Activation of Stat-3 as one of the early events in tobacco chewing-mediated oral carcinogenesis


  • Jatin K. Nagpal M.Sc.,

    1. Molecular Oncology and Medical Biotechnology Division, Institute of Life Sciences, Chandrasekharpur, Bhubaneswar, India
    Search for more papers by this author
  • Rajakishore Mishra M.Sc.,

    1. Molecular Oncology and Medical Biotechnology Division, Institute of Life Sciences, Chandrasekharpur, Bhubaneswar, India
    Search for more papers by this author
  • Bibhu R. Das Ph.D.

    Corresponding author
    1. Molecular Oncology and Medical Biotechnology Division, Institute of Life Sciences, Chandrasekharpur, Bhubaneswar, India
    2. Clinical Reference Laboratory, Speciality Ranbaxy Limited, Mumbai, India
    • Molecular Oncology and Medical Biotechnology Division, Institute of Life Sciences, Nalco Square, Chandrasekharpur, Bhubaneswar 751 023, India
    Search for more papers by this author
    • Fax: 91-674-430933



The Jak/Stat signaling pathway transmits signals from many cytokines and growth factor receptors to target genes in the nucleus. Constitutive activation of Stat-3 recently has been observed in many tumor cells, and dysregulation of the Stat signaling pathway has been proposed to be implicated in malignant transformation. In the current study for the first time to the authors's knowledge, the expression of STAT-3 was analyzed in various stages and sites of squamous cell carcinoma of the head and neck (HNSCC).


Tissue samples from 90 patients of tobacco chewing-mediated HNSCC representing various stages, sites, and differentiation states were selected for studying STAT-3 protein and RNA expression. In vivo localization of STAT-3 was studied by immunohistochemistry of paraffin embedded sections. The presence of STAT-3 and its phophorylated and activated form pSTAT-3 was checked by Western blotting. mRNA expression was analyzed by reverse transcriptase-polymerase chain reaction (RT-PCR). Apoptosis analysis was conducted by in situ ENA nick end labeling assay and hematoxylin and eosin staining.


Overall, 58.9% of HNSCC tumors showed very high Stat-3 protein accumulation, and 23.3% showed intermediate accumulation whereas 17.8% of HNSCC tumors were negative for Stat-3. No Stat-3 was detected in normal samples, and only one of eight premalignant lesions showed intermediate Stat-3 accumulation. On immunoblotting, very high protein accumulation was detected in T1 and T2 classification, moderate in T3 and T4 (P = 0.033, chi-square test), whereas no Stat-3 was detected in normal samples. Similar trend also was found in Stat-3 mRNA expression by RT-PCR analysis which was high in T1 and T2 (early stages), moderate in T3 and T4 (late stages), and no expression in normal samples. The mean apoptotic indices were 1.75, 1.88, and 1.66 for normal, premalignant lesions, and HNSCC cases, respectively.


Stat-3 activation is an early event in head and neck carcinogenesis though its role in blocking the apoptosis in vivo in solid tumors was not observed. Cancer 2002;94:2393–400. © 2002 American Cancer Society.

DOI 10.1002/cncr.10499

Squamous cell carcinoma of the head and neck (HNSCC) is the sixth most common malignancy worldwide. The increase in the number of cases of HNSCC every year in India and South East Asian countries as a whole is very disturbing. Whereas significant progress has been made in defining the molecular mechanism of oral carcinoma progression, the overall survival percentage has not changed in recent years. Among the more pressing problems in clinical management is the lack of early detection because of the absence of a potential diagnostic marker. Hence, knowledge of molecular alterations in various stages of oral tumorigenesis will greatly help in identifying putative biomarkers for early diagnosis and as novel targets for therapeutic intervention.

Cytokines and growth factors play central roles in the regulation of a wide array of cellular functions in eukaryotic cells by affecting target cells through the JAK-STAT pathway.1, 2 The STAT proteins (signal transducers and activators of transcription) were identified in the last decade as transcription factors that were critical in mediating virtually all cytokine driven signaling.3–5 In addition to their central roles in normal cell signaling, recent studies have demonstrated that diverse oncoproteins can activate specific STATs (particularly Stat-3 and Stat-5) and that constitutively activated STAT signaling directly contributes to oncogenesis.6 STAT-3 has been shown to be activated by membrane-associated JAK tyrosine kinase as well by other non-membrane-associated family members of Src, Fes, and Abl.7 The inactive cytoplasmic STATs are activated by phosphorylation that then translocate to the nucleus and activate target gene transcription.1, 2, 8 Recently Bromberg et al.7 have shown for the first time to our knowledge that activated STAT-3 by itself can mediate cellular transformation, thus qualifying Stat-3 as a protooncogene. In normal cells, ligand-dependent activation of the STATs is a transient process, lasting for several minutes to several hours. In contrast, in many cancerous cell lines and tumors, where growth factor dysregulation is frequently at the heart of cellular transformation, STAT proteins are persistently tyrosine phosphorylated or activated.5, 9 It has been shown by many researchers that Stat-3 inhibits apoptosis by modulating apoptotic regulatory proteins.10–14 Investigations using an antisense gene therapy approach,11 transfection with dominant negative Stat-3 constructs,2, 15 and JAK kinase inhibitor AG49013,16 show inhibition of growth and decrease in levels of BclXL/Bax.

In the current study, we have investigated the expression and activation levels of Stat-3 in various stages of HNSCC and compared it with noncancer controls. Furthermore, apoptotic index (AI) of the cells was analyzed and correlated with Stat-3 expression.


Tumor Specimens

Patients (n = 90) with HNSCC, who were treated at A.H. Regional Cancer Research and Treatment Centre, Cuttack, Orissa, during the last 3 years were included in this study. All these patients had a tobacco chewing habit for more than 10 years. The samples were collected after obtaining informed consent from the patients at the time of surgery. Samples were immediately snap-frozen in liquid nitrogen and stored at −70 ° C. The medical records of these patients were reviewed to obtain information regarding location of tumors, age, gender, and stages of tumor. Staging of the tumors was conducted according to the (American Joint Committee on Cancer [AJCC]/International Union Against Cancer [UICC]) TNM classification after brief histologic studies. Samples of normal and premalignant lesions were obtained from patients without cancer undergoing nononcologic surgical procedures.


All tissue specimens were fixed in 4% buffered formalin, processed routinely, and embedded in paraffin. Five-micrometer-thick sections were cut and mounted on silanized glass slides, air-dried, and heated 1 hour at 58 °C in an oven. For immunostaining mouse monoclonal antibody, Stat-3 (F-2), dilution 1:200 (Santa Cruz Biotechnology, Santa Cruz, CA) was used. Sections were dewaxed and rehydrated, and endogenous peroxidase was quenched by 30-minute incubation in 0.3% H2O2 in deionized water. Sections were heated in citrate buffer 0.1M (pH 6.0), in pressure for 20 minutes. Sections were cooled and washed in phosphate-buffered saline (pH 7.4). Primary antibody was applied and incubated at 4 °C overnight. The peroxidase reaction was developed using 3-3′ diaminobenzidine (Sigma Chemical Co., St. Louis, MO) for the chromogenic substrate. The slides were counterstained with hematoxylin. For negative control, buffer was used in place of primary antibody.

Western Blot Analysis

Whole cell extracts (25 μg/lane) were electrophoresed through 8% sodium dodecyl sulfate (SDS)-polyacrylamide gel and were transferred onto a Hybond-C-super nitrocellulose membrane (Amersham, Buckinghamshire, UK). Prestained molecular weight markers (Life Technologies, Inc. Gaithersburg, MD) were included. Membranes were blocked for 30 minutes in Tris-buffered saline (TBS, pH 7.5) with 0.5% Tween-20 (TBST) and 5% nonfat dry milk. After blocking, membranes were incubated for 60 minutes with a Stat-3 (F-2) and phosphorylation (Tyr-705) specific p-Stat-3 (B-7) mouse monoclonal antibody (Santa Cruz Biotechnology). After incubation with horseradish peroxidase-conjugated secondary antibody, the membranes were developed by using the enhanced chemiluminescence (ECL) detection system (Amersham).

Reverse Transcriptase-Polymerase Chain Reaction Analysis

RNA was extracted from normal and tumor tissues by LiCl-urea method with slight modification.17 Briefly, 0.5 g of tissue was homogenized in 5 mL of lysis buffer (6 M urea, 3 M LiCl, 50 mM sodium acetate, 200μg/mL heparin, and 0.1% SDS). It was centrifuged at 16,000×g for 20 minutes and then extracted twice with equal volume of phenol and chloroform and finally ethanol precipitated. The RNA pellet was air-dried and dissolved in diethylpyrocarbonate-treated water. Reverse transcriptase-polymerase chain reaction (RT-PCR) was performed using Perkin-Elmer Kit (Oak Brook, IL). The sequence of the human primers used were as follows: Stat-3 (116 base pair [bp]): sense, 5′-TCT CCT ACT TCT GCT ATC TTT GAG-3′; antisense, 5′-ATG GGT CTC AGA GAA CAC ATC-3′; glyceraldehyde-3-phosphate dehydrogenase (180 bp): sense, 5′-ACC CAC TCC TCC ACC TTT G-3′; antisense, 5′-CTC TTG TGC TCT TGC TGG G-3′.

The conditions used for the RT-PCR were 42 °C for 15 minutes, at 99 °C for 5 minutes, and at 5 °C for 5 minutes for 1 cycle, 95 °C for 45 seconds, at 60 °C for 45 seconds for 35 cycles, followed by a 7-minute extension at 72 °C. The products were checked in 2% agarose gel, along with the 100-bp ladder (Promega, Madison, WI).

AI Determination

The AI was determined by the method described.18 Hematoxylin and eosin-stained sections were examined microscopically at ×400. Morphologic features used to identify apoptotic nuclei included overall shrinkage and homogenous dark basophilia.

For in situ ENA nick end labeling (TUNEL) assay, the in situ cell death detection kit (Boehringer Mannheim, Mannheim, Germany) was used according to the manufacturer's instructions.

Statistical Analysis

The statistical analysis for determining the significance of STAT-3 staining with various clinical parameters was performed using the chi-square test. Apoptotic indices were presented as mean ± standard error. P value less than 0.05 was considered significant.


Stat-3 Expression/Activation in HNSCC

The first objective was to check the extent of Stat-3 immunoreactivity and its in vivo localization in the tissue samples of 90 patients of tobacco chewing-mediated HNSCC representing different stages, sites, and differentiation states, by immunohistochemistry (Table 1). For comparison, eight normal samples from noncancer control and eight premalignant lesions were selected and used. All of the eight normal samples were found to be negative for Stat-3 expression. Whereas of the eight premalignant lesions examined seven were negative for Stat-3 expression, and one of the samples showed intermediate cytoplasmic staining (Table 2). With 90 tumor samples studied, 58.9% (53 of 90) of samples showed very high Stat-3 accumulation, 23.3% (21 of 90) showed intermediate Stat-3 accumulation, and 17.8% (16 of 90) of primary tumors were negative for Stat-3. Both cytoplasmic and nuclear localization of the Stat-3 was detected in HNSCC primary tumors (Fig. 1A). No statistically significant correlation was observed between Stat-3 expression and various clinicopathologic parameters such as gender, age, site of cancer, and grade of tumors.

Table 1. Clinicopathologic Profile of Oral Carcinoma Patients with STAT-3 Immunoreactivity
Serial no.Clinical parametersNo. of casesStrongStat-3 interWeak/absentP value
  1. NS: not significant; WDSCC: well differentiated squamous cell carcinoma; MDSCC: moderately differentiated squamous cell carcinoma; PDSCC: poorly differentiated squamous cell carcinoma.

  2. The following scale was used to express the extent of positivity: 0, < 5%; 1, > 5–25%; 2, > 25–50%; 3, > 50–75%; and 4, > 75%.

  3. The intensity of STAT-3 was scored as follows: 0, negative; 1+, weak; 2+, moderate; 3+, as strong. The final score, obtained by multiplying the extent of positivity and intensity scores, ranged from 0 to 12. Scores of 0 to 4 were defined as “markedly reduced” or “no expression”; scores 5–8 were defined as “intermediate expression”; and scores of 9–12 were defined as “strong expression.”

2≤ 50 yrs47251309NS
> 50 yrs43280807
3Buccal mucosa (cheek)47300809NS
Angle of mouth (maxilla and mandible)26130706
Table 2. STAT-3 Immunoreactivity in Normal, Premalignant, and HNSCC Tumors
Serial no.CharacteristicNo. of casesStrongStat-3 interWeak/absentP value
  • HNSCC: squamous cell carcinoma of the head and neck; NA: not applicable.

  • a

    Chi-square test not applicable.

  • b

    Chi-square = 6.851; degrees of freedom = 2.

3HNSCC cases
 T1 + T2392906040.033b
 T3 + T451241512
Figure 1.

(A) STAT-3 immunostaining with F-2 antibody showing the cytoplasmic staining (original magnification ×200). (B) STAT-3 immunostaining showing nuclear staining (original magnification ×200). (C) TUNEL-stained apoptotic cells in normal epithelium (arrow) (original magnification ×400). (D) TUNEL-stained apoptotic cells in HNSCC (arrow) (original magnification ×200).

Of the HNSCC patients representing early classification (T1 + T2) of carcinogenesis, 74.4% showed high Stat-3 staining, whereas only 47.1% of HNSCC patients representing late classification (T3 + T4) of carcinogenesis showed high Stat-3 staining. This was confirmed further by immunoblotting analysis. High Stat-3 accumulation was observed in all stages of HNSCC in contrast with normal samples in which Stat-3 was totally absent. Also, the level of Stat-3 was maximum in Stage I and II, whereas its level decreased in AJCC/UICC Stage III and IV. Chi-square test (P = 0.033) clearly shows the gradual decrease in Stat-3 accumulation with the progression of tumorigenesis. When phosphorylation (Tyr-705) specific antibody was used, high levels of activated Stat-3 was observed in all stages of HNSCC with normal epithelium showing absence of phosph Stat-3 (Fig. 2).

Figure 2.

Expression of stage specific Stat-3 (top) and phospho Stat-3 (bottom) oncogene product in squamous cell carcinoma of the head and neck (HNSCC). Twenty-five micrograms of total cell lysates from noncancer controls and HNSCC patients was immunoblotted with both Stat-3 and phospho specific Stat-3 (Tyr-705) antibody to detect both native Stat-3 and activated form of Stat-3.

To compare the Stat-3 mRNA expression level between normal epithelium and tumor samples, we performed RT-PCR analysis. A high Stat-3 mRNA expression was detected in Stages I, II, and IV of HNSCC, with Stage III showing reduced expression. No mRNA expression was detected in normal epithelium (Fig. 3), suggesting that the Stat-3 pathway is blocked in normal oral mucosa. Various activities that are damaging to the oral mucosa such as chewing of carcinogenic agents, such as tobacco paste, pan masala, and betel quid, lead to expression and activation of Stat-3 pathway, which might play a critical role in process of carcinogenesis.

Figure 3.

Reverse transcription-polymerase chain reaction analysis of stage specific mRNA expression of STAT-3 in noncancer controls as well as HNSCC patients. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used to normalize for mRNA integrity and equivalent loading. (bp) base pair.

Stat-3 Activation and Apoptosis

We examined how the AI varies in correlation to Stat-3 expression and activation. Both hematoxylin and eosin staining and TUNEL assay were performed on 8 normal samples, 8 premalignant lesions, and 30 HNSCC samples showing strong Stat-3 staining (Fig. 1B). The mean AI was found to be 1.75 in normal samples in comparison with a mean AI of 1.66 in tumor samples. In premalignant lesions, the mean AI was found to be 1.88 (Table 3). No significant change was observed in AI of normal (Stat-3 negative) and tumor samples (Stat-3 positive) suggesting either no or a weak role of the Stat-3 pathway in blocking apoptosis in solid head and neck tumors.

Table 3. Apoptotic Indices in Normal, Premalignant and HNSCC Tumors
Serial no.CharacteristicNo. of casesAI (mean ± SE)
  1. HNSCC: squamous cell carcinoma of the head and neck; AI: apoptotic index; SE: standard error.

1Normal081.75 ± 0.313
2Premalignant081.88 ± 0.398
3HNSCC cases301.67 ± 0.175


Increasing evidence supports the critical role of Stat-3 in transformation and tumor progression. Extensive surveys of primary tumors and cell lines derived from tumors indicate that inappropriate activation of specific STATs occurs with surprisingly high frequency in a wide variety of human cancers.19 Delineation of the critical pathway involved in the carcinogenesis is the key for the pharmaceutical intervention. Grandis et al.,10, 11, 20 in their pioneering work on head and neck carcinoma have reported that constitutive activation of Stat-3 is linked to both production of transforming growth factor (TGF)-α and expression of epidermal growth factor receptor in HNSCC cells in vitro, as well as in mice model. Our study complements the findings of Grandis et al. in solid head and neck tumors. We observed that 82.2% of solid HNSCC tumors showed Stat-3 expression and activation whereas 17.8% of the tumors were negative for Stat-3. Both nuclear and cytoplasmic localization of Stat-3 was observed, suggesting that Stat-3 participates in signal transduction as well as activating transcription of other genes. Future studies of the downstream molecules activated by Stat-3 will be highly rewarding in reducing the morbidity and mortality caused by HNSCC. In contrast, none of the eight normal epithelium samples analyzed showed Stat-3 presence. Only one of the eight premalignant lesions showed intermediate cytoplasmic Stat-3 accumulation, thus agreeing with the hypothesis of Grandis et al.11 that Stat-3 activation may be an early event in squamous epithelial carcinogenesis. This early and specific activation of Stat-3 in the head and neck carcinogenesis also qualifies it as a potential diagnostic marker.

In the current study for the first time to our knowledge we have analyzed the stage specific expression and activation of Stat-3. Both in vivo immunolocalization analysis and immunoblotting showed high accumulation of activated Stat-3 in low classifications, i.e., T1 and T2, of tumors. In higher classifications, i.e., T3 and T4, there was a gradual reduction in accumulation of activated Stat-3 (P = 0.033). It suggests that Stat-3 performs its role (as signal transducers and activator of transcription) in the early stages and is subsequently degraded, as the cells proceeds toward malignant transformation. In the normal epithelia, accumulation of Stat-3 was not detected, thereby indicating a check on the Stat-3 pathway in normal epithelium. Analysis of Stat mRNA expression showed no mRNA expression in normal epithelia, high mRNA levels in classifications T1, T2, and T4 with reduced mRNA levels in classification T3. It will be interesting to know the involvement of other extracellular stimulus that removes this blockade in addition to TGF-α, as observed by Grandis et al.10, 11 Because in India tobacco chewing in the form of paste, pan masala, and betel quid are the primary causes of head and neck carcinoma, studies are in progress using the in vitro cell culture system to know the primary stimulus that leads to the activation of the Stat-3 pathway.

Emerging evidence suggests that Stat-3 inhibits apoptosis by modulating apoptotic regulatory proteins such as BclXL and Bax. We compared normal epithelium samples and premalignant lesions (with no Stat-3 accumulation) with tumor samples (with high Stat-3 accumulation) for presence of apoptotic cells. It was observed that the AI of normal, premalignant lesions and tumor samples were 1.75, 1.88, and 1.66, i.e., without having any significant difference. However, the current finding indicates for the first time the role of Stat-3 in blocking apoptosis in any solid tumor. All previous studies in this regard suggested an antiapoptotic property of Stat-3 and were performed in cell lines or in mice system whereas our results suggest that in an in vivo situation Stat-3 by itself has no or a weak role in blocking apoptosis. Thus, it is in contrast with the reports of the role of Stat-3 in blocking apoptosis in head and neck,10, 11 myeloma cells,12, 21 leukemic Sezary cells,16 293T, Hep2, NIH3T3 cells,7 fibroblasts,22, 23 breast carcinoma cells,17, 24, 25 and prostate carcinoma cells.2 However, from the current investigation, it may be concluded that activation of the Stat-3 pathway is a critical mediator of the tobacco chewing-mediated oral carcinogenesis though its role in blocking the apoptosis in vivo in solid tumors was not observed in HNSCC.


J.K.N. and R.M. thank Council of Scientific and Industrial Research, New Delhi, for research fellowship. The authors are grateful to the director and his colleagues of A. H. Regional Cancer Centre, Cuttack and Prof. K. S. Panda, Panda Nursing Centre, Cuttack for their cooperation in obtaining tumor samples