Expression of blood group-related glycoconjugates in the junctional and other oral epithelia of rodents.код для вставкиСкачать
THE ANATOMICAL RECORD 241:310-318 (1995) Expression of Blood Group-Related Glycoconjugates in the Junctional and Other Oral Epithelia of Rodents I.C. MACKENZIE, E. DABELSTEEN, G. RITTMAN, L. JUNGGREN, A N D H. TOH Dental Branch, University of Texas Houston Health Science Center, Houston, Texas (I.C.M., G.R.); School of Dentistry, University of Copenhagen, Copenhagen, Denmark (E.D., L.J.); Department of Oral Anatomy, Fukuoka Dental College, Fukuoka, Japan (H.T.) ABSTRACT Background: The junctional epithelium (JE) attaches the gingiva to the non-vital tooth surface and has other unusual properties which protect the underlying periodontal tissues. The J E differs from other gingival and oral epithelia in its unusual expression of cytokeratins typical of both stratifying and of simple epithelia, a phenotypic pattern possibly related to its specialized functions. Methods: The patterns of differentiation of rodent gingival and other epithelia were examined using monoclonal antibodies against various glycoconjugates which are expressed on epithelial cell surfaces and provide an alternative marker system for regionally-differing patterns of cell maturation. Results: Markers that are typical of basal cells in other stratifying epithelia were expressed by all cell strata of JE. J E lacked differentiation markers typical of other stratifying oral epithelia but showed suprabasal expression of markers typically expressed by simple epithelia and specialized epithelia, such as taste buds. Conclusions: The phenotype of rodent J E differs from that of other oral epithelia and the pattern of differentiation assessed by its expression of glycoconjugates parallels that for other phenotypic markers, such as cytokeratins. Differentiation of rodent J E is similar to that of human JE. The functional significance of these patterns of expression is not yet clear but the markers characterizing this unusual epithelium in rodents may be associated with its behavior in periodontal disease and of value to experimental studies of its development. o 1995 Wiley-Liss, Inc. Key words: Gingiva, Junctional epithelium, Rodents, Mucosa, Differentiation, Blood group antigens Oral mucosal epithelia show a regional diversity of structure that is presumably related to regional differences in function (Meyer et al., 1984).A unique feature of the oral mucosa is its perforation by teeth, and the patterns of organization of the gingival epithelia are particularly interesting because of their post-natal development during tooth eruption and their unusual relationship to the tooth. Ultrastructurally, the epithelia of the human gingival region have been classified as (a) the junctional epithelium (JE) which extends coronally from the region of the cemento-enamel junction, (b) the oral sulcular epithelium (OSE) which is continuous with the junctional epithelium and extends coronally from it to the region of the gingival crest, and (c) the oral gingival epithelium (OGE) which covers the oral surface of the gingiva from the gingival crest to the muco-gingival junction (Schroeder and Listgarten, 1977). The function of the gingiva is to provide a seal between the mineralized tooth surface and the surrounding soft tissues, and histological and ultrastruc0 1995 WILEY-LISS, INC tural studies indicate that the J E is a highly unusual epithelium that attaches to the non-vital calcified surface of the tooth and lacks the patterns of differentiation typical of other oral epithelia (Schroeder and Listgarten, 1977; Marks et al., 1994). The gingiva of rodents is similar to human gingiva in its pattern of organization into three distinct epithelial regions (Listgarten, 1975). Rodent J E is also ultrastructurally characterized by its lack of overt patterns of differentiation (Yamasaki et al., 1979; Shimono et al., 1989; Takata et al., 1986), the formation of specialized attachment to the tooth (Hormia et al., 1992; Sawada et al., 1990), and regeneration after its experimental removal (Schroeder and Listgarten, 1977; Takata et al., 1986). Received May 31, 1994; accepted August 8, 1994. Address reprint requests to Ian C. Mackenzie, Dental Branch, University of Texas Houston Health Science Center, 6516 John Freeman Avenue, Houston, TX 77030-3402. 311 DIFFERENTIATION OF GINGIVAL EPITHELIA Stratified squamous epithelia are continuously renewed by cell proliferation with the molecular markers that characterize regionally-specific patterns of differentiation being typically expressed when newly formed cells emigrate from the basal into the suprabasal strata (Purkis et al., 1990; Dabelsteen et al., 1991). Such expression can be demonstrated by staining with antibodies directed against cell differentiation products, such as cytokeratins or cell surface carbohydrates, and the expression of these markers provides a measure both of the regionally-differing phenotypes of oral epithelia, and of the stage of differentiation of particular cells within a n epithelium (Morgan et al., 1987; Purkis et al., 1990; Dabelsteen et al., 1991). Some differentiation products are common to the three human gingival epithelia but each epithelial region also expresses some markers that are absent from the other regions (Morgan, 1987; Steffensen e t al., 1987; Mackenzie et al., 1989; Mackenzie et al., 1991). The J E is thus not simply a n undifferentiated epithelium (Schroeder and Listgarten, 1977): evidence for its unusual pattern of cell maturation is found in the suprabasal expression of some keratins and cell surface carbohydrates absent from other gingival epithelia (Mackenzie et al., 1989; Mackenzie et al., 1991). Although structural studies indicate that there is a similar pattern of gingival organization in other mammals (Marks et al., 1994), the expression of differentiation markers has not been extensively investigated in non-human species. Lectin binding studies (Takata e t al., 1990) indicate that rodent J E lacks the suprabasal binding of lectins that characterize differentiation in the other gingival epithelia, but that there are differences in the lectin binding properties of basal and suprabasal J E cells indicative of some type of differentiation process. Rodent J E also appears similar to human J E in its suprabasal expression of some cytokeratins typically expressed only in simple epithelia (Mackenzie, 1989a, b). Cell surface glycoconjugates of the ABH series of antigens, are expressed on oral epithelial cells (Dabelsteen et al,. 1992), and staining with monoclonal antibodies (mAbs) with known specificities for human carbohydrate epitopes indicates that the pattern of expression of cell surface carbohydrates is consistently related both to the region of origin and the stage of differentiation of human epithelia (Mandel et al., 1992). Rodents are useful for experimental studies of the development and function of the dento-gingival region (Altman et al., 1987; Hemmerle and Frank, 1991; Hewage and Heany, 1990; Sashima e t al., 1991) and phenotypic markers that would permit more detailed studies of developmental and pathological changes of the gingival epithelia would be of value. The patterns of expression of cell surface carbohydrates in rodent gingiva were therefore examined for comparison with (a)those for various other rodent epithelia and (b) with the patterns reported for human gingiva. were sectioned through the molar region in the frontal plane to provide a range of oral and other epithelia for examination. Tissues were fixed in buffered 4% formaldehyde prior to brief decalcification in 1% SSFA, dehydrated in graded alcohols, passed through xylene, and wax embedded. Sections were cut a t 6 pm, mounted on gelatine-subbed slides, and hydrated prior to staining. In preliminary work, a wide range of mAbs was tested for binding to murine tissues. The mAbs used in this study were selected for their differential staining of gingival epithelia and are listed in Table 1. These mAbs have defined specificities for various carbohydrate determinants associated with human Type I1 or Type I11 chain peripheral cores; information about their origins and specificities, and about the staining methods that were used, have been given in detail in previous publications (Dabelsteen and Clausen, 1992; Dabelsteen et al., 1991; Mackenzie et al., 1989; Gao et al., 1989).In brief, mAbs were applied to sections as the undiluted supernatant and were incubated overnight at 4°C. After three washes in phosphate buffered saline, pH7.4, sections were incubated with a 1:lOO dilution of polyclonal rabbit antibodies directed against murine immunoglobulins and conjugated to f luorescein isothiocyanate, alkaline phosphatase, or peroxidase (DAKO). To unmask antigenic determinants blocked by sialic acid, some sections of each specimen were incubated in neuraminidase (Sigma, Type 11) prior to staining. Fluorescent specimens were examined by epi-illumination with appropriate barrier filters. Specimens for histochemical demonstration of mAb binding were processed to display enzyme activity, as previously described (Mackenzie et al., 1989; Mackenzie et al., 1991), and examined by transmitted light. Controls for staining specificity consisted of (a) omission of the primary antibody (b) staining with irrelevant primary antibodies of similar isotypes, and (c) comparison of regional binding differences between different antibodies. For each strain or species at least five different gingival samples were stained with each primary antibody: most specimens were stained using both fluorescent and histochemical methods for demonstrating mAb binding. RESULTS For each of the mAbs used, the staining methods provided relatively clear binding patterns with little non-specific background staining. The strongest binding was typically to the cell surface region of epithelial cells and, occasionally, to vascular elements within the connective tissue. Some mAbs showed relatively strong binding to material on the epithelial surface, possibly blood or salivary proteins, and most showed binding to remnants of enamel matrix when present. Essentially similar staining patterns were observed for both the Balb/C and the C3H mice and, except where noted, the following description applies equally to both strains. MATERIALS AND METHODS Murine Gingival Epithelia Specimens were collected from young adult C3H/Fej or Balb/C mice, and from Sprague-Dawley rats. Gingival tissues were dissected from the jaws en bloc together with the molar teeth and the adjacent alveolar bone. Some mouse heads were examined whole and The mAbs against Type I1 chain determinants showed epithelial binding patterns that varied both regionally and with cell maturation (Table 1).The mAbs against Lewis antigens X and Y (Le", Ley) both consistently bound to a subpopulation of cells in the J E and, 312 I.C. MACKENZIE ET AL. TABLE 1. Summary of staining data for junctional (JE), oral sulcular (OSE), oral gingival (OGE), palatal (PI, buccal (B) and simple (S) epithelia and taste buds (TB). Staining is indicated as absent (-) or basally (B), parabasally (PB) or suprabasally (SB) located in each epithelium. Staining for simple epithelium and taste buds is recorded only as either present ( + ) or absent (-1. The data shown are for C3H mice. Some small differences between C3H and Balb/C mice were noted and are discussed in the text. The pattern for the rat was essentially similar to that for mice but some differences were noted and are discussed in the text. Determinant Type I1 Chain Lac Sa-Lac H Le” Ley Lea Leb B Type I11 Chain Tn Sa-Tn T Sa-T H mAb JE 1B2 1B2( + N) BE2 SH1 AH6 SB B+SB - SB SB SB - Serotec B+SB OSE OGE P B S TB + + + + + + + - - - - - PB SB PB SB PB SB PB SB - - - - - - - - SB B SB B+PB SB B+PB SB B Lu35 SB - - - - TKHZ HH8 HH8( + N) MBRl - SB B B SB SB B+PB B+PB SB SB B+PB B+PB SB SB B B SB B+SB B+SB - in most sections, binding could be clearly localized to the upper part of the J E in the region where i t abutted the stratum corneum of the OSE (Fig. lA,B). Usually, staining for Ley was stronger and more extensive than that for Le”. In some sections there appeared to be weak binding to a few suprabasal cells in the deeper (most apical part) of the OSE and, occasionally, weak staining of granular cells of the OGE was present (Fig. 1A). Typically, however, there was no staining of either the OSE or the OGE. Some structures in the sub-gingival connective tissue were occasionally stained (Fig. lA,B) but i t was difficult to determine whether these were vascular elements or epithelial rests. The mAb against N-acetyl-lactosamine (Lac) showed moderate binding to the area of J E that stained for Le” and Ley (Fig. 1C). Staining after treatment with neuraminidase indicated that sialated N-acetyl-lactosamine (Sa-Lac) was present in all epithelial regions. In the OSE and OGE there was weak staining restricted mainly to cells of the parabasal regions, but the J E showed weak binding to basal cells and strong binding to suprabasal cells (Fig. 1D). The antibody against BGA “B” showed a similar pattern of binding to the full thickness of the J E and restriction to the basal and parabasal cells of other epithelia (Fig. 1E). Staining for Type I1 chain “H” was present at the cell surfaces of suprabasal cells of the OSE and OGE but was absent from J E (Fig. 1F). Epithelial staining for Type I11 chain determinants similarly differed with anatomical region and degree of maturation. Weak staining for Tn antigen was occasionally present on the basal and parabasal cells of OSE and OGE but was usually restricted to the region of the J E that stained for Le” and Ley (Fig. 1G). Sialated Tn (Sa-Tn) was present on all suprabasal cells of the OSE and OGE but was absent from the J E except for weak staining of the most coronal portion (Fig. 1H). Staining for unsialated (T) and sialated T-antigen (Sa-T) was present on basal and parabasal cells of the OSE and OGE. Staining for both forms was present on all strata of the J E but was strongest on basal cells. Staining for Type I1 chain “H” was present on the su- + + + + - prabasal cells of OSE and OGE but was absent from the J E. Other Murine Epithelia The buccal and palatal epithelia of C3H mice showed patterns of glycoconjugate expression that were essentially similar to those of OSE and OGE. No major differences were noted between these epithelia except that palatal and oral gingival epithelia differed consistently from buccal and oral sulcular epithelia in the level of transition of staining from “basal” to “suprabasal” markers. The former epithelia, as in the rat (see below), showed this transition at a higher level in the epithelium. Balb/C mice differed from C3H mice (Fig. 11) in lacking expression of Type I11 chain “H” in the buccal epithelium. The epithelium of the dorsal surface of the tongue showed binding patterns that varied in relation to the structure of the filiform papillae: the interpapillary regions and the posterior cell columns of the papillae showed patterns essentially similar to buccal and palatal epithelia, respectively. However, the suprabasal cells of the anterior column of the papillae differed from the other regions in its positive staining for T, Tn, and Le” (Fig. 2B). In Balb/C, but not C3H mice, this region also stained for Ley. The simple epithelia lining the ducts of glands and the nasal surface of the palate showed markedly different staining patterns from the stratifying epithelia lining the oral cavity. Typically, they showed strong staining for Lac, Sa-Lac, Le” (Fig. 2A), Ley, T, and Tn antigens. Weaker staining was seen for Lewis antigens A and B (Lea, Leb) and B antigen but there was no staining for Type I1 or I11 chain H, or sialated Tn antigens. The staining of these epithelia was usually fairly uniform but some mAbs, such as those against Ley and Lex, stained subpopulations of cells (Fig. 2A). C3H mice lacked the staining for B antigen seen in a subpopulation of basally-positioned cells in the simple epithelia of Balb/C mice. Taste buds showed a pattern of staining for Type I1 chain determinants that was more similar to simple and junctional epithelia than to Fig. 1. (A-H) Sections of the murine gingiva buccal to the first molar tooth and of the mucogingival region (I) stained with mAbs against various determinants of the ABO blood group antigen series. For orientation, the position occupied by the enamel of the tooth crown prior to decalcification is marked (E) on each figure. (A) Staining for Ley is restricted mainly to the upper part of the J E facing the OSE. A few suprabasal cells in the deeper part of the OSE often showed some staining (arrow) and, occasionally, there was weak staining of stratum granulosum cells in the OGE. Some cell clusters (*) in the connective tissue, possibly vascular elements, were also stained, as were remnants of the decalcified enamel matrix. (B) The pattern of staining for Le” was similar to that for Ley but more restricted to the upper part of the J E and absent from the OGE. (C) Staining for Lac was similarly restricted to the upper part of the J E . The plane of the section shown passes obliquely through the J E and thus shows staining of a wider band of cells. (D)After neuraminidase treatment, more extensive staining for Lac is seen. There is weak staining of parabasal cells in OGE and OSE weak basal and strong suprabasal staining of the JE. (E) Moderately strong staining for BGA “B’ was seen in the basal and parabasal cells of the OGE with weaker staining of the OSE. The full thickness of the J E is stained with the strongest staining seen in the coronal region. (F)Staining for Type I1 chain “ H is restricted to suprabasal cells in the OGE and OSE. (GI Staining for Tn is similarly restricted to the upper part of the J E with a pattern similar to that for Le“ and Ley. (H) Staining for sialated-Tn shows a similar but more extended pattern of suprabasal staining of OGE and OSE than for “H” (F).(I) BalbiC differed from C3H mice in the lack staining of the buccal epithelium for Type I1 Chain “ H . This section, through the region of mucogingival junction (arrow), shows suprabasal staining of the OGE but no staining of the buccal epithelium (B). 314 I.C. MACKENZIE ET AL. Fig. 2. Sections of C3H mouse tissues. (A) Simple epithelium lining of ducts of a sublingual salivary gland shows staining of a subpopulation of cells for Le”. (B) Dorsum of tongue stained for Ley shows strong staining of the suprabasal cells in the anterior columns of the filiform papillae. (C) A taste bud at the lateral border of the tongue stained with hematoxylin and eosin to show its general structure. (D) Section adjacent to that shown in (C) shows suprabasal staining of the mucosal epithelium but lack of staining of the taste bud for Type I1 chain “H”. (E & F) Strong staining for Le” and L e y is restricted to the taste bud. the other oral epithelia. In particular, they stained strongly for Le” and Ley (Fig. 2E,F). They did not stain for Type I11 chain determinants. staining of basal as well as parabasal cells in the OSE and OGE (Fig. 3B). As in the mouse, staining for Le” and Ley was restricted to the junctional epithelium, and staining for Type I1 and Type I11 chain H was restricted to suprabasal cells of the OSE and OGE and was absent from the J E (Fig. 3C,D). Staining for Sa-T was present on the basal and parabasal cells of the OSE and OGE but, unlike the mouse, suprabasal staining of the J E was absent and a loop of basal cells, similar to that seen stained with BGA “B”, passed up onto the surface of the tooth (Fig. 3E). A further difference between the staining patterns for rat and mouse was seen in the lack of rat mucosal staining for T antigen. In the rat, the region of the muco-gingival junction clearly showed the different levels of transition from “basal” to “suprabasal” markers in masticatory and lining mucosae. In the buccal epithelium, staining for Type I1 and I11 chain “ H was restricted to parabasal cells, whereas in the OGE i t was seen relatively high in Rat Epithelia In the rat, only epithelia of the gingival region were examined and these generally showed similar staining patterns to those of the mouse. Being larger, however, rat gingiva showed regional patterns of staining variation more clearly. In the OSE and OGE, staining for BGA “B” was restricted to basal and parabasal cells. The coronal portion of the J E showed strong staining of all strata for BGA “B” but in the apical region suprabasal staining was weaker and only a “loop” of basal cells passing up onto the attachment with the tooth surface was strongly stained (Fig. 3A). Staining for SaLac was also strongest in the coronal part of the J E where the full thickness of the epithelium was stained but staining differed from mouse in that there was Fig. 3. Sections of rat gingival tissues with the position of the enamel space marked (E). (A) Staining for BGA “B” is present on the basal and parabasal cells of the OGE and OSE and there is staining of the full thickness of JE with particularly strong staining of the coronal region, as in the mouse. (B) Staining for “Lac” after neuraminidase treatment is similar to that for the mouse with weak staining of basal and parabasal cells in the OGE and OSE and staining of the full thickness of the J E with strongest staining of the coronal region. (C) Staining for Type I1 chain “ H shows strong suprabasal staining of the OGE and OSE and lack of J E staining. (D) A similar pattern of staining is seen for Type I1 chain “H’. (El Staining for “T” antigen after neuraminidase treatment is restricted to basal and parabasal cells in the OGE and OSE. In the JE, staining is restricted to a small loop of basal cells a t the apical tip which includes cells adjacent to the tooth (arrow). 316 I.C. MACKENZIE ET AL. Fig. 4. Section of the mucogingival junction of rat. The region of junction between OGE and buccal (Bt epithelium is marked by arrows. Small but consistent differences in the levels of expression of basal and differentiation markers are seen between these epithelia. Both Type I1 chain and Type I11 chain “ H (A,B) are expressed in the the epithelium (Fig. 4A,B). Reciprocally, staining for the “basal” markers BGA “B” and Sa-T was more closely restricted to basal cells in the buccal epithelium and extended further suprabasally in the OGE (Fig. 4C,D). OGE at a higher level than in the buccal epithelium where expression is confined to the parabasal region. Reciprocally, the expression of the basal markers BGA “ B and Sa-T (C,D) extends well above the basal region in OGE but is more basally restricted in buccal epithelium. quite well characterized (Dabelsteen and Clausen, 1992). The ABH series of antigens within the ABO blood group system are termed “blood group antigens” because of their initial discovery as agents associated with the agglutination of human red blood cells. SubDISCUSSION sequently, i t has been shown that such antigens are The various staining methods employed produced expressed by various other mammalian tissues and clear patterns of epithelial staining and, for each of that, phylogenetically, their expression on mucous the mAbs used, demonstrated consistent regional dif- cells, nervous receptors, and epithelia in fact precedes ferences in the epithelial staining patterns. Lack of their expression on erythrocytes which is found only in epithelial staining with the irrelevant mAbs used humans (Oriol et al., 1992). Human epithelia show exas controls, and the consistent regional distribution pression of short carbohydrate structures on basal epof staining with mAbs against known human glycocon- ithelial cells and a stepwise elongation is associated jugates, were indicative of recognition of specific with cell maturation (Dabelsteen et al., 1991). Antiepitopes. Synthesis of cell surface carbohydrates is ge- gens of the ABH series have been previously demonnetically determined by the presence or absence of var- strated on rodent mucosal epithelia but their sequence ious glycosyl transferases (Clausen et al., 1992) which of expression in relation to differentiation differs from control the stepwise addition of specific carbohydrate that in human tissues: the longest structures are seen linkages to the terminal regions of protein or lipid core on basal cells and expression of shorter structures ocstructures (Mandel et al., 1992). Such antigenic carbo- curs during maturation (Reibel e t al., 1994). The mohydrate moieties may be carried on various types of lecular structures recognized in murine tissues by the protein or lipid cores and the formation of these mole- mAbs used in the present study are not yet well defined cules depends upon complex patterns of interaction be- and characterization of these rodent antigens will retween substrate availability and the cellular expres- quire further work. However, the staining patterns insion of particular glycosyl transferases (Clausen et al., dicate that they recognize epitopes that are consis1992). For human tissues, the particular core struc- tently expressed by rodent epithelia and thus provide tures and epitopes recognized by these mAbs have been suitable markers of epithelial differentiation. DIFFERENTIATION O F GINGIVAL EPITHELIA In terms of their patterns of epithelial expression in mice, these antigens appear to provide three general groups of markers. One group (Sa-Lac, B, T, Sa-T) is expressed by basal and early differentiating epithelial cells, by simple epithelia and, variably, by taste buds. This group of antigens is expressed by all cells of the JE. A second group (Leb, Sa-Tn, Type 11, and Type I11 chain H) is expressed by differentiating suprabasal cells of stratifying epithelia and, except for Leb, is not expressed by simple epithelia and taste buds. This group is not expressed by JE. A third group (Lac, Lex, Ley, Lea, Tn) is not expressed by stratifying epithelia but is expressed by simple epithelia and usually by taste buds. This group of antigens is expressed suprabasally by differentiating cells in JE. The J E pattern of differentiation, assessed by its staining for carbohydrates, thus clearly differs from that of the other gingival epithelia and is generally distinguished by (a) its expression on all cell strata markers that are typically expressed only on the immature cells of other stratifying epithelia, (b) its lack of expression of markers that are expressed suprabasally in other stratifying epithelia and which characterize “oral” patterns of differentiation, and (c) its expression of markers typically expressed by simple epithelia but not by stratifying epithelia. It is interesting that this pattern for carbohydrate marker molecules parallels the pattern of expression of cytokeratins in gingival epithelia. In terms of cytokeratin expression, the phenotype of OGE corresponds to that of masticatory mucosa and that of the OSE to lining mucosa (Morgan et al., 1991; Mackenzie et al., 1991) whereas, in both humans and rodents, J E shows full-thickness staining for cytokeratins found in basal cells of other stratifying epithelia, lack of staining for most cytokeratins expressed suprabasally in other oral epithelia, and staining for cytokeratins usually expressed only in simple epithelia (Mackenzie et al., 1991; Mackenzie, 1989). This pattern is similar to that seen for the specialized epithelial cells in taste buds which were found to share several differentiation markers with simple epithelia and have been shown previously to express simple epithelial keratins (Toh et al., 1992). The arrangement of hard and soft tissues a t the region of junction between the oral mucosa and the teeth is a n anatomical peculiarity. Elsewhere in the body, epithelia form a continuous covering of the underlying tissues which is rapidly restored when continuity is disrupted, for example, by trauma. During tooth development (Ten Cate, 1989), odontogenic epithelium differentiates into the enamel organ, which defines the morphology of the tooth crown and secretes enamel, and then grows apically a s Hertwig’s epithelial root sheath to define root morphology. At the completion of development, the root sheath disintegrates into the rests of Malassez and the enamel organ collapses to become the reduced enamel epithelium (REE) which covers the crown of the unerupted tooth. This pattern of development results in a “free edge” of epithelium at the region of the cemento-enamel junction of the tooth. As the tooth erupts, the REE, which forms the primary epithelial attachment to the tooth, proliferates and fuses with the oral mucosa. Subsequently, the cells against the surface of the tooth lose the appearance of reduced ameloblasts and acquire the typical appear- 317 ance of JE. It is not entirely clear whether the cells that initially form the J E are derived from proliferation and transformation of REE cells, as ultrastructural evidence suggests (Schroeder and Listgarten, 1977), or from downgrowth of oral epithelial cells (McHugh, 1957). However, whatever its initial cellular derivation, J E is regenerated, apparently from the adjacent oral mucosal epithelium, after its surgical removal (Schroeder and Listgarten, 1977). As yet, little is known about the connective tissue or other factors that are associated with the formation and reformation of J E . The pattern of cell maturation occurring in other stratified oral epithelia is associated with the formation of a relatively inert surface barrier but the retention of basal markers through the full thickness of J E indicates its lack of normal maturation. As previously pointed out by Squier (1981),development of a basal lamina against the tooth surface requires “basal” or immature cell properties and normal maturation of J E cells would inhibit adherent interactions with the tooth. Cell surface glycoconjugates have been associated with intercellular adhesion (Clausen et al., 1992) and the development of strong intercellular bonds, as occurs during the maturation of other stratifying epithelia, would prevent the type of cell movement necessary for replacement of the JE. Phagocytic and lysomal activities of J E cells also play a role in the maintenance of gingival health (Takata et al., 1986) and it appears that such functions would similarly be impaired by normal patterns of differentiation. However, the J E is not undifferentiated: it shows a n alternative pathway of differentiation with suprabasal synthesis of macromolecules characteristic of simple epithelia and taste buds whose cells are also metabolically active in the mature state. It may be significant that both taste buds and J E , but not other oral epithelia (Nagata et al., 1992), have a high level of innervation. The expression of some glycoconjugates, such as Lex, may facilitate or direct leukocyte traffic through the J E (Dabelsteen and Clausen, 1992). J E properties such as attachment, permeability, and adhesion may help stabilize the integrity of the dentogingival attachment (Page and Schroeder, 1982; Marks et al., 1994) but i t is apparent that the epithelial discontinuity produced by tooth eruption is the cause of vulnerability to inflammation which often progresses, in both humans and other mammals, to breakdown of the supporting structures of the teeth (Page and Schroeder, 1982). Inflammation generates cytokine networks that typically stimulate proliferation and migration of epithelia (Mauch e t al., 1994) and the process of periodontal pocket formation is invariably associated with apical migration of the JE. It is unclear what mechanisms normally prevent downward migration of the “free edge” of epithelium from the cemento-enamel junction onto the root of the tooth (Mackenzie, 1989a,b), but some evidence suggests that the J E phenotype is also associated with “passive” or non-migratory properties which may make i t resistant to migratory stimuli derived from the inflammatory response (Mackenzie, 1987,1989a,b). Further information about the properties of this epithelium, and its mechanisms development and regeneration, would therefore be of value, and the identification of markers characterizing 318 I.C. MACKEN ZIE ET AL. the J E phenotype in rodents should enable further experimental studies of these questions. ACKNOWLEDGMENTS This work was supported in part by NIH/NIDR Grants DE05190 and DE10317. LITERATURE CITED Altman, L.C., C.L. Nelson, B. Povolny, P. Fleckman, B.A. Dale, R.V. Maier, C. Soderland, and C. Baker 1987 Culture and characterization of rat junctional epithelium. J . Periodont Res., 23:91-99. Braga, A.M., and C.A. Squier 1980 Ultrastructure of regenerating junctional epithelium in the monkey. J. Periodontol., 51:386392. Clausen, H., E.C. Bennett, and E. Dabelsteen 1992 Carbohydrates of the cell surface: Molecular aspects of glycosyltransferases and their genes. APMIS Supplement 27,100:9-17. Dabelsteen, E., and H. Clausen 1992 Carbohydrate Pathology. APMIS Supplement 27, Vol 100. Munksgaard, Copenhagen. Dabelsteen, E., U. Mandel, and H. Clausen 1991 Cell surface carbohydrates are markers of differentiation in human oral epithelium. Critical Reviews in Oral Biology and Medicine, 2:493-507. Gao, Z., I.C. Mackenzie, A.T. Cruchley, D.M. Williams, I. Leigh, and E.B. Lane 1989 Cytokeratin expression of the odontogenic epithelia in dental follicles and developmental cysts. J. Oral Pathol., 18:63-67. Gao, Z., and I.C. Mackenzie 1992 Patterns of phenotypic expression of human junctional, gingival and reduced enamel epithelia in vivo and in vitro. Epith. Cell Biol., Ir156-167. Hemmerle, J., and R.M. Frank 1991 Bacterial invasion of periodontal tissues after experimental immunosuppression in rats. J . Biol. Buccal., 19t271-282. Hewaee, S., and T.G. Heaney 1990 Creation of a chimaeric periodonti<m in the rat by isotbpic tooth germ transplantation. Archs Oral Biol., 35r681-684. Hormia, M., I. Virtanen, and V. Quaranta 1992 Immunolocalizationof integrin asp4in mouse junctional epithelium suggest an anchoring function to both the internal and the external basal lamina. J . Dent. Res., 718:1,503-1,508. Listgarten, M.A. 1972 Ultrastructure of the dento-gingival junction after gingivectomy. J. Periodont. Res., 7:151-160. Listgarten, M.A. 1975 Similarity of epithelial relationships in the gingiva of rat and man. J. Periodontal., 46:677-680. Mackenzie, I.C. 1987 Nature and mechanisms of regeneration of the junctional epithelial phenotype. J . Periodont. Res., 22:243-245. Mackenzie, I.C. 1990 Formation and sequential changes of the periodontium. In: Contemporary Periodontics. H.M. Goldman, D.W. Cohen, and R.J. Genco, eds. C.V. Mosby, St. Louis, pp. 55-62. Mackenzie, I.C. 1989b Factors influencing the stability of the gingival sulcus. In: Periodontology Today. B. Guggenheim, ed. Karger, Basel, pp. 1-8. Mackenzie, I.C. 1993 Epithelial-mesenchymal interactions in the development and maintenance of epithelial tissues. In: The Keratinocyte Handbook. B. Lane, I. Leigh, and F. Watt, eds. Cambridge University, pp. 243-258. Mackenzie, I.C., E. Dabelsteen, and U. Mandel 1989 Expression of blood group antigen-related carbohydrates by human gingival epithelia. J . Periodont. Res., 24r289-297. Mackenzie, I.C., G. Rittman, Z. Gao, I. Leigh, and B. Lane 1991 Patterns of cytokeratin expression in human gingival epithelia. J. Periodont. Res., 26t468-478. Mackenzie, I.C., and Z. Gao 1993 Patterns of cytokeratin expression in the epithelia of inflamed human gingiva and periodontal pockets. J . Periodont. Res., 28.49-59. Mandel, J., O.W. Petersen, H. Sorensen, P. Vedtofte, S. Hakomori, H. Clausen, and E. Dabelsteen 1992 Simple mucin-type carbohy~~ % drates in oral stratified squamous and salivary gland epithelia. Carbohydrate Pathology, 97:713-721. Marks, S.C. Jr., M.D. McKee, S. ZalZal, and A. Nanci 1994 The epithelial attachment and the dental junctional epithelium: Ultrastructural features in porcine molars. The Anatomical Record, 238:l-14. Mauch, C., T. Oono, B. Eckes, and T. Krieg 1994 Cytokines and Wound Healing. In: Epidermal Growth Factors and Cytokines. T.A. Luger and T. Schwarz, eds. Marcel Dekker, Inc., New York, pp. 325-344. McHugh, W.D. 1963 Some aspects of the development of gingival epithelium. Periodontics, 1:239-245. Meyer, J.,C.A. Squier, S.J. Gerson 1984 The Structure and Function of Oral Mucosa. Pergamon Press, Oxford. Morgan, P.R., I.M. Leigh, P.E. Purkis, I.D. Gardner, G.N.P. Van Muijen, and E.B. Lane 1987 Site variation in keratin expression in human oral epithelia-an immunocytochemical study of individual keratins. Epithelia, 1:31-43. Nagata, E., T. Kondo, N. Ayasaka, M. Nakata, and T. Tanaka 1992 Immunohistochemical Study of Nerve fibres with Substance P- or Calcitonin Gene-Related Peptide-Like Immunoreactivity in the Junctional Epithelium of Developing Rats. Archs Oral Biol., 37: 655-662. Oriol, R., R. Mollicone, P. Coullin, A.M. Dalix, and J.J. Candelier 1992 Genetic regulation of the expression of ABH and Lewis antigens in tissues. APMIS Suppl. 27, vol. 100:28-38. Page, R.C., and H.E. Schroeder 1982 Periodontitis in Man and Other Animals. A Comparative Review. S. Karger, Basel. Purkis, P.E., J.B. Steel, I.C. Mackenzie, W.J.B. Nathrath, I.M. Leigh, and E.B. Lane 1990 Antibody markers of basal cells in complex epithelia. J . Cell Sci., 97:39-50. Reibel, J., E. Dabelsteen, S. Hakomori, W.W. Young, and I.C. Mackenzie 1984 The distribution of blood group antigens in rodent epithelia. Cell Tissue Res., 237.111-116. Sawada, T., T. Yamamoto, T. Yanagisawa, S. Takuma, H. Hasegawa, and K. Watanabe 1990 Electron-immunocytochemistry of laminin and type-IV collagen in the junctional epithelium of rat molar gingiva. J. Periodont. Res., 25:372-376. Sashima, M., M. Satoh, and A. Suzuka 1991 Age-related development of the long-junctional epithelium in the senescence-accelerated mouse. J . Periodont. Res., 70:1,462-1,466. Schroeder, H.E., and M.A. Listgarten 1977 Fine structure of the developing epithelial attachment of human teeth. In: Monographs in Developmental Biology, Vol. 2 A. Wolsky, ed. Karger, Basel. Shimono, M., T. Inoue, Y. Hamada, Y. Abiko, and S. Hashimoto 1989 A cytochemical study of dense granules in the rat junctional epithelium. J . Periodont. Res., 24:186-191. Squier, C.A. 1981 Keratinization of the sulcular epithelium-a pointless pursuit? J. Periodontol., 52r426-429. Steffensen, B., D.E. Lopatin, R.G. Caffesse, and C.T. Hanks 1987 Blood group substances as differentiation markers in human dento-gingival epithelium. J. Periodont. Res., 22r451-455. Takata, T., H. Nikai, N. Kjuhin, and H. Okamoto 1986 Ultrastructure of regenerated junctional epithelium after surgery of the rat molar gingiva. J . Periodontol., 57:776-783. Takata, T., H. Nikai, M. Miyauchi, H. Ito, J . Kobayashi, and N. Ijuhin 1990 Lectin binding of rat gingival epithelia. J . Periodont. Res., 25:151-155. Ten Cate, A.R. 1989 Oral Histology. Development, Structure and Function, Fourth Edition. Mosby, St. Louis. Thaller, C., and G. Eichele 1988 Characterization of retinoid metabolism in the developing chick limb bud. Development, 103r473483. Toh, H., G. Rittman, and I.C. Mackenzie 1992 Kertain expression in taste bud cells of the circumvallate and foliate papillae of adult mice. Epith. Cell Biol., 2.126-133. Yamasaki, A,, H. Nikai, K. Niitani, and N. Ijuhin 1979 Ultrastructure of the junctional epithelium of germfree rat gingiva. J . Periodontol., 50.641-648.