One of the really tough conundrums about how life got started on Earth is the “water paradox”: Liquid water is necessary for life as we know it here on Earth, and yet a lot of chainlike molecules that all living things need (like proteins, RNA, and DNA) cannot link up spontaneously in liquid water. 

Let’s consider proteins.  Living organisms can’t make proteins without the help of … already-existing proteins.  So how could proteins possibly have arisen in reasonable amounts to help set up the beginning of life on Earth, somewhere between 4.4 and 3.8 billion years ago?

Well, it just got a whole lot easier to explain.  A research team at Purdue University led by chemistry professor R. Graham Cooks used a simple but clever setup to convincingly solve this conundrum and show how this could have happened — quite easily, actually — on the early Earth.  Their findings were reported October 3 in the Proceedings of the National Academy of Sciences.  (Behind a paywall, but I thank Dr. Cooks for kindly sending me a reprint so I could check it out.)

We’re confident that amino acids — the building blocks of proteins — were available on the early Earth because of a bunch of observations:

But protein formation?  Not so much.  There was one report in 2020 of the first protein being found in a meteorite, which generated some buzz but ultimately didn’t survive peer review and hasn’t appeared in any journal.  If it had, believe me, I’d have been all over that one. 

So we’ve been looking very hard for a demonstration of how proteins could have formed from amino acids somehow on the early Earth, but until now, no such luck.

Why is it so hard to make proteins out of amino acids in water?

In order to get started, a “condensation” reaction has to occur; that is, two amino acids need to link together and kick out a water molecule:

The two molecules on the left side are both the same kind of molecule: a generic amino acid.  All amino acids look like this, the only difference among them being the “R”.  “R” can be lots of different things, but life on Earth, with rare exceptions, only uses 20 specific things for “R”. 

To link two amino acids together, the “amino” (—NH2) end of one amino acid reacts with the “acid” (—COOH) end of another one.   This gives us OC—NH (the “peptide bond”, shown in red on the right above), and it also produces a water molecule (HOH, in blue).  But keep in mind that this reaction can also go backwards.  And within a body of water, this backward reaction happens way, way, way more often than the forward, just because there’s so dang much water around.

I wanted to find out exactly how obnoxiously difficult it is to make a peptide bond in water, so I used a couple of online tools called eQuilibrator and Calistry to help me out.  Let’s take the simplest amino acid — glycine — where the “R” is just a hydrogen (H) and try to link two of them together.  Under garden-variety conditions (for those keeping score at home: pH 7.5, 25°C, and 0.25 M salt), it turns out that when this reaction settles to its final equilibrium state, only two out of every 360 quadrillion glycine molecules will have paired up.  Sheesh, we’re never going to get life started this way.

As if things weren’t already hard enough, amino acids spend most of their time in the “zwitterionic” form at most pH levels you’d encounter on Earth, where they can’t even react with each other:

This means that even the super-crappy equilibrium state mentioned above will take a really long time to reach.  Now you see what an uber-tough nut the “water paradox” is to crack. 

So people have concocted explanations to get around this problem in ways that would fit in with conditions on early Earth, proposing that amino acids could be coerced to link together in water under very special circumstances like near natural mineral or clay catalysts, special salts, etc.  But these things start to feel like reaches and don’t provide intuitively satisfying answers.

Intelligent-design people would be happy to hear me say that, I suppose, but now I have to burst their bubble.

Cooks’ group provided the part we’d been missing all along — small water droplets — and showed that linking amino acids together within them turns out not to be hard at all.  They made a dilute solution of glycine or alanine in water.  After two hours, nothing had happened, of course.  But when they made fine spray out of those very same solutions, glycine-glycine (Gly-Gly) or alanine-alanine (Ala-Ala) dipeptides formed in easily detectable amounts in a matter of milliseconds.

Their first crack at making the droplets was with nano-electrospray ionization, or nESI.  Here we suck electrons out of a liquid (a dilute solution of glycine in water, in this case) by applying a strong voltage between the emitter and the detector.  The liquid takes on a net positive charge, so it’s literally pulled over to the negatively charged detector plate as a mist:

The drops get even smaller on their way over to the detector because the positive charges within them repel each other, and this makes the droplets break up.  Normally, nESI makes droplets that start out around 0.2 μm (millionths of a meter) wide and get smaller from there.

But now you might say, isn’t putting a charge onto the molecules kind of cheating, though?  That doesn’t routinely happen in nature!  And if you say that, you are right.  It’s not exactly fair.  Lots of people have made peptide bonds under crazy, non-Earth-like conditions.  Big deal.

So then they repeated the experiment without applying any voltage at all, making the spray by physically pushing the liquid through the emitter with a syringe.  The droplets weren’t quite as small, but it worked anyway.  The net charge hadn’t been responsible for the effect after all; it was all about the droplets. 

But it gets better.  They sprayed two glycine-containing mists together and ended up with not only Gly-Gly but also Gly-Gly-Gly and Gly-Gly-Gly-Gly.  When they sprayed two jets containg Gly-Gly together, they got Gly-Gly-Gly-Gly and Gly-Gly-Gly-Gly-Gly-Gly!  They mixed glycine and alanine sprays together and got mixed peptides like Gly-Gly-Ala, Ala-Gly-Ala, etc. 

Hey, this is getting protein-like! 

Presumably if you keep slamming peptide-containing droplets together, as in sea spray, waterfall mist, and the like, the peptides can keep getting longer and longer and give you some primitive protein material to work with.

But wait a second.  How could these reactions work so fast in little droplets of water but not at all in bulk water?  I mean, water is water, isn’t it?  The trick seems to be that they happen only at the surface of each droplet.  One side of that surface is all water, true, but the other side is air, with very little water.  It becomes a whole lot easier to make water — that is, to run a condensation reaction like

Gly + Gly → Gly-Gly + H2O

— when there is no water around.  On the surface of a droplet, these molecules can stay dissolved in water and yet have access to a place with no water.  And there, my friends, is your solution to the “water paradox”.

There are plenty of other reasons that reactions can behave very differently at an air-water interface, and I mention that because this is actually a pretty hot topic in chemistry now.  It’s known that the —OH groups of water molecules can stick up at the surface, not being part of the (weak) bonding network they would have within the liquid, and that alone provides a different environment:

There’s even an electric field across the air-water interface, and while that isn’t at all well-understood, it certainly can influence chemical reactions quite a bit.

The Purdue researchers observed that smaller droplets are better, because smaller droplets have a higher surface-to-volume ratio, so that means more surface area to do reactions compared to the volume to be filled, and so products like Gly-Gly can get more concentrated and be easier to detect.  

Can water droplets in nature get so tiny?  Of course they can!  Any droplet suspended in air will evaporate completely over time, so within a spray or mist we can generally have a full range of droplet sizes at any one time.  

It should be pointed out that, incredibly enough, the total surface area of sea spray on Earth is actually larger than the planet’s entire air-sea interface!  Not to mention all the droplets in clouds, waterfalls, rapids, etc.  So this is a hugely significant venue for new classes of chemical reactions occurring in nature, many of which we probably don’t even know about yet.

The same principle that applies to proteins here could very well also apply to DNA, RNA, polysaccharides, phosphates, and anything else biologically important that forms by condensation.  Hopefully our Purdue team will give some of those things a try next! 

But for now, one more unfathomable step in the emergence of life on Earth just got a whole lot more fathomable.  It’s always mesmerizing to watch the waves crash on the shore, and maybe part of the reason why is that we’re looking at the very beginnings of where we came from.