGF-1-mTORC1 axis affects the subcellular localization of SRPK2 in breast cancer cells. To determine the role the IGF-1 mTORC1 pathway and SRPK2 dynamics, breast cancer cells were serum starved overnight in the presence of either mTORC1 inhibitor, rapamycin, or vehicle and exposed to IGF-1 for various time points. IGF-1 induced nuclear localization of SRPK2 at 6 and 12 hours of exposure in both MCF-7 and MDA-MB-231 cell lines (Figure S2). This was not apparent at hours 2 and 18 of IGF-1 exposure (Figure S2). At a six-hour time point, there was a substantial increase in nuclear SRPK2 upon IGF-1 exposure in both breast cancer cell lines (Figure 2A-B), which was abrogated upon rapamycin treatment. Recent work has illustrated the potential of mTORC1 signaling to promote the phosphorylation of SRPK2 and its contribution to its nuclear localization (8, 29). Thus, we aimed to explore if IGF-1 contributed to phosphorylation of SRPK2. Upon IGF-1 exposure for 6 hours, there was an increase in phosphorylated SRPK2 in both cell lines and this increase in phosphorylation was attenuated by rapamycin treatment (Figure 2C). We further investigated the subcellular localization of phosphorylated SRPK2 upon IGF- 1 exposure by performing cell fractionation in which breast cancer cells were treated with a vehicle or rapamycin and exposed to IGF-1. Rapamycin reduced the amount of nuclear phosphorylated SRPK2 and increased cytosolic, compared to control in response to IGF-1 exposure (Figure 2D). Thus, our current findings demonstrate that IGF-1 induces nuclear localization and phosphorylation of SRPK2 through mTORC1.
3. SRPK2 contributes to IGF-1-induced FASN expression and de novo fatty acid synthesis.
SRPK2 has been demonstrated to regulate lipogenesis through post-transcriptional RNA processing(8, 29). Thus, we proposed that IGF-1 induced FASN regulation was through the mTORC1-SRPK2 axis. To investigate an IGF-1-SRPK2 mediated FASN regulation, breast cancer cells were treated with vehicle or SRPN-340, a small molecule inhibitor of SRPK2. Interestingly, the inhibition of SRPK2 decreased total mRNA expression of both FASN and sterol-CoA desaturase-1 (SCD-1), another lipogenic enzyme in the de novo fatty acid synthesis pathway but had no significant impact on gene expression of glycolytic enzymes, glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and phosphofructokinase platelet (PFK) (Figure 3A). Additionally, SRPK2 inhibition had no significant impact on any gene expression in the non-transformed mammary epithelial cell line, MCF-10A, suggesting this a cancer cell specific pathway (Figure 3A). We further investigated if SRPK2 also contributed to IGF-1 induced FASN protein expression by knocking down SRPK2 by RNAi. Surprisingly, we observed a specific decrease in FASN protein expression upon SRPK2 knockdown only in the MDA-MB-231 breast cancer cells and not in the MCF-7 cells (Figure 3B). Since MDA-MB-231 cells are triple negative breast cancer (TNBC) cells, we next asked if the IGF-1-SRPK2 axis modulated FASN protein expression in other breast cancer cell lines. To this end, MDA-MB-453 (TNBC, luminal androgen receptor +), SUM-159 (TNBC, mesenchymal stem-like), BT-549 (TNBC, mesenchymal) and SKBR3 (HER2+) cells were transfected with control or SRPK2 specific siRNA followed with IGF-1 exposure. In all of the cell lines, we observed significant decreases in FASN protein expression upon SRPK2 knockdown, suggesting that this observed response was not specific to a single cell line.
Given its role in FASN mRNA and protein expression, we next explored the role of SRPK2 in de novo palmitate synthesis in cancer cells in response to IGF-1 exposure. To investigate the role of SRPK2 in IGF-1 indued palmitate synthesis, we knocked down SRPK2 by RNAi and incorporated [U-13 C] glucose into lipid stripped media to measure de novo palmitate synthesis. SRPK2 knockdown resulted in significant decreases in total palmitate upon IGF-1 exposure in MDA-MB-231 cells only (Figure 3C) (14). Additionally, we observed a significant reduction of full labeled palmitate from [U-13 C] glucose (C16:0) in MDA-MB-231 cell lines and not in the MCF-7 (Figure 3D). These results elucidate the role of a potential IGF-1-SRPK2-FASN axis in triple negative breast cancer cells.
4. The IGF-1-mTORC1-SRPK2 axis contributes to FASN mRNA stability
SRPK2 induces lipogenic gene expression through pre-mRNA splicing, primarily mediated through the phosphorylation of RNA binding proteins, serine/arginine rich splicing factors (SRSFs), and can result in mRNA stabilization (8, 9). Thus, we sought to investigate if IGF-1 could mediate FASN mRNA stability through our observed IGF-1-mTORC1- SRPK2 pathway. To ask if IGF-1 induces mRNA stability through SRPK2, breast cancer cells were exposed to IGF-1 with a vehicle or SRPK2 inhibitor, SRPN-340, prior to treatment with actinomycin-D, a transcriptional inhibitor, for 0, 3, and 5 hours (8). In response to IGF-1 exposure, we observed a significant decrease in mRNA stability in both FASN and SCD-1 upon SRPK2 inhibition, which was specific only to the MDA-MB-231 cells (Figure 4A). Additionally, there was no significant effect on the mRNA stability of glycolytic PFK and GAPDH in the MDA-MB-231 cells (Figure 4A). Intron retention is alternative splicing event, which can result in mRNA decay (9). Thus, we investigated if this observed decrease in mRNA stability was due to an increase in intron retention. To this end, exposed MDA-MB-231 cells to IGF-1 with either control or SRPK2 siRNA. Interestingly, we found that IGF-1 exposure significantly reduced intron retention, which was rescued by SRPK2 knockdown in the MDA-MB-231 cells (Figure 4B). These findings propose that IGF-1 regulates FASN through abrogating IR, which is dependent upon SRPK2.
5. The IGF-1mTORC1-SRPK2 axis regulates FASN through SRSF-1
Given our results demonstrating IGF-1 induced SRPK2 nuclear localization, we proposed that IGF-1 also mediates nuclear localization of SRSF-1, a common SRPK2 substrate and SR protein involved in pre-mRNA processing (8, 10). SRSF-1 was expressed in breast cancer cells with a p-eGFP-C1 vector (EGFP-SRSF-1), which has been shown previously to be highly homologous with endogenous SRSF-1 localization and function (15, 30). Pre-mRNA splicing machinery including SR proteins, hnRNPs, and snRNPs are stored in membrane-free structures, termed nuclear speckles (16). Interestingly, we observed consistent decreases in EGFP-SRSF-1 distribution as well as more diffusion within the nucleus upon IGF-1 exposure (Figure 5B). Co-transfecting cells with an SRPK2 siRNA, attenuated the IGF-1 induced decrease the observed SRSF-1 speckle-like size. (Figure 5A). Additionally, treatment with rapamycin had similar effects (Figure 5B). These results suggest that the IGF-1-mTORC1-SRPK2 axis alters SRSF-1 nuclear localization. When released from speckles, SR proteins are able to initiate pre-mRNA splicing of nascent RNA transcripts through their RNA recognition motifs (RRMs) (16). To determine if IGF-1 was mediating the binding of SRSF-1 to FASN pre-mRNA, we performed RNA-immunoprecipitation. Intriguingly, we observed that IGF-1 exposure significantly increased the binding of SRSF-1 to FASN mRNA, which as abrogated by SRPK2 knockdown (Figure 5C). Considering SRPK2 knockdown enhanced FASN IR in response to IGF-1 exposure (Figure 4B), we proposed that SRSF-1 could be mediating FASN regulation downstream of SRPK2. To investigate the role of IGF-1 induced SRSF-1 FASN regulation, we knocked down SRSF-1 by RNAi in breast cancer cells and performed RT-qPCR for FASN mRNA expression. As expected, knocking down SRSF-1 greatly reduced FASN mRNA expression, which little effect on glycolytic expression of PFK (Figure 5C). However, these effects were specific to MDA-MB-231 cells and not in MCF-7. These results suggest that the IGF-1-SRPK2 axis in MDA-MB-231 cells mediates FASN mRNA regulation is through SRSF-1.