Evaluation of superficial oral squamous cell malignancy based on morphometry and immunoexpression of cytokeratin 13 and cytokeratin 17

OBJECTIVE

Evaluation of combined morphometry and immunoexpression of cytokeratin 13 (CK13) and cytokeratin 17 (CK17) for cytological identification of superficial oral squamous cells.

STUDY DESIGN

Smears from 11 tongue squamous cell carcinoma patients were processed by liquid-based cytology, stained via the Papanicolaou method and divided into multiple specimens by cell transfer. Morphometric indices, including nuclear area, nuclear perimeter, nuclear circular rate, largest-to-smallest dimension ratio of the nucleus and nucleocytoplasmic ratio, were measured using a computerized analysis system. CK13 and CK17 were detected by immunostaining. Morphometric values were compared between cell populations with distinct staining and immunoexpression patterns.

RESULTS

Most orange G-stained superficial cells were negative for CK13 (99.4%) and CK17 (98.6%). For light green-stained superficial cells, loss of CK13 was associated with greater cellular atypia in the nuclear area, nuclear perimeter and nucleocytoplasmic ratio (p < 0.01), while expression of CK17 was related to higher-grade cellular atypia in the same parameters (p < 0.01) as well as the nuclear circular rate (p < 0.05).

CONCLUSION

Immunoexpression of CK13 and CK17 in light green-stained superficial cells was associated with more severe morphological atypia. Combined morphometry and immunoexpression of CK13 and CK17 might be useful for cytological diagnosis of this cell population.

Reduced expression of cytokeratin 4 and 13 is a valuable marker for histologic grading of esophageal squamous intraepithelial neoplasia

Histologic evaluation of low-grade or high-grade intraepithelial neoplasia (LG-IN or HG-IN) of the esophagus is important for estimating the risk of progression to invasive carcinoma. Discrimination between LG-IN and HG-IN, or neoplasia and non-neoplastic lesion (NNL), however, is occasionally difficult. This study was designed to evaluate whether cytokeratin expression can be used for discrimination of these lesions. Esophageal Iodine-unstained lesions (n=154), less than 10 mm, were classified into HG-IN, LG-IN, and NNL. These lesions together with 154 foci of normal esophageal epithelium (NEE) were examined by immunohistochemistry for cytokeratins (CK4 and CK13), p53 overexpression, and the MIB-1 labeling index. The ratios of CK4- and CK13-positive staining were scored from 1 to 3. The CK4- and CK13-positive staining ratios were decreased in NNL (73% and 78%), LG-IN (55% and 69%), and HG-IN (33% and 48%), compared to NEE (91% and 95%). The differences between LG-IN and HG-IN, neoplasia and NNL, and among these three lesions and NEE were statistically significant (p < 0.005). The cytokeratin scores correlated with the MIB-1 labeling index (both: p < 0.0001), but not with p53 overexpression. CK4 and CK13 immunohistochemistry could be an objective method for evaluating the risk for progression to invasive carcinoma.

Localization of keratins 13 and 14 in the lingual mucosa of rats during the morphogenesis of circumvallate papillae

We used fluorescence immunohistochemistry, analysis of differential interference contrast (DIC) images and confocal laser-scanning microscopy in the transmission mode, after staining specimens with toluidine blue, to examine the localization of keratin 13 (K13) and keratin 14 (K14) in the lingual epithelium of fetal and juvenile Sprague-Dawley rats during the prenatal and postnatal morphogenesis of circumvallate papillae. No immunoreactivity specific for K13 and K14 was detected in the lingual epithelium of fetuses on day 15 after conception (E15), at which time the primitive rudiment of the circumvallate papillae was detectable by the thickening of several layers of cuboidal epithelial cells. On E17 and E19, the developing circumvallate papillae were clearly recognizable, consisting of a central papilla and the surrounding sulcus. No immunoreactivity specific for K13 and K14 was evident in the lingual epithelium around these structures at this time. K14-specific immunoreactivity was first detected in the basal layer of the epithelium of the circumvallate papillae on postnatal day 0 (P0) and K13-specific immunoreactivity was detected on P7. Morphogenesis of the circumvallate papillae progressed significantly from P0 to P14, and immunoreactivity specific for K13 and K14 was clearly recognizable after P7. The respective patterns of K13-specific and K14-specific immunoreactivity differed during the development of the circumvallate papillae: K13-specific immunoreactivity was generally evident in cells of the intermediate layer of the epithelium, while K14-specific immunoreactivity was detected in cells of the basal and suprabasal layers. The present results are discussed in the context of the previously determined localization of K13 and K14 in the dorsal epithelium of the anterior part of the rat tongue during its morphogenesis.

In vivo behavior of complete human oral mucosa equivalents: characterization in athymic mice

BACKGROUND AND OBJECTIVE

The interest in tissue engineering as a way to achieve repair of damaged body tissues has led to the carrying out of many studies whose results point to the potential effectiveness of these methods. In a previous study, we reported the obtaining of complete autologous oral mucosa equivalents (CAOMEs), characterized by oral immature keratinocytes and stem cells on an autologous plasma and fibroblast scaffold. The purpose of this study is to show their behavior in vivo, by using them as free grafts in experimental animals, and to demonstrate their potential capacity to regenerate oral mucosa.

MATERIAL AND METHODS

We engineered CAOMEs, as previously described. All CAOMEs thus obtained were used as free grafts in nu/nu mice. To assess their evolution in vivo, we studied their histological and immunohistochemical features by using AE1/AE3 pancytokeratin, the 5/6 cytokeratin pair, cytokeratin 13, laminin 5, collagen IV, vimentin, p-63 and Ki-67, at 7, 14 and 21 d.

RESULTS

The structure became progressively closer to that of oral mucosa samples. Cytokeratin 5/6 staining became increasingly intense in the basal and suprabasal layers, and cytokeratin 13 was exclusively positive in the superficial layers. The basal membrane was completed in 21 d. Vimentin showed a correct formation of the chorion. The increasingly positive staining of p-63 and Ki-67 indicated that the regeneration process was taking place.

CONCLUSION

The present study shows the potential regenerative capacity of the CAOMEs by their ability to reach maturity similar to that seen in oral mucosa.

In vitro engineering of complete autologous oral mucosa equivalents: characterization of a novel scaffold

BACKGROUND AND OBJECTIVE

Restoration of oral mucosa defects by means of in vitro-cultured equivalents has become a valid alternative in the field of oral and periodontics surgery. Although different techniques have been described, none has been able to provide an equivalent with an autologous scaffold for the epithelium. The purpose of this study was to obtain complete autologous oral mucosa equivalents (CAOME) using the patient's own fibroblasts and plasma and to characterize these equivalents both morphologically and immunohistochemically.

MATERIAL AND METHODS

We acquired cell types (keratinocytes and fibroblasts) from the same mucosal samples, which were taken from healthy patients who underwent oral surgery. To construct the CAOME, a small sample of blood was obtained from the patient and subsequently processed to obtain a fibrin glue scaffold. All CAOME thus obtained were stained using the standard hematoxylin and eosin method to study their morphological characteristics. To establish the type of cells in the epithelial layer, CAOME were stained with pancytokeratin AE1/AE3, cytokeratins 5/6 and 13, p-63 and Ki-67. Finally, laminin 5 and collagen IV were used to reveal the presence of a basal membrane.

RESULTS

The CAOME featured a monolayer of cube-shaped epithelial cells similar to that found on the basal layer of the oral mucosa. Close to the epithelial layer lay the fibrin and fibroblasts-embedded scaffold. The CAOME was positive to pancytokeratin AE1/AE3, cytokeratin 5/6 and p-63. No reaction was found to cytokeratin 13 and Ki-67. There was staining to laminin 5 but not to collagen IV.

CONCLUSIONS

It is possible to engineer a CAOME with an epithelium of basal-like and immature keratinocytes, which could potentially reconstruct in vivo loss of tissue.

Loss of cytokeratin 13 expression in squamous cell carcinoma of the tongue is a possible sign for local recurrence

Cytokeratin (CK) 13 is an intermediate filament protein that is expressed in a cell-type-specific manner, in the tongue and occasionally in tongue squamous cell carcinoma (SCC). Correlations between the clinical features of patients with SCC and CK13 expression in the tumor are here investigated along with CK13's utility as a marker for tongue cancer status. Samples from 121 patients with SCC of the tongue were examined by immunohistochemistry with antibodies against CK13. Correlations between the expression level of CK13 in the tumor and the patients' clinical features were statistically analyzed by univariate and multivariate methods. Univariate analysis showed a more relevant number of local recurrence (P = 0.04) in CK13-negative staining patients. In addition, CK13-negative cases were associated with local recurrence by multiple logistic regression analysis (OR: 3.36; 95% CI: 1.044-10.78; P = 0.04). Our results suggest that the loss of CK13 expression indicates tumors with a high potential for recurrence, and thus CK13 could be useful for determining the best course of treatment.

[Bile salts induce differentiation in cultured human normal esophageal mucosal epithelial cells]

OBJECTIVE

To investigate the effects of bile salts on the proliferation and differentiation of human normal esophageal mucosal epithelial cells on cultured.

METHODS

Normal human esophageal mucosa was obtained during operation. The esophageal epithelial cells were isolated, cultured, and treated with 6 different conjugated bile salts [glycocholate (GC), glycochenodeoxycholate (GCDC), glycodeoxycholate (GDC), taurochenodeoxycholate (TCDC), and taurodeoxycholate (TDC), all of the concentration of 50 micromol/L, and taurocholate (TC) of the concentration of 20 micromol/L], and their mixtures the concentration of 50 micromol/L respectively. One, three, and five days later MTT assay was applied to detect the cell proliferation. The cell cycle was assayed by flow cytometry with propidium iodide staining 1 and 3 d after treating with the bile salts. The cytokeratin 13 (CK13) in the differentiated cells and cytokeratin 14 (CK14) in the proliferating cells were detected by immunocytochemical assay. The concentration of intercellular calcium ([Ca(2+)]i) was analyzed by Laser Scanning Confocal Microscope (LSCM) in cells with TC, mixed bile salts and GC.

RESULTS

The cultured esophageal epithelial cells treated by the bile salts, except those treated by GC for 1 - 3 days, became larger and shuttle-like, with the cell proliferation inhibited, the percentages of cells in G(0)-G(1) phase increased and percentages of cells in S phase decreased. The percentages of CK14 positive cells were increased, but the percentages of CK13 positive cells were decreased time-dependently in cells treated for 1 - 5 days. The most obvious effects were seen in those cells treated with TC. The percentage of CK13 positive cells reached 74% +/- 8% in those cells treated with TC for 5 days, higher significantly than the percentage in the control cell (22% +/- 7%), (P < 0.01). The [Ca(2+)]i increased significantly only several minutes after treatment of TC and mixed bile salts, however, GC failed to cause the increase of [Ca(2+)]i.

CONCLUSION

GCDC, GDC, TC, TCDC, TDC and their mixture all induce differentiation of cultured human normal esophageal mucosal epithelial cells, but nor does GC. Increased [Ca(2+)]i is related to the TC-induced differentiation in those esophageal mucosal epithelial cells.

Antimicrobial barrier of an in vitro oral epithelial model

OBJECTIVE

Oral epithelia function as a microbial barrier and are actively involved in recognizing and responding to bacteria. Our goal was to examine a tissue engineered model of buccal epithelium for its response to oral bacteria and proinflammatory cytokines and compare the tissue responses with those of a submerged monolayer cell culture.

DESIGN

The tissue model was characterized for keratin and beta-defensin expression. Altered expression of beta-defensins was evaluated by RT-PCR after exposure of the apical surface to oral bacteria and after exposure to TNF-alpha in the medium. These were compared to the response in traditional submerged oral epithelial cell culture.

RESULTS

The buccal model showed expression of differentiation specific keratin 13, hBD1 and hBD3 in the upper half of the tissue; hBD2 was not detected. hBD1 mRNA was constitutively expressed, while hBD2 mRNA increased 2-fold after exposure of the apical surface to three oral bacteria tested and hBD3 mRNA increased in response to the non-pathogenic bacteria tested. In contrast, hBD2 mRNA increased 3-600-fold in response to bacteria in submerged cell culture. HBD2 mRNA increased over 100-fold in response to TNF-alpha in the tissue model and 50-fold in submerged cell culture. Thus, the tissue model is capable of upregulating hBD2, however, the minimal response to bacteria suggests that the tissue has an effective antimicrobial barrier due to its morphology, differentiation, and defensin expression.

CONCLUSIONS

The oral mucosal model is differentiated, expresses hBD1 and hBD3, and has an intact surface with a functional antimicrobial barrier.

Immunohistochemical expression of keratins 13 and 14 in the lingual epithelium of rats during the morphogenesis of filiform papillae

We examined the immunofluorescence of keratins 13 (K13) and 14 (K14) and differential interference contrast (DIC) images during the morphogenesis of filiform papillae and the keratinization of the lingual epithelium of rats on semi-ultrathin sections of epoxy resin-embedded samples by laser-scanning microscopy. We also examined semi-ultrathin sections of epoxy resin embedded, toluidine blue stained samples by light microscopy to obtain details of cell histology and morphology. No immunoreactivity specific for K13 and K14 was detected on the lingual epithelium of foetuses on days 13, 15 and 17 after conception (E13, E15 and E17), during which time the number of layers of cuboidal cells in the lingual epithelium increased from one to several. Immunoreactivity specific for K13 and K14 was first detected on the lingual epithelium of foetuses on E19. The immunoreactivity specific for K13 appeared in the suprabasal cells of the papillary and interpapillary cell columns and immunoreactivity specific for K14 was detected in the basal and suprabasal cells of the papillary and interpapillary cell columns. The lingual epithelium was composed of stratified squamous cells. The rudiments of filiform papillae were compactly arranged and interpapillary cell columns were very narrow. Filiform papillae developed gradually from postnatal day 0 (PO) to 21 (P21). The width of interpapillary spaces also increased during this period. Immunoreactivity specific for K13 and K14 was distinct at all postnatal stages examined. Thus, the patterns of immunoreactivity of K13 and K14 differed as the filiform papillae developed.