Protocol Online logo
Top : Forum Archives: : Immunology and Histology

shape-shifting antibodies? - (Dec/10/2007 )

Pages: 1 2 Next

I am writing on a short research paper for my Immunology class and Ican't seem to find anything about shape-shifting antibodies...i'dlike to discuss the probability of engineering an antibody that can conform it's 3-D structure to a pathogen, no matter how many times this pathogen can change its structure or mutates.
If e.g the HIV-virus can mutate constantly(and therefore make it diffucult to come up with appropriately designed complimenary ab's) can we design an ab that can adapt its shape accordingly?
Adapting is a big topic in evolution.
Like a chameleon adjusting to its environment; -an antibody adjusting constanltly and instanteously.Mimickry in antibodies.
So, there is still a lot that I have to read and find about in order to base this idea.I can't find anything on this topic.
I am trying to find more info about chem. enhancers that might increase the interaction between ab and antigen.
What do you think? Do you suggest any other approaches?

Thank you,


if you find it, patent it - that's like the holy grail, which is why there isn't much out there. good luck - very interesting topic, but very tough


what would keep it from shifting into a shape that would attack something normal?


Smart antibody? hmm, no one has done it yet. Would a bi-specific antibody gives you more utility in this case? If one site on the antigen mutated, your antibody can still bind to the other site. If both remain, you got stronger binding, if the epitopes are located in proper position.


very interesting
I don't know whether this helps you but just trying
to make single chain antibodies or to generate bicistronic constructs people use a tetramer of GGGS which is very flexible
so would engineering this sequence in the variable region of the antibody would make it attain more flexible shape so that it can bind to different mutant forms of the antigen?
try looking for this linker region details and papers
all the best


Thank you everybody for your input so far, especially to genehunter and leelaram.

Dear leelaram,I will definetly look into the use of 'tetramer of GGGS'.
I needed a starting point somewhere and this really helps.Also the construction of a bi-specific antibody as genehunter suggested, with a tetramer of 'GGGS' incorporated in the variable region and then start a trial experiment to investigate if this engineered ab is starting to adapt to any 3D conformities.
Well obviously i am not trying to find the cure,(or holy grail of that matter) and even if my paper is going to end in a negative result
I have always been interested in this topic and I am very happy and enthusiastic now that I have found an area to start researching about which is related in trying to solve this issue.
I thank you dearly, as I am sitting here in the library trying to find ressources and materials.



I don't think that a bi-specific would get you where you want to go here. I think doubling your specificity is not helpful, when what you want is to increase your adaptability. unless you could make an antibody that was multi-specific? it would probably be easiest to start by engineering antibodies against classes or groups of molecules - then making them into a multimer. to make one antibody specific to unlimited things would be damn near impossible at our current level.

I'm thinking the way to go is to put the antibody on a chip. use a random-number program to determine specificity, and therefore conformation of the VH and VL regions. to be honest, it might be easier to modify APC? or, boost the innate immune system


The idea sounds interesting but whether it's feasible is another thing. You might expect antibodies with such flexibility to have been developed in nature if it were a realistic possibility. Antibody diversity is already large but microbes still manage to avoid the immune system. The other point is the acquired immune system is designed to be antigen specific and such 'flexible' antibodies might give rise to more non-specific effects and autoimmunity.

Maybe the direction to take is to use a multiunit vaccine with a number of different targets to avoid selecting out mutated viruses in the same way that chemotherapies for cancer/HIV are used in combination to avoid the development of resistance.

Or to predict the likely mutations in the target antigen and development a cocktail of antibodies to take in these possibilities.

Good luck with your paper.



'Ceri' and 'mdfenko' are right. What would prevent such an adaptable antibody from attacking something of it's own ??(i.e., host's own protein).

What 'leelaram' is suggesting is not 'adaptability' but 'a hinge region'. A combination of glycines and serine amino acids give localized region of high 'flexibility' - like a hinge or a knee joint. This has nothing to do with 'adaptability', as what you are looking for.

Antigen (epitope) recognition comes from the rich diversity of the side chains of the aminoacids in the hyper variable region of the antibody chain. This is what the body tries to manipulate to get the best fit to the epitope. Glycines don't have a side-chain, only a -H.

The best I can think of is to engineer 'natively unfolded regions' into these hyper-variable regions of the antibody chain. These natively-unfolded regions are known to be highly adaptable. BUT, this comes at the cost of specificity and binding affinity.

best wishes for your project.



Hi all,
I think you people are mistaken about the flexible hinge region in variable region
If we can engineer a GGGS sequence in the variable region of the antibody specific for an antigen, then because of the flexibility of GGGS sequence only point mutations within the epitope can still be recognized by the antibody but it won't be non-specific because the backbone is still specific to the antigen.


Pages: 1 2 Next