MiR-148a-3p Promotes Colorectal Cancer Cell Ferroptosis by Targeting SLC7A11

Simple Summary The role of miR-148a-3p in colorectal cancer (CRC) is still debated. In this study, the in vitro antineoplastic effect of miR-148a-3p overexpression in the CRC model is reported. The antitumor activity of miR-148a-3p occurs through apoptosis, mitochondrial impairment, lipid peroxidation, and ferroptosis sustained by the ACSL4/TFRC/Ferritin axis. Bioinformatic analysis and transfection experiments with miR-148a-3p mimics and inhibitors revealed that the cytotoxicity might be related to the downregulation of SLC7A11. These findings, extending knowledge on functional and molecular mechanisms, unveil the oncosuppressor role of miR-148a-3p, pointing out its potential as a diagnostic and therapeutic biomarker in CRC. Abstract Ferroptosis, an iron-dependent form of cell death, and dysregulated microRNA (miRNA) expression correlate with colorectal cancer (CRC) development and progression. The tumor suppressor ability of miR-148a-3p has been reported for several cancers. Nevertheless, the role of miR-148a-3p in CRC remains largely undetermined. Here, we aim at investigating the molecular mechanisms and regulatory targets of miR-148a-3p in the CRC cell death mechanism(s). To this end, miR-148a-3p expression was evaluated in SW480 and SW620 cells and normal colon epithelial CCD 841 CoN cells with quantitative real-time polymerase chain reaction (qRT-PCR). Data reported a reduction of miR-148a-3p expression in SW480 and SW620 cells compared to non-tumor cells (p < 0.05). Overexpression of miR-148a selectively inhibited CRC cell viability (p < 0.001), while weakly affecting normal CCD 841 CoN cell survival (p < 0.05). At the cellular level, miR-148a-3p mimics promoted apoptotic cell death via caspase-3 activation (p < 0.001), accumulation of mitochondrial reactive oxygen species (ROS) (p < 0.001), and membrane depolarization (p < 0.001). Moreover, miR-148a-3p overexpression induced lipid peroxidation (p < 0.01), GPX4 downregulation (p < 0.01), and ferroptosis (p < 0.01), as revealed by intracellular and mitochondrial iron accumulation and ACSL4/TFRC/Ferritin modulation. In addition, levels of SLC7A11 mRNA and protein, the cellular targets of miR-148a-3p predicted by bioinformatic tools, were suppressed by miR-148a-3p’s overexpression. On the contrary, the downregulation of miR-148a-3p boosted SLC7A11 gene expression and suppressed ferroptosis. Together, these in vitro findings reveal that miR-148a-3p can function as a tumor suppressor in CRC by targeting SLC7A11 and activating ferroptosis, opening new perspectives for the rationale of therapeutic strategies through targeting the miR-148a-3p/SLC7A11 pathway.


Introduction
The incidence of colorectal cancer (CRC) in 2020 still represented 10% of the total cancer cases and was the second cause of tumor-dependent mortality in the world, with 3.2 million new cases expected by 2040 [1,2].Despite the rapid development of early screening methods and improved therapies, the prognosis for CRC is still not satisfactory.Among the different tools suitable for tumor screening, microRNAs (miRNAs) represent crucial biomarkers in CRC patients [3].
Ferroptosis is a novel form of programmed cell death featuring iron overload, which, via depletion of reduced glutathione (GSH) and/or the inhibition of glutathione peroxidase 4 (GPX4), enhances intracellular reactive oxygen species (ROS) and lipid peroxidation, causing membrane impairment and severe cell damage [24,25].Inactivation of GPX4, responsible for biomembrane preservation from peroxidation damage, leads to an accumulation of lipid peroxide species [26,27].To date, ferroptosis has emerged as a considerable strategy in the treatment of many cancers, although the malignancies susceptible to ferroptosis have not been clearly defined [28].However, recent studies reveal an association between ferroptosis and CRC carcinogenesis [29,30].The induction of ferroptosis by intracellular Fe 2+ and ROS accumulation, along with reduced GPX4 and GSH content in CRC cells, contribute to the clinical treatment of this neoplasia [31,32].On the contrary, inhibition of ferroptosis may lead to CRC progression and drug resistance [33].More recently, the induction of ferroptosis has been described as a consequence of the inhibition of solute carrier family seven-member 11 (SLC7A11), a multichannel transmembrane protein [34][35][36], resulting in a potential antineoplastic therapeutic approach in different tumors.Of interest, SLC7A11 expression is positively associated with microsatellite instability, a phenomenon of molecular fingerprinting related to CRC prognosis [37,38].MiR-148a-3p directly targeted SLC7A11 and inhibited its expression.Based on these results, the miR-148a-3p/SLC7A11 axis might be a novel therapeutic target for colon cancer.Although miR-148a has been reported to be associated with a good prognosis for patients with CRC and its expression is associated with tumor metastasis, the specific mechanism of action remains unclear.In this study, we aimed at investigating the biological functions of miR-148a-3p in SW480 and SW620 CRC cells and the possible role of the miR-148a-3p/SLC7A11 axis in the definition of a novel potential molecular candidate for CRC therapies.

Viability and Cytotoxicity
Cell viability and cytotoxicity were evaluated by cell counting kit-8 and Cytotoxicity LDH assays (CK04 and CK12, both from Donjindo Molecular Technologies, Inc., Rockville, MD, USA), respectively, following the manufacturer's instructions.Absorbances (450 and 490 nm) were detected by a microplate reader (model 680, Bio-Rad, Hercules, CA, USA).Experiments were conducted with n = 4 replicates, and results were reported as %.

Apoptotic Cell Death
The analysis of the apoptotic mechanism was performed by the FITC annexin V apoptosis detection kit (556547, BD Pharmigen, Franklin Lakes, NJ, USA), while the caspase-3/7 detection was assessed by the Nucview 488 caspase-3 substrate and annexin V kit (30067, Biotium, Fremont, CA, USA), as already described [39,40].For both assays, fluorescence intensities were recorded by a BD Accuri C6 cytometer (BD Biosciences, San José, CA, USA), collecting at least 20,000 events for each sample, and the data were analyzed by FlowJo V10 software (Williamson Way, Ashland, OR, USA).

Mitochondrial State
Mitochondrial integrity and ROS levels were assessed by staining cells for 20 min with 5 µM MitoTracker Green FM (M7514, Invitrogen, Waltham, MA, USA) and MitoSOX Red Mitochondrial Superoxide Indicator (M36008, Invitrogen, Waltham, MA, USA) fluorescent probes.The JC-1 stain (MT09, Donjindo Molecular Technologies, Inc., Rockville, MD, USA) was used to measure mitochondrial potential membranes, as previously reported [41].The detection of ferrous ions (Fe 2+ ) in mitochondria was carried out with the Mito-FerroGreen (M489; Donjindo Molecular Technologies, Inc., Rockville, MD, USA) fluorescent probe, following the supplier's instructions.Briefly, after transfection, cells were stained with a 5 µM working solution and incubated for 20 min at 37 • C.After each staining procedure, fluorescent images were collected by an EVOS M5000 microscope (Thermo Scientific, Rockford, IL, USA), while fluorescent signals were recorded by a BD Accuri C6 cytometer (BD Biosciences, San José, CA, USA).Data analysis was performed by FlowJo V10 software (Williamson Way, Ashland, OR, USA).

Ferroptotic Mechanism
A cell ferrous iron colorimetric assay (E-BC-K881-M, Elabscience Biotechnology Inc., Houston, TX, USA) was performed to measure the intracellular Fe 2+ content, following the manufacturer's instructions.After transfection, 1 × 10 6 cells were centrifuged for 10 min at 1500× g at 4 • C and then incubated with 80 µL of reagent for 10 min at 37 • C. The 593 nm absorbance was measured with the microplate reader (model 680, Bio-Rad, Hercules, CA, USA), and the Fe 2+ concentration was derived from the cell ferrous iron standard curve.Ferroptosis was investigated by using the fluorescent probe FerroOrange (F374, Dojindo Molecular Technologies, Inc., Rockville, MD, USA), according to the supplier's guidance.Cells were incubated for 20 min at 37 • C with 1 µM FerroOrange working solution before fluorescent image acquisition by an EVOS M5000 microscope (Thermo Scientific, Rockford, IL, USA).Fluorescence intensity was recorded by a BD Accuri C6 cytometer (BD Biosciences, San José, CA, USA), and analysis was carried out by FlowJo V10 software (WilliamsonWay, Ashland, OR, USA).

Statistical Analysis
GraphPad Prism software version 9.1.2(GraphPad Software Inc., La Jolla, CA, USA) was used to analyze the data.Results were presented as mean ± standard deviation (SD).Using test t or ANOVA and Tukey's, the significance was examined and defined as statistically significant as a p < 0.05.

Statistical Analysis
GraphPad Prism software version 9.1.2(GraphPad Software Inc., La Jolla, CA, USA) was used to analyze the data.Results were presented as mean ± standard deviation (SD).Using test t or ANOVA and Tukey's, the significance was examined and defined as statistically significant as a p < 0.05.

MiR-148a + Triggered Mitochondrial Damage in CRC
Mitochondrial impairment was analyzed using different approaches.Firstly, double staining with MitoTracker and MitoSOX fluorescent probes was used to evaluate mitochondrial integrity and the occurrence of mitochondrial oxidative stress, respectively (Figures 3 and S3).Results revealed that miR-148a + promoted mitochondrial ROS accumulation and decreased mitochondrial integrity (p < 0.001 vs. miR-NC) (Figures 3 and S3).Fluorecence and cytometric analysis of mitochondrial membrane potential showed the miR-148a + capacity to provoke an impressive membrane depolarization rate (p < 0.001 vs. miR-NC) (Figures 3 and S3).Results revealed that miR-148a + promoted mitochondrial ROS accumulation and decreased mitochondrial integrity (p < 0.001 vs. miR-NC) (Figures 3 and S3).Fluorecence and cytometric analysis of mitochondrial membrane potential showed the miR-148a + capacity to provoke an impressive membrane depolarization rate (p < 0.001 vs. miR-NC) (Figures 3 and S3).

MiR-148a + Promoted Lipid Peroxidation in CRC
The accumulation of oxidative damage to mitochondria led to the exploration of the occurrence of lipid peroxidation (Figures 4 and S4), closely involved in mitochondrial detriment.

MiR-148a + Promoted Lipid Peroxidation in CRC
The accumulation of oxidative damage to mitochondria led to the exploration of the occurrence of lipid peroxidation (Figures 4 and S4), closely involved in mitochondrial detriment.

MiR-148a + Provoked Ferroptosis in CRC
Mitochondrial oxidative imbalance and lipid peroxidation are critical events in ferroptosis.Therefore, iron-dependent cell death was investigated (Figures 5 and S5).

MiR-148a + Provoked Ferroptosis in CRC
Mitochondrial oxidative imbalance and lipid peroxidation are critical events in ferroptosis.Therefore, iron-dependent cell death was investigated (Figures 5 and S5).
To further confirm SLC7A11 as a miR-148a target, CRC cells were transfected with a-miR-148a, and then SLC7A11 expression was evaluated (Figures 6 and S6).Results showed a marked upregulation of SLC7A11 mRNA and protein levels in a-miR-148atransfected cells (p < 0.01 vs. a-miR-NC) (Figures 6C,D,K,L and S6), validating the miR-148a capacity to modulate SLC7A11 expression in both CRC lines.Further functional experiments were performed with a-miR-148a to corroborate the miR-148a-mediated ferroptosis as a cell death mechanism and its oncosuppressor role in the in vitro CRC model.To further confirm SLC7A11 as a miR-148a target, CRC cells were transfected with a-miR-148a, and then SLC7A11 expression was evaluated (Figures 6 and S6).Results showed a marked upregulation of SLC7A11 mRNA and protein levels in a-miR-148a-transfected cells (p < 0.01 vs. a-miR-NC) (Figures 6C,D,K,L and S6), validating the miR-148a capacity to modulate SLC7A11 expression in both CRC lines.Further functional experiments were performed with a-miR-148a to corroborate the miR-148a-mediated ferroptosis as a cell death mechanism and its oncosuppressor role in the in vitro CRC model.

Discussion
In this study, we provide the first evidence on the antiproliferative role of miR-148a-3p, showing that the overexpression of miR-148a-3p by mimic transfection exerts antitumor effects and mediates lipid peroxidation and ferroptotic cell death via the ACSL4/TFRC/Ferritin axis by targeting SLC7A11 in CRC cells.
CRC still represents one of the main causes of massive cancer-related deaths, and its invasion and metastasis significantly affect the prognosis of CRC patients [1,2].Further studies on the underlying mechanisms of CRC initiation and progression are required to reduce mortality caused by this neoplasia.Undoubtedly, abnormal expression of miRNAs correlates with CRC pathological stage and prognosis [43][44][45][46].
MiR-148a-3p has been described as participating in various biological processes in human cancer.Several reports indicated the miR-148a-3p ability to repress proliferation and invasion in esophageal and bladder cancer, inhibit progression of epithelial ovarian cancer through directly inhibiting the expression of c-Met, as well as predict patient drug response and inhibit breast cancer progression [47][48][49][50][51].The role of miR-148a-3p in CRC is still debated since contrasting data reported that both elevated and downregulated miR-148a-3p levels were detected in tissues from patients with advanced colorectal adenoma [20,52].Here, we found decreased miR-148a-3p levels in SW480 and SW620 CRC cells compared to their non-tumor counterpart, consistent with previous data reporting miR-148a-3p expression was significantly downregulated in colon adenocarcinoma, thus contributing to elucidating its suppression as a crucial factor in tumor progression and poor survival in CRC [53,54].We report that miR-148a-3p overexpression induced selective CRC cytotoxicity, triggering caspase-3-dependent apoptotic cell death via Bcl-2 downregulation, in line with evidence showing that miR-148a-3p overexpression reduced the progression of colon adenocarcinoma [55].

Discussion
In this study, we provide the first evidence on the antiproliferative role of miR-148a-3p, showing that the overexpression of miR-148a-3p by mimic transfection exerts antitumor effects and mediates lipid peroxidation and ferroptotic cell death via the ACSL4/TFRC/Ferritin axis by targeting SLC7A11 in CRC cells.
CRC still represents one of the main causes of massive cancer-related deaths, and its invasion and metastasis significantly affect the prognosis of CRC patients [1,2].Further studies on the underlying mechanisms of CRC initiation and progression are required to reduce mortality caused by this neoplasia.Undoubtedly, abnormal expression of miRNAs correlates with CRC pathological stage and prognosis [43][44][45][46].
MiR-148a-3p has been described as participating in various biological processes in human cancer.Several reports indicated the miR-148a-3p ability to repress proliferation and invasion in esophageal and bladder cancer, inhibit progression of epithelial ovarian cancer through directly inhibiting the expression of c-Met, as well as predict patient drug response and inhibit breast cancer progression [47][48][49][50][51].The role of miR-148a-3p in CRC is still debated since contrasting data reported that both elevated and downregulated miR-148a-3p levels were detected in tissues from patients with advanced colorectal adenoma [20,52].Here, we found decreased miR-148a-3p levels in SW480 and SW620 CRC cells compared to their non-tumor counterpart, consistent with previous data reporting miR-148a-3p expression was significantly downregulated in colon adenocarcinoma, thus contributing to elucidating its suppression as a crucial factor in tumor progression and poor survival in CRC [53,54].We report that miR-148a-3p overexpression induced selective CRC cytotoxicity, triggering caspase-3-dependent apoptotic cell death via Bcl-2 downregulation, in line with evidence showing that miR-148a-3p overexpression reduced the progression of colon adenocarcinoma [55].
The search for targeted therapy in CRC aimed at interfering with critical cell mechanisms such as cell growth and proliferation, angiogenesis, migration, and differentiation focuses on miRNA's ability to penetrate cells and inhibit target pathway(s), preventing cancer growth and causing apoptosis [43][44][45][46].The capacity of miR-148a-3p to induce mitochondrial injury and aberrant ROS production has been reported in different cell models [56][57][58].The results of this study provided the first evidence that overexpression of miR-148a-3p caused mitochondrial dysfunction along with oxidative stress, resulting in mitochondrial impairment and membrane depolarization in SW480 and SW620 cells.Mechanistically, miR-148a-3p mimic induces lipid peroxidation and GPX4 decreases, resulting in ferroptosis death via the ACSL4/TFRC/Ferritin signaling pathway.MiR-148a-3p overexpression increased MDA and mitochondrial and intracellular Fe 2+ ion accumulation in CRC cell lines, thereby promoting ferroptotic cell death.The effects of ferroptosis accompanied by elevated ROS, intracellular ferrous iron, MDA, and GSH levels, as well as the involvement of the GPX4/TFRC/Ferritin axis mediated by miRNAs, were already reported as antiproliferative strategies in CRC [59,60].Interestingly, inhibition of miR-148a-3p by agomir transfection led to a reduction in MDA and lipid peroxidation and intracellular and mitochondrial iron accumulation, resulting in a lack of miR-148a-3p-mediated ferroptosis.Results also clarified that miR-148a-3p directly targets SLC7A11 and represses its expression in vitro, promoting ferroptosis via SLC7A11 in CRC.
SLC7A11 mediates the Cys/Glu antiporter activity in the xc-system, a non-sodiumdependent transporter protein complex that outputs intracellular Glu and inputs extracellular Cys, a precursor for GSH production [61].Several reports have elucidated that the p53-SLC7A11 axis, along with the accrual of coenzyme ubiquinone produced by the mevalonate pathway, may be linked to the ferroptosis mechanism [62].Similarly, SLC7A11 inhibition-derived ferroptosis has been described as an escape mechanism for tumor growth in CRC [36,63,64].However, none of these studies explored the miR-148a-3p/ferroptosis association as an antineoplastic tool in CRC.We reported that miR-148a-3p overexpression induced mitochondrial iron accumulation, lipid peroxidation, and ferroptosis via SLC7A11 modulation, sustaining the oncosuppressor action of miR-148a-3p in CRC.
Of interest, the levels of miR-148a-3p, the most abundant exosomal miRNA in human and bovine milk, are influenced by diet [65,66], and elevated levels of exosomederived miR-148a-3p have also been characterized in milk from Mediterranean buffalo, Bubalus bubalis, compared to commercial cow milk [8].Different population studies evaluated the role of the milk-abundant miR-148a-3p in breastfed children's immunity and described the relationship between miR-148a-3p levels and food supplementation [67,68].Recently, a randomized controlled trial assessed in breast milk the relationship among a diet integrated with Limosilactobacillus reuteri and omega-3 polyunsaturated fatty acids, the levels of miRNAs related to the immune system, and the frequency of infant regulatory T cells (Treg) [67].Results correlated colostrum miR-148a-3p expression with activated Treg cells at 24 months in breast-fed infants, elucidating the role of miR-148a-3p in infant immune function [67].Additionally, a previous trial evaluating probiotic supplementation for mothers in the perinatal period demonstrated a 40% relative risk reduction in the incidence of atopic dermatitis (AD) in infancy at 2 years of age [68].Human breast milk samples contained some highly expressed miRNAs, including miR-148a-3p, thus displaying a pivotal role in the prevention of AD in infancy, enhanced by probiotic ingestion [68].
The relationship between modifiable lifestyles, including diet, and the onset and prevention of CRC is constantly increasing [69].Understanding the mechanistic relationship between miR-148a-3p nutritional modulation and susceptibility to CRC development could be important for both cumulative risk reduction and the design and implementation of future public health programs and behavioral interventions.
The present work sheds light on the mechanism behind miR-148a-3p functioning in CRC and unveils its value as a novel therapeutic target for CRC treatment.Previous studies have shown the potential of circulating miR-148a-3p as a noninvasive biomarker for tumor diagnosis and monitoring.A study performed on 137 CRC patients and 145 healthy subjects showed the important dysregulation of serum miR-148a-3p in CRC patients and indicated miR-148a-3p diagnostic ability with high sensitivity and specificity for CRC detection [70].Here, we showed that the overexpression of miR-148a-3p, by promoting ferroptosis, could represent a considerable molecular target for CRC and suggested the importance of including miR-148a-3p in a combined miR-panel for effective discrimination of CRC from non-cancerous subjects.However, other roles of miR-148a-3p in the development and prognosis of CRC warrant further investigation.
This study has several limitations to take into account.First, the reported findings only relate to the in vitro culture model, which cannot adequately represent the complexity of factors affecting in vivo CRC homeostasis.Therefore, preclinical tumor models are imperative to confirm our results.Second, the genetic profile proper of each cell line mimicking the heterogeneity and complexity of in vivo CRC, specifically SW480 as primary colon adenocarcinoma (Dukes' type B) and SW620 as lymph node metastasis from SW480's primary tumor site (Dukes' type C), could explain a selective responsiveness to miR-148a-3p action.Indeed, differences in the tumor microenvironment due to the different cell models might translate into distinct susceptibilities, leading to non-translating clinical results.Third, the search and characterization of novel targets will need to be improved to better elucidate the role of miR-148a-3p in the progression and treatment of CRC.Lastly, the encapsulation of miR-148a-3p in effective and safe nanostructured carriers will allow site-specific effects able to impair tumor cells without influencing the healthy counterpart, leading to the development of personalized medicine, a promising strategy in CRC diagnosis and treatments [71].

Conclusions
Altogether, this study sheds light on the role of miR-148a-3p in CRC development and progression.Herein, we describe the tumor suppressive ability of miR-148a-3p in CRC SW480 and SW620 cells by inducing mitochondrial stress, lipid peroxidation, mitochondrial and intracellular iron accumulation, and ferroptotic cell death by SLC7A11 modulation.Overall, our study provides novel insights into the molecular pathogenesis of CRC, showing the potential of miR-148a-3p in the setting of novel strategies for the diagnosis and therapy of CRC.
MiR-148a + Triggered Mitochondrial Damage in CRC Mitochondrial impairment was analyzed using different approaches.Firstly, double staining with MitoTracker and MitoSOX fluorescent probes was used to evaluate mitochondrial integrity and the occurrence of mitochondrial oxidative stress, respectively (Figures 3 and S3).

Figure 3 .
Figure 3. MiR-148a + -related mitochondrial impairment.Representative fluorescent images and cytofluorimetric evaluation of (A,B,E,F) mitochondrial integrity (green) and ROS content (red) and (C,D,G,H) mitochondrial membrane potential in SW480 and SW620 cells.Scale bars = 100 µm.The data are expressed as the mean ± SD of n = 3 experiments.‡ p < 0.001 vs. miR-NC.

Figure 3 .
Figure 3. MiR-148a + -related mitochondrial impairment.Representative fluorescent images and cytofluorimetric evaluation of (A,B,E,F) mitochondrial integrity (green) and ROS content (red) and (C,D,G,H) mitochondrial membrane potential in SW480 and SW620 cells.Scale bars = 100 µm.The data are expressed as the mean ± SD of n = 3 experiments.‡ p < 0.001 vs. miR-NC.

Figure 4 .
Figure 4. MiR-148a + -derived lipid peroxidation.(A,B,F,G) Representative fluorescent images and FACS analysis of lipid peroxide levels, evaluation of (C,H) MDA and (D,I) GSH/GSSG content, and (E,J) western blotting detection of GPX4 protein levels in SW480 and SW620 cells.Scale bars = 100 µm.The data are expressed as the mean ± SD of n = 3 experiments.Western blotting results are reported as arbitrary units (AU).† p < 0.01 vs. miR-NC; ‡ p < 0.001 vs. miR-NC.

Figure 4 .
Figure 4. MiR-148a + -derived lipid peroxidation.(A,B,F,G) Representative fluorescent images and FACS analysis of lipid peroxide levels, evaluation of (C,H) MDA and (D,I) GSH/GSSG content, and (E,J) western blotting detection of GPX4 protein levels in SW480 and SW620 cells.Scale bars = 100 µm.The data are expressed as the mean ± SD of n = 3 experiments.Western blotting results are reported as arbitrary units (AU).† p < 0.01 vs. miR-NC; ‡ p < 0.001 vs. miR-NC.