Practical Questions for Protein Crystallization
I have come across quite a few papers and books about protein crystallization. Unfortunately, I have not yet found a resource that answers the more practical questions about crystallization:
Typical Questions:
1) What should I read to understand the basic theory of protein crystallization?
2) How much protein do I need for crystallization?
(Note: Peter did a good post on this topic: Risky business or, how much protein sample do you need?)
3) How much will it cost to attain a protein crystal?
4) What is the ideal buffer for my protein?
5) Which method should I use for crystallization?
6) What concentration should I use when setting up my crystallization trails?
7) What size should my drops be during expansion set ups?
When should I expect to see crystals?
9) How can I tell that I have salt crystals and not protein crystals?
10) What screen(s) should you use and in what order?
Have I left any questions out?
Please feel free to add more questions below in the comments section or send me an email: sean@p212121.com
If you are new to commenting on blogs, your comment will be ‘held for moderation’ and not appear immediately. Having this setting has already blocked thousands of spam comments – thanks for your patience.
Bosco
June 30th, 2009 at 7:40 PM #
As a theorist, I found Lattman and Loll’s “Protein Crystallography: A Concise Guide” a really clear and concise explanation of the principles of crystallography.
Sean
June 30th, 2009 at 7:47 PM #
Thanks Bosco, but am unsure how this relates to this post. I am looking for questions about protein crystallization.
Jinny
July 3rd, 2009 at 12:42 PM #
Where are you from? Is it a secret?
DrSNo
July 3rd, 2009 at 1:02 PM #
3) How much will it cost to attain a protein crystal?
X-tal condition already known and just soaking in ligand? Cheap! Some say $5000. Include the time it takes to write paper and publish. Depends a lot on personnel costs. One structure I solved, took ~4 years from start of project to publication. 20+ constructs, 20+ purification with each purification requiring 7 different columns. Cost?? Too much (personnel costs being the highest). Impact factor, the highest. I think the structural genomic centers are down to $50000/structure. The biological value of most of those structures are very limited. How many structures of almost identical proteins from various bacteria do we need?
DrSNo
July 3rd, 2009 at 1:12 PM #
(1) Biggest Q, why do you need the structure, is it worth all the effort?
(2) you don’t have to know anything (much) about crystallography to solve a structure. Most structures are readily solved. Just click, click, click with your mouse and you might be done solving a structure in a day. Wladek Minor (UVA) solves a structure using HKL3000 in 30 min (seen him do it twice, and it was the same structure!). Clearly there can be numerous obstacles along the way. But before doing all the work, we imagine the easiest path possible.
Sean
July 3rd, 2009 at 1:49 PM #
@Jinny: No secret, I didn’t think anyone really cared, so thanks for asking! I am currently living in Toledo, OH. Hmm, maybe I should put up a bio…
Sean
July 4th, 2009 at 11:37 AM #
@DrSNo: Thanks for the additional questions!
“you don’t have to know anything (much) about crystallography to solve a structure.”
I guess it depends on the structure, but there are some structures that are quite difficult that even those who study crystallography for years are unable to solve.
I would say that Wladek Minor is an expert in crystallography and has spent years studying the subject, the result: click, click, click – done. My fear is that there is a tendency to just simply click through the programs without considering the results of each step.
Raj Gosavi
July 21st, 2009 at 10:40 PM #
Here are some questions that can be added to the above list. In my experience these have been crucial too.
1. What temperature is optimum for crystallization?
2. Which tags and where (N-term or C-term or both) should be added and how to choose them?
3. How to design proper construct for proteins (N-term truncations or C-term truncations or both)?
4. Should I try to co-crystallize the protein with any substrate (peptide, ligand, nucleic acid, protein, etc..)?
Sean
July 25th, 2009 at 7:55 PM #
@DrSNo Did he already know the structure? I believe that I could solve a structure about that fast using FREE programs. I will have to play around and time myself one of these days