Upregulation of SLC25A32 in Tumorous Tissues of Patients with Non-Metastatic Colorectal Cancer: A Pilot Study

Abstract

The SLC25A32 gene plays a pivotal regulatory role in one-carbon metabolism and redox balance by functioning as a transporter for tetrahydrofolate and flavin adenine dinucleotide (FAD). The probable contribution of SLC25A32 toward facilitating the proliferation of cancerous cells remains inadequately explored. This pilot study aims to survey the expression of SLC25A32 in non-metastatic colorectal tissues and its potential association with clinicopathological characteristics. Fresh tissue samples were collected from 30 colorectal cancer patients and their adjacent normal tissues. Total RNA was extracted, and cDNA was synthesized for qRT-PCR. In silico analysis was also performed to further investigate the functional significance of SLC25A32 in colorectal cancer by examining its expression in normal tissues, identifying protein interactions, and searching for pathogenic mutations. Our results revealed a significant up-regulation of SLC25A32 expression in tumorous tissues compared to non-tumorous tissues. The elevated expression of SLC25A32 was associated with tumor anatomic site and size. In silico analysis revealed that higher levels of SLC25A32 expression were correlated with reduced overall survival rates. Furthermore, enrichment analysis showed that this gene remarkably enriched in the process of Folate metabolism. The involvement of SLC25A32 in tumorigenesis may be due to its regulation of mitochondrial folate and FAD metabolism, which have been linked to cancer progression. Changes in SLC25A32 expression or function may contribute to cancer development, making it a potential therapeutic target for cancer treatment.

Introduction

Metabolic reprogramming refers to the phenomenon whereby neoplastic cells alter their metabolism, resulting in an elevated demand for energy to support their rapid growth and proliferation [1]. The property of neoplastic cells has been subjected to thorough investigation in the context of human malignancies. Otto Warburg's initial observations over nine decades ago provided the foremost proof supporting metabolic reprogramming. Specifically, Warburg demonstrated that neoplastic cells produce ATP via glycolysis, independent of sufficient oxygen levels or functional mitochondria [2]. This phenomenon runs counter to the oxidation of glucose in its entirety to provide energy for mitochondrial respiration, which highlights the significance of mitochondria in tumorigenesis [3]. ATP synthesis is the primary function of mitochondria. However, mitochondria play an important role in other activities such as producing reactive oxygen species (ROS) and redox metabolites, regulating cell signaling, programmed cell death, and biosynthesis of metabolites. folate-mediated 1-carbon metabolism is a complex and interconnected metabolic pathway that involves the activation and transfer of one-carbon units for a variety of cellular processes, including the biosynthesis of nucleotides, biosynthesis of amino acids, methylation of DNA, and mitochondrial protein translation [4].

Folate metabolism occurs in two parallel and complementary pathways, localized in the cytosol and mitochondria [5]. Solute carrier (SLC) proteins constitute a vast family of transmembrane transporters that play an essential role in facilitating the transport of nutrients, ions, and waste products across cellular or organelle plasma membranes. These proteins are intricately involved in active transport processes, which are crucial for maintaining optimal cellular homeostasis and supporting vital physiological functions [6, 7]. SLC25A32 constitutes one of the mitochondrial transporters belonging to the solute carrier 25 protein family. SLC25A32 transport the THF species in mitochondrial one-carbon metabolism, as evidenced by in vitro experimentation [8], genetic analysis of mice [9], and analysis of cell culture [10–12]. It has been hypothesized that SLC25A32 functions as a carrier for the coenzyme FAD [13], and when this gene undergoes mutations in humans, it results in riboflavin-responsive exercise intolerance [14, 15]. Although the dependence of cancer cells on these metabolic pathways for cell proliferation is established, the contribution of SLC25A32 to the growth of cancer cells has yet to be investigated [16]. The literature provides evidence that the amplification of the gene encoding SLC25A32 is a frequent occurrence in various human cancers such as breast, prostate, ovarian serous cystadenocarcinoma, and liver hepatocellular carcinoma. Additionally, it has been found to be linked with a decrease in patient survival [16]. The present investigation sought to evaluate the expression level of SLC25A32 in tumoral tissues of colorectal cancer (CRC) and explore its potential association with clinicopathological parameters relevant to this medical condition.

Material and Methods

The ethical approval for this research was granted by the Ethics Committee of Research Affairs at Ahvaz Jundishapur University of Medical Sciences in Ahvaz, Iran, under grant number CMRC-0004 and with a corresponding code of ethics IR.AJUMS.MEDICINE.REC.1400.003. The experimental procedures were conducted within the Department of Medical Genetics and Cellular and Molecular Research Centre of Ahvaz University of Medical Sciences.

Patient’s Recruitment

During the period from October 2021 to December 2022, thirty patients' Tissue samples were collected at Imam Khomeini Hospital and Apadana Hospital in Ahwaz, who underwent surgical treatment for colon and rectal adenocarcinoma. Informed Written Consent was obtained from all subjects before joining the project. The adjacent tissues were taken about 5 cm away from the tumor tissue [17–19]. The clinicopathologic characteristics of enrolled patients have been summarized in Table 1.

Table 1.

Clinicopathological characteristics in 30 CRC patients

Characteristic	Patients (n = 30)	Colorectal cancer
		Colon	Rectal
Gender
Male	17	12	5
Female	13	9	4
Age(years)
≤ 60	10	7	3
> 60	20	14	6
Mean age (range)	61.0 (37–84)	59.4 (42–71)	64.7 (37–84)
Tumor diameter (cm)
≥ 5.0	18	13	5
< 5.0	12	8	4
Tumor differentiation
Well	12	7	5
Moderate	14	10	4
Poor	4	4	0
T stage
pT1	2	1	1
pT2	10	7	3
pT3	13	8	5
pT4	5	5	0
N stage
N0	18	13	5
N1	8	4	4
N2	4	4	0
AJCC stage
I	10	8	2
II	8	5	3
III	12	8	4

RNA extraction

Total RNA was extracted from 50 to 100 mg of tissue using the YTzol Pure DNA/RNA/protein reagent (Cat No. YT9064, YTA, Iran) according to its manufacturer’s instructions. The quality and quantity of extracted RNA were performed using 1% agarose gel electrophoresis and Nanodrop ND- 2000c spectrophotometer (Nanodrop Technologies, Wilmington, DE, USA) respectively. The RNAs were kept at a temperature of − 80°c before use.

Complementary DNA (cDNA) Synthesis and Real-Time Polymerase Chain Reaction (qPCR)

Complementary DNA was synthesized using the cDNA Synthesis Kit (Cat No. YT4500, YTA, Iran) In accordance with the manufacturer's guidelines. Table 2 displays the SLC25A32 specific primers (Primer Bank ID: 208973265c1) [20] and GAPDH primers as reference genes [21]. Real-time polymerase chain reaction (qPCR) analysis was conducted utilizing the RealQ Plus 2 × Master Mix Green high ROX™ (Cat No. A325402, Ampliqon, Denmark) as the reagent for the study. The methodology employed involved the utilization of the A325402 kit, manufactured by Ampliqon, Denmark, and following the guidelines stipulated by the company.

Table 2.

Used primer in realtime-PCR

Gene	Primer sequence (5′–3′)	Amplicon size (bp)
SLC25A32	F: TCCCCACACCGACAATATAAAG	116
	R: CCATGCGATGTTCCAAACAGC
GAPDH	F: GTGAACCATGAGAAGTATGACAAC	123
	R: CATGAGTCCTTCCACGATACC

In silico functional analysis

Study of Gene Expression in Normal Tissues

The GTEx Portal, (https://gtexportal.org), was employed to assess gene expression in the normal tissue of CRC subjects. The Genotype-Tissue Expression (GTEx) project is an extensive, accessible resource facilitating the examination of tissue-specific gene expression and regulation across 54 non-pathological tissue sites from a cohort comprising 1000 individuals [22].

Gene Expression Analysis

To verify our observation and analyze the prognostic value of SLC25A32 in CRC, we performed some in silico studies using a well-cited cancer dataset. Colorectal cancer (TCGA, PanCancer Atlas, Cell 2018) was downloaded from the cBioPortal Cancer dataset (https://www.cbioportal.org/datasets) [23]. This included 594 patients of CRC with 534 data of mutation, 592 data of Copy Number Alteration (CNA), and 592 data of RNA seqV2. The data on the protein was not included in this package. The expression of genes was also compared relative to normal samples. Moreover, we analyzed the mRNA expression levels of the SLC25A32 gene in CRC and normal tissues in the GEPIA database (http://gepia.cancer-pku.cn) [24].

The Prognostic Values of SLC25A32

The prognostic values of the SLC25A32 were assessed utilizing the Kaplan–Meier Plotter tool, an online database including gene expression data and clinical data (http://kmplot.com). This study compared hazard ratios (HRs), 95% confidence intervals (CIs), and log-rank P-values between groups of high and low expression levels [25].

PPI Network and Functional Enrichment Analysis

A protein–protein interaction (PPI) network was constructed for overlapping functional genes using the STRING database (https://string-db.org/) [26]. To create the PPI network, only interactions scoring above 0.7 were included. Unconnected nodes were disqualified from the network. The PPI pairs list was downloaded for analysis. Subsequently, Metascape (http://metascape.org/) was employed to conduct Gene Ontology (GO) in order to identify the biological functional roles of interactor proteins [27]. Metascape is a web-based portal specifically designed to deliver a complete set of gene list annotations and analysis tools with the purpose of serving experimental biologists. Significant results were selected with a P-value of < 0.05.

Statistical Analyses

The experimental protocol was carried out in triplicate for each study. The statistical data were presented in the form of Mean ± SEM, with computation carried out using the GraphPad Prism software (Version 9.3, San Diego, CA, USA). The statistical difference between the case and control groups was assessed using the Student's T-test. A Spearman Rank Correlation test was employed to conduct a correlation analysis of the available data. The statistical significance threshold was considered with a P-value of < 0.05.

Results

Clinical Characteristics of the Patients

This pilot study examined thirty patients diagnosed with CRC, As indicated in Table 2, with 70% (n = 21) having colon cancer and 30% (n = 9) having rectal carcinoma. The study population consisted of 56.7% (n = 17) male and 43.3% (n = 13) female patients, with a median age of 61 years (range, 37–84 ± 10.56). The tumors were categorized as well differentiated (40%), moderately differentiated (46.6%), and poorly differentiated (13.4%), and there were no metastatic patients included in the study.

The SLC25A32 Gene was Upregulated in Tumorous Tissues of Colorectal Cancer

As indicated in Fig. 1A, the expression level of SLC25A32 in tumoral tissues was significantly higher than the corresponding value in adjacent normal tissues. The mean fold change was 1.527 ± 0.1581 (P_value = 0.0043). In fact, SLC25A32 expression levels in normal surrounding tissues were negative or weak.

Fig. 1.

mRNA expressions of SLC25A32 by RT-qPCR. A SLC25A32 expression analysis in CRC tissues and adjacent normal tissues (mean fold change 1.527 ± 0.1581, P_value = 0.0043). B comparison of SLC25A32 gene expression level in colon (n = 21) and rectal (n = 9) tumor tissues (Mean Fold changes: 1.317 in colon vs 2.018 in rectal, P_value = 0.0399). SLC25A32 expression in samples where the tumor size was over 5 cm vs in samples with a tumor size smaller than 5 cm (mean fold changes: 2.13 vs 1.25, P_value = 0.0253). Reported data are presented as mean ± SEM. The asterisks * and ** represent the P-values < 0.05 and < 0.01 respectively

The Expression Level of SLC25A32 was Correlated With the Tumor Anatomic Site and Tumor Size

As shown in Fig. 1B, there is a significant difference in the expression of SLC25A32 between rectal and colon samples (the mean fold changes were 2.018 in the rectum vs 1.317, in the colon) (P_value = 0.0399). This data was also confirmed through our in-silico analysis (data not shown). Additionally, there was a significant correlation between the elevated expression level of SLC25A32 and tumor size (Fig. 1C) (P_value = 0.0253). The expression levels of SLC25A32 were observed to be significantly higher in tumor samples with sizes above 5 cm as compared to those with sizes smaller than 5 cm (Mean Fold changes were 2.13 vs 1.25, respectively). No significant correlation was found between the expression level of SLC25A32 and age (P-value = 0.8285), gender (P-value = 0.6095), tumor differentiation (P-value = 0.6109), pathological tumor stage (P-value = 0.8553), pathological lymph node stage (P-value = 0.6627), and TNM stage (P-value = 0.7343).

A Variety of Genetic Alterations were Observed in Patients With Colorectal Cancers

The extracted data from cBioPortal showed that a variety of gene alterations can be found in tumoral samples of colorectal. These included upregulation of SLC25A32 in 75% (n = 442) of the samples, gene amplification in 2.1% (n = 11) of the samples, and 9 mutations were also observed in 1.7% (n = 9) of the recorded cases.

The Detected mutations were H67Y, T224I, Y243N, A231V, I228M, S15G, Y176H, A28V, and I65L. Except for S15G, all mutations were located in the “Mitochondrial carrier” domain of the SLC25A32 protein (Fig. 2A). Functional impact of these mutations mutation has been presented in Table 3.

Fig. 2.

In-silico analysis of the SLC25A32 gene in various online databases. (A) Schematic presentation of nine missense mutations of SLC25A32 in CRC. Data from cBioPortal has been obtained (www.cbioportal.org). (B) The expression level of SLC25A32 in normal tissues of the colon and intestine. Data were extracted from the GTEX portal (https://gtexportal.org). (C) Representation of the prognostic value of differential expressed SLC25A32 in COAD. Survival curves were plotted for colon cancer patients (n = 550). Data were analyzed using Kaplan-Meier Plotter (http://kmplot.com). Patients with upregulation of SLC25A32 are indicated on the red line, and patients with low expression levels of SLC25A32 are presented on the black line. COAD: colon adenocarcinoma, HR: hazard ratio. (D) PPI network for SLC25A32. PPI network complex containing 4 nodes and edges. SLC25A32 strongly interacts with 3 proteins FLAD1, RFK and SLC46A1 (medium confidence = 0.7)

Table 3.

Reported mutation of SLC25A32 in CRC

Protein change	Cancer type detailed	Functional impact*
A231V	Colon adenocarcinoma	Deleterious and probably damaging
Y243N		Deleterious and probably damaging
T224I		Deleterious and benign
I228M		Deleterious and benign
S15G		Tolerated and benign
H67Y	Rectal adenocarcinoma	Deleterious and probably damaging
I65L		Tolerated and benign
A28V	Mucinous adenocarcinoma of the colon and rectum	Deleterious and probably damaging
Y176H		Tolerated and benign

*Functional impact of mutation was predicted using SIFT and PolyPhen-2, respectively

Protein SLC25A32 is Remarkably Expressed in Normal Tissues of the Colon and Small Intestine

As presented in Fig. 2B, SLC25A32 is notably expressed within the normal tissues of the sigmoid colon, transverse colon, and small intestine. The associated TPM (Transcription per million) was 16.38, 11.96, and 14.09 respectively.

Prognostic Value of SLC25A32 Gene

According to the findings presented in Fig. 2C, it was observed that a total of 146 patients exhibited elevated levels of SLC25A32 gene expression, while 405 patients displayed reduced levels of its expression. The findings indicate a significant decrease in overall survival rates among patients exhibiting heightened expression levels, as exhibited by a hazard ratio of 1. 42 (1.06–1.92) and a log-rank P-value of 0. 0202.

The SLC25A32 Protein Interacts with Some Proteins in the Folate Pathway

Using the STRING online database, a total of 4 genes with the highest node degree were selected (Fig. 2D). SLC25A32 strongly interacts with SLC46A1 (on 17q11.2), FLAD1 (on 1921.3), and RFK (on 9q21.13). Pathway and process enrichment analyses have been carried out with the GO Biological Processes. The result showed that these genes are functionally enriched in the water-soluble vitamin metabolic process (GO:0006767).

Discussion

The mitochondrial folate transporter-SLC25A32 gene- has attracted increasing attention in recent years due to its potential role in tumorigenesis as well as its regulatory function in folate metabolism [16]. Despite its significance, the available literature on SLC25A32 is limited, and the precise role of this gene has not been comprehensively explored.

In this study, the expression level of the SLC25A32 gene was evaluated in tumorous tissues of patients with non-metastatic colorectal cancer. The obtained data showed that the expression level of SLC25A32 was significantly increased in tumoral tissues relative to adjacent normal tissues. This differential expression pattern for SLC25A32 was consistent with results recorded in the Gene Expression Omnibus (GEO), Cancer Genome Atlas (TCGA), cBioPortal, and GEPIA databases.

The data extracted from in-silico analysis also showed that SLC25A32 has notable expression in the colon and small intestine. All of these findings are consistent with the possible role of SLC25A32 in CRC. A recently conducted study by Wettergren in 2021, suggested that increased expression of SLC25A32 in colorectal cancer patients (stage I-III) and is a significant correlation between the elevated SLC25A32 expression and lower risk of disease progression in individuals of advanced colorectal cancer [28]. In a 2022-published study by Veja, he showed that 54 genes have been upregulated in colorectal cancer and have a strong correlation with the development of CRC. Among these genes, SLC25A32, SLC22A3, CA12, CA4, CA1, and SLC4A4 were identified as the most significant genes in colorectal carcinogenesis and tumor progression [29]. We also found that the expression level of SLC25A32 was significantly higher in rectal samples than in the colon samples. There was also a positive association between the expression level of SLC35A2 and tumor size. The observed correlation provides evidence to support the hypothesis that SLC25A32 may have a potential role in tumor progression. However, no significant association was found between the expression levels of SLC25A32 expression and tumor differentiation and stage. This is partly due to the fact that our study was carried out on 30 patients, which may explain the lack of any statistically significant association between the expression of SLC25A32 and various clinicopathological characteristics.

Using in-silico analysis, we found that there is a significant reduction in the overall survival (OS) rates of CRC patients with elevated expression levels of SLC25A32. Also, the prognostic value of SLC25A32 has been documented in several human cancers. In breast and ovarian cancers, there is evidence suggesting an association between the amplification of SLC25A32 and decreased survival rates among patients [16]. We also found 8 missense mutations in SLC25A32 (H67Y, T224I, Y243N, A231V, I228M, Y176H, A28V, and I65L) in the functional domain of SLC25A32. All of them were predicted as pathogenic. Gene amplification of SLC25A32 was also found in 2.1% of the CRC samples of the cBioPortal Database. Previous studies showed that mutations in the SLC25A32 gene are involved in the pathogenesis of neural tube defects (NTD) [9], hypoketotic hypoglycemia [30], acyl-CoA dehydrogenase deficiency (MADD) and glutaric aciduria type II (GAII) [13].

In the PPI network, SLC25A32 interacts closely with FLAD1, RFK, and SLC46A1, which are enriched in the water-soluble vitamin metabolic process and folate metabolism. ETC Complex II requires FAD as an essential cofactor, and the process of fatty acid oxidation plays a pivotal role in sustaining the bioenergetics of mitochondria and maintaining energy homeostasis in the cell [31]. Recently, mutations in genes involved in riboflavin (Rf) transport and metabolism, that is, Rf transporter 1 gene (SLC52A1), FAD synthase gene (FLAD1), and mitochondrial FAD transporter gene (SLC25A32) have been shown to cause MADD(-like) clinical and biochemical phenotypes [32]. On the other hand. It has been shown that inhibition of SLC25A32 may lead to respiratory chain dysfunction in enzyme complex II, which depends on FAD. This in turn can lead to increased production of reactive oxygen species and reduced glutathione [16]. FLAD1, which catalyzes the adenylation of flavin mononucleotide (FMN) to form flavin adenine dinucleotide (FAD) coenzyme, is closely related to the survival of cancer cells and the prognosis of malignant tumor [33]. RFK catalyzes the phosphorylation of riboflavin (vitamin B2) to form flavin mononucleotide (FMN). Knockdown of RFK expression has been shown to induce apoptosis, while its upregulation increases FAD and FMN levels [34]. Therapeutically, reducing mitochondrial FAD concentration by targeting SLC25A32 and its related genes may have promising clinical applications as a stand-alone drug or in combination with other approved cancer drugs. This approach can lead to increased levels of oxidative stress and decreased growth of tumor cells [16].

Conclusion

In conclusion, this study is the first study providing novel experimental and in silico evidence to support the significant role of SLC25A32 in the etiology of non-metastatic colorectal cancer and probably in other human malignancies. It is therefore imperative to undertake further investigation to elucidate the intricate mechanisms underlying its involvement. There is also an urgent need to explore the potential intrinsic and extrinsic factors that may modulate the expression level of SLC25A32. Finally, it is necessary to mention this point that our study was a pilot study and only 30 cases of CRC have been enrolled in our analysis. Such limitation in sample size may adversely impact the credibility of the results. Accordingly, to generalize its application in clinics, it is better to confirm data on a larger sample size and provide insight into the potential effectiveness of SLC25A32 as a new therapeutic target. Finally, it is necessary to remind that our study was a pilot study and while it is true that the sample size of 30 patients in the study may be considered relatively small, it is important to note that pilot studies often have smaller sample sizes as they serve as preliminary investigations to assess the feasibility and potential effects of a particular intervention or treatment. Pilot studies are valuable in providing initial insights and generating hypotheses for further research. However, even though the sample size may be small, it can still provide valuable preliminary data on the potential effects of the intervention being studied. Besides, while statistical significance and generalizability are important considerations in research, it is crucial to interpret the findings of a study within the context of its specific objectives and limitations. The small sample size in this pilot study should be acknowledged, but it does not necessarily invalidate the potential insights and implications that can be drawn from the findings. Further research with larger sample sizes can build upon these initial findings to provide more robust evidence.

Funding

This research was supported by a faculty research grant from School of Medicine, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran (Grant No. CMRC-0004).

Declarations

Conflict of interest

The authors declare that they have no conflicts of interest.

Ethical Approval

The ethical approval for this research was granted by the Ethics Committee of Research Affairs at Ahvaz Jundishapur University of Medical Sciences in Ahvaz, Iran, with a corresponding code of ethics IR.AJUMS.MEDICINE.REC.1400.003. This study was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki.

Informed Consent

Freely-given and informed consent was obtained from all individual participants included in the study.