Author contributions: L.A., M.L., M.S., and D.O. designed research; L.A., M.L., E.C., P.F., A.A.-B., and D.O. performed research; F.A.P., A.R.L., and G.Z. contributed new reagents/analytic tools; L.A., M.L., E.C., J.-M.E., M.S., and D.O. analyzed data; and M.S. and D.O. wrote the paper.
Despite exhibiting different phenotypes, the UV-sensitive syndromes trichothiodystrophy (TTD) and xeroderma pigmentosum (XP) result from the same mutated genes encoding specific subunits of the DNA-repair/transcription factor TFIIH. The widely accepted notion is that XP cancer proneness results from mutations interfering with DNA repair, whereas TTD clinical hallmarks (namely, hair anomalies, physical and mental retardation, and premature aging) are because of transcriptional alterations. The present study identifies a TTD-specific overexpression of matrix metalloproteinase 1, leading to collagen I degradation in patient skin. This finding explains the clinical features distinguishing TTD from XP (i.e. bone alterations, pregnancy abnormalities and likely lack of skin cancer despite the accumulation of unrepaired DNA lesions) and the features that TTD shares with osteogenesis imperfecta caused by COL1 mutations.
Keywords: NER-defective disorders, TFIIH, transcription, MMP-1, collagen degradationMutations in the XPD subunit of the DNA repair/transcription factor TFIIH result in distinct clinical entities, including the cancer-prone xeroderma pigmentosum (XP) and the multisystem disorder trichothiodystrophy (TTD), which share only cutaneous photosensitivity. Gene-expression profiles of primary dermal fibroblasts revealed overexpression of matrix metalloproteinase 1 (MMP-1), the gene encoding the metalloproteinase that degrades the interstitial collagens of the extracellular matrix (ECM), in TTD patients mutated in XPD compared with their healthy parents. The defect is observed in TTD and not in XP and is specific for fibroblasts, which are the main producers of dermal ECM. MMP-1 transcriptional up-regulation in TTD is caused by an erroneous signaling mediated by retinoic acid receptors on the MMP-1 promoter and leads to hypersecretion of active MMP-1 enzyme and degradation of collagen type I in the ECM of cell/tissue systems and TTD patient skin. In agreement with the well-known role of ECM in eliciting signaling events controlling cell behavior and tissue homeostasis, ECM alterations in TTD were shown to impact on the migration and wound-healing properties of patient dermal fibroblasts. The presence of a specific inhibitor of MMP activity was sufficient to restore normal cell migration, thus providing a potential approach for therapeutic strategies. This study highlights the relevance of ECM anomalies in TTD pathogenesis and in the phenotypic differences between TTD and XP.
The extracellular matrix (ECM) is a complex structural network that surrounds and supports cells within connective tissues. It generally includes three major types of macromolecules: fibrous proteins (collagens and elastic fibers), glycoproteins (such as fibronectins and laminins), and glycosaminoglycans/proteoglycans. The type, amount and composition of the ECM give tissues their unique physical and biological properties (1). Notably, the ECM is not only a structural scaffold but also exhibits important functional roles in controlling key cellular events (e.g., adhesion, migration, proliferation, differentiation, and survival) involved in tissue homeostasis, development, inflammation, and tissue repair (2, 3). Thus, inherited or acquired structural alterations as well as metabolic disturbances of the ECM may cause cellular and tissue changes leading to the onset and progression of specific disorders, such as those affecting the connective tissues (osteogenesis imperfecta, Ehlers-Danlos syndrome, and Marfan’s syndrome) or demanding ECM degradation (tumor invasion and metastasis) (4, 5). Recently, a reduced synthesis of the ECM component collagen type VI (COL6) was shown in confluent fibroblast cultures from patients with trichothiodystrophy [TTD; Online Mendelian Inheritance in Man (OMIM) #601675] (6), an autosomal recessive disorder characterized by symptoms affecting several tissues and organs (7). The most relevant features of TTD are hair abnormalities, ichthyotic skin, physical and mental retardation, neurodegeneration, and signs of premature aging. Moreover, about 50% of patients show cutaneous photosensitivity that is associated with an altered cellular response to UV light caused by a defect in nucleotide excision repair (NER), the DNA repair pathway that removes a wide spectrum of DNA lesions, including UV-induced damage. The genes responsible for the photosensitive form of TTD encode the XPB, XPD, and p8/TTDA subunits of the general transcription factor IIH (TFIIH). All of the mutations associated with TTD result in reduced cellular levels (8, 9) and impaired functioning of TFIIH in NER and basal transcription (10 –12). Furthermore, they may interfere with the role of TFIIH in transcription regulation (13 –16).
The engagement of TFIIH in distinct cellular processes has provided a rationale for explaining the variety of clinical entities resulting from mutations in the XPD and XPB genes, which are also found in patients with the cancer-prone disorder xeroderma pigmentosum (XP). Notably, premalignant skin lesions and cutaneous tumors have never been reported in TTD. It was suggested that pathological features of XP are associated with mutations that mainly affect the DNA repair activity of TFIIH, whereas those typical of TTD also impair transcription (12, 17 –19).
To assess whether mutated TFIIH in TTD might alter the transcription of ECM-related genes (thus affecting the strength and architecture of the ECM and contributing to the wide clinical spectrum of TTD), we compared gene-expression profiles of primary fibroblasts from TTD patients and their parents. In TTD, we found a cell type-specific transcription overexpression of the matrix metalloproteinase 1 (MMP-1), a gene encoding a zinc-containing endopeptidase that plays a key role in physiological and pathological tissue remodeling (20 –23). The MMP-1 transcription up-regulation results in high secretion of the active MMP-1 enzyme and subsequent degradation of collagen type I (COL1) in cell/tissue cultures, as well as in patient skin. We investigated the molecular mechanisms responsible for this transcription deregulation, which is specific for TTD and not XP mutations. Moreover, we demonstrated that the abundant secretion of active MMP-1 is responsible for altered migration features of TTD cells, a defect that can be recovered by using a specific MMP inhibitor.
A high-throughput gene-expression profiling by RealTime ready Custom Panels was applied to RNA pools obtained by mixing equal amounts of RNA from primary dermal fibroblasts of either 11 TTD patients (TTD pool) or 11 healthy donors (normal pool). Ten TTD parents were included in the normal donor sample to minimize gene-expression variations resulting from different genetic backgrounds. The analyzed TTD patients are representative of different types and combinations of mutated XPD alleles (TTD/XP-D), including the changes arg112his and arg722trp, which represent the most frequent alterations observed in TTD (Table S1). The TTD and normal pools were investigated for the expression of 83 genes encoding proteins related to the ECM and involved in cell-to-cell and cell-to-matrix interactions (Table S2). Using a ±2.5-fold change cut-off value, we identified eight ECM genes deregulated in TTD, three being up- and five down-regulated (Fig. S1). Intriguingly, five of the deregulated genes encode proteins belonging to the family of MMPs, which are proteolytic enzymes responsible for ECM turnover and degradation. The remaining three genes encode a member of the EGF-containing fibulin-like family of extracellular matrix glycoproteins (EFEMP1), the elastin (ELN) protein, and the structural component laminin β3 (LAMB3), respectively.
The expression of these eight ECM-related genes was further investigated by individually analyzing several fibroblast strains from healthy donors and TTD or XP patients mutated in XPD (Fig. S2). MMP-1 is the only gene transcriptionally deregulated in all TTD/XP-D cell strains, irrespective of whether they are homozygotes, hemizygotes, or compound heterozygotes, with a 2- to 14-fold expression increment compared with normal fibroblasts ( Fig. 1A and Fig. S3A). Strikingly, this finding holds true for TTD cells mutated in either the XPB or TTDA gene as well (Fig. S3A), indicating that all of the genetic alterations responsible for repair-defective TTD patients are associated with a substantial and specific increase in MMP-1 expression. In contrast, normal levels of MMP-1 mRNA are found in fibroblasts from XP/XP-D patients and a rare case showing the clinical symptoms of XP in combination with those of the NER-deficient disease Cockayne syndrome (XP/CS).
Increased transcription and intracellular protein levels of MMP-1 in primary dermal fibroblasts but not in keratinocytes of TTD/XP-D patients. (A) Real-time quantification of MMP-1 mRNA levels in fibroblasts from five healthy (C3PV, C377RM, C5BO, and TTD2PV mother and father -TTD2m and TTD2f-: white bars), eight TTD/XP-D (TTD2PV, TTD8PV, TTD11PV, TTD12PV, TTD20PV, TTD22PV, TTD23PV, and TTD24PV: black bars), and four XP/XP-D (XP15PV, XP16PV, XP17PV, and XP1BR: gray bars) donors. The suffix PV in the strain code is omitted for sake of brevity. MMP-1 expression was normalized to the expression of the GAPDH housekeeping gene and then to the corresponding value in C3PV cells. ***P < 0.001; Student’s t test. (B) Immunoblot analysis of the MMP-1 proenzyme (pro–MMP-1) in total lysates of fibroblasts from five normal (C3PV, mother and father of TTD12PV or TTD2PV patients: white bars), eight TTD/XP-D (black bars), and two XP/XP-D (gray bars) donors (strain codes in A). Protein levels were normalized to the amount of the loading control γ-tubulin. (C) Real-time quantification of MMP-1 mRNA levels in the normal K490 and K1609 and the TTD24PV keratinocytes. MMP-1 expression was normalized to the GAPDH expression and then to the corresponding value in K490 keratinocytes. (D) Immunoblot analysis of pro–MMP-1 in proliferating (day 0) and differentiating (2 and 4 d in calcium-containing media) normal K490 and TTD24PV keratinocytes. Protein levels were quantified and normalized to the amount of γ-tubulin. The values reported in all of the panels represent the mean of at least three independent experiments. Bars indicate the SE.
Immunoblot analysis of whole-cell extracts revealed that TTD fibroblasts contain higher amounts of the 52- and 57-kDa MMP-1 proenzyme compared with fibroblasts from normal donors, healthy TTD parents, and XP/XP-D patients ( Fig. 1B ). This finding attests that the MMP-1 transcription up-regulation specifically detected in TTD primary dermal fibroblasts results in abnormally high levels of the MMP-1 protein.
To clarify whether the observed MMP-1 overexpression also affects epidermal cells, we analyzed primary keratinocytes actively proliferating in vitro or induced to differentiate by increasing calcium levels, which mimics the naturally occurring calcium gradient found in the human epidermis (24). We observed that TTD24PV proliferating as well as differentiating keratinocytes contain an MMP-1 transcript and protein levels similar to those observed in the corresponding normal keratinocytes ( Fig. 1 C and D ), thus demonstrating that MMP-1 overexpression in TTD is cell-type–specific.
The phorbol myristate acetate (PMA) stimulates the expression and production of MMP-1 in human fibroblasts through the binding of the c-Jun factor to the AP-1 site of the MMP-1 promoter (25, 26). Conversely, all-trans- or 9-cis-retinoic acid (RA) isomers reduce MMP-1 expression via the formation of a large repressor complex containing c-Jun that binds directly to the AP-1 site and apparently tethers the retinoid receptors (RARs and RXRs) to the complex (27). To elucidate the mechanistic basis of MMP-1 transcription up-regulation in TTD, we analyzed the MMP-1 expression in normal and TTD fibroblasts treated with PMA and RA either singly or in combination ( Fig. 2A and Fig. S3B). In normal fibroblasts, the steady-state level of MMP-1 progressively increases and peaks at 12-h PMA treatment and then slowly decreases. In contrast, the addition of RA to the culture medium does not alter the MMP-1 basal expression but slightly lowers the PMA-induced MMP-1 expression. Also in TTD fibroblasts the MMP-1 expression drastically increases during PMA treatment and is unaffected by RA. Interestingly, the PMA-induced expression of MMP-1 is further increased (rather than slightly reduced) by the presence of RA in the combined treatment. Overall, these findings indicate that in TTD cells the RA-activated signaling pathway erroneously impacts on the transcriptional machinery acting on the MMP-1 promoter. Because it is well known that XPD mutations prevent TFIIH-dependent transcriptional transactivation by RAR (13), we investigated the recruitment at MMP-1 promoter of key components of the transcriptional machinery, namely the hypophosporylated RNA polymerase II (RNApol IIA), the TFIIH complex, and the transcription factors c-Jun and RARβ. The analysis was performed by ChIP assays in basal conditions and during the PMA-induced transcription up-regulation of MMP-1. Both in normal and TTD fibroblasts, the highest occupancy of RNApol IIA and TFIIH (visualized by the presence of its XPB subunit) is observed on the transcription start site of the MMP-1 promoter, whereas c-Jun and RARβ are localized mainly on the AP-1 binding site at position −77/−72 ( Fig. 2 B and C ). The PMA treatment promotes the recruitment of RNApol IIA, c-Jun, and TFIIH at the MMP-1 promoter (compare the level at 4- and 8-h treatment with that of untreated cells in Fig. 2C ), with a twofold higher efficiency in TTD than in normal fibroblasts, thus paralleling the respective increment of MMP-1 expression. Conversely, the occupancy of RARβ in normal fibroblasts is reduced following transcription activation, according to the previously described repressive role of RARs on MMP-1 transcription (27). Of relevant note, the basal level of RARβ in TTD fibroblasts is reduced compared with normal as well as XP/XP-D fibroblasts (Fig. S3C). Overall, these results demonstrate that the increased expression of MMP-1 in TTD fibroblasts is associated with an enhanced recruitment of c-Jun and a reduced occupancy of the RARβ repressor complex on the AP-1 binding site of the MMP-1 promoter.
Reduced binding of retinoid receptors on MMP-1 promoter in TTD/XP-D primary dermal fibroblasts. (A) Real-time quantification of MMP-1 mRNAs in C3PV and TTD23PV fibroblasts in basal condition (0) and after 4, 8, 12, 16, and 24 h of PMA, RA, or PMA+RA treatments. The MMP-1 expression was normalized to the GAPDH expression and then to the basal level in C3PV fibroblasts. (B) Schematic organization of the genomic fragment included between nucleotides −1,000 and +250 of MMP-1 gene (GenBank accession no. NG_011740.1). The arrow indicates the transcription start site. The white, gray, and black boxes indicate AP1-, Ets1-, and potential TIE-binding sites, respectively. Horizontal bars show the position of the fragments amplified by ChIP assays (amplicon locations: a, −1,000/−887; b, −361/−188; c, −121/+55; and d, −5/+86). (C) RNApol IIA, TFIIH (XPB), c-Jun, and RARβ occupancy on MMP-1 promoter in basal condition (0) and after 4 or 8 h of PMA treatment in C3PV and TTD23PV fibroblasts. The occupancy identified by ChIP analysis is expressed as percentage of the input after subtraction of the corresponding background value (IgG antibody). Values in A and C are the mean of three independent experiments. Statistically significant differences between corresponding TTD23PV and C3PV values are indicated (*P < 0.01, **P < 0.005, ***P < 0.001; Student’s t test). Bars indicate the SE.
The MMP-1 proenzyme is secreted by the cells in its latent form and requires extracellular activation by proteolytic cleavage (21). Thus, we searched for the presence of the proteolytically cleaved forms of MMP-1 (46–42 kDa) in the extracellular environment of TTD and normal fibroblasts. Immunoprecipitation experiments with anti–MMP-1 antibodies revealed a progressive accumulation of the proenzyme as well as the 46- and 42-kDa isoforms in culture media conditioned by TTD fibroblasts ( Fig. 3A , Left). These proteins are not observed in the culture media of normal fibroblasts even at 72 h, when they are clearly detectable in TTD culture media ( Fig. 3A , Right). Compared with normal, TTD fibroblasts secrete twice the amount of MMP-1 protein, as evaluated by using the MMP-1 Human Biotrak ELISA System that directly measures the amount of soluble MMP-1 ( Fig. 3B ). We then investigated whether the secreted forms of MMP-1 are active and capable of triggering the cleavage of interstitial collagen triple helix, mainly the fibrillar COL1. In culture media conditioned by TTD fibroblasts but not normal fibroblasts, zymogram analysis revealed the presence of enzymatic activities with the molecular weights of the active recombinant MMP-1 isoforms that are also capable of degrading COL1 ( Fig. 3C ). Accordingly, a strong reduction of the α1 subunit of COL1 (COL1A1) was detected uniquely in TTD cells ( Fig. 3D ) despite the nearly normal COL1A1 expression (Table S2). Conversely, the levels of collagen types V (COL5) and VII (COL7), which are not MMP-1 substrates, are within the range detected in normal and XP/XP-D cells ( Fig. 3D ).
TTD fibroblasts secrete abnormally high level of active MMP-1 that in turn degrades COL1. (A) MMP-1 immunoprecipitation analysis in media conditioned for 24, 48, and 72 h by TTD23PV fibroblasts (Left), or in media conditioned for 72 h by C3PV, TTD8PV, TTD20PV, TTD23PV, or TTD24PV fibroblasts (Right). The suffix PV in the strain code is omitted for sake of brevity. Arrows indicate the MMP-1 proenzyme (57 kDa) and its proteolitically cleaved forms (46 and 42 kDa). Asterisks indicate the IgG heavy chains. (B) Total amount of active MMP-1 in media conditioned for 72 h by C3PV, TTD2PV, TTD12PV, or TTD23PV cells measured by MMP-1 Human Biotrak ELISA System. (C) Zymogram analysis of the active recombinant MMP-1 protein (rMMP1) and media conditioned for 72 h by normal (C3PV and B119) or TTD/XP-D (TTD2PV, TTD12PV, TTD23PV, and TTD24PV) cells. The SDS/PAGE gel was embedded with COL1 and stained with Coomassie R-250. The MMP-1 enzymatic activity is attested by the COL1 degradation revealed by the white bands (arrows). (D) Immunoblot analysis of COL1A1, COL5A1, and COL7A1 in total cell lysates of normal (C3PV), TTD/XP-D (TTD2PV, TTD8PV, TTD11PV, TTD12PV, TTD22PV, TTD23PV, and TTD24PV), and XP/XP-D (XP16PV and XP17PV) fibroblasts. γ-Tubulin is the loading control.
The effect of MMP-1 overexpression on COL1 distribution was evaluated in more sophisticated in vitro systems, such as tridimensional (3D) tissue-like matrices stained with Masson trichrome goldner to disclose collagen fibers. Collagen is undetectable in the area around 53–60% of TTD fibroblasts versus 9–19% of normal fibroblasts in 4-wk-old 3D COL1 lattices (representative results in Fig. 4A ). This analysis was extended to whole reconstructed skin equivalents in which TTD or normal primary keratinocytes were grown on 3D COL1 lattices containing either normal or TTD fibroblasts. The presence of TTD fibroblasts is associated with areas devoid of collagen, independently of the cocultivated keratinocytes ( Fig. 4B ). Taken together, these results demonstrate that TTD dermal fibroblasts secrete high levels of active MMP-1, which are responsible for COL1 impoverishment in the ECM of monolayer or 3D fibroblast cultures as well as TTD skin equivalents.
COL1 reduction in fibroblast-populated collagen lattices and skin equivalents generated with TTD fibroblasts. (A) Masson trichrome staining of collagen lattices prepared with fibroblasts from the normal donor C3PV (Normal F) and the patient TTD24PV (TTD F). Collagen fibers are light blue. (Scale bar, 20 μm.) (B) Masson trichrome staining of skin equivalents in which TTD or normal primary keratinocytes (TTD K, Normal K) were grown on collagen lattices containing either normal or TTD fibroblasts (Normal F, TTD F). Keratinocytes are black; collagen fibers are light blue. Red arrowheads point to the area of collagen depletion in 3D lattices. (Scale bar, 50 μm.)
To clarify whether the high secretion of active MMP-1 and subsequent degradation of COL1 in the ECM might alter the migratory features of TTD fibroblasts, we compared the wound closure in monolayers of several TTD fibroblasts with that of normal and XP fibroblasts by wound-healing assays. In the time-frame of 36 h, TTD fibroblasts are able to close the wound, whereas normal and XP cells do not (Fig. S4A). Over this period, all TTD fibroblasts show a higher migration rate compared with normal and XP cells (Fig. S4B).
We then analyzed the effect of normal and TTD cells on the wound-healing response of XP17PV fibroblasts up to 36 h after wound generation ( Fig. 5 , Top). In this setting, TTD but not normal fibroblasts are able to speed up the migration rate of XP17PV cells, demonstrating that TTD cells secrete molecules capable of influencing cell migration. Next, we performed wound-healing assays in the presence of galardin (GAL), a synthetic inhibitor of MMPs (28, 29). The addition of GAL to the culture medium is sufficient to reduce the migration speed of TTD23PV fibroblasts, whereas it has no effect on the migration of XP cells ( Fig. 5 , Middle). Moreover, GAL is also able to slow down the migration of XP17PV fibroblasts surrounded by TTD23PV cells ( Fig. 5 , Bottom), thus demonstrating that the metalloprotease activity secreted by TTD fibroblasts influences XP cell migration. By showing a causal relationship between migration ability and MMP activity, these findings demonstrate that the MMP-dependent degradation of the ECM is the main event responsible for the high migration rate of TTD cells.
Increased MMP-1 levels account for the high migration rate of TTD fibroblasts. Wound-healing assays of XP17PV or TTD23PV single cultures or XP17PV fibroblasts surrounded by C3PV or TTD23PV fibroblasts [XP17PV (C3PV) and XP17PV (TTD23PV), respectively]. The assays were performed in media with or without the MMP inhibitor GAL. The mean numbers of migratory cells in the central region of the wound at different time points after wound generation, are indicated. The values represent the mean of three independent experiments. Bars indicate the SE. On the right, representative images at 36 h after wound formation. (Scale bar, 200 μm.)
Finally, to clarify whether MMP-1 transcription deregulation in TTD also affects the ECM in vivo, we analyzed the presence and distribution of COL1A1 in the skin of the 18-y-old patient TTD8PV showing a moderately affected phenotype. Because premature aging is a relevant feature of TTD (7, 30), the patient skin was compared with that of four normal donors of different ages, namely a 19-, 21-, 29-, and 58-y-old ( Fig. 6 ). In parallel, we analyzed the distribution of COL5, which is not a substrate of MMP-1. Immunohistochemistry staining with anti-COL1A1 antibody on formalin-fixed paraffin-embedded skin sections clearly show the presence of COL1A1 in the dermis of normal donors and its reduction in aged skin as a consequence of MMP-dependent dermal collagen fragmentation (31). In TTD skin, despite a normal distribution and a strong signal of COL5A1, a very weak COL1A1 staining is found.
Reduced levels of COL1 in TTD skin tissue. Immunohistochemistry staining with anti-COL1A1 and anti-COL5A1 antibodies on formalin-fixed paraffin-embedded skin sections of an 18-y-old TTD patient (TTD8PV) and four normal donors of different ages (i.e., 19-, 21-, 29-, and 58-y-old). A hair follicle is observed in the TT8PV skin section stained with anti-COL1A1. Positive immunostaining appears brown, and nuclei are counterstained blue with hematoxylin. Both collagens are present throughout the dermis and along the basement membrane. (Scale bar, 20 μm.)
Overall, these results demonstrate that MMP-1 overexpression in TTD dermal fibroblasts leads to high secretion levels of active MMP-1 that, in turn, degrades interstitial collagen and affects the ECM not only in cell-culture systems but also in patient skin.
The ECM is a dynamic structure that interacts with cells and generates signals controlling cell behavior and tissue homeostasis. The diversity and functional importance of the ECM is illustrated by the occurrence of numerous genetic and acquired connective tissue disorders in which bone, cartilage, skin, muscle, brain, eye, and cardiovascular systems are differentially affected (5). In the present study, we demonstrate that TTD-specific mutations affecting the general TFIIH are responsible for the overexpression of MMP-1 gene in primary dermal fibroblasts and, in turn, for the loss of the interstitial component COL1 in the patient skin dermis. The dissection of the underlying mechanistic defect led to the identification of a reduced occupancy of the RARs on the AP-1 binding site of the MMP-1 promoter ( Fig. 2C and Fig. S3C). The RAR interaction with the c-Jun component of the AP-1 complex was previously demonstrated to repress MMP-1 transcription (27). Thus, the reduced amount of the RAR repressor on the MMP-1 promoter offers a clue to explain the MMP-1 overexpression in TTD cells that are otherwise characterized by a malfunctioning basal transcription factor. It was previously shown that TFIIH phosphorylates specific nuclear hormone receptors, including RARs, via its cdk7 kinase subunit, thus leading to transcriptional transactivation of hormone-responsive genes (13). This raises the possibility that TTD mutations, which are known to impair the stability of the TFIIH complex, might differentially affect the transcriptional activity of RARs and their binding affinities for other transcription factors, cofactors, and chromatin remodelers. Interestingly, the MMP-1 transcription deregulation occurs specifically in TTD dermal fibroblasts and not in keratinocytes, which have different embryonic origins (mesodermal and ectodermal, respectively). This finding suggests the involvement of additional cell/tissue-specific transcription factors likely interacting with RARs, which may contribute to the increased levels of MMP-1 mRNA in TTD fibroblasts. Furthermore, the lack of correlation among MMP-1 overexpression, reduced level of TFIIH, and mutation sites, which are known to differentially affect RAR phosphorylation (13), suggests that other TFIIH-dependent signaling events must contribute to the reduced levels of RARs on the MMP-1 promoter.
MMP-1 transcriptional deregulation results in the increased secretion of active MMP-1 isoforms and consequent degradation of COL1 ( Fig. 3 C and D ), the major structural component of interstitial ECM that provides mechanical strength to the skin. The in vivo relevance of this event was demonstrated by showing that COL1 degradation occurs not only in monolayer and 3D fibroblast cultures ( Fig. 4A ) but also in skin equivalent, a system mimicking the in vivo cell environment ( Fig. 4B ). Even more remarkably, COL1-depletion was observed in TTD patient skin as well ( Fig. 6 ). Because the correct balance between deposition and degradation of ECM collagens is altered in TTD, the outcome is a high migration rate of TTD fibroblasts, as demonstrated by wound-healing assays (Fig. S4). During the process of in vivo wound healing, normal dermal fibroblasts, which are the primary collagen-producing cells in the dermis, contribute to the repair by filling the wound with abundant provisional ECM, a key component of the granulation tissue. The persistent overexpression of MMP-1 in TTD dermal fibroblasts would result in excessive COL1 proteolysis within granulation tissue and likely in the scar tissue, thus protracting the period of healing. Indeed, a prolonged time of wound repair in TTD patients is anecdotally reported by several parents. Remarkably, we showed that the altered migration features of TTD cells are recovered by inhibiting the metalloprotease activity ( Fig. 5 ), suggesting a new potential target for therapeutic approaches.
Besides a key structural role in skin, COL1 is the main protein-building block in bone, and is important in other connective tissues, such as tendons and ligaments. Accordingly, reduced amounts of normal COL1 because of a mutated COL1A1 or COL1A2 allele, are typically found in the mildest form of osteogenesis imperfecta (OI type I), which is characterized by fragile bones and reduced bone density often associated with slight spinal curvature, loose joints, muscle weakness, lax ligaments, easy bruising, blue sclerae, and early loss of hearing (32). Notably, skeletal abnormalities, including kyphosis, osteopenia and osteosclerosis, abnormalities in joints, muscle tone and deep tendon reflex, and sensorineural hearing loss are reported in several TTD patients just as are bone fragility, kyphosis and reduced mineral density of vertebrae in the TTD mouse model (30, 33, 34).
ECM alterations might also explain other TTD-related features. Recent molecular epidemiological studies on pregnancy and neonatal abnormalities in the mothers of patients with XPD mutations showed that 94% of TTD pregnancies (16 of 17) have preterm delivery, pre-eclampsia, hemolysis, elevated liver enzymes and low platelets syndrome, prematurity, and low birth weight. In contrast, none of 17 XP pregnancies had these complications (35, 36). It is worthwhile considering that the human placenta, whose function is to facilitate nutrient exchanges between the fetus and the mother, contains abundant ECM components and well-preserved endogenous growth factors. Because the placenta consists of both fetal and maternal tissue, we are tempted to speculate that the pregnancy complications in the mothers of TTD patients could be partially ascribed to the fetal ECM alterations affecting the structure and functionality of the placenta. This observation is supported by the finding that gestational abnormalities were found only when the pregnancy resulted in TTD-affected individuals and not in their unaffected siblings (37). Overall, our previous and present studies suggest how the ECM alterations might contribute to TTD etiopathogenesis.
It is yet to be ascertained whether ECM alterations may also contribute to other features that distinguish TTD from the XP phenotype, such as cancer proneness. It is well known that ECM remodeling plays a role in cancer initiation, progression, and invasion (3, 38). Excess ECM production or reduced ECM turnover are prominent in cancer development and various collagens, including collagen type I, II, III, V, and IX, show increased deposition during tumor formation. Increased MMP1 levels were shown in dermal fibroblasts, dermal equivalents, and skin sections from XP patients mutated in the XPC gene, leading to the suggestion that MMP-1 overespression could be a worsening actor toward exacerbated premature skin aging and tumor susceptibility in these patients. Nevertheless, no evidence of ECM alterations was reported in the analyzed XP-C cells and tissues (39). Further studies are needed to ascertain whether the reduced levels of COL1 observed in the present study together with the previously reported lowered amount of COL6A1 secreted by TTD dermal fibroblasts (6) might create a microenvironment not conducive for cancer cells to survive and grow.
Patient fibroblast strains used in this study are listed in Table S1. Normal primary fibroblasts were from 4 genetically unrelated healthy donors (C3PV, C5BO, C377RM, and B119) and 10 phenotypically normal TTD parents (mother and father of the patients TTD7PV, TTD8PV, TTD11PV, TTD2PV/TTD3PV, and TTD12PV/TTD15PV). Experiments on keratinocytes were performed on the normal cell strains K490 and K1690 and the only available TTD keratinocyte strain (TTD24PV). Fibroblast-populated collagen lattices and reconstructed epidermis were prepared as detailed in SI Materials and Methods. Gene-expression analysis was performed by RealTime ready Custom Panels (Roche) and quantitative real-time RT-PCR. Primer sequences are listed in Table S3. Immunoprecipitation, immunohistochemistry, and immunoblot experiments were performed according to standard protocols. ChIP assays were previously described (6). Zymogram analysis were obtained in nonreducing conditions on 10% (wt/vol) SDS/PAGE gels embedded with 0.8 mg/mL rat tail type I collagen (Sigma). Wound-healing assays were performed using a culture-insert (Ibidi) according to the manufacturer’s instructions. Full details are provided in SI Materials and Methods.
This work was supported by Associazione Italiana per la Ricerca sul Cancro Grant IG 13537 (to M.S.); Consiglio Nazionale delle Ricerche-Centre National de la Recherche Scientifique collaboration program (D.O. and E.C.); Ligue Contre le Cancer Grant 1FI11327OWGG (to J.-M.E. and E.C.); and the Consiglio Nazionale delle Ricerche/Ministro dell'Istruzione, dell'Università e della Ricerca “Progetto d’interesse invecchiamento” and the European Molecular Biology Organization (European Molecular Biology Organization-Short Term Fellowship ASTF 15-2008) (to D.O.).
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
1. Frantz C, Stewart KM, Weaver VM. The extracellular matrix at a glance. J Cell Sci. 2010; 123 (Pt 24):4195–4200. [PMC free article] [PubMed] [Google Scholar]
2. Brown NH. Extracellular matrix in development: Insights from mechanisms conserved between invertebrates and vertebrates. Cold Spring Harb Perspect Biol. 2011; 3 (12):3. [PMC free article] [PubMed] [Google Scholar]
3. Lu P, Takai K, Weaver VM, Werb Z. Extracellular matrix degradation and remodeling in development and disease. Cold Spring Harb Perspect Biol. 2011; 3 (12):3. [PMC free article] [PubMed] [Google Scholar]
4. Järveläinen H, Sainio A, Koulu M, Wight TN, Penttinen R. Extracellular matrix molecules: Potential targets in pharmacotherapy. Pharmacol Rev. 2009; 61 (2):198–223. [PMC free article] [PubMed] [Google Scholar]
5. Bateman JF, Boot-Handford RP, Lamandé SR. Genetic diseases of connective tissues: Cellular and extracellular effects of ECM mutations. Nat Rev Genet. 2009; 10 (3):173–183. [PubMed] [Google Scholar]
6. Orioli D, et al. XPD mutations in trichothiodystrophy hamper collagen VI expression and reveal a role of TFIIH in transcription derepression. Hum Mol Genet. 2013; 22 (6):1061–1073. [PubMed] [Google Scholar]
7. Stefanini M, Botta E, Lanzafame M, Orioli D. Trichothiodystrophy: From basic mechanisms to clinical implications. DNA Repair (Amst) 2010; 9 (1):2–10. [PubMed] [Google Scholar]
8. Vermeulen W, et al. Sublimiting concentration of TFIIH transcription/DNA repair factor causes TTD-A trichothiodystrophy disorder. Nat Genet. 2000; 26 (3):307–313. [PubMed] [Google Scholar]
9. Botta E, et al. Reduced level of the repair/transcription factor TFIIH in trichothiodystrophy. Hum Mol Genet. 2002; 11 (23):2919–2928. [PubMed] [Google Scholar]
10. Dubaele S, et al. Basal transcription defect discriminates between xeroderma pigmentosum and trichothiodystrophy in XPD patients. Mol Cell. 2003; 11 (6):1635–1646. [PubMed] [Google Scholar]
11. Nonnekens J, et al. Mutations in TFIIH causing trichothiodystrophy are responsible for defects in ribosomal RNA production and processing. Hum Mol Genet. 2013; 22 (14):2881–2893. [PubMed] [Google Scholar]
12. Compe E, Egly JM. TFIIH: When transcription met DNA repair. Nat Rev Mol Cell Biol. 2012; 13 (6):343–354. [PubMed] [Google Scholar]
13. Keriel A, Stary A, Sarasin A, Rochette-Egly C, Egly JM. XPD mutations prevent TFIIH-dependent transactivation by nuclear receptors and phosphorylation of RARalpha. Cell. 2002; 109 (1):125–135. [PubMed] [Google Scholar]
14. Drané P, Compe E, Catez P, Chymkowitch P, Egly JM. Selective regulation of vitamin D receptor-responsive genes by TFIIH. Mol Cell. 2004; 16 (2):187–197. [PubMed] [Google Scholar]
15. Compe E, et al. Neurological defects in trichothiodystrophy reveal a coactivator function of TFIIH. Nat Neurosci. 2007; 10 (11):1414–1422. [PubMed] [Google Scholar]
16. Zhou X, et al. Abnormal XPD-induced nuclear receptor transactivation in DNA repair disorders: Trichothiodystrophy and xeroderma pigmentosum. Eur J Hum Genet. 2013; 21 (8):831–837. [PMC free article] [PubMed] [Google Scholar]
17. Fan L, et al. XPD helicase structures and activities: Insights into the cancer and aging phenotypes from XPD mutations. Cell. 2008; 133 (5):789–800. [PMC free article] [PubMed] [Google Scholar]
18. Fuss JO, Tainer JA. XPB and XPD helicases in TFIIH orchestrate DNA duplex opening and damage verification to coordinate repair with transcription and cell cycle via CAK kinase. DNA Repair (Amst) 2011; 10 (7):697–713. [PMC free article] [PubMed] [Google Scholar]
19. Lehmann AR. The xeroderma pigmentosum group D (XPD) gene: One gene, two functions, three diseases. Genes Dev. 2001; 15 (1):15–23. [PubMed] [Google Scholar]
20. Wilhelm SM, et al. Human fibroblast collagenase: Glycosylation and tissue-specific levels of enzyme synthesis. Proc Natl Acad Sci USA. 1986; 83 (11):3756–3760. [PMC free article] [PubMed] [Google Scholar]
21. Brinckerhoff CE, Matrisian LM. Matrix metalloproteinases: A tail of a frog that became a prince. Nat Rev Mol Cell Biol. 2002; 3 (3):207–214. [PubMed] [Google Scholar]
22. Pardo A, Selman M. MMP-1: The elder of the family. Int J Biochem Cell Biol. 2005; 37 (2):283–288. [PubMed] [Google Scholar]
23. Bertini I, et al. Structural basis for matrix metalloproteinase 1-catalyzed collagenolysis. J Am Chem Soc. 2012; 134 (4):2100–2110. [PMC free article] [PubMed] [Google Scholar]
24. Stanley JR, Yuspa SH. Specific epidermal protein markers are modulated during calcium-induced terminal differentiation. J Cell Biol. 1983; 96 (6):1809–1814. [PMC free article] [PubMed] [Google Scholar]
25. Hall M-C, et al. The comparative role of activator protein 1 and Smad factors in the regulation of Timp-1 and MMP-1 gene expression by transforming growth factor-beta 1. J Biol Chem. 2003; 278 (12):10304–10313. [PubMed] [Google Scholar]
26. Sampieri CL, et al. Activation of p38 and JNK MAPK pathways abrogates requirement for new protein synthesis for phorbol ester mediated induction of select MMP and TIMP genes. Matrix Biol. 2008; 27 (2):128–138. [PubMed] [Google Scholar]
27. Schroen DJ, Brinckerhoff CE. Inhibition of rabbit collagenase (matrix metalloproteinase-1; MMP-1) transcription by retinoid receptors: Evidence for binding of RARs/RXRs to the -77 AP-1 site through interactions with c-Jun. J Cell Physiol. 1996; 169 (2):320–332. [PubMed] [Google Scholar]
28. Mirastschijski U, Haaksma CJ, Tomasek JJ, Ågren MS. Matrix metalloproteinase inhibitor GM 6001 attenuates keratinocyte migration, contraction and myofibroblast formation in skin wounds. Exp Cell Res. 2004; 299 (2):465–475. [PubMed] [Google Scholar]
29. Green KA, et al. Profibrinolytic effects of metalloproteinases during skin wound healing in the absence of plasminogen. J Invest Dermatol. 2008; 128 (8):2092–2101. [PubMed] [Google Scholar]
30. Stefanini M, Ruggieri M. Trichothiodystrophy. In: Ruggieri M, Pascual-Castroviejo I, Di Rocco C, editors. Neurocutaneous Disorders, Phakomatoses and Hamartoneoplastic Syndromes. Springer; Vienna: 2008. pp. 821–845. [Google Scholar]
31. Fisher GJ, Varani J, Voorhees JJ. Looking older: Fibroblast collapse and therapeutic implications. Arch Dermatol. 2008; 144 (5):666–672. [PMC free article] [PubMed] [Google Scholar]
32. Byers PH, Pyott SM. Recessively inherited forms of osteogenesis imperfecta. Annu Rev Genet. 2012; 46 :475–497. [PubMed] [Google Scholar]
33. Faghri S, Tamura D, Kraemer KH, Digiovanna JJ. Trichothiodystrophy: A systematic review of 112 published cases characterises a wide spectrum of clinical manifestations. J Med Genet. 2008; 45 (10):609–621. [PMC free article] [PubMed] [Google Scholar]
34. Diderich KEM, et al. Bone fragility and decline in stem cells in prematurely aging DNA repair deficient trichothiodystrophy mice. Age (Dordr) 2012; 34 (4):845–861. [PMC free article] [PubMed] [Google Scholar]
35. Tamura D, et al. Effect of mutations in XPD(ERCC2) on pregnancy and prenatal development in mothers of patients with trichothiodystrophy or xeroderma pigmentosum. Eur J Hum Genet. 2012; 20 (12):1308–1310. [PMC free article] [PubMed] [Google Scholar]
36. Moslehi R, et al. Phenotype-specific adverse effects of XPD mutations on human prenatal development implicate impairment of TFIIH-mediated functions in placenta. Eur J Hum Genet. 2012; 20 (6):626–631. [PMC free article] [PubMed] [Google Scholar]
37. Moslehi R, et al. Adverse effects of trichothiodystrophy DNA repair and transcription gene disorder on human fetal development. Clin Genet. 2010; 77 (4):365–373. [PMC free article] [PubMed] [Google Scholar]
38. Cox TR, Erler JT. Remodeling and homeostasis of the extracellular matrix: Implications for fibrotic diseases and cancer. Dis Model Mech. 2011; 4 (2):165–178. [PMC free article] [PubMed] [Google Scholar]
39. Fréchet M, et al. Overexpression of matrix metalloproteinase 1 in dermal fibroblasts from DNA repair-deficient/cancer-prone xeroderma pigmentosum group C patients. Oncogene. 2008; 27 (39):5223–5232. [PubMed] [Google Scholar]
Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
