Skip to main content
NCBI home page
Protein Science : A Publication of the Protein Society logoLink to Protein Science : A Publication of the Protein Society
. 2002 Oct;11(10):2512–2521. doi: 10.1110/ps.0212402

Bacterial expression and membrane targeting of the rat complement regulator Crry: A new model anticomplement therapeutic

Deborah A Fraser 1, Claire L Harris 1, Richard AG Smith 2, B Paul Morgan 1
PMCID: PMC2373695  PMID: 12237472

Abstract

Inappropriate or unregulated activation of complement can contribute to pathology in inflammatory diseases. Previous studies have shown that soluble recombinant regulators of complement are effective in animal models and some human diseases. However, limitations include cost, rapid clearance, and unwanted systemic effects. To avoid some of these problems, bacterial expression of regulators has been optimized and methods for the addition of a membrane-targeting moiety to the complement regulator developed. When administered directly to sites of inflammation, membrane-targeted human regulators are retained and inhibit complement-activation locally. To test the efficacy of membrane-targeted complement regulators in vivo, we have undertaken the expression and membrane targeting of the rat-complement regulator Crry. A soluble recombinant form of Crry, containing only the first four short consensus repeats, was expressed in a mammalian expression system and shown to be functional as a fluid phase regulator. To generate the quantities required for testing in vivo, Crry was expressed in bacteria and refolded successfully. Refolded protein had full-complement regulatory activity in vitro. Attachment of a membrane address tag conferred membrane-binding capacity and greatly increased complement regulatory function in vitro. This novel anticomplement agent can now be applied to rat models of arthritis and other inflammatory diseases.

Keywords: Complement, CR1-related gene Y (Crry), anticomplement therapy, targeting, rat complement regulatory protein, complement receptor type 1 (CR1), bacterial expression

The complement system is an important host-defense mechanism. However, excessive or inappropriate activation of complement can contribute to the pathogenesis of acute and chronic disorders as diverse as hyperacute graft rejection and rheumatoid arthritis (RA) (Whaley et al. 1993). Recombinant soluble forms of the naturally occurring complement regulators are therefore promising candidates for therapeutic intervention in many situations. Numerous studies have investigated the therapeutic efficacy of recombinant soluble complement regulators in animal models of inflammatory diseases such as ischemia/reperfusion injury; most have used soluble forms of human complement receptor type-1 (CR1), which has been shown to be a highly potent anticomplement agent in man and rodents (Weisman et al. 1990; Hill et al. 1992; Lindsay et al. 1992; Chavez-Cartaya et al. 1995). CR1 is a large polymorphic molecule with its most common isoform containing 30 structural units, known as short consensus repeats (SCR), linked to the plasma membrane by a transmembrane domain (Wong et al. 1989). This large molecule acts on the C3 and C5 cleaving enzymes of the activation pathways, directly causing decay of the enzyme complex (decay acceleration) and indirectly causing irreversible inactivation by acting as a cofactor for the C3b/C4b-cleaving plasma enzyme factor I (cofactor activity). The three amino-terminal SCRs in CR1 comprise one of three active sites within the 30-SCR native protein, and are sufficient to retain significant complement inhibitory activity, including both decay acceleration and cofactor functions. A truncated form of CR1 comprising these three SCRs, has been expressed and refolded from Escherichia coli and shown to inhibit complement activation (Dodd et al. 1995). Modification of this small CR1-derived molecule by adding at the carboxl terminus a membrane-targeting moiety consisting of two sequentially linked outer-cell membrane ligands or addressins (giving a product termed APT070) markedly enhanced function in in vitro complement activation assays and increased the circulating half-life in vivo (Smith et al. 2001).

A major problem with using sCR1 or derivatives in animal models of disease is that these human proteins are strongly immunogenic in rats and other experimental animals; neutralizing antibodies appear after a single week, rendering impossible investigations in chronic disease models (Piddlesden et al. 1994). To gain a better understanding of the potential that targeted soluble complement regulators have in treating human disease and to assess the long-term effects in subacute and chronic models, we have undertaken to develop soluble complement regulators derived from the relevant experimental species. Rats and mice express analogs of each of the human C regulators. The functional analog of human CR1 in rodents is a powerful membrane complement regulator, termed Crry that, like CR1, has both decay acceleration and cofactor function, making it an ideal model complement regulator (Kim et al. 1995). In the rat, Crry is expressed as two isoforms, comprising six or seven SCRs, respectively, and is linked to the plasma membrane via a transmembrane domain (Quigg et al. 1995). Soluble recombinant forms of rat Crry containing the five amino-terminal SCRs have been generated previously in Pichia pastoris and shown to be functional in vitro (He et al. 1997). To generate in an economically viable manner the large amounts of recombinant soluble Crry necessary for characterizing this protein in vivo, we undertook to express a soluble recombinant form of rat Crry in a bacterial expression system. The SCR structural unit represents a considerable challenge for folding in a bacterial system. Three SCRs from CR1 have been refolded successfully with retention of activity (7). Structural clues led us to believe that the four amino-terminal SCRs of Crry would be sufficient for activity and might be more amenable to refolding than the five SCRs previously made in yeast and shown to retain function. We first expressed these four SCRs in a eukaryotic system and showed strong complement regulatory activity in glycosylated and unglycosylated proteins, the latter an important control for subsequent prokaryotic expression (De Bernardez Clark 1998). Expression and refolding from E. coli was optimized to obtain high yields of fully active protein that was subsequently modified by addition of a carboxy-terminal membrane targeting moiety, the addressin. The targeted protein showed markedly increased potency in complement inhibition assays when compared with the untargeted parent molecule. This new agent can now be tested in rat models of inflammatory disease without the problems of immunogenicity encountered when using derivatives of human complement regulators.

Results

Production and characterization of soluble recombinant Crry in CHO cells

cDNA encoding the following constructs was cloned into the pDR2ΔEF1α expression vector: sCrry, encoding the four amino-terminal SCRs of rat Crry; sCrry-Cys, encoding the amino-terminal SCRs of rat Crry with an additional 8 amino acids at the carboxyl terminus consisting of a Ser/Gly spacer region and carboxy-terminal Cys residue; sCrry-Cys(N|ZeG), encoding a construct identical to sCrry-Cys except for an amino acid substitution (N247|ZeG). A schematic of these constructs is shown in Figure 1 ▶. These constructs were transfected into CHO cells, stable cell lines were generated, and the supernatant harvested. Crry was purified from CHO cell supernatant by affinity purification. Purified proteins were analyzed by SDS-PAGE and Western blot (Fig. 2 ▶) by use of the monoclonal anti-Crry antibody TLD1C11 to probe. Under nonreducing conditions, sCrry migrated as a doublet of Mr 28 and 32 kD. sCrry-Cys migrated as a diffuse band of 28–33 kD, and further minor bands at 60, 64, and >90 kD. sCrry-Cys(N|ZeG) had bands at 30, 60, and a very faint band at >90 kD. Under reducing conditions, sCrry-Cys(N|ZeG) (unglycosylated) migrated as a single band of 30 kD, whereas sCrry and sCrry-Cys migrated as doublets of 30 and 32 kD, indicating that these samples contained both glycosylated (32 kD) and unglycosylated (30 kD) forms of Crry. Densitometric analysis of silver-stained SDS-PAGE gels showed that the higher Mr aggregates in nonreduced gels of sCrry-Cys and sCrry-Cys(N|ZeG) represented <5% of the total protein in the sample (data not shown).

Fig. 1.

Fig. 1.

Open in a new tab

Schematic representation of Crry constructs. Constructs expressed in mammalian (A) and bacterial (B) expression systems are shown. All constructs were engineered to contain the four amino-terminal SCRs of rat Crry (represented as circles). The single N-linked glycosylation site is indicated (CHO). Constructs engineered to contain an eight amino acid spacer sequence consisting of serine and glycine residues and a carboxy-terminal cysteine residue are shown (–C). Modification of the carboxy-terminal cysteine residue of bacterially expressed sCrry-Cys with the membrane address tag APT542 is also shown (+++ arrow).

Fig. 2.

Fig. 2.

Open in a new tab

Western blot analysis of purified Crry constructs. sCrry (1), sCrry-Cys (2), and sCrry-Cys(N|ZeG) (3) were purified from supernatant from cells transfected with plasmids encoding these constructs by affinity chromatography. Purified proteins were subjected to SDS-PAGE on a 10% gel under both nonreducing (left) and reducing (right) conditions and Western blotted. Blots were probed with monoclonal anti-Crry antibody (TLD1C11) and bound antibody detected using HRPO-conjugated secondary antibody. Bands were visualized using ECL. Molecular weight markers are indicated on the left and arrows show the glycosylated (32 kD) and unglycosylated (30 kD) forms of sCrry-Cys.

In vitro functional analysis of recombinant forms of soluble Crry expressed in CHO

To confirm the complement inhibitory activity of sCrry and investigate functional differences due to the modification of sCrry with a 7 amino acid spacer and carboxy-terminal cysteine residue, the capacities of sCrry and sCrry-Cys to inhibit the classical pathway of complement were investigated using a haemolytic assay (Fig. 3A ▶). Both sCrry and sCrry-Cys were powerful inhibitors of lysis, with equivalent activity, indicating that the spacer and carboxy-terminal cysteine region had no adverse effects on the function of sCrry-Cys. Formation of disulphide-bonded dimers and higher multimers, apparent on SDS-PAGE (Fig. 2 ▶), did not adversely affect function. The influence of glycosylation was investigated by comparing the ability of sCrry-Cys and sCrry-Cys(N|ZeG) to inhibit the classical pathway of complement in a haemolytic assay (Fig. 3B ▶). Both glycosylated and unglycosylated forms of soluble Crry were potent complement inhibitors, with identical levels of activity when compared directly, showing that glycosylation had no effect on the complement-inhibitory function of these molecules and indicating that lack of glycosylation in bacterial expression systems would not compromise activity. Specific activity of sCrry-Cys in these separate assays was closely comparable, showing the reproducibility of the method.

Fig. 3.

Fig. 3.

Open in a new tab

In vitro complement regulatory function of Crry expressed in a mammalian system. Antibody-sensitized erythrocytes were incubated in CFD with rat serum and different concentrations of (A) sCrry (•) and sCrry-Cys (▪), or (B) sCrry-Cys (▪) and sCrry-Cys(N|ZeG) (cross) expressed in a mammalian system. BSA was used as control in both cases (♦). Haemolysis was assessed by release of haemoglobin to the supernatant and lysis was calculated as a percentage of the lysis obtained by incubation of EA with rat serum and equivalent concentrations of the control protein. Results represent the mean value +/− SD of three determinations.

Production of soluble recombinant Crry in E. coli

cDNA encoding the four amino-terminal SCRs of rat Crry modified with a 7 amino acid carboxy-terminal spacer region and Cys residue (sCrry-Cys) was cloned into the bacterial expression vector pET26b. This construct was transformed into E. coli BL21(DE3), and large amounts of sCrry-Cys were then generated by fermentation. sCrry-Cys was expressed as insoluble inclusion bodies that were isolated, washed, and solubilized. Solubilized protein, renatured by rapid dilution into cold refold buffer, was concentrated to 3 mg/mL by ultrafiltration and buffer exchanged into PBS. The final protein yield from 2 L of bacterial fermentation culture was 200 mg as determined by Coomassie protein assay (Pierce). sCrry-Cys was finally purified by anion exchange chromatography. Samples from the pre- and postinduction fermentation, isolation, solubilization, and refolding of sCrry-Cys from bacterial inclusion bodies were analyzed by SDS-PAGE stained with Coomassie blue (Fig. 4 ▶). Numerous bands were present in the fermentation samples, but an additional band at the predicted Mr for sCrry-Cys (30 kD) was present only in the postinduction sample (Fig. 4 ▶, lane2). This band was the major protein in the bacterial pellet before (Fig. 4 ▶, lane 4) and after (lane5) solubilization and was retained following refolding (lane 6) and after gel filtration (lane 7). Protein was further purified and concentrated by anion exchange chromatography. Negligible amounts of dimer were present in freshly purified sCrry-Cys and in samples stored at −20°C, although samples stored at 4°C did show a gradual increase in dimer formation. The molecular mass of purified sCrry-Cys was assessed by MALDI-Tof analysis; a single major peak of molecular mass 29299 Da was obtained.

Fig. 4.

Fig. 4.

Open in a new tab

Bacterial expression, isolation, and refolding of sCrry-Cys. Expression of sCrry-Cys (30 kD) was induced in bacteria grown to high-cell densities by fermentation (BioFlo3000). Samples of the pre-induction (lane 1) and 1-h post induction (lane 2) cultures are shown. Inclusion bodies of induced sCrry-Cys were isolated from bacterial cell pellets by high-pressure homogenization, and the pellets washed to remove bacterial contaminants. Samples of the supernatant (lane 3) and pellet (lane 4) after the final wash are shown. sCrry-Cys was solubilized from washed inclusion bodies to 5 mg/mL in equilibration buffer (lane 5) and refolded by a 1/40 dilution into 0.02 M ethanolamine, 1 mM EDTA (pH 11.0) (lane 6). Refolded sCrry-Cys was buffer exchanged into PBS by gel filtration (lane 7). Samples at each stage were subjected to SDS-PAGE on a 4%–12% Bis-Tris gradient gel (Novex) under nonreducing conditions and stained with Coomassie blue. Molecular weight markers are shown at left. The arrow indicates the anticipated size of the sCrry-Cys protein.

In vitro functional analysis of soluble recombinant Crry expressed in E. coli

The ability of sCrry-Cys, generated and refolded from bacterial inclusion bodies, to inhibit the classical pathway of complement in a haemolytic assay was compared with the sCrry-Cys expressed in CHO cells. These molecules were of equivalent potency in this system when compared directly, showing that fully active protein had been refolded from the bacteria (Fig. 5 ▶).

Fig. 5.

Fig. 5.

Open in a new tab

In vitro complement regulatory function of Crry expressed in a bacterial system. Sensitized erythrocytes were incubated in CFD with rat serum and different concentrations of sCrry-Cys either expressed in a bacterial system and refolded from inclusion bodies (▴) or expressed in CHO cells (▪). Haemolysis was assessed by release of haemoglobin to the supernatant and lysis was expressed as a percentage of the lysis obtained in the presence of the control protein, BSA (♦). Results represent the mean value +/− SD of three determinations.

sCrry-Cys refolded from bacterial inclusion bodies was also tested for cofactor activity by incubation of human C3ma with rat factor I and soluble Crry as cofactor and was compared with the function of human sCR1 as cofactor. Samples were separated by SDS-PAGE and either Western blotted or silver stained to identify C3 cleavage products as an index of active cofactor (Fig. 6 ▶). Western analysis (Fig. 6A ▶) showed that the known cofactor sCR1 yielded a 43-kD fragment, indicative of cofactor activity (Fig. 6A ▶, lane 1). The four amino-terminal SCRs of rat Crry yielded the same fragment (Fig. 6A ▶, lane 2), showing that it was able to function as a cofactor for the factor I-mediated cleavage of C3ma. No significant cleavage was obtained when either factor I or the cofactor was omitted. Silver staining and densitometry to quantify cleavage (Fig. 6B ▶) showed that under the stated experimental conditions, sCR1 caused 82% cleavage of the C3 α-chain, whereas the same amount of sCrry-Cys caused 44% cleavage.

Fig. 6.

Fig. 6.

Open in a new tab

Analysis of cofactor activity. Methylamine inactivated C3 (C3ma) was incubated with rat factor I in the presence of the known cofactor sCR1or with sCrry-Cys, samples separated by SDS-PAGE on a 10% reducing gel and either Western blotted (A) or silver stained (B). (A). Western blotting (Lane 1) sCR1; (lane 2) sCrry-Cys. Arrows show the α and β chains of C3 and the 43-kD fragment generated from the α-chain upon factor I cleavage. Controls included C3ma alone (lane 3), C3ma incubated with factor I (lane 4), C3ma with sCR1 (lane 5), and C3ma with sCrry-Cys (lane 6). Blots were probed with polyclonal α-C3c and detected using HRPO-conjugated secondary antibody. Bands were visualized using ECL. Molecular weight markers are shown at left. (B) Silver staining: (Lane 1) sCR1; (lane 2) sCrry-Cys. Arrows show the α and β chains of C3 and the 43-kD fragment generated from the α-chain upon factor I cleavage. sCR1 andsCrry-Cys are also arrowed. Lanes 3 and 4 are controls in which sCR1 (lane 3) or sCrry-Cys (lane 4) were incubated with C3ma in the absence of factor I. Gels were stained and analyzed by densitometry. Boxes adjacent to the gel lanes show the results of densitometry for each band. The C3 β chain was used as an internal standard to correct for loading differences.

Production and in vitro functional analysis of membrane-targeted recombinant Crry (sCrry-APT542)

sCrry-Cys was modified at its carboxyl terminus by the addition of a membrane address tag, APT542, by use of thiol-interchange chemistry. SDS-PAGE analysis (Fig. 7 ▶) of soluble Crry before (Fig. 7 ▶, lane 1) and after (lane 2) conjugation to APT542 showed that addition of a membrane address tag increased the Mr of sCrry-Cys from 30 to ∼31 kD. As some unmodified sCrry-Cys remained in the reaction, the membrane-targeted protein (sCrry-APT542) was purified from the untargeted molecule, sCrry-Cys, by use of hydrophobic interaction chromatography, which exploits the change in the distribution of hydrophilic/hydrophobic regions induced by addition of the amphiphilic tag. SDS-PAGE analysis of purified sCrry-APT542 under nonreducing conditions revealed a single band at the predicted molecular weight of ∼31 kD (Fig. 7 ▶, lane 4).

Fig. 7.

Fig. 7.

Open in a new tab

Production and purification of membrane-targeted Crry. sCrry-Cys was modified at the carboxyl terminus by the addition of the membrane address tag APT542, with the resultant molecule termed sCrry-APT542. Tagged Crry was separated from the untagged molecule by hydrophobic interaction chromatography (HIC). Samples of sCrry-Cys before (lane 1) and after (lane 2) conjugation to APT542, and sCrry-Cys (lane 3) and sCrry-APT542 (lane 4) after separation by HIC were subjected to nonreducing SDS-PAGE on 4%–12% Bis-Tris gradient gels (Novex). Gels were stained with Coomassie blue. Arrows indicate the positions of sCrry-Cys and sCrry-APT542; molecular weight markers are shown at left.

The ability of sCrry-APT542 to inhibit the classical pathway of complement was compared with untargeted sCrry-Cys by use of nonwash (Fig. 8A ▶) and wash (Fig. 8B ▶) haemolytic assays. In a nonwash assay, soluble Crry containing a membrane-targeting moiety (sCrry-APT542) was 100-fold more active than the untargeted protein (sCrry-Cys). The wash assay showed that sCrry-APT542 was able to protect sheep erythrocytes from complement-mediated lysis even after two wash steps, indicating that the protein had bound firmly to the cells. Such anti-haemolytic activity was not seen with the unmodified sCrry-Cys after two wash steps.

Fig. 8.

Fig. 8.

Open in a new tab

In vitro complement regulatory function of membrane-targeted Crry. The complement inhibitory activities of sCrry-APT542 (cross) and sCrry-Cys (▴) were compared in a standard classical pathway assay (A) and in an assay in which the inhibitors were preincubated with the target cells and the cells washed to remove free inhibitor prior to exposure to complement (wash assay; B). Antibody-sensitized sheep erythrocytes were used as target in both assays. Lysis was expressed as a percentage of lysis obtained by incubation of EA with rat serum and the control protein, BSA (♦). Results represent the mean value +/− SD of three determinations.

Discussion

We report the generation of soluble recombinant rat Crry in eukaryotic and prokaryotic systems, introduction of a membrane-targeting moiety into the expressed protein, and characterization of the complement regulatory activity of the proteins. The eventual aim of the work is to develop powerful anticomplement therapies for testing in both acute and chronic disease models in rats. The value of using complement regulators in therapy is highlighted by the successful use of human proteins such as sCR1 in animal models and in human disease (Weisman et al. 1990; Hill et al. 1992; Lindsay et al. 1992; Chavez-Cartaya et al. 1995). However, studies to date using animal models have focused on acute diseases such as experimental demyelination (Piddlesden et al. 1994), glomerulonephritis (Couser et al. 1995) and myasthenia gravis (Piddlesden et al. 1996) due to the short half-lives and immunogenic nature of human complement regulators in rodents. Development of therapeutic agents for chronic conditions such as rheumatoid arthritis requires sustained complement inhibition over a longer time course than was possible in previous studies. To investigate long-term therapy with complement regulators in these models, the problem of immunogenicity can be addressed by the use of rodent-complement regulatory proteins. The problem of rapid clearance may be addressed by targeting the inhibitor to the site of complement activation in disease. Several strategies have been devised for targeting recombinant complement regulatory proteins to the cell surface. One approach involved modifying sCR1 by the incorporation of the sialylated tetrasaccharide sialyl lewisx (sLex) moiety at N-glycosylation sites to give a derivative termed TP20 (Rittershaus et al. 1999). TP20 is targeted to the endothelium by binding E-selectin and P-selectin (Lasky 1992), which are up-regulated during the early inflammatory response (Springer 1990). Another involved generation of fusion proteins that incorporate a cell-specific antibody or antibody fragment to target complement regulatory proteins to specific sites. Zhang and colleagues have in vitro successfully targeted both CD59 and DAF to the hapten dansyl incorporated onto the surface of CHO cells, by the generation of chimeras comprising CD59 or DAF linked to an anti-dansyl antibody fragment (Zhang et al. 1999, 2001). Such targeting may accomplish several therapeutically useful objectives, including a reduction in the dose of inhibitor required and reducing or eliminating systemic complement inhibition. A third strategy for targeting soluble proteins to tissues has been developed by incorporation of a tag comprising linear combinatorial array of individual ligand units, termed sequential membrane addressins (SMAs), which interact with a molecular address on a cell membrane. The first generation of these tags consists of a lipid moiety that interacts with the hydrophobic interior of the plasma membrane and a short positively charged peptide that is able to bind to the negatively charged phospholipid headgroups (Smith et al. 1998a; Smith and Smith 2001). Individually, SMAs provide only weak interactions with the membrane, but, in combination, they bind more strongly to cellular addresses. The tag can be synthetically linked via a carboxy-terminal cysteine to the soluble protein of choice, giving the technology a broad range of potential applications. The first agent on the basis of these principles is the anticomplement therapeutic APT070, which consists of the first three SCRs of human CR1 covalently attached at the carboxyl terminus to a 2-SMA membrane address tag (APT542) ( Smith et al. 1998a; Smith and Smith 2001). APT070 has been shown to have significant clinical benefit in antigen-induced arthritis in the rat, a model for human inflammatory arthritis (Linton et al. 2000), and has also been used effectively in ischemia/reperfusion injury (Dong et al. 1999). Generation of this type of membrane-targeted recombinant rat proteins could therefore provide useful therapeutic agents for testing in complement-mediated disease models in the rat.

The amino-terminal five SCRs of rat Crry have been expressed previously as soluble proteins in yeast and adenoviral expression systems and are shown to be functional in vitro (He et al. 1997; McGrath et al. 1999; Quigg et al. 2000). To establish principles for bacterial expression of Crry, we first generated in CHO cells a truncated soluble rat Crry comprising of only the four amino-terminal SCRs. Modified forms of the protein were made in which the single glycosylation site was eliminated to investigate the role of carbohydrate (not present in bacterially expressed proteins) and with a carboxy-terminal free cysteine necessary for addition of the addressin tag. Functional comparison of these soluble Crry proteins in a classical pathway haemolytic assay showed that the four SCR molecule was, as predicted, a powerful inhibitor of complement and that neither glycosylation nor addition of a carboxy-terminal cysteine had any effect on complement regulatory activity. The role of the single N-linked carbohydrate group in Crry had not been examined previously, but here we show conclusively that it is not relevant to function. SDS-PAGE analyses showed that addition of a carboxy-terminal cysteine caused a significant proportion of the expressed protein to dimerize and a smaller proportion to form higher oligomers, yet these events had no detectable effect on function. In addition to haemolytic assays, we also tested function of soluble rat Crry in a factor I cofactor assay. It has been assumed that rat Crry, like its counterpart in the mouse (28), has both decay accelerating and cofactor activities, but no formal test of this assumption has been reported. Here, we show that soluble recombinant rat Crry is an efficient cofactor for the factor I-mediated cleavage of methylamine-inactivated C3.

On the basis of these data, we proceeded to express the four amino-terminal SCRs of Crry with the Cys modification in bacteria. Prior to this study, the only SCR-containing complement regulatory protein reported to be expressed in an active form from bacteria was a truncated form of human CR1 comprising the amino-terminal three SCRs. This protein was successfully overexpressed in E. coli, and expression optimized for large-scale manufacture (Dodd et al. 1995; Smith 1998b). We used a protocol based on that used for the three SCR protein to express the even more complicated four SCR Crry. Typical protein yields from bacterial expression range from 5–100 mg protein per litre of culture (Durbin 1998), and as a consequence of these high expression levels, most proteins accumulate as dense, insoluble, and inactive protein aggregates of protein within inclusion bodies in the cell (Lilie et al. 1998). Isolation, solubilization, and renaturation of complex proteins in active form from these aggregates represents a significant hurdle (De Bernardez Clark 1998). Each SCR unit consists of 60–70 amino acids with four invariant cysteines that form two intradomain disulfide bonds, and integrity of the SCRs is essential for function (Campbell et al. 1988). Despite this complexity, the four SCRs of Crry were successfully refolded and renatured from the inclusion bodies, and the resultant protein had complement-inhibitory activity equivalent to that of the same molecule expressed in a mammalian system.

Membrane-targeted recombinant Crry was generated from the bacterially expressed protein by coupling a membrane-targeting peptide (APT542) to the engineered cysteine residue at the carboxyl terminus by standard thiol-interchange chemistry. The complement inhibitory activity of the soluble recombinant form of CR1 consisting of SCRs 1–3, when modified with this peptide (termed APT070), increased over 100-fold compared with the unmodified construct (Dodd et al. 2000). We showed a similar increase in activity for tagged Crry in nonwash assays, and the activity was retained in wash assays, showing that soluble Crry modified with the APT542 tag was able to incorporate into cell membranes. This novel agent, Crry-APT542, can now be used in rat models of human disease without the problems of immunogenicity that have accompanied use of agents based on human CR1 and with the expectation of strong anticomplement activity and prolonged half-life in the animal. Results from such studies will provide a meaningful assessment of the likely benefit of targeted human complement regulators in target diseases. Our priorities include arthritis and demyelination, both situations in which complement inhibitors have been used previously to good effect within the limitations of the agents. The capacity to treat over extended periods opens many more doors and enables testing in chronic disease models more relevant to the human condition.

Materials and methods

Chemicals and reagents

Chemicals and reagents were obtained from Fisher Scientific or Sigma unless otherwise stated. pDR2ΔEF1α was a gift from Dr I. Anegon (INSERM U437); pET26b and E. coli BL21(DE3) were obtained from Novagen. Restriction enzymes, T4 DNA ligase, lipofectamine, and DNA MW markers were purchased from Life Technologies. Oligonucleotide primers were synthesized in-house using an Applied Biosystems 392 RNA/DNA synthesizer, and DNA sequencing was carried out on an Applied Biosystems 373A DNA sequencer, utilizing the ABI Prism dye-terminator kit. All tissue culture reagents and plastics were from Life Technologies. PBS is 8.1 mM Na2PO4, 1.5 mM KH2PO4, 137 mM NaCl, 2.7 mM KCl (pH 7.4). BBS is 100 mM H3BO3, 25 mM Na2B4O7.10H2O, 75 mM NaCl (pH 8.4). CFD is 2.8 mM barbituric acid, 145.5 mM NaCl, 0.8 mM MgCl2, 0.8 mM CaCl2, 0.9 mM sodium barbital (pH 7.2) (Oxoid Ltd.).

Sera, antibodies, and other proteins

Normal human serum (NHS) was obtained by venipuncture from healthy volunteers. Normal rat serum (NRS) was obtained from the local animal facility. Sheep erythrocytes in Alsever's solution were from TCS Microbiology. Primary antibodies were as follows: monoclonal mouse anti-rat Crry, TLD1C11, was kindly provided by Professor W. Hickey, Dartmouth, New Hampshire, USA. Polyclonal rabbit anti-human C3c antiserum was provided by Dr. M. Fontaine (Rouen, France). Secondary antibodies were as follows: goat anti-mouse Ig conjugated to HRPO was obtained from BioRad and goat anti-rabbit immunoglobulin conjugated to HRPO was obtained from Jackson Laboratories. Rabbit anti-sheep erythrocyte antibody (Amboceptor) was from Behring Diagnostics GmbH. APT070 for haemolytic assay controls was provided by Adprotech Ltd. For cofactor assays, recombinant soluble human CR1 (sCR1) was provided by T Cell Sciences Inc. Human C3 was made in house by standard methods. Rat factor I was prepared and purified from rat serum as described previously (Farries et al. 1990).

Construction of soluble recombinant rat Crry cDNA mammalian expression vectors

cDNA encoding the four amino-terminal SCRs of rat Crry (sCrry) was provided in the expression vector pDR2ΔEF1α by Dr Neil Rushmere (UWCM, Cardiff, UK). This was used as the cDNA template from which sCrry-Cys cDNA was amplified by PCR using the following primers: (1) 5`-GCGTCTAGAACCGCTTT GCCGTGAGACTGG-3`; (2) 5`-TTACCCGGGTTAACATGAT CCTCCTCCTCCTGTCCGGATTTCACCTTGAAGGAGC-3`. These primers introduced restriction sites for XbaI and SmaI into the cDNA sequence (shown in bold) for subsequent cloning into pDR2ΔEF1α. Primer (2) also encoded for a 7 amino-acid linker sequence and carboxy-terminal cysteine residue (sequence: GSGGGSGC). A two-step PCR was used to generate sCrry-Cys(N|ZeG) cDNA, which was identical to sCrry-Cys except for two adjacent point mutations that encoded for a glycine substitution at the putative N-glycosylation site, N247. Step 1 of the two-step PCR was carried out as follows: Primers 1 (above) and 3, 5`-ACAAGCTTTTGCCTTTAGAC were used to amplify a 805-bp fragment of DNA (fragment 1) encoding most of the first four SCRs of Crry, but with A |Ze G base mutations at positions 631 and 632 (shown in bold) within the Crry gene (Quigg et al. 1995). Primers 2 (above) and 4, 5`-TAGTGTCTAAAGGCAAAAGC were used to amplify a 184-bp fragment of DNA (fragment 2) also encoding for the point mutations, the remainder of SCR 4, a carboxy-terminal 7 amino-acid spacer region (as before) and carboxyl terminus cysteine residue. Fragments 1 and 2 were visualized by agarose gel electrophoresis, and gel purified. In Step 2, fragments 1 and 2 were used as template DNA in a PCR. By binding each other via a 20-bp overlapping sequence, they were also used to self-prime the reaction to generate a full-length sCrry-Cys(N|ZeG) cDNA template. This template was then used in a PCR using primers 1 and 2 to amplify the full sequence of sCrry-Cys(N|ZeG) cDNA. The amplified cDNAs were subcloned into the pGEM-T easy vector (Promega), from which they were digested with XbaI and SmaI, purified using the Qiaquick PCR purification kit (QIAGEN), and ligated into the corresponding sites of pDR2ΔEF1α. Plasmids containing the sCrry-Cys or sCrry-Cys(N|ZeG) cDNA were PCR screened using primers 1 and 2 as described above, isolated, and the fidelity of DNA was confirmed by automated sequencing.

Construction of soluble recombinant rat Crry cDNA bacterial expression vector

sCrry-Cys cDNA was amplified by PCR from a plasmid containing sCrry cDNA using the following primers: (5) 5`-GCCATC TACTCATATGCAGTGCCCAGC-3`, which added a NdeI restriction site (bold) immediately upstream of the nucleotides encoding the initiating methionine of the recombinant protein and primer 2 (above). The PCR product was subcloned into the plasmid pGEM-T Easy as before, digested with NdeI and NcoI, purified, and ligated into the bacterial expression vector pET26b digested at the corresponding sites. Sequencing of the insert confirmed that no errors had been introduced by PCR.

Expression and purification of soluble Crry from a mammalian expression system

Chinese hamster ovary (CHO) cells were transfected with pDR2ΔEF1α expression plasmids containing sCrry, sCrry-Cys, and sCrry-Cys(N|ZeG) cDNA using Lipofectamine (Life Technologies) according to the manufacturer's instructions. Medium from stably transfected CHO cells was tested to confirm the presence of soluble recombinant Crry by Western analysis. Proteins were purified from the supernatant by passage over affinity columns prepared by coupling the monoclonal anti-Crry antibody TLD1C11 to CNBr-activated Sepharose 4B (Amersham Pharmacia) according to the manufacturer's instructions. The column was washed with PBS and protein eluted with 50 mM Diethylamine (pH 11.5). Eluted protein was neutralized with Tris, concentrated by ultrafiltration using a 10-kD molecular weight cut-off membrane, and dialyzed into PBS.

Bacterial expression of sCrry-Cys by fermentation

High-bacterial culture densities were obtained by fermentation of E. coli in a Bioflo 3000 Bioreactor (New Brunswick Scientific) with a 2-L bioreactor culture vessel. The fermenter was prepared according to the manufacturer's instructions. Briefly, the temperature was set at 37°C for the duration of the fermentation, and the pH was monitored. Maximum values for agitation and airflow were set to 750 rpm and 2 L/min, respectively. The vessel, containing 2 L of sterile NZCYM medium (Sigma) supplemented with kanamycin (50 μg/mL) and 0.1% polyethylene glycol (anti-foaming agent), was then inoculated with 20 mL of an overnight culture from a single colony of chemically competent E. coli BL21(DE3) transformed with the bacterial expression vector pET26b containing sCrry-Cys cDNA. The Bioflo 3000 was set to monitor the amount of dissolved O2 within the culture and automatically adjust agitation and airflow within the set maximum values to maintain a dissolved O2 content of 50% for maximum bacterial growth density. Cultures were grown for 4 h until the bacteria were in their log phase of growth (OD600 = 5–8). Protein expression was then induced by adding sterile filtered isopropyl β D thiogalactopyranoside (IPTG) to a final concentration of 1 mM. The fermentation culture was harvested at 3 h post induction, and centrifuged at 10,000g for 10 min. Cell pellets were stored at −40°C.

Isolation, solubilization, and refolding of sCrry-Cys from bacterial inclusion bodies

A frozen cell pellet from 2 L of culture was thawed at 37°C in a shaker-incubator and washed twice in 5 mL of 50 mM Tris, 1 mM EDTA, and 50 mM NaCl (pH 8.0) per gram of pellet. The pellet was then resuspended in 5 mL of ice-cold wash buffer per gram pellet and blended using an electric blender until only fine particles remained. The cells were then disrupted by two passes at 12,000 Psi through an Emulsiflex C5 High Pressure Homogenizer (Glen Creston) at 4°C. The homogenized pellet was immediately centrifuged at 10,000g for 10 min, and the supernatant discarded. The pellet was washed three times in ice-cold wash buffer. Inclusion bodies were solubilized by resuspension of the pellet to 5 mg/mL in equilibration buffer (8 M urea, 20 mM Tris, 1 mM EDTA, and 50 mM 2-mercaptoethanol at pH 8.5) and refolded using a method based on that of Dodd et al. (1995). Solubilized sCrry-Cys inclusion bodies were refolded by rapid dilution 1/40 into cold 0.02 M ethanolamine, 1 mM EDTA (pH 11.0), and left static for 24 h at 4°C. The solution was then concentrated by ultrafiltration to a final volume of 750 mL and buffer exchanged into PBS. Soluble recombinant Crry was purified by anion exchange. Protein was loaded onto a Source 15Q column (Amersham Pharmacia) in 20 mM Tris (pH 8.9), and eluted with a linear gradient to 1 M NaCl, 20 mM Tris (pH 8.9), over 30 column volumes. Protein was run on SDS-PAGE to confirm purity and analyzed by MALDI-Tof (Bruker Relex 3 spectrometer; Bruker UK Ltd.) to confirm molecular mass.

Modification of sCrry-Cys with a membrane address tag

sCrry-Cys was modified by derivatization of the free carboxy-terminal cysteine with the sulfydryl-reactive peptide, N-(myristoyl)GSSKSPSKKKKKKPGDC-(S-2-thiopyridyl) C-amide (termed APT542, also known as MSWP-1) (Adprotech Ltd.) (Smith et al. 1998a,b) to produce sCrry-APT542, which was purified to homogeneity by hydrophobic-interaction chromatography. Protein was loaded onto a Macroprep-Methyl (Bio-Rad) column in 3 M ammonium sulfate, 1 M NaCl, 0.1 M sodium phosphate (pH 6.0), and eluted with a step gradient in 0.2 M NaCl decrements, to 3 M ammonium sulfate, 0.1 M sodium phosphate (pH 6.0). Fractions containing sCrry-APT542 were identified by SDS-PAGE.

SDS-PAGE and immunoblotting

Samples of soluble recombinant Crry were resolved by SDS-PAGE using 4%–12% Bis-Tris gradient gels (Novex) according to the manufacturer's instructions. Protein bands were visualized by staining with Coomassie Blue R-250. For Western blotting, proteins were transferred to nitrocellulose membrane (Schleicher and Schuell, Dassel, Germany). The membrane was blocked with 5% (w/v) nonfat dried milk and incubated with primary antibody overnight. Membranes were washed with PBS/0.1% Tween 20 and incubated with secondary antibody conjugated to HRPO in PBS/5% (w/v) nonfat dried milk for 2 h. Membranes were washed three times in PBS/Tween and three times in PBS before visualization of the bands using ECL (Perbio Science) and autoradiographic film (Kodak).

Haemolytic assays

Functional analysis of soluble recombinant Crry was carried out by haemolytic assay as described previously (Linton et al. 2000). Values of the lysis obtained in the presence of an inhibitor were determined as a percent of the lysis seen in the presence of the inactive control protein, BSA, with each assay performed in triplicate. A mean value of percent lysis was then plotted against the concentration of inhibitor used, and the standard deviation of each data point calculated.

Cofactor assays

Methylamine-inactivated human C3 (C3ma) was prepared by incubating C3 (0.5 mg/mL) with 0.1 M methylamine for 2 h at 37°C in BBS (pH 8.0). C3ma (50 μg/mL) was incubated with rat factor I (10 μg/mL) and cofactor/test sample (40 μg/mL) for 16 h at 37°C. Samples were separated by SDS-PAGE and either silver stained or Western blotted using a polyclonal anti-C3c antiserum (1/1000) to probe and goat anti-rabbit immunoglobulin conjugated to HRPO (1/2000) to detect. Silver-stained gels were scanned, and the absolute integrated optical density (IOD) for each band was determined using LabWorks 4.0 gel imaging and analysis software (UVP Inc.).

Acknowledgments

This work was supported by the BBSRC, The Wellcome Trust (ref. no. 016668), and Adprotech Ltd. We thank Jeremy Bright, Wendy Bruce-Johnson, and Sean Gallagher (Adprotech Ltd.) for helpful advice, discussion and assistance.

The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.

Article and publication are at http://www.proteinscience.org/cgi/doi/10.1110/ps.0212402.

References

  1. Campbell, R.D., Law, S.K., Reid, K.B., and Sim, R.B. 1988. Structure, organization, and regulation of the complement genes. Annu. Rev. Immunol. 6 161–195. [DOI] [PubMed] [Google Scholar]
  2. Chavez-Cartaya, R.E., DeSola, G.P., Wright, L., Jamieson, N.V., and White, D.J. 1995. Regulation of the complement cascade by soluble complement receptor type 1. Protective effect in experimental liver ischemia and reperfusion. Transplantation 59 1047–1052. [DOI] [PubMed] [Google Scholar]
  3. Couser, W.G., Johnson, R.J., Young, B.A., Yeh, C.G., Toth, C.A., and Rudolph, A.R. 1995. The effects of soluble recombinant complement receptor 1 on complement-mediated experimental glomerulonephritis. J. Am. Soc. Nephrol. 5 1888–1894. [DOI] [PubMed] [Google Scholar]
  4. De Bernardez Clark, E. 1998. Refolding of recombinant proteins. Curr. Opin. Biotech. 9 157–163. [DOI] [PubMed] [Google Scholar]
  5. Dodd, I., Mossakowska, D.E., Camilleri, P., Haran, M., Hensley, P., Lawlor, E.J., McBay, D.L., Pindar, W., and Smith, R.A. 1995. Overexpression in Escherichia coli, folding, purification, and characterization of the first three short consensus repeat modules of human complement receptor type 1. Protein Expr. Purif. 6 727–736. [DOI] [PubMed] [Google Scholar]
  6. Dodd, I., Oldroyd, R., Powell, S., Affleck, L., Bamber, L., Gallagher, S., Rowling, P., Ragnauth, C., Smith, G., Pratt, J.R., et al. 2000. Develpment of a membrane-targeted complement inhibitor for clinical use. Immunopharmacology 49 63. [Google Scholar]
  7. Dong, J., Pratt, J.R., Smith, R.A., Dodd, I., and Sacks, S.H. 1999. Strategies for targeting complement inhibitors in ischaemia/reperfusion injury. Mol. Immunol. 36 957–963. [DOI] [PubMed] [Google Scholar]
  8. Durbin, R. 1998. Gene Expression Systems. Academic Press, Carlsbad, CA.
  9. Farries, T.C., Seya, T., Harrison, R.A., and Atkinson, J.P. 1990. Competition for binding sites on C3b by CR1, CR2, MCP, factor B and Factor H. Complement Inflammat. 7 30–41. [DOI] [PubMed] [Google Scholar]
  10. He, C., Alexander, J.J., Lim, A., and Quigg, R.J. 1997. Production of the rat complement regulator, Crry, as an active soluble protein in Pichia pastoris. Arch. Biochem. Biophys. 341 347–352. [DOI] [PubMed] [Google Scholar]
  11. Hill, J., Lindsay, T.F., Ortiz, F., Yeh, C.G., Hechtman, H.B., and Moore, Jr., F.D. 1992. Soluble complement receptor type 1 ameliorates the local and remote organ injury after intestinal ischemia-reperfusion in the rat. J. Immunol. 149 1723–1728. [PubMed] [Google Scholar]
  12. Kim, Y.U., Kinoshita, T., Molina, H., Hourcade, D., Seya, T., Wagner, L.M., and Holers, V.M. 1995. Mouse complement regulatory protein Crry/p65 uses the specific mechanisms of both human decay-accelerating factor and membrane cofactor protein. J. Exp. Med. 181 151–159. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Lasky, L.A. 1992. Selectins: Interpreters of cell-specific carbohydrate information during inflammation. Science 258 964–969. [DOI] [PubMed] [Google Scholar]
  14. Lilie, H., Schwarz, E., and Rudolph, R. 1998. Advances in refolding of proteins produced in E. coli. Curr. Opin. Biotechnol. 9 497–501. [DOI] [PubMed] [Google Scholar]
  15. Lindsay, T.F., Hill, J., Ortiz, F., Rudolph, A., Valeri, C.R., Hechtman, H.B., and Moore, Jr., F.D. 1992. Blockade of complement activation prevents local and pulmonary albumin leak after lower torso ischemia-reperfusion. Ann. Surg. 216 677–683. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Linton, S.M., Williams, A.S., Dodd, I., Smith, R., Williams, B.D., and Morgan, B.P. 2000. Therapeutic efficacy of a novel membrane-targeted complement regulator in antigen-induced arthritis in the rat. Arthritis Rheum. 43 2590–2597. [DOI] [PubMed] [Google Scholar]
  17. McGrath, Y., Wilkinson, G.W., Spiller, O.B., and Morgan, B.P. 1999. Development of adenovirus vectors encoding rat complement regulators for use in therapy in rodent models of inflammatory diseases. J. Immunol. 163 6834–6840. [PubMed] [Google Scholar]
  18. Piddlesden, S.J., Storch, M.K., Hibbs, M., Freeman, A.M., Lassmann, H., and Morgan, B.P. 1994. Soluble recombinant complement receptor 1 inhibits inflammation and demyelination in antibody-mediated demyelinating experimental allergic encephalomyelitis. J. Immunol. 152 5477–5484. [PubMed] [Google Scholar]
  19. Piddlesden, S.J., Jiang, S., Levin, J.L., Vincent, A., and Morgan, B.P. 1996. Soluble complement receptor 1 (sCR1) protects against experimental autoimmune myasthenia gravis. J. Neuroimmunol. 71 173–177. [DOI] [PubMed] [Google Scholar]
  20. Quigg, R.J., Lo, C.F., Alexander, J.J., Sneed, 3rd, A.E., and Moxley, G. 1995. Molecular characterization of rat Crry: Widespread distribution of two alternative forms of Crry mRNA. Immunogenetics 42 362–367. [DOI] [PubMed] [Google Scholar]
  21. Quigg, R.J., He, C., Hack, B.K., Alexander, J.J., and Morgan, B.P. 2000. Production and functional analysis of rat CD59 and chimeric CD59-Crry as active soluble proteins in Pichia pastoris. Immunology 99 46–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Rittershaus, C.W., Thomas, L.J., Miller, D.P., Picard, M.D., Geoghegan-Barek, K.M., Scesney, S.M., Henry, L.D., Sen, A.C., Bertino, A.M., Hannig, G., et al. 1999. Recombinant glycoproteins that inhibit complement activation and also bind the selectin adhesion molecules. J. Biol. Chem. 274 11237–11244. [DOI] [PubMed] [Google Scholar]
  23. Smith, G.P. and Smith, R.A.G. 2001. Membrane-targeted complement inhibitors. Mol. Imuunol. 38 249–255. [DOI] [PubMed] [Google Scholar]
  24. Smith, R.A.G., Dodd, I., and Mossakowska, D.E.I. 1998a. International Patent Publication, WO 98/02454
  25. Smith, R.A.G., Dodd, I., Rowling, P., Cox, V., Mossakowska, D.E.I., Oldroyd, R., and Lachmann, P. 1998b. Cell surface engineering using a complement regulatory molecule modified with a synthetic myristoyl-electrostatic switch derivative. Mol. Immunol. 35 400. [Google Scholar]
  26. Smith, R.A.G., Dodd, I., Oldroyd, R.G., Harry, J., Clarke, C., Rolan, P., and Dawes, L. 2001. Preclinical and clinical progression of a membrane targeted complement regulator therapeutic. Mol. Immunol. 38 122. [Google Scholar]
  27. Springer, T.A. 1990. Adhesion receptors of the immune system. Nature 346 425–434. [DOI] [PubMed] [Google Scholar]
  28. Weisman, H.F., Bartow, T., Leppo, M.K., Boyle, M.P., Marsh, Jr., H.C., Carson, G.R., Roux, K.H., Weisfeldt, M.L., and Fearon, D.T. 1990. Recombinant soluble CR1 suppressed complement activation, inflammation, and necrosis associated with reperfusion of ischemic myocardium. T. Assoc. Am. Physician. 103 64–72. [PubMed] [Google Scholar]
  29. Whaley, K., Loos, M., and Weiler, J.M. 1993. Complement in health and disease. Kluwer Academic Publishers, Dordrecht, The Netherlands.
  30. Wong, W.W., Cahill, J.M., Rosen, M.D., Kennedy, C.A., Bonaccio, E.T., Morris, M.J., Wilson, J.G., Klickstein, L.B., and Fearon, D.T. 1989. Structure of the human CR1 gene. Molecular basis of the structural and quantitative polymorphisms and identification of a new CR1-like allele. J. Exp. Med. 169 847–863. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Zhang, H.F., Yu, J., Bajwa, E., Morrison, S.L., and Tomlinson, S. 1999. Targeting of functional antibody-CD59 fusion proteins to a cell surface. J. Clin. Invest. 103 55–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Zhang, H., Lu, S., Morrison, S.L., and Tomlinson, S. 2001. Targeting of functional antibody-decay-accelerating factor fusion proteins to a cell surface. J. Biol. Chem. 276 27290–27295. [DOI] [PubMed] [Google Scholar]

Articles from Protein Science : A Publication of the Protein Society are provided here courtesy of The Protein Society

RESOURCES