Patent Grant
7544780
1. Field of the Invention
The invention relates generally to protein molecular biology and microbiology. More specifically, the present invention relates to recombinant prokaryotic collagen-like or triple helical proteins and methods of producing collagen-like materials in prokaryotic organisms.
2. Description of the Related Art
Collagens are abundant extracellular matrix proteins that are essential structural elements of connective tissues in human and animals. The vertebrate collagens are classified into 19 types that are grouped in two major categories known as fibrillar, i.e., types I-III, V, and XI and nonfibrillar, i.e., types IV, VI-X, and XII-XIX collagens. In addition, collagens can alter cell function by interacting with specific cellular receptors (1). The repeating sequence Gly-Xaa-Yaa (GXY), in which a sterically small glycine residue occupies every third position, because only glycine is small enough to be accommodated at the center of the triple helix and in which X is often occupied by proline and Y by hydroxyproline, is a unique feature of the collagen polypeptide (2-4). Proline hydroxylation together with glycosylation of the polypeptide, are essential factors in stabilizing the collagen triple helix and the formation of collagen networks, respectively.
Long tracks of repeated GXY sequences fold into left-handed polyproline type II-like chains and three such chains cooperatively twist around a central axis to form a right-handed rope-like superhelix (2,5-7). The three chains are linked by hydrogen bonds between the backbone group, —NH (Gly) and the backbone carbonyl group of residue X of another chain, and by water-mediated interactions. In humans, mutations that affect collagen triple helix formation and fiber assembly have serious pathological consequences often leading to death.
In some collagen types, the C-terminal and N-terminal propeptides are removed during secretion from cells (7-9). The resulting mature collagen molecules are deposited in the extracellular matrix in the form of fibers, networks and beaded filaments (8). One group of mammalian proteins has collageneous subdomains, but they are not conventional collagens. This group includes several proteins that fulfill rudimentary host defense functions, including complement factor C1q (10) and some mammalian lectins (11). These proteins form characteristic lollipop-like structures with stalks made from their collagenous domains and globular heads made from the non-collagenous regions.
Collagen-like molecules also have been found in lower eukaryotes, such as mussels, worms, and sponges (12), and collagen-like sequences have been deduced from analyses of the genomes of prokaryotes (13-16). Moreover, DNA tracks encoding collagen-like sequences have been found in the genomes of bacteria and phages. However, these organisms appear to lack proline hydroxylases and since hydroxyproline in the past has been considered a critical residue for triple helix formation, it is unclear if the prokaryotic GXY repeated motifs result in proteins that can form stable collagen-like triple helices. Recent studies of model synthetic peptides demonstrate that a GXY sequence with certain so called “guest” residues other than proline and hydroxyproline in the X and Y positions are capable of forming a stable triple helix. Furthermore, type I collagen expressed in tobacco plants, although virtually unhydroxylated, does form a triple helix.
Collagens can act as cell adhesion substrates, organize the cytoskeleton and promote cellular contractility and motility by their ability to interact with integrins and cellular adhesion receptors (17-19). Integrins are large glycoproteins and are expressed as αβ heterodimers on the cell surface (17,19-21). There are 14 distinct α subunits and eight β subunits in mammals that combine to form 24 known heterodimers (22-23).
There are four major subunits of integrin that act as collagen receptors, including α1β1 and α2β1, which are the most widely expressed, and α10b1 and α11b1, which are more distinctly distributed (24-26). While α1β1 and α2β1 favor Col IV and Col I, respectively (27-28), each of these heterodimers is known to bind to both types of collagens (30). Even though α1⊕1 and α2β1 integrins interact with several types of collagen proteins, they appear to possess distinct recognition abilities. For example, α1β1 integrin can bind to type XIII collagen, whereas α2b1 integrin cannot (30).
The extracellular domains of integrins interact with extracellular matrix (ECM) proteins in a metal ion-dependent manner (31). Recent studies demonstrate that the so called I-domains of the α subunits of α1β1, α2β1, α10β1, and α11β1 integrins mediate the interactions of these ECM receptors with collagens and control cell adhesion activity (32-34). The cytoplasmic segments of integrins interact with elements of the cytoskeleton and the signaling molecules, and can trigger intracellular signaling pathways. For example, integrin ligation induces tyrosine phosphorylation of FAK, PYK2, p72SYK, ILK-1, CAS, paxillin, SRC/FYN, and Shc (35-40). Furthermore, signaling events mediated by these molecules are important in an array of biological processes, including cell-migration, cell proliferation and differentiation, angiogenesis, and cancer cell metastasis (35,38,40).
Denatured collagen, gelatin, is widely used in the cosmetic and pharmacological industries, for example, as a pill coating or as a stabilizer. Collagen is usually obtained from bovine skin or other animal products. Unfortunately, these animal protein products can be contaminated by viruses and prions, such as occurs in mad cow disease. Mammalian collagens have been shown to induce autoimmune diseases in animal models. An artificial collagen product would be a desirable alternative to animal based collagen products.
The Streptococcal collagen-like proteins, Scl1 and Scl2, also known as SclA and SClB, are the best-characterized members of the prokaryotic family of collagen-like proteins (41-45). The two related proteins contain long segments of repeated GXY sequences and are located on the cell surface of the human bacterial pathogen, Streptococcus pyogenes or a group A Streptococcus (GAS). The Scl1 and Scl2 proteins have a similar primary structure, which allows for the assignment of four common domains (42). The amino terminal signal sequence and carboxyl terminal cell-wall associated regions are conserved between Scl1 and Scl2, whereas the variable (V) and the collagen-like (CL) regions differ significantly in length and primary sequence. In addition, Scl1, but not Scl2, contains a linker (L) region between the collagen-like and the cell-wall regions, which is composed of highly conserved tandem repeats. This model was recently supported by an extensive genome-based sequence analysis that further established the presence of putative collagen-like domains in prokaryotic proteins (16).
Group A Streptococcus (GAS) are extracellular pathogens that can attach to and invade various human cell types using cellular receptors such as CD44, CD46, and integrins (46-50). Productive adherence is the first step required for pathogenic bacteria to colonize, invade, divide, and secrete virulence factors (51). Integrins are normally located on the basal side of polarized cells and, therefore, may not be immediately accessible for the interactions. It has been postulated that local trauma is required for streptococci to invade deeper anatomical sites. A few distinct routes, through which S. pyogenes emigrate into the underlying tissues, have been identified.
One model proposes that bacterial invasion is dependent on the presence of M1 protein on the cell surface. In this mechanism, streptococci efficiently invaded HeLa (epithelial) cells by a zipper-like mechanism mediated by host cell microvilli and the resulting endocytosis was accompanied by actin polymerization (52). In a paracellular model, the M3-serotype GAS strain producing hyaluronic acid (HA) capsule interacts with CD44 to promote the formation of lamellipodia in keratinocytes in a Rac-1-dependent manner. This event also disrupts intercellular cell-cell adhesion junctions, thereby allowing the pathogen to emigrate onto the baso-lateral surface of cultured keratinocytes (50).
More recently, a study demonstrated that human umbilical vein endothelial cells (HUVECs) directly uptake GAS strain expressing surface protein Sfb1 through “caveolae” (53). A primary component of these caveolae is a membrane bound protein caveolin-1, whose functions are dependent on sphingolipids and cholesterol (54-55). Pre-treatment of HUVECs with methyl-β-cyclodextrin and filipin, drugs that disrupt caveolae and membrane-microdomain by removing lipid moieties, abolished invasion of HUVECs by GAS (53). It was previously shown that integrins associate with caveolin-1 (56-58).
Three other streptococcal proteins have been reported to interact with various integrins. For example, a secreted cysteine protease (SpeB) variant that contains an RGD sequence motif and is expressed by the serotype M1 strains that cause invasive disease was shown to bind integrins αvβ3 and αIIbβ3 (48,59). In addition, two FN-binding streptococcal cell surface proteins, SfbI/F1 and M1, were shown to bind to α5β1 integrin via FN (52, 60-61).
Streptococcal cell wall structures that molecularly mimic components of the human body have long been postulated to be a factor in postinfectious autoimmune disease such as rheumatic fever and poststreptococcal glomerulonephritis. In addition, microbial infections are presumed to play a triggering role in several other autoimmune diseases. Interestingly, autoimmune diseases are often associated with elevated levels of anti-collagen antibodies present in patients' sera. Furthermore, autoimmune diseases can be induced in experimental animals treated with collagen. The discovery that Scl proteins have structural similarity to collagens and during infection could induce antibodies cross-reacting with host collagens adds another dimension to GAS-induced autoimmunity.
The inventors have recognized a need in the art for improvements in methods of producing a new source of collagen and in methods of employing prokaryotic GXY sequences in function- and structure-related studies. Specifically, the prior art is deficient in prokaryotic-like collagens and in determining their role in employing host cell-specific receptors, e.g., the collagen-binding integrins. The present invention fulfills this long-standing need and desire in the art.
The present invention is directed to a recombinant triple helical protein. This recombinant protein comprises a prokaryotic protein or one or more domains of a prokaryotic protein comprising a collagen-like peptide sequence of repeated Gly-Xaa-Yaa triplets having non-modified amino acids. This recombinant triple helical protein may further comprise one or more domains of a biologically active mammalian protein or peptide fragment therefrom.
The present invention is directed to a related recombinant collagen-like protein. The recombinant collagen-like protein comprises one or more domains having a peptide sequence of one or more of SEQ ID NOS: 1-42.
The present invention is directed also to a chimeric collagen-like protein. The collagen-like protein comprises one or more domains having a peptide sequence of one or more of SEQ ID NOS: 1-42 and one or more domains of a human collagen protein.
The present invention also is directed to an expression vector comprising a DNA encoding the recombinant triple helical protein. The present invention also is directed to a related expression vector comprising a DNA encoding the recombinant collagen-like protein.
The present invention also is directed to a host cell comprising an expression vector described herein.
The present invention is directed further to a method of producing a triple helical protein. The method comprises introducing into a prokaryotic host cell an expression vector comprising a DNA sequence encoding the triple helical protein described herein, culturing the host cell under conditions suitable to express the protein and isolating the expressed protein to produce a triple helical protein. Additionally, the method may comprise purifying the triple helical protein. Furthermore, the method may comprise heating the triple helical protein and producing a gelatin-like material therefrom. In a related method the expression vector comprises DNA encoding a triple helical protein having one or more domains having a peptide sequence of one or more of SEQ ID NOS: 1-42.
The present invention is directed further to another related method of producing a collagen-like protein having one or more domains having a peptide sequence of one or more of SEQ ID NOS: 1-42. The method comprises those steps described supra for a triple helical protein.
The present invention is directed further still to a method of producing a gelatin-like material. The method comprises introducing into a prokaryotic host cell an expression vector comprising a DNA sequence encoding a protein having one or more domains having a peptide sequence of one or more of SEQ ID NOS: 1-42, and culturing the prokaryotic host cell under conditions suitable to express the protein. The expressed protein is heated to produce the gelatin-like material.
The present invention is directed further yet to an antibody directed against a recombinant streptococcal collagen-like (rScl) protein. The rScl protein binds to α2β1 integrin. Also, the present invention is directed to a related antibody directed against p176 which binds α2β1 integrin. The present invention is directed further to another related antibody directed against a domain in p176 that binds an α2-I domain of α2β1 integrin.
The present invention is directed further still to a method for screening for compounds that inhibit streptococcal collagen-like (Scl) protein binding to or interaction with an integrin. The method comprises measuring Scl binding to or interaction with the integrin in the presence of a test compound and measuring Scl binding to or interaction with the integrin in the absence of the test compound in a control sample. Binding of the integrin to Scl in the presence of the test compound is compared with binding in the control sample. A reduction in binding of the integrin to Scl in the presence of the test compound compared to control correlates with inhibition of integrin binding to or interaction with Scl by the test compound.
Other and further aspects, features, and advantages of the present invention will be apparent from the following description of the presently preferred embodiments of the invention. These embodiments are given for the purpose of disclosure.
In one embodiment of the present invention there is provided a recombinant triple helical protein comprising a prokaryotic protein or one or more domains of a prokaryotic protein comprising a collagen-like peptide sequence of repeated Gly-Xaa-Yaa triplets having non-modified amino acids. Further to this embodiment the recombinant triple helical protein may comprise one or more biologically domains of a mammalian collagen or peptide therefrom. An example of a mammalian collagen is human collagen. In an aspect of this further embodiment, the prokaryotic collagen-like protein or one or more collagen-like domains thereof are fused to the one or more mammalian collagen domains.
In addition, in this embodiment the recombinant triple helical protein may comprise a peptide designed to add or improve a biological function of the recombinant triple-helical protein without disrupting the triple helical structure thereof. In one aspect the peptide comprises an amino acid sequence comprising at least one mutation in one or more of SEQ ID NOS: 1-42 or a peptide fragment thereof. In another aspect the peptide comprises a peptide fragment thereof or comprising an amino acid sequence generated via computer simulation.
In yet other aspects of this embodiment the collagen-like peptide sequence is from one or more streptococcal collagen-like proteins. The streptococcal collagen-like protein(s) may be from a Group A Streptococcus. A representative example of a Group A Streptococcus is Streptococcus pyogenes. Additionally, in this aspect the streptococcal collagen-like proteins may be a variant of S. pyogenes Scl1 or of S. pyogenes Scl2 or a combination thereof. Further to this aspect the peptide sequence comprises one or more of SEQ ID NOS: 1-42.
In a related embodiment there is provided a recombinant collagen-like protein comprising one or more domains having a peptide sequence of one or more of SEQ ID NOS: 1-42. Further to this related embodiment the recombinant triple-helical protein further may comprise one or more biologically active domains of a mammalian collagen. An example is human collagen. Additionally, in an aspect of this related embodiment, the one or more of SEQ ID NOS: 1-42 may fused to the one or more human collagen domains. Again further to this embodiment the recombinant collagen-like protein may comprise a synthetic peptide as described supra.
In another related embodiment, there is provided a chimeric collagen-like protein comprising one or more domains having a peptide sequence of one or more of SEQ ID NOS: 1-42; and one or more biologically active domains of a mammalian collagen or peptide therefrom. The mammalian collagen may be human collagen. The collagen-like protein may be a fusion protein or a synthetic protein. Further to this embodiment the chimeric collagen-like protein may comprise a peptide as described supra.
In another embodiment of the present invention there is provided an expression vector comprising a DNA sequence encoding the recombinant triple helical protein described supra. In a related embodiment there is provided an expression vector comprising a DNA sequence encoding the collagen-like protein described supra.
In yet another embodiment of the present invention there is provided a host cell comprising and expressing an expression vector having a DNA sequence encoding the triple helical protein described supra. In a related embodiment there is provided an expression vector having a DNA sequence encoding the collagen-like protein described supra. In both these related embodiments the host cell may be a prokaryotic cell. Representative examples of useful prokaryotic cells are well known in the art and include Escherichia coli, Streptococcus and Bacillus.
In yet another embodiment of the present invention there is provided a method of producing a triple helical protein comprising introducing into a prokaryotic host cell an expression vector comprising a DNA sequence encoding the triple helical protein described supra; culturing the prokaryotic host cell under conditions suitable to express the protein; and isolating the expressed protein thereby producing the triple helical protein.
In an aspect of this embodiment the method may further comprise purifying the triple helical protein. In another aspect the method may further comprise heating the triple helical protein and producing a gelatin-like material therefrom. Representative examples of useful prokaryotic cells are well known in the art and include Escherichia coli, Streptococcus and Bacillus.
In a related embodiment there is provided a method of producing a triple helical protein comprising introducing into a prokaryotic cell an expression vector comprising a DNA sequence encoding a protein comprising one or more domains having a peptide sequence of one or more of SEQ ID NOS: 1-42; culturing the prokaryotic host cell under conditions suitable to express the protein; and isolating said expressed protein thereby producing the triple helical protein. In this embodiment the method may further comprise the purifying and heating steps as described supra. Representative examples of useful prokaryotic cells are well known in the art and include Escherichia coli, Streptococcus and Bacillus.
In another related embodiment there is provided a method of producing a collagen-like protein comprising introducing into a prokaryotic host cell an expression vector comprising a DNA sequence encoding the collagen-like protein comprising one or more domains having a peptide sequence of one or more of SEQ ID NOS: 1-42; culturing the prokaryotic host cell under conditions suitable to express the protein; and isolating the expressed protein thereby producing the collagen-like protein. In this embodiment the method may further comprise purifying and heating steps as described supra. Additionally, the collagen-like proteins and the prokaryotic host cell are as described supra.
In still another embodiment of the present invention there is provided a method of producing a gelatin-like material, comprising introducing into a prokaryotic host cell an expression vector comprising a DNA sequence encoding a protein comprising one or more domains having a peptide sequence of one or more of SEQ ID NOS: 1-42; culturing the prokaryotic host cell under conditions suitable to express the protein; and heating the expressed protein thereby producing the gelatin-like material. Representative examples of useful prokaryotic cells are well known in the art and include Escherichia coli, Streptococcus and Bacillus.
In still another embodiment of the present invention there is provided an antibody directed against a recombinant streptococcal collagen-like (rScl) protein. In an aspect of this embodiment the rScl protein may bind to α2β1 integrin. More specifically, in this aspect the rScl protein binds to an α2-I domain of the α2β1 integrin. An example of such an α2-I domain binding rScl is p176. Thus, in a related aspect of this embodiment there is provided an antibody directed against p176. Furthermore, in another related aspect the antibody may be directed against a domain in p176 that binds an α2-I domain of α2β1 integrin.
In still another embodiment of the present invention there is provided a method of screening for therapeutic compounds that inhibit a Streptococcal collagen-like protein (Scl) binding to or interacting with an integrin. This method generally comprises measuring Scl binding to or interaction with the integrin in the presence of a test compound; measuring Scl binding to or interaction with the integrin in the absence of the test compound in a control sample; comparing binding of the integrin to Scl in the presence of the test compound with binding in the control sample; and correlating a reduction in binding of the integrin to Scl in the presence of the test compound compared to control with inhibition of integrin binding to or interacting with Scl by the test compound, thereby screening for the therapeutic compound.
In an aspect of this embodiment the streptococcal collagen-like protein may be P176. In another aspect the Scl protein may bind to or interact with an α2β1 integrin. Further to this aspect the Scl protein may bind to or interact with an α2-I domain of said α2β1 integrin.
The following abbreviations are used herein: (r)Scl, (recombinant)streptococcal collagen-like; GAS, group A Streptococcus; FAK, focal adhesion kinase; CAS, crk associated substrate; JNK, c-Jun N-terminal kinase; FN, fibronectin; Col I, type I collagen; Col IV, type IV collagen.
Provided herein are prokaryotic triple helical proteins, prokaryotic collagen-like proteins and gelatin-like materials comprising these proteins. Generally, these proteins are recombinant proteins and may comprise a prokaryotic protein or one or more domains of a prokaryotic protein comprising a collagen-like peptide sequence of non-modified evolutionarily conserved Gly-Xaa-Yaa (GXY) sequence repeats that can form structurally conserved collagen-like triple helices, despite the lack of hydroxyprolines. As such, the GXY triplets may comprise a non-modified proline in the X or Y position.
More particularly, the present invention provides recombinant streptococcal collagen-like (Scl) proteins, derived from variants of Scl1 and Scl2, i.e., Scl1.41 and Scl2.28 comprising such non-modified conserved GXY triplet sequences. One of skill in the art will appreciate that several sequence characteristics of Scl1 and Scl2 can promote and stabilize the triple helix folding of the CL regions of the two proteins. For example, the number of continuous GXY repeats is high and often exceeds the number found in collagenous domains of mammalian proteins. Secondly, 30% and 37% of GXY repeats in Scl1.41 and Scl2.28, respectively, contain prolines, primarily at the X position.
In collagens, the cyclic ring of prolines stabilizes each collagen chain by steric repulsion. Without being bound by theory, the proline residues in Scls likely have the same effect. Also, Scl proteins have a high content of charged amino acids, i.e., 32% and 29.5% in Scl1.41 and Scl2.28, respectively, especially lysine and arginine with long ionizable side chains that can directly interact with residues of a neighboring polypeptide chain. Further, Scls contain triplets GPR, GER, GPA, GDR, GKD, and GEK, which stabilize a triple helix. None of the GXY triplets which destabilize the triple helix structure of model synthetic peptides are present in the CL region of the rScl proteins.
Scls also contain triplets that stabilize the triple helix by formation of alternative hydrogen bonding and hydration patterns. For example, the GEK triplet, which was proposed to stabilize the triple helix through formation of indirect water-mediated bridges, is the most frequent triplet in Scl1.41CL. In addition, threonines in the Y position of GET and GKT, found in both Scl1 and Scl2, may substitute for hydroxyprolines in the formation of hydrogen bonds.
Additionally, the lollipop-like structure of the recombinant Scl proteins resembles the observed structural organization of mammalian cell surface proteins with collagenous domains. Scl proteins likely form lollipop-like structures on the surface of streptococcal cells as do the recombinant polypeptides of the present invention. Scl proteins likely interact with molecules in the environment of the streptococci.
The structural organization of the Scl proteins is particularly suited for a ligand binding where the stalk-forming CL region projects the globular V region away from the bacterial surface, facilitating possible interactions between the V regions and their potential targets. The CL regions may themselves interact with various molecules as found with mammalian collagens and proteins with collagenous domains. Furthermore, interactions between the CL and V regions may be required in the efficient assembly of the triple helix.
A list of sequences of collagen-like domains from Scl1 and Scl2 proteins are presented in Table 1. These sequences contain the repeating amino acid sequence Gly-Xaa-Yaa (GXY) in which glycine occupies every third position which is critical to allow the encoded proteins to form a homotrimer. The recombinant collagen-like proteins may comprise one or more domains having a peptide sequence from SEQ ID NOS: 1-42.
The present invention also provides chimeric triple helical proteins or chimeric collagen-like protein. The recombinant triple helical proteins or recombinant collagen-like proteins described herein may additionally comprise at least one biologically active domain from a mammalian collagen, such as, but not limited to human collagens. For example, the protein may comprise one or more of SEQ ID NOS.: 1-42 and one or more domains from a mammalian, e.g., human, collagen protein.
The chimeric protein may be a fusion protein. Methods of constructing vectors suitable to transfect and express DNA encoding a fusion protein are standard in the art and well known to those of ordinary skill in the art. Alternatively, the protein may be a synthetic protein.
It is further contemplated that a peptide may be added to the triple helical recombinant proteins, the recombinant collagen-like proteins or the chimeric collagen-like proteins described herein. Addition of these peptides may alter structure and thereby function of the chimeric protein in a specifically selected manner. For example a new function may be added or an existing function improved. However the triple helical structure inherent in all the proteins described herein is not disrupted by the presence of the this peptide. The peptides may comprise at least one amino acid mutation in at least one of SEQ ID NOS.: 1-42 or in a peptide fragment thereof. Alternatively, the sequence of the peptide may be determined via computer simulation.
As such, the present invention further provides methods of producing a triple helical protein or a collagen-like protein. The methods may utilize expression vectors comprising a DNA that encodes the triple helical or collagen-like proteins of the present invention. Additionally, the present invention provides a host cell, such as a prokaryotic cell, for example, although not limited to, Escherichia coli, a Streptococcus or a Bacillus. It is standard in the art to construct such expression vectors and to transform a host cell to express the product of the DNA contained therein.
Generally, an expression vector as described herein is introduced into the prokaryotic host cell. The prokaryotic cell is cultured under conditions suitable to induce expression of the protein. The expressed protein is isolated and may further be purified. Also provided is a method of producing a gelatin-like material from the recombinant proteins presented herein. The triple helical and/or the collagen-like proteins may be heated to produce the gelatin-like material.
Furthermore, the present invention contemplates that designer collagens or gelatin-like substances may be produced for the specific needs of the user using the recombinant triple helical proteins, collagen-like proteins or chimeric proteins described herein. For example a collagen or gelatin may be designed to effectively function in any biological system. The problems associated with mammalian collagens, such as bovine collagens, can be avoided. This effectively would change and improve the commercial use of collagens.
The present invention further demonstrates that a member of the prokaryotic collagen-like proteins, the Scl1.41 variant of S. pyogenes, interacts with the I-domain of α2β1 integrin, but not α1β1 integrin. This interaction induces cell adhesion and signaling activities through FAK and CAS in human lung fibroblast MRC-5 cells. Thus, the Scl1.41 recombinant protein P17. (SEQ ID NO.: 44) can interact with an integrin in a productive manner, preferably via a binding site in the CL domain shown in SEQ ID NO.: 15.
Furthermore it is contemplated that cell adhesion to P176 is predominantly mediated by integrins. First, removal of Ca2+ and Mg2+ metal ions, which are required for collagen-integrin binding, inhibited cell adhesion activity. Second, the addition of an anti-β1 integrin antibody completely inhibited adhesion of these cells onto P176. Third, cell adhesion to P176 was specifically mediated by the α2β1 integrin, a known type I collagen receptor, because an anti-α2β1 integrin monoclonal antibody blocked cell adhesion, whereas an anti-α1β1 antibody did not. Fourth, the specificity of the α2β1-P176 interaction was further supported by the fact that C2C12-α2+ cells adhered to P176 and an anti-α2β1 integrin antibody or EDTA inhibited this adhesion. Finally, SPR analysis showed that P176 interacts with the I-domain of α2 integrin, which is an important characteristic of a collagen-integrin binding (34).
However, integrin ligation is not a general property of Scl proteins and several other recombinant proteins based on different Scl sequences did not support cell adhesion, such as the Scl2.28 recombinant protein P163, for example. This suggests that specific sequences present in some but not other Scl proteins are recognized by the integrins. It has been demonstrated that the α1β1, and α2β1 integrins interact with specific sequences in mammalian collagens. The substrate specificity of the two integrins appears to be similar, but not identical. Thus, α1β1 but not α2β1, integrin requires a hydroxyproline residue in the binding site for full activity. Consistent with this observation α2β1, but not α1β1, mediates cell adhesion to the P176 protein that does not contain hydroxy-proline residues.
Given that integrins are involved in streptococci host cell invasion and that fibronectin binding a MSCRAMM peptide (SEQ ID NO.: 46) on gram-positive bacteria can recruit soluble fibronectin which is then recognized by α5β1 integrin thereby initiating cellular invasion, it could be contemplated that integrin interaction through a Scl protein would play a role in the infections process of S. pyogenes. Additionally, the α2β1 integrin is also a prominent collagen receptor on platelets and it is possible that Scl proteins may induce a platelet signaling through interacting with the platelet integrin. The Scl protein-integrin interaction described herein suggest that these bacterial proteins are mimicking and being recognized as collagens when it comes to α2β1 integrin. It is further contemplated that Scl proteins could also behave as collagens in other systems and that this molecular mimicry allows the bacteria to manipulate host biology at a number of different levels.
Thus the present invention provides antibodies directed against integrin-binding or integrin-interacting rScl proteins described herein. Preferably, cell adhesion is mediated by α2β1 integrin. Most preferably, the rScl proteins bind to the I-domains of the integrins, such as to the α2-I domain of α2β1 integrin. The antibody may be directed against P176 or to the domain in P176 that binds α2β1 integrin or, more specifically, binds the α2-I domain of α2β1 integrin. It is contemplated that antibodies may be directed against rScls comprising a P176 α2-I domain binding site or a P176-like CL domain or P176-like α2-I domain binding site.
The present invention further provides means to screen for therapeutic compounds or drugs, including proteins or peptides, that inhibit P176 from binding or interacting with integrin I-domains. Generally, a potential therapeutic compound may be screened by comparing the binding of α2-I to P176 in the presence and absence, as control, of the potential therapeutic compound. Inhibition of α2-I binding in the presence of the test compound would screen for potential therapeutic efficacy.
The following examples are given for the purpose of illustrating various embodiments of the invention and are not meant to limit the present invention in any fashion.
Materials and Methods
Recombinant Proteins
Recombinant proteins were obtained using the Strep-tag II expression and purification system (IBA-GmbH). Fragments encoding sequences of the scl1.1 allele were amplified from the serotype M1 strain MGAS6708 identical to SF370 used for genome sequencing. The scl2.28 sequence was amplified from the serotype M28 strain MGAS6274. DNA fragments were ligated to the Escherichia coli vector pASK-IBA2, and clones were confirmed by sequencing. Recombinant proteins were expressed in E. coli and purified by affinity chromatography on a Strep-Tactin-Sepharose. Thus, pSL163 was constructed which encodes a P163 polypeptide. The identity of each purified protein was confirmed by amino-terminal sequencing, and mass determinations were done by electrospray ionization mass spectrometry measurements.
Western immunoblot analysis was performed with proteins separated on SDS-PAGE, transferred onto a nitrocellulose membrane (Amersham Biosciences) and probed with specific, affinity-purified polyclonal rabbit antibodies raised against a synthetic peptide derived from the V region of Scl1.1. Horseradish peroxidase-conjugated goat anti-rabbit affinity-purified immunoglobulin G, heavy and light chains, (Bio-Rad) was used as the secondary antibody, and SuperSignal chemiluminescent substrate (Pierce) was used in signal detection. Prestained broad-range marker proteins (Bio-Rad) were used as molecular mass standards.
To obtain recombinant P176, the DNA sequence of the scl1.M41 allele (accession number AY452037) was amplified and cloned into an E. coli vector pASK-IBA2, resulting in plasmid pSL176. In a domain swapping experiment, plasmid pSL163 was used as a core sequence. First, a DNA fragment encoding the variable V region of P176 was amplified and substituted for the corresponding region of pSL163, resulting in the pSL181/P181 construct. Next, the pSL182 plasmid was obtained by replacing the DNA sequence of the collagen-like region in pSL181, originally from pSL163, with a corresponding DNA sequence of pSL176, which lacked a linker region. These plasmid constructs were verified by DNA sequencing. As above, recombinant proteins were purified by affinity chromatography on Strep-Tactin-Sepharose and their identity was confirmed by amino-terminal sequencing.
Gel Migration Analyses
The ability of the Scl proteins to form oligomers was examined by gel electrophoresis. Protein samples were heat-denatured at 90° C. for 5 min and then placed on ice for 2 min before loading, whereas native protein samples were not heated and were kept at 4° C. Denatured or native P144 and P163 proteins were electrophoretically separated in 10-20% gradient polyacrylamide gels with or without 0.1% SDS and visualized by staining.
The resistance of triple helix collagen to trypsin digestion was used as an indicator of macromolecular structure. Native and heat-denatured samples of recombinant P144 dissolved in 25 mM HEPES, pH 8.0, were digested with 10 μg/ml trypsin using an enzyme:protein molar ratio of 1:10 at 15° C. Aliquots were removed after 15, 30, and 60 min, and the reaction was stopped with phenylmethylsulfonyl fluoride. Samples were analyzed by SDS-PAGE using undigested P144 and P157 as size markers. Resistance of the triple-helical domain to trypsin was measured from band intensities and compared with an undigested sample.
Circular Dichroism (CD) Spectroscopy
The triple helical conformation of recombinant rScls was analyzed by circular dichroism, as described (62). Spectra were recorded on a Jasco J720 spectropolarimeter equipped with a variable temperature unit. Samples were dissolved in 1% phosphate-buffered saline, pH 7.4. Thermal transition profiles were recorded at 220 nm, as described above, with a temperature slope of 20° C./h or of 10° C./h. Measurements were taken with a 0.5-cm path length and data were integrated for 1 sec at 0.2-nm intervals with a bandwidth of 1 nm. A wavelength scan was performed for each protein before unfolding (4° C., 25° C.) and after unfolding (50° C.), or after subsequent refolding (4° C., renatured). The melting temperatures (tm) were given as mean±S.D. of tm values from several measurements. Secondary structure compositions were estimated using three deconvolution programs, CD Estima, CONTIN, and SELCON, and the results were averaged.
Electron Microscopy
The structural organization of the recombinant proteins was viewed by electron microscopy of the rotary shadowed rScls, as previously (63). Protein samples (100 μg/ml) were dialyzed against 0.1 M ammonium bicarbonate and then mixed with glycerol to a final glycerol concentration of 70% v:v. Samples were sprayed onto freshly cleaved mica sheets and rotary shadowed with carbon/platinum at an angle of 6 degrees. The replicas were backed with carbon at 90 degrees and placed on copper grids. Photomicrographs were taken with a Philips 410 electron microscope operated at 80 KV with a 30-μm objective aperture.
The microscope was calibrated using a carbon grating replica. The height of the goniometer stage was adjusted for each grid square before taking pictures and the lenses were saturated before taking each photo. Molecule contours were traced on a digitizing tablet at a final magnification of ×181,500 and measurements were generated with Bioquant software.
Reagents and Antibodies
Cell culture media, human fibronectin, type I and type IV collagen, control mouse IgG, and anti-α1β1 (TS2/7), anti-α2β1 (P1E6), anti-α3β1 (P1B5), anti-α5β1 (P1D6) integrin antibodies were purchased from Gibco-BRL/Invitrogen (Carlsbad, Calif.). Anti-FAK (2A7), anti-paxillin and anti-phospho-tyrosine (4G10) mouse monoclonal antibodies (mAbs) were purchased from Upstate Biotechnology Inc. (Lake Placid, N.Y.). Anti-phospho-JNK and anti-phospho-FAK (Y397) antibodies were purchased from Cell Signaling Technology (Beverly, Mass.). The anti-αvβ3 (LM609) was obtained from Chemicon (Temecula, Calif.). Anti-β1 (4B4) integrin antibody was procured from BD Biosciences (San Jose, Calif.).
Cells and Cell Culture
Human lung fibroblast cell lines MRC-5 and WI-38 were obtained from ATCC (CCl-171; CCL-75) and cultured in DMEM supplemented with 10% FBS, 2.4 mM L-glutamine and antibiotics. Parental mouse myoblast C2C12 (control cells) and C2C12 cells stably transfected with the human α2-integrin subunit (C2C12-α2+), provided by D. Gullberg (Uppsala, Sweden), were maintained in 10% FBS in DMEM in the absence or presence of 10 mg/ml puromycin, respectively. Methods for growth and maintenance of adherent cells have been described (56,64).
Fluorescent Activated Cell Sorting (FACS)
Cells were washed in PBS, pH 7.4, then non-enzymatically dissociated with 2 mM EDTA in PBS and washed twice in PBS containing Ca2+ and Mg2+. Cells were passed through a cell strainer and enumerated. 2×106 cells were then incubated with saturating amounts of control mouse IgG or anti-human integrin monoclonal antibodies, e.g., anti-α1β1 (TS2/7), anti-α2β1 (P1E6), anti-α3β1 (P1B5), anti-αvβ3 (LM609), anti-α5β1 (P1D6) and anti-β1 (4B4) for 30 minutes at 4° C. Secondary labeling was performed for 1 hour at 4° C. with a goat-anti mouse IgG conjugated to fluorescein isothiocyanate (FITC) (Molecular Probes, Oreg.). Cells were washed twice with PBS, fixed with 0.5% p-formaldehyde (PFA) and analyzed with the FACS Calibur flow cytometer (Becton Dickinson, San Jose, Calif.).
Cell Adhesion and Spreading Assays
Cell adhesion to 24-well plates coated with specified substrates has been described (64). Briefly, the test substrates, i.e., Scl recombinant proteins, were prepared at a concentration of 12.5, 25, 50 and 100 nM in TBS (20 mM Tris, pH.7.4, 150 mM NaCl, 0.02% sodium azide) and coated onto sterile microplates overnight at 4° C. Human fibronectin, type IV collagen, type I collagen and BSA (1.25, 2.5, 5 and 10 μg/ml) were included as controls. BSA was subjected to 0.22 μm filtration and heat inactivation. After washing with PBS, plates were saturated with 0.5% BSA for 60 min in a 37° C. CO2 incubator.
Human lung fibroblast MRC-5 and WI-38 or C2C12 cells starved overnight in serum-free media were detached in PBS containing 2 mM EDTA and 0.0025% trypsin. Cells were collected and resuspended in M199 defined media containing appropriate antibiotics, 0.2% BSA and 1×ITS (insulin, transferrin and selenium-A), supplemented with 2 mM Mg2+ and 1 mM each of Ca2+/Mn2+. Saturated plates were washed once with PBS and 100 μl of cell suspension (0.15×106 cells/ml) were added and allowed to adhere.
After 45 or 90 min, unattached cells were gently removed and washed with PBS before fixing in 3% PFA for 10 min. The cells were then washed with cold TBS, refixed in 20% methanol for 10 min, and stained in 0.5% crystal violet for 5 min. Plates were thoroughly washed with water and air-dried. The dye was then eluted in 100 μl of 100 mM sodium citrate and absorbance was measured at 590 nm.
To visualize cell spreading, PFA-fixed cells were stained with standard eosin and hematoxylin and pictures were captured in an inverted light microscope under 20× magnification. Alternatively, cells grown on coverslips under similar conditions were stained with phalloidin-TRITC (Sigma, P-1951) and 4,6-diamidino-2-phenylindole dihydrochloride (DAPI; Molecular Probes, D-1306) to identify stress fibers and nuclei, respectively. A Zeiss Axiovert-125 fluoroscope was used for this purpose.
Cell Adhesion Blocking Assay
The cell adhesion blocking assay has been described (64). In brief, 70% confluent human lung fibroblast (MRC-5), parental C2C12 and C2C12-2+ cells were washed in PBS, detached non-enzymatically in 2 mM EDTA, and washed. Cells were passed through a cell strainer and enumerated. Approximately 0.1×106 cells were incubated with 1, 5 and 10 mg/ml of anti-a1b1, anti-a2b1, anti-a3b1, anti-a5b1, anti-avb3 and anti-b1 integrin monoclonal antibodies on ice for 30 min. Cells were washed, resuspended in defined media, and seeded onto dishes pre-coated with 100 nM P176 and subjected to cell adhesion assays. The use of single dose (10 mg/ml) anti-integrin monoclonal antibody for C2C12-a2+ cell adhesion blocking assay was considered optimal. Cold (4° C.) PBS, pH 7.4 containing 1 mM of Ca2+ and Mg2+ was used for washing cells. Statistical analyses were performed as described (64).
Recombinant I-domain of α2 Integrin Subunit
Cloning, expression, and purification of the I-domain of α2 integrin have been described previously (65). In brief, a DNA fragment encoding the I-domain of α2 integrin subunit was amplified by PCR from a human hepatoma cDNA library and subcloned into the expression vector pQE30 (Qiagen Inc., Chatsworth, Calif.). The accuracy of the DNA sequence was verified by dideoxy DNA sequencing (Lonestar Lab). Large-scale expression and purification of the recombinant α2-I were carried out using Ni+2-chelating affinity chromatography, as described (62).
Surface Plasmon Resonance (SPR) Analysis
Measurements were performed at room temperature in a BIACORE 3000 instrument (BIAcore, Uppsala, Sweden) as described previously (65). Briefly, 990 response units (RU) of P176Scl and 717 RU of P181Scl proteins were immobilized onto the flow cells on a CM5 chip. Different concentrations of α2-I proteins in HBS (25 mM HEPES, 150 mM NaCl, pH 7.4) buffer containing 5 mM β-mercaptoethanol, 1 mM MgCl2 and 0.05% octyl-D-glucopyranoside were passed over these surfaces at 20 μl/min for 6 min. Regeneration of the Scl protein surface was achieved by running 10 μl of a solution of 0.01% SDS. Binding of α2-I to a reference flow cell, which had been activated and deactivated without the coupling of proteins, was also measured and was subtracted from the binding to Scl protein-coated chips.
SPR sensorgrams from different injections were overlaid using the BIAevaluation software (BIAcore AB). Data from the equilibrium portion of these sensorgrams were used for Scatchard analysis. Based on the correlation between the SPR response and change in soluble I-domain protein binding to the immobilized Scd proteins, values for the binding ratio, nbound, and the concentration of free protein, [P]free, were calculated using the equations described previously (65). Scatchard analysis was performed by plotting nbound/[P]free against nbound, in which the negative reciprocal of the slope is the dissociation constant, KD and the X-intercept is the number of binding interactions, n.
Biochemical Methods
Cells were serum-starved overnight, detached and maintained in suspension (Sus) for 45 minutes at room temperature. Cells were then replated onto dishes coated with indicated substrates. After 30 or 60 min cells were washed with cold PBS, pH 7.4, and solubilized in cell extraction buffer (50 mM HEPES, pH 7.5, 150 mM NaCl, 1% Triton X-100, 10 mM sodium fluoride, 1 mM sodium pyrophosphate, 2 mM sodium orthovanadate and with various protease inhibitors freshly added) for 30 minutes on ice. Extracts were centrifuged at 21,000×g for 30 minutes at 4° C. to remove insoluble material.
Protein concentrations of the resulting lysates were determined by the Bio-Rad DC protein assay. For immunoprecipitation analyses, cell lysates were pre-adsorbed with mouse IgG-agarose beads at 4° C. for 1 hour. Immunoprecipitation, immunoblotting, and detection protocols were performed as described previously (64,66). For immunoprecipitation, 2-3 mg of antibodies were used for each sample. For immunoblotting analyses, antibodies were prepared at a concentration of 0.5-2 mg/ml in 1×TBS, pH 7.4, with 3% BSA.
Characterization of Recombinant Scl Proteins
Previous studies demonstrated that recombinant proteins expressed in eukaryotic systems could successfully be used to study the structure and organization of mammalian proteins with collagenous domains. A series of recombinant polypeptides generated in a prokaryotic (E. coli) expression system derived from either the Scl1.1, i.e., a Scl1 protein from a serotype M1 GAS, or the Scl2.28, i.e., a Scl2 protein from a M28-type GAS, protein that contain collagen-like regions composed of continuous GXY repeats was examined (
These two related proteins were selected for structural studies because they were both expressed by the parental GAS isolates and they differed significantly in their primary amino acid sequence. The collagen-like region (CL) is composed of 50 and 79 GXY repeats in Scl1.1 and Scl2.28, respectively. The amino-terminal segment of each protein, called the variable (V) region, is composed of noncollagenous sequence and consists of 67 amino acids in Scl1.1 and 73 amino residues in Scl2.28.
To examine the structural organization of Scl1, three recombinant proteins, P144, P157, and P158, were constructed representing the combined VCL regions, the CL region, and the V region, respectively (
Furthermore, both P144 and P158 reacted with the V region-specific antibodies, whereas P157 did not (
The recombinant P176 protein has an 84 amino-acid long V region followed by a CL region that includes 62 GXY triplets. It adopted a triple-helical conformation and formed lollipop-like structures when viewed by EM (
Multimolecular Organization of Scl Proteins
To assess the ability of Scl1 and Scl2 to assemble into polymeric structures, the electrophoretic mobilities of P144, P157, and P163 in polyacrylamide gel run in the absence of SDS before and after heat denaturation were compared (
In subsequent experiments, the mobility of native and heat-denatured samples of P144 and P163 were compared in SDS-PAGE using molecular mass standards (
To test the formation of triple helix by Scl proteins, the susceptibility of recombinant P144 to trypsin was examined (
Circular Dichroism (CD) Spectroscopy of Recombinant Scl Proteins
Collagen triple helices have characteristic CD spectra with a positive ellipticity maximum at 220 nm. To examine the secondary structure composition of Scl1 and Scl2, recombinant proteins were analyzed by CD spectroscopy (
The peak of the shoulder, however, was of a negative value probably due to the contribution of the secondary structure of the noncollagenous V domain. Similarly small amplitudes were previously reported for some mammalian proteins with collagenous domains including surfactant proteins and complement component C1q. The CD spectrum of the Scl2.28, recombinant P163, also had the characteristics of a collagen triple helix with ellipticity maximum at 220 nm (˜2×103 degree cm2 dmol1) (
The CD spectrum of P158 (10 μM) did not include the characteristics of a collagen triple helix (
When samples of P144 (8.1 μM), P157 (29.8 μM) and P163 (7.7 μM) were heated to 50 ° C., the CD spectra recorded at this temperature (
The thermal unfolding of the collagen-like triple helix of P144 and P163 appears to be readily reversible. When the protein samples were cooled, the ellipticity at 220 nm gradually increased (
The CD spectra of P176 at 4° C. and 25° C. showed ellipticity maxima at 220 nm, which is consistent with a collagen triple helix-like structure (
Electron Microscopy of Scl Proteins
Electron microscopy of rotary shadowed samples has been used for studying the structural organization of collagens and collagen-like mammalian proteins. Examination of a rotary shadowed preparation of Scl1, e.g., recombinant P144, revealed a two-domain lollipop-like structure (
The measured contour length, i.e., 279 particles measured, of the collagenous tail in P144 is 45.5±4.3 nm. Based on a translation of 0.286 nm per residue observed in a collagen triple helix, the 150-residue stretch encompassing 50 GXY repeats in the CL region was calculated to be 43 nm long, which correlated well with the measured length. Rod-shaped particles (n=190) of similar length 45.6±4.4 nm were seen in the P157 preparation, further confirming that the CL region of Scl1 alone formed a triple helix.
The shape of the globular domain in P144 was more heterogenous with some molecules appearing spherical and others elongated extending along the axis of the collagenous tail. The diameter of the globular domain (n=279) measured in the middle of the head perpendicularly to the axis of the tail was 8.9±2.2 nm. Considering size overestimation by 2.5 to 5 nm due to metal decoration, a diameter of 3.9-6.4 nm appeared more realistic. The latter estimate was close to the theoretical value of 3.8 nm calculated for a spherically shaped protein with a molecular mass of a trimeric V region using the equation: d=2(3vMr/4πNA)1/3, where Mr=25,072, NA is Avogadro's number, and v=0.73 cm3/g is the assumed partial specific volume.
The dimensions of P163 also agreed with theoretical predictions calculated for the Scl2.28 protein. Specifically, the measured length of the collagenous domain (n=253) of 66.2±8.9 nm correlated well with the calculated value of 67.8 nm. Similarly, a diameter of the globular domain (n=124) of 7.4±0.9 nm agreed with the calculated value of 3.9 nm, considering a metal decoration component. Examination of a rotary shadowed preparation of P176 (
Computer Modeling
The trimeric structure of the collagen-like region of Scl1.41 and Scl2.28 was homology-modeled using SwissModel, based on the coordinates of a (Pro-Pro-Gly)10 collagen-like peptide. The target-template alignment was guided by superposition of Gly residues. Energy minimization in vacuo was carried out using GROMOS96 to relieve short contacts and improve local stereochemistry. Interactions involving the side chains have not been modeled explicitly.
The amino acid sequence of the collagenous domains of Scl1.41 and Scl2.28 was fitted onto the structure of a regular polyGPP triple helix (
Although the CL regions in Scl1.41 and Scl2.28 are both composed of continuous GXY repeats, they show remarkable primary sequence differences. The five most frequent triplets in Scl1.1 GEK, GPQ, GEA, GET, and GPA, account for 48% of all triplets in this protein but represent only 10% of the triplets in Scl2.28. Similarly, the GKD, GAQ, GPA, GER, and GLP triplets make up 55.9% of the collagen-like segment in Scl2.28 but account for only 16% of the triplets in Scl1.1. Fifty GXY motifs of Scl1.1CL contain 21 distinct triplets, seven of which are not found in Scl2.28. Conversely, 12 out of 26 distinct triplets are specific to Scl2.28CL.
In addition, those GXY motifs that are common to both Scd proteins are arranged in different orders in Scl1.1. CL and Scl2.28CL. Despite the significant sequence variation in the CL region of the two proteins, these could both successfully be modeled on a polyGPP structure. Thus the ability to form a collagen-like triple helix structure is not unique to a particular Scl sequence but may be the property of other prokaryotic proteins containing repeated GXY motifs.
Collagen-like Protein Promotes Cell Adhesion and Spreading
The expression levels of various integrin chains on MRC-5 and WI-38 cells were determined using fluorescence activated cell sorting (FACS). Both cell types expressed collagen binding α1β1 and α2β1 integrins at comparable levels, as well as α3β1, α5β1 and αv1 which presumably partners with β5 (
Next, cell adhesion assays were performed on immobilized recombinant P163 and P176 proteins or known adhesive extracellular matrix (ECM) proteins. P176 served as a substrate for the attachment of both MRC-5 and WI-38 cells, similar to fibronectin (FN), type IV collagen (Col IV), and type I collagen (Col I). The extent of cell attachment to these substrates depended on the amount of protein used to coat the wells and the incubation time of the cells (
The cell attachment results were complemented with cell spreading assays. Detached WI-38 and MRC-5 cells were allowed to reattach onto dishes coated with FN, Col I and P176 substrates for 45 or 90 minutes. P176 induced considerable cell spreading, as did the positive control adhesion substrates (
Cell spreading was further examined by evaluating cell morphology in fluorescent-stained MRC-5 cells (
Cell Adhesion Activity of P176 is Mediated via its Collagen-like Region
To delineate the regions of the P176 protein responsible for it ability to support cell adhesion and spreading activity, two chimeric recombinant proteins, P181 and P182, were generated by exchanging domains between the putative integrin-binding variant P176 and the non-interacting variant P163 (
Both P181 and P182 were able to form collagen-like triple helices, as determined by CD (
P176 Interacts Specifically with the α2β1 Collagen-binding Integrin
To determine which collagen-binding integrin(s) recognizes P176, we tried to inhibit cell attachment using a panel of monoclonal antibodies directed against the extracellular segments of human integrins. The ability of MRC-5 cells to attach onto plates coated with P176 was measured in the presence of increasing concentrations of anti-α1β1, -α2β1, -α3β1, -αvβ3, -α5β1, or -β1 integrins adhesion-blocking monoclonal antibodies, or in the presence of EDTA, a metal ion chelating agent (
To further examine the role of the α2β1 integrin in cell adhesion to P176, collagen receptor deficient C2C12 myoblast cells, and C2C12 cells stably expressing the human wild-type α2 integrin subunit (designated C2C12-α2+) were used. Upon expression in C2C12 cells, wild-type2 polypeptide combines with the endogenous β1 subunit, to form a functional α2β1 integrin (24). Col I, Col IV, FN, and VN were included as positive controls in cell adhesion assays. As expected, Col I and Col IV failed to promote adhesion of parental C2C12 cells, in contrast, both ECM proteins supported adhesion of C2C12-α2+ cells (
If C2C12-α2+ adheres to P176 through integrins(s), does adhesion blocking anti-integrin antibodies inhibit their interactions? The attachment of both parental and C2C12-α2+ cells to BSA (negative control) and P163 were insignificant (
These findings further support the hypothesis that cell adhesion is mediated through α2β1. Furthermore, because P176 interacts directly with α2β1 integrin, the data may support a caveolae-mediated entry route for certain S. pyogenes strains. Both the paracellular and the caveolar mechanisms may be advantageous for the propagation of GAS and for its pathogenicity.
The I-domain of α2 Integrin (α2-I) Interacts with P176 but not P181
The I-domain of the α components of collagen binding integrins have been demonstrated to directly bind to specific site in triple helix collagens (62). To study if the I-domain of a2 integrin subunit binds to P176, α2-I was expressed as a recombinant protein and its binding to P176 and P181 examined by surface plasmon resonance (SPR) spectroscopy. Different concentrations of α2-I were passed over Bia-core chips to which the recombinant Scl proteins had been coupled.
A concentration dependent binding of α2-I to P176 was noted with rapid on and off rates and defined equilibrium (
P176 Induces Tyrosine Phosphorylation of p125FAK, p130CAS, Paxillin, and JNK
Cell adhesion-induced tyrosine phosphorylation of p125FAK, p130CAS, paxillin proteins and JNK are considered to be integrin-mediated signaling events (36). The phosphorylation state of these proteins was evaluated to determine whether P176 induces α2β1 integrin signaling in this manner. Cells kept in suspension were used as a negative control and cells adhering to Col I and FN substrates were used as positive controls.
Cells were incubated with the various substrates for 30 min, a time required for FAK activation, and the lysates were analysed as shown in
Immune complexes were analyzed by immunoblotting with an anti-phospho-tyrosine antibody. Like p125FAK, both p130CAS and paxillin were phosphorylated in response to adhesion onto P176, FN, and Col I (
Similarly, phospho-JNK immunoblotting analysis of total lysates showed that JNK was phosphorylated in response to adhesion onto P176, FN and Col I (
The data indicate that fibroblast cells attach and spread on P176 and that these events are associated with the formation of focal adhesion contacts where vinculin and talin localize. P176-mediated integrin clustering and signaling is consistent with the time-dependent activiation of FAK and JNK protein kinases observed in fibroblasts plated on P176. In fibroblast and endothelial cells, p38 MAP kinase and ATF-2 are activated in response to α2β1 integrin ligation (67-68). Thus P176 promotes tyrosine phosphorylation of integrin signaling molecules in fibroblasts. This is consistent with the hypothesis that Scl1.41 is a bacterial collagen that induces signaling through the cell adhesion receptor α2β1 integrin.
The following references are cited herein:
Any patents or publications mentioned in this specification are indicative of the levels of those skilled in the art to which the invention pertains. Further, these patents and publications are incorporated by reference herein to the same extent as if each individual publication was specifically and individually incorporated by reference.
One skilled in the art will appreciate readily that the present invention is well adapted to carry out the objects and obtain the ends and advantages mentioned, as well as those objects, ends and advantages inherent herein. Changes therein and other uses which are encompassed within the spirit of the invention as defined by the scope of the claims will occur to those skilled in the art.
This is a divisional of U.S. Ser. No. 11/245,689, filed Oct. 7, 2005, now U.S. Pat. No. 7,238,783, which is a divisional of and claims benefit of priority under 35 U.S.C. Å 120 of U.S. Ser. No. 10/830,792, filed Apr. 23, 2004, now U.S. Pat. No. 6,953,839, which claims benefit of priority under 35 U.S.C Å 119(e) of U.S. Ser. No. 60/464,816, filed Apr. 23, 2003, now abandoned.
This invention was produced in part using funds obtained through grants AR44415 and AI50666 from the National Institutes of Health. Consequently, the federal government has certain rights in this invention.
| Number | Date | Country | |
|---|---|---|---|
| 20080014613 A1 | Jan 2008 | US |
| Number | Date | Country | |
|---|---|---|---|
| 60464816 | Apr 2003 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 11245689 | Oct 2005 | US |
| Child | 11800288 | US | |
| Parent | 10830792 | Apr 2004 | US |
| Child | 11245689 | US |
