Clinical significance of USP7 and EZH2 in predicting prognosis of laryngeal squamous cell carcinoma and their possible functional mechanism

Abstract

Background: Laryngeal squamous cell carcinoma (LSCC) is one of the most prevalent cancer types in the world. The ubiquitin specific protease 7 (USP7), a kind of deubiquitylating enzyme, has been reported to play multifaceted roles in different tumor types. EZH2 has been found to be highly expressed in various malignantcells and high expression of EZH2 is closely related to tumor growth infiltration, lymph node involvement, clinical stage, and poor prognosis. The aim of this study was to investigate the expression and function of USP7 and EZH2 in LSCC. Case presentation: Immunohistochemical staining and histochemical staining were performed to explore the expression of USP7 and EZH2 in both LSCC tissues and adjacent normal laryngeal tissues. Chi-square test, univariate analysis, and multivariate analysis were conducted to statistically evaluate the clinical significance of USP7 and EZH2. Conclusions: USP7and EZH2 affects LSCC evolution; USP7 and EZH2 were upregulated in LSCC tissues, which can serve as independent prognostic predictors, and potential therapeutic targets for LSCC.

Introduction

Laryngeal carcinoma is one of the most common malignant tumors in otolaryngology, accounting for about 20% of head and neck tumors, 95% of which are mainly squamous cell carcinoma, mostly occurring in people over 40 years old [1]. Men’s rate is significantly higher than that of women, especially among middle-aged and elderly men aged 60-70 years [2]. 177,442 new cases worldwide each year account for 1% of cancer, and 94,713 deaths per year in 2018 [2]. Despite advances in diagnosis and treatment, the survival rate of patients with laryngeal cancer has not increased over the past 30 years, but has declined [3]. Patients with laryngeal cancer have various treatments, such as surgery, radiotherapy and chemotherapy, but the treatment effect and prognosis need to be improved [4].

Local recurrence, cervical lymph node metastasis, and poor differentiation are the main reasons for the treatment and prognosis of patients with laryngeal squamous cell carcinoma (LSCC). Early stage of laryngeal carcinoma is difficult find because it is asymptomatic. Most patients visit a doctor and get a diagnosis in an advanced stage of LSCC. Meanwhile, infiltration and metastasis have occurred, and the treatment effect is poor, which seriously endangers human health. At present, the treatment of laryngeal carcinoma includes surgery, chemotherapy, radiotherapy and comprehensive treatment including surgical resection combined with radiotherapy [1]. Surgical methods include partial laryngectomy, total laryngectomy, etc. However, the scope of the surgical incision is large, and the recovery after surgery is poor, the comorbidities are serious: the loss of laryngeal function; some patients who need to have a full laryngectomy may wear a tube for life [5]. All these cause patients psychosocial stress.

Targeted therapy is a new tumor treatment method discovered in recent years from the molecular biologic level to study the pathogenesis and treatment of tumors. It is a target for cell-related receptors, key genes and regulatory molecules. Active components such as antibodies are targeted to cancer cells for the purpose of treating tumors. After targeted therapy, if only surgery is performed, the scope of tumor invasion gets reduced and the removed tissue will be less, so part of the laryngeal function can retained which brings better surgical prognosis to patients. In order to improve the therapeutic effect of LSCC and preserve laryngeal function, exploring the relationship between tumor markers and clinicopathologic features of LSCC plays an important role in understanding the clinicopathologic features and treatment of LSCC.

USP7 (ubiquitin specific peptidase 7) is a deubiquitinating enzyme that regulates many key proteins such as tumor suppressor genes, DNA repair proteins, immune responders, viral proteins, and epigenetic regulators [6]. USP7 abnormalities can cause tumor and viral diseases, therefore, it has become a potential therapeutic target [7]. USP7, also known as herpes virus-associated ubiquitin-specific protease (HAUSP), is reported to regulate DNA damage repair [8], DNA replication, cell mitosis [9], viral infection [10]. USP7 has also been reported to be involved in tumor progression. The overexpression of USP7 has been observed in cancer, myeloma cells, neuroblastoma, breast cancer, non-small cell lung cancer, and ovarian cancer [11-15]. USP7 inhibitor can restrain progression of colorectal tumor and non-small cell lung cancer [16,17].

Enhancer of Zeste homolog 2 (EZH2), which is part of the core subunit of the polycomb repressive complex 2 (PRC2), catalyzes the transcription of its downstream target gene by catalytic methylation modification of histone H3K27 (H3K27me3) [18]. H3K27me3 always inhibits transcription, regulates gene activity at the chromosome level and it is a critical epigenetic event during tissue development and stem cell fate determination [19]. EZH2 silences the expression of multiple pathway-related genes including Wnt, thereby promoting cellular EMT, inhibiting cellular senescence and cell differentiation [20,21]. EZH2 has been found to be highly expressed in various malignant tumor cells, such as breast cancer, prostate cancer, bladder cancer, lung cancer, gastric cancer, pancreatic cancer and nasopharyngeal carcinoma [22-27]. The high expression of EZH2 is closely related to tumor growth infiltration, lymph node involvement, clinical stage and poor prognosis. Whether EZH2 has this role in LSCC is yet to be fully explored, but there is increasing evidence of the importance of EZH2 in LSCC. High EZH2 expression was shown to be associated with more malignant forms of LSCC [28].

Material and methods

Specimens

We collected specimens from all 160 patients who were diagnosed with LSCC at the First Affiliated Hospital of Bengbu Medical University, from January 2012 to January 2018. Patients underwent radical resection and peripheral mesenteric lymph node dissection. All patients were sporadic cases who had compete clinical, pathological and follow-up data (per 4 months intervals by phone, mail or out-patient follow-up) until October 2018. We excluded patients who had no complete data or history of received preoperative chemotherapy and radiotherapy. The study was approved by the ethics committee of the First Affiliated Hospital of Bengbu Medical University and performed obeying the guidance of the Declaration of Helsinki. Overall survival (OS) time was calculated from patients’ surgery to death of LSCC or last follow-up (months). Tumor differentiation grade was defined according to World Health Organization. Clinical stages were rejudged based on AJCC/UICC 8th Edition new staging rules in Oropharyngeal Squamous Cell Carcinoma [29]. Clinicopathologic characteristics are listed in Table 1.

Table 1.

Patients characteristics

Patient characteristics	Frequency (n)	Percentage (%)
Gendel
Male	112	70.0
Female	48	30.0
Age (years)
< 60	89	55.6
≥ 60	71	44.4
Site
Glottic	96	60.0
Supraglottic	60	37.5
Subglottic	4	2.5
Tumor size
< 3 cm	91	56.9
≥ 3 cm	69	43.1
Pathologic differentiation
Well	66	41.3
Moderate	57	35.6
Poor	37	23.1
Lymphatic invasion
Negative	98	61.3
Positive	62	38.8
pTNM
I + II	75	46.9
III + IV	85	53.1

Immunohistochemical analysis

Immunohistochemical (IHC) staining was performed to explore the expression of USP7 and EZH2 in both LSCC tissues and adjacent normal larynx tissues. Immunohistochemical analysis was performed with the Elivision^TM Plus detection kit (Lab Vision, USA), according to the manufacturer’s instructions. Collected tissues were fixed in 10% buffered formalin, embedded in paraffin, and then continuously sectioned (thickness, 4 μm). Then, all sections were deparaffinized and dehydrated with xylene and graded. Subsequently, all sections were washed for 10 min in phosphate-buffered saline (PBS, pH 7.2), incubated in 3% hydrogen peroxide for at room temperature 10 min to block endogenous peroxidase activity, placed in citrate buffer (pH 6.0), and heated to 95°C for 30 min for antigen repair. After three washes with PBS, all sections were blocked with goat serum at room temperature for 20 min, followed by incubation with rabbit monoclonal antibody against primary antibodies USP7 (ab4080, Abcam, USA) and EZH2 (ab186006, Abcam, USA) at 4°C overnight. Finally, all sections were counterstained with hematoxylin, dehydrated, air-dried, and mounted.

Evaluation of staining

The staining results were interpreted by two independent pathologists who were blinded to the clinical data and were judged by semi-quantitative points. To overcome the intra-tumoral heterogeneity of antigen expression, 10 visual fields from different areas of each tumor were examined. In cases of disagreement between the observers, consensus was achieved by re-examining the sections. Staining was scored according to the intensity and extent, < 4 as negative expression, ≥ 4 as positive expression. The staining intensity score was graded for four stages as follows: 0, none; 1, weak; 2, moderate; and 3, strong. The extent of positive staining was graded as follows: 1, ≤ 10%; 2, 11%-50%; 3, 51%-75%; and 4, > 75%.

Statistics

Chi-square test, univariate analysis, and multivariate analysis were conducted to statistically evaluate the clinical significance of USP7 and EZH2 in LSCC. The SPSS software package (version 24.0, IBM, USA) was used for the statistical analysis. The associations between protein expression and clinicopathologic parameters were evaluated by chi-square tests. Kaplan-Meier method was used to generate survival curves, and the differences were compared by log-rank test. A Cox multivariate proportional hazard regression model was introduced to determine the independent effect on OS. Beta coefficients and 95% confidence intervals (CI) were used for analysis. Differences were considered significant if P < 0.05.

Results

Expression of USP7 and EZH2 in different grade of LSCC

We selected typical histochemical staining of LSCC on high-differentiation (Figure 1A and 1D), moderate differentiation (Figure 1B and 1E), and low differentiation (Figure 1C and 1F). Representative IHC result showed the negative protein expression of USP7 in normal laryngeal tissues (Figure 2A). Representative IHC result showed the high protein level of USP7 in LSCC tissues, which predominately located in the nucleus and the expression level in high-differentiation, moderate differentiation, and low differentiation LSCC is increasing (Figure 2C, 2E and 2G). We investigated the expression profile and subcellular localization of EZH2 in samples of benign and malignant LSCC tissue and found abundant expression in the nucleus and cytoplasm of highly differentiated LSCC epithelial cells (Figure 2D). EZH2 is over-expressed in the nucleus of low differentiated LSCC cells (Figure 2H). Moderate differentiation LSCC EZH2 expression is between high and low (Figure 2F). We observed substantially increased levels of EZH2 expression in the nucleus and cytoplasm of malignant LSCC as compared with normal tissue (Figure 2B). Moreover, a positive correlation was observed between nuclear and cytoplasmic EZH2 staining intensity in malignant LSCC epithelial cells.

Figure 1.

Presentation of histochemical staining of high-differentiation, moderate differentiation, and low differentiation. A: Highly differentiated squamous cell carcinoma is similar to normal squamous epithelium, but a variety of keratinized forms of keratinized beads are common, and infiltration occurs from the basal layer (100 magnification); B: In moderate differentiation, obvious nuclear polymorphism and less keratinization, irregular growth, lack of polarity, and a significant increase in the nucleoplasmic ratio, and the appearance of obvious eosinophilic nucleoli and mitotic figures (100 magnification); C: Poorly differentiated squamous cell carcinoma is dominated by immature cells, with a large number of normal or abnormal mitotic figures, and very little keratinization (100 magnification); D: High-differentiation of LSCC corresponding to A (400 magnification); E: Eoderate differentiation of LSCC corresponding to B (400 magnification); F: Low differentiation of LSCC corresponding to C (400 magnification). In D-F we can see a clear trend: the nucleolus becomes larger, the nucleoplasmic ratio increases, the mitotic figures are more numerous, and keratinization becomes less or even disappears.

Figure 2.

The presentation of immunochemical staining of normal epidemal laryneal tissue and LSCC. A: Negative staining of USP7 in laryneal epithelium. The cytoplasm in the granular layer is light yellow (400 magnification); B: Negative staining of EZH2 in squamous epithelium of the larynx, the nucleus in the granular layer is light yellow (400 magnification); C: Negative staining of USP7 in hyperplastic squamous cells. There is flaky expression in the cytoplasm, as well as staining of individual nuclei (400 magnification); D: Negative staining of EZH2 in hyperplastic squamous cell, mainly expressed in the nucleus and cytoplasm, but the degree of staining is light (400 magnification); E: Positive staining of USP7 in moderate differentiation, diffusely expressed in the cytoplasm and nucleus with brown-staining (400 magnification); F: Positive staining of EZH2 in moderate differentiation, diffusely expressed in the nucleus with brown staining (400 magnification); G: Positive staining of USP7 in poorly differentiation, full expression in the cytoplasm and nucleus, dark brown (400 magnification); H: Positive staining of EZH2 in poor differentiation, full expression in nucleus, dark-brown staining (400 magnification).

Correlations among USP7, EZH2 and clinicopathological characteristics

Although USP7 showed a generally increased level in LSCC tissue, we classified patients into a low-USP7 group and a high-USP7 group based on the IHC scores to better investigate its clinical effects. According to the chi-square tests, high USP7 level was associated with pathologic differentiation (P < 0.001), lymphatic invasion (P < 0.001), and advanced TNM stage (P < 0.001) (Table 2). The correlations between high USP7 level with unfavorable clinicopathologic characteristics indicated that USP7 may promote tumor progression of LSCC. High EZH2 level was associated with pathologic differentiation (P < 0.05) and advanced TNM stage (P < 0.001) (Table 2). The correlations between high EZH2 level with unfavorable clinicopathologic characteristics indicated that EZH2 may promote tumor progression of LSCC. Furthermore, correlation between USP7 and EZH2 showed a positive correlation with these two factors expression (r = 43.361, P < 0.001) (Table 3).

Table 2.

Relationship between expression of USP7 and EZH2 and clinicopathologic characteristics of laryngeal squamous cell carcinoma

Characteristic	USP7		P value	EZH2		P value
	Negative	Positive		Negative	Positive
Gender
Male	27	85	0.297	33	79	0.395
Female	8	40		11	37
Age (years)
< 60	24	65	0.081	31	58	0.020
≥ 60	11	60		13	58
Site
Glottic	17	79	0.163	28	68	0.846
Supraglottic	16	44		15	45
Subglottic	2	2		1	3
Tumor size
< 3 cm	24	67	0.114	30	61	0.075
≥ 3 cm	11	58		14	55
Pathologic differentiation
Well	25	41	< 0.001	24	42	0.020
Moderate	7	50		16	31
Poor	3	34		4	33
Lymphatic invasion
Negative	31	67	< 0.001	32	66	0.066
Positive	4	58		12	50
pTNM
I + II	30	45	< 0.001	31	44	< 0.001
III + IV	5	80		13	72

Table 3.

Correlation between USP7 and EZH2

Variable	USP7		r	P
	Negative	Positive
EZH2
Negative	25	10	43.361	< 0.001
Positive	19	106

Prognostic effect of USP7 and EZH2 for LSCC patients

We plotted the survival curves using Kaplan-Meier methods based on different variables. Importantly, high USP7 level was correlated with unfavorable clinical outcomes of ESCC patients (Figure 3C). The mean overall survival time of patients in low-USP7 group was 62.37 ± 20.196 months, while only 36.24 ± 16.334 months in the patients with higher USP7 levels (Table 4, P < 0.001). High EZH2 level was correlated with unfavorable clinical outcomes of LSCC patients (Figure 3D). The mean overall survival time of patients in low-EZH2 group was 54.27 ± 22.573 months, while it was only 37.28 ± 17.322 months in the patients with higher EZH2 levels (Table 4, P < 0.001).

Figure 3.

Kaplan-Meier analysis of the survival rate of patients with LSCC. The y-axis represents the percentage of patients; the x-axis, their survival time in months. A: Overall survival of patients in relationship to lymphatic invasion (LNM) (log-rank = 30.432, P < 0.001); B: Overall survival of patients in relationship to pathologic Tumor/lymph node/metastasis staging (pTNM) (log-rank = 8.256, P = 0.004); C: Overall survival of all patients in relationship to USP7 expression (log-rank = 27.772, P < 0.001); D: Overall survival of all patients in relationship to EZH2 expression (log-rank = 17.338, P < 0.001).

Table 4.

Results of univariate analyses of overall survival (OS) time

Variable	n	Mean OS (months)	Log-rank	P-value
USP7
Negative	35	62.37 ± 20.196	27.772	< 0.001
Positive	125	36.24 ± 16.334
EZH2
Negative	44	54.27 ± 22.573	17.338	< 0.001
Positive	116	37.28 ± 17.322
Gender
Male	112	41.96 ± 20.193	0.081	0.776
Female	48	41.94 ± 20.816
Age (years)
< 60	89	43.30 ± 21.007	0.295	0.587
≥ 60	71	40.27 ± 19.430
Site
Glottic	96	38.93 ± 19.362	1.649	0.438
Supraglottic	60	45.98 ± 21.144
Subglottic	4	54.25 ± 19.805
Tumor size
< 3 cm	91	43.12 ± 20.971	0.074	0.786
≥ 3 cm	69	40.42 ± 19.463
Pathologic differentiation
Well	66	46.23 ± 23.076	5.036	0.081
Moderate	57	41.42 ± 17.675
Poor	37	35.16 ± 17.150
Lymphatic invasion
Negative	98	43.57 ± 21.888	30.432	< 0.001
Positive	62	39.40 ± 17.413
pTNM
I + II	75	43.20 ± 23.785	8.256	0.004
III + IV	85	40.86 ± 16.741

Multivariate analysis was performed to further identify independent prognostic factors of LSCC. The pTNM showed no significantly independent effect (Table 5, P = 0.234). However, lymphatic invasion was independently correlated with a poor overall survival (HR = 2.605, 95% CI 1.534-4.423, P < 0.001). Furthermore, low USP7 expression was an independent protective factor for the clinical outcome (HR = 4.338, 95% CI 1.639-11.477, P = 0.003). Low EZH2 expression was also an independent protective factor for the clinical outcomes (HR = 2.130, 95% CI 1.063-4.269, P = 0.033) (Table 5).

Table 5.

Multivariate analysis of factors affecting overall survival

Variables	Categories	P value	Multivariate analysis
			Harzard Ratio	95% CI
Lymphatic invasion	Negative/Positive	< 0.001	2.605	1.534-4.423
pTNM	I + II/III + IV	0.234	0.711	0.406-1.246
USP7	Negative/Positive	0.003	4.338	1.639-11.477
EZH2	Negative/Positive	0.033	2.130	1.063-4.269

Discussion

Laryngeal carcinoma is one of the most common malignant tumors in otolaryngology. Most patients get diagnosed at advanced stage when infiltration and metastasis have occurred, and the prognosis is poor [4]. The etiology and pathogenesis of laryngeal cancer are not fully understood, from epidemiology, smoking, drinking, air pollution, contact reflexes, human papillomavirus infection, EB virus infection, gastroesophageal reflux, bile reflux and sex hormones are closely related to the occurrence and development of LSCC [30]. From molecular biology, it is believed that the occurrence of laryngeal cancer is related to the activation of oncogenes such as Bcl-2, c-Myc, EGFR and the inactivation of tumor suppressor genes such as p53, Rb and p21 [31-34]. In order to improve the therapeutic effect of laryngeal cancer and the preservation of laryngeal function, exploring the relationship between tumor markers and clinicopathologic features of laryngeal cancer plays an important role in treatment.

Ubiquitination modifications can regulate almost all physiologic or pathologic processes. USP7/HAUSP, is a deubiquitinating enzyme that regulates many key proteins such as tumor suppressor genes, DNA repair proteins, immune responders, viral proteins, and epigenetic regulators [6,9,35]. The role of biological activity, USP7 abnormalities can cause tumor and viral diseases, therefore, it has become a potential therapeutic target [36]. Ubiquitin specific proteases (USP) are among the most studied members of the DUBs family, including more than 60 members. Current studies have found that several USP family members are closely related to tumorigenesis and metastasis, such as USP1 and glioblastoma (GBM) [37]; USP2, USP11 and breast cancer [38,39], USP7 with prostate cancer [40]; and USP4 with colorectal cancer [41].

In view of the important role of EZH2 protein in tumorigenesis, it can be used as a treatment for tumors. A potential rough point provides a new research direction for the precise treatment of tumor diseases. However, the exact mechanism that plays a role in the growth and metastasis of EZH2 protein tumors remains the focus of future research. Expression array studies in lymphoma cells treated with EZH2 inhibitors have shown mostly increases in gene expression [42,43], as would be expected given its silencing role. What is the relationship between EZH2-mediated histone methylation and DNA methylation and will targeting one be sufficient to overcome tumor suppressor gene silencing? Further understanding of this process may help guide combination treatments with EZH2 inhibitors.

In our study high USP7 level was correlated with unfavorable clinical outcomes of ESCC patients and high EZH2 level was also correlated with unfavorable clinical outcomes of LSCC patients. Both USP7 and EZH2 level were associated with pathologic differentiation (P < 0.001). Furthermore, correlation between USP7 and EZH2 showed a positive correlation with these two factors expression. USP7 and EZH2 may promote tumor progression of LSCC.

Studies have reported that USP7 has many substrates involved in the development of cancer, the most common of which is the USP7-MDM2-P53 molecular axis [44]. Another important substrate of USP7 is the PTEN (Phosphatase and tensin homolog) tumor suppressor, whose deubiquitination will prevent protein degradation [45,46]. However, the clinical significance of USP7 seems distinct among various tumor types. For example, high expression of USP7 was correlated with poor prognosis in lung squamous cell carcinoma and large cell carcinoma [47]. Initial studies have found that USP7 can interact with P53 and stabilize its protein expression, but further study found that USP7 preferentially forms a stable complex with MDM2 in cells, which promotes the degradation of P53 protein [48]. The high expression of USP7 in tumor cells is mainly due to the increased stability of MDM2 and promotes tumor development. The regulation of p53 by USP7 is very complex. On the one hand, USP7 deubiquitinates p53, which prevents p53 from being deubiquitinated and is degraded by the proteasome, thereby increasing the intracellular level of p53. On the other hand, USP7 can also deubiquitinate the negative regulatory proteins of p53, such as MDM2 and MDM4, and increase the intracellular level of p53 negative regulatory protein, which leads to a decrease in the intracellular level of p53 and a decrease in activity.

EZH2 can play a role in tumor invasion and metastasis by affecting cytoskeletal polymerization independent of its histone methyltransferase activity [49]. P53 alteration may be involved in dysregulated EZH2 expression, and aberrant expression of EZH2 may play a role in carcinogenesis of oral squamous cell carcinoma [50]. P53 can inhibit the promoter of EZH2 and inhibit the overexpression of EZH2. On the other hand, low expression of EZH2 can effectively prolong the G2/M phase of the cell cycle and impede cell proliferation [51].

Targeted therapy has the advantage of localization. It has a higher concentration in the tumor site and has a longer retention time. It can directly regulate tumor cells, inhibit the growth of tumor cells, reduce volume of the tumor, and reduce the selection of other chemotherapy drugs. In addition to killing tumor cells, it can also protect other normal cells in the body from obvious toxic side effects. Regarding the study of small molecule inhibitors of USP7 and in the field of anti-tumor research, many companies have launched USP7 drug development program [17,52]. In recent years, USP7’s new selective P5091 series of small molecule inhibitors have been identified through a proprietary screening platform, which has the characteristics of stabilizing p53 expression, inhibiting cancer cell proliferation and activity in disease cell models [13,53].

Whether USP7 and EZH2 promote the invasion and metastasis of LSCC through the interaction mechanism of p53, we still need a more complete experimental system to verify, and the specific mechanism is waiting for us to discover.

Conclusions

Our findings imply that USP7and EZH2 affects LSCC evolution; USP7 and EZH2 are upregulated in LSCC tissues, which can serve as an independent prognostic predictor. USP7 and EZH2 are both target biomarkers of LSCC and provide the possibility for developing inhibitors to treat LSCC.

Disclosure of conflict of interest

None.