SUMMARY. This paper reports the development of an enhanced chemiluminescent (ECL) assay for measuring the total antioxidant (AO) capacity of serum, saliva and a fluid collectable from the gum margin called gingival crevicular fluid (GCF). The theory behind the assay is explained, and the optimum conditions for the assay, and for storage of reagents and clinical samples is described. Calibration lines were linear (R~0'99; P

The expression of keratins, CEA, EMA, and rat liver antigen (RLA) and the presence of Ki67+ proliferating cells were studied in the epithelial linings of 50 odontogenic cysts using an indirect immunoperoxidase method on acetone-fixed frozen sections. All cysts were positive with monoclonal antibodies of broad keratin specificity (CK1, AE1-3), and between 40 and 100 per cent of epithelial cells expressed keratins 13 and 19. Keratins 7, 8, and 18 were rarely expressed although surface cells in areas of mucous metaplasia often expressed keratins 7 and 18. Expression of keratin 10/11 was related to the presence of a well-ordered epithelial lining and was detected in isolated cells in 4/32 non-keratinizing cysts and in the upper suprabasal cell layers of 17/18 keratocysts. Although CEA, EMA, and RLA were detected in the epithelium of all specimens, the pattern of expression of CEA and EMA differed between cyst types. Ki67+ proliferating cells were most prevalent in keratocyst epithelia, where they were usually found within lower suprabasal layers which were negative or weakly positive for keratins 10/11 and 13. These results indicate differences in keratin, CEA, and EMA expression between cyst types which appear to be dependent on epithelial differentiation/structure rather than cyst type or histogenesis. Although these differences may not be of diagnostic significance, the consistent expression of both keratins 13 and 19 may provide a useful marker of odontogenic epithelium in general.

The influence of eight decalcifying agents on the immunoreactivity of formalin-fixed, paraffin-embedded tissue for immunoglobulins, lysozyme, factor VIII-related antigen and keratin was studied using the unlabelled antibody peroxidase-antiperoxidase (PAP) method. Limited studies were also performed on tissues fixed in acid-formalin mixtures. All tissues were stained using an indirect immunoperoxidase method with mouse monoclonal antibodies to IgM, kappa and lambda light chains. The results suggest that, with controlled trypsinization of sections, it was possible to obtain optimal immunostaining for all tested antigens, with adequate preservation of histological structure, after decalcification in neutral EDTA or 10% aqueous acetic or formic acid. Tissue treated with agents containing mineral acids exhibited variable immunoreactivity and impaired counterstaining.

No abstract