|Home | About | Journals | Submit | Contact Us | Français|
Neutralizing antibodies are a critical component in the protection or recovery from viral infections. In the absence of available vaccines or antiviral drugs for many important human viral pathogens, the identification and characterization of new human monoclonal antibodies (hmAbs) able to neutralize viruses offers the possibility for effective pre- and/or post-exposure therapeutic modalities. Such hmAbs may also help in our understanding of the virus entry process, the mechanisms of virus neutralization and in the eventual development of specific entry inhibitors, vaccines and research tools. The majority of the more recently developed antiviral hmAbs have come from the use of antibody phage-display technologies using both naïve and immune libraries. Many of these agents are also enveloped viruses possessing important neutralizing determinants within their membrane-anchored envelope glycoproteins and the use of recombinant, soluble versions of these viral glycoproteins is often critical in the isolation and development of antiviral hmAbs. This chapter will detail several methods that have been successfully employed to produce, purify and characterize soluble and secreted versions of several viral envelope glycoproteins which have been successfully used as antigens to capture and isolate human phage-displayed monoclonal antibodies.
For the vast majority of viral pathogens there is a paucity of drug-based therapies. Rather, it has been the development of vaccines that has been the mainstay of prevention and intervention strategies for combating human and animal viral diseases. There are presently 15 viral vaccines approved for human use, excluding various subtypes, and the majority of these are live-attenuated formulations (reviewed in (1)). In general terms, these live-attenuated viral vaccines are highly effective because they elicit a balanced immune response in the recipient host; stimulating both cell-mediated and humoral immunity. However, for many viruses especially those that have associated highly pathogenic characteristics as with Biological Safety Level-4 (BSL-4) restricted agents or retroviruses such as human immunodeficiency virus type 1 (HIV-1), the use of live attenuated vaccines are not feasible.
A number of studies have demonstrated the importance of neutralizing antibodies in the protection or recovery from viral infections (2, 3). Indeed, as obligate intracellular parasites, viruses pose significant challenges for the development of effective antiviral therapeutics. Neutralizing polyclonal antibodies have a long history of being effective against some viruses and more recently, monoclonal antibodies (mAbs) have also shown success. The humanized mAb Synagis (palivizumab), which is currently the only mAb against a viral disease approved by the U.S. Food and Drug Administration (FDA), has been widely used as a prophylactic measure against respiratory syncytial virus (RSV) infections in neonates and immune-compromised individuals and is more cost-effective and efficacious than the original polyclonal product (4). Most recently, the anti-RSV palivizumab has been improved, and motavizumab has been shown to potently inhibit viral replication in the upper respiratory tract in a cotton rat model (5). Virus-neutralizing antibodies can also be administered passively to acutely infected individuals and be highly efficacious. The mechanism of passively administered antibody therapy can be viewed as that of an antiviral drug; suppressing infection and permitting the host to mount an effective immune response (6). Today, passively administered antibody is routinely used as an effective antiviral therapy or prophylactic for hepatitis B, varicella-zoster, rabies virus, measles virus, and others (reviewed in (2)). In most cases their use is a first-line therapy as a post-exposure measure or in circumstances where vaccination is not possible. However, serum polyclonal antibody preparations have associated problems related to toxicity and potential allergic reactions, as well as lot to lot variation and uncertain dosing regimes (7).
The major advances in furthering the development of specific mAbs, has been through the use of bacterial phage display platforms with combinatorial antibody libraries (8, 9). Further, these phage libraries can be prepared to encode human antibodies as Fabs which contain the light chain and the first two domains of the heavy chain or single-chain variable domain fragments (scFvs) containing the variable domains of the light and heavy chains, and this technology has been complemented by innovative affinity maturation strategies to improve antibody binding (reviewed in (10)). These techniques in human phage-display antibody platforms have facilitated the rapid identification and isolation of specific human mAbs, eliminating the immunization, hybridoma development, and humanization processes. In the absence of available vaccines or antiviral drugs, the identification and characterization of new human monoclonal antibodies (hmAbs) able to neutralize viruses offers the possibility for effective pre- and post-exposure therapeutic modalities. Such antibodies may also help in our understanding of the virus entry process and its underlying mechanisms, the viral neutralization mechanisms and in the eventual development of specific entry inhibitors, vaccines and research tools. There have been many recent examples of the development and isolation of hmAbs using phage-display methodologies reactive against important human viral pathogens including HIV-1 (11-16), the paramyxoviruses, Hendra virus (HeV) and Nipah virus (NiV) (17), and the human SARS coronavirus (18).
Many of these viral pathogens are also enveloped viruses, and it is almost without exception that all neutralizing antibodies to enveloped viruses are directed against the virus’ envelope glycoproteins and traditionally the antibody response has been the immunologic measure of vaccine efficacy (19). All known viral envelope glycoproteins are homo- or heterooligomers in their mature and functional forms (20) and multimeric proteins, like these, generally interact over large areas which often translate into important structural differences between monomeric subunits and the mature oligomer. This feature can also impart significant differences in antigenic structure which has been shown for a number of proteins such as the trimeric influenza HA glycoprotein (21) and HIV-1 gp120/gp41 (22). In addition, some viruses pose significant additional challenges such as antigenic variation of their structural proteins that are important neutralization determinants. Perhaps the best characterized example of this particular problem is with primary HIV-1 isolates that exist across the many varied HIV-1 subtypes (23). In efforts to circumvent this issue, further improvements and enhancements in the techniques of phage-displayed antibody library panning have been developed in order to better select for broadly reactive mAbs or for mAbs reactive to particular subunits of a multi-subunit viral glycoprotein such as sequential antigen panning (SAP) and competitive (CAP) antigen panning methodologies (14, 15).
It is often critical that the antigens used for the panning and isolation of hmAbs from phage libraries be produced and/or purified using methods whereby they retain a near native structure and conformation, such as an oligomeric configuration. A useful approach to develop viral membrane glycoproteins suitable for panning phage libraries or as antigens for eliciting antibody responses that recognize their native form is to engineer soluble and secreted versions of the molecules. Often, this approach yields a quaternary structure similar to their native counterparts and for animal viruses, eukaryotic expression systems are typically employed such as recombinant bacculovirus or vaccinia virus, or transient or stable expression in cell culture (22, 24-36). This chapter will detail several methods that have been successfully employed to produce, purify and characterize soluble and secreted versions of several viral envelope glycoproteins which have been successfully used as antigens to capture and isolate human phage-displayed monoclonal antibodies.
Filter sterilization is recommended for all buffers.
All buffers should be sterile filtered and degassed.
The methods described below outline the steps for recombinant viral glycoprotein production using vaccinia virus for HIV-1 gp140 and HeV and NiV soluble G glycoprotein (sG) and affinity purification of the glycoproteins. In addition, recombinant viral glycoprotein production and affinity purification from stable HeV and NiV soluble F glycoprotein (sF) secreting 293T cell lines will be detailed. Additional purification methods are also detailed, including size exclusion gel filtration and analysis and separation of oligomeric species of the purified viral glycoproteins; sucrose gradient ultracentrifugation analysis; and Blue Native Polyacrylamide Gel Electrophoresis (BN-PAGE) analysis.
The design and construction of recombinant vaccinia viruses encoding HIV-1 gp140 and HeV and NiV sG as well as the characterization of the recombinant, soluble and secreted viral glycoproteins have been detailed elsewhere(24, 40). In this section for the purposes of additional illustration there will be an emphasis on the production and purification of soluble and secreted forms of the HeV and NiV F glycoprotein, including the analysis of oligomeric characteristics as well as the generation of stable cell lines secreting recombinant soluble F (sF).
This is the first step of purification. Depending on the protein’s nature or the tag engineered to the protein, different approaches can be carried out for affinity purification. Here, we describe the use of (1) Lentil lectin purification of the highly glycosylated HIV-1 gp140 and (2) S-protein agarose purification for the N-terminally S-peptide tagged HeV/ NiV sG.
All the buffers should be degassed and sterile filtered.
Several considerations may be required while constructing the expression plasmid e.g. Kozak translation initiation sequence and an ATG start codon for proper initiation of translation, promoter/enhancer and termination sequences, the expression plasmid employed, the type of antibiotic selection, the cell line employed, codon optimization of the viral RNA sequence, mutation and modifications on the amino acid sequence to enhance expression and secretion, the tag required in facilitating protein purification, etc. The sF construct described here has been optimized based on all the above concerns. Figure 1 diagrams the construct of our sF GCN.
The vector employed was a promoter modified pcDNA 3.1 Hygro (+). The enhanced CMV promoter was imported from phCMV 1 vector that allows high expression level of the human and mouse codon optimized sF. The Hygromycin selection marker allows the selection of transfected 293T cell which is resistant to the commonly used Geneticin antibiotic. Replacing the C-terminal transmembrane and cytoplasmic tail domain of F is the trimeric GCN4 motif (41, 42) that allows stabilization of the protein trimeric structure (26, 43) and hence enhances the expression and secretion of the protein. This is followed by the 15 amino acid S-peptide that facilitates purification and immuno-detection of the sF by S-protein agarose and anti S-peptide antibody. A factor Xa cleavage site (IEGR) is also engineered upstream to the S-peptide tag allowing removal of the tag by enzymatic digestion. Standard cloning procedures were employed to construct the plasmid.
Before generating a stable cell line, several optimizations may be required to obtain the best expression level. This can be investigated in transient expression using various cell lines to obtain the best result. We describe here the transient transfection of the sF construct in 293T cells using lipid-based transfection procedure. The expression level can be analyzed using immuno or affinity precipitations from the cell lysate and supernatant follow by SDS-PAGE and Western blotting to access the secretion level. Figure 2 shows a representative result of transient transfection of various sF construct to assess the secretion level by western blotting. The addition of GCN tail and codon optimization by Geneart, Inc. greatly enhanced expression and secretion of the sF.
Once the best expression construct has been optimized with the best cell line, a stable cell line can be generated. In order to select a stable cell clone expressing the soluble glycoprotein, the minimum concentration of the appropriate selection antibiotic required to kill the non transfected host cell line needs to be determined. Because natural resistance varies among cell lines, we recommend testing a range of antibiotic concentrations on 25% confluent cells. Choose the concentration that prevents growth within 2-3 days and kills all cells within 5-7 days. The procedures described here have been optimized for 293T cells.
Once a stable cell line has been established, it can then be grown into roller bottles for larger scale protein production. The following protocol describes the large scale production of secreted sF from 293T cells in roller bottles. For protein expression, addition of selection antibiotic is not necessary. Because of the weak adherent nature of 293 derivative cells, gelatin treatment of the roller bottle surface is necessary to avoid the formation of cell clumps which will decrease the protein secretion level.
This step is essentially similar to section 22.214.171.124 except replacing all buffers to the list in section 2.5.3 and omitting the PBS wash in step 5. The protein concentration and yield can then be estimated. We use the Bradford method (44) to estimate the protein concentration at 595 nm OD. At this point, a small portion of the purified protein can be analyzed on SDS-PAGE follow by coomassie staining (Figure 3). Proceed to section 3.3 for molecular weight purification.
Following S-affinity purification, the concentrated purified protein can be analyzed on SDS-PAGE followed by coomassie staining to assess the purity and yield. In most cases, there will be presence of small amount of contaminating proteins especially when purification was done from serum medium. Hence, further purification is needed.
Gel filtration chromatography separates molecules on the basis of size. It is often reserved for the final step of purification to achieved higher purity. This method can also be used to estimate the molecular masses of globular proteins in their native condition base on their migration through a chromatographic matrix packed in a column. Pre-packed columns with various matrix of different fractionation range are available commercially from GE healthcare.
The approximate molecular mass of a certain protein can be estimated from a calibrated curve obtained from a series of standards protein with known molecular masses. Standard calibration protein kits can be obtained commercially from e.g. GE healthcare, Sigma, etc with well described calibrating protocols. Once the molecular weight of the different oligomeric species has been determined, a preparative chromatography can be carried out to isolate the desired oligomeric species.
Here, we describe the use of a calibrated Superdex 200 10/300 GL gel filtration column to estimate the approximate molecular mass of the different oligomeric species of the soluble viral glycoprotein using example of HeV and NiV sF GCN. We also describe the use of HiLoad 16/60 Superdex 200 prep grade gel filtration column XK 16 to separate the different oligomeric species of affinity purified sF.
Before analyzing the soluble viral glycoprotein, calibrate the column with a series of protein standards. The Superdex 200 10/300 GL gel filtration column employed here has been calibrated with the Molecular weights (29 KDa -669 KDa) calibration kit. As depicted in Figure 4, the calibration curve generated was used to estimate the approximate size of the HeV sF GCN. Well described calibrating protocols are included in commercially available gel filtration molecular weights calibration kit. Below is a simple procedure used to determine the approximate size of the sF.
Native oligomeric forms of the purified soluble glycoprotein can also be analyzed by sucrose gradient centrifugation. As shown in Figure 6, two species (large aggregate and trimer) of HeV sF GCN were observed in the sucrose gradient. This is consistent with the gel filtration profile as shown in Figure 4.
Native gel analysis allows the visualization of the soluble glycoprotein on polyacrylamide gel electrophoresis in its native form. This allows the estimation of the size of the native oligomeric species. It is most helpful in analyzing the fractions separated from gel filtration and sucrose gradient.
The Blue native (BN) gel system is based on the originally described system by Schagger and von Jagow (45). Many modifications have been made based on the original method for example in observing the native forms of HIV-1 Env (30, 46, 47). We find that the commercially available BN-PAGE system from Invitrogen Corp. is most suitable for our soluble glycoprotein analysis. It requires little or no optimization and is included with well described step by step protocol. In this BN-PAGE system, the Coomassie ® G-250 binds to proteins and confers a negative charge while maintaining the proteins in their Native state. Below are the procedures we use to analyze our soluble glycoproteins using the BN-PAGE system followed by western blotting.
The G-250 in the Cathode buffer stains the proteins during electrophoresis. Therefore, further staining is not required. However, this depends on the sensitivity required. More sensitive staining protocol is available by the manufacturer.
This work was supported in part by Middle Atlantic Regional Center of Excellence (MARCE) for Biodefense and Emerging Infectious Disease Research, NIH AI057168 and AI054715 grants C.C.B.
1Tris is an inhibitor for DTSSP cross-linker. If DTSSP will be used with the protein in downstream analysis, HEPES can be used as an alternate neutralization buffer.
2Arginine prevents protein aggregation and is used in protein refolding. It also helps in elution of proteins and antibody. Addition of arginine is optional, but we find that it is necessary for sF purification.
3Cells should be very dense. Estimation of the following days of incubation for cell growth is based on each T-150 flask containing greater than 3.0 × 107 cells. To obtain the maximum yield of pure protein, the cells must be very dense when seeded into roller bottles and when infected with purified recombinant vaccinia virus.
4Ensure cell clumps are broken up by repeatedly pipetting the suspension up and down. An even cell suspension is necessary for an even monolayer in the roller bottles. Formation of cell clumps in the roller bottles will decrease the protein yield.
5A 10 mL pipette without a cotton stopper is connected to a tube that connects to a 10% CO2, 90% air tank with a pressure controller and indicator, a 0.2 μm filter membrane should be connected to the tubing.
6Pre-wet the filter unit membrane with PBS, 0.1% Triton® X-100 or the appropriate washing buffer before filtering the supernatant. This can be done by filtering 50 to 100 mL of wash buffer prior to sample filtering. We find that this procedure will decrease protein lost during filtration. This step is essential for purification of sF.
7Gelatin treatment is optional. However for 293 and derivatives cells, gelatin treated surface allows the cells to adhere better during the medium changing procedures of transfection and hence yields better transfection results. Coat 2 mL per well of sterile 0.1% Gelatin onto a 6-well tissue culture plate. Incubate the plate at 37°C for 1 hr or 4°C overnight. Rinse twice with sterile PBS before using.
8Harvest the DMEM medium and follow procedures in section 126.96.36.199. Use this medium for S-column purification as described in section 3.2.5. There will still be contaminating serum protein at this stage. Apply the neutralized elution to another round of S-column purification in smaller scale. Scale down all buffers for 2 mL bed volume of S-Agarose. Use 10 mL Poly-Prep Chromatography Columns (Bio-Rad Laboratories, Inc). The flow rate is entirely depending on gravity. Depending on the expected protein yield, several columns may be required. It is estimated that 1 mL bed volume of S-Agarose is able to capture at least 1 mg sF. We use 3 columns for sF GCN and more than 6 mg can be recovered from supernatant of 10 roller bottles.