This project has received funding from the European Union&#39;s Horizon 2020 research and innovation program under the Marie Sk&#322;odowska-Curie grant agreement No 701647.&#160;Many thanks for the assistance and support of beamline scientists at the Swiss Light Source beamline X10SA-PXII.

The authors have no conflicts of interest to disclose.

The method presented shows how to optimize the crystallization of endothiapepsin from large crystals (&#8805;&#160;100 &#181;m longest dimension), grown in sparse-matrix 96-well screens, to micro-crystals, grown in centrifuge tubes (300 &#181;L volume) via batch. The idea behind the protocol is that the steps taken to optimize endothiapepsin could also be used for other proteins. Ultimately, answering the problem of creating large volumes (&#62;100 &#181;L) of micro-crystals (10-20 &#181;m) for serial crystallography experiments at XFELs and synchrotrons.
The protocol divides the task of large volume micro-crystallization into three steps: (1)&#160;Optimizing crystal morphology, (2)&#160;Transitioning to batch, and (3)&#160;Scaling. In Step 1, the range of crystal forms that a protein can create should be explored in vapor diffusion plates. Conditions that give rise to single, box-like crystals that diffract to the required resolution should be the goal. In Step 2, selected conditions can then be transformed from vapor diffusion into batch. Here, the optimization criterion is crystal growth time and to find conditions that give rise to protein crystals within 24 h. A morphogram can also be plotted giving the experimenter an idea of the location of the solubility line and nucleation zone boundaries. This morphogram is of great utility in Step 3, Scaling. The morphogram will give an indication of the whether nucleation alone can increase Xn and drive down Xs. As the volume of the experiment is increased, Xn and Xs can be continually assessed as the key criteria of scaling success.
In the case of Endothiapepsin, Step 1 unearthed what potentially was a previously unknown crystal morphology for endothiapepsin. This morphology had the same spacegroup as those previously reported but, importantly for serial crystallography, a more box-like shape. Single crystals also seemed to grow from single nucleation points, unlike the fans created from other conditions (Figure 4). For the selected condition, Step 2 was already partially satisfied as crystal growth occurred in &#60; 24 h. The morphogram indicated that both a straight or seeded-batch protocol might be successful when scaling in Step 3. Initial scaling in straight batch, created a condition that produced crystals with an Xn and Xs range of 3.6 &#177; 1.2 x 106 crystals&#183;mL-1 and 42 &#177; 4.1 &#181;m, respectively. These crystals, although acceptable for some serial crystallography experiments, were deemed too large. So additional optimizations were performed. The final protocol produced crystals with a concentration and size range of 3.1 x 106 crystals&#183;mL-1 and 15 &#177; 3.9 &#181;m, respectively. This was more than ideal for the planned experiments.
The method focuses on the transformation of &#39;soluble&#39; protein crystals grown in vapor diffusion plates to batch. The reason for this focus is that the vast majority of soluble protein crystals are grown via vapor diffusion26. However, the concepts presented could also be applied to soluble protein crystals grown using other methods, such as micro-batch. The concepts may also be applicable to membrane protein crystals grown in LCP; as this too is a batch crystallization process.
A key aspect of the protocol is the process of transforming the conditions of crystals grown in vapor diffusion plates such that they can be grown in batch. For this transformation, the method uses the criterion proposed by Beale et al. (2019)26. Crystals grown via a batch process, even in vapor diffusion plates, will form rapidly (&#60; 24 h). This criterion is an approximation based on the speed of vapor diffusion drop equilibration, and is most true for PEG-based precipitant conditions. However, crystallization conditions will contain a wide variety of compounds that will influence the equilibration time. The equilibration of salt-based crystallization conditions, e.g. highly concentrated ammonium chloride, can happen in 1-2 days. Therefore, the 24 h criterion may not be true for salt-based conditions. Salt-based conditions also can have more complex phase diagrams26, 30 that may not conform to the archetype presented in this protocol. A reduction in the time criterion for salt-based conditions to 12 or 6 h may be necessary if scaling into larger volumes proves impossible.
Another limitation of this method is its apparent complexity. The protocol that was followed to optimize the micro-crystallization of endothiapepsin actually changed the original condition from the sparse-matrix screen relatively little. The first hit observed in the PACT screen was 0.1 HEPES pH 7.0, 0.2 M MgCl2, and 20% (w/v) PEG 6,000. The final scaled crystallization buffer was 0.1 Tris-HCl pH 7.0, 0.15 M MgCl2, and 40% (w/v) PEG 6,000. It is also very possible that the change in buffer from HEPES to Tris-HCl, and MgCl2 concentration, contributed little to the success of the process. Leaving the increase in PEG 6,000 concentration the sole optimization, and one that could have been achieved quite simply.
This assessment, however, is also too simplistic. It not only discounts the problems encountered during scaling (i.e., the use of seeds and quenching), but also the fact that just because this protein proved straightforward, there is no guarantee the next will also prove to be. The steps advised in the protocol, were devised because optimizing the scaling of protein crystallization volumes can be very protein expensive. Over the seven endothiapepsin scaling trials that are shown, 100 mg of protein were consumed. Admittedly some of these steps were performed to show their consequences in the light of this protocol. Even so, 100 mg of a protein, plus potentially a further 50 mg for protein consumed during an experiment (Table 1), can be a significant investment in either time or money.
Fortunately, it is not clear that this mass of required sample is ubiquitous across all proteins. Endothiapepsin was highly soluble, and therefore required a large protein concentration to reach supersaturation. In others (currently under optimization), supersaturation can be reached at 10 or even 5 mg/mL. Such variables are protein specific and need to be embraced when they appear.
Other limitations of the method include its reliance on complex equipment such as liquid handling robots for screen and plate creation, and imagers to automatically image plates when required. Alternative routines have been offered to limit the need of some of these pieces of equipment, but the protocol will be more time consuming to follow without them. The protocol also suggests testing the diffraction of optimized crystals. For crystallographers without regular access to a synchrotron, these tests could prove challenging. Controls at every step may not be necessary, but these tests are strongly recommended once a hit has been identified, and pre-and post-scaling. Non-diffracting crystals at an XFEL are, unfortunately, not an uncommon occurrence. Given this, it is better to err on the side of caution regarding assumptions about crystal diffraction.
Ultimately, this protocol and results presented here will offer a guide, ideas, and an example to those struggling with producing samples for serial crystallography experiments. Hopefully, as serial crystallography is further developed, the sample demands of the technique will be reduced such that the need for protocols like this will be reduced. However, even in this event, the strategies presented here will still be useful to those wishing to explore the crystallization space of their protein.

Room temperature (RT) macromolecular crystallography is&#160;popular again within the structural biology community. The development of X-ray Free Electron Laser (XFEL) light sources has spurred the development of RT sample delivery approaches1,2,3,4, and these methods have now been applied to synchrotrons5,6,7,8. Not only do RT methods open up the possibility of pump-probe experimental strageties9,10,11,12, but there is also mounting evidence that they promote alternative conformational states within proteins13,14,15,16,17.
However, the principal reason why cryo-methods gained traction over RT approaches in the late 1990s was the slowing of radiation damage by sub-zero crystal temperatures18. Cryo-methods19 began to allow for the collection of a complete dataset from a single protein crystal. Modern RT methods at XFELs and synchrotrons solved the problem of single-crystal radiation damage by the development of rapid (&#62; 100 Hz) crystal delivery strategies1,2,3,4. These methods allow for the collection of a complete dataset from thousands of individually exposed crystals. These RT delivery approaches therefore require the production of large quantities of solutions containing homogenous micro-crystals (&#62; 100 &#181;L of &#60; 50 &#181;m crystals). However, since cryo-methods tend to only require single crystals, methods to create such micro-crystalline slurries are currently not ubiquitous across protein crystallography laboratories.
There are examples in the literature of how to do parts of the micro-crystallization optimization procedure for serial crystallography samples. Here, a distinction should be made between membrane and soluble proteins. Protocols to optimize the growth of micro-membrane protein crystals grown in monoolein (or some other lipid), for lipidic cubic phase (LCP), have been well described20,21,22. However, methods for the micro-crystallization of soluble proteins, including membrane proteins grown in non-LCP&#160;conditions, are generally lacking. Previous studies have focused on specific parts of the process, such as micro-crystal screening23,24, enhancing nucleation24, and scaling using free-interface diffusion25, but not a complete method.
However, a method was recently described26 that attempts to offer a complete protocol. Like many aspects of protein crystallography, it is not new. Many of the ideas proposed were already described by Rayment (2002)27. The method aims to show crystallographers how to perform the conversion from a single, crystal grown using vapor diffusion, to a batch methodology to grow thousands of micro-crystals. The method focuses on vapor diffusion as a common starting point, as 95% of all Protein Data Bank (PDB) depositions come from crystals grown in vapor diffusion plates26. Vapor diffusion is, however, not the ideal method for micro-crystallization26, so a methodology is described to convert vapor diffusion to batch crystallization. Once crystals can be grown in batch, scaling routes to larger volumes become more practicable. Given the vagaries of protein crystallization, the authors would stress that this method is not failsafe. However, the protocol should, at least, provide an insight into the &#39;crystallization space&#39; of a protein.
This method relies on the protein crystallization phase diagram and how an understanding of that diagram can act as a guide during micro-crystallization optimization. A protein phase diagram is commonly depicted as an x/y plot with precipitant and protein concentrations on the x and y axes, respectively (Figure 1A). From the pure water point (bottom left corner - Figure 1A), the concentration of both protein and precipitant increases until the solubility line is reached. The solubility line marks the point of supersaturation (purple line - Figure 1A). When a protein is supersaturated, the solution becomes thermodynamically unstable and will begin separating into two phases: &#39;protein-rich&#39; and a stable saturated solution. This separation can occur anywhere beyond the solubility line and its kinetics are dependent upon the properties of the protein and the components of the solution.
When the protein and precipitant concentrations are too great, the protein will decompose unstably out of solution and result in amorphous precipitate (pink region - Figure 1A). However, ordered phase separation can occur in the nucleation region [see Garcia-Ruiz (2003)28 for detailed description] and crystal nucleants have the propensity to form (green region - Figure 1A). Nucleation and growth removes protein from the solution and moves the drop into the metastable region where growth can continue until the solubility line is reached [see McPherson and Kuznetsov (2014)29 for detailed discussion]. The diagram is, for the vast majority of crystallization conditions, a gross oversimplification30. Regardless of this however, the diagram is still of great utility for micro-crystallographers as the mapping of the diagram allows for the solubility line and the kinetics of nucleation to be determined.
In terms of creating micro-crystals, the two factors during crystallization that need to be optimized are the number of crystals (Xn) and their mean, longest dimension (Xs). Xn will be proportional to the number of nucleation events (n) (Eq. 1).
<img alt="Equation 1" src="/files/ftp_upload/61896/61896fig01v2.jpg" />&#160; &#160; &#160;Eq. 1
Xs is proportional to the concentration of free protein above the solubility line (Ps) divided by Xn (Eq. 2).
<img alt="Equation 2" src="/files/ftp_upload/61896/61896eq02.jpg" />&#160; &#160; &#160;Eq. 2
In a perfect situation, every nucleation event would yield a possible crystal and every one of these crystals, would have equal access to the available protein in solution. Figure 2 is a graphical representation from an ideal scenario of the relationship between Xn and Xs. Practically, the principal control a crystallographer has over Xn and Xs is by influencing the amount of nucleation or by the addition of seed crystals. The micro-crystallographer must judge how to increase Xn such that a suitable crystal concentration and crystal size can both be created.
The majority of crystallization techniques require a &#39;transitionary period&#39; (Figure 1B). For example, in a vapor diffusion experiment, upon mixing the protein and precipitant solutions, the concentrations of each will change as the drop equilibrates with the well solution. One hopes that these changes will gradually transition the drop into the nucleation zone where the propensity for crystallization will increase. As crystals begin to nucleate and grow, the amount of protein in solution will begin to fall, decreasing the probability of further nucleation. The ultimate amount of nucleation will be protein and condition specific, and also dependent upon the depth of penetration of into the nucleation zone. Given the limited nucleation zone penetration of methods that require a transitionary step, the level of nucleation will ultimately be limited to the rate of nucleation at the metastable-nucleation region boundary.
Due to the importance of being able to enhance the level of nucleation for a micro-crystallographer, it is important to move to a batch crystallization methodology. Batch can take greater advantage of the whole nucleation region (Figure 1C). In batch methods, the idea is to mix the protein and precipitant together such that a supersaturated solution is created without the need of any changes in component concentrations. Nucleation should be possible immediately upon mixing. Batch methods therefore allow for the entire nucleation zone to be theoretically reached. Any increase in nucleation kinetics beyond the metastable-nucleation boundary can then be utilized.
If the basal-level of crystal nucleation is not enough to generate a large Xn, micro-seeding methods can be used. In micro-seeding, pre-grown crystals are broken up to create a slurry of crystalline fragments which can act as a scaffold for fresh crystal growth31,32. Micro-seeding has been widely used in serial crystallographic sample preparation as a way to increase Xn without the need of increasing crystal nucleation (Figure 1C).
The transition from vapor diffusion to batch can be visualized on a phase diagram as moving the experimental starting point from either the non-supersaturated or metastable regions to the nucleation zone. This can be done by increasing the protein and/or precipitant concentrations, and/or the ratio of the two within the drop (Figure 1D), and observing which conditions yield crystals appearing rapidly (&#60; 24 h)26. Complete vapor diffusion drop equilibration can take days or weeks33. Therefore, by looking for conditions that show rapidly appearing crystals, batch conditions can be found without having to move to alternative crystallization screening formats such as micro-batch34,35,36,37.
Once the nucleation zone has been found, a batch condition has been found and a morphogram - here, a rough phase diagram - can be created. The morphogram is of great utility when contemplating whether to use a seeded-batch or straight batch protocol. By plotting the Xn as a function of the protein and precipitant concentration, an assessment of the nucleation kinetics can be made26. If Xn remains low across the whole nucleation region, seeded-batch may be required to make Xn large enough to limit crystal growth. This assessment is the first step in the process of scaling to larger volumes (&#62; 100 &#181;L).
This method was designed such that it could be conducted in the majority of crystallization laboratories by using standard vapor diffusion crystallization equipment. Many studies have also been conducted which describe techniques to facilitate many parts of this process, should the equipment be available. These include, but are not limited to, dynamic light scattering (DLS)25,27, non-linear imaging20,24,25, powder diffraction20,24,27, and electron microscopy26 [see Cheng et al. (2020)40 for a nice review].
The aim of this work is to provide a visual demonstration of the method to transition from small volume (&#60; 500 nL) vapor diffusion crystallization to large volume (&#62; 100 &#181;L) batch crystallization. Endothiapepsin from Cryphonectria parasitica has been used as an example system for demonstrating this translation. The type of experiment and sample delivery method that the micro-crystals are required for will influence the ideal Xs output26. For mixing experiments requiring a millisecond time resolution41 or gas-dynamic virtual nozzles42, a final Xs of &#60; 5 &#181;m may be desirable. In this case, the goal was to produce protein crystals that diffract to approximately 1.5 &#197;, for a photon-activated pump-probe experiment, and using a fixed-target delivery approach.
To give an illustration of the sample requirements of such a serial crystallography experiment using endothiapepsin, Table 1 shows the experimental parameters of a hypothetical experiment. The sample information was based upon the protocol described below. Given some conservative estimates on hit rates and data collection requirements, 50 mg is the total sample consumption estimate for the whole experiment.
Figure 3 shows a flow-chart of the complete optimization process from initial small volume vapor diffusion crystallization to large scale batch. For the majority of serial crystallography projects, this protocol will begin at Step 2: &#39;transitioning to batch&#39;, since the target protein will already have been crystallized. However, Step 1 has been included for completeness and to remind readers of its importance. Finding a condition that gives rise to a well diffracting, single, large crystal is the best starting point for micro-crystal optimization. In Step 2, this condition can then be optimized from vapor diffusion to batch, and a morphogram of the nucleation and metastable regions can be plotted. Once this has been done, scaling the batch condition to larger volumes can be performed in Step 3. By the end of the flow-chart, a crystallographer will have created a repeatable, large-volume (&#62; 100 &#181;L), micro-crystallization, batch protocol for endothiapepsin. This method can then be applied to their particular protein of interest.

<table border="1" fo:keep-together.within-page="1" fo:keep-with-next.within-page="always">
	<tbody>
		<tr>
			<td>Swissci 96-well 2-Drop plates</td>
			<td>Molecular Dimensions</td>
			<td>MD11-002</td>
			<td>96-well 2-drop crystallisation plate</td>
		</tr>
		<tr>
			<td>Swissci 96-well 3-Drop plates</td>
			<td>Molecular Dimensions</td>
			<td>MD11-003</td>
			<td>96-well 3-drop crystallisation plate</td>
		</tr>
		<tr>
			<td>mosquito LCP liquid handling robot</td>
			<td>sptlabtech</td>
			<td>mosquito LCP</td>
			<td>Crystallisation robot</td>
		</tr>
		<tr>
			<td>ClearVue Sheets</td>
			<td>Molecular Dimensions</td>
			<td>MD6-015</td>
			<td>96-well crystallization plate seals</td>
		</tr>
		<tr>
			<td>Safe-Tube 1.5 mL</td>
			<td>Eppendorf</td>
			<td>30120086</td>
			<td>1.5 mL centrifuge tubes</td>
		</tr>
		<tr>
			<td>Scaple</td>
			<td>Swan and Morton</td>
			<td>No. 3 scalple and No. 3 handle</td>
			<td>Scalple for cutting open plate seals</td>
		</tr>
		<tr>
			<td>MS 3 Vortex</td>
			<td>IKA</td>
			<td>3319000</td>
			<td>Vortex for mixing solution and making seed stocks</td>
		</tr>
		<tr>
			<td>24-well XRL Plate</td>
			<td>Molecular Dimensions</td>
			<td>MD3-11</td>
			<td>24-well hanging-drop plates</td>
		</tr>
		<tr>
			<td>Tube revolver/rotator</td>
			<td>Thermo Fischer Scientific</td>
			<td>88881001</td>
			<td>Tube revolver for mixing solution during scaling</td>
		</tr>
		<tr>
			<td>Eppendorf Research plus pipettes</td>
			<td>Eppendorf</td>
			<td></td>
			<td>Range of manual pipettes, 0.5-10, 1-20, 10-100, 100-1000 &#181;L</td>
		</tr>
		<tr>
			<td>Eppendorf pipette tips</td>
			<td>Eppendorf</td>
			<td></td>
			<td>Range of tip sizes for manual pipettes</td>
		</tr>
		<tr>
			<td>Suparen 600</td>
			<td>Prochem AG</td>
			<td>Suparen 600</td>
			<td>Endothiapepsin solution</td>
		</tr>
		<tr>
			<td>Sodium Acetate</td>
			<td>Sigma-Aldrich</td>
			<td>241245-1KG</td>
			<td>Sodium Acetate</td>
		</tr>
		<tr>
			<td>Tris</td>
			<td>Merck</td>
			<td>8382T014</td>
			<td>Tris</td>
		</tr>
		<tr>
			<td>Magnesium Chloride</td>
			<td>Sigma-Aldrich</td>
			<td>M2670-1kg</td>
			<td>Magnesium Chloride</td>
		</tr>
		<tr>
			<td>PEG 6,000</td>
			<td>Sigma-Aldrich</td>
			<td>81255-1kg</td>
			<td>PEG 6,000</td>
		</tr>
		<tr>
			<td>Ethelyene glycol</td>
			<td>Sigma-Aldrich</td>
			<td>324558-1L</td>
			<td>Ethelyene glycol for cyro-protecting the crystals</td>
		</tr>
		<tr>
			<td>PACT Premier HT screen</td>
			<td>Molecular Dimensions</td>
			<td>MD1-36</td>
			<td>PACT Premier 96-well crystal screen</td>
		</tr>
		<tr>
			<td>DOW CORNING high vacuum grease</td>
			<td>Molecular Dimensions</td>
			<td>MD6-02</td>
			<td>Grease for sealing 24-well plates</td>
		</tr>
		<tr>
			<td>Hirschmann 22 x 22 mm glaser cover slides</td>
			<td>Hirschmann</td>
			<td>8000104</td>
			<td>Cover slides for sealing 24-well sitting drop plates</td>
		</tr>
		<tr>
			<td>Crystal pins</td>
			<td>PSI</td>
			<td>Manufactured inhouse</td>
			<td>Thin-film supports for micro-crystals.</td>
		</tr>
		<tr>
			<td>1-1.3 mm SiLibeads Type S</td>
			<td>Faust</td>
			<td>6239547</td>
			<td>Glass beads for making mico-seed stocks</td>
		</tr>
		<tr>
			<td>Macbook Pro</td>
			<td>Apple</td>
			<td>Macbook Pro</td>
			<td>Computer for performing data analysis</td>
		</tr>
		<tr>
			<td>CCP4 software suite</td>
			<td>CCP4</td>
			<td></td>
			<td>Diffraction pattern data processing software</td>
		</tr>
		<tr>
			<td>Excel</td>
			<td>Microsoft</td>
			<td>Microsoft Office</td>
			<td>Plotting tool for phase diagram</td>
		</tr>
		<tr>
			<td>Hausser Scientific Bright-Line counting chamber</td>
			<td>Thermo Fischer Scientific</td>
			<td>02-671-51B</td>
			<td>Tool to calculate crystal concentration</td>
		</tr>
		<tr>
			<td>PACT Premier</td>
			<td>Molecular Dimensions</td>
			<td>MD1-29-ECO</td>
			<td>Sparse-matrix crystallization screen</td>
		</tr>
		<tr>
			<td>Rock Imager</td>
			<td>Formulatrix</td>
			<td>Rock Imager</td>
			<td>Temperature controlled crystal plate storage and imager</td>
		</tr>
		<tr>
			<td>Rock MakerWeb</td>
			<td>Formulatrix</td>
			<td>Rock MakerWeb</td>
			<td>Crystal plate creation and image storage stoftware</td>
		</tr>
		<tr>
			<td>Formulator</td>
			<td>Formulatrix</td>
			<td>Formulator</td>
			<td>96-well crystal screen creation liquid handling robot</td>
		</tr>
		<tr>
			<td>Leica MZ16 Microscope</td>
			<td>Leica</td>
			<td>Leica MZ16</td>
			<td>Light microscope</td>
		</tr>
		<tr>
			<td>LAS V4.6</td>
			<td>Leica</td>
			<td>LAS V4.6</td>
			<td>Software for Leica microscopes</td>
		</tr>
		<tr>
			<td>Spectra/Por 3.5 kDa dialysis tubing</td>
			<td>Spectrumlabs</td>
			<td>Spectra/Por 3 Dialysis Membrane</td>
			<td>3.5 kDa dialysis membrane</td>
		</tr>
		<tr>
			<td>Dialysis tubing closures</td>
			<td>Spectrumlabs</td>
			<td>Spectra/Por 3 Duniversal Closures</td>
			<td>Clips to seal the dialysis tubing ends</td>
		</tr>
		<tr>
			<td>Amicon 10 kDa centrifugal concentrator</td>
			<td>Merck-Millipore</td>
			<td>Amicon Ultra-15 10 kDa centrifugal concentrator</td>
			<td>10 kDa centrifugal filter</td>
		</tr>
		<tr>
			<td>5810 R swing bucket centrifuge</td>
			<td>Eppendorf</td>
			<td>5810 R Centrifuge</td>
			<td>Swing bucket centrifuge</td>
		</tr>
	</tbody>
</table>

optimizing the growth of endothiapepsin crystals for serial crystallography experiments

Here, a protocol is presented to facilitate the creation of large volumes (&#62; 100 &#181;L) of micro-crystalline slurries suitable for serial crystallography experiments at both synchrotrons and XFELs. The method is based upon an understanding of the protein crystal phase diagram, and how that knowledge can be utilized. The method is divided into three stages: (1) optimizing crystal morphology, (2) transitioning to batch, and (3) scaling. Stage 1 involves finding well diffracting, single crystals, hopefully but not necessarily, presenting in a cube-like morphology. In Stage 2, the Stage 1 condition is optimized by crystal growth time. This strategy can transform crystals grown by vapor diffusion to batch. Once crystal growth can occur within approximately 24 h, a morphogram of the protein and precipitant mixture can be plotted and used as the basis for a scaling strategy (Stage 3). When crystals can be grown in batch, scaling can be attempted, and the crystal size and concentration optimized as the volume is increased. Endothiapepsin has been used as a demonstration protein for this protocol. Some of the decisions presented are specific to endothiapepsin. However, it is hoped that the way they have been applied will inspire a way of thinking about this procedure that others can adapt to their own projects.

Watch this Scientific Journal Video about Optimizing the Growth of Endothiapepsin Crystals for Serial Crystallography Experiments at JoVE.com

Optimizing the Growth of Endothiapepsin Crystals for Serial Crystallography Experiments

The aim of this article is to give the viewer a solid understanding of how to transform their small-volume, vapor-diffusion protocol, for growing large, single protein crystals, into a large-volume batch micro-crystallization method for serial crystallography.

Here, a protocol is presented to facilitate the creation of large volumes (&#62; 100 &#181;L) of micro-crystalline slurries suitable for serial ...

optimizing-growth-endothiapepsin-crystals-for-serial-crystallography

Research

JoVE Journal

Biochemistry

2.1K Views.  Paul Scherrer Institut. The aim of this article is to give the viewer a solid understanding of how to transform their small-volume, vapor-diffusion protocol, for growing large, single protein crystals, into a large-volume batch micro-crystallization method for serial crystallography.

Video: Optimizing the Growth of Endothiapepsin Crystals for Serial Crystallography Experiments

Preparation and Delivery of Protein Microcrystals in Lipidic Cubic Phase for Serial Femtosecond Crystallography

Membrane proteins (MPs) are essential components of cellular membranes and primary drug targets. Rational drug design relies on precise structural information, typically obtained by crystallography; however MPs are difficult to crystallize. Recent progress in MP structural determination has benefited greatly from the development of lipidic cubic phase (LCP) crystallization methods, which typically yield well-diffracting, but often small crystals that suffer from radiation damage during traditional crystallographic data collection at synchrotron sources. The development of new-generation X-ray free-electron laser (XFEL) sources that produce extremely bright femtosecond pulses has enabled room temperature data collection from microcrystals with no or negligible radiation damage. Our recent efforts in combining LCP technology with serial femtosecond crystallography (LCP-SFX) have resulted in high-resolution structures of several human G protein-coupled receptors, which represent a notoriously difficult target for structure determination. In the LCP-SFX technique, LCP is recruited as a matrix for both growth and delivery of MP microcrystals to the intersection of the injector stream with an XFEL beam for crystallographic data collection. It has been demonstrated that LCP-SFX can substantially improve the diffraction resolution when only sub-10 &#181;m crystals are available, or when the use of smaller crystals at room temperature can overcome various problems associated with larger cryocooled crystals, such as accumulation of defects, high mosaicity and cryocooling artifacts. Future advancements in X-ray sources and detector technologies should make serial crystallography highly attractive and practicable for implementation not only at XFELs, but also at more accessible synchrotron beamlines. Here we present detailed visual protocols for the preparation, characterization and delivery of microcrystals in LCP for serial crystallography experiments. These protocols include methods for conducting crystallization experiments in syringes, detecting and characterizing the crystal samples, optimizing crystal density, loading microcrystal laden LCP into the injector device and delivering the sample to the beam for data collection.

We describe procedures for the preparation and delivery of membrane protein microcrystals in lipidic cubic phase for serial crystallography at X-ray free-electron lasers and synchrotron sources. These protocols can also be applied for incorporation and delivery of soluble protein microcrystals, leading to substantially reduced sample consumption compared to liquid injection.

Membrane proteins (MPs) are essential components of cellular membranes and primary drug targets. Rational drug design relies on precise structural ...

Automated Acoustic Dispensing for the Serial Dilution of Peptide Agonists in Potency Determination Assays

As with small molecule drug discovery, screening for peptide agonists requires the serial dilution of peptides to produce concentration-response curves. Screening peptides affords an additional layer of complexity as conventional tip-based sample handling methods expose peptides to a large surface area of plasticware, providing an increased opportunity for peptide loss via adsorption. Preventing excessive exposure to plasticware reduces peptide loss via adherence to plastics and thus minimizes inaccuracies in potency prediction, and we have previously described the benefits of non-contact acoustic dispensing for in vitro high-throughput screening of peptide agonists1. Here we discuss a fully integrated automation solution for non-contact acoustic preparation of peptide serial dilutions in microtiter plates utilizing the example of screening for peptide agonists at the mouse glucagon-like peptide-1 receptor (GLP-1R). Our methods allow for high-throughput cell-based assays to screen for agonists and are easily scalable to support increased sample throughput, or to allow for increased numbers of assay plate copies (e.g., for a panel of more target cell lines).

Peptide adsorption to plasticware during traditional tip-based serial dilutions can significantly impact potency determination and confound the understanding of structure-activity relationships used for lead identification and lead optimization phases of drug discovery. Here methods for automated acoustic non-contact serial dilution of peptide samples are described.

As with small molecule drug discovery, screening for peptide agonists requires the serial dilution of peptides to produce concentration-response curves. ...

Microcrystallography of Protein Crystals and In Cellulo Diffraction

The advent of high-quality microfocus beamlines at many synchrotron facilities has permitted the routine analysis of crystals smaller than 10 &#181;m in their largest dimension, which used to represent a challenge. We present two alternative workflows for the structure determination of protein microcrystals by X-ray crystallography with a particular focus on crystals grown in vivo. The microcrystals are either extracted from cells by sonication and purified by differential centrifugation, or analyzed in cellulo after cell sorting by flow cytometry of crystal-containing cells. Optionally, purified crystals or crystal-containing cells are soaked in heavy atom solutions for experimental phasing. These samples are then prepared for diffraction experiments in a similar way by application onto a micromesh support and flash cooling in liquid nitrogen. We briefly describe and compare serial diffraction experiments of isolated microcrystals and crystal-containing cells using a microfocus synchrotron beamline to produce datasets suitable for phasing, model building and refinement.
These workflows are exemplified with crystals of the Bombyx mori cypovirus 1 (BmCPV1) polyhedrin produced by infection of insect cells with a recombinant baculovirus. In this case study, in cellulo analysis is more efficient than analysis of purified crystals and yields a structure in ~8 days from expression to refinement.

Microcrystallography of Protein Crystals and In Cellulo Diffraction

A protocol is presented for X-ray crystallography using protein microcrystals. Two examples analyzing in vivo-grown microcrystals after purification or in cellulo are compared.

The advent of high-quality microfocus beamlines at many synchrotron facilities has permitted the routine analysis of crystals smaller than 10 &#181;m in ...

Growing Protein Crystals with Distinct Dimensions Using Automated Crystallization Coupled with In Situ Dynamic Light Scattering

The automated crystallization device is a patented technique1 especially developed for monitoring protein crystallization experiments with the aim to precisely maneuver the nucleation and crystal growth towards desired sizes of protein crystals. The controlled crystallization is based on sample investigation with in situ Dynamic Light Scattering (DLS) while all visual changes in the droplet are monitored online with the help of a microscope coupled to a CCD camera, thus enabling a full investigation of the protein droplet during all stages of crystallization. The use of in situ DLS measurements throughout the entire experiment allows a precise identification of the highly supersaturated protein solution transitioning to a new phase &#8211; the formation of crystal nuclei. By identifying the protein nucleation stage, the crystallization can be optimized from large protein crystals to the production of protein microcrystals. The experimental protocol shows an interactive crystallization approach based on precise automated steps such as precipitant addition, water evaporation for inducing high supersaturation, and sample dilution for slowing induced homogeneous nucleation or reversing phase transitions.

Growing Protein Crystals with Distinct Dimensions Using Automated Crystallization Coupled with In Situ Dynamic Light Scattering

Here we present a protocol for controlled production of protein microcrystals. The process uses an automated device allowing controlled manipulation of several crystallization parameters. The protein crystallization is carried-out by controlled and automated addition of crystallization solutions while monitoring and investigating the radius distribution of particles in the crystallization droplet.

The automated crystallization device is a patented technique1 especially developed for monitoring protein crystallization experiments with the aim to ...

Characterization of Glycoproteins with the Immunoglobulin Fold by X-Ray Crystallography and Biophysical Techniques

Glycoproteins on the surface of cells play critical roles in cellular function, including signalling, adhesion and transport. On leukocytes, several of these glycoproteins possess immunoglobulin (Ig) folds and are central to immune recognition and regulation. Here, we present a platform for the design, expression and biophysical characterization of the extracellular domain of human B cell receptor CD22. We propose that these approaches are broadly applicable to the characterization of mammalian glycoprotein ectodomains containing Ig domains. Two suspension human embryonic kidney (HEK) cell lines, HEK293F and HEK293S, are used to express glycoproteins harbouring complex and high-mannose glycans, respectively. These recombinant glycoproteins with different glycoforms allow investigating the effect of glycan size and composition on ligand binding. We discuss protocols for studying the kinetics and thermodynamics of glycoprotein binding to biologically relevant ligands and therapeutic antibody candidates. Recombinant glycoproteins produced in HEK293S cells are amenable to crystallization due to glycan homogeneity, reduced flexibility and susceptibility to endoglycosidase H treatment. We present methods for soaking glycoprotein crystals with heavy atoms and small molecules for phase determination and analysis of ligand binding, respectively. The experimental protocols discussed here hold promise for the characterization of mammalian glycoproteins to give insight into their function and investigate the mechanism of action of therapeutics.

We present approaches for the biophysical and structural characterization of glycoproteins with the immunoglobulin fold by biolayer interferometry, isothermal titration calorimetry, and X-ray crystallography.

Glycoproteins on the surface of cells play critical roles in cellular function, including signalling, adhesion and transport. On leukocytes, several of ...

An All-in-one Sample Holder for Macromolecular X-ray Crystallography with Minimal Background Scattering

Macromolecular X-ray crystallography (MX) is the most prominent method to obtain high-resolution three-dimensional knowledge of biological macromolecules. A prerequisite for the method is that highly ordered crystalline specimen need to be grown from the macromolecule to be studied, which then need to be prepared for the diffraction experiment. This preparation procedure typically involves removal of the crystal from the solution, in which it was grown, soaking of the crystal in ligand solution or cryo-protectant solution and then immobilizing the crystal on a mount suitable for the experiment. A serious problem for this procedure is that macromolecular crystals are often mechanically unstable and rather fragile. Consequently, the handling of such fragile crystals can easily become a bottleneck in a structure determination attempt. Any mechanical force applied to such delicate crystals may disturb the regular packing of the molecules and may lead to a loss of diffraction power of the crystals. Here, we present a novel all-in-one sample holder, which has been developed in order to minimize the handling steps of crystals and hence to maximize the success rate of the structure determination experiment. The sample holder supports the setup of crystal drops by replacing the commonly used microscope cover slips. Further, it allows in-place crystal manipulation such as ligand soaking, cryo-protection and complex formation without any opening of the crystallization cavity and without crystal handling. Finally, the sample holder has been designed in order to enable the collection of in situ X-ray diffraction data at both, ambient and cryogenic temperature. By using this sample holder, the chances to damage the crystal on its way from crystallization to diffraction data collection are considerably reduced since direct crystal handling is no longer required.

A novel sample holder for macromolecular X-ray crystallography along with a suitable handling protocol is presented. The system allows crystal growth, crystal soaking and in situ diffraction data collection at both, ambient and cryogenic temperature without the need of any crystal manipulation or mounting.

Macromolecular X-ray crystallography (MX) is the most prominent method to obtain high-resolution three-dimensional knowledge of biological macromolecules. ...

X-Ray Crystallography to Study the Oligomeric State Transition of the Thermotoga maritima M42 Aminopeptidase TmPep1050

The M42 aminopeptidases form functionally active complexes made of 12 subunits. Their assembly process appears to be regulated by their metal ion cofactors triggering a dimer-dodecamer transition. Upon metal ion binding, several structural modifications occur in the active site and at the interaction interface, shaping dimers to promote the self-assembly. To observe such modifications, stable oligomers must be isolated prior to structural study. Reported here is a method that allows the purification of stable dodecamers and dimers of TmPep1050, an M42 aminopeptidase of T. maritima, and their structure determination by X-ray crystallography. Dimers were prepared from dodecamers by removing metal ions with a chelating agent. Without their&#160;cofactor, dodecamers became less stable and were fully dissociated upon heating. The oligomeric structures were solved by the straightforward molecular replacement approach. To illustrate the methodology, the structure of a TmPep1050 variant, totally impaired in metal ion binding, is presented showing no further breakdown of dimers to monomers.

X-Ray Crystallography to Study the Oligomeric State Transition of the Thermotoga maritima M42 Aminopeptidase TmPep1050

This protocol has been developed to study the dimer-dodecamer transition of TmPep1050, an M42 aminopeptidase, at the structural level. It is a straightforward pipeline starting from protein purification to X-ray data processing. Crystallogenesis, data set indexation, and molecular replacement have been emphasized through a case of study, TmPep1050H60A H307A variant.

The M42 aminopeptidases form functionally active complexes made of 12 subunits. Their assembly process appears to be regulated by their metal ion ...

Derivatization of Protein Crystals with I3C using Random Microseed Matrix Screening

Protein structure elucidation using X-ray crystallography requires both high quality diffracting crystals and computational solution of the diffraction phase problem. Novel structures that lack a suitable homology model are often derivatized with heavy atoms to provide experimental phase information. The presented protocol efficiently generates derivatized protein crystals by combining random microseeding matrix screening with derivatization with a heavy atom molecule I3C (5-amino-2,4,6-triiodoisophthalic acid). By incorporating I3C into the crystal lattice, the diffraction phase problem can be efficiently solved using single wavelength anomalous dispersion (SAD) phasing. The equilateral triangle arrangement of iodine atoms in I3C allows for rapid validation of a correct anomalous substructure. This protocol will be useful to structural biologists who solve macromolecular structures using crystallography-based techniques with interest in experimental phasing.

This article presents a method to generate protein crystals derivatized with I3C (5-amino-2,4,6-triiodoisophthalic acid) using microseeding to generate new crystallization conditions in sparse matrix screens. The trays can be set up using liquid dispensing robots or by hand.

Protein structure elucidation using X-ray crystallography requires both high quality diffracting crystals and computational solution of the diffraction ...

Crystallization and Structural Determination of an Enzyme:Substrate Complex by Serial Crystallography in a Versatile Microfluidic Chip

The preparation of well diffracting crystals and their handling before their X-ray analysis are two critical steps of biocrystallographic studies. We describe a versatile microfluidic chip that enables the production of crystals by the efficient method of counter-diffusion. The convection-free environment provided by the microfluidic channels is ideal for crystal growth and useful to diffuse a substrate into the active site of the crystalline enzyme. Here we applied this approach to the CCA-adding enzyme of the psychrophilic bacterium Planococcus halocryophilus in the presented example. After crystallization and substrate diffusion/soaking, the crystal structure of the enzyme:substrate complex was determined at room temperature by serial crystallography and the analysis of multiple crystals directly inside the chip. The whole procedure preserves the genuine diffraction properties of the samples because it requires no crystal handling.

A versatile microfluidic device is described that enables the crystallization of an enzyme using the counter-diffusion method, the introduction of a substrate in the crystals by soaking, and the 3D structure determination of the enzyme:substrate complex by a serial analysis of crystals inside the chip at room temperature.

The preparation of well diffracting crystals and their handling before their X-ray analysis are two critical steps of biocrystallographic studies. We ...

Workflow and Tools for Crystallographic Fragment Screening at the Helmholtz-Zentrum Berlin

Fragment screening is a technique that helps to identify promising starting points for ligand design. Given that crystals of the target protein are available and display reproducibly high-resolution X-ray diffraction properties, crystallography is among the most preferred methods for fragment screening because of its sensitivity. Additionally, it is the only method providing detailed 3D information of the binding mode of the fragment, which is vital for subsequent rational compound evolution. The routine use of the method depends on the availability of suitable fragment libraries, dedicated means to handle large numbers of samples, state-of-the-art synchrotron beamlines for fast diffraction measurements and largely automated solutions for the analysis of the results.
Here, the complete practical workflow and the included tools on how to conduct crystallographic fragment screening (CFS) at the Helmholtz-Zentrum Berlin (HZB) are presented. Preceding this workflow, crystal soaking conditions as well as data collection strategies are optimized for reproducible crystallographic experiments. Then, typically in a one to two-day procedure, a 96-membered CFS-focused library provided as dried ready-to-use plates is employed to soak 192 crystals, which are then flash-cooled individually. The final diffraction experiments can be performed within one day at the robot-mounting supported beamlines BL14.1 and BL14.2 at the BESSY&#160; II electron storage ring operated by the HZB in Berlin-Adlershof (Germany). Processing of the crystallographic data, refinement of the protein structures, and hit identification is fast and largely automated using specialized software pipelines on dedicated servers, requiring little user input.
Using the CFS workflow at the HZB enables routine screening experiments. It increases the chances for successful identification of fragment hits as starting points to develop more potent binders, useful for pharmacological or biochemical applications.

Crystallographic fragment screening at the Helmholtz-Zentrum Berlin is performed using a workflow with dedicated compound libraries, crystal handling tools, fast data collection facilities&#160;and largely automated data analysis. The presented protocol intends to maximize the output of such experiments to provide promising starting points for downstream structure-based ligand design.

Fragment screening is a technique that helps to identify promising starting points for ligand design. Given that crystals of the target protein are ...