Recombinant proCVF

Abstract

Recombinant proCVF exhibits substantially the same activity as CVF and is useful for lowering complement activity.

Description

BACKGROUND OF THE INVENTION

[0001] 1. Field of the Invention

[0002] The present invention relates to recombinant pro-cobra venom factor (proCVF), DNA encoding recombinant proCVF, plasmids comprising such DNA, and transformed microorganisms containing such DNA. The present invention also relates to various methods of making and using recombinant proCVF.

[0003] 2. Discussion of the Background

[0004] The third component of complement, C3, plays a pivotal role in both the classical and alternative pathways of complement activation, and many of the physiologic C3 activation products have important functions in the immune response and host defense (Müller-Eberhard, H. J., 1988, Annu. Rev. Biochem. 57:321). In the alternative pathway, the activated form of C3, C3b, is a structural subunit of the C3 convertase. This bimolecular enzyme consists of C3b and Bb, the activated form of factor B. This enzyme is formed by the binding of C3b to factor B that is subsequently cleaved by factor D, resulting in the formation of the C3 convertase, C3b,Bb, and the release of the activation peptide Ba. The C3 convertase activates C3 by cleaving the molecule into C3b and the anaphylatoxin, C3a. The C3b molecule will bind in close proximity to the C3 convertase. Eventually, the bound C3b will allow for the activation of C5 into C5b and the anaphylatoxin, C5a. C5 activation occurs by the same C3b,Bb enzyme that can cleave CS when it is bound to an additional C3b molecule. The C5-cleaving enzyme is called C5 convertase. It is a trimolecular complex composed of (C3b)2,Bb. Inasmuch as the activation of both C3 and C5 occurs at the identical active site in the Bb subunit, the enzyme is also called C3/C5 convertase; and only one EC number has been assigned (EC 3.4.21.47).

[0005] Cobra venom contains a structural and functional analog of C3 called cobra venom factor (CVF). This molecule can bind factor B in human and mammalian serum to form the complex, CVF,B (Hensley, P. M., et al, 1986, J. Biol. Chem. 261:11038), which is also cleaved by factor D into the bimolecular enzyme CVF,Bb and Ba (Vogel. C.-W., et al, 1982, J. Biol. Chem. 257:8292). The bimolecular complex CVF,Bb is a C3/C5 convertase that activates C3 and C5 analogously to the C3/C5 convertase formed with C3b (Vogel, C.-W., 1991, In Handbook of Natural Toxins, Vol. 5, Reptile and Amphibian Venoms. A. T. Tu, ed. Marcel Dekker, New York, p. 147). Although the two C3/C5 convertases C3b,Bb and CVF,Bb share the molecular architecture, the active site-bearing Bb subunit, and the substrate specificity, the two enzymes exhibit significant functional differences. The CVF,Bb enzyme is physicochemically far more stable than C3b,Bb (Vogel. C.-W., et al, 1982, J. Biol. Chem. 257:8292; Medicus, R. G., et al, 1976, J. Exp. Med. 144:1076), it is resistant to inactivation by the regulatory proteins factors H and 1 (Lachmann, P. J., et al, 1975, Clin. Exp. Immunol. 21:109; Nagaki, K., et al, 1978, Int. Arch. Allergy Appl. Immunol. 57:221), it exhibits different kinetic properties (Vogel. C.-W., et al, 1982, J. Biol. Chem. 257:8292; Pangburn, M. K., et al, 1986, Biochem J. 235:723), and it does not require additional C3b for C5 cleavage (Miyama, A., et al, 1975, Biken J. 18:193; Von Zabern, et al, 1980, Immunobiology 157:499).

[0006] CVF and mammalian C3 have been shown to exhibit several structural similarities including immunologic cross-reactivity (Alper, C.A. et al, 1983, Science 191:1275; Eggertsen, G. A., et al, 1983, J. Immunol. 131:1920; Vogel, C.-W., et al, 1984, J. Immunol. 133:3235; Grier, A. H., et al, 1987, J. Immunol. 139:1245), amino acid composition (Vogel, C.-W., et al, 1984, J. Immunol. 133:3235; Vogel, C.-W., et al, 1985, Complement 2:81), circular dichroism spectra, and secondary structure (Vogel, C.-W., et al, 1984, J. Immunol. 133:3235), electron microscopic ultrastructure (Vogel, C.-W., et al, 1984, J. Immunol. 133:3235; Smith, C. A., et al, 1982, J. Biol. Chem. 257:9879; Smith, C. A., et al, 1984, J. Exp. Med. 159:324), and amino-terminal amino acid sequence (Vogel, C.-W., et al, 1984, J. Immunol. 133:3235; Lundwall, A., et al, 1984, FEBS Lett. 169:57). Nevertheless, significant structural differences exist between the two molecules. Whereas C3 is a two-chain molecule with an apparent molecular mass, dependent on the species, of 170 to 190 kDa (Eggertsen, G. A., et al, 1983, J. Immunol. 131:1920; DeBruijn, M. H. L., et al, 1985, Proc. Natl. Acad. Sci. USA 82:708; Alsenz, J., et al, 1992, Dev. Comp. Immunol. 16:63; Vogel, C.-W., et al, 1984, Dev. Comp. Immunol. 8:239), CVF is a three-chain molecule with an apparent molecular mass of 149 kDa (Vogel, C.-W., et al, 1984, J. Immunol. Methods 73:203) that resembles C3c, one of the physiologic activation products of C3 (Vogel, C.-W., 1991, In Handbook of Natural Toxins, Vol. 5, Reptile and Amphibian Venoms. A. T. Tu, ed. Marcel Dekker, New York, p. 147; Vogel, C.-W., et al, 1984, J. Immunol. 133:3235). Another significant structural difference between C3 and CVF lies in their glycosylation, CVF has a 7.4% (w/w) carbohydrate content consisting mainly of N-linked complex-type chains with unusual α-galactosyl residues at the non-reducing termini (Vogel, C.-W., et al, 1984, J. Immunol. Methods 73:203; Gowda, D. C., et al, 1992, Mol. Immunol. 29:335). In contrast, human and rat C3 exhibit a lower extent of glycosylation with different structures of their oligosaccharide chains (Hase, S., et al, 1985, J. Biochem. 98:863; Hirani, S., et al, 1986, Biochem. J. 233:613; Miki, K., et al, 1986, Biochem J. 240:691).

[0007] The multifunctionality of the C3 protein, which interacts specifically with more than 10 different plasma proteins or cell surface receptors, has spurred significant interest in a detailed structure/function analysis of the molecule. For some ligands of C3 the binding sites have been assigned to more or less defined regions of the C3 polypeptide including factor H (Ganu, V. S., et al, 1985, Complement 2:27), properdin (Daoudaki, M. E., et al, 1988, J. Immunol. 140:1577; Farries, T. C., et al, 1990, Complement Inflamm. 7:30), factor B (Fishelson, Z., 1981, Mol. Immunol. 28:545), and the complement receptors CR1 (Becherer, J. D., et al, 1988, J. Biol. Chem. 263:14586), CR2 (Lambris, J. D., et al, 1985, Proc. Natl. Acad. Sci. USA 82:4235; Becherer, J. D., et al, 1989, Curr. Top. Microbiol. Immunol. 153:45), and CR3 (Becherer, J. D., et al, 1989, Curr. Top. Microbiol. Immunol. 153:45; Wright, S. D., et al, 1987, Proc. Natl. Acad. Sci. USA 84:1965; Taniguchi-Sidle, A., et al, 1992, J. Biol. Chem. 267:635). The elucidation of structural differences between C3 and CVF, two closely related molecules that share some properties (e.g., formation of a C3/C5 convertase) but differ in others (e.g., susceptibility to regulation by factors H and I) can be expected to help identify functionally important regions of the C3 molecule.

[0008] The inventors have recently discovered that CVF actually exists in two forms, CVF1 and CVF2. It is desirable to obtain large quantities of CVF1 and CVF2 for a number of reasons. However, the isolation of large quantities of the peptides from cobras is problematic to say the least. Thus, it is desirable to clone the genes which encode CVF1 and CVF2. It is also desirable to provide molecules that exhibit the activity of CVF and can be conveniently produced in large quantities.

SUMMARY OF THE INVENTION

[0009] Accordingly, it is one object of the present invention to provide novel molecules which exhibit the activity of CVF and which are conveniently produced in large quantities.

[0010] It is another object of the present invention to provide novel sequences of DNA which encode such a molecule exhibiting the activity of CVF.

[0011] It is another object of the present invention to provide plasmids which comprise a sequence of DNA which encodes a molecule exhibiting the activity CVF.

[0012] It is another object of the present invention to provide transformed microorganisms which contain heterologous DNA encoding such a molecule exhibiting the activity of CVF.

[0013] It is another object of the present invention to provide a method for producing large quantities of such a molecule.

[0014] It is another object of the present invention to provide various methods of using such a molecule.

[0015] These and other objects, which will become apparent during the following detailed description, have been achieved by the inventors' discovery that recombinant proCVF exhibits substantially the same activity as CVF.

BRIEF DESCRIPTION OF THE DRAWINGS

[0016] A more complete appreciation of the invention and many of the attendant advantages thereof will be readily obtained as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings, wherein:

[0017] FIG. 1 depicts a map of clones used for the sequencing of CVF1. The upper portion shows a schematic drawing of CVF1 cDNA in which the positions and numbers of amino acid residues of the α-, γ-, and β-chains are indicated. The lower portion shows the relative positions of the five cDNA clones that were used to sequence the molecule;

[0018] FIG. 2 shows the cDNA and derived amino acid sequence of CVF1. The NH2- and C-termini of the α-,γ-, and β-chains, functionally important regions, and known ligand binding sites are indicated. Amino acid residue numbering starts at the NH2-terminus of the pro-CVF1 molecule;

[0019] FIG. 3 provides a comparison of CVF1 and C3 sequences at the factor B binding site. Comparisons were made with a sequence analysis program (Devereux, J. R., et al, 1984, Nucleic Acids Res. 12:387). Identical amino acid residues are boxed and shaded, whereas conservative replacements are shaded only. Amino acid residue numbering is based on the CVF1 sequence as shown in FIG. 2. The percent sequence identity and similarity with the CVF1 sequence is shown on the right;

[0020] FIG. 4 provides a comparison of CVF1 and C3 sequences at the properdin binding site. comparisons were made with a sequence analysis program (Devereux, J. R., et al, 1984, 25 Nucleic Acids Res. 12:387). Identical amino acid residues are boxed and shaded, whereas conservative replacements are shaded only. Amino acid residue numbering is based on the CVF1 sequence as shown in FIG. 2. The percent sequence identity and similarity with the CVF1 sequence is shown on the right;

[0021] FIG. 5 provides a comparison of CVF1 and C3 sequences at the thioester site. Comparisons were made with a sequence analysis program (Devereux, J. R., et al, 1984, Nucleic Acids Res. 12:387). Identical amino acid residues are boxed and shaded, whereas conservative replacements are shaded only. Amino acid residue numbering is based on the CVF1 sequence as shown in FIG. 2. The percent sequence identity and similarity with the CVF1 sequence is shown on the right;

[0022] FIG. 6 provides a comparison of C3 sequences at the convertase cleavage site with the N-terminus of the γ-chain of CVF1. Comparisons were made with a sequence analysis program (Devereux, J. R., et al, 1984, Nucleic Acids Res. 12:387). Identical amino acid residues are boxed and shaded, whereas conservative replacements are shaded only. Amino acid residue numbering is based on the CVF1 sequence as shown in FIG. 2. The percent sequence identity and similarity with the CVF1 sequence is shown on the right;

[0023] FIG. 7 provides a comparison of CVF1 and C3 sequences at the factor H and CR2 binding sites. The upper panel shows the factor H orientation site. The lower panel shows the dis-continuous factor H binding site that includes the CR2 binding site (residues 1180-1191) with the highly conserved LYNVEA sequence in all mammalian C3 proteins. Comparisons were made with a sequence analysis program (Devereux, J. R., et al, 1984, Nucleic Acids Res. 12:387). Identical amino acid residues are boxed and shaded, whereas conservative replacements are shaded only. Amino acid residue numbering is based on the CVF1 sequence as shown in FIG. 2. The percent sequence identity and similarity with the cobra sequence is shown on the right;

[0024] FIG. 8 shows hydophilicity/hydrophobicity plots of CVF1 and cobra and human C3 proteins. The plots were generated using a sequence analysis program (Devereux, J. R., et al, 1984, Nucleic Acids Res. 12:387). Hydrophilic regions are shown above, hydrophobic regions below the line. The locations of functionally important sites are indicated;

[0025] FIG. 9 provides a comparison of cobra. human. and mouse C3 with: (a) the N-terminal CVF1 α-chair (b) the N-terminal CVF1 β-chain; and (c) the N-terminal CVF1 γ-chain. Comparisons were made with a sequence analysis program (Devereux, J. R., et al, 1984, Nucleic Acids Res. 12:387). Identical amino acid residues are boxed and shaded, whereas conservative replacements are shaded only. Amino acid residue numbering is based on the CVF1 sequence as shown in FIG. 2. The percent sequence identity and similarity with the cobra sequence is shown on the right;

[0026] FIG. 10, shows the partial cDNA sequence of CVF2;

[0027] FIG. 11 is a schematic which illustrates the construction of the full-length clone pCVF-FL3Δ;

[0028] FIG. 12 is a schematic which illustrates the construction of the full-length clone pCVF-VL5.

[0029] FIG. 13 is a schematic diagram outlining the strategy for polyhedrin-directed expression of recombinant proteins.

[0030] The nonessential polyhedrin gene and viral flanking sequences are cloned into a plasmid vector. The polyhedrin gene promoter is modified for the insertion of foreign genes to produce polyhedrin-fused or nonfused proteins. The transfer vector containing the foreign gene is cotransfected with linearized wild-type Baculovirus DNA into insect cells. In a fraction of the transfected cells, the polyhedrin gene will be replaced with the recombinant DNA via homologous recombination. Recombinant virus is purified by visual screening of a plaque assay, where recombinant plaques are morphologically distinct from wild-type plaques;

[0031] FIG. 14 graphically illustrates the time course of expression rate (left axis) and cell viability (right axis);

[0032] FIG. 15 shows the results of the electrophoresis of recombinant proCVF under reducing conditions: coomassie blue stained 7.5% PAGE gel; lane 1: natural CVF (non-reduced form); lane 2: recombinant proCVF (non-reduced form), lane 3: natural CVF (reduced form, γ-chains (32 kD) are out of range); lane 4: recombinant proCVF (reduced form); lane 5: recombinant proCVF (reduced form, after prolonged incubation);

[0033] FIG. 16 shows the results of a series of agglutination reactions with CVF and proCVF: ConA, Canavalia ensiformis;

[0034] GNA, Galanthus nivalis; SNA, Sambucus nigra; PNA, peanut agglutinin; MAA, Maackia amurensis; and DSA, Datura stramonium;

[0035] FIG. 17 illustrates the effect of tunicamycin on the expression of recombinant proCVF;

[0036] FIG. 18 illustrates the hemolytic activity of recombinant proCVF expressed using the transfer vector pAC-CVF-secr;

[0037] FIG. 19 illustrates the hemolytic activity of recombinant proCVF expressed using the transfer vector PAcGP67-CVF;

[0038] FIG. 20 graphically illustrates the temperature stabilities of recombinant CVF and proCVF; and

[0039] FIG. 21 shows the amino acid sequence of pre-pro-CVF and pre-proCVF-3′His. For pre-pro-CVF-3′His a stretch of 6 histidine residues was added to the C-terminus of pre-pro-CVF.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0040] In a first embodiment, the present invention provides recombinant proCVF. In the context of the present invention, recombinant proCVF includes recombinant proCVF1 and proCVF2, each either glycosylated, partially unglycosylated or totally unglycosylated. The present invention also provides recombinant proCVF to which 1 to 8, preferably about six histidine residues have been added to the 3′ terminus.

[0041] It is also to be understood that the term recombinant proCVF includes those proCVF1 and proCVF2 molecules in which up to 10 amino acid residue deletions, insertions, substitutions, or combinations thereof have been made, so long as the molecule retains at least 10%, preferably at least 30%, more preferably at least 75% of the specific activity of natural CVF from Naja naja for in vitro anticomplementation as measured by the method of Ballow et al (M. Ballow et al, J. Immunol., vol. 103, p. 944 (1969)). It is also to be understood that the term recombinant proCVF includes chimeric molecules in which an amino acid sequence of a particular chain (α, β. . .) or segment of proCVF1 has been substituted for the analogous chain or segment of proCVF2 (or vice versa), with the activity proviso set forth above.

[0042] In a preferred embodiment the recombinant proCVF has the amino acid sequence:

[0043] (a) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2;

[0044] (b) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus;

[0045] (c) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues inserted before position 1 added to the amino terminus of proCVF;

[0046] (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2;

[0047] (e) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with a methionine residue added to the amino terminus;

[0048] (f) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus;

[0049] (g) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the peptide of sequence II added to the amino terminus;

[0050] (h) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus;

[0051] (i) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus;

[0052] (j) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus and a methionine residue added to the amino terminus;

[0053] (k) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus and a methionine residue added to the amino terminus of the histidine residues;

[0054] (l) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus and with 1 to 8 histidine residues added to the carboxy terminus; and

[0055] (m) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 and with 1 to 8 histidine residues added to the amino terminus and the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus of the histidine residues;

[0056] wherein sequence I (gp67 signal peptide) is: MLLVNQSHQGFNKEHTSKMVSAIVLYVLLAAAAHSAFA; and sequence II (peptide encoding for M start codon, 6 histidine residues, and enterokinase cleavage site) is MRGSHHHHHHGMASMTGGQQMGRDLYNNNNK.

[0057] In an especially preferred embodiment, the recombinant proCVF is recombinant proCVF1.

[0058] In a second embodiment, the present invention provides the DNA encoding proCVF. The present DNA may be any which encodes any of the present recombinant proCVF molecules, optionally with 1 to 8 histidine residues added to the 3′-terminus. Thus, in preferred embodiments, the present DNA encodes a recombinant proCVF which has the amino acid sequence:

[0059] (a) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2;

[0060] (b) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus;

[0061] (c) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues inserted before position 1 added to the amino terminus of proCVF;

[0062] (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2;

[0063] (e) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with a methionine residue added to the amino terminus;

[0064] (f) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus;

[0065] (g) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the peptide of sequence II added to the amino terminus;

[0066] (h) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus;

[0067] (i) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus;

[0068] (j) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus and a methionine residue added to the amino terminus;

[0069] (k) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus and a methionine residue added to the amino terminus of the histidine residues;

[0070] (l) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus and with 1 to 8 histidine residues added to the carboxy terminus; and

[0071] (m) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 and with 1 to 8 histidine residues added to the amino terminus and the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus of the histidine residues;

[0072] wherein sequence I (gp67 signal peptide) is: MLLVNQSHQGFNKEHTSKMVSAIVLYVLLAAAAHSAFA; and sequence II (peptide encoding for M start codon, 6 histidine residues, and enterokinase cleavage site) is MRGSHHHHHHGMASMTGGQQMGRDLYNNNNK.

[0073] Preferably, the present DNA encodes recombinant proCVF1. In a preferred embodiment the present DNA encodes recombinant pre-pro CVF. In a particularly preferred embodiment, the present DNA encodes pre-proCVF1.

[0074] The cDNA sequence for CVF1 is 5948 nucleotides long, containing a single open reading frame of 4926 nucleotides, coding for a single pre-pro protein of 1642 amino acid residues. The complete cDNA and amino acid sequence for CVF1 have been reported in D. C. Fritzinger et al, Proc. Natl. Acad. Sci. USA, vol. 91, pp. 12775-12779 (1994) and D. C. Fritzinger et al, Proc. Natl. Acad. Sci. USA, vol. 92, p. 7605 (1995). The reported sequence has a 5′ untranslated region of 3 nucleotides and a 3′ untranslated sequence of 1019 nucleotides, including a 20 base poly-A tail. In especially preferred embodiments, the present DNA has the sequence corresponding to from about position 4 to about position 4929 in the CDNA sequence shown in FIG. 2 or the sequence corresponding to from about position 70 to about position 4929 in the CDNA sequence shown in FIG. 2. Of course, it should be understood that the present DNA sequence encoding proCVF encompasses those derived from the sequence shown in FIG. 2 by any number of additions, deletions and/or substitutions, so long as the encoded proCVF possesses substantially the same anticomplementation activity as natural CVF from Naja naja, as described above in the context of the polypeptides.

[0075] There are several lines of evidence that support the conclusion that the cDNA is indeed the cDNA for CVF1. First of all, the derived protein sequences at the N-termini of α-, β-, and γ-chains match those of the N-termini of the protein, with a single mismatch in each sequence (data not shown). Secondly, the location of glycosylation sites is similar to that found in the protein, with 2 or 3 sites found in the α-chain, a single site in the β-chain, and no sites found in the γ-chain. Finally, the similarity to the sequence of cobra C3 implies that we have sequenced a C3 related protein. Since the mRNA used for this sequence was isolated from the venom gland of cobras, and the sequence is different from that of cobra C3, it is likely that the sequence is that of CVF1.

[0076] From the cDNA sequence, it is clear that CVF1, like cobra and other C3 proteins, is transcribed and translated as a single pre-pro-protein, which is then processed to form the mature protein. In the case of CVF1, this processing includes the removal of the signal sequence from the N-terminus of the α-chain, the removal of the 4 arginines and the “C3a” region that lie between the α- and γ-chains, the removal of the “C3d.g” region that lies between the C-terminus of the γ-chain and the N-terminus of the β-chain, and the glycosylation that occurs on the 2 (or 3) sites in the α-chain and the single site that occurs in the β-chain. While it is not known if all 3 sites in the a-chain are glycosylated, it seems likely that the close proximity of the two sites at positions 131 and 136 would not allow the glycosylation of one if the other is already glycosylated.

[0077] As stated above, CVF1 shows a great deal of homology to cobra C3, both at the protein and at the nucleic acid level. One of the goals in sequencing both cobra C3 and CVF1 was to determine if the two proteins are derived from the same gene (through differential processing at either the RNA or protein level) or from different genes. Comparing the CVF cDNA sequence to that of cobra C3 shows that the two proteins are derived from different, though closely related genes. The main reason for this conclusion is that the comparison of the two nucleic acid sequences shows that the similarity is spread throughout the molecule, with differences not localized to discreet regions. If CVF1 were a product of differential processing at the protein level, it would be expected that the cDNA sequences would be identical throughout. If the differential processing takes place at the RNA level, then one would expect to see portions of the sequence that are identical, interspersed with regions that have little or no similarity to one another. Since the two cDNAs are highly similar throughout their lengths, it is most likely that they are derived from two different genes that are closely related to one another.

[0078] The Thioester site and Factor H binding site of CVF1 and cobra C3 are remarkably similar, even though neither is present in the mature CVF1 protein. The degree of similarity found in this region, where there is no selective pressure to maintain the homology, is further proof that the CVF1 gene arose quite recently. The similarity between the two genes is also evident in the “C3a” region, that also is not present in the mature protein, and in the first 200 nucleotides of the 3′ untranslated region, again implying that CVF1 and C3 only recently diverged from one another.

[0079] Recently, protease activities have been characterized in cobra venom that are able to cleave human C3 into a form that resembles C3b functionally, but has a similar subunit structure to CVF1 (O'Keefe, M. C., et al, 1988, J. Biol. Chem. 263:12690). Since this activity appears to be specific, and not just a random protease, it is possible that this protease serves in the maturation pathway of CVF1. Comparing the venom protease cleavage sites in human C3 to the processing sites in CVF1 shows that the enzyme cleaves human C3 at a position 11 amino acid residues downstream from the actual CVF1 processing site at the N-terminus of the γ-chain, though the venom protease site appears to be in the middle of one of the proposed Factor B binding sites. The second venom protease cleavage site is in a position similar to the C-terminus of the γ-chain, though this position has not been mapped in CVF1. The third venom protease cleavage site is in position 71 amino acids downstream from the N-terminus of the β-chain.

[0080] Given the complete structure of CVF1, and knowing the binding sites for certain regulatory proteins on C3, it should be possible to account for some of the unique properties of CVF1 in activating complement. For example, it is known that, while Factors H and I are able to regulate the activation of complement by dissociating C3b,Bb (the C3 convertase), and by cleaving C3b, CVF1 is resistant to this regulation. Mapping the Factor H binding site on CVF1 shows that the binding site is in the “C3d.g” domain that is removed during the maturation of the protein. Therefore, Factor H is unable to bind to the CVF1 containing C3/C5 convertase, preventing Factor I from cleaving the CVF moiety of the convertase. It is also interesting to speculate on the intrinsic stability of the CVF1 containing C3/C5 convertase compared to the enzyme that contains C3. Comparing the Factor B binding sites to the two proteins should provide some insight into the increased stability of the CVF1,Bb complex. One difference between the factor B binding site is the replacement of the serine at position 721 of CVF1 with an acidic amino acid in C3s.

[0081] A partial sequence of the DNA encoding of CVF2 is shown in FIG. 10. A full length sequence encoding pre-pro-CVF2 may be constructed by ligating the 3′ end of any DNA sequence which encodes for a polypeptide having the amino acid sequence of from position about −22 to position about 299, as shown in FIG. 2 to the 5′ end of any DNA sequence encoding a polypeptide having the amino acid sequence as shown in FIG. 10.

[0082] The DNA of the present invention for CVF2 may comprise any DNA sequence that encodes: pre-pro-CVF2, corresponding to the amino acid sequence in which the carboxy-terminus of the amino acid sequence of from position about −22 to position about 299 in FIG. 2 is bonded to the amino-terminus of the amino acid sequence of from position about 1 to position about 1333 in FIG. 10; or pro-CVF2, corresponding to the amino acid sequence in which the carboxy-terminus of the amino acid sequence of from position about 1 to position about 299 in FIG. 2 is bonded to the amino-terminus of the amino acid sequence of from position about 1 to position about 1333 in FIG. 10.

[0083] In another embodiment, the present invention provides plasmids which comprise a DNA sequence encoding proCVF or pre-proCVF. Any plasmid suitable for cloning or expression may be used, and the DNA may be inserted in the plasmid by conventional techniques. Suitable plasmids and the techniques used to insert the DNA of the present invention into such plasmids are well known to those skilled in the art. For expression purposes, the DNA should be inserted downstream from a promoter and in the proper reading frame.

[0084] In another embodiment, the present invention provides transformed hosts which contain a heterologous DNA sequence encoding proCVF or pre-proCVF. Again, suitable hosts and the means for transforming them are well know to those skilled in the art. Examples of suitable prokaryotic hosts include: E coli, B. subtilis, etc. In the present case, it may be desirable to express the present genes in eukaryotic hosts such as CHO, NIH 3T3 cells, yeast or COS cells. Expression of recombinant proCVF in eukaryotic hosts may be carried out using a broad variety of methods, e.g., transient expression by transfection of cells with recombinant plasmids, development of stable cell lines, expression in cells infected with a recombinant virus, etc.

[0085] In yet another embodiment, the present invention provides a method for preparing proCVF by culturing a transformed host comprising a heterologous DNA sequence encoding proCVF or pre-proCVF. The exact conditions required for the culturing will depend of course on the identity of the transformed host. However, selection of culture conditions is well within the abilities of the skilled artisan.

[0086] It should be noted that although CVF1 and CVF2 are glycosylated as naturally occurring, it has been discovered that proCVF retains its activity even in the unglycosylated state. Thus, an active product may be obtained even if produced by a host incapable of effecting the proper glycosylation.

[0087] Further, proCVF may be processed from the pre-pro-form by treatment with either whole cobra venom or the purified proteases from cobra venom, as described in the Doctoral thesis of M. Clare O'Keefe, Georgetown University, 1991. Thus, active proCVF may be obtained even when produced by a host incapable of the proper post-translational processing. Of course, in some expression systems proCVF will be secreted by the host even though the DNA encodes pre-proCVF.

[0088] Natural native cobra venom factor (CVF) has been used extensively as a research agent to deplete the complement activity in the plasma of laboratory animals in vitro and in vivo, (I. R. Leventhal, et al, Transplantation Proceedings, vol. 25, pp. 398-399 (1993); C.-W. Vogel, et al, J. Immunol. Methods, vol. 73, pp. 203-220 (1984); and K. I. Gaede, et al, Infection and Immunity, vol. 63, pp. 3697-3701 (1995), all of which are incorporated herein by reference). Due to its ability to exhaustively activate complement, injection of CVF into vertebrate animals leads to consumption of complement. This provides for a model system to study the involvement of complement in any biological or pathological mechanism by comparing normal animals with complement-depleted, i.e. CVF-treated, animals. Since recombinant pro-CVF exhibits the same activity of depleting complement activity in serum or plasma as natural CVF, recombinant pro-CVF can be used like natural CVF in a large variety of studies where animals are to be depleted of their complement activity.

[0089] Other examples of complement depletion using CVF are reported in the table below.

Subject Studied                  Reference
Uptake of mycobacteria by monocytes Swartz et al., Infect.
                                 Immun., 56:2223-2227
                                 (1988)
Renal xenograft rejection        Kemp et al., Transplant
                                 Proc., 6:4471-4474
                                 (1987)
Feline leukemia                  Kraut et al., Am. J.
                                 Vet. Res., 7:1063-1066
                                 (1987)
Cardiac xenograft survival       Adachi et al.,
                                 Transplant Proc.,
                                 19:1145-1148 (1987)
Antitumor mechanism of monoclonal Welt et al., Clin.
antibody                         Immunol. Immunopathol.
                                 45:215-229 (1987)
Pulmonary vascular permeability  Johnson et al., J. Appl.
                                 Physiol., 6:2202-2209
                                 (1986)
Glomerular injury and proteinuria Rehan et al., Am. J.
                                 Pathol. 111:57-66 (1986)
Fowlpox virus infection          Ohta et al., J. Virol.,
                                 2:670-673 (1986)
Endotoxin-induced lung injury    Flick et al., Am. Rev.
                                 Respir. Dis. 135:62-67
                                 (1986)
Immunologically mediated otitis media Ryan et al., Clin.
                                 Immunol. Immunopathol.,
                                 40:410-421 (1986)
Antigen-induced arthritis        Lens et al., Clin. Exp.
                                 Immunol., 3:520-528
                                 (1984)
Humoral resistance to syphilis   Azadegan et al., Infect.
                                 Immun., 3:740-742 (1984)
Acute inflammation induced by Escherichia Kopaniak and Movat, Am.
coli                             J. Pathol., 110:13-29
                                 (1983)
Cutaneous late-phase reactions   Lemanske et al., J.
                                 Immnunol., 130:1881-1884
                                 (1983)
Bleomycin-induced pulmonary fibrosis Phan and Thrall, Am. J.
                                 Pathol., 107:25-28
                                 (1982)
Delayed hypersensitivity reactions Jungi and Pepys,
                                 Immunology, 42:271-279
                                 (1981)
Vitamin D2-induced arteriosclerosis Pang and Minta, Artery,
                                 2:109-122 (1980)
Macrophage activation by Corynebacterium Ghaffar, J.
                                 Reticuloendothel. Soc.,
                                 27:327-335 (1980)
Allergic encephalomyelitis       Morariu and Dalmasso,
                                 Ann. Neurol., 5:427-430
                                 (1978)
Effect of complement depletion on IgG and Martinelli et al., J.
IgM response                     Immunol., 121:2043-2047
                                 (1978)
Myocardial necrosis after coronary artery Maroko et al., J. Clin.
occlusion                        Invest., 3:661-670
                                 (1978)
Resistance to ticks              Wikel and Allen,
                                 Immunology, 34:257-263
                                 (1978)
Lung clearance of bacteria       Gross et al., J. Clin.
                                 Invest., 62:373-378
                                 (1978)
Immune complex disease in the lung Roska et al., Clin.
                                 Immunol. Immunopathol.,
                                 8:213-224 (1977)
Migration of T and B lymphocytes into Spry et al., Immunology,
lymph                            32:947-954 (1977)
Leukocyte circadian variation    Hoopes and McCall,
                                 Experientia, 2:224-226
                                 (1977)
Initial gingivitis               Kahnberg et al., J.
                                 Periodont. Res., 5:269-
                                 278 (1976)

[0090] A more recent example where complement depletion in laboratory animals has become important is the suppression of hyperacute rejection in xeno-transplantation, because complement has been shown to be a major player in hyperacute rejection. Another more recent application is the depletion of complement activity in gene therapy when retroviruses are used as a vehicle for gene transfer to the target cells. Retroviruses are lysed by complement and can survive in complement-depleted serum.

[0091] In addition to the use of recombinant proCVF to deplete complement in laboratory animals, recombinant pro-CVF may also be used as a therapeutic agent in humans for the treatment of cancer. One application is the covalent coupling of CVF to monoclonal antibodies with specificity for a tumor surface antigen. By coupling proCVF to such an antibody, complement activation is targeted to the tumor cell which is subsequently lysed by the complement system. Thus, procvF may be used for antibody targeting to tumor cells. Since, proCVF is insensitive to factor H control, this method will lead to the selective destruction of the cancer cells. In addition, single chain proCVF molecule may be used in a fusion chimeric protein with an scFv-fragment of an antibody for the same purpose.

[0092] The property of the CVF,Bb enzyme to exhaustively activate complement has also been exploited for the selective killing of tumor cells by coupling of CVF to monoclonal antibodies with specificity for surface antigens of tumor cells. Antibody conjugates with CVF will target CVF to the cell surface, at which the CVF,Bb enzyme forms from complement factors B and D of the host complement system. The antibody-bound and, therefore, cell surface-bound CVF,Bb enzyme will continuously activate C3 and C5 and elicit complement-dependent target cell killing. Antibody conjugates with CVF have been shown to kill human melanoma cells (Vogel and Müller-Eberhard, Proc. Natl. Acad. Sci. USA, 78:7707-7711 (1981); Vogel et al., Modern Trends in Human Leukemia VI, Neth et al, eds, Springer Verlag, Berlin, pp. 514-517 (1985)), human lymphocytes and leukemia cells (Muller et al., Br. J. Cancer, 54:537 (1986); Müller and Müller-Ruchholtz, Immunology, 173:195-196 (1986); Müller and Müller-Ruchholtz, Leukemia Res., 11:461-468 (1987)), and human neuroblastoma cells (Juhl et al., Proc. Am. Assoc. Cancer Res., 30:392 (1989); Juhl et al., Mol. Immuno., 27:957-964 (1990)).

[0093] An additional clinical use for proCVF is the use of proCVF to deplete complement in patients undergoing xenotransplantation to suppress the hyperacute rejection of the foreign organ. Another clinical use is the temporary depletion of complement in patients undergoing gene therapy using retroviral vectors. In addition, there is a host of other diseases where complement is known to be involved in the pathogenesis of disease, and where depletion of complement by proCVF might be of clinical use such as the diseases with circulating immune complexes (e.g. rheumatoid arthritis, lupus erythematosus, septic shock, adult respiratory distress syndrome, ischemia-reperfusion injury, and thermal injury from burns (see: F. D. Moore, Jr., et al, in Therapeutic Immunology, K. F. Austen, et al, Eds., Blackwell Science, Cambridge Mass., 1996, which is incorporated herein by reference).

[0094] Thus, the present invention also provides a method for complement depletion in animals by administration of proCVF. The animal may be any vertebrate, such as reptiles, fish, birds (chickens, turkeys, etc.), and mammals such as guinea pigs, mice, rats, pigs, baboons, chimps, dogs, cats, horses, cows, and humans. The proCVF may be administered by injection (intravenous or intraperitoneal) or by slow drip intravenous administration. The proCVF will typically be administered in the form of a sterile saline solution containing 10 to 2000 U/ml of proCVF, preferably 100 to 500 U/ml of proCVF, where a unit of proCVF has the same activity as a unit measured by the method of Ballow (see M. Ballow et al, J. Immunol., vol. 103, p. 944 (1969) which is incorporated herein by reference).

[0095] In the case of a guinea pig an appropriate single dose for complete decomplementation for 7 days is about 300-800 U/kg of body weight. In the case of humans, an appropriate dosage of proCVF is 1 to 1000 U/kg of body weight, preferably 30 to 80 U/kg of body weight, for complete deomplementation. for some applications, multiple injections might be useful to prolong the period of decomplementation, e.g., four 80 U/kg injections in an interval of 60 hours divided into 4 injections spaced over a period of 24 hours.

[0096] Other features of the invention will become apparent in the course of the following descriptions of exemplary embodiments which are given for illustration of the invention and are not intended to be limiting thereof.

EXAMPLES

[0097] I. CVF1 and CVF2:

[0098] Materials and Methods

[0099] Materials:

[0100] Solutions for RNA isolation, λgt11 cloning, and hybridization probe labeling were obtained from Amersham (Skokie, Ill.). In addition, an RNA isolation kit was purchased from Stratagene (La Jolla, Calif.). Reagents for cDNA preparation were obtained either from Gibco-BRL (Gaithersburg, Md.), or from Amersham. Oligo dT-cellulose was from Boehringer Mannheim (Indianapolis, Ind.), or from Invitrogen (San Diego, Calif.). Restriction enzymes were from either Pharmacia, from New England Biolabs (Beverley, Mass.), or from Gibco-BRL. Plasmids pUC18 and pUC19 were purchased from Boehringer Mannheim (Indianapolis, Ind.), while M13mpl8 and M13mpl9 were purchased from New England Biolabs. DNA modification enzymes were obtained from Pharmacia, New England Biolabs, or Gibco-BRL, and DNA sequencing reagents were obtained from United States Biochemicals (Cleveland, Ohio). Reagents required for PCR amplification of venom gland library ensens were obtained from Perkin-Elmer Cetus. Oligonucleotides for screening the libraries were obtained from Clontech (Palo Alto, Calif.), or were synthesized, using an Applied Biosystems #380 DNA synthesizer. A GeneCleanII kit, containing reagents used for isolation of DNA from agarose gels, and for the purification of PCR products, was obtained from Bio101 (La Jolla, Calif.). Nitrocellulose and nylon membranes for plaque lifts were obtained from Schliecher and Scheull (Keene, N.H.). Rabbit anti-goat IgG (Alkaline Phosphatase conjugated) was obtained from Sigma (St. Louis, Mo.). [α-32P]dATP, [α-32P]dCTP, and [α-35S ]dATP were obtained from Amersham.

[0101] Methods:

[0102] RNA Isolation from Cobra Venom Glands:

[0103] Adult cobras (Naja naja, 1.5-2 meters in length) were anesthetized with katamine (70 μg/kg i.m.) and with halothane/oxygen by intubation, essentially as described (Vogel, C.-W., et al, 1985, Dev. Comp. Immunol. 9:311). Venom glands were removed and immediately frozen in liquid nitrogen. For RNA preparation, approximately 1 gram of tissue was suspended (while frozen) in 20 ml of a solution of 4 M guanidinium thiocyanate and 1.14 M β-mercaptoethanol, and the RNA extracted according to the instructions supplied with the Amersham RNA Extraction Kit. This procedure was based on a published procedure (Han, J.H., et al, 1987, Biochemistry 26:1617). Poly-A containing RNA was then isolated by chromatography over oligo-dT cellulose, (Jacobson, A., 1987, In Methods in Enzymology, Vol. 152. S. L. Berger and A. R. Kimmel, eds. Academic Press, Orlando, Fla., p. 254). Whole RNA was also prepared using the Stratagene RNA isolation kit, in which the organs were homogenized in the presence of guanidinium isothiocyanate and β-mercaptoethanol, followed by phenol extraction and isopropanol precipitation (Chomczynski and Sacchi, 1987). Following extraction, poly A+ RNA was prepared by chromatography over oligo-dT cellulose, as described above.

[0104] CDNA Synthesis and Cloning:

[0105] Cobra venom gland CDNA was synthesized (Krug, M. S., et al, 1987, In Methods Enzymology. Vol. 152. S. L. Berger and A. R. Kimmel, eds. Academic Press. Orlando, Fla., p. 316; Gubler, U., 1987, In Methods in Enzymology, vol. 152. S. L. Berger and A. R. Kimmel. eds. Academic Press, Orlando. Fla.. p. 330) using the CDNA synthesis kit from Amersham. CDNA was synthesized using both oligo-dt and random hexamers as the primers. CDNA was then prepared for cloning into λgt11 (Wu. R., et al, 1987, In Methods in Enzymology. vol. 152. S. L. Berger and A. R. Kimmel, eds. Academic Press, Orlando, Fla.. p. 343), and the recombinant λ clones were packaged (Hohn, B., et al 1977, Proc. Natl. Acad. Sci. USA 74:3259). E. coli Y1090 (r, m+) was used as the host for recombinant λgt11.

[0106] In addition, cDNA was prepared using Superscript (RNase H—) MMLV Reverse Transcriptase (Gerard et al, 1989). In this case, double stranded CDNA was sized on a 1% Agarose gel in TAE buffer. cDNA greater than 4.5 kb was excised from the gel, and the DNA extracted from the agarose using the GeneCleanII kit from Bio101. This cDNA was then cloned into the plasmid pSPORT (Chen and Segburg, 1985), and the recombinant plasmids transformed into E. coi DH5α competent cells.

[0107] screening of λgt11 Libraries:

[0108] Libraries were screened (Young, R. A., et al, 1983, Science 222:778; Huynh, T. V., et al, 1985, In DNA Cloning vol. 1. A Practical Approach. D. M. Glover. ed. IRL Press, Oxford, p 49). The primary antibody for screening the venom gland library was goat anti-CVF antiserum. Further plaque purification was done as described above, using successively lower plaque densities. Later screening was done by the hybridization protocol (Wahl, G. M., et al, 1987, In Methods in Enzymology, vol. 152. S. L. Berger and A. R. Kimmel, eds. Academic Press, Orlando, Fla., p. 415), using the clones derived from the antibody screening as probes. These probes were labeled with [α-32P]dATP or [α-32P]dCTP (Feinberg, A. P., et al, 1983, Anal. Biochem. 132:6). pSPORT libraries were screened using other cDNA clones as a probe.

[0109] Subcloning and DNA Sequence Analysis:

[0110] Clones containing CVF inserts were grown up on agarose plates and their DNA prepared as described (Maniatis, T., et al, 1982, Molecular Cloning-A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.). Inserts were prepared by EcoR1 digestion, followed by agarose gel electrophoresis. In some cases, λgt11 inserts were isolated using the polymerase chain reaction (PCR) on a Savant Model TC49 thermal cycler. In this case, λgt11 amplimers from Clontech were used as primers, and the inserts were amplified using the protocol supplied with the amplimers. This consisted of 30 cycles with a 15 sec. denaturing stop at 94° C., 15 sec. of annealing at 58° C. and a 1 minute extension at 72° C. in a 50 μl reaction mixture containing 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 2 mM MgCl2, 1 mM in each deoxynucleoside triphosphate, 1 μM in each primer, 1.25 U of Amplitaq® polymerase, and approximately 100 ng DNA to be amplified. Following amplification, the inserts were purified with the Bio101 GeneCleanil kit, the DNA digested with EcoRI, and electrophoresed through an agarose gel. In all cases, fragments were eluted from the gel using the IBI Electroeluter (Model UEA) or by using the GeneCleanII kit from Bio101. The DNA inserts were then ligated into pUC18 (Yanisch-Perron, C., et al, 1985, Gene 33:103) and transformed into E. coli JM105 to facilitate the production of large quantities of the insert. Subfragments of the CVF inserts were subcloned into M13mp18 or mp19 (Yanisch-Perron, C., et al, 1985, Gene 33:103) for sequence analysis. Sequencing was performed using the dideoxy-chain termination technique (Sanger, F., et al, 1977, Proc. Natl. Acad. Sci. USA 74:5463). The Sequenase Version 2.0 sequencing kit from U.S. Biochemicals (Tabor, S., et al, 1989, Proc. Natl. Acad. Sci. USA 86:4076) was used as the source of enzymes, chemicals and primers for sequencing. The DNA sequence was assembled and analyzed using the group of sequence analysis programs written by the Genetics Computer Group of the Wisconsin Biotechnology Center (Devereux, J. R., et al, 1984, Nucleic Acids Res. 12:387).

[0111] Results

[0112] Screening of the Cobra Venom Gland Library.

[0113] Poly-A+ RNA from cobra venom glands was used for the preparation of cDNA that was cloned into the EcoRI site of λgt11. Libraries were prepared from cDNA that had been primed with oligo-dT and with random hexamers. Each library contained at least 5×106 clones. Initially, 5×105 clones from the random primed library were screened using CVF specific antisera to detect clones producing CVF containing fusion proteins. In the first round of screening, a single positive clone (CVF5, 1.1 kb, FIG. 1) was isolated. Sequence analysis of this clone revealed that it contained a single open reading frame of 639 nucleotides comprising the C-terminal 213 amino acid residues of the β-chain, and an 3′-untranslated region of 502 nucleotides. This represented approximately 12% of the mature protein.

[0114] To obtain clones representing the rest of the CVF message, several strategies were used. First, the oligo-dT primed library was screened by using hybridization, using CVF5 as a probe. This resulted in the isolation of several clones, one of which (CVF18, 2.6 kb) was used for further sequence analysis. CVF18 contains the 3′-end of the CVF message, with an open reading frame of 1002 nucleotides, and 3′ untranslated region of 994 nucleotides. Hybridization screening of an oligo-dT primed venom gland library yielded the clone CVF106 (3.5 kb). Clones containing the 5′ end of the CVF cDNA were isolated by screening the random primed λgt11 venom gland library by hybridization, using upstream restriction fragments of sequenced DNA as probes. By this means, two additional clones, CVF65 (3.4 kb) and CVF72 (2.0 kb) were isolated for sequencing. FIG. 1 shows the placement of the clones used for sequencing on the CVF1 mRNA.

[0115] Structure of the Cobra Venom Factor cDNA.

[0116] The CVF1 CDNA is 5948 nucleotides in length. It contains a single open reading frame of 4926 nucleotides, coding for a prepro-protein of 1642 amino acid residues (FIG. 2). The cDNA has a 5′ untranslated region of 3 nucleotides, and a 3′ untranslated region of 1019 nucleotides, including a poly-A tail of 20 bases. The coded pre-pro-protein has a signal sequence of 22 residues with a core rich in hydrophobic amino acids. The signal sequence is followed by the 627 amino acid a-chain. The α-chain has three glycosylation sites at residues 131, 136, and 187. Immediately following the C-terminus of the α-chain, there are 4 arginine residues, and a 68 amino acid peptide resembling the C3a anaphylatoxin. There is a single glycosylation site at position 640, though this site is not present in the mature protein. The γ-chain begins at position 710, and extends for approximately 300 amino acid residues. The position of the C-terminus of the γ-chain is unknown, and is apparently heterogeneous. The γ-chain contains no glycosylation sites. The β-chain of CVF begins at position 1242, and extends for 378 residues to the end of the open reading frame. The β-chain contains a single glycosylation site at position 1324.

[0117] The G+C composition of the open reading frame for CVF1 is 43.5% (for the whole cDNA: 42.4%). This is approximately the same as cobra C3, though lower than that for sequenced mammalian C3s.

[0118] Homology to Cobra and Other C3 Proteins.

[0119] The CVF1 sequence was compared to C3 sequences from cobra, human, and mouse. CVF1 shows a high degree of homology to cobra C3 at both the nucleic acid and protein level. At the protein level, CVF1 is nearly 85% identical to cobra C3 (greater than 91% similar if conservative replacements are allowed), while the nucleic acid sequences of the two messages are greater than 93% identical. CVF1 also shows a high, though lesser degree of homology to human and mouse C3 sequences. For example, the protein sequence of CVF is nearly 50% identical to that of human C3 (more than 69% similar if conservative replacements are allowed), while the nucleic acid sequence is nearly 57% identical. Comparing the CVF sequence to that of mouse C3, we find that the protein sequences are more than 51% identical (70% similar), and the nucleic acid sequences are nearly 58% identical. Dotplot comparisons of the CVF1 protein sequence with that of cobra and human C3 show that the homologies are spread throughout the molecule.

[0120] The homology between CVF1 and mammalian C3s is markedly higher than the average at certain ligand binding sites. For example, at the Factor B binding site, near the N-terminus of the γ-chain, CVF is 90% similar (though it is only 30-40% identical) to the homologous regions of other sequenced C3s (FIG. 3). The homology is also quite high at the properdin binding site (FIG. 4), where greater than 53% of the amino acid residues are identical, and about 80% of the amino acids are similar.

[0121] Interestingly, some sites are well conserved, even though they are not present in the mature protein. The best example of this is the sequence around the internal thioester site, where approximately 70% of the sequence is identical, and 80% is similar (FIG. 5). At the Factor H binding site, which is also not present in the mature protein, the homology is not noticeably greater than in the rest of the protein (FIG. 6). However, there is a stretch of 9 amino acid residues in the second portion of the discontinuous Factor H binding site (1192-1200 in CVF1) that is strictly maintained in all of the sequences examined, including rat and rabbit (data not shown), implying conservation of at least part of the Factor H binding site.

[0122] Additional clones encoding a distinct CVF were also isolated. The protein encoded by this gene is referred to as CVF2, and a partial sequence for CVF2 is shown in FIG. 10. The clones containing the DNA of CVF2 were obtained in the same series of experiments which gave the DNA for CVF1.

[0123] Expression of CVF clones in eukaryotic cells. Transient expression studies of CVF are done by transfecting CHO of NIH 3T3 cells with CVF sequences cloned into the mammalian expression vector pMT2, containing the SV40 origin of replication and early gene enhancer, the adenovirus major late promoter fused to the adenovirus tripartite leader, a hybrid intron splice site, and the SV40 polyadenylation signal. The CVF cDNA is ligated into the unique EcoRI site that is between the intron and the polyA addition site, and recombinant plasmids are transformed into E. coli DH5α. Recombinant clones are checked for the orientation of the insert by restriction analysis. Plasmids containing the CVF insert in the proper orientation are isolated and purified by two rounds of isopycnic CsCl2 gradient centrifugation, and transformed into COS cells by calcium phosphate mediated transfection. The transformed cells are then grown for 24 hrs, and both the cells and the media are assayed for the CVF production by Western analysis as described above.

[0124] For production of larger quantities of CVF, the baculovirus expression system is used. In this system, a plasmid containing the gene to be expressed is co-transfected into Spodopera frugiperda (Sf9) cells, along with the wild type Autographica californica nuclear polyhedrosis virus (AcNPV). Following transfection, up to 90% of the wild type viruses acquire the gene to be expressed by homologous recombination.

[0125] II. Recombinant ProCVF:

1.	Methods and Materials
1.1	Buffers and Solutions
	CAPS blotting	10	mM	CAPS (2-
	buffer			[Cyclohexylamino]-
				1-propanesulfonic acid)
		10%	(v/v)	Methanol,
				adjust to pH 11.0
				with NaOH
	Coomassie R250	0.25%	(w/v)	Coomassie brilliantblue
				R250
	staining sol'n	45%	(v/v)	Methanol
		45%	(v/v)	H2O
		10%	(v/v)	Acetic acid,
				filtered through paper filter
				before usage.
	Coomassie R250	45%	(v/v)	Methanol
	destaining sol'n	45%	(v/v)	H2O
		10%	(v/v)	Acetic acid
	Colloidal	0.1%	(w/v)	Coomassie Brilliant Blue
				G (Sigma)
	Coomassie stain	2%	(v/v)	Phosphorous acid
	premix	15%	(w/v)	Ammonium sulfate,
				stored at 4° C. in the dark,
				shake well before usage.
	Colloidal	80%	(v/v)	Colloidal Coomassie
				stain premix
	Coomassie	20%	(v/v)	Methanol
	staining sol'n			should be prepared fresh
				and shaken well before
				usage
	Colloidal	50%	(v/v)	Methanol
	Coomassie	40%	(v/v)	H2O
	detain. sol'n	10%	(v/v)	Acetic acid
	DNA loading	20%	(w/v)	Ficoll 400
	buffer (5x)	100	mM	EDTA
		0.025%	(w/v)	Bromphenolblue
		0.025%	(w/v)	Xylenxyanol FF
	Virus Extraction	0.1	M	Tris-HCl, pH 7.5
	Buffer	0.1	M	Na2EDTA
		0.2	M	Potassium chloride
	GVBS++	2.5	mM	Sodium barbital
				(Sodium-5,5-
				diethylbarbituric
				acid)
		143	mM	Sodium chloride
		0.75	mM	Magnesium chloride
		0.15	mM	Calcium chloride
		0.1%	(w/v)	Gelatine,
				pH 7.5
	LB Medium	1%	(w/v)	Tryptone
		0.5%	(w/v)	Yeast extract
		1%	(w/v)	Sodium chloride,
				adjust pH to 7.0. For agar
				plates, add 1.5% (w/v) agar
	Silverstain	30%	(v/v)	Ethanol
	fixing sol'n	15%	(v/v)	Acetic acid
	Silverstain	25%	(v/v)	Ethanol
	incubation sol'n	0.5	M	Sodium acetate
		2.5	mM	Sodium thiosulfate
		0.1%	(v/v)	Glutaraldehyde
				(25% aq. sol.)
	Silverstain	0.1%	(w/v)	Silver nitrate
	staining sol'n	0.006%	(v/v)	Formaldehyde solution
				(37%)
	Silverstain	2.5%	(w/v)	Sodium carbonate
	developing sol'n	0.006%	(v/v)	Formaldehyde solution
				(37%)
	Silverstain stop	50	mM	Na2EDTA
	sol 'n
	TAR buffer (50x)	24.2%	(w/v)	Tris base
		5.71%	(v/v)	Acetic acid
		10%	(v/v)	0.5M EDTA, pH 8.0,
				adjust pH to ˜8.5.
	TBE buffer (10x)	10.8%	(w/v)	Tris base
		5.5%	(w/v)	Boric acid
		4%	(v/v)	0.5M EDTA, pH 8.0
	TE buffer	1%	(v/v)	1M Tris, pH 7.4,
				7.6 or 8.0
		0.2%	(v/v)	0.5M EDTA, pH 8.0
	TBS (Tris	20	mM	Trig base
	buffered saline)	500	mM	Sodium chloride,
				adjust to pH 7.5
	TBS (5x)	100	mM	Tris base
		1.5	M	Sodium chloride,
				adjust to pH 7.5,
				dilute to
				1x before usage
	Transfection	25	mM	Hepes, pH 7.1,
	Buffer	140	mM	Sodium chloride
		125	mM	Calcium chloride
	TSS	85%	(v/v)	LB-medium
		10%	(w/v)	PEG-8000
		5%	(v/v)	DMSO
		50	mM	Magnesium chloride,
				pH 6.5
	VBS (Veronal	2.5	mM	Sodium barbital
	buffer saline)			(Sodium-5,5-
				diethylbarbituric
				acid)
		143	mM	Sodium chloride, pH 7.5
	VBS++ (Veronal	2.5	mM	Sodium barbital
	buffer saline)			(Sodium-5,5-
				diethylbarbituric acid)
		143	mM	Sodium chloride
		0.75	mM	Magnesium chloride
		0.15	mM	Calcium chloride,
				pH 7.5

[0126] 1.2 Enzymes

[0127] New England Biolabs:

[0128] KpnI, SacI, BanII, NotI, BamHI, DraI, Sai1I, Bsu36I, mungbean nuclease

[0129] MBI Fermentas:

[0130] Ec1 136 II, Mun I, SmaI, EcoRI, Kpn2I, Bsp119I, CIAP (calf intestinal alkaline phosphatase)

[0131] Amersham:

[0132] PshAI

[0133] 1.3 Construction of full-length CVF CDNA clones

[0134] For the expression of recombinant proCVF two clones were constructed using partial CVF CDNA clones from a λgt11-library. The resulting pCVF-FL3Δ represents the entire CVF CDNA including the signal sequence. pCVF-VL5 also represents the entire sequence, but the signal sequence of CVF was truncated.

[0135] Experimental:

[0136] Two of the parental clones (CVF72 and pCVF106) for the construction of full-length CVF cDNA were described previously (D. C. Fritzinger et al, Molecular Cloning and Derived Primary Structure of Cobra Venom Factor, PNAS, vol. 91, pp. 12775-12779 (1994)) and above. CVF72 is a λgt11-clone containing a 2 kb 5′-end fragment of the CVF cDNA. pCVF106 is a clone in pSPORT1 vector (Gibco BRL) containing a 3 kb 3′-end fragment of CVF. pCVF65/9 is a pUC18 clone containing a EcoRI-fragment from the λgt11-clone CVF65, which was also described previously (D. C. Fritzinger et al, Molecular Cloning and Derived Primary Structure of Cobra Venom Factor, PNAS, vol. 91, pp. 12775-12779 (1994)). pCVF65/9 spans the CVF cDNA sequence from bp 849 to bp 3350 of the CVF cDNA sequence.

[0137] First the insert of λgt11-clone CVF72 was subcloned into pUC18 vector: λ-DNA was prepared using standard methods from CVF72 and digested with SacI and KpnI. A resulting 4 kb fragment with 1 kb vector sequences on each site of the CVF insert was separated by agarose gelelectrophoresis, eluted from the gel, ligated into SacI, KpnI-digested pUC18 and transformed into E. coli DH5a resulting in the plasmid construct pCVF72. pCVF72 was digested with Ecl136II and BanII, treated with mungbean nuclease, religated and transformed into E. coli DH5α. This procedure removed the λ-vector sequence and the CVF signal sequence from the 5′-end of the CVF insert and generated a new Ecl136II-site at the 5′end of the CVF cDNA sequence. This plasmid construct was named pCVF72mut.

[0138] The plasmid clones pCVF65/9 and pCVF106 were both digested with SalI and MunI. The proper fragments were separated by agarose gelelectrophoresis, eluted from the gel, ligated together and transformed into E. coli DH5α resulting in the plasmid construct pCVF-MK1.

[0139] For the construction of pCVF-FL3Δ, pCVF-MK1 was digested with BamHI and NotI and treated with CIAP. The 4.3 kb CVF fragment spanning bp 1711 to the end of the CVF sequence was separated by agarose gelelectrophoresis and eluted from the gel. pCVF72 was digested with DraI. DraI digests the pUC18 vector about 1 kb upstream of the CVF insert creating a blunt end. Subsequently, the construct was digested with BamHI, the proper fragment separated by agarose gelelectrophoresis and eluted from the gel. This fragment was ligated to the pCVF-MK1 fragment. An ATP-concentration of 5 mM prevents dimerization. The resulting fragment was digested with SalI, subcloned in NotI/SalI precut psportI vector and transformed into E. coli DH5α resulting in the vector pCVF-FL3. To remove the λ-gt11 sequence, the vector has to be partially digested with EcoRI, because the CVF cDNA sequence has 3 EcoRI-sites itself. This was realized by digestion with a high amount of enzyme at room temperature for a short time (30 s). The truncated sequence missing the 1 kb λ-fragment was separated by gelelectrophoresis and religated. A unique SmaI-site is removed by the correct truncation. The background of other truncated molecules was reduced by digestion with SmaI. The fragments were transformed in E. coli DH5α. Plasmid DNA from a polyclonal culture was linearized by digestion with KpnI. Plasmids with a correct length of 10 kb were separated by agarose gelelctrophoresis, eluted from the gel, religated and again transformed in E. coli DH5α. Monoclonal cultures were screened by plasmid-miniprep, restriction mapping with several restriction enzymes and sequencing. A correct clone was obtained and named pCVF-FL3Δ. The CVF insert can be isolated by digestion with NotI and Kpn2I.

[0140] FIG. 11 shows an overview of this cloning strategy.

[0141] For the construction of pCVF-VL5 the plasmid-clones pCVF-M1 and pCVF72mut were both digested with BamHI. pCVF72mut was treated with CIAP to prevent selfligation. The proper fragments were separated by agarose gelelectrophoresis, eluted from the gel, ligated together and transformed into E. coil DH5α resulting in the plasmid construct pCVF-VL5. The CVF insert, spanning bp 72 to 5948 of the CVF CDNA, can be isolated by digestion with Ec1136I and NotI. Ecl136I creates a blunt end at the 5′-end of CVF.

[0142] FIG. 12 shows an overview of this cloning strategy. 1.4. Expression of Recombinant ProCVF in the Baculovirus Expression System

[0143] 1.4.1 Construction of ProCVF—Expression Vectors

[0144] Three vectors were constructed for secreted expression of proCVF. All vectors use the strong baculovirus polyhedrin promoter. pAc-CVF-secr is designed for secreted expression of full-length CVF with its natural signal sequence. For pAc-CVF-secr-3′His a stretch of six histidine residues is added to the 3′-end of proCVF to faciliate purification. Since the signal sequence of pre-proCVF differs from signal sequences in other species, it is not known whether it is recognized correctly in insect cell dependent expression systems. For pAcGP67-CVF, the secretory signal sequence of pre-proCVF was exchanged against the signal sequence from the baculovirus gp67 glycoprotein.

[0145] Construction of pAc-CVF-secr

[0146] The baculovirus transfer vector pVL1393 (PharMingen) was digested with XmaI and EagI, treated with CIAP and purified using agarose gelelctrophoresis. pCVF-FL3Δ was digested with NotI and Kpn2I and ligated into the digested pVL1393 vector. Kpn2I/XmaI and EagI/NotI have compatible cohesive ends. Correct clones were selected using restriction mapping and sequencing. A clone named pAc-CVF-secr was used for cotransfection of Sf9 insert cells.

[0147] Construction of pAc-CVF-3′HIS

[0148] pAc-CVF-secr was modified by placing a (His)6-Tag directly at the C-terminus of proCVF using the polymerase chain reaction. The following set of primers was used:

[0149] CVF106-OPS: GAGGAATTCAAGGTGC (base 3347-3362 of CVF CDNA)

[0150] CVF-3′HIS: AAGTTTAGCGGCCGCTTA(ATG)6 AGTAGGGCAGCCAAACTCAGT

[0151] NotI-site STOP (His)6—C-terminus of proCVF—

[0152] The following temperature program was used for the PCR:

[0153] First a denaturation step of 94° C. for 4 minutes. Subsequently, 30 cycles with 94° C. for 1 minute, 35° C. for 1 minute, 72° C. for 2 minutes. Finally 72° C. for 15 minutes. PCR was performed in a Hybaid OmniGene thermocycler under “tubecontrol”.

[0154] The 1.6 kb product was isolated from the reaction by ethanol precipitation and digested with Bsp119I and NotI. The resulting 373 bp fragment was isolated using 2% agrose gelelectrophoresis and ligated into Bsp119I/NotI-digested and CIAP-treated pCVF-FL3Δ vector. Correct clones were characterized by restriction mapping and sequencing. A correct clone was named pCVF-FL3Δ-3′HIS. This clone was digested with PshAI and NotI. The resulting 105 bp fragment was isolated using 2% agarose gelelectrophoresis and ligated into the PshAI/NotI-digested and CIAP-treated pAc-CVF-secr vector. Correct clones were characterized by restriction mapping and sequencing. A correct clone, named pAc-CVF-secr-3′HIS, is used for cotransfection into insect cells.

[0155] Construction of pAcGP67-CVF

[0156] The baculovirus transfer vector pAcGP67a (PharMingen) was digested with BamHI and blunted with mungbean nuclease. Subsequently, the the linearized vector was digested with NotI, treated with CIAP and purified using agarose gelelectrophoresis. pCVF-VL5 was digested with NotI and Ec1136II and ligated into the digested pAcGP67a vector. Correct clones were selected using restriction mapping and sequencing. A clone named pAcGP67-CVF was used for cotransfection of Sf9 insect cells.

[0157] 1.4.2 Insect Cell Culture

[0158] Cells

[0159] Spodoptera frugiperda (Sf9) cells (ATCC CRL 1711): Sf9 was cloned by G. E.Smith and C. L.Cherry in 1983 from the parent line IPLB-SF 21 AF, which was derived from pupal ovarian tissue of the fall armyworm Spodoptera frugiperda, by Vaughn, et al in 1977. The cell line is highly susceptible to infection with Autographa california MNPV and other Baculoviruses. (Ref.: J. L. Vaughn, et al, In Vitro, vol. 13, pp. 213-217 (1977); G. E. Smith et al, Proc. Natl. Acad. Sci. USA, vol. 82, pp. 8404-8408, (1985)).

[0160] Media

[0161] Grace's insect medium (T. D. C. Grace, Establishment of four strains of cells from insect tissues grown in vitro, Nature, vol. 195, pp. 788-789 (1962)) is the most common for growth of lepidopteran cells. TMN-FH medium (W. F. Hink, Established insect cell line from cabbage looper, Trichoplusia ni, Nature, vol. 226, pp. 466-467 (1970)) is Grace's basal medium supplemented with lactalbumin hydrolysate and yeastolate. Cell culture grade fetal bovine serum (FBS) is added to 10% (v/v) to make complete TNM-FH. Complete TMN-FH containing 50 μg/ml gentamicin sulfate was purchased from BioWhittaker. Protein-free medium was purchased from BioWhittaker (Insect Xpress). 10 μg/ml gentamycin (Gibco BRL) was used routinely in the medium of stock cultures. The addition of 2.5 μg/ml amphotericinB (“Fungizone”, Gibco BRL) is optional and is not done routinely.

[0162] Sf9 cells are shear sensitive and can be damaged permanently by handling during routine subculturing. When the surfactant, Pluronic F-68 (Gibco BRL) is added to a final concentration of 0.1% (w/v), shear sensitivity is significantly reduced. Especially for shaker-cultures and for culturing cells in serum-free medium, Pluronic F-68 is used.

[0163] Culturing Insect Cells

[0164] Carbon dioxide is not required. Cells are maintained at a constant 27±1.0° C. For expression of recombinant protein, the incubation temperature should be a constant 27±0.1° C. For monolayer cultures a B6120 Incubator (Heraeus), and for suspension an incubation skaker (Innova 4300, New Brunswick Scientific) with digital temperature control were used.

[0165] Fresh cell culture medium should be equilibrated to room temperature before use. As the cells divide, some might be either loosely attached or suspended in the medium (“floaters”). This is a normal occurrence and is often seen in older cultures and cultures which are “overgrown.” If “floaters” constitute more than 5% of the culture, the old medium containing the “floaters” is removed and replaced with fresh medium before subculturing.

[0166] Cells which take up trypan blue are considered nonviable. Sf9 cell density is determined using a hemacytometer (Neubauer, Germany). Cell viability can be checked by mixing 10 μl of trypan blue (0.4% stock solution made up in buffered isotonic salt solution, pH 7.2) and 10 μl of cells and examining under a microscope at low magnification. Cell viability should be at least 98% for healthy log-phase cultures.

[0167] Population doubling times for these cells will vary depending on growth conditions; as a general guide, healthy suspension cultures double in 18-22 hours. If cell doubling time exceeds 24 hours then there may be a problem with the cell viability, media, temperature, oxygenation, etc. Cells may be spun down at 1200 rpm for 10 minutes and resuspended in fresh media at a density of 1×106 cells/ml.

[0168] Thawing Sf9 Cells

[0169] A vial containing Sf9 insect cells is removed from liquid nitrogen and rapidly thawed with gentle agitation in a 37° C. water bath. When the contents are almost thawed, the outside of the vial is quickly decontaminated by treating with 70% ethanol. The vial is dried, and the 1 ml cell suspension is directly transfered into 4 ml of cold (+4° C.) complete TNM-FH media in a pre-wet 25 cm2 flask. The flask is transferred to a 27° C. incubator and the cells are allowed to attach for 30-45 minutes. Thawed cells do not always appear round; some may be amorphous or have a “wrinkled” appearance. In addition, there is usually a significant portion of debris associated with the cells. Moreover, some of the cells will not attach. Nonattached cells are termed “floaters” and should represent no more than 5% of the flask culture population when the cells are property maintained. The debris and floaters are reduced when cells are subcultured.

[0170] After the cells are attached, the media and any floaters are gently removed and transfered to a fresh 25cm2 flask as a back up. 5ml of fresh 27° C. media are added to the first flask, and both flasks flasks are incubated at 27° C. 24 hours later, the media in both flasks are changed. Viability of the cells will be greater than 70% when revived in this manner. The cells should be checked daily until a confluent monolayer has formed. Once a confluent monolayer has formed, cells can be subcultured.

[0171] Culturing Sf9 Cells in Monolayer Culture

[0172] All medium is removed from a confluent 175 cm2 flask and 10 ml of fresh media are added (2 ml for a 25 cm2 flask). To dislodge the cells, the flask is placed on end, the medium is drawn into a sterile pipette and rapidly discharged from the pipette while sweeping the tip of the pipette across the monolayer from side to side. The procedure should start at the bottom and end at the top of the flask. Alternatively, a soft cell scraper (Nunc #179707) can be used to dislodge the cells.

[0173] The 10 ml (2 ml) of culture produced should not be split in a ratio higher than 7. For a new culture, flask 2.0 ml (0.5 ml) of the culture are transfered into a new 150 cm2 flask containing 25 ml of fresh medium. The flask is gently rocked to wet the growth surface and distribute cells evenly. It is incubated at 27° C., and cells are allowed to grow to confluency.

[0174] It is important that cells are not passaged before confluency as they may be more difficult to dislodge, causing decreased viability.

[0175] Culturing Sf9 Cells in Suspension Culture

[0176] The cultures obtained from 3 to 4 confluent 175 cm2 flasks are combined into a 1000 ml spinner flask. The total medium volume in the spinner flask should be at least 100 ml at starting time, and the cell density should be at about 1×106 cells/ml. The 100 ml spinner is incubated at 27° C. with constant stirring at 80 rpm. Volumes >100 ml are stirred at 100-120 rpm in the presence of 0.1% Pluronic F-68 to increase aeration by diffusion and to provide protection from shearing.

[0177] The viability of the cells is checked every 24 hours. optimally, the cell density should be about 1×106 cells/ml with a viability of >98%. When the cells reach a density of ˜2×106 cells/ml, an equal volume of fresh medium is added, thus dropping the cell density to ×106 cells/ml again. This process is continued until reaching a 500 ml of culture at a density of ˜2×106 cells/ml. This culture is now ready for infection and large scale production of recombinant protein. For routine maintenance, spinner cultures should be subcultured when the cell density reaches about ˜2.5×106 cell/ml. Cells should be subcultured before their density reaches 4×106 cell/ml. However, the density of the cells should not drop below 1×106 cell/ml.

[0178] Adapting Sf9 Cells to Serum-free Conditions

[0179] Sf9 cells may be slowly adapted to serum-free medium by slowly decreasing the ratio of TMN-FH to protein-free medium stepwise. First, an almost confluent plate of Sf9 cells is split and two thirds TMN-FH/ one third of protein-free medium is added to the plate. Two passages later, the medium is changed to half and half TMN-FH and protein-free medium. Another two passages later, the ratio is changed to one quarter TMN-FH and three quarters of protein-free medium. After this, the medium is changed to 10% TMN-FH and 90% protein-free medium. Finally, pure protein-free medium can be used. Two to three passages on every step of this procedure should be allowed. Cells seem to be more healthy and grow quicker using this gentle adapting procedure instead of a rapid one.

[0180] Freezing Sf9 Cells from Complete Medium

[0181] Cells must be frozen in logarithmic growth phase (1.0-2.5×106 cells/ml) at 98% viability. Cells are removed from flasks and separated by centrifugation at 1200 rpm for 10 minutes (Sorvall RC-5B centrifuge with an SS-34 rotor, or equivalent).

[0182] The cells are gently resuspended in a volume that will result in a cell density of 1×107 cells/ml of 90% fetal bovine serum, 10% dimethylsulfoxide (DMSO) and aliquoted into cryogenic tubes. The tubes are stored at −20° C. for 1 hour and then transferred to an ultra-low freezer (−80° C.) overnight. The tubes are then transferred to liquid nitrogen for storage.

[0183] Freezing Sf9 Cells from Serum-free Medium

[0184] Cells are ready to freeze as soon as they are dividing regulary. Cells are detached from the flask by gentle sloughing with medium from a sterile pasteur pipet. Mid to late logarithmic cells (80-90% confluency, 2-3 days old) with a viability of ≧90%, as determinated by trypan blue dye exclusion, are recovered by centrifugation (1200 rpm, 5 minutes). The old (conditioned) medium should be saved. The cell pellet is resuspended to a cell density of 3×106 cells/ml in ice-cold, sterile, filtered freezing medium consisting of 45% conditioned medium, 45% fresh growth medium, and 10% dimethylsulfoxide (DMSO) Cells are then placed at 20° C. for 1 hour, transferred to an ultra-low freezer (−80° C.) overnight and finally stored in liquid nitrogen.

[0185] 1.4.3. Production of Recombinant Virus

[0186] Like many other eukaryotic systems, the BEVS has historically been less convenient and much more time-consuming than bacterial expression systems, especially the construction of the recombinant viruses. This limitation was largely overcome by the development of viruses having Bsu36I restriction sites positions within the essential gene, ORF1629, downstream of the AcNPV polyhedrin gene, and in the upstream ORF603, such that digestion releases a fragment containing a sequence necessary for virus growth. The necessary ORF1629 sequence is repaired and supplied by the transfer plasmid used for cotransfection. See: P. A. Kitts, et al, BioTechnique, vol. 14, pp. 810-817 (1993). The vast majority of survivors of cotransfection contain the repaired virus with the target gene, thus minimizing the need to screen large numbers of viruses. FIG. 13 summarizes this method. In addition, the polyhedrin gene in some of these virus strains was replaced with the β-galactosidase gene (lacZ) This facilitates selection between recombinant and wild type virus by using the chromogenic substrate X-gal. β-galactosidase expressing viruses will result in blue plaques during virus purification. Recombinant viruses will remain colorless. This “wild-type” virus is named wild-type (lacZ).

[0187] Wild-type Viral DNA Isolation and Linearization

[0188] Purification of Extracellular Virus DNA

[0189] Relatively pure viral DNA can be obtained from extracellular virus particles (ECV), which are separated from infected cell culture medium by centrifugation. Spodoptera frugiperda (Sf9) cells infected with AcMNPV (at 2×106 cells/ml) will yield about 1 μg of purified ECV DNA per ml of culture medium after 5-7 days p.i. The major problems encountered during purification are degradation of the DNA by mechanical shearing and contamination by nucleases. Also, if the DNA concentration is too high during purification, much of it can be lost during phenol extraction. These difficulties can be avoided with the following procedure adapted from Smith, G. B., and Summers, M. D. (G. E. Smith et al, Virology, vol. 123, pp. 393-406 (1982)).

[0190] Preparation of AcMNPV DNA

[0191] Approximately 500 ml of infected Sf9 cells (at 2×106 cells/ml) are required for the DNA preparation. For the infection, the cells are pelleted in a sterile tube by spinning for 5-10 minutes at 1000 rpm. The amount of titered virus needed to infect the cells at a Multiplicity of Infection (MOI) of 5-10 is calculated: 1$\mathrm{ml}\mathrm{of}\mathrm{virus}=\frac{\mathrm{MOI}\left(\mathrm{plaque}\mathrm{forming}\mathrm{units}/\mathrm{cell}\right)\times \mathrm{number}\mathrm{of}\mathrm{cells}}{\mathrm{titer}\left(\mathrm{pfu}/\mathrm{ml}\right)}$

[0192] The cells are gently resuspended in approximately 10 ml of complete TNM-FH containing the appropriate amount of virus (from the equation above) and incubated for 1 hour at room temperature with gentle rocking. Subsequently, the cells are transferred to a spinner flask with 500 ml fresh, complete TNM-FH to achieve again approximately 2×106 cells/ml. At 48 or more hours postinfection (hpi) the virus is harvested by pelleting the cells at 10,000 rpm for 10 minutes at +4° C. The virus-containing supernatant is transferred to ultracentrifuge tubes, and the virus is pelleted by spinning at 100,000 ×g for 30 minutes at +4° C. The viral pellets are resuspend in ˜1 ml of 0.1X TE. Half of the virus is overlayed onto each of two linear 25-56% (wt/vol) sucrose gradients prepared in 0.1X TE (typically 11 ml gradients in Beckman SW-41 tubes). The tubes are centrifuged at 100,000 ×g for 90 minutes at +4° C. Subsequently, the broad viral band is removed using a Pasteur pipette. The sucrose is diluted by adding at least 2 volumes of 0.1X TE, and the virus is repelleted by spinning for 60 minutes at 100,000 ×g.

[0193] The virus is resuspended in 4.5 ml Extraction Buffer (0.1 M Tris, pH 7.5, 0.1 M Na2EDTA, 0.2 M KCl), and the sample is digested with 200 μg proteinase K for 1-2 hours at 50° C. Subsequently, 0.5 ml of 10% Sarkosyl are added and incubated at 50° C. for at least two hours (or overnight, if convenient). This solution is extracted twice with phenol-chloroform/isoamyl alcohol (25:24:1). To minimize shearing of the viral DNA, extraction is performed by gently inverting the tubes just fast enough to mix the phases for several minutes. The phases are separated by low speed centrifugation and the aqueous phase is carefully removed by using a wide mouth 5 or 10 ml pipette. 10 ml of cold 100% ethanol are added to precipitate the viral DNA, and it is incubated at −80° C. for 30 minutes (or overnight at −20° C. if the DNA is not visible). The DNA is pelleted by spinning at 2500 rpm for 20 minutes, washed once with cold 90% ethanol, and resuspended in 0.1X TE. To facilitate the resuspension, incubation at 65° C. for about 10 minutes or overnight at +4° C.: might be necessary. Linearization of Wild-type Baculovirus DNA

[0194] Wild-type (lacZ) viral DNA is prepared as described above. 10 μg of viral DNA are incubated with 20 u of Bsu36I (New England Biolabs) for 2 hours at 37° C. Subsequently, another 20 u of Bsu36I are added followed by incubation overnight at 37° C. After inactivating the enzyme at 70° C. for 15 minutes, the digest was stored at 4° C. Aliquots of the digested and undigested viral DNA were run on a 0.5% agarose gel to check that the digest is complete. Uncut (circular) viral DNA does not enter the gel. For the digested viral DNA, a 3.3 kb band should be visible. At high concentrations, an additional 1.1 kb band can be detected.

[0195] Calcium Phosphate Mediated Transfection of Sf9 Cells

[0196] After cloning the CVF-CDNA into the transfer vector, at least 10 μg of highly purified plasmid DNA are prepared using standard techniques. Spodoptera frugiperda cells are sensitive to some contaminants found in crude plasmid preparations, which cannot be removed by phenol extraction or ethanol precipitation. A consistently reliable method for plasmid purification is CsCl-ethidium bromide gradient centrifugation. Impure preparations of plasmid DNA are toxic to the cells, and many cells may lyse shortly after transfection. This results in an apparently lower recombination frequency and increased difficulty in detecting recombinant viruses. In addition, the viral DNA quality (i.e., minimize nicking and shearing) is important too.

[0197] Plasmids containing foreign genes are cotransfected with wild-type (lacZ) ACMNPV DNA by the calcium phosphate precipitation technique, (F. L. Grahm et al, Virology, vol. 52, pp. 456-467 (1973)) as modified for insect cells (J. P. Burand, et al, Virology, vol, 101, ppg. 286-290 (1980); E. B. Carstens, et al, Virology, vol. 101, pp. 311-314 (1980); and K. N. Potter, J. Invertebr. Pathol., vol. 36, pp. 431-432 (1980)). The foreign gene is transferred to the AcMNPV genome in a subpopulation of the transfected cells by homologous recombination.

[0198] 2×106 Sf9 log Phase Sf9 cells (1.5-2.5×106 cells/ml; ≧98% viable) are seeded in complete TNM-FH in a 60 mm plate. The cells are allowed to attach for at least 30 minutes. 0.1 μg linearized AcMNPV (lacZ) DNA and 3-5 μg of plasmid DNA (the transfer vector containing the foreign gene) were mixed in a 1.5 ml polypropylene tube in a volume of 10 μl. 0.75 ml of Transfection Buffer are added, and pipetted once to mix. AcMNPV DNA is˜130 kb in length and is therefore very easy to shear. The Transfection Buffer may be stored at +4° C., but should be warmed to room temperature before transfection. The plate is tipped at a 45° angle, and all of the medium is aspirated with care from the cells using a Pasteur pipette. 0.75 ml of TNM-FH complete medium are added. The DNA solution is added dropwise to the cells. Then the plates are incubated at 27° C. on a side/side rocking platform slowly for 4 hours (setting 2.5 for a Bellco #774020020 side/side rocking platform). Following the incubation period, the medium is removed, and the cells are carefully rinsed with fresh complete TNM-FH. 3 ml of complete TNM-FN are added, and the cells are incubated at 27° C. In addition, one plate is infected with wild type virus (positive control), and one plate is just incubated with medium.

[0199] The cells are checked 4 days posttransfection to visually confirm a successful transfection. This is done using an inverted phase microscope a 250-400×magnification. Nearly all of the cells infected with wild type virus should contain viral occlusions, which appear as refractive crystals in the nucleus of the insect cell. This sign of infection should be absent for the cotransfected cells since the polyedrin gene is replaced. Other positive signs of virus infection include a 25-50% increase in the diameter of the cells, a marked increase in the size of the cell nuclei relative to the total cell volume (the nuclei may appear to “fill” the cells), and cell lysis and debris. In the late phase of infection, cells start to float. 10-50% of the cells in the cotransfected plate should have these signs.

[0200] Plaque Purification of Recombinant Virus

[0201] A 5% solution (w/v) of Baculovirus agarose (Invitrogen) in distilled water is prepared and autoclaved to dissolve the agarose and to sterilize the solution. The solution is incubated at 50° C. until needed. Aliquots of 45 ml of TNM-FH complete medium are stored at 50° C. until needed. 10-fold dilutions of virus inoculum (media harvested from transfections) are prepared in the range of the expected titer (1 ml of diluted virus for each plate). It is essential that the viral inoculum is vortexed vigorously prior to the preparation of the dilutions. Routinely, dilutions of 10−3 to 10−5 are plated when transfection is performed.

[0202] Sf9 cells at a density of 5×106 cells/100 mm plate are seeded in complete medium. Duplicate plates should be used for each virus inoculum to be tested. Plates are rocked at room temperature for at least 30 minutes on a side/side rocking platform in order to distribute the cells evenly. Use a setting of 5 for a Bellco #774020020 side/side rocking platform. The cells should be examined to confirm that they have attached. The plates are removed from the rocker and kept at room temperature for at least 30 minutes before using.

[0203] All but 2 ml of medium are removed from the cells once they have firmly attached. 1 ml of each dilution is added dropwise to the appropriately labelled plate. Following addition of the viral dilutions, the plates are incubated at room temperture on a slowly rocking platform for 1 hour (setting of 2.5 for a Bellco #774020020 side/side rocking platform). During this incubation period, a water bath is heated to a temperature of 46° C. and placed in the laminar flow hood. Just prior to the end of the 1-hour incubation, the 5 ml of the autoclaved agarose are added to a 45 ml aliquot of prewarmed medium, mixed, and placed in the 46° C. water bath. When plating involves selection between recombinant and wildtype virus, the chromogenic substrate 5-bromo-4-chloro-3-indolyl-β-D-galactoside (X-gal) or halogenated indolyl-β-D -galactosidase (Blu-gal, Bluo-gal) is incorporated into this pre-aliquotted medium at a concentration of 150 μg/ml. At the completion of the incubation period, the medium is completely removed from the plates. Then, working from the edge, 10 ml of the agarose/medium mixture are gently poured onto one of the plates from which the medium was removed. Care must be taken not to move move the plates until the agarose has set. The plates are incubated in a humid environment at 27° C. for 5-6 days, or until plaques well-formed. For 60 mm plates, 2×106 cell, an inoculation volume of 1 ml, and an agarose overlay of 4 ml are used.

[0204] Trypan Blue Overlay

[0205] On day 3-5 of the plaque assay, a second agarose overlay containing Trypan Blue is prepared. 1 ml of a sterile 1% Trypan blue solution (equilibrate to 40°-42° C.) is added to 12.5 ml of 1% agarose (also equilibrate to 40°-42° C.) and mixed well. The plates are overlayed with 1 ml of the Trypan Blue/agarose mixture per 60 mm plate (2 ml per 100 mm plate) and incubated overnight at 27° C. to allow the dye to diffuse into the dead cells. The number of blue plaques is counted to determine the viral titer. See H. Piwinica-Worms, in Current Protocols in Molecular Biology, Supplement 10, F. M. Ausubel et al Eds., John Wiley, New York, 1990.

[0206] The Trypan Blue method is useful for titering virus, but is not useful for selecting a recombinant plaque from a plate with wild-type and recombinant plaques—there is no distinction between plaque types. In addition, Trypan Blue is a known mutagen and not recommended for isolating recombinant virus.

[0207] Neutral Red Overlay:

[0208] On day 6-7 of the plaque assay, a second agarose overlay containing Neutral Red is prepared. 50 μl of a sterile Neutral Red solution (20 mg/ml) is added to 10 ml of 1.5% agarose (equilibrate to 46°-50° C.), mixed well, and kept at 46°-50° C. until ready to use. The plates are ovelayed with 2 ml of the Neutral Red/agarose mixture per 100 mm plate and incubated overnight at 27° C. to allow the dye to diffuse into the dead cells. The plaques will appear as morphologically distinct opaque areas on the pink/red monolayer. The number of plaques is counted to determine the viral titer.

[0209] The titer (pfu/ml) is calculated as follows:

[0210] pfu/ml=(1/dilution)×number of plaques.

[0211] The following formula can then be used to determine the Multiplicity of Infection (MOI): 2$\mathrm{ml}\mathrm{of}\mathrm{inoculum}\mathrm{needed}=\frac{\mathrm{MOI}\left(\mathrm{pfu}/\mathrm{cell}\right)\times \mathrm{number}\mathrm{of}\mathrm{cells}}{\mathrm{titer}\mathrm{of}\mathrm{virus}\left(\mathrm{pfu}/\mathrm{ml}\right)}$

[0212] Neutral Red is a known mutagen and not recommended for isolating recombinant virus.

[0213] Visual Screening for Recombinant Plaques

[0214] When plaques are distinct (at least 6 days postinfection), the plates are examined using a dissecting microscope with a magnification of 30-40×. Plates which have been infected with a dilution of virus resulting in well-seperated plaques in the parallel Neutral Red overlay are investigated first.

[0215] The plate is placed upside down on a nonreflective dark surface (e.g., black velvet or paper) and illuminated from the side using an intense light source (slide projector). The angle of the light is adjusted until the plaques can be observed (usually, a 45° angle or greater is best). Against a black, nonreflective background recombinant plaques will be of a dull milky-white color. Non-recombinant wild-type (lacZ) plaques are blue in color when the chromogenic substrate X-gal is added to the agarose. Plaques from both the recombinant and the wild-type (lacZ) strain do not develope polyhedrin occlusion bodies. Recombinant and wild-type (lacZ) plaques can not be distinguished without X-gal. Occlusion body positive wild-type plaques will not become blue but look shiny, almost crystal-like against a black, nonreflective background.

[0216] If plaques are dubious, the plate is scanned at a magnification of 30-40× and any plaques suspected to be recombinant, occlusion body negative are circled. Circled plaques are reexamined under an inverted phase microscope at 200-400×. Viral plaques are observed as a clear area in the cell monolayer which is ringed by infected cells which are morphologically distinct from the uninfected cells. They are generally larger in diameter, display at marked increase in the size of the nuclei relative to the total cell volume and show signs of cell lysis. The entire plaque area should be examined for the presence or absence of occlusion bodies. To avoid several rounds of screening, it is important that only recombinants are selected. Blue color of wild-type (lacZ) sometimes developes late. The sligthest hint for blue color under the microscope should be considered.

[0217] When several putative recombinant plaques have been located, the circled plaques are checked again using a dissecting scope. A tiny dot is placed within each circle, directly over the plaque to be picked. See: R. Dulbecco, et al, Proc. Natl. Acad. Sci. U.S.A., vol. 38, pp. 747-751 (1952); W. F. Hink, et al, J. Invertebr. Pathol., vol. 22, pp. 168-174 (1973); and H. H. Lee, et al, J. Virol., vol. 27, pp. 754-767 (1978).

[0218] Purification of the Recombinant Virus

[0219] Several 25 cm2 plates are seeded with 1×106 Sf9 cells each. The total volume of the plate should not exceed 3 ml. Using a sterile pasteur pipette and bulb, the agarose over the recombinant plaque is carefully removed, and the agarose plug containing the plaque is transferred to one of the plates. These steps are repeated for all putative plaques. Another plate acts as a cells-only control. The plates are incubated at 27° C. for 4 days. If cells grow confluent and start to float, the floaters and the medium are transferred to a new 75 cm2 plate. On day 3, the plates are visually screened for the presence of occlusion bodies. Any plate containing occlusion bodies is not plaque pure and requires additional rounds of purification.

[0220] The plates that are occlusion-negative should be kept and incubated at 27° C. until all of the cells lyse. The medium from these wells is harvested and stored at +4° C. This is the P1 virus stock for the generation of large-scale, high-titer virus stock. Each P1 stock is checked for the presence of wild-type (lacZ) by a plaque assay including X-gal selection. Only P1 stocks that are neither occlusion positive nor develope blue plaques in a plaque assay can be considered recombinant. An aliquot of the P1 may be used to carry out PCR analysis of the putative recombinant virus or Western analysis for the recombinant protein.

[0221] Virus Propagation

[0222] Preparing Large-scale, High-titer Virus Stocks

[0223] Once a recombinant virius has been identified, 100 μl of the P1 viral stock each are added to two 25 cm2 flasks seeded with 2×106 Sf9 cells. The flasks are incubated at 27° C. for 4-5 days, or until the cells are 90% lysed. This is the P2 inoculum. Two 1 ml aliquots of the P2 virus stock are removed. One may be placed at −80° C. for long-term storage, while the other is stored at +4° C. The remaining P2 virus stock obtained from the cultures in the two 25 cm2 flasks is used to infect 500 ml of Sf9 cells seeded at a density of 1.5-2.5×106 cells/ml. Cells are transferred in fresh medium to a spinner flask, and virus is added. 5 ml of the cell/virus suspension is transferred to a 25 cm2 flask to monitor the infection process. The culture is incubated at 27° C. with constant stirring (120 RPM) for 5-7 days. The progress of the infection is regularly checked by observing the 25 cm2 flask, or alternatively, by examining aliquots of the infected cell suspension with a microscope. When cells are 90% lysed, the culture medium (for future inoculum or protein purification) is collected by centrifugation, and the supernatant is transferred to a sterile bottle. This is now the P3 virus stock which can be titered (plaque assay for dilutions 10−5 through 10−8 as described above) and used to determine a time course of protein expression. The P3 virus stock is stored at +4° C.

[0224] It is recommended that all stocks be protected from light in order to ensure maintenance of titer (Bio/Techniques, vol. 116, no. 3, pp. 508-513).

[0225] Time Course for Production of Recombinant Protein

[0226] To optimize the level of protein production it is essential that an initial time course of expression be carried out. A 100 ml spinner flask is seeded with 50 ml of Sf9 cells at a density of 2×106 cells/ml and infected with P3 high-titer viral stock at a MOI of 5.1 ml aliquots of cells are removed every 12-24 hours over a period of 5 days and analyzed for the presence of the protein of interest. When the time point at which maximal expression is obtained, large-scale protein expression can be carried out

[0227] Seeding Densities

[0228] The chart below gives approximate seeding densities for typical vessel sizes. Infection at these densities will usually give high virus titers (≧1×108 PFU/ml).

		Minimum	Incubate in
Type of Vessel	Cell Density	Virus Volume	Final Volume
96-well plate	2.0 × 104/well	10	μl	100	μl
24-well plate	6.0 × 105/well	200	μl	500	μl
60 mm2 plate	2.5 × 106/flask	1	ml	3	ml
25 cm2 flask	3.0 × 106/flask	1	ml	5	ml
75 cm2 flask	9.0 × 106/flask	2	ml	10	ml
150 cm2 flask	1.8 × 107/flask	4	ml	20	ml
150 cm2 flask	1.5-2.0 × 106/ml	(based on	(based on
spinners (all)		MOI*)	size)
(*MOI = 0.5-1.0 for high-titer stocks and 5.0-10.0 for time course/protein expression).

[0229] 1.4.4 Production of Recombinant ProCVF

[0230] Large Scale Expression of Recombinant ProCVF

[0231] For large scale expression of recombinant proCVF,a 500 ml culture in 1 liter flasks of protein-free cultured Sf9 insect cells with a density of ˜2×106 cells/ml are infected with a baculovirus strain recombinant for CVF at a Multiplicity of Infection (MOI) of 5-10. The cultures obtained from 3 to 4 confluent 175 cm2 flasks are combined into a 1000 ml spinner flask. The total medium volume in the spinner flask should be at least 100 ml at starting time, and the cell density should be at about 1×106 cells/ml. The 100 ml spinners are incubated at 27° C. with constant stirring at 80 rpm. Volumes >100 ml are stirred at 100-120 rpm in the presence of 0.1% Pluronic F-68 to increase aeration by diffusion and to provide protection from shearing. The viability of the cells is checked every 24 hours. Optimally, cell density should be about 1×106 cells/ml with a viability of >98%. When the cells reach a density of ˜2×106 cells/ml an equal volume of fresh medium is added, thus dropping the cell density to 1×106 cells/ml again. This process is continued until reaching a 500 ml of culture at a density of ˜2×106 cells/ml. This culture is now ready for infection and large scale production of recombinant protein.

[0232] The amount of titered virus needed to infect the cells at a Multiplicity of Infection (MOI) of 5-10 is calculated: Approximately 500 ml of infected Sf9 cells (at 2×106 cells/ml) are required for the DNA preparation. 3$\mathrm{ml}\mathrm{of}\mathrm{virus}=\frac{\mathrm{MOI}\left(\mathrm{plaque}\mathrm{forming}\mathrm{units}/\mathrm{cell}\right)\times \mathrm{number}\mathrm{of}\mathrm{cells}}{\mathrm{titer}\left(\mathrm{pfu}/\mathrm{ml}\right)}$

[0233] The progress of the infection is regularly checked every 24 hours by examining aliquots of the infected cell suspension with a microscope. When the cells are 60% lysed (3.5-4 days after infection), the culture medium (for future inoculum or protein purification) is collected by centrifugation, and the supernatant is transferred to a sterile bottle.

[0234] Purification of Recombinant ProCVF Using Affinity Chromatography

[0235] This method was developed for isolation of recombinant proCVF from insect cell cultures infected with baculovirus strains expressing proCVF with the signal peptide from gp67a glycoprotein. Yields were found to be low for this strain.

[0236] At the fourth day of infection, the culture supernatant was collected by centrifugation, and phenylmethylsulfonyl fluoride (PMSF) was added to a final concentration of 2 mM. The supernatant was cooled to 4° C., diluted 1:1 with water, adjusted to pH 7.2, and filtered through a 0.45 μm cellulose acetate membrane (Sartorius). This solution was directly applied to a Highload-S cation exchange column (110×15 mm)(Bio-Rad) equlibrated in 4.3 mM phosphate (pH 7.2). Recombinant proCVF was eluted with a linear (0-300 mM) NaCl gradient. Fractions containing recombinant proCVF were identified by SDS-PAGE and Western blotting, pooled, concentrated, and dialyzed against 4.3 mM phosphate buffer (pH 7.2) using Amicon ultrafiltration system.

[0237] 5 mg of polyclonal rabbit anti-CVF antibody 1090 were immobilized on NHS-activated Sepharose® High Performance (1 ml HiTrapw affinity column, Pharmacia) according to the manufacturer's manual. For affinity absorption of recombinant proCVF, the recombinant proCVF pool from above is slowly (1 ml/min) applied to the column using a syringe or, for greater volumes, a peristaltic pump (P1, Pharmacia). The column is intensively washed with buffer A (0.1 M Tris-HCl, 100 mM NaCl pH 7.5). Recombinant proCVF is eluted with buffer B (0.1 M glycine, pH 2.5), and 0.5 ml fractions are immediately neutralized with 50 μl Buffer C (1M Tris-HC1, pH 9.0). Fractions containing recombinant proCVF were identified by SDS-PAGE and Western blotting and pooled. Purified recombinant proCVF was aliquoted and stored at −20° C.

[0238] Purification of Recombinant ProCVF Using Ion-exchange Chromatography

[0239] At the fourth day of infection, the culture supernatant was collected by centrifugation, and phenylmethylsulfonyl fluoride (PMSF) was added to a final concentration of 2 mM. The supernatant was cooled to 4° C., diluted 1:1 with water, adjusted to pH 7.2, and filtered through a 0.45 μm cellulose acetate membrane (Sartorius). This solution was directly applied to a Highload-S cation exchange column (110×15 mm) (Bio-Rad) equilibrated in 4.3 mM phosphate (pH 7.2). Recombinant proCVF was eluted with a linear (0-300 mM) NaCl gradient. Fractions containing recombinant proCVF were identified by SDS-PAGE and Western blotting, pooled, and directly applied to a Highload-Q anion exchange column (5×100 mm) (Bio-Rad) equilibrated in 4.3 mM phosphate (pH 7.2). Recombinant proCVF was eluted with a linear (0-500 mM) NaCl gradient. Fractions containing recombinant proCVF were identified by SDS-PAGE and Western blotting and pooled. Purified recombinant proCVF was aliquoted and stored at −20° C.

[0240] Protein Charaterization Methods

[0241] Western Blot Analysis

[0242] Cobra venom factor is immunogenic and, therefore, it is fairly easy to generate polyclonal antisera. Antisera has been raised in goats (C.-W. Vogel et al, J. Immunol Methods, vol. 73, pp. 203-220 (1984)), rabbits (C.OW. Vogel et al, J. Immunol Methods, vol. 133, ppg. 3235-3241 (1984)), and mice (A. H. Grier et al, J. Immunol., vol. 139, pp. 1245-1252 (1987)).

[0243] For detection of recombinant cobra venom factor, a polyclonal rabbit anti-CVF antisertun (AK-1900) was used. The antiserum was cleaned and concentrated by fractionated ammonium sulfate precipitation. Cold saturated ammonium sulfate solution was added to the antiserum to 28% saturation, and it was incubated on ice for thirty minutes. Precipitated protein was separated by centrifugation (3000 g, 40° C., 20 min.) and discharged. Ammonium sulfate was added to the supernatant up to a saturation of 50%. Precipitated antibodies were isolated by centrifugation and dissolved in PBS (10 mM sodium phophate, 140 mM NaCl, pH 7.4). The concentrated antibody is used in a dilution of 1:125.

[0244] Protein (1 μg) is electrophoresed on SDS-polyacrylamide gels under non-reducing or reducing conditions (U. K. Laemmli, Nature, vol. 227, pp. 680-685 (1970)) and electro-transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, Mass.) using CAPS blotting buffer. The membrane is blocked for at least 2 hours in 10% (w/v) milkpower/TBS. Subsequently, the membrane is washed twice with TBS for 5 minutes each. 20 μl of rabbit antiCVF antibody (AK1900) are added in 25 ml 5% (m/v) milkpower/TBS to the membrane and incubated for 1 hour. The membrane is washed two times with TBS and incubated with 25 ml 5% milkpower/TBS containing 2 μl anti-rabbit alkaline phosphatase conjugate (Sigma). After this, the membrane is washed twice with TBS and once with 0.1 Tris-buffer pH 9.5 (100 mM sodium chloride). The color reaction is performed in the same buffer containing 50 mM magnesium chloride, 250 μl BCIP (0.5% (w/v) in DMF) and 2500 μl NBT (0.1% (w/v) in Tris buffer) until bands become visible.

[0245] Assays For Cobra Venom Factor

[0246] Two assays have been developed to determine CVF activity. The first assay is based on the anticomplementary activity of CVF. The sample containing CVF (or recombinant proCVF) is incubated with a defined volume of normal serum (human or guinea pig). Subsequently, the remaining complement homolytic activity in the serum is determined in a hemolytic assay that uses sensitized sheep erythrocytes as targets (M. Ballow et al, J. Immunol., vol. 103, pp. 944-952 (1969); C. G. Cochrane et al, J. Immunol., vol. 105, pp. 55-69 (1970).

[0247] Sheep erythrocytes from whole sheep blood (Behringwerke) are separated by centrifugation (720 g, 4° C., 10 min.). The supernatant is discharged, and the erythrocytes are washed with GVPS++. Centrifugation and washing is repeated three to four time or until the supernatant remains clear. Resuspended erythrocytes are adjusted to a concentration of 5×108 cells/ml (20 μl lysed 1 ml H2O results in an OD412 =1.3) and incubated with rabbit anti-sheep antibodies (diluted 1:100, Sigma) at 37° C. for 30 minutes. Subsequently, the antibody-sensitized erythrocytes (ESA) are separated again by centrifugation, washed three times with GVPS++and adjusted to a concentration of 5×108 cells/ml. 10 μl sample containing CVF (or recombinant proCVF) is incubated with 10 μl serum (guinea pig or human) for an appropriate time (30 minutes to 3 hours) at 37° C. Controls included normal serum only, heat-inactivated serum, heat-treated CVF or heat-treated recombinant proCVF, VPS++buffer only, and 100% lysis using H2O. After the incubation time, 100 μl GVPS++and 30 μl ESA are added and incubated in 2 ml reaction tubes (Eppendorf, round bottom) in a Thermomixer 5437 (Eppendorf) at 37° C. with moderate shaking (#8). The incubation period can vary from 10 minutes to 30 minutes. Lysis in the control reaction with serum only should be about 80% of total lysis. Subsequently, the reaction is stopped by addition of 1 ml GVPS++, and unlysed erythrocytes are sedimented by centrifugation (2000 g, 4° C., 2 minutes) and released hemoglobin is spectrophotometrically determinated in the supernatant (412 nm).

[0248] A second assay for CVF is based on fluid-phase C5 cleavage and bystander lysis of unsensitized erythrocytes (C.-W. Vogel, et al, J. Immunol. Methods, vol. 73, pp. 203-220 (1984)). The sample containing CVF (or recombinant proCVF) is incubated with normal guinea pig serum and guinea pig erythrocytes. Guinea pigs are narcotized with Ketamin and Rumpun. About 2 ml whole blood is taken by heart puncture and added to 1 ml ACD-buffer. The erythrocytes are washed three times with GVPS++. Centrifugation and washing is repeated three to four times or until the supernatant remains clear. Resuspended erythrocytes are adjusted to a concentration of 5×108 cells/ml (20 μl in 1 ml H2O results in an OD412=1.3). 20 μl of the CVF containing sample are incubated with 20 μl of normal guinea pig serum and 20 μl of a guinea pig erythrocytes suspension for an appropriate time (30 minutes to 3 hours, depends on CVF concentration). Controls include normal serum only, heat-inactivated serum, heat-treated CVF/recombinant proCVF, addition of EDTA or Mg-EDTA, addition of specific CVF-inhibitor from cobra serum, VPS++buffer only, and 100% lysis using H2O. Subsequently, the reaction is stopped by the addition of 1 ml GVPS++and unlysed erythrocytes are sedimented by centrifugation (2000 g, 4° C., 2 min.) and released hemoglobin is spectrophotometrically determined in the supernatant (412 nm).

[0249] N-terminal Sequencing

[0250] N-terminal sequencing was performed on an ABI-476A Protein Sequencer. The N-terminus was sequeced following SDS-PAGE and electroblotting to ProBlott membrane and was found to be the same as that of the CVF α-chain (or the N-terminus of proCVF). This demonstrates that the signal peptide is probably removed, but that there is nearly no (<10%) cleavage in the (Arg)4 linkage between the CVF α- and γ-chain.

[0251] Glycosylation Analysis

[0252] Glycosylation is analysed using DIG Glycan Differentiation Kit (Boehringer Mannheim), based on lectin affinity staining. This kit is based on detection of carbohydrate structures with lectins of different carbohydrate specifity, listed in Table 1. The lectins are digoxigenin labeled and are detected by an anti-digoxigenin antibody conjugated with alkaline phosphatase. Alkaline phosphatase triggers a color reaction with chromogenic substrates.

[0253] Protein (1 μg) is electrophoresed on SDS-polyacrylamide gels under non-reducing or reducing conditions (U. K. Laemmli, Nature, vol. 227, pp. 680-685 (1970)) and electro-transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore). Blocking and detection of carbohydrate structures is performed according to the manufacturer's guidelines: All steps are performed at room temperature with moderate shaking. Only the color reaction is performed without shaking. The membrane is blocked for at least 30 minutes. Subsequently, the membrane is washed twice with TBS for 10 minute each, and once with TBS containing 1 mM magnesium chloride, 1 mM calcium chloride, and 1 mM manganese chloride (TBS*). The appropriate amount of lectin is added in 10 ml TBS* to the membrane and incubated for 1 hour. The membrane is washed three times with TBS and incubated with 10 ml TBS containing 10 μl anti-digoxigenin alkaline phosphatase conjugate. After this, the membrane is washed twice with TBS and once with 0.1 Tris-buffer pH 9.5 (50 mM magnesium chloride, 100 mM sodium chloride). The color reaction is performed in the same buffer containing 37.5 μl BCIP (50 mg/ml) and 50 μl NBT (75 mg/ml) for about 10 minutes.

[0254] For dot-blot analysis, 3 μl protein solution (1 μg) was directly applied to hydrophilic cellulose nitrate, membranes (BA-S 83 supported, Schleicher & Schull). This method in combination with a multiple incubation chamber (PR 150 Mighty Small Deca-Probe, Hoefer) is preferred for parallel analysis of several proteins with several different lectins. After sample application and blocking, the membrane is transferred into the incubation device. That divides the membrane into ten isolated lanes allowing parallel side-by-side incubation with up to ten lectins without the need to slice it into strips. The glycosylation analysis is performed as described above. Only 2 ml of lectine solution are necessary for each chamber.

[0255] After incubation with the lectins and a primary washing step the incubation device is not necessary anymore. It is more convenient to handle the following incubation steps with the non-separated membrane. All steps are performed according to the manufacturer's guidelines.

TABLE 1
Lectin                       Specificity
ConA                         α-Man
Concanavalia ensiformis      binds to mannose containing
agglutinin                   carbohydrates
GNA                          Man α(1-3) Man (α1-3 α1-6
Galanthus nivalis agglutinin α1-2)
                             binds to terminal mannose
                             linked to mannose,
SNA                          indicating high-mannose type
Sambusus nigra agglutinin    structure
                             Neu NAc α(2-6) Gal/GalNAc
MAA                          binds to sialic acid linked
Maachia amurensis agglutinin to galactose in both N- and
                             O-glycans
DSA                          Neu NAc α(2-3) Gal
Datura stramonium agglutinin binds to sialic acid linked
                             to galactose in both N- and
PNA                          O-glycans
Peanut agglutinin agglutinin
                             binds to Galβ(1-4) GlcNAc in
                             complex N-glycans and
                             GlcNac-Ser/Thr core in O-
                             glycans
                             Galβ(1-3) GlcNAc
                             binds to core disaccharide
                             of O-glycans

[0256] 2. Results

[0257] Expression and Purification of Recombinant ProCVF

[0258] Cell pellets and supernatants of insect cell cultures infected with CVF containing recombinant baculoviruses were analyzed by Western blot analysis using a polyclonal anti-CVF antiserum. For recombinant proCVF expression, an anti-CVF reactive 200 kDa protein could be detected which was absent in wild type-baculovirus infected cells or uninfected cultures. The highest amount of recombinant proCVF is obtained by expressing full-length CVF CDNA with its natural secretory signal sequence (Ac-CVF-secr and Ac-CVF-secr-3′His) in serum-free cultured Sf9 insect cells. Maximum expression rates were about 2 mg/l recombinant proCVF. Optimal expression was performed for 4 days at 27° C. in monolayer culture or spinner flasks with a multiplicity of infection (MOI) of 5-10. After 4 days, recombinant proCVF expression leveled. Prolonged expression times did not increase the yield of recombinant protein, but insect cells started to lyse intensively making purification more difficult (FIG. 14). Recombinant proCVF was purified by two-step ion-exchange chromatography. With this procedure, the yield of recombinant proCVF was approximately 1 mg from 1 liter of culture supernatant and the purity was >90%.

[0259] Chain Structure of Recombinant ProCVF

[0260] Purified recombinant proCVF analyzed electrophoretically under reducing and nonreducing conditions, differed from the behavior of natural CVF. Natural CVF resolves in three bands of 68.5 kDa, 48.5 kDa, and 32 kDa. The recombinant proCVF expressed and purified under normal conditions, is produced as the single-chain proCVF (FIG. 15) with a molecular weight of about 190-200 kDa.

[0261] The N-terminus was sequenced following SDS-PAGE and electroblotting to ProBlott membrane and was found to be the same as that of the CVF a-chain. This result strongly suggests that the signal peptide is removed but that there is near no (<10%) cleavage in the (Arg)4 linkage between the CVFα- and γ-chain.

[0262] Incubation of the cell-free supernatant after the expression at 4° C. for 4 days resulted in an in vitro-processing up to 100%. The resulting protein analyzed electrophoretically behaved similar to human C3 resolving into bands of 115 kDa and 65 kDa. The 115 kDa chain normally resolves to a double band indicating unspecific proteolysis. N-terminal sequencing of the separated 115 kDa bands was not possible to perform, probably due to heterogeneous N-termini.

[0263] Glycosylation of Recombinant ProCVF

[0264] Natural CVF from Naja naja kaouthia contains nearly exclusively asparagine-linked oligosaccharides. The major N-linked oligosaccharide is a symmetrical fucosylated biantennary complex-type structure terminating with an unusual α-galactosyl residue (D. C. Gowda et al, Mol. Immunol., vol. 29, pp. 335-342 (1992)).

[0265] For recombinant proCVF there is a strong reaction with Canavalia ensiformis agglutinin (ConA) and Galanthus nivalis agglutinin (GNA) indicating N-linked oligosaccharides of “high-mannose”-type (FIG. 16). A positive reaction with peanut agglutinin (PNA) indicates a O-linked glycosylation of the simple galactose-β(1-3)-N-acetylgalactosamine type. No complex O-glycosylation was detected by reaction with Datura stramonium agglutinin (DSA). No sialic acid was found in either natural and recombinant proCVF. These results are consistent with the glycosylation patterns normally found in insect cells. However, natural and recombinant proCVF differ in their glycosylation structure. The N-linked oligosaccharides seem to be more simple for the recombinant one. In contrast to natural CVF, recombinant proCVF seems to have a reasonable amount of O-glycosylation.

[0266] Glycosylation of natural CVF is not required for is complement-activating function. However, there are hints that the oligosaccharide structure contributes to the thermal stability of the molecule, because the deglycosylation causes CVF to be more sensitive to temperature (D. C. Gowda et al, J. Immunol., vol. 152, pp. 2977-2986 (1994).

[0267] Treatment of insect cells with tunicamycin, a strong inhibitor of N-linked glycosylation, results in a complete stop of section (FIG. 17) decreased expression, intracellular accumulation, and degradation of recombinant proCVF. These results are consistent with several others found for insect and mammalian cells. Glycosylation seems to be necessary for proper secretion of some proteins. Treatment of mammalian cell lines, expressing complement factors C2, C4 and B, with tunicamycin also results in a complete inhibition of secretion (W. J. Matthews, Jr. et al, Biochem J., vol. 204, pp. 839-846 (1982)).

[0268] Functional Activity of Recombinant ProCVF

[0269] To determine whether recombinant proCVF can activate complement, two assay systems were used. One assay is based on fluid-phase C5 cleavage and bystander lysis of guinea pig erythrocytes by complement activation in guinea pig serum from CVF. The other assay depends on complement consumption by incubation of guinea pig or human normal serum with CVF/recombinant proCVF. Remaining complement activity in the serum is measured by lysis of antibody-sensitized sheep erythrocytes.

[0270] The activity for proCVF expressed using the transfer vector pAC-CVF-secr and PAcGP67-CVF was found to be identical. Both assays demonstrate, that the recombinant proCVF has the same complement-activating activity of the natural protein (FIGS. 18 and 19). Even the thermal stability seems to be the same (FIG. 20).

[0271] However, the activity for proCVF expressed using the transfer vector pAc-CVF-secr-3′His differs from the activity of natural CVF. While the complement-consumption activity was present in at least the same level as that for the natural protein, activity in the bystander lysis assay was not detectable. These results demonstrate, that the modification of proCVF by addition of six histidine residues at the C-terminus of the proCVF deletes the ability of proCVF to cleave C5 in the fluid-phase.

[0272] 3. Conclusion

[0273] CVF is expressed in the baculovirus system as a single chain proprotein. Production of a broad spectrum of proteins in eukaryotes occurs via an intricate cascade of biosynthetic and secretory processes. Often these proteins initially are synthesized as parts of higher molecular weight, but inactive, precursor proteins. Specific endoproteolytic processing of these proteins is required to generate a mature and biologically active form. Such endoproteolysis generally occurs at cleavage sites consisting of particular sequence motifs of basic amino acids, often paired basic residues. Examples for enzymes with a cleavage specificity for basic amino acids are kexin from Saccharomyces cerevisiae or the mammalian enzyme furin (W. J. Van de Ven et al, Crit. Rev. Oncog., vol. 4, pp. 115-136 (1993)). Furin is expressed in a wide variety of tissues, perhaps even in all tissues. In all likelihood, it is the enzyme responsible for the proteolytic bioactivation of a wide variety of precursor proteins (e.g., proC3).

[0274] Surprisingly, for proCVF the lack of proteolytic processing does not have an effect on activity. This is a very rare, unpredictable event, since there are only a few cases for functionally active proproteins (B. P. Dunker et al, Biochem. Biophys. Res. Commun., vol. 203, pp. 1851-1857 (1994); N. Moguilevsky et al, Eur. J. Biochem., vol. 197, pp. 605-614 (1991); J. I. Paul et al, J. Biol. Chem., vol. 265, pp. 13074-13083 (1990).

[0275] It should be understood that proCVF may be prepared by expressing a single DNA sequence which encodes, e.g., pre-pro-CVF1, pro-CVF1, pre-pro-CVF2, or pro-CVF2, with any post translational processing being carried out as described above.

[0276] Obviously, numerous modifications and variations of the present invention are possible in light of the above teachings. It is therefore to be understood that, within the scope of the appended claims, the invention may be practiced otherwise than as specifically described herein.

Claims

1. A polypeptide, comprising an amino acid sequence, said amino acid sequence being selected from the group consisting of: (a) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2; (b) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (c) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues inserted between position −1 and position 1, (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2; (e) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with a methionine residue added to the amino terminus; (f) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus; (g) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the peptide of sequence II added to the amino terminus; (h) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (i) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus; (j) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus and a methionine residue added to the amino terminus; (k) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus and a methionine residue added to the amino terminus of the histidine residues; (l) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus and with 1 to 8 histidine residues added to the carboxy terminus; and (m) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 and with 1 to 8 histidine residues added to the amino terminus and the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus of the histidine residues; wherein sequence I (gp67 signal peptide) is: MLLVNQSHQGFNKEHTSKMVSAIVLYVLLAAAAHSAFA; and sequence II (peptide encoding for M start codon, 6 histidine residues, and enterokinase cleavage site) is MRGSHHHHHHGMASMTGGQQMGRDLYNNNNK.

2. The polypeptide of claim 1, which is: (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2.

3. An isolated segment of DNA comprising a DNA subsequence, said DNA subsequence encoding a polypeptide, said polypeptide comprising an amino acid sequence, said amino acid sequence being selected from the group consisting of: (a) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2; (b) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (c) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues inserted between position -1 and position 1; (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2; (e) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with a methionine residue added to the amino terminus; (f) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus; (g) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the peptide of sequence II added to the amino terminus; (h) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (i) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus; (j) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus and a methionine residue added to the amino terminus; (k) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus and a methionine residue added to the amino terminus of the histidine residues; (l) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus and with 1 to 8 histidine residues added to the carboxy terminus; and (m) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 and with 1 to 8 histidine residues added to the amino terminus and the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus of the histidine residues; wherein sequence I (gp67 signal peptide) is: MLLVNQSHQGFNKEHTSKMVSAIVLYVLLAAAAHSAFA; and sequence II (peptide encoding for M start codon, 6 histidine residues, and enterokinase cleavage site) is MRGSHHHHHHGMASMTGGQQMGRDLYNNNNK.

4. The DNA of claim 3, wherein said polypeptide is (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2.

5. A vector, comprising a segment of DNA, said segment of DNA comprising a DNA subsequence, said DNA subsequence encoding a polypeptide, said polypeptide comprising an amino acid sequence, said amino acid sequence being selected from the group consisting of: (a) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2; (b) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (c) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues inserted between position −1 and position 1; (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2; (e) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with a methionine residue added to the amino terminus; (f) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus; (g) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the peptide of sequence II added to the amino terminus; (h) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (i) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus; (j) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus and a methionine residue added to the amino terminus; (k) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus and a methionine residue added to the amino terminus of the histidine residues; (l) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus and with 1 to 8 histidine residues added to the carboxy terminus; and (m) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 and with 1 to 8 histidine residues added to the amino terminus and the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus of the histidine residues; wherein sequence I (gp67 signal peptide) is: MLLVNQSHQGFNKEHTSKMVSAIVLYVLLAAAAHSAFA; and sequence II (peptide encoding for M start codon, 6 histidine residues, and enterokinase cleavage site) is MRGSHHHHHHGMASMTGGQQMGRDLYNNNNK.

6. The vector of claim 5, wherein said polypeptide is: (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2.

7. A transformed host, comprising a heterologous segment of DNA, said segment of DNA comprising a DNA subsequence, said DNA subsequence encoding a polypeptide, said polypeptide comprising an amino acid sequence, said amino acid sequence being selected from the group consisting of: (a) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2; (b) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (c) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues inserted between position −1 and position 1; (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2; (e) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with a methionine residue added to the amino terminus; (f) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus; (g) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the peptide of sequence II added to the amino terminus; (h) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (i) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus; (j) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus and a methionine residue added to the amino terminus; (k) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus and a methionine residue added to the amino terminus of the histidine residues; (l) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus and with 1 to 8 histidine residues added to the carboxy terminus; and (m) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 and with 1 to 8 histidine residues added to the amino terminus and the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus of the histidine residues; wherein sequence I (gp67 signal peptide) is: MLLVNQSHQGFNKEHTSKMVSAIVLYVLAAAAHSAFA; and sequence II (peptide encoding for M start codon, 6 histidine residues, and enterokinase cleavage site) is MRGSHHHHHHGMASMTGGQQMGRDLYNNNNK.

8. The transformed host of claim 7, wherein said polypeptide is: (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2.

9. A method of preparing proCVF, comprising culturing a transformed host, said transformed host comprising a heterologous segment of DNA, said segment of DNA comprising a DNA subsequence, said DNA subsequence encoding a polypeptide, comprising an amino acid sequence, said amino acid sequence being selected from the group consisting of: (a) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2; (b) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (c) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues inserted between position −1 and position 1; (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2; (e) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with a methionine residue added to the amino terminus; (f) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus; (g) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the peptide of sequence II added to the amino terminus; (h) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (i) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus; (j) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus and a methionine residue added to the amino terminus; (k) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus and a methionine residue added to the amino terminus of the histidine residues; (l) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus and with 1 to 8 histidine residues added to the carboxy terminus; and (m) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 and with 1 to 8 histidine residues added to the amino terminus and the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus of the histidine residues; wherein sequence I (gp67 signal peptide) is: MLLVNQSHQGFNKEHTSKMVSAIVLYVLLAAAAHSAFA; and sequence II (peptide encoding for M start codon, 6 histidine residues, and enterokinase cleavage site) is MRGSHHHHHHGMASMTGGQQMGRDLYNNNNK.

10. The method of claim 9, wherein said polypeptide is: (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2.

11. An antibody, which binds specifically to a polypeptide, said polypeptide comprising an amino acid sequence, said amino acid sequence being selected from the group consisting of: (a) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2; (b) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (c) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues inserted between position −1 and position 1; (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2; (e) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with a methionine residue added to the amino terminus; (f) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus; (g) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the peptide of sequence II added to the amino terminus; (h) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (i) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus; (j) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus and a methionine residue added to the amino terminus; (k) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus and a methionine residue added to the amino terminus of the histidine residues; (l) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus and with 1 to 8 histidine residues added to the carboxy terminus; and (m) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 and with 1 to 8 histidine residues added to the amino terminus and the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus of the histidine residues; wherein sequence I (gp67 signal peptide) is: MLLVNQSHQGFNKEHTSKMVSAIVLYVLAAAAHSAFA; and sequence II (peptide encoding for M start codon, 6 histidine residues, and enterokinase cleavage site) is MRGSHHHHHHGMASMTGGQQMGRDLYNNNNK.

12. The antibody of claim 11, wherein said polypeptide is: (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2.

13. The antibody of claim 11, which is a monoclonal antibody.

14. A method of decomplementation, comprising administering to an animal in need thereof an effective amount of a polypeptide, said polypeptide comprising an amino acid sequence, said amino acid sequence being selected from the group consisting of: (a) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2; (b) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (c) from position about −22 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues inserted between position −1 and position 1; (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2; (e) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with a methionine residue added to the amino terminus; (f) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus; (g) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the peptide of sequence II added to the amino terminus; (h) from position about 1 to position about 1620 of the 5 amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus; (i) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus; (j) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the carboxy terminus and a methionine residue added to the amino terminus; (k) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with 1 to 8 histidine residues added to the amino terminus and a methionine residue added to the amino terminus of the histidine residues; (l) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 with the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus and with 1 to 8 histidine residues added to the carboxy terminus; and (m) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2 and with 1 to 8 histidine residues added to the amino terminus and the signal peptide of the baculovirus glycoprotein gp67 (sequence I) added to the amino terminus of th e histidine residues; wherein sequence I (gp67 signal peptide) is: MLLVNQSHQGFNKEHTSKMVSAIVLYVLAAAAHSAFA; and sequence II (peptide encoding for M start codon, 6 histidine residues, and enterokinase cleavage site) is MRGSHHHHHHGMASMTGGQQMGRDLYNNNNK.

15. The method of claim 14, wherein said polypeptide is: (d) from position about 1 to position about 1620 of the amino acid sequence shown in FIG. 2.

16. The method of claim 14, wherein said animal is a reptile, fish, bird, or mammal.

17. The method of claim 14, wherein said animal is a mammal selected from the group consisting of guinea pigs, mice, rats, pigs, baboons, chimps, dogs, cats, horses, cows, and humans.

18. The method of claim 14, wherein said animal is suffering from septic shock, ischemia-reperfusion injury, thermal injury, arthritis, lupus, respiratory distress syndrome, or a tissue rejection.

19. The method of claim 14, wherein said polypeptide is administered in an amount of 1 to 1000 U/kg of body weight.

20. The method of claim 14, wherein said polypeptide is administered in an amount of 30 to 80 U/kg of body weight.