Streak seeding is a method to introduce pre-formed crystal nuclei into a drop to control nucleation and alter the way in which crystals grow. It uses the basic conditions under which crystals normally grow, modified slightly. Also referred to as micro-seeding, since the nuclei are usually invisibly small.
For this protocol, I recommend using hanging drop vapor diffusion rather than the sitting drop version that is often more frequently these days. Two reasons for this: it's easier to streak the seed stock over the flat cover slip than the sitting drop depression; it's easier to re-seal hanging drop plates with glass cover slips than sitting drop trays with sealing tape. If you don't pre-equilbrate your drops this will matter less.
Step 1.: Set up drops in conditions that are 60-80% of the precipitant composition for a normal condition that produces crystals - you specifically do not want crystal nuclei to form of their own accord in the drop. Conditions for spontaneous nucleation are usually higher than those required for crystal growth.
Step 2.: Let the drops sit for a while, mostly sealed (don't push down too hard). This removes the initial burst of solvent (water) during equilibration when the drop is first set up - the drop is much more dilute initially which might dissolve the microseeds. This pre-equilibration time varies from a couple of hours for AmSO4 conditions at room temperature to overnight for high percentages of PEG at 4 degrees C. Generally you want it to be significantly less than the amount of time it takes for a crystal to appear, since you don't want even the faintest chance of spontaneous nucleation before you add your own nuclei.
Step 3.: Select a drop or crystal to be used for seed stock. The crystals need not look that good, because all you're going to use are microscopic fragments. You can even use xtals used for data collection although radiation damage makes this a less attractive proposition in general. Extract the crystal, wash it thoroughly to remove precipitate etc, and put it in a small volume of stabilizing solution (e.g. well buffer or something a little more concentrated than well buffer). Smash up the crystal as thoroughly as possible using (e.g.) a pipette tip, syringe needle or other tool. The idea is to create as many very small fragments as possible. You don't have to make the thing homogenous. Some people then quickly spin down this seed stock to remove the larger crystal fragments and use only the supernatant as the seed stock - careful not to spin the seed stock down too hard or you'll pellet the seeds at the bottom of the tube.
Note: the quick and dirty proof-of-concept experiment could involve you running your cat whisker through an existing drop containing a large number of very small crystals and using this "loaded" whisker to streak into new drops. While I don't usually recommend this - too much chance that precipitate/skin/denatured protein or other undesirables get transferred as well - it might work as an initial trial. It's less likely to work if the crystals in that drop are large.
Step 4.: Take this concentrated seed stock and put it in a larger volume of stabilizing solution (e.g. 100-500 microliters) and mix well. This dilution can be varied but an initial attempt might be ~5x. The amount of time that the seeds remain "useful" depends on how stabilizing your "stabilizing solution" is. Sometimes they can remain good for a few days but I tend to create seeds just before I use them and don't assume they are useful beyond a few hours. For my first seeding attempt on each project I lean towards diluting the seed stock less, to make sure I get hits.
Step 5.: Open a drop that is pre-equilibrated. Dip a clean cat whisker into the seed stock. Run the cat whisker through the drop starting from outside the drop and ending outside the drop in one smooth motion. Reseal the drop. Repeat the same procedure with all the other drops. You might want to rinse the cat whisker in stabilizing buffer between loading it with seeds. You could also use the loaded cat whisker for more than one drop (e.g. 2 or 3 in rapid succession) which would be expected to transfer progressively fewer seeds to the subsequent drops.
Step 6.: Make sure all the drops are sealed and then put them away. Check the following morning and afternoon for signs of crystals forming along the streak seeding track. Many crystals take a few to several days to grow, even from streak seeding. If there's nothing, the condition you used was too low (seeds dissolved) or your seed stock was too dilute or you didn't pre-equilibrate for long enough. If there's too many nuclei your seed stock was too concentrated.
Step 7.: You can seed SeMet conditions with native seed stock, and vice versa. Because the seed nuclei are extremely small they contribute only minimally to the final volume of the crystal, whose composition is dictated mostly by the protein in the drop and not the seed.
Major variables:
Hampton Research being the vendor of many things crystallization-related have a couple of useful tools to make your streak seeding experiment easier. Although I prefer obtaining actual cat whiskers from actual crazy cat ladies the generically-named Seeding Tool can work in it's place. I suspect that Enrico Stura used Chinchilla whiskers in his earliest experiments (see: Stura and Wilson (1991) Applications of the streak seeding technique in protein crystallization. J. Crystal Growth 100:270-282) but wild Chinchillas are especially elusive in New Jersey. Hampton's Seed Bead is an efficient smasher of your fragile and hard-to-obtain protein crystals to generate a seed stock.
While this protocol speaks to using crystals from the same condition and crystal form to grow more of the same type, Microseed Matrix Screening has become more popular as a method to provoke crystal formation in entirely unrelated conditions. This makes me nervous apropos clogging the nano needle in our crystallization robot, but there are certainly some success stories with this, if you can generate large quantities of seed stock.
Last modified: October 2013 by Phil Jeffrey