Results

A detailed description of the proband (III.1, Figure 1a,B) and her daughter (IV.4, Figure 1b,C) was provided in two previous reports (^{Beighton, 1972;Corbett et al, 1994}). The following description of the family is based on our own (F. M. P.) interview and collection of clinical information from the family and has been recently updated. III.1 had early, generalized CL. Her skin laxity and associated facial features including sagging facial skin and ears, hooked nose, everted nostrils, and long philtrum progressively worsened during childhood, necessitating a facial plastic surgery at the age of 6. Despite the surgery, facial features reappeared in adolescence and progressively worsened during adulthood (Figure 1a). Her respiratory symptoms began in childhood and following frequent chest infections, bronchiectasis was confirmed by bronchography at 21 y of age. At 29, an inguinal hernia was diagnosed. By the age of 36 y, emphysema was confirmed (Figure 1c), as was an -1 antitrypsin M/Z genotype. Emphysema in III.1 was progressive and aggravated by heavy intermittent cigarette smoking. Steady deterioration of respiratory function (pre-transplant forced expiratory volume in 1 s (FEV₁)=0.36, 17% predicted) (^{Corbett et al, 1994}) culminated in end-stage respiratory failure with right-heart failure, hypercapnia, hypoxemia, and respiratory acidosis. Consequently, at the age of 51 y, she received a double lung transplant, followed by routine immunosuppression with cyclosporine and azothioprine. She died at the age of 61 y because of renal failure associated with immunosuppressant treatment.

Figure 1.

Clinical characterization of family cutis laxa (CL)-1. (A) Prematurely redundant and sagging facial skin of the proband III.1 at age 40 y. (B) The pedigree of CL-1 indicates autosomal dominant inheritance. Hatched symbols indicate individuals (I.2, II.9, and III.11) who were reportedly affected but were not available for examination. Small diamonds (IV.2 and IV.3) indicate miscarriages; large diamonds denote multiple unaffected individuals within the same generation. (C) Posterioanterior (PA) chest X-ray (viewed in AP position) of III.1 shortly before lung transplantation at the age of 51 y. Hyperexpanded lung fields indicate emphysema. Bilateral basal shadows reveal lung fibrosis. Note the enlarged left pulmonary artery along the left heart border demonstrating pulmonary hypertension. (D) Inelastic skin and generalized CL in individual IV.4 at birth.

Full figure and legend (115K)

After one normal pregnancy and two miscarriages, at the age of 29, a daughter (IV.4) was born to III.1 with obvious congenital CL (Figure 1d). Between 6 and 18 mo of age, systolic heart murmur caused by a small ventricular septal defect was noted (^{Beighton, 1972}). At 23 y, lung function tests indicated signs of emphysema (FEV₁=2.04, 62% predicted). IV.4 was also an -1 antitrypsin M/Z heterozygote and a smoker at the time (^{Corbett et al, 1994}). But, she declined regular clinical evaluation. The mother (II.5) of III.1 had congenital CL and later developed an inguinal hernia, uterine prolapse, dyspnea, and bronchiectasis (^{Beighton, 1972}). In addition, the maternal grandmother (I.2), the maternal uncle (II.9), and a cousin (III.11) of III.1 were allegedly affected by CL but these individuals were not available for examination (Figure 1b).

Elastic staining of skin biopsy sections from III.1 revealed a normal quantity, but abnormal morphology of elastic fibers with metachromasia after hematoxylin–eosin staining (Figure 2a), and fragmented and clumped appearance (Figure 2a,B), especially in the upper dermis (Figure 2b). Electron microscopic evaluation of dermal elastic fibers showed a lack of association of elastin with peripheral microfibrils and aberrant branching and clubbing (Figure 2d). Similar abnormalities were noted on a bronchial biopsy specimen (not shown).

Figure 2.

Histological and electron microscopic evaluation of dermal elastic fibers in patient III.1. (A) Hematoxylin–eosin staining of a skin section shows metachromatic, fragmented, and clumped (arrows) elastic fibers. (B) Elastic van Giesson (EVG) staining of a skin section from III.1 indicates disorganization of elastic fibers in the upper (arrows) and deep (arrowheads) dermis. (C) EVG staining of a control skin section shows normal, candelabra-like organization of elastic fibers in the papillary dermis (arrows). Transmission electron microscopy of dermal elastic fibers (El) in III.1 (D) demonstrates diminished microfibrillar component (arrows) on the periphery of elastic fibers compared with a normal control (E). Collagen bundles (Col) appear normal. Magnification bars are 100 m (A–C) and 2 m (D–E), respectively.

Full figure and legend (256K)

To search for disease-causing alleles CL-1, exon-by-exon direct sequence analysis of genomic amplimers in the elastin gene was completed in patient III.1 (not shown). No mutations were identified. Metabolic labeling and immunoprecipitation of TE were therefore conducted with cultured dermal fibroblasts from III.1 and IV.4 to determine whether any protein abnormality could be detected in these families. An abnormal, 120 kDa protein was detected in both III.1 and IV.4 (Figure 3a) but not in normal control fibroblast samples (not shown). A corresponding decrease of normal TE protein was also observed in III.1 and IV.4 cells (Figure 3a).

Figure 3.

Abnormal tropoelastin (TE) and elastin mRNA in cutis laxa (CL)-1. (A) Skin fibroblasts from affected individuals III.1 and IV.4 were used to assess TE synthesis by metabolic labeling and immunoprecipitation. In addition to normal TE (68 kDa), a prominent 120 kDa protein (arrow) was detected in CL-1 samples only. Note that fibronectin (FN) binds directly to the immunoprecipitation matrix and thus is present without the use of a primary antibody (^{Kuusela et al, 1984}). Molecular weight marker positions are shown in kilodaltons (kDa). (B) Total RNA was isolated from fibroblasts from CL patients III.1 and IV.4 and a normal control (CON) and was analyzed by RNA-blot analysis using human elastin (^{Olson et al, 1995}) and -actin cDNA as probes. In addition to the normal 3.5 kb elastin mRNA (elastin), fibroblasts from patients III.1 and IV.4 also contained a 5 kb elastin mRNA (arrow). Molecular weight marker positions are shown in kilobases (kb). (C) Total mRNA from III.1 and a normal control was analyzed by 3' rapid amplification of cDNA ends (3'-RACE) using an upstream gene-specific primer complementary to exon 30. The mutation-specific 3'-RACE product (arrow) was sequenced to determine the structure of the 3'-end of the mutant mRNA. Molecular weight marker positions are shown in base pairs.

Full figure and legend (75K)

To investigate whether a transcript of the abnormal 120 kDa protein was present in CL-1, total RNA was blotted with an elastin cDNA H-11 probe (^{Olson et al, 1995}), which identified an abnormal transcript of 5 kb (Figure 3b). To study the structure of the mutant elastin mRNA in family CL-1, a 3'-RACE (rapid amplification of complementary DNA ends) experiment was conducted using an exon 30-specific oligonucleotide primer. A mutation-specific 3'-RACE product was recovered from the samples obtained from individuals III.1 and IV.4 (Figure 3c) and was found to contain exons 30, 31, 33, 9, 10, and 452 bp of intron 10 using direct sequence analysis. This product lacked exon 32, which is normally subject to high-frequency alternative splicing.

To characterize the entire mutant open reading frame, a mutation-specific downstream oligonucleotide primer was designed complementary to the junction of exons 33 and 9. Sizes and direct DNA sequences of amplification products between this mutant oligonucleotide and upstream sequences suggested the following structure of the mutant message: exons 1–33, 9–33, 9, 10, and 452 bp of intron 10 (Figure 4a). This structure was further confirmed by RT-PCR experiments specific for each duplicon (supplemental data). Nucleotides 3–65 of intron 10 encoded a 21 amino acid C-terminal missense peptide sequence before ending in a stop codon. The mutant mRNA had a 4014 bp open reading frame and encoded a protein with a predicted molecular weight of 116 kDa, similar to that observed for the mutant TE protein. Alternative splicing patterns were not altered by the tandem duplication (supplemental data).

Figure 4.

A tandem duplication in patients III.1 and IV.4. (A) Domain structure encoded by the normal elastin message is compared with that of the mutant mRNA detected in the cutis laxa (CL)-1 family. TE consists of alternating hydrophobic domains (open) and cross-linking domains (diagonally hatched), flanked by an N-terminal signal peptide and a C-terminal cysteine-containing, charged peptide (horizontally hatched). Exons 23 and 32 (asterisks) are normally subject to high-frequency alternative splicing. In the mutant mRNA, a 452 bp segment of intron 10 is present (shaded) that generates a missense peptide sequence encoded by the first 63 nucleotides. A portion of this unique sequence (underlined) was used to raise a mutant-specific polyclonal rabbit antibody. Note that rather than being a simple tandem duplication, the mutant mRNA is a product of a complex rearrangement, a duplication of the region encoded by exons 9—33; a third copy of exons 9–10 and intron 10 is added to the end of the mRNA. (B) PCR amplification of the breakpoint region from genomic DNA of a normal control (CON) and of CL patients III.1 and IV.4 using a sense primer complementary to intron 32 and an antisense primer complementary to intron 9. Size standard positions are shown in kilobase pairs. Panel C. Sequence analysis of the breakpoint-specific genomic PCR product identified a chimeric intron consisting of 155 bp of intron 33 and 95 bp of intron 8 (double arrows). The locations of exons 33 and 9 (boxes) and of the amplification primers (arrows) are indicated. (D) Southern blot analysis of the duplication breakpoint region. Genomic DNA samples from a normal control (CON) and from CL patient III.1 were digested with restriction enzymes BamHI, EcoRI, and HindIII. DNA fragments were subjected to Southern blot analysis using a breakpoint-specific PCR product (shown in panel B) as a probe. Size standard positions are shown on the left in kilobases. In addition to fragments identical to the control, duplication-specific 6, 5, and 5 kb fragments were detected in BamHI, EcoRI, and HindIII digests, respectively (arrowheads).

Full figure and legend (67K)

To verify the presence of the duplication at the DNA level, the breakpoint region from genomic DNA of patients III.1 and IV.4 was amplified using oligonucleotide primers complementary to introns 32 (sense) and 9 (antisense) (Figure 4b). Direct DNA sequencing of this 507 bp product demonstrated that the duplication junction was 155 bp downstream of the 5'-end of intron 33 and 95 bp upstream of the 3'-end of intron 8 creating a hybrid intron of 250 bp (Figure 4c). The two introns did not share significant homology with each other, and database searching did not indicate any common repetitive DNA sequence in either intron 8 or 33. Both introns lacked topoisomerase I and II cleavage consensus sequences at the duplication junction positions.

To provide further evidence of a chromosomal DNA duplication, genomic DNA from patient III.1 and a normal control was analyzed using Southern blot analysis. A 507 bp genomic PCR product containing the duplication breakpoint was used as a probe. Mutation-specific hybridization signals were found in samples digested with a variety of restriction enzymes (Figure 4d). The clean background obtained with this probe further supports our earlier finding in that introns 33 and 9 are free of common repetitive sequences.

To demonstrate disease specificity of the observed duplication to family CL-1, 136 normal control individuals including ten donors from England were screened for the presence of the breakpoint-specific PCR products and were found to be negative for the duplication.

In retrospect, the presence of a genomic duplication in CL-1 explains why direct sequence analysis of genomic amplimers containing individual exons failed to identify this mutation. In this assay, both the mutant and normal alleles served as amplification templates, generating products that were identical in sequence. Thus, the duplication escaped detection because of a technique based on screening for sequence differences.

The synthesis of the duplicated TE molecule was investigated using immunoprecipitation and western blotting. A monoclonal TE antibody was used to immunoprecipitate TE from cell lysates and conditioned media of confluent cultures of fibroblasts derived from III.1, IV.4, and a normal control. The immunoprecipitates were then analyzed by immunoblotting using a polyclonal anti-human elastin antibody (Figure 5). Cell lysates of CL cells showed the presence of both normal and mutant TE, confirming the results of our earlier metabolic labeling experiments (Figure 3a). No band at the molecular weight of the mutant product was seen in the control cell lysate. In the conditioned media, a small amount of the mutant product was also observed, indicating that some of the aberrant TE protein was secreted.

Figure 5.

Partial secretion of mutant tropoelastin (TE) in dermal fibroblasts. TE was immunoprecipitated from cell lysates and conditioned media from cultured skin fibroblasts. The protein was detected by western blotting using a different anti-TE antibody. The larger, mutant protein was detected in the lysate and media (arrows), indicating that the mutant protein was present in the cell and secreted into the medium. The presence of the mutant protein apparently did not affect the synthesis and secretion of the normal TE. Standard sizes and the calculated size of TE are shown in kilodaltons to the left.

Full figure and legend (30K)

Since the mutant TE contained a unique and abnormal extra coding region, immunofluorescence experiments were conducted to determine whether the mutant TE could be localized in the fibroblast cultures. For these experiments, a polyclonal rabbit antibody was raised against a peptide in the unique C-terminal sequence encoded by intron 10 (Figure 4a). Although the antibody did not recognize either the mutant protein or any other non-specific proteins in immunoblot experiments, both the crude serum and an affinity-purified preparation recognized the protein by indirect immunofluorescent staining of fibroblast cultures. Permeabilized, control fibroblasts showed only weak background staining when incubated with the mutant TE-specific intron 10 antibody (Figure 6a). In contrast, permeabilized fibroblast cultures from the CL patient III.1 showed strong intracellular staining (Figure 6b). The staining pattern was punctate and peri-nuclear consistent with intracellular organelles of the secretory pathway. When cultures were grown to post-confluency and left non-permeabilized, control cells were again negative for the intron 10 antibody (Figure 6c), whereas faint fibrillar staining was seen in CL cultures, indicating the presence of the mutant protein product in the extracellular matrix (ECM) (Figure 6d). An extensive network of filamentous staining was seen with the anti-TE antibody surrounding both control and CL cells, a pattern typical of elastic fibers assembled by cultured fibroblasts (Figure 6e,F).

Figure 6.

Immunofluorescence localization of the mutant tropoelastin (TE) in dermal fibroblasts. Subconfluent cultures control (A) and cutis laxa (CL) patient III.1 (B) fibroblasts were fixed and permeabilized for intracellular localization of the mutant TE protein using the intron 10 antibody. Whereas only background staining was seen in the control cells (A), a prominent, punctate, peri-nuclear staining was seen in the patient fibroblasts (B). Post-confluent fibroblast cultures from control (C) and CL-1 patient III.1 (D) were fixed and stained with the intron 10 antibody (specific for the mutant protein). Controls were negative (C), whereas the extracellular matrix of CL cells was positive for the mutant protein (D). Nuclei of control cells were counterstained to demonstrate the presence of the cell layer because staining with the intron 10 antibody was negative. Post-confluent fibroblast cultures from control (E) and CL-1 patient III.1 (F) were fixed and stained with an anti-TE antibody (recognizing both normal and mutant proteins). The abundance and intensity of elastic fibers were similar in both patient and control cultures. To indicate cell density, nuclei were also stained.

Full figure and legend (183K)

Discussion

This study demonstrates that a partial tandem duplication in ELN causes autosomal dominant CL with severe chronic obstructive pulmonary disease (COPD). The following lines of evidence support the conclusion that this duplication is the disease-causing defect: (1) both affected individuals studied in the family carried the same duplication; (2) the duplication was absent in 136 normal control individuals; (3) the gene defect resulted in a major change in the reading frame and in the synthesis of mutant TE with partial ECM incorporation and partial intracellular retention of the abnormal protein; and (4) tissue samples from CL patients with the ELN duplication contained abnormal elastic fibers as characterized by a lack of association of the amorphous elastin with microfibrillar components of the fiber.

Detailed clinical investigations identified early-onset emphysema in two members of the family, III.1 and IV.4, which, in III.1, culminated in end-stage respiratory failure requiring a bilateral lung transplant. In addition to carrying an autosomal dominant mutation in ELN, both III.1 and IV.4 had two additional risk factors for COPD: smoking and the -1 antitrypsin M/Z genotype. Although smoking is an established risk factor of COPD, the -1 antitrypsin M/Z genotype, which has a frequency of 5% in Northern Europeans, confers only a modest relative risk (1.5) of COPD (^{Dahl et al, 2002}). Moreover, smoking status does not appear to interact with the M/Z genotype to decrease lung function (^{Dahl et al, 2002}). Only a minority of smokers and M/Z heterozygotes develop emphysema, which usually occurs after the fourth decade of life. Thus, the primary cause of early and progressive COPD in family CL-1 is the elastin gene mutation that our studies uncovered, with smoking and reduced -1 antitrypsin levels exacerbating an existing lung disease.

Severe emphysematous lung disease is frequently associated with autosomal recessive forms of cutis laxa (ARCL) and leads to significant mortality in infancy (^{Beighton, 1972}). One form of ARCL associated with infantile lethal lung disease is caused by a homozygous mutation in the FBLN5 gene (^{Loeys et al, 2002}). In contrast, ADCL is generally described as a milder disorder with little systemic pathology (^{Beighton, 1972}). Thus, our studies have identified COPD as a newly recognized feature of ADCL caused by ELN defects, and establish ELN as a candidate gene for the genetic risk of COPD. Furthermore, these findings indicate the importance of lung function testing in the clinical management of ADCL patients and of counseling against smoking or other lifestyle choices that may contribute to the deterioration of lung function.

Previous studies have described small in-frame deletions and insertions in the elastin gene (^{Urban et al, 2000;Urban et al, 2001}), some of which were shown to be unrelated to disease (^{Urban et al, 2001}). Most of these variants were caused by duplication or deletion of repeated peptide segments. Interestingly, neither the length of individual exons (particularly those encoding hydrophobic domains) nor the number of exons constituting the elastin open reading frame was conserved in vertebrate evolution (^{Boyd et al, 1991}). In fact, progressive loss of exons was demonstrated in the evolution of primates (^{Szabo et al, 1999}). Furthermore, multiple exons were shown to be alternatively spliced, resulting in a normal size heterogeneity of TE (^{Parks and Deak, 1990}). Taken together, these observations indicate that in contrast to fibrillar collagens, where the length of the individual procollagen polypeptides is crucial for the assembly of functionally competent collagen fibrils (^{Byers, 2001}), the length of TE monomers is not critical for the deposition of functional elastic fibers. Despite relatively little constraint in TE length, the duplication mutation in ELN reported in this study leads to a severe form of dominant CL, indicating that excessive differences in the size of TE monomers may be detrimental to elastic fiber function.

Analysis of TE synthesis and secretion in fibroblasts by immunoprecipitation experiments indicated that the majority of duplicated mutant protein was associated with the cell layer, with a relatively small amount being secreted into the medium. Immunostaining of cultured CL fibroblasts using a mutant-specific antibody confirmed this finding in that mutant TE was partially retained in an intracellular compartment and partially incorporated into a fibrillar ECM. These results suggest that both inefficient secretion and the binding of mutant TE to the ECM may contribute to the observed scarcity of mutant TE in conditioned media. Notably, secretion and matrix incorporation of normal TE was not impaired, as indicated by our immunoprecipitation and immunostaining experiments, and no quantitative or qualitative defects of elastic fibers were noted in vitro.

All three ADCL mutations described in ELN to date (^{Tassabehji et al, 1998;Zhang et al, 1999}) are caused by single nucleotide deletions in exons 30 or 32. These frameshift mutations, depending on the alternative splicing of exon 32, may result in mRNA with a 3'-terminally extended open reading frame. The duplicated mRNA described in this study shares two features with the mutant mRNA described previously in that; (1) there is a missense peptide sequence located at the C-terminus of the predicted mutant proteins, and (2) none of the predicted mutant proteins contains an intact C-terminal domain encoded by exon 36. Either one of these features, or both together, may contribute to a disruption of normal elastic fiber structure. For example, in vitro results have demonstrated that the C-terminal domain, which contains only 2 cysteines in elastin and a positively charged RKRK motif, is important for an interaction of TE with microfibril-associated glycoprotein, MAGP-1 (^{Brown-Augsburger et al, 1994}). Antibodies raised against the C-terminal domain of TE inhibited elastin deposition (^{Brown-Augsburger et al, 1996}) but deletion of this region permitted association of recombinant TE with ECM (^{Kozel et al, 2003}). Although the effect of C-terminal missense peptides on elastic fiber assembly or stability has not yet been investigated, such hydrophilic sequences may predispose the normally hydrophobic elastin polymer to proteolytic attack. Both -1 antitrypsin M/Z genotype and smoking cause decreased elastase inhibitor levels (^{Gadek et al, 1979;Dahl et al, 2002}), leading to an imbalance of elastase/elastase inhibitor levels known to be critical in the pathogenesis of COPD. Extreme sensitivity to cigarette smoking and to the M/Z genotype supports the hypothesis that incorporation of mutant TE into the elastic fibers in this CL family results in increased susceptibility to degradation by elastases. The progressive nature of CL and COPD in family CL-1 further corroborates this elastase-susceptibility hypothesis.

Histological and electron microscopic analyses of skin and bronchial biopsy specimens from III.1 showed quantitatively normal but qualitatively abnormal elastic fibers. Significantly, electron micrographs indicated a lack of microfibrils around the periphery of the elastic fibers. Microfibrils are critical for connecting elastic fibers to cells (^{Davis, 1993}) and for providing mechanical coupling and key signals for tissue development, integrity, and function (^{Bunton et al, 2001}). Although our studies do not provide a detailed mechanistic explanation for the loss of microfibrillar content that we observed, we hypothesize that incorporation of abnormal TE could result in a dominant disruption of the elastin–microfibrillar interaction, leading to a defect in elastic fiber function through a dominant-negative mechanism. Alternatively, increased proteolytic degradation of elastin as discussed previously may also result in the loss of microfibrillar elements in CL.

Methods

Patient samples and cell lines

Dermal fibroblast cultures were established from punch biopsy specimens obtained from III.1 (CL-1) at the age of 51 y and IV.4 (CL-1) age 24 y as well as a normal control individual age 31 y following informed consent. Family CL-1 was of English ancestry. Fibroblasts were maintained in Dulbecco's modified Eagle medium supplemented with 10% fetal bovine serum unless indicated otherwise. Skin and bronchial biopsy specimens were also taken from III.1 at the same time for histology and electron microscopy. Control DNA samples were isolated from blood samples of normal, unrelated volunteer participants in our study. The racial composition of this normal panel of 136 individuals was as follows: 61 Caucasian (including ten individuals from England to match the ethnicity of family CL-1), 40 Asian, seven Pacific Islander, three African, 24 more than one race, and one unknown. This study was conducted according to the Declaration of Helsinki and was approved by the Human Studies Committees of the University of Hawaii and of the Washington University in St Louis.

DNA isolation, direct sequence analysis, and isolation of the duplication breakpoint

After nuclear isolation, DNA was purified by proteinase K digestion and phenol extraction (^{Herrmann and Frischauf, 1987}). Each exon of ELN with flanking intronic sequences was amplified using PCR as described earlier (^{Urban et al, 1999}). Amplimers were directly sequenced on both strands using the BigDye dye terminator cycle sequencing chemistry (Applied Biosystems, Foster City, California) and a model 310 Genetic Analyzer (Applied Biosystems). A breakpoint-specific PCR product was recovered using 100 ng of genomic DNA in an amplification reaction with a sense primer complementary to intron 32 5'-TGCAGGCAGAAAGTGATGAG-3' and an antisense primer complementary to intron 9 5'-GCCTCAGTCTCCCAAAGCAA-3'. The resulting 507 bp product was then subjected to direct DNA sequence analysis. The same PCR assay was used to screen DNA samples from 136 unrelated, normal control individuals.

Southern and RNA blot analyses

Southern and RNA blot analyses were conducted using standard techniques as previously described (^{Olson et al, 1995;Urban et al, 2000}).

SMART RACE and cloning of the mutant mRNA

Total RNA (1 g) from individuals III.1, IV.4, and a normal control was reverse transcribed using a SMART RACE cDNA Amplification Kit (Clontech, Palo Alto, California). The RT reaction mixture was used as the template for primary PCR with one human TE exon 30-specific primer E30F1 5'-GCCTAGTGGGAGCCGCTGGGCTCGGAG-3' and a universal primer mix. A secondary PCR was performed using another exon 30-specific primer E30F2 5'-GAGTTCCAGGTGTTGGGGGCCTTGGAG-3' and a nested universal primer. PCR products were separated on the 1.2% agarose gel to display the abnormal bands, which were purified with a Gel Extraction kit (Qiagen, Valencia, California) and further cloned using a TOPO TA Cloning kit (Invitrogen, Carlsbad, California) for sequencing.

Metabolic labeling, immunoprecipitation, and western blotting

Fibroblasts from the CL patients or a normal control individual were plated in six-well tissue culture plates at an initial density of 2 10⁵ per well and used at 2 d post-confluency. For autoradiography, each well of cells was metabolically labeled with 50 Ci of [4,5-3H]-L-leucine (1 mCi per mL; ICN Pharmaceuticals, Irvine, California) in 1 mL of leucine-free medium containing 5% dialyzed fetal bovine serum for 4 h. Media were collected and cell layers were washed three times with cold phosphate-buffered saline (PBS) prior to lysis in 1 mL of cold lysis buffer (25 mM Tris-HCl (pH 7.5), 5 mM EDTA (pH 7.5), 250 mM NaCl, 0.1% Triton X-100) containing protease inhibitors. TE was immunoprecipitated from both media and lysate samples using a monoclonal (BA-4) anti-elastin antibody (^{Wrenn et al, 1986}) and Staphylococcus aureus (Pansorbin cells) to collect the immune complexes. Immunoprecipitates were separated by sodium dodecyl-sulfate polyacrylamide gel electrophoresis fixed, treated with EN³HANCE, (New England Nuclear, Boston, Massachusetts) and exposed to X-ray film. For immunoblotting, media on each well of cells were changed to 1.25 mL of fresh, unlabeled media for 24 h. Media and lysates were collected and immunoprecipitated as described above, and then separated by sodium dodecyl-sulfate polyacrylamide gel electrophoresis and transferred to nitrocellulose. A polyclonal antibody raised against recombinant human TE (a gift from Dr Robert Mecham) was used to detect normal and mutant TE proteins on the blot. Donkey anti-rabbit horseradish peroxidase-conjugated F(ab')₂ (Amersham Life Science, Arlington Heights, Illinois) and ECL detection reagents (Pierce, Rockford, Illinois) were used according to the manufacturers' directions.

Mutant-specific antibody and immunofluorescent staining

A polyclonal rabbit antiserum was raised against a C-terminal peptide from the mutant TE allele (MQLSGQRADGRDSPTTFWPR) encoded by an intron 10 sequence in the mutant mRNA (Figure 4). The peptide was chemically synthesized and conjugated to keyhole limpet hemocyanin using an amino group-specific chemistry. Antisera were assayed against the peptide using an ELISA and were further purified using affinity chromatography using immobilized peptide antigen.

For intracellular immunolocalization of TE, confluent cultures of III.1 and control fibroblasts in four-well LabTek Chamber Slides (Nunc no. 177437, Thomas Scientific, Swedesboro, New Jersey) were washed with PBS and fixed with 2% paraformaldehyde in PBS for 30 min. Following several washes in PBS, reactive aldehydes were quenched by incubating the cells in 50 mM NH₄Cl for 30 min. The cell layers were then rinsed in PBS containing 1% bovine serum albumin and 0.1% saponin and further permeabilized for 2 15 min in the saponin permeabilization buffer. The permeabilized cells were then incubated with primary antibody for 1 h at room temperature. All subsequent antibody dilutions and washes were with the permeabilization buffer. After washing 4 5 min, the cells were incubated with goat anti-rabbit fluorescein-conjugated IgG (Cappel, West Chester, Pennsylvania) diluted 1:200 for 1 h. Nuclei were stained with propidium iodide to visualize the cell layer in some experiments. Cell layers were then washed 4 5 min, then rinsed once with PBS, before mounting in 50% glycerol in PBS containing 1 mg per mL p-phenylenediamine, and visualized with an Axioskop microscope (Zeiss, Thornwood, New York). To investigate the formation of extracellular elastic fibers, post-confluent III.1 and control fibroblast cultures were treated as described above, but with the exclusion of saponin. For localization of normal TE in the matrix, the monoclonal BA-4 antibody was used followed by a goat anti-mouse fluorescein-conjugated IgG (Cappel).

Supplementary Material

The following material is available from http://www.blackwellpublishing.com/products/journals/suppmat/JID/JID23758/JID23758.htm

Figure S1

The mutant allele with a tandem duplication is subject to similar alternative splicing as the normal allele

Supplementary data

References

1. Beighton, P: The dominant and recessive forms of cutis laxa. J Med Genet 1972 9: 216–221,  | PubMed | ISI | ChemPort |
2. Boyd, CD, Christiano, AM, Pierce, RA, Stolle, CA, Deak, SB: Mammalian tropoelastin: Multiple domains of the protein define an evolutionarily divergent amino acid sequence. Matrix 1991 11: 235–241,  | PubMed | ISI | ChemPort |
3. Brown-Augsburger, P, Broekelmann, T, Mecham, L, et al: Microfibril-associated glycoprotein binds to the carboxyl-terminal domain of tropoelastin and is a substrate for transglutaminase. J Biol Chem 1994 269: 28443–28449,  | PubMed | ChemPort |
4. Brown-Augsburger, P, Broekelmann, T, Rosenbloom, J, Mecham, RP: Functional domains on elastin and microfibril-associated glycoprotein involved in elastic fibre assembly. Biochem J 1996 318: 149–155,  | PubMed | ChemPort |
5. Bunton, TE, Biery, NJ, Myers, L, Gayraud, B, Ramirez, F, Dietz, HC: Phenotypic alteration of vascular smooth muscle cells precedes elastolysis in a mouse model of Marfan syndrome. Circ Res 2001 88: 37–43,  | PubMed | ISI | ChemPort |
6. Byers, PH: Disorders of collagen biosynthesis and structure. In: Scriver CR, Beaudet AL, Sly WS, Valle D (eds). The Metabolic and Molecular Bases of Disease, 2001; 8th edn. New York: McGraw-Hill, p 5241–5285,
7. Corbett, E, Glaisyer, H, Chan, C, Madden, B, Khaghani, A, Yacoub, M: Congenital cutis laxa with a dominant inheritance and early onset emphysema. Thorax 1994 49: 836–837,  | PubMed | ISI | ChemPort |
8. Dahl, M, Tybjaerg-Hansen, A, Lange, P, Vestbo, J, Nordestgaard, BG: Change in lung function and morbidity from chronic obstructive pulmonary disease in alpha1-antitrypsin MZ heterozygotes: A longitudinal study of the general population. Ann Intern Med 2002 136: 270–279,  | PubMed | ISI | ChemPort |
9. Damkier, A, Brandrup, F, Starklint, H: Cutis laxa: Autosomal dominant inheritance in five generations. Clin Genet 1991 39: 321–329,  | PubMed | ISI | ChemPort |
10. Davis, EC: Smooth muscle cell to elastic lamina connections in developing mouse aorta. Role in aortic medial organization. Lab Invest 1993 68: 89–99,  | PubMed | ISI | ChemPort |
11. Gadek, JE, Fells, GA, Crystal, RG: Cigarette smoking induces functional antiprotease deficiency in the lower respiratory tract of humans. Science 1979 206: 1315–1316,  | PubMed | ISI | ChemPort |
12. Hayden, JG, Talner, NS, Klaus, SN: Cutis laxa associated with pulmonary artery stenosis. J Pediatr 1968 72: 506–509,  | PubMed | ISI | ChemPort |
13. Herrmann, BG, Frischauf, AM: Isolation of genomic DNA. Methods Enzymol 1987 152: 180–183,  | PubMed | ISI | ChemPort |
14. Kozel, BA, Wachi, H, Davis, EC, Mecham, RP: Domains in tropoelastin that mediate elastin deposition in vitro and in vivo. J Biol Chem 2003 278: 18491–18498, 10.1074/jbc.M212715200 | Article | PubMed | ISI | ChemPort |
15. Kuusela, P, Vartio, T, Vuento, M, Myhre, EB: Binding sites for streptococci and staphylococci in fibronectin. Infect Immun 1984 45: 433–436,  | PubMed | ISI | ChemPort |
16. Loeys, B, Van Maldergem, L, Mortier, G, Coucke, P, Gerniers, S, Naeyaert, JM, De Paepe, A: Homozygosity for a missense mutation in fibulin-5 (FBLN5) results in a severe form of cutis laxa. Hum Mol Genet 2002 11: 2113–2118, 10.1093/hmg/11.18.2113 | Article | PubMed | ISI | ChemPort |
17. Olsen, DR, Fazio, MJ, Shamban, AT, Rosenbloom, J, Uitto, J: Cutis laxa: Reduced elastin gene expression in skin fibroblast cultures as determined by hybridizations with a homologous cDNA and an exon 1- specific oligonucleotide. J Biol Chem 1988 263: 6465–6467,  | PubMed | ISI | ChemPort |
18. Olson, TM, Michels, VV, Urban, Z, et al: A 30 kb deletion within the elastin gene results in familial supravalvular aortic stenosis. Hum Mol Genet 1995 4: 1677–1679,  | PubMed | ISI | ChemPort |
19. Parks, WC, Deak, SB: Tropoelastin heterogeneity: Implications for protein function and disease. Am J Respir Cell Mol Biol 1990 2: 399–406,  | PubMed | ISI | ChemPort |
20. Sephel, GC, Byers, PH, Holbrook, KA, Davidson, JM: Heterogeneity of elastin expression in cutis laxa fibroblast strains. J Invest Dermatol 1989 93: 147–153, 10.1111/1523-1747.ep12277389 | Article | PubMed | ISI | ChemPort |
21. Szabo, Z, Levi-Minzi, SA, Christiano, AM, Struminger, C, Stoneking, M, Batzer, MA, Boyd, CD: Sequential loss of two neighboring exons of the tropoelastin gene during primate evolution. J Mol Evol 1999 49: 664–671,  | PubMed | ISI | ChemPort |
22. Tassabehji, M, Metcalfe, K, Hurst, J, et al: An elastin gene mutation producing abnormal tropoelastin and abnormal elastic fibres in a patient with autosomal dominant cutis laxa. Hum Mol Genet 1998 7: 1021–1028, 10.1093/hmg/7.6.1021 | Article | PubMed | ISI | ChemPort |
23. Tsuji, A, Yanai, J, Miura, T, et al: Vascular abnormalities in congenital cutis laxa–report of two cases. Acta Paediatr Jpn 1990 32: 155–161,  | PubMed | ChemPort |
24. Urban, Z, Michels, VV, Thibodeau, SN, et al: Isolated supravalvular aortic stenosis: Functional haploinsufficiency of the elastin gene as a result of nonsense-mediated decay. Hum Genet 2000 106: 577–588, 10.1007/s004390050028 | PubMed | ISI | ChemPort |
25. Urban, Z, Michels, VV, Thibodeau, SN, Donis-Keller, H, Csiszar, K, Boyd, CD: Supravalvular aortic stenosis: A splice site mutation within the elastin gene results in reduced expression of two aberrantly spliced transcripts. Hum Genet 1999 104: 135–142, 10.1007/s004390050926 | Article | PubMed | ISI | ChemPort |
26. Urban, Z, Zhang, J, Davis, EC, et al: Supravalvular aortic stenosis: Genetic and molecular dissection of a complex mutation in the elastin gene. Hum Genet 2001 109: 512–520, 10.1007/s00439-001-0608-z | PubMed | ISI | ChemPort |
27. Weir, EK, Joffe, HS, Blaufuss, AH, Beighton, P: Cardiovascular abnormalities in cutis laxa. Eur J Cardiol 1977 5: 255–261,  | PubMed | ISI | ChemPort |
28. Wrenn, DS, Griffin, GL, Senior, RM, Mecham, RP: Characterization of biologically active domains on elastin: Identification of a monoclonal antibody to a cell recognition site. Biochemistry 1986 25: 5172–5176,  | Article | PubMed | ISI | ChemPort |
29. Zhang, MC, He, L, Giro, M, Yong, SL, Tiller, GE, Davidson, JM: Cutis laxa arising from frameshift mutations in exon 30 of the elastin gene (ELN). J Biol Chem 1999 274: 981–986, 10.1074/jbc.274.2.981 | Article | PubMed | ISI | ChemPort |

Acknowledgments

We are grateful to the families whose cooperation made this study possible. This study was supported by National Institutes of Health grants HL73703 (Z. U.) and HL60394 (E. C. D.), by core facilities funded by NIH Grant RR16453 (Z. U.), by a Canadian Institutes of Health Research Grant MOP-57663 (E. C. D.), and by the Medical Research Council and the Ehlers Danlos Association of the United Kingdom (F. M. P.).