Hi Amit,

Yes, there will likely be antibodies made against the tags linked to your protein. When using recombinant proteins as the immunogen to make antibodies, we always remove the tags before injecting. I think the pET vector should contain a thrombin and/or enterokinase cleavage site to enable tag removal. That would be the best if you want antibodies specific for your protein of interest. To purify antibodies specific for your protein, you will still need an untagged version of it, otherwise you will select antibodies that not only recognize your protein, but the tags as well. So, my advice is to remove the tags before proceeding with the immunization.

Alternatively, you can use a peptide sequence from your protein (be sure to blast the sequence to make sure it shows little homology to other proteins). This is a commonly used approach to make antibodies as well.

yes poly clonal by definition may cross react to other protein, for theis reason you should test it out its absolute specificity by preabsorbing with the immunising agent.

My suggestion is to use another tagged protein (p.e. GST) for the purification of the antiserum. This is a good strategy to improve the specificity of your antibodies and to eliminate antibodies cross-reacting with the different tags

You can also do a serial dilution of the polyclonal antibody and select the dilution in which the specific band is more prominent than the non-specific ones.

The immune response to an antigen generally involves the activation of multiple B-cells all of which target a specific epitope on that antigen. As a result a large number of antibodies are produced with different specificities and epitope affinities these are known as polyclonal antibodies.

Monoclonal antibodies represent a single B lymphocyte generating antibodies to one specific epitope.

The answer to your question is yes. they will cross react and you have to either use a different tag for purification of serum

Hi Amit,

Yes, there will likely be antibodies made against the tags linked to your protein. When using recombinant proteins as the immunogen to make antibodies, we always remove the tags before injecting. I think the pET vector should contain a thrombin and/or enterokinase cleavage site to enable tag removal. That would be the best if you want antibodies specific for your protein of interest. To purify antibodies specific for your protein, you will still need an untagged version of it, otherwise you will select antibodies that not only recognize your protein, but the tags as well. So, my advice is to remove the tags before proceeding with the immunization.

Alternatively, you can use a peptide sequence from your protein (be sure to blast the sequence to make sure it shows little homology to other proteins). This is a commonly used approach to make antibodies as well.

Yes certainly polyclonal antibodies cross react. By pre treating the polyclonal antiboides with tags you can get antibodies to your protein upto some extent only. Because antibodies formed against conformational epitopes of both tag and your protein will always be there and cross react.

the short answer to your question is yes, it is pretty likely that cross-reacting abs will be induced. And therefore, you definitively have to adsorb the serum or already better affinity purified abs (e.g. purified via protein A, G) by using an affinity column. This affinity column can be constructed by covalently linking another recombinant protein that is expressed and tagged in the same way as your antigen. With respect to the antigenicity of tags, in general homopolymeric amino acid (aa) sequences such as the oligo-his tag are less immunogenic than other peptide sequences. But commonly there are few aa between the (his-)tag and the protein sequence. These foreign linker aa are encoded by the restriction enzyme sequence(s) used for cloning. Many "anti-his" abs are directed to these vector encoded flanking sequences. So you have to be careful by immunoadsorption and you should use not only the same vector, expression system but also the same cloning sites for your control protein. And after (perhaps repeated) immunoadsorption, you have to check your adsorbed abs for cross-reactivity to other recombinant proteins tagged in the same way (e.g. by western blotting).

The answers above provide very useful information. You should keep in mind that monoclonal antibodies also can cross react. Monoclonal does NOT equal monospecific. Depending on the binding site's affinity for its target and its ability to recognize other similar epitopes, monoclonal antibodies also can be very dirty. Simply changing to a monoclonal antibody is not necessarily a solution for cross reactivity problems.

From my understanding, any kind of antibodies may cross react with other proteins. For tagged proteins, it aslo depend on the protein loading, for example, if you load with 1 ug tagged protein, antibody may cross react with it; on the otehr hand, if you only load 1 ng protein, there may be no cross reaction.

The following article from Dr. Saper, the former eidtor of JCN may help your question.

Thanks,

You have full text access to this content

Magic peptides, magic antibodies: Guidelines for appropriate controls for immunohistochemistry

Clifford B. Saper MD, PhD, Paul E. Sawchenko PhD

Article first published online: 25 AUG 2003

DOI: 10.1002/cne.10858

Copyright © 2003 Wiley-Liss, Inc.

Issue Journal of Comparative Neurology

Journal of Comparative Neurology

Volume 465, Issue 2, pages 161–163, 13 October 2003

Additional Information(Show All)

How to CitePublication History

SEARCH

Search Scope

Search StringSearch

Advanced >Saved Searches >

ARTICLE TOOLS

Get PDF (44K)

Save to My Profile

E-mail Link to this Article

Export Citation for this Article

Get Citation Alerts

Request Permissions

More Sharing ServicesShare|Share on citeulikeShare on connoteaShare on deliciousShare on www.mendeley.comShare on twitter

Abstract

Article

Cited By

Get PDF (44K)

Since the earliest days of immunohistochemistry, the field has been troubled by the tendency for these methods to give spurious results. There are a number of reasons why various antibody preparations may provide variable and even erroneous results, and a series of steps one can take to control for this. Unfortunately, these are not as widely known as we feel they should be, even as we enter the fourth decade in which these methods have been widely available to neuroanatomists. Hence, we provide this editorial to help guide aspiring authors through the rocky shoals of immunolocalization in the nervous system.

WHAT AN ANTIBODY “SEES”

Antibodies are large protein molecules of about 150,000 kD molecular weight. Native antibodies have two identical binding sites, although these can be dissociated from the remainder of the molecule and each other to form isolated binding domains (Fc fragments). Each antibody binding region recognizes a chemical “epitope.” The epitope may be found on any type of molecule (including entirely synthetic ones, such as dinitrophenol, a classic experimental antigen). When the epitope is on a protein, it may be a particular molecular configuration that is present in the native protein (perhaps a site that is exposed only in a particular tertiary configuration) but is not present when the protein has been fixed with formaldehyde (or glutaraldehyde, or another fixative). Conversely, an epitope may be exposed in the fixed molecule that is not present in the native configuration.

An antiserum is a preparation containing the immune globulins from the serum (the supernatant after blood has been allowed to clot and the cells have been centrifuged out), usually of an animal that has been immunized with a specific antigen. The animal will have a wide array of immune globulins with highly varied binding sites in its blood, depending on its previous immune experience. Only a very small proportion of the immune globulins in that serum will be directed against the antigen used for immunization. These immune globulins will in turn recognize different parts of the antigen. Once the antigen has been fixed, many if not all of these antigenic sites will be destroyed. In fact, when animals have been immunized against small peptides, serum from many animals will not recognize anything in the specific immunizing sequence when applied to fixed brain tissue. Among the antisera that stain something in the brain, it is typical to find that many will stain spurious things, and often only a minority will stain the original molecule. The spurious staining may represent antibody that the animal just happened to be making against an unrelated antigen (e.g., a virus the animal encountered) and that cross-reacts with something in the brain, or it may represent antibody against the antigen used for immunization that cross-reacts with something else when applied to fixed tissue. When the antiserum is mixed with an excess of the immunizing peptide, antibodies of the latter type will be adsorbed to the peptide and will not stain. Antibodies of the former type will continue to stain. Hence, the first step in authenticating a staining pattern is to carry out an adsorption control (see below).

A monoclonal antibody is a single clone of immune globulins that is raised against a specific antigen. The immunized animal is killed, B cells that make immune globulins are isolated, and these are fused with neoplastic B cells from a myeloma tumor. The resulting cells are immortal (divide continuously), and each continuously dividing cell clone makes a single type of antibody molecule that recognizes a single epitope. The antibodies made by each clone are screened for recognition of the original antigen, and ones that do recognize the antigen are called monoclonal antibodies against that antigen. Because this clone of antibodies recognizes the antigen, it should always be blocked in adsorption testing (see below), although, if the antibody-secreting cells are raised in a host animal (instead of cultured in vitro), the ascites fluid that contains the antibody may be contaminated by host immune globulins. However, even the purest monoclonal antibodies provide no greater assurance than polyclonal antisera that the epitope that is recognized in the tissue belongs to the original antigen. They require the same types of additional controls and careful interpretation.

Interestingly, many polyclonal sera may functionally act as monoclonal preparations, because it is possible that only a single clone of antibodies (of the many present in the serum) recognizes the fixed antigen in the tissue. The probability that one or more clones in an animal's serum will recognize the fixed antigen is predicted by a Poisson distribution. Because the failure rate for producing any clones that stain tissue is high (as many as 50% or more of sera from immunized animals fail to stain anything in fixed tissue and hence have zero clones capable of recognizing the antigen in fixed tissue), the Poisson distribution predicts that most antisera that stain tissue probably only have one or two clones of antibodies that actually provide the staining, and the rest of the antibodies in the serum are irrelevant, except for producing artifact. As a result, some investigators affinity purify antisera by running them over a column to which the antigen is fixed to a support matrix. The nonbinding antibodies pass through the column, and then the clones of antibody that bind the antigen are released. Such antisera are less prone to artifact (the adsorption test has essentially been performed in the purification step) and tend to have lower background staining. However, quite often the highest affinity clones, which are difficult to elute, are lost in the process, and the other caveats that apply to monoclonal antibodies still are relevant.

WHAT IS AN ADEQUATE SET OF CONTROLS FOR AN ANTIBODY?

The key issue in designing controls for antibody staining is this: Does the antibody actually reveal the antigen of interest, or is the binding an artifact? The gold standard for deciding this issue is to stain tissue with and without the antigen of interest. This can be accomplished by staining tissue from a wild-type mouse and one in which the antigen of interest has been eliminated by transgenic engineering. If the antibody fails to stain the transgenic tissue, then the antigen is present either on the molecule whose gene was deleted or on some molecule that is “downstream” of that protein in a metabolic or enzymatic pathway. It is still possible, therefore, for an antibody that passes this test to stain something other than what was intended, but this is very unlikely. On the other hand, if the antibody fails this test, it cannot be used for immunolocalization in nervous tissue. There have been several examples of antibodies that have passed less stringent tests and yet failed the knockout test. Hence, this test is recommended for all antibodies for antigens whose localization is not already well characterized. Unfortunately, it is directly applicable at this point only in mice, and staining in tissues from other species may still represent other antigens (not present in mice) and hence requires additional controls.

If a knockout animal is not available for testing, a less stringent but still reasonable set of tests can be carried out. It is first necessary to perform a Western blot of the tissue of interest, to show that the antibody recognizes only one antigen in the tissue and that it is of the appropriate molecular weight. Although some antibodies that recognize two or more bands may identify the same target molecule in different states of posttranslational modification (or in oligomeric forms), the use of such an antibody requires verification with antibodies against a different part of the same original molecule to demonstrate that the staining is genuine.

The antibody should then be subjected to an adsorption test. It is necessary to perform this test carefully, and many investigators are not aware of its nuances. The investigator virtually never knows the absolute amount of antigen that constitutes an “excess.” This is because the amount of antigen necessary to block, e.g., 50% of staining can be calculated only if the investigator knows the Kd of the antibody for both the tissue antigen and the soluble one (which may not be the same, insofar as the soluble antigen is usually not reacted with fixative). Some investigators use a large amount of antigen (e.g., 10–100 μM/ml of diluted antibody), but this may not be enough if the antiserum binds particularly tightly to the tissue antigen or loosely to the soluble antigen or if it is present in the tissue in overwhelming abundance.

A more sensible approach to adsorption testing is to dilute out the primary antiserum to the point where it just stains the sites of interest in the tissue. Most antisera will go from dense staining to light or imperceptible staining with a dilution of about tenfold. If the investigator starts with the dilution that just obliterates staining and then doubles the antibody concentration, this will provide a concentration of antibody at which a 50% or greater reduction in antibody availability will then eliminate staining. Adding a large excess of antigen (e.g., 10–100 μM/ml of diluted antibody) at this point should serve to block staining. It is important to note that, even for an antibody that has been characterized extensively under other circumstances, it is important to perform this test under tissue and fixation conditions identical to those used in the actual localization experiment.

It is common to accept adsorption controls as minimally acceptable documentation in support of antibody specificity. However, there are circumstances in which they are clearly not sufficient. For example, if the antigen used for immunization contains an amino acid sequence that is shared among a group of structurally related proteins or peptides, it is possible that the antibody may recognize any or all of the members of the group in staining tissue. In fact, instances of spurious cross-reactivity have been documented in cases in which peptides share just a few amino acids in common or may even generate a tertiary epitope that is similar. Hence, adsorbing the antibody against known structurally related antigens can provide additional security against cross-reactive staining.

Under some circumstances, such stringent controls might not be necessary. For example, in some cases, an antiserum may be well characterized, and its staining pattern may be well known in the species and tissue in which the experiment is performed. In that case, it might not be necessary to repeat all of the controls again. Under other circumstances, the antigen may be a large protein that is not readily available as an antigen for adsorption studies. In such cases, the best control that may be available can be omitting the primary antiserum. In the end, the validity of the controls in a specific study will always have to be decided by the referees of that paper, and it is always wise to err on the side of using more controls than necessary, rather than fewer. In particular, when applying an antibody to a new species, especially one that is not closely related to the one in which it was originally characterized, it is almost always necessary to perform a full set of controls, including immunoblots with tissue from that species.

MAGIC ANTIBODIES

A cornerstone of scientific publication is that the methods must contain adequate information to allow another investigator to repeat the experiment for the purpose of verification. For investigators who use antibodies, this entails a special duty. First, the manuscript must contain adequate information concerning the antigen. Some companies are well-known for keeping the exact sequence of the immunizing peptide as a “trade secret.” Investigators should be aware that such antisera are not fit for use in serious scientific work, nor will we allow publication of studies using such antibodies in the Journal of Comparative Neurology. The full sequence of the antigen must be available for the publication to be considered by this Journal. This means that the author must either get this information from the purveyor of the antibody or must sequence the blocking peptide that is sold with the antibody. “Magic antibodies” against “magic antigens” are not science.

Second, full information must be given about the antiserum that is used. Each antiserum is, in a sense, unique. When an animal is bled, there is a unique combination of antibody clones expressed on that particular day. The same animal, bled on a different day, may express a very different set of antibody clones. Hence, it is important to know whenever possible both the code number for the animal and the bleed. The catalog number for the antiserum may not be sufficient. Because bleeds from rabbits are limited in volume, a given manufacturer may run out of a particular bleed, and substitute another bleed from the same animal, or a bleed from a different animal, without changing the catalog number. Quite often the new bleeds have not been checked to make sure that they stain the same thing. In practice, it may be impossible to get full information at this level from some manufacturers; they may not have accurate records available. Hence, it is very important for investigators to keep a small supply of whatever antiserum they use for publications, in case questions arise in the future because of a failure to replicate a staining pattern. This policy allows checking new antisera against the old one for staining patterns. Similarly, if the investigator prepares a new antiserum, it is important to make samples of this available to other investigators who want to replicate the work. Fortunately, this issue is not a problem for monoclonal antisera, which by nature are the same every time.

DON'T BELIEVE EVERYTHING YOU SEE

The most important thing to understand about antibodies is that they are biological agents, not standard chemical reagents. Antibodies may bind to a wide variety of antigens other than the one that they were raised to recognize, and there is no way to be sure that the pattern they stain really represents that antigen. Hence, it is important to have sufficient controls to convince a skeptical observer that the staining pattern observed really does represent the antigen, and wherever possible the results should be checked with other techniques. Methods that provide verification of the expression of the same molecule in the same cells (e.g., in situ hybridization or single-cell RT-PCR) are particularly powerful. Ultimately, this will be a matter for our referees to decide on a case-by-case basis. In the end, all authors and readers should be aware that antibody staining is fallible, and such staining should always be considered to label “antigen-like” molecules rather than staining the actual antigen.

Processing of the antigen/antibody depends upon the use of antibodies raised. Different antibodies from an antiserum react in different assays and hence we see difference in antibody titers estimated by different methods. Results may/ may not overlap or interfere, particularly in functional assays. Cross reactivity exists in nature and may even be observed in assays using antibodies of desired purity; hence, results have to be evaluated accordingly.

I agree with Tina Leisner's opinions.

If an untagged version of immunogen unavailable, one may want to use another protein from the same vector as immunogen to run subtractive immunization in the animal prior to the normal immunization with the target tagged protein for polyclonal antibody production. This way may reduce the cross-reactivity to the tags. There are some protocols one can find from the internet.

Yes.

Yes it will but it is not a problem. What we do in my lab is to make the same protein in GST-fusion and His-Tag. It is a litle more cloning but works well. Then when you get your serum, if you immunized with the His-tag fusion protein you then affinity purify using the GST fusion protein. We have even immunized goat with multiple fusion proteins for different protein and purify by affinity sequencially.

Some key point: Crosslink your fusion protein for affinity purification on NHS-agarose or magnetic beads.

Purify small amount of antibody at a time. We do this from 1 ml of serum which will give you 50 to 200 µg of affinity purified antibody. The reason is that the serum has almost an infinite life in a frezzer while the affinity purified antibody have usually short life. 2 to 4 weeks at 4C. 2 to 4 month at -20 in 50% glycerol. Some affinity purified antibody will be destroyed by freezing at -20 or -80 even in aliquots so better do a small test.

In my hand the substraction does not work.

A short way to do your affinity purification is by using small spin column. Incubate you serum with you covalent beads. 2 hour at room temp or overnight at 4C. Wash by spining twice with PBS. Transfer to the spin column and spin for 10 sec at 3000 rpm. increase time if your liquid has not gone through. Wash 2 more time with PBS and then elute rapidly with glycine (100 mM ph 2.8). 2 elution with 250 µl will usually work. When you elute you want to collect in 1M tris ph 9.5 (25 µl of 500 µl of glycine will neutralise).

This way the antibody is only exposed to low Ph for less than a minute and the recovery is good.

If interested I can post the full protocol.

Dom

The polyclonal antibodies that you generate will be potentially against any portion of the protein/antigen, with more immunogenic regions of the protein being more likely to produce antibodies directed against that domain. You have three choices to remove the antibodies that are generated against the tag.

#1 You can try to dilute out the polyclonal antiserum and hope that the antibodies against the tag are at a low concentration and are therefore, preferentially lost during the dilution. You have little control in this approach, if the antibodies against the tag are in high concentration, it will not work.

#2 You can clone your protein/antigen into another expression vector that has a different tag. Express this new fusion protein, purify the protein, and covalently couple it to a column (such as Sepharose). You can then pass your polyclonal antiserum over this column. The antibodies against the first tag will not bind and will flow through to be discarded. You can then elute the antibodies that bound to the fusion protein and they should be purified against the desired protein.

#3. Clone a different protein into the pET32a+ vector, express and purify this protein. Covalently couple it to a column. Pass your antibody over this column and the antibodies against the tag will bind and the antibodies against the desired protein flow through.

My lab has used both approaches #2 and #3 to purify polyclonal antibodies successfully. You can find examples of both approaches if you search Pubmed for papers by DR Hyde. Good luck.

Thanks everyone for your suggestions.