Protein Crystallization Hits

Sorting through my past years' crystallization reports I'm particularly glad about the fact that all of the crystal images are associated with a description of the crystallization precipitation cocktail. In the same image. Of course one can always go back to the original notebook and dig out the particular crystallization condition used, but time tends to cast doubts and having the well solutions directly associated with the crystals shown in  an image is very practical. Here's a simple way to pull out primary screen parameters for association with a particular crystallization experiment:

Let's say the crystallization was done with the JCSG+ screen, formulation from well E4

• Go to E-Screen Builder 
• Select Vendor (Emerald BioSystems),  pick screen: JCSG+
• Click on E4 - et voila, E4 contains 1260 mM Ammonium sulfate, 200 mM Lithium sulfate and 100 mM Tris at pH 8.5.
• From here you can take a screenshot, or use the 'Snipping Tool' to copy and paste this image right next to the image of the crystal in your report.

Finding a particular protein crystallization screen condition is simple: 1. select vendor, 2. select screen, 3. select hit. The crystallization formulation is displayed for you to (4) cut and paste into a crystallization report.

Easy, isn't it?

Peter

For an academic researcher paper writing is often the culmination of hundreds, if not many thousands of hours of work. The data is collected, a story is shaping up - how do you go about transmitting your newly found wisdom to your peers?

I have witnessed firsthand several different paper writing styles and practices and am the first to admit that there's no single best way to this. There's some good advice out there, though. For example, Bill Wells suggests in his aptly titled paper "Me Write Pretty One Day: How to Write a Good Scientific Paper"  'a few small steps" to make "...scientific writing clear, straightforward, and digestible". As I'm starting to write a paper myself I find it useful to look at this paper to remind myself of how to write clearly, to define what's my point and then adhere to the 'show, don't tell' rule. Wells guides his readers through 'The Anatomy of a Paper' and gives a lot of good advise on how to tell our story.

As a reader of a protein crystallization paper I expect to see the following:

1. The amino acid sequence of the protein that's expressed.

2. Description of a typical expression and purification experiment. What's the yield & purity? Are there functional assays that provide checkpoints for anybody who wishes to repeat your work?

3. A detailed description of what happened between eluting the protein from the last column and before setting it up for crystallization: temperature(s), storage conditions and time, dialysis (time, volume, MWCO used, volume ratio), concentrating device, centrifugation, filtration, addition of ligands.

4. The crystallization experiment/regime: composition of precipitant solution buffers, crystallization type, volumes, trays, dispensing methodology (stirred or mixed?), screen used, time it took for crystals to show up. Any 'out of the norm' observations, such as 'crystals formed at the air liquid interface'

5. Crystallization result: how many crystals, crystal habit, size, crystal quality (X-ray diffraction limit and any 'Table 1' associated data that resulted from a complete data set). Describe reproducibility (only 1 drop out of 10 produced crystals?, only 1 out of 10 crystals diffracted?) and any insight gained while handling the crystals (did the crystals bend when looping out, do they tend to break? Do they 'melt' when the temperature is changed?).

6. X-ray diffraction experiment: treatment of crystal (cryo used, orientation crystal was mounted) before and during exposure to X-rays. Diffraction equipment used, exposure time, sweep angle.

7. Two images of the crystals: close-up to see the crystal habit and an overview of the entire crystallization experiment. An X-ray image showing diffraction spots (crystal oriented along a special axis would rock).

There's really no good reason to hold back with any such details in a protein crystallization paper (I won't hold it against you when I review your paper). After all, one of the reasons you're writing the crystallization paper is that you aim to provide instructions to a crystallizer who seeks to enable repetition of your work in the future. This is a neat way you can 'pay back' to all scientists that provided you with insightful tips in their papers and that have helped you succeed with your own protein crystallization research project. Pass on the baton.

Cheers, Peter

We were excited to host a UV microscope demo from JANSi here last week. UV images of crystallization setups can be used to distinguish protein from salt crystals and we wanted to see first-hand how much of a difference this particular UV microscope for protein crystal inspection from JANSi  makes.

For recent reports on the utility of UV absorption and fluorescence, check out Harindarpal Gill's paper:

where these microscopes are compared: PRS-1000 , UVEX, MUVIS, QDI-2010 UV microscopes from Korima, JANSi, Formulatrix and CRAIC Technologies, respectively. One could use this paper as a 'shopping guide' to identify the instrument that best fits one's particular need.

The fundamental idea is that tryptophan containing proteins (>80% of proteins have Trp) absorb UV (280 nm)and fluoresce (320 nm), while salt crystals do not.  Hence, such UV microscopes can be used as a first stage process filter to avoid screening  many salt crystals with scarce X-ray beamtime. Having a tool to quickly resolve the salt vs. protein crystal issue is very useful.

So, did it work? Of course it works! Salt crystals appear dark and protein crystals appear bright.

The surprising thing for me though was something different: I re-discovered a crystal that I had seen before but was triaged because (i) it was a solitary object (I'm expecting showers of small needles) and (ii) within the crystallization trial there were many other hits that looked better (usually having many objects within a single setup, some with nicer facets). Now that I know that none of what we have tested with X-rays turned out any viable diffraction, I may actually go back and check out this solitary object: 

So, contrary to my expected utility for UV microscopes (namely to decrease the number of objects that need to be screened with X-rays) in this case UV microscopy can be used to identify additional hits that would have otherwise been overlooked. Rather than a filter, UV microscopes have utility in focusing your attention on the likely winners.

I was surprised by that,

Peter 

Over the past two years I've systematically reduced the amount of printed paper in my office as well as in my private life. The idea behind this effort is that 'going digital' with all my documents will make it simpler to store and quickly find information that I need. This is work in progress where the crucial software elements that stuck are Sync Toy (general file housekeeping), PICASA (automatic archiving of images) and Google Desktop (searches everything I have on my PC; yes - I'm a PC).

The most recent addition to this toolset is Mendeley. This is a PDF file organization tool that's devised specifically for researchers to archive, annotate and share their PDF articles. To say the least, I'm very impressed with Mendeley's utility. Within a short period of time Mendeley has helped me to aggregate and organize all of my (currently 642 and growing library of) PDF documents. The functionality goes well beyond a traditional Reference Manager. Features are here, the ones that I like particularly are:

• making notes within PDFs,
• sharing libraries over the web,
• backing up my own collection of articles,
• cross-talk with Zotero (a Firefox-based reference manager),
• cross-platform compatibility (I have Mendeley installed on Windows 7 and on iPhone/iPod touch, and use it via the web browser interface - works seamlessly so far) and
• search functionality.

Here's a quick primer to Mendeley:

Mendeley Teaching Presentation

Mendeley : the best research tool since streak-seeding?

There's even a social media facette too; here's my Mendeley profile - friend me if you'd like: 

Oh - and did I mention that Mendeley is available for free?

Cheers,

Peter

Working with membrane proteins has its own joys. The main reason for this is that you're dealing with amphiphiles that show interesting behavior in solution. Trouble is, that such lipids and detergent are usually very expensive. Hence, micro methods that reduce the amount of detergent and membrane proteins for initial screening purposes are highly welcome. Daniel Gutmann et al. have devised rapid and cheap protocols to screen membrane protein samples for

• Solubilization in different detergents,
• Finding detergents for crystallization, and
• Finding the best buffer to combine detergents.

Gutmann, D.A.P. et al., (2007)
A high-throughput method for membrane protein solubility screening: The ultracentrifugation dispersity sedimentation assay.
Protein Science, 16:1422-1428

Their main point is to replace size exclusion chromatography with ultracentrifugation. Running a sizing column usually requires multiple hundred mL of detergent solution, while spinning a protein sample in an ultracentrifuge tube and assaying the supernatant with SDS-PAGE can be done with 5 uL. That's a smart thing to do if you have access to 100,000 g and a suitable rotor holding small tubes. The protocol, glamorously called "ultracentrifugation dispersity sedimentation assay" is hence very similar to a conventional pull-down assay at high rotor speed. Ultracentrifugation is required to separate the non-sedimenting properly detergent solubilized membrane protein from heave aggregated membrane proteins.

I was glad to see that the authors showed images of the membrane protein crystals that they grew (using dodecyl maltoside as a detergent), together with diffraction images. But why on earth are they not disclosing the nature of the membrane protein? Can you imagine a paper describing in detail the "Expression and purification of membrane proteins" in the Materials and Methods section while disguising the nature of the three target proteins as MP-A, MP-B and MP-C? Well, MP stands for - you guessed it - membrane protein; some kind of ABC transporter. Here is it. 

"I'm not telling you what I'm working on but I need the paper". Wiley InterScience, Protein Society & PROTEIN SCIENCE - how could you let this one slip through?

So much for the flak.

Other than that - great micro method contribution to help us do more with less.

Peter