**3.1 Seeding**

The supersaturation level needed for the nucleation and the growth of best diffracting crystals is often different. Nucleation takes place when the supersaturation level is very high, while crystal-growing process appears when the conditions are slightly changing to lower supersaturation level [33]. Therefore, the seeding strategy has been developed as one of the most powerful tools for the optimization of the size and diffractive quality of single crystal. Another benefit is that it reduces the amount of the used protein and saves time needed for the spontaneous nucleation. This is also suitable for the examination of the slightly changing crystallization conditions, such as testing different pH ranges, adding new additives or even new precipitants. Several seeding strategies such as macroseeding, microseeding and/or their modifications can be applied.

In order to perform macroseeding, a single seed crystal has to be selected; its size is not relevant. The solubility of the crystal has to be investigated by testing of different precipitant concentrations to establish the conditions where the crystal is stable. This method is more demanding as multiple washes of the crystal are needed before the crystal is put into the suitable equilibration condition. The washes are

aimed to eliminate additional nuclei. The main disadvantage of the method is that during this handling, the crystal seeds can be damaged and can form only the microcrystals in the new equilibration condition, which are not suitable for the X-ray diffraction analysis [33].

Microseeding can be described as a delivery of microseeds into a new equilibration condition. Before the experiment, the seed stock has to be prepared by crushing crystals using vortex, sonication or seed beads [34]. Once the seed stock is prepared, the dilution series have to be done in order to stabilize the nuclei presented in the seed stock and enable them to form suitable crystals in the new crystallization drop.

Another way of seeding strategy is the streak seeding. The drop containing the protein crystal is opened and the crystal is touched using an animal (cat, horse, rabbit or chinchilla) whisker. The nuclei can be attached to the whisker and transfer into a new crystallization drop. The sitting drop technique is recommended for this experiment, as it does not dry as quickly as the drops in hanging drop technique [33].

Cross seeding is specifically designed for crystallization of related proteins. These related proteins are different recombinantly modified protein variants, homologous proteins, and chemically modified proteins, macromolecular complexes with ligand or even misfolded proteins that are crystallizing difficult [35]. Some of the related proteins do not crystallize in the same conditions; however, sometimes this method can help and save the protein solution and time.

Random microseeding is a modified microseeding strategy using random screening kits. It is reported to create extra additional hits as well as produce better quality protein crystals for diffraction experiments [36].

#### **3.2 Co-crystallization with ligands and crystallization chaperones**

The co-crystallization of macromolecules with ligands have been developed because some proteins crystallize efficiently with ligands considering its structure stabilization role. The use of ligands has benefits not only for improving crystallization but also for obtaining stable and functional protein during purification because they enhance thermal stability of molecules. The study of small ligands stabilization roles for proteins in aqueous solution is still in progress. There exist online libraries of ligands as compounds used for crystallization purposes to prepare the best condition for protein stabilization [37]. The use of ligands for crystallization promotes the possibility of obtaining good-quality protein crystals. Ligands are regularly used to expand the screening capacity of commercially available screens [38]. On the contrary, it can be challenging to crystalize protein-ligand complex and so, the use of stabilizing additives is necessary. The process of co-crystallization can be further affected by temperature used for the complex formation prior its crystallization and by concentration of protein or used ligand [39].

Another "compounds" used for stabilization of proteins are amino acids and their derivatives such as glycine ethyl ester or glycine amide. There is noted considerable benefit of the stabilization role of amino acids for protein purification as well as crystallization [40, 41]. Hence, small molecule additives such as coenzymes, prosthetic groups, inhibitors and small molecules, mentioned in [42], are used for successful crystallization of macromolecules. Usage of these additives helps forming hydrogen bonds and crosslinks between molecules that helps formation of stable crystals with regular lattice order of molecules [38].

Additionally, the antibodies and nanobodies are used to improve protein crystallization. Antibodies such as conventional IgG, Fab-fragments (fragment antigen binding), scFv (single-chain variable fragments) and Camelid hcAb (heavy chain IgG) were used to reinforce protein crystallization, mainly for membrane protein

#### *Advanced Biocrystallogenesis DOI: http://dx.doi.org/10.5772/intechopen.97162*

co-crystallization. Latterly, nanobodies replaced the use of antibodies because production of antibodies is limited and obtained yield tends to be conformationally heterogeneous with limited solubility [43, 44]. On the contrary, nanobodies are extremely stable and soluble antibody single-domain fragments that stabilize unstructured proteins by binding to them, form crystal contacts and thus speed up crystal growth. For crystallization of selected protein, many conformationally different nanobodies can be used to enhance the probability of best quality crystal hits [45].

## **3.3 Protein engineering**

Protein engineering is one of the most used strategies in molecular biology to produce sufficient amounts of soluble protein. Protein biocrystallogenesis is likewise using protein engineering for crystallization of proteins that are hard to crystallize owing to have highly flexible regions, high surface entropy or less common high surface hydrophobicity or proteins are not able to be produced soluble [46]. To conquer these issues some strategies such as designing modified proteins, mutation of surface residues or designing fusion proteins are used [47].

The modifications of proteins include designing only some parts of protein such as conserved domains, and removing flexible parts or internal loops that prevent crystal formation. This can be fulfilled by creation of genetic constructs based on e.g. computation selection of expression clone libraries to find the clones with fragments that would encode the concerned part of the protein [47]. In the literature, several strategies for protein modification have been mentioned such as colony filtration blot technique [48], the open reading frame selection through the ESPRIT automation to find soluble complexes [49] and fluorescence screening based on protein fusion with GFP protein [47, 50].

Next strategy to enhance crystal formation is design of protein mutants by sitespecific mutation or chemical modification. Mutagenesis is ordinarily used to produce soluble protein assuming that wild type is not soluble. The substitution of Lys, Gln or Glu that are frequently located on the molecule surface [51] with Ala that has lower surface entropy is a commonly used strategy [52]. There are plenty of protein structures that were solved mainly due to mutagenesis [47]. In addition, many chemical modifications as reductive methylation of Lys that reduce protein solubility and reductive carboxymethylation or substitution of Cys that on the contrary enhance solubility, stability and have benefits for protein function and oligomerization [52]. Different strategy is to introduce Cys at the surface to strengthen formation of symmetrical dimers that seem to crystallize more easily [52, 53].

Design of fusion proteins or fusion tags is another approach to produce soluble proteins and subsequently solve protein structures. In many cases the fusion partner such as short *Strep*-tag or poly-histidine tag and some larger tags as glutathione S-transferase (GST) and Maltose-binding protein (MBP) are removed before crystallization [54–56]. This is done to remove possible highly flexible regions that can block formation of protein contacts during crystallization. However, it was shown that in two different approaches fusion proteins manage to crystalize target proteins. First, one is based on increasing protein surface area by incorporation of fused protein such as MBP, Green fluorescent protein (GFP), Barnase or T4 lysozyme (T4L) [56]. Second is aimed to merge interacting proteins or protein and peptide with linker, mostly to stabilize one of fused proteins or to support their interaction [56, 57].

## **3.4 Crystallization in living cells**

The formation of native protein crystals in living cells has been detected over the last few decades as a natural process. This approach is ideal for proteins that cannot

be crystallized applying basic crystallization techniques. Recently, X-ray sources have been exceedingly improved, such as microfocus beamlines, X-ray free-electron lasers, or even serial crystallography as a new data collection strategy [58, 59]. These advances in the field of X-ray crystallography enable data collection from smaller crystals and thus enable the development of *in cellulo* crystallization as a powerful advanced crystallization approach.

Recombinant or fusion recombinant proteins can be crystallized in a number of living cells, including plant cells, mammalian cells or insect cells [60]. The bottleneck of this strategy is the detection of the protein crystals inside the cells. Crystals can occur in different compartments of the cell, such as cytosol, endoplasmic reticulum (ER) or peroxisomes [60, 61]. The crystals can be detected using various methods, e.g. bright-field microscopy techniques or transmission electron microscopy (TEM), which is also used for the observation of crystal growth and subsequent optimization. The crystal can be isolated from the cell before the diffraction experiment or directly measured in the cell [60, 62].

#### **3.5 Automation**

The expansion of the crystallographic field and running extensive research led to need for facilitation of performing crystallization experiments. This was achieved by automation that supported faster progress of experiments thanks to development of a variety of robots [63]. There are robots adjusted to establishment of crystallization conditions, arrangement of drops, for carrying out the vapour diffusion experiments specifically hanging drop, microbatch, free-interface diffusion and random micro-seeding seeding methods with sitting drop method as the most preferred method. Most widely used crystallization robots are from the Oryx series (Douglas Instruments Ltd.), the Mosquito (TTP Labtech, Royston, UK) and TOPAZ system (Fluidigm Corp., San Francisco CA, USA) [64]. Automation devices allow crystallization of nanovolumes into an array of plates for 96 or more conditions [28]. Each robot has its own specification, from which can be chosen preferred experimental method suitable for protein of interest.
