br the IC values for FU increased by fold for
the IC50 values for 5-FU increased by 2-fold, for Oxa increased by 12-fold, and for Dox by 5-fold, compared to SW48-5FU cells, respectively (Table 1). Consistently, colon cancer WiDr α-CEHC that carry homozygous R273H presented substantial resistance to these drugs, as compared either to SW48-Dox and TP53-Dox, or to SW48-5FU and TP53-5FU cells (Table 1). Further, we employed small interfering RNA (siRNA) to silence METTL3 expression (siMETTL3), and found that siMETTL3 treatments (100 nM) substantially sensitized the response of drug-resistant TP53-Dox cells to anticancer drugs. To wit, the IC50 values for Dox decreased by 7-fold in TP53-Dox cells treated with siMETTL3 in combination with Dox (Fig. 1B, C, Table 1). Also, based on reports that neplanocin A (NPC) could decrease METTL3 activity indirectly in m6A modification by inhibition of S-adenosylhomocysteine (SAH) hydrolase , we examined the responsiveness of TP53-Dox cells to drugs after NPC treatments (20 nM). We found that NPC, like siMETTL3, could also sensitize TP53-Dox cells to anticancer drugs, decreasing the IC50 values for Dox by 12-fold, for Oxa by 2-fold and for 5-FU by 2-fold,
respectively, comparing to vehicle control (Fig. 1B, C, Table 1).
In parallel experiments, treatments with the GCS inhibitor PDMP that sufficiently inhibited GCS, and consequently decreased glyco-sphingolipid production, significantly sensitized the responsiveness of TP53-Dox cells, but not SW48-Dox cells, to these same three drugs (Fig. 1C, Table 1). To wit, the IC50 values decreased, respectively, for Dox by 12-fold (Fig. 1C), and for Oxa or 5-FU, by 2-fold, in TP53-Dox cells treated with PDMP in individual combination with each of these three anticancer drugs (Table 1).
We further examined the effects of p53 R273H on apoptosis. Cells were pretreated with NPC (20 nM) or vehicle, and then exposed to Dox (100 nM, 48 h) to induce apoptosis. Flow cytometry assay with a FITC BrdU kit showed that Dox induced apoptosis in SW48-Dox cells, whereas TP53-Dox cells displayed resistance to the apoptosis, as the apoptotic cells were approximately 4-fold fewer than those seen in SW48-Dox (5.3% vs. 25.6% of total cells) (Fig. 2A, B). NPC treatments significantly increased Dox efficacy in inducing apoptosis in TP53-Dox cells, but not in SW48-Dox cells. In TP53-Dox cells treated with NPC,
apoptotic cells significantly increased, by approximately 5-fold (25% vs. 5% of total cells), comparing to vehicle treatments (Fig. 2A, B). Further, Western blotting analysis indicated that cleaved PARP, a DNA nick sensor that was cleaved from PARP by activated caspase-3 or caspase-7 in apoptosis, was significantly increased, by approximately 8-fold in TP53-Dox cells treated with NPC and Dox in combination (Fig. 2C). These results indicate that drug resistance of cancer cells hetero-zygously carrying R273H mutant TP53 highly correlates to activity of METTL3, an enzyme responsible for m6A RNA methylation.
3.2. m6A modification in pre-mRNA transited codon 273 determines the mutant protein expression in cells heterozygously carrying TP53 R273H mutation
We next wished to identify the roles of m6A modification in reg-ulating protein expression of the R273H p53 mutant. TP53-Dox cells (heterozygous for the missense codon for R273) expressed mutant p53 protein, which could not be phosphorylated, with consequent loss of transactivation of p53 target genes or p53-responsive genes involved in cell proliferation arrest and apoptosis [19,31]. The levels of phos-phorylated p53 (pp53), as well as levels of p21, Bax and PUMA, in TP53-Dox cells, particularly those exposed to Dox (100 nM for last 48 h), were significantly lower than in SW48-Dox cells under the same conditions, due to the fact that in the latter cells, DNA damage stress could upregulate wt p53 expression (Fig. 3A, B). Interestingly, NPC treatments (20 nM) significantly increased the levels of pp53, by more than 7-fold (0.52 vs 0.07), and of p21, PUMA and Bax, by approxi-mately 3-fold, in TP53-Dox cells, as compared to vehicle, respectively (Fig. 3A). Treatments with PDMP, which inhibits GCS activity and the downstream synthesis of GSLs including globotriaosylceramide (Gb3) , decreased METTL3 protein levels by approximately 3-fold (0.31 vs 0.81), increased the pp53 levels by 6-fold, and increased p21 or PUMA and Bax levels by 3-fold, compared to vehicle treatments in TP53-Dox cells. The increases in protein levels of pp53, p21, Bax and PUMA brought about by NPC and PDMP were also observed for the treated TP53-Dox cells absent Dox exposure (Fig. 3B), although these differ-ences were not significant (Fig. 3B), and overall the levels of these proteins in these cells were lower than in those subjected to Dox ex-posure that cause DNA damage stress (Fig. 3A, B). Importantly, treat-ments with NPC or PDMP were without significant effect on levels of pp53, or of p21, Bax or PUMA, in the wt p53 SW48-Dox cells (Fig. 3A, B).