I was joking the other day that we could start an organization of Protein Purification Protocol Scouts – where we “earn badges” by learning new techniques – well, today, I’m working towards my Hydrophobic Interaction Chromatography (HIC) “badge.” And I feel it is my scoutly duty to release this PSA about the HYDROPHOBIC EFFECT – what it is and what it isn’t – because it’s easy to get confused, especially when “unfair” words like “fearing” are used.
You know how in soccer (or “football” as most of the world calls it), the defense “forms a wall” in front of the goal to defend it during free kicks? The players in the wall have their movement limited and aren’t very “comfy.” Thankfully they only have 1 “side” of the goal to protect – but what if the goal wasn’t “one-sided”? What if the ball could be kicked in from any of the 4 sides? You’d need an even more extensive wall – and this would take more of the defensive players “out of commission” in terms of their freeness to do other things – like score their own goals. The bigger the goal, the more players you need for the wall, but, at least you still just have a single goal. But what if you had lots of goals? And each of them had to be surrounded on all 4 sides? Well, that kinda sucks…
But – if you could squish all of those goals together to form one giant goal then you’d only have to surround a single goal – and although it’d take more players to defend than for a single smaller goal, it takes a lot fewer players than if you had to defend lots of single small goals, thanks to the surface area (the potential ball entry points) growing smaller than the volume (the space inside the goal where you don’t want the ball to go).
Take a cube for instance. The surface area (S.A.) of a cube with sides of length s is 6s^2 (each side has an area of s^2 and you have 6 of them per cube) and the volume is s^3. So if your cube has a side length of 2…
S.A. = 6s^2 = 6(2^2) = 6(4) = 24
And the volume?
V = s ^3 = 2^3 = 8
What if you increased the side length to 4?
S.A. = 6s^2 = 6(4^2) = 6(16) = 96
And the volume?
V = s^3 = 4^3 = 64
By doubling the side length, you’ve quadrupled the surface area, but octupled? (multiplied by 8) the volume. The S.A./V ratio for the smaller cube was 24:8 = 3:1. But for the bigger cube it’s 96:64 = 3:2. You can see that the volume is growing faster than the surface area. So we can protect more goal volume with fewer players by squishing the small goals together. And this is similar to a biochemical phenomenon called the HYDROPHOBIC EFFECT.
Water molecules order themselves into “cages” called clathrates around “hydrophobic” molecules that don’t offer binding opportunities (at the extreme things like fats and oils, which don’t readily mix with water) – kinda like those defensive walls formed around goals in soccer.
There are a lot of misconceptions about this effect. The word “hydrophobic” literally means “afraid of water” – but I hate this term – and you know I’m all for anthropomorphizing molecules – that’s not my issue here. Actually, issue(s) – I have a lot of issues 😛 but, when it comes to hydrophobes , here are a couple…
Firstly, I don’t think it’s “fair” to call them “afraid,” especially because, as we’ll see, the hydrophobes aren’t the ones with the problem in this relationship…. I prefer the interpretation “water-avoiding” – but even this doesn’t well characterize them – a better description would be water-avoided. Because it’s not that hydrophobes are repulsed by water or anything – and water’s not even repulsed by them – but they don’t offer much in the way of favorable interactions. In contrast, water molecules offer lots of attractive interaction opportunities for other water molecules, so water molecules would much rather spend their time hanging out with other water molecules.
So, the hydrophobes really don’t have much say in the interactions named after them – instead it’s the water optimizing itself that is the real source of the “hydrophobic effect”
Water molecules are really “sticky” towards each other because they’re highly polar – basically atoms (like the 2 hydrogens and the oxygen in H2O) have smaller parts called subatomic particles – positive-charged protons & neutral neutrons in a dense central nucleus with a cloud of negatively-charged electrons whizzing around. Atoms can form bonds by sharing electrons, but they don’t always share fair. Oxygen is much more electronegative (electron-hogging) than hydrogen, so it pulls their shared electrons closer to itself, making the O partly negative (δ-) and the Hs partly positive (δ+). And opposite charges attract, so the H’s of one water molecule can hang out with the O of another. Each water molecule can form up to 4 “hydrogen bonds” (H-bonds) with other molecules.
Unlike the strong covalent bonds holding the Hs to the O in each individual water molecule, these inter-molecular bonds are weaker, so they can stick and unstick. As long as water molecules have sufficient energy to temporarily break free of the bonds, the water molecules can move around & explore, breaking and forming interactions with other water molecules as they travel. These water molecules can occupy many different “states” and the term we use to describe this is high entropy (aka “disorder” or “randomness”)
But they can’t interact readily with hydrophobic molecules, which are characterized by being nonpolar (electrons are evenly distributed so there aren’t partly or fully charged regions) and thus don’t offer tantalizing charge opportunities. So each water molecule that has to be next to part of a hydrophobe has part of its stickiness “hidden” and is limited in its binding opportunities – it can occupy fewer “states” and thus has lower entropy (is less disordered)
We use a term called free energy, G, to describe how “comfy” a molecule is – it takes into account entropy (S) (that disorder) as well as something called enthalpy (H), which has to do with bond energy. Molecules interact spontaneously in ways that make them comfier (reduce the free energy) and we describe this using the equation
ΔG = ΔH – TΔS
This says that the change in (abbreviated delta, Δ) free energy equals the change in enthalpy (do the new interactions have more or less energy than the old ones) minus temperature (in Kelvin) times the change in entropy (do the molecules have more freedom now?) Negative G is “good” (means a reaction is favorable) – and you can get to it if the new bonds are much less energetic (easier to hold together) and/or the molecules gain freedom of movement.
Say you have a sea of water molecules and you toss in a hydrophobe. Some of the water molecules will have to hang out with it – there’s no getting around that – and because there’s only so much space around a water molecule, it’ll have to break up some of its water-water bonds to do this. And this requires putting in energy (without getting a better bond in return) so you have a + ΔH
Now imagine you keep dropping in hydrophobes. Each time a water molecule swaps an interaction with water for an interaction with the hydrophobe, it has to “spend energy,” so you keep racking up “enthalpic penalties.” and it loses binding opportunities – so you have “entropic penalties” as well. But, if those hydrophobes all cluster together, fewer water molecules will have to give up the “better” opportunities offered by water.
But the hydrophobes have no intrinsic desire to clump themselves together – instead it is the water molecules around them that kind of shepherd them together – by “reaching out” to other water molecules in their network (remember each water molecule can form up to 4 H-bonds), they draw together (kind like how surface tension can lead to drops of water staying spherical – when hydrophobes combine, water molecules get released from the clathrate cage, leading to an increase in entropy (+ ΔS). And this offsets the energy you have to put in (ΔH) to break up the individual cages when you merge them. So the hydrophobic effect is ENTROPY-driven.
I’ve been talking about hydrophobes as whole molecules – but you can also have molecules where parts are hydrophobic but other parts are hydrophilic (can form positive interactions with water). And this is the case with proteins. The hydrophobic exclusion effect is actually the main driving force for protein folding.
Proteins are written as long chains of “letters” called amino acids – there are 20 common amino acids, and they have a generic “backbone” so they can link together and unique “side chains” sticking off (kinda like charms in a charm bracelet). Some of the charms are hydrophobic, and you can think of them kinda like “soccer goals” – if you can get the goals together, you don’t need to take as many players “away from the game” to guard it.
When proteins are being made, the water molecules have to shift around to accommodate them, and this drives the hydrophobic areas together (merging the goals), usually inside the interior of the protein’s 3D structure. But the water has to make some compromises, so there are some hydrophobic patches left accessible – different proteins have different levels of such accessible “hydrophobicity”
We’ve been looking at the hydrophobic effect as between water and one type of hydrophobe. But, since it’s the water that’s “in control” it can shepherd different types of hydrophobes together in order to maximize water-water interactions and minimize water-hydrophobic-region interactions. So hydrophobic parts of different molecules can “stick together” – so we can get hydrophobic parts of protein to stick to hydrophobic beads, which is the theory behind Hydrophobic Interaction Chromatography (HIC).
This was my first time trying out this chromatography technique which is basically the opposite of the more common ion exchange chromatography. You’re still getting proteins to selectively stick to little beads (resin) inside of a column but, Instead of getting proteins to stick to oppositely charged beads based on electrostatic interactions at low salt concentrations, then disrupting those interactions by increasing the salt, in HIC you get proteins to stick based on the HYDROPHOBIC EFFECT at high salt concentrations, and then get them to “unstick” by decreasing the salt.
The hydrophobic effect is more pronounced at high concentrations of certain salts because the salt kinda distracts the water molecules, which leaves “less water” for solvating the protein – this exposes hydrophobic regions and makes it easier for the water cages to merge so the protein can stick to the hydrophobic groups stuck to the resin. There are different types of HIC resins with different hydrophobic groups sticking off the beads – the one I’m trying is phenyl sepharose which as a ring structure, but there are also other options like octyl & butyl, which are straight-chains.
We exploit this salt effect in crystallography as well, adjusting salt concentrations to “salt in” or “salt out” to get proteins to group up, and you can learn more about that here: http://bit.ly/2PEpwgf