HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 38, 35371-35377, September 20, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/38/35371    most recent M205143200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Santos, M. Articles by Jorcano, J. L. Search for Related Content PubMed PubMed Citation Articles by Santos, M. Articles by Jorcano, J. L. Social Bookmarking                      What's this?

Severe Abnormalities in the Oral Mucosa Induced by Suprabasal Expression of Epidermal Keratin K10 in Transgenic Mice^*

Mirentxu Santos, Ana Bravo^§, Ceferino López^§, Jesús M. Paramio^¶, and José L. Jorcano

From the  Project on Cell and Molecular Biology and Gene Therapy, CIEMAT Av. Complutense 22, E-28040 Madrid, Spain and the ^§ Department of Animal Pathology, Veterinary School, University of Santiago de Compostela, 27002 Lugo, Spain

Received for publication, May 24, 2002, and in revised form, July 8, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Previous studies have demonstrated that keratin K10 plays an important role in mediating cell signaling processes, since the ectopic expression of this keratin induces cell cycle arrest in proliferating cells in vitro and in vivo. However, apart from its well known function of providing epithelial cells with resilience to mechanical trauma, little is known about its possible roles in nondividing cells. To investigate what these might be, transgenic mice were generated in which the expression of K10 was driven by bovine K6 gene control elements (bK6hK10). The transgenic mice displayed severe abnormalities in the tongue and palate but not in other K6-expressing cells such as those of the esophagus, nails, and hair follicles. The lesions in the tongue and palate included the cytolysis of epithelial suprabasal cells associated with an acute inflammatory response and lymphocyte infiltration. The alterations in the oral mucosa caused the death of transgenic pups soon after birth, probably because suckling was impaired. These anomalies, together with others found in the teeth, are reminiscent of the lesions observed in some patients with pachyonychia congenita, an inherited epithelial fragility associated with mutations in keratins K6 and K16. Although no epithelial fragility was observed in the bK6hK10 oral epithelia of the experimental mice, necrotic processes were seen. Collectively, these data show that the carefully regulated tissue- and differentiation-specific patterns displayed by the keratin genes have dramatic consequences on the biological behavior of epithelial cells and that changes in the specific composition of the keratin intermediate filament cytoskeleton can affect their physiology, in particular those of the oral mucosa.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Keratin intermediate filaments (KIFs)¹ are present in the cytoplasm of all epithelial cells as heteropolymers of type I and type II keratin polypeptides. Type I and type II keratin genes display highly regulated expression patterns in a pairwise and differentiation-specific fashion (1-3). The role of KIF in epithelial cells and tissues remained elusive until the discovery, through studies with transgenic mice and the finding of mutations affecting keratin proteins in dominantly inherited epithelial fragility syndromes (4-8), that keratins impart mechanical resilience to cells. This appears to be a function shared by the majority of the keratin family. Therefore, the changes in keratin expression observed during differentiation (or in certain situations such as in tumor growth or wound healing involving stratified epithelia) probably indicate subtle, cell type-specific differences in function among these polypeptides. Further keratin-specific functions should not be discarded.

Previous studies have demonstrated that K10 has specific functions. This keratin replaces K14 as skin keratinocytes enter the terminal differentiation program and become postmitotic (9). In addition, K10 expression is severely reduced under hyperproliferative situations, such as in wound healing and epidermal tumors. We have previously demonstrated that forced K10 expression in cultured cells induces cell cycle arrest through a mechanism that requires a functional retinoblastoma gene (10). This process seems to take place by impairing the activation of Akt and protein kinase C and leads to reduced cyclin D1 expression (10, 11). Moreover, ectopic human K10 (hK10) expression in the basal layer of the epidermis of transgenic mice (making use of the bovine basal keratin bK5 promoter (bK5hK10 mice)), also inhibits cell proliferation and dramatically impairs tumor development (12). Collectively, these results indicate that K10 may play a role in the induction and/or maintenance of postmitotic status of suprabasal epidermal cells (10-12). However, no evidence was provided of other functions of K10 in cells that normally do not divide in vivo but where this protein is normally expressed (9). A way of exploring these possible functions is to express K10 ectopically in postmitotic suprabasal cells of transgenic mice.

Keratin K6 (K6) is a type II keratin under elaborate control. It displays constitutive and inducible expression in various types of complex epithelia. K6 is constitutively expressed in the suprabasal cells of the paw pad and sole of the foot, the nail bed, esophagus, trachea, oral cavity, and the outer root sheath of the hair follicles (13-16). In addition, K6 is induced after injury and in diseases involving altered proliferation or differentiation in humans and mouse skin epidermis (15, 17-22). The complexity of K6 expression is further increased by the fact that there are six functional K6 genes in humans, three in mice (13, 17, 18), and putatively three in cows.² The significance of this diversity is unclear. Finally, inherited mutations affecting the K6 genes in humans are associated with type I (23) and type II (24) pachyonychia congenita. These genetic disorders are characterized by severe dystrophy of the nail plate and differ principally in the involvement of other stratified epithelia (6, 19).

The expression pattern of the K6 genes makes their regulatory regions appropriate to direct the expression of selected transgenes to stratified epithelia in transgenic mice (15, 20, 21). We have previously reported two lines of transgenic mice expressing human keratin K10 (hK10) under the control of the bovine keratin K6 promoter (bK6hK10 mice) (25). Although they displayed no overt phenotype, a clear delay was found in tumor development when these mice were subjected to skin chemical carcinogenesis protocols (25). However, this is a relatively minor effect compared with that observed in bK5hK10 transgenic mice, which are almost completely resistant to tumor development (12). This difference is probably attributable to the expression of hK10 in different cell compartments. The three mK6 genes are normally absent from interfollicular epidermis, but they are rapidly induced upon hyperproliferative stimuli in suprabasal keratinocytes (13, 15, 16). Only one, namely mK6a, is expressed in the basal layer of the hyperproliferative epidermis (13, 16), where bK5 is expressed (26). In contrast, the bovine bK6 regulatory elements drive the expression of the transgene, similarly to the endogenous mK6b gene, in the suprabasal layers of the hyperproliferative epidermis (13, 15, 16). In this compartment, the keratinocytes display a very limited proliferative activity compared with the basal layer cells. In addition, differences in the level of K10 expression may also contribute toward explaining the observed differences in tumorigenic susceptibility between bK5hK10 and bK6hK10 transgenic mice. In support of this, heterozygous bK5hK10 mice do not display overt epidermal abnormalities, whereas hypoplasic and hyperkeratotic epidermises have been observed in homozygous bK5hK10 transgenic mice in parallel with increased expression of the transgene (12). This is also in agreement with our observations demonstrating that the effects of keratin K10 are clearly related to its expression level (10, 11).

In this work, we have tried to address the possible functions of K10 in nonproliferative cells by studying the consequences of hK10 expression in tissues normally expressing K6. As previously reported, bK6hK10 animals display no obvious phenotype even in homozygosis; we have thus generated new bK6hK10 transgenic mice lines bearing a higher copy number of the transgene in order to increase the expression of hK10 in those cells in which bK6 is active. In this context, it is important to point out that the bK6 promoter has an expression pattern very similar to the endogenous mouse keratin K6b (mK6b) (13-16).

All these high copy number transgenic mice display a clear phenotype that affects the oral mucosa and is characterized by the necrosis of the suprabasal cells of the tongue, palate, and gingival epithelium, in association with acute inflammation. This leads to severe shedding of the epithelium, causing perinatal death by impeding suckling. Alterations were also seen in the incisors, but no significant anomalies were observed in nails or hair. Our results indicate that the ectopic expression of K10 in postmitotic suprabasal cells provokes dramatic alterations in their biological behavior and indicate that alterations of the specific expression pattern of keratins in a given epithelium affect the physiological status of the tissue, providing clear evidence of the functional diversity of these proteins.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Transgene Construction and Generation of Transgenic Mice-- To generate the bK6hK10 construct (13) (Fig. 1A), the 9-kb fragment containing the bovine bK6 gene regulatory sequence (15) was inserted, using Asp718 digestion, 5' of the human K10 gene in the plasmid hK10-MC (25). This plasmid contains the full-length hK10 gene including the ATG site and the polyadenylation site 0.5 kb downstream (27). The specificity of the construct was monitored by transfection into K6-expressing and -nonexpressing cells (see below and Fig. 1B). Transgenic mice were produced by microinjection into (C57Bl/10× BALB/c)F₂ mouse embryos as previously described (12, 15, 25, 26). For genotyping and identification of the transgenic mice, genomic DNA was extracted from mouse tails, digested with BamHI, electrophoresed, and transferred onto nylon membrane (GeneScreen Plus; PerkinElmer Life Sciences) for Southern blotting. The number of copies of the transgene (see Fig. 1D) was estimated using a phosphorimaging scanner (Bio-Rad) after normalization. A randomly primed labeled specific probe was generated using 1 kb of the 3'-untranslated fragment of the hK10 gene.

Cell Culture and Transfection-- Bovine mammary gland BMGE+H cells and mouse MCA3D keratinocytes were cultured in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum as previously reported (10, 11, 28). Transfections using the calcium phosphate method were performed using cells cultured on glass coverslips as previously described (10, 11, 28). The expression of the transfected bK6bhK10 construct and the endogenous keratin K6 protein were analyzed by double immunofluorescence using mouse monoclonal antibody K8.60 (diluted 1:40; Sigma) against K10 and a rabbit polyclonal antibody against K6 (diluted 1:600; BabCo, Covance, CA). Secondary antibodies and immunofluorescence visualization were as previously described (10, 11, 28).

Histological and Immunohistochemical Analysis-- Freshly collected tissues were fixed immediately in 10% formaldehyde and left for at least 24 h before being embedded in paraffin prior to sectioning. 4-µm sections were stained with hematoxylin-eosin. The antibodies used for immunohistochemistry were K8.60 monoclonal antibody (Sigma), which recognizes mouse and human K10, and a rabbit polyclonal anti-K6 (BabCo). Sections were incubated with a biotinylated anti-mouse or anti-rabbit antibody and then with streptavidin conjugated to horseradish peroxidase (DAB kit; Vector Laboratories). Control immunostainings using the secondary antibody in the absence of the primary antibody were routinely performed. Antibody localization was determined using 3,3-diaminobenzidine as the chromogenic substrate for peroxidase (3,3-diaminobenzidine kit; Vector Laboratories).

Transmission Electron Microscope Analysis-- Tongues from 0-3-day-old transgenic and control mice were fixed in 2.5% gluteraldehyde in 0.1 M phosphate buffer (pH 7.5) and postfixed in 1% osmium tetroxide prior to dehydration and embedding in Epon 812 resin. Semithin sections were stained with 1% toluidine blue for field selection. Ultrathin sections were stained with uranyl acetate and lead citrate and analyzed as previously described (12).

Western Blot Analysis-- Newborn mice were sacrificed, and excised tongues were snap-frozen in liquid nitrogen. Total protein extracts were prepared as previously reported (12, 25). Protein concentrations were determined using a modified Bradford Assay Kit (Bio-Rad). Equal amounts of protein were electrophoresed in 8.5% SDS-PAGE gels and electroblotted onto nitrocellulose. The membranes were incubated with the primary antibody, followed by donkey anti-mouse horseradish peroxidase or donkey anti-rabbit horseradish peroxidase (1:2000; Jackson Immunoresearch Laboratories). WestPicoSignal (Pierce) was used to detect the bands according to the manufacturer's recommendations. The primary antibodies used included rabbit polyclonal antiserum directed against K6, K5, K13, and K10 (BabCo) and RCK107 monoclonal antibody to react with K14 (29).

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Generation of Transgenic Mice Expressing bK6hK10-- To monitor the proper expression of the bK6hK10 construct (Fig. 1A), transient transfection experiments were performed in MCA3D mouse keratinocytes, which express K6. Synthesis and incorporation of hK10 into the endogenous keratin cytoskeleton was observed in all transfection experiments, with no sign of cytoskeletal disruption (arrows in Fig. 1, B and B'). Similar results were obtained using bovine BMGE+H cells. No expression of the construct was detected when using cell lines that did not express K6 (VeroC, PtK2, and NIH3T3; not shown). Since the absence of keratin clumping and the proper incorporation of hK10 into the keratin cytoskeleton was observed in all experiments, these results indicate that K10 does not cause the collapse of the endogenous cytoskeleton, contrary to the reported effects of increased hK16 (30-32) and mutant mK6 (14) expression.

View larger version (79K): [in this window] [in a new window]   Fig. 1.   Structure of the transgene, expression in cultured cells, and gross abnormalities in bK6hK10 transgenic mice. A, structure of the bK6hK10 transgene used in these studies, including the bK6 gene regulatory region (shadowed) and the hK10 gene showing the exons and introns. B, expression of bK6hK10 in MCA3D mouse keratinocyte cells 48 h after transfection. Transfected cells were analyzed by double immunofluorescence for hK10 (B) and endogenous K6 (B') expression. Note that in all cases K10 is integrated into the endogenous keratin cytoskeleton (at least 200 cells were scored per experiment; experiments were performed in triplicate). C, appearance of transgenic (right) and wild type (left) 3-day-old littermate. Note the runt appearance of the transgenic animal. The arrows indicate the empty stomach in the transgenic compared with the milk-containing stomach in its wild type littermate. D, example of Southern blot showing the identification of the transgenic animals and the estimated copy number of the transgene after normalization and phosphorimaging scan. E, abnormalities in transgenic tongue; note the leukoplakia lesions observed from the midregion to the posterior region of the dorsal tongue in the transgenics (denoted by lp). E', normal tongue in wild type littermate.

The linearized bK6hK10 construct was subsequently injected into fertilized oocytes. We have previously reported that in animals with 15 or fewer copies of the bK6hK10 transgene, no overt phenotype (besides a delay in tumor formation) is observed (25). Given that the expression levels of transgenes under the control of the bK5 and bK6 promoter regions are frequently proportional to the copy number of the transgene integrated (12, 25, 33) (data not shown), attempts were made to generate transgenic mice with more than 25 copies of the transgene. Two founders were obtained bearing 50 and 75 copies, respectively (see Fig. 1D). These animals displayed no overt phenotype. However, further analysis indicated that this was attributable to mosaicism in the expression of the transgene, as reported in other transgenic mice models (data not shown; see Ref. 33 for a careful discussion). As expected, however, the offspring derived from these founders displayed no such mosaicism in K10 expression (see below) and showed severe phenotypic alterations, which finally led to death between days 3 and 5 after birth. Both transgenic mice lines had similar phenotypic alterations and are collectively described as high copy number transgenics.

Overexpression of bK6hK10 Leads to Severe Necrosis and Acute Inflammation of the Oral Mucosa-- Transgenic bK6hK10 pups appeared normal at birth, but within a few days they were smaller and weaker and had less milk in their stomachs (sometimes none) in comparison with their control littermates (Fig. 1C, arrows). These pups generally died between 3 and 5 days postpartum, weighing about half that of their littermate controls (1.4 and 2.9 g, respectively, on average). At death, the skin and nails of these mice appeared normal (Fig. 1C and data not shown) and showed no obvious developmental anomalies other than reduced size and frail appearance.

Since the bK6 transgene is constitutively expressed in several oral epithelia (15) and since one possible reason for the observed mortality of the bK6hK10 transgenics could be poor feeding, the anatomy of the oral cavity of these mice was examined. The dorsal surface of the tongue and, to a lesser extent, the ventral surface of the upper palate were covered with white plaques from the midregion to the pharynx (denoted by lp in Fig. 1D and data not shown). Nontransgenic littermates (Fig. 1D') showed no signs of these lesions. Similar features have been reported for mice lacking keratins mK6a and mK6b (13, 16) and are similar to the oral leukoplakia seen in some pachyonychia congenita patients (23, 34).

Histological studies of sections from the tongue of bK6hK10 mice demonstrated severe damage from the midregion to the posterior region of the dorsal epithelium. Extensive areas with consistent features of coagulative necrotic changes were seen, with loss of cell cytoplasm and degenerative nuclei. Some tissue organization was still present, however, as seen by the remaining papillae profiles, although these were heavily infiltrated by neutrophils (Figs. 2A" and 3C; compare with controls in Figs. 2A and 3A and no-lesion tongue in Fig. 3B). Similar features were also observed in the palate (Figs. 2B' and 3D; compare with controls in Figs. 2B and 3A) and in the gingival epithelium of the incisors (Fig. 8A'; compare with control in Fig. 8A). In all cases, the apparent thickening of the unaffected epithelial cell compartments was observed, giving rise to a moderate hyperplasic phenotype, probably caused by inflammation and infiltration of the tissue. The lesions of the tongue primarily affected the filiform papillae, which are the most abundant papillary type in mice. In these, three distinct programs of epithelial differentiation exist (Fig. 2A'), giving rise to the anterior (ac), posterior (pc), and buttress (bc) columns (35, 36). Interestingly, the data obtained with mK6a/b-null mice suggest that these filiform papillae are particularly sensitive to mechanical stress and alterations of the keratin cytoskeleton (13, 16). The bK6 promoter is active in the anterior and buttress columns (Fig. 3B), as previously demonstrated by -galactosidase staining in bK6LacZ transgenic mice (15). In agreement, the immunohistochemical analysis of the tongue of newborn transgenic mice showed a coincident expression of mK6 (Fig. 3A; wild type tongue denoted by t) and the transgene (Fig. 3B; nonlesional region of the tongue). Consistently, K10 expression was also observed in the necrotic suprabasal regions of the epithelium in the tongue (Fig. 3C) and palate (Fig. 3D).

View larger version (81K): [in this window] [in a new window]   Fig. 2.   Histological abnormalities in the epithelium of the tongue and palate of bK6hK10 transgenic mice. Hematoxylin-eosin-stained sections of the tongue (A, A', and A") and palate (B and B') in wild type (A, A', and B) and transgenic mice (A" and B'). See the common filiform papillae arrangement of the dorsal epithelium in the tongue of the wild type (A) compartmentalized in the anterior column, posterior column, and buttress column (ac, pc, and bc, respectively in A'). The transgenic suprabasal layers of the epithelium in the tongue (A") and palate (B') appeared necrotic and infiltrated by neutrophil leukocytes, whereas the basal layers remained unaffected. A moderate increase in the number of cell layers in this region can also be observed in the transgenic samples. Dashed lines denote the epithelial boundaries in A" and B'. Bar, 100 µm.

View larger version (131K): [in this window] [in a new window]   Fig. 3.   Immunohistochemical analysis of K6 and K10 expression in the epithelium of the tongue and palate of bK6hK10 transgenic mice. Peroxidase-hematoxylin-stained sections of the tongue (A-C) and palate (A and D) in wild type (A) and transgenic mice (B-D). The mK6 expression is restricted to the suprabasal epithelium of the palate (p in A) and the anterior and buttress column of the tongue (t in A). K10 in the transgenics is strongly expressed in the same regions of the nonlesional regions of the tongue (B) as well as in the necrotic plaque of the tongue (C) and the palate (D). Dashed lines denote the epithelial boundaries. Bars, 100 µm.

The injured superficial areas expanded the tongue and palate epithelia with a mixture of neutrophils and cell debris disrupted from the several unaffected rows of basaloid cells normally attached to the basement membrane (Figs. 2A" and 4B, semithin section). This shedding process of the necrotic suprabasal areas was probably induced by the intensive discharge of gelatinase and other proteases present in the granules of the abundant neutrophils. No major alterations were observed in the subjacent muscle of the tongue (Fig. 4B). Most of the pathological changes in the lingual, gingival, and palate epithelia are similar to those described for mice lacking mK6a and mK6b genes (13, 16). In these deficient mice, the lesions are associated with a decrease in (or even the absence of) KIF in the anterior compartment, which induces an important increase in the size of intercellular spaces (13, 16). Although a similar absence of KIF was not expected, it could not be ruled out a priori that other changes in the keratin cytoskeleton might occur in the bK6hK10 mice. To investigate such possible changes, ultrastructural analyses were performed.

View larger version (185K): [in this window] [in a new window]   Fig. 4.   Ultrastructural analysis of the tongue of bK6hK10 transgenic mice. Semithin toluidine blue-stained section (B) and electron microscope appearance (A, C, and D) of the tongue epithelium in wild type (A) and transgenic mice (B, C, and D). A, note the characteristic features of the filiform papillae in the wild type, organized in three compartments, AC with abundant keratohialin granules, BC with heavily packed bundles of KIF, and PC. SC, stratum corneum. B, toluidine blue-stained semithin section of the lesional tongue of the bK6bhK10 transgenic mice. The large arrow denotes the shedding region. C, high magnification obtained by electron microscopy of the superficial region marked by the upper rectangle in B. Note that this necrotic plaque still preserves part of the papillary structure and that anterior column cells (AC) have enlarged keratohialin granules, whereas buttress column cells (BC) display abundant keratin bundles. D, high magnification obtained by electron microscopy of the region marked by the lower rectangle in B. The asterisks indicate the abnormally swollen and clear cytoplasm in AC cells. The open arrows point to the keratohialin granules characteristic of this compartment. The black arrows show numerous neutrophil leukocytes infiltrating the cytolytic AC areas. Note that posterior column and basal cells are unaffected (PC and BC, respectively). C, conjunctive cells in the submucosa. In B, m marks the muscular layer. Dashed lines in B and D denote the epithelial boundaries and the basement membrane.

Ultrastructural Pathology of the Tongue of bK6hK10 Transgenic Mice-- The dorsal epithelium of the tongue in control and transgenic mice was studied by transmission electron microscopy. In wild-type animals, differentiating keratinocytes in the anterior column (AC) showed electron dense granules similar to those found in the granular layer of the epidermis (37). The buttress column (BC) keratinocytes had a flattened shape and large bundles of densely packed KIF but no keratohyalin granules. Finally, the keratinocytes of the posterior column (PC) were rounded and had a well organized keratin cytoskeleton, although less bundled than in BC keratinocytes. An example of the different regions of a control filiform papilla, as observed with the electron microscope, is provided in Fig. 4A. Samples taken from the same region in bK6hK10 transgenic mice (semithin section in Fig. 4B) showed dramatic alterations in the AC and BC keratinocytes, readily visible through their cytolytic and distended appearance and unusually clear and swollen cytoplasm (Figs. 4D and 5A, asterisks). The AC cells appeared to have a normal complement of keratohyalin granules, but these were larger than those of wild type mice (Fig. 4C; wild type in Fig. 4A). Desmosomes and KIF were found in these cytolytic swollen cells (data not shown and Fig. 5A, arrows). Similar cytolytic events were observed in BC keratinocytes (Fig. 4C), which also displayed normally arranged KIF bundles and desmosomes (Fig. 5B). No major abnormalities were observed in the PC or basal keratinocytes (Figs. 4D and 5A), which do not express the transgene (Fig. 3, B and C). Neutrophil leukocytes were seen infiltrating the cytolytic areas of the AC and BC (arrows in Fig. 4D, pmn in Fig. 5A, and data not shown). Despite the severe cytolytic changes observed in these regions, the affected cells maintained intact desmosomes (white arrows in Fig. 5B), and the intercellular spaces were not enlarged (Fig. 4, C and D, and Fig. 5A). Since the hK10 expression is restricted to the AC and BC in agreement with the expression of endogenous mK6 (Fig. 3, compare B with A), these findings suggest that these cells are particularly susceptible to changes in the expression, either quantitative or qualitative, of keratin polypeptides. In this regard, the blisters occurring in mK6a/b-deficient mice have been attributed to the absence of KIF in these cells (13, 16), whereas KIF and desmosomes normally arranged were clearly observed even in cytolytic cells (Fig. 5, A and B). Data obtained with mK6a/b-deficient mice have been interpreted as an indication that the cells in the AC region of the filiform papillae are particularly sensitive to the mechanical stress of suckling (13, 16). However, some animals were maintained by parenteral feeding, and similar lesions still occurred (not shown), indicating that mechanical stress is not the only origin of the observed tongue and palate anomalies. Nonetheless, it is worth mentioning that some differences exist between the lesions observed in mK6a/b null mice and those of our transgenic animals. In particular, desmosomes and KIF are normally arranged, and intercellular edema is absent from the tongue lesions in bK6hK10 mice, whereas in K6a/b-deficient mice KIF is decreased or absent in the AC, and the intercellular spaces are enlarged (13, 16). This clearly points to a different etiology in these apparently similar lesions; in the deficient mice there is a clear epithelial fragility process, whereas in the present case the defects are more related to an acute inflammatory response.

View larger version (96K): [in this window] [in a new window]   Fig. 5.   Presence of keratin filaments and normal desmosomes in altered AC and BC cells of bK6hK10 transgenic mice. Electron microscopy analysis of the epithelium of the tongue in transgenic mice. A, cytolytic cells in the anterior column showing swollen and clear cytoplasm (asterisks), keratohialin granules (kh), and KIF (arrows); N, nucleus; pmn, neutrophils. Note that the underlying posterior column cells remain unaffected. B, intercellular space between two BC cells showing a close cellular union and desmosomes (white arrows); f, KIF bundles; m, mitochondria.

The Expression of Other Keratins Is Not Altered in the Tongue of bK6hK10 Transgenic Mice-- The above results do not rule out the possibility that the ectopic expression of hK10 can affect the composition of the keratin cytoskeleton of the tongue keratinocytes. To investigate this possibility, protein extracts prepared from whole tongues of low copy number (15 copies) nontransgenic and high copy number (75 copies) transgenic mice (Fig. 6) were studied by Western blotting. As a control, protein extracts from human skin were included. The anti-K10 antibody detects a single polypeptide in the extracts from human skin and transgenic tongues. In addition, the level of expression of hK10 was increased in those extracts from transgenic animals with high transgene copy number. The amounts of the characteristic keratin polypeptides of the suprabasal tongue keratinocytes, namely K13 and K6 (36), were not significantly different between the transgenic and control samples. Finally, compared with the controls, the protein levels of the basal keratins K5 and K14 were not altered in transgenic samples. Collectively, these results indicate that the expression of hK10 does not produce major alterations in the keratin expression profile. It is worth mentioning that the hK10 expression levels in high copy number transgenic mice are very similar to those of endogenous hK10 in human epidermis, indicating that the phenotype found in these mice cannot be attributed to an aberrant excessive overexpression of the transgenic protein.

View larger version (35K): [in this window] [in a new window]   Fig. 6.   Absence of major alterations in the keratin expression pattern of tongue epithelium. Analysis of keratin polypeptides in tongue epithelium. Western blots were performed using 10 µg of total protein extracts from human skin and nontransgenic and transgenic mice, with low and high copy numbers. The antibodies specifically reacting with the indicated keratins were used. Note that no significant changes were observed in the levels of mK13, mK14, mK5, or mK6 keratins among the different murine tongue samples.

bK6bhK10 Transgenic Mice Have No Hair or Nail Abnormalities-- Keratin K6 is constitutively expressed in the outer root sheath of the hair follicles and nail bed epithelium. Possible alterations in these cutaneous structures were therefore studied. No abnormalities were found between control and transgenic mice in the hair or interfollicular skin by day 4 after birth (Fig. 7, A and A', respectively). To see whether the absence of such defects could be due to a lack of hK10 transgene expression, the expression of K6 and K10 was studied in samples from wild type and transgenic mice. The results confirmed the expression of hK10 in the outer root sheath of transgenic mouse hairs (Fig. 7C), coincident with endogenous mK6 expression (Fig. 7B). No expression of K10 was detected in the hair follicles of control mice (Fig. 7D). No defects were seen in the nails of the transgenic mice (data not shown); nor were defects seen in hair follicles or nails of K6a/b-deficient mice (13, 16). This can be explained in that these mice die too early to develop any possible alterations. In agreement with this hypothesis is the appearance of hair and nail lesions in pachyonychia congenita patients several months after birth. However, in the mK6a/b-deficient mice, the absence of hair and nail abnormalities might also be attributed to the expression of the recently described mK6hf gene (13), which appears to be specifically expressed in the hair follicles and nail beds and may compensate for mK6a/b deficiency (13). The present results also point to the possibility that quantitative and/or qualitative alterations in the normal characteristic keratin expression pattern may cause functional anomalies in certain keratinocytes (e.g. AC and BC tongue keratinocytes), but not in others (e.g. outer root sheath keratinocytes).

View larger version (119K): [in this window] [in a new window]   Fig. 7.   Absence of alterations in interfollicular epidermis and hair follicles of bK6hK10 transgenic mice. Hematoxylin-eosin (A and A') or Peroxidase-hematoxylin-stained (B-D) dorsal skin sections from wild type (A, B, and D) and transgenic mice (A' and C). The morphology of the epithelium and anagen hair follicles is similar in the transgenic (A') and wild type (A) animals. K10 in the transgenic skin is expressed in the suprabasal layer of the epidermis (C) and the outer root sheath of hair follicles (arrows in C), coinciding with the expression of K10 (D) and mK6 (B) in the wild type. Bar, 100 µm.

As with the hair follicles, no alterations were detected in the esophagus or forestomach (not shown). Again, the absence of abnormalities may be due to the early death of bK6hK10 transgenic mice, which precludes the detection of lesions that might occur later. Supporting the early lethality hypothesis is the appearance of esophageal alterations only 2 months after birth in K4-deficient mice (38), in clear contrast with the absence of alterations in the tongue of these null mice. However, it cannot be ruled out that the levels of hK10 transgene expression may vary among the different K6-expressing cells in the different tissues, thus promoting phenotypic changes (or not) or that these cells may display different intrinsic susceptibility to possible K10-induced effects.

Tooth Abnormalities in bK6bhK10 Transgenic Mice-- Different tissues from bK6hK10 mice were analyzed, and no ectopic expression of K10 besides that observed in K6-expressing cells was seen. The findings in oral mucosa closely resemble those observed in patients suffering from pachyonychia congenita. One of the variants of this disease, the Jackson-Lawler form or pachyonychia congenita type II, is characterized by the presence of neonatal teeth (24, 39-41). We therefore investigated the presence of abnormalities in the teeth of the transgenic mice. These animals displayed severe teeth anomalies such as defects in position and precocious eruption of the incisors, in some cases associated with microdontia (Fig. 8A' and data not shown; wild type in Fig. 8A). Hematoxylin-eosin sections revealed a clear decrease (even absence) of the thickness of the dentine layer (de in Fig. 8B, wild type; transgenic shown in Fig. 8B') associated with degenerative changes in the ameloblast and odontoblast layers (am and od in Fig. 8, B and B', wild type and transgenic, respectively). In addition, some degenerative changes were also observed (see inset in Fig. 8A'). Collectively, these abnormalities are clearly suggestive of specific roles of keratin-expressing cells in the process of tooth growth. In this regard, it has been demonstrated that keratins participate in the assembly of amelogenin during amelogenesis, supporting a possible involvement of these proteins in tooth development (42). This important and complex issue will be the subject of future studies.

View larger version (146K): [in this window] [in a new window]   Fig. 8.   Alterations in the incisors of bK6hK10 transgenic mice. Hematoxylin-eosin stained sections of the lower jaw (A and A') and incisor (B and B') in newborn wild type (A and B) and transgenic mice (A' and B'). Note the precocious eruption of the incisor in the transgenic (A') in comparison with the wild type (A). The gingival epithelium displays similar necrotic changes in suprabasal layers as described for the tongue and palate epithelium (arrows in A'). The inset in A' shows the degenerative region in the transgenic incisor. Higher magnification of these degenerative areas of the transgenic incisor (B'), compared with wild type littermate (B). Note the affected ameloblast (am) and odontoblast (od) layers as well as the absence of the dentine layer (denoted by de in B). p, dental pulp; od, odontoblast layer; de, dentine; en, enamel; am, ameloblast layer; ep, enamel pulp. Bars, 50 µm.

Many of the phenotypic features observed in the bK6hK10 mice resemble those characterized in pachyonychia congenita patients. This disease is an autosomal dominant ectodermal dysplasia caused by mutations in keratins K6, K16, and K17, whose major hallmark is hypertrophic nail dystrophy (6). Two main types of this disorder have been characterized: type I or the Jadassohn-Lewandowsky form, in which pachyonychia occurs in conjunction with palmoplantar keratoses and oral leukokeratoses (6, 23), and type II or the Jackson-Lawler form, characterized by hair abnormalities and neonatal teeth but no oral lesions (6, 24). The phenotype reported here displays some characteristics coincident with type I (tongue and palate lesions) and others with type II (incisor anomalies). However, no alterations in nails (a common feature of types I and II), the palmoplantar epidermis (type I), or hair abnormalities (type II) were seen, suggesting that the alterations found in bK6hK10 transgenics do not represent a pachyonychia-like phenotype.

Similar to other keratins, K10 is involved in providing mechanical resilience to epidermal cells, as demonstrated by the presence of disruptive K10 mutations in epidermolytic hyperkeratosis (or BCIE; see Ref. 6 and references therein) and reinforced by the phenotype displayed by newborn K10-deficient mice (43). Therefore, it is difficult to explain why the ectopic expression of this keratin in oral epithelia leads to alterations reminiscent of those due to the changes in the keratin cytoskeleton associated with epithelial fragility. One possible explanation is that the abnormal KIF cytoskeleton observed in pachyonychia congenita keratinocytes and characterized by densely aggregated keratin filament bundles (41) might also be produced by increased expression of K10. In this regard, we have previously reported that K10 forms highly bundled filaments in transfected cells and in transgenic mouse epidermis (12). However, we did not detect altered KIF in the affected cells of the AC and BC compartments of the tongue of bK6hK10 transgenic mice (Figs. 4 and 5). Further, the phenotypic alterations in these mice are frequently associated with an acute inflammatory response, with high leukocyte infiltration (Figs. 2 and 3). No such alterations have been reported in affected tissues of pachyonychia congenita patients, and only mild infiltration was found in mK6a/b-deficient mice (16). Determining whether this inflammatory response is a cause or a consequence of the disruption of the tongue epithelia requires further analysis. However, the implication of K10 in the modulation of Akt-dependent signaling (11, 12) suggests that the presence of ectopic K10 might also affect the expression or activity of certain inflammatory cytokines in the affected epithelia. Among them, tumor necrosis factor- is of particular interest, since it has recently been demonstrated that the alterations in the keratin cytoskeleton may modulate the activity and response to this molecule (44-46). These aspects will be addressed in future experiments using bK5hK10 and bK6hK10 transgenic mice.

	ACKNOWLEDGEMENTS

We are greatly indebted to J. Martínez for excellent animal care, I. de los Santos for assistance in histology preparations, and S. Moreno for help in photography.

	FOOTNOTES

^* This work was supported in part by Spanish MCYT Grant PB94-1230 and CAM Grant 08.1/0054/2001.1.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed. Tel.: 34-91-3466438; Fax: 34-91346484; E-mail: jesusm.paramio@ciemat.es.

Published, JBC Papers in Press, July 15, 2002, DOI 10.1074/jbc.M205143200

² J. L. Jorcano, unpublished results.

	ABBREVIATIONS

The abbreviations used are: KIF, keratin intermediate filament; hK, human keratin; mK, mouse keratin; bK, bovine keratin; AC, anterior column; BC, buttress column; PC, posterior column.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				G. Fitsialos, A.-A. Chassot, L. Turchi, M. A. Dayem, K. LeBrigand, C. Moreilhon, G. Meneguzzi, R. Busca, B. Mari, P. Barbry, et al. Transcriptional Signature of Epidermal Keratinocytes Subjected to in Vitro Scratch Wounding Reveals Selective Roles for ERK1/2, p38, and Phosphatidylinositol 3-Kinase Signaling Pathways J. Biol. Chem., May 18, 2007; 282(20): 15090 - 15102. [Abstract] [Full Text] [PDF]

				B. K. Bloor, N. Tidman, I. M. Leigh, E. Odell, B. Dogan, U. Wollina, L. Ghali, and A. Waseem Expression of Keratin K2e in Cutaneous and Oral Lesions: Association with Keratinocyte Activation, Proliferation, and Keratinization Am. J. Pathol., March 1, 2003; 162(3): 963 - 975. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/38/35371    most recent M205143200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Santos, M. Articles by Jorcano, J. L. Search for Related Content PubMed PubMed Citation Articles by Santos, M. Articles by Jorcano, J. L. Social Bookmarking                      What's this?