SLC5A3 is important for cervical cancer cell growth

Novel molecular targets for cervical cancer must be identified. This study examined the role of SLC5A3, a myo-inositol transporter, in the pathogenesis of cervical cancer. Through boinformatics analysis, we showed that the SLC5A3 mRNA levels were upregulated in cervical cancer tissues. The upregulated SLC5A3 mRNA levels were negatively correlated with survival and progression-free interval. Genes co-expressed with SLC5A3 were enriched in multiple signaling cascades involved in cancer progression. In primary/established cervical cancer cells, SLC5A3 shRNA/knockout (KO) exerted growth-inhibitory effects and promoted cell death/apoptosis. Furthermore, SLC5A3 knockdown or KO downregulated myo-inositol levels, induced oxidative injury, and decreased Akt-mTOR activation in cervical cancer cells. In contrast, supplementation of myo-inositol or n-acetyl-L-cysteine or transduction of a constitutively active Akt1 construct mitigated SLC5A3 KO-induced cytotoxicity in cervical cancer cells. Lentiviral SLC5A3 overexpression construct transduction upregulated the cellular myo-inositol level and promoted Akt-mTOR activation, enhancing cervical cancer cell proliferation and migration. The binding of TonEBP to the SLC5A3 promoter was upregulated in cervical cancer. In vivo studies showed that intratumoral injection of SLC5A3 shRNA-expressing virus arrested cervical cancer xenograft growth in mice. SLC5A3 KO also inhibited pCCa-1 cervical cancer xenograft growth. The SLC5A3-depleted xenograft tissues exhibited myo-inositol downregulation, Akt-mTOR inactivation, and oxidative injury. Transduction of sh-TonEBP AAV construct downregulated SLC5A3 expression and inhibited pCCa-1 cervical cancer xenograft growth. Together, overexpressed SLC5A3 promotes growth of cervical cancer cells, representing as a novel therapeutic oncotarget for the devastating disease.


Introduction
Globally, cervical cancer is estimated to be associated with more than 320,000 mortalities every year [1,2]. The incidence (13.1%) and mortality (6.1%) rates of cervical cancer in women are the fourth highest among cancers [3,4].
SLC5A3, which encodes a Na + /myo-inositol (MI) cotransporter, is located at q22 on chromosome 21 and contains one promoter and two exons spanning 2157 nucleotides [11,12]. The SLC5A3 protein comprises 718 amino acid residues and is expressed in different human tissues [12]. The expression of SLC5A3 is reported to be upregulated in patients with Down's syndrome [12][13][14]. Andronic et al. demonstrated that hypotonic stress upregulated SLC5A3 expression and MI transport in HEK293 cells, suggesting that SLC5A3 regulates mammalian cell hypotonic volume [15].

Ivyspring International Publisher
Some studies have examined the function of SLC5A3 in the pathogenesis of human cancer. For example, SLC5A3-dependent MI transport, which promotes nutrient dependency, is required for acute myeloid leukemia (AML) cell proliferation [16]. SLC5A3 silencing decreased MI contents and arrested AML cell proliferation [16]. This study demonstrated that SLC5A3 overexpression promoted cervical cancer cell growth.

Materials and methods
Reagents CCK-8, puromycin, cell culture medium, serum, MI, N-Acetyl-L-cysteine (NAC) as well as RNA reagents, fluorescence dyes, antibodies and the caspase inhibitors were provided by Dr. Cui [17] or purchased from Sigma (St. Louis, Mo).

Human tissues
The cervical cancer tissues along with the matched adjacent normal cervical epithelial tissues were from twenty (20) cervical cancer patients (written-informed consent, administrated at authors' institution). The immunohistochemistry (IHC) staining in four μm-thick tissue slides were reported previously [20].

CRISPR/Cas9
Cells were transfected with lentiviral particles with Cas9-expressing construct and stable cells were formed following selection. The two different sequences encoding small-guide (sgRNA) targeting SLC5A3 were each inserted into a lenti-CRISPR/ Cas9-KO-puro construct (Genechem). sg-1 targets ATCCCAATTTACATCCGGTC with AGG PAM sequence. sg-2 targets TCCCAATTTACATCCGGTCA with GGG PAM sequence. The construct was transfected to HEK-293 cells to generate lentiviral particles. The particles were then added to Cas9 stable cells, and puromycin added to select stable colonies. SLC5A3 KO screening was carried out.

Immunofluorescence and ChIP assays
In the cell culture plate, the glass slide was first fixed and washed with PBS for three times. Next, the 0.5% Triton X-100 was added for permeabilization. Different fluoresce dyes were added to the glass slide and the redundant fluoresce dyes washed off four times with PBS. Water absorbent paper was utilized to suck up the liquid and the fluoresce signalings were observed and collected under a fluorescence microscope (Zeiss). TonEBP chromosome immunoprecipitation (ChIP)'s protocols were described previously [21].

Myo-inositol (MI) detection and superoxide dismutase (SOD) activity
In brief, MI contents were detected through a MI assay kit (ab252896, abcam, Shanghai, China). The SOD activity in fresh tumor tissues was measured under the SOD ELISA kit (Thermo-Fisher Invitrogen, Shanghai, China).

Statistical difference
All the in vitro cellular experiments were repeated five times (n = 5). The statistical methods were described early [23]. P values < 0.05 indicated statistically significant. "N. S." stands for non-statistical difference (P > 0.05).

SLC5A3 is upregulated in cervical cancer
The Cancer Genome Atlas (TCGA) database and Genotype-Tissue Expression (GTEx) dataset, which comprise RNA sequencing data of 10 healthy tissues, three paracancerous tissues, and 306 cervical cancer tissues, were analyzed. The SLC5A3 mRNA expression levels in tumor tissues were markedly higher than those in paracancerous and healthy tissues (P < 0.05) ( Figure 1A). Analysis of TCGA cervical cancer datasets revealed that patients with cervical cancer exhibiting SLC5A3 upregulation were associated with poor overall survival (P = 0.065) ( Figure 1B) and progression-free interval (PFI) (P = 0.001, Figure 1C). The receiver operating characteristic curve analysis suggested that SLC5A3 upregulation can predict the 1-5 year survival rates of cervical cancer patients ( Figure 1D).
Immune cell infiltration analysis based on transcriptome and other omics data revealed that SLC5A3 upregulation was correlated with the infiltration of different immune cells ( Figure 1E). The LinkedOmics functional model was used to examine genes co-expressed with SLC5A3 in TCGA cervical cancer cohort. The top 50 significant genes positively correlated with SLC5A3 were retrieved (Figure 1F-G). Kyoto Encyclopedia of Genes and Genomes analyses revealed that genes co-expressed with SLC5A3 were enriched in multiple pathways involved in cancer progression, including Hippo, FOXO, JAK-STAT, and extracellular matrix pathways ( Figure 1H). Thus, bioinformatics analysis revealed that SLC5A3 is upregulated in cervical cancer.

SLC5A3 is upregulated in clinical cervical cancer tissues and patient-derived or established cervical cancer cells
Next, the clinical specimens of cervical cancer (n = 20) were obtained. The SLC5A3 mRNA levels in cancer tissues ("T") were higher than those in healthy epithelial tissues (N") ( Figure 2A). The protein levels of SLC5A3 were upregulated in the cancer tissue lysates prepared from four representative clinical specimens ( Figure 2B). Western blotting analysis results of all 20 sets of tissue lysates were combined, which revealed significant SLC5A3 protein upregulation in cancer tissues ( Figure 2C). SLC5A3 mRNA and protein expression was significantly upregulated in the primary human cervical cancer cells (pCCa-1/-2/-3 cells, which were obtained from Dr. Cao [18]) and the immortalized HeLa cells ( Figure  2D-F), but were low in the cervical epithelial cells (HCerEpC1 [18] and Ect1/E6E7 cells) (Figure 2D-F).
Transduction with sh-SLC5A3 (seq3) significantly induced apoptosis and increased the number of TUNEL-positive nuclei in pCCa-2 and pCCa-3 primary cervical cancer cells and HeLa cells ( Figure  4G) but did not promote CASP3 activation ( Figure  4H) or induce cell apoptosis (TUNEL assays, Figure  4I) in HCerEpC1 and Ect1/E6E7 epithelial cells.
To investigate the contribution of MI downregulation and oxidative injury to the anti-cervical cancer effects of SLC5A3 depletion, cells were exogenously treated with MI or the anti-oxidant N-acetyl cysteine (NAC). Treatment with MI (2.5 mM) or NAC (400 μM) significantly mitigated the SLC5A3 knockdown/KOinduced downregulation of pCCa-1 cell viability ( Figure 6E) and upregulation of pCCa-1 cell death and apoptosis (Figure 6F-G). However, treatment with MI or NAC alone did not decrease pCCa-1 cell viability and induce death/apoptosis (Figure 6E-G). These results indicate that SLC5A3 depletion promotes cervical cancer cell death by downregulating MI and ROS production.

TonEBP-SLC5A3 promoter binding is upregulated in cervical cancer
The upregulated SLC5A3 mRNA and protein levels in cervical cancer contribute to cancer cell growth. Next, the potential mechanism of SLC5A3 upregulation was examined by focusing on the transcriptional mechanism. Johnson et al. demonstrated that TonEBP (NFAT5) is an important transcription factor of SLC5A3 and that TonEBP expression was correlated with SLC5A3 expression [35]. TonEBP (NFAT5) was the second top differentially expressed gene (DEG) that was positively correlated with SLC5A3 expression in TCGA cervical cancer cohort (Figure 1F-G). The pCCa-1 primary cells were transduced with lentiviral particles encoding TonEBP shRNAs (sh-TonEBP-seq1 and sh-TonEBP-seq2). Stable TonEBP knockdown cells were obtained after selection. Transduction with TonEBP shRNAs markedly downregulated the expression of TonEBP (Figure 9A-B) and SLC5A3 (Figure 9A-B) in pCCa-1 cells.
TonEBP cDNA sequence-encoding lentiviral particles were transduced into pCCa-1 cells to obtain stable TonEBP-overexpressing cells ("TonEBP-OE" cells). The TonEBP Mrna and protein levels, as well as the SLC5A3 mRNA and protein levels, were upregulated in TonEBP-OE pCCa-1 cells (Figure 9C-D). These results indicate that TonEBP is a key transcription factor of SLC5A3 in cervical cancer cells.
The results of ChIP assay ( Figure 9E) revealed that the binding of TonEBP to the proposed SLC5A3 promoter [35] was significantly upregulated in the cervical cancer tissues of four patients but was relatively low in matched adjacent non-cancerous cervical epithelial tissues ( Figure 9E). The binding of TonEBP to the SLC5A3 promoter was strong in pCCa-1/pCCa-2/pCCa-3 primary cells and immortalized HeLa cells (Figure 9F) but weak in HCerEpC1 and Ect1/E6E7 cells (Figure 9F). Therefore TonEBP-SLC5A3 promoter DNA binding increasing in cervical cancer is key for SLC5A3 upregulation.

Discussion
Human papillomavirus (HPV) vaccines, cervical cancer screening, and advances in diagnosis and adjuvant treatments have contributed to the decreased incidence of cervical cancer, especially in developed countries [3,36,37]. Currently, cervical cancer does not feature among the top 10 malignant tumors in the United States. However, in China and Figure 11. SLC5A3 KO suppresses cervical cancer xenograft growth. The koSLC5A3 pCCa-1 cells (with sg1) orCas9C control were s.c. injected to the nude mice, after 60 days pCCa-1 xenograft volumes (A) and weights (B) as well as animal body weights (C) were recorded. In the fresh pCCa-1 xenograft tissues listed mRNAs/proteins were measured (D, E, G and J); Myo-inositol (MI) contents (F), TBAR intensity (H) and SOD activity (I) were tested. *P < 0.05 versus "Cas9C" group. other developing countries, cervical cancer is the second most common tumor among women and is associated with high cancer-related mortality rates each year [3,36,37].
We hypothesized that SLC5A3 is an important oncogenic gene for cervical cancer. Bioinformatics analysis revealed that SLC5A3 is upregulated in cervical cancer tissues. The upregulated SLC5A3 expression was correlated with both poor survival and poor PFI. SLC5A3 upregulation was confirmed in clinical cancer tissues and different cervical cancer cells. SLC5A3 knockdown or KO suppressed cervical cancer cell viability, proliferation, and migration and induced cell death and apoptosis. In contrast, SLC5A3 overexpression promoted cancer cell proliferation and migration. Transduction with sh-SLC5A3 AAV constructs suppressed pCCa-1xenograft growth in nude mice. Similarly, SLC5A3 KO also inhibited pCCa-1 cervical cancer xenograft growth.
MI is used to treat reproductive and metabolic disorders [38]. Previous studies have demonstrated that MI decreases body mass index and increases insulin sensitivity in female patients with polycystic ovary syndrome [39]. Additionally, MI regulates steroidogenesis and modulates androgen and estrogen contents [39]. Lin et al. demonstrated that SLC5A3 is a key metabolic factor for AML using an in vivo CRISPR screening platform [40]. SLC5A3 is critical for AML cell proliferation. CRISPR-mediated SLC5A3 KO markedly suppressed AML orthotopic xenograft growth in vivo and induced apoptosis [40]. SLC5A3-induced MI import promotes AML cell proliferation [40]. Wei et al. demonstrated that MI promotes nutrient dependency in AML and that MI imported via SLC5A3 maintained AML cell proliferation [16]. This study demonstrated that cellular MI contents were upregulated in SLC5A3 knockdown/KO cervical cancer cells but were upregulated in SLC5A3-overexpressing cells. MI was also downregulated in sh-SLC5A3 AAV-injected or koSLC5A3 (sg1)-injected cervical cancer xenograft tissues. MI supplementation alleviated SLC5A3 KO-induced apoptosis and cell death in cervical cancer cells. Thus, SLC5A3-dependent MI import is critical for cervical cancer cell growth. Figure 12. TonEBP silencing inhibits pCCa-1 cervical cancer xenograft growth. The nude mice bearing pCCa-1 xenografts were intratumorally injected with TonEBP shRNA-expressing AAV ("AAV-shTonEBP-seq1") or AAV-shC, every 48h for 10 days; Weekly pCCa-1 xenograft volumes (A) and body weights of the mice (D) were recorded; Daily pCCa-1 xenograft growth was estimated (B). All pCCa-1 xenografts were isolated at Day-42 and xenograft weights were recorded (C). At Day-20, one pCCa-1 xenograft per group was isolated; Listed mRNAs/proteins (E, F and H) and myo-inositol (MI) contents (G) were tested. *P < 0.05 versus "AAV-shC" group.
Akt-mTOR cascade hyperactivation is a driving factor for cervical cancer growth and progression [33,34,41]. Various pharmacological inhibitors or monoclonal antibodies against the Akt-mTOR cascade-related factors exerted therapeutic effects on cervical cancer [33,34,41]. This study proposed that SLC5A3 is essential for Akt-mTOR signaling activation. SLC5A3 knockdown/KO significantly decreased the phosphorylation of Akt1 and S6K in primary cervical cancer cells. The Akt-mTOR cascade was inhibited in sh-SLC5A3 AAV-injected cervical cancer xenograft tissues and SLC5A3 KO xenograft tissues. In contrast, SLC5A3 overexpression upregulated the phosphorylation of Akt1 and S6K. The reactivation of Akt-mTOR by caAkt1 suppressed SLC5A3 KO-induced cervical cancer cell death. Thus, SLC5A3 mediates cervical cancer cell growth, at least partially, by promoting Akt-mTOR activation.
SLC5A3, which is widely expressed in different human tissues, is important for cellular osmoregulation [11]. Previous studies have reported that SLC5A3 regulates inflammatory cell infiltration during the progression of sporadic inclusion body myositis [29]. Zhou et al. revealed the potential anti-oxidant activity of SLC5A3 and demonstrated that SLC5A3 knockdown augmented oxidative stress in human neuroblastoma cells [31]. Long non-coding RNA NORAD upregulated SLC5A3 expression by sponging microRNA-204-5p and consequently alleviates oxidative injury and suppressed cell death in neuroblastoma cells [31]. The ROS and oxidative stress levels were upregulated in SLC5A3 knockdown cervical cancer cells and SLC5A3 knockdown/KO xenograft tissues. The anti-oxidant NAC suppressed SLC5A3 KO-induced cytotoxicity against cervical cancer cells. Thus, SLC5A3 depletion exerts anticervical cancer effects by inducing oxidative stress.
This study demonstrated that TonEBP is the key transcription factor for SLC5A3 in cervical cancer. TonEBP was the second top DEG that was positively correlated with SLC5A3 in TCGA cohort. The SLC5A3 mRNA and protein expression levels were downregulated upon TonEBP knockdown but were upregulated upon ectopic TonEBP overexpression in cervical cancer cells. The binding of TonEBP to the SLC5A3 promoter was upregulated in various cervical cancer tissues/cells and may contribute to SLC5A3 upregulation. TonEBP knockdown suppressed SLC5A3 expression and inhibited pCCa-1 tumor growth in vivo. Thus, TonEBP upregulates SLC5A3 expression and promotes cervical cancer cell growth.

Conclusion
The upregulated SLC5A3 promotes cervical cancer cell growth.