Introduction

Epidermolysis bullosa simplex (EBS) is a dominantly inherited skin disorder caused by mutations in keratin (K) 5 or K14 genes that is characterized by cytoplasmic protein aggregates and extensive cytolysis. It is widely regarded as a mechanobullous disease, resulting from an altered cytoskeleton unable to provide support against mechanical stress. Previously, we and others have generated K5 and K14 mutant mice serving as models for EBS (Lloyd et al., 1995; Cao et al., 2001; Peters et al., 2001). Of these, K5^-/- mice develop the most severe condition owing to the absence of keratin filaments in the basal epidermis, followed by a complete detachment of the epidermis from the underlying dermis. The resulting cytolysis causes death of neonates during the first hour after birth with full penetrance (Peters et al., 2001).

Although it is well established that EBS is mainly caused by mechanical trauma of an epidermis that expresses mutant keratins, additional EBS pathomechanisms have recently emerged. Tumor necrosis factor- (TNF-) has been hypothesized to contribute to the severe Dowling–Meara form of EBS, caused by a point mutation at Arg 125 in K14 (Yoneda et al., 2004). TNF- is a proinflammatory cytokine that induces apoptosis in some cell types (Yeh et al., 1998; Hsu and Twu, 2000). It exerts its effects by binding to TNF- receptors TNF-R1 and TNF-R2. The cytotoxic effects of TNF- are mediated by TNF-R1, which has a cytoplasmic death domain that interacts with an adaptor protein, TNF- receptor-associated death domain, following ligand binding. TNF- receptor-associated death domain subsequently interacts with additional adapter proteins receptor-interacting protein, TNF-R-associated factor 2, and Fas-associated death domain, which recruits and activates pro-caspase-8. Caspase-8 then oligomerizes and is activated by self-cleavage, and initiates a protease cascade that leads to apoptosis. Yoneda et al. (2004) have suggested that TNF- mediates cytotoxicity in a keratinocyte model of EBS, based on HaCaT keratinocytes transiently transfected with a K14 mutant (K14Arg125Cys). Using this cell culture model of EBS, they reported that K14 aggregates led to the release of TNF- and the subsequent activation of the TNF- receptor by an autocrine/paracrine pathway, which resulted in cell death.

In view of these reports, we considered the hypothesis that additional mechanisms and not only mechanical weakness contribute to EBS. Identification of such mechanisms would offer potential targets for novel therapy approaches. Given the significant problems accompanying gene therapy approaches of EBS, alternative efforts are indeed necessary (Kirfel et al., 2002; Arin and Roop, 2004). One clinical study has recently suggested that the systemic application of tetracycline to patients suffering from the severe Dowling–Meara form of EBS leads to a significant reduction of the blistering and less fragile epidermis (Retief et al., 1999). Tetracyclines can act as antibiotics by inhibiting bacterial protein synthesis, but recent work has shown that they are pluripotent drugs that affect many cellular functions: minocycline reduces inflammation and protects against focal cerebral ischemia (Yrjanheikki et al., 1999); tetracyclines affect prion infectivity through a direct interaction with proteinase-resistant forms of the prion protein (Forloni et al., 2002); doxycycline modulates smooth muscle cell growth, migration, and matrix remodeling after arterial injury (Bendeck et al., 2002).

Here, we report on the systemic treatment of K5^-/- mice serving as an animal model for EBS, by doxycycline. We demonstrate that the systemic application of physiological concentrations of doxycycline prolonged the life of neonatal K5^-/- mice. Affymetrix analysis revealed that this was accompanied by changes at the transcriptional level. Proinflammatory cytokines IL-6 and IL-1 were significantly upregulated at the mRNA and protein level in K5^-/- mice skin. In contrast to other studies, however, the mRNA and protein levels of TNF- remained unchanged in mice and in a cell culture model. At the same time, genes involved in tissue remodeling were repressed. We hypothesize that mutations in genes that cause EBS and related disorders not only lead to cell fragility, but cause local inflammation, possibly by mediating a stress response.

Results

Absence of TNF signaling in K5^-/- mice and in a cell culture model of EBS

EBS is a blistering cutaneous disease featuring protein aggregates. Recently, TNF--mediated cytotoxity in a keratinocyte model of EBS was reported (Yoneda et al., 2004). In that study, HaCaT keratinocytes transiently transfected with K14 (R125C) were found to release TNF- with a subsequent activation of the TNF- receptor by an autocrine/paracrine pathway, which resulted in cell death. A well-established EBS mouse model, K5^-/- mice (Peters et al., 2001) and a HaCaT keratinocyte line expressing mutant K14 (R125C) serving as EBS cell models (Werner et al., 2004) were used to investigate whether TNF- was induced in these models. First, we examined the potential role of TNF-, which is expressed in the epidermis, in K5-deficient mouse skin (Kondo and Sauder, 1997). We quantified TNF- protein expression by ELISA (Figure 1g) and found no significant difference of TNF- expression between K5^-/- (2029.8 pg/mg total protein) and wild-type (wt) control animals (19213 pg/mg total protein), P=0.49 (Figure 1g). Therefore, neither the extensive cytolysis nor K14 aggregates present in K5^-/- epidermis stimulated TNF- secretion. Consistent with these data, immunofluorescence analysis of frozen skin sections using TNF- antibodies showed no difference between K5^-/- and wt back skin (Figure 1a and b). Second, we used mouse MGU74Av2 microarrays (Affymetrix, Santa Clara, CA) for large-scale gene expression profiling in mouse skin between K5^-/- and wt controls (GSE760). A comprehensive analysis of genes relevant to the TNF- pathway failed to reveal differentially expressed genes related to the TNF- pathway when mRNA expression profile from K5^-/- and wt mice was compared (Table 1).

Figure 1.

TNF- expression in an EBS mouse and a cell culture model. (a) Immunofluorescence of TNF- in K5^-/- and (b) wt skin (bar=10 m); TNF- immunofluorescence in cultured MCF7 cell lines, (c) K14 and (d) TNF- staining in stably transfected mutant K14 ArgR125Cys MCF-7 cell line, (e) K14 and (f) TNF- staining in stably transfected wt K14wt MCF-7 cell line. Bar=10 m. (g) Quantification of TNF- protein expression in K5^-/- versus wt epidermis. Amount of TNF- protein in K5^-/- epidermis (black bar, 2029.8 pg/mg total protein, n=6) and wt (white bar, 19213 pg/mg total protein, n=6). Student's t-test showed no TNF- difference between K5^-/- and wt epidermis, P=0.49. (h) Quantification of TNF- protein expression. Measurements were taken in K14 ArgR125Cys versus wt K14wt MCF-7-transfected cell lines using ELISA (n=3).

Full figure and legend (207K)

Table 1 - Expression of genes related to the TNF- pathway analyzed by MGU74Av2 microarray.

Full table

Given that these data were in contrast to a previous study using HaCaT keratinocytes transiently transfected with the most frequent EBS mutant K14 Arg125Cys, we reasoned that the failure to induce TNF signaling in K5^-/- mice might be due to the presence of wt K14 (Yoneda et al., 2004). To extend our in vivo analysis to a cell culture model, we switched to HaCaT and to MCF-7 cells, which were both stably transfected with the mutant K14 Arg125Cys created previously (Werner et al., 2004). In both cell types, we found no changes in TNF- by immunofluorescence analysis (Figure 1c–f). Next, TNF- levels secreted into the medium and inside cells were measured with a solid-phase ELISA. Repeated measurements revealed that TNF- was not increased in the media supernatants of K14 mutant (16.6 pg/ml) compared to those from control transfectants (20.1 pg/ml; Figure 1h). The expression level of TNF- protein in mutant-transfected cells (44 pg/ml) was lower as in mock (47.7 pg/ml)- or K14 wt-transfected cells (57 pg/ml; Figure 1h). Collectively, our data have demonstrated that neither the absence of K5 nor the presence of mutant K14 or of keratin aggregates stimulate TNF- secretion in both EBS models. To our knowledge, TNF- levels have not been examined in EBS patients.

Increased IL-6 and IL-1 mRNA levels in K5^-/- skin

The absence of TNF signaling does not exclude a role of other proinflammatory cytokines. Therefore, we performed a microarray analysis from neonatal K5^-/- and wt skin. This revealed that 85 genes were up- and 31 were downregulated upon loss of K5. Of these, the mRNA for IL-6 was increased 3.46-fold in K5^-/- skin versus wt skin, the one encoding chemokine (C–X–C motif) ligand 1 was upregulated 4.05-fold (Table 2). This was confirmed by TaqMan real-time quantitative reverse transcription (RT)-PCR (Figure 2a). IL-6 was significantly upregulated in the skin of K5^-/- mice (8.47-fold increase, P<0.05). The transcript of chemokine (C–X–C motif) was little in wt, but statistics show no significant difference comparing with K5^-/- skin (Figure 2b, P=0.26). IL-1 was also increased in the skin of K5^-/- mice (5.72-fold increase, P<0.05; Figure 2a). Furthermore, we analyzed IL-6 and IL-1 at the protein level in K5^-/- and wt epidermis. In the former, the content of IL-1 was increased significantly (8.800.18 pg/mg total protein, P<0.05) compared with wt controls (7.250.15 pg/mg total protein). Owing to the very low levels of IL-6, however, statistic analysis showed that there was no significant difference of IL-6 expression between K5^-/- (0.0860.036 pg/mg total protein) and wt control mice (0.0570.028 192 13 pg/mg total protein, P=0.52, Figure 2c).

Figure 2.

IL-6 and IL-1 induction in K5^-/- mice. (a) Changes in TaqMan real-time relative quantitative RT-PCR with glyceraldehyde-3-phosphate dehydrogenase as reference gene. Both IL-1 and IL-6 were increased in the skin of K5^-/- mice (5.72- and 8.47-fold increase, P<0.05), white: controls (n=6), black: K5 ^-/- (n=6). (b) C–X–C motif real-time relative quantitative RT-PCR with glyceraldehyde-3-phosphate dehydrogenase expression, P=0.26. (c) The content of IL-1 in K5^-/- epidermis (black bar, 447.311 pg/mg total protein, n=6) and wt (white bar, 3627.4 pg/mg total protein n=6). P<0.05. The content of IL-6 in K5^-/- epidermis (black bar, 69.010.9 pg/mg total protein, n=6) and wt (white bar, 20.17.6 pg/mg total protein, n=6) P< 0.05.

Full figure and legend (18K)

Table 2 - Selected upregulated genes in K5^-/- compared to wt skin.

Full table

Extended survival of K5^-/- mice following small molecule treatment

K5, together with K14 and K15, forms the keratin cytoskeleton of basal epidermis. Previously, we reported that K5^+/- mice die in the first hour after birth (n=200) with full penetrance because the inability to form a keratin cytoskeleton in their basal epidermis causes extensive fragility. In a first attempt to improve this condition, pregnant K5^+/- females were systemically treated with doxycycline, starting at E13.5. This was based on our previous finding that the formation of keratin aggregates typical of EBS started at E13.5 in embryonic stage (Lu et al., 2005). Treatment with the drug prolonged the survival time of 34 K5^-/- pups up to 8 hours (Table 3).

Table 3 - Surviving time of doxycycline-treated and -untreated K5 knockout neonatal mice.

Full table

To analyze the mechanisms responsible for the prolonged survival at the molecular level, we used Affymetrix Genechips mouse MGU74Av2 for large-scale gene expression profiling in epidermis and in skin after treatment with doxycycline and compared it to vehicle-treated controls (GSE760). In K5^-/- epidermis, 159 genes were differentially expressed after doxycycline treatment. Fifty-nine genes were upregulated after doxycycline treatment and 100 genes were suppressed. Here, we focus on genes relevant to inflammation and tissue repair (Table 4). Most importantly, in K5^-/- epidermis, doxycycline downregulated matrix metalloproteinase 13 (MMP-13, also called collagenase 3) mRNA expression (0.4-fold decrease), and upregulated various procollagens (procollagen type III 1, 39.95-fold increase; procollagen type I 1, 18.38-fold increase; procollagen type VI 3, 15.03-fold increase). Moreover, doxycycline repressed the expression of genes encoding IL-1 (0.28-fold decrease), IFN-activated gene 202B (0.24-fold decrease), chemokine ligand 1 (0.21-fold decrease), serine protease 12 (0.11-fold decrease). On the other hand, it activated the serine proteinase inhibitor gene (6.41-fold increase).

Table 4 - Differentially expressed genes in the microarray profiling in K5^-/- epidermis following doxycycline treatment.

Full table

Next, TaqMan real-time relative quantitative RT-PCR was performed with glyceraldehyde-3-phosphate dehydrogenase as the reference gene to validate the microarray results (Figure 3a and b). In line with the microarray data, mRNA expression of MMP-13 and IL-1 were significantly downregulated in the epidermis of doxycycline-treated K5^-/- mice (0.263, 0.339, 0.295, P<0.05; Figure 3a). On the other hand, real-time RT-PCR result did not confirm the downregulation of IFN-activated gene 202B and the upregulation of procollagen. The statistical data showed no significant changes (Figure 3b) because of significant deviation between samples. This result was consistent with previous data on the doxycycline treatment following myocardial infarction. In this setting, left ventricular remodeling occurred in the post-infarction period by inhibition of collagenase activity without changing collagen (Villarreal et al., 2003). However, there was no significant difference when we checked the content of IL-1 in K5^-/-epidermis after doxycycline versus no treatment by ELISA (data was not shown). Using Western blotting, we found a slightly decreased amount of active MMP-13 in the epidermis after doxycycline treatment (Figure 4). We hypothesize that these changes represent the initiation of "matrix remodeling" and contributes to a reduced bullae formation and prolongs the life of K5^-/- mice. These results indicate that doxycycline can partially rescue K5^-/- mice and support the previous clinical study (Retief et al., 1999; Weiner et al., 2004).

Figure 3.

Validation of gene expression changes in K5^-/- after doxycycline treatment using TaqMan real-time relative quantitative RT-PCR. (a) MMP-13 and IL-1 were decreased in the epidermis of doxycycline-treated K5^-/- mice (0.263, 0.295, P<0.05). (b) No mRNA expression changes of IFN and procollagen. White bar, doxycycline untreated (n=6); black bar, doxycycline treated (n=6).

Full figure and legend (11K)

Figure 4.

Analysis of MMP-13 in protein extracts of skin of K5^-/- mice. (a) Equal amounts (30 g) of protein extracts, isolated from epidermis (epi) of K5^-/- mice treated/untreated with doxycycline were assayed by immunoblot analysis for the presence of latent (60 kDa) and active (48 kDa) and MMP-13. Coomassie Blue staining (lower panel) of a gel run in parallel was used as loading control. (b) Protein levels normalized to Coomassie Blue staining, normalization of this representative experiment is shown.

Full figure and legend (69K)

Discussion

TNF- is not involved in blister formation in EBS

TNF- is a multifunctional cytokine and belongs to a family of ligands that can bind to a family of receptors. The pleiotropic actions of TNF range from proliferative responses such as cell growth and differentiation to inflammatory effects and the mediation of immune responses, to destructive cellular outcomes such as apoptotic and necrotic cell death mechanisms (MacEwan, 2002). Although some skin disease like psoriasis – a disease of keratinocyte proliferation and inflammation, which follow infiltration by activated T-helper cells and mononuclear cell and the release of TNF-, EBS is clearly a well-established mechanobullous disease resulting from mutations in epidermal keratins with subsequent skin fragility and keratinocyte cytolysis, particularly following mild mechanical stress (Fuchs and Cleveland, 1998). Recently, however, the secretion of TNF- was reported from keratinocytes transiently transfected with K14 Arg125Cys. The subsequent stimulation of TNF-R1 and caspase-dependent apoptosis in an autocrine/paracrine pathway were suggested to play an important role in the mechanism linking keratin aggregates to cell death (Yoneda et al., 2004). Using the same mutant K14 ArgR125Cys stably transfected into the same keratinocytes, we were unable to identify any changes in TNF- release by several, highly sensitive methods. In support, the analysis of K5^-/- mice by microarray analysis and quantitative analysis of TNF- at the protein level have not shown any difference compared with wt mouse samples. Therefore, our in vitro and in vivo data strongly argue that TNF- is not involved in EBS. Possibly, the setting of the previous experiment – transiently transfected cells – has contributed to the apparent difference.

We note, that on the other hand, IL-6 and IL-1 are significantly upregulated in K5^-/- skin. IL-1 is an important mediator of inflammation and tissue damage in multiple organs, both in experimental animal models of disease and in human diseases. The IL-1 family consists of two agonists, IL-1 and IL-1, two receptors, biologically active IL-1RI and inert IL-1RII, and a specific receptor antagonist, IL-1Ra. Epidermal keratinocytes have been shown to produce IL-1 and IL-1 and their receptors. IL-6 is similar to IL-1 in that it stimulates T- and B-cell proliferation and production of acute-phase reactants (Kishimoto, 1989), and it has been found to stimulate keratinoncyte growth in vitro (Gallucci et al., 2004). Both IL-1 and IL-6 mediate local inflammation. This is consistent with local inflammation that can accompany rare and severe cases of EBS (Fine and Johnson, 1988). We posit that cytokine secretion in EBS can result from cytolysis, a process known to involve mechanical and/or oxidative stress in keratinocytes. Furthermore, EBS keratinocytes are in an activated state, as indicated by the expression of K6 (Fuchs and Cleveland, 1998).

Doxycycline can repress inflammation and remodeling tissues in an EBS mouse model

In the past 20 years, remarkable advances have been made in our understanding of major subtypes of EBS. These include the more complete characterization of the breadth, severity, and relative frequency of cutaneous and extracutaneous manifestations within each EBS, and elucidation of the molecular basis of most EBS subtypes (Fuchs et al., 1994; Fine and Eady, 1999). Although eventually some of this knowledge may result in the successful development of beneficial gene therapies, at present there are no treatments available that appear to result in long-term reduction in skin blistering or improvement of mechanical skin fragility in EBS. This was particularly disappointing, because even those patients with the most localized forms of EBS experience tremendous alterations in lifestyle and quality of life. Malkinson and co-workers described their experience with oral tetracycline treatment of a father and son affected with the Dowling–Meara subtype of EBS (Retief et al., 1999). Tetracycline was initially administered to a patient to treat acne, and significant reduction in both the number of blisters and skin fragility were noted. A dose-dependent response was observed, with an effective dose of 1,500 mg/day. An identical response was observed in the affected parent. Most importantly, a long-standing benefit was observed, with recurrence of disease activity as the dosage of tetracycline was reduced. This study did not address the mechanism of tetracycline affecting skin blister formation in a keratin disorder such as EBS.

In this study, we applied doxycycline to K5^-/- mice, an animal model of EBS, to get a first insight into the mechanism of tetracycline action in diseases including EBS. Doxycyline was chosen due to several reasons. First, it is used widely in tet-on or tet-off systems in mice, where the correct dose and additional parameters have been optimized (Gossen and Bujard, 1992; Boy et al., 2006). Second, doxycycline can be administered orally and is rapidly and almost completely absorbed from the gastrointestinal tract and the interference by food with absorption of doxycycline is not significant. The most important finding in our study is that doxycycline extended the survival time of K5^-/- mice from 1 to 8 hours. This was accompanied by changes in the transcriptional activity of 160 genes. This change puts a cautionary note on the widespread use of tetracycline derivatives in conditional transgenesis in mice.

By microarray and Taqman realtime PCR analysis, the mRNA-encoding MMP-13 decreased after doxycycline treatment. This demonstrated that doxycycline might downregulate MMP-13 activity that controls matrix degradation and thereby reduce bullae formation. This is also consistent with observation from patients in which tetracycline suppressed the formation of new bullae (Retief et al., 1999; Weiner et al., 2004). Doxycyclines and other derivatives of tetracycline are potent MMP inhibitors (Sapadin and Fleischmajer, 2006). Studies with chemically modified tetracyclines have shown that the antibiotic and anti-MMP activities lie in different regions of the molecule; with the antibiotic activity residing at the dimethylamino group at the carbon-4 position of the A ring, whereas the anti-MMP activity resides in the carbon-11 carbonyl oxygen and carbon-12 hydroxyl groups (Ryan et al., 1996). On the basis of this anti-MMP activity, they have been used to reduce tissue degradation in aortic aneurysms (Petrinec et al., 1996; Boyle et al., 1998; Curci et al., 1998, 2000; Kaito et al., 2003), periodontal disease (Ryan et al., 1996), as well as arthritis (Greenwald et al., 1987, 1998; Brandt, 1995; Steinmeyer et al., 1998; Shlopov et al., 1999), and are used to prevent tumor cell invasion, metastasis (Seftor et al., 1998), tumor angiogenesis (Tamargo et al., 1991), arterial remodeling (Bendeck et al., 2002), and ocular surface repair (Smith and Cook, 2004).

Using transcriptional profiling, our study has revealed that the positive effect of doxycycline on blister formation is not only exerted by dowregulation of MMP-13, but also by a significant downregulation of IL-1, decreasing inflammation. Of note, the Dowling–Meara subtype of EBS is characterized by the presence of inflammatory blister (Fine and Eady, 1999). Moreover, EBS with circinate erythema, accompanied by local inflammation has been described (Gu et al., 2003). It will be informative to investigate inflammatory cytokines in such patients. The effect of tetracycline derivatives are in line with an inhibited induction of IL-1-converting enzyme to reduce inflammation and to protect against focal cerebral ischemia (Yrjanheikki et al., 1999). Moreover, doxycycline was reported to inhibit IL-1 in the corneal epithelium and to be most effective in the treatment of inflammatory ocular conditions (Solomon et al., 2000; Smith and Cook, 2004). Although our results were not confirmed in protein level, this might be due to the oral application of doxycycline to pregnant K5^+/- females starting from E13.5. Possibly, the level of the drug decreases quickly in newborn K5^-/- embryos that might prevent further improvement. We did not succeed in administering drugs to neonatal K5^-/- pups. In the clinical report, a long-term improvement was observed, with recurrence of disease activity as the doses of tetracycline was reduced (Retief et al., 1999).

Recently, a number of reports have revealed that well-known antibiotics play multiple roles beyond their antibacterial activities: tetracyclines reduce prion infectivity through a direct interaction with the proteinase-resistant form of the prion protein (Forloni et al., 2002), and penicillin and ceftriaxone have potential as neurotherapeutics for treating neurodegenerative disease by acting at the trancriptional level (Rothstein et al., 2005). Given the large number of drugs already approved by the FDA, a systematic screen of these compounds for additional targets may not only unravel their mode of action but also provide novel therapy options in diseases difficult to treat.

Materials and Methods

Animal treatment

K5^+/- and K5^-/- mice (BALB/c) were originally generated in our lab (Peters et al. 2001). All mice were kept under the rule of the animal license 50.203.2-BN 24, 20/04. Experiments were approved by the Regierungspraesidium Koeln (Germany). To get doxycycline-treated or -untreated neonatal pups, doxycycline (50 g/ml doxycycline with 5% sucrose) was administered by oral route to pregnant K5^+/- females (mated to K5^+/- males) starting from E13.5. Five percent sucrose solution was administered to the control group.

Total RNA isolation

Skin was removed from neonatal mice and immediately frozen in liquid nitrogen. RNA was extracted with TRIzol reagent (Life Technologies-BRL, Karlsruhe, Germany) according to the protocol of the supplier. Briefly, 50–100 mg of tissue samples were homogenized in 1 ml TRIzol reagent with a power homogenizer and then extracted with 0.2 ml chloroform. The aqueous phase was precipitated by 0.5 ml isopropyl alcohol. The RNA pellets were washed with 75% ethanol and dissolved in RNase-free water. RNA concentration was determined by UV spectrophotometer, and then RNA quality was checked on a 0.8% agarose gel. Total RNA samples were stored at -80°C.

Microarray analysis

Mouse MGU74Av2 microarrays (Affymetrix, Santa Clara, CA) were used for large-scale gene expression profiling in epidermis or skin after doxycycline treatment versus controls. The following six groups of tissues were used in the study. Three groups were K5^-/- epidermis, K5^-/- skin, and wt skin treated with doxycycline, another three groups were control. Six animals were included in each group. All the total RNA samples were isolated from tissues taken immediately after birth, and were pooled for later microarray experiments. Array hybridization was carried out at the core facility of the University of Münster. Scanned images were analyzed to identify the differentially expressed genes with MA5.0 software (Affymetrix).

Real-time RT-PCR

Quantification of five selected transcripts was performed by relative quantitative real-time RT-PCR (ABI PRISM 7700 SDS; PE Biosystems, Foster City, CA) (Table 5) (Chen et al., 2001). We carried out relative quantification of starting mRNA copy numbers according to the threshold cycle (Ct) method (Fink et al., 1998). Glyceraldehyde-3-phosphate dehydrogenase was used as the endogenous reference gene as it does not exhibit significant expression changes between EBS and control mice, or doxycycline-treated and -untreated mice. The reaction was performed in a MicroAmp Optical 96-Well Reaction Plate. Eurogentec Reverse Transcriptase qPCR Master Mix (cat. no. RT-QPRT-032X, Eurogentec, Scraing, Belgium) was used for a one-tube-one-step RT-PCR in a total reaction volume of 12.5 l. Relative quantification started from each reaction containing 10 ng of total RNA. Reaction conditions were optimized for primers, probes, and MgCl₂ concentration as outlined. Cycling conditions were 48°C for 30 minutes for RT step, followed by 95°C for 10 minutes, and a two-step PCR with 60 cycles of 95°C for 15 seconds and 59°C for 1 minute.

Table 5 - Sequences and concentration ratios of primers and hydrolysis probes of reference genes with labeling dyes (F for FAM as reporter dye, T for TAMRA as quencher dye), which were used for real-time RT-PCR.

Full table

Cell culture

Four cell lines were used throughout this study: human HaCaT keratinocytes stably transfected with wt K14 or with K14(Arg125Cys), human MCF7-wt K14-, and MCF7-K14Arg125Cys-transfected (Werner et al., 2004). Cells were grown on polystyrene flasks or dishes at 37°C in a 95% O₂/5% CO₂ atmosphere. Cells were maintained in DMEM with 10% fetal calf serum, penicillin/streptomycin (2 mM), and L-glutamine (100 U/ml).

ELISA

The content of TNF-, IL-6, and IL-1 in mouse skin or epidermis, as well as TNF- in cell culture medium and lysate were measured with a solid-phase ELISA kits (Biosource, Camarillo, CA), according to the manufacturer's instruction. Briefly, mice were killed immediately after birth. Skin and epidermis of K5^-/- mice was obtained by using tweezers. For preparation of wt epidermis and whole-cell homogenates, we used the method described by Arnott et al. (2004). The total protein concentration was determined using the BCA protein assay. Identical amounts of total proteins were applied for ELISA detection. The absorbance of each well was measured at 450 nm with a BioRad spectrophotometer.

Immunofluorescence analysis

Preparation of tissue samples and immunofluorescence analysis were carried out as before (Peters et al., 2001). The following primary antibodies were used: AF 138 against K5 (1:600, Hiss Diagnostics, Freiburg, Germany), -CK14.2 against K14 (1:700, kindly provided by L. Langbein, German Cancer Research Center, Heidelberg, Germany), M18 against TNF- (1:10, Santa Cruz Biotechnology, Santa Cruz, CA), Image analysis and processing were performed using the AxionVision 4.3 (Carl Zeiss, Goettingen, Germany) and Adobe Photoshop 6.0 software. Secondary antibodies used for immunofluorescence studies in recommended dilutions were Alexa 488A- or Alexa 594-conjugated IgG (Molecular Probes, Leiden, The Netherlands). Immunofluorescence analysis of cells was described by our lab (Werner et al., 2004).

Immunoblot analysis

For preparation of mouse skin extracts, frozen mouse tissue was minced and extracted in 2% SDS/phosphate-buffered saline solution, sonicated, and incubated overnight at 4°C. The homogenate was cleared by centrifugation at 14,000 r.p.m. at 4°C for 15 minutes and processed for immunoblot analysis. Protein concentration was assessed using the BCA^® Protein assay according to the manufacturer's instructions (Pierce, Rockford, IL). Proteins were separated on SDS-polyacrylamide gels (10% acrylamide) with 10 g of the samples. Gels were run in duplicate, one gel was stained with Coomassie Blue (Serva, Heidelberg, Germany), the other was transferred to nitrocellulose membranes and transfer efficiency controlled by Ponceau Red staining (Sigma-Aldrich, St Louis, MO). Membranes were blocked with 5% milk in TBST for 1 hour at room temperature. The MMP-13 antibody K73/6, kindly provided by Dr R M Hembry (School of Biological Sciences, University of East Anglia, Norwich, UK), was used at a dilution of 1:500 in TBST. After several washings with TBST, bound primary antibody was revealed using a peroxidase-conjugated secondary antibody and specific immunocomplexes were detected using an ECL western blotting detection kit (Amersham, Freiburg, Germany). Quantification of signal intensities was performed using ImageQuant software supplied to "Personal Densitometer" from Molecular Dynamics (Krefeld, Germany). Signal intensities of specific bands were normalized to the intensity of the total proteins detected by Coomassie Blue staining.

Conflict of Interest

The authors state no conflict of interest.

References

1. Arin MJ, Roop DR (2004) Inducible mouse models for inherited skin diseases: implications for skin gene therapy. Cells Tissues Organs 177:160–168 | Article | PubMed | ISI |
2. Arnott CH, Scott KA, Moore RJ, Robinson SC, Thompson RG, Balkwill FR (2004) Expression of both TNF-alpha receptor subtypes is essential for optimal skin tumour development. Oncogene 23:1902–1910 | Article | PubMed | ISI | ChemPort |
3. Bendeck MP, Conte M, Zhang M, Nili N, Strauss BH, Farwell SM (2002) Doxycycline modulates smooth muscle cell growth, migration, and matrix remodeling after arterial injury. Am J Pathol 160:1089–1095 | PubMed | ChemPort |
4. Boy J, Leergaard TB, Schmidt T, Odeh F, Bichelmeier U, Nuber S et al. (2006) Expression mapping of tetracycline-responsive prion protein promoter: Digital atlasing for generating cell-specific disease models. Neuroimage 33:449–462 | Article | PubMed |
5. Boyle JR, McDermott E, Crowther M, Wills AD, Bell PR, Thompson MM (1998) Doxycycline inhibits elastin degradation and reduces metalloproteinase activity in a model of aneurysmal disease. J Vasc Surg 27:354–361 | Article | PubMed | ISI | ChemPort |
6. Brandt KD (1995) Modification by oral doxycycline administration of articular cartilage breakdown in osteoarthritis. J Rheumatol Suppl 43:149–151 | PubMed | ChemPort |
7. Cao T, Longley MA, Wang XJ, Roop DR (2001) An inducible mouse model for epidermolysis bullosa simplex: implications for gene therapy. J Cell Biol 152:651–656 | Article | PubMed | ISI | ChemPort |
8. Chen J, Sochivko D, Beck H, Marechal D, Wiestler OD, Becker AJ (2001) Activity-induced expression of common reference genes in individual cns neurons. Lab Invest 81:913–916 | Article | PubMed | ISI | ChemPort |
9. Curci JA, Mao D, Bohner DG, Allen BT, Rubin BG, Reilly JM et al. (2000) Preoperative treatment with doxycycline reduces aortic wall expression and activation of matrix metalloproteinases in patients with abdominal aortic aneurysms. J Vasc Surg 31:325–342 | Article | PubMed | ISI | ChemPort |
10. Curci JA, Petrinec D, Liao S, Golub LM, Thompson RW (1998) Pharmacologic suppression of experimental abdominal aortic aneurysms: acomparison of doxycycline and four chemically modified tetracyclines. J Vasc Surg 28:1082–1093 | Article | PubMed | ChemPort |
11. Fine JD, Eady RA (1999) Tetracycline and epidermolysis bullosa simplex: a new indication for one of the oldest and most widely used drugs in dermatology? Arch Dermatol 135:981–982 | Article | PubMed | ChemPort |
12. Fine JD, Johnson L (1988) Evaluation of the efficacy of topical bufexamac in epidermolysis bullosa simplex. A double-blind placebo-controlled crossover trial. Arch Dermatol 124:1669–1672 | Article | PubMed | ChemPort |
13. Fink L, Seeger W, Ermert L, Hanze J, Stahl U, Grimminger F et al. (1998) Real-time quantitative RT-PCR after laser-assisted cell picking. Nat Med 4:1329–1333 | Article | PubMed | ISI | ChemPort |
14. Forloni G, Iussich S, Awan T, Colombo L, Angeretti N, Girola L et al. (2002) Tetracyclines affect prion infectivity. Proc Natl Acad Sci USA 99:10849–10854 | Article | PubMed | ChemPort |
15. Fuchs E, Cleveland DW (1998) A structural scaffolding of intermediate filaments in health and disease. Science 279:514–519 | Article | PubMed | ISI | ChemPort |
16. Fuchs E, Coulombe P, Cheng J, Chan YM, Hutton E, Syder A et al. (1994) Genetic bases of epidermolysis bullosa simplex and epidermolytic hyperkeratosis. J Invest Dermatol 103:25S–30S | Article | PubMed | ChemPort |
17. Gallucci RM, Sloan DK, Heck JM, Murray AR, O'Dell SJ (2004) Interleukin 6 indirectly induces keratinocyte migration. J Invest Dermatol 122:764–772 | Article | PubMed | ISI | ChemPort |
18. Gossen M, Bujard H (1992) Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proc Natl Acad Sci USA 89:5547–5551 | Article | PubMed | ChemPort |
19. Greenwald RA, Golub LM, Lavietes B, Ramamurthy NS, Gruber B, Laskin RS et al. (1987) Tetracyclines inhibit human synovial collagenase in vivo and in vitro. J Rheumatol 14:28–32 | PubMed | ChemPort |
20. Greenwald RA, Golub LM, Ramamurthy NS, Chowdhury M, Moak SA, Sorsa T (1998) In vitro sensitivity of the three mammalian collagenases to tetracycline inhibition: relationship to bone and cartilage degradation. Bone 22:33–38 | Article | PubMed | ISI |
21. Gu LH, Kim SC, Ichiki Y, Park J, Nagai M, Kitajima Y (2003) A usual frameshift and delayed termination codon mutation in keratin 5 causes a novel type of epidermolysis bullosa simplex with migratory circinate erythema. J Invest Dermatol 121:482–485 | Article | PubMed | ISI | ChemPort |
22. Hsu HY, Twu YC (2000) Tumor necrosis factor-alpha -mediated protein kinases in regulation of scavenger receptor and foam cell formation on macrophage. J Biol Chem 275:41035–41048 | Article | PubMed | ChemPort |
23. Kaito K, Urayama H, Watanabe G (2003) Doxycycline treatment in a model of early abdominal aortic aneurysm. Surg Today 33:426–433 | Article | PubMed | ChemPort |
24. Kirfel J, Peters B, Grund C, Reifenberg K, Magin TM (2002) Ectopic expression of desmin in the epidermis of transgenic mice permits development of a normal epidermis. Differentiation 70:56–68 | Article | PubMed | ChemPort |
25. Kishimoto T (1989) The biology of interleukin-6. Blood 74:1–10 | PubMed | ISI | ChemPort |
26. Kondo S, Sauder DN (1997) Tumor necrosis factor (TNF) receptor type 1 (p55) is a main mediator for TNF-alpha-induced skin inflammation. Eur J Immunol 27:1713–1718 | Article | PubMed | ISI | ChemPort |
27. Lloyd C, Yu QC, Cheng J, Turksen K, Degenstein L, Hutton E et al. (1995) The basal keratin network of stratified squamous epithelia: defining K15 function in the absence of K14. J Cell Biol 129:1329–1344 | Article | PubMed | ISI | ChemPort |
28. Lu H, Hesse M, Peters B, Magin TM (2005) Type II keratins precede type I keratins during early embryonic development. Eur J Cell Biol 84:709–718 | Article | PubMed | ChemPort |
29. MacEwan DJ (2002) TNF receptor subtype signalling: differences and cellular consequences. Cell Signal 14:477–492 | Article | PubMed | ISI | ChemPort |
30. Peters B, Kirfel J, Bussow H, Vidal M, Magin TM (2001) Complete cytolysis and neonatal lethality in keratin 5 knockout mice reveal its fundamental role in skin integrity and in epidermolysis bullosa simplex. Mol Biol Cell 12:1775–1789 | PubMed | ISI | ChemPort |
31. Petrinec D, Liao S, Holmes DR, Reilly JM, Parks WC, Thompson RW (1996) Doxycycline inhibition of aneurysmal degeneration in an elastase-induced rat model of abdominal aortic aneurysm: preservation of aortic elastin associated with suppressed production of 92 kD gelatinase. J Vasc Surg 23:336–346 | Article | PubMed | ISI | ChemPort |
32. Retief CR, Malkinson FD, Pearson RW (1999) Two familial cases of epidermolysis bullosa simplex successfully treated with tetracycline. Arch Dermatol 135:997–998 | Article | PubMed | ChemPort |
33. Rothstein JD, Patel S, Regan MR, Haenggeli C, Huang YH, Bergles DE et al. (2005) Beta-lactam antibiotics offer neuroprotection by increasing glutamate transporter expression. Nature 433:73–77 | Article | PubMed | ISI | ChemPort |
34. Ryan ME, Ramamurthy S, Golub LM (1996) Matrix metalloproteinases and their inhibition in periodontal treatment. Curr Opin Periodontol 3:85–96 | PubMed | ChemPort |
35. Sapadin AN, Fleischmajer R (2006) Tetracyclines: nonantibiotic properties and their clinical implications. J Am Acad Dermatol 54:258–265 | Article | PubMed |
36. Seftor RE, Seftor EA, De Larco JE, Kleiner DE, Leferson J, Stetler-Stevenson WG et al. (1998) Chemically modified tetracyclines inhibit human melanoma cell invasion and metastasis. Clin Exp Metastasis 16:217–225 | Article | PubMed | ISI | ChemPort |
37. Shlopov BV, Smith GN Jr, Cole AA, Hasty KA (1999) Differential patterns of response to doxycycline and transforming growth factor beta1 in the down-regulation of collagenases in osteoarthritic and normal human chondrocytes. Arthritis Rheum 42:719–727 | Article | PubMed | ChemPort |
38. Smith VA, Cook SD (2004) Doxycycline-a role in ocular surface repair. Br J Ophthalmol 88:619–625 | Article | PubMed | ChemPort |
39. Solomon A, Rosenblatt M, Li DQ, Liu Z, Monroy D, Ji Z et al. (2000) Doxycycline inhibition of interleukin-1 in the corneal epithelium. Invest Ophthalmol Vis Sci 41:2544–2557 | PubMed | ISI | ChemPort |
40. Steinmeyer J, Daufeldt S, Taiwo YO (1998) Pharmacological effect of tetracyclines on proteoglycanases from interleukin-1-treated articular cartilage. Biochem Pharmacol 55:93–100 | Article | PubMed | ChemPort |
41. Tamargo RJ, Bok RA, Brem H (1991) Angiogenesis inhibition by minocycline. Cancer Res 51:672–675 | PubMed | ISI | ChemPort |
42. Villarreal FJ, Griffin M, Omens J, Dillmann W, Nguyen J, Covell J (2003) Early short-term treatment with doxycycline modulates postinfarction left ventricular remodeling. Circulation 108:1487–1492 | Article | PubMed | ISI | ChemPort |
43. Weiner M, Stein A, Cash S, de Leoz J, Fine JD (2004) Tetracycline and epidermolysis bullosa simplex: a double-blind, placebo-controlled, crossover randomized clinical trial. Br J Dermatol 150:613–614 | Article | PubMed | ChemPort |
44. Werner NS, Windoffer R, Strnad P, Grund C, Leube RE, Magin TM (2004) Epidermolysis bullosa simplex-type mutations alter the dynamics of the keratin cytoskeleton and reveal a contribution of actin to the transport of keratin subunits. Mol Biol Cell 15:990–1002 | Article | PubMed | ChemPort |
45. Yeh WC, Pompa JL, McCurrach ME, Shu HB, Elia AJ, Shahinian A et al. (1998) FADD: essential for embryo development and signaling from some, but not all, inducers of apoptosis. Science 279:1954–1958 | Article | PubMed | ISI | ChemPort |
46. Yoneda K, Furukawa T, Zheng YJ, Momoi T, Izawa I, Inagaki M et al. (2004) An autocrine/paracrine loop linking keratin 14 aggregates to tumor necrosis factor alpha-mediated cytotoxicity in a keratinocyte model of epidermolysis bullosa simplex. J Biol Chem 279:7296–7303 | Article | PubMed | ISI | ChemPort |
47. Yrjanheikki J, Tikka T, Keinanen R, Goldsteins G, Chan PH, Koistinaho J (1999) A tetracycline derivative, minocycline, reduces inflammation and protects against focal cerebral ischemia with a wide therapeutic window. Proc Natl Acad Sci USA 96:13496–13500 | Article | PubMed | ChemPort |

Acknowledgments

These studies were supported by GRK 804. Part of this work was supported by the Deutsche Forschungsgemeinschaft (FOR 367, TP 6 to T.M.M.), by DEBRA UK (to T.M.M.) and by the Bonner Forum Biomedizin (to T.M.M.). Lu H received a scholarship from Deutsche Forschungsgemeinschaft GRK804 from 2002 to 2005. We also thank Dr Loeffek and Dr Roth for discussion.