Vapor diffusion crystallography methods harness the power of evaporation to remove water from a protein solution, raising the protein’s concentration to a supersaturated state where there’s not enough water to go around, so the protein molecules bind to each other instead (hopefully as crystals and not just clumps of “precipitate”).

To power the evaporation, we seal our a drop of our protein above a “reservoir” of a more concentrated solution. The solution is more concentrated in terms of precipitants (the chemicals that lower the protein’s solubility to promote crystallization) but less “concentrated” in terms of water molecules (because there’s more “other stuff”) and it’s the water that’s going to do the moving because it’s volatile (evaporates easily) whereas the other stuff’s usually non-volatile, so it’s stuck.

Molecules want to be free – they try to escape the liquid state, where they’re “chained” by bonds to other molecules, to be a gas (where they can move around as they wish) but, below the boiling point, only molecules close to the surface have a chance to break free. 

But we perform vapor diffusion in sealed wells -> if a molecule breaks free, it’s still trapped in the well. It tries get as “free” as it can by moving away from the drop’s surface, where there’s lots of other waters -> it travels towards the reservoir, where the concentration of water molecules is lower. But when it gets there, it gets “pulled in” by the reservoir and condenses back to a liquid. 

Initially, we have a net movement of water from the drop to the reservoir, but eventually, the amounts of water in the air above the drop (whose presence we measure as vapor pressure) are the same as the amounts of water above the reservoir, so there’s no “benefit” to moving towards the reservoir – molecules continue to leave the drop (and the reservoir) but other molecules are entering at the same rate and we say that we’ve reached a vapor pressure equilibrium. Note that the vapor-pressure equilibrium is different from the dissolved/undissolved equilibrium, which is referring to whether our protein’s dissolved.

We can understand this movement of water in terms of Raoult’s Law, which tells us that adding solute to a solution decreases the evaporation and thus the vapor pressure. It’s a colligative property, so the more solute – any solute – , the lower the vapor pressure. This is because the solute

  1. takes up some of the valuable real estate near the surface, where molecules have a chance of escaping from & 
  2. increases the entropy (randomness) of the solution so there’s less benefit to becoming a gas 

↑ solute leads to ↓ evaporation and ↓ vapor pressure

more on Raoult’s law here: blog: 

The consequence the extent of evaporation from the drop depends on the difference in solvent concentration between the drop & reservoir (more specifically the difference in their vapor pressures). And Raoult’s law shows the direct connection between the amount of solute dissolved (and correspondingly the decrease in the solvent fraction). The bigger the difference between the drop & the reservoir, the more dramatic the effect (although many things deviate from the ideal assumptions of the law so beware things might not behave quite like you predict).

So we have water moving from the drop to the reservoir. It really is like a drop in the bucket – Because there’s so much more liquid in the reservoir, the drop can’t change it much, so the final composition of the mother liquor is basically the same as the initial composition of the reservoir. 

And we can call this movement of water in its gas form “vapor diffusion.” Diffusion is when molecules move from where there’s more of something to where there’s less (more here) to even things out.

It’s easiest to picture if you think of putting a drop of food coloring in water and watching the dye spread out. In vapor diffusion crystallization, water molecules are going from the drop containing our protein to the reservoir – and because these 2 solutions are separated, the water molecules move as gas (water vapor).

It can get confusing because precipitant concentrations are higher in the reservoir. But it’s not the precipitants that are moving (typically) – it’s the water. And from the water’s perspective, it’s less “concentrated” when there’s less other molecules around.

Another thing that can be confusing – what about the protein concentration? There’s none in the reservoir but a lot in the drop! The protein isn’t volatile so it can’t move, and what the water’s trying to “even out” isn’t really the concentrations of the different molecules, instead it’s equilibrating an indirect measure, vapor pressure.

Basically, at an air-water interface there will be some water molecules going into and out of the drop. And instead of moving from where there are more molecules to where there are less molecules, they’re moving from where there’s higher vapor pressure to where there’s lower vapor pressure. And the vapor pressure is going to be higher where there’s a lower concentration of solute (e.g. precipitant). 

for example, say you have a 10mg/mL protein sample & a crystallization cocktail with 1.0M precipitant). If you mix 1uL of each in your drop, you dilute both of them so your drop solution starts at 5mg/mL protein & 0.5M precipitant. Your reservoir is pure cocktail, so it’s a 1.0M precipitant (and no protein)

Concentration’s the number of molecules in a specific volume, so when you decrease the volume, you increase the concentration (even though you’re not changing the number of molecules). Because the drop & the reservoir are separated from each other, the only way they can transfer molecules is through the vapor phase (they have to evaporate from the drop (go from liquid to gas), diffuse down to the reservoir, and condense (go back to liquid)

The only molecules that can do this are those that are volatile – in our drops, this usually is just the water – so it’s the water that’s moving. But because it’s leaving, the volume of the drop’s decreasing, so the concentrations of protein & precipitant increase. And hopefully you get crystals!

So that’s the theory – how does it work in practice? There are 2 main methods that differ by where the drop is. If the drop is on the “roof” of the well (held there by surface tension) we call it hanging drop and if the drop is sitting in a depression on a ledge or pedestal (like a dunk tank except it never dunks) we call it sitting drop. The theory behind the 2 is very similar, but they can give you different results, and there are both practical and scientific reasons to choose one over the other (and the protein’s the final judge so you might want to try both!) 

From a technical/practical standpoint: 

  • Sitting drops are easier for robots to set up – with hanging drops the robot dispense the drops onto the seal that we then flip over by hand and seal above the well – and if you don’t line it up just right the robotic imager’s not very good at finding them…
  • Sitting drops are more “stable” in terms of not falling if you accidentally bump the plate
  • BUT hanging drops are easier to harvest from (fish the crystals out of).

From a scientific standpoint:

  • they have different equilibration kinetics (how fast things happen). 
    • With hanging drops, the drop’s surface area is larger, so evaporation is quicker, so equilibration is quicker (think of an egg yolk sitting in a cup versus on a plate – the cup “blocks” part of it from air contact and prevents it from spreading out)

This is especially important if your sample contains surfactants – surfactants (SURFace ACTing ageNTS) include soaps & detergents, and they lower the surface tension of water – surface tension is a measure of the intermolecular forces “gluing together” the surface of a liquid – it’s why our drops stay spherical even if they’re “upside down”

Water has a high surface tension because water molecules have an uneven charge distribution that makes them “sticky.”

Surfactants get between the water molecules linked together at the surface, lowering the surface tension, so the drop spreads out – as it spreads out, more water molecules are at the surface and able to escape -> vapor pressure equilibrium is reached more quickly 

You might think this is a good thing – less wait time, right? BUT if the protein concentration rises too quickly, you can get lots of crystal “births” whose seeds quickly deplete resources so none of them can grow very big. and/or you can get “defective” crystals because the growth happens too quickly and there’s not enough time for “quality control”

So sitting drop might be a better choice if your solution contains surfactants

Either method can be done in a variety of different drop sizes and plate sizes (e.g. 24-well (good for getting big crystals) or 96-well format (good for screening)). 

more on crystallization:

more on crystallography here:

Leave a Reply

Your email address will not be published.