Immunogen design

Antigens used for antibody production

  • You can use synthetic peptides, recombinant proteins, immunogens, and others. We do not recommend the us of whole cells for antibody production.
  • Arbitrary selection of the antigenic peptide or use of a recombinant fusion protein usually will result in a 65-70% success rate using standard methodologies. If the antigenic peptide is chosen using standard bioinformatic algorithms, you can increase your success rate to 75-80%. If this optimized peptide is then conjugated using Invitrogen’s proprietary approach, you can increase your success rate to about 85-90%.

A recent collaborative study showed that approximately 450 polyclonal peptide antibodies out of 500 individual protein targets recognized their target by western blot analysis.

Peptide conjugation

  • Peptides are normally too short to be immunogenic. A carrier protein enhances immunogenicity. MAP and KLH are the two most popular conjugation methods. Other available conjugates are OVA (Ovalbumin), BSA (Bovine Serum Albumin), and THY (Thyroglobulin).
  • Conjugation is usually done through the sulfhydryl group of a Cys designed into the peptide.
  • For peptides < 6 kDa, conjugation will be required.
  • If studying invertebrate species, you should use carrier proteins such as ovalbumin and KLH (from sea snail). KLH has no homology to vertebrate proteins

We utilize a proprietary multiple conjugation method where we conjugate peptides through both termini and through key internal sites. This methodology presents the peptides in alternate, multiple conformations, making them more antigenic and producing antibodies with greater utility across assays. Our conjugation scheme provides a better matching of epitopes between the presented peptide and the way the peptides are presented in the native protein.

We offer the following options for peptide conjugation to carrier proteins:

  • KLH—keyhole limpet hemocyanin is a copper-containing protein, isolated from the hemolymphs of a mollusk. It exists in five different aggregated states (in Tris buffer pH 7.4), which readily dissociate with moderated pH change. Subunit molecular mass ca. 450 kDa.
  • BSA—bovine serum albumin, subunits of 67 kDa. Popular protein used in immunoassays to block non-specific binding sites. It should not be used as carrier if future assays involve BSA (for instance as blocking agent).
  • OVA—ovalbumin is a glycoprotein and the major protein of egg white. It has a pH of 4.6 and molecular weight of 45 kDa.
  • MAP—multiple antigenic peptide. This system uses peptides attached at many positions to a polylysine core. MAPs containing 2 to 16 copies of synthetic peptide molecule can be produced. MAPs have a high molar ratio of peptide antigen to core molecule and do not require the use of a carrier protein to elicit an antibody response.

MAP versus KLH

  • Eight copies of your peptide are synthesized on a branched lysine or MAP carrier core.
  • MAP peptides are suitable for direct injection (with adjuvant) for antibody production.
  • Since the peptide is linked through the C-terminus, the MAP technique favors N-terminal or internal peptides and is not recommended for peptides from the extreme C-terminus. The carrier attachment may also cause some steric hindrance. Use KLH conjugates for C-terminal peptides.
  • The average peptide length is about 15 residues.
  • MAP peptides behave much like large proteins.
  • MAP peptides may give a slightly higher titer due to their structure. Irrelevant antibodies can attach to the peptide branches.

Insert-tag incompatibility

Many fusion vectors offer the convenience of either N-terminal or C-terminal epitope tags for purification and detection of the recombinant peptide. Epitope fusions may reduce the number of steps required to purify a recombinant protein, however these tags may also interfere with the structure and function of the recombinant peptide. The effect of epitope tags cannot be predicted as the folding of these peptide fusions cannot be predicted. In rare instances, epitope tags cannot be detected by common molecular detection methods due to interactions with the fusion polypeptide that mask the antigenic amino acids.

The Tag-On-Demand™ method uses adenoviral-based stop suppression technology to allow expression of an untagged (i.e. native) or C-terminally tagged recombinant protein of interest in mammalian cells from a single expression vector. Using a single clone, you cangenerate tagged protein for purposes of verifying expression and untagged protein for functionality. You may use one of the Tag-On-Demand™ Gateway® vectors or other compatible expression vectors available from Invitrogen.

If you wish to express your gene as a fusion to a C-terminal tag, you must clone in frame with the tag.

Peptide design

There are many important factors to consider when choosing a suitable part of a protein sequence for successful antibody production. The sequence, amino acid composition, and length of a peptide will influence whether correct assembly, purification, and subsequent solubilization are feasible.

Peptide length and purity

The purity of a crude peptide typically decreases as the length increases. Purity >70% is sufficient for generating or testing antibodies. However, purities greater than 95% are often required for biological activity studies. Typically, peptides of 10-15 residues in length are used for raising antisera and the yield of peptide for sequences less than 15 residues is often sufficient.

Typical yields for different synthesis scales

Scale Yield (Crude) Yield (Purified)
20 20 mg 10-12 mg
50 50 mg 20-25 mg
100 100 mg 50-51 mg


Solubility is strongly influenced by composition. Peptides containing a large percentage of hydrophobic residues, such as Leu, Val, Ile, Met, Phe, and Trp, will often have limited solubility in aqueous solution and may be completely insoluble. To help ensure solubility, the hydrophobic amino acid content should be below 50% and there should be at least one charged residue (e.g., Asp, Glu, Lys, and Arg) for every five amino acids.

Cys, Met, or Trp residues are often problematic in synthesis because these residues are susceptible to oxidation and/or side reactions. If possible, choose sequences that contain a minimum of these residues. Alternatively, conservative replacements can be made for some residues.


Peptide solubility is dependent on the amino acid content of the peptide. Test a small amount of peptide for solubility in water, PBS, or various solvents prior to dissolving the entire peptide amount. For most peptides, solubilization in PBS or PBS-azide is used. Depending on the amino acid content, the use of mild acids or bases may be required (e.g., ammonia or acetic acid). If the peptide is not water soluble, employ one of the suggestions below:

  • For peptides that are acidic—due to the presence of aspartic and/or glutamic acid residues—add a small amount of 5% ammonium hydroxide.
  • For peptides that are basic—due to the presence of histidine, lysine, and/or arginine residues—add a small amount of 5% acetic acid. Avoid basic conditions when reconstituting peptides that contain cysteine.
  • For peptides that are hydrophobic—due to the presence of isoleucine, leucine, phenylalanine, and/or valine residues, dissolve the peptide in a minimal amount of acetonitrile, isopropyl alcohol, ethanol, DMF or DMSO in addition to PBS or as a substitute.


  • Most peptides are stable at -20°C indefinitely, especially if they are lyophilized and stored in a dessicator.
  • Allow lyophilized peptides to come to room temperature before exposing them to air. This will minimize moisture-related effects. When lyophilization is not possible, the next best method of storage is in small, working size aliquots at -20°C or -80°C.

Post-translational modifications

Higher eukaryotes perform a variety of post-translational modifications, including methylation, sulfation, phosphorylation, lipid addition,and glycosylation. Such modifications may be of critical importance to the function of an expressed protein. Secreted proteins, membrane proteins, and proteins targeted to vesicles or certain intracellular organelles are likely to be glycosylated. The most common and best studied is N-linked glycosylation, where oligosaccharides are uniquely added to asparagine found in Asn-X-Ser/Thr recognition sequences in proteins. Another type of glycosylation is O-linked glycosylation, which involves either simple oligosaccharide chains or glycosaminoglycan chains (1). When expressing and purifying a glycosylated protein in a heterologous expression system, it may be desirable to quickly determine whether the protein is glycosylated properly. Protocols for carbohydrate analysis of proteins have been published to allow the molecular biologist to characterize glycosylated proteins of interest (2). The following sections discuss glycosylation patterns found in eukaryotic cells.

Glycosylation in mammalian cells

N-linked glycoproteins contain standard branched structures, which are composed of mannose (Man), galactose, N-acetylglucosamine (GlcNAc) and neuramic acids. O-linked glycoproteins are composed of various number of sugars including galactose, N-acetylglucosamine, N-acetylgalactosamine, and neuramic acids.

Glycosylation in insect cells

The nature of N-linked glycosylation in insect cells (Sf21, Sf9, High Five™) is dependent on the protein expressed and the host cell line. N-linked glycosylation is generally of the high-mannose type. O-linked glycosylation is similar, although not identical, to mammalian cells, depending on localization and type of protein. Drosophila N-linked glycosylation is less complex in that it is not trimmed and sialylated. Thus Drosophila proteins have a high mannose content. Drosophila can also add O-linked glycosylation. Mimic™ Sf9 Insect Cells are modified Sf9 cells that stably express a variety of mammalian glycosyltransferases. These enzymes allow for production of biantennary, terminally sialyated N-glycans from insect cells. The cells can be used to produce more mammalian-like proteins in both baculovirus and stable insect expression systems.

Glycosylation in yeast

S. cerevisiae N-linked glycoproteins contain only highly branched and extended high mannose structures (hyperglycosylation). S. cerevisiae O-linked glycoproteins are composed of less than four mannose residues. Pichia N-linked glycosylation consists mostly of short chain Man (3) GlcNAc residues and is closer to the typical mammalian high-mannose glycosylation pattern. Pichia O-linked oligosaccharides are present but are not major components of the total soluble glycoprotein of Pichia. Following you will find: an outline of the two basic types of N-linked glycosylation (1,2), a table of glycosylation inhibitors that can be used in vivo (2), and a table of enzymes which can be used to analyze carbohydrate structure on proteins. For further information about glycosylation in eukaryotes, see reference 4.

Effects of certain antibiotics on glycosylation structure in eukaryotes (2)

Type of inhibitor Target enzyme Effects on oligosaccharide structure Effects on electrophoretic mobility in SDS-PAGE
Tunicamycin GlcNAc transferase
  • Prevents first synthetic step of the core oligosaccharide
  • Glycosylation of Asn residues does not occur
Proteins migrate faster and show less heterogeneity
Deoxynojirimycin Castanospermine Glucosidase I and/or II
  • Prevents removal of first glucose residue, inhibiting further processing
  • Sensitive to Endoglycosidase H (Endo H) digestion
Dependent on processing Proteins appear to be a smaller size
Deoxymannojirimycin α-mannosidase I
  • Prevents removal of mannose residues on the αI-3 arm of the high mannose structure
  • Blocks activity of GlcNac T II (no modification of the α1-6 arm)
  • Sensitive to Endo H
Proteins appear to be a smaller size
Swainsonine α-mannosidase II
  • Prevents removal of mannose residues on the α1-6 arm of the high mannose structure
  • Blocks activity of GlcNac T II (no modification of the α1-6 arm)
  • Addition of GlcNAc, Gal and NeuAc on the α1-3 arm may occur
  • Sensitive to Endo H
Dependent on the extent of processing

Enzymes used to analyze gycoproteins

Enzyme Type of Enzyme Specificity Reference
Endoglycosidase D Endo Cleaves various high mannose glycans
 7, 11
Endoglycosidase F Endo Cleaves various high mannose glycans
 6, 3  
Endoglycosidase H  Endo
Cleaves various high mannose glycans
 12, 13
β-galactosidase Exo Removes terminal galactosides from Gal-β1, 3-GlcNAc; Gal-β1,4-GlcNAc; Galβ1,3 GalNAc

 5, 8, 9
Peptide:N-Glycosidase F
Glycoproteins between Asn and GlcNAc
Sialidases (Neuraminidases)
 NeuAc-α2,6-Gal; NeuAc-α2,6-GlcNAc;
or NeuAc-α2,3-Gal 

This table is adapted from Molecular Biology Labfax by T.A. Brown with permission from BIOS Scientific Publishers Ltd


  1. Alberts, B. et al. (1989) Molecular Biology of the Cell, second edition, Garland Publishing, Inc., New York, pp. 433-475.
  2. Ausubel, F.M. et al. eds. (1990) Current Protocols in Molecular Biology. Greene Publishing Associates and Wiley-Interscience, New York. pp. 17.0.1 to 17.14.9.
  3. Tarentino, A.L. et al. (1985) Biochemistry 24: 4665.
  4. Varki, A. and Freeze, H.H. (1994) Subcellular Biochem. 22: 71.
  5. Distler, J.J. and Jourdian, G.W. (1973) J. Biol. Chem. 248: 6772.
  6. Elder, J.H. and Alexander, S. (1982) Proc. Natl. Acad. Sci. USA 79: 4540.
  7. Kobata, A. (1979) Anal. Biochem. 100: 1.
  8. Niemann, H. et al. (1984) EMBO J. 3: 665.
  9. O’Sullivan, M.J. and Marks, V. (1981) Meth. Enzymol. 73: 147.
  10. Plummer, T.H. et al. (1984) J. Biol. Chem. 259: 10700.
  11. Taniguchi, T.L. et al. (1986) J. Biol. Chem. 261: 1730.
  12. Tarentino, A.L. and Maley, F. (1974) J. Biol. Chem. 249: 811.
  13. Trimble, R.B. and Maley, F. (1984) Anal. Biochem. 141: 515.
  14. Jarvis D.L. et al. (1995) Virology 212(2): 500-511.
  15. Duman, J.G. et al. (1998) Biotechnology And Applied Biochemistry 28 (Pt 1): 39-45.
  16. Gemmill, T.R. and Trimble, R.B. (1999) Biochimica Et Biophysica Acta 1426: 227.