Oncogenic PIK3CA mutations increase dependency on the mRNA cap methyltransferase, RNMT, in breast cancer cells

Basic mechanisms in gene expression are currently being investigated as targets in cancer therapeutics. One such fundamental process is the addition of the cap to pre-mRNA, which recruits mediators of mRNA processing and translation initiation. Maturation of the cap involves mRNA cap guanosine N-7 methylation, catalysed by RNMT (RNA guanine-7 methyltransferase). In a panel of breast cancer cell lines, we investigated whether all are equivalently dependent on RNMT for proliferation. When cellular RNMT activity was experimentally reduced by 50%, the proliferation rate of non-transformed mammary epithelial cells was unchanged, whereas a subset of breast cancer cell lines exhibited reduced proliferation and increased apoptosis. Most of the cell lines which exhibited enhanced dependency on RNMT harboured oncogenic mutations in PIK3CA, which encodes the p110α subunit of PI3Kα. Conversely, all cell lines insensitive to RNMT depletion expressed wild-type PIK3CA. Expression of oncogenic PIK3CA mutants, which increase PI3K p110α activity, was sufficient to increase dependency on RNMT. Conversely, inhibition of PI3Kα reversed dependency on RNMT, suggesting that PI3Kα signalling is required. Collectively, these findings provide evidence to support RNMT as a therapeutic target in breast cancer and suggest that therapies targeting RNMT would be most valuable in a PIK3CA mutant background.

Basic mechanisms in gene expression are currently being investigated as targets in cancer therapeutics. One such fundamental process is the addition of the cap to pre-mRNA, which recruits mediators of mRNA processing and translation initiation. Maturation of the cap involves mRNA cap guanosine N-7 methylation, catalysed by RNMT (RNA guanine-7 methyltransferase). In a panel of breast cancer cell lines, we investigated whether all are equivalently dependent on RNMT for proliferation. When cellular RNMT activity was experimentally reduced by 50%, the proliferation rate of nontransformed mammary epithelial cells was unchanged, whereas a subset of breast cancer cell lines exhibited reduced proliferation and increased apoptosis. Most of the cell lines which exhibited enhanced dependency on RNMT harboured oncogenic mutations in PIK3CA, which encodes the p110a subunit of PI3Ka. Conversely, all cell lines insensitive to RNMT depletion expressed wild-type PIK3CA. Expression of oncogenic PIK3CA mutants, which increase PI3K p110a activity, was sufficient to increase dependency on RNMT. Conversely, inhibition of PI3Ka reversed dependency on RNMT, suggesting that PI3Ka signalling is required. Collectively, these findings provide evidence to support RNMT as a therapeutic target in breast cancer and suggest that therapies targeting RNMT would be most valuable in a PIK3CA mutant background.

Background
Basic mechanisms in gene expression are currently being investigated as targets in cancer therapeutics [1,2]. In eukaryotes, gene expression is dependent on the mRNA cap, a structure that protects nascent transcripts from nucleases and recruits complexes which mediate mRNA processing, including splicing, export and translation initiation [3][4][5]. The mRNA cap forms during transcription and remains on the transcript throughout its lifetime. The basic cap structure is m7G(5 0 )ppp(5 0 )Xm (X denotes the first transcribed nucleotide), 7-methylguanosine linked to the first transcribed nucleotide via 5 0 to 5 0 triphosphate bridge. A series of enzymes catalyse cap formation initiating with RNGTT (RNA guanylyltransferase and 5 0 phosphatase), which catalyses the addition of the inverted guanosine cap to the nascent transcript. Subsequently, RNMT (RNA guanine-7 methyltransferase) and its activating subunit, RAM (RNMT-activating miniprotein), are recruited to transcribing RNA pol II [6][7][8]. RNMT-RAM methylates the guanosine cap on the N-7 position to create the cap structure, m7G(5 0 )ppp(5 0 )X [9,10]. RNMT -RAM also has a major role in enhancing the transcription of the most active genes via interactions with protein complexes of the transcription machinery [11]. RNMT -RAM can also be involved in the process of recapping, during which partially capped or uncapped transcripts receive a cap in the cytoplasm [12]. Thus, RNMT -RAM has a potent role in gene expression, promoting transcription and translation. Cellular signalling pathways can regulate mRNA cap formation on specific target mRNAs, resulting in the regulation of their expression. For example, c-Myc and E2F1 (and potentially other transcription factors) increase RNA pol II phosphorylation which promotes mRNA cap formation via recruitment of the capping enzymes [13][14][15][16][17]. c-Mycdependent mRNA cap methylation also requires upregulation of SAHH, the enzyme which hydrolyses the inhibitory biproduct of methylation reactions [18]. Upregulation of mRNA cap formation is critical for c-Myc function; inhibition of this process is synthetic lethal with deregulated c-Myc [15,18]. Formation of the mRNA cap is also regulated during the cell cycle by CDK1-dependent phosphorylation and activation of RNMT [19]. As CDK1 and c-Myc are implicated in tumorigenic events and cap-dependent mRNA translation is deregulated in many cancers, this raises the potential of utilizing the mRNA capping enzymes as therapeutic targets to selectively inhibit protein synthesis in cancer cells [20,21]. In particular, because RNMT ligands have low polarity, there may be an opportunity to create active site inhibitors with a low charge that are able to cross the plasma membrane [22].
When considering RNMT as a therapeutic target, is it valuable to determine which oncogenes drive a heightened state of dependency on RNMT. This gives an important indication of which cancer types and patient populations are most likely to response to RNMT inhibitors. In this study, we investigated the impact of RNMT depletion on the proliferation of a panel of breast cancer cell lines. We found that cells with activating mutations in PIK3CA, which are present in around 35% of breast cancers, exhibit enhanced dependency on RNMT for proliferation [23]. Cells expressing wild-type (WT) PIK3CA had their dependency on RNMT increased by the exogenous expression of oncogenic PIK3CA mutants, which increase PI3Ka activity. Conversely, cells expressing oncogenic PIK3CA mutations had their enhanced RNMT dependency reversed by the use of low doses of PI3Ka inhibitors, suggesting that PI3Ka signalling is mediating this dependency. Taken together, these data demonstrate that breast cancer cells with activating mutations in PIK3CA are selectively sensitive to RNMT inhibition.

Tissue culture
Cells were cultured at 378C and 5% CO 2 . The carcinoma cells were purchased from the American Type Culture Collection (ATCC) and IMECs (immortalized normal mammary epithelial cell lines) kindly provided by Dr James DiRenzo (Dartmouth Medical School). Experiments were typically performed within four weeks of culture. Breast carcinoma cell lines were maintained in RPMI, 10% FBS, 2 mM L-glutamine and 1% Anti-anti. IMEC cells were maintained in DMEM/F12, 2 mM L-glutamine, 5 mg ml 21 insulin, 10 ng ml 21 EGF, 0.5 mg ml 21 hydrocortisone and 1% Antianti. pBMN-IRES neo-constructs containing Myc-PIK3CA, RNMT-GFP and mutants or empty vector control were transduced into cells by retro-viral infection and selected using 0.5 mg ml 21 G418 for 7 days.

PIK3CA sequencing
RNA was extracted using the GeneJET RNA Purification Kit (ThermoFisher Scientific). Approximately 5-10 mg of RNA was digested with 2 ml of Turbo DNAse (ThermoFisher Scientific) to remove genomic DNA and purified using the RNA Clean & Concentrator Kit (Zymo Research), and 1 mg of RNA was used in the reverse transcription reaction using RevertAid H Minus Reverse Transcriptase (ThermoFisher Scientific) and the following RT primer: tcaaaccctgtttgcgttta. About 2 ml of reverse transcription reaction was used as a template in the PCR with KOD Hot Start DNA Polymerase (Merck) with primers PIK3CA_fwd2 tgctttgggacaaccataca and PIK3CA_rev3 cctatgcaatcggtctttgc. The PCR product was purified with the QIAquick PCR Purification Kit (Qiagen) and sequenced using the following primers: PIK3CA_ FWD2 tgctttgggacaaccataca; 669R cacagtcatggttgattttcaga; 1166R cctgggattggaacaaggta; 1288F ggcatggggaaatataaacttg; 1664R tctcctgctcagtgatttcaga; 2143R gcctcgacttgcctattcag; 2312F caggcttgaagagtgtcgaa; 2636R gaactgcagtgcacctttca; PIK3CA_REV3 cctatgcaatcggtctttgc.

Statistical analysis
Statistical significance was assessed by one-way analysis (ANOVA) followed by Dunnett's multiple comparison tests using GraphPad PRISM 5.0. A value of p 0.05 is denoted with *, p 0.01 denoted with **, p 0.001 denoted with ***.

Cell extract preparation
Cell lysis was performed at 48C. Culture media were removed, cells were washed twice with ice-cold PBS and lysed in ice-cold F buffer, comprising 10 mM Tris (pH 7.05), 50 mM NaCl, 30 mM Na-pyrophosphate, 50 mM NaF, 5 mM ZnCl 2 , 10% glycerol, 0.5% Triton X-100, 1 mM EGTA, royalsocietypublishing.org/journal/rsob Open Biol. 9: 190052 1 mM EDTA, and 1 mM sodium orthovanadate) supplemented with 0.1 TIU (trypsin inhibitor unit) aprotinin, 1 mM pepstatin, 10 mM leupeptin and 1 mM DTT immediately before use. For analysis of phosphorylated protein, lysis buffer was supplemented with Sigma Phosphatase Inhibitors (cocktail mixtures 2 þ 3). Cell lysates were collected by scraping and the soluble fraction was collected following centrifugation at 16 000 g for 10 min at 48C. Protein concentration was determined using the Bradford method and extracts were normalized for protein content. Typically, 5-20 mg of cell extract was analysed. Band intensity was quantitated using Image J software.

Antibodies
Anti-RNMT, RAM and AKT antibodies were developed in-house and raised against full-length recombinant human proteins in sheep and sera purified against the antigen.   (table 1). Cell lines were purchased from ATCC (American Type Culture Collection) and used within four to six weeks of culture to reduce passage-dependent effects. Known mutations of cancer-associated genes in these cell lines were extracted from the COSMIC database (table 1). In addition, a low-passage, non-transformed TERT-IMEC (TERT-immortalized mammary epithelial cell line) was analysed [24]. RNMT expression was reduced by transfection of three independent RNMT siRNAs and a non-targeting siRNA control. All cell lines harbouring PIK3CA-activating mutations (MCF7, JIMT-1 and T47D, marked with a red asterisk), and one cell line expressing WT PIK3CA (HCC-1806), exhibited reduced proliferation in response to transfection of all three RNMT siRNAs (figure 1a). By contrast, the proliferation of IMECs and other breast cancer cell lines expressing WT PIK3CA (BT-549, MDA-MB-231, CAMA-1 and ZR-75-1) was unaffected by RNMT siRNA transfection. The PIK3CA coding region sequence was verified for all cell lines (electronic supplementary material, figure S1). Combining three independent experiments, the average cell number 96 h after RNMT siRNA 1 transfection was calculated relative to the cell number in a control transfection (figure 1b). In agreement with the growth curves in figure 1a, in response to RNMT siRNA transfection, there was a significant reduction in the cell number for MCF7, HCC-1806, JIMT-1 and T47D cell lines compared with IMEC. However, BT-549, CAMA-1, ZR-75-1 and MDA-MB-231 cell lines did not exhibit a significant loss in cell number following RNMT siRNA transfection. This difference in sensitivity to RNMT siRNA transfection did not correlate with the level of RNMT reduction. All cell lines exhibited an equivalent reduction in RNMT expression in response to transfection of all three RNMT siRNAs (figure 1c). As observed previously, expression of the RNMT-activating subunit, RAM, was also reduced in response to reduced RNMT expression (figure 1c) [10,25]. RNMT and RAM are co-translated, and their interaction protects the two proteins from proteasome-mediated degradation.
Since we observed a similar proliferative defect following transfection of three independent RNMT siRNAs, it is likely that this impact is specifically due to a reduction in RNMT levels. To strengthen this conclusion, an RNMT-GFP construct resistant to RNMT siRNA 1 was introduced into HCC1806 cells by retroviral infection. As expected, transfection of RNMT siRNA 1 inhibited endogenous RNMT expression but did not inhibit expression of RNMT-GFP (figure 2a). Consistent with figure 1, depletion of endogenous RNMT significantly inhibited the proliferation of control HCC1806 cells, but this effect was reversed by the expression of RNMT-GFP (figure 2b). This confirms that the proliferative defect observed following RNMT siRNA transfection is probably due to a reduction in RNMT expression and not a consequence of siRNA off-target effects The reduction in cell number that we observed in response to reduced RNMT expression could be due, in part, to an increase in apoptosis. Inhibition of RNMT and RAM has been observed to induce apoptosis in HeLa cells, a cervical cancer cell line [10,26]. We analysed apoptosis in a subset of our breast cancer cell panel by detecting fragments of PARP, a nuclear antigen cleaved during apoptosis (figure 2c). All four cell lines investigated were competent to undergo apoptosis; all exhibited cleaved PARP in response to incubation with Staurosporine, a pro-apoptotic agent. Cell lines exhibiting enhanced dependency on RNMT (MCF7 and HCC-1806), also exhibited PARP cleavage in response to reduced RNMT expression, indicating that apoptosis was taking place. MDA-MB-231 and ZR-75-1 cells, which do not exhibit a proliferative defect in response to RNMT siRNA transfection, also did not exhibit PARP cleavage in response to reduced RNMT expression (figure 2c). These data demonstrate that RNMT depletion induces a cytotoxic response in RNMT-dependent breast cancer cell lines.

RNMT dependency does not correlate with RNMT expression, activity or rate of protein synthesis
To investigate the molecular mechanisms governing dependency on RNMT, we investigated whether cells sensitive to royalsocietypublishing.org/journal/rsob Open Biol. 9: 190052  We investigated whether cellular dependency on RNMT correlated with RNMT expression or activity, measuring both basal levels and levels following RNMT siRNA transfection. RNMT and RAM expression were analysed by western blot performed on four independent samples (figure 3a,b). Expression of RNMT and RAM varied little across the cell panel, and a correlation between RNMT and RAM expression and dependency on RNMT was not observed. Cellular N7 cap guanosine methyltransferase activity, which is dependent on RNMT expression, also did not correlate with dependency on RNMT (figure 3c,d ) [27]. For each cell line, the mRNA cap methyltransferase royalsocietypublishing.org/journal/rsob Open Biol. 9: 190052 activity assay was performed on titration of cell extracts to determine the linear range of the assay (electronic supplementary material, figure S2). In this assay, a guanosinecapped transcript was incubated with cell extracts and the methyl donor, SAM (S-adenosyl methionine). Following incubation, the proportion of transcripts with an N7 methylguanosine cap (m7G(5 0 )ppp(5 0 )X) was determined by thin layer chromatography. Across the cell panel, only two cell lines exhibited significantly different N7 cap guanosine methyltransferase activity compared with IMEC; however, one was a cell line sensitive to RNMT siRNA transfection (HCC-1806), and the other was insensitive (CAMA-1) (figure 3c). Furthermore, the reduction in N7 cap guanosine methyltransferase activity following RNMT siRNA transfection did not vary significantly across the cell panel ( figure 3d). All cell lines exhibited a 40-60% reduction in N7 cap guanosine methyltransferase activity in response to RNMT siRNA transfection, consistent with RNMT being equivalently depleted (figure 1c). Given that RNMT -RAM expression is required for efficient cap-dependent translation, a reasonable hypothesis was that cells with higher protein synthesis or proliferation rates are more sensitive to loss of RNMT. An approximation of the protein synthesis rate was determined by measuring the rate of radio-labelled amino acid incorporation into cellular proteins. Although HCC1806, JIMT-1, CAMA-1 and ZR-75-1 cells exhibited higher basal net protein synthesis rates than IMEC, this did not correlate with dependency on RNMT (figure 3e). The reduction in protein synthesis in response to RNMT siRNA transfection also did not vary significantly across the cell panel ( figure 3f ). Furthermore, basal cell doubling time did not correlate with sensitivity to a reduction in RNMT (figure 3g).

RNMT dependency is induced by expression of oncogenic PIK3CA mutants
Three of the breast cancer cell lines sensitive to a reduction in RNMT have oncogenic hotspot activating PIK3CA mutations; MCF7 expresses PIK3CA E545 K, T47D expresses PIK3CA H1047R, and JIMT-1 expresses PIK3CA C420R (table 1; electronic supplementary material, figure S1). Conversely, all of the cell lines insensitive to RNMT inhibition expressed PIK3CA WT. No other cancer-associated mutations in the cell panel correlated with sensitivity to RNMT depletion. PIK3CA encodes the p110a subunit of the heterodimeric lipid kinase, PI3 K, which, upon stimulation by growth factors, activates downstream kinases including AKT and PDK1 (table 1). These effector kinases regulate major processes, including cell proliferation, growth and survival [28,29]. PIK3CA mutations are found in approximately 35% of breast cancer patients, with C420R, E545K and H1047R being the most common variants [23,30,31]. No other cancer-associated mutations in the cell panel correlated with sensitivity to RNMT depletion. Oncogenic PIK3CA mutants can confer growth advantage and render cells exquisitely sensitive to inhibition of PI3Ka [32 -34]. Indeed, MCF7, T47D and JIMT-1 cell lines were found to have the highest sensitivity to the PI3Ka inhibitor, BYL719, in the breast cancer cell line panel, demonstrating that these cell lines require PI3Ka for proliferation and survival [35] (electronic supplementary material, figure S3).    PIK3CA and were demonstrated to be insensitive to RNMT siRNA transfection (figures 1 and 2). As expected, the oncogenic mutants increased PI3K signalling, as assessed by AKT phosphorylation (figure 5b,f ). As with ZR-75-1 cells, expression of oncogenic PIK3CA mutants did not affect RNMT expression or the efficacy of RNMT siRNA ( figure 5c,g). However, in both cell lines investigated, transfection of two independent RNMT siRNAs caused a   To extend these findings, similar experiments were performed using isogenic human mammary epithelial cells in which an endogenous copy of the PIK3CA gene was replaced by either PIK3CA E545K or WT control. The benefit of these cell lines is that the PIK3CA mutant is expressed at endogenous levels in an untransformed mammary epithelial cell line. RNMT expression was reduced by transfection of two independent siRNAs resulting in a similar level of knockdown (figure 6a). Following RNMT reduction, there was a significant reduction in p110a E545K-expressing cells but not in p110a WT-expressing cells ( figure 6b).
To extend the finding that oncogenic PIK3CA mutants sensitize cells to a reduction in RNMT, RNMT knockdown was performed in three more breast cancer cell lines (HCC38, HCC1569 and HCC1500) and one lung carcinoma cell line (NCI-H460). The two cell lines with WT PIK3CA (HCC38 and HCC1569) were insensitive to a reduction in RNMT, whereas the two cell lines with mutations in PIK3CA (NCI-H460 and HCC1500) exhibited reduced proliferation in response to RNMT siRNA transfection ( figure 6c,d). This is consistent with cancer cells harbouring oncogenic PIK3CA mutations exhibiting enhanced dependency on RNMT.

Inhibition of PI3K signalling reduces RNMT dependency
To further investigate the hypothesis that oncogenic PI3K p110a signalling increases dependency on RNMT, we used inhibitors of PI3Ka activity. Phospho-AKT 473 served as a readout of PI3K inhibitor target engagement (figure 7a,c). GDC-0941 is a pan PI3K inhibitor which inhibits p110 a, b, d and g [37] (figure 7a). As expected, reducing RNMT expression in vehicle-treated T47D and MCF-7 cells significantly reduced cell proliferation (figure 7b). However, incubation with a sub-lethal dose of GDC-0941 suppressed this proliferative defect. BYL719 selectively inhibits p110a [37] (figure 7c). As expected, in vehicle-treated T47D and JIMT cells reducing RNMT expression resulted in reduced proliferation (figure 7d). However, incubation with a sublethal dose of BYL719 suppressed this proliferative defect. Therefore, suppression of PI3K p110a activity desensitizes these cell lines to RNMT inhibition. Collectively, these results

Discussion
Breast cancer is the most prevalent cancer in women worldwide, causing over 500 000 deaths annually [35,38]. Although the development of hormone and HER2-targeted therapies has greatly improved the prognosis of patients with luminal A/B and HER2 breast cancer, respectively, the emergence of resistance remains a major problem, limiting the effectiveness of drug treatment. Moreover, currently, there are limited treatment options for triple negative breast cancer, which represent 10-20% of breast cancer cases [39 -41]. The PI3K-AKT -mTOR pathway is a growth and survival pathway that is frequently deregulated in breast cancer due to genetic alterations in the pathway, including activating mutations in the oncogene PIK3CA and deletion of the tumour suppressor PTEN. Consequently, several inhibitors of this pathway are currently in clinical development for the treatment of breast cancer [42,43]. Among new targets being explored for the treatment of cancer are those that target the basic mechanisms of gene expression [1,2]. Deregulation of common oncogenes (e.g. c-Myc, PIK3CA, Ras and Raf ) and loss of common tumour suppressors (e.g. p53, PTEN and SWI/SNF) are associated with deregulated or enhanced gene expression. Since the formation of the mRNA cap is a key step in gene expression, we considered the mRNA cap methyltransferase, RNMT, which completes the basic functional mRNA cap, as a therapeutic target. Specifically, we were interested in identifying oncogenes or tumour suppressors, which sensitize cells to RNMT inhibition. The aim was to determine which breast cancer subtypes or gene signatures may be responsive to therapies targeting RNMT.
To address whether different breast cancer cell lines exhibit differential sensitivity to inhibition of RNMT, we reduced RNMT expression in a panel of breast cancer cell lines and non-transformed cells by transfection of RNMT siRNA (specific inhibitors of RNMT are not currently available). Using this technique, we reduced cellular cap methyltransferase activity by 50 -60% across the cell panel ( figure 3d). We observed that a subset of breast cancer cell lines exhibits enhanced sensitivity to inhibition of RNMT, compared with a non-transformed control cell line (figures 1 and 6). This differential dependency on RNMT did not correlate with RNMT expression level or activity, both in log-phase cells and following RNMT siRNA transfection (figures 1 and 3). However, oncogenic mutations in the gene PIK3CA, which encodes the p110a catalytic subunit of PI3K, were found in  The mRNA cap protects transcripts from degradation and recruits protein complexes involved in RNA processing and translation [47,48]. RNMT also has a major role in enhancing the transcription of the most highly expressed genes [11]. In previous studies in which levels of RNMT or its activator RAM were altered, genes responded differentially at the mRNA level and translation rate [11,14,19,49,50]. This results in cellular proteins and signalling pathways being differentially dependent on RNMT for expression. Why are genes differentially dependent on RNMT? Some transcripts may be dependent on high concentrations of RNMT to be methylated correctly. Certain guanosine-capped transcripts may have limited access to RNMT, because of the chromatin context, the speed of transcription or RNA secondary structure. Alternatively, some transcripts may not require their mRNA cap to be methylated on the cap guanosine, such as by using IRES (internal ribosome entry site)-dependent translation initiation. Although RNMT-catalysed methylation of the mRNA cap guanosine increases affinity for cap-binding complexes, it is not absolutely required. Some transcripts may be largely independent of the cap guanosine methylation for translation initiation. In experiments performed here, when cellular RNMT activity was reduced by approximately 50%, the global translation rate (approximated by amino acid incorporation into proteins) was reduced by only 5 -10% in most cell lines investigated, indicating that RNMT is present in excess for the expression of most genes (figure 3f ). The mechanism by which oncogenic PIK3CA increases dependency on RNMT remains an open question. Oncogenic PIK3CA mutants (and downstream activation of mTORC1driven cap-dependent translation) may alter the expression of proteins which render cells sensitive to RNMT inhibition. Alternatively, RNMT inhibition may affect protein expression similarly in all breast cancer cell lines, regardless of PIK3CA status; cells expressing PIK3CA may have increased dependency on a subset of RNMT-dependent proteins as a mechanism of oncogene addiction. We investigated whether the expression or phosphorylation of direct or indirect AKT substrates, which are involved in mRNA translation, were differentially expressed in IMECs and the breast cancer cell lines (electronic supplementary material, figure S7). The expression and phosphorylation level of AKT, eIF4E BP-1 and P70-S6 kinase did not correlate with sensitivity to RNMT suppression. We also investigated whether expression of the p85 and p110 subunits of PI3K were supressed in response to RNMT siRNA transfection, in MCF7 and HCC1806 (sensitive to RNMT siRNA transfection), and MDA-MB-231 and ZR-75-1 (insensitive to RNMT siRNA transfection). Expression of p85 and p110 was not robustly up or downregulated following 48 and 72 h RNMT siRNA transfection (electronic supplementary material, figure S8a). This experiment was also repeated with three independent RNMT siRNAs with similar results (not shown). This suggests that PIK3CA oncogenic mutations increase dependency on RNMT independently of increasing translation of p85 and p110.
In HeLa cells and other lines, repression of capping enzymes results in repression of c-Myc expression and RNA pol II phosphorylation [11,15]. c-Myc is induced by mitogenic signalling (including via the PI3K pathway) to promote cell proliferation, including increasing RNA pol II phosphorylation, which reflects global transcription levels. We therefore investigated whether, in the cell lines sensitive to RNMT siRNA transfection, c-Myc and/or RNA pol II phosphorylation is repressed when RNMT expression is reduced. In MCF7, HCC-1806, JIMT-1 and T47D cells, c-Myc expression and RNA pol II phosphorylation were not consistently repressed in response to RNMT siRNA transfection (electronic supplementary material, figure S8b). Deep proteomic analyses of a large panel of breast carcinoma cell lines will be useful in identifying the full spectrum of RNMT-dependent proteins in PIK3CA-mutant breast cancer.
In this study we provide evidence to suggest RNMT should be explored as a therapeutic target in breast cancers harbouring PI3K p110a mutations. The development of specific tool compounds, which inhibit RNMT function and the elucidation of the function of RNMT in vivo, will be important next steps in these investigations.
Data accessibility. This article has no additional data.