rXXXX American Chemical Society A dx.doi.org/10.1021/cg101460z | Cryst. Growth Des. XXXX, XXX, 000–000 ARTICLE pubs.acs.org/crystal Hetero- vs Homogeneous Nucleation of Protein Crystals Discriminated by Supersaturation Published as part of the Crystal Growth & Design virtual special issue on the 13th International Conference on the Crystallization of Biological Macromolecules (ICCBM13). Giovanna Tosi, † Simona Fermani, † Giuseppe Falini,* ,† Jos  e A. Gavira,* ,‡ and Juan Manuel Garcia Ruiz ‡ † Dipartimento di Chimica “G. Ciamician”, Alma Mater Studiorum Universit  a di Bologna, Via Selmi 2, 40126 Bologna, Italy ‡ Laboratorio de Estudios Crystalogr  afico, Instituto Andaluz de Ciencias de la Tierra CSIC/Universidad de Granada, Av. Fuentenueva s/n, 18002 Granada, Spain 1. INTRODUCTION The studies conducted on protein crystallization generally focus on how to maximize the probability of nucleation, 1,2 i.e. nucleation frequency, J (see Garcia-Ruíz, 2003, for details), 3 with this being the critical step. There are two main ways to increase nucleation frequency, J: either by changing the relation between the surface and volume energies, according to GibbsThomp- son expression, or by introducing in the system any material that decreases the energy nucleation barrier, facilitating the crystal formation at lower supersaturation values. 4 In practice, the simple way to change the surface to volume energies ratio is adding preformed seeds into the droplets, from previous experi- ments or from homologous proteins. 5 Heterogeneous nucleation can be driven by subjecting the drops to materials of various nature that are either supporting the drop or added into it. 416 A number of materials that fits the required criteria, i.e. surface ordered at the molecular level, presence of ionizable groups, local concentration of cavities, nano- and mesoscopic structures, have recently been tested by Georgieva and co-worker. 6 Among them, fragments of human hair allowed them to crystallize not only model proteins but also a polysaccaride-specific Fab fragment and the potato protease inhibitor. 6 Thakur et al. have shown that other materials, such as dried seaweed, horse hair, cellulose, and hydroxyapatite, can be used as nucleating substrates, able to increase the crystallization success of several model proteins, especially when a mix of those materials is used. 7 Two kinds of layer silicate, semisynthetic micromica and natural chlorite, have also been tested as heterogeneous nucleant materials. Added as powder to a hanging drop, they were able to aid protein crystal- lization. In particular, micromica powder has shown a strong influence even in the absence of precipitants. 8 Also, microporous synthetic zeolite molecular sieves have turned out to be able to selectively facilitate new crystal forms which did not grow spontaneously from solution, acting as a crystallization catalyst. 9 Zeolite-mediated crystallization also improved the crystal quality in almost all of tested proteins (5 of 6). 9 Poly vinyldiene-fluoride membranes with pore sizes of 400 nm have been successfully used as heterogeneous nucleant surfaces: they promoted the growth of perfect lysozyme crystals, allowing to obtain two distinct polymorphic forms. 10 Nucleation of lysozyme crystals has also been promoted by using glass slide surfaces chemically modified with (3-aminopropyl)triethoxysilane, poly(2-hydro- xyethylmethacrylate), and poly-L-glutamic acid. 11 Moreover, polymer-induced nucleation has been used for the production of high-quality lysozyme crystals and to induce selective nuclea- tion of multiple macromolecule crystal forms. 12 Surfaces Received: November 4, 2010 Revised: March 14, 2011 ABSTRACT: In protein crystallization trials there is always some ambiguity in defining if the nucleation event occurred by a homogeneous or a heterogeneous process. One of the reasons comes from the difficulty to compare experiments with or without a heterogeneous nucleant under the same chemical conditions. The outcome is that it is difficult to figure out the values of supersaturation addressing homo- versus heterogeneous nuclea- tion. Recently it has been shown that the crystallization mushroom allows performance of sitting drop vapor diffusion experiments under identical physicalchemical conditions and that functionalized surfaces are favorable heterogeneous substrates for protein crystallization. With the aim to discriminate between nucleation processes, three model proteins, lysozyme, glucose isomerase, and thaumatin, have been crystallized at different starting concentrations, using positively and negatively charged surfaces, in the crystallization mushroom. The results show that (i) the heterogenenous nucleation does not crucially affect the nucleation process under conditions of high supersaturation and that (ii) the nucleating action of the functionalized surfaces is mainly related to their superficial charge distribution more than to their absolute charge, according to both classical and colloidal nucleation theories.