Vapour Diffusion Under Oil

I sought a crystallization method which maximized the number of drops I could set up in a given time while minimizing cost. I have found that vapour diffusion under oil (VDUO) is by far the fastest, cheapest and most successful technique to maximize crystallization hits for those who do not have access to a robotic crystallization system. The typical VDUO experiment is over 10 times cheaper and can be setup 4.5 times faster than your typical hanging drop vapour diffusion experiment (Limbro plates/cover slips). In addition, using VDUO setups has further benefits since you always have direct access to the drops. I experimented with various 96-well plates, oils, pippette tips, and how the VDUO experiment was assembled until I arrived at a cheap, quick, stable, and successful crystallization technique.

How to Setup VDUO Experiments

(Prepare the Screen) Transfer 400 ul of each crystallization condition from the screens into the related 96 well polypropylene assay block well. For example, Hampton I and II can fit in one 96-well assay block. Cover with a Hampton Robolid and store at room temperature. Other grid screens and screens around crystallization hits can also be prepared by pippetting stock solutions into the various wells of the assay block, covering with the Robolid, and mixing by inverting the assay block several times.

(Prepare the Plate) Remove a number of Costar 3795 96-well plates from their plastic sleeve and put at 4 degrees for anytime over 5-10 minutes. I like to store a number of these plates at this temperature so they are always at the ready. Fill the multichannel pipette basin with Fisher Silicon oil. Allow several of the 96-well plates to warm to room temperature, and label them with the relevant information (screen name, protein, date, etc.). Using the 50-300 UL 8-channel pipette, pippette 100ul of Fisher Silicon oil into each well of the 96-well plates. Several plates can be 'oiled' at once and the tips don't have to be changed for each transfer. The plate can be returned to 4 degrees to cool the oil if the plate is destined for incubation at 4 degrees (5).

(Adding the Protein) Spin the protein at 14,000 g for 10 minutes to remove particulates. Using a second Costar 3795 96-well plate (the storage plate), and a single channel pipette, pipette 50-300 UL of protein into each well of one column. Using the 0.5-10 UL 8-channel pipette, pipette 1 column of protein (1-10ul per well) from the storage plate to each column of each oiled plate. The tips do not have to be changed for each transfer. The plates can be stored for a short period at 4 degrees at this point if the plates are destined for incubation at this temperature.

(Starting Crystallization) Using the 0.5-10 UL 8-channel pipette, pipette one column (1-10ul per well) from the 96-well assay block containing the crystallization solutions to the corresponding column of the 96-well plate containing the oil and protein. When the all of the columns are finished, hold the plate over a light box to visualize the drops. If any of the drops have not merged take a fresh micro pippette tip and place tip at the surface of the oil to fill tip by capillary action with oil, then push the drops together.

(Using HDUV Crystals) Examine the plates immediately after setup, the next day, and ever two days afterwards. Since the conditions within the drops continually changes over time, plates should be monitored frequently to catch/use crystals soon after they grow to full size. Crystals can be harvested from drops using a cryo loop and transferred to various solutions (heavy atom, cryoprotectant, etc) prior to data collection.

VDUO Test - VDUO parameters for crystallization of lysozyme

• Using VDUO has numerous advantages over hanging drop vapour diffusion
  • No More Cold Room Setups - The 96-well plate with silicon oil can be incubated at 4 degrees prior to initiation of screening. Protein and precipitants can be added at room temperature and the plate quickly returned to 4 degrees. In addition the silicon oil acts as a thermal regulator, smoothing out any short term rapid temperature fluctuations.
  • Easy Manipulation - The open nature of the drops allows you to perform complex manipulations without the usual evaporation worry one has with with assembly of hanging drop vapour diffusion experiments.
    
    • Easy Seeding - Crystal plates can be assembled with drops which are unable to nucleate but will support crystal growth. The nuclei from a crushed crystal ( with serial dilution's) can then be added to each drop ( 10ul drop, 1ul seed stock).
    • Easy Additive Screening - Your Hampton detergent and additive screening was never so simple! Just assemble your 96 well plates as above. Typically one protein condition, one crystallization condition (usually 4.5 + 4.5). Then at your leisure add 1ul of the detergent or additive agent from Hampton to the corresponding drop.
    • Cheap, Cheap, Quick, Quick. - The low cost and speed of VDUO gives you the ability to sample larger areas of 'crystallization space'. If you don't get hits with +/- substrates/inhibitors, the best advice is to change the protein variable! Change/remove tags, mutate protein, change protein source (different species). Screen, Screen, Screen, Move ON!
      
      
• This is an aggressive crystallization strategy. The drops last approximately 2 weeks before the appearance of salt crystals in several of the conditions. Applying paraffin oil over drops which have positive results will extend the life of the drop another 2 weeks. Crystals should be screened relatively soon after they form. Questionable crystals after 2 weeks should be tested to rule out salt by attempting to crush them with a drawn out capillary. Protein crystals should be soft and not crunch when pushed on. We have not encountered any difficulty converting crystallization conditions from hanging drop to VDUO and visa-versa so crystallization hits can be reattempted with hanging drops for longer shelf life if need be.
• Fisher silicon oil and Hampton silicon oil are not the same. Hampton silicon oil yields faster evaporation and can be fully mixed with Hampton paraffin oil to yield what is termed 'Als oil'. Fisher Silicon oil is immissible with Hampton paraffin oil. The rate of diffusion from the drops is highly dependent on temperature. Silicon oil from Fisher and incubation at room temperature gives just the right time frame. If you want diffusion I would recommend using Hampton Silicon oil at 4 degrees and Fisher Silicon oil at room temperature. I would switch to batch under oil at higher temperatures (25-40 degrees) since drops under Silicon based oils dry out too fast at these temperatures.
• Not all 96-well plates are created equal. Several I have tried resulted in drops which did not remain spherical under oil due to interactions of the drop with the plate surface. Use Costar 3795 Non-treated polystyrene plates. They are cheap and result in spherical drops.
• Static is not your friend. 96-well plates which are freshly removed from their packing sleeve tend to have a buildup of static electricity. This causes the drops to adhere to the side wall instead of dropping to the bottom of the well. Cycling of the plates between 4 degrees and room temperature seems to help dissipate the charge. Storage of the plates, out of their sleeve at 4 degrees is the easiest.
• Despite your best efforts some of the drops will not merge when the plate is first set up. This is highly dependent on the size of the drops. 3+3 drops almost always results in one drop, while 1 + 1 yields ~ 20% unmerged drops. Apparently some of the drops also carry a net charge, and can be manipulated by moving a finger under the well in question. Very often the drops will move around and then merge. Otherwise one can push the drops around with a micro-pippette tip (after letting it take up some oil). For those with money, one can get an adapter for centrifuges which allow you to spin the plates (500 rpm/1 minute) which brings the drops together.
• Remember that the drops continually concentrate over time, so it is difficult to pinpoint the exact drop conditions when one wishes to harvest the crystal. Usually the crystal will be stable in the mother liquor 'reservoir solution' that was combined with the protein at the time of setup.
• Crystallization screens are good for about 1 month in the 96-well Assay blocks covered by the Hampton Robolid. The more proteins you screen each month, the more economical this method of crystallization becomes.