Fibronectin, an Extracellular Adhesion Molecule

Michael Ward '99 and David Marcey


I. Background

Fibronectin (FN) is involved in many cellular processes, including tissue repair, embryogenesis, blood clotting, and cell migration/adhesion. Fibronectin exists in two main forms: 1) as an insoluble glycoprotein dimer that serves as a linker in the ECM (extracellular matrix), and; 2) as a soluble disulphide linked dimer found in the plasma (plasma FN). The plasma form is synthesized by hepatocytes, and the ECM form is made by fibroblasts, chondrocytes, endothelial cells, macrophages, as well as certain epithelial cells.

Fibronectin sometimes serves as a general cell adhesion molecule by anchoring cells to collagen or proteoglycan substrates. FN also can serve to organize cellular interaction with the ECM by binding to different components of the extracellular matrix and to membrane-bound FN receptors on cell surfaces. The importance of fibronectin in cell migration events during embryogenesis has been documented in several contexts, e.g.: 1) mesodermal cell migration during gastrulation can be blocked by injection of Arg-Gly-Asp (RGD) tripeptides that block cellular FN receptors (integrins); 2) injection of anti-FN antibodies into chick embryos blocks migration of precardiac cells to the embryonic midline, and; 3) the patterns of FN deposition in developing vertebrate limbs determines the patterns of precartilage cell adhesion to the ECM, thereby specifying limb-specific patterns of chondrogenesis. 

II. Introduction to Fibronectin Structure

Fibronectin's structure is rod-like, composed of three different types of homologous, repeating modules, Types I, II, and III. These modules, though all part of the same amino acid chain, can be envisioned as "beads on a string," each one joined to its neighbors by short linkers.

Below is a schematic of the FN protein, showing the repeated arrangement of the three module types, as well as key binding sites. Twelve type I modules make up the amino-terminal and carboxy-terminal region of the molecule, and are involved mainly in fibrin and collagen binding. Only two type II modules are found in FN. They are instrumental in binding collagen. The most abundant module in fibronectin is Type III, which contains the RGD FN receptor recognition sequence along with binding sites for other integrins and heparin. Depending on the tissue type and/or cellular conditions, the fibronectin molecule is made up of 15-17 type III modules. In addition, there is a module that does not fall into any of these catagories, called IIICS. This module, along with EDB and EDA (both type III modules), is regulated through alternative splicing of FN pre-mRNA. Fibronectin molecules can form two disulphide bridges at their carboxy-termini, producing a covalently-linked dimer.

The FN fragment shown at left is composed of four Type III modules, numbers 7-10. Note the integrin binding tripeptide Arg-Gly-Asp (RGD). Also note the four amino acids in the ninth Type III module, which assist in the binding of fibronectin to the integrin receptor. 

III. Type I Module Structure

The Type I module of fibronectin (F1) is made up of ~45 amino acids, and is found in the amino-terminal and carboxy terminal regions of the full-length protein. The segment of fibronectin at left is the 4-5 F1 module pair. Each F1 module, like all FN modules, is constructed of antiparallel beta sheets < >.

Two antiparallel beta strands make up the top sheet , which folds over a bottom sheet , composed of three antiparallel beta strands < >. These sheets interact through hydrophobic bonding that considerably stabilizes the module.  One such interaction can be visualized in the hydrophobic stacking of two highly conserved aromatic residues on opposite sheets < >.

In addition to hydrophobic interactions between beta sheets, each F1 module is stabilized by two disulfide bridges , one between opposite sheets and one between beta strands of the same sheet < >">>.

Interactions between adjacent modules is important to fibronectin structure. In the F1 case shown, the linker between the fourth and fifth module pair has hydrophobic interactions with the turns between strand B and C and the turns between D and E of 5F1 < >. Such interactions prevent rotation between these F1 modules. Although neighboring modules always interact, the interactions are not always identical to the example just illustrated (4-5F1).

The first five F1 modules (1-5F1) of the N-terminal domain are important in ECM assembly (deletion of any of these modules prevents fibronectin from contributing to the ECM. The 4-5 F1 modules, along with the 10-12 F1 modules, are involved in fibrin binding, an important event during the formation of blood clots. Finally, F1 modules that surround the two F2 modules assist in collagen binding. 

IV. Type II Module Structure

The Type II module (F2) shown at left is a type II module from the PDC109b protein. Like other FN modules, the F2 module is found in a wide array of proteins, and is highly conserved. So, though the crystal structure of the FN2 module of fibronectin has not yet been determined, it is likely that it will be very similar to the FN2 module of PDC109b. The two FN2 modules of fibronectin, along with surrounding F1 modules, are exclusively involved in collagen binding. ~60 amino acids in length, the F2 module includes four beta strands in its core < >. Two antiparallel beta strands make up the top sheet , which folds perpendicularly over the two antiparallel beta strands of the bottom sheet>. Again, the folding of these beta sheets is stabilized by two disulfide linkages>. However, note that in this case the disulphide linkages are not between beta strands. A cavity in the side of the module made up of aromatic and hydrophobic residues < >, in conjunction with a charged serine>, is thought to bind specific leucine and/or isoleucine sequences in collagen. 

V. Type III Module Structure

At left is a segment of fibronectin that contains four type III modules, 7F3, 8F3, 9F3, and 10 F3. Perhaps the best characterized and most studied, type III modules make up a large part of the fibronectin protein, each module being ~ 90 amino acids in length. In fact, over 67 proteins have been identified as containing F3 modules. Like other fibronectin modules, type III module cores are made up of overlapping beta sheets < >. In F3 modules, the top sheet contains four antiparallel beta strands and the bottom sheet is three-stranded >. Unlike F2 or F1 modules, disulphide bridges do not stabilize F3 structure. Instead, this occurs solely through hydrophobic interactions in the module core. The 10 F3 module contains the Arg-Gly-Asp (RGD) the FN receptor (integrin) binding motif < >. This tripeptide protrudes from the protein and is exposed to solution. The 10 F3 module alone does not bind to the integrin receptor with the same affinity as intact fibronectin. This observation is due to the fact that a synergistic site on the 9F3 module participates in RGD binding to integrin < >. It is thought that similar sites exist in 8F3. It has recently been discovered that fibronectin can bind to another member of the integrin family, aIIb-B3 integrin, through a solvent exposed loop in 9F3 > .

In addition to binding various forms of integrins, F3 modules can participate in heparin binding. One of the most intriguing features of F3 modules in fibronectin is that their structure is modulated through alternative pre-mRNA splicing. Whereas F1 and F2 modules are encoded by one exon each, two exons encode most F3 modules. Through the alternative splicing of certain type III exons, multiple fibronectin mRNAs can arise from a single gene. Splicing occurs primarily in three regions: EDB (also called EIIIB or EDII)), EDA (also called EIIIA or EDI), and IIICS (also called V or variable). EDB and EDA can each be spliced out through exon skipping, while IIICS is spliced through a more complex mechanism, which is species and tissue specific. In humans there are potentially 20 different forms of fibronectin, most arising from alternative splicing of the IIICS module. The IIICS module encodes two integrin binding sites, and is involved in heparin binding, so by changing the structure of this molecule through alternative splicing, the function of fibronectin can be changed. A good example of how fibronectin is regulated by alternative splicing is found during embryonic chondrogenesis, or the formation of cartilage during development. In pre-cartilagenous mesenchymal cells, fibronectin containing the EIIIA exon is expressed. However, as these cells differentiate into cartilage, EIIIA is removed through mRNA splicing. Only minus-EIIIA fibronectin is found in mature chondrocytes. 

VI. References

Baron M. Main AL. Driscoll PC. Mardon HJ. Boyd J. Campbell ID.  1992. 1H NMR Assignment and Secondary Structure of the Cell Adhesion Type III Module of Fibronectin.  31: 2068-2073.

Baron M. Norman D. Willis A. Campbell ID.  1990.  Structure of the Fibronectin Type I Module.  Nature. 345: 642-646.

Constantine KL. Ramesh V. Banai L. Trexler M. Patthy L. Llinas M.  1990.  Sequence-Specific 1H NMR Assignments and Structural Characterization of Bovine Seminal Fluid Protein PDC-109 Domain b.  Biochemistry.30: 1663-1672.

Downie SA. Newman SA.  1995.  Different Roles for Fibronectin in the Generation of Fore and Hind Limb Precartilage Condensations. Develop. Bio. 172: 519-530.

Ffrench-Constant C.  1995.  Alternative Splicing of Fibronectin--Many Different Proteins but Few Different Functions.  Exp. Cell. Res. 221: 261-271.

Gehris AL. Oberlender SA. Shepley KJ. Tuan RS. Bennett VD.  1996.  Fibronectin mRNA Alternative Splicing is Temporally and Spatially Regulated During Chondrogenesis in Vivo and in Vitro.  Develop. Dyn. 206: 219-230.

Main AL. Harve TS. Baron JB. Campbell ID.  1992.  The Three-Dimensional Structure of the Tenth Type III Module of Fibronectin:  An Insight into RGD-Mediated Interactions.  Cell.  71: 671-678.

Potts JR and Campbell ID.  1994.  Fibronectin Structure and Assembly.  Curr. Cell Bio.  6: 648-655.

Potts JR and Campbell ID.  1996.  Structure and Function of Fibronectin Modules.  Matrix Bio.  15: 313-320. 

Back to Molecule Index