1.1 Methods

There are several steps in a successful flash-cooling experiment:

1.1.1 Choice of a suitable crystal:

In general, as for almost every crystallographic experiment, crystals should be of high quality. The experimenter will learn how to define quality for each crystal system from practical experience. Low quality crystals can be used for preliminary tests. For some systems, small crystals will freeze better than large ones and smooth ones better than rough. The absence of cracks is important as well, as cracks usually widen upon freezing.

1.1.2 Choice of a cryo-protectant:

When searching for a cryo-protectant (Petsko, 1975; Sutton, 1991), one should first consider the mother liquor composition. For instance, if the crystal was grown using 2-methyl-2,4-pentanediol (MPD) as the precipitant, it is most likely ready for freezing without any additional cryo-protectant. Thus the original mother liquor of any kind should be the first to be tested (see below). On the other hand, crystals that grow in salt should be washed in a solution containing a cryo-protectant, or may need to go through a complete mother liquor exchange with a cryo-protected solution in order to be cooled to cryo-temperatures. In general the best choice of cryo-protectant will be one that most closely resembles the composition of the mother liquor unless that is mostly salt. If the crystal grows in PEG, then ethylene glycol, the "monomer" of PEG, would be a good first choice. One may then try glycerol and even low molecular weight PEG. If the crystal grows in low MPD concentration, a higher MPD concentration would be the first option and so on. The nature of the search is trial and error. Some crystals will be more sensitive to one protectant than to another. A good place to start the search is with one of the chemicals suggested in the following list: Other methods of "cryo-protection" have been described in Petsko, 1975; Ray et al., 1991; Wierenga et al., 1992.

1.1.3 Solvent exchange:

The most critical portion of the solution in flash-freezing is the part surrounding the crystal but some cases call for treatment of the interior part of the solution as well. The basic guidelines are similar to all protein crystal washes and soaks, but the details will depend on the characteristics of each case. For most cases a brief dip in a solution of cryo-protectant (wash of few seconds) will suffice and should be tried first. In other cases a long soak will be needed. Some crystals will not survive a drastic change in solvent content, and will have to be taken through a gradual change of the solvent, either by serial soaks or dialysis. Crystals to which a cryo-protectant cannot be introduced in their native form might require cross-linking with glutaraldehyde prior to the introduction of the cryo-protectant. It is also possible to move the crystal physically from its aqueous environment into oil, which seals the crystal and minimizes the external excess of solvent (Hope, 1988; Hope et al., 1989).

Before using any crystals, the cryo-protectant should be tested and the minimal concentration for vitrification must be determined. This is done following these steps: Construct a loop, approximately 1 x 1mm, of very thin fishing line (nylon). This loop will be used only to test the suitability of the mother liquor for flash-cooling, not for data collection. Starting with the original mother liquor, dip the loop into the solution and then plunge it rapidly into liquid nitrogen and inspect the mother liquor drop while still immersed in the liquid nitrogen. If the frozen drop in the loop turns opaque (milky color) while still immersed in the liquid nitrogen, the composition of the mother liquor is not adequate for flash-cooling. Incrementally increase the concentration of the chosen cryo-protectant and repeat until the frozen drop in the loop stays clear while still immersed in the liquid nitrogen. It should be noted that this test usually overestimates the minimal adequate concentration of cryo-protectant but provides a good starting point.

Once the adequate concentration of the additive is established, crystals are tested next. Briefly wash an expendable crystal in that solution and test its diffraction in the X-ray beam. If a brief wash seems to fail even when the cryo-protectant concentration is set higher, soak an expendable crystal and look for morphological changes. If none appear, test the diffraction of soaked crystal. Soaking should be done in the environment the crystal was grown, i.e. temperature and atmosphere. The length of soak depends on the temperature and the viscosity of the mother liquor; at room temperature 4-6 hours, at 4deg.c the crystal should be left for at least twice as long as for room temperature. If possible, compare with the mosaicity recorded for capillary-mount crystals. If the first soaked crystal shows visible degradation in the cryo-protected solution under the microscope, try to move the crystal from one solution to another with gradually higher concentration of the cryo-protectant up to the optimal concentration (serial soaks), or by means of dialysis (dialysis buttons for example). If the crystal seems to reject any amount of a specific cryo-protectant, start over with a different one. Once the crystal/cryo-protectant passes the visible test but the diffraction shows an unacceptable increase in mosaicity, try increasing the protectant concentration. If no improvement is achieved, start over with another protectant. The final determination of a successful cryo-protection of a crystal will be made when the diffraction is assessed. Again, if no other way to introduce a cryo-protectant succeeds, the crystal may require cross-linking. The process of testing a cryo-protectant is depicted in the flow-chart below.

Figure 1.2: Flow chart for cryo-Protectant Search

1.1.4 Choice of the mounting assembly:

The mounting assembly has three parts: the support, the pin and the base. The term "support" refers to the interface between the crystal and the pin. The pin is attached to the base and the base is held on the goniometer head. In the design of the support, on which the crystal is suspended, one should consider the mechanical stability of the crystal and the requirement to minimize the size of the drop in which the crystal is suspended. When conditions allow, the support can be as simple as a very thin glass rod (Dewan and Tilton, 1987), thinner than the crystal itself. The crystal can be taken on the tip of the rod. A touch of vacuum grease applied to the rod's tip will assist in "fishing" out the crystal. This type of support calls for a tough crystal but when successful provides a "dry" mount and, in cases where the crystal extends out of the rod, allows exposure of the crystal to X-rays without any of the support being in the beam. In the other extreme, a highly sensitive crystal which is either very thin or soft will not survive mounting on a rod and will break by mere touch or withdrawal from the solvent surface. In these cases, a flat glass spatula can be built, and for the most extreme cases, a "sandwich" spatula can be built to prevent the surface tension of the drop from breaking the crystal. This method, although very challenging, is still the only way that allows the collection of diffraction data from crystals of intact ribosomes and ribosomal particles (Hope et al., 1989).

Figure 1.3: The complete mounting assembly.
This chapter will deal primarily with the most popular and successful method of crystal support for flash-cooling - the loop (Teng, 1990). The basic knowledge was acquired in different labs, and in particular, the labs of Dr. P. Sigler of Yale University and Dr. S. Harrison of Harvard University (especially acknowledging Dr. Greg van-Düyne and Dr David Rodgers for sharing their practical experience). Loops are relatively easy to construct, can be made to fit a specific crystal size, can greatly improve the dexterity of handling and mounting the crystal, and are suitable to deal with a wide range of crystal cases. A loop can be made from any thin flexible fiber, but there are several considerations to be noted. The addition to the diffraction pattern by loop material must be minimal. Materials such as some polymers that add rings to the diffraction and metallic loops which add distinct reflections should be avoided. When searching for loop material one should test it by exposing the bare loop to the X-ray beam and observing the result. Loops can be made from fibers of dental floss, nylon, rayon or silk. The glue with which the fiber's twists are affixed should also add as little to the diffraction as possible and should be tested at the same time as the fiber. If the crystal is mechanically strong, a loop smaller than the crystal will substitute for a glass rod. In other more common cases the loop should have dimensions that will accommodate the crystal but minimize the amount of liquid carried with it in the mounting stage.

The support, as defined above, is suspended on a pin in the X-ray beam. It is glued to the tip of the pin, which in turn can be permanently attached to the base. One popular option is depicted in fig.1.3.

This mounting assembly is made of a steel base and a pin of G22 hypodermic tubing (Small Parts Inc., (305) 557-8222, #HTX-22-36 or any syringe needle such as TERUMOreg. 22GX). The pin is inserted through a hole drilled in the base and affixed in place by soldering, with glue or by friction. Since the base is made of ferromagnetic metal, it is held on the goniometer head with a piece of adhesive magnetic strip (fig. 1.4). This assembly is easily put on the goniometer head with one hand, which facilitates rapid transfers to the camera or other data collection device.

Figure 1.4: Mounting assembly attached to the goniometer head

1.1.5 Nylon loop construction:

It is highly recommended to use polyamide nylon 66, 0.025 mm in diameter, such as is used in eye surgery (Goodfellow Corp.(USA), (800) 821-2870, #AM325710/1). This fiber appears to make a negligible contribution to the background of the diffraction pattern and is durable and flexible. The recommended glue is a cyanoacrylate adhesive, also known as "crazy glue".

  1. Build a loop-making tool, such as the one shown in figure 1.5, from transparent material such as perspex, that can be used under a light microscope (Also see the ultimate loop making machine in fig. 1.7).

    Figure 1.5: Loop Making Machine
  2. Tack on a short segment of stainless steel wire to a wooden dowel of approximately 2mm in diameter and bend the tip to a hook (fig. 1.6). The width of the wire should correlate with the desired width of the final loop.

  3. Seal one end of a common disposable glass micropipette with some clay or wax, or use any glass rod approximately 2mm in diameter and 5cm in length. Cut about one centimeter of fiber. Apply a touch of crazy glue to the sealed side, fix both ends of the fiber and let harden.

  4. Place the hook and the glass capillary face to face in the tool as shown in fig. 1.6

Now when all is in place the loop is formed, follow the steps depicted in fig. 1.6:

Figure 1.6: Loop Construction
  1. Hook the loop and start twisting (fig. 1.6a, 1.6b).

  2. Apply glue all along the twists and as close as possible to the loop. Avoid fixing the loop to the hook (fig. 1.6c).

  3. Let the glue harden and snip the loop-stem far from the loop (fig. 1.6d) and remove carefully from the hook.

Figure 1.7: This multiple-loop machine, invented by Paul Pepin at Dr. P. Sigler's lab and commissioned from the Gibbs machine-shop at Yale University, makes 15 loops of various sizes at the same time.
To complete the mounting assembly construction, the loop is glued to the tip of the pin (in the opening of the pin if a hypodermic tubing is used). This glue must be different from the glue used to affix the twists of the loop or it will dissolve and the loop will untwist. The recommended glue to use is crazy glue gel which has a different solvent than the liquid variety.

1.1.6 Materials for Loop Construction

Hypodermic tubing: Small Parts Inc., (305) 557-8222 #HTX-22-36
Loop fiber: Goodfellow Corp.(USA), (800) 821-2870 #AM325710/1
Cryo-vials: Corning, 2 ml disposable cryogenic vials #25724-2
Cryo-storage: Nalgene, aluminum cryo-canes #5015-0001
Nalgene, PVC cryo-sleeves #5016-0001

1.1.7 Flash-cooling technique and transfer to the camera:

There are three basic methods of flash-cooling:
  1. Rapid exposure to a low temperature gaseous nitrogen stream at the camera or other data collection device.
  2. Plunging the suspended crystal into liquid nitrogen prior to transfer to the cold gaseous nitrogen stream.
  3. Plunging the suspended crystal into liquid propane prior to transfer to the cold gaseous nitrogen stream.
The process must be rapid in all methods in order to ensure the amorphous solidification of the solvent in and around the crystal and to allow isotropic freezing of the entire loop contents. It is highly recommended to run a mock experiment with an empty loop for any camera configuration and flash-cooling/transfer technique. The loop can hold a drop of the solution that the crystal is later mounted from. If, at the end of the transfer, the drop in the loop is anything but "crystal clear", something is faulty in the transfer process.

Method 1: Exposure to cryo-conditions in the gaseous-liquid interface is less efficient than in the liquid-liquid interface, but may be sufficient in many cases. It is advantageous in its simplicity and the fact that further transfers are unnecessary. Lengthy exposure of the unfrozen crystal to air should be avoided, or the crystal will degrade or dry out. Position the low temperature device nozzle in its final position. Block the cold gas stream with a small piece of cardboard. Mount the crystal and place the mounting assembly on the goniometer. Rapidly remove the cardboard.

Method 2: This method requires the spindle is either pointing down or has the ability to be swung to that position. Swing the spindle to a position between 12 and 2 o'clock (unless utilizing a fixed angle spindle pointing down). Position the low temperature device nozzle slightly further away from its final position to allow room for the transfer vial. Mount the crystal and plunge it to a cryo-vial containing liquid nitrogen. Bring the cryo-vial containing the mounting assembly in liquid nitrogen under the goniometer head and place the assembly on the magnetic strip. Remove the cryo-vial rapidly and adjust the position of the nozzle.

Method 3: This method is advantageous over method 2 by providing more comfortable storage and transfer possibilities but requires use of flammable materials. Assuming the use of the mounting assembly suggested above, the appropriate vial to use is the standard 2 ml cryo-vial by Corning (2 ml disposable cryogenic vials, #25724-2). Propane should be of high quality to assure solidification at liquid nitrogen temperature. The propane can be withdrawn in the liquid form or liquefied in a separate vessel, according to the available propane source. A small vessel suitable to contain liquid nitrogen should be fitted with an insert that will hold a few cryo-vials in an upright position. The rim of the vials should be about 1.5 cm bellow the rim of the vessel.

Figure 1.8: Loop in a Cryo-Vial
The frozen propane vials can be stored under liquid nitrogen in cryo-storage aluminum canes (Nalgene, aluminum cryo-canes, #5015-0001) inserted in plastic sleeves (Nalgene, PVC cryo-sleeves, #5016-0001).
Press here to go to the next section.