Crystallization screening is the process of evaluating methods, reagents, and other chemical and physical variables with the objective of producing crystals and/or identifying the variables which are positively or negatively associated with crystallization of the sample.

At the time of this writing, up to 40% of samples screened for crystallization will produce some kind of crystalline result and 10% of samples will produce a crystal suitable for X-ray diffraction analysis. About 75% of proteins screened require optimization. Optimization is the systematic manipulation and evaluation of variables which influence the crystallization of the sample.

Primary Screen
Primary screens are front line screens used for initial screening. If one does not have the knowledge or desire for a specific bias or focus on a reagent class, one may choose a sparse matrix screen composed of salts, polymer, organics, buffers at various pH levels, and mixtures thereof. Or one may have knowledge that a specific reagent class or mixture is desired and choose a screen biased for salt, polymer, polymer and salt, or other formulations.

Secondary Screen
Secondary screens are follow up screens to primary screens. The score from a primary screen such as Index may indicate crystals or promising results in polymer and polymer – salt mixtures. In such an instance, one may choose a secondary screen such as PEGRx 1 and PEGRx 2, as well as PEG/Ion and PEG/Ion 2. Or, primary screens may show promising scores in salt based reagents, where secondary screens such as SaltRx 1 and SaltRx 2 would be appropriate for follow up screening.

Grid Screen
Grid Screens are simple, logical methods for systematically screening on a pH versus precipitant (reagent) grid.1 For example, the pH range 4 to 9 might be screened in 1 pH increments across the 6 wells of the X-axis of a 24 well crystallization plate, while a reagent, such as Polyethylene glycol 6,000 might be screened in 4 concentrations (5, 10, 20, 30% w/v) across the 4 wells of the Y-axis of a 24 well crystallization plate. The method depends on the ability to identify preliminary crystallization conditions while coarsely or finely sampling two variables, typically pH and reagent concentration. Grid Screening can be used as a primary or secondary screen strategy and is most often employed in optimization of initial crystallization conditions (hits). Grid Screens can be designed to cover a broad range of pH and reagent concentration in big steps, casting a broad net to identify an initial promising pH and reagent concentration (hit). Subsequently, successively finer grids can be generated to identify the optimal pH and reagent concentration for crystallization. The Grid Screen strategy was an original approach to protein crystallization, prior to the development and popularization of sparse matrix screening.

Sparse Matrix Screen
Sparse Matrix Screens are composed of a sampling of reagent formulations that have previously crystallized a protein.2 The formulations found in a Sparse Matrix Screen have emerged over time from the accumulated wisdom and experience of generations of many crystal growers. Initial ideas are assembled, formulated, and tested against previously crystallized and not yet crystallized proteins. Duds are dropped and winners move onto subsequent rounds of testing. Testing also employs formulations from the literature as well as databases, such as the Protein Data Bank (PDB)3 , Biological Macromolecule Crystallization Database (BMCD)4-6, in house data, or data shared through centers and collaborators. When data mining, one must carefully review the data, as screens have existed long enough now that they themselves are within the database, and one must avoid getting caught in some local minima; one should also avoid cherry picking formulations to create a screen that, while looking good on paper, produces redundant hits, rather than sample an appropriate and balanced chemical space of home run conditions as well as singles; one needs both to win the crystallization game. And though one should appreciate and respect data mining, one must also remind oneself to look outside the box, for new chemicals and formulations. The unprecedented success of polyethylene glycols, detergents, salt libraries (Tacsimate), small molecular libraries, and numerous other reagents would not have happened had it not been from looking outside the box, and a bit of dumb luck.7-11

To continue reading the rest of Crystal Growth 101 - Crystallization Screening, please click here.

Please visit the dedicated Crystal Growth 101 webpage for more topics.

References and Readings
1. A protein crystallization strategy using automated grid searches on successively finer grids. Patricia C. Weber. Methods: A Companion to Methods in Enzymology Vol. 1, No. 1, August, pp. 31-37, 1990.
2. Sparse matrix sampling: a screening method for crystallization of proteins. J. Jancarik and S.-H. Kim. J. Appl. Cryst. (1991). 24, 409-411.
3. H.M. Berman, J. Westbrook, Z. Feng, G. Gilliland, T.N. Bhat, H. Weissig, I.N. Shindyalov, P.E. Bourne (2000) The Protein Data Bank Nucleic Acids Research, 28: 235-242.
4. The Biological Macromolecular Crystallization Database: A Tool for Developing Crystallization Strategies. Gary L. Gilliland and Dorothy M. Bickham. Methods, A Companion to Methods in Enzymology, Vol. 1, No. 1, August, pp. 6-11, 1990.
5. Biological Macromolecule Crystallization Database, Version 3.0: new features, data and the NASA archive for protein crystal growth data. Gilliland GL, Tung M, Blakeslee DM, Ladner JE. Acta Crystallogr D Biol Crystallogr. 1994 Jul 1;50(Pt 4):408-13.
6. The Bimolecular Crystallization Database version 4: expanded content and new features. Tung, M and Gallagher, DT. Acta Crystallographica D65, 18-23. 2009.
7. Crystallization of Proteins from Polyethylene Glycol. Alexander McPherson. The Journal of Biological Chemistry. Vol. 251, No. 20, Issue of October 25, pp. 6300-6303, 1976.
8. The effects of neutral detergents on the crystallization of soluble proteins. Alexander McPherson, Stanley Koszelak, Herbert Axelrod, Duilio Cascio. J. Cryst. Growth 76, 547-553. August 1986.
9. A comparison of salts for the crystallization of macromolecules. Alexander McPherson. Protein Science, 2001 Feb; 10(2) 418-422.
10. Searching for silver bullets: an alternative strategy for crystallizing macromolecules. J Struct Biol 2006 Dec, 156(3) 387-406.
11. Protein crystallization and dumb luck. Bob Cudney. The Rigaku Journal, Volume 16, No. 1, pp. 1-7, 1999.
12. Efficient Factorial Designs and the Analysis of Macromolecular Crystal Growth Conditions. Charles W. Carter, Jr. Methods, A Companion to Methods in Enzymology, Vol. 1, No. 1, August, pp. 12-24, 1990.
13. Glycerol concentrations required for cryoprotection of 50 typical protein crystallization solutions using standard area-detector X-ray images, Garman, E.F. and Mitchell, E.P., J. Appl. Cryst. (1996) 29, 584- 587.
14. The development and application of a method to quantify the quality of cryoprotectant solutions using standard area-detector X-ray images. McFerrin and Snell, J. Appl. Cryst. (2002). 35, 538-545.
15. Rapid crystallization of chemically synthesized hammherhead RNAs using a double screening procedure. William G. Scott, John T. Finch, Richard Grenfell, Jan Fogg, Terry Smith, Michael J. Gait and Aaron Klug. J. Mol. Biol. (1995) 250, 327–332.
16. A Highly Effective 24 Condition Matrix for the Crystallization of Nucleic Acid Fragments. Berger I1, Kang CH, Sinha N, Wolters M, Rich A. Acta Cryst. Section D. (1996) Vol. D52 Part 3, 465-468.
17. Crystallization of intact monoclonal antibodies. Harris, L.J., Skaletsky, E., McPherson, A. Proteins. 1995 Oct;23(2):285-9.
18. Screening and optimization strategies for macromolecular crystal growth. Cudney, R., Patel, S. Weisgraber, K., Newhouse, Y., McPherson, A. Acta Crystallogr D Biol Crystallogr. 1994 Jul 1;50(Pt 4):414-23.
19. Crystallization of biological macromolecules. Alexander McPherson. 1999. Cold Spring Harbor Laboratory Press. Pp. 271-329.