Antigen Design & Sera Purification Tech Sheet
Antibodies to small peptides have become an essential tool in life science research, with applications including gene product detection and identification, protein processing studies, diagnostic tests, protein localization, active site determination, protein homology studies and protein purification. While it is quite easy to generate anti-peptide antibodies, it is important to carefully consider the ultimate use for the antibody and the sequence used to ensure success. This tech sheet will briefly explore peptide selection and design, coupling strategy, and carrier proteins which are important factors in anti-peptide antisera generation. Serum purification will also be discussed. For more complete coverage of antigen design, please refer to the References1,2.
Peptide Selection and Design
The first step in the process is the selection of the appropriate peptide sequence. At this step the ultimate use for the antibody must be considered. If the antibody is needed to probe a specific protein domain then the choice is simple. For example, if one is studying proteolytic processing of an N-terminal precursor, antibodies against the N-terminal region of interest would be raised. Likewise if the goal is to monitor the phosphorylation state of a specific sequence, antibodies to the phosphorylated sequence can be used.
If the goal is to raise antibodies that will recognize the protein in its native state, the problem becomes more complex. Anti-peptide antibodies will always recognize the peptide. However, the same antibody may not recognize the sequence within the folded intact protein. Sequence epitopes in proteins generally consist of 6-12 amino acids and can be classified as continuous and discontinuous. Continuous epitopes are composed of a contiguous sequence of amino acids in a protein. Anti-peptide antibodies will bind to these types of epitopes in the native protein provided the sequence is not buried in the interior of the protein. Discontinuous epitopes consist of a group of amino acids that are not contiguous but are brought together by folding of the peptide chain or by the juxtaposition of two separate polypeptide chains. Anti-peptide antibodies may or may not recognize this class of epitope depending on whether the peptide used for antisera generation has secondary structure similar to the epitope and/or if the protein epitope has enough continuous sequence for the antibody to bind with a lower affinity.
When examining a protein sequence for potential antigenic epitopes, it is important to choose sequences which are hydrophilic, surface-oriented, and flexible3. Most naturally occurring proteins in aqueous solutions have their hydrophilic residues on the surface and their hydrophobic residues buried in the interior. Antibodies bind to epitopes on the surface of proteins. Additionally, it has been shown that epitopes have a high degree of mobility4.
Because the C-termini of proteins are often exposed and have a high degree of flexibility they are usually a good choice for generating anti-peptide antibodies directed against the intact protein. If the protein is an integral membrane protein and the C-terminus is part of the transmembrane segment, this sequence will be too hydrophobic and not a good choice.
Like the C-terminus, the N-terminus is also frequently exposed and on the surface of the protein making it an ideal candidate for antibody generation. If a protein sequence is derived from the cDNA sequence, the leader sequence should not be included in the sequence selected for antibody generation.
Algorithms for predicting protein characteristics such as hydrophilicity/hydrophobicity and secondary structure regions such as alpha-helix, beta-sheet and beta-turn aid selection of a potentially exposed, immunogenic internal sequence for antibody generation.
Hydrophilicity plots as described by Hopp and Woods5 assign an average hydrophilicity value for each residue in the sequence. The highest point of average hydrophilicity for a series of contiguous residues is usually at or near an antigenic determinant. A slightly different algorithm described by Kyte and Doolittle6 evaluates the hydrophilic and hydrophobic tendencies of the sequence. This profile is useful for predicting exterior vs. interior regions of the native protein. Secondary structure can be identified by the use of algorithms developed by Chou and Fasman7 or Lim8. Surface regions or regions of high accessibility often border helical or extended secondary structure regions. In addition, sequence regions with beta-turn or amphipthic helix character have been found to be antigenic9.
Many commercial software packages such as MacVectorTM, DNAStarTM, and PC-GeneTM incorporate these algorithms. To be sucessful, none of the algorithms should be used alone. Combined use of the predictive methods may result in a success rate as high as 86% in predicting antigenic determinants9,10.
Once the protein region of interest has been identified, the length of the peptide must be selected. There are two differing thoughts on the topic of peptide length. One suggests that long peptides (20-40 amino acids in length) are optimal because it increases the number of possible epitopes. The other suggests that smaller peptides are sufficient, and their use ensures that the site-specific character of anti-peptide antibodies is retained. Clearly, any peptide selected must be chemically synthesizable and should be soluble in aqueous buffer for conjugation to the carrier protein. Peptides longer than 20 residues in length are often more difficult to synthesize with high purity because there is greater potential for side reactions, and they are likely to contain deletion sequences. On the other hand, short peptides (10 amino acids) may generate antibodies that are so specific in their recognition that they cannot recognize the native protein or do so with low affinity. The typical length for generating anti-peptide antibodies is in the range of 10-20 residues. Peptide sequences of this length minimize synthesis problems, are reasonably soluble in aqueous solution and may have some degree of secondary structure.
A factor that is often over-looked when designing a synthetic peptide is the method of coupling the peptide to the carrier protein. For example, N-terminal sequences should be coupled through the C-terminal amino acid and vice versa for C-terminal sequences. Internal sequences can be coupled at either end. Another consideration for internal sequences is to acetlyate or amidate the unconjugated end as the sequence in the native protein molecule would not contain a charged terminus.
The most common coupling methods rely on the presence of free amino (alph-amino or Lys), sufhydryl (Cys), or carboxylic acid groups (Asp, Glu, or alpha-carboxyl). Coupling methods should be used that link the peptide to the carrier protein via the carboxy- or amino-terminal residue. The sequence chosen should not have multiple residues that may react with the chosen chemistry. If multiple reactive sites are present, try to shorten the peptide or choose the sequence so they are all localized at either the amino or the carboxyl terminus of the peptide. For internal sequences the end furthest from the predicted epitope is normally favored as this avoids potential masking problems.
The EDC (1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride) or carbodiimide method is routinely used in the Sigma-Genosys laboratory unless otherwise stated by the researcher. Carbodiimides can activate the side chain carboxylic groups of aspartic and glutamic acid as well as the carbooxyl terminal group to make them reactive sites for coupling with primary amines. The activated peptides are mixed with the carrier protein to produce the final conjugate. If the carrier protein is activated first, the EDC method will couple the carrier protein through the N-terminal alpha amine and possibly through the amine in the side-chain of Lysine, if present in the sequence.
The m-Maleimidobenzoyl-N-hydroxysuccinimide ester (MBS) is a heterobifunctional reagent that can be used to link peptides to carrier proteins via cysteines. The coupling takes place with the thiol group of cysteine residues. If the chosen sequence does not contain Cys it is common to place a Cys residue at the N- or C-terminus to obtain highly controlled linking of the peptide to the carrier protein. For synthesis purposes we recommend that the placement of cysteine be at the N-terminus of the peptide if possible.
Glutaraldehyde is a bifunctional coupling reagent that links two compounds through their amino groups. Glutaraldehyde provides a highly flexible spacer between the peptide and carrier protein for favorable presentation to the immune system. Unfortunately, glutaraldehyde is a very reactive compound and will react with Cys, Tyr and His to a limited extent. The result is a poorly defined conjugate. The glutaraldehyde method is particularly useful when a peptide contains only a single free amino group at its amino terminus. If the peptide contains more than one free amino gorup, large multimeric complexes can be formed, which are not well defined, but are highly immunogenic.
Selecting the Protein Carrier
Conjugation to a carrier protein is important because peptides are small molecules, that alone do not tend to be immunogenic, thus possibly eliciting a weak immune response. The carrier protein contains many epitopes that stimulate T-helper cells, which help induce the B-cell response. Many different carrier proteins can be used for coupling to synthetic peptides. The most commonly selected carriers are keyhole limpet hemacyanin (KLH) and bovine serum albumin (BSA). The higher immunogenicity of KLH often makes it the preferred choice. Another advantage of choosing KLH over BSA is that BSA is used as a blocking agent in many experimental assays. Because antisera raised against peptides conjugated to BSA will also contain antibodies to BSA, false positives may result. Although KLH is large and immunogenic, it may precipitate during cross-linking, making it difficult to handle in some cases.
Ovalbumin (OVA) is another useful carrier protein. It is a good choice as a second carrier protein when verifying whether antibodies are specific for the peptide alone and not the carrier. Rabbit Serum Albumin (RSA) may be used when the antibody response to the carrier protein must be kept to a minimum. Rabbits immunized with RSA conjugate are less likely to raise antibodies to the carrier, as the RSA is recognized as "self." If the RSA conjugate were injected into another host, the protein would not be recognized as self.
It is important to recognize that the immune system reacts to the peptide-protein carrier as a whole and that there will be a portion of response directed against the conjugated peptide as well as the linker and the carrier protein1. When screening by ELISA it is advisable to use a peptide conjugate prepared using a different carrier protein. This is not necessary if performing ELISA assays where the plates are coated directly with unconjugated peptide.
Multiple Antigenic Peptides (MAPs)
The MAP system represents a unique approach to anti-peptide antibody generation11. The system is based on a small immunogenically inert branched lysine core onto which multipe peptides are synthesized in parallel. The result after synthesis is a three-dimensional molecule, which has a high molar ratio of peptide antigen to core molecule and therefore does not require the use of a carrier protein to induce an antibody response. Each core molecule may contain four identical peptides. In theory, MAP has an advantage when compared to its monomeric counterpart attached to a carrier protein in that the lysine core of a MAP is small compared with the peptide antigen. Therefore, the concentration of antigen is at a maximum. The result is a highly immunogenic MAP, which exhibits significantly higher titers when compared to its monomeric counterpart attached to a carrier protein.
It should be noted that there are some synthesis concerns when making a MAP complex. The branched nature of the lysine core allows for multiple copies of the peptide to be synthesized; however, steric hindrance becomes a problem during the synthesis of long peptides, resulting in some arms of the dendrimer being deletion products. The high molecular weight of the complex does not lend itself to good quality control measures (mass spec and/or analytical HPLC). An indirect synthesis of the MAP can eliminate analysis problems. In the indirect method, the peptide is first synthesized, purified then analyzed using mass spec and analytical HPLC. The peptide antigen is then coupled through a Cys to a functionalized lysine core.
Choice of Host
When attempting to raise an antibody, choose an animal that is genetically very different from the source of immunogen. In order to achieve maximum immune response, it is important to avoid self-recognition of the immunogen by the host animal. As an example, when raising antibodies against a human protein, it is more suitable to use a rabbit or mouse host than a monkey. For highly conserved mammalian proteins, raising antibodies in the avian (chicken) system is often a preferred alternative.
Adjuvant, Immunization, & Sera Collection
Sigma-Genosys routinely uses Freund's adjuvant for immunization purposes. The first injection is given in Complete Freund's adjuvant. Adjuvant is combined with the antigen to improve the immune response so that less vaccine is needed to produce more antibodies. The adjuvant allows a slow release of the antigen which allows for continual stimulation. Injections are routinely performed subcutaneously at multiple sites. A pre-immune bleed should be drawn from each host animal to produce a baseline to which the production bleeds can be compared. The drawn sera will contain a number of different types (IgG, IgM, IgA) and subclasses (Ig1, Ig2a, Ig2b, Ig3). Sodium azide (0.1%) can be added to the sera. Sodium azide is a broad-spectrum enzyme inhibitor and acts as an antimicrobial agent. Sodium azide should not be added to sera when using in cell culture or in vivo studies.
If a high background is observed in assays using the antisera, various purification techniques are available. It is important to first check that the background is non-specific and not due to the response against the peptide. This can be determined by performing a competitive peptide blocking study. Peptide blocking studies check that the response against the target protein is not a background artifact.
Ammonium Sulfate Precipitation
Ammonium sulfate precipitation is a commonly used method for removing protein from solution. The method is a fairly crude, non-specific purification that removes the majority of plasma proteins and leaves the immunoglobulin fraction. When in solution, proteins form hydrogen bonds with water through their exposed polar and ionic groups. Adding small ions such as ammonium or sulfate removes water molecules from the protein, resulting in precipitation of the protein out of solution. It should be stated that ammonium sulfate precipitation will not result in highly purified antibodies. The contaminants will consist of other high-molecular-weight proteins and proteins that are trapped in the large flocculent precipitates. It is recommended that ammonium sulfate precipitation be used as part of a purification scheme involving further purification steps.
Protein A or Protein G purification removes the IgG fraction based on the specificity of these proteins for the Fc portion of the IgG. Protein A is produced from Staphylococcus aureus. It has the capacity to bind at least two molecules of IgG. The binding is specific to the Fc portion and does not affect the antigen binding sites. Protein G is isolated from Group G streptococci and binds the Fc region of the IgG in a similar manner to Protein A. Protein A and G have differing binding efficiencies for IgG from different species. It is important to check this before deciding which method to use. For example, Protein G works well with sheep Ig, but Protein A does not. Neither Protein A nor Protein G will bind to chicken Ig. Care should be taken when eluting the antibody from the column to avoid denaturation of the antibody.
The most commonly used method to purify antigen-specific antibodies from crude sera is immunoaffinity purification. Unlike Protein A or G, the non-specific Ig fraction is not retained. In this procedure, peptide antigen is bound covalently to a solid support. The antibodies within the polyclonal sample that are specific for the peptide antigen bind to the support column. The unbound antibodies are removed from the column by washing and the specific antibodies are eluted from the column. The product of immunoaffinity purification is highly specific antibodies. Immunoaffinity purification can occasionally cause denaturation of the antibody due to the conditions used to elute the bound antibody from the column. It is important to compare the response generated by the purified sample against the response generated by the crude sera.
Van Regenmortel, M.H.V., 1988, Synthetic Polypeptides as Antigens, Elsevier, Amsterdam.
Halow, E., 1988, Antibodies; A Laboratory Manual, Cold Spring Harbor, New York.
Van Regenmortel, M.H.V., 1986, Trends in Biochemistry, 11:36-39.
Westof, E., 1984, Nature, 411: 123-126.
Hopp, T.P. and Woods, K.R., 1981, Proc. Natl. Acad. Sci. U.S.A., 78: 3824-3828.
Kyte, J. and Doolittle, R.F., 1982, J. Mol. Biol., 157: 105-132.
Chou, P.Y. and Fasman, G.D., 1974, Biochemistry, 13: 222-245.
Lim, V.I., d1974, J. Mol. Biol., 88: 873-894.
Parker, J.M.R. and Hodges, R.S., 1991, Peptide Res., 4: 347-354.
Parker, J.M.R. and Hodges, R.S., 1991, Peptide Res., 4:355-363.
Tam, J.P., 1988, Proc. Natl. Acad. Sci. U.S.A., 85:5409-5413.