Normal squamous epithelium. Strong expression of transforming growth factor β type II receptors in middle layers of the epithelium and moderate intensity in basal and parabasal layers. No expression in upper keratin layer (×250).
Transition from nonneoplastic epithelium to verrucous carcinoma. A, Nonneoplastic epithelium (right) shows cytoplasmic expression of transforming growth factor β type II receptor (TβR-II) in middle and upper layers. Verrucous carcinoma (left) shows membranous expression of TβR-II. B, Membranous TβR-II expression in central regions of verrucous carcinoma (×250). C, Competing peptide control.
Transforming growth factor β type II receptor location within verrucous carcinoma (VC). A, Cytoplasmic location in the periphery (right) and membranous location in the center (upper left) (×250). B, Membranous expression in VC component (upper left) and cytoplasmic expression in the squamous cell carcinoma (SqCC) component (×100). C, Cytoplasmic staining in de novo SqCC (×250).
Detection of transforming growth factor β type II receptor messenger RNA by in situ hybridization. A, Verrucous carcinoma. B, Well-differentiated squamous cell carcinoma. C, Negative control (×250).
Central region of verrucous carcinoma; predominantly cytoplasmic location of transforming growth factor β type I receptor in verrucous carcinoma (×100).
Anderson M, Muro-Cacho C, Cordero J, Livingston S, Muñoz-Antonia T. Transforming Growth Factor β Receptors in Verrucous and Squamous Cell Carcinoma. Arch Otolaryngol Head Neck Surg. 1999;125(8):849-854. doi:10.1001/archotol.125.8.849
To study the intracellular location of transforming growth factor β type II receptors (TβR-II) in verrucous carcinoma (VC) and squamous cell carcinoma (SqCC), and to evaluate their role in the biological behavior of both neoplasias.
Ten VC and 10 well-differentiated SqCC specimens were analyzed by immunohistochemistry and in situ hybridization for the expression and intracellular location of TβR-II. Receptor expression was evaluated in areas of invasion and in areas of transformation of VC into SqCC. TβR-II expression was compared with expression of the type I receptor (TβR-I).
Formalin-fixed, paraffin-embedded tissue sections from VCs and well-differentiated SqCCs, operated on at the H. L. Moffitt Cancer Center and Research Institute from May 1987 to January 1998, were selected for the study.
While in all VCs TβR-II was found to be located along the membrane of the neoplastic keratinocytes, TβR-II expression in SqCC was observed predominantly in a cytoplasmic location. This cytoplasmic location of TβR-II was also seen in areas of transition from VC to SqCC. Expression of TβR-I was found in a cytoplasmic location in both tumor types.
The membranous location of TβR-II in VC exposes the receptor to the growth inhibitory control of TGF-β and may explain why VC tumors are less aggressive clinically. The marked reduction of membranous TβR-II and their predominant cytoplasmic location diminishes TGF-β growth inhibition and may contribute to the transformation of VC into the more aggressive SqCC.
VERRUCOUS carcinoma (VC) is a rare, well-differentiated, slow-growing, exophytic neoplasm of squamous epithelium, characterized microscopically by the presence of broad rete ridges, minimal atypia, low to no mitotic activity, and no metastatic potential.1- 3 Incomplete resection, however, leads to multiple recurrences and routine histopathological evaluation fails to identify those tumors that will progress to the biologically aggressive SqCC. Verrucous carcinoma differs from SqCC in that it is slow growing and exophytic, producing a warty clinical appearance, and in that endophytic growth usually results in superficial invasion with little risk of metastasis. Unsuccessful treatment, however, leads to multiple local recurrences and transformation into the biologically aggressive SqCC.
Transforming growth factor β (TGF-β) is a potent cytokine that regulates growth, differentiation, and extracellular matrix deposition in many cell types.4 One of the most striking in vitro effects of TGF-β is the reversible growth arrest, of nonneoplastic epithelial cells, in the G1 phase of the cell cycle.5 Resistance to the in vitro antiproliferative effects of TGF-β correlates with tumorigenicity in vivo. The events leading to TGF-β–induced growth arrest have not been fully elucidated, but they appear to include complex formation of the TGF-β signal transducers Smad2 or Smad3 with Smad4, and eventual activation of target genes,6 including the induction of inhibitors of cyclin/Cdk complexes,7 such as p27Kip and p15INK4. TGF-β induces these responses through interactions with the TGF-β type I (TβR-I) and type II (TβR-II) cell surface receptors.8 In most cases, the mechanisms of resistance to the antiproliferative effects of TGF-β are unknown. However, a decrease in the number of TβR-II binding sites has been associated with loss of responsiveness to TGF-β and increased tumorigenicity in some experimental models.9- 14 These observations have led to the hypothesis that escape from TGF-β–mediated growth control is an important aspect of the malignant phenotype.15
Several groups have examined the expression of TβR-II, by immunohistochemistry, in breast, head and neck, and prostate carcinomas.16- 18 In these human cancers, the expression of TβR-II was predominantly cytoplasmic. In the present study, we examined by immunohistochemistry the expression of TβR-II receptors in 10 VCs. In contrast to SqCC, where receptors are found in a cytoplasmic location, in VC, TβR-II are located predominantly in the cellular membrane. This different location in TβR-II may contribute to the differences in biological behavior between these 2 tumor types.
Ten well-differentiated SqCC and 10 VC specimens of the head and neck were obtained from the tissue archives of the Pathology Department at the H. Lee Moffitt Cancer Center and Research Institute, Tampa, Fla. The diagnosis of VC was made when a well-differentiated tumor had broad rete ridges, minimal atypia, and no stromal invasion. The diagnosis of well-differentiated SqCC was made following standard criteria, including keratinization, intracellular bridges, mild to moderate atypia, occasional mitoses, and peritumoral stromal invasion. In 3 VC specimens, areas of transition to SqCC were observed.
Five-micrometer sections from formalin-fixed, paraffin-embedded tissues were cut and placed on poly-L-lysine–coated slides. The tissue sections were subjected to deparaffination in xylene, followed by hydration through a series of decreasing alcohol concentrations, following standard procedures. Endogenous peroxidase was quenched with a 3% solution of hydrogen peroxide for 20 minutes at 37°C, and the slides were washed in deionized water for 5 minutes. Antigen retrieval was performed by placing the slides in a clear plastic container with a vented top containing citrate buffer (4.5 mL of 0.1-mol/L citric acid; 21.5 mL of 0.1-mol/L sodium citrate; and 225 mL of deionized water), in a microwave oven set on high twice, each for 5 minutes. The slides were allowed to cool for 10 to 20 minutes, rinsed in deionized water, placed in phosphate-buffered saline (PBS) for 5 minutes, and drained. Blocking serum was applied and the slides were incubated in a humid chamber for 20 minutes at room temperature. After blotting, the following primary antibodies were applied at room temperature at a 1:100 dilution: anti–TβR-II (C-terminal domain) and anti–TβR-I (Santa Cruz Biotechnology Inc, Santa Cruz, Calif). After 1 hour, the slides were rinsed with PBS and placed in PBS for 5 minutes. For detection, the Vectastain ABC Kit, rabbit IgG, Elite series (Vector Laboratories, Inc, Burlingame, Calif) was used following the manufacturer's specifications. The biotinylated secondary antibody was applied for 20 minutes at room temperature in a humid chamber. At the end of this incubation, the slides were rinsed, and placed in PBS for 5 minutes, followed by the addition of the avidin-biotin complex (ABC). The slides were incubated in a humid chamber for 30 minutes at room temperature, rinsed, and placed in PBS for 5 minutes. Diamino benzidine, prepared following the manufacturer's recommendations, was applied to the slides and color development was monitored. When the desired intensity was reached (2-5 minutes), the slides were rinsed in water and counterstained with modified Mayer hematoxylin for 30 seconds. The slides were then washed in running water for 10 minutes, dehydrated, cleared, and mounted with resinous mounting medium. The intensity of staining was scored as 0 (no expression), 1+(<25% of cells), 2+(25%-50% of cells), and 3+(>50% of cells). The location of a receptor was reported as membranous when signal was observed only on the plasma membrane. Cytoplasmic expression was reported when diffuse signal was observed in the cytoplasm with no enhancement on the plasma membrane. Omission of the primary antibody and irrelevant isotype-matched antibody were used as negative controls. For TβR-II, the immunizing peptide was available for competition.
Paraffin-embedded samples were incubated at 60°C for 20 minutes to melt the paraffin and deparaffinized with xylene twice for 10 minutes each time. Samples were rehydrated in absolute ethanol 3 times for 3 minutes and rinsed in deionized water 3 times. The tissue slides were then incubated in 0.2% pepsin at room temperature for 30 minutes. The TβR-II sense and antisense (negative control) biotinylated oligonucleotide probes (50 ng/µL) were denatured by incubation at 90°C for 8 minutes and applied to the sections. Sense probe is 5′-CTCAGAAGAATATAACACCAGCAATCCTGA-3′; antisense probe is 5′-TCAGGATTGCTGGTGTTATATTCTTCTGAG-3′. Sections were then coverslipped, placed in a humid chamber, and incubated at 42°C overnight. After incubation, the coverslips were soaked with 2 times 0.3-mmol/L sodium chloride and 0.03-mmol/L sodium citrate (SSC) and removed. The tissue slides were then washed in 2 times SSC/0.2% bovine serum albumin (BSA) for 30 minutes at 60°C followed by rinses in tris-buffered saline (TBS) twice for 3 minutes each. The streptavidin-alkaline phosphatase (1:100) in tris–magnesium chloride/BSA was applied to each slide and incubated for 30 minutes at room temperature, followed by 3 rinses of TBS for 3 minutes each. The slides were soaked in TBS/magnesium chloride, pH 9.5, for 5 minutes before the addition of the 5-bromo-4-chloro-3-indolyl-phosphate/nitroblue tetrazolium substrate and incubated in the dark for 1 hour until the color developed. After the color developed, the slides were rinsed with deionized water and counterstained with nuclear fast red for approximately 10 minutes. Crystal/Mount aqueous mounting medium was applied and the slides were heated at 70°C for 20 minutes followed by the application of Cytoseal 60 mounting media (Stephens Scientific, Kalamazoo, Mich) and coverslip.
Normal squamous epithelium showed strong expression of TβR-II in the middle layers and mild to moderate intensity in basal and parabasal layers (Figure 1). The expression of TβR-II, in these areas, was mostly cytoplasmic, with occasional increase in signal intensity in the membrane. No expression was observed in the upper, acellular, keratin layer.
To examine the localization of TGF-β receptors in VC, tissue samples from 10 VC specimens were subjected to immunohistochemical analysis. In areas of transition from nonneoplastic epithelium to VC, a shift from cytoplasmic location to a predominantly membranous expression of TβR-II (Figure 2, A) was observed. This membranous pattern of expression of TβR-II was maintained in cells of the central regions of all 10 VC specimens examined (Figure 2, B). In the deeper regions of VC, the receptor was found predominantly in a cytoplasmic location. Incubation of the primary antibody with the competing peptide abolished signal (Figure 2, C).
To determine if the TβR-II membranous pattern observed in VC is also found in conventional SqCC, we performed the immunohistochemical studies on 10 well-differentiated, keratinizing SqCC specimens. The pattern of expression of TβR-II in well-differentiated SqCC samples was predominantly cytoplasmic (Table 1). While all VC specimens showed intense membranous pattern, only 2 of 10 SqCC samples showed a focal area of weak membranous staining (Table 1). In deeper regions of VC (Figure 3, A), in transitions from VC to SqCC (Figure 3, B), and in SqCC arising de novo (Figure 3, C), the location was predominantly cytoplasmic.
To compare the expression of TβR-II protein to the presence of TβR-II messenger RNA (mRNA) in VC and well-differentiated SqCCs, we performed in situ hybridization. Abundant TβR-II mRNA was detected in both VC and SqCC samples(Figure 4). Interestingly, as it was the case for the protein, stronger mRNA signal intensity was observed in a perimembranous location in central areas of VC (Figure 4, A). The intracellular location of the mRNA correlated with that of the protein, suggesting that mRNA transcription and protein synthesis occur in areas of the cell where the protein exerts its function. These phenomena have been reported by several authors in other models.19,20
To examine if the membranous pattern of expression was unique for TβR-II, we also evaluated the expression of another TGF-β receptor, TβR-I. In contrast to the pattern observed for TβR-II, immunohistochemical analysis showed a predominantly cytoplasmic location of TβR-I in both VC and SqCC samples (Figure 5). This suggests that the predominant membranous location of TβR-II plays an important role in TGF-β–mediated growth inhibition.
Alterations of critical genes in early stages of tumor development may explain why squamous epithelial cells, with the potential to form fully malignant tumors (SqCC), develop instead into tumors with low to moderate malignant potential (VC). Although several groups have found p53 mutations and cyclin D1 overexpression in these tumors, these alterations do not distinguish VC from SqCC at the molecular level.21,22 In an effort to understand the different biological behavior of these 2 tumors, and the mechanisms responsible for the transformation of VC into SqCC, we have explored the integrity of the TGF-β signaling pathway.
TGF-β induces reversible growth arrest, in the G1 phase of the cell cycle, of nonneoplastic epithelial cells.5,7,23 Resistance to the in vitro antiproliferative effects of TGF-β correlates with tumorigenicity in vivo. Although the mechanisms responsible for this resistance are unknown, a decrease in the number of TβR-I and TβR-II binding sites has been associated with loss of responsiveness to TGF-β and tumorigenicity.9- 14 Thus, the breast carcinoma cell line MCF7(−) is TβR-II negative, resistant to TGF-β antiproliferative effects, and tumorigenic. Transfection with TβR-II complementary DNA (cDNA) increases the number of TβR-II receptors, restores TGFβ sensitivity, and decreases tumorigenicity.13 Recently, Wang et al24 demonstrated that reexpression of TβR-II in colorectal cancer cell lines, which have frameshift mutations associated with microsatellite instability, reversed both in vivo and in vitro malignant properties. These studies reveal a direct relationship between TβR-II levels, TGF-β resistance, and tumorigenicity, and suggest a role for TβR-II as a tumor suppressor gene.
TβR-I and TβR-II form heterotetramers in which TβR-II is constitutively phosphorylated. On TGF-β binding, TβR-II transphosphorylates TβR-I and initiates downstream events, including the formation of a complex between the TGF-β signal transducer proteins Smad2 or Smad3 and Smad4, translocation of this complex to the nucleus, and activation of target genes.6,7 Normal function of this signaling pathway requires the presence of the receptors in the cell surface. However, in human breast and ovarian carcinoma cell lines, TβR-II has been found in a cytoplasmic location by biochemical methods.17,25 Furthermore, Koli and Arteaga17 reported that, in the MCF-7 breast carcinoma line, the cytosolic TβR-II is an incompletely processed receptor that cannot bind iodine 125 TGF-β, while the membranous TβR-II does.
In this study, we have examined the expression of TβR-II, using immunohistochemistry, and found that, in well-differentiated head and neck SqCC, TβR-II is found predominantly in the cytoplasm. Cytoplasmic expression has also been observed in proliferating layers of normal squamous epithelium. In contrast, our analysis of 10 VC specimens showed that TβR-II is located in the cellular membrane in the central areas of the tumor. This membranous pattern of TβR-II protein changes to a cytoplasmic expression in deeper areas, where the tumor is morphologically indistinguishable from SqCC, and in areas of transformation of VC into SqCC. Based on this evidence, it is highly likely that VC cells, with membranous TβR-II expression, are still capable of responding to the antiproliferative effects of TGF-β. In contrast, the more aggressive SqCC may have become resistant to TGF-β–mediated inhibitory control due to the cytosolic location of TβR-II. A predominantly cytoplasmic location of the receptor in the normal squamous epithelium is consistent with this hypothesis. The receptor is found in the cytoplasmic location of proliferating layers. Nonproliferating superficial layers have minimal or no expression. Since only paraffin blocks were available, we were not able to investigate the functional status of TβR-II. However, the in situ hybridization experiments strongly suggest that abnormal RNA transcription may be responsible for the lack of functional receptor. Therefore, assessment of the intracellular location of TβR-II may aid in predicting the biological behavior of VC on biopsy material. Thus, a change of membranous to cytoplasmic location in a recurrent tumor might be considered a sign of progression to SqCC. Similarly, a disappearance of the receptor might indicate a tendency to invasiveness and aggressive behavior, as suggested by studies of colon,26 prostate,18 and head and neck cancers,27 where loss of expression of TβR-II is associated with disease aggressiveness.
Accepted for publication January 11, 1999.
Presented at the National American Association for Cancer Research meeting, New Orleans, La, March 28-April 1, 1998.
This work was supported in part by American Cancer Society grants RPG-98-039-01-CNE and IRG 202.
Reprints: Teresita Muñoz-Antonia, PhD, H. Lee Moffitt Cancer Center and Research Institute, 12902 Magnolia Dr, Tampa, FL 33612 (e-mail: antoniaT@moffitt.usf.edu).