Expert: Florence M Rollwagen - 5/9/2009

Question
Hi, thank you so much for taking the time to look at my
question.  I am a third year Biology undergraduate student
and am trying to get some ideas of how to design good
experiments.
What is the best way of testing where a protein is
expressed? I know you could tag the protein, but then how
would you find out where it was expressed without having to
extract tissue from every cell in the body??! I have also
heard of expression studies where a promoter for a gene of
interest is taken from another cell and then inserted into
the new cell where the gene of interest has been replaced
with GFP. So the GFP would be expressed wherever the
original gene would have been expressed Would this work?
Also again, how do you know where it is without testing
every tissue in the body? Once you know where the gene is,
would you then do a Western blot, or is the Western blot
INSTEAD of the protein tagging?
I was also wondering what is the best way of testing the
function of a gene product? Would it be to KO or knock down
the gene and then maybe test that you knocked out the
correct thing by doing a rescue experiment? Is that the
best way of testing that a KO experiment had worked or
would qtPCR be better?  Also for a gain of function
experiment, what would be the best way of testing that
worked? What qtPCR be good or could you use a comparative
hybridization microarray between the tested tissue and a
control?
Again, thank you so much,
Gemma  

Answer
Hi Gemma:  thanks for your (interesting) question.

You are definitely running before walking in this question, but no matter, it shows you are thinking, which is always good.

You are basically asking two related questions:  1.  What’s the best way to find out where a protein is synthesized in the body, and, 2.  What does it do?  Also, in your questions, you’re exchanging “gene” with “protein”.  I’ve answered as if all you have is the protein which you are trying to learn more about.

1.  What is the best way of testing where a protein is expressed? I know you could tag the protein, but then how would you find out where it was expressed without having to extract tissue from every cell in the body??!

Usually when dealing with a new protein, an extract is made of each tissue (brain, liver, etc) and a Western blot run on the homogenates.  If this is a protein that you’ve isolated, you will need an antibody.  Making it yourself would be time consuming and unless you had an immunologist as collaborator, pretty near impossible.  Fortunately, there are services where you can send your protein and they’ll make the antibody.  Here’s an example: http://vir.sgmjournals.org/cgi/reprint/69/4/955.pdf

2.  I have also heard of expression studies where a promoter for a gene of interest is taken from another cell and then inserted into the new cell where the gene of interest has been replaced with GFP. So the GFP would be expressed wherever the original gene would have been expressed.  Would this work?

Yes, it would work, but would be of no use in determining where the protein was made in the body.  Because you would be inserting the gene artificially, it would not resemble the natural location.  Remember, you don’t have the gene, all you have is the protein.

3.  Also again, how do you know where it is without testing every tissue in the body? Once you know where the gene is, would you then do a Western blot, or is the Western blot INSTEAD of the protein tagging?

Yes, you have to test every tissue in the body.  Protein tagging would be of little use in learning where in the body the protein is synthesized.  If you wanted to find out where it is bound (like a hormone) tagging would work.

4.  I was also wondering what is the best way of testing the function of a gene product?

5.  Would it be to KO or knock down the gene and then maybe test that you knocked out the
correct thing by doing a rescue experiment?

Knockouts and knockdowns are VERY hard to make, and VERY laborious.  When a scientist says she made a knockout mouse, know that it takes over a year and needs a team of scientists.  So KO wouldn’t be my first choice.  You have a couple of decisions to make here.  Is your protein used internally by the cell, or is it exported?  If it is used internally, you could use RNAi (RNA interference) in tissue culture to reduce the amount of protein being made.  (you’d have to guess at the specific nucleic acid sequence from the protein sequence)  If the protein is an exported product (like albumin or thyroxin) you could use your antibody to tie up the protein in circulation and block its activity that way.  In either case, you would have to run many controls to be sure that you have effectively inhibited the protein.

6.  Is that the best way of testing that a KO experiment had worked or would qtPCR be better?  

Again, I don’t know whether you’re referring to the protein of the nucleic acid.  Quantitative PCR works for DNA, not for protein.

7.  Also for a gain of function experiment, what would be the best way of testing that worked?

You cannot at this point be asking gain of function questions.  If all you have is the protein, you’d have to guess at the nucleic acid sequence, which would always present problems due to the redundancy of the genetic code.

8.  What qtPCR be good or could you use a comparative hybridization microarray between the tested tissue and a control?

Again, once you have the nucleic acid, then you could work on these experiments.

As I said above, you've conflated gene with protein.  Decide which one you want to study, and we can talk some more on methods.

Hope this helps!

FM Rollwagen, PhD