Expression of cell-cycle regulators p53, p27, and p21 in mucoepidermoid and adenoid cystic carcinomas of the minor salivary glands

Abstract

Background 

The orderly progression of cells through different phases of the cell cycle is orchestrated by various molecules and proteins. p53 plays a crucial role in stress response and tumor suppression. It is a unique cellular processor that integrates signals from various pathways in order to coordinate cell proliferation, differentiation, and apoptosis. p21^WAF1/CIP1 and p27 have key positions in cell-cycle regulations, leading to arrest of cell proliferation. They enable a repair process of DNA damage. In several human tumors, a loss of these proteins is associated with poor clinical outcome. Despite recent progress in molecular medicine, there is still a paucity of data on the involvement of cell-cycle regulatory proteins in the pathogenesis of salivary gland tumors.

Aim 

The present work was performed to assess the immunohistochemical expression of p53, p27, and p21 in 15 patients with mucoepidermoid carcinomas (MECs) and adenoid cystic carcinomas (ACCs) of the minor salivary glands and their possible correlation with histopathologic grading.

Results 

The present study included eight patients with ACC and seven patients with MEC. Most of the ACC patients (62.5%) showed positive reaction for p53, whereas only 25% were immunoreactive to p21. Among the seven examined MEC patients, the overall expression of p53 was detected only in those with high-grade tumors. With respect to p21, a moderate reaction was detected in a patient with low-grade MEC. In contrast, immunoreactivity to the p27 protein was detected only in low-grade tumors.

Conclusion 

The cell-cycle regulatory protein p53 was highly expressed in high-grade neoplasms, whereas the reverse was noticed in the case of p27. These findings have a strong prognostic impact. p21 overexpression may be a useful prognostic marker when evaluated in conjunction with p53 status.

Introduction

Salivary gland tumors are relatively uncommon, comprising no more than 1% of all tumors and 3% of all head and neck neoplasias (^{Russo et al., 2005}). Their morphologic diversity and relative rarity frequently pose a challenge in their diagnosis and treatment (^{Papadogeorgakis et al., 2011}). As the clinical course and final outcome of many patients with salivary gland tumors cannot be reliably predicted on the basis of histomorphologic features, it is highly desirable to find new prognostic markers aimed at better characterizing tumor aggressiveness (^{Nagler et al., 2003}).

Mucoepidermoid carcinoma (MEC), representing 5% of all salivary gland tumors and 20% of the malignant forms, is the most frequently occurring primary malignancy of the salivary gland in both adults and children. This tumor is characterized by squamoid cells, mucus-producing cells, and cells of intermediate type, with a marked variation in prognosis (^{Miyabe et al., 2009}).

Adenoid cystic carcinoma (ACC) is a malignant salivary gland tumor characterized by a slow but relentless progression plagued by local recurrences, late metastases, and ultimately fatal outcome (^{Khafif et al., 2005}). Histologically, ACC is a biphasic tumor comprising ductal and myoepithelial components. The three major growth patterns are tubular, cribriform, and solid. These growth patterns form the basis for histologic grading, and the quantity of the solid component is considered to be the most significant adverse prognosticator (^{Da Cruz Perez et al., 2006}). However, it is often difficult to predict the prognosis of ACC by histologic findings alone; thus, there is a need to explore additional parameters for predicting the prognosis (^{Ciccolallo et al., 2009}).

Despite recent progress in molecular medicine, there is still a paucity of data on the involvement of cell-cycle regulatory proteins in the pathogenesis of head and neck tumors, particularly of salivary glands (^{Kelsch et al., 1997; Russo et al., 2005}).

Cell proliferation control is ensured by a group of proteins named cyclin-dependent kinases (CDKs), the activation of which is dependent on phosphorylation and cyclin association. In parallel, these CDKs are negatively controlled by two distinct groups of inhibitory proteins, the cyclin-dependent kinase inhibitors (CDKIs) (^{Affolter et al., 2005; Hernández-Zavala et al., 2005}). CDKIs fall into two families, INK4 and CIP/KIP, on the basis of their structural and functional properties (^{Hirai et al., 1995; Choi et al., 2001; Chung et al., 2008}).

The CIP/KIP family of genes includes p21Waf1, p27Kip1, and p57Kip2. They possess structural similarity – that is, a 60-residue protein homology, which is very important for their inhibitory activities. These proteins are potent inhibitors of a wide range of cyclin–CDK complexes implicated in G1 and S phase and are said to be universal CKIs (^{Chung et al., 2008; Croucher et al., 2010}).

Progression through the cell cycle is regulated by a variety of proteins that play an important role in allowing the correct order and timing of events involved in chromosomal repair, duplication, and separation (^{Kapuy et al., 2009}). CDKs and their regulatory partners, cyclins, form heterodimeric protein complexes; they appear and degrade during predetermined steps in the cell cycle (^{Schoelch et al., 1999}). These proteins allow for orderly progression of the cell cycle and also act to phosphorylate fundamental cell cycle proteins (^{Martinsson-Ahlzén et al., 2008}).

The p53 gene is a tumor suppressor gene that acts as the ‘guardian of the genome’. Many diverse cellular events, including DNA damage and hypoxia, activate the p53 gene. The p53 protein functions as a transcription factor, regulating downstream genes involved in cell-cycle arrest, DNA repair, and programmed cell death (^{Ryan, 2011}). Loss of p53 function confers genomic instability, impaired apoptosis, and diminished cell-cycle restraint. Therefore, p53 mutations are responsible for certain critical features of malignancy. Alteration of p53 is the most common mutation in human cancer (^{Shi et al., 2010}). Loss of p53 function has been demonstrated in about one half of all human cancers, including malignant salivary gland tumors (^{Affolter et al., 2005}).

The p21 protein is the gene product of the WAF1/CIP1 gene, which plays an important role in regulating the G1–S transition of the cell cycle (^{Chung et al., 2008}). In response to DNA damage, wild-type p53 accumulates and binds to the promoter region of the p21^WAF1/CIP1 gene. This induces the expression of p21^WAF1/CIP1, which inhibits the activity of the cyclin/CDK complex to block cell-cycle progression. The levels of p21^WAF1/CIP1 may also be elevated by nondependent mechanisms (^{Huo et al., 2004}). Further, p21^WAF1/CIP1 is associated with terminal differentiation and cell senescence, and thus may play a role in cell maturation and cell death (^{Schwandner et al., 2002}).

p27 is a member of the universal CKDI family (^{Dai et al., 2013}). p27 blocks the cell cycle at the G1/S phase checkpoint and is highly expressed in cells arrested at G0 and G1 phases. Enhanced p27 expression is induced by cell–cell contact and by specific growth factors such as transforming growth factor-β and cyclic AMP (^{Miskimins et al., 2001}). The prognostic value of p27 expression has been reported in various human tumors (^{Lloyd et al., 1999; Seki et al., 2009}) The low p27 expression was associated with an unfavorable prognosis for squamous cell carcinoma of the tongue. It was found that p27 can play a pathogenic role in the malignant transformation of tumors of salivary gland origin (^{Ben-Izhak et al., 2009}).

This work aimed to assess the immunohistochemical expression of p53, p27, and p21 in minor salivary gland MEC and ACC and their possible correlation with histopathologic grading.

Materials and methods

Samples

Fifteen cases of ACC and MEC were selected from the archival files of the Pathology and Oral Pathology Departments, Faculties of Medicine and Dentistry, Alexandria University, Egypt, between 2005 and 2010.

Methods

Serial sections of 4-μm thickness were cut from each paraffin-embedded tissue block. Sections were stained with hematoxylin and eosin for confirmation of the histopathological diagnosis and for application of immunohistochemical (IHC) assays.

Antibodies and immunohistochemical assays

IHC study was performed to detect the expression of p53, p21, and p27 using monoclonal anti-p53 (Clone DO7; Dako, Glostrup, Denmark), p21 (Clone IB4; Novocastra, New Castle, UK), and p27 (Clone D10; Novocastra) antibodies. Sections were deparaffinized and dehydrated. Microwave unmasking of antigens was performed for 20 min in 0.01 mol/l citrate buffer at 98°C (pH 6). The sections were then left to cool for 1 h. Endogenous peroxide was subsequently blocked with 3% hydrogen peroxide for 10 min, followed by washing for 5 min with PBS. The specimens were incubated overnight at 4°C with anti-p53, p21, and p27 antibodies diluted at 1 : 100, 1 : 40, 1 : 40, respectively. They were then washed three times in PBS for 5 min each and incubated for 30 min with labeled-polymer-conjugated secondary antibody (Envision Kit; Dako, Carpinteria, California, USA). Finally, they were washed and developed with 3,3-diaminobenzidinetetrahydrochloride for 5 min, lightly counterstained with hematoxylin, dehydrated, and mounted.

Scoring

Tumor cells that showed positive nuclear reaction were semiquantitatively evaluated in at least 1000 cells, which were examined at ×40 magnification, and the number of positive tumor cells from the total number of neoplastic cells present in the same area was recorded. The biomarker immunoreactivity was classified as reported in the studies by ^{Aoki et al. (2004)} and ^{Croucher et al. (2010)} as follows:

Table: No title available.

Statistical analysis

Fisher’s exact test was used to evaluate the association between the cell cycle proteins and the two studied tumors, namely, MEC and ACC. Correlation between tumor grade and IHC results was determined using the Spearman correlation rank test. Data analysis was carried out using the SPSS statistical package, release 5.0.1 (SPSS Inc., Chicago, Illinois, USA). Results were considered statistically significant when the P value was less than 0.05.

Results

In the present work, 15 malignant salivary gland tumors were analyzed immunohistochemically for the expression of the cell-cycle regulators p53, p21, and p27.

The present work included eight patients with ACC of different patterns and seven patients with MEC.

Of the eight cases of ACC, four were predominantly cribriform (Fig. 1a), three were tubular-trabecular (Fig. 1b), and one was of solid type (Fig. 1c). Among the patients with MEC, four had low-grade tumors (Fig. 1d) and three had high-grade tumors (Fig. 1e).

Fig. 1: Different patterns of adenoid cystic carcinoma (ACC) and mucoepidermoid carcinoma (MEC). (a) Cribriform pattern of ACC revealing typical cylindrical-like spaces surrounded by hyperchromatic basaloid cells (H&E, ×400). (b) Tubular-trabecular pattern of ACC revealing numerous tubules and trabeculae (H&E, ×200). (c) Solid pattern of ACC exhibiting solid nests of darkly stained anaplastic basaloid cells in a fibrohyalinized stroma (H&E, ×200). (d) Low-grade MEC showing nests of clear mucus-secreting cells intermingled with a few epidermoid cells (H&E, ×200). (e) High-grade MEC revealing sheets of highly anaplastic epidermoid cells (H&E, ×400).

Among the examined malignant salivary gland tumors, 53.33% (8/15) showed positive reaction for p53, whereas 40% (6/15) were p27 positive. However, p21 was detected in only 20% (3/15) of tumors (Table 1).

Table 1: Immunohistochemical results of 15 specimens of adenoid cystic carcinoma and mucoepidermoid carcinoma

With regard to the ACC patients, 62.5% showed positive immune signals for the p53 protein and 25% were immunoreactive for p21. Intense nuclear brownish reaction for p53 was detected in the cribriform (Fig. 2a) and solid patterns, whereas a moderate to weak reaction was noted in the tubular pattern. p21 was detected only in the cribriform and tubular patterns of ACC. The reaction was moderate (Fig. 2b).

Fig. 2: Pattern of staining of p53 and p21 in adenoid cystic carcinoma (ACC). (a) Cribriform pattern showing strong expression of the p53 nuclear protein (×200). (b) Tubular-trabecular pattern exhibiting moderate immunosignals to p21 (×200).

Among the seven examined MEC patients, the expression of p53 was detected only in high-grade tumors. Intense nuclear reaction was observed in the anaplastic epidermoid cells in two patients (Fig. 3a), whereas moderate reaction was detected in one case. With regard to p21, a moderate reaction was detected in only one patient with low-grade MEC (Fig. 3b). Interestingly, the high-grade MECs were completely negative for p21 protein.

Fig. 3: Pattern of staining of p53 and p21 in mucoepidermoid carcinoma (MEC). (a) High-grade MEC revealing intense expression of p53 in epidermoid cells (×400). (b) Low-grade MEC showing moderate immunosignals for p21 (×200).

In this work, p27 protein was detected only in low-grade tumors, whereas it was completely negative in high-grade ones. In case of ACC, the immunoreaction was expressed mainly in the tubular-trabecular pattern. Only one patient with cribriform pattern showed weak reaction (Fig. 4a). In case of MEC, the p27 protein was detected only in low-grade tumors (Fig. 4b). However, it was completely negative in the high-grade ones.

Fig. 4: Pattern of staining of p27 in adenoid cystic carcinoma (ACC) and mucoepidermoid carcinoma (MEC). (a) Cribriform pattern of ACC revealing weak reaction to p27(×400). (b) Low-grade MEC exhibiting moderate reaction to the p27 protein in the clear mucus-secreting cells (×200).

No statistically significant difference was detected between ACC and MEC with respect to their IHC results for any of the studied cell-cycle proteins (P=0.57, 1.0, 0.44, respectively) (Table 2).

Table 2: Expression of p53, p27, and p21 in adenoid cystic carcinoma and mucoepidermoid carcinoma

Tumor grade is positively correlated only with p53 expression; there is no correlation with either p21 or p27 (Table 3).

Table 3: Correlation^a of tumor grade and immunohistochemical results of tumors to the studied cell-cycle proteins

Discussion

The significance of CIP/KIP protein expression in malignant neoplasms of the salivary glands is still unclear. The few reports in literature are contradictory (^{Affolter et al., 2005}).

p53 is a tumor suppressor gene located on the short arm of chromosome 17, whose protein product acts as a nuclear transcription factor with various functions, including the arrest of cell-cycle progression. (^{Das et al., 2008}). Mutation of this gene inactivates this tumor suppressor activity and is related to tumor progression and survival (^{Fisher, 2001}). There are conflicting results on p53 as a prognostic marker in head and neck malignancies, and the role of p53 mutations in the development and progression of salivary gland neoplasms remains largely unknown (^{Nagler et al., 2003}). According to the literature, involvement of the p53 mutation seems to play an important role in salivary gland malignancies in general at later rather than earlier stages of tumor progression and recurrence (^{Nagao et al., 1998; Kiyoshima et al., 2001; Mutoh et al., 2001}).

Our results revealed that 62.5% of ACCs exhibited immunopositivity for the tumor suppressor p53 with different grades of intensities. This is in agreement with the results of ^{Röijer et al. (1997)} and ^{Al-Rawi et al. (2010)}, who reported moderate p53 staining in their studied cases of ACC and correlated the associated immunoreactivity to the large size of the tumor, high grade, and extent of invasion. The same observation was suggested by ^{Kiyoshima et al. (2001)} and ^{Da Cruz Perez et al. (2004)}. Many other authors found that p53 analysis may be a useful prognostic indicator of aggressive disease (^{Zhu et al., 1997; Chau et al., 2001}; and ^{Wegner et al., 2007}).

p53 immunopositivity in MEC varies greatly from 8 to 87% (^{Kiyoshima et al., 2001; Abd-Elhamid and Elmalahy, 2010}). In the present study, 42.9% of MEC cases revealed p53 expression. The immunopositivity was mainly detected in high-grade tumors. This is in agreement with the results of ^{Yin et al. (2000)}. They stated that grade II and grade III MECs had a slightly higher rate of p53 immunopositivity compared with grade I MECs. They suggested that p53 is an inefficient prognostic marker as their patients with p53-negative tumors showed better survival.

The CDKI p21/waf1 is regulated by p53-dependent and p53-independent pathways (^{Ng et al., 1999}). The degree of p21 immunoreactivity is assumed to reflect the degree of expression of the normal p21 gene and serves as a useful tool for assessing its function (^{Goan et al., 2005}). p21 expression appears to vary in different human malignancies (^{Huang and Tao, 2000}). Its expression has been reported to be increased in cutaneous squamous cell carcinoma (^{Stoyanova et al., 2012}), non-small-cell lung carcinoma (^{Groeger et al., 2000}), head and neck cancer (^{Kapranos et al., 2001}), and hepatocellular carcinoma (^{Zhang et al., 2009}) but decreased in colorectal carcinoma (^{Al-Maghrabi et al., 2012}) and ovarian carcinoma (^{Yan et al., 2004}).

In the present study, 80% of the examined malignant salivary gland tumors showed negative reaction to p21. Only 25% of ACCs were positive. This is commensurate with the results of ^{Affolter et al. (2005)}. Seventy-two percent of their studied patients were p21 negative on IHC.

In the present work, only one of the studied MEC samples was immunopositive for p21; it was a low-grade one. In contrast, ^{Etemad-Moghadam et al. (2007)} detected p21 in 22.5% of their examined MEC specimens, with low expression in all grades. They claim that the loss of p21 develops in the earlier stages of tumorigenesis but not in later stages.

p27 regulates progression from G1 to S phase of the cell cycle by inhibiting cyclin-D or E-dependent kinase activity (^{Boudová et al., 2003}). Several clinical reports have shown that low p27 expression is associated with poor prognosis in patients with breast cancer (^{Filipits et al., 2009}), colon cancer (^{Dai et al., 2003}), gastric cancer (^{Takano et al., 2000}), and oral cancer (^{Kudo et al., 2000}).

Our results revealed that p27 expression was detected in low-grade malignant salivary gland tumors. ^{Choi et al. (2001)} reported that low p27 was seen more often in high-grade MECs than in low-grade ones. They stated that the loss of p27 expression was related to the loss of its inhibitory role in cell-cycle progression, as well to rapid tumor growth and high frequency of metastasis. Moreover, ^{Keikhaee et al. (2007)} and ^{Affolter et al. (2005)} showed that downregulation of p27 was observed in ACC and was well correlated to metastasis.

Conclusion

This study illustrated that cell-cycle regulatory proteins are dysregulated during the development of salivary gland cancer. p53 was highly expressed in high-grade neoplasms and thus has a strong prognostic impact. p21 was detected only in a few patients and may be a useful prognostic marker when evaluated in conjunction with p53 status. However, p27 was associated with loss of expression in high-grade tumors and was expressed only in low-grade ones. It could be a possible parameter for poor prognosis in malignant salivary gland tumors.

Acknowledgements

Conflicts of interest

There are no conflicts of interest.