Immunohistological optimization of detection of bromodeoxyuridine-labeled cells in decalcified tissue.

Mice were injected with a range of bromodeoxyuridine (BrdU) concentrations from 0.01 mg to 10 mg, and their jaws were fixed in buffered formalin or modified Carnoy. After EDTA or formic acid decalcification, a range of DNA denaturation schedules was assessed and immunohistological detection of BrdU-containing nuclei was performed using the Sera Lab anti-BrdU antibody MAS 250b. For Carnoy-fixed tissue, denaturation in 1 N HCl for 8 min at 60 degrees C was capable of adequately detecting an injected dose of 0.05 mg but not a dose of 0.01 mg BrdU, whereas pepsin/HCl treatment gave only weak staining after injection of 1 mg BrdU. In comparison, formalin fixation required pre-treatment with 0.2-0.4% pepsin/HCl at 37 degrees C for comparable staining intensity, but could still not adequately detect a dose of 0.1 mg BrdU. There was little detectable difference in staining between EDTA- and formic acid-decalcified tissues after injection of 10 mg BrdU.

Patterns of immunoreactivity to an anti-fibronectin polyclonal antibody in formalin-fixed, paraffin-embedded oral tissues are dependent on methods of antigen retrieval.

The influence of antigen retrieval (AR) technique on immunoreactivity in formalin-fixed, paraffin-embedded tissues is recognized. In focal reactive overgrowths of oral mucosa, we noted that patterns of immunoreactivity to a fibronectin polyclonal antibody were dependent on methods of AR. To establish these patterns we investigated eight pyogenic granulomas and eight fibroepithelial polyps. In the absence of AR no immunoreactivity was observed. After alpha-chymotrypsin AR a band of intense immunoreactivity was associated with vascular endothelial cells, with either minimal or no staining of connective tissue. After microwave AR immunoreactivity was observed in connective tissue, especially in the subepithelial region, but there was no specific vascular staining. After autoclave AR immunoreactivity was observed in connective tissue and also in epithelial nuclei. To determine if these results represented either changes induced by tissue fixation and processing or exposure of epitopes normally masked in vivo, we investigated frozen sections from two fibroepithelial polyps and one pyogenic granuloma. In frozen sections immunoreactivity was observed around vascular endothelial cells and within connective tissue, especially in the subepithelial region, but not in epithelia. This study provides indirect evidence that alpha-chymotrypsin and heat-mediated AR protocols expose different masked epitopes, and reinforces the need to undertake appropriate pilot studies for immunohistochemical investigation of archival tissue.

RT-PCR investigation of fibronectin mRNA isoforms in malignant, normal and reactive oral mucosa.

This study aimed to establish patterns of cellular fibronectin mRNA splice variants in normal oral mucosa, oral squamous cell carcinoma, oral leukoplakias with and without atypia, and focal reactive overgrowths of oral mucosa. Particular emphasis was placed on evaluation of either the EDA or EDB domains as markers of malignancy. Total RNA was extracted from normal oral mucosa, oral squamous cell carcinoma, oral leukoplakias with and without atypia, reactive epulides, fibroepithelial polyps and denture-related hyperplasia. Reverse transcriptase polymerase chain reaction (RT-PCR) was used to identify different fibronectin transcripts at three splice sites (EDA, EDB and IIICS). All the tissues investigated produced EDA+, EDA-, EDB+ and EDB- splice variants, and this study did not support RT-PCR-based detection of either EDA or EDB domains as markers of malignancy in oral tissues. Variations in IIICS splice patterns were observed, although these were not specific to any lesion group. In particular, there were differences in either the inclusion or omission of the domain coding for the CS-5 binding site for alpha 4 beta 1 integrin, whereas the CS-1 binding site for alpha 4 beta 1 integrin was typically present when additional domains were included at the IIICS splice site. In conclusion, complex patterns of fibronectin splice variant transcripts exist in normal and pathological oral mucosa. This may reflect the multiple biological functions identified for fibronectin proteins, although the significance of different specific fibronectin splice variants has yet to be fully elucidated.

Iodo- and bromodeoxyuridine are excised at different rates from DNA of mouse tongue keratinocytes in vitro.

The development of antibodies to iododeoxyuridine (IdU) and bromodeoxyuridine (BrdU) has increased interest in the use of these compounds in cell biology. However, little is known about their metabolism in vivo and in vitro. In this paper we show that, contrary to the impression gained from earlier studies, these compounds are not stably incorporated into DNA of mouse tongue epithelial cells in vitro but are removed, with turnover times of 1.0 days (BrdU) and 3.8 days (IdU).

The degradation of 5-iododeoxyuridine and 5-bromodeoxyuridine by serum from different sources and its consequences for the use of the compounds for incorporation into DNA.

Enzymes are present in sera that convert 5-iododeoxyuridine (IdU) to deoxyuridine (dU) and 5-iodouracil (IU). Although in the presence of serum 5-bromodeoxyuridine (BrdU) is not subject to extensive debromination it is converted to 5-bromouracil (BrU) at approx. 50% of the rate for IdU. These conversions are likely brought about by the enzymes thymidylate synthetase and thymidine phosphorylase. In vivo and in culture the dU enters DNA as thymidine 5'-monophosphate (dTMP) via the de novo pathway. Deoxyuridine is often found as a contaminant of [3H]IdU and [3H]BrdU. For these reasons, complications can arise in the interpretation of experimental work using these radioactive compounds. The problems may be overcome by purifying the compounds by high performance liquid chromatography (HPLC) before use together with identification of the DNA components with which the 3H is associated by chromatographic analysis.

A topographical study of the circadian rhythm in labelling index of mouse gingival and floor-of-mouth epithelium, including changes in labelling activity with individual cell position on the epithelial ridges.

A continuous strip of epithelium from the mandibular teeth to the ventral surface of the tongue of B6D2F-1 mice was examined autoradiographically after tritiated thymidine flash-labelling. Five areas were defined: area 1, the gingival sulcus epithelium adjacent to tooth enamel; area 2, the free gingival margin epithelium; area 3, the attached gingiva; area 4, the floor of mouth with undulating basement membrane; area 5, the floor of mouth with flat basement membrane. Data for the circadian variation in the proportion of DNA synthetic cells were recorded into a microcomputer, which enabled a large number of cells to be scored. The topographical position of each basal cell along the rete ridges and the incidence of labelling were noted. In each of the five areas a statistically significant circadian variation in labelling index (LI) was demonstrated, with a peak at 04.00-06.00 h and a trough at 20.00 h, although area 1 was slightly out of phase with the rest. The 24-h average LI values were almost double those obtained from a single flash-labelling at 10.00 h. The peak to trough ratio in LI was greatest in area 5 and fell towards area 1. Within the attached gingiva, cells deepest in the epithelial ridges had a larger peak to trough ratio than more superficial basal cells. For a group of mice labelled at 10.00 h the mean LI of the basal epithelial cells in areas 1-5 was 7.5 +/- (3.0)% (SD). Various aspects of the distribution of DNA synthesis in relation to topography were examined.(ABSTRACT TRUNCATED AT 250 WORDS)

Rapid detection of human herpes simplex virus type 1 in saliva.

Herpes simplex virus type 1 (HSV1) can be rapidly identified in saliva from patients with acute herpetic gingivostomatitis, by in vitro amplification using the polymerase chain reaction and specific primers. Amplification of DNA results in a product of 110 bp length corresponding to the region 1381-1490 bp of the HSV1 thymidine kinase gene. The specificity of the reaction was demonstrated in three ways: (i) the presence of a Sma 1 restriction enzyme site in the amplified product sequence; (ii) Southern blot using a biotinylated HSV1-specific oligonucleotide probe and (iii) direct sequencing of amplified product. At high titres of virus (> 5 x 10(5) virions/ml saliva), saliva may be added directly to the amplification assay for detection purposes. However, at lower titres of HSV1 viral DNA must be purified from saliva before in vitro amplification. HSV was identified in the saliva from symptomatic patients with acute herpetic gingivostomatitis and was absent in saliva collected from controls.

Differential expression of protease inhibitor and small proline-rich protein genes between normal human oral tissue and odontogenic keratocysts.

The technique of differential hybridization was used to compare gene transcription between normal oral mucosa and odontogenic keratocyst lining. Protease inhibitors, elafin and stefin-B as well as beta-actin and two epithelial-specific small proline-rich (spr) proteins, which we have named SPRC and SPRK and which are distinct from salivary proline-rich proteins, were differentially expressed. Increased abundance of alpha I(I) collagen and elafin transcripts was demonstrated in the keratocyst, with decreased abundance of stefin B, SPRC and cytokeratins 4 and 13 transcripts compared to normal palatal mucosa. The deduced protein sequences of SPRC and SPRK were described and compared, and the relative abundance of their respective cDNAs in palatal and keratocyst libraries determined. Identification of factors controlling transcription of these genes could advance our understanding of the development of odontogenic keratocysts.

Yeast artificial chromosome cloning and chromosomal localization of the abundant odontogenic keratocyst protein elafin.

An imbalance in human leucocyte elastase (HLE) activity is widely recognized to play an important pathological role in a number of human diseases. An earlier report has described greater transcription of elafin, an endogenous inhibitor of HLE, in epithelia of odontogenic keratocysts of the jaw than in normal oral mucosa. The elafin gene was now localized to chromosome 20q11.2-13.1 using a combination of somatic cell-hybrid panel screening and fluorescence in situ hybridization using a biotinylated DNA probe prepared from isolated yeast artificial chromosomes. No other positive fluorescent signals were observed. This eliminates the elafin gene as a candidate gene for naevoid basal-cell carcinoma syndrome, as the gene for this syndrome localizes to chromosome 9q23.1-31. The elafin yeast artificial chromosome DNA is to be subcloned to identify polymorphic microsatellite markers that will establish whether this gene is frequently amplified in oral neoplastic tissue.

A mucoepidermoid tumour of the tongue.

A case of mucoepidermoid tumour arising in minor salivary glands of the tongue and assessed histologically to be of high-grade malignancy is described. The diagnostic difficulties involved in distinguishing the neoplasm from a squamous cell carcinoma are considered. From published figures it would appear that salivary gland neoplasms in the tongue are much more likely to be malignant than benign.

An unusual epithelial neoplasm of gingiva resembling the keratoacanthoma.

An unusual, tumour-like mass protruding from labial and palatal gingivae and surrounding the unerupted canine in a 12-year-old child is described. Clinically, the lesion was felt to be aggressive but histologically, the most likely diagnosis was keratoacanthoma, probably of the nodulovegetating type. Following excision of the mass, healing has been uneventful. Whether or not the histological diagnosis of keratoacanthoma is accepted, rather than an unusual odontogenic neoplasm, this case illustrates the existence of an entity comprising well-differentiated islands and cysts of squamous epithelium with a rapid growth pattern.

Atypical fibroxanthoma of oral mucosa: a variant of malignant fibrous histiocytoma.

Atypical fibroxanthoma, a more recently recognised variant of malignant fibrous histiocytoma, is described arising at the site of an earlier squamous cell carcinoma, in a patient treated with radiotherapy 21 years previously. Distinction between atypical fibroxanthoma and a poorly-differentiated squamous cell carcinoma proved to be difficult by ordinary light microscopy and it was only after immunohistochemical staining that the second lesion was confirmed as atypical fibroxanthoma. The natural history, differential diagnosis and histological pitfalls of this poorly characterised mesenchymal lesion are discussed.

Differences in in vitro growth of epithelium from inflammatory and developmental odontogenic cysts.

Ninety-three odontogenic cysts, 42 of inflammatory and 51 of developmental origin, were grown in vitro from explants and/or cell suspensions. There was little difference in the success rate of culturing epithelium from explants of dentigerous cysts (N = 28) or odontogenic keratocysts (N = 23) (approximately 75% and 87%, respectively) and the dentigerous cyst grew particularly well from suspensions (N = 11) (91%) compared with the keratocyst (N = 19) (58%). Epithelium from developmental odontogenic cysts grew much better in vitro than did cysts of inflammatory origin (56 to 58% from explants and 19 to 25% from suspension). From this work there is little evidence to support previous statements that the dentigerous cyst cannot be grown from explants, or that the odontogenic keratocyst has 'aggressive' growth characteristics.