The mTORC2 component SIN1 post-transcriptionally regulates TYMS levels and modulates P53 activity in response to 5-FU chemotherapy

Abstract

Transcriptional and translation control of thymidylate synthase (TYMS) is poorly understood, particularly in response to chemotherapeutic drugs such as 5-Fluorouracil (5-FU) and its derivatives. The current study addressed this gap by demonstrating a biphasic response in TYMS protein levels upon 5-FU treatment. Indeed, we observe an initial reduction within the first few hours, followed by a marked increase at 24 h. These changes occurred independently of transcriptional regulation, as TYMS mRNA levels remained stable during the early phase and showed only a moderate increase later. We further showed that neither autophagy nor proteasomal degradation contributed to this dynamic, but instead it is driven by a change in its translation. Using thermal proteome profiling, we identified SIN1, a key component of the mTORC2 complex, as a key regulator of TYMS protein levels. Our functional studies revealed that SIN1 depletion negatively alters TYMS levels and dynamics and sensitizes cancer cells to 5-FU-mediated cell death. Moreover, our data show that SIN1 protein is not only required for stress mediated increase in P53 activity, but also for TYMS translation. These findings uncover a novel mechanism controlling TYMS protein levels and suggest that targeting SIN1 may represent a promising strategy to enhance the therapeutic efficacy of 5-FU-based treatments.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12964-025-02640-y.

Introduction

Thymidylate synthase (TYMS) plays a key role in DNA synthesis. TYMS protein catalyzes the conversion of deoxyuridine monophosphate (dUMP) to deoxythymidine monophosphate (dTMP), playing a central role in maintaining nucleotide pools necessary for cell proliferation [1, 2]. TYMS RNA and protein levels were found to be significantly up-regulated in various cancers and associated with the poor prognosis of patients [3, 4]. Given its essential role in DNA replication, TYMS is considered as a primary target of chemotherapeutic agents developed to inhibit nucleotide biosynthesis in rapidly proliferating cancer cells [5, 6]. One of the most widely used inhibitors of TYMS is 5-fluorouracil (5-FU). Upon entry into the cell, 5-FU is converted into metabolites: fluorodeoxyuridine monophosphate (FdUMP), fluorodeoxyuridine triphosphate (FdUTP), and fluorouridine triphosphate (FUTP). Incorporation of these metabolites during DNA and RNA synthesis leads to DNA and RNA damage and death of rapidly proliferating cells. Moreover, the binding of 5-FdUMP to TYMS leads to the formation of an inactive complex, altering TYMS activity, which in turn causes deoxynucleotide (dNTP) imbalance and DNA damage [7]. Despite the widespread clinical use of 5-FU and its derivatives, resistance to treatment remains a significant challenge. Indeed, studies have shown that prolonged treatment with 5-FU or its derivative leads to a marked increase in TYMS protein levels [8, 9]. The exact mechanism behind this increase is still not clear. In addition to its role in thymidylate biosynthesis, thymidylate synthase (TYMS) functions as an RNA-binding protein that auto regulates its expression. Indeed, TYMS binds to two cis-acting elements on its own mRNA: a stem–loop structure in the 5′ UTR and a sequence within the coding region, which in turn reduces translation efficiency. This translational autoregulation allows TYMS to control its protein levels in response to substrate availability and exposure to inhibitors such as 5-FU. Moreover, studies also indicate that TYMS can bind other cellular mRNAs, including p53 and members of the myc family, repressing their translation and impacting cell cycle control, apoptosis, and drug resistance [10–12]. Whether this is the only mechanism that controls TYMS protein dynamics is currently unknown. Our study addresses this gap by examining how 5-FU and its derivatives influence TYMS protein dynamics. We have shown that TYMS exhibits a biphasic response, with protein levels significantly reduced within the first few hours of treatment, followed by a marked increase at 24 hours. These changes occurred independently of transcriptional changes, as TYMS mRNA levels remained stable in the early phase of treatment and showed only a moderate increase at later time points. Moreover, neither autophagy nor proteasomal degradation contributed to this dynamic. Using thermal proteome profiling, we identified MAPKAP1 (SIN1 protein), a key component of the mTORC2 complex [13], as a key regulator of TYMS levels. Using a small hairpin targeting the SIN1 protein, we showed that downregulation of SIN1 alters TYMS protein levels and dynamics following 5-FU or FuDR treatment, demonstrating a key role for SIN1 in modulating the cellular response to 5-FU and its derivative. Moreover, SIN1 depletion sensitizes cancer cells to FuDR-mediated cell death. Furthermore, our data show that SIN1 depletion alters P53 transcriptional activity and alters TYMS translation following genotoxic stress. These findings highlight a novel mechanism of TYMS regulation and suggest that targeting SIN1 may provide new therapeutic avenues to modulate TYMS levels and enhance the efficacy of 5-FU treatments.

Results

TYMS protein levels exhibit a biphasic response to 5-FU and its derivatives

To examine the immediate cellular response to 5-fluorouracil (5-FU), we used two breast cancer cell lines, MCF7 and MDA-MB-231. We treated these cell lines with 100 µM 5-FU for a short duration (2, 4, and 6 hours). Interestingly, our data revealed that short treatment of cells with 5-FU led to a significant decrease in thymidylate synthase (TYMS) protein levels in both cell lines (Fig. 1 A-D). These changes also occur at lower concentrations of 5-FU (Sup. Figure 1 A-D). We next examined the impact of longer treatment with 5-FU. Our data revealed that prolonged treatment of cells (24 hours) with 5-FU led to a significant increase in the levels of thymidylate synthase (TYMS) protein in both cell lines, suggesting that 5-FU metabolites have a similar effect on cells irrespective of the origin of breast cancer cells (Fig. 1 E-G). Moreover, our data are in line with previous findings showing an increase in TYMS levels following prolonged treatment with 5-FU in colorectal cancer cells [8, 14].

Fig. 1.

5-fluorouracil and derivatives alter TYMS protein levels in a time-dependent manner. A-B Lysates from MDA-MB-231 and MCF-7 cells treated with 100 µM 5-FU for the indicated durations (h: hours) and then stained with the indicated antibodies. C-D Bar graphs showing the results of densitometric analysis of TYMS protein levels under different conditions obtained in A and B, respectively. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: Dunnett test. MDA-MB-231: F (3, 8) = 4.809, P < 0.05. MCF7: (F (3, 8) = 17.83, P < 0.001). E, F Lysates from MDA-MB-231 and MCF-7 cells treated with 100 µM 5-FU for 24 h and stained with the indicated antibodies. G Bar graphs showing the results of densitometric analysis of TYMS protein levels under different conditions obtained in E and F. Unpaired T test, *, P < 0.01. H-I Lysates from MDA-MB-231 and MCF-7 cells treated with 50 µM FuDR for the indicated durations and then stained with the indicated antibodies. J Bar graphs showing the results of densitometric analysis of TYMS protein levels under different conditions obtained for each cell line in H and I, respectively. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: Dunnett test. MDA-MB-231: F (3, 8) = 30.35, P < 0.001. MCF7: F (3, 8) = 9.587, P < 0.001). K-L Lysates from MDA-MB-231 and MCF-7 cells treated with 50 µM FuDR for 24 h and then stained with the indicated antibodies. The bar graphs show the results of densitometric analysis of TYMS protein levels under different conditions obtained for each cell line. The data are presented as the means ± SEMs. Unpaired T test, *, P < 0.001

5-FU is converted into the nucleotide metabolites fluorodeoxyuridine monophosphate (FdUMP) and fluorodeoxyuridine triphosphate (FdUTP). Subsequent metabolism of FdUMP and FdUTP leads to DNA and RNA damage, respectively. FdUMP is also able to bind the nucleotide-binding domain of thymidylate synthase (TYMS), leading to the formation of a ternary complex containing FdUMP, TYMS, and 5,10‑methylenetetrahydrofolate, which in turn blocks TYMS activity [2, 7]. Next, we used 5-fluorouridine (FUR) and 5-fluorodeoxyuridine (FuDR), which are intracellularly metabolized and converted into FdUTP and FdUMP, respectively. Surprisingly, short-duration treatment of cells with FuDR and FuR similarly alters TYMS protein, with a decrease at early and a significant increase at later time points (Fig. 1 H-L and Sup. Figure 2 A-D). Upon binding of FdUMP to the TYMS protein, the complexes migrate more slowly, as indicated by the appearance of a second band adjacent to the free protein (Fig. 1). Moreover, it is worth noting that while a 2-hour treatment with FuDR results in approximately a 50% decrease in TYMS levels, the same treatment with FuR leads to only a 20% reduction in this protein. Consistent with the metabolism of 5-FU, this observation suggests that a minor portion of FuR is also converted into FdUMP, which subsequently binds to TYMS.

To rule out the possibility that the decrease in TYMS levels at the early time point is a result of a change in its thermodynamics upon binding to 5-FU or its derivatives, a drug affinity responsive target stability assay (DART) was carried out to examine the binding of TYMS to FUR and FuRD [15]. This assay confirmed the direct interaction between TYMS and FuDR. Indeed, TYMS proteolytic stability is significantly decreased in the presence of FuDR. In contrast, FuR had no effect, demonstrating that a decrease in TYMS protein upon treatment with 5-FU or derivatives does not result from drug-protein interaction (Sup. Figure 2 E-F). Together, these data highlight time-dependent changes in TYMS protein upon treatment with pyrimidine analogues.

The decrease in TYMS protein levels is a result of changes in its translation following 5-FU Treatment

The dynamic changes in TYMS protein level upon 5-FU treatment might be driven by (I) changes in protein stability, (II) changes in mRNA translation, or (III) alterations in the transcription level of the gene encoding TYMS protein. We thus analyzed the mechanism that drives the decrease in TYMS protein upon short treatment with 5-FU. We used proteasome inhibitor bortezomib or Mg132 and the autophagy inhibitors bafilomycin or chloroquine to examine changes in TYMS protein degradation [16, 17]. Simultaneous treatment of cells with 5-FU and these inhibitors did not prevent the decrease in TYMS protein levels but promoted it further, especially in cells treated with chloroquine (Fig. 2 A-B and Sup. Figure 3 A-B). These results suggest that alterations in TYMS protein could result from changes in the transcriptional activity of the gene encoding the protein. To test this possibility, we analyzed the levels of mRNA encoding TYMS upon short and prolonged treatment with 5-FU or FuDR. Interestingly, this analysis showed no changes in TYMS mRNA levels after 2 hours of treatment with 5-FU or FuDR, and only a moderate increase occurred after 24 hours (Fig. 2 C), suggesting that the decrease seen in TYMS protein levels at early hours could be a result of changes in translation. To examine this possibility, we used a non-canonical amino acid, β-ethynylserine (βES) for metabolic labelling of the newly synthesized proteins [18, 19]. Following the labelling in cells, proteins that incorporated βES were biotinylated via click chemistry, affinity-enriched, resolved by SDS-PAGE, and TYMS nascent polypeptide chain levels were inspected using western blot analysis. As indicated in Fig. 2 D-E, in vehicle-treated cells, newly synthesized TYMS protein was detected within two hours. In contrast, TYMS was nearly undetectable in cells treated with FuDR or FuR, demonstrating that the decrease in TYMS protein within the first two hours of treatment with drugs is a result of the change in its translation.

Fig. 2.

5-fluorouracil-mediated TYMS decrease is driven by a change in its translation. A Lysates from MDA-MB-231 cells pre-treated with 10 nM bafilomycin A or 100 nM bortezomib for 2 h, followed by 100 µM 5-FU for 2 h. B Bar graphs showing TYMS protein levels under different conditions obtained in A. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: Dunnett test. F (3, 8) = 11.97, P < 0.005. C Bar graph showing Tyms mRNA levels in MDA-MB-231 cells following treatment with 100 µM 5-FU or 50 µM FuDR for the indicated durations. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: Dunnett test. 5-FU (F (2, 8) = 2.48, P = 0.163), FuDR (F (2, 8) = 4.25, P = 0.07). ns: not significant. D Visualization of newly synthesized protein in MDA-MB-231 cells treated or not with 50 µM FuDR or 50 µM FuR during two hours, combined with β-ethynylserine (βES). Gels show the total input (5%) and the newly synthesized proteins after enrichment. PARP1 and AKT were used as internal controls. CHX: Cycloheximide is used as a protein synthesis inhibitor as a control. n = 3 separate biological replicates. E The bar graph shows the amount of newly synthesized TYMS protein under the conditions shown in D. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: F (2, 6) = 7.37, P < 0.05

The PI3K/AKT/mTOR signalling pathway exhibits a biphasic response to 5-FU and its derivatives

The mTOR signalling pathway is a central regulator of cell growth and metabolism. This pathway exists under two protein complexes, mTORC1 and mTORC2. mTORC1 is an energy sensor and controls protein translation, whereas mTORC2 controls cell proliferation and cytoskeletal remodelling. Overactivation of this signalling pathway promotes cancer growth and resistance to therapy [19]. In this respect, studies have shown that inhibition of the PI3K/Akt/mTOR signaling pathway enhances the antitumor effect of 5-FU [20, 21]. However, the dynamic activity and the contribution of this signaling pathway to 5-FU resistance remain poorly described.

Using phosphorylated P70S6 kinase at Thr389 and phosphorylated AKT1 at Ser473 as a readout of mTORC1 and mTORC2 activity [22], respectively, we analyzed the impact of the 5-FU or its derivative on the phosphorylation levels of these two proteins. In both cell lines, 5-FU treatment led to a marked decrease in phospho-p70 S6 kinase levels, with a marked effect in MCF7 cells. In contrast, the same treatment moderately increased AKT1 Ser473 phosphorylation in both cell lines (Sup. Figure 4 A-D). Consistently, short-duration treatment of cells with FuDR leads to similar changes in mTORC1 and mTORC2 activity, albeit with different magnitudes in the analyzed cell lines, highlighting a cell type-dependent effect of 5-FU and its derivative on the mTORC activity (Sup. Figure 4 E-H).

Given the role of mTORC in translation [23], these findings suggest that the decrease in TYMS level upon short treatment with 5-FU and derivatives may be directly linked to a reduction of mTORC1 activity. To test this hypothesis, we first examined the impact of a panel of inhibitors targeting PI3K/AKT or mTORC1. As indicated in supplementary figure 5, application of PI3K/AKT inhibitor did not affect the basal TYMS protein levels. In contrast, prolonged treatment of cells with rapamycin alone led to a marked decrease in TYMS protein (Sup. Figure 5). Short-term treatment of cells with rapamycin primarily affects mTORC1, whereas prolonged treatment inhibits mTORC1 and mTORC2 [24]. To further exclude the role of mTORC1 in the regulation of TYMS levels, we transiently inhibited mTORC1 with rapamycin. As shown in Supplementary Figure 6 A-B, combined treatment with rapamycin and 5-FU for 2 hours had no marked effect on TYMS protein levels. Therefore, our data suggest that the observed changes in TYMS dynamics may be mediated by mTORC2.

Next, we analyzed the role and activity of mTORC1/2 upon prolonged treatment with FuDR. Consistent with the above results, P70S6 kinase phosphorylation is significantly lower following prolonged treatment with FuDR (Fig. 3 A, D). Moreover, this experiment confirmed that longer treatment with rapamycin alone significantly alters TYMS protein levels. When combined with FuDR, a marked decrease in TYMS levels is also observed in MCF7 cells (Fig. 3 D-F). Nevertheless, TYMS protein levels remain higher in MDA-MB-231 cells simultaneously treated with FuDR and rapamycin, suggesting that these cells may resist lower rapamycin concentrations upon stress (Fig. 3 A, D). Furthermore, a further increase in AKT1 phosphorylation following prolonged treatment with rapamycin was not accompanied by a marked increase in TYMS, suggesting that changes in TYMS protein are independent of the PI3/AKT signalling axis. Moreover, these data highlight an additional separate mechanism that controls TYMS protein level.

Fig. 3.

Changes in TYMS levels are mTORC1 independent. A Lysates from MDA-MB-231 cells treated or not with 50 µM 5-FuDR or 50 nM Rapamycin for 24 h were stained with the indicated antibodies. n = 3 separate biological replicates. B-C Bar graphs showing the results of densitometric analysis of the protein levels under different conditions obtained in A. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: Dunnett test. p-AKT1 (F (3, 8) = 4.06, P < 0.05), p-P70SK (F (3, 8) = 17.5, P < 0.001), TYMS (F (3, 8) = 16.34, P < 0.001). D Lysates from MCF7 cells treated or not with 50 µM 5-FuDR, 50 nM Rapamycin for 24 h were stained with the indicated antibodies. n = 3 separate biological replicates. E–F Bar graphs showing the results of densitometric analysis of the protein levels under different conditions obtained in D. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: Dunnett test. p-AKT1 (F (3, 8) = 18.5, P < 0.0005), p-P70SK (F (3, 8) = 76.9, P < 0.0001), TYMS (F (3, 8) = 124.1, P < 0.0001)

Thermal proteome profiling reveals a potential role for mTORC2 in cellular response to 5-FU

To identify cellular targets of 5-fluorouracil (5-FU), we applied the Proteome Integral Solubility Alteration (PISA) assay with mass spectrometric readout, which monitors shifts in apparent protein thermal stability across the cellular proteome. Such shifts typically arise from direct protein-small-molecule binding but can also reflect conformational changes driven by post-translational modifications or protein-protein interactions [25]. A short exposure of MDA-MB-231 cells to 5-FU (200 µM, 2 h) produced a marked decrease in TYMS abundance, prompting us to profile the early proteome response with PISA. Over this brief interval, intracellular conversion of 5-FU to FuR and FuDR is expected to be minimal [26], so our PISA experiment primarily reports the effects of very low concentrations of FuR and FuDR.

In contrast to prior studies employing prolonged 5-FU treatments or direct FuDR exposure [25, 27], our PISA analysis revealed limited alterations in proteome thermal stability: relative to vehicle, only 16 proteins exhibited significant changes (9 decreased and 7 increased stability; FDR-adjusted p < 0.05; Fig. 4 A). Despite the small set, pathway analysis indicated enrichment for protein-DNA interaction functions (Fig. 4 B), consistent with recent observations for FuDR [27], which may reflect either sufficient intracellular accumulation of FuDR during the short 5-FU exposure or a subset of functionally related protein targets engaged by both 5-FU and FuDR (these explanations are not mutually exclusive). Notably, the mTORC2 signalling pathway was also implicated: SIN1 (MAPKAP1), a core mTORC2 subunit, showed a significant decrease in thermal stability (Fig. 4 A, Sup. Table 1).

Fig. 4.

Proteome Integral Solubility Alteration analysis of 5-fluorouracil in MDA-MB-231 cells. A Volcano plot of PISA results expressed as log-transformed values of the ratio of protein soluble amount per protein of 5-FU-treated compared to control-treated cells, and relative p-value. Each dot represents an individual protein, identified in the final analysis. The most significant hits are labeled in orange font (FDR ≤ 0.05). Bleu dots represent identified proteins with no significant change (FDR p > 0.05). B Bar graph showing log10 p-value and GO terms of the biological processes affected by the significant identified hits using EnrichR KG. C Western blot analysis of SIN1 isoform level in MCF7 and MDA-MB-231. 3 biological replicates. D Independent validation of PISA proteomics results. Thermal stability analysis of SIN1 isoforms in MDA-MB-231 cells (Short and long). Cells were incubated for 2 h with vehicle or 5-FU (200 µM), then PISA. Three independent biological replicates/conditions were shown. Total AKT and ERK were used as a loading control. E Bar graph summarizing SIN1 isoform levels, as presented in panel D. The data are presented as the means ± SEMs. Unpaired T test, Short, *, P < 0.05. Long, P = 0.11

SIN1 has been reported to regulate responses to growth factors, genotoxic stress, and DNA damage as part of the mTORC2 complex, and it exists as several transcript isoforms encoding distinct proteoforms [13]. Consistent with this, our data indicate that MCF7 and MDA-MB-231 cells express different levels of individual SIN1 isoforms (Fig. 4 C). The specific functions of these isoforms in mTORC2 regulation and stress responses remain incompletely defined; for example, some isoforms appear essential for mTORC2 signaling, whereas others may modulate separate processes [13, 28]. To determine which isoform(s) were thermally affected, we performed isoform-resolved PISA followed by immunoblotting of soluble fractions in MDA-MB-231 cells. As shown in Fig. 4D, 5-FU treatment destabilized both the short and long SIN1 isoforms, with a more pronounced effect on the short isoform, confirming that 5-FU perturbs SIN1 thermal stability.

SIN1 protein is required for the cellular response to 5-FU metabolite and protects cells from genotoxic stress

Our data above showed that SIN1 is among the top proteins whose function could be immediately affected at the early point of treatment with 5-FU. Previous studies have shown that SIN1 is required for mTORC2 complex assembly and AKT-Ser473 phosphorylation in response to growth factors and DNA-damaging agents [29, 30]. To examine the role of SIN1 in response to pyrimidine analogues, we used a lentiviral knockdown method to achieve a stable reduction in the mRNA encoding SIN1 in MCF7 and MDA-MB-231. As indicated in the Supplementary figure 7 A-D, F, this approach significantly reduces the levels of mRNA encoding SIN1 and all its proteoforms. Analysis of mTORC1/2 activity showed that basal mTORC1 is slightly higher in Sh-SIN1 cells, as seen with the increase in P70S6 kinase phosphorylation. In contrast, the inverse occurs for mTORC2, as seen with markedly lower levels of AKT1 Ser473 phosphorylation. Interestingly, our data show that SIN1 partial depletion leads to reduced levels of TYMS protein, and short-term treatment of cells with FuDR led to almost undetectable levels of this protein in the cells, highlighting a role for SIN1 in controlling the levels of TYMS (Sup. Figure 7 D-E). We next examined the impact of SIN1 depletion on the cellular response to prolonged 5-FU and FuDR treatments. Interestingly, this experiment revealed that long-term treatment of cells with FuDR led to the complete inability of Sh-SIN1 cells to activate mTORC2, as seen with markedly low levels of AKT1 phosphorylation (Fig. 5 A-B, F-G). Moreover, this analysis also showed that long term stress induced increase in the level of TYMS protein is markedly altered in cells in which SIN1 was knocked down, suggesting a key role for SIN1 in regulating TYMS protein under both basal and stress conditions (Fig. 5 D, and I). Furthermore, SIN1 seems to be required for WT P53 stability and activity, as shown by lower P53 and P21 protein levels in Sh-SIN1 treated cells with FuDR (Fig. 5 F, H).

Fig. 5.

SIN1 controls AKT activity and TYMS levels following genotoxic stress. A Lysates from MDA-MB-231 cells stably expressing Sh control or small hairpin against SIN1, treated or not with 50 µM 5-FuDR or 100 µM 5-FU for 24 h, were stained with the indicated antibodies. n = 3 separate biological replicates. B-D Bar graphs showing the results of densitometric analysis of the protein levels under different conditions obtained in A. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: Dunnett test. p-AKT1 (F (5, 12) = 16.02, P < 0.001), p-P70SK (F(5,12) = 2.86, p = 0.06), P53 (F (5, 12) = 3.498, P < 0.05, P21 (F (5, 12) = 26.09, P < 0.0001), TYMS (F (5, 12) = 19.39, P < 0.001). E Bar graphs showing the percentage of Sh Ctrl and Sh SIN1 MDA-MB-231 cells at SubG1 after 72 h of treatment with FuDR. Data are presented as the means ± SEMs. ANOVA (F (3, 8) = 45.1, P < 0.0001). F Lysates of MCF7 cells stably expressing Sh control or small hairpin against SIN1 treated or not with 50 µM 5-FuDR for 24 h, then stained with the indicated antibodies. G-I Bar graphs showing the results of densitometric analysis of the protein levels under different conditions obtained in F. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: Dunnett test. p-AKT1 (F (3, 8) = 483.1, P < 0.0001), p-P70SK (F (3, 8) = 11.58, P < 0.01), P53 (F (3, 8) = 616.3, P < 0.0001), P21 (F (3, 8) = 251.3, P < 0.0001), TYMS (F (3, 8) = 97.82, P < 0.0001). J Bar graphs showing the percentage of MCF7 Sh Ctrl and Sh SIN1 cells at different phase of the cell cycle after 72 h of treatment with FuDR. Data are presented as the means ± SEMs. ANOVA (F (3, 18) = 11.89, **P < 0.001, *P < 0.05)

Having shown the crucial role of SIN1 in mTORC2/AKT activation and control of TYMS protein, we next examined whether SIN1 depletion affects cellular sensitivity to short and long-term treatment with pyrimidine analogs. Compared to Sh control cells, SIN1 depletion in MDA-MB-231 cells sensitizes them to FuDR-mediated apoptosis, as seen with the increase in the percentage of SubG1 cells (Fig. 5 E). While no distinct Sub-G1 peak was observed in MCF7 cells. Nevertheless, SIN1-depleted MCF7 cells exhibited a moderate reduction in the G1 fraction upon FuDR treatment, suggesting altered cell-cycle dynamics (Fig. 5 J). Consistent with this, colony-forming ability of SIN1-depleted MDA-MB-231 and MCF7 cells was markedly reduced upon treatment with a lower concentration of FuDR, indicating that SIN1 knockdown sensitizes cells to FuDR (Sup. Figure 8 A-D). Together, these results demonstrate that SIN1 is required for the cellular response to 5-FU and FuDR.

Intact mTORC2-P53 signalling axis is required for cellular response to genotoxic stress

Our data above demonstrate that SIN1 plays an important role in regulating TYMS protein levels following stress. Moreover, similar results were observed in both MCF7 and MDA-MB-231 cells, suggesting a conserved function of SIN1 despite significant differences between these two breast cancer cell lines. To further examine whether the role of SIN1 is preserved in other cell models, we analysed SIN1 function in HEK293T cells. Consistent with our observations in MCF7 and MDA-MB-231 cells, SIN1 depletion in HEK293T cells resulted in a marked reduction in AKT1 phosphorylation as well as a substantial decrease in TYMS protein levels under both basal and stress conditions (Sup. Figure 9 A-D). These findings indicate that the requirement for SIN1 to maintain AKT activity and TYMS protein levels is not restricted to a particular lineage or p53 genotype, but is conserved across distinct cellular backgrounds. However, upon stress, the magnitude of change in mTORC1/2 activity and P21 levels was not the same, which may be linked to differences in the expression and activity of stress response proteins such as P53. Indeed, MCF7 expresses wild-type P53, whereas MDA-MB-231 harbors’ a mutant form of P53. Several mutant forms of P53 have been described in various cancers, and the changes in their transcriptional activity appear unclear [31]. While the role of P53 in regulating PI3K/AKT signalling upon stress has been established [32], its involvement in TYMS regulation has not been previously addressed.

To investigate the role of p53 in this conserved SIN1-dependent response, we knocked down P53 in both breast cancer cell lines and analysed their response to FuDR treatment. Consistent with previous studies, P53 depletion inhibited both basal and stress-induced AKT phosphorylation (Fig. 6 A, C, Sup. Figure 9 E). Interestingly, our data also reveal that P53 regulates TYMS protein levels under both basal and stress conditions. Indeed, in cells with stable P53 knockdown, we observed a dramatic reduction in TYMS protein levels (Fig. 6 A-D)

Fig. 6.

P53 controls AKT phosphorylation and TYMS levels following genotoxic stress. A, C Lysates from MDA-MB-231 and MCF7 cells stably expressing Sh control or small hairpin against p53, treated or not with 50 µM during 24 h, were stained with the indicated antibodies. B-D Bar graphs showing the results of densitometric analysis of AKT phosphorylation and TYMS protein levels under different conditions obtained in A and C. The data are presented as the means ± SEMs. B ANOVA, multiple comparisons: Dunnett test. p-AKT1 (F (3, 8) = 5.51, P < 0.05), TYMS (F (3, 8) = 240.97, P < 0.0001). D ANOVA, multiple comparisons: Dunnett test. p-AKT1 (F (3, 8) = 291.1, P < 0.001), TYMS (F (3, 8) = 25.51, P < 0.0001)

We next examined whether the decrease in TYMS levels upon SIN1 depletion is directly related to changes in P53 activity. Our results show that overexpression of either wild-type or dominant-negative P53 in SIN1-depleted cells restored basal and stress-induced AKT phosphorylation. However, TYMS protein levels remained low in both control and cells treated with FuDR, suggesting that intact SIN1 is necessary for P53-mediated regulation of TYMS (Sup. Figure 9 F).

To further investigate this mechanism, we analyzed TYMS mRNA levels under basal and stress conditions. As shown in Fig. 7 A, SIN1 depletion had no significant effect on TYMS transcript levels, whereas P53 depletion led to a marked reduction in basal TYMS mRNA expression. Interestingly, FuDR treatment increased TYMS transcript levels even in P53-depleted cells, although they remained lower than in Sh-control cells. These data suggest that the remaining active P53 induced upon FuDR treatment may be sufficient to partially induce TYMS transcription. Moreover, our data show that SIN1 is required for the full activation of P53 transcriptional activity. Indeed, upon FuDR treatment, the expression of known direct P53 target genes was significantly reduced in SIN1-depleted cells that express either the WT or the mutant p53 (Fig. 7 B-G). In MDA-MB-231 cells, which carry mutant p53, p21 protein was detectable under basal conditions and following p53 knockdown, consistent with p53-independent regulation, such as post-transcriptional stabilization in cells with active RAS signaling [33]. Despite the impaired p53 transcriptional function, SIN1 depletion modestly altered mutant p53 activity (Fig. 7 E-F), indicating that SIN1 influences the p53 axis even in a mutant context. Together, these results highlight that SIN1 can contribute to cellular stress responses via both p53-dependent and p53-independent mechanisms.

Fig. 7.

SIN1 controls P53 transcriptional activity following genotoxic stress. A-D Bar graphs showing mRNA levels of TYMS and P53 target genes under the indicated conditions in MCF7 cells. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: Dunnett test. TYMS F (5, 12) = 17.18, P < 0.0001. CDKN1A F (5, 12) = 274.7, P < 0.0001. GAD45A F (5, 12) = 205.0, P < 0.0001. SESN2 F (5, 11) = 12.43, P = 0.0003. E–G Bar graphs showing mRNA levels of P53 target genes under the indicated conditions in MDA-MB-231 cells. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: Dunnett test. CDKN1A F(3,8) = 6.94, P = 0.012. GAD45A F (3, 8) = 22.4, P < 0.001. SESN2 F (3, 8) = 2.22, P = 0.16

SIN1 controls TYMS translation

Our data above show that SIN1 depletion alters TYMS protein levels without marked changes in its mRNA levels, suggesting a role for SIN1 in the regulation of TYMS translation (Fig. 7A and Sup. Figure 7G). To examine this possibility, we performed metabolic labeling to assess the levels of newly synthesized TYMS protein. As shown in Fig. 8 A-B, newly synthesized TYMS protein is markedly reduced in SIN1-depleted cells under basal conditions and following FuDR treatment, demonstrating a role for SIN1 in the regulation of TYMS translation.

Fig. 8.

SIN1 controls TYMS translation. A Visualization of newly synthesized protein in MDA-MB-231 stably expressing sh control of small hairpin against SIN1. Cells treated or not with 25 µM FuDR during 24 h, combined with β-ethynylserine (βES). Gels show the total input on the left (5%) and the newly synthesized proteins after enrichment (right). PARP1 was used as an internal control. n = 3 separate biological replicates. B The bar graph shows the amount of newly synthesized PARP1 and TYMS proteins under the conditions shown in D. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: PARP1 F(3,8) = 1.67, p = 0.24. TYMS F(3,8) = 7.75, p < 0.005. C-F Line plot depicting polysome abundance across gradient fractions in MDA-MB-231 Sh-control and Sh1 SIN1 cells treated or not with 50 µM FuDR during 18 h. Representative of three biological replicates per condition. G-H Bar graph showing the quantification of the global RNA-ribosome association shown in A-D. The results summarize the area under the curve (AUC) as an indication of global translation. Data are presented as mean ± SD; n = 3 independent replicates per condition. (I-J) Bar graphs showing TYMS mRNA levels in light and heavy polysomal fractions in MDA-MB-231 Sh control and Sh-SIN1 cells treated or not with FuDR. RNA quantified by real-time PCR. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: control F(3,8) = 9.18, p < 0.005, FuDR F(3,8) = 8.26, p < 0.005

To assess whether SIN1 affects translational activity, we performed polysome profiling to examine global ribosome distribution. As shown in Fig. 8 C-H, the abundance of light polysome fractions was comparable across all conditions (Fig. 8 G). In contrast, FuDR treatment led to the reduction in heavy polysome fractions, consistent with decreased translational activity (Fig. 8 H). Under basal conditions, no difference was observed between sh-control and sh-SIN1 cells. However, upon FuDR treatment, sh-SIN1 cells exhibited a more pronounced reduction in heavy polysomes compared with sh-control cells, indicating increased sensitivity of global translation to FuDR in cells that express low levels of SIN1 (Fig. 8 H). This selective sensitivity suggests that SIN1-depleted cells are more vulnerable to FuDR treatment. Moreover, this observation suggests that SIN1 may function as an important regulator of translational capacity upon stress.

To determine if SIN1 depletion affects TYMS translation specifically, we next analyzed TYMS mRNA distribution across light and heavy polysome fractions. In sh-control cells, TYMS mRNA was similarly distributed between the light and heavy polysomes under both basal and FuDR-treated conditions (Fig. 8 I-J), indicating active and ongoing translation. In contrast, SIN1-knockdown resulted in a significant shift of TYMS mRNA from heavy to light polysome fractions, both in the presence and absence of FuDR (Fig. 8 I-J). This redistribution is consistent with the metabolic labeling data (Fig. 8 A-B) and reduced association of TYMS mRNA with actively translating ribosomes. Together, these results indicate that SIN1 depletion selectively impairs TYMS translational engagement and exacerbates the inhibitory effects of FuDR on global translation, without affecting basal polysome distribution. To examine the biological significance of SIN1-TYMS interaction, we ectopically expressed TYMS in both Sh control and SIN1-depleted cells. As indicated in Fig. 9 A-B, overexpression of TYMS led to a marked decrease in P53 activity upon FuDR treatment, as indicated by reduced levels of p21 protein. Interestingly, this effect is more pronounced in SIN1 depleted cells. Strikingly, cell cycle analysis revealed that TYMS overexpression in SIN1 depleted cells resulted in a significant increase in the proportion of cells arrested in G1 following FuDR treatment (Fig. 9 C). Although the precise molecular mechanism requires further investigation, our findings suggest that an intricate balance among SIN1 expression, TYMS levels, and p53 activity is required to ensure an appropriate cellular response to stress.

Fig. 9.

Increase in TYMS levels alters P53 activity upon genotoxic stress. A Lysates from MCF7 cells stably expressing Sh control or small hairpin against SIN1 transiently transfected with TYMS expressing plasmid, treated or not with 50 µM during 24 h, were stained with the indicated antibodies. n = 3 separate biological replicates. B Bar graphs showing the results of densitometric analysis of P53 protein levels under different conditions obtained in A. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: Dunnett test. P53 F(7,17) = 44.62, P < 0.0001. C Bar graph displaying the proportion of cells at different stages in the cell cycle analyzed by flow cytometry. MCF7 sh control and sh SIN1 transfected as in A, then treated with or without 50 µM during 72 h. ANOVA, multiple comparisons: Dunnett test. G1 phase F (F (7, 24) = 25.65, P < 0.0001. See supplementary Table 2 for detailed statistics. D Lysates from MCF7 cells treated or not with FuDR for the indicated durations (hours). N = 3 biological replicates per condition. E Bar graph displaying the amount of SIN1 isoforms under the condition obtained in D. The data are presented as the means ± SEMs. ANOVA, multiple comparisons: Dunnett test. Long F(3,8) = 39.67, P < 0.001. Short Long F (3,8) = 26.06, P < 0.001

Discussion

Resistance to 5-Fluorouracil (5-FU) is directly linked to elevated levels of its target protein, thymidylate synthase (TYMS). Several mechanisms controlling TYMS levels have been proposed. It was first suggested that FdUMP interaction with TYMS releases its RNA, which in turn promotes translation [10–12]. Using colorectal cells, Varghese et al. reported that forkhead box transcription factor, FOXM1, could be involved in the positive regulation of TYMS [34]. Despite being controversial, data from Intuyod et al. showed that FOXM1 knockdown had only a moderate effect on TYMS levels [35]. This raises the question of which factor directly controls TYMS levels and dynamics. In the current study, we provide the first mechanistic evidence that TYMS protein dynamics are regulated not only through transcriptional induction, but also at the level of translation in response to 5-fluorouracil (5-FU) and its derivative, fluorodeoxyuridine (FuDR). Our findings challenge the traditional view that TYMS regulation in the context of 5-FU treatment occurs mainly at the transcriptional level. Instead, we show here that in breast cancer cells, 5-FU and FuDR treatment results in a biphasic response in TYMS protein levels degradation within the first two hours, followed by a significant accumulation at 24 hours. Importantly, these changes occur independently of transcriptional changes, as TYMS mRNA levels remain stable during the initial phase and increase modestly at later time points.

The absence of changes in TYMS mRNA in the early phase of treatment suggests that the observed decrease in TYMS protein levels is not a result of transcriptional regulation, but rather changes in translation. This hypothesis was further supported by metabolic labelling experiments, which demonstrated a significant decrease in TYMS translation. It is very well established that mTORC1 plays a crucial role in the regulation of translation. However, emerging evidence also suggests a role for mTORC2 in this process. Indeed, mTORC2 association with the ribosome is required for cotranslational phosphorylation of AKT, its stability, and signalling [36, 37]. Our data show that 5-FU treatment negatively alters mTORC1 activity as seen by decreased P70S6 kinase phosphorylation at both early and later time points, suggesting that mTORC1 is not involved in TYMS dynamics. Using thermal protein profiling, we next identified mTORC2 component SIN1 among the top proteins that showed changes in thermal stability after two hours of treatment with 5-FU. Previous studies reported that 5-FU significantly alters the thermal stability of a large number of proteins, including TYMS [25, 27]. Our findings reveal a more restricted effect, with only 16 proteins showing detectable changes in thermal stability. Several factors could account for these discrepancies. First, the compounds, the cell lines, and the treatment duration used in all the studies differ. Gaetani et al. used 5-FU in A-498 kidney cells, Liang et al. employed 5-FU derivatives (FuR and FuDR), whereas we tested 5-FU in MDA-MB-231 cells. Such differences in drug derivatives, cell type, and treatment duration could strongly influence the extent of protein stabilization and destabilization. In addition, it is well established that 5-FU metabolism varies between cell lines [38]. Finally, technical factors could contribute. Mass spectrometry-based thermal profiling results can vary with instrument performance and detection thresholds, which may affect the number of proteins identified across studies. Thus, the smaller set of thermally altered proteins observed in our case may reflect both biological and experimental differences, as well as variability in protein detection.

Although the precise mechanism underlying the changes in SIN1 thermal stability upon 5-FU treatment is not addressed in this study, it is reasonable to speculate that processes such as protein-protein interactions, post-translational modifications, complexation with other proteins, or alterations in subcellular localization may contribute. Indeed, some of these factors have previously been reported to influence protein thermal stability [39, 40]. In this respect, two phosphorylation sites on SIN1 have already been identified, and their modification appears to vary depending on the specific cellular stimuli. Moreover, some SIN1 isoforms predominantly localize to the plasma membrane, while others are mainly found at the centrosome [28, 29, 41].

Our data show that the mTORC2 component, SIN1, is required for p53 transcriptional activity in both MCF7 and MDA-MB-231 cells, which express WT and mutant p53, respectively. This observation is in line with recent findings by Chen et al. [32], who identified a P53-phosphoinositide signalosome that recruits and activates nuclear AKT in an mTORC2-dependent manner. Interestingly, this mechanism was evident in the presence of both wild-type and mutant p53. Indeed, it has been reported that wild-type p53 regulates nuclear AKT activity in an on/off manner following cellular stress, whereas mutant p53 promotes a dose-dependent increase in basal AKT activity [32]. These findings and ours suggest that mTORC2 provides a conserved signaling input that affects the transcriptional output of p53 irrespective of mutational status. These findings not only reinforce the role of mTORC2 as a stress-responsive signaling factor but also highlight a potential vulnerability that may be exploited for therapeutic use.

Supporting a conserved SIN1-dependent mechanism across diverse cellular backgrounds, we also observed that SIN1 depletion in HEK293T cells resulted in a similar reduction in AKT1 phosphorylation and TYMS protein levels, indicating that SIN1 plays a lineage-independent role in maintaining stress-responsive AKT-TYMS regulation.

While our data support a SIN1-AKT-TYMS regulatory axis involving p53, we cannot exclude contributions from p53-independent pathways. Indeed, 5-FU can also trigger nucleolar stress and RNA damage, which activate distinct signaling pathways that lead to cell cycle arrest or cell death [42]. Thus, some effects of SIN1 depletion on the cellular stress response may also involve these p53-independent mechanisms.

Furthermore, although FuDR is a widely used analog to study 5-FU induced stress, its metabolism and downstream signaling are different from the native 5-FU, which may affect the magnitude or kinetics of p53 activation.

Interestingly, our data also provide the first evidence that SIN1 controls TYMS protein levels and dynamics. Indeed, SIN1 knockdown significantly alters TYMS protein levels without affecting its transcription, both under basal conditions and following FuDR treatment. These findings point to translation as the dominant regulatory mechanism controlling TYMS protein in response to this chemotherapeutic agent. Indeed, our data show that newly synthesized TYMS protein is markedly lower in SIN1 depleted cells under both basal conditions and following FuDR treatment, consistent with a shift of TYMS mRNA away from actively translating heavy polysomes (Fig. 8A-B, I-J). Importantly, our results establish a functional link between SIN1 and TYMS protein homeostasis, providing a foundation for future studies aimed at dissecting the molecular mechanism underlying this regulation.

Importantly, enforced TYMS expression altered P53 activity upon FuDR treatment, with a stronger effect in SIN1-depleted cells (Fig. 9 A-C). This observation suggests that high TYMS levels may help suppress DNA damage signaling, potentially by providing sufficient nucleotide pools for DNA repair or replication, which in turn lowers P53 activity. Indeed, the cell-cycle data support this model. The percent of cells at G1 of the cell cycle phase was most pronounced in SIN1 depleted cells overexpressing TYMS, pointing to a scenario where TYMS-driven nucleotide supply modifies checkpoint responses in the absence of protective mTORC2 signaling. Reduced p53 activity in this setting could impair the canonical G1/S checkpoint regulation, paradoxically leading to an accumulation of cells unable to properly transition into S phase. In line with this hypothesis, the lowest percentage of cells at the S phase of the cycle is seen in SIN1-depleted cells expressing exogenous TYMS treated with FuDR (Fig. 9 C).

SIN1 possesses several known isoforms; however, their expression patterns and exact functions remain poorly defined, especially in the context of cancer [13]. Some of these isoforms may be involved in the regulation of mTORC2 activity, RAS/MAPK signaling pathway, and JNK activity. Therefore, additional research is required to address the exact functions of each isoform, both in cancer development and progression, as well as in mediating resistance to therapy. Furthermore, it would be interesting to study how individual isoforms, as well as combinations of different isoforms, influence cellular responses to stress and growth factor signalling. In line with this hypothesis, for instance, our data show that SIN1 isoform levels exhibit dynamic variation in response to stress over the course of 48 hours (Fig. 9 D-E).

Although a comprehensive analysis of global and selective translational changes following 5-FU treatment is lacking, our results are consistent with recent findings implicating translational changes in response to 5-FU as a key component of the cellular response to 5-FU [43]. In our study using breast cancer cells, we observed a decrease and an increase in mTORC1 and mTORC2, respectively. In contrast, a previous study in colorectal cancer cells reported increased mTORC1 activity upon 5-FU treatment. This discrepancy could be attributed to differences in stress response pathways and translational control mechanisms between different cancer types. Nevertheless, both studies consistently highlight the impact of 5-FU on factors that affect translation machinery linked to pro-survival pathways. Together, these observations suggest that modulation of translation is a critical determinant of cell fate in response to 5-FU across diverse cancer types. Therefore, co-targeting translation machinery with 5-FU could be a strategy to enhance therapeutic efficacy, potentially overcoming resistance mechanisms.

Materials and methods

Reagents

DMEM High Glucose (4.5 g/l), with L-Glutamine, Sodium Pyruvate (Capricorn Scientific, #DMEM-HPA), Fetal Bovine Serum (Sigma Aldrich, # F7524-500ML), Penicillin-Streptomycin (Thermo Fisher Scientific, #15140122). The following inhibitors were used: 5-fluorouracil (Sigma Aldrich, #F6627), 5-Fluoro-2'-deoxyuridine (Thermo Scientific Chemicals, #L16497.ME), 5-fluorouridine (Thermo Scientific Chemicals, #J62083.03), Rapamycin (Thermo Scientific Chemicals, #J62473.MC).

Cell culture

HEK293T, MDA-MB-231, and MCF7 cells were generously provided as a gift from Dr. Anna Marusiak, IMol Institute. The cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 U/mL streptomycin. Cells were maintained at 37 °C in a humidified, 5% CO2 incubator. The concentrations and duration of treatment of cells with the above inhibitors were all described in the figure legends. In each experiment, control cells were treated with an equivalent volume of solvent (DMSO).

Transient transfection

Transient transfection was carried out using jetPRIME® DNA and siRNA transfection reagent (Polyplus, # 101000046). Cells were plated overnight in 12-well plates at 60% confluency. The transfection mixture containing the buffer, jetPRIME reagent, and plasmids was prepared according to the manufacturer’s instructions (1µg DNA/well). The following plasmids were used. Empty vector c-Flag pcDNA3 was a gift from Stephen Smale (Addgene plasmid # 20011; http://n2t.net/addgene:20011; RRID: Addgene_20011). pcDNA3 flag p53 was a gift from Thomas Roberts (Addgene plasmid # 10838; http://n2t.net/addgene:10838; RRID:Addgene 10838). p53 (dominant negative R175H mutant)-pcw107-V5 was a gift from David Sabatini & Kris Wood (Addgene plasmid # 64638; http://n2t.net/addgene:64638; RRID: Addgene_64638).

Proteome integral solubility alteration analysis of 5-FU

PISA assay was carried out as described previously [25]. Briefly, MDA-MB-231 cells were seeded into a 10 cm culture dish at a confluence of 80 %. Twenty-four hours later, cells were treated for two hours with 200 μM 5-FU or DMSO as a control, three biological replicates for each condition. After treatment, cells were detached using Trypsin-EDTA solution (Thermo Fisher Scientific, #15090046) and washed twice with phosphate-buffered saline (PBS). Cell pellets were collected and resuspended in 500 µL of PBS. After homogenization, each biological replicate was split into eight 0.2 mL PCR tubes, 60 μl/tube, one for each temperature point. The following temperatures were used for PISA (43°C, 44.3°C, 46.2°C, 48.5°C, 51.8°C, 54.3°C, 55.9°C, 57 °C). Thermal treatment was carried out for 3 minutes using a C1000 Touch Thermal Cycler (BioRad). Samples were then incubated at room temperature for 5 min. Cell lysates were obtained by freeze-thaw cycles in liquid nitrogen three times and then thawing at 30 °C. Aliquots of the same biological replicates were recombined. Soluble cell lysates were obtained after two rounds of centrifugation at 20000 g for 20 min at 4 °C. Soluble fractions from each condition were collected,

Protein concentrations were determined using the Pierce™ BCA Protein Assay Kit (Thermo Scientific™ #23225), and BSA standards were used in parallel according to the manufacturer’s instructions. Twenty μg of each sample was digested using trypsin overnight at 37 °C with vortexing at 1000 rpm. Digested samples were then labeled with TMT as follows. First, three layers of the resin C18 mesh were packed inside 200 µL pipette tips, which were inserted onto the lid of the Eppendorf tube called STAGE Tips-Column. Columns were first conditioned using 150 µL methanol at 1200 g for 2 minutes, then washed with 100 µL (50% acetonitrile/0.1% formic acid) at 1200 g for 2 minutes. The resin was then equilibrated twice with 150 µL 0.1% formic acid at 1200 g for 2 minutes.

Ten 10 µg equivalent volumes of the digested samples were loaded onto the STAGE Tips-Column and centrifuged at 1200 g for 2 minutes. STAGE Tips- Columns were washed twice with 150 µL of 0.1% formic acid (1200 g, 2 min).

For peptide labeling, TMTs were first resuspended in 2 µL acetonitrile, followed by the addition of 200 µL of freshly prepared 50 mM HEPES, pH 8. Peptide labeling was done by loading 200 µL TMT solution onto STAGE Tips, followed by 300 g centrifugation for 10 minutes. Labeled peptides were washed 3 times with 150 µL of 0.1% formic acid (1200 g, 2 min), then eluted from the resin using 60 µL 60 % acetonitrile (1200 g, 2 min). A volume of 55 µL was taken from each sample and combined, then dried in a centrifugal vacuum concentrator at 40 °C (SpeedVac). Dried samples were reconstituted in 0.1% Trifluoroacetic acid and fractionated using Pierce High pH Reversed-Phase Peptide fractionation kit (# 84868, Thermo Scientific™), following manufacturer instructions. Fractionated peptides were dried in a centrifugal vacuum concentrator at 40 °C (SpeedVac), reconstituted in 0.1% formic acid, and measured using LC-MS/MS.

LC–MS/MS measurements and data analysis

The peptide fractions were resuspended in 0.1% TFA and 2% acetonitrile in water. Chromatographic separation was performed on an Easy-Spray Acclaim PepMap column (50 cm length × 75 µm inner diameter; Thermo Fisher Scientific) at 55 °C by applying 120 min acetonitrile gradients in 0.1% aqueous formic acid at a flow rate of 300 nl/min. An UltiMate 3000 nano-LC system was coupled to a Q Exactive HF-X mass spectrometer via an easy-spray source (all Thermo Fisher Scientific). The Q Exactive HF-X was operated in TMT mode with survey scans acquired at a resolution of 60,000 at m/z 200. Up to 15 of the most abundant isotope patterns with charges 2–5 from the survey scan were selected with an isolation window of 0.7 m/z and fragmented by higher-energy collision dissociation with normalized collision energies of 32, while the dynamic exclusion was set to 35 s. The maximum ion injection times for the survey scan and dual MS (MS/MS) scans (acquired with a resolution of 45,000 at m/z 200) were 50 and 120 ms, respectively. The ion target value for MS was set to 3e6, and for MS/MS, it was set to 1e5, and the minimum AGC target was set to 1e3.

The data were processed with MaxQuant v. 1.6.17.0, and the peptides were identified from the MS/MS spectra searched against Uniprot Human Reference Proteome (UP000005640) using the built-in Andromeda search engine [44–46]. Reporter ion MS2-based quantification was applied with reporter mass tolerance = 0.003 Da and min. reporter PIF = 0.75. Cysteine carbamidomethylation was set as a fixed modification, and methionine oxidation, glutamine/asparagine deamination, as well as protein N-terminal acetylation were set as variable modifications. For in silico digests of the reference proteome, cleavages of arginine or lysine followed by any amino acid were allowed (trypsin/P), and up to two missed cleavages were allowed. The FDR was set to 0.01 for peptides, proteins, and sites. Match between runs was enabled. Other parameters were used as pre-set in the software. Reporter intensity corrected values for protein groups were loaded into Perseus v. 1.6.10 [47]. Standard filtering steps were applied to clean up the dataset: reverse (matched to decoy database), only identified by site, and potential contaminant (from a list of commonly occurring contaminants included in MaxQuant) protein groups were removed. Reporter intensity corrected values were log2 transformed, and protein groups with values across all samples were kept. The values were normalized by median subtraction within TMT channels. To determine proteins whose thermal stability was affected by the supplementation of 5-FU compared to the vehicle control, Student’s t-test (2-sided, permutation-based FDR = 0.05, S0 = 0.1, n = 3) was performed. The table was exported from Perseus and formatted to its final form in Microsoft Excel 2016.

Western blotting

Vehicle and control-treated cells were first washed with PBS. Protein lysates were prepared using lysis buffer containing 50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1.0% NP-40, 0.1% SDS, 25 units/ml Benzonase, and 100 µM PMSF (#36978, Thermo Scientific™). Total protein extracts were isolated by centrifugation at 12000 g for 10 minutes at 4°C. Protein concentrations were determined using a BCA assay (Thermo Scientific™ #23225). Equal amounts of proteins (10–20 µg) were then resolved using SDS-PAGE and transferred to a Polyvinylidene fluoride (PVDF) membrane. Upon transfer, membranes were blocked with tris-buffered saline containing 0.05 % Tween 20 supplemented with 5 % non-fat milk for 1 hour, followed by overnight incubation in primary antibodies. Subsequently, membranes were washed three times with TBST, followed by incubation in an appropriate dilution of secondary antibody at room temperature for 1 hour. Antibody dilutions were prepared following the manufacturer's instructions. The chemiluminescence signal was detected using Amersham ImageQuant 800 systems. The following antibodies were purchased from Invitrogen: phospho-AKT1 (Ser473) # 700256; AKT Pan # MA5–14916, Phospho-p70 S6 Kinase (Thr389) #710095, P70 S6 Kinase #MA5-15141, P21 Monoclonal Antibody (R.229.6) # MA5-14949, GAPDH # MA5-15738. SIN1 antibody was purchased from Bethyl Laboratories #A300-910A-T. Densitometry analysis of proteins was performed using ImageJ quantification software to measure the relative band intensity. For the TYMS protein, both slower and faster migrating bands were included for quantification of the total TYMS levels.

Packaging shRNA-encoding lentivirus

HEK293T cells were plated overnight in 6-well plates at 50% confluency. The medium was replaced before transfection. A mixture containing Packaging plasmid psPAX2, 1 μg (addgene#12260), Envelope plasmid VSV-G, 0.2 μg (Addgene #12259), and 1μg of each sh-RNA (Control and targeting Sh-RNA) were resuspended in 100 μL of Opti-MEM (Gibco, #31985062). The second mixture containing 6.6 µg polyethylenimine (PEI) (Sigma Aldrich, #913375) and 100 μL of Opti-MEM was prepared in a separate tube. The two mixtures were combined, vortexed, and incubated for 20 minutes at RT. The mixture was then added to each well dropwise. The medium was replaced after 4 hours. Forty-eight hours later, lentiviral supernatants were collected and centrifuged for 5 minutes at 12000 g, then stored at −80°C until further use. The following shRNAs were used: SIN1 (shRNA1: Addgene #13483, and shRNA2: Addgene #13484). ShRNA control Empty (Sigma Aldrich, # SHC001), P53 (shp53 pLKO.1 puro: Addgene, #19119).

Lentivirus infection and selection

MCF7 and MDA-MB-231 cells were plated overnight in 6-well plates at 50% confluency. For virus infection, 1 mL of fresh medium was mixed with 1 mL of lentiviral supernatant, then directly added to the cells and left overnight. Upon medium change, cells were incubated for an additional 24 hours. The selection was carried out using puromycin (Sigma Aldrich, # P4512). The following concentrations were used: MCF7 3 µg/ml, MDA-MB-231 2 µg/ml) for 96 hours. Real-time PCR and western blot were used to verify the knockdown efficiency.

RNA Isolation

Vehicle and control-treated cells were washed with PBS and resuspended in appropriate volumes of TRI Reagent (Thermo Fisher Scientific, #AM9738). RNA isolation was carried out according to the manufacturer’s instructions. Briefly, cells were lysed in TRI Reagent by incubating for 5 min at room temperature. The mixture was then supplemented with an appropriate volume of chloroform, followed by vigorous shaking for 20 seconds. Samples were incubated at room temperature for an additional 5 minutes and then centrifuged at 15,000 × g for 15 min at 4°C. The aqueous solutions containing RNAs were transferred to a new tube. RNA precipitation was carried out by adding one volume of isopropanol and incubating at −20°C for 1 hour, followed by centrifugation at 12,000 × g for 20 min at 4°C. The RNA pellet was washed with 75% ethanol and centrifuged at 12,000 × g for 10 min at 4°C. RNA concentrations were determined, and cDNA synthesis was carried out as described below.

cDNA synthesis and quantitative real-time PCR

RNA samples prepared above were used for cDNA synthesis using the High-Capacity cDNA Reverse Transcription Kit (#4368814, ThermoFisher), 500ng of RNA per reaction. Then, Real-time PCR was carried out using PowerUp™ SYBR™ Green Master Mix (ThermoFisher, #A25776) according to the manufacturer’s instructions. Briefly, a mixture containing 10 ng of cDNA, 7 μL SYBR™ Green Master Mix, 300 nM forward and reverse primers, and appropriate volumes of nuclease-free water was prepared and analyzed in duplicates using a Light Cycler 480 Real-Time PCR System. The following steps were used: First, an activation step at 95 °C for 2 min, followed by 40 cycles at 95 °C for 30 s, 61 °C for 30 s, and 72 °C for 30 s. A melting curve analysis of the PCR products was performed to verify the specificity of each pair of primers. Expression levels of the indicated genes were analyzed and quantified relative to Gapdh. The following primers were used:

• Forward-TYMS-CTGCTGACAACCAAACGTGTG,

• Reverse-TYMS-GCATCCCAGATTTTCACTCCCTT,

• Forward-SIN1- GGTGGACACCGATTTCCCC,

• Reverse-SIN1- CGCTTCACTGCCTTCAGTAAGA.

• Forward-GAPDH-GGAGCGAGATCCCTCCAAAAT,

• Reverse-GAPDH-GGCTGTTGTCATACTTCTCATGG,

• F-CDKN1A- CGATGGAACTTCGACTTTGTCA

• R-CDKN1A- GCACAAGGGTACAAGACAGTG

• F-GADD45A- CCCTGATCCAGGCGTTTTG

• R-GADD45A- GATCCATGTAGCGACTTTCCC

• F- SESN2-CCTCTGGGCGAGTAGACAAC

• R- SESN2-GGAGCCTACCAGGTAAGAACA

• F-TP53-GAGGTTGGCTCTGACTGTACC

• R-TP53-TCCGTCCCAGTAGATTACCAC

Drug Affinity Responsive Target Stability (DARTS)

DART assay was carried out as described previously [15]. Briefly, cell lysate was prepared using NP-40 lysis buffer containing 50 mM Tris-HCl pH 8.0, 150 mM NaCl, 1% NP-40, and 1 mM PMSF. Lysates were then cleared by centrifugation at 10000 g for 10 minutes at 4°C. Protein concentrations were first determined using a BCA assay (Thermo Scientific™ #23225). Equal amounts of lysates (100–150 µg) were treated with DMSO or the indicated inhibitors and incubated on a rotating shaker for 30 minutes at RT. Mixtures were collected by short pulse centrifugation, supplemented with control solution or pronase (Roch, #10165921001), and digested for 15 minutes at RT. Digestion was stopped by adding Laemmli Sample Buffer and incubating at 95 °C for 5 min. Samples were then analyzed using a regular Western blot and stained with the indicated antibodies as described above.

Metabolic labeling of proteins in MDA-MB-231 cells with β-ethynylserine (βES)

Metabolic labeling was carried out as described previously [18, 19]. An equal number of MDA-MB-231 cells were seeded in a T-75 flask and grown overnight in complete medium. Labeling experiments were performed by adding 1 mM βES at 37 °C in a complete medium for 60 min. Control samples were simultaneously pre-treated with 10 µg/ml cycloheximide to block translation. One hour later, cells were treated with DMSO, 50 µM FuDR, or 50 µM FuR for an additional 2 hours. At the end of incubation, cells were washed with PBS and resuspended in 200 μL of lysis buffer containing 2 % SDS in PBS. Samples were sonicated using a probe sonicator (1× 10 seconds) at room temperature.

Bioorthogonal conjugation of βES-labeled proteins to Azide-PEG3-biotin in cell lysate, enrichment of biotinylated proteins, and targeted detection by Western blotting

Lysates prepared after labeling cells with β-ethynylserine above were used as follows. Lysates were diluted with 200 μL PBS, then supplemented with 40 mM iodoacetamide and incubated at room temperature for 40 min under shaking with a thermomixer (900–1000 rpm at room temperature). Proteins were then precipitated with 400 μL methanol and 100 μL chloroform, then collected by centrifugation (4000× g, 3 min) at RT. The protein pellet was washed twice with methanol and dissolved in 50 μL PBS containing 2 % SDS, followed by 150 μL PBS (0.5 % SDS final).

Copper-catalyzed azide-alkyne click (CuAAC) reaction was performed for 1 hour at RT by adding 14 µL CuAAC reaction mixture to the 200 μL proteins. The final concentrations of CuAAC in the reaction mixture are as follows: 1 mM CuSO4, 200 μM THPTA, 1 mM TCEP, and 100 μM Azide-PEG3-biotin. The click reaction was stopped by adding 2 μL of 500 mM EDTA. Proteins were then precipitated with 1 volume of methanol and ¼ volume of chloroform, then collected by centrifugation (4000× g, 3 min) at RT. The protein pellet was washed twice with methanol and dissolved in 100 μL PBS containing 2 % SDS, and diluted with 300 μL of PBS. Finally, enrichment of biotine-βES-protein conjugates was achieved using Pierce™ NeutrAvidin™ Agarose beads (Thermo Scientific™, #29201), following the manufacturer's instructions. Bound proteins were eluted using SDS sample loading buffer, separated using SDS-PAGE, transferred onto nitrocellulose membrane, and probed with protein-specific antibodies TYMS, AKT, and PARP1. PARP1 was used as an internal control because it shows high and relatively stable expression under the conditions described above, making it a robust and easily detectable reference for this experiment.

Cell cycle analysis (SubG1 measurement)

MDA-MB-231 cells were harvested following 72 h treatment with DMSO or 50 µM FuDR. Cells were first trypsinized, washed twice with PBS, and fixed in 70% ethanol at −20°C for 24 hours. Fixed cells were then resuspended in 400 mL of PBS supplemented with 0.2 mg/ml DNase-free RNase A (Thermo Scientific™, EN0531), 10 µg/ml propidium iodide (Invitrogen™, P1304MP), and incubated at 4 °C for 2 hours protected from light. Stained cells were analyzed using a NovoCyte Flow Cytometer System with 1–3 Lasers. DNA content determination was carried out using the Novoexpress built-in cell cycle analysis module.

Colony formation assay

MCF7 and MDA-MB-231 cells stably transfected with shRNA control or shRNA against SIN1 were plated into 24-well plates, each containing 1 mL of culture medium (4000 cells/mL). After 24 hours in culture, the medium was changed with fresh medium supplemented with vehicle (DMSO) or the indicated concentrations of 5-fluorodeoxyuridine (FuDR). Cells were grown for 10 to 12 days at 37 °C and 5 % CO2. Cell colonies were first fixed with methanol and then stained with 0.5 % (w/v) crystal violet (Sigma Aldrich, Cat# C0775). Culture plates were imaged using GelDoc (Bio-Rad).

Polysome fractionation

Polysome fractionation was performed as previously described [48–50]. Briefly, MDA-MB-231 Sh control and Sh-SIN1 cells were treated with 50µM for 18 hours. Total and polysomal RNA were isolated from confluent 10 cm of each condition in triplicate. Cells were washed with PBS and incubated for 10 min at 37 °C in medium supplemented with 50 µg/mL cycloheximide (CHX) (Sigma-Aldrich). Immediately afterward, cells were transferred to ice, washed with ice-cold PBS containing 10 µg/mL CHX, and lysed in 500 µL polysome lysis buffer (PLB) containing 10 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1.5 mM MgCl₂, 0.1% Triton X-100, 20 mM DTT, 40 U/mL RiboLock RNase Inhibitor (Thermo Fisher Scientific), and 150 µg/Ml CHX (Sigma-Aldrich). Lysates were incubated on ice for 40 min, vortexed every 10 min, and centrifuged at 10,000 rpm for 10 min at 4 °C. A total of 450 µL of the cleared supernatant from each lysate was layered onto linear 10%−60% (w/v) sucrose gradients (Thermo Fisher Scientific) prepared in 25 mM Tris-HCl (pH 7.4), 25 mM NaCl, 5 mM MgCl₂, and 2 mM DTT in nuclease-free water. Samples were centrifuged using an SW41T rotor (Beckman) at 37,000 rpm for 2 h 30 min at 4 °C and fractionated using a BioComp Gradient Station (BioComp). Fractions of 700 µL were collected and kept immediately on ice. Fractions 10 to 12 were pooled as Light polysomes and fractions 13–15 as Heavy polysomes. TRIzol reagent (Invitrogen) was added to each pooled sample at a ratio of 1.2 volumes. RNA was isolated with the Direct-zol™ RNA Microprep Kit (ZymoResearch), including in-column DNase treatment, following the manufacturer’s instructions. RNA was eluted from the column, and equal volumes of RNA from each pooled fraction were reverse-transcribed using the High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific). The resulting cDNA was used for TYMS expression analysis by RT-qPCR.

Statistical analysis

Statistical analysis was performed using GraphPad Prism (http://www.graphpad.com). A minimum of three biological replicates were carried out for each experiment. An unpaired, two-sided test was used to compare the means of two groups. One-way ANOVA with Dunnett test post hoc test was applied to compare the means of three or more groups. All results in the graphs are expressed as means ± SEM. P < 0.05 was considered to be significant.

Authors’ contributions

Conceptualization, A.A, Methodology; A.A, R.S, Proteomic sample preparation and data analysis: A.A, A.E.S, M.Z, R.S; Investigation, A.A, A.E.S, N.G, V.L; original draft preparation, AA; Writing review and editing, A.A, R.S. Funding acquisition, A.A.

Funding

This work was supported by the National Science Centre, Poland, NCN SONATA BIS grants to A. AZZI, number: 2022/46/E/NZ3/00144.

Data availability

Proteomic data that support the findings of this study are available at ProteomeXchange database (PRIDE), Project accession: PXD072095.

Declarations

Ethics approval and consent to participate

Not applicable.

Competing interests

The authors declare no competing interests.