Abstract

To elucidate the molecular mechanisms of impaired elastic fiber formation in recessive cutis laxa, we have investigated two disease-causing missense substitutions in fibulin-5, C217R and S227P. Pulse-chase immunoprecipitation experiments indicated that S227P mutant fibulin-5 was synthesized and secreted by skin fibroblasts at a reduced rate when compared with the wild-type protein. Both mutants failed to be incorporated into elastic fibers by transfected rat lung fibroblasts. Purified recombinant fibulin-5 with either mutation showed reduced affinity for tropoelastin in solid-phase binding assays. Furthermore, S227P mutant fibulin-5 also showed impaired association with fibrillin-1 microfibrils. The same mutation triggered an endoplasmic reticulum (ER) stress response, as indicated by the strong co-localization of this mutant protein with folding chaperones in the ER, including calreticulin, immunoglobulin-binding protein and protein disulfide isomerase, and by increased rates of apoptosis in patient fibroblasts. Histological analysis of skin sections from a cutis laxa patient with a homozygous S227P mutation showed a lack of fibulin-5 in the extracellular matrix and a concomitant disorganization of dermal elastic fibers. By electron microscopy, elastic fibers in the skin of this patient showed a failure of elastin globules to fuse into a continuous elastic fiber core. We conclude that recessive cutis laxa mutations in fibulin-5 result in misfolding, decreased secretion and a reduced interaction with elastin and fibrillin-1 leading to impaired elastic fiber development. These findings support the hypothesis that fibulin-5 is necessary for elastic fiber formation by facilitating the deposition of elastin onto a microfibrillar scaffold via direct molecular interactions.

INTRODUCTION

Fibulin-5 is an extracellular matrix (ECM) protein consisting of six calcium-binding epidermal growth factor (cbEGF)-like modules and a C-terminal fibulin domain (1,2). Several lines of evidence show that this protein is essential for elastic fiber formation. Fibulin-5 null mice show marked disorganization of elastic fibers, causing skin laxity, airspace enlargement in the lungs and arterial tortuosity (3,4). A similar disease phenotype was observed in patients with recessive cutis laxa caused by homozygous missense mutations in the fibulin-5 gene (FBLN5) (5,6). A cutis laxa patient with a heterozygous, partial, tandem duplication of fibulin-5 has also been reported (7). Finally, heterozygous missense mutations in FBLN5 have been associated with age-related macular degeneration (8) and acquired cutis laxa (9), suggesting a contribution of these variants to complex diseases associated with elastic fiber abnormalities.

The ability of fibulin-5 to bind integrin receptors on the cell surface (3) and key components of elastic fibers, including elastin (4), lysyl oxidase-like-1 (10) and fibrillin-1 (11), suggest a role for fibulin-5 in cell-directed assembly of elastic fibers. However, the exact role of fibulin-5 in this complex hierarchical process (12,13) has yet to be elucidated. In this study, we investigate the impact of FBLN5 missense mutations associated with cutis laxa on fibulin-5 synthesis and elastic fiber formation. Shared functional characteristics of these mutations provide insight into the molecular mechanisms of cutis laxa and related systemic lesions. The characterization of specific molecular deficits associated with these mutations also improves our understanding of the role of fibulin-5 in elastic fiber formation.

RESULTS

To investigate the effect of cutis laxa-causing mutations on fibulin-5 synthesis and secretion, we conducted pulse-chase experiments using skin fibroblasts from a patient with homozygous mutation S227P, located in the fourth cbEGF-like domain of fibulin-5 (6). Following a 15 min pulse with 35S-cysteine and a chase for up to 60 min, newly synthesized fibulin-5 could be seen to be rapidly secreted by wild-type fibroblasts (Fig. 1). In contrast, mutant fibroblasts showed a significant reduction in the amount of fibulin-5 within the cell lysate immediately following labeling, suggesting a reduced rate of synthesis (Fig. 1B). Moreover, secretion of fibulin-5 by the cutis laxa fibroblasts also appeared to be impaired, as indicated by a reduced rate of elimination of the protein from cell lysate (Fig. 1B) and a decreased rate of accumulation in the media (Fig. 1C).

Figure 1.

Impaired synthesis and secretion of fibulin-5 in cutis laxa. Skin fibroblasts from a patient with a homozygous mutation in fibulin-5 (S227P) and from a wild-type control (WT) were pulse-labeled for 15 min using 35S-cysteine and then incubated for 0, 15, 30 and 60 min in unlabeled media before immunoprecipitating fibulin-5 from cell lysates and conditioned media. (A) Phosphorimaging of immunoprecipitated fibulin-5 following SDS–PAGE. (B) Quantitative analysis of fibulin-5 in triplicate lysate samples. Mutant samples contained significantly less fibulin-5 than wild-type at 0 min (P<0.01, t-test) and at 15 min (P<0.05). (C) Quantitative analysis of fibulin-5 in triplicate medium samples showed a significantly reduced amount of fibulin-5 in mutant cultures at 60 min (P<0.01, t-test).

Figure 1.

Impaired synthesis and secretion of fibulin-5 in cutis laxa. Skin fibroblasts from a patient with a homozygous mutation in fibulin-5 (S227P) and from a wild-type control (WT) were pulse-labeled for 15 min using 35S-cysteine and then incubated for 0, 15, 30 and 60 min in unlabeled media before immunoprecipitating fibulin-5 from cell lysates and conditioned media. (A) Phosphorimaging of immunoprecipitated fibulin-5 following SDS–PAGE. (B) Quantitative analysis of fibulin-5 in triplicate lysate samples. Mutant samples contained significantly less fibulin-5 than wild-type at 0 min (P<0.01, t-test) and at 15 min (P<0.05). (C) Quantitative analysis of fibulin-5 in triplicate medium samples showed a significantly reduced amount of fibulin-5 in mutant cultures at 60 min (P<0.01, t-test).

Because the pulse-chase experiments demonstrated that secretion of some mutant fibulin-5 occurred, we investigated the ability of fibulin-5 mutants to participate in elastic fiber formation in transfected rat fetal lung fibroblasts (RFL-6 cells). In addition to S227P, a recently described fibulin-5 mutation, C217R (5), was also evaluated. In RFL-6 cells transfected with wild-type fibulin-5, tropoelastin and fibulin-5 localized to different intracellular compartments (Fig. 2). Consistent with our previous findings (9), these data suggest that the intracellular pools of these two proteins are spatially separated. Although mutations C217R and S227P did not alter this specific subcellular localization, the intracellular compartments containing mutant S227P were enlarged when compared with wild-type (Fig. 2), consistent with the accumulation of this mutant in the secretory pathway. In the ECM, wild-type fibulin-5 co-localized with tropoelastin in elastic fibers; however, mutant C217R showed no ECM deposition and S227P fibulin-5 formed extracellular globular aggregates distinct from the elastic fibers (Fig. 2).

Figure 2.

Defective association of mutant fibulin-5 with elastic fibers. RFL-6 cells were transfected to express wild-type (WT) or mutant (C217R, S227P) fibulin-5 with a C-terminal Myc-tag. Anti-Myc mouse monoclonal and anti-elastin rabbit polyclonal primary antibodies were used to investigate co-localization of the fibulin-5 (red) and (tropo)elastin (green) in permeabilized (intracellular) and non-permeabilized (extracellular) cultures. Both wild-type and mutant fibulin-5 were localized to a different subcellular compartment from tropoelastin. Wild-type fibulin-5 showed strong co-localization with elastin in fibrillar structures of the ECM. In contrast, mutant fibulin-5 showed a complete lack of deposition into the ECM (C217R) or formed globular extracellular deposits separate from elastic fibers (S227P). Magnification bars: 50 µm.

Figure 2.

Defective association of mutant fibulin-5 with elastic fibers. RFL-6 cells were transfected to express wild-type (WT) or mutant (C217R, S227P) fibulin-5 with a C-terminal Myc-tag. Anti-Myc mouse monoclonal and anti-elastin rabbit polyclonal primary antibodies were used to investigate co-localization of the fibulin-5 (red) and (tropo)elastin (green) in permeabilized (intracellular) and non-permeabilized (extracellular) cultures. Both wild-type and mutant fibulin-5 were localized to a different subcellular compartment from tropoelastin. Wild-type fibulin-5 showed strong co-localization with elastin in fibrillar structures of the ECM. In contrast, mutant fibulin-5 showed a complete lack of deposition into the ECM (C217R) or formed globular extracellular deposits separate from elastic fibers (S227P). Magnification bars: 50 µm.

To investigate the cause of impaired association of fibulin-5 with elastic fibers, we tested whether the cutis laxa mutations disrupt the ability of fibulin-5 to directly bind tropoelastin. Previous studies have demonstrated that fibulin-5 specifically binds tropoelastin in solid-phase binding assays (4) and that the affinity of fibulin-5 for tropoelastin can be altered by a missense variant, G202R, in fibulin-5 (9). Therefore, we conducted binding assays with purified recombinant tropoelastin in the solid phase and purified recombinant fibulin-5 in the solution phase. Both S227P and C217R mutant fibulin-5 showed significantly reduced affinity for tropoelastin (Fig. 3) compared with wild-type fibulin-5, thus explaining the impaired ability of these mutants to participate in elastic fiber formation in transfected RFL-6 cells.

Figure 3.

Reduced binding of mutant fibulin-5 to tropoelastin. Solid-phase binding assay using immobilized tropoelastin and purified recombinant wild-type (WT) or mutant (C217R, S227P) fibulin-5 in the solution phase. Wells coated with BSA only were used as negative controls. Significantly reduced binding of tropoelastin by mutant fibulin-5 compared with wild-type was observed at concentrations of 30–50 nm (t-test, P<0.01).

Figure 3.

Reduced binding of mutant fibulin-5 to tropoelastin. Solid-phase binding assay using immobilized tropoelastin and purified recombinant wild-type (WT) or mutant (C217R, S227P) fibulin-5 in the solution phase. Wells coated with BSA only were used as negative controls. Significantly reduced binding of tropoelastin by mutant fibulin-5 compared with wild-type was observed at concentrations of 30–50 nm (t-test, P<0.01).

We also investigated the impact of the S227P fibulin-5 mutation on the association of fibulin-5 with microfibrils by dual staining of wild-type and patient skin fibroblasts for fibulin-5 and fibrillin-1 (Fig. 4). These two proteins showed overlapping localization to fibrillar structures of the ECM in wild-type cells. Conversely, fibrillin-1 microfibrils made by the mutant cells demonstrated greatly diminished staining for fibulin-5 (Fig. 4).

Figure 4.

Mutant fibulin-5 fails to interact with fibrillin-1 microfibrils. Fibulin-5 (green) and fibrillin-1 (red) showed overlapping distribution in ECM fibrils produced by wild-type cells (WT, arrowheads). In contrast, fibrillin-1 microfibrils made by S227P mutant fibroblasts showed diminished fibulin-5 staining (S227P, arrowheads). Magnification bars: 50 µm.

Figure 4.

Mutant fibulin-5 fails to interact with fibrillin-1 microfibrils. Fibulin-5 (green) and fibrillin-1 (red) showed overlapping distribution in ECM fibrils produced by wild-type cells (WT, arrowheads). In contrast, fibrillin-1 microfibrils made by S227P mutant fibroblasts showed diminished fibulin-5 staining (S227P, arrowheads). Magnification bars: 50 µm.

Impaired secretion and defective intermolecular interactions are frequently caused by misfolding of proteins. Misfolding also triggers an endoplasmic reticulum (ER) stress response, which is associated with the recruitment of folding chaperones to mutant proteins (14). We therefore studied ER stress in skin fibroblasts from the cutis laxa patient with the S227P mutation by immunostaining for fibulin-5 and for the folding chaperones, calreticulin (CRT), immunoglobulin binding protein (BiP) and protein disulfide isomerase (PDI) (Fig. 5). No significant co-localization of any of the chaperones was observed with wild-type fibulin-5. In contrast, the mutant fibroblasts showed strong intracellular punctate co-localization of fibulin-5 with all three chaperones, indicating that substitution of a proline at the 227 position results in a significant disruption of fibulin-5 folding (Fig. 5). We also investigated the abundance of CRT, BiP and PDI in the wild-type and mutant cells and found no significant difference (data not shown). In addition to a staining pattern consistent with ER, nuclei were also strongly positive for BiP in both wild-type and mutant cells. This is consistent with earlier reports of both ER and nuclear localization of BiP (15,16).

Figure 5.

ER stress response to mutant fibulin-5 in fibroblasts. Dermal fibroblasts from a cutis laxa patient with homozygous mutation S227P in fibulin-5 (S227P) and a normal control (WT) were stained for fibulin-5 (FBLN5, green) and for the chaperones (red), CRT, BiP and PDI. Mutant fibulin-5 showed strong co-localization with all three chaperones. In contrast, no significant co-localization of fibulin-5 with any of the chaperones was found in wild-type cells. Magnification bars: 50 µm.

Figure 5.

ER stress response to mutant fibulin-5 in fibroblasts. Dermal fibroblasts from a cutis laxa patient with homozygous mutation S227P in fibulin-5 (S227P) and a normal control (WT) were stained for fibulin-5 (FBLN5, green) and for the chaperones (red), CRT, BiP and PDI. Mutant fibulin-5 showed strong co-localization with all three chaperones. In contrast, no significant co-localization of fibulin-5 with any of the chaperones was found in wild-type cells. Magnification bars: 50 µm.

Cellular stress responses triggered by misfolded proteins often result in the activation of cell death pathways (14). Thus, we investigated whether the ER stress we observed in S227P mutant fibroblasts also results in an increased rate of apoptosis. Indeed, cells with the S227P mutation showed a significantly greater number of apoptotic cells compared with wild-type fibroblasts, as measured by terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end-labeling (TUNEL) staining (Fig. 6). Thus, both chaperone localization and apoptosis assays confirmed that mutation S227P disrupted fibulin-5 folding.

Figure 6.

Increased apoptosis in S227P fibroblasts. TUNEL staining (green) was used to detect apoptotic cells. Nuclei were counterstained with propidium iodide (red). Wild-type cells showed a very low level of apoptosis with most microscopic fields lacking TUNEL-positive nuclei (A). Cutis laxa fibroblasts with the S227P mutation often showed apoptotic cells (B). The rate of apoptosis (C), based on counting of positive and negative nuclei, was significantly increased in cutis laxa fibroblasts (S227P) compared with wild-type (WT). Fisher's exact test was used for statistical analysis. Magnification bars: 50 µm.

Figure 6.

Increased apoptosis in S227P fibroblasts. TUNEL staining (green) was used to detect apoptotic cells. Nuclei were counterstained with propidium iodide (red). Wild-type cells showed a very low level of apoptosis with most microscopic fields lacking TUNEL-positive nuclei (A). Cutis laxa fibroblasts with the S227P mutation often showed apoptotic cells (B). The rate of apoptosis (C), based on counting of positive and negative nuclei, was significantly increased in cutis laxa fibroblasts (S227P) compared with wild-type (WT). Fisher's exact test was used for statistical analysis. Magnification bars: 50 µm.

Taken together, our in vitro studies showed that cutis laxa mutations impair the ability of fibulin-5 to interact with tropoelastin and fibrillin-1 microfibrils and to participate in elastic fiber formation. To investigate the impact of fibulin-5 mutation on elastic fiber development in vivo, we performed Hart's elastin staining on a skin biopsy specimen from a cutis laxa patient with the homozygous S227P mutation and an age-matched control (Fig. 7). Control skin sections showed normal elastic fiber architecture with robust horizontal elastic fibers in the deep dermis and fine, vertical, tree-like fibers in the superficial, papillary dermis (Fig. 7A and B). In contrast, skin from the patient showed elastin assembled into clumped aggregates (Fig. 7C and D), similar to earlier observations of the same patient's skin using a different (Verhoeff van Giesson) elastin stain (6,17). In addition, we observed an almost complete lack of the vertical elastic fiber system in the papillary dermis of the patient's skin (Fig. 7C and D). To further study the effect of the S227P mutation on fibulin-5 secretion and assembly in vivo, we immunostained the skin biopsy sections for fibulin-5. In wild-type skin, robust staining of fibulin-5 was seen in both the vertical (Fig. 7E and F) and horizontal elastic fibers (Fig. 7I and J). Pre-immune serum staining of adjacent sections showed only low level of background staining (Fig. 7M and N), indicating that our fibulin-5 antibody was specific. The patient's skin showed no detectable fibulin-5 in the dermal ECM (Fig. 7G, H, K and L). Because Hart's staining for elastin showed that elastic fibers were present in the skin of the patient, the lack of fibulin-5 staining could not be caused by an absence of elastic fibers but rather was caused by the inability of the S227P mutant fibulin-5 to interact with the elastic fibers.

Figure 7.

The impact of fibulin-5 mutation S227P on elastic fiber formation in vivo. (A and B) Hart's elastin stain of skin from a healthy donor (WT, 9-month-old) showed robust horizontal elastic fibers (white arrowhead) in the deep dermis and abundant fine vertical tree-like structures in the papillary dermis (blue arrowheads). (C and D) In contrast, skin from a cutis laxa patient (S227P, 6-month-old) had abnormally clumped elastin deposits in the deep dermis (white arrowhead) and greatly diminished elastic fibers in the papillary dermis (blue arrowheads). (E and F) Both papillary and (I and J) deep dermal elastic fibers were positive for fibulin-5 immunostaining (green). Negative control WT skin sections stained with pre-immune serum showed low background in the deep dermis (M and N). Cutis laxa skin showed complete lack of staining in the papillary dermis (G and H) and only background staining in the deep dermis (K and L). Nuclei in images (E–L) were counterstained with propidium iodide (red). Magnification bars: (A and C), 20 µm; (B, D, E, G, I and K), 50 µm; (F, H, J and L), 100 µm.

Figure 7.

The impact of fibulin-5 mutation S227P on elastic fiber formation in vivo. (A and B) Hart's elastin stain of skin from a healthy donor (WT, 9-month-old) showed robust horizontal elastic fibers (white arrowhead) in the deep dermis and abundant fine vertical tree-like structures in the papillary dermis (blue arrowheads). (C and D) In contrast, skin from a cutis laxa patient (S227P, 6-month-old) had abnormally clumped elastin deposits in the deep dermis (white arrowhead) and greatly diminished elastic fibers in the papillary dermis (blue arrowheads). (E and F) Both papillary and (I and J) deep dermal elastic fibers were positive for fibulin-5 immunostaining (green). Negative control WT skin sections stained with pre-immune serum showed low background in the deep dermis (M and N). Cutis laxa skin showed complete lack of staining in the papillary dermis (G and H) and only background staining in the deep dermis (K and L). Nuclei in images (E–L) were counterstained with propidium iodide (red). Magnification bars: (A and C), 20 µm; (B, D, E, G, I and K), 50 µm; (F, H, J and L), 100 µm.

To gain further insight into the role of fibulin-5 in elastic fiber formation, skin from a 7-month-old cutis laxa patient with the S227P mutation and an age-matched unaffected child was subjected to electron microscopic analysis (Fig. 8). The distribution and organization of collagen bundles and dermal cells appeared normal in both individuals. In contrast, elastic fibers in the patient consisted of adjoining globules of elastin (Fig. 8B and D) rather than a continuous elastin core as seen in healthy individuals (Fig. 8A and C). Moreover, the elastin globules lacked a peripheral mantle of microfibrils and, instead, distinct microfibril bundles could be seen interspersed among the elastin deposits (Fig. 8D). Electron microscopic analysis of skin from a different, affected member of the same cutis laxa family was reported earlier with similar findings (18). These observations support the notion that fibulin-5 is important for the assembly of continuous elastin polymer and for promoting the interaction of microfibrils and elastin.

Figure 8.

Ultrastructure of dermal elastic fibers. Electron microscopic analysis of the skin from an unaffected 7-month-old child (A and C) and a cutis laxa patient with the S227P fibulin-5 mutation at the same age (B and D). In normal skin, elastin forms a continuous, solid core of the elastic fiber (A). In contrast, elastic fibers in the dermis of the cutis laxa patient consist of globules of elastin (E) with adjacent bundles of microfibrils (MF) (B). At higher magnification, the integration of elastin (E) and MF in normal elastic fibers can be seen (C), whereas the two components are distinct in the patient (D). Magnification bars: (A and B), 1.0 µm; (C and D), 0.2 µm. COL, collagen.

Figure 8.

Ultrastructure of dermal elastic fibers. Electron microscopic analysis of the skin from an unaffected 7-month-old child (A and C) and a cutis laxa patient with the S227P fibulin-5 mutation at the same age (B and D). In normal skin, elastin forms a continuous, solid core of the elastic fiber (A). In contrast, elastic fibers in the dermis of the cutis laxa patient consist of globules of elastin (E) with adjacent bundles of microfibrils (MF) (B). At higher magnification, the integration of elastin (E) and MF in normal elastic fibers can be seen (C), whereas the two components are distinct in the patient (D). Magnification bars: (A and B), 1.0 µm; (C and D), 0.2 µm. COL, collagen.

DISCUSSION

These results provide new insights into the molecular mechanisms of recessive cutis laxa. We demonstrate that a fibulin-5 mutation, S227P, results in reduced synthesis and secretion of fibulin-5. Furthermore, studies on this and another cutis laxa-related mutation, C217R, in transfected RFL-6 cells showed that these fibulin-5 mutants cannot interact with elastic fibers and therefore are null mutations with respect to their function in elastic fiber formation. This was also confirmed in patient-derived skin fibroblasts and skin tissue sections. Finally, solid-phase binding and immunostaining assays suggest that a reason for impaired association of these mutants with elastic fibers is their reduced binding to tropoelastin and fibrillin-1 microfibrils.

We have shown that mutation S227P caused ER stress associated with the recruitment of folding chaperones and with increased apoptosis of patient-derived cells. These observations raise the possibility that the disease mechanisms in recessive cutis laxa, in addition to the consequences of a lack of functional fibulin-5, may also involve decreased cell survival, especially in cell types characterized by high fibulin-5 expression. Missense mutations in other inherited diseases have also been shown to contribute to the disease phenotype through the combined effect of loss of protein function and misfolding-related toxicity. For example, patients with alpha-1-antitrypsin deficiency develop emphysema because of an impaired protease–antiprotease balance in the lungs (19). In addition, patients with the common Pi Z allele can also have liver cirrhosis and hepatoma as a result of liver damage associated with the accumulation of misfolded mutant protein in hepatocytes (20), the main cell type responsible for the synthesis of this protein.

We have recently reported a heterozygous fibulin-5 G202R missense allele in a patient with acquired cutis laxa in combination with a mutation in the elastin gene (G773D) that in itself could cause dominant cutis laxa (9,21). In our previous study, G202R fibulin-5 showed increased binding to tropoelastin in solid-phase binding assays, it co-localized with tropoelastin in the secretory pathway and was capable of enhancing elastin deposition in a dominant fashion (9). Thus, we concluded that it mitigated the adverse outcomes of the co-existing elastin mutation. In contrast, the present study demonstrates for the first time that missense fibulin-5 mutations in recessive cutis laxa interfere with the secretion and matrix deposition of fibulin-5, leading to impaired elastin polymerization and association with microfibrils. In summary, these investigations highlight the mechanistic differences between acquired and inherited forms of cutis laxa.

Limited functional analysis of missense alleles in fibulin-5 associated with cutis laxa and age-related macular degeneration has recently been published (22). In that study, transfection of mutant constructs into COS7 cells showed no secretion of either C217R or S227P mutants into the media. In contrast, our pulse-chase results from patient-derived fibroblasts showed only a partial deficiency in the synthesis and secretion of the S227P mutant fibulin-5. We conclude that analysis of fibulin-5 mutations is preferable in either patient-derived cell lines or in cells with the capacity of forming elastic fibers, because overexpression of mutant constructs or the absence of binding partners of fibulin-5 may impact the results of functional studies.

In a broader context, functional studies of fibulin-5 mutations also provide insights into the role of this protein in elastic fiber formation, which is known to occur on a scaffold of microfibrils (23). More recently, it was recognized that cells deposit elastic fibers in a highly dynamic and hierarchical fashion, which is initially characterized by the formation of globular elastin aggregates that are later fused into elastin fibrils associated with active cell movements (12,13). The results of the present study show that missense mutations in fibulin-5 disrupt interactions with both elastic fibers and fibrillin-1 microfibrils. Moreover, our electron microscopic observations of a lack of microfibrillar sheaths around elastin deposits in the skin of a patient with a fibulin-5 mutation suggest that fibulin-5 is important for the interaction of elastin with microfibrils. Finally, the presence of multiple globular elastin deposits and the absence of a continuous elastic fiber core in the same patient indicate that fibulin-5 is also necessary for the fusion of these elastin globules into larger scale fibers.

Interestingly, a dissociation of the microfibrils and elastin components of the elastic fiber was also observed in the skin of patients with autosomal dominant cutis laxa caused by mutations in the elastin gene (ELN) (24,25). Taken together with the present study, these reports suggest that disruption of the physical continuity of microfibrils and elastin in the ECM may be a common mechanistic feature of cutis laxa.

MATERIALS AND METHODS

Pulse-chase and immunoprecipitation

Patient fibroblasts were plated onto 60 mm culture dishes (100 000 cells/dish) and grown to confluence for 3 days. Cell cultures were washed using warmed phosphate-buffered saline (PBS) three times, incubated in 1.5 ml pre-warmed Dulbecco's modified Eagle's medium (DMEM) deficient in cysteine at 37°C for 90 min. Then, 20 µl 35S-cysteine (10 mCi/ml, MP Biomedicals, CA, USA) was added to medium and cells were incubated at 37°C for 15 min. After the labeling period, cells were washed three times using PBS and incubated in serum-free DMEM at 37°C. After the initial pulse, and following 15, 30 and 60 min of chase, conditioned medium was collected. After centrifugation to remove cell debris, one-tenth volume of 10× immunoprecipitation buffer (500 mm Tris, pH 7.5, 50 mm EDTA, 5% NP-40) was added to the conditioned medium. Cells were harvested by scraping in PBS, followed by centrifugation. Cells were resuspended in lysis buffer TSD (50 mM Tris, pH 7.5, 1% SDS and 5 mM DTT). Conditioned medium and cell lysates were boiled for 10 min and centrifuged to remove debris. Supernatants were transferred to new tubes for immunoprecipitation and 1.2 ml TNN (50 mM Tris, pH 7.5, 250 mM NaCl, 5 mM EDTA, 0.5% NP-40) was added to cell lysates. Protease inhibitor cocktail (Sigma, St Louis, MO, USA) and 2.5 mg/l phenylmethylsulfonyl fluoride (PMSF, Sigma) were added to the conditioned medium and cell lysates. All samples were pre-cleared using 25 µl protein G-Agarose (Santa Cruz, CA, USA) at 4°C for 1 h. Immunoprecipitation was performed using 5 µl (1:200) of fibulin-5 antibody (Table 1) at 4°C for 1 h. Antigen–antibody complexes were precipitated using 25 µl protein G-Agarose and washed three times with TNN containing protease inhibitor cocktail and PMSF. Immunoprecipitates were solubilized in 2× Laemmli loading buffer by heating at 95°C for 5 min. Samples were electrophoresed in 7.5% SDS–PAGE gels. Gels were fixed for 30 min (20% methanol, 10% acetic acid) and then washed in water for 10 min and dried. Dried gels were exposed to Tritium screen (Amersham, Piscataway, NJ, USA) for 1–4 days. Images were visualized on a Typhoon imaging system (Amersham) and analyzed using the software ImageQuant 5.2.

Table 1.

Antibodies used in this study

Antigen Host (type) Dilution Label Source Cat # 
Fibulin-5 Rabbit (poly) 1:500 – Dr Elaine Davis – 
Fibrillin-1 Mouse (mono) 1:1000 – Chemicon MAB1919 
Rat tropoelastin Goat (poly) 1:1000 – Elastin Products Company RA75 
C-myc Mouse (mono) 1:1000 – Invitrogen R950 
CRT Mouse (mono) 1:1000 – BD Biosciences 612136 
BiP/GRP78 Mouse (mono) 1:1000 – BD Biosciences 610978 
PDI Mouse (mono) 1:1000 – BD Biosciences 610946 
Goat IgG Donkey 1:500 AF 488 Invitrogen A11055 
Mouse IgG Goat 1:500 AF 350 Invitrogen A11045 
Mouse IgG Rabbit 1:500 AF 350 Invitrogen A21062 
Mouse IgG Donkey 1:500 AF 594 Invitrogen A21203 
Mouse IgG Goat 1:500 Fluorescein Sigma F9887 
Rabbit IgG Goat 1:500 AF 350 Invitrogen A21068 
Rabbit IgG Donkey 1:500 AF 488 Invitrogen A21206 
Rabbit IgG Goat 1:500 Fluorescein Sigma F0257 
Antigen Host (type) Dilution Label Source Cat # 
Fibulin-5 Rabbit (poly) 1:500 – Dr Elaine Davis – 
Fibrillin-1 Mouse (mono) 1:1000 – Chemicon MAB1919 
Rat tropoelastin Goat (poly) 1:1000 – Elastin Products Company RA75 
C-myc Mouse (mono) 1:1000 – Invitrogen R950 
CRT Mouse (mono) 1:1000 – BD Biosciences 612136 
BiP/GRP78 Mouse (mono) 1:1000 – BD Biosciences 610978 
PDI Mouse (mono) 1:1000 – BD Biosciences 610946 
Goat IgG Donkey 1:500 AF 488 Invitrogen A11055 
Mouse IgG Goat 1:500 AF 350 Invitrogen A11045 
Mouse IgG Rabbit 1:500 AF 350 Invitrogen A21062 
Mouse IgG Donkey 1:500 AF 594 Invitrogen A21203 
Mouse IgG Goat 1:500 Fluorescein Sigma F9887 
Rabbit IgG Goat 1:500 AF 350 Invitrogen A21068 
Rabbit IgG Donkey 1:500 AF 488 Invitrogen A21206 
Rabbit IgG Goat 1:500 Fluorescein Sigma F0257 

Mono, monoclonal; poly, polyclonal; AF, Alexa Fluor.

Immunostaining

Cells were plated on coverslips in 6-well plates. Confluent cultures were washed with PBS once and were fixed in 4% paraformaldehyde in PBS at 4°C for 30 min. After several washes in PBS, cell cultures were incubated in 3% bovine serum albumin (BSA) in either Tris-buffered saline (TBS, 20 mm Tris/HCl, pH 7.4, and 0.15 M NaCl) for non-permeabilized cells or TBS/0.1% Triton X-100 for permeabilized cells for 1 h. Cell cultures were then incubated with primary antibody (Table 1) for 1 h. After several washes, cell cultures were incubated with secondary antibody (Sigma) (Table 1). Affinity-purified secondary antibodies (Invitrogen, Grand Island, NY, USA) (Table 1) were applied in dual labeling. In some experiments, nuclei were stained with propidium iodide or hoechst 33258 (Sigma). Cell cultures were washed and then mounted with Gel/Mount (Biomeda, Foster City, CA, USA) and visualized with an Axioskop BX60 microscope (Zeiss, Thornwood, NY, USA). Pseudo-colors were applied in co-localization.

Solid-phase binding assay

Human fibulin-5 cDNA was cloned into pcDNA3.1/myc-His (Invitrogen) to provide a C-terminally tagged fibulin-5. Mutations were introduced into expression vectors by inverse-PCR mutagenesis using Pfu Ultra DNA polymerase (Stratagene, La Jolla, CA, USA). Recombinant fibulin-5 was expressed in COS7 cells and purified using Ni-column chromatography. Tropoelastin (100 ng/well) was coated onto 96-well plates in TBS at 4°C for 4–8 h. Non-specific binding was blocked by incubating with 6% BSA in TBS/1% Triton X-100 overnight at 4°C. Fibulin-5 was added at various concentrations and was incubated for 2 h at room temperature (RT). In negative controls, fibulin-5 was replaced with 10% BSA. Anti-c-Myc primary antibody (Invitrogen) and horseradish peroxidase-conjugated secondary antibody were then added in turn. Wells were washed three times with TBS/1% Triton X-100 after each step. Color was developed using tetramethylbenzidine (Sigma) and stopped using 2 M sulfuric acid. Plates were read at a wavelength of 450 nm.

TUNEL assay

Apoptotic cells were stained using ApopTag fluorescein direct in situ apoptosis detection kit (Chemicon, Temecula, CA, USA). Fibroblasts were plated on coverslips in 6-well plates. Confluent cultures were fixed in 1% paraformaldehyde in PBS for 10 min at RT. Following two washes in PBS, cell cultures were fixed and permeabilized in pre-cooled ethanol:acetic acid (2:1) for 5 min at −20°C and washed in PBS three times. For positive controls, cell cultures were incubated in DN buffer (30 mm Tris-HCl, pH 7.2, 4 mM MgCl2, 0.1 mm DTT) for 5 min at RT. Then, 150 µl DNaseI working solution (1000 U/ml) was applied to positive control samples for 10 min at RT. Cell cultures were rinsed with five changes of water for 3 min each change. Samples were subsequently incubated in equilibration buffer for 5 min, and then 120 µl working strength terminal deoxynucleotidyl transferase (TdT) enzyme was applied to each coverslip and incubated at 37°C for 1 h in a humidified chamber. The reaction was stopped by incubation in working strength stop/wash buffer for 10 min. Coverslips were mounted using Antifade (Chemicon) containing 0.5 µg/ml propidium iodide and visualized with an Axioskop BX60 microscope (Zeiss). To calculate the rate of apoptosis, 15 fields, arranged in a matrix of 3×5, of 10× magnification were taken from each slide. Numbers of total cells (propidium iodide positive) and apoptotic cells (fluorescein positive) were counted. Fisher's exact test was used for statistical analysis.

Histology, immunohistochemistry and electron microscopy

Paraffin-embedded skin samples were cut to 5 µm-thick sections, deparaffinized in Histo-Clear (National Diagnostics, Atlanta, GA, USA) and then rehydrated. For Hart's staining, sections were immersed in 0.25% potassium permanganate solution for 5 min, cleaned in 5% oxalic acid and then soaked in working concentration Resorcin–Fuchsin solution (Poly Scientific, Bay Shore, NY, USA) for 4 h. After washing in water, sections were counterstained with tartrazine, dehydrated in ethanol, cleared in Histo-Clear and mounted. For immunocytochemistry, sections were incubated with 0.05 U/ml porcine pancreatic elastase (Sigma) for 30 min. After blocking with 1.5% normal serum for 1 h, sections were incubated with primary antibody at 4°C overnight. After several washes, sections were incubated with fluorescein-conjugated secondary antibody. Nuclei were stained using propidium iodide. Sections were mounted with Gel/Mount and were visualized with an Axioskop BX60 microscope (Zeiss). Images of immunohistochemistry were processed using Adobe Photoshop. Images captured in the blue channel, as background noise or auto-fluorescence, were subtracted from images in the green channel. For electron microscopy, skin biopsies were fixed in glutaraldehyde, stained sequentially en bloc with OsO4 and uranyl acetate, then dehydrated and embedded in Epon (26). Thin sections (60 nm) were cut, placed on formvar-coated grids and counterstained with 7% methanolic uranyl acetate and lead citrate. Sections were viewed using a Tecnai 12 transmission electron microscope at 120 kV and images were digitally captured.

ACKNOWLEDGEMENTS

We thank Dr Sofie De Schepper and Professor Dr Jean-Marie Naeyaert for providing skin biopsy materials and Dr Dieter P. Reinhardt and Dr Louis J. Muglia for critical reading of the manuscript. This study was supported by NIH grant HL073703 (Z.U.) and a National Marfan Foundation grant (E.C.D.). E.C.D. is a Canada Research Chair. B.L.L. is a senior clinical investigator from the Fund for Scientific Research, Flanders.

Conflict of Interest statement. The authors declare no conflict of interest.

REFERENCES

1
Kowal
R.C.
Richardson
J.A.
Miano
J.M.
Olson
E.N.
EVEC, a novel epidermal growth factor-like repeat-containing protein upregulated in embryonic and diseased adult vasculature
Circ. Res.
 , 
1999
, vol. 
84
 (pg. 
1166
-
1176
)
2
Nakamura
T.
Ruiz-Lozano
P.
Lindner
V.
Yabe
D.
Taniwaki
M.
Furukawa
Y.
Kobuke
K.
Tashiro
K.
Lu
Z.
Andon
N.L.
, et al.  . 
DANCE, a novel secreted RGD protein expressed in developing, atherosclerotic, and balloon-injured arteries
J. Biol. Chem.
 , 
1999
, vol. 
274
 (pg. 
22476
-
22483
)
3
Nakamura
T.
Lozano
P.R.
Ikeda
Y.
Iwanaga
Y.
Hinek
A.
Minamisawa
S.
Cheng
C.F.
Kobuke
K.
Dalton
N.
Takada
Y.
, et al.  . 
Fibulin-5/DANCE is essential for elastogenesis in vivo
Nature
 , 
2002
, vol. 
415
 (pg. 
171
-
175
)
4
Yanagisawa
H.
Davis
E.C.
Starcher
B.C.
Ouchi
T.
Yanagisawa
M.
Richardson
J.A.
Olson
E.N.
Fibulin-5 is an elastin-binding protein essential for elastic fibre development in vivo
Nature
 , 
2002
, vol. 
415
 (pg. 
168
-
171
)
5
Hadj-Rabia
S.
Devillers
M.
Jobard
F.
Bodemer
C.
Megarbane
H.
Pauchard
J.Y.
Blanchet-Bardon
A.
Megarbane
A.
Sommer
P.
Fischer
J.
Clinical and molecular study of 15 patients with cutis laxa
4th European Meeting on Elastin
Lyon, France
6
Loeys
B.
Van Maldergem
L.
Mortier
G.
Coucke
P.
Gerniers
S.
Naeyaert
J.M.
De Paepe
A.
Homozygosity for a missense mutation in fibulin-5 (FBLN5) results in a severe form of cutis laxa
Hum. Mol. Genet.
 , 
2002
, vol. 
11
 (pg. 
2113
-
2118
)
7
Markova
D.
Zou
Y.
Ringpfeil
F.
Sasaki
T.
Kostka
G.
Timpl
R.
Uitto
J.
Chu
M.L.
Genetic heterogeneity of cutis laxa: a heterozygous tandem duplication within the fibulin-5 (FBLN5) gene
Am. J. Hum. Genet.
 , 
2003
, vol. 
72
 (pg. 
998
-
1004
)
8
Stone
E.M.
Braun
T.A.
Russell
S.R.
Kuehn
M.H.
Lotery
A.J.
Moore
P.A.
Eastman
C.G.
Casavant
T.L.
Sheffield
V.C.
Missense variations in the fibulin 5 gene and age-related macular degeneration
N. Engl. J. Med.
 , 
2004
, vol. 
351
 (pg. 
346
-
353
)
9
Hu
Q.
Reymond
J.L.
Pinel
N.
Zabot
M.T.
Urban
Z.
Inflammatory destruction of elastic fibers in acquired cutis laxa is associated with missense alleles in the elastin and fibulin-5 genes
J. Invest. Dermatol.
 , 
2006
, vol. 
126
 (pg. 
283
-
290
)
10
Liu
X.
Zhao
Y.
Gao
J.
Pawlyk
B.
Starcher
B.
Spencer
J.A.
Yanagisawa
H.
Zuo
J.
Li
T.
Elastic fiber homeostasis requires lysyl oxidase-like 1 protein
Nat. Genet.
 , 
2004
, vol. 
36
 (pg. 
178
-
182
)
11
Freeman
L.J.
Lomas
A.
Hodson
N.
Sherratt
M.J.
Mellody
K.T.
Weiss
A.S.
Shuttleworth
A.
Kielty
C.M.
Fibulin-5 interacts with fibrillin-1 molecules and microfibrils
Biochem. J.
 , 
2005
, vol. 
388
 (pg. 
1
-
5
)
12
Czirok
A.
Zach
J.
Kozel
B.A.
Mecham
R.P.
Davis
E.C.
Rongish
B.J.
Elastic fiber macro-assembly is a hierarchical, cell motion-mediated process
J. Cell. Physiol.
 , 
2006
, vol. 
207
 (pg. 
97
-
106
)
13
Kozel
B.A.
Rongish
B.J.
Czirok
A.
Zach
J.
Little
C.D.
Davis
E.C.
Knutsen
R.H.
Wagenseil
J.E.
Levy
M.A.
Mecham
R.P.
Elastic fiber formation: a dynamic view of extracellular matrix assembly using timer reporters
J. Cell. Physiol.
 , 
2006
, vol. 
207
 (pg. 
87
-
96
)
14
Rao
R.V.
Bredesen
D.E.
Misfolded proteins, endoplasmic reticulum stress and neurodegeneration
Curr. Opin. Cell. Biol.
 , 
2004
, vol. 
16
 (pg. 
653
-
662
)
15
Matsumoto
A.
Hanawalt
P.C.
Histone H3 and heat shock protein GRP78 are selectively cross-linked to DNA by photoactivated gilvocarcin V in human fibroblasts
Cancer Res.
 , 
2000
, vol. 
60
 (pg. 
3921
-
3926
)
16
Morris
J.A.
Dorner
A.J.
Edwards
C.A.
Hendershot
L.M.
Kaufman
R.J.
Immunoglobulin binding protein (BiP) function is required to protect cells from endoplasmic reticulum stress but is not required for the secretion of selective proteins
J. Biol. Chem.
 , 
1997
, vol. 
272
 (pg. 
4327
-
4334
)
17
de Schepper
S.
Loeys
B.
de Paepe
A.
Lambert
J.
Naeyaert
J.M.
Cutis laxa of the autosomal recessive type in a consanguineous family
Eur. J. Dermatol.
 , 
2003
, vol. 
13
 (pg. 
529
-
533
)
18
Van Maldergem
L.
Vamos
E.
Liebaers
I.
Petit
P.
Vandevelde
G.
Simonis-Blumenfrucht
A.
Bouffioux
R.
Kulakowski
S.
Hanquinet
S.
Van Durme
P.
, et al.  . 
Severe congenital cutis laxa with pulmonary emphysema: a family with three affected sibs
Am. J. Med. Genet.
 , 
1988
, vol. 
31
 (pg. 
455
-
464
)
19
DeMeo
D.L.
Silverman
E.K.
Alpha1-antitrypsin deficiency. 2: genetic aspects of alpha(1)-antitrypsin deficiency: phenotypes and genetic modifiers of emphysema risk
Thorax
 , 
2004
, vol. 
59
 (pg. 
259
-
264
)
20
Rudnick
D.A.
Perlmutter
D.H.
Alpha-1-antitrypsin deficiency: a new paradigm for hepatocellular carcinoma in genetic liver disease
Hepatology
 , 
2005
, vol. 
42
 (pg. 
514
-
521
)
21
Kelleher
C.M.
Silverman
E.K.
Broekelmann
T.
Litonjua
A.A.
Hernandez
M.
Sylvia
J.S.
Stoler
J.
Reilly
J.J.
Chapman
H.A.
Speizer
F.E.
, et al.  . 
A functional mutation in the terminal exon of elastin in severe, early-onset chronic obstructive pulmonary disease
Am. J. Respir. Cell. Mol. Biol.
 , 
2005
, vol. 
33
 (pg. 
355
-
362
)
22
Lotery
A.J.
Baas
D.
Ridley
C.
Jones
R.P.
Klaver
C.C.
Stone
E.
Nakamura
T.
Luff
A.
Griffiths
H.
Wang
T.
, et al.  . 
Reduced secretion of fibulin 5 in age-related macular degeneration and cutis laxa
Hum. Mutat.
 , 
2006
, vol. 
27
 (pg. 
568
-
574
)
23
Rosenbloom
J.
Abrams
W.R.
Mecham
R.
Extracellular matrix 4: the elastic fiber
Faseb J.
 , 
1993
, vol. 
7
 (pg. 
1208
-
1218
)
24
Szabo
Z.
Crepeau
M.W.
Mitchell
A.L.
Stephan
M.J.
Puntel
R.A.
Yin Loke
K.
Kirk
R.C.
Urban
Z.
Aortic aneurysmal disease and cutis laxa caused by defects in the elastin gene
J. Med. Genet.
 , 
2006
, vol. 
43
 (pg. 
255
-
258
)
25
Urban
Z.
Gao
J.
Pope
F.M.
Davis
E.C.
Autosomal dominant cutis laxa with severe lung disease: synthesis and matrix deposition of mutant tropoelastin
J. Invest. Dermatol.
 , 
2005
, vol. 
124
 (pg. 
1193
-
1199
)
26
Davis
E.C.
Smooth muscle cell to elastic lamina connections in developing mouse aorta. Role in aortic medial organization
Lab. Invest.
 , 
1993
, vol. 
68
 (pg. 
89
-
99
)