Introduction

Fibrillins are physiologically secreted as extended thread-like monomers into the extracellular matrix by many cell types. The mature functional entity, however, is constituted by higher order aggregates which are called microfibrils.¹ The assembly process of forming these supramolecular structures is a continuous process that probably requires a number of molecules and molecular events on its way from monomeric fibrillin to mature supramolecular microfibrils. It is likely that the molecular assembly of microfibrils is disturbed by at least some mutations in the gene for fibrillin-1 (FBN1) leading to Marfan syndrome (MFS) and other microfibrillopathies. This chapter summarizes the current knowledge of fibrillin assembly into microfibrils and discusses this understanding in the light of FBN1 mutations.

Methodological Approaches to Study Microfibril Assembly

To facilitate the discussion of the molecular assembly mechanisms of monomeric fibrillin and other molecules into supramolecular microfibrils, we introduce in this chapter arbitrary groups of events based on methodological accessibility of the involved processes. Early events take place within minutes to several hours, intermediate events within several days, and late events within weeks to months (Fig. 1). In the following, an overview is given about the methods used to monitor assembly in each of these stages.

Figure 1

Assembly of Microfibrils Involves Many Steps. After secretion of monomeric fibrillin, early assembly events takes place within minutes to hours and include for example disulfide-bond formation. After a few days (intermediate stages) a typical fibrillin (more...)

The early events in microfibril assembly are typically analyzed by metabolic labeling techniques and biochemical approaches. Fibrillins are rich in cysteine residues (˜13 %), a property that facilitates pulse chase metabolic labeling of cultured cells or organ cultures with radiolabeled [³⁵S]cysteine. Analysis of radiolabeled culture medium and extracted extracellular matrices by autoradiography after gel electrophoresis allows conclusions on molecular events based on changes of the molecular size or concentrations of fibrillin monomers or multimers. With this approach, it is possible to analyze the role of profibrillin processing, and of fibrillin multimerization mediated by formation of reducible and nonreducible cross-links. Biochemical analyses of assembly intermediates and of molecular properties important for fibrillin assembly have been hampered by the propensity of fibrillin to form reducible and nonreducible cross-links in tissues. Therefore, extraction procedures for fibrillin from tissues usually include reducing agents resulting in denaturation of the protein.² Although it is in principal possible to purify fibrillin monomers from cell culture sources using nondenaturing buffer systems, the required methods are time consuming and costly.³ To solve this dilemma, various recombinant expression techniques have been developed over the past few years for production of fibrillin fragments, such as expression in eukaryotic cells, in bacteria, as well as expression with in vitro translation systems. Recombinant fragments of fibrillin-1 and fibrillin-2 produced by these systems have been extensively used in various biochemical assembly assays.

Within a few days fibrillin assembles into a loosely structured network in cell culture assays (Fig. 1).³ Primary cells of mesenchymal origin such as dermal fibroblasts, smooth muscle cells, chondrocytes, ligament cells and established cell lines such as MG-63 (human osteosarcoma) and others secrete fibrillin into the culture medium and assemble microfibrillar networks (Fig. 1). These fibrillin networks can be easily visualized by indirect immunofluorescence with specific mono- or polyclonal antibodies against the specific fibrillin isoform. This type of assembly assay is often used to demonstrate quantitative and qualitative differences of the assembled fibrillin network in various situations. Additionally, potential modifiers and inhibitors can be analyzed presuming that these reagents are not toxic for the cells.

After a few weeks of cell culture, “beads-on-a-string” microfibrils can be extracted from the extracellular layer of cells in culture and visualized by transmission electron microscopy after rotary shadowing or negative staining of the samples (Fig. 1).^4,5 Microfibrils extracted from cell culture systems are very similar, if not identical, to microfibrils extracted from tissue sources such as ocular tissues,^6,7 placenta extracts,⁸ or bovine nuchal ligaments.⁹ Identification of antibody epitopes for fibrillin has led to a relatively precise localization and arrangement of fibrillin monomers within the beaded microfibrils as described in detail in chapter “Organization and Biomechanical Properties of Fibrillin Microfibrils”. While extracted microfibrils can be used for example to test the structural integrity in various situations, they are of limited use to analyze dynamic assembly mechanisms. Beaded microfibrils do not represent the final stage of microfibrils in tissues. Tissue microfibrils often do not appear as beaded structures but simply as bundles of extended thread-like aggregates (Fig. 1). It is likely that components that are peripherally associated with tissue microfibrils are released by the extraction procedures, and thus converting the thread-like structure to the beaded form. Analysis of tissue microfibrils is helpful to uncover mechanisms relevant to assembly in mouse models with modified fibrillin genes.¹⁰¹²

Microfibril Assembly Is Often Disturbed in Individuals with MFS

Pulse chase experiments with dermal fibroblasts from individuals with MFS identified defects in fibrillin synthesis, secretion and aggregation into the extracellular matrix, indicating that at least for some fibrillin mutations early assembly events are compromised.^13,14 In a detailed study, five different groups of MFS fibroblasts have been identified based on the parameters of fibrillin synthesis and deposition into the extracellular matrix.¹⁵ Despite differences in the amount of secreted fibrillin, fibroblasts in three of the five groups (43 of 55 analyzed) showed impaired incorporation of the mutant fibrillin into extracellular aggregates at various levels, suggesting functional problems in early assembly steps.¹⁵ Consequently, the intermediate assembly stages of the relevant groups, assessed by indirect immunofluorescence of fibrillin, showed considerably reduced or thinner microfibrillar networks as compared to normal controls.¹⁶ Originally, immunofluorescence studies with skin biopsies from individuals with MFS or with dermal fibroblasts cultivated from such tissues samples revealed that the intermediate microfibrillar networks in cell culture or the mature microfibrils in skin were reduced in amount as compared to age-matched controls.¹⁷¹⁹ One possible interpretation of these data is that fibrillin assembly is impaired at a stage upstream of the intermediate stage. However, reduced or qualitatively altered fibrillin staining patterns in skin or dermal fibroblasts from individuals clinically diagnosed with MFS are not always observed. In one study, about 70 % (16 of 23) of affected skin samples and 89 % (16 of 18) of the respective fibroblast cultures showed altered fluorescence patterns.¹⁹ In contrast, in another study, none (0 of 7) of the analyzed fibroblasts cultures established from individuals with MFS showed alterations in the immunostainable fibrillin patterns.²⁰ Therefore, in terms of fibrillin assembly, it is likely that only a subset of fibrillin mutations resulting in MFS disturbs the assembly mechanism. In the severe neonatal form of MFS, which is caused by mutations in the central region of fibrillin-1 coded by exons 24–32, reduced immunostainable fibrillin-1 depositions have also been reported.²¹²³ Additionally, in these samples, fibrillin fibers appeared short, fragmented and frayed, indicating that the central part of the fibrillin-1 molecule has a role in elongation and/or alignment of fibrillin within microfibrils (see also Fig. 2).

Figure 2

Regions in Fibrillin-1 Potentially Important for Assembly. Shown is a schematic model of fibrillin-1 composed of various modules. “S” represent thiol groups in cysteine residues potentially involved in intermolecular disulfide bond formation. (more...)

A number of studies have addressed the ultrastructural appearance of beaded microfibrils obtained from long term cell cultures of MFS fibroblasts. Four types of abnormalities have been characterized when compared with normal controls: The interbead domains were diffuse and poorly defined,^5,24,25 and cell cultures produced nearly no beaded structures at all,^24,26 the interbead domains appeared frayed,^24,25 and variable interbead periodicities were observed.^24,25 All of these consequences on the morphology of beaded microfibrils could potentially arise from various defects in fibrillin assembly.

In summary, it is clear for early assembly steps, analyzed by pulse chase experiments, that a subset of the MFS mutations directly affect assembly events. Based on the correlation of this subset with intermediate consequences, it is likely that other abnormalities observed in the intermediate and the late stages are also results of problems in the assembly process. All of the observed consequences on the early, intermediate and late assembly stages are not conclusive as to which molecular event may be compromised. However, not all MFS mutations result in defects in the assembly mechanism, but this observation is not addressed in this chapter.

Role of Propeptide Processing in Fibrillin Assembly

Originally, pulse chase experiments with human skin fibroblasts suggested that fibrillin-1 is secreted from the cells as a precursor protein of ˜350 kDa, which is processed to a mature ˜320 kDa form.^13,27 Conversion of profibrillin-1 to fibrillin-1 usually is completed within several hours. Only the processed form of fibrillin-1 is deposited into the extracellular matrix, suggesting that profibrillin-1 conversion to mature fibrillin-1 plays a regulatory role in assembly into higher order aggregates.¹³ The reduction of about 30 kDa upon processing implied that a relatively small fragment of fibrillin-1 was removed from either one or both terminal ends, indicating that the cleavage sites are located within the unique terminal domains (Fig. 2). The conversion of profibrillin-1 to fibrillin-1 was experimentally inhibited by calcium depletion of the culture medium, indicating that calcium-dependent enzymes are responsible for the conversion.^26,28,29 A mutation in fibrillin-1 (R2726W) associated with isolated skeletal features of MFS was shown to interfere with normal processing at the C-terminal end and thus disturbed the incorporation of the mutated protein into the extracellular matrix.²⁷ Since the amino acid residue altered by this mutation occurs immediately adjacent to a sequence motif (RKRR), that matches the tribasic recognition site for proprotein proteases of the furin/PACE type (RX(K/R)R), it has been suggested that processing is mediated by this type of endoproteases.²⁷ Furin/PACE type proteases are responsible for cleavage of propeptides of a variety of secreted proteins (for review see reference 30). The predicted propeptide removed from the C-terminal end of fibrillin-1 by processing spans 140 amino acid residues. Essentially similar conclusions were drawn based on the analysis of a mutation harboring a premature stop codon (W2756ter), which also prevents profibrillin-1 processing in the C-terminal domain.²⁸ It is interesting that the W2756ter mutation caused over-glycosylation of the mutated fibrillin-1 due to intracellular retention. On the other hand, elimination of the N-glycosylation site (NET) directly C-terminal of the furin/PACE recognition motif in the C-terminal domain increased processing by 15–20 % (Fig. 2).³¹ These data indicate a regulatory role of this N-glycosylation site for processing of profibrillin-1.

Direct evidence for utilization of this recognition sequence was obtained by analysis of a recombinant fibrillin-1 fragment comprising the last calcium binding epidermal growth factor (cbEGF) like module and the unique C-terminal domain.³² This recombinant fragment was cleaved during or after secretion directly after the tribasic recognition sequence (RKRR↓STNET, Fig. 2). In another recombinant system, processing was demonstrated by specific antibodies against the C-propeptide.²⁹ In addition, site directed mutagenesis of the furin/PACE recognition sequence at various positions inhibited processing and thus further substantiated that processing is mediated by furin/PACE-type proteases.^29,32,33 Furthermore, it was shown that various members of the furin/PACE family, namely furin/PACE, PACE4, PC1/3, and PC2, were able to mediate processing and that an inhibitor for furin/PACE-type proteases prevented processing as well as matrix deposition.³³ The importance of profibrillin-1 processing as a crucial assembly requirement is further highlighted by the fact that the tribasic furin/PACE recognition sequence in the C-terminal domain is conserved in human fibrillin-2 (RQKR) and fibrillin-3 (RPRR) as well as in fibrillins from other species, indicating that those fibrillin isoforms are probably processed in an identical or similar fashion.³⁴³⁸

The discussion at what topological site the C-terminal processing takes place is controversial. Pulse chase methods have led to the conclusion that the site of profibrillin-1 processing is extracellular.^13,27 n contrast, experiments with secretion blocking agents and recombinant constructs indicated intracellular cleavage of the profibrillin-1 carboxy-terminal region in an early secretory pathway compartment.²⁹ Although furin/PACE-type proteases are typically localized to the trans-Golgi network, there is evidence that these proteins cycle between the cell surface, where they would be able to process extracellular proteins, and the Golgi apparatus (for review see reference 30). For understanding of the assembly mechanism, it is important to clarify these controversial issues in the future. Dependent on the local environment for the initial assembly steps (secretory pathway versus extracellular matrix), the consequences of mutations in fibrillin-1 in the region of the processing site may lead to a different outcome in terms of how they interfere with assembly.

Recognition sequence motifs for furin/PACE-type proteases do not only occur in the unique C-termini of fibrillins but also in the unique N-termini of all fibrillins.^34,35,3941 For fibrillin-1, N-terminal sequencing of the full length authentic protein, isolated from cell culture medium of dermal fibroblasts, produced a sequence starting directly after the tribasic recognition sequence motif RAKR↓R45GGG (Fig. 2).⁴² In addition, it has been shown that recombinant N-terminal fragments expressed in mammalian cells are cleaved at the same position.^43,44 On the other hand, the cleavage site for the signal peptide, responsible for secretion of fibrillin-1 into the endoplasmic reticulum, was experimentally determined using recombinant systems. The preferred site of cleavage is located between glycine at position 24 and the adjacent alanine (G24↓A25) and a secondary minor site between alanine at position 27 and the adjacent asparagine (A27↓N28).^29,42 Therefore, removal of the propeptide by furin/PACE-type proteases only removes 20 or 17 amino acid residues, respectively. Due to this very small difference, experimental analyses to address whether processing of the N-terminal propeptide is also a prerequisite for the fibrillin-1 assembly have been hampered. No mutation leading to MFS has been detected within the cleaved propeptide.⁴⁵ It is not clear whether other mutations, which are located relatively close C-terminal to the processing site, have any adverse effect on profibrillin-1 processing at the N-terminus. Interestingly, fibrillin deposition into the extracellular matrix was markedly diminished by over-expression of a construct in which exon 2 of the coding sequence for fibrillin-1 was deleted.⁴⁶ This deletion predicts a mutant polypeptide consisting of 100 amino acid residues composed of the authentic fibrillin-1 sequence 1–55 followed by 45 out of frame nonsense residues. If this mutant polypeptide is processed like the wild-type fibrillin-1 after position 44, then only 11 functional residues would be able to disrupt normal fibrillin assembly. However, a potential “toxic” effect of the nonsense sequence (resulting from the frameshift), that is normally not part of the fibrillin-1 molecule, also needs to be considered.

Role of Intermolecular Cross-Link Formation in Fibrillin Assembly

Fibrillin can be immunoprecipitated from the medium and the cell layer of fibroblast and smooth muscle cell cultures in the form of monomers and higher molecular weight disulfide-bonded aggregates.⁴ These results are in agreement with metabolic labeling assays of organ cultures, demonstrating rapid formation of disulfide-bonded aggregates as one of the initial steps in fibrillin assembly.⁴² These concepts are supported by the observation that reducing agents are necessary to extract fibrillin from tissues.² Based on the structures of individual modules contained in fibrillins, it is believed that almost all cysteine residues contribute to intramolecular disulfide-bonds.^47,48 Only a few candidate cysteine residues are potentially available for intermolecular disulfide-bonds (Fig. 2): 4 cysteine residues in the unique N-terminal domain, 2 cysteine residues within the unique C-terminal domain, and, additionally, the first hybrid (Fib-like) module contains one unpaired cysteine residue. All of these cysteine residues are conserved between the fibrillin isoforms and between species.^3438,41,4951 It has been shown using a recombinant approach that the 4 cysteine residues in the unique N-terminal domain are involved in intramolecular disulfide-bonds, while cysteine 204 in the first hybrid domain occurs as a free thiol on or close to the surface of the molecule (Fig. 2).⁴² Currently, it is still unclear whether the 2 cysteine residues in the C-terminal domain can contribute to intermolecular disulfide bridges. It is very likely that more than one cysteine residue participates in intermolecular disulfide bonding, since molecular aggregates with more than two fibrillin molecules require more than one connecting residue per molecule. It is tempting to speculate that the two cysteine residues in the C-terminal domain, in addition to cysteine 204, can fulfill such a bridging function. The two candidate cysteine residues in the C-terminal domain are located very close to the furin/PACE processing site (Fig. 2). Thus, they potentially could be involved in regulatory mechanisms connecting profibrillin processing with intermolecular disulfide bond formation in early assembly stages. At present, however, experimental evidence is lacking for this hypothesis. No mutations leading to MFS have been reported for either cysteine 204 or for the cysteine residues in the unique N- and C-terminal domains.⁴⁵ It is possible that loss of these cysteine residues is not compatible with life.

Other experimental evidence, however, suggests that other cysteine residues may be involved in intermolecular disulfide bond formation. Reducible homodimer formation was observed for recombinant fragments of fibrillin-1 and fibrillin-2 spanning from the proline and glycine-rich modules, respectively, to the second 8-Cys/TB module.⁵² The authors hypothesized that cysteine residues in the second 8-Cys/TB module are involved in formation of intermolecular disulfide bridges (Fig. 2). Interestingly, in this study, the homodimer formation occurred early during biosynthesis in the endoplasmic reticulum.⁵² Another study has utilized recombinant fragments of the proline-rich region of fibrillin-1 and the glycine-rich region of fibrillin-2 including flanking domains.⁵³ These fragments also had the tendency to dimerize, although dimer formation was assigned to the extracellular compartment. The authors suggested that molecular recognition might be mediated through the proline and glycine domains, respectively, and that the dimers become covalently stabilized by a hypothetic disulfide exchange mechanism involving cysteine residues in the first 8-Cys/TB module (Fig. 2).⁵³ In general, disulfide reshuffling mechanisms for extracellular matrix proteins have not been extensively studied. However, some protein disulfide isomerases, which mediate disulfide bond exchanges, can be found on the cell surface and in the extracellular matrix (for review see ref. 54). It may be even possible that disulfide isomerases exist, that specifically mediate disulfide exchanges in fibrillins and microfibrils.

Further support for dimer formation comes from ultrastructural studies of microfibrils. In some cases interbead regions appear as two defined strands^43,55 rather than the 6–8 strands, which are believed to form the mature microfibrils.^6,7,56 However, it is not clear at present whether the two-stranded interbead regions represent early assembly intermediates or rather degradation products.

Typically, extraction of microfibrillar components from mature tissues using reducing conditions results in relatively small amounts, indicating the presence of additional nonreducible cross-links in microfibrillar aggregates. Nonreducible ϵ(γ-glutamyl)lysine cross-links have in fact been identified in microfibrils extracted from various tissues.⁵⁷⁵⁹ A family of Ca²⁺-dependent enzymes, the transglutaminases, catalyze the formation of such cross-links between a γ-carboxyamide group of a glutamine and a lysine residue (for review see reference 60). In ocular tissue, the zonular fibers have been shown to be a target for a member of this family, transglutaminase 2.⁶¹ The analysis of extracted microfibrils from human amnion indicated that 10–15 % of the lysine residues in microfibrils are involved in cross-links.⁵⁸ One cross-link region between a N-terminal region of the fibrillin-1 molecule starting at position 580 and a C-terminal region starting at position 2312 has been identified (Fig. 2).⁵⁸ Based on biomechanical approaches, it has been suggested that transglutaminase cross-links play a major role in strengthening microfibrillar networks.⁶² In terms of assembly mechanisms, it is conceivable that transglutaminase cross-links may facilitate correct molecular alignments at fixed positions during early microfibrillar assembly stages as a prerequisite for further assembly. At present however, the time course for the formation of transglutaminase cross-links is not known and needs to be clarified. In addition to fibrillin-1, the microfibril associated glycoprotein (MAGP)-1 (see below) was also characterized as a substrate for transglutaminase.⁶³ Thus, it is possible that besides homotypic fibrillin-1 transglutaminase cross-links, heterotypic fibrillin-1-MAGP-1 cross-links may be present in microfibrils. However, the transglutaminase cross-links in microfibrils have been localized to the interbead domain of beaded microfibrils, while MAGP-1 was primarily localized to the beads.^58,64 Mutations that disrupt transglutaminase cross-link sites are predicted to result in serious consequences for microfibril assembly and stability, which would likely contribute to the pathogenesis of MFS. To advance the understanding of the role of transglutaminase cross-links in MFS, it is important to exactly identify the amino acid residues involved.

Interaction Epitopes Important for Fibrillin Assembly

A relatively large number of protein ligands have been immunolocalized to microfibrils (for review see reference 65). However, on the molecular level, only a few fibrillin-protein interactions have been identified. Fibulin-2 interacts with high affinity with fibrillin-1 and are found on some but not all microfibrils.⁴⁴ Similar observations have been made with versican,⁶⁶ and LTBP-1 and -4.⁶⁷ While the precise functional roles of these microfibril-associated proteins is not clear at present, it appears unlikely that they fulfill an essential role in the assembly mechanism, since (i) they are not present in all microfibrils, and (ii) they are not regularly spaced in microfibrils. If a microfibril-associated protein serves an important role in the assembly process, a regular arrangement of this ligand within microfibrils would be expected. A periodic labeling pattern similar to the pattern of fibrillins have been demonstrated for MAGP-1 and -2.^64,68 While MAGP-1 codistributes with microfibrils in most if not all tissues where they occur,^69,70 MAGP-2 appears to be expressed in a more tissue-specific fashion and is thus not a good candidate for an assembly mediator.⁶⁸ MAGP-1 can interact with fibrillin-1 either directly,⁷¹ or mediated by the proteoglycan decorin.⁷² The interaction epitope of MAGP-1 with fibrillin-1 is localized at the N-terminal region of the fibrillin-1 molecule (Fig. 2).⁷¹

During elastic fiber synthesis, the microfibrils appear before the amorphous elastin core and are believed to act as a scaffold for the deposition of tropoelastin.⁷³ Therefore it is unlikely that direct⁷⁴ or indirect interactions, possibly mediated by MAGP-1,^63,71,75 of fibrillins with tropoelastin play a critical role in the assembly of microfibrillar structures. Moreover tropoelastin is not present in nonelastic tissues such as the ciliary zonules in the eye where abundant bead-on-a-string structures can be found.

In summary, although various proteins ligands have been identified to interact with fibrillins, at present there is no conclusive evidence that one or more of these fibrillin binding ligands play an important role in the fibrillin assembly process. These ligands rather may have modulating functional roles in the biology of microfibrils.

In another scenario, the molecules that assist in the assembly process are not permanently associated with microfibrils but rather transiently “catalyze” the assembly process. Such molecules would bind to fibrillins or other structural microfibrillar proteins probably at an early stage of the assembly process and release as soon as the specific assembly event has been completed. In this light, it is important to mention that three regions of fibrillin-1 have been found to interact with heparin/heparan sulfate with high affinity (Fig. 2).⁷⁶ In a cell culture assembly assay with dermal fibroblasts, these glycosaminoglycans can very efficiently inhibit microfibrillar assembly leading to the hypothesis that binding of fibrillin to highly sulfated heparan side chains of proteoglycans is important to nucleate the assembly process (Fig. 3).⁷⁶ This hypothesis is further supported by the fact that inhibition of sulfation or formation of heparan sulfate chains compromises microfibrillar assembly.^72,76 Since assembly of microfibrils is believed to take place close to the cell surface, it is possible that heparan sulfate containing proteoglycans located on the plasma membrane, such as syndecans or glypicans, may fulfill such a catalyzing function. Binding of fibrillins to cell surface glycosaminoglycans may be necessary to generate high local concentrations required for assembly or for the alignment of fibrillin molecules in the correct spatial orientation, which would allow formation of intermolecular disulfide bonds. Alternatively, binding of fibrillin to heparan sulfate chains potentially confer conformational changes necessary to expose assembly epitopes.

Figure 3

Hypothetical Sequence of the Fibrillin Assembly Steps into Microfibrils. Various cell types such as fibroblasts or smooth muscle cells synthesize and secrete fibrillin. Processing of the terminal ends by furin/PACE type proteases occur either intra- or (more...)

Binding of fibrillins to cell surface integrin receptors has been clearly defined. Fibrillin-1 and -2 interact via the fourth 8-Cys/TB domain, which contains a RGD sequence, with αvβ3 integrin as shown with recombinant polypeptides⁷⁷ and with purified fibrillin from tissues using denaturing agents (Fig. 2).⁷⁸ In addition, members of the β1 integrin dimers and possibly α5β1 integrin may also play a role in the cellular interactions of fibrillins.^78,79 Potentially, fibrillin-integrin interactions may play a distinct role in the assembly process similar to the role of α5β1 in fibronectin assembly (for review see ref. 80). Interactions of fibronectin with the α5β1 integrin introduce conformational changes in the fibronectin molecules resulting in exposed assembly epitopes. Much of the discussion about the role of cell surface receptors in fibrillin assembly is of speculative nature. More research efforts are necessary to answer the underlying questions and to clarify potential genotype-biochemical phenotype relationships in terms of cellular interaction of mutated fibrillin-1 in individuals with MFS. At present no point mutation, other than the generation of a stop signal, has been found within the RGD sequence of fibrillin-1 that would allow conclusions on these important questions.⁴⁵

The static organization of fibrillin-1 in microfibrils has been examined by several groups and various techniques (see the chapter “Organization and Biomechanical Properties of Fibrillin Microfibrils” in this volume). Labeling of microfibrils with specific antibodies, high resolution structure of cbEGF modules, and analysis of transglutaminase cross-links have lead to various models of fibrillin alignment in microfibrils.^{43,47,56,58,81} Despite the controversy whether fibrillin molecules are arranged in a nonstaggered or in a staggered fashion, common to all models is a head-to-tail arrangement of fibrillin molecules originally proposed by Sakai and coworkers in 1991.⁸¹ Mapping of monoclonal antibody epitopes in fibrillin-1 molecules and correlation with the epitopes in microfibrils revealed that the N- and the C-terminal ends of the fibrillin-1 molecules are located closely together in or close to the bead structures.^43,56,81 To address the question whether fibrillin molecules directly interact with each other or whether they need adapter molecules such as MAGP-1 for example, a recombinant approach has been utilized.⁸² These studies demonstrated homotypic self-interaction properties of fibrillin-1 and heterotypic interaction properties between fibrillin-1 and -2 in a N- to C-terminal fashion.⁸² The interactions were of high affinity in the low nanomolar range. Aggregation of full length recombinant fibrillin-1 via its terminal regions further supported these data, emphasizing that at least early assembly events do not require adapter molecules.⁸² Heterotypic interactions between fibrillin-1 and fibrillin-2 in a similar fashion suggested that both isoforms can be colocalized within the same microfibril. Immunogold localization of fibrillin-1 and -2 within the same microfibril further supported this hypothesis.⁸³ Interestingly, fibrillin-2 was not able to homotypically interact with itself, suggesting that fibrillin-2 alone might use a different mechanism to assemble into microfibrils.⁸² For these studies, relatively large recombinant polypeptides have been utilized. Interestingly, homo- and heterotypic interactions could not be observed using smaller subfragments, suggesting that the assembly epitopes are stabilized by long range structural effects.⁸² In the light of mutations leading to MFS, this observation may indicate that in addition to the mutations located in close vicinity to the assembly epitopes, mutations located farther away may result in conformational changes of the assembly epitopes. Further work with mutation constructs will be necessary to clarify this hypothesis.

Mouse Models and Fibrillin Assembly

The contribution of the central part of the fibrillin-1 molecule to microfibrillar assembly was analyzed in different mouse models. In one model a sequence coding for residues 770–1042 (coded by exons 19–24) of the fibrillin-1 molecule was genetically deleted, and due to transcriptional interference by the neo cassette, the mutated fibrillin-1 was expressed at low levels.¹⁰ Immunohistochemical analysis of fibrillin-1 assembly by homozygous mutant dermal fibroblasts revealed fewer and more primitive appearing microfibrils. However, immunoelectron microscopy of skin sections of homozygous mutant animals demonstrated the presence of beaded microfibrils.¹⁰ Despite the uncertainty whether these microfibrils represented fully functional aggregates, these data clearly demonstrated that deletion of the region between the cbEGF module 8 and 8-Cys/TB module 3 does not preclude microfibrillar assembly.

The Tsk mutation is a genomic in-frame duplication of a relatively large central part of the mouse fibrillin-1 gene resulting in the “tight skin” phenotype.⁸⁴ This mutation predicts a larger fibrillin-1 molecule with 984 additional amino acid residues and a molecular mass of ˜418 kDa instead of the normal ˜350 kDa fibrillin-1 product. The assembly properties of this mutant fibrillin-1 has been analyzed by two groups: Kielty and coworkers analyzed beaded microfibrils isolated from skin of the heterozygous Tsk/+ mouse and found the occurrence of two mutually exclusive populations of beaded microfibrils.¹¹ One population appeared normal and was suggested to be composed of normal fibrillin-1, while the other population displayed longer than normal bead-to-bead periodicities, as well as an altered morphology and organization, and thus was suggested to contain the abnormally long fibrillin-1. The authors concluded a molecular selection based on the length of the molecules and suggested that lateral alignment of fibrillin molecules is a crucial assembly step which precedes or facilitates linear polymerization. Based on analysis of Tsk/Tsk fibroblasts, Gayraud and coworkers also found that the abnormally long fibrillin-1 can form homopolymeric beaded microfibrils with abnormal morphologies such as irregularities in the size and shape of the globular beads, as well as in the distances between the beads.¹² However, these authors demonstrated that the longer Tsk fibrillin-1 is also able to copolymerize together with normal wild-type fibrillin-1 into abnormal beaded microfibrils. It was concluded that the central long stretch of cbEGF-like domains, which is duplicated in the abnormal Tsk fibrillin-1, has an important role in the alignment of fibrillin-1 molecules within microfibrils and in the elongation process.¹² Collectively, these data further strengthen the idea that primary assembly epitopes closer to the terminal ends of the fibrillin-1 molecules, which are not affected by the mutated Tsk fibrillin-1, are important for the initial assembly processes while regions in the center of the molecule may be important for lateral alignment and elongation.

Future Directions

Despite the advances over the last years in understanding the assembly mechanism of fibrillins into microfibrils, only a fragmented picture has evolved. This is in part due to the lack of feasible and modifiable assays to monitor fibrillin assembly. Therefore, the development of such assays will be crucial for more in depth understanding of the principle mechanisms. The major goal of such assay systems will be to precisely define (i) the assembly epitopes in fibrillins, (ii) the structural components important for assembly, and (iii) catalysts, which only transiently participate in microfibrillar assembly. As our understanding of the assembly mechanism increases, so will the potential to correlate the genotype with the biochemical phenotype in MFS. The hope is that from a detailed understanding of this correlation, more mechanism-based approaches for the development of therapeutic strategies will evolve.

Acknowledgements

This work was supported by the Deutsche Forschungsgemeinschaft (Grants SFB367-A1, Re1021/3-2, and Re1021/4-2). We are grateful to Lynn Y. Sakai and Douglas R. Keene for generously providing parts of Fig. 1.

References

1.

Low FN. The extracellular portion of the human blood-air barrier and its relation to tissue space. Anat Rec. 1961;139:105–124.

2.

Gibson MA, Kumaratilake JS, Cleary EG. The protein components of the 12- nanometer microfibrils of elastic and nonelastic tissues. J Biol Chem. 1989;264:4590–4598. [PubMed: 2647740]

3.

Sakai LY, Keene DR. Fibrillin: Monomers and microfibrils. Methods Enzymol. 1994;245:29–52. [PubMed: 7760739]

4.

Kielty CM, Shuttleworth CA. Synthesis and assembly of fibrillin by fibroblasts and smooth muscle cells. J Cell Sci. 1993;106:167–173. [PubMed: 8270622]

5.

Kielty CM, Shuttleworth CA. Abnormal fibrillin assembly by dermal fibroblasts from two patients with Marfan syndrome. J Cell Biol. 1994;124:997–1004. [PMC free article: PMC2119967] [PubMed: 8132720]

6.

Wright DW, Mayne R. Vitreous humor of chicken contains two fibrillar systems: An analysis of their structure. J Ultrastruct Mol Struct Res. 1988;100:224–234. [PubMed: 2468720]

7.

Wallace RN, Streeten BW, Hanna RB. Rotary shadowing of elastic system microfibrils in the ocular zonule, vitreous, and ligament nuchae. Curr Eye Res. 1991;10:99–109. [PubMed: 1709401]

8.

Keene DR, Maddox BK, Kuo HJ. et al. Extraction of extendable beaded structures and their identification as fibrillin-containing extracellular matrix microfibrils. J Histochem Cytochem. 1991;39(4):441–449. [PubMed: 2005373]

9.

Kielty CM, Cummings C, Whittaker SP. et al. Isolation and ultrastructural analysis of microfibrillar structures from foetal bovine elastic tissues. J Cell Sci. 1991;99:797–807. [PubMed: 1770007]

10.

Pereira L, Andrikopoulos K, Tian J. et al. Targeting of fibrillin-1 recapitulates the vascular phenotype of Marfan syndrome in the mouse. Nat Genet. 1997;17(2):218–222. [PubMed: 9326947]

11.

Kielty CM, Raghunath M, Siracusa LD. et al. The tight skin mouse: Demonstration of mutant fibrillin-1 production and assembly into abnormal microfibrils. J Cell Biol. 1998;140(5):1159–1166. [PMC free article: PMC2132699] [PubMed: 9490728]

12.

Gayraud B, Keene DR, Sakai LY. et al. New insights into the assembly of extracellular microfibrils from the analysis of the fibrillin 1 mutation in the tight skin mouse. J Cell Biol. 2000;150(3):667–680. [PMC free article: PMC2175205] [PubMed: 10931876]

13.

Milewicz D, Pyeritz RE, Crawford ES. et al. Marfan syndrome: Defective synthesis, secretion, and extracellular matrix formation of fibrillin by cultured dermal fibroblasts. J Clin Invest. 1992;89:79–86. [PMC free article: PMC442822] [PubMed: 1729284]

14.

Aoyama T, Tynan K, Dietz HC. et al. Missense mutations impair intracellular processing of fibrillin and microfibril assembly in Marfan syndrome. Hum Mol Genet. 1993;2:2135–2140. [PubMed: 8111384]

15.

Aoyama T, Franke U, Dietz HC. et al. Quantitative differences in biosynthesis and extracellular deposition of fibrillin in cultured fibroblasts distinguish five groups of Marfan syndrome patients and suggest distinct pathogenetic mechanisms. J Clin Invest. 1994;94:130–137. [PMC free article: PMC296290] [PubMed: 8040255]

16.

Brenn T, Aoyama T, Francke U. et al. Dermal fibroblast culture as a model system for studies of fibrillin assembly and pathogenetic mechanisms: Defects in distinct groups of individuals with Marfan's syndrome. Lab Invest. 1996;75(3):389–402. [PubMed: 8804362]

17.

Godfrey M, Olson S, Burgio RG. et al. Unilateral microfibrillar abnormalities in a case of asymetric Marfan syndrome. Am J Hum Genet. 1990;46:661–671. [PMC free article: PMC1683667] [PubMed: 2180285]

18.

Godfrey M, Menashe V, Weleber RG. et al. Cosegregation of elastin-associated microfibrillar abnormalities with the Marfan phenotype in families. Am J Hum Genet. 1990;46:652–660. [PMC free article: PMC1683653] [PubMed: 2180284]

19.

Hollister DW, Godfrey M, Sakai LY. et al. Immunohistologic abnormalities of the microfibrillar-fiber system in the Marfan syndrome. N Engl J Med. 1990;323:152–159. [PubMed: 2194127]

20.

Halliday D, Hutchinson S, Kettle S. et al. Molecular analysis of eight mutations in FBN1. Hum Genet. 1999;105(6):587–597. [PubMed: 10647894]

21.

Raghunath M, Superti-Furga A, Godfrey M. et al. Decreased extracellular deposition of fibrillin and decorin in neonatal Marfan syndrome fibroblasts. Hum Genet. 1993;90:511–515. [PubMed: 8428751]

22.

Godfrey M, Raghunath M, Cisler J. et al. Abnormal morphology of fibrillin microfibrils in fibroblast cultures from patients with neonatal Marfan syndrome. Am J Pathol. 1995;146(6):1414–1421. [PMC free article: PMC1870897] [PubMed: 7778680]

23.

Wang M, Price C, Han J. et al. Recurrent mis-splicing of fibrillin exon 32 in two patients with neonatal Marfan syndrome. Hum Mol Genet. 1995;4:607–613. [PubMed: 7633409]

24.

Kielty CM, Phillips JE, Child AH. et al. Fibrillin secretion and microfibril assembly by Marfan dermal fibroblasts. Matrix Biol. 1994;14(2):191–199. [PubMed: 8061930]

25.

Kielty CM, Davies SJ, Phillips JE. et al. Marfan syndrome: Fibrillin expression and microfibrillar abnormalities in a family with predominant ocular defects. J Med Genet. 1995;32(1):1–6. [PMC free article: PMC1050170] [PubMed: 7897619]

26.

Raghunath M, Kielty CM, Kainulainen K. et al. Analyses of truncated fibrillin caused by a 366 bp deletion in the FBN1 gene resulting in Marfan syndrome. Biochem J. 1994;302:889–896. [PMC free article: PMC1137314] [PubMed: 7945217]

27.

Milewicz DM, Grossfield J, Cao SN. et al. A mutation in FBN1 disrupts profibrillin processing and results in isolated skeletal features of the Marfan syndrome. J Clin Invest. 1995;95:2373–2378. [PMC free article: PMC295860] [PubMed: 7738200]

28.

Raghunath M, Kielty CM, Steinmann B. Truncated profibrillin of a Marfan patient is of apparent similar size as fibrillin: Intracellular retention leads to over-N-glycosylation. J Mol Biol. 1995;248:901–909. [PubMed: 7760331]

29.

Ritty TM, Broekelmann T, Tisdale C. et al. Processing of the Fibrillin-1 Carboxyl-terminal Domain. J Biol Chem. 1999;274(13):8933–8940. [PubMed: 10085138]

30.

Nakayama K. Furin: A mammalian subtilisin/Kex2p-like endoprotease involved in processing of a wide variety of precursor proteins. Biochem J. 1997;327( Pt 3):625–635. [PMC free article: PMC1218878] [PubMed: 9599222]

31.

Ashworth JL, Kelly V, Rock MJ. et al. Regulation of fibrillin carboxy-terminal furin processing by N-glycosylation, and association of amino- and carboxy-terminal sequences. J Cell Sci. 1999;112:4163–4171. [PubMed: 10547375]

32.

Lönnqvist L, Reinhardt DP, Sakai LY. et al. Evidence for furin-type activity-mediated C-terminal processing of profibrillin-1 and interference in the processing by certain mutations. Hum Mol Genet. 1998;7:2039–2044. [PubMed: 9817919]

33.

Raghunath M, Putnam EA, Ritty T. et al. Carboxy-terminal conversion of profibrillin to fibrillin at a basic site by PACE/furin-like activity required for incorporation in the matrix. J Cell Sci. 1999;112:1093–1100. [PubMed: 10198291]

34.

Zhang H, Apfelroth SD, Hu W. et al. Structure and expression of fibrillin-2, a novel microfibrillar component preferentially located in elastic matrices. J Cell Biol. 1994;124:855–863. [PMC free article: PMC2119952] [PubMed: 8120105]

35.

Nagase T, Nakayama M, Nakajima D. et al. Prediction of the coding sequences of unidentified human genes. XX. The complete sequences of 100 new cDNA clones from brain which code for large proteins in vitro. DNA Res. 2001;8(2):85–95. [PubMed: 11347906]

36.

Tilstra DJ, Li L, Potter KA. et al. Sequence of the coding region of the bovine fibrillin cDNA and localization to bovine chromosome 10. Genomics. 1994;23:480–485. [PubMed: 7835900]

37.

Yin W, Smiley E, Germiller J. et al. Primary structure and developmental expression of Fbn-1, the mouse fibrillin gene. J Biol Chem. 1995;270:1798–1806. [PubMed: 7829516]

38.

Zhang H, Hu W, Ramirez F. Developmental expression of fibrillin genes suggests heterogeneity of extracellular microfibrils. J Cell Biol. 1995;129:1165–1176. [PMC free article: PMC2120487] [PubMed: 7744963]

39.

Maslen CL, Corson GM, Maddox BK. et al. Partial sequence of a candidate gene for the Marfan syndrome. Nature. 1991;352:334–337. [PubMed: 1852207]

40.

Corson GM, Chalberg SC, Dietz HC. et al. Fibrillin binds calcium and is coded by cDNAs that reveal a multidomain structure and alternatively spliced exons at the 5' end. Genomics. 1993;17:476–484. [PubMed: 7691719]

41.

Pereira L, D'Alessio M, Ramirez F. et al. Genomic organization of the sequence coding for fibrillin, the defective gene product in Marfan syndrome. Hum Mol Genet. 1993;2:961–968. [PubMed: 8364578]

42.

Reinhardt DP, Gambee JE, Ono RN. et al. Initial steps in assembly of microfibrils. Formation of disulfide-cross-linked multimers containing fibrillin-1. J Biol Chem. 2000;275(3):2205–2210. [PubMed: 10636927]

43.

Reinhardt DP, Keene DR, Corson GM. et al. Fibrillin 1: Organization in microfibrils and structural properties. J Mol Biol. 1996;258:104–116. [PubMed: 8613981]

44.

Reinhardt DP, Sasaki T, Dzamba BJ. et al. Fibrillin-1 and fibulin-2 interact and are colocalized in some tissues. J Biol Chem. 1996;271:19489–19496. [PubMed: 8702639]

45.

Robinson PN, Booms P, Katzke S. et al. Mutations of FBN1 and genotype-phenotype correlations in Marfan syndrome and related fibrillinopathies. Hum Mutat. 2002;20(3):153–161. [PubMed: 12203987]

46.

Eldadah ZA, Brenn T, Furthmayr H. et al. Expression of a mutant human fibrillin allele upon a normal human or murine genetic background recapitulates a Marfan cellular phenotype. J Clin Invest. 1995;95:874–880. [PMC free article: PMC295574] [PubMed: 7860770]

47.

Downing AK, Knott V, Werner JM. et al. Solution structure of a pair of calcium-binding epidermal growth factor-like domains: implications for the Marfan syndrome and other genetic disorders. Cell. 1996;85:597–605. [PubMed: 8653794]

48.

Yuan X, Downing AK, Knott V. et al. Solution structure of the transforming growth factor β-binding protein-like module, a domain associated with matrix fibrils. EMBO J. 1997;16(22):6659–6666. [PMC free article: PMC1170270] [PubMed: 9362480]

49.

Kanwar YS, Ota K, Yang Q. et al. Isolation of rat fibrillin-1 cDNA and its relevance in metanephric development. Am J Physiol. 1998;275(5):F710–F723. [PubMed: 9815129]

50.

Biery NJ, Eldadah ZA, Moore CS. et al. Revised genomic organization of FBN1 and significance for regulated gene expression. Genomics. 1999;56(1):70–77. [PubMed: 10036187]

51.

Yang Q, Ota K, Tian Y. et al. Cloning of rat fibrillin-2 cDNA and its role in branching morphogenesis of embryonic lung. Dev Biol. 1999;212(1):229–242. [PubMed: 10419698]

52.

Trask TM, Ritty TM, Broekelmann T. et al. N-terminal domains of fibrillin 1 and fibrillin 2 direct the formation of homodimers: A possible first step in microfibril assembly. Biochem J. 1999;340:693–701. [PMC free article: PMC1220300] [PubMed: 10359653]

53.

Ashworth JL, Kelly V, Wilson R. et al. Fibrillin assembly: Dimer formation mediated by amino-terminal sequences. J Cell Sci. 1999;112:3549–3558. [PubMed: 10504303]

54.

Turano C, Coppari S, Altieri F. et al. Proteins of the PDI family: Unpredicted nonER locations and functions. J Cell Physiol. 2002;193(2):154–163. [PubMed: 12384992]

55.

Hanssen E, Franc S, Garrone R. Atomic force microscopy and modeling of natural elastic fibrillin polymers. Biol Cell. 1998;90(3):223–228. [PubMed: 9726120]

56.

Baldock C, Koster AJ, Ziese U. et al. The supramolecular organization of fibrillin-rich microfibrils. J Cell Biol. 2001;152(5):1045–1056. [PMC free article: PMC2198817] [PubMed: 11238459]

57.

Bowness JM, Tarr AH. Epsilon(gamma-Glutamyl)lysine crosslinks are concentrated in a noncollagenous microfibrillar fraction of cartilage. Biochem Cell Biol. 1997;75(1):89–91. [PubMed: 9192078]

58.

Qian RQ, Glanville RW. Alignment of fibrillin molecules in elastic microfibrils is defined by transglutaminase-derived cross-links. Biochemistry. 1997;36:15841–15847. [PubMed: 9398316]

59.

Thurmond FA, Koob TJ, Bowness JM. et al. Partial biochemical and immunologic characterization of fibrillin microfibrils from sea cucumber dermis. Connect Tissue Res. 1997;36(3):211–222. [PubMed: 9512889]

60.

Lorand L, Graham RM. Transglutaminases: Crosslinking enzymes with pleiotropic functions. Nat Rev Mol Cell Biol. 2003;4(2):140–156. [PubMed: 12563291]

61.

Raghunath M, Cankay R, Kubitscheck U. et al. Transglutaminase activity in the eye: Cross-linking in epithelia and connective tissue structures. Invest Ophthalmol Vis Sci. 1999;40(12):2780–2787. [PubMed: 10549636]

62.

Thurmond FA, Trotter JA. Morphology and biomechanics of the microfibrillar network of sea cucumber dermis. J Exp Biol. 1996;199:1817–1828. [PubMed: 9319729]

63.

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: 7961786]

64.

Henderson M, Polewski R, Fanning JC. et al. Microfibril-associated glycoprotein-1 (MAGP-1) is specifically located on the beads of the beaded-filament structure for fibrillin-containing microfibrils as visualized by the rotary shadowing technique. J Histochem Cytochem. 1996;44(12):1389–1397. [PubMed: 8985131]

65.

Kielty CM, Sherratt MJ, Shuttleworth CA. Elastic fibres. J Cell Sci. 2002;115(Pt 14):2817–2828. [PubMed: 12082143]

66.

Isogai Z, Aspberg A, Keene DR. et al. Versican interacts with fibrillin-1 and links extracellular microfibrils to other connective tissue networks. J Biol Chem. 2002;277(6):4565–4572. [PubMed: 11726670]

67.

Isogai Z, Ono RN, Ushiro S. et al. Latent transforming growth factor β-binding protein 1 interacts with fibrillin and is a microfibril-associated protein. J Biol Chem. 2003;278(4):2750–2757. [PubMed: 12429738]

68.

Gibson MA, Finnis ML, Kumaratilake JS. et al. Microfibril-associated glycoprotein-2 (MAGP-2) is specifically associated with fibrillin-containing microfibrils but exhibits more restricted patterns of tissue localization and developmental expression than its structural relative MAGP-1. J Histochem Cytochem. 1998;46(8):871–886. [PubMed: 9671438]

69.

Gibson MA, Cleary EG. The immunohistochemical localization of microfibril- associated glycoprotein (MAGP) in elastic and nonelastic tissues. Immunol Cell Biol. 1987;65:345–356. [PubMed: 3315986]

70.

Kumaratilake JS, Gibson MA, Fanning JC. et al. The tissue distribution of microfibrils reacting with a monospecific antibody to MAGP, the major glycoprotein antigen of elastin-associated microfibrils. Eur J Cell Biol. 1989;50(1):117–127. [PubMed: 2693088]

71.

Jensen SA, Reinhardt DP, Gibson MA. et al. MAGP-1, Protein interaction studies with tropoelastin and fibrillin-1. J Biol Chem. 2001;276(43):39661–39666. [PubMed: 11481325]

72.

Trask BC, Trask TM, Broekelmann T. et al. The microfibrillar proteins MAGP-1 and fibrillin-1 form a ternary complex with the chondroitin sulfate proteoglycan decorin. Mol Biol Cell. 2000;11(5):1499–1507. [PMC free article: PMC14862] [PubMed: 10793130]

73.

Mecham RP, Davis E. Elastic fiber structure and assemblyIn: Yurchenco PD, ed.Extracellular Matrix Assembly and StructureNew York:Academic Press,1994281–314.

74.

Trask TM, Crippes Trask B, Ritty TM. et al. Interaction of tropoelastin with the amino-terminal domains of fibrillin-1 and fibrillin-2 suggests a role for the fibrillins in elastic fiber assembly. J Biol Chem. 2000;275(32):24400–24406. [PubMed: 10825173]

75.

Brown-Augsburger P, Broekelmann T, Rosenbloom J. et al. Functional domains on elastin and microfibril-associated glycoprotein involved in elastic fibre assembly. Biochem J. 1996;318:149–155. [PMC free article: PMC1217601] [PubMed: 8761465]

76.

Tiedemann K, Bätge B, Müller PK. et al. Interactions of fibrillin-1 with heparin/heparan sulfate: Implications for microfibrillar assembly. J Biol Chem. 2001;276(38):36035–36042. [PubMed: 11461921]

77.

Pfaff M, Reinhardt DP, Sakai LY. et al. Cell adhesion and integrin binding to recombinant human fibrillin-1. FEBS Lett. 1996;384:247–250. [PubMed: 8617364]

78.

Sakamoto H, Broekelmann T, Cheresh DA. et al. Cell-type specific recognition of RGD- and nonRGD-containing cell binding domains in fibrillin-1. J Biol Chem. 1996;271:4916–4922. [PubMed: 8617764]

79.

D'Arrigo C, Burl S, Withers AP. et al. TGF-beta1 binding protein-like modules of fibrillin-1 and -2 mediate integrin-dependent cell adhesion. Connect Tissue Res. 1998;37(12):29–51. [PubMed: 9643645]

80.

Schwarzbauer JE, Sechler JL. Fibronectin fibrillogenesis: A paradigm for extracellular matrix assembly. Curr Opin Cell Biol. 1999;11(5):622–627. [PubMed: 10508649]

81.

Sakai LY, Keene DR, Glanville RW. et al. Purification and partial characterization of fibrillin, a cysteine-rich structural component of connective tissue microfibrils. J Biol Chem. 1991;266:14763–14770. [PubMed: 1860873]

82.

Lin G, Tiedemann K, Vollbrandt T. et al. Homo- and heterotypic fibrillin-1 and -2 interactions constitute the basis for the assembly of microfibrils. J Biol Chem. 2002;277(52):50795–50804. [PubMed: 12399449]

83.

Charbonneau NL, Dzamba BJ, Ono RN. et al. Fibrillins can coassemble in fibrils, but fibrillin fibril composition displays cell-specific differences. J Biol Chem. 2003;278(4):2740–2749. [PubMed: 12429739]

84.

Siracusa LD, McGrath R, Ma Q. et al. A tandem duplication within the fibrillin 1 gene is associated with the mouse tight skin mutation. Genome Res. 1996;6:300–313. [PubMed: 8723723]