Patent Application
20050079585
The activation of the complement system can be achieved by three different pathways, the classical antibody dependent activation pathway, the alternative activation pathway and thelectin-activation pathway. The alternative activation pathway as well as the lectin pathway were shown not to be dependent on antibodies. All pathways share a similar cascade-like organization, wherein a protease acts on zymogenes of a subsequent protease.This cascade results in an amplification of the initiation signals. The central step of the complement cascade resides in the formation of a C3convertase, which cleaves C3 to C3b and C3a (
The complement protein C3 is the central component of all activation pathways. It is predominantly expressed in the liver as a 1663 amino acid precursor protein (Alper et al., 1969). After the 22 amino acid signal sequence has been cleaved off, the precursor protein is proteolytically cleaved into two chains by removal of four arginine residues. The resulting achain has a molecular weight of 115 kDa and the β-chain has a molecular weight of 73kDa (DeBruijn and Fey, 1985). The chains are linked by a disulfide bridge and by non-covalent interactions (Dolmer and SottrupJensen, 1993; Janatova, 1986). Furthermore, the resulting 188 kDa protein carries a carbohydrate chain on each chain, which consists of 5 to 9 manose residues and two N-acetylglucosamine residues (Hirani et al., 1986).
C3 is cleaved between the amino acids Arg26 and Ser727 by the C3-convertases. The 9 kDa C3a, which results from the cleavage, is an anaphylatoxin and causes an increase in chemotaxi as well as an increase in the permeability of the blood capillaries. By cleavage of the 179 kDa-C3b between the amino acid Cyg88 and Glu991 a highly reactive thioester is released, by the use of which C3b binds on the cell surfaces via transacetylation (Tack et al., 1980). Furthermore, several binding sites for different complement proteins are exposed by the cleavage, which explains the various interactions of the C3,b molecule. Several regulatory complement proteins interact with C3b, which comprises binding sites for CR1 or Factor H, which act as ce factors for the cleavage by Factor I. Factor I cleaves C3b between Arg1281 and Ser1282, and Arg2298 and Ser1299, whereby the fragments C3f and C3bi emerge, the latter of which is inactive and unable to bind Factor B and C5 (Lachmann et al., 1982; Davis et al., 1982). C3bi, however, is capable to remain on the surface of pathogens, where it is recognized by CR3, which occurs on macrophages and killer cells. Subsequently, CR3 mediates the destruction of pathoges (Newman et al., 1984). In case CR1 acts as a cefactor for the protease, Factor I can additionally cleave between amino acids Arg932 and Glu933, thereby forming C3dg and C3c (Ross et al., 1982). C3dg is also capable to remain on the surface and is recoglized by CR2 (CD21), which is, expressed on B-lymphocytes and dendritic cells (Law and Dodds, 1997). The binding of C3dg to the complement receptor CR2 leads to the activation of B cells (Bohnsack and Cooper, 1988).
After the degradation of C3 by a proteasefrom the venom of the cobra Naja siamensis the fragment C3o is formed, which no longer contains the amino acids 730739. However, C3o is capable of binding Factor B (O'Keefe et al., 1988). In contrast, the cleavage product of the Factor I proteolysis C3c cannot form a convertase. Based on the comparison of C3c and C3o, one region in C3o of the amino acid sequence 933EGVQKEDIPP appeared to be responsible for binding to Factor B. In further studies, the amino acids937KED were mutated to alanine. However, no changes in the binding characteristics of Factor B to C3b could be shown (Taniguchi-Sidle und Isenman, 1994).
Activated complement proteins cannot distinguish between external substances and substances which occur naturally in the body. Thereby, it is ersured that for example self-reactive B-cells can be eliminated. Thus, a plurality of regulatory mechanisms is necessary for protecting healthy cells which occur naturally in the body.
The regulation is effected by short half life of the activated complement proteins on the one hand and by plasma proteins such as the C1-Inhibitor (C1-Inh), Factor H and Factor I, as well as membrane-bound proteins such as the DecayAccelerating-Factor (DAF, CD55), the Membrane-Cofactor-Protein (MCP, CD46) and the Complement Receptor 1 (CR1, CD35) on the other hand, which regulate the complement cascade on specific levels.
C1-Inh controls the activation of C1 by binding to activated C1r and C1s which results in the dissociation of C1q. The time period for the cleavage of C2 andC4 by activated C1 is restricted to a few minutes by C1-Inh (Mollnes und Lachmann, 1988). The C4binding protein (C4bp) binds to C4b and separates it from C2b. Additionally, it acts as a co-factor for the cleavage of C4b and C3b by Factor I (Scharfstein et al., 1978). The C3-convertase of the classical pathway is inactivated in the same manner by DAF, which exists on all peripheral cells of the blood, epithet and endothel (Lublin and Atkinson, 1989, Lublin and Atkinson, 1990).
C3b represents the central Dmponent of all three activation pathways. C3b is regulated by Factor H, CR1, DAF as well as by MCP. Here, Bb is competitively displaced by CR1, FactorH and DAF from the complex of the C3-convertase C3bBb (Makrides et al., 1992). Subsequently, C3b is cleared by Factor I and inactivated (Pangburn and Muller-Eberhard, 1984). MCP directly attacks C3b and is also a co-factor for the cleavage by Factor I. Protectin (CD95) is a further membrane-bound regulatory protein. It inhibits the polymerization of C9 by binding to C8 and C9 (Mollnes and Lachmann, 1988).
Besides the regulation for the activation, an additional transcriptional control of the complement genes exists. For example, several genes of the complement proteins are upregulated by cytokine and IFNy-activated transcription factors after damaging a tissue (Volanakis, 1995).
The strict regulatory mechanisms prevent an attack of the complement system on cells which occur naturally in the body. However, body tissue can be damaged by unregulated activation triggered by diverse diseases. In this situation, the activation of the complement is not the primary reason for disease. However, the resulting damaging of the tissue is mediated by the complement. Diseases which are connected with the activation of the cqnlement can be divided into three groups: Chronical diseases, acute diseases and incompatibility towards biomaterials. The group of acute diseases comprise for example asthma (Regal et al., 1993; Regal and Fraser, 1996), sepsis (Hack et al., 1989; Hack et al., 1992), hyperacute rejection in connection with transplantations or xenotransplantations (Bach et al., 1995; Baldwin et al., 1995), pneumonia (Eppinger et al., 1997) and cardiac infarction (Kilgore et al., 1997), as well as a massive C3a-accumulation, which occurs in connection with the cardiopulmonale bypassoperation (Kirklin et al., 1983; Homeister et al., 1992). The chronical diseases comprise, for example, systemic lupus erythematodes (SLE) (Belmont et al., 1986; Buyon et al., 1992), glomerulonephritis (Couser et al., 1985; Couser et al., 1995), rheumatoide arthritis (Kemp et al., 1992; Wang et al., 1995), Alzheimer's disease (Rogers et al., 1992; Morgan et al., 1997), myastenia gravis (Lennon et al., 1978; Piddlesden et al., 1996) and multiple sclerosis (Piddlesden et al., 1994; Williams et al., 1994) as well as organ rejection after transplantations or xenotransplantations (Baldwin et al., 1995; Dalmasso, 1997). The group of incompatibilities towards biomaterials was described in connection with operation material at a cardiopulmonal bypass (Craddock et al., 1977; Mollnes, 1997), with depositions of blood platelets (Gyongyossy-Issa et al., 1994) and with conducting hemodialysis (Cheung et al., 1994; Mollnes, 1997).
A reduced protein concentration of a complement protein or mutations which lead to a total loss of the protein are the reason for many complement-associated diseases. Factor I-deficiency results in a very small content of C3 and other complement proteins of the cascade in the blood. Thisleads to diverse diseases, such as a monthly occuring meningitis which is associated with menstruation (Gonzales-Rubio et al., 2001). Factor Ideficiency by gene mutation is associated with the hemolyticuremic syndrom (Zipfel et al., 2001). An unrestricted activity in the classical activation by depletion of C1, C2 or C4 leads for example to a higher disposition towards systemic lupus erythematodes (Morgan and Walport, 1991). A depletion of a component from the alternative activation such as Factor B or Factor D leads to a higher susceptibility towards infections (Morgan and Walport, 1991).
Complement-associated diseases occur both with an increased and decreased complement activation. In case the regulation is disturbed or the activation is prevented, effetive complement modulators are needed.
The group of complement inhibitors for therapeutic use comprises proteins such as the C1-Inhibitor and the soluble complement regulators sCR1 (soluble CR1), sMCP or sDAF, antibodies against C5 or C3 and smaller moleales such as the peptide Compstatin or RNA-aptamers. Several complement inhibitors are tested in clinical phases I, II or III, such as the C5Inhibitor Pexelizumab, a monoclonal antibody for use at cardiopulmonal bypass (Whiss, 2002) or the soluble complermnt receptor sCR1 (Zimmerman et al., 2000).
The C1-Inhibitor is the only plasma protein which has been tested in in vivo-studies (Struber et al., 1999; Horstick, 2002). The serine protease is a suicide inhibitor of the serpinefamily which inhibits activated C1s and C1q by binding to the active site (Sim et al., 1979). The disadvantages of these molecules relate to the sole inhibition of the classical activation pathways as well as in the susceptibility of the protein towards the inactivation by elastase. For this reason, elastase-resistent C1-Inhibitor mutants were generated (Eldering et al., 1993).
The recombinant complement inhibitors embrace soluble regulators such as sCR1, sMCP and sDAF (Christiansen et al., 1996). The soluble complement receptor sCR1 ats as C3- and C5-convertase-inhibitor and has been tested successfully in diverse animal models such as for myasthenia gravis (Piddlesden et al., 1996), multiple sclerosis (Piddlesden et al., 1994) or asthma (Regal et al., 1993). By altering the condition of expression, it was possible to increase the short half-life of approx. 8 h in vivo up to 70 h. It is proposed that a different glycosylation pattern is responsible for the increased half-life (Weismann et al., 1990; Zimmerman et al., 2000).
The complement receptors MCP and DAF act as complement inhibitors both in vitro and in vivo, for example in the model of reverse passive Arthusreaction (Moran et al., 1992; Christiansen et al., 1996). sDAF accellerates the decomposition of both the classical and the alternative C3- and C5-convertases. However, sDAF does not act as a co-factor for the cleavage of Factor I (Kinoshita et al., 1985). In contrast, sMCP acts as cofactor for the cleavage of C3b and C4b by Factor I. However, it does not act on the convertases (Liszewski and Atkinson, 1992).
Protectin (CD59) is a further membrane protein which protects naturally occuring cells of the body from MAC-mediated damage. It binds to C5b-8 and prevents the formation of a pore in the membrane by binding of C9 (Davies, 1996). Its soluble counter-part, sCD59, showed inhibition in vitro (Sugita et al., 1994).
A further group of complement inhibitors consists of antibodies, wherein C5 in particular represents an attractive target protein, since its concentration in tl serum is clearly lower than the one of C3. Monoclonal antibodies combine the advantage of specifity and high affinity with a relatively long halflife and the ease of production in large amounts. One prerequisite for the therapeutic application is the hman origin of the antibodies which prevents an immune response, for example the human anti-mouse-antibody-response. Several antibodies against C3 (Kemp et al., 1994), C3a (Burger et al., 1988; Elsner et al., 1994) or against C5a (Ames et al., 1994; Park et al., 1999) have been developed. Some have been tested in different animal models, for example for nephritis (Wang et al., 1996), collageninduced arthritis (Wang et al., 1995), myocardial ischemia und repezfusion (Vakeva et al., 1998).
Anaphylatoxin-receptor-antagonists (Konteatis et al., 1994; Pellas et al., 1998; Heller et al., 1999) and RNA-aptamers, which inhibit the C5-cleavage (Biesecker et al., 1999) belong to the group of complement inhibitors with low molecular weight. Compstatin, a C-3 Inhibitor, binds to native C3 and prevents its cleavage in C3b. By application of Compstatin, the hyperacute rejection of transplants in an ex-vivo pig-to-human-liver-transplantation was prevented (Fiane et al., 1999a; Fiane et al., 1999b).
Cobra Venom Factor (CVF) is a potent complement modulator of natural origin. CVF is a 149 kDa-glycoprotein from the venom of the cobra species Naja, Ophiophagus and Hemachatus (Müller-Eberhard and Fjellstrom, 1971). The nontoxic protein consists of three chains, the 68 kDa a-chain, the 48 kDa Pchain and the 32 kDa ychain which are linked by disulfide bridges. Additional intramolecular disulfide bridges exist both in the sand Unchain (one in the uchain and six in the β-chain; Vogel et al., 1996). The γ-chain can exhibit different sizes due to different processing on the C-terminus (Vogel and Müller-Eberhard, 1984). Two carbohydrate residues are attached to the α-chain and one to the β-chain in form of complex, abound oligosaccharide chains (Vogel and Müller-Eberhard, 1984; Grier et al., 1987).
The percentual composition of the secondary structure of CVF was determined by circulardichroism. The composition shows a high analogy to the composition of the secondary structures of the human three-chain C3-derivate C3c. For CVF 11% helices, 47% β-sheets and 18% β-loops were determined. The C3cmolecule also has 11% helices and 47% β-sheets. In contrast, human C3 consists of 24% hlices and 32% β-sheets (Vogel et al., 1984). In the primary structure of the pre-pro-CVF the α-chain is encoded first, followed by the γ-chain and subsequently by the mchain. On the C-terminus of the γ-chain 4 arginine residues are located, followed by aC3a-homologous region. Subsequent to the γ-chain, a C3d-homologous region is located. Both the signal peptide and the arginine residues and the C3a and C3d-homologous regions are removed post-translationally, thereby generating the three-chain structure. The venom protease, which is thought to be responsible for the modification also cleaves C3 in a CVF-similar structure (O'Keefe et al., 1988).
CVF shares an identity of 85% and a similarity of 92% on the protein level with cobra C3 (coC3). With human C3 he identity amounts to 51% and the similarity to 70% (Fritzinger et al., 1992; Fritzinger et al., 1994; Vogel et al., 1996). Moreover, both proteins have a chain structure of the same kind.
This high similarity is also reflected by the fact that CVF—as C3b—can bind to Factor B and forms a convertase by the Factor D initiated cleavage of B in Bb and Ba. In contrast to C3Bb, the CVF dependent convertase CVFBb is a C3- and C5-convertase. By the resistence of CVFBb towards Factor H and FactorI, a convertase is formed with a much higher half-life of 7 h (Vogel and Müller-Eberhard, 1982) under physiological conditions. In comparison, C3bBb has a half-life of 1.5 min (Medicus et al., 1976).
In addition to the increased stability, the CVFdependent convertase CVFBb cleaves C3 and C5 also in fluid phase, whereas the dependent convertase C3bBb is only active when bound to the cell surface (Vogel et al., 1996). CVF unifies all the above characteristics and leads to a permanent activation of the complement system and to deconplementation resulting thereof.
The decomplementing characteristic of CVF offers a variety of applications. After decomplementation the synthesis of complement protein takes approx. 7 days; during this time e.g. the function of the complement system in the immune response in vivo as well as in the pathogenesis of diseases can be studied (Cochrane et al., 1970; Ryan et al., 1986).
In various xenotransplantation models, such as liver transplantation from guinea pigs to rats, heart transplantations from hamsters to mice as well as islet cell transplantations from rats to mice, CVF was successfully employed (Chrupcala et al., 1994; Chrupcala et al., 1996; Lin et al., 2000; Oberholzer et al., 1999) In all these cases, hyperacute rejection of transplab could be prevented by CVF. Different studies demonstrate that also for diseases like arthritis (Lenset al., 1984), arteriosclerosis (Pang and Minta, 1980) and encephalomyelitis (Morariu and Dalmasso, 1978) CVF can be therapeutically employed.
The problem with therapeutic applications of CVF, however, predominantly resides in the strong immunogenic character of CVF. CVF contains a foreign peptide structure and complex, N-bound oligosaccharide chains with terminal galactosyl residues, which have a significant immunogenic potential (Taniguchiet al., 1996). Consequently, CVF is not suitable for repetitive application. With a relative high portion of carbohydrate structures (7.4%, Vogel and Müller-Eberhardt, 1984) CVF differs clearly from human C3 which orA has 1.7% (Hirani et al., 1986). Activity analyses in complement consumption-assays and bystander lysis-assays of CVF deglycosylated by n-glycanase showed that the oligosaccharide chains of CVF are not necessary for both C3-convertase and C5-convertase activity. A reduction of the immunogenicity, however, cannot be achieved by deglycosylation since deglycosylated CVF is still strongly immunogenic due to its foreign amino acid composition.
In an attempt to reduce the immunogenicity of CVF, the CVF e chain was replaced by the corresponding human C3β-chain (Kolln et al., 2001). The resulting hybrid protein, however, is still strongly immunogenic and thus inagpropriate for therapeutic uses.
From these published results, no information is available which could aid in designing a human C3-derivative i) capable of forming a stable C3-convertase comparable to CVFBb, ii) and suitable for therapeutic applications. Furthermore, published CVF/cobra C3 hybrids (Wehrhahn, 2000) are not suitable to provide any valuable alta with regard to the tertiary structure of a C3derivative required for effective binding of Factor B and for increasing the half life of the resulting C3-convertase.
The identity of cobra C3 with human C3 is too low to allow any specific structural conclusions.
Accordingly, there exists a need to identify polypeptides that exhibit complement-depleting activity and to develop methods of preparing these compounds recombinantly as therapeutics.There also exists a need to identify polypeptides having redued or eliminated immunogenicity, which polypeptides can be used therapeutically for treating complement-associated disorders and disorders affected by complement activation, respectively.
The present invention satisfies this need, and provides related advantages as well.
The invention provides isolated polypeptides having complement modulating activity. Specifically, the invention resides in the provision of isolated polypeptides having complement-depleting properties, i.e. that effect an efficient consumption of complement in human serum. Thus, the invention provides molecules capable of effectively inhibiting the complement system by depletion.
More specifically, the invention relates to polypeptides which are derivatives of the human complement component C3 (referred to herein as ‘human C3’, or ‘C3’, respectively), where the carboxy terminal (C terminal) part of the polypeptide is replaced by a carboxyterminal part or fragment of Cobra Venom Factor (CVF). The amino acid sequences df human C3 and CVF are depicted as SEQ ID NO:2 and SEQ ID NO:4, respectively.
The number of amino acids of CVF which replace the C terminal C3 fragment is either equal or less than the number of amino acids which have been ‘removed’ in relation to the natie human C3 sequence. The polypeptides of the invention, which are in some respect hybrid polypeptides of C3 and CVF or chimeric proteins, respectively, require at least the presence of amino acids 1575 to 1617 of the CVF sequence shown as SEQ ID NO:4, inorder to retain complement depleting activity.
In addition, without departing from the spirit of the invention, it may be desired for specific purposes, however, to attach additional non-C3 and non-CVF amino acids to the carboxy terminus of the hybrid proteins.
The polypeptides of the invention being at least 70% identical to the sequence of human C3 are less immunogenic than CVF (SEQ ID NO:4) or CVF in which the α-chain is replaced by the corresponding human C3 β-chain.
Also provided are methods of modulating or depleting complement in a human subject and methods of therapeutical treatment, respectively, by administering to the human an effective amount of a C3-derivative of the invention.
A further embodiment is apharmaceutical composition comprising a protein of the invention.
The invention also provides nucleic acid moleculesencoding novel C3-derivatives.
Vectors comprising nucleic acid molecules encoding novel C-3 derivatives and isolated host cells transfected with said vectors are provided as well.
The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the recessary fee.
FIGS. 6(A+B)shows the cloning strategy used for generating construct H5 expressing a hybrid protein of 1663 amino acids in length, whichamino terminally comprises amino acids 1 to 1388 of C3 followed by amino acids 1368 to 1642of CVF at the carboxy terminus.
(A) Activation of purified human factor B. Purified hC3 (lane 1), CVF (lane 3) and H5 (lane 6) were incubated for 2 h in the presence of factor D, factor B, and Mg2+. Inhibition of convertase formation by EDTA (hC3: lane 2; nCVF: lane 4) and incubation of factor B with factor D and Mg2+ (lane 5) were performed as controls. After separation of the reaction mixture by 10% SDS-PAGE under non-reducing conditions, generation of the cleavage products Bb and Ba was analyzed by western blotting using antifactor B antibodies.
(B-D) Catalytic activity of nCVFBb, hC3Bb, and H5Bb complexes. Shown is the time-dependent generation of fluorescent amido-methylcoumarin (AMC) from BocLeu-Gly-Arg-AMC by the different C3 convertases (B: nCVFBb; C: hC3Bb; D: H5Bb) after preincubation at 37° C. for 60 min (squares), 120 min (triangles), 180 min (dots), and 240 min (crosses) in the absence of the fluorogenic substrate. New formation of convertase complexes was inhibited by addition of EDTA prior to preincubation. Convertase activity of complexes without preincubation (diamonds) was used as control. After addition of the fluorogenic peptide to the samples, the timedependent release of AMC was followed by measuring fluorescence at 465 nm. (E) Stability of nCVFBb and H5Bb complexes. Based on the slopes in
The invention provides isolated polpeptides or proteins, respectively, that are biologically active derivatives of human complement component C3, that are able to continuously activate human complement system in human serum and thus to temporarily achieve inactivation of the complement system.
As used herein, the term “human complement component C3” (‘human C3’ or ‘C3’, respectively) refers to a polypeptide comprising the amino acid sequence of human complement component C3 shown in
As used herein, the term ‘CVF’ rfers to a polypeptide comprising the amino acid sequence of Cobra Venom Factor as shown in
As used herein, the terms “comprising,” “having,” “encoding,” and “containing,” and derivatives of these terms, are intended to be open-ended. “consisting” is intended to be closed-ended.
The invention specifically relates to a polypeptide or protein having a length of 1638 to 1663 amino acids, which is a derivative of human complement component C3 (human C3), the amino acid sequence of which is shown in SEQ ID NO:2, wherein the carboxy terminal part of at least 68 amino acid residues of human C3 is replaced by a partial sequence of Cobra Venom Factor (CVF), the amino acid sequence of which is shown in SEQ ID NO:4, which partial sequence comprises at least the 68 carboxy terminal amino acid residues of CVF or a fragment thereof lacking 1 to 25 carboxy terminal amino acids, wherein said protein has at least 70% identity to human C3 or a fragment thereof, the fragment comprising at least amino acids 1 to 1638 of SEQ ID NO:2. Thus, the invention provides a protein comprising a derivative of human complement component C3 (human C3), the human C3 having an amino acid sequence set forth as SEQ ID NO: 2, wherein the carboxy terminal part of at least 68amino acids of said human C3 is replaced by a partial sequence of Cobra Venom Factor (CVF), the CVF having an amino acid sequence set forth as SEQ ID NO: 4, wherein the partial sequence of CVF comprises at least at least 68 carboxy terminal amino acids ofCVF or a fragment thereof, said fragment lacking 1 to 25 carboxy terminal amino acids, and wherein said proteinhas at least 70 percent identity to said human C3 or a fragment of said human C3 comprising at least amino acids 1 to 1638 of the amino acid segence set forth as SEQ ID NO: 2. An alignment of the C3 and CVFsequences is shown in
In addition, without departing from the spirit of the invention, it may be desired for specific purposes, however, to attach additional non-C3 and non-CVF amino acids to the carboxy terminus of the hybrid proteins.
Decomplementing activity of C3/CVF hybrid proteins is observed with polypeptides where the C3 achain was replaced by the corresponding carboxy terminal amino acids of the CVF chain (including the γ- and the β-chain of CVF). However, due to the high immunogenicity of such polypeptides, a higher degree of humanization is desired. Thus, according to a preferred embodiment, the polypeptides of the invention comprise an amino terminal C3 fragment containing the amino acids forming the Schain (amino acids 23 to 667 of SEQ ID NO:2) as well as additional amino acids of the C3 chain following at the carboxy terminal end of the β-chain, i.e. from amino acid 668 towards the carboxy terminus of the peptide. At least 68 amino acids of the C3 sequence are replaced by amino acids of the corresponding CVF sequence. Reference is made in this respect to
According to a preferred embodiment of the invention, the protein or polypeptide, which is a derivative of human complement component C3 (human C3), has an amino acid sequence, which is selected from the group consisting of:
The most preferred constructs represented by SEQ ID NO:6 and SEQ ID NO:8 are hereinafter also termed ‘H5’ and ‘H6’, respectively. The identity of the amino acid sequences with human C3 amino acid sequence is approx. 90.7% (91%) for H5, and 96.3% (96%) for H6. Reference is made in this regard to
In order to retain decomplementing activity, the proteins require the presence of at least a stretch of 43 CVF amino acids in the carboxy terminal region, namely amino acids 1575 to 1617 of SEQ ID NO:3, which replace amino acids 1596 to 1638 of the C3 sequence (cf.
The invention also relates to anucleic acid encoding a protein having a length of 1638 to 1663 amino acids, which is a derivative of human complement omponent C3 (human C3), the amino acid sequence of which is shown in SEQ ID NO:2, wherein the carboxy terminal part of at least 68 amino acids of human C3 is replaced by a partial sequence of Cobra Venom Factor (CVF), the amino acid sequence of which is shown in SEQ ID NO:4, which partial sequence comprises at least the 68 carboxy terminal amino acids of CVF or a fragment thereof lacking 1 to 25 carboxy terminal amino acids, wherein said protein has at least 70% identity to human C3 or a fragment thereof, the fragment comprising at least aminoacids 1 to 1638 of SEQ ID NO:2.
According to a preferred embodiment, the nucleic acid is a nucleic acid encoding a protein having the amino acid sequence shown in SEQ ID NO:6. Preferably, the nucleic acid has the nucltide sequence shown in SEQ ID NO:5.
According to another embodiment, the nucleic acid is a nucleic acid encoding a protein having the amino acid sequence shown in SEQ ID NO:8. Preferably, the nucleic acid has the nucleotide sequence shown in SEQ ID NO:7.
According to a further embodiment, the nucleic acid is a nucleic acid encoding a protein having the amino acid sequence shown in SEQ ID NO:10. Preferably, the nucleic acid has the nucleotide sequence shown in SEQ ID NO:9.
Further provided is a nucleic acidencoding a protein having the amino acid sequence shown in SEQ ID NO:12. Preferably, the nucleic acid has the nucleotidesequence shown in SEQ ID NO:11.
As used herein, the term “nucleic acid molecule” refers to a polynucleotide of natural or synthetic orgin, which can be single- or double stranded, can correspond to genomic DNA, cDNA or RNA, and can represent either the sese or antisense strand or both.
The term “nucleic acid molecule” is intended to include nucleic acid molecules that contain one or mre non-natural nucleotides, such as nucleotides having modifications to the base, the sugar, or the phosphate portion, or having one or more non natural linkages, such as phosphothioate linkages. Such modifications can be advantageous in increasing the stability of the nucleic acid molecule, particularly when usedin hybridization applications.
Furthermore, the term “nucleic acid molecule” is intended to include nucleic acid molecules modified to contain a detectable moiety, such as a radiolabel, a fluorobrome, a ferromagnetic substance, or a detectable binding agent such as biotin. Nucleic acid molecules containing such moieties are useful as probes for detecting the presence or expression of C3-derivative nucleic acid molecule.
The invention is also directed to expression of the protein in suitable host cells. In this context, a vector is provided which comprises a nucleic acid, which nucleic acid encodes a protein having a length of 1638 to 1663 amino acids, which is a derivative of human complement component C3 (human C3), the amino acid sequence of which is shown in SEQ ID NO:2, wherein the carboxy terminal part of at least 68 amino acids of human C3 is replaced by a partial sequence of Cobra Venom Factor (CVF), the amino acid sequence of which is shown in SEQ ID NO:4, which partial sequence comprises at least the 68 carboxy terminal amino acids of CVF or a fragment thereof lacking 1 to 25 carboxy terminal amino acids, wherein said protein has at least 70% identity to human C3 or a fragment thereof, thefragment comprising at least aminoacids 1 to 1638 of SEQ ID NO:2.
Preferably, the vector comprisesa nucleic acid, which nucleic acid encodes a protein having an amino acid sequence, which amino acid is selected from the group consisting of:
According to a preferred embodiment, the vector comprises a nucleic acid, which nucleic acid has a nucleotide sequence selected from the group consisting of:
The constructs represented by SEQ ID NO:5 and SEQ ID NO:7 are hereinafter also termed ‘H5’ and ‘H6’, respectively.
In the afore mentioned expression vectors, the nucleic acid molecules are operatively linked to a promoter of gene expression.As used herein, the term “operatively linked” is intended to mean that the nucleic acid molecule is positioned with respect to either the endogenous promoter, or a heterologous promoter, in such a manner that the promoter will direct the transcription ofRNA using the nucleic acid molecule as a template. The invention providesnucleic acid molecules which are operatively linked to said promoter as well as vectors comprising said promoter driven nucleic acid molecules.
Methods for operatively linking a nuchic acid to a heterologous promoter are well known in the art and include, for example, cloning the nucleic acid into a vector containing the desired promoter, or appending the promoter to a nucleic acid sequence using polymerase chain reaction (PCR). A nnleic acid molecule operatively linked to a promoter of RNA transcription can be used to express C3-derivative transcripts and polypeptides in a desired host cell or in vitro transcription-translation system. The choice of promoter to operatively link to a invention nucleic acid molecule will depend on the intended application, and can be determined by those skilled in the art. Exemplary promoters suitable for mammalian cell systems include, for example, the SV40 early promoter, the cytomegalovirus (CMV) promoter, the mouse mammary tumor virus (MMTV) steroid-inducible promoter, and the Moloney murine leukemia virus (MMLV) promoter. Exemplary promoters suitable for bacterial cell systems include, for example,T7, T3, SP6 and lac promoters.
Exemplary vectors of the invention include vectors derived from a virus, such as a bacteriophage, a baculovirus or a retrovirus, and vectors derived from bacteria or a combination of bacterial sequences and sequences from other organisms, such as a cosmid or a plasmid. The vectors of the invention will generally contain elements such as an origin of replication compatible with the intended hast cells; transcription termination and RNA processing signals; one or more selectable markers compatible with the intended host cells;and one or more multiple cloning sites. Optionally, the vector will further contain sequences encoding tag sequences, such as GST tags, and/or a protease cleavage site, such as a Factor Xa site, which facilitate expression and purification of the encodedpolypeptide.
The choice of particular elements to include in a vector will depend on factors such as the intended host cells; the insert size; whether expression of the inserted sequence is desired; the desired copy number of the vector; the desired selecton system, and the like. The factors involved in ensuring compatibility between a host cell and a vector for different applications are well known in the art.
Also provided are cells containing an isolated nucleic acid molecule encoding a C3-derivative of the invention. The isolated nucleic acid molecule will generally be contained within a vector. The cells of the invention can be used, for example, for molecular biology applications such as expansion, subcloning or modification of the isolated nucleic acid molecule. For such applications, bacterial cells, such as laboratory strains ofE. coli, are useful, and expression of the encoded polypeptide is not required. The cells of the invention can also advantageously be used to recombinantly express and isolate the encoded polypeptide. For such applications, bacterial cells (e.g. E. coli), insect cells (e.g. Drosophila), yeast cells (e.g. S. cerevisiae), and vertebrate cells (e.g. mammalian primary cells and established cell lines; and amphibian cells, such as Xenopus embryos and oocytes) , can be utilized.
The polypeptides described herein, which are referred to, for example, as C3-derivatives or hybrid proteins, are therapeutic compounds that can be administered to individuals, in particular to human subjects, for decomplementation.
According to a preferred embodiment, the invention provides the use of said polypeptides in a method fortreating a patient suffering from complement-associated disorders or disorders affected by complement activation, comprising adinistering an effective amount of said protein or polypeptide, respectively. Specifically, the complement-associated disorder includes but is not limited to asthma, systemic lupus erythematodes, glomerulonephritis, rheumatoid arthritis, Alzheimer's diseam, multiple sclerosis, myocardial ischemia, reperfusion, sepsis, hyperacute rejection, transplant rejection), cardiopulmonary bypass, myocardial infarction, angioplasty, nephritis, dermatomyositis, pemphigoid, spinal cord injury, and Parkinson's disease
The invention thus relates to the use of the polypeptides or proteins of the invention for preparing a pharmaceutical composition for decomplementation, for treating a patient suffering from complement-associated disorders or disorders affected by coplement activation (see above).
The invention further provides apharmaceutical composition comprising the protein or polypeptide of the invention. Specifically, a pharmaceutical composition of the invention comprises an effective amount of a protein having an aino acid sequence, said amino acid sequence being selected from the group consisting of:
The therapeutic compounds can be administered to a mammal or human, respectively, by routes known in the art including, for example, intravenously, intramuscularly, subcutaneously, intraorbitally, intracapsularly, intraperitoneally intracisternally, intra-articularly, intracerebrally, rectally, topically, intranasally, or transdermally. Preferred routes for human administration are intravenous administration. The pharmaceutical compositions of the invention are thus formulated forsaid administration routes, and, according to a preferred embodiment, comprise the compound and a pharmaceutically acceptable carrier depending on the route of administration of the compound and on its particular physical and chemical characteristics. Phamaceutically acceptable carriers are well known in the art and include sterile aqueous solvents such as physiologically buffered saline, and other solvents or vehicles such as glycols, glycerol, oils such as olive oil and injectable organic esters. A pharmceutically acceptable carrier can further contain physiologically acceptable compounds that stabilize the compound, increase its solubility, or increase its absorption. Such physiologically acceptable compounds include carbohydrates such as glucose, sucrse or dextrans; antioxidants, such as ascorbic acid or glutathione; chelating agents; andlow molecular weight proteins.
The following examples are intended to illustrate but not limit the present invention.
1 Materials and Methods
1.1 Materials
1.1.1 Chemicals
General chemicals were purchased from Sigma (Taufkirchen, Germany), Merck (Darmstadt, Germany), Fluka (Buchs, Switzerland), Invitrogen (Leek, Netherlands), Biomol Feinchemikalien (Hamburg, Germany), Gibco BRL (Eggenstein, Germany), Applichem (Darmstadt, Germany), Roth (Karlsruhe, Germany), Millipore (Eschborn, Germany), Calbiochem (Schwalbach, Germany) and Peqlab (Erlangen, Germany).
1.1.2 Enzymes, Proteins and Antibodies
Restriction enzymes, Mung Bean nuclease, T4-Ligase, Calf intestinal alkaline phosplatase (CIAP) and the respective buffers were purchased from MBI Fermentas (St. LeonRot, Germany) and at NEB (Frankfurt, Germany). Thermus aquaticus DNA-polymerase was purchased from AGS (Heidelberg, Germany) or Roche (Mannheim, Germany). Lysozyme was also purchased from Roche (Mannheim, Germany). RNase was purchased from Sigma (Taufkirchen, Germany). CVF, C3 and Factor B are commercially available from Calbiochem (Schwalbach, Germany).
Antisera with specificity for CVF, C3 and Factor B can be generated following standard immunization procedures. Factor D was purchased from Sigma (Taufkirchen, Germany). StrepTactin was purchased from IBA (Gottingen, Germany). Antiserum against C3 from goat was purchased from Cappel (Eschwege, Germany). A monoclonal antibody against C3d was purchased from Quidel (Heidelberg, Germany). The monoclonal anti-Strep-tagII antibody was purchased from IBA (Gottingen, Germany). The secondary antibody antirabbit, anti-mouse or anti-goat, conjugated with alkaline phosphatase or perokdase was also purchased from Sigma (Taufkirchen, Germany) or Cappel (Eschwege, Germany).
1.1.3 Affinity Matrix and Particles
Ni-NTA-agarose was purchased from Qiagen (Hilden, Germany). Strep-Tactin-Sepharose was purchased from IBA (Göttingen, Germany) and protein A/G-agarose was purchased from Santa Cruz Biotechnology (Santa Cruz, Calif., USA).
1.1.4 Bacteria and Yeast Strains
The bacteria strainE. coli DHS(X (Promega, Mannheim, Germany) was used for amplification of plasmids. In Table 1, the genotypes of the strains are listed.
1.1.5 Vectors
The commercially available vectors which were used are listed. The vectors which are not commercially available are depicted in
pUC18
Vector pUC18 (MBI Fermentas, St. LeonRot, Germany) was used for cloning.
pEGFP-N1
Vector pEGFP-N1 (Clontech, Heidelberg, Germany) was used for transfection of mammalian cell lines in order to determine the transfection efficiency.
pcDNA3
Vector pcDNA3 (Invitrogen, Leek, Netherlands) was used for eukaryotic expression of proteins
pUC18CVF*/pCDNA3CVF
Plasmids pUC18CVF* and pcDNA3CVF contain the cDNA of CVF (Genebank Accession No. U09969). In pUC18CVF*, a BglII-restriction site was introduced in Position 1793 (Mutation A1797C) and a HindIII restriction site was deleted in Position 2380 (A2380T, G2381C). Plasmid pcDNA3CVF is shown in
pUC18hC3/pcDNA3hC3
The plasmids contain the cDNA of human C3 (
1.1.6 Cell Lines and Culture
GMEM medium, DMEM nedium, penicillin G, streptomycin, geneticin and trypsin/EDTA were purchased from Gibco BRL (Eggenstein, Germany). Adenosine, guanosine, cytosine, uracil, thymidine, -L aspartic acid, and L-glutamic acid were purchased from SigmaAldrich (Steinheim, Germany). Culture flasks were purchased from Greiner (Frickenhausen, Germany), Nunc (Wiesbaden, Germany) or Sarstedt (Numbrecht, Germany). Fet al calf serum was purchased from Biochrom (Berlin, Germany).
Cell lines CHO, COS-7 and HEK293 were used for expression. CHO-cells were cultured in GMEM with 10% FC5 and other additives. HEK293 and COS-7-cells were cultured in DMEM with 10% FCS.
1.1.7 Solutions
1.1.8 Media
1.1.9 Oligonucleotides
The oligonucleotides which were used were synthesized by Metabion (Martinsried, Germany) (Table 2).
1.2 Methods
1.2.1 Densitometric Determination of Concentration
For densitometric determination of the concentration, 45 dilutions of known concentration of native CVF or human C3 (depending on the sample) were applied in addition to different volumina of the protein containing sample to be determined. The concentrations of the proteins of the calibration serious was chosen so that they were located in the range of the protein amount which was expected in the sample. The gel was subjected to wetblotting or semidry-blotting procedures and the proteins were subsequently stained using immunoprinting. Then, the membrane was scanned and the concentration of the sample was determined using the program Imagemaster iD Elite Version 2.01 (Amersham Pharmacia Biotech, Freiburg, Germany)
1.2.2 ELISA
ELISA was performed in order to detect proteins in supernatant. For this purpose, lpg of a protein, which was diluted in 100 μl incubation solution were immobilized on the well of an ELISA plate overnight at 4° C. Subsequently, the wells of the plate were washed three times with washing buffer (200 μl). Then, the wells were blocked with 200 μl 5% milk powder in PBS for 5 hours at room temperature. After washing three times, 200 μl of the supernatant of transient expression or purified diluted proteins in PBS, respectively, were introduced into the wells and incubated overnight under agitation at 4° C. After this, the samples were washed three times with washing buffer and incubated with 100μl of a 1:1000 dilution of the respective antibody in 2,5% milk powder in PBS at room temperature. Subsequently, the samples were washed three times and incubated for one hour with 100 μl of a 1:1000 dilution of an respective peroxidase-conjugate in a 2,5% milk powder in PBS at room temperature. Finally, the samples were washed three times, and then 100 μl detection buffer (for POD) were added. After staining the wells, extinction at 405 nm was measured wih a micro titer plate-photometer (ELISA-Reader, SLT-Instruments, Grodingen, Austria, Easy Reader EAR 400AT).
1.2.3 Expression of CVF, C3 and the Hybrids in Mammalia
For expression of recombinant CVF, C3 and of the hybrids of the invention, different cell lines are available.
COS-7-cells were originally obtained from kidney of monkeys and contain the SV-40 origin of replication. They express the large T-antigen of the SV-40 virus, which facilitates efficient replication. As a consequence, these plasmids show hihg copy numbers in the cell. Thus, the cells are suitable for transient expression. CHO-cells from the ovaria of Chinese hamster contain an RNA polymerase gene having a nucleus localisation signal. Therefore, they can be used for preparation of stably tranfected cell lines. HEK293-cells are human embryonic kidney cells and can be used for transient or stable expression. The HEK.EBNA-cells which were employed constitutively express the EpsteinBarr-Virus (EBV) Nuclear Antigen 1 (EBNA-1, EBV nuclear antigen 1). EBNA-1 was identified as the gene which is mainly responsible for immortalization of cells by the EBV (Lupton and Levine, 1985). For expression, vector pcDNA3 was available, which facilitates an efficient expression in mammalia via the CMV-promotor. The neomycin gene is available as resistence for the expression under selection pressure. The expression wasperformed as desribed in 1.2.9.
1.2.4 Partial Purification of the Recombinant CVF from the Supernatant of the Transient Expression
For separation of the low molecular components of the culture supernatants or column fractions, these supernatants or fractions were transferred into a dialysis tube (SpectraPor CE 100, MWCO 100 kDa, Roth, Karlsruhe, Germany) and dialysed overnight against PBS.
The concentration of dialysed samples was performed with Centricon-units (MWCO 100 kDa, Millipore, Eschborn, Germany). For this purpose, centrifugation was performed at 1,000×g (4° C.) until the desired final volume was reached.
For partial purification and for separation ofinterfering components, respectively, as well as of low molecular components, column chromatographic methods were employed. The supernatant (2.5 ml) from transient expression was loaded onto a PD10-column according to the manufacturer's instructions (Amresham Pharmacia Biotech, Freiburg, Germany). PBS was used as a running buffer. The fractions were examined for recombinant protein by Western blotting and subsequent immunoprinting. For partial purification, 2 ml supernatant of the transient expression weE loaded onto a 1 ml EconoPac-column (Biorad, Munich, Germany) using a Trisbuffer (50 mM, pH 7.5). The same buffer with 500 mM NaCl added thereto was used for elution. The fractions were also examined for protein content and dialysed against PBS.
For control purposes, the identic procedure was employed with the supernatant of nontransfected cells.
1.2.5 Purification and Detection of Proteins using the Strep-tag System
1.2.5.1 The Strep-tag System
For purification and immobilization of the recombinant proteins, CVF, C3 and the hybrids consisting of CVF and C3 were supplied with a Strep-tag-fusion peptide. The Strep-tag is a synthetic peptide consisting of 9 amino acids (AWRHPQFGG), which binds to streptavidin with an affinity of 2.7×10 (Schmidt et al., 1996). It uses the binding pocketfor biotin. As C-terminal fusion partner for proteins, it can be used for purification and detection. An N-terminal fusion is also possible since the system has been improved. The resulting Strep-tagII (WSHPQFEK) binds with a lower affinity to Streptavidin (Schmidtet al., 1996), however, a derivate was found by a selection round with randomly mutated Streptavidin, which in turn has a sufficient high affinity (Skerra and Schmidt, 2000). With a dissociation constant of 1 μM, the agarose-immobilized Streptavidin-derivate Strep-Tactin can be used for purification. Strep-tactin-conjugates or anti-Strep-tagII antibodies can be utilized for detection in immunoprinting or in ELISA analysis.
1.2.5.2 Detection of Strep-tag Fusionproteins
For detection of strep-tagII-fusionproteins in supernatant of transient expression, an ELISA was conducted. For this purpose, 3 μg strep-tactin (diluted in 40 μl incubation solution) were immobilized, overnight at 4° C. on the surface on the wells ofan ELISA plate (Greiner, Frickenhausen, Gemany). Subsequently, the procedure described above in connection with the ELISA was performed.
1.2.6 Purification of Hydrid H5
Supernatant (500 ml) obtained from stably transfected cells was adjusted to pH 7.5, passed trough a 0.45 μm cellulose acetate membrane and loaded onto a Poros HQ/M anion exchange column equilibrated with 50 mM Tris, pH 7.5 using AKTA purifier (Amersham Bioscience, Freiburg, Germany). The recombinant protein was eluted using a linear (0-500 mM) NaCl gradient. Fractions (2 ml) were analyzed using 7.5% SDS-PAGE and western blotting, pooled and dialyzed against phosphate buffered saline (PBS). The pooled sample was diluted (1:9) in 50 mM sodium phosphate, 0.55 M sodium sulfate buffer, pH 7.0, filtered (0.2 μm) and applied to a thiophilic resin (1.5 ml, BD Bioscience, Heidelberg, Germany) equilibrated with 50 mM sodium phosphate, 0.5 M sodium sulfate buffer, pH 7.0. After extensive washing of nonadsorbed proteins with the equilibration buffer (>30 column volumes), elution was performed using 50 mM sodium phosphate buffer, pH 7.0. Fractions (1.5 ml) were analyzed by 7.5% SDS-PAGE and western blotting. Fractions containing H5 were pooled, dialyzed against 100 mM Tris, 150 mM NaCl, pH 8.0 (buffer), loaded onto Strep-Tactin sepharose (2 ml, IBA, Göttingen, Germany) equilibrated with buffer W, washed with 10 ml buffer W, and eluted with buffer W containing 2.5 mM desthiobiotin. Protein concentration and purity of the fractions were analyzed by 7.5% SDS-PAGE. Pooled fractions were dialyzed against PBS and employed for further characterization.
1.2.7 Purification of Hybrid His-H6
For purification of HisH6 using IMAC, imidazol was added to 50 ml stable supernatant and incubated over night on a shaig unit. The matrix was then centrifuged (700×g, 10 min, 4° C.), the supernatant removed, and the matrix was resuspended in 50 ml PBS. Following centrifugation (700×g, 10 min, 4° C.), the matrix was again resuspended in 5 ml PBS and loaded onto a flowthrough column. Subsequently, bound proteins were eluted from the column in 1 ml portions using 3 ml 300 mM imidazole in PBS. Then, protein content was determined in the fractions by SDSPAGE followed by subsequent Western blot and immunoprinting. Suitable fracions were combined, dialysed against PBS and used for further analysis.
1.2.8 Complement Methods
1.2.8.1 Preparation of Sensitized Sheep Erythrocytes
Sheep whole blood (1 ml, Behringwerke, Marburg) was resuspended in 13 ml cold GVBS++ and centrifuged (10 min, 1,000×g, 4° C.). Supernatant was removed, and the erythrocytes were again resuspended in 14 ml GVB++. This procedure was repeated until supernatant became clear. Subsequently, the erythrocytes were resuspenden in approx. 5 ml GVBS++. Erythrocytes were adjusted by diluting with GVBS++ to give an absorption of 1.9 (5×108 cells(ml) at 412 nm for 30 μl erythrocyte suspension in 1 ml ddH2O. 2 μl antiserum against sheep erythrocytes (anti sheep red blood cell stroma, Sigma, Taufkirchen) were added to eachof 1 ml of the adjusted erythrocytes. Sensitizing was performed for 1 h in a water bath at 37° C., while inverting the reaction tube regularly after 10 min. The sensitized erythrocytes were washed 3 times with 2 ml GVBS++ and centrifuged (3 min, 1,000×g, 4° C.). The erythrocytes could be stored up to three days. Prior to each use, OD412 was adjusted to 1.9.
1.2.8.2 Isolation of Guinea Pig-erythrocytes
Guinea pig-erythrocytes were isolated from whole blood, which was obtained from isofluoraanarcotized guinea pigs by punction of the eyes. Approximately 1 ml blood was taken and immediately transferred into a tube containing 1 ml icecold ACD solution. The ACD solution serves as anti-coagulant. The erythrocytes were separated by centrifugation (1,000×g, 4 min,4° C.), resuspended in 14 ml GVBS++ and again centrifuged (1,000×g, 4 min., 4° C.). This procedure was repeated 3 to 4 times until the supernatant remained clear. Finally, the erythrocytes were resuspended in approximately 5 ml GVBS++. The erythrocytes were diluted with GVBS++ until 30 μl of the erythrocyte suspension in 1 ml ddH2O gave an absorption of 1.9 (5×108 cells/ml) at 412 nm.
1.2.8.3 Complement Consumption Assay
This test is based on the complement consuming effect of CVF. If a CVF containing ample is incubated with human serum, the complement proteins are consumed depending on the CVF activity (Ballow and Cochrane, 1969; Cochrane et. al., 1970). The remaining complement activity of the serum can be detected subsequently using sensitized sheep erythrocytes.
First, quantification of human serum was performed by serum titration, which led to hemolysis of sheep erythrocytes by 7-090%. For this purpose, different serum concentrations (serum value) were provided in 2 ml reaction tubes (double measurments) and filled up to 40 μl with GVBS++. Additionally, controls were prepared, which contained 40 μl GVBS++ (buffer-control) only, or 40 μl ddH2O (complete lysis) only, respectively. All reactions were incubated for 30 min at 37° C. under agitation (Thermaixer 5437, Eppendorf, Hamburg, Germany) Then, 100 μl cold GVBS++ or 100 μl ddH2O (upon complete lysis), respectively, and 30 μl sensitized sheep erythrocytes were added. Further incubation took place for 30 min, as described above. Subsequently, the samples were kept on ice. 850 μl cold VBS++ or 850 μl ddH2O (upon complete lysis), were added, respectively. The supernatants were transferred into cuvettes, and optical density was measured at 412 nm. Hemolysis was subsequently calculated according to the following formula:
For complement consumption assay, the amount of serum determined in prior tests and te samples (max. 20 μl) were provided in a 2 ml reaction tube (double measurements) and supplemented with GVBS++ to give 40 μl. Additionally, reaction mixes of the following controls were prepared: determined amount of serum, supplemented with GVBS++ to give 40 μl (serum control, 4 to 5 samples); 40μl GVBS++ (buffer control) and 40 μl ddH2O (complete lysis). All reaction mixes were incubated for 3 hours at 37° C. under agitation. Then, 100 μl cold GVBS++ or 100 μl ddH2O upon complete lysis, respectively, and30 μl adjusted sensitized sheep erythrocytes were added, and incubation took place for 30-40 min, as described above. After 15 min, serum control as well as a reaction mix of the complete lysis were measured according to the following principle. The sample were kept on ice, 850 μl cold VBS++ or 850 μl ddH2O at complete lysis, respectively, were added and centrifuged (4° C., 2,000×g, 2 min.). Supernatants were transferred into cuvettes, and the optical density was measured at 412 nm. Lysis was calculated afcording to the following formla:
In case the value for the serum control was clearly below 80% of the complete lysis value, a second serum control was taken after 10 further minutes of incubation and measured as described above. Once a value of 70-80% hemolysis was obtained, all reaction mixes were measured and evaluated according to the same principle In order to facilitate comparison of different series of measurements, values were referred to the corresponding serum control.
1.2.8.4 Solid Phase Complement Consumption Assay
For characterization of complement consuming activity of the recombinant strep-tagII-C3/CVF hybrids, a solid phase assay was performed. Here, proteins were bound via their strep-tag to an ELISA plate, which was covered with streptactin. Subsequently, activity can be determined by adding serum. For this purpose, 3 μg strep-tactin (diluted in 40 μl incubation solution) were immobilized overnight at 4° C. on the surface of the wells of an ELISA plate (Greiner, Frickenhausen, Germany). Then, the wells were washed 3 times with 200 μl washing buffer and subsequently blocked with 200μl 3% BSA in PBS for 5 hours at room temperature. After washing the wells again (3 times), different volumes of the supernatants or of the samples were provided in the wells. Protein concentration in the supernatants was previously determined densitometricaly. After over night incubation at 4° C. under agitation, the wells were washed 3 times with washing buffer. Subsequently, GVBS++ and the amounts of serum determined in serum titration were added, giving a volume of 60 μl. The ELISA plate was sealed with parffin and fixed to the bottom of an incubation shaker. Subsequently, incubation took place for 3 hours at 37° C., 150 rpm. Supernatants were transferred into 2 ml reaction tubes, and following addition of 100μl GVBS++ and 30 μl sensitized erythrocytes the camplement consumption assay was conducted as described.
1.2.8.5 Bystander Lysis-assay
This test for detecting hemolytic activity is based on fluid C5-activation and can be determined by lysis of norsensitized guinea pig erythrocytes (Vogel, 1985). In thismethod, the extent of complement activation is determined by photometric measurement of released hemoglobin. Different amounts of the protein to be analyzed were diluted in 20 μl GVBS++ mixed with 20 μl guinea pig serum (Sigma, Taufkirchen, Germany) and 2 μl guinea pig erythrocytes (5×108 cells/ml) in a 2 ml reaction tube and incubated for 3 hours at 37° C. under agitation (Thermomixer 5436, Eppendorf, Hamburg, Germany). The reaction was stopped by adding 1 ml icecold VBS++-buffer. The erythrocytes were centrifuged (2,000×g, 4° C., 2 min) and released hemoglobin in the supernatant was determined by measurement of extinction at 412 nm. Reaction mixes with 20μl erythrocytes and 40 μl ddH2O (complete lysis) or 20 μl guinea pig serum, 20 μl erythrocytes and 20 μl GVBS++ (serum control), respectively, served as controls.
1.2.8.6 Determination of the Stability of C3 Convertases
500 ng of each native CVF (nCVF), hC3, and the derivative H5 were mixed with 950 ng of factor B and 8 ng of factor D in a volume of 60 μl VBS (2.5 mM Na-5,5-diethyl-barbituric acid, 143 mM NaCl, pH 7.4). After addition of MgCl2 to a final concentration of 10 mM, the samples were incubated for 2 h at 37° C. to allow for convertase formation. Subsequently, all samples were supplementedwith EDTA to a final concentration of 10 mM to inhibit further formation of convertases. Thereafter, the samples were incubated at 37° C. and after different periods of incubation 20 μl aliquots were added to 150 μM Boc-Leu-Gly-Arg-7-amido-4-methylcoumarin acetate (Sigma, Taufkirchen, Germany) in 180 μl VBS. The timecourse of fluorophore release was determined in black FIAPlates (96 K, Greiner Bio-One, Frickenhausen, Germany) using an excitation filter of 370 nm and an emission filter of 465 nm in amicroplate reader Genios, Tecan, Creilsheim, Germany). Values after 60 min of fluorophore release were defined as 100%. The slope of the graph was used to determinethe enzymatic activity of the sample.
1.2.9 Tissue Culture and Expression Methods
1.2.9.1 Culturing and Passaging of COS7 and HEK293 Cells
COS-7 cells or HEK293 cells were cultured in an incubator (Heraeus Instruments Begasungsbrutschrank 6060) in awater saturated atmosphere (5% CO2) at 37° C. Growth was performed in DMEM medium (Gibco/BRL, Eggenstein, Germany) which was supplemented with 10% FCS (Biochrom, Berlin, Germany). As soon as cells grew confluently in the tissue culture flasks (75 cm, Cellstar, Greiner Labortechnik, Frickenhausen, Germany), they were passaged into a new culture flask (approximately every 3 days). Cell supernatant was removed and the cells were washed with PBS. 4 μl trpysin/EDTA (Gibco BRL, Eggenstein, Germany) were added, and the cells were incubated for 5 min in an incubator. Detachment of the cells from the bottom was suported by gentile knocking and was controlled under the microscope. When the cells were almost completely detached, the procedure was stopped by adding 8 ml serum-containing medium. The suspension was transferred into a 15 ml-reaction tube, and the cells were sedimented by centrifugation (5 min, 1,000×g, RT). The supernatant was removed, the pellet was resuspended in 10 ml serumcontaining medium and 1 to 3 ml of the cell suspension were transferred into a new tissue culture flask and supplemented with serum containing medium to give a final volume of 13 ml.
1.2.9.2 Culturing and Passaging CHO-cells
Growth of cells was performed in serumcontaining GMEM-medium (10% FCS). Culturing and passaging of CHGcells was performed in an analogous manner as described for COS7-cells.
1.2.9.3 Transfection
For expression in mammalian systems, expression vector pcDNA3 (Invitrogen, Leek, the Netherlands) comprisingthe corresponding genes was introduced into the cells by the GenePorter reagent (PeqLab, Erlangen).
DNA (1 to 4 μg) and 10 to 15 μl GenePorter transfection reagent were diluted in 500 μl serum-free medium and pooled. The reaction mix was incubated for 45 min at room temperature. At the same time, the cells which were passaged in a 6well plate approx. 24 hours earlier (300-500 μl of cell suspension with 2 ml serum containing medium per well; TC-plate, 6-well, Greiner Labortechnik, Frickenhausen, Germany), were washed with PBS. Then, the DNA GenePorter mix was carefully added dropwise to the cells, and the 6-well plate was placed in the incubator. After 3 hours, the medium was replaced against 2 ml serumcontaining medium, and the cells were kept for 2 to 3 days in the incubator for cell growth.
When using Nutridoma HU (100×; Roche Diagnostics, Mannheim, Germany), the cells were added to Nutridoma HU after transfection in serum-free medium and kept growing in the incubator for 2 to 3 days.
A reaction mix without DNA was prepared as negative control. If applicable, a reaction mix with 1 μg pEGFP-N1 was prepared as positive control. Plasmid pEGFPN1 encodes the green fluorescence protein (GFP) and can be used for determining transfection efficiency since the GFP expressing cells can be detected using a fluorescence microscope. The supernatant of the pEGFPN1 transfected cells was already removed after 24 hours, cells were washed with 2 ml PBS, and 500 μl trypsin/EDTA were added. Detachment of cells was performed for 5 min in the incubator. Then, 2 ml PBS were added and the cell suspension was transferred into a 50 ml reaction tube for centrifugation (5 min., 1,000×g, RT). The supernatant was removed and the cell pellet was resuspended in 2 ml PBS. 10μl thereof were provided in a Neubauer-counting chamber, and the number of fluorescening cells and the total number of all cells were counted in the outer four squares.
Calculation of the Cell Number/ml:
Calculation of Transfection Efficiency:
1.2.9.4 Expression Under Selection Pressure
In order to increase yields, generation of stably expressing lines was desired, wherein the reaction mixes from transient expression were kept in culture by adding an antibiotic and by sub culturing. Depending on resistance, 10 μl/ml culture medium were added to an antibiotic stock solution (G418), or 5 μl/ml culture medium were added to zeocine.
1.2.9.5 Expression in Serum or Proteinfree Medium
For culturing in serumfree or proteinfree medium, cells were step-wise adapted with SCF30- or Mampf3-medium (Promocell, Heidelberg, Germany) with 1 mM L-glutamine. The percentual portion of the serum-free or protein-free medium was increased by 25% in every second passage. Otherwise, passaging took place as described.
1.2.9.6 Monoclonalization
In order to obtain a homogeeous cell population, monoclonalization was conducted, wherein the different cells were monoclonalized, expanded and subsequently examined with respect to their expression level by Western blotting and immuno printing.
For this purpose, all cells were iirubated for 3 to 4 passages after transfection under selection pressure in order to guarantee that a large portion of cells contained the resistance and therefore expressed the target protein. This procedure allows to equate the cell number per ml, which is counted thereafter using a Neubauer counting-chamber, with the number of cells, which express the target protein. Subsequently, cell density was set to 1 cell per 100 μl medium with 10% FCS, and 100μl of this dilution were introduced into each of the 48-96 wells of the 96 well plates (Greiner Labortechnik, Frickenhausen, Germany). After approximately one week, the medium was removed, and 100 μl medium with 10% FCS were added. After approximately 2 weeks, the wells were examined under the microscope for one colony per well. Some were selected, and the cells were detached with 25 μl trypsine for 5 min at 37° C. and completely transferred into the well of a 24-well-plate (Greiner Labortechnik, Frickenhausen, Germany), which was completely filled with 500 μl medium containing 10% FCS. After one week, this precedure was repeated with 100 μl trypsine, and all cells were placed into a well of a 6-wellplate. After another 3 to 4 days, 100 μl supernatant per well were taken and examined by Western blotting andsubsequent immuno printing. The cell population, which finally showed the strongest band, was further expressed under selection pressure and cryoconserved, if applicable.
Concept and Establishment of a Solid Phase Assay (cf. Example I, 1.2.8.4)
The assay comprises immobilizing the hybrids as Strep-tagII fusion proteins to Strep-Tactin, which is bound to the surface of ELISA plates. The subsequent addition of buffer and serum should facilitate the conduction of the complementconsumption-assays in the ELISA plate.
For immobilization, a Strep-tagII was selected, which is a peptide consisting of 8 amino acids (WSHPQFEK). With a dissociation constant of 1 μM to Strep-Tactin, the Strep-tag is suitable for directed immobilization as well as for proteinpurification and detection purposes.
Fusion of an Affinity Tag to CVF and C3
The presence of the affinity tag is essential for the performance of the developed solidphase-assay. The proteins which shall be analyzed need a fusion tag; therefore, an enterinase-cleavage site was inserted into the cDNA of CVF and C3 between the signal sequence and the Iterminus of the Strep-tagII which allows the cleavage of the affinity tags lateron.
Using the CVF-cDNA two amplification products were generated in two PCR reactions using the oligonucleotides S35 and AS26 and the oligonucleotides S23 and AS34. The amplification products were hybridized using PCR. The amplification product was digested using the restriction enzymes KpnI and Eco72I and ligated into a pcDNA3CVF-vector which was digested analogously. The cDNA of human C3 was treated in an analogous manner. Using the oligonucleotides S01 and AS36 and S37 and AS03, amplification products were generated and hybridized. For insertion, the restriction sites NotI andBpull02I were used.
Alternatively, His tags were used as affinity tags, which in analogy to the Strep-tags—were inserted between the signal sequence and the N-terminus.
Briefly, for H6, two amplification products were generated using St-hC3 with the oligonucleotides S01 and AS61 and with the oligonucleotides S62 and AS03. The amplification products were hybridized by PCR. Then, the amplification product was inserted via the restriction sites NotI and Bpull02I in an analogously digested vector. The successful transient expression of the hybrid HisH6 was verified in a sandwichELISA and in an immunoblot. The densitometric quantification which was performed on the basis of the immunoblot resulted in yields of 1-2 mg/l.
Subsequently, St-CVF, St-C3 or the hybrids His-H6 or St-H6 hybrids, respectively, were successfully expressed in CHocells. A densitometric quantification was performed on the basis of an immunoblot and resulted in yields, which are comparable to the respective yields of wildtype proteins.
Evaluation of Assay Conditions
For establishing this solid-phase-assay which should offer the possibility of directly characterizing proteins after transient expression, an evaluation of the conditions was performed. The basis for evaluation is the adartion of the complement consumption assay to a solid phase-format. In order to determine the most suitable conditions, native CVF was used in a complement consumption assay in an ELISA plate under variation of different parameters.
For incubating the samples in the ELISA plate agitation at 37° C., a common incubation shaker was used. For optimizing the conditions, the reaction mixtures were measured at different rotation velocities for ensuring sufficient admixture of the samples, which is necessary for the assay. Further reaction mixtures were analyzed with different pre-incubation periods for obtaining representative values for the expected protein concentration range of 20200 ng protein.
The best results were obtained at a rotation velocity of 150 rpm and a pre-incubation time of 3 hours, since the samples with different protein concentrations represent significant distinguishable values.
Activity Studies of the Strep-tag-Fusion Proteins
A prerequesite for the solid phase assay is the accessibility of the Strep-tag-fusion protein, which was examined in an ELISA analysis. The fusion proteins St-CVF and St-C3 were selectively immobilized to Strep-Tactin on the surface on an ELISA plate. Detection of the target proteins was performed using polyclonal sera. Both proteins could be immobilized and detected, thus demonstrating the accessibility of the Strep-tag.
The concentration of the two proteins in the supernatant of the transient expression was quantified densitometrically on the basis of immunoblot. Polyclonal sera against C3 and CVF were used. The quantified recombinant proteins were then immobilized in comparable concentrations to Strep-Tactin and utilized in a complement consumption assay under evaluated conditions.
The activity of the recombinant CVF exprssed in mammalia could be demonstrated through significant reduction of hemolysis compared to human C3.
The positive results which were achieved also confirm that the established solid phase complement consumption assay is an effective method for characterizing hybrids consisting of CVF and human C3.
Discussion
The solid phase-assay enables the binding of CVF and of the hybrids via an affinity tag to a protein immobilized on a suitable surface. In this manner, the interfering components could be separated and a subsequent complement consumption assay with the immobilized proteins enables the characterization of the hybrids.
The Strep-tagII was introduced via oligonucleotides between the signal sequence and the sterminus of the CVF and C3, respectively. Additionally, an enterokinase-restriction site was inserted which allows the cleavage of the Strep-tagII lateron.
All fusion proteins were successfully expressed in CHOcells, and the accessibility of the fusion peptide was confirmed by ELISA analysis. The fusion of the Strep-tag did neither influence secretion nor expression yield, as shown by densitometric quantification.
The performing of the complement-consumption-assay in a solid-phase-system requires an adaption of the assay conditions. For this reason, an evaluation of different parameters was performed. Different concentrations of nCVF were subjected to complement consumption-assays under different conditions in an ELISA plate. Both the rotation velocity as well as the time period of the pre incubation was varied. Subsequent studies under optimized conditions (150 rpm, 3 h pre-incubation with St-CVF) confirmed the general feasability of this assay system. A significant complementconsuming activity was demonstrated for St-CVF. In contrast, the control protein St-C3 did not show an activity.
The establishing of solid-phase-assays therefore allows for the first time the efficient characterization of transiently expressed proteins comprising a Strep-tag-fusion peptide. The successful characterization of te recombinant CVF confirmed that a further processing of the twachain CVF is not required for activity.
Generation and Expression of H5
Cloning and Expression of Construct H5
A hybrid was constructed, which contains th human β-chain and additionally the humanized Factor B and Factor H-binding sites as well as the cleavage sites for Protease Factor I. In order to generate the construct, BglII restriction sites were used. Construct H5 is shown in
For cloning of construct H5, a fragment consisting of pUC18 and the 3′terminus of CVF was obtained from pUC18CVF* utilizing Ecll36I and BglII restriction. Subsequently, said frgment was ligated with a fragment obtained from pcDNA3hC3 by EcoRI restriction, followed by mung bean nuclease digestion and BglII restriction. The latter fragment had a size of 1870 bp contained the 5′ terminus of C3 cDNA. The vector ligated in this way(referred to as H2Δ2307bp) contained 1800 bp of the C3 5′ terminus and 1000 bp of the CVF 3′ terminus.
In order to complete construct H5, vector H2Δ2307bp was digested with BglII. The middle region of the C3cDNA was isolated from plasmid pUC18hC3 via its BglII restriction sites and inserted into the vector. The resulting construct H5 was then inserted into an analogously digested pcDNA3-vector via the EagI-restriction sites (
Expression in CHOcells was confirmed in an ELISA via Strep Tactin and in an immunoblot. Quantification which was performed on the basis of an immunoblot resulted in yields of 12 mg/l, wherein the polyclonal serum was employed agaist C3. Since hybrid H5 has 90% identity compared to human C3, it can be presumed that the polyclonal serum detects both proteins with a variance that is lower than the one of densitometric quantification.
Determination of the densitometric concentration wa confirmed by a sandwich ELISA. Here, a monoclonal C3dantibody and the antibody fragment C3-1, respectively, were immobilized on the surface of an ELISA plate and then incubated with the recombinant proteins. The detection was performed using a polyclonal C3-antiserum. In addition to the samples, various concentrations of human C3 were employed. The evaluation of the ELISA analysis confirmed the concentrations obtained from the densitometric quantification.
Characterization of Hybrid H5
Following successful transient expression, hybrid H5 was used in a solid phase-assay, where CVF and human C3 served as controls (
Hybrid H5 clearly showed complement-consuming activity which is comparable to CVF. Despite of ist degree of 90% humanization, hybrid H5 completely retains complement-consuming activity.
To analyze the functional activity of derivative H5 more in detail, we established a stably transfected CHO cell line and purified the protein using thiophilic and streptactin resins (for details see Example I, 1.2.6). A detailed analysis of the complement consuming activity of H5 over a broad concentration range confirmed the CVF-like activity observed in analyses of transiently expressed H5. As evident from
Furthermore, we analyzed the formation and stability of the convertase generated by the derivative H5 (for details see Example I, 1.2.8.6). As shown in
These results confirm the fact that, in addition to the CVFα-chain, also the CVF-γ-chain as well as the C3a- and C3d-homologous regions can be substituted against human C3 without loss of complement-consuming activity.
Expression and Characterisation of Hybrid H6
Cloning and Expression of Hybrid H6
Due to the fact that the complement-consuming activity of the hybrid H5 was successfully detected, a further minimizing of the CVF region was envisaged. If the C-termini of CVF and human C3 are compared, it is obvious that in the depicted region upstream of the Bsp1407I-restriction sites the identify between human C3 and CVF is 56%, whereas the C-termini of the two proteins merely exhibits in an identity of 44% (
For the construction of the ew hybrid H6, the Bsp1407I-restriction site was used. The C-terminal region of CVF was selected, whereas the region upstream of the Bsp1407Irestriction site was humanized. Hybrid H6 has 96.3% identity to human C3 (
For the cloning of construct H6 a 350 bp-fragment from pUC18CVF* was amplified by PCR using the oligonucleotides S50 and AS51. This fragment contains the 3′-terminus of CVF. The amplification product and the vector pcDNASthC3 were digested with Bsp1407I and XbaI and ligated (
Expression of hybrid H6 was successfully confirmed in an ELISA via Strep-Tactin and in an immunoblot.
The concentration of hybrid H6 in the supernatant of the transient expression was quantified densitometrically on the basis of an immunoblot, wherein poyclonal serum against C3 was employed. This yielded concentrations of approx.1-2 mg/l supernatant.
Characerization of Hybrid H6
In addition to densitometric determination, a sandwich ELISA was performed for quantifying recombinant C3 and hybrid H6. The quantification resulted in identical concentrations. Then, hybrid H6 was examined in a solid phaseassay for complement-consuming activity (
The activity of hybrid H6 was shown in the solid phaseassay. Humanization of the CVF-molecule was successfully performed, whereby the complement-consuming activity was retained. H6 was purified for further characterization.
Cloning and Expression of Hybrid His-H6
For a detailed characterization of hybrid H6, purification was performed according to established techniques using affinity chromatographical methods on the basis of the Histag-system. For this purpose, the insertion of the Histag was performed in analogy to the insertion of Strep-tag between signal sequence and N-terminus.
Briefly, two amplification products were generated using St-hC3 with the oligmnucleotides S01 and AS61 and with the oligonucleotides S62 and AS03. The amplification products were hybridized by PCR. Then, the amplification product was inserted via the restriction sites NotI and Bmll02I in an analogously digested vector. The successful transient expression of the hybrid HisH6 was verified in a sandwich-ELISA and in an immunoblot. The densitometric quantification which was performed on the basis of the immunoblot resulted in yields of 1-2 mg/l.
Purification and Characterisation of Hybrid H6
For the purification of the hybrid, stable expressing CHO cells were monoclonalized and expanded. Subsequently, imidazole was added to the supernatant of the CHGcells in a final concentration of 20 mM and the mixture was incubated with Ni-NTA-matrix. The elution fractions were then analyzed in an immunoblot. The fractions were pooled, dialyzed and quantified in a Sandwich-ELISA using a monoclonal antibody. Densitometric determination was performs on the basis of an immunoblot and it confirmed the quantification achieved with the ELISA. Protein concentrations of 34 μg/ml were obtained. For determining purity, the samples were separated by SD-SPAGE, and the gel was subjected to silver staining. Itwas determined that the purification procedure via the Histag resulted in a restricted purification, since a part of the protein did not bind to the matrix. Furthermore, the fractions contained a strong protein background. For subsequent studies, however, sufficient concentrations of the hybrids were obtained.
For further characterization of hybrid H6, first the C3 convertase activity was determined in a complement consumption assay in solution (
In a Bystander Lysistest it was determined whether hybrid H6 also has the ability (compared to CVF) to activate the complement system via a C5-convertase-activity in fluid phase. The guinea pig serum employed is activated by CVF and the guinea pig erythrocytes are lysed by subsequent formation of the membrane attack complex. Subsequently, the released hemoglobin can be measured in the supernatant at 412 nm. It was demonstrated that hybrid H6 does not exert significant fluid phase C5convertase activity.
Discussion
Recombinant CVF was deteted in the supernatant of all three mammalian cell lines (CHO, CO57 and HEK293). The densitometrically quantified yields were approx. 0.53 mg/l.
The recombinant CVF with an apparent molecular weight of 210 kDa had the structure of two chains. Here, theCVF-α-chain as in the native protein is a singular chain, whereas the γ and the β-chain are expressed together with regions which are homologous to the C-3 fragments C3d and C3a as one chain. The processing is analogous to the processing of the twochain human C3; only 4 arginine residues are removed in the two proteins. Further processing of CVF in the cobra by a venom protease (O'Keefeet al., 1988) does not occur in mammalian and insect cells. However, the recombinant twochain CVF from insect cells exhibits a comparable activity to native CVF (Kock, 1996).
The modification of CVF by the generation and characterization of different hybrids from CVF and human C3 led to the expression of hybrids, the majority of which consists of human C3. In order to analyze in how far the expression systems are also suitable for such hybrids, human C3 was expressed in addition to CVF in CHO, HEK293- and COS-7-cells. The recombinant protein was also detected in the cell supernatant. Recombinant C3 showed an apparent molecular weight of 210 kDa and expression yields ranged from 13 mg/l, which is in compliance with literature values (Feckeet al., 1998). For the secretory expression of the proteins, the native signal sequences of CVF and human C3 were used.
Due to the fact that CHO-cells allow for stable integration and further exhibit higher yields compared to HEK293cells, CHO-cells were selected for further studies.
Successful expression of CVF and human C3 in mammalian cells can be considered as a useful basis for furthr expression of hybrids of CVF and human C3.
Following successful expression of CVF and human C3, it was possible to establish mammalian cells as expression systems for these hybrids. Subsequently, a method had to be developed for analyzing the complement-consuming activity of the hybrids. For this purpose, recombinant CVF in the supernatants of transient expression was studied with regard to its decomplementing activity in complement consumption-assays. However, no difference to control supernatants without CVF could be detected. Neither by a transient expression with the serum replacement substance Nutridoma nor by culturing the cells in serum-free medium or by culturing in proteinfree medium an activity of the recombinat CVF could be detected. The densitometrically quantified yields of expression with different replacment substances ranged from 0.3-1.5 mg/l. Thus, they were slightly lower compared to the yields obtained by use of serum containing medium. The results indicated that direct characterization of rCVF in supernatants of transient expression is not feasable. Consequently, further strategies where analyzed to characterize CVF and diverse hybrids after transient expression without the need to apply time- and cost-consuming purification procedure for increasing the concentration of C3/CVF-hybrid molecules. Different techniques for concentrating and purification were examined in order to achieve a separation of the components which prevent characterization of rCVF in supernatants. Subsequent analpes of the samples in complement consumption assays, however, did not reveal any success of the applied strategies. The decomplementing activity in the samples did not differ significantly from corresponding controls, but the results pointed to the present of high molecular weight compounds in the samples interfering with the complement consumption assays. Based on these data, solid-phase assays were developed for complement consumption analyses. One strategy employed a Strep-tag fused to the N-terminus of the recombinant proteins.
Hybrid H5 only comprises the functional relevant 5chain of CVF and shares identity with human C3 of 90.7%. Based on the data available for human C3, it comprises the Factor B and Factor H-binding sites as well as the cleavage sites for Factor I.
The hybrid was generated, fused with a Strep-tag and was successfully expressed in CHOcells.
Subsequently, complement consuming activity was characterized in a solid phase-assay. It was confirmed that H5 has an activity comparable to CVF.
For the first time, a C3/CVF hybrid protein was provided where the CVF-portion could be reduced to 9% while retaining a CVFlike activity. The hybrid only comprises the CVFP-chain and exhibits decomplementing activity.
Up to now, the C3-region, which is homologous to the CVF-chain has only insufficiently been studied. Therefore, only a few binding sites are identified. A known binding site for the complement receptor CR3 is located in the region of amino acids 13611381 (Wright et al., 1987). CR3, which is present on macrophages and killer cells, binds to C3bi which has bound on the surface of pathogenes and mediates the destruction of the pathogene (Newman et al., 1984). In the conducted in vitro-assay, the binding of CR3 does not have any influence. In the vicinity of the CR3region, the binding site for properdin with the amino acids 14241432 is in the analogous C3-region (Daoudaki et al., 1988. properdin binds and stabilizes the alternative C3-convertase C3bBb (Fearon et al., 1975). A binding site is also postulated in the CVFmolecule. The identity of the binding sites of C3 and CVF amounts to 70%, which is clearly higher than the identity of the whole proteins (Fritznger et al., 1994). However, even an identity of 70% is in accordance with crucial structural differences in the CVF-β-chain and the C3-region homologous to the CVF-β-chain.
Considering the available data for human C3, H5 should be inactive. It comprises the human cleavage sites for Factor I and an additional cleavage site for Factor H. Furthermore, the increased stability of the CVF-dependent convertase as a result of the stronger binding of CVF to Factor B, should be lost by humanization in H5 since all postulated binding regions for Factor B are located in regions replaced by C3sequences. However, since no loss ofactivity was observed, it is possible that structural differences between H5 and C3 protect the hybrid HS from a regulation by Factor H and Factor I. Further, it would be conceivable that an additional region which is located in the region of the CVF-β-chain is responsible for the stronger binding to Factor B.
Quantification of hybrid HS by antibodybased procedures did not pose any problems. Since the identity with C3 amounts to approx. 91%, it can be assumed that the polyclonal serum against C3 recognizes the hybrid HS with a coiparable reactivity. Therefore, quantification based on densitometric immunoblot analysis or ELISA with polyclonal anti-C3 sera was considered to provide reliable results.
This new molecule is of therapeutic relevance since it should clearly exhibit a lover immunogenicity compared to CVF or CVF in which the α-chain is replaced by the corresponding human C3β-chain. Nevertheless, it shows the same complementconsuming activity. It should be possible to apply this molecule in low concentrations.
After having demonstrated that hybrid H5 exerts complement activating activity, the analogous region of the human C-3cDNA was compared to the homologous regions of the CVFcDNA. After analyzing the identities of CVF and human C3 in the analogous terminal regions, a further construct was generated, H6. Hybrid H6 corresponds to human C3 in the first 1527 amino acids; the C-terminus of the protein is CVF-sequence and exhibits an identity of 96.3% to human C3.
After cloning, the hybrid was transiently expressed in CHO cells and yields of 1-2 mg/l were obtained. Subsequent analysis of the activity of the hybrid H6 in a solid phaseassay demonstrated a substantial complement-consuming activity.
Upon quantification, activity decreased only by 50% when utilizing comparable amaunts of H6 and CVF in the solid phaseassay. [0198] With hybrids H5 and H6, two humanized CVFmolecules were generated, which can be used in a therapeutic application. The hybrid H5 exhibits a complement-consuming activity comparable to that of CVF. Hybrid H6 shows 96% identity to human C3 and, most probably, this molecule is not immunogenic. However, H6 shows a loss of approx. 50% in activity compared to the CVFmolecule.
Since hybrid H6 could also be identified as a C3derivative containing less than 4% foreignamino acid residues, further data were collected. First, a Histag fusion protein was generated. Hybrid H6 was provided with a Histag using oligonucleotides as already done for the cloning of the Strep-tag. The His-tag fusion protein was successfully expressed in CHOcells given yields of 1-3 mg/l. After generating a stable cell-line, hybrid His-H6 was enriched using IMAC. However, the majority of the protein could not be immobilized and was detected in the flow-through. Additionally, the elution fractions were contaminated by a strong protein background. Nevertheless, sufficient amounts of the protein were obtained for further analyses.
The protein was densitometrically determined in the immunoblot and quantification was confirmed using immobilized anti-C3 antibodies in a Sandwich-ELISA. Concentrations of 3-4 mg/l were determined. In a subsequent complement-consumption-assay, in which up to 80 ng protein was employed, a loss of activity of only 32% compared to CVF was demonstrated. The results show that thegenerated molecule exhibits a clear complement-consuming activity despite its 96% humanization. Therefore, the requirements of low dose applications are fulfilled. Therefore, CVF is an attractive complement modulator.
Additionally, the purified protein wa employed in a Bystander Lysis-Assay for the determination of fluidphase-C5-convertase activity. In contrast to CVF, hybrid H6 did not exert significant fluid phase-C5-convertase activity. The Bystander Lysis-activity of CVF leads to a fast and massive acumulation of C5a, which can cause severe tissue damages (Till et al., 1982; Schmid et al., 1997). Therefore, a loss in C5-convertase activity caused by humanizing seems to represent an advantage.
Using cassette mutagenetic C3-derivatives were generated which are capable of forming stable C3-convertases. Specific sequences of CVF were utilized for the replacement of corresponding C-3regions (cf. alignment in
All C3/CVF hybrids as well as CVF and human C3 as such were utilized in parallel complement consumption assays.
The assays confirmed that the complementconsuming activity of both CVF and H5, the latter of which has 90.7% identity to human C3, are comparable. The activity of H6 having 96% identity to human C3 was slightly decreased incomparison to CVF (
The therapeutic application of a complement modulator or inhibitor is attractive for treating several complementassociated diseases or diseases affected by complement activation such as asthma (Regal et al. 1993), systemic lupas erythematodes (Belmont et al., 1996), glomerulonephritis (Couseret al, 1985), rheumatoid arthritis (Kemp et al., 1992), Alzheimer's disease (Rogerset al., 1992), multiple sclerosis (Piddlesdenet al., 1996), sepsis (Hack et al., 1989), hyperacute rejection and transplant rejection (Bachet al., 1995, Baldwin et al., 1995), cardiopulmonary bypass, myocardial infarction, angioplasty, nephritis, dermatomyositis, pemphigoid, spinal cord injury, and Parkinson's disease.
Comparison of complement inhibitorswith respect to production costs and dosage requirements for therapeutic applications indicates a preference for the C3/CVF hybrid proteins. The generated C-3 derivatives are, as enzymes, superior to other complement inhibitors. The 32% activity loss observed for H6 can be compensated by a higher application dose of approx. 400 μg/kg. However, this dose is still low compared to other inhibitors known in the art which require up to 80fold higher doses to be applied. The current invention provides human C3-derivatives that are capable of forming C3convertases exerting an extended CVF,Bb-like half-life of several hours, compared to 1.5 minutes of the naturally occurring C3convertase, thus escaping the physiological degradation mechanisms. The high degrees of identity to human C3 should allow repetitive therapeutic applications of the polypeptides of the invention.
List of Abbreviations
This application is based on, and claims the benefit of, U.S. Provisional Application No.60/484,797, filed Jul. 3, 2003, and entitled Complement Depletion using recomlinant Human C3-Derivatives, and which is incorporated herein by reference
| Number | Date | Country | |
|---|---|---|---|
| 60484797 | Jul 2003 | US |
