PROTEASE RESISTANT MUTANTS OF STROMAL CELL DERIVED FACTOR-1 IN THE REPAIR OF TISSUE DAMAGE

Abstract

The present invention is directed stromal cell derived factor-1 peptides that have been mutated to make them resistant to digestion by the proteases dipeptidyl peptidase IV (DPPIV) and matrix metalloproteinase-2 (MMP-2) but which maintain the ability of native SDF-I to attract T cells. The mutants may be attached to membranes formed by self-assembling peptides and then implanted at sites of tissue damage to help promote repair.

Claims

1-20. (canceled)

21. A fusion protein comprising the formula: A—(L).sub.n—(R).sub.q, wherein: A is an isolated mutant form of stromal cell derived factor-1 (SDF-1) peptide comprising the formula mSDF-1 or X.sub.p-mSDF-1, wherein mSDF-1 comprises the amino acid sequence of at least amino acids 1-8 of SEQ ID NO:52 and which is optionally extended at the C terminus of amino acids 1-8 of SEQ ID NO: 52 by all or any portion of the remaining sequence of SEQ ID NO:52, shown as amino acids 9-68 of the full-length of SEQ ID NO: 52 and the amino acids 1-8 comprises a mutation at the fourth and/or fifth amino acids from the N terminus of the amino acids 1-8 of SEQ ID NO:52, and wherein: a) X is a proteinogenic amino acid or a protease protective organic group; b) p is an integer from 1 to 4; and c) the mutant form of SDF-1 peptide has chemoattractant activity for T cells, is inactivated by dipeptidyl peptidase IV (DPPIV) at a rate that is less than one-half of the rate at which native SDF-1 is inactivated, and is inactivated by MMP-2 at a rate that is less than one-half of the rate at which native SDF-1 is inactivated; L is a linker sequence of 3-9 amino acids; R is a self-assembling peptide comprising the amino acid sequence of any one of SEQ ID NOs: 1-51; n is an integer from 0-3; and q is an integer from 1-3.

22. The fusion protein of claim 21, wherein A comprises: i) at least amino acids 1-8 of SEQ ID NO:53 and which is optionally extended at the C terminus of amino acids 1-8 of SEQ ID NO: 53 by all or any portion of the remaining sequence of SEQ ID NO:53, shown as amino acids 9-68 of the full-length of SEQ ID NO: 53; ii) at least amino acids 1-8 of SEQ ID NO:54 and which is optionally extended at the C terminus of amino acids 1-8 of SEQ ID NO: 54 by all or any portion of the remaining sequence of SEQ ID NO:54, shown as amino acids 9-68 of the full-length of SEQ ID NO: 54; iii) at least amino acids 1-8 of SEQ ID NO:55 and which is optionally extended at the C terminus of amino acids 1-8 of SEQ ID NO: 55 by all or any portion of the remaining sequence of SEQ ID NO:55, shown as amino acids 9-68 of the full-length of SEQ ID NO: 55; or iv) at least amino acids 1-8 of SEQ ID NO:56 and which is optionally extended at the C terminus of amino acids 1-8 of SEQ ID NO: 56 by all or any portion of the remaining sequence of SEQ ID NO:56, shown as amino acids 9-68 of the full-length of SEQ ID NO: 56.

23. The fusion protein of claim 21, wherein A comprises: i) at least amino acids 1-17 of SEQ ID NO:53 and which is optionally extended at the C terminus of amino acids 1-17 of SEQ ID NO: 53 by all or any portion of the remaining sequence of SEQ ID NO:53, shown as amino acids 18-68 of the full-length of SEQ ID NO: 53; ii) at least amino acids 1-17 of SEQ ID NO:54 and which is optionally extended at the C terminus of amino acids 1-17 of SEQ ID NO: 54 by all or any portion of the remaining sequence of SEQ ID NO:54, shown as amino acids 18-68 of the full-length of SEQ ID NO: 54; iii) at least amino acids 1-17 of SEQ ID NO:55 and which is optionally extended at the C terminus of amino acids 1-17 of SEQ ID NO: 55 by all or any portion of the remaining sequence of SEQ ID NO:55, shown as amino acids 18-68 of the full-length of SEQ ID NO: 55; or iv) at least amino acids 1-17 of SEQ ID NO:56 and which is optionally extended at the C terminus of amino acids 1-17 of SEQ ID NO: 56 by all or any portion of the remaining sequence of SEQ ID NO:56, shown as amino acids 18-68 of the full-length of SEQ ID NO: 56.

24. The fusion protein of claim 21, wherein A comprises the sequence of any one of SEQ ID NOs: 53-56.

25. The fusion protein of claim 21, wherein X is serine.

26. The fusion protein of claim 25, wherein p is 1.

27. The fusion protein of claim 21, wherein R comprises the amino acid sequence of SEQ ID NO: 35.

28. The fusion protein of claim 27, wherein q is 1.

29, The fusion protein of claim 21, wherein L comprises the amino acid sequence of any one of SEQ ID NOs: 57-59.

30. The fusion protein of claim 29, wherein n is 1.

31. A method of treating a patient to promote the repair of damaged tissue, the method comprising administering to the patient the fusion protein of claim 21.

32. The method of claim 31, wherein the patient is treated for a disease or condition selected from stroke, limb ischemia, tissue damage due to trauma, and diabetic ulcer.

33. A biologically-compatible membrane comprising the fusion protein of claim 21.

34. A method of treating a patient to promote the repair of damaged tissue, the method comprising administering to the patient the biologically-compatible membrane of claim 33.

35. The method of claim 34, wherein the patient is treated for a disease or condition selected from stroke, limb ischemia, tissue damage due to trauma, and diabetic ulcer.

36. The method of claim 34, wherein the biologically-compatible membrane is injected or implanted at the site of tissue damage.

37. The method of claim 34, wherein the patient is treated for damage to cardiac tissue and the biologically-compatible membrane is injected or implanted into the myocardium of the patient.

Description

DESCRIPTION OF THE INVENTION

[0033] The present invention is based upon the concept that the recovery of damaged tissue, e.g., damaged cardiac tissue, is promoted by exposing the tissue to SDF-1 that has been mutated to make it resistant to MMP-2 and/or DPPIV cleavage and which is delivered by means of a membrane formed by spontaneously assembling peptides. The self-assembling peptides have been described in U.S. Pat. Nos. 5,670,483 and 6,548,630 (hereby incorporated by reference in their entirety). Methods of attaching factors to membranes and the use of the membranes in delivering therapeutic agents to cardiac tissue have also been described (see published US applications 20060148703 and 20060088510, hereby incorporated by reference in their entirety). The same procedures for making and using membranes may be applied to the present invention.

Description of Self-Assembling Peptides

[0034] The peptides used for self-assembly should be at least 12 residues in length and contain alternating hydrophobic and hydrophilic amino acids. Peptides longer than about 200 amino acids tend to present problems with respect to solubility and membrane stability and should therefore be avoided. Ideally, peptides should be about 12-24 amino acids in length.

[0035] The self-assembling peptides must be complementary. This means that the amino acids on one peptide must be capable of forming ionic bonds or hydrogen bonds with the amino acids on another peptide. Ionic bonds would form between acidic and basic amino acid side chains. The hydrophilic basic amino acids include Lys, Arg, His, and Orn. The hydrophilic acidic amino acids are Glu and Asp. Ionic bonds would form between an acidic residue on one peptide and a basic residue on another. Amino acids that form hydrogen bonds are Asn and Gln. Hydrophobic amino acids that may be incorporated into peptides include Ala, Val, Ile, Met, Phe, Tyr, Trp, Ser, Thr, and Gly.

[0036] Self-assembling peptides must also be “structurally compatible.” This means that they must maintain an essentially constant distance between one another when they bind. Interpeptide distance can be calculated for each ionized or hydrogen bonding pair by taking the sum of the number of unbranched atoms on the side-chains of each amino acid in the pair. For example, lysine has five and glutamic acid has four unbranched atoms on their side chains. An interaction between these two residues on different peptides would result in an interpeptide distance of nine atoms. In a peptide containing only repeating units of EAK, all of the ion pairs would involve lysine and glutamate and therefore a constant interpeptide distance would be maintained. Thus, these peptides would be structurally complementary. Peptides in which the variation in interpeptide distance varies by more than one atom (about 3-4 angstroms) will not form gels properly. For example, if two bound peptides have ion pairs with a nine-atom spacing and other ion pairs with a seven-atom spacing, the requirement of structural complementarity would not have been met. A full discussion of complementarity and structural compatibility may be found in U.S. Pat. Nos. 5,670,483 and 6,548,630.

[0037] It should also be recognized that membranes may be formed from either a homogeneous mixture of peptides or a heterogeneous mixture of peptides. The term “homogeneous” in this context means peptides that are identical with one another. “Heterogeneous” indicates peptides that bind to one another but which are structurally different. Regardless of whether homogenous or heterogeneous peptides are used, the requirements with respect to the arrangement of amino acids, length, complementarity, and structural compatibility apply. In addition, it should be recognized that the carboxyl and amino groups of the terminal residues of peptides can either be protected or not protected using standard groups.

Making of Peptides

[0038] The self-assembling and protease resistant SDF-1 peptides of the present invention can be made by solid-phase peptide synthesis using standard N-tert-butyoxycarbonyl (t-Boc) chemistry and cycles using n-methylpyrolidone chemistry. Once peptides have been synthesized, they can be purified using procedures such as high pressure liquid chromatography on reverse-phase columns. Purity may also be assessed by HPLC and the presence of a correct composition can be determined by amino acid analysis. A purification procedure suitable for mSDF-1 peptides is described in the Examples section.

[0039] Fusion proteins may either be chemically synthesized or made using recombinant DNA techniques. The full sequences of these proteins are described herein and examples are provided of DNA sequences that can be used in producing them.

Binding of SDF-1 to Self-Assembling Peptides

[0040] Several strategies may be used for attaching protease resistant SDF-1 to self-assembling peptides. One strategy is non-covalent binding which has previously been shown to be effective in delivering PDGF-BB, a growth factor, to tissues (Hsieh, et al., J. Clin. Invest. 116:237-248 (2006)).

[0041] A second attachment strategy is the biotin-sandwich method (Davis, et al., Proc. Nat'l Acad. Sci. USA 103:8155-8160 (2006)) in which a protease resistant SDF-1 is biotinylated and bound to biotinylated peptides using tetravalent streptavidin as a linker. To accomplish this, the protease resistant SDF-1 may be coupled to the 15 amino acid sequence of an acceptor peptide for biotinylation (referred as AP; Chen, et al., Nat. Methods 2:99-104 (2005)). Because the active site of SDF-1 is situated near the amino terminus, fusion proteins should be made by incorporating the extra sequences at the C-terminus. The acceptor peptide sequence allows site-specific biotinylation by the E. coli enzyme biotin ligase (BirA; Chen, et al., Nat. Methods 2:99-104 (2005)). Many commercial kits are available for biotinylating proteins. However, many of these kits biotinylate lysine residues in a nonspecific manner, and this may reduce mSDF-1 activity as it has been shown that the N-terminal lysine of SDF-1 is crucial for receptor binding and activity (Crump, et al, EMBO 1 16:6996-7007 (1997)). Biotinylated self-assembling peptides are made by MIT Biopolymers laboratory and when mixed in a 1 to 100 ratio with native self-assembling peptides, self-assembly of nanofibers should not be disturbed (Davis, et al., Proc. Nat'l Acad. Sci. USA 103:8155-8160 (2006)).

[0042] A third targeting strategy is direct incorporation of protease resistant SDF-1 peptides into self-assembling nanofibers by construction of a fusion protein of mutated SDF-1 with a self-assembling peptide. For example an mSDF-1 may be coupled to the 16 amino acid sequence of SEQ ID NO:35. This “RAD” portion of the fusion protein will incorporate into the nanofiber scaffold while assembling.

Formation of Membranes

[0043] The self-assembling peptides and fusion proteins described herein will not form membranes in water, but will assemble in the presence of a low concentration of monovalent metal cation. The order of effectiveness of these cations is Li.sup.+>Na.sup.+>K.sup.+>Cs.sup.+ (U.S. Pat. No. 6,548,630). A concentration of monovalent cation of 5 mM should be sufficient for peptides to assemble and concentrations as high as 5 M should still be effective. The anion associated with the monovalent cation is not critical to the invention and can be acetate, chloride, sulfate, phosphate, etc.

[0044] The initial concentration of self-assembling peptide will influence the final size and thickness of membranes formed. In general, the higher the peptide concentration, the higher the extent of membrane formation. Formation can take place at peptide concentrations as low as 0.5 mM or 1 mg/ml. However, membranes are preferably formed at higher initial peptide concentrations, e.g., 10 mg/ml, to promote better handling characteristics. Overall, it is generally better to form membranes by adding peptides to a salt solution rather than adding salt to a peptide solution.

[0045] The formation of membranes is relatively unaffected by pH or by temperature. Nevertheless, pH should be maintained below 12 and temperatures should generally be in the range of 4-90° C. Divalent metal cations at concentrations equal to or above 100 mM result in improper membrane formation and should be avoided. Similarly, a concentration of sodium dodecyl sulfate of 0.1% or higher should be avoided.

[0046] Membrane formation may be observed by simple visual inspection and this can be aided, if desired, with stains such as Congo Red. The integrity of membranes can also be observed microscopically, with or without stain.

Pharmaceutical Compositions and Dosages

[0047] Membranes with attached protease resistant SDF-1 peptides or fusion proteins may be incorporated into a pharmaceutical composition containing a carrier such as saline, water, Ringer's solution and other agents or excipients. The dosage form will generally be designed for implantation or injection, particularly into cardiac tissue but topical treatments will also be useful, e.g., in the treatment of wounds. All dosage forms may be prepared using methods that are standard in the art (see e.g., Remington's Pharmaceutical Sciences, 16th ed. A. Oslo. ed., Easton, Pa. (1980)).

[0048] It is expected that the skilled practitioner will adjust dosages on a case by case basis using methods well established in clinical medicine. The optimal dosage will be determined by methods known in the art and will be influenced by factors such as the age of the patient, disease state and other clinically relevant factors.

EXAMPLES

Example 1

Biological Effects and Protease Resistance of SDF-1 Mutants

SDF-1 Purification and Expression

[0049] The DNA sequence of mature SDF-1α may be cloned from human cDNA into pET-Sumo vector and an extra N-terminal serine residue may be incorporated to facilitate cleavage by Sumo protease (yielding an SDF-1 form of 69 AA). Fusion proteins may be made by incorporating RAD or AP sequences in reverse primers. Sumo-SDF-1 fusion proteins are expressed in Rosetta DE3 E coli and grown to an optical density of 1.5 (600 nm) at 37° C. Cells are induced with 0.25 mM isopropyl β-D-thiogalactoside for 4 h and harvested by centrifugation. As described below, SDF-1α may be purified by a 3-step procedure; all steps being performed at 21° C.

[0050] Cells from a 4-L growth were lysed in 300 ml lysis buffer (6M Guanidine, 20 mM phosphate (pH 7.8), 500 mM NaCl) and homogenized. Debris is collected by centrifugation at 3000 g. The first purification step consisted of capture of the poly-histidine tag present in the SUMO-SDF-1α fusion protein with Nickel-NTA. Nickel-NTA resin was washed with wash buffer (8M Urea, 500 mM NaCl, 20 mM phosphate (pH 6.2)) and the bound protein was eluted at pH 4. Further purification and oxidative refolding were performed on a Cation Exchange HPLC column. The sample was adjusted to binding buffer (8M Urea, 30 mM 2-mercaptoethanol, 1 mM EDTA, 50 mM Tris pH8) and loaded on the HPLC column. Refolding of Sumo-SDF-1 was performed on the column with a 2 h run of refolding buffer (50 mM Tris pH8, 75 mM NaCl, 0.1 mM reduced Glutathione and 0.1 mM oxidized Glutathione). Sumo-SDF-1 was eluted with a step gradient (0.5 to 1M NaCl) and concentrated. The SUMO-SDF-1 fusion protein was cleaved by Sumo Protease 1 (1U/50 μg protein) in 50 mM Tris pH 8.0, 500 mM NaCl. The sample was adjusted to 0.1% trifluoroacetic acid (TFA) and loaded on a C18 Reversed Phase HPLC column for the final purification step. The column was subjected to a linear gradient from 30 to 40% acetonitrile in 0.1% TFA. The fractions containing SDF-1 were lyophilized and resuspended. Activity of purified SDF-1 was tested by migration of Jurkat T-lymphocyte cell line.

Modification of SDF-1 Constructs

[0051] SDF-1 fusion constructs were modified by insertional mutagenesis with one of three sequences: one sequence is susceptible to MMP-2 cleavage (MMP cleavage site or MCS), another sequence contains the same amino acids but in a random order (scrambled sequence, or SCR), and the third sequence contains 6 glycines as a linker.

Mutations of the MMP Cleavage Sites in Chemokines

[0052] SDF-1 is cleaved by MMP-2 in its active site at the N-terminus, leaving an N-terminal tetrapeptide and inactive SDF-1(5-68). Specific mutagenesis of 4 different amino acids was performed in order to render SDF-1 resistant to MMP-2 cleavage, based on substrate sequences of MMP-2 described by Netzel-Arnett et al (Biochemistry 32:6427-6432 (1993)). The four different constructs were expressed and purified as described for SDF-1. Of the 4 different mutations, SDF-1(L5W) and SDF-1(L5E) showed minimal activity on T-cell migration. In contrast, SDF-1(S4V) and SDF-1(L5P) showed bioactivity comparable to native SDF-1. Because SDF-1(L5P) was more difficult to purify, SDF-1(S4V) was selected for further experiments.

Effect of Mutations on Protease Susceptibility and Chemoattractant Activity

[0053] The mutated forms of SDF-1 were examined in an assay of migration of Jurkat T cells at a concentration of 100 nM. This assay indicated that both SDF-1(S4V) and SDF-1(L5P) retained most of the activity of unmutated SDF-1 in promoting T cell migration. This activity was greatly reduced in SDF-1(L5W) mutants and SDF-1(L5E) mutants.

[0054] The susceptibility of the peptides to cleavage by MMP-2 was determined by incubating the mutants with the enzyme for one hour and then examining the incubation product by SDS-PAGE. This revealed that, unlike SDF-1, the mutants did not undergo a positional shift indicative of cleavage. MMP-2 incubation was also found to reduce the chemoattractant activity of SDF-1 but not SDF-1(S4V) as shown by a Jurkat T-cell migration assay. These results suggest that the S4V variant of SDF-1 retains chemokine bioactivity but is resistant to activation by MMP-2.

In Vivo Data

[0055] A blinded and randomized study was performed to evaluate the effect of different SDF-1 forms on cardiac function after myocardial infarction in rats. Ejection fraction was measured with a Millar catheter system for measurement of intraventricular pressures and ventricular volumes. Both SDF-1(S4V)-6G-RAD and SDF-1(S4V)-MCS-RAD significantly increased cardiac function 4 weeks after myocardial infarction in rats compared to MI only group. This indicates that both MMP-2 resistance (SDF-1(S4V)) and attachment to membranes are necessary for successful cardiac repair therapy.

Example 2

Experiments with Truncated Forms of SDF-1

[0056] 3 truncated forms of SDF-1 were synthesized commercially; all include the first 17 amino acids of native SDF-1. Two variants of SDF-1 17AA were designed to be more resistant to MMP-2, based on our prior work with the entire SDF-1 protein:

TABLE-US-00003 SDF-1 17AA: KPVSLSYRCPCRFFESH (SEQ ID 64) SDF-1(S4V) 17AA: KPVVLSYRCPCRFFESH (SEQ ID 65) SDF-1(L5P) 17AA: KPVSPSYRCPCRFFESH (SEQ ID 66)

[0057] Migration experiments were performed with the Jurkat T-lymphocyte cell line. Truncated SDF-1 17AA was 500 times less potent than native SDF-1 but maximal migration induced was similar to native SDF-1. Therefore, if 500 times higher concentrations were used compared to full-length protein, the same migratory response of T-lymphocytes should be observed. The mutated SDF-1(S4V) 17AA and SDF-1(L5P) 17AA were three times less potent than SDF-1 17AA without mutation. This is a similar shift to that seen between native SDF-1 and SDF-1(S4V).

[0058] Cleavage experiments of the peptides with MMP-2 were performed: 2 nmole of SDF-1 17AA, SDF-1(S4V) 17AA, and SDF-1(L5P) 17AA were incubated with MMP-2 for 1 h at RT. Proteins were run on an SDS-PAGE showing cleavage of SDF-1 17AA, but not of SDF-1(S4V) 17AA or SDF-1(L5P) 17AA. Thus, these truncated proteins may be useful therapeutically, as they are still bioactive and also MMP-2 resistant.

[0059] All references cited herein are fully incorporated by reference. Having now fully described the invention, it will be understood by those of skill in the art that the invention may be practiced within a wide and equivalent range of conditions, parameters and the like, without affecting the spirit or scope of the invention or any embodiment thereof.