N-glycosylation of GSTO1 promotes cervical cancer migration and invasion through JAK/STAT3 pathway activation

doi:10.21203/rs.3.rs-4813636/v1

Download PDF

Research Article

N-glycosylation of GSTO1 promotes cervical cancer migration and invasion through JAK/STAT3 pathway activation

https://doi.org/10.21203/rs.3.rs-4813636/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background Protein glycosylation is strongly associated with tumor progression. Glutathione S-transferase omega 1 (GSTO1) is a member of the glutathione S-transferase family. The significance of GSTO1 N-glycosylation in the progression of cervical cancer (CC) has remained elusive. In this study, we investigated the functional significance of GSTO1 N-glycosylation in CC progression.

Methods We employed immunohistochemistry to detect the relative expression of evaluating the link between GSTO1 in CC and benign tissues and the overall survival (OS) and progression-free survival (PFS) in CC patients. In vitro and in vivo experiments to detect CC cell proliferation or metastatic ability after GSTO1 downregulation. NetNGly1.0 Server database predicts potential N-glycosylation modification sites of GSTO1 (Asn55, Asn135, Asn190). Investigating GSTO1 N-glycosylation's function in cellular migration, invasion and epithelial–mesenchymal transition (EMT), we mutated the N-glycosylation sites of GSTO1 through lentivirus-based insertional mutagenesis. Detection of signalling pathways associated with N-glycosylation-modified GSTO1 by enrichment analysis and Western blot.

Results Compared to normal cervical tissue, CC tissue showed significantly higher GSTO1 expression. Further, high GSTO1 levels were a poor predictor of OS and PFS. Both cell and animal experiments suggested that down-regulation of GSTO1 inhibited cell proliferation and metastasis. Glycosylation modification of targeted mutant GSTO1 at positions 55, 135 and 190 significantly inhibits migration and invasion of CC cells. GSTO1 N-glycosylation fixed point mutation inhibits EMT process in CC cells. Mechanistically, N-glycosylated GSTO1 promoted the expression of JAK/STAT3 pathway related markers.

Conclusions GSTO1 N-glycosylation is associated with CC progression and may promote EMT via JAK/STAT3 signaling.

GSTO1

N-glycosylation

JAK/STAT3

EMT

Cervical cancer

Cervical cancer (CC) ranks as the world's fourth most prevalent cancer among women ^[1] and stands as the primary reason for cancer-related fatalities in women in developing nations, accounting for nearly 8% of all female cancer deaths every year ^{[2, 3]}. In 2020, 604,100 new cases of CC and 341,800 deaths were reported worldwide ^[4]. Patients with early-stage cervical cancer have a relatively good prognosis, while advanced disease is associated with a greater chance of metastasis and a relatively poor prognosis ^[5]. Therefore, improving the prognosis of patients with advanced and progressive CC remains a challenge.

Glycosylation, a common posttranslational modification, is altered in many cancers ^[6]. Glycoconjugates primarily consist of N-linked and O-linked glycoproteins, where glycans are covalently linked to a polypeptide structure, mainly through the connection of a nitrogen atom (N) in asparagine (Asn) or an oxygen atom (O) in serine (Ser) or threonine (Thr) ^{[7, 8]}. These are known as N-glycans or O-glycans, respectively. Glycosylation plays a significant role in tumorigenesis and tumor progression, including signal transduction in tumor cells as well as tumor angiogenesis, invasion, metastasis, immunity, and metabolism ^[9]. Alterations in glycosylation influence the structure and function of proteins, causing cells to acquire new features closely related to tumor growth, progression, and metastasis ^[10]. Thus, glycosylation-related factors represent potential therapeutic targets and biomarkers for cancer.

Glutathione S-transferase omega 1 (GSTO1), belonging to the gene group known as glutathione S-transferase (GST), catalyzes the binding of glutathione to various exogenous and endogenous electrophilic substrates, including carcinogenic drugs, therapeutic agents, and oxidative stress products ^[11]. Emerging evidence indicates that GSTO1 polymorphisms increase the risk of cutaneous malignant melanoma ^[12], breast cancer ^[13], pancreatic cancer ^[14], and clear cell renal cell carcinoma ^[15]. However, the functional relevance of GSTO1 in CC progression and therapeutic response remains uncertain. GSTO1 has been reported to regulate JAK/STAT3 signaling in non-small cell lung cancer ^[16]. Further, studies have reported that the JAK/STAT pro-inflammatory pathway is activated in HPV-associated CC ^[17]. Therefore, we speculated that GSTO1 N-glycosylation could regulate the progression of CC through JAK/STAT3 signaling and assessed this through a series of experiments.

The JAK/STAT3 signaling plays a central role in various types of cancers ^[18]. Activation of this process enhances tumorigenic and metastatic capacities, facilitates the transformation of cancer stem cells, and fosters chemoresistance by encouraging the epithelial-mesenchymal transition (EMT) ^[19]. EMT is a critical regulator of cancer progression involved in the regulation of invasion, spread, and survival ^[20].

To date, the role of GSTO1 N-glycosylation in cervical cancer has not been reported. In this study, we found that GSTO1 could be modified by N-glycosylation at positions Asn55, Asn135, and Asn190. Our study showed that N-glycosylated GSTO1 inhibits the proliferation, migration, and invasion of CC cells in vitro. In addition, we elucidated the molecular mechanism through which GSTO1 N-glycosylation modulates CC migration and invasion.

Cell culture and lentiviral transfection

Fuxiang Biotechnology Co. supplied the HeLa and C33A cells. (Location: Shanghai, China). HeLa cells were cultivated using DMEM (Thermo Fisher Scientific, Waltham, MA, USA), whereas C33A cells were cultivated in MEM (Gibco-BRL, Gaithersburg, MD, USA), each infused with 10% fetal bovine serum (FBS; Bioind, Kibbutz Beit-Haemek, Israel). All cells were maintained at 37°C and 5% CO₂ in an incubator.

The shGSTO1 and shNC constructs, in conjunction with the lentivirus package plasmids, underwent co-transfection into HEK-293 cells to produce the shRNA lentivirus. Post-filtration and enrichment, CC cells underwent infection with the lentivirus. Subsequently, the infected cells underwent puromycin selection (1.0 µg/mL) for an additional 4–5 cycles, followed by the creation of the stable transfected HeLa and C33A cell lines.

Lentiviruses containing GSTO1 55-position N-glycosylation mutant vector gene sequences (N55Q group), 135-position N-glycosylation mutant vector gene sequences (N135Q group), and 190-position N-glycosylation mutant vector gene sequences (190Q group), as well as GSTO1 wild-type vector gene sequences (WT group), and empty vector gene sequences (vector group) were transfected into CC cells. Subsequently, the infected cells underwent a selection process using puromycin (1.0 µg/mL) for an additional 4–5 passages. The stably transfected HeLa and C33A cell lines were thus established.

IHC

Samples of clinical cervical tissue were gathered at Shihezi University Medical College's First Affiliated Hospital (Xinjiang, China), adhering to the guidelines sanctioned by the Medical Ethics Committee and the Declaration of Helsinki. A total of 102 human CC samples were formalin-fixed and paraffin-embedded. Clinicopathological information on 102 cases was collected until December 2019, and FIGO staging 2018 edition has not yet been administered, so in this study FIGO staging was grouped according to the 2009 edition. Hematoxylin- and eosin-stained sections were observed under a microscope, and the most representative necrosis-free areas were selected. Tissue sections were stained with antibodies against GSTO1 (1:100) and Ki67 (1:200). Immunohistochemistry (IHC) staining for GSTO1 was scored semiquantitatively. The staining intensity of tumor cells was divided into three levels: 0 (negative), 1 (weak), 2 (moderate), and 3 (strong), depending on the proportion of stained cells compared to the overall cell count (< 5%, 0 points; 5–25%, 1 point; 26–50%, 2 points; 51–75%, 3 points; and > 75%, 4 points). The ultimate score was calculated by multiplying the intensity score by the percentage score. The immune response to GSTO1 was categorized into negative (0 points), weak (1–6 points), or positive (≥ 6 points) groups. GSTO1 expression was classified as low if it was negative or feeble, and as high if it was positive.

qRT-PCR

The complete RNA was isolated utilizing the Trizol reagent (Invitrogen, Carlsbad, CA). The RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Waltham, MA, USA) facilitated the synthesis of cDNA from the isolated RNA. The application of GSTO1 primers (forward: CCTGAGCTACCCAGTGTAGA and reverse: TTTCGGCATACTCTTTCAGA) followed the guidelines provided by the manufacturer. The presence of GSTO1 was identified through quantitative reverse transcription-polymerase chain reaction (qRT-PCR) on a Bio-Rad GSTO1 sequence detection system (Hercules, CA), utilizing the SYBR Green PCR kit (QIAGEN, Hilden, Germany). GAPDH served as the internal standard for control.

Western blot

The total protein content was extracted by lysing cells using RIPA lysis buffer (Solarbio, China) and protease inhibitors (Solarbio). Post-denaturation, the protein specimens underwent separation using 6–12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and were then moved onto a polyvinylidene difluoride membrane (Millipore, USA). After blocking with protein rapid sequester, the membrane, together with the primary antibodies, including GSTO1 (1:2000, Abcam), E-Cadherin (1:3000, Abcam), N-Cadherin (1:3000, Abcam), Vimentin (1:2000, Abcam), JAK (1:2500, Abcam), p-JAK (1:2000, Abcam), STAT3 (1:3000, Abcam), p-STAT3 (1:1000, Abcam), and β-actin (1:10000), were incubated at 4°C for 14–16 h. The polyvinylidene fluoride membrane was then washed with TBST thrice and incubated with anti-mouse, anti-rabbit horseradish peroxidase-coupled secondary antibodies at 20 ℃ for 2 h. The cells were then washed five times with TBST. The protein bands were visualized using an electrochemical luminescence kit (Biosharp, China), and the intensity was quantified using ImageJ software.

EdU staining

Stably transfected cells were seeded into 96-well plates (1×10⁴ cells per well) and cultured for 24 h. EdU (5-ethyl-20-deoxyuridine) staining was performed according to manufacturer instructions. EdU-positive cells (Cy3 vs. DAPI, %) were recorded in five random views under a light microscope.

CCK-8 assays

After growing for 24 hours, 1×10³ cells were distributed into 96-well plates. Following this, each well was treated with a solution of CCK-8 (Dojindo, Rockville, MD, USA) and left to incubate for 2 hours. The absorbance readings of each well at 450 nm were then measured using a Microplate A reader (Bio-Rad, Hercules, CA, USA).

Wound-healing migration assay

The cells were spread out and grown in 12-well plates. When the cells were approximately 100% confluent, the plates were scraped with a pipette tip and washed to remove the debris. The serum-containing medium was then discarded, and the cells were cultured for 24 h in a cell incubator. At 0 and 24 h, cell migration to the wound was photographed under a microscope.

Transwell assay

For the migration experiment, 1×10⁴ cells were plated in chambers with a non-coated membrane (24-well insert; pore size, 8 mm; Corning Inc). For the invasion tests, 1×10⁴ cells were plated in chambers containing Matrigel-coated membranes (24-well inserts; pore size, 8 mm; Corning Inc.). After 24 h of culture, cells on the inferior surface of the membrane were stained with crystal violet and then photographed and counted.

Animal experiments

A total of 1×10⁶ shGSTO1, shNC, or control HeLa cells were subcutaneously injected into the axillae of 5-week-old female BALB/c nude mice. The tumor volume was calculated as follows: volume = (length × width^²)/2, after which the length and width of the tumors were measured every 3 days. Nude mice were euthanized after 6 weeks, whereafter tumors were collected and histologically examined for the expression of GSTO1 and Ki-67 markers.

Statistical analyses

IBM SPSS software (version 26.0) was used to perform statistical analyses. Count data, expressed as a percentage of incidence, were compared between the groups using a chi-square test. The t-test was used for intergroup comparison of data that met the normal distribution, and the rank-sum test was used to compare data that did not the meet normal distribution. Kaplan-Meier survival curves and log-rank tests were used to analyze the relationship between GSTO1 expression and patient overall survival (OS). Univariate and multivariate Cox regression analyses were used to assess independent risk factors affecting the prognosis of cervical cancer. Statistical significance was set at p < 0.05.

GSTO1 is upregulated in CC

To determine the expression pattern of GSTO1 in CC, 102 CC samples were evaluated through IHC staining. The baseline clinicopathological data of patients with CC is shown in Table S1. IHC analysis revealed higher GSTO1 protein expression in CC tissues than in normal cervical tissues (Fig. 1A-B). Based on GSTO1 protein expression, samples were classified into two groups: a high expression group with scores ≥ 6 and a low expression group with scores < 6. The correlation between GSTO1 expression and the clinicopathological features of patients with CC is shown in Table 1. GSTO1 was associated with FIGO stage, differentiation, tumor size, DSI, LVSI, and LNM. Meanwhile, GSTO1 expression was not associated with age or type.

Table 1

The association of GSTO1 protein expression and patient clinical characteristics
Patient characteristics	N (%)	GSTO1 Expression		\(\:{\varvec{x}}^{2}\)	P
Patient characteristics	N (%)	High	Low	\(\:{\varvec{x}}^{2}\)	P
Age (year)
<50	52(51.0)	21	31	0.061	0.805
≥ 50	50(49.0)	19	31	0.061	0.805
FIGO Stage
≤ⅠB1	29(28.4)	19	10	11.759	0.001
≥IB2	73(71.6)	21	52	11.759	0.001
Histological type
Squamous cell carcinoma	79(77.5)	29	50	0.924	0.337
Adenocarcinoma	23(22.5)	11	12	0.924	0.337
Differentiation
Well differentiated	21(20.6)	13	08	5.711	0.017
Moderately and poorly differentiated	81(79.4)	27	54	5.711	0.017
Tumor size(cm)
<2.5	53(52.0)	27	26	6.366	0.012
≥ 2.5	49(48.0)	13	36	6.366	0.012
DSI
<1/2	55(53.9)	30	25	11.768	0.001
≥ 1/2	47(46.1)	10	37	11.768	0.001
LVSI
Yes	25(24.5)	03	22	10.291	0.001
No	77(75.5)	37	40	10.291	0.001
LNM
Yes	14(13.7)	02	12	4.231	0.040
No	88(86.3)	38	50	4.231	0.040
FIGO, International Federation of Gynecology and Obstetrics; DSI, Depth of stromal invasion; LVSI, Lymphovascular space infiltrates; LNM, Lymph node metastasis.

High GSTO1 expression in cancer tissue indicates poor prognosis in patients with CC

Kaplan-Meier survival curves showed that the 5-year cumulative OS was lower in the high GSTO1 expression group (48.4%, 30/62) than in the low GSTO1 expression group (62.5%, 25/40), as was the 5-year cumulative PFS (41.9%, 26/62 versus 57.1%, 24/40) (Fig. 1C). According to the univariate Cox regression model, age (HR: 2.567, 95% CI: 1.116 ~ 5.964, p = 0.027), FIGO stage (HR: 16.673, 95% CI: 2.180 ~ 127.499, p = 0.007), differentiation (HR: 11.574, 95% CI: 1.504 ~ 89.074, p = 0.019), and GSTO1 expression (HR: 6.222, 95% CI: 1.846 ~ 20.975, p = 0.003) were the key factors affecting OS in patients with CC (Table 2). Further, multivariate Cox regression analysis showed that age (HR: 2.711, 95% CI: 1.159 ~ 6.343, p = 0.021), FIGO stage (HR: 8.304, 95% CI: 1.083 ~ 63.644, p = 0.042), and GSTO1 expression (HR: 3.559, 95% CI: 1.039 ~ 12.198, p = 0.043) were independent risk factors affecting OS in patients with CC (Table 2).

Table 2

Univariate and multivariate analysis of OS in patients with CC
Characteristics	Univariate			Multivariate
Characteristics	HR	95%CI	P	HR	95%CI	P
Age (< 50 vs. ≥50)	2.576	(1.116 ~ 5.964)	0.027	2.711	(1.159 ~ 6.343)	0.021
FIGO stage (≤ⅠB1 vs ≥ IB2)	16.673	(2.180 ~ 127.499)	0.007	8.304	(1.083 ~ 63.644)	0.042
Type (Squamous carcinma vs Adenocarcinoma)	0.731	(0.276 ~ 1.935)	0.529
Differentiation(Well vs Moderate and poor)	11.574	(1.504 ~ 89.074)	0.019	2.581	(0.353 ~ 23.012)	0.326
Tumor size (<2.5cm vs ≥ 2.5cm)	2.038	(0.871 ~ 4.770)	0.101
DSI (<1/2 vs ≥ 1/2)	1.985	(0.880 ~ 4.477)	0.098
LVSI (Yes vs No)	0.537	(0.237 ~ 1.218)	0.137
LNM (Yes vs No)	0.501	(0.170 ~ 1.470)	0.208
GSTO1 (Low vs High)	6.222	(1.846 ~ 20.975)	0.003	3.559	(1.039 ~ 12.198)	0.043

As per the univariate Cox regression model, age (HR: 2.276, 95% CI: 1.094 ~ 4.736, p = 0.028), FIGO stage (HR: 6.931, 95% CI: 1.644 ~ 29.220, p = 0.008), differentiation (HR: 8.916, 95% CI: 1.214 ~ 65.490, p = 0.032), tumor size (HR: 2.139, 95% CI: 1.005 ~ 4.552, p = 0.048), DSI (HR: 2.642, 95% CI: 1.264 ~ 5.523, p = 0.010), and GSTO1 expression (HR: 6.122, 95% CI: 1.856 ~ 20.188, p = 0.003) were the key factors affecting PFS in patients with CC (Table 3). Multivariate Cox regression analysis showed that age (HR: 2.643, 95% CI: 1.231 ~ 5.676, p = 0.013) and GSTO1 expression (HR: 4.082, 95% CI: 1.223 ~ 13.621, p = 0.022) were independent risk factors affecting PFS in patients with CC (Table 3).

Table 3

Univariate and multivariate analysis of PFS in patients with CC
Characteristics	Univariate			Multivariate
Characteristics	HR	95%CI	P	HR	95%CI	P
Age (< 50 vs. ≥50)	2.276	(1.094 ~ 4.736)	0.028	2.643	(1.231 ~ 5.676)	0.013
FIGO stage (≤ⅠB1 vs ≥ IB2)	6.931	(1.644 ~ 29.220)	0.008	3.114	(0.655 ~ 14.811)	0.153
Type (Squamous carcinma vs Adenocarcinoma)	0.545	(0.209 ~ 1.420)	0.214
Differentiation(Well vs Moderate and poor)	8.916	(1.214 ~ 65.490)	0.032	3.561	(0.473 ~ 26.820)	0.218
Tumor size (<2.5cm vs ≥ 2.5cm)	2.139	(1.005 ~ 4.552)	0.048	1.155	(0.511 ~ 2.607)	0.730
DSI (<1/2 vs ≥ 1/2)	2.642	(1.264 ~ 5.523)	0.010	1.782	(0.883 ~ 3.813)	1.782
LVSI (Yes vs No)	0.732	(0.345 ~ 1.556)	0.418
LNM (Yes vs No)	0.632	(0.221 ~ 1.8122)	0.394
GSTO1 (Low vs High)	6.122	(1.856 ~ 20.188)	0.003	4.082	(1.223 ~ 13.621)	0.022

GSTO1 downregulation inhibits the proliferation and metastasis of CC cell lines

To further investigate the influence of GSTO1 on CC cell proliferation and metastasis, we suppressed GSTO1 in two CC cell lines (HeLa and C33A) via lentiviral transfection with GSTO1 shRNA (shGSTO1) and a negative control (shNC). Western blot and qRT-PCR were used to confirm GSTO1 protein and mRNA expression, which revealed significant suppression of GSTO1 in cells stably transfected with shGSTO1 relative to the control and shNC cells (Fig. 2A-B).

The effect of GSTO1 downregulation on the proliferation of HeLa and C33A cells was determined using CCK-8 assays. Compared with the control and shNC cells, the shGSTO1 group cells exhibited a significantly reduced proportion of positive nuclei in both cell lines (Fig. 2C). The same trend was observed via EdU assays in HeLa and C33A cells (Fig. 2D). To further explore the effect of GSTO1 on the proliferative capacity of CC cells in vivo, lentivirus-shGSTO1, lentivirus-shNC, or control HeLa cells were subcutaneously inoculated on the left side of four nude mice to construct a subcutaneous xenograft animal model. Our results showed that the tumor volume in the shGSTO1 group was significantly reduced compared to that in the control and shNC groups (Fig. 2E). HE staining was used to examine the histological morphology of sections, confirming the presence of tumors. IHC showed that, compared to the control and shNC groups, the level of GSTO1 in tumor tissue in the control and shNC groups was significantly lower (Fig. 2F).

The effects of GSTO1 downregulation on the invasion and migration of HeLa and C33A cells were assessed via cell scratch tests and Transwell assays. Scratch tests revealed a significantly lower mobility of shGSTO1 group cells relative to that of control and shNC group cells at 24 h. Similar results were observed in C33A cells (Fig. 2G). Transwell migration and invasion experiments showed that the number of HeLa and C33A cells transfected with shGSTO1 passing through the chamber basement membrane was significantly lower than that in the control and shNC groups (Fig. 2H-I).

GSTO1 is N-glycosylated

HeLa and C33A cells were treated with the N-glycosylation inhibitor tunicamycin, whereafter lysates were subjected to Western blot analysis. Total protein was also treated with PNGase F (recombinant glycosidase) or Endo H (endoglycosidase H), whereby GSTO1 protein was detected. We found that GSTO1 protein levels were significantly suppressed after treatment with tunicamycin (Fig. 3A). GSTO1 levels were significantly reduced after treatment with PNGase F or Endo H (Fig. 3B). These results suggested that GSTO1 is N-glycosylated.

To identify GSTO1 glycosylation sites, we searched for evolutionarily conserved NXT motifs within the GSTO1 amino acid sequence, using the NetNGlyc 1.0 Server database for prediction (N-X-S/T, X! = P). N-glycosylation was predicted to occur at positions N55 (p = 0.67), N135 (p = 0.62), and N190 (p = 0.69) (Figure S1A). To determine the presence of N-glycosylation on these potential residues, we mutated them to glutamine (Q), thus generating N55Q, N135Q, N190Q, or 3Q (Fig. 3C). Sequencing results indicated the successful construction of GSTO1 N-glycosylation mutants. CC cell lines were transfected with vectors, GSTO1 wild-type (WT), N55Q, N135Q, N190Q, or 3Q, whereafter cell lysates were analyzed via Western blot. Our results showed that the N55Q, N135Q, N190Q, and 3Q mutants exhibited significantly lower protein levels than the WT (Fig. 3D).

Mutation of N-glycosylation sites in GSTO1 suppresses the proliferation and metastasis of CC cells

Tumor cells exhibit an enhanced ability to proliferate in vivo and ex vivo. Thus, we examined the effect of GSTO1 N-glycosylation on the proliferative capacity of CC cells through EdU (Fig. 4A) and CCK-8 assays (Fig. 4B). CC cells harboring the N55Q, N135Q, N190Q, and 3Q mutants exhibited compromised proliferation compared to WT cells. This suggests that GSTO1 N-glycosylation is associated with the proliferative capacity of CC cells.

Further, scratch healing assays showed a reduced cell migration ability in the mutant groups (Fig. 4C). Transwell assays confirmed that GSTO1 mutants exerted an inhibitory effect on the migration and invasion abilities of HeLa and C33A cells (Fig. 4D-E).

N-glycosylated GSTO1 promotes CC progression through the JAK/STAT3 signaling pathway

The specific regulatory mechanism of GSTO1 N-glycosylation in CC remains unknown. To further determine the possible roles of GSTO1 and its N-glycosylation in CC, we subjected it to KEGG pathway enrichment analysis and found a significant association with the JAK/STAT signaling pathway (Figure S1B).

First, we detected the expression of JAK/STAT3 pathway factors JAK, STAT3, p-JAK, and p-STAT3 via Western blot analysis. As shown in Fig. 5A, cells harboring GSTO1 N-glycosylation mutants exhibited lower p-JAK and p-STAT3 protein levels, with no significant changes in total levels of JAK and STAT3.

To determine whether CC cell proliferation and migration were promoted via the JAK/STAT3 pathway, we treated cells with the JAK/STAT3 inhibitor WP1066. Western blot analysis showed that WP1066 significantly inhibited the phosphorylation of JAK and STAT3 (Fig. 5B). Further, EdU and CCK-8 assay results indicated lower proliferation in WP1066-treated cells (Fig. 5C-D). Scratch healing and Transwell assays (Fig. 5E-G) extended the suppressive effect of WP1066 to CC migration and invasion.

To further verify that GSTO1 N-glycosylation affects the invasion and metastasis of CC through JAK/STAT3 signaling, the JAK/STAT3 signaling activator colivenlin was added to cells harboring GSTO1 simultaneously mutated at three GSTO1 N-glycosylation sites. Western blot analysis showed that colivenlin significantly increased the levels of p-JAK and p-STAT3 (Fig. 6A). EdU and CCK-8 assay results indicated that colivenlin promoted CC cell proliferation (Fig. 6B-C). Scratch healing and transwell assays (Fig. 6D-F) demonstrated that colivenlin enhanced the migration and invasion of HeLa and C33A cells.

N-glycosylated GSTO1 regulates the EMT of CC cells via JAK/STAT3 signaling

To verify whether N-glycosylated GSTO1 affects the EMT of HeLa and C33A cells, the expression of EMT-related factors E-cadherin, N-cadherin, and Vimentin was determined via Western blot analysis. As shown in Fig. 7, compared to the vector and WT groups, cells harboring GSTO1 N-glycosylation mutants exhibited significantly higher E-cadherin levels, in parallel to a downregulation of N-cadherin and Vimentin (Fig. 7A).

To verify whether JAK/STAT3 signaling promotes the EMT, EMT-related factor expression was detected via western blotting after treatment with a JAK/STAT3 signaling pathway inhibitor. JAK/STAT3 signaling inhibition significantly enhanced E-cadherin levels in HeLa and C33A cells while downregulating N-cadherin and Vimentin (Fig. 7B).

Next, we used colivelin in CC cell lines to explore whether GSTO1 N-glycosylation affects the EMT via JAK/STAT3 signaling. As shown in Fig. 7C, N-cadherin and Vimentin were upregulated in the colivelin-treated group, while E-cadherin was suppressed, when compared to the respective levels in the 3Q groups. Therefore, colivelin reversed the effect of N-glycosylation-defective GSTO1 on the EMT.

CC ranks first in incidence among malignant tumors of the female reproductive system ^[21]. Approximately one-fifth of patients with CC are from China, emphasizing the threat that this disease poses to Chinese women ^{[22, 23]}. More than 50% of patients in China are initially diagnosed with locally advanced CC, which is usually characterized by lymph node and distant metastases, poor therapeutic efficacy, a high rate of recurrence, and a five-year survival rate of less than 60% ^[24]. Currently, the molecular pathogenesis of CC is not well understood. The treatment of locally advanced disease is based on the combination of radiotherapy and chemotherapy, and there is a lack of prognostic biomarkers in clinical practice.

GSTO1 is a member of the omega class of the GST family, with a unique cysteine site and multiple reductase activities ^[25]. Previous studies have demonstrated that GSTO1 alters the structure of β-actin, heat shock protein 70, heat shock protein 7c, and prolactin-inducible proteins through deglutamylation, which in turn modulates the cellular stress response, thereby effectively controlling the ontogeny, differentiation, and proliferation of cancer cells ^[26]. It has been shown that GSTO1 is widely expressed in various tumours and is correlated with cancer occurrence and development. GSTO1 is highly expressed in colon cancer ^[27], non-small cell lung cancer ^[16], and cutaneous malignant melanoma ^[12]. First, IHC confirmed the increased expression of GSTO1 in CC compared to normal cervical tissues. Further, GSTO1 protein expression was significantly correlated with FIGO stage, differentiation, tumor size, DSI, LVSI, and LNM. Kaplan-Meier survival analysis demonstrated that, in patients with CC, high GSTO1 expression was associated with a poorer OS and PFS. In addition, a univariate factorial Cox model showed that high GSTO1 expression, age, FIGO stage, and differentiation were key factors affecting the OS of patients with CC. Further, high GSTO1 expression, age, FIGO stage, differentiation, tumor size, and DSI were identified as key factors affecting PFS. Importantly, high GSTO1 expression was an independent risk factor for survival in patients with CC. This evidence suggests that GSTO1 may assume a significant part in promoting CC progression and has potential as a new prognostic biomarker for CC.

Studies have suggested that GSTO1 plays a key role in the invasion and metastasis of many cancers. High GSTO1 expression in colon cancer was shown to promote the efficacy of cisplatin-based chemotherapy in patients ^[27]. Further, GSTO1 was significantly upregulated in non-small-cell lung cancer ^[16] and cutaneous melanoma ^[12], promoting invasion and metastasis. However, previous studies have not investigated whether GSTO1 enhances the proliferation, invasion, and migration of CC cells. In the present study, EdU and CCK8 assays confirmed that downregulation of GSTO1 inhibited the proliferation of HeLa and C33A cells. Further, cell scratch and Transwell assays confirmed that downregulation of GSTO1 significantly inhibited CC cell invasion and migration. The tumor-suppressive effect of GSTO1 downregulation was also confirmed in a xenograft nude mouse model.

Protein glycosylation is a common post-translational modification that plays an essential role in cellular localization, intercellular signaling, proliferation, migration, and differentiation. As such, it is implicated in the development of a wide range of pathologies, including autoimmune diseases, inflammatory diseases, and tumors ^[28]. Glycosylation site can be classified as N-glycosylated, O-glycosylated, C-mannosylated, or glycosylphosphatidylinositol-anchored linkages, of which N-glycosylated and O-glycosylated linkages are the most common ^{[29, 30]}. Numerous studies have confirmed that altered protein glycosylation is associated with tumorigenesis, cancer progression, and metastasis. CD44 O-glycosylation promotes tumor metastasis by binding to endothelial E-selectin in colon carcinoma cells ^[31]. FUCA1-mediated protein deglycosylation of fucoidan glycosylation is associated with distant tumor metastasis and has prognostic value ^[32]. N-glycosylation of SGK196 inhibited the invasion and metastasis of breast cancer cells via PI3K/AKT/GSK3β signaling ^[33]. Here, we identified three N-glycosylation sites within the GSTO1 protein in CC cells, namely, Asn55, Asn135, and Asn190. Mutating any of these led to a decrease in GSTO1 expression, and N-glycosylation was required for maintaining their expression. Further, EdU proliferation and CCK-8 assays verified that the GSTO1 N-glycosylation mutants significantly inhibited HeLa and C33A cell proliferation. Cell scratch and Transwell assays confirmed that N-glycosylation-defective GSTO1 significantly inhibited the invasion and migration of CC cells.

The JAK/STAT3 signaling pathway is a common pathway for the signal transduction of multiple cytokines and growth factors, implicating the development of various and contributing to their proliferation, migration, and apoptosis ^{[34, 35]}. JAK/STAT signaling activation promotes colorectal carcinogenesis and malignancy ^{[36, 37]}. Further, apoptosis and cell cycle blockade of lung cancer cells are induced through modulation of JAK/STAT signaling ^{[38, 39]}. Inhibition of OTX1 was shown to suppress proliferation, migration, and apoptosis of ovarian cancer through inhibition of JAK/STAT signaling ^[40]. IL4I1 promotes malignant lung adenocarcinomas through the JAK/STAT signaling pathway and immune escape ^[41]. EYA2 binds to DACH1 to regulate the expression of SOCS3, which in turn inhibits hepatocellular carcinoma progression through JAK/STAT blockade ^[42]. STAT3 assumes a significant part in cervical carcinogenesis, and inhibition of the JAK/STAT pathway may be necessary to promote CC cell death. We found that GSTO1 was significantly associated with the JAK/STAT signaling in CC through enrichment analysis, then speculated that GSTO1 N-glycosylation might regulate the progression of CC through the JAK/STAT pathway. GSTO1 N-glycosylation mutations downregulated p-JAK and p-STAT3 in CC cells, with suppression of CC proliferation, invasion, and migration noted after treatment with JAK/STAT3 inhibitors. CC cells harboring GSTO1 N-glycosylation mutants were also treated with a JAK/STAT3 signaling activator. JAK/STAT3 activation counteracted the inhibitory effect of defective GSTO1 N-glycosylation on the malignant behavior of CC cells. Taken together, activation of JAK/STAT3 signaling promoted the proliferation and metastasis of CC cells, whereas GSTO1 N-glycosylation promoted malignancy via regulating the JAK/STAT3 pathway.

The EMT is a central mechanism of tumor metastasis, referring to the transformation of epithelial cells into a mesenchymal phenotype ^{[43, 44]}. EMT induces tumor metastasis by promoting reorganization of tumor metabolism and disrupting the extracellular matrix ^{[45, 46]}. It is regulated by several signaling pathways, such as Notch ^[47], PI3K/AKT ^{[48, 49]}, and Hippo ^[50]. Common features of EMT include the conversion of E-cadherin to N-cadherin and modification of the cytoskeleton ^[43]. In this study, we found that GSTO1 N-glycosylation-targeted mutations significantly inhibited the expression of N-cadherin and suppressed the EMT. The JAK/STAT3 inhibitor WP1066 suppressed the expression of N-cadherin and Vimentin, suggesting that JAK/STAT3 signaling influences the invasion and migration of CC cells by regulating EMT. We thus propose that GSTO1 N-glycosylation activates the JAK/STAT3 pathway to regulate the EMT and promote CC migration and invasion.

In the present study, we found that GSTO1 downregulation inhibits the proliferation and metastasis of CC cells. Further, we identified three N-glycosylation sites in GSTO1, the mutation of which significantly inhibited CC cell proliferation and metastasis. Mechanistically, we show that GSTO1 N-glycosylation stimulates JAK/STAT3 signaling to promote the EMT (Fig. 8). However, this study has some limitations. We detected changes in the JAK/STAT3 signaling pathway and EMT-related factors in cells harboring GSTO1 N-glycosylation mutants. However, the regulatory mechanism of JAK/STAT3 signaling within the EMT remains to be further explored.

In summary, the current findings support GSTO1 as a biomarker for CC prognosis, highlighting the importance of its N-glycosylation in CC progression mediated via the JAK/STAT3-regulated EMT.

cervical cancer

GSTO1

glutathione S-transferase omega 1

GST

glutathione S-transferase

IHC

immunohistochemical

overall survival

PFS progression free survival

EMT

epithelial-mesenchymal transition

DSI

cervical deep stromal invasion

LVSI

lymph vascular space invasion

LNM

lymph node metastasis

qRT-PCR

quantitative real-time-polymerase chain reaction

PNGase F

recombinant glycosidase

Endo H

endoglycosidase H.

Ethics approval and consent to participate

This study was approved by the Medical Ethics Committee of the First Affiliated Hospital of Shihezi University (KJ2020-065-01 and A2020-115-01).

Consent for publication

Not applicable

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare that we have no conflicts of interest with the content of this article.

Funding

This work was supported by the National Natural Science Foundation of China (grant number 82072893). Corps key areas of science and technology research programme (2023AB055). Tianshan Ying Talents Leading Talents in Science and Technology Innovation-High-level Leading Talents (CZ001201). The Key R & D Program of Xinjiang Uyghur Autonomous Region (2022B03018-2).

Authors’ contributions

All experimental results of this paper were jointly completed by Panpan Yu, Zouyu Zhao and Qianyu Sun, and the writing of this paper was completed by Panpan Yu. Therefore, Panpan Yu, Zouyu Zhao and Qianyu Sun are regarded as the co-first authors. Ping Yang conceived these ideas. Panpan Yu and Ping Yang designed the experiment. Panpan Yu, Zouyu Zhao carried out the experiment. The data were analyzed by Panpan Yu, Zouyu Zhao and Qianyu Sun, Bowen Diao, Chongfeng Sun, Yan Wang, Hui Qiao and Hong Li. Panpan Yu provide key materials. Panpan Yu wrote the manuscript. Ping Yang oversaw the study. All authors have read and approved the paper for publication.

Acknowledgments

We thank all recruited patients and the staff of the Department of Gynecology and the staff of the Department of Pathology, First Affiliated Hospital, Shihezi University, Shihezi, China for sample collection.

Zhao M, Pu X, Ma G, Zhang M, Luo L, Gu R, et al. Quality of life in patients with cervical cancer between the Han nationality and ethnic minorities in the Yunnan Province of China. BMC Womens Health. 2023;23(1):217.
Arbyn M, Weiderpass E, Bruni L, de Sanjosé S, Saraiya M, Ferlay J, et al. Estimates of incidence and mortality of cervical cancer in 2018: a worldwide analysis. Lancet Glob Health. 2020;8(2):e191–203.
Fernandes A, Viveros-Carreño D, Hoegl J, Ávila M, Pareja R. Human papillomavirus-independent cervical cancer. Int J Gynecol Cancer. 2022;32(1):1–7.
Sung H, Ferlay J, Siegel R, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209–49.
Xu X, Shen L, Li W, Liu X, Yang P, Cai J. ITGA5 promotes tumor angiogenesis in cervical cancer. Cancer Med. 2023;12(10):11983–99.
Bangarh R, Khatana C, Kaur S, Sharma A, Kaushal A, Siwal S, et al. Aberrant protein glycosylation: Implications on diagnosis and Immunotherapy. Biotechnol Adv. 2023;66:108149.
Čaval T, Alisson-Silva F, Schwarz F. Roles of glycosylation at the cancer cell surface: opportunities for large scale glycoproteomics. Theranostics. 2023;13(8):2605–15.
Hu M, Zhang R, Yang J, Zhao C, Liu W, Huang Y, et al. The role of N-glycosylation modification in the pathogenesis of liver cancer. Cell Death Dis. 2023;14(3):222.
Krug J, Rodrian G, Petter K, Yang H, Khoziainova S, Guo W, et al. Maschauer, N-glycosylation regulates intrinsic IFN-γ resistance in colorectal cancer: implications for immunotherapy. Gastroenterology. 2023;164(3):392–406. e5.
Grzesik K, Janik M, Hoja-Łukowicz D. The hidden potential of glycomarkers: Glycosylation studies in the service of cancer diagnosis and treatment. Biophys Acta Rev Cancer. 2023;1878(3):188889.
Zhang Y, Cao K, Niu Q, Deng J, Zhao L, Khalil M, et al. Alpha-class glutathione S-transferases involved in the detoxification of aflatoxin B1 in ducklings. Food Chem Toxicol. 2023;174:113682.
Wang L, Yue H, Peng X, Zhang S. GSTO1 regards as a meritorious regulator in cutaneous malignant melanoma cells. Mol Cell Probes. 2019;48:101449.
Li J, Peng L, Li H, Cai Y, Yao P, Chen Q, et al. Network and pathway-based analysis of candidate genes associated with esophageal adenocarcinoma. J Gastrointest Oncol. 2023;14(1):40–53.
Liu W, Zhang Z, Zhang ZM, Hao P, Ding K, Li Z. Integrated phenotypic screening and activity-based protein profiling to reveal potential therapy targets of pancreatic cancer. Chem Commun (Camb). 2019;55(11):1596–9.
Radic T, Coric V, Bukumiric Z, Pljesa-Ercegovac M, Djukic T, Avramovic N et al. GSTO1* CC genotype (rs4925) predicts shorter survival in clear cell renal cell carcinoma male patients. Cancers (Basel). 2019; 11(12): 2038.
Wang DP, Wu LH, Li R, He N, Zhang QY, Zhao CY, et al. A novel aldisine derivative exhibits potential antitumor effects by targeting JAK/STAT3 signaling. Mar Drugs. 2023;21(4):218.
Shen M, Xu Z, Xu W, Jiang K, Zhang F, Ding Q, et al. Inhibition of ATM reverses EMT and decreases metastatic potential of cisplatin-resistant lung cancer cells through JAK/STAT3/PD-L1 pathway. J Exp Clin Cancer Res. 2019;38(1):149.
Ramesh V, Brabletz T, Ceppi P. Targeting EMT in cancer with repurposed metabolic inhibitors. Trends Cancer. 2020;6(11):942–50.
Wang K, Zhang FL, Jia W. Glutathione S–transferase ω 1 promotes the proliferation, migration and invasion, and inhibits the apoptosis of non–small cell lung cancer cells, via the JAK/STAT3 signaling pathway. Mol Med Rep. 2021;23(1):71.
Zhang J, Jia J, Zhao L, Li X, Xie Q, Chen X, et al. Down-regulation of microRNA‐9 leads to activation of IL‐6/Jak/STAT3 pathway through directly targeting IL‐6 in HeLa cell. Mol Carcinog. 2016;55(5):732–42.
Siegel RL, Miller KD, Wagle NS, Jemal A. Cancer statistics, 2023. CA Cancer J Clin. 2023;73(1):17–48.
Lu H, Li L, Cheng Y, Yang Z, Cao X, Zhang H, et al. Timely Estimates of 5-Year Relative Survival for Patients With Cervical Cancer: A Period Analysis Using Cancer Registry Data From Taizhou, Eastern China. Front Public Health. 2022;10:926058.
Mauricio D, Zeybek B, Tymon-Rosario J, Harold J, Santin AD. Immunotherapy in Cervical Cancer. Curr Oncol Rep. 2021;23(6):61.
Castle PE, Einstein MH, Sahasrabuddhe VV. Cervical cancer prevention and control in women living with human immunodeficiency virus. CA Cancer J Clin. 2021;71(6):505–26.
Sokulsky LA, Goggins B, Sherwin S, Eyers F, Kaiko GE, Board PG, et al. Foster, GSTO1-1 is an upstream suppressor of M2 macrophage skewing and HIF‐1α‐induced eosinophilic airway inflammation. Clin Exp Allergy. 2020;50(5):609–24.
Li S, Wang L, Xu Z, Huang Y, Xue R, Yue T, et al. ASC deglutathionylation is a checkpoint for NLRP3 inflammasome activation. J Exp Med. 2021;218(9):e20202637.
Paul S, Bhardwaj M, Kang SC. GSTO1 confers drug resistance in HCT–116 colon cancer cells through an interaction with TNFαIP3/A20. Int J Oncol. 2022;61(5):136.
Zhang W, Yang Z, Gao X, Wu Q. Advances in the discovery of novel biomarkers for cancer: spotlight on protein N-glycosylation. Biomark Med. 2020;14(11):1031–45.
Staudacher E, Van Damme EJM, Smagghe G. Glycosylation-The Most Diverse Post-Translational Modification. Biomolecules. 2022;12(9):1313.
Eichler J. Protein glycosylation. Curr Biol. 2019;29(7):R229–31.
Lin NY, Lee JJ, Chen ST, Lin JA, Lin CH, Lin HY, et al. Truncation of GalNAc-type O-glycans suppresses CD44-mediated osteoclastogenesis and bone metastasis in breast cancer. Mol Cancer Res. 2023;21(7):664–74.
Zhang C, Liu J, Chao F, Wang S, Li D, Han D, et al. Alpha-L-Fucosidase Has Diagnostic Value in Prostate Cancer With Gray-Zone PSA and Inhibits Cancer Progression via Regulating Glycosylation. Front Oncol. 2021;11:742354.
Xu C, Zhang M, Bian L, Li Y, Yao Y, Li D. N-glycosylated SGK196 suppresses the metastasis of basal-like breast cancer cells. Oncogenesis. 2020;9(1):4.
Zhang K, Yin W, Ma L, Liu Z, Li Q. HSPB8 facilitates prostate cancer progression via activating the JAK/STAT3 signaling pathway. Biochem Cell Biol. 2023;101(1):1–11.
Karuppaiya V, Annamalai A, Krishnamurthy S, Soundarapandian K. Dieckol prevents prostate cancer cell proliferation by transcriptionally attenuating JAK/STAT3 signaling pathway. Environ Toxicol. 2024;39(3):1187–96.
Liang Q, Ma D, Zhu X, Wang Z, Sun TT, Shen C, et al. RING-finger protein 6 amplification activates JAK/STAT3 pathway by modifying SHP-1 ubiquitylation and associates with poor outcome in colorectal cancer. Clin Cancer Res. 2018;24(6):1473–85.
Chen Y, Sun L, Li D, et al. Aspirin Inhibits Carcinogenesis of Intestinal Mucosal Cells in UC Mice Through Inhibiting IL-6/JAK/STAT3 Signaling Pathway and Modulating Apoptosis and Proliferation. Turk J Gastroenterol. 2022;33(9):731–42.
Deng Y, Li Y. Linderalactone inhibits human lung cancer growth by modulating the expression of apoptosis-related proteins, G2/M cell cycle arrest and inhibition of JAK/STAT signalling pathway. J BUON. 2019;24(2):566–71.
Prabhu KS, Bhat AA, Siveen KS, Kuttikrishnan S, Raza SS, Raheed T, et al. Sanguinarine mediated apoptosis in Non-Small Cell Lung Cancer via generation of reactive oxygen species and suppression of JAK/STAT pathway. Biomed Pharmacother. 2021;144:112358.
Zhang Q, Li X, Ren L, Gu X, Xiao N, Li N. OTX1 silencing suppresses ovarian cancer progression through inhibiting the JAK/STAT signaling. Tissue Cell. 2023;82:102082.
Zhu J, Li Y, Lv X. IL4I1 enhances PD-L1 expression through JAK/STAT signaling pathway in lung adenocarcinoma. Immunogenetics. 2023;75(1):17–25.
Liu ZK, Li C, Zhang RY, Wei D, Shang YK, Yong YL, et al. EYA2 suppresses the progression of hepatocellular carcinoma via SOCS3-mediated blockade of JAK/STAT signaling. Mol Cancer. 2021;20(1):79.
Manfioletti G, Fedele M. Epithelial-Mesenchymal Transition (EMT) 2021. Int J Mol Sci. 2022;23(10):5848.
Radhakrishnan K, Truong L, Carmichael CL. An unexpected role for EMT transcription factors in hematological development and malignancy. Front Immunol. 2023;14:1207360.
Wang X, Eichhorn PJA, Thiery JP. TGF-β, EMT, and resistance to anti-cancer treatment. Semin Cancer Biol. 2023;97:1–11.
Saitoh M. Transcriptional regulation of EMT transcription factors in cancer. Semin Cancer Biol. 2023;97:21–9.
Diao B, Sun C, Yu P, Zhao Z, Yang P. LAMA5 promotes cell proliferation and migration in ovarian cancer by activating Notch signaling pathway. FASEB J. 2023;37(9):e23109.
Chen H, Ma L, Yang W, Li Y, Ji Z. POLR3G promotes EMT via PI3K/AKT signaling pathway in bladder cancer. FASEB J. 2023;37(12):e23260.
Maharati A, Moghbeli M. PI3K/AKT signaling pathway as a critical regulator of epithelial-mesenchymal transition in colorectal tumor cells. Cell Commun Signal. 2023;21(1):201.
Yang L, Li J, Zang G, Song S, Sun Z, Li X, et al. Pin1/YAP pathway mediates matrix stiffness-induced epithelial–mesenchymal transition driving cervical cancer metastasis via a non‐Hippo mechanism. Bioeng Transl Med. 2022;8(1):e10375.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

N-glycosylation of GSTO1 promotes cervical cancer migration and invasion through JAK/STAT3 pathway activation

Status:

Version 1

Abstract

Figures

Introduction

Experimental Procedures

IHC

Results

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1