Viral infections in the mouth

Oral hairy leukoplakia (OHL) and Kaposi's sarcoma (KS) are commonly encountered in the HIV-infected patient. A unique feature of OHL is non-cytolytic high level of replication of Epstein–Barr virus (EBV) in the glossal epithelium. The expression of viral-encoded anti-apoptotic proteins concomitant to replicative proteins probably underlies this phenomenon. The question of whether OHL arises from activation of EBV latent in the tongue, or from superinfection by endogenous EBV shed via non-glossal sites or by exogenous EBV remains unresolved. Human herpesvirus 8 (HHV8) is now seen as necessary but not sufficient cause of KS. Expression of HHV8-encoded oncogenic proteins in endothelial cells probably explains the aberrant proliferation of these cells in KS lesions. Studies into why KS is so commonly observed at the palate in HIV-infected patients may provide important clues to its pathogenesis.     

Epstein-Barr virus infection and expression in oral lesions

OBJECTIVE: The prevalence of Epstein-Barr virus (EBV) and the recently discovered Kaposi's sarcoma associated herpes virus, human herpesvirus 8 (KSHV or HHV8), was determined within oral lesions common to HIV infection including OHL, pseudoOHL (PHL), oral lymphoma, oral aphthous ulcers, and an oral Kaposi's sarcoma. METHODS: DNA and RNA were extracted from oral lesions. EBV and HHV8 genomes were detected by Southern blot and polymerase chain reaction (PCR), and viral expression was analyzed using PCR amplification of cDNA. RESULTS: Multiple EBV strains were detected within OHL with recombination across repeat sequences generating new viral variants. EBV expression in OHL included expression of some viral genes, usually expressed in latent infections, that induce the EBV receptor. EBV replication was detected only within OHL lesions but not within adjacent Kaposi's tissue or oral aphthous ulcers while HHV8 was only detected within the Kaposi's lesions. CONCLUSIONS: These findings indicate that the OHL lesion is unique with viral replication and superinfection with additional EBV strains. Expression of the EBV receptor within the OHL lesion may promote superinfection which then activates EBV replication. The consistent detection of EBV replication only within OHL lesions and the detection of HHV8 only within Kaposi's sarcoma, strengthens the etiologic link between EBV and HHV8 infection to these specific pathologies.     

Expression of keratin 14 and 19 mRNA and protein in normal oral epithelia, hairy leukoplakia, tongue biting and white sponge nevus

This study was undertaken to analyze keratin gene expression at both the mRNA and protein level in oral hairy leukoplakia (OHL). Comparisons were made with normal lingual epithelium from a similar site, tongue biting, normal buccal mucosa and another condition which disturbs oral epithelial differentiation, white sponge nevus. Combined immnunocytochemical and in situ hybridization studies for keratins 14 and 19 were carried out on 2 specimens of OHL from HIV-positive males and one sample each of the other cases. Keratin 14 protein expression was uniform throughout all the epithelia. In normal epithelia and in lesions other than OHL, keratin 14 mRNA was most strongly expressed in basal cells with weaker but still significant amounts in the spinous cell layer. In both cases of OHL there was weaker basal cell expression of keratin 14 mRNA and frequent absence in koilocytoid cells. Keratin 19 protein expression was heterogeneous in the basal layer ol all specimens with suprabasal staining of occasional groups of cells. Its mRNA was uniformly distributed in all cases. The findings indicate that keratin mRNA expression does not always parallel that of protein and that, in the case of keratin 14, expression may be influenced by the presence of EBV.     

Bromodeoxyuridine incorporation and Ki 67 expression in oral hairy leukoplakia

OBJECTIVES: Oral hairy leukoplakia (HL) is an acan-thotic, hyperparakeratotic lesion characterised by the presence of a replicative Epstein-Barr virus (EBV) infection in the superficial and adjoining layers of the epithelium. EBV or its gene products are capable of modifying epithelial differentiation. The aim of this study was to establish whether the presence of EBV was associated with an alteration in cell turnover by assessing bromodeoxyuridine (BrdU) incorporation and Ki 67 expression in lesional tissue and control mucosa.
METHODS: Biopsies of HL together with age, site and sex matched controls ( n = 7 and 8 respectively) were incubated in 200 μM BrdU in vitro , fixed in methacarn and processed to paraffin wax. Following acid hydrolysis, incorporated BrdU and Ki 67 were identified in serial 5 fim sections using a three-stage immunoperoxidase technique and cell density expressed as the number of positive cells per mm basement membrane length.
RESULTS: Overall, there was no difference in the number of BrdU positive cells per mm basement membrane length between control and HL tissue. However, within HL alone, the presence'of focal EBV replication was associated with a significant reduction in the number of basal cells incorporating BrdU compared to adjacent EBV free areas (P < 0.05). There was no significant difference between Ki 67 positive cells in control and HL tissue and no evidence of a reduction of Ki 67 positive cells in areas associated with EBV replication.
CONCLUSIONS: These results suggest that there is no evidence of a generalised alteration of the proliferative capacity of basal cells in HL, although the focal reduction in BrdU incorporation may reflect subtle changes on cell turnover by EBV infection.     

Epstein-Barr virus: Biology and disease

Epstein—Barr virus is a human herpes virus which, whilst found as a widespread asymptomatic infection, is also associated with certain tumours of lymphoid and epithelial origin including Burkitt's lymphoma (BL), immunoblastic lymphoma (IBL), Hodgkin's Disease (HD) and nasopharyngeal carcinoma (NPC). A unique characteristic of EBV is its ability to infect and transform primary resting B lymphocytes in vitro into permanently growing lymphoblastoid cell lines (LCLs); this effect is associated with constitutive expression of a limited set of viral genes. Interestingly, the pattern of EBV gene expression observed in LCLs in vitro is also a feature of IBLs, a tumour associated with immunosuppression. The other EBV associated tumours display a more restricted pattern of EBV latent protein expression. B cell lines can be activated in vitro into the virus replicative cycle, where a large number of viral genes associated with EBV DNA replication and virus assembly are synthesised. Whilst EBV can be detected in throat washings from seropositive individuals, the only in vivo situation where full virus replication can be reliably observed is hairy leukoplakia (HL), a benign lesion of lingual epithelium frequently found in AIDS patients. Thus, the relative contribution of lymphoid cells and epithelial cells to latent EBV infection/persistence vs replication in vivo remains controversial. Recent studies suggest that HL represents a focus of EBV replication in the absence of a truly latent infection and this supports the contention that EBV persistence resides in the lymphoid compartment. These aspects together with the role of EBV in oral diseases and the effect of certain EBV genes on the control of epithelial cell growth and differentiation will be discussed.     

Prevalence of Epstein-Barr virus and human papillomavirus in oral mucosa of HIV-Infected patients

Epstein-Barr virus (EBV) has been implicated in the genesis of oral hairy leukoplakia (OHL). Initially, OHL was also associated with human papillomavirus (HPV) as evidenced by staining with antiserum to papillomavirus common structural antigens and reports of two HPV-positive OHL as detected by in situ DNA hybridization. The aims of this study were to determine the prevalence of EBV and HPV DNA in OHL and normal oral mucosa and to explain the basis for the staining of OHL tissues with antibodies to papillomavirus common structural antigens. EBV DNA was detected by in situ hybridization in 47 of 47 cases of OHL from human immunodeficiency virus (HIV)-seropositive individuals and in 1 of 10 biopsies of clinically normal buccal mucosa from the same group of individuals. Twenty-five of 35 OHL specimens stained with antibody to papillomavirus common structural antigens. There was no staining of two EBV-containing lymphoblastoid lines, indicating that the staining with anti-papillomavirus antibody was not due to antigenic cross-reactivity with EBV antigens. HPV DNA was detected by polymerase chain reaction amplification in 10 of 18 OHL specimens and in 6 of 10 normal buccal mucosa specimens. Our results indicate that EBV and HPV are present frequently in OHL and that HPV can be found regularly in histologically normal mucosa.     

A simple and rapid technique for the detection of Epstein-Barr virus DNA in HIV-associated oral hairy leukoplakia biopsies

A method of generating nucleic acid probes by polymerase chain reaction (PCR) for the detection of Epstein-Barr virus (EBV)-DNA by in situ hybridization in oral hairy leukoplakia (OHL) lesions is described. This method has the advantage over older methods of being cheaper, quicker and retaining sensitivity and specificity. Purified PCR products of Epstein-Barr virus DNA of 110 bp and 328 bp were labelled with biotin by nick translation or random primer labelling and were compared in in situ hybridization experiments with probes prepared by incorporation of biotin-labelled nucleotides in the PCR reaction mixture, with EBV viral DNA as a template. These probes were applied to 18 OHL tongue biopsies known to be positive for EBV-DNA, using a commercially available biotin-labelled BamHI "V" fragment EBV-DNA probe. To determine the specificity of the probes, we applied them to 20 normal tongue tissue samples and to 12 biopsies taken from keratotic tongue lesions from patients without risk factors for HIV infection and known to be negative for EBV-DNA. Clear positive signals for EBV-DNA were detected in all 18 cases of OHL biopsies using the amplimer of 328 bp labelled by PCR and random primer labelling. However, nick translation labelling was less efficient and sensitive. All control specimens were negative for EBV-DNA.     

Epstein-Barr virus detection in non-Hodgkin's lymphoma of the oral cavity: an immunocytochemical and in situ hybridization study

OBJECTIVE: The purpose of this study was to histologically characterize a series of oral non-Hodgkin's lymphomas (NHLs) and to investigate latent and lytic Epstein-Barr virus (EBV) infection in these. STUDY DESIGN: The revised European-American Lymphoma classification system (41) was used to categorize 58 cases of oral NHL, which included 9 immunosuppression-related NHLs. EBV infection was determined by in situ hybridization for Epstein-Barr virus-encoded RNA and by immunohistochemistry for the EBV antigens latency membrane protein, Epstein-Barr nuclear antigen-2 (EBNA2) and Z EBV replication activator protein. RESULTS: Most tumors were B-cell lymphomas (78%), but the proportion of T-cell lymphomas was surprisingly high (22%). The most common histologic subtypes were diffuse large B-cell lymphomas (45%), peripheral T-cell lymphomas (19%), and follicle center lymphomas (14%). Two thirds of the known immunosuppression-related NHLs were T-cell lymphomas. All of the immunosuppression-related tumors were EBV-infected, whereas the EBV infection rate in the NHLs of the remaining patients presumed to be immunocompetent was only 9%. Most EBV-positive tumors expressed neither of the latent antigens (ie, latency membrane protein and Epstein-Barr nuclear antigen-2), and coexpression of the 2 was observed only in immunosuppressed patients. Z EBV replication activator protein expression, which is indicative of replicative infection, occurred only in immunosuppressed individuals. CONCLUSIONS: Diffuse large B-cell lymphomas were the most common histologic subtype of oral NHLs, but T-cell lymphomas were relatively common and frequently occurred in states of immunosuppression. EBV may play a limited role in the initiation of lymphoma in the immunocompetent patient, but the virus may be of importance in progression of the disease in those patients with more aggressive tumors, as immunosuppression occurs.     

Epstein-Barr virus in oral hairy leukoplakia scrapes: identification by PCR

Oral hairy leukoplakia (OHL) is a lesion associated with a compromised immune system, and its diagnosis is determined by the demonstration of the presence of Epstein-Barr virus (EBV) in lesional tissue. The purpose of this article was to develop a simple technique to help the diagnosis of OHL, using PCR as an alternative technique to evidence EBV in scrapings. DNA samples were obtained by scraping the lateral border of the tongue of 38 adult patients: 29 HIV-positive patients (4 with clinical evidence of OHL; 4 with history of OHL, but without lesion at the moment the samples were collected; and 21 without clinical evidence of OHL), and 9 healthy volunteers for the control group. DNA was extracted from scrapes and amplified by PCR using specific primers for EBV. Of the 29 cases of HIV-positive patients, 22 (75.86%) were positive for EBV: 2 patients with clinical evidence of OHL, 4 patients with history of OHL, but without lesion at the moment the samples were collected, and 16 patients without clinical evidence of OHL. In the control group, samples of 5 (55.56%) healthy volunteers presented amplification for EBV. We concluded that the use of PCR in oral scrapes suggests a high sensitivity but low specificity for the diagnosis of OHL.     

Hairy leukoplakia: Epstein-Barr virus receptors on oral keratinocyte plasma membranes

Hairy leukoplakia (HL) is an oral white lesion associated with, and probably caused by, the Epstein-Barr virus (EBV) among persons who are seropositive for infection with human immunodeficiency virus. A unique feature of HL is its localization to the lateral portion of the tongue. To determine site differences for EBV receptors according to epithelial phenotype, these receptors were mapped in oral mucosa with the use of monoclonal antibodies HB5 and B2(specific for the Complement Fraction 3d/EBV receptor on B lymphocytes). Immunoperoxidase and immunofluorescence techniques were employed with the use of both cytologic suspensions and frozen tissue sections of oral epithelium. Pericellular plasma membrane immunoreactants were localized to upper spinous layer cells of the parakeratin phenotype; basal and parabasilar layers as well as all strata of orthokeratinized epithelia were negative. Those cells harboring EBV DNA as detected by in situ hybridization corresponded to cells with C3d/EBV receptors.     

Histopathologic and ultrastructural studies of oral mucosa with Candida infection

Eighteen oral mucosal biopsies with Candida infection were studied with light and electron microscopy. Under light microscopy, candidal infected oral mucosa was classified with epithelial hyperplasia, 15 cases and epithelial dysplasia, three cases. Four of 15 epithelial hyperplasias showed marked parakeratosis, and high grade acanthosis with many eosinophilic cells in the spinous cell layers. Epithelial dysplasia was characterized by atrophy of the spinous cell layers and increased nucleocytoplasmic ratio in the basal cell layers. Ultrastructurally, candidal infected oral mucosa showed numerous small desmosomes and the interdigitation of cytoplasmic membranes between spinous cells in both epithelial hyperplasia and epithelial dysplasia. Moreover, eosinophilic spinous cells, observed predominantly in epithelial hyperplasia showed intricate arrangement of dense tonofibrils. These ultrastructural findings seemed to give rise to mechanical strength between spinous cells in oral mucous epithelium with Candida infection. Results in this study suggest that excessive hyperplasia of candidal infected oral mucosa might be a protective reaction to the invasion of candidal pseudohyphae, but not associated with precancerous conditions.     

Epstein-Barr virus BMRF-2 and BDLF-3 expression in hairy leukoplakia

Hairy leukoplakia (HL) is a lesion found on the side of the tongue of immunocompromised individuals, including those with human immunodeficiency virus (HIV) infection. The lesion has unique histopathologic features and is characterised by high-level Epstein-Barr virus (EBV) replication, multiple EBV strains, and extensive inter-and intra-strain recombination. Expression of EBV genes spanning the entire viral life cycle from latency-associated genes to late, replicative genes has been detected in the lesion. HL thus provides a unique opportunity to study EBV expression in oral epithelium, and to study expression of novel EBV genes. We therefore constructed a cDNA library from an HL biopsy and detected expression of two genes not previously described in vivo: BMRF-2 and BDLF-3. Sequence analysis of the cDNAs revealed few amino acid changes from the B95-8 sequence. Expression of both genes was localized to the lower prickle cell layer of the tongue epithelium. BMRF-2 protein expression was primarily detected in the cell nuclei of the upper prickle cell layer. BDLF-3 protein expression was observed in the perinuclear space and Golgi compartment. The function of these proteins is currently under investigation.     

Reactive pocket epithelium in untreated chronic periodontal disease: possible derivation from developmental remnants of the enamel organ and root sheath

The pathological lining epithelium of destructive periodontitis was studied by analysis of the expression of intermediate filament proteins in biopsies of untreated advanced periodontitis. The cytokeratin (CK) pair 8/18 characteristic of simple epithelia was expressed consistently in a distribution pattern confined to the reactive pocket epithelium. The pattern of CK8/18 expression was complex with two broad presentations evident. In two-thirds of the advanced disease biopsies, the entire pathological lining epithelium was strongly reactive for both CK8 and CK18. In the remainder, the more superficial lining epithelium was mixed with foci of reactive and unreactive cells, with the deeper epithelium uniformly reactive. Only occasional highly localised reactivity for the simple keratins (CK8/18) was found in the lining epithelia of biopsies from minimally inflamed periodontal tissues. The pathological lining epithelium of advanced periodontitis was further characterised by the co-expression in basal layers of CK14, and of CK13 but not CK4, which are characteristic of suprabasal layers of stratified squamous epithelia. Cytokeratin 17, a marker of high turnover and migrating epithelial cells was extremely variable with no clear association between expression pattern and location of the epithelium ordisease status. There was no reactivity for CK10/11 typical of cornifying cells nor of vimentin, the characteristic intermediate filament of mesenchymal cells. The intermediate filament protein profile of the reactive lining epithelium was indistinguishable from the reactive epithelium present in three of five biopsies of periapical granulomas containing hyperplastic epithelium from activation of the developmental remnants of Hertwig's sheath, known as the cell rests of Malassez. The data reported are compatible with a contribution by remnants of developmental epithelium, including the reduced enamel epithelium and the cell rests of Malassez, to the reactive lining epithelium of the subgingival pocket in the pathogenesis of chronic periodontitis.     

Lack of evidence for local immune activity in oral hairy leukoplakia and oral wart lesions

BACKGROUND: Oral warts, caused by human papillomavirus (HPV), and oral hairy leukoplakia (OHL) caused by Epstein-Barr virus (EBV), are common oral manifestations in HIV-infected persons. Although both conditions occur most often with reduced blood CD4+ T-cell numbers, oral warts and OHL rarely occur simultaneously, suggesting that dysfunctions in other secondary local immune parameters are also involved. The present study evaluated tissue-associated proinflammatory and T-helper cytokine and chemokine mRNA expression and the presence of T cells in each lesion. METHODS: Biopsies were taken from lesion-positive and adjacent lesion-negative sites of HIV+ persons with oral warts or OHL and lesion-negative sites from HIV+ persons who were oral HPV or EBV DNA-positive (matched controls). Cytokine/chemokine mRNA expression was quantified by real-time polymerase chain reaction. CD3, CD4, and CD8 cells were identified by immunohistochemistry. RESULTS: No differences were detected in tissue-associated cytokine/chemokine mRNA expression in warts or OHL when compared to lesion-negative sites. Immunohistochemical analysis of T cells showed CD8+ cells exclusively, but few cells were present in either lesion. No differences were detected between lesion-positive and -negative control sites of each pathologic condition. CONCLUSION: Little evidence was found for local immune reactivity to either oral warts and OHL, suggesting that CD4+ T cells are a primary host defense against both oral warts and OHL, but with nonimmune factors potentially responsible for the divergent prevalence of each.     

Expression of alpha-defensin-1 in chronic hyperplastic candidosis

BACKGROUND: Chronic hyperplastic candidosis (CHC) represents a chronic opportunistic candida infection. We clarified the presence, localization and participation of alpha-defensin-1 in host response against chronic candidal stimulus. METHODS: Immunohistochemically stained CHC biopsies (n = 10) were compared to candida negative idiopathic leukoplakia (n = 10). RESULTS: In CHC alpha-defensin-1 was detected in neutrophils intravascularly, in lamina propria and in the epithelium, in part in intraepithelial microabscesses. Staining intensity of individual neutrophils varied and was associated with peri- and extracellular staining, in particular in the superficial epithelial cell layers. In controls only very few homogeneously staining neutrophils were detected intravascularly without any extracellular alpha-defensin-1 deposition. CONCLUSIONS: Neutrophils form microabscesses and respond to Candida by activation and release of alpha-defensin-1 to peri- and extracellular matrix. This together with the epithelial cell migration from the basal layer to epithelial surface leads to alpha-defensin-1 rich protective shield in the most superficial epithelial cell layers.     

Prevalence of oral hairy leukoplakia in 120 pediatric patients infected with HIV-1

Oral hairy leukoplakia (OHL) is an EBV (Epstein-Barr virus) opportunistic infection found in HIV-infected patients. It is an asymptomatic lesion that has an important prognostic value in AIDS. Differently from what takes place with HIV adult patients, OHL has been described in the literature as having a very small prevalence in pediatric patients. Therefore, the aim of this study was to investigate the prevalence of OHL in HIV pediatric patients using cytopathology. The sample consisted of 120 patients who were submitted to oral examination and had material scraped from both sides of their tongues. The diagnostic criterion was based on the identification of nuclear alterations. Clinical OHL was identified in two (1.67%) patients. The cytopathology revealed twenty (16.7%) cases of subclinical OHL. Our results show that in pediatric patients the prevalence of OHL may be larger than that described in the literature.     

Human papillomavirus and Epstein Barr virus in oral hairy leukoplakia among HIV positive Venezuelan patients

Oral hairy leukoplakia (OHL) is commonly found in individuals infected with HIV and represents the most frequent oral manifestation. The purpose of this study was to detect the presence of Human Papillomavirus (HPV) and Epstein Barr Virus (EBV) in OHL of HIV+ Venezuelan patients. We evaluated 21 HIV+ adult patients with clinically present OHL lesions: 11 under antiretroviral therapy, 10 without therapy, and 10 oral mucosal samples as controls. Nested-PCR was used to detect EBV and HPV infection. The INNO-LiPA HPV Genotyping v2 was applied to determine the HPV genotype. The EBV genome was found in 16/21 (76%) of the HIV+ patients with OHL. No difference was observed in EBV+ and EBV- patients related to antiretroviral therapy viral load and CD4+ Tcell coant. HPV-DNA was observed in 7/21 HIV positive cases (33%). The HPV genotypes detected were: 6, 11, 31, 33, 52, and 56/74. The most frequently HPV found was genotype 6 in 7/7, while two cases were HPV-11 and two HPV-52. Of the positive cases, 5/7 (71%) presented co-infection with more than one HPV genotype and 4/7 (57%) had HPV coinfection with high and low risk types. No case was EBV or HPV positive in the control group. In this study, a higher EBV prevalence was observed in OHL-HIV+ patients, confirming the etiologic role in this entity. A considerable number of cases were positive for HPV infection, and many patients presented coinfection with more than one HPV genotype as well as the presence of high oncogenic risk HPV in OHL.     

Human immunodeficiency virus-positive individuals with oral hairy leukoplakia are able to mount cytotoxic T lymphocyte responses to Epstein-Barr virus

OBJECTIVE: Oral hairy leukoplakia (OHL) is a white lesion of the tongue that is caused by Epstein-Barr virus (EBV) and occurs mainly in people infected with human immunodeficiency virus (HIV). The aim of this study was to determine whether the presence of OHL reflects the absence of EBV-specific cytotoxic T lymphocyte (CTL) activity. SUBJECTS AND METHODS: EBV-specific CTL responses were measured in HIV-positive homosexual men with OHL, HIV-positive homosexual men without OHL, and HIV-negative homosexual men. Also, the phenotypes of cells responsible for EBV-specific responses were studied. RESULTS: Eighty percent (8/10) of HIV-positive subjects with OHL, 52% (12/23) of HIV-positive subjects without OHL, and 83% (15/18) HIV-negative subjects had a positive anti-EBV CTL response (P = 0.004, Kruskal-Wallis test). Two HIV-positive subjects showed a greater anti-EBV CTL response after developing OHL than before the appearance of OHL Additional experiments showed that CD8-positive T cells and CD4-positive T cells were responsible for the EBV-specific CTL responses. CONCLUSION: Our data show more EBV-specific CTL activities in HIV-positive individuals with OHL than in HIV-positive individuals without OHL. Whether the presence of EBV-specific CTL contributes to resolution of OHL remains to be clarified.     

Detection of Epstein-Barr virus in oral scrapes in HIV infection, in hairy leukoplakia, and in Healthy non-HIV-infected people

Epstein-barr Virus (EBV) has been implicated in the pathogenesis of oral hairy leukoplakia (OHL). EBV is normally detected by lesional biopsy. The objectives of this study were to examine oral scrapes containing squamous epithelial cells from healthy non-HIV-infected people with and without clincial lesions of OHL, and from healthy non-HIV-infected controls, for EBV-DNA by the polymerase chain reaction (PCR). EBV-DNA was detected in 65% of HIV-infected individuals it was found as frequently in those without OHL as in those with. Moreover, EBV-DNA was not detected in all HIV-infected individuals, nor in all OHL as in those with. Moreover, EBV-DNA was not detected in all Hiv-infected individuals, nor in all OHL> The results suggest that the presence or absenc e of detectable EBV-DNA in oral scrapes, though a guide, cannot be regarded as absolutely reliable in the diagnosis or exclusion of OHL.     

Electron microscopic study of junctional and oral gingival epithelia in the juvenile and adult beagle dog

A previous study demonstrated structural differences in the junctional epithelium between juvenile and adult dogs. In juveniles the junctional epithelium showed some resemblances to the oral gingival epithelium, and a cuticular structure at the dento-gingival junction was of particular interest. The oral gingival epithelium is considered to be less permeable than the junctional epithelium. As cytoplasmic filaments are held to be the main component in the process of keratinization, and to have a stabilizing influence on the cells and tissues, the present investigation was designed to study the relative amounts of cytoplasmic filaments in the junctional and the oral epithelia of beagle dogs during juvenile and adult stages. In addition, the ultrastructure of the dento-gingival junction was characterized. Six beagle dogs were used. The material consisted of gingival biopsies sampled when the dogs were 3 and 12 months old, respectively. On these occasions the gingiva was in excellent health. The biopsies were prepared for electron microscopic analysis and three randomly selected fields were recorded photographically from each of the following epithelial strata: basal and granular cell layers of the oral epithelium and basal and superficial cell layers of the junctional epithelium. Morphometric analysis was performed in order to estimate the density of cytoplasmic filaments of the cells in these epithelial strata. The amount of cytoplasmic filaments was considerably lower in the cells of the junctional epithelium than in those of the oral epithelium. In the oral epithelium the amount increased from basal towards superficial cells, whereas no such increase was seen in the junctional epithelium. The pattern was the same in the juvenile and the adult stage. At the dento-gingival junction all dogs had a thick, laminated layer of dental cuticle material in the juvenile stage. In the adult stage a similar structure was seen only infrequently, and it never attained the thickness observed in the juvenile stage.