MATERIALS AND METHODS

Electron microscopy

Biopsies were taken from fetal (E19) and neonatal mice and minced to < 0.5 mm³. They were fixed for 4 h in cacodylate-buffered 2.5% glutaraldehyde. Half of the samples were postfixed with 1% aqueous osmium tetroxide (OsO₄), containing 1.5% potassium ferrocyanide and the other half with 0.5% ruthenium tetroxide (RuO₄, Polysciences, Warrington, PA)/0.25% aqueous potassium ferrocyanide at pH 6.2 for 1 h in the darkness at 5°C (^{Fartasch et al. 1993;Fartasch & Ponec 1994}). After rinsing in buffer for 3 h and dehydration in graded ethanol series, the OsO₄-fixed samples were embedded in Glycidether 100 (Sigma, Deisenhofen) and the RuO₄-fixed samples were embedded in low viscosity Epon (Electron Microscopic Science, Fort Washington, PA). Thin sections were examined before and after double-staining with ethanolic uranyl acetate and lead citrate with a Jeol 100 CX electron microscope. Serial semithin sections of OsO₄-fixed samples were stained with 1% methylene blue.

SC preparation and lipid analysis

Neonatal mice were killed and total trunk skin was dissected immediately. The skin was spread upside down on a precooled (–20°C) glass Petri dish. Subcutaneous tissue was removed by gentle scraping with a scalpel. The remaining skin was floated on 0.5% trypsin/phosphate-buffered saline for 30 min at 37°C in a plastic Petri dish, followed by sonication for 10 min and three washes with phosphate-buffered saline. Subsequently, the SC was separated from the epidermis with forceps. The individual SC preparations were homogenized, lyophilized, and weighed. Lipids were extracted in chloroform–methanol–water (1:2:0.6, vol/vol/vol) for 24 h at 37°C. For separation of ceramides total lipid extracts were applied to thin-layer silica gel 60 plates (Merck, Darmstadt, Germany), and the chromatograms were developed twice with chloroform–methanol–acetic acid (190:9:1, vol/vol/vol).

Fatty acids were separated according to^{Ponec et al. (1988)}. After application of total lipid extracts to high-performance thin-layer chromatography plates (Merck, Darmstadt, Germany) the chromatograms were developed using chloroform–methanol (9:1, vol/vol) (up to 10 mm) as the first and chloroform–diethylether–ethyl acetate (80:4:16, vol/vol/vol) (up to 90 mm) as the second solvent system.

For quantitation of cholesterol, glucosylceramide, and sphingomyelin, residual phospholipids were degraded by alkaline hydrolysis with 2 ml methanolic sodium hydroxide (50 mM) for 2 h at 37°C. After neutralization with acetic acid, lipid extracts were desalted by reversed-phase chromatography on LiChroprep RP18. Neutral and anionic lipids were separated by anion-exchange chromatography on DEAE-Sepharose CL6B (Pharmacia Freiburg, Germany) and applied to thin-layer silica gel 60 plates. The chromatograms were developed using chloroform–methanol–aqueous CaCl₂ (0.22%) (60:35:8, vol/vol/vol) for separation of glucosylceramide and sphingomyelin and using chloroform–methanol–acetic acid (190:9:1, vol/vol/vol) for separation of cholesterol.

For quantitative analytical thin-layer chromatography determination, increasing amounts of standard lipids N-stearoyl-sphingosine (kind gift from Beiersdorf, Germany), cholesterol, palmitic acid and sphingomyelin (all from Sigma Deisenhofen, Germany) and glucosylceramide (purified from Gaucher spleen, Sandhoff lab) were applied. After development, plates were air-dried, sprayed with 8% (wt/vol) H₃PO₄ containing 10% (wt/vol) CuSO₄, charred at 180°C for 10 min and quantitated by photodensitometry (Shimadzu Kyoto, Japan).

Statistical evaluation of data was performed using a two-tailed Student's t test.

Sodium dodecyl sulfate (SDS)–polyacrylamide gel electrophoresis and western blot

Total protein was extracted from neonatal skin in SDS–polyacrylamide gel electrophoresis sample buffer (50 mM sodium phosphate, pH 6.8, 5% SDS, 40 mM dithiothreitol, 5 mM ethylenediamine tetraacetic acid, 5 mM ethyleneglycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid and 15% glycerol). The samples were homogenized and heated for 5 min at 95°C. Insoluble material was removed by centrifugation. Total protein was determined by the BioRad protein quantitation kit and extracts were loaded equally. Separation of total protein extracts was carried out by standard procedures (10% SDS–polyacrylamide gel electrophoresis). Proteins were electrotransferred to polyvinylidene difluoride membranes (Millipore, Eschborn, Germany) by wet blotting with 20 mM sodium borate/1 mM ethylenediamine tetraacetic acid, pH 9.2. Membranes were stained with 0.2% Coomassie/40% methanol/5% acetic acid and destained with 40% methanol/5% acetic acid. Primary antibodies (Hiss Diagnostics, Freiburg, Germany) were diluted as follows: loricrin anti-serum (AF 62, 1:20 000) and involucrin anti-serum (1:30 000). Immunodetection was performed with horseradish peroxidase-conjugated antirabbit secondary antibodies (Dianova, Hamburg, Germany, 1:30 000) using Super Signal (Pierce, St. Augustin, Germany) as a substrate.

Indirect immunofluorescence

Sectioning and immunofluorescence analysis of the skin samples was performed as previously described (^{Reichelt et al. 1997}). Primary antibodies (Hiss Diagnostics, Freiburg) were diluted 1:500 (AF 62, loricrin anti-serum) and 1:200 (involucrin anti-serum). The secondary antibody, Cy3-conjugated goat anti-rabbit IgG (Dianova, Hamburg, Germany), was diluted 1:800.

RNA preparation and northern blot

Trunk skin of neonatal mice was prepared immediately after decapitation and either frozen in liquid nitrogen or directly used for RNA extraction according to^{Chomczynski & Sacchi (1987)}. Thirty micrograms of RNA were loaded per lane on 1% agarose gels (2% agarose gels for Figure 7) in the presence of formamide/formaldehyde. After separation RNA was transferred onto Genescreen membrane (NEN, Dreieich, Germany) by capillary transfer using 10 SSPE as buffer (^{Sambrook et al. 1989}).

Figure 7.

Northern blot analysis revealed different mRNA sizes for involucrin and loricrin in different mouse strains. Trunk skin RNA of neonatal wild-type C57Bl/6 (lanes 1 and 3) and BALB/C (lanes 2 and 4) mice were hybridized with probes specific for involucrin (lanes 1 and 2) and loricrin (lanes 3 and 4). The mRNA for both proteins were about 0.1 kb larger in C57Bl/6 mice than in BALB/C.

Full figure and legend (2K)

Hybridization was performed overnight at 37°C for the acid sphingomyelinase probe and at 42°C for the other probes (for detailed conditions, see^{Magin et al. 1998}). Blots were washed 3 with 300 ml of 0.1 SSPE/0.1% SDS at 55°C for the acid sphingomyelinase probe and at 65°C for the other probes. A probe of about 200 bp length for mouse loricrin was derived from the 3' noncoding region (kind gift of Dennis Roop, Houston, TX). As a probe for mouse involucrin a 1.3 kb HindIII fragment derived from a cDNA clone was used (kind gift of Philippe Djian, Meudon-Bellevue, France). Acid sphingomyelinase was probed with a 1.4 kb fragment of the 3' human cDNA clone sequence which showed 81% sequence identity to the mouse sequence. The TRI-GAPDH-Mouse anti-sense template (Ambion, Frankfurt, Germany) was used as a control.

RESULTS

Ultrastructural alterations of the epidermis of K10-deficient mice

Previous ultrastructural analysis of K10-deficient mice had been focused on the living layers of the epidermis and investigated the localization of keratin intermediate filaments and aggregates of K1/10 in relation to keratohyalin granules (^{Reichelt et al. 1997}). To examine the ultrastructure of the SC and SG/SC transition layer, postfixation with osmium tetroxide and ruthenium tetroxide were used. Whereas osmium staining permitted assessment of routine ultrastructural features, including LB, ruthenium tetroxide revealed the organization of SC intercellular lipids.

In neonatal (not shown) and fetal (E19) homozygous K10-deficient mice intraepidermal bullae resulted from separation of edematous cells from one another (Figure 1a). Intact SG cells next to cytolysing cells showed LB with regular morphology which fused with the upper SG membrane and appeared to secrete their lipid content (Figure 1b). The epidermis of neonatal homozygous K10-deficient mice showed 12–15 SC layers whereas the wild type exhibited only approximately 10 layers. Especially the matrix of the corneocytes of the lower SC showed electron-dense aggregations of filament-like structures and an irregular staining pattern. Keratin clumping was present in bullous as well as in nonbullous areas (Figure 1c). There were variously sized clear spaces around the nuclei of the upper stratum spinosum and SG cells. Peripheral to the cytolytic clear spaces (Figure 1c, C), the cells showed thick shells of irregularly clumped keratin filaments.

Figure 1.

Ultrastructural analysis revealed normal extrusion of LB in K10-deficient mice. Homozygous fetal K10-deficient skin (E19) displayed intraepidermal bullae (B) resulting from separation of cytolytic cells of the upper SG and spinous layer (a, OsO₄ staining, scale bar: 10 m). Intact SG cells next to cytolytic areas were still capable of secreting LB lipids (arrow) at the SG/SC interface (b, KH: keratohyalin, OsO₄ staining, scale bar: 1 m).The epidermis of neonatal homozygous K10-deficient mice showed cytolytic disruption of some of the SG and stratum spinosum cells (c, C). Clumping of keratins was mostly found in the periphery of the cells (c, arrows, OsO₄ staining, scale bar: 2 m). In regions where the SG and stratum spinosum cells lacked cytolytic clear spaces LB secretion and intercellular formation of LB sheets at the SG/SC interface was normal (d, arrows; RuO₄-staining; scale bar: 0.1 m). Mature lamellar lipid layers were found in the intercellular spaces of the horny layer (e, RuO₄-staining; scale bar: 0.1 m).

Full figure and legend (88K)

The quantity of LB in the cytosol of the SG cells of homozygous mice varied greatly. In regions with severe cytosolic swelling, LB were conspicuously reduced. In some areas, where the corneocytes showed a CE and an intact lipid envelope, LB sheets were present in the intercellular space of the SG/SC interface (Figure 1d). As in wild-type epidermis the LB sheets were found in the extracellular domains of the lower SC, whereas fully processed, mature lamellar lipid layers were observed in the extracellular domains of the middle and upper part of the SC (Figure 1e).

In neonatal heterozygous mice, the extrusion of LB contents and the subsequent transformation of LB-derived membrane sheets into mature lamellar membrane unit structures with a characteristic pattern of repeating electron-lucent and electron-dense lamellae was normal.

Disruption of the keratin cytoskeleton is accompanied by alterations of the composition of SC lipids

The SC lipids in animal models for EH (^{Fuchs et al. 1992;Bickenbach et al. 1996;Takahashi & Coulombe 1996}) have not yet been determined. After isolating the epidermis of neonatal K10-deficient and wild-type mice, we investigated the lipid content of SC fractions. We found an enormous increase in total lipid per mg SC dry weight in homozygous neonatal K10-deficient mice (89.1 g) compared with wild-type littermates (48.3 g). The value of heterozygous animals (62.6 g) was intermediary between the results of homozygotes and controls. The increase comprised all major SC lipids, namely cholesterol, free fatty acids, and ceramides (Figure 2a). The analysis of individual ceramide subpopulations revealed that the overall increase in ceramides was based on the increase in ceramide 2 [or NS using the nomenclature of^{Motta et al. (1993)} as revised by^{Robson et al. (1994)}] (Figure 2b). In percentage of total SC lipid, ceramides 1 (EOS), 3 (NP), 4 (AS), and 5 (AS) were decreased in hetero- and homozygous K10-deficient mice. Individual ceramides differ from each other according to the hydroxylation of either sphingoid bases (S = sphingosine, p = phytosphingosine) or N-acyl fatty acids (N = nonhydroxy, A = -hydroxy, O = -hydroxy) and whether the -position is further acylated (i.e., esterified = E) (^{Madison et al. 1990;Doering et al. 1999}). The ceramide precursors glucosylceramide and sphingomyelin were both reduced in SC preparations (Figure 2c). At present, it is not possible to decide whether the changes in lipids are due to the loss of cytoarchitecture in cytolysing cells (^{Porter et al. 1996}), or whether they result from an altered epidermal differentiation program as indicated by the expression of K6 and K16 (^{Reichelt et al. 1997;Porter et al. 1998}).

Figure 2.

Altered SC lipid composition in K10-deficient mice. Levels of the main SC lipids (cholesterol, free fatty acids, and total ceramides) are shown (a). Separate levels of five individual murine SC ceramides (b). Note the increased amount of ceramide 2 in homozygous K10-deficient mice and the mainly unchanged levels of the other ceramides in relation to SC dry weight. Residual amounts of sphingomyelin and glucosylceramides in the SC are indicated (c). Each value is the mean of four different animals SEM. The lipid levels of heterozygous and homozygous mice differed significantly from wild-type levels at ***p 0.001, **p 0.005, or *p 0.08. n.s., not significant.

Full figure and legend (19K)

Selective changes in CE precursor proteins

Recently, we reported a strong increase in keratohyalin in the upper epidermis of K10-deficient mice which pointed to an altered terminal differentiation (^{Reichelt et al. 1997}). In order to elucidate whether other proteins expressed during terminal differentiation were also altered as found in the human condition (^{Kanitakis et al. 1987;Ishida-Yamamoto et al. 1995}), we analyzed the expression of involucrin and loricrin.

By immunofluorescence analysis of frozen sections the distribution of involucrin showed no significant differences between K10-deficient mice and wild-type littermates (Figure 4d–f). By western blotting (Figure 3, lanes 4–6), however, we found an increase in involucrin which was accompanied by a significantly increased expression of the corresponding mRNA in neonatal epidermis of homozygous and to a minor extent also in heterozygous K10-deficient mice (Figure 5a).

Figure 4.

Immunofluorescence staining of loricrin and involucrin. (a–c) Show the expression of loricrin and (d–f) the expression of involucrin in wild-type (a, d), heterozygous (b, e) and homozygous (c, d) neonatal K10-deficient epidermis. Loricrin was detected in the granular layer in all genotypes whereas involucrin was localized in the SC. The basal layer is outlined by arrowheads, asterisks show areas of cytolysis in the granular layer of homozygous epidermis. Scale bar: 40 m.

Full figure and legend (28K)

Figure 3.

Western blotting revealed an increase in involucrin in K10-deficient mice. Equal amounts of total protein extracts from neonatal skin were loaded. Lanes 1 and 2 show the expression of loricrin in wild-type and heterozygous K10-deficient mice, respectively, which resulted from a BALB/C and C57Bl/6 cross. Note that two allelic size variants were expressed. Homozygous neonatals on a C57Bl/6 background expressed exclusively the smaller sized variant (lane 3). The expression of involucrin in wild-type (lane 4), heterozygous (lane 5) and homozygous (lane 6) K10-deficient epidermis from C57Bl/6 mice was different among the genotypes. The amount of involucrin was increased in heterozygous and to an even greater extent in homozygous K10-deficient mice. Size markers are from top to bottom: 97 kDa (right), 66 kDa and 56 kDa (left). Equal amounts of protein were loaded (see Materials and Methods).

Full figure and legend (4K)

Figure 5.

Northern analysis of epidermal differentiation markers. Total RNA of trunk skin of neonatal wild-type (lane 1), heterozygous (lane 2) and homozygous (lane 3) K10-deficient mice were hybridized with probes specific for involucrin (a), loricrin (b), acid sphingomyelinase (c), and GAPDH (d). Involucrin mRNA was clearly increased in heterozygous and massively in homozygous skin.

Full figure and legend (7K)

In contrast, loricrin was not altered at the protein or mRNA expression level as shown by western blotting (Figure 3, lanes 1–3), by immunofluorescence staining (Figure 4a–c) or northern blot analysis (Figure 5b). Interestingly, we discovered allelic size variants of both loricrin and involucrin in distinct mouse strains (Figure 3 and Figure 6). We examined BALB/C, C57Bl/6, and 129/Sv mice and found that the latter two expressed a 95 kDa involucrin whereas in BALB/C mice the protein was about 5 kDa smaller. Loricrin was about 5 kDa larger in BALB/C than the corresponding protein in C57Bl/6 or 129/Sv. Mice that were derived from a cross between BALB/C and C57Bl/6 expressed both size variants. These size variants are also reflected at the mRNA level (Figure 7). At present, we do not know, however, whether the size difference seen for both mRNAs is due to coding or noncoding sequences.

Figure 6.

Size variations of involucrin and loricrin in different mouse strains. Western blots of total protein extracts from neonatal trunk skin were probed with specific anti-sera. Littermates from a cross between C57Bl/6 and BALB/C mice expressed both size variants as it is shown for a wild-type (lane 1) and a heterozygous K10-deficient animal (lane 2). A homozygous K10-deficient neonate on a C57Bl/6 background expressed the large variant (lane 3). In BALB/C mice (lane 4) the involucrin-specific anti-serum detected a band of about 95 kDa whereas C57Bl/6 (lane 5) and 129/Sv (lane 6) mice showed a protein of 100 kDa whereas BALB/C mice expressed a larger loricrin (lane 7) than C57Bl/6 (lane 8) or 129/Sv mice (lane 9) where the protein migrated at about 53 kDa. The size marker on the left (97 kDa) corresponds to lanes 1–6 and the size markers on the right (from top to bottom: 66 kDa and 56 kDa) correspond to lanes 7–9. Equal amounts of protein were loaded (see Materials and Methods).

Full figure and legend (6K)

DISCUSSION

It is well documented that the SC lipids are important factors in the formation of a functional epidermal permeability barrier (^{Elias & Menon 1991}) which is known to be impaired in EH (^{Frost et al. 1968}). Most importantly, our study has revealed that as a consequence of disrupting the gene for K10, the composition of SC lipids became altered. Notably, these changes did not lead to an impaired processing of LB contents except in cytolytic cells. SC lipids are concentrated in lamellar bodies where they form lamellar stacks. These characteristic organelles which also contain numerous hydrolases increase in number as the keratinocytes differentiate and are finally actively extruded from terminally differentiating cells. In contrast to the alterations in LB morphology and secretion that have been described for several disorders of cornification (^{Ghadially et al. 1992}; for review see^{Fartasch 1997}) our electron microscopy analysis of K10-deficient epidermis showed that the number and extrusion of LB were not markedly altered in homozygous K10-deficient mice. Also, we found regular lamellar sheets and a normal lamellar layer structure in the extracellular compartment. These data demonstrate that hyperkeratosis needs not necessarily be accompanied by visible alterations in lipid structures of the SC, although its lipid content and composition may be clearly affected.

We found that levels of ceramides 1, 3, 4, and 5 remained unchanged in K10-deficient mice whereas the other SC lipids, including ceramide 2, were increased when compared with wild-type littermates. These data are in accordance with those demonstrating a decrease in all ceramide subpopulations as a percentage of total lipid in EH (^{Paige et al. 1994}). The significant increase in ceramide 2 in our mice cannot be explained at the moment. Ceramide 2 is the major epidermal ceramide but it is not restricted to this tissue like, for example, ceramide 1, which is exclusively found in the epidermis. Moreover, ceramide synthesis of basal keratinocytes is limited to ceramide 2 production whereas generation of the full series of ceramides found in SC is restricted to differentiated keratinocytes of the stratum spinosum and granulosum (^{Madison et al. 1990}). Therefore, the observed imbalance in ceramide composition may result from an altered keratinocyte differentiation. It will be interesting to analyze further the ratio of free ceramides to CE-bound -hydroxyceramides, the latter of which are thought to play a part in the organization of the intercellular lipids and were found in a covalent linkage with involucrin (^{Swartzendruber et al. 1987;Marekov & Steinert 1998}).

Despite a seemingly intact ultrastructure of the extracellular lamellar layer we would predict an increased transepidermal water loss in K10-deficient mice. An increased transepidermal water loss might result from the occurrence of large-scale blisters (^{Porter et al. 1996}) as well as from numerous microlesions not visible to the eye (this study, see Figure 1a).

SC ceramides are predominantly derived from glucosylceramide and sphingomyelin (^{Elias et al. 1979;Lampe et al. 1983}). Here, we reported a reduction of both ceramide precursors in SC preparations of K10-deficient mice. Although the amounts of both precursors are low in the SC in comparison with SG cells, their decrease in the SC may reflect a more effective enzymatic processing in K10-deficient mice. This change did not result from an increase in the mRNA level of acid sphingomyelinase, an enzyme that is responsible for the release of ceramides from sphingomyelin. It remains to be determined whether its activity or subcellular distribution are altered in K10-deficient mice.

Whereas we observed an increase in total extractable lipids per mg SC dry weight in the K10-deficient mice, a similar study performed in samples of EH patients reported no such changes (^{Paige et al. 1994}). In acute barrier disruption, however, an increase in lipid synthesis has been observed (^{Holleran et al. 1991;Harris et al. 1997}). This apparent discrepancy can be resolved by the following observation: we have previously noted a considerable decrease in the suprabasal keratins of K10-deficient mice (^{Porter et al. 1996}). Given that about 70% of the SC dry weight consists of keratin, it easily follows that the overall change in lipids per mg SC dry weight could at least in part result from a proportional decrease in keratin.

Recently, a more than 2-fold elevation of involucrin immunostaining was reported in EH patients (^{Kanitakis et al. 1987;Ishida-Yamamoto et al. 1995}). A similar analysis in K10-deficient mice did not reveal major alterations of involucrin's distribution pattern or staining intensity. Immunofluorescence staining of cryosections is, however, certainly not a quantitative technique. By western and northern blotting a significant increase in involucrin in K10-deficient mice was very obvious, along with a similar increase in profilaggrin (^{Reichelt et al. 1997}; Reichelt et al. in preparation). At the same time, the amount of loricrin remained constant. As we have shown before, the disruption of the intermediate filament cytoskeleton is accompanied by an increase in K6 and 16 (^{Reichelt et al. 1997;Porter et al. 1998}) which are considered as markers for altered keratinocyte differentiation. The fact that we observed an increase in one CE protein whereas another one was not altered indicated that the changes were not simply a result of epidermal disruption but result from selective mechanisms triggered by the disruption of keratin 10. It will be interesting to investigate how changes in the keratin cytoskeleton lead to changes in epidermal lipids and CE proteins.

Finally, we have noted size variants of loricrin and involucrin in BALB/C and C57Bl/6 mouse strains. Two-dimensional gel electrophoresis data (not shown) do exclude phosphorylation to account for the size difference. The presence of differently sized mRNAs coding for both proteins in the two mouse strains does favor the existence of allelic variants, perhaps with varying extent of glycine repeats. A final answer has to await the cloning and sequencing of the corresponding cDNA.

References

1. Bickenbach, JR, Longley, MA, Bundman, DS, Dominey, AM, Bowden, PE, Rothnagel, JA, Roop, DR: A transgenic mouse model that recapitulates the clinical features of both neonatal and adult forms of the skin disease epidermolytic hyperkeratosis. Differentiation 1996 61:129–139,  | Article | PubMed | ISI | ChemPort |
2. Candi, E, Tarcsa, E, Digiovanna, JJ, Compton, JG, Elias, PM, Marekov, LN, Steinert, PM: A highly conserved lysine residue in the head domain of type II keratins is essential for the attachment of keratin intermediate filaments to the cornified cell envelope through isopeptide crosslinking by transglutaminases. Proc Natl Acad Sci USA 1998 95:2067–2072,  | Article | PubMed | ChemPort |
3. Candi, E, Tarcsa, E, Idler, WW, Kartasova, T, Marekov, LN, Steinert, PM: Transglutaminase cross-linking properties of the small proline-rich 1 family of cornified cell envelope proteins. Integration with loricrin. J Biol Chem 1999 274:7226–7237,  | Article | PubMed | ISI | ChemPort |
4. Cheng, J, Syder, AJ, Yu, QC, Letai, A, Paller, AS, Fuchs, E: The genetic basis of epidermolytic hyperkeratosis: a disorder of differentiation-specific epidermal keratin genes. Cell 1992 70:811–819,  | Article | PubMed | ISI | ChemPort |
5. Chipev, CC, Yang, JM, DiGiovanna, JJ, Steinert, PM, Marekov, L, Compton, JG, Bale, SJ: Preferential sites in keratin 10 that are mutated in epidermolytic hyperkeratosis. Am J Hum Genet 1994 54:179–190,  | PubMed | ISI | ChemPort |
6. Chomczynski, P, Sacchi, N: Single-step method of RNA isolation. Anal Biochem 1987 162:156–159,  | Article | PubMed | ISI | ChemPort |
7. Corden, LD, McLean, WH: Human keratin diseases: hereditary fragility of specific epithelial tissues. Exp Dermatol 1996 5:297–307,  | PubMed | ChemPort |
8. Doering, T, Holleran, WM, Potratz, A, Vielhaber, G, Elias, PM, Suzuki, K, Sandhoff, K: Sphingolipid activator proteins are required for epidermal permeability barrier formation. J Biol Chem 1999 274:11038–11045,  | Article | PubMed | ISI | ChemPort |
9. Elias, PM, Menon, GK: Structural and lipid biochemical correlates of the epidermal permeability barrier. Adv Lipid Res 1991 24:1–26,  | PubMed | ChemPort |
10. Elias, PM, Brown, BE, Fritsch, P, Goerke, J, Gray, GM, White, J: Localization and composition of lipids in neonatal mouse stratum granulosum and stratum corneum. J Invest Dermatol 1979 73:339–348,  | Article | PubMed | ISI | ChemPort |
11. Fartasch, M: Epidermal barrier in disorders of the skin. Microscopy Res Technique 1997 38:361–372,  | ISI | ChemPort |
12. Fartasch, M, Ponec, M: Improved barrier structure formation in air-exposed human keratinocyte culture systems. J Invest Dermatol 1994 102:366–374,  | Article | PubMed | ISI | ChemPort |
13. Fartasch, M, Bassukas, ID, Diepgen, TL: Structural relationship between epidermal lipid lamellae, lamellar bodies and desmosomes in human epidermis: an ultrastructural study. Br J Dermatol 1993 128:1–9,  | PubMed | ISI | ChemPort |
14. Freinkel, RK, Traczyk, TN: Lipid composition and acid hydrolase content of lamellar granules of fetal rat epidermis. J Invest Dermatol 1985 85:295–298,  | Article | PubMed | ISI | ChemPort |
15. Frost, P, Weinstein, GD, Bothwell, JW, Wildnauer, R: Ichthyosiform dermatoses. Arch Dermatol 1968 98:230–233,  | Article | PubMed | ISI | ChemPort |
16. Fuchs, E, Weber, K: Intermediate filaments: structure, dynamics, function, and disease. Annu Rev Biochem 1994 63:345–382,  | Article | PubMed | ISI | ChemPort |
17. Fuchs, E, Esteves, RA, Coulombe, PA: Transgenic mice expressing a mutant keratin 10 gene reveal the likely genetic basis for epidermolytic hyperkeratosis. Proc Natl Acad Sci USA 1992 89:6906–6910,  | PubMed | ChemPort |
18. Ghadially, R, Williams, ML, Hou, SY, Elias, PM: Membrane structural abnormalities in the stratum corneum of the autosomal recessive ichthyoses. J Invest Dermatol 1992 99:755–763,  | Article | PubMed | ISI | ChemPort |
19. Grayson, S, Johnson Winegar, AG, Wintroub, BU, Isseroff, RR, Epstein, Eh Jr, Elias, PM: Lamellar body-enriched fractions from neonatal mice: preparative techniques and partial characterization. J Invest Dermatol 1985 85:289–294,  | Article | PubMed | ISI | ChemPort |
20. Harris, IR, Farell, AM, Grunfeld, C, Holleran, WM, Elias, PM, Feingold, KR: Permeability barrier disruption coordinately regulates mRNA levels for key enzymes of cholesterol, fatty acid, and ceramide synthesis in the epidermis. J Invest Dermatol 1997 106:783–787,
21. Holleran, WM, Feingold, KR, Man, MQ, Gao, WN, Lee, JM, Elias, PM: Regulation of epidermal sphingolipid synthesis by permeability barrier function. J Lipid Res 1991 32:1151–1158,  | PubMed | ISI | ChemPort |
22. Huber, M, Rettler, I, Bernasconi, K et al: Mutations of keratinocyte transglutaminase in lamellar ichthyosis. Sci 1995 267:525–528,  | ISI | ChemPort |
23. Ishida-Yamamoto, A, Iizuka, H, Manabe, M et al: Altered distribution of keratinization markers in epidermolytic hyperkeratosis. Arch Dermatol Res 1995 287:705–711,  | Article | PubMed | ChemPort |
24. Kanitakis, J, Zambruno, G, Viac, J, Thivolet, J: Involucrin expression in keratinization disorders of the skin—a preliminary study. Br J Dermatol 1987 117:479–486,  | PubMed | ISI | ChemPort |
25. Lampe, MA, Williams, ML, Elias, PM: Human epidermal lipids: characterization and modulations during differentiation. J Lipid Res 1983 24:131–140,  | PubMed | ISI | ChemPort |
26. Landmann, L: Epidermal permeability barrier: transformation of lamellar granule-disks into intercellular sheets by a membrane-fusion process, a freeze-fracture study. J Invest Dermatol 1986 87:202–209,  | Article | PubMed | ISI | ChemPort |
27. Langbein, L, Heid, HW, Moll, I, Franke, WW: Molecular characterization of the body site-specific human epidermal cytokeratin 9: cDNA cloning, amino acid sequence, and tissue specificity of gene expression. Differentiation 1993 55:57–71,  | Article | PubMed | ISI | ChemPort |
28. Madison, KC, Swartzendruber, DC, Wertz, PW, Downing, DT: Sphingolipid metabolism in organotypic mouse keratinocyte cultures. J Invest Dermatol 1990 95:657–664,  | Article | PubMed | ISI | ChemPort |
29. Maestrini, E, Monaco, AP, McGrath, JA, et al: A molecular defect in loricrin, the major component of the cornified envelope, underlies Vohwinkel's syndrome. Nat Genet 1996 13:70–77,  | Article | PubMed | ISI | ChemPort |
30. Magin, TM: Lessons from keratin transgenic and knockout mice. In: Subcellular Biochemistry, Intermediate Filaments 1998 Vol. 31 New York: Plenum Press, pp 141–172,
31. Magin, TM, Schröder, R, Leitgeb, S, Wanninger, F, Zatloukal, K, Grund, C, Melton, DW: Lessons from keratin 18 knockout mice: formation of novel keratin filaments, secondary loss of keratin 7 and accumulation of liver-specific keratin 8-positive aggregates. J Cell Biol 1998 140:1441–1451,  | Article | PubMed | ISI | ChemPort |
32. Marekov, LN, Steinert, PM: Ceramides are bound to structural proteins of the human foreskin epidermal cornified cell envelope. J Biol Chem 1998 273:17763–17770,  | Article | PubMed | ISI | ChemPort |
33. Ming, ME, Daryanani, HA, Roberts, LP, Baden, HP, Kvedar, JC: Binding of keratin intermediate filaments (K10) to the cornified envelope in mouse epidermis: implications for barrier function. J Invest Dermatol 1994 103:780–784,  | Article | PubMed | ISI | ChemPort |
34. Moll, R, Franke, WW, Schiller, DL, Geiger, B, Krepler, R: The catalog of human cytokeratins: patterns of expression in normal epithelia, tumors and cultured cells. Cell 1982 31:11–24,  | Article | PubMed | ISI | ChemPort |
35. Motta, S, Monti, M, Sesana, S, Caputo, R, Carelli, S, Ghidoni, R: Ceramide composition of the psoriatic scale. Biochim Biophys Acta 1993 1182:147–151,  | PubMed | ISI | ChemPort |
36. Nemes, Z, Marekov, LN, Steinert, PM: Involucrin cross-linking by transglutaminase 1. Binding to membranes directs residue specificity. J Biol Chem 1999 274:11013–11021,  | Article | PubMed | ISI | ChemPort |
37. Paige, DG, Morse Fisher, N, Harper, JI: Quantification of stratum corneum ceramides and lipid envelope ceramides in the hereditary ichthyoses. Br J Dermatol 1994 131:23–27,  | PubMed | ISI | ChemPort |
38. Ponec, M, Weerheim, A, Mommaas, AM, Nugteren, DH: Lipid composition of cultured human keratinocytes in relation to their differentiation. J Lipid Res 1988 29:949–960,  | PubMed | ISI | ChemPort |
39. Porter, RM, Leitgeb, S, Melton, DW, Swensson, O, Eady, RAJ, Magin, TM: Gene targeting at the mouse cytokeratin 10 locus: severe skin fragility and changes of cytokeratin expression in the epidermis. J Cell Biol 1996 132:925–936,  | Article | PubMed | ISI | ChemPort |
40. Porter, RM, Reichelt, J, Lunny, DP, Magin, TM, Lane, EB: The relationship between hyperproliferation and epidermal thickening in a mouse model for BCIE. J Invest Dermatol 1998 110:951–957,  | Article | PubMed | ISI | ChemPort |
41. Reichelt, J, Bauer, C, Porter, R, Lane, E, Magin, M: Out of balance: consequences of a partial keratin 10 knockout. J Cell Sci 1997 110:2175–2186,  | PubMed | ISI | ChemPort |
42. Rentrop, M, Nischt, R, Knapp, B, Schweizer, J, Winter, H: An unusual type-II 70-kilodalton keratin protein of mouse epidermis exhibiting postnatal body-site specificity and sensitivity to hyperproliferation. Differentiation 1987 34:189–200,  | PubMed | ISI | ChemPort |
43. Robson, KJ, Stewart, ME, Michelsen, S, Lazo, MD, Downing, DT: 6-Hydroxy-4-sphingenin in human epidermal ceramides. J Lipid Res 1994 35:2060–2068,  | PubMed | ISI | ChemPort |
44. Rothnagel, JA, Dominey, AM, Dempsey, LD et al: Mutations in the rod domains of keratins 1 and 10 in epidermolytic hyperkeratosis. Science 1992 257:1128–1130,  | PubMed | ISI | ChemPort |
45. Sambrook, J, Fritsch, EF, Maniatis, T: Molecular Cloning: 1989 A Laboratory Manual. Cold Spring Harbor. Cold Spring Harbor Laboratory Press, pp 7.39–37.52,
46. Steinert, PM, Marekov, LN: The proteins elafin, filaggrin, keratin intermediate filaments, loricrin, and small proline-rich proteins 1 and 2 are isodipeptide cross-linked components of the human epidermal cornified cell envelope. J Biol Chem 1995 270:17702–17711,  | Article | PubMed | ISI | ChemPort |
47. Steinert, PM, Marekov, LN: Direct evidence that involucrin is a major early isopeptide cross-linked component of the keratinocyte cornified cell envelope. J Biol Chem 1997 272:2021–2030,  | PubMed | ISI | ChemPort |
48. Swartzendruber, DC, Wertz, PW, Madison, KC, Downing, DT: Evidence that the corneocyte has a chemically bound lipid envelope. J Invest Dermatol 1987 88:709–713,  | Article | PubMed | ISI | ChemPort |
49. Takahashi, K, Coulombe, PA: A transgenic mouse model with an inducible skin blistering disease phenotype. Proc Natl Acad Sci USA 1996 93:14776–14781,  | Article | PubMed | ChemPort |
50. Williams, ML: Lipids in normal and pathological desquamation. Adv Lipid Res 1991 24:211–262,  | PubMed | ChemPort |

Acknowledgments

We are very grateful to Dr. Philippe Djian and Dr. Dennis Roop for involucrin and loricrin cDNA clones. Supported by SFB 284 (K.S. and T.M.M.).