2023 年 98 巻 4 号 p. 161-169
Paired box 6 (PAX6) is a member of the PAX family and plays an essential role in cancer cell cycle progression, colony formation, proliferation and invasion. Its expression is upregulated in many cancers including breast cancer, but the process of PAX6 mRNA translation has rarely been studied. We found that PAX6 translation level increased in MCF-7 breast cancer cells treated with the chemotherapeutic drug adriamycin (ADM), which might be attributable to internal ribosome entry site (IRES)-mediated translation. By modifying a bicistronic luciferase plasmid that is widely used to examine IRES activity, we found that the 469-base 5′-UTR of PAX6 mRNA contains an IRES element and that core IRES activity is located between nucleotides 159 and 333. Moreover, PAX6 IRES activity was induced during ADM treatment, which may be the main reason for the elevated level of PAX6 protein. We also found that cymarin, a cardiac glycoside, acts as an inhibitor of PAX6 protein expression by impairing its IRES-mediated translation. Furthermore, MCF-7 cell proliferation was suppressed during treatment with cymarin. These results provide novel insights into the translation mechanism of PAX6 in breast cancer cells and suggest that cymarin is a promising candidate for the treatment of breast cancer via targeting the expression of PAX6.
Paired box 6 (PAX6) is a member of the PAX family and encodes a transcription factor that is essential for the correct development of the eye, central nervous system and pancreas (Williamson et al., 2020). In recent years, more and more studies have shown that PAX6 plays an essential role in malignant tumors: its abnormal expression and dysfunction are closely related to cancer cell cycle progression, colony formation, proliferation and invasion (Lang et al., 2007; Blake and Ziman, 2014). PAX6 expression is upregulated in many cancers, including non-small cell lung cancer (Wu et al., 2019), pancreatic cancer (Ahmad et al., 2015) and breast cancer (Xia et al., 2015). It has been shown that PAX6 can act in an oncogenic manner in breast cancer, and can regulate breast cancer cell behavior, including cell proliferation, apoptosis and autophagy (Zhang et al., 2020). However, the mechanisms behind the regulation of PAX6 expression at the translational level have not been fully revealed.
Protein expression is regulated at multiple levels, notably at the step of translation initiation (Kozak, 1992). There are two basic mechanisms of initiating translation, namely cap-dependent and cap-independent translation. Normal physiological conditions favor cap-dependent translation (Shatsky et al., 2010). The protein complex eIF4F (comprised of cap-binding protein eIF4E, RNA helicase eIF4A and scaffolding protein eIF4G) interacts initially with the 5′ leader of the mRNA and, via the interaction of eIF3 with eIF4G, recruits the 40S ribosomal subunit (Pestova and Kolupaeva, 2002). An alternative and less understood mechanism is mediated by cap-independent activation via an internal ribosome entry site (IRES). The IRES element can recruit the ribosome directly without recognition of the 5′ cap structure and initiate translation (Komar and Hatzoglou, 2011). Recruitment requires the assistance of IRES trans-acting factors (ITAFs) and is dependent on the specific active conformations of the IRES (Martinez-Salas et al., 2013). Poliovirus and encephalomyocarditis virus were the first biological systems found to utilize internal ribosome entry mechanisms to initiate translation of their polypeptides, and IRESs were also identified in a set of cellular mRNAs, such as Aurora A kinase, the anti-apoptotic factor BCL-2, the RNA editing enzyme ADAR1 and P53 (Sherrill et al., 2004; Ray et al., 2006; Dobson et al., 2013; Yang et al., 2015). Like viral IRES-containing mRNAs, cellular mRNAs containing IRES elements are preferentially translated under conditions of inhibition of cap-dependent initiation, such as endoplasmic reticulum stress, hypoxia, mitosis and cellular differentiation (Spriggs et al., 2010).
In this study, we demonstrated that the expression of PAX6 in the breast cancer cell line MCF-7 is controlled at the level of translation initiation. There is a strong IRES located in the 5′-untranslated region (UTR) of PAX6 mRNA, and we analyzed the influence of the chemotherapeutic drug adriamycin (ADM) on translational regulation of the PAX6 IRES in MCF-7 cells. Furthermore, we found that cymarin can reduce PAX6 protein expression and IRES activity, and can also inhibit MCF-7 cell proliferation. Our data provide a framework for understanding how PAX6 mRNA is translated via an IRES, which causes abnormal expression in cancer cells.
Previous studies have demonstrated that PAX6 plays a crucial role in cancer cell proliferation and apoptosis. To investigate whether the expression of PAX6 was affected during cellular stress, we measured the endogenous protein and mRNA levels of PAX6 in human breast cancer MCF-7 cells treated with ADM, which is applied as a chemotherapy drug in breast cancer. PAX6 protein expression under 0.2 and 0.4 μg/ml ADM was two-fold higher than that in cells not treated with ADM (Fig. 1A). To understand whether ADM-induced PAX6 expression was regulated at the transcriptional or translational level, we performed qPCR to examine PAX6 mRNA level in MCF-7 cells that were treated with ADM. The result (Fig. 1B) showed that ADM had no significant effect on PAX6 mRNA expression. Therefore, we concluded that ADM-induced PAX6 expression is regulated at the translational level.
By searching the NCBI database, we found that the entire 5′-UTR of PAX6 is long and GC-rich (469 bases; NM_001127612.3), consistent with the RNA structural features of an IRES. To determine whether the PAX6 5′-UTR has an IRES, we performed transient transfection assays on HEK293T and MCF-7 cells using constructs based on the classical bicistronic vector pRF (Fig. 2A). The upstream cistron, the Renilla luciferase (RL) gene, will be translated by normal cap-dependent translation, whereas the second cistron, the firefly luciferase (FL) gene, will be translated only if there is an IRES element inserted between the two cistrons. The ratio of FL to RL activity is a measurement of the IRES activity of any sequence inserted between the two cistrons. The 5′-UTR of EMCV mRNA, which is known to harbor a strong IRES element, was utilized as a positive control, while unmodified pRF was used as a negative control. As indicated in Fig. 2B, the construct containing the PAX6 5′-UTR was active in the HEK293 and MCF-7 cell lines, suggesting that there is an IRES element in PAX6 mRNA.
To exclude the possibility that FL expression was induced by a cryptic promoter in the 5′-UTR of PAX6, the cytomegalovirus (CMV) promoter-deleted vector pBR-PAX6-F was constructed (Fig. 3A) and transfected into HEK293T cells and the RL and FL activity was analyzed. As shown in Fig. 3B, the FL activity measured in pBR-PAX6-F-transfected cells is comparable to that measured in cells transfected with the pBRF plasmid, and is much lower than the activity produced by pR-PAX6-F. This result clearly illustrated that the PAX6 5′-UTR is devoid of cryptic promoter activity.
As mRNA secondary structure is critical in IRES-dependent translation, the secondary structure of the PAX6 5′-UTR, which was found to contain several stem–loop elements, was estimated by Mfold (Zuker, 2003) under the following conditions: 37 ℃, 1 M NaCl, and no divalent ions (Fig. 4A). To further determine the core IRES region within the PAX6 5′-UTR, we deleted the stem–loop elements progressively and then inserted a series of deleted fragments into the bicistronic vector pRF (Fig. 4B). HEK293T cells were transfected with each of these plasmids. Upon assay of their luciferase activities, we discovered that the 1–158 and 159–333 sequences showed higher IRES activity than the full-length 5′-UTR, while the 334–469 sequence showed much lower IRES activity (Fig. 4C). Notably, the sequence between 159 and 333 exhibited the highest IRES activity (about 5-fold higher than the full-length 5′-UTR), indicating that this region, which contains two stem–loops, is particularly important for PAX6 IRES activity.
Substantial evidence indicates that IRES-dependent protein synthesis is often activated under cellular stress conditions (Subkhankulova et al., 2001; Young et al., 2008; Wang et al., 2009). To determine whether the elevated IRES activity increases PAX6 protein synthesis during drug stress, we measured the reporter activity in MCF-7 cells treated with ADM using constructs containing the PAX6 IRES. We observed that the IRES activity increased gradually in cells treated with ADM from 0.05 to 0.4 μg/ml (Fig. 5), which was similar to the enhancement of PAX6 protein expression level (Fig. 1A). These results show a positive correlation between PAX6 translational efficiency and PAX6 IRES activity. Therefore, the IRES activity of the PAX6 5′-UTR was the main cause of the rise of PAX6 protein expression during stress conditions.
A previous study of viral IRES elements used high-throughput screening to identify compounds that regulate viral IRES-dependent translation (Novac et al., 2004). Recently, a set of ~14,000 natural compounds were tested with a bicistronic reporter gene system to identify positive or negative modulators of cellular IRES-directed translation. A cardiac glycoside, cymarin, showed high potency and low cytotoxicity to regulate cellular IRES-mediated translation (Didiot et al., 2013). Therefore, we investigated whether cymarin had an effect on PAX6 IRES-mediated translation. To accomplish this, a dual-luciferase assay was performed in MCF-7 cells that had been transfected with pR-PAX6-F and then treated with dimethyl sulfoxide (DMSO) or an increasing concentration of cymarin. In these cells, RL was expressed under the control of the cap and FL was expressed under the control of the cellular PAX6 IRES element. The results showed that cymarin inhibited PAX6 IRES-dependent protein synthesis in a dose-dependent fashion in MCF-7 cells (Fig. 6A), with a more than four-fold reduction of PAX6 IRES-mediated translation relative to cap-dependent translation at the concentration of 1 μg/ml. To confirm that cymarin has an effect on endogenous PAX6 expression, PAX6 protein and mRNA level were measured in MCF-7 cells. The cells were treated with DMSO or cymarin from 0.25 to 1.0 μg/ml for 24 h. The results revealed that PAX6 protein expression decreased as cymarin concentration increased (Fig. 6B), whereas the PAX6 mRNA level remained unchanged (Fig. 6C). These results indicate that cymarin preferentially inhibits PAX6 expression by blocking IRES-directed translation in MCF-7 cells.
PAX6 protein expression level has been shown to have the potential to regulate cell proliferation in different types of cancer, including breast cancer. It has been found that a decrease of PAX6 protein expression inhibits MCF-7 cell proliferation (Zong et al., 2011). To confirm the influence of PAX6 protein expression on MCF-7 cell proliferation, we constructed the PAX6 expression vector and transfected MCF-7 cells with it, and then performed a CCK8 assay to examine cell proliferation. The results showed that PAX6 protein level increased in MCF-7 cells after the transfection (Fig. 7A), and the overexpression of PAX6 protein could increase MCF-7 cell proliferation compared with the control cells (Fig. 7B). Because cymarin reduced PAX6 protein expression in MCF-7 cells, to investigate whether cymarin can influence MCF-7 cell growth, an MTT cell proliferation assay was performed on MCF-7 cells for five successive days. The results showed that when treated with different concentrations of cymarin, MCF-7 cell proliferation was inhibited to different degrees (Fig. 7C). The proliferation decreased by 47.8% in MCF-7 cells at the concentration of 1 μg/ml cymarin.
PAX6 is a member of the PAX gene family and has been demonstrated to play a crucial role in various cancers; it is associated with tissue proliferation and is expressed abnormally in many cancers, including breast cancer (Muratovska et al., 2003). Previous research suggested that PAX6 acts as a promoter in breast cancer in vitro and in vivo, since decreased PAX6 protein expression resulted in remarkable suppression of cell viability, DNA synthesis and colony formation in vitro and of tumorigenesis in vivo (Zong et al., 2011). Moreover, Liu et al. (2022) showed that reducing TNFRSF9 expression could upregulate PAX6 protein level in MCF-7 cells and increase MCF-7 cell proliferation, and in this paper we also showed that overexpression of PAX6 protein could promote MCF-7 cell proliferation. These results indicated that the expression level of PAX6 protein is important in cellular proliferation, cell cycle progression, colony formation and cellular invasion in breast cancer, although relatively little was known of the mechanism of its translation.
Translational control provides cells with the plasticity and flexibility to respond to rapid changes in the environment, and we found that the translational level of PAX6 was enhanced, allowing its increased expression in a breast cancer cell line treated with ADM for 24 h. Several cellular mRNAs have been identified as containing IRES elements that allow them to be translated at times of cellular stress, such as chemotoxic stress (Yang et al., 2010; Dai et al., 2015). In this paper, we cloned the PAX6 5′-UTR into a bicistronic luciferase reporter and found that the PAX6 5′-UTR exhibited efficient IRES activity, which was not due to a cryptic promoter. Moreover, PAX6 IRES activity was induced by ADM in the MCF-7 cell line, indicating that PAX6 expression is regulated through an IRES-dependent translational mechanism in response to chemotoxic stress.
The activity of eukaryotic IRESs depends not only on nucleotide sequence but also on RNA secondary structure, and several studies have demonstrated that structural diversity can influence the activity of IRESs in response to different cellular stress (Leppek et al., 2018). Therefore, it is essential to characterize their RNA structure. To identify the core region of the PAX6 IRES element, we performed deletion experiments according to the predicted secondary structure of the PAX6 5′-UTR, and found that the region of 159–333 nucleotides from the 5′ terminus, which contains two stem–loops, was essential for IRES activity, while the 334–469 region showed much lower IRES activity, suggesting that it acts as an inhibitor of PAX6 IRES activity. As ITAFs can function as either positive or negative regulators of IRES-mediated translation (Yang and Wang, 2019), the different regions performing different IRES activity may be associated with secondary structure that can influence the combination of the trans-acting factors, although further study is needed to test this hypothesis.
Numerous reports have confirmed that cardiac glycosides have anti-proliferation and apoptotic effects in several cancer cells, including breast cancer cells (Newman et al., 2008; Diederich, 2017). In this report, we found that the cardiac glycoside cymarin could inhibit MCF-7 cell proliferation in a dose-dependent manner. Furthermore, we found that cymarin can act as an inhibitor of PAX6 IRES-mediated translation. Cymarin-treated MCF-7 cells that expressed the bicistronic reporter genes were used to validate the compound’s effect on the PAX6 IRES, which showed that cymarin could inhibit PAX6 IRES-mediated translation in a concentration-dependent manner. In addition, treatment of MCF-7 cells with cymarin induced a clear decrease of PAX6 protein, but had no influence on its mRNA level. These findings suggested that inhibition of IRES-mediated translation initiation is a strategy to inhibit abnormal expression of PAX6 protein in tumor cells, and cymarin-induced PAX6 protein reduction may be the reason for the inhibition of MCF-7 cell proliferation.
Taking these results together, we provide evidence for a novel mechanism of IRES-dependent PAX6 translation. This IRES-mediated translation mechanism may be important for regulating PAX6 expression, especially in the adaptation response of cancer cells to drug stress. Moreover, cymarin showed a specific inhibition of PAX6 IRES-mediated translation and MCF-7 cell proliferation. Future studies on the role of PAX6 IRES-interacting proteins need to be conducted, which would address how the ribosome is recruited to the PAX6 IRES and could provide new insights into the targeting of PAX6 in tumor cells.
All cell lines were purchased originally from ATCC and cultured in Dulbecco’s modified Eagle’s medium (Thermo Fisher Scientific) supplemented with 10% fetal bovine serum. Cells were kept at 37 ℃ in a humidified atmosphere with 5% CO2. For drug treatment, MCF-7 cells were cultured with or without increasing concentrations (0.05 to 0.4 μg/ml) of ADM (Thermo Fisher Scientific) or cymarin (Sigma-Aldrich) for 24 h.
Plasmid constructionThe open reading frame of human PAX6 cDNA (NM_000280) was amplified and cloned into the vector GV657 (GeneChem) and the empty vector was used as a control. The bicistronic plasmid pRF (encoding Renilla and firefly luciferases) and pR-EMCV-F (positive control) were purchased from Geneseed Biotech. The PAX6 5′-UTR sequence (NM_001127612.3) was synthesized by Sangon Biotech and inserted into pRF between the Renilla and firefly luciferase coding sequences. The CMV promoter was deleted from pRF to yield pBRF. pBR-PAX6-F was constructed by inserting the PAX6 5′-UTR into pBRF. To detect the core regions for IRES activity, PAX6 5′-UTR deletions of different lengths and flanked by restriction sites were synthesized by Sangon Biotech. The deletion products were digested with BamHI and EcoRI, and then inserted into pRF. PCR was performed using PrimeSTAR Max DNA polymerase (Takara) following the manufacturer’s recommendations, with a denaturation step at 98 ℃ for 3 min followed by 32 cycles of 98 ℃ for 10 s, 68 ℃ for 20 s and 72 ℃ for 1 min, and then a 5-min extension at 72 ℃. The primer sequences used are listed in Table 1.
| Name | Sequences of oligonucleotide primers |
|---|---|
| pR-PAX6-F (1–158) | Forward: GGATCCACCCTCTTTTCTTATCATTGACATTTAA |
| Reverse: GAATTCGATTCCTGGGAGCGGAG | |
| pR-PAX6-F (159–333) | Forward: GGATCCTGAGAATTGCTCTCACACACCA |
| Reverse: GAATTCGGCATCCTTTCTGGTTGTCA | |
| pR-PAX6-F (334–469) | Forward: GGATCCTCATAAAGGGGGAAGACTTTAACTAG |
| Reverse: GAATTCGGCATCCTTTCTGGTTGTCACA |
Cells were transfected using Lipofectamine 2000 (Thermo Fisher Scientific) as specified by the manufacturer. Briefly, cells were seeded in 24-well plates 24 h before transfection with 1 μg of bicistronic constructs. After 24 h, lysates were prepared from transfected cells using 1 × passive lysis buffer (Promega). Firefly and Renilla luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer’s instructions. Light emission was measured over 10 s using an Optocomp I luminometer. These experiments were repeated more than three times.
Western blottingCells were washed with phosphate-buffered saline, lysed in 150 μl RIPA buffer (Beyotime Institute of Biotechnology) for 10 min at 4 ℃ and centrifuged at 12,000 × g for 10 min. The obtained proteins were separated by 10% SDS-PAGE and transferred to polyvinylidene difluoride membranes. The membranes were incubated with 5% milk in a solution of Tris-buffered saline containing 0.1% Tween-20 (TBST) overnight at 4 ℃. The blots were washed three times in TBST, incubated with polyclonal antibodies directed against PAX6 (1:1000; ab195045; Abcam) and β-actin (1:5000; sc-69879; Santa Cruz) for 4 h at room temperature, washed, and incubated with the respective secondary antibodies (1:1000; 7074 and 7076; Cell Signaling Technology). Proteins were visualized using an ECL kit (Abbkine) following the manufacturer’s directions and quantified by ImageJ software (National Institutes of Health).
RNA extraction and quantitative (q) PCRTotal RNA was isolated from treated cells using Trizol reagent (Thermo Fisher Scientific) and was reverse transcribed using a reverse transcription system (Takara) to generate a cDNA template according to the manufacturer’s instructions. The mRNA levels of PAX6 were analyzed using qPCR, which was performed with SYBR Premix Ex Taq (Takara) according to the manufacturer’s instructions and quantified with a Roche LightCycler 480. The PCR involved a denaturation step at 95 ℃ for 10 min, followed by 40 cycles of 95 ℃ for 10 s and 60 ℃ for 30 s. The primer sequences used for qPCR are listed in Table 2.
| Name | Sequences of oligonucleotide primers |
|---|---|
| PAX6 | Forward: ATGGGCGGAGTTATGATACCTAC |
| Reverse: GGAACTTGAACTGGAACTGACAC | |
| ACTB | Forward: GCGTGACATTAAGGAGAAGC |
| Reverse: CCACGTCACACTTCATGATGG |
Cells were transiently transfected with DNA and Lipofectamine with Plus Reagent (Thermo Fisher Scientific) according to the manufacturer’s protocol. Renilla and firefly luciferase activities were detected by the dual-luciferase reporter assay system (Promega) according to the instructions of the manufacturer.
Cell proliferation analysisMethyl thiazolyl tetrazolium (MTT, Genview) and Cell Counting Kit-8 (CCK8, Meilunbio) were used to detect cell proliferation. For MTT assay, 3,000 cells per well were seeded into 96‑well plates and maintained at 37 ℃ with different concentrations of cymarin. After 1, 2, 3, 4 and 5 days, cells were treated with MTT solution (5 mg/ml). The crystals were then dissolved in DMSO and the absorbance was recorded at 490 nm. For CCK8 assay, cells were transfected with PAX6 expression vectors or control vectors in 6-well plates for 12 h and re-seeded in 96-well plates at a density of 2,000 cells per well. After 24, 48 and 72 h, CCK8 medium was added to the wells and the plates were incubated at 37 ℃ for 1 h, after which the absorbance was measured at 450 nm.
Statistical analysisEach experiment was performed at least three times. All experimental data were analyzed using GraphPad Prism 8.0 (GraphPad) and are presented as mean values ± SEM. Statistical analysis was performed using a t-test (* P < 0.05; ** P < 0.005; *** P < 0.0005).
This research was supported by the Scientific Research Project of Wuxi Health Commission (Q202028).
