You put some salts in, some proteins precipitate out, you put some other salts in and things go the other route – you do the Hoffmeister series Hokey Pokey and you salt things ion and out. What the heck’s this all about?! (no one really knows, but its effect shows!…) And how does ionic strength impact molecular interactions? I hope you have the ionic strength to handle today’s post, complete with a bumbling biochemist PSA on salt concentrations in binding experiments…

I’ve been trying to optimize this binding experiment, and one of the conditions I tested was using salt concentration to manipulate the “ionic atmosphere.” Might sound like some witchy spell-casting, but it really just has to do with changing to charge “fogginess” around molecules. Since opposite charges attract, when a salt dissociates (e.g. table salt (NaCl) dissociates into Na⁺ & Cl⁻), the charged particles (ions) can surround and “shield” oppositely-charged parts of other molecules (like parts of proteins). This can make the shielded molecules less “available” to binding partners, weakening their affinity

This leads to the biochemistry equivalent of the whole “the dose makes the poison” phenomenon, whereby anything (even water) can kill you if you have too much – “the salt concentration makes the binder” – basically, “anything” will bind “anything” at low salt concentrations (which is why you should be suspicious when papers describe doing pull down assays, etc. at low salt concentrations and claim they have “specific binding”). 

I’ll get into the more technical specifics in a second but an easy way to think about it is that water really likes water. So, if you stick a bunch of water molecules together, they’ll bind to each other tightly, forming strong interconnected networks that give you surface tension and stuff. If you want to break up those networks, you have to give water something that it like more than water. Pure water typically likes itself more than biochemical molecules like proteins and nucleic acids (DNA & RNA). So if you stick those things in water, they’ll be “excluded” by the water – the water network forms a “solvation shell” around them and, because the water tries to maximize contacts with other water molecules, it kinda tightens around them, leading them to minimize the surface area exposed to the water. 

This “hydrophobic effect” leads to proteins folding up into compact shapes in which only their hydrophilic (water-liked) parts are on the surface, exposed to water, and all their hydrophobic (water-excluded) parts are hidden in the center). If you stick multiple copies of a molecule in water (which is basically always!), each molecule can either get a full water coat (solvation shell) of their own, in which case we say it’s “dissolved” – or a bunch of the molecules can glob together and share a coat, minimizing their combined water-exposed surface. In this latter case, we say it is insoluble.

You can think of intermolecular (between molecule) interactions as a sort of “middle ground” whereby a couple of molecules (e.g. protein A & protein B) bind together in a specific manner and share a coat, but remain functional and able to unbind and they’re not just “globbed together” like you’d see with a precipitate. These “middle ground” interactions are crucial to the functioning of basically all biological molecules, so our bodies (and those of all organisms) rely on salts to (among other things) help make them possible (and in the lab we sometimes have to figure out the ideal salt concentrations to do so through experiments).

How does salt come into play? First, a quick terminology note: In everyday talk, “salt” usually refers to one specific salt, sodium chloride (NaCl) (aka table salt). But there are lots of different salts, not all of which act the same as we’ll see. Their unifying factor is that they’re a neutral combo of a positive ion (cation) and a negative ion (anion). (e.g. in table salt, NaCl, Na⁺ is cation & Cl⁻ is anion).

Proteins are also (usually) ions, but more complicated ones. Proteins are folded-up chains of amino acid “letters,” each of which has 1 of 20 unique “side chains” or “R groups.”  Most of these side chains are neutral but some are ➕ charged (“basic residues”) or ➖ charged (“acidic residues”). Because water likes charged things, these amino acids are hydrophilic and proteins fold to keep them on the surface, so you end up with a protein surface coated w/➕ & ➖ charges.

Opposite charges attract,  so these charged residues interact w/salt’s cations & anions. These interactions “shield” proteins from each other, and, when shield’s strong enough (which we can describe as the IONIC STRENGTH of the solution) proteins can’t “see” their potential protein partners.

BUT if there’s not enough salt around, proteins can “see” each other. If the proteins are already dissolved, this can lead to specific protein-protein interactions which you might not see at a low salt concentration (low ionic strength). And this brings me back to that whole “salt concentration makes the binder” PSA – when things are really foggy (high ionic strength) only the highest affinity (strongest) intermolecular relationships can survive. You can also think about it as salt “out-competing” the weaker binders. 

Often scientists do experiments called “pull-downs” or “immunoprecipitations” (IPs) to see what a molecule of interest binds to. They mix antibody-coated beads with a cellular lysate (the insides of broken-open (lysed) cells), wash the beads to remove non-specific binders, and then look to see what’s bound (e.g. with Western blots or mass spec). The washing is key, and a lot of the time it’s done at too-low salt concentrations. And same with some binding experiments where purified components are mixed, but the salt’s too low. 

Note: this is for charge-based (electrostatic) interactions (which can involve full or partial charges). Increasing ionic strength weakens charge-based interactions, BUT it strengthens hydrophobic interactions.

That was looking at proteins which are already dissolved, but what about proteins which aren’t dissolved? The line between solubility and insolubility is thin, leading for opportunity to get things to precipitate out or go in!

If you start at a low ionic strength (low salt) where your proteins can see (and bind to) each other really well, and add (certain) salts, you “loosen up” the water shell around the proteins because you distract the waters a bit with the salt ions and give the proteins a chance to dissolve. We call this SALTING IN – adding salt when solution is at ⬇️ ionic strength causes protein to go INto solution

BUT there are only so many protein sites the salt can interact with, so this effect tops out. And it doesn’t just top out. Instead of just stopping having any positive effect on solubility, you reach point where adding more salt causes protein to come OUT of solution – you’ve reached the SALTING OUT region. Here, salt starts competing w/protein for water (they both need water coats (hydration layers or solvation shells) to stay dissolved) & salt’s a better competitor so it wins the water & proteins bind to each other & huddle together to share water that’s left over.

In crystallography, we can use this salting out as a strategy for getting proteins to (hopefully) form orderly crystal structures which we can beam x-rays at to get a look at the protein structure. 

Salting out can also be used as a protein purification technique (especially common in the past) because the amount of salt needed to precipitate out a protein is protein-specific, so you can selectively precipitate proteins from impure samples using the “right” amount of a salt like ammonium sulfate, then re-dissolve it to get pure protein. This can work if the salted-out proteins are no longer soluble BUT they’re not fully denatured, so you can “reverse it” (don’t have to unboil an egg). Alternatively, you can use salting-out conditions you know will *not* precipitate your protein of interest, so that you can get rid of other stuff but keep your protein happy. 

I used salting out during a purification in undergrad, but I haven’t used it for purification since and don’t think it’s very common anymore now that we have more advanced protein chromatography techniques and the ability to use things like affinity tags to express a protein of interest with a “tag” that we can get to bind specifically to certain beads and wash the rest of the stuff off. In the early days of biochemistry, though, they were isolating native proteins and they didn’t even really know what they were going to find a lot of the time. So they’d do things like keep adding salt, and testing the various salted-out parts to try to find the fraction with some activity they’re interested in – and then look more closely at that fraction to find the molecular culprit. Sorry that got technical and I don’t have time to discuss it all here, but check out  for an example of how it was used in the discovery of DNA Polymerase. 

Now, for a little complication… As I hinted at before, not all salts are created equal. If we go back to our definition of a salt (a neutral combo of a cation(s) and anion(s)) we see that there’s a lot of room for variation, and the identity of the ions matters. To help understand why, let’s review what an ion is.

Atoms are the basic units of elements and the difference between elements (e.g. carbon (C) vs nitrogen (N)) is the # of protons they have. Protons are little positively charged things & they’re one of 3 key subatomic particles – they hang out with neutral neutrons in a dense central atomic nucleus and then negatively-charged electrons (which have an equal but opposite charge despite being itty-bitty-er) whizz around them in an electron cloud. If a neutral molecule loses an electron, # of protons > # electrons, so it becomes positively-charged (cationic) and if a neutral molecule gains an electron, # of protons < # of neutrons, so it becomes negatively-charged (anionic). 

“How charged” a molecules is depends on how many electrons it has relative to its neutral state – we call this the VALENCE, and we write it in superscript to the right of the element’s letters. If the charge is +1 or -1, we call it MONOVALENT and if it’s more, we call it POLYVALENT

Since opposite charges attract, cations are attracted to anions & vice versa – so even when Na (which has one electron it doesn’t want in a shell all to itself) gives up an electron to Cl (which just needs one to complete its shell) to give you Na⁺ & Cl⁻, those ions hang out together through an “ionic bond” which is really just a strong attraction – unlike covalent bonds (like those linking together amino acids) which involve electron sharing (i.e. orbitals unite!)

A salt is a neutral combo of cation(s) & anion(s) – such as NaCl (Na⁺ + Cl⁻) or CaCl₂ (Ca²⁺ + 2Cl⁻)

When you put a salt in water (assuming there’s enough water around – that is, you’re under the solubility limit), the salt will dissociate (the cations & anions will separate), so if we want to know how they’ll interact with other molecules, especially when it comes to solubility, we sometimes need to start thinking of the ions individually. But there are also things where the identity doesn’t matter as much, so let’s look at this “generic stuff” first.

In the presence of dissolved salt, opposite charges still attract, but now there are more potential partners because the ions are free to move around and “discover” other ions that were previously far away in the solid form. This results in an IONIC ATMOSPHERE where cations cluster near anions and anions cluster near cations. This has the effect of “shielding” the charges of the individual ions. The more ions there are, the more dense the fog, the stronger the IONIC STRENGTH. 

Not all ions contribute equally – in addition to concentration, ionic strength depends on the square of the valences of the ions – so ions with a higher charge contribute more (e.g. the same amount of Ca²⁺ will raise ionic strength more than that amount of Na⁺) So, we get “more bang for the buck” in terms of ionic strength using salts with polyvalent ions such as CaCl₂.

This ionic strength is what I was talking about before, when looking at binding affinity. IONIC STRENGTH depends on the amount of ALL the ions present. It adds up the contribution of each ion to the atmosphere. At a higher ionic strength, there are more ions competing for the water. This lowers the “chemical activity” of water, so it’s like there’s less water (another way to think about this is that the fog is hiding the water from the proteins).

Note: “chemical activity” is an “alternative” to concentration which takes into account how many molecules it “seems like there are.” The activity depends on the ionic “environment” of the solution (how many other charges are around to interact with – the more other charges around, the more it’ll get “side-tracked” and be unavailable). Increasing the ionic strength lowers the chemical activity of the various ions in the solution – it “weakens” the contribution of any one ion, so it “counts less” when we consider how the solution behaves)

“more salt” -> higher ionic strength -> lower chemical activity

Let’s tie this back into our binding talk… An addendum to the “salt concentration makes the binder” saying is that “binding partner concentrations make the binder” – if you stick a ton of copies of thing A and thing B together, even if they only kinda like each other they’re going to bind. Even if the binding’s just short-lived they’ll quickly bump into another molecule and bind it and do this over and over so that they’re usually bound. Adding salt increases the ionic strength, lowering the chemical activity of the binding partners so that it’s like there’s less of them, so there’s less binding. 

We can also tie chemical activity into solubility and the salting-in/salting-out curve. At low ionic strength, our protein will have a higher activity. So it sees there’s lots of copies of itself around and binds to them (insoluble aggregate). But as you raise the ionic strength by adding salt, our protein has a lower activity it’s like there’s less of it around so it feels free to come out of solution. Alternatively, it can’t find “partners” as easily – but there’s still a lot of water around, so it partners with those & becomes more soluble. But as you raise the ionic strength even more, you’re also reducing the activity of the water – salt & proteins both want water shells, but the salt’s a better competitor, so it gets the water and the proteins bind to each other instead.

You might expect solutions with the same ionic strength to have the same precipitating power. But this is *not* the case. Not all ions contribute equally. Ionic strength does *not*  depend on the identity of the ions, just their concentration & valence (so the same amounts of CaCl2 & MgCl2 have the same effect on ionic strength). Solubility on the other hand, *does* depend on the identity of the ions. The relationship isn’t straight forward, and scientists are still trying to figure out the “why,” but there are ion-specific effects.

The Hoffmeister series orders ions according to their ability to dissolve proteins from hen egg whites. This series is empirical (based on experimental observations) and doesn’t hold true for all proteins – it’s more like a quick rule of thumb. Different proteins will show more specific effects because the ions can also interact directly with the protein. So, it’s important to try out multiple salts, even if they have the same ionic valence.

But here’s the series:

anions (ordered from most precipitating to least precipitating): F⁻ ≥ SO₄²⁻> H₂PO₄⁻> H₃CCOO⁻> Cl⁻> NO₃⁻> Br⁻> ClO₃⁻> I⁻>ClO

cations (ordered from most precipitating to least precipitating): NH₄⁺> K⁺> Na⁺ >Li⁺ >Mg²⁺ >Ca²⁺ 

Traditionally, these differences have been explained in terms of ions being “kosmotropes” which “bring order to water” or “chaotropes” which “bring chaos.” Ions to the left of Cl⁻ & Na⁺ are classified as kosmostropes and tend to contribute more to salting-out. Ions to the right of Cl⁻ and Na⁺ are classified as chaotropes and contribute more to salting-in.  

Why? It’s usually been talked about in terms of affecting the “bulk water” (so most of the water in a solution) but newer research is pointing to its affects being more specific to the solvation shell as well as interactions with the proteins themselves. But here’s the theory…

As we discussed, the reason why water’s really “sticky” is that it’s highly POLAR (has partly ➕ & partly ➖ parts that are attracted to their opposites on other water molecules, allowing them to form strong networks. They can also stick to partially or fully charged parts of “non-waters,” so you can get things (solutes) to dissolve in it if they can integrate themselves into network.

CHAOTROPES are solutes that are *not* very good at integration – they only weakly bind water, but they do bind it, and in doing so they disrupt water networks and make them more fluid. Now water molecules are on the move & looking for binding partners – if there’s protein around, they can latch on. This can break up protein-protein bonds, causing the protein to start to unfold (denature). As it unfolds it reveals parts its hydrophobic core it’d been hiding. If the chaotrope weren’t there, the protein might clump these parts together because water wouldn’t want to bind to it, BUT the chaotrope makes it harder for water molecules to find one another, leading the water to settle with binding less-polar protein parts. 

so, in summary: chaotropes increase solubility of less-polar things that might otherwise not be able to compete w/water-water attractions 

We can use chaotropes like guanidine hydrochloride & urea to denature proteins & nucleic acids (DNA & RNA) but keep them soluble

CHAOTROPES bring chaos to water, keeping water networks “loosey-goosey” so they can more easily coat things (think of trying to paint w/fresh pant vs old sticky paint)

KOSMOTROPES, on the other hand, interact strongly with water, so they *can* integrate well into water networks & strengthen them. This pulls water away from protein ⏩ proteins compensate by clumping together (aggregating) & coming out of solution (precipitating) to ⬇️ surface area & “share” water. The proteins are no longer soluble BUT they’re not fully denatured, so you can “reverse it” (don’t have to unboil an egg)

Amount of kosmotrope needed to do this is protein-specific, so you can selectively precipitate proteins from impure samples using the “right” amount of a kosmotrope like ammonium sulfate, then re-dissolve it to get pure protein 

As for the whole Hoffmeister series thing – well, it’s complicated… As I mentioned, it was empirically determined with a mixture of egg white proteins, not taking into account pH or anything. Apparently the order can be reversed for some proteins, the counter-ion really matters, etc. Here’s a good explanation 

But I think that’s enough for now 😅 Sorry if it was too much, but I was learning for myself so thought I’d share in case anyone did care!

more on topics mentioned (& others) #365DaysOfScience All (with topics listed) 👉⠀

Leave a Reply

Your email address will not be published.