Sage Journals: Discover world-class research

Abstract

Tumor necrosis factor receptor–associated factor 1, an adaptor protein of tumor necrosis factor 2, is involved in classical nuclear factor (NF)-κB activation and lymphocyte recruitment. However, less is known about the expression and association of tumor necrosis factor receptor–associated factor 1 with cancer stem cell markers in oral squamous cell carcinoma. This study aimed to investigate the expression of tumor necrosis factor receptor–associated factor 1 and stem cell characteristic markers (lin28 homolog B, B cell-specific Moloney murine leukemia virus integration site 1, and aldehyde dehydrogenase 1) in oral squamous cell carcinoma and analyze their relations. Paraffin-embedded tissues of 78 oral squamous cell carcinomas, 39 normal oral mucosa, and 12 oral dysplasia tissues were employed in tissue microarrays, and the expression of tumor necrosis factor receptor–associated factor 1, B cell-specific Moloney murine leukemia virus integration site 1, aldehyde dehydrogenase 1, and lin28 homolog B was measured by immunohistostaining and digital pathological analysis. The expression of tumor necrosis factor receptor–associated factor 1 was higher in the oral squamous cell carcinoma group as compared with the expression in the oral mucosa (p < 0.01) and oral dysplasia (p < 0.001) groups. In addition, the expression of tumor necrosis factor receptor–associated factor 1 was associated with those of B cell-specific Moloney murine leukemia virus integration site 1, aldehyde dehydrogenase 1, and lin28 homolog B (p = 0.032, r² = 0.109; p < 0.0001, r² = 0.64; and p < 0.001, r² = 0.16) in oral squamous cell carcinoma. The patient survival rate was lower in the highly expressed tumor necrosis factor receptor–associated factor 1 group, although the difference was not significant. The clustering analysis showed that tumor necrosis factor receptor–associated factor 1 was most related to aldehyde dehydrogenase 1. These findings suggest that tumor necrosis factor receptor–associated factor 1 has potential direct/indirect regulations with the cancer stem cell markers in oral squamous cell carcinoma, which may help in further analysis of the cancer stem cell characteristics.

Keywords

Tumor necrosis factor receptor–associated factor 1 cancer stem cell head neck cancer inflammation

Introduction

Oral squamous cell carcinoma (OSCC) is a common malignant tumor of the head and neck. In the last 50 years, its clinical survival rate is less than 50% even after treating with systematic combined therapy, including surgery, chemotherapy, and radiotherapy.^1,2 Hence, the molecular factors that play important roles in OSCC carcinogenesis, progression, and targeted therapy need to be identified. Recent evidence has shown that chronic inflammation and mutation play important roles in cancer development and cancer cell origin.³

Tumor necrosis factor receptor–associated factors (TRAFs) are adapter proteins that connect several signaling pathways and are involved in nuclear factor (NF)-κB and interferon regulatory factor (IRF) signaling activation. TRAF1 is different from six other TRAFs. It has no ring structure and no rearranged zinc finger area.⁴ By interacting with downstream signaling molecules, TRAFs are involved in the activation of NF-κB signaling. TRAF1 expression is upregulated in nasopharyngeal carcinoma and Hodgkin’s lymphoma.^5,6 However, the expression of TRAF1 in OSCC remains to be determined.

Lin28 homolog B (Lin28B) is an RNA-binding protein and a paralog of Lin28A in human species. When combined with Nanog, Oct4, and Sox2, Lin28B can induce human somatic cells into pluripotent cells; when combined with let-7 signaling, Lin28B plays an important role in the regulation of key cancer stem–like properties in OSCC.^7–9 Lin28B expression is upregulated in OSCC, which can promote oral cancer cell migration, invasion, colony formation, epithelial–mesenchymal transition (EMT), and in vivo proliferation.¹⁰ B cell-specific Moloney murine leukemia virus integration site 1 (BMI-1), a key component of the PcG family, participates in DNA repair, chromosome reconstruction, gene translation, and cell apoptosis. BMI-1 is highly expressed in several primary malignant types of cancer and lowly expressed in normal tissues;¹¹ in specific, BMI-1 is highly expressed in OSCC and predicts poor progression and clinical outcome.^12,13 Aldehyde dehydrogenase 1 (ALDH1), a member of the ALDH family, can catalyze the conversion of acetaldehyde to acetic acid and is highly expressed in cancer stem cells.^14–16 Current studies have shown that ALDH1 is a cancer stem cell marker in lung, pancreatic, prostate, and head and neck cancers.¹⁷

To date, no study has performed a comprehensive analysis of the expression of TRAF1 in human OSCC and its associations with Lin28B, BMI-1, and ALDH1 expression in OSCC and patients’ clinical outcome. In this study, we examined the expression of TRAF1 in OSCC and its associations with Lin28B, ALDH1, and BMI-1 in OSCC by employing tissue microarrays. We also analyzed their roles in the prognosis of OSCC through relevance analysis between pathologic features and clinical outcome.

Methods and materials

Ethics statement

This study was approved by the Ethics Committee of the School and Hospital of Stomatology, Wuhan University, and all specimens were processed in accordance with the World Medical Association Declaration of Helsinki (Version 2008) and with the National Institutes of Health Guidelines regarding the use of human tissues. Each individual in this study had signed and approved the informed consent before surgery.

Patient samples

We determined the expression of TRAF1, Lin28B, ALDH1, and BMI-1 in primary surgically resected oral cancer tissues from the Department of Oral and Maxillofacial Surgery, School and Hospital of Stomatology, Wuhan University, from January 2008 to August 2010. These tissue samples were formalin-fixed, paraffin-embedded, and archived. All cases were diagnosed and confirmed for OSCC by pathological examination after surgery. The clinical stages of OSCC were classified in accordance with the guidelines of the Union for International Cancer Control (UICC, 2002).¹⁸ The pathological grade, tumor, node, metastasis (TNM) stage, and the survival data are presented in Table 1.

Table 1.

Clinicopathological characteristics of HNSCC used in this study.

Case	Grade	TNM	Alive or dead	Survival time (months)
1	I	T2N0M0	Dead	10
2	I	T1N1M0	Alive	42
3	I	T1N1M0	Alive	45
4	I	T2N0M0	Alive	48
5	I	T2N0M0	Alive	50
6	I	T1N0M0	Alive	56
7	I	T2N1M0	Alive	55
8	I	T1N0M0	Alive	48
9	I	T2N0M0	Alive	39
10	I	T2N1M0	Alive	37
11	I	T3N0M0	Alive	36
12	I	T1N0M0	Dead	6
13	I	T2N0M0	Alive	34
14	I	T1N0M0	Alive	34
15	II	T2N0M0	Alive	44
16	II	T3N1M0	Dead	9
17	II	T3N0M0	Dead	12
18	II	T2N1M0	Dead	18
19	II	T1N0M0	Alive	56
20	II	T3N0M0	Alive	51
21	II	T3N0M0	Alive	53
22	II	T2N0M0	Alive	51
23	II	T2N1M0	Dead	6
24	II	T1N1M0	Alive	48
25	II	T2N1M0	Alive	65
26	II	T2N1M0	Dead	8
27	II	T2N0M0	Alive	37
28	II	T3N0M0	Alive	48
29	II	T2N1M0	Alive	42
30	II	T2N1M0	Dead	17
31	II	T2N1M0	Dead	10
32	II	T2N0M0	Dead	7
33	II	T2N0M0	Alive	53
34	II	T2N0M0	Alive	54
35	II	T3N1M0	Dead	30
36	II	T2N0M0	Dead	16
37	II	T2N1M0	Dead	17
38	II	T2N0M0	Dead	7
39	II	T2N0M0	Dead	7
40	II	T3N1M0	Alive	18
41	II	T2N1M0	Dead	27
42	II	T3N1M0	Alive	32
43	II	T3N0M0	Alive	35
44	II	T2N0M0	Alive	34
45	II	T1N0M0	Dead	3
46	II	T1N0M0	Alive	34
47	II	T2N1M0	Alive	33
48	II	T2N2M0	Dead	19
49	II	T2N0M0	Dead	18
50	II	T2N0M0	Alive	33
51	II	T2N0M0	Alive	33
52	II	T1N0M0	Alive	32
53	II	T2N0M0	Alive	32
54	II	T3N0M0	Dead	21
55	II	T2N1M0	Dead	29
56	II	T2N0M0	Alive	34
57	II	T2N0M0	Dead	23
58	II	T1N0M0	Alive	34
59	II	T2N1M0	Alive	34
60	II	T2N0M0	Dead	9
61	II	T2N0M0	Dead	14
62	II	T2N0M0	Alive	33
63	II	T2N1M0	Alive	36
64	II	T2N0M0	Alive	34
65	II	T1N0M0	Dead	34
66	II	T2N0M0	Dead	22
67	II	T2N0M0	Alive	34
68	II	T2N1M0	Alive	10
69	II	T2N1M0	Alive	35
70	II	T3N1M0	Alive	36
71	III	T3N0M0	Alive	35
72	III	T2N1M0	Alive	34
73	III	T2N1M0	Alive	33
74	III	T2N0M0	Alive	33
75	III	T2N0M0	Alive	33
76	III	T2N0M0	Dead	12
77	III	T1N0M0	Alive	35
78	III	T2N1M0	Dead	6

HNSCC: head and neck squamous cell carcinoma; TNM: tumor, node, metastasis.

Tissue microarrays

Custom-made tissue microarrays (T12-412 TMA and T12-412 TMA2) were constructed by selecting OSCC specimens from above paraffin-embedded tissues, including 78 OSCCs (eluted tissue dots were excluded), 39 normal tissues, and 12 dysplastic tissues. Clinicopathologic information, including T category, lymph node metastasis, TNM stage, histological grade, and follow-up information was available for all specimens.

Immunohistochemical staining

Immunohistochemistry was performed using the streptavidin–peroxidase method.^19,20 Immunohistochemical studies for tissue arrays were performed using the following antibodies: Lin28B (1:200) from Proteintech (Chicago, USA) and Rabbit anti-Human Polyclonal TRAF1 (1:100), Rabbit anti-Human Polyclonal BMI-1 (1:800), and Rabbit anti-Human Polyclonal ALDH1 (1:800) from GeneTex (Irvine, USA). Before using the gradient concentration, it was first tested in paraffin tissue sections. Detailed tissue-section procedures, including antigen retrieval, primary antibody conjugation, secondary antibody conjugation, and immunostaining visualization, were conducted as previously described.¹⁰ In brief, the tissue sections were deparaffinized, hydrated, antigen retrieved, conjugated with primary and then secondary antibodies, and finally visualized by diaminobenzidine and hematoxylin.

Staining assessment

As described earlier, high-resolution tissue array staining images were automatically scanned by the Aperio ScanScope CS scanner (Vista, CA, USA).²¹ The staining spots for TRAF1, Lin28B, ALDH1, and BMI-1 were outlined and then analyzed using Aperio Quantification software (Vista, CA, USA) (Version 9.1). Then, the staining intensities of the nucleus and membrane were calculated as previously described.¹⁹ With these methods, each tissue spot and staining was scored. Finally, we used the scores for subsequent analysis. Using the median score values of TRAF1, the expression was divided into TRAF1-high and TRAF1-low groups, and this cutoff method was also used in the group division for Lin28B, ALDH1, and BMI-1.^10,19,21

Hierarchical clustering

The staining scores were converted into the database (Organized Notepad pattern) that Cluster 3.0 needed, and then the data were adjusted in Cluster 3.0 by log transformation and generated scaled value data.²² Then, the result was visualized in TreeView (Eisen et al., 1998) (Version 1.1.6), with biomarkers shown on the vertical axis and specimens on the horizontal axis. The expression levels for each sample were aggregated by the Euclidean distance.²³

Statistical analysis

Most of the statistical analyses were performed using GraphPad Prism (GraphPad Software, San Diego, USA) (Version 5.01). One-way analysis of variance (ANOVA), followed by the Tukey test, was used for group difference comparison; log-rank test was used for survival rate difference; and linear regression analysis was used for expression level associations. Clustering analysis was used for the protein expression scattering and associations. Statistical significance was considered at p < 0.05.

Results

Increased expression of TRAF1, ALDH1, BMI-1, and Lin28B in OSCC

TRAF1 was mainly expressed in the cell cytoplasm in OSCC; its expression was significantly higher in OSCC (n = 78) specimens than in normal oral mucosa (OM; Figure 1(a), (b), and (i); n = 39, p < 0.01) and in the dysplasia tissues (p < 0.001). However, the difference between the mucosa and dysplasia tissues was not significant (Figure 1(i)). BMI-1 was mainly expressed in the cell nucleus; its expression was similar in the normal mucosal, dysplastic, and OSCC groups (Figure 1(c), (d), and (j)). ALDH1 was mainly expressed in the cytoplasm; its expression was significantly higher in OSCC and dysplasia tissues (Figure 1(e), (f), and (k); p < 0.001) than in normal OM tissues (Figure 1(k)). Lin28B was mainly expressed in the cytoplasm; its expression was higher in OSCC tissues than in dysplasia and normal tissues (Figure 1(g), (h), and (l); p < 0.05).

Figure 1.

Expression of (a and b) TRAF1, (c and d) BMI-1, (e and f) ALDH1, and (g and h) Lin28B in normal oral mucosa and oral squamous cell carcinoma (SCC). (a, c, e, and g) TRAF1, BMI-1, ALDH1, and Lin28B selected from serial sections of a normal oral epithelium; (b, d, f, and h) TRAF1, BMI-1, ALDH1, and Lin28B selected from serial sections of an OSCC sample. (i–l) Scattering of histoscores for protein expression in normal oral mucosa (Muc), dysplasia (Dys), and oral squamous cell carcinoma.

Expression of TRAF1, BMI-1, ALDH1, and Lin28B in OSCC analyzed by pathological grades and TNM stages

Considering that different pathological grades are often associated with different clinical outcomes, we examined the expression levels of TRAF1, BMI-1, ALDH1, and Lin28B in pathological grade I, II, and III. TRAF1 expression was similar in the three groups (Figure 2(a)). BMI-1 expression was significantly higher in the high-grade group than in the clinical-grade II group (Figure 2(b); p < 0.01), but it was higher in the grade I group than in the grade-II group (Figure 2(b); p < 0.001). Meanwhile, the difference in BMI-1 expression between grade I and III groups was not significant. ALDH1 expression was similar in the three pathological groups (Figure 2(c)). Lin28 expression level was higher in grade III than in grades I and II, the difference was not significant (Figure 2(d)).

Figure 2.

(a, b, c, and d) The top row showed expression levels for TRAF1, BMI-1, ALDH1, and Lin28B in pathological groups (grade I, II, and III). (e, f, g, and h) The lower row showed expression levels for TRAF1, BMI-1, ALDH1, and Lin28B in clinical groups (I–II stage, III–IV stage).

Different clinical stages were also involved in the patients’ final outcome. In accordance with the guidelines of American Joint Committee on Cancer (2010, 7th edition), we divided the samples into I–II and III–IV groups, and then compared their difference in protein expression levels. Interestingly, TRAF1 expression was similar in the two clinical-stage groups (Figure 2(e); p = 0.08). BMI-1 expression was lower in the III+IV group (Figure 2(f); p = 0.21), while ALDH1 (Figure 2(g); p = 0.31) and Lin28B (Figure 2(h); p = 0.47) expression levels were similar in both groups.

Survival curves of TRAF1, BMI-1, ALDH1, and Lin28B in OSCC

The expression of TRAF1 was divided into TRAF1-low and TRAF1-high groups by the median expression levels on the basis of the method used in our previous work.^10,19,21 This group sorting method was also used for the expression of BMI-1, ALDH1, and Lin28B. We found through the Kaplan–Meier analysis that patients with high TRAF1 expression had worse overall survival rates, but the difference was not significant (Figure 3(a); p = 0.42). The same findings were obtained in the expression of BMI-1 (Figure 3(b)) and Lin28 (Figure 3(d)) in oral cancer tissues (p = 0.56 and p = 0.6, respectively). The higher expression levels of ALDH1 did not indicate a worse overall survival rate than the lower expression group (Figure 3(c)).

Figure 3.

Kaplan–Meier survival curves for overall survival rates for TRAF1 low/high expression. (a) BMI-1 low/high expression. (b) ALDH1 low/high expression. (c) and Lin28B low/high expression. (d) in OSCCs.

Associations of TRAF1, Lin28B, BMI-1, and ALDH1 in OSCC

We also analyzed the associations of TRAF1, Lin28B, BMI-1, and ALDH1 proteins in OSCC by Pearson’s correlation coefficient test. Results showed that TRAF1 was associated with BMI-1 (Figure 4(a); p < 0.05, r² = 0.109), ALDH1 (Figure 4(b); p < 0.0001, r² = 0.64), and Lin28B (Figure 4(c); p < 0.001, r² = 0.16); BMI-1 was associated with ALDH1 (Figure 4(d); p < 0.0001, r² = 0.30) and Lin28B (Figure 4(e); p < 0.01, r² = 0.09); and ALDH1 was associated with Lin28B (Figure 4(f); p < 0.001, r² = 0.26). However, though the p values were less than 0.05, only the r² of TRAF1 and ALDH1 was greater than 0.5 and closer to 1. Hence, we performed the hierarchical clustering of TRAF1, Lin28B, BMI-1, and ALDH1, which showed similar result that TRAF1 was most related to ALDH1.(Figure 4(g)).

Figure 4.

(a–f) Associations of TRAF1, Lin28B, BMI-1, and ALDH1 in OSCC. (g) Analysis of potential relations among TRAF1, Lin28B, BMI-1, and ALDH1 by clustering analysis.

Discussion

Chronic inflammation is common in patients with oral cancer.²⁴ The accumulated reactive oxygen species and nitrogen intermediates that are produced by inflammatory cells can damage cell DNA and increase cell malignancy.²⁵ The cytokines and enzymes (epidermal growth factor (EGF), vascular endothelial growth factor (VEGF), fibroblast growth factor 2 (FGF2), matrix protease, cysteine cathepsin, and heparanase) produced by inflammatory cells can promote cancer development. In addition, activated NF-κB can invoke Wnt signaling and, hence, promote cell dedifferentiation, which improves the stem cell characteristics for malignant cells.²⁶ Although the exact relation between inflammation and cancer cell origin is unclear, a possibility of inflammation-induced cancer stem cell exists.²⁷ In this study, we found TRAF1, an upstream molecular factor of NF-κB associated with the cancer stem cell marker of BMI-1, ALDH1, and Lin28B in oral cancer may have a potential relation in their molecular mechanism functions.

TRAF1 was originally discovered in 1994,²⁸ and the human homolog type was found a year later.²⁹ Since then, it has been established that TRAF1 is a unique type of TRAF because it lacks the ring domain in the N-terminal regions and possesses multiple functions in cytokine signaling. The promoter area of TRAF1 reportedly contains several binding sites that can be activated by NF-κΒ, as found by electrophoretic mobility shift assay studies and luciferase functional analyses. Importantly, NF-κB activation can also upregulate TRAF1.³⁰ TRAF1 can also inhibit tumor necrosis factor (TNF)-α-induced cell apoptosis by binding to the TNFR1-TNFR-associated death domain protein (TRADD) complex.^5,6,31 TRAF2 and TRAF5 are similar in structure and function, and participate in NF-κβ, mitogen-activated protein kinase (MAPK), and c-Jun N-terminal kinase (JNK) signaling activation. TRAF3 could not activate the JNK or NF-κB pathways. TRAF4 over-expression or deletion in animal models does not result in tumor development. TRAF6 deficiency can lead to the abnormal formation of skin appendices and development of the nervous system.³²

Emerging data indicate that the NF-κB-related inflammation state in local tissues is associated with stem cell–like properties in carcinogenesis.³³ Stem cell markers, including BMI-1, ALDH1, and Lin28B, are associated with head and neck tumorigenesis.^6,7,17 About 40.9% of oral leukoplakia patients with BMI-1 positivity develop oral cancer.¹² ALDH1-high cancer cells exhibit a high tumor sphere–forming ability.¹⁷ The reprogramming factor Lin28B promotes cancer stem–like properties in oral cancer.⁷ Inflammation site cytokines, such as transforming growth factor (TGF)-β, leukemia inhibitory factor, (LIF), basic fibroblast growth factor (b-FGF), and interleukin (IL)-6, can directly or indirectly upregulate the stem cell marker characteristics, including Lin28, Sox2, Nanog, and Oct4.^10,34 Hence, we analyzed the association of ALDH1, Lin28, and BMI-1 expression with the expression of the NF-κB upstream regulator TRFA1 in oral cancer tissues.

In this study, high expression levels of TRAF1, ALDH1, and Lin28B were observed in oral cancer, but the difference in expression of each protein was not significant between the clinical stages and pathological stages. Considering that the samples were dotted and stained in the same supporter and collected by a digital scanner in tissue arrays, we speculated that no significant association exists with the patients’ pathological information. We also found that high expression levels of TRAF1 were associated with the patients’ poor prognosis, but the association was not significant (p = 0.42), which was also found in the Lin28B (p = 0.6) and BMI-1 (p = 0.56). Absence of such significant differences might be due to the small quantity in this batch of samples or due to the variations, such as sample sites, visual field of statistics (for field, dot in tissue array is less than the traditional section), and follow-up time. Hence, the statistical result of the prognosis value of TRAF1, BMI-1, ALDH1, and Lin28B in OSCC in this study needs to be further integrated and analyzed with other sample database and by molecular mechanism studies.

Interestingly, the expression of TRAF1 showed a significant association with the expression of Lin28B (p < 0.001), BMI-1 (p = 0.032), and ALDH1 (p < 0.0001) in oral cancer tissues. This association was also found in the expression level clustering (Figure 4). Our early study and other groups had reported that Lin28B can promote cell malignancy,^10,35 cancer progression, and inflammation-induced carcinogenesis,³³ BMI-1 with oral carcinogenesis,¹³ and ALDH1+ cells show cancer stem cell characteristics.^14,15,17 Considering the above mentioned points, we speculated potential relation of TRAF1 with stem cell subpopulations in oral cancer, which needs to be investigated further by molecular mechanism studies.

In conclusion, TRAF1 is highly expressed in OSCC tissues and associated with cancer stem cell markers of ALDH1, Lin28, and BMI-1. TRAF1 as a NF-κB upstream regulator may be involved in inflammation-induced cancer stem cell characteristics.

Footnotes

Acknowledgements

T-F.W. and Y-C.L. contributed equally to this work.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This study was supported by the Fundamental Research Funds for the Central Universities (2042015kf0075) to Dr. Tian-fu Wu, National Natural Science Foundation of China (81272963, 81472528) to Prof. Zhi-Jun Sun, and National Natural Science Foundation of China (81272964, 81472529) to Prof. Wen-Feng Zhang.

References

Warnakulasuriya

Global epidemiology of oral and oropharyngeal cancer. Oral Oncol 2009; 45(4–5): 309–316.

Jemal

Bray

Center

. Global cancer statistics. CA Cancer J Clin 2011; 61(2): 69–90.

Shigdar

Bhattacharya

. Inflammation and cancer stem cells. Cancer Lett 2014; 345(2): 271–278.

Lee

Choi

TRAF1 and its biological functions. Adv Exp Med Biol 2007; 597: 25–31.

Guo

Sun

Wang

. TRAF1 is involved in the classical NF-kappaB activation and CD30-induced alternative activity in Hodgkin’s lymphoma cells. Mol Immunol 2009; 46(13): 2441–2448.

Carpentier

Beyaert

TRAF1 is a TNF inducible regulator of NF-kappaB activation. FEBS Lett 1999; 460(2): 246–250.

Chien

Wang

Chu

. Lin28B/Let-7 Regulates expression of Oct4 and Sox2 and reprograms oral squamous cell carcinoma cells to a stem-like state. Cancer Res 2015; 75(12): 2553–2565.

Wang

Zhao

Deng

. Lin28: an emerging important oncogene connecting several aspects of cancer. Tumour Biol 2016; 37(3): 2841–2848.

Wang

Zhou

Guo

. Lin28 promotes Her2 expression and Lin28/Her2 predicts poorer survival in gastric cancer. Tumour Biol 2014; 35(11): 11513–11521.

10.

Jia

Xiong

. Increased expression of Lin28B associates with poor prognosis in patients with oral squamous cell carcinoma. PLoS ONE 2013; 8(12): e83869.

11.

Wei

Wang

. Bmi-1 is essential for the oncogenic potential in CD133(+) human laryngeal cancer cells. Tumour Biol 2015; 36(11): 8931–8942.

12.

Liu

Feng

Shen

. Two stem cell markers, ATP-binding cassette, G2 subfamily (ABCG2) and BMI-1, predict the transformation of oral leukoplakia to cancer: a long-term follow-up study. Cancer 2012; 118(6): 1693–1700.

13.

Kang

Kim

. Elevated Bmi-1 expression is associated with dysplastic cell transformation during oral carcinogenesis and is required for cancer cell replication and survival. Br J Cancer 2007; 96(1): 126–133.

14.

Xue

Huang

. Distinct prognostic values of ALDH1 isoenzymes in breast cancer. Tumour Biol 2015; 36(4): 2421–2426.

15.

Charafe-Jauffret

Ginestier

Bertucci

. ALDH1-positive cancer stem cells predict engraftment of primary breast tumors and are governed by a common stem cell program. Cancer Res 2013; 73(24): 7290–7300.

16.

Yalcintepe

Altinel

Albeniz

. Characterization of Lin(-)ALDH (bright) population using Ehrlich ascites tumor cells in mice. Tumour Biol 2014; 35(10): 10363–10373.

17.

Seino

Shigeishi

Hashikata

. CD44(high)/ALDH1(high) head and neck squamous cell carcinoma cells exhibit mesenchymal characteristics and GSK3beta-dependent cancer stem cell properties. J Oral Pathol Med 2016; 45(3): 180–188.

18.

Sobin

Wittekind

Gospodarowicz

. TNM Classification of Malignant Tumours (UICC). Hoboken, New Jersey: JohnWiley & Sons, 2002.

19.

Liu

Mao

. C4.4A as a biomarker of head and neck squamous cell carcinoma and correlated with epithelial mesenchymal transition. Am J Cancer Res 2015; 5(12): 3505–3515.

20.

Mazzotta

Partipilo

De Summa

. Nuclear PARP1 expression and its prognostic significance in breast cancer patients. Tumour Biol 2016; 37(5): 6143–6153.

21.

L-L

Huang

C-F

. PD-1 blockade attenuates immunosuppressive myeloid cells due to inhibition of CD47/SIRPalpha axis in HPV negative head and neck squamous cell carcinoma. Oncotarget 2015; 6(39): 42067–42080.

22.

Deng

. Targeting STAT3 signaling reduces immunosuppressive myeloid cells in head and neck squamous cell carcinoma. Oncoimmunology 2016; 5(5): e1130206.

23.

Huang

Deng

Zhang

. Expression of LC3, LAMP2, KEAP1 and NRF2 in salivary adenoid cystic carcinoma. Pathol Oncol Res 2016; 22(1): 109–114.

24.

Liu

Pan

. Germline competency of parthenogenetic embryonic stem cells from immature oocytes of adult mouse ovary. Hum Mol Genet 2011; 20(7): 1339–1352.

25.

Hanahan

Weinberg

RA.

Hallmarks of cancer: the next generation. Cell 2011; 144(5): 646–674.

26.

Schwitalla

Fingerle

Cammareri

. Intestinal tumorigenesis initiated by dedifferentiation and acquisition of stem-cell-like properties. Cell 2013; 152(1–2): 25–38.

27.

Wong

Quigley

. Inflammation and Hras signaling control epithelial-mesenchymal transition during skin tumor progression. Genes Dev 2013; 27(6): 670–682.

28.

Rothe

Wong

Henzel

. A novel family of putative signal transducers associated with the cytoplasmic domain of the 75 kDa tumor necrosis factor receptor. Cell 1994; 78(4): 681–692.

29.

Mosialos

Birkenbach

Yalamanchili

. The Epstein-Barr virus transforming protein LMP1 engages signaling proteins for the tumor necrosis factor receptor family. Cell 1995; 80(3): 389–399.

30.

Wang

Mayo

Korneluk

. NF-kappaB antiapoptosis: induction of TRAF1 and TRAF2 and c-IAP1 and c-IAP2 to suppress caspase-8 activation. Science 1998; 281(5383): 1680–1683.

31.

Duckett

Gedrich

Gilfillan

. Induction of nuclear factor kappaB by the CD30 receptor is mediated by TRAF1 and TRAF2. Mol Cell Biol 1997; 17(3): 1535–1542.

32.

TNF receptor associated factors (TRAFs). Adv Exp Med Biol 2007; 597: 72–77.

33.

Iliopoulos

Hirsch

Struhl

An epigenetic switch involving NF-kappaB, Lin28, Let-7 MicroRNA, and IL6 links inflammation to cell transformation. Cell 2009; 139(4): 693–706.

34.

Niwa

Ogawa

Shimosato

. A parallel circuit of LIF signalling pathways maintains pluripotency of mouse ES cells. Nature 2009; 460(7251): 118–122.

35.

Alajez

Shi

Wong

. Lin28b promotes head and neck cancer progression via modulation of the insulin-like growth factor survival pathway. Oncotarget 2012; 3(12): 1641–1652.

Expression and associations of TRAF1,BMI-1,ALDH1,and Lin28B in oral squamous cell carcinoma

Abstract

Keywords

Introduction

Methods and materials

Ethics statement

Patient samples

Tissue microarrays

Immunohistochemical staining

Staining assessment

Hierarchical clustering

Statistical analysis

Results

Increased expression of TRAF1, ALDH1, BMI-1, and Lin28B in OSCC

Expression of TRAF1, BMI-1, ALDH1, and Lin28B in OSCC analyzed by pathological grades and TNM stages

Survival curves of TRAF1, BMI-1, ALDH1, and Lin28B in OSCC

Associations of TRAF1, Lin28B, BMI-1, and ALDH1 in OSCC

Discussion

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

References