The Thrice-Told Story of Watson and Crick

I'd like to get into a little history.  This is an area where I'm more of a fan than a scholar, but at least I try to be a responsible fan of history.  What does it mean to be a 'responsible fan'?  From my conversations with actual historians I think this requires an ability to admit to multiple narratives, many of which could be conflicting, at the same time.  History has a long tradition of being used as a turf war to 'prove' worldviews right or wrong.  This happens both in the teaching of history and in the way it's written down.  The old adage, "those who don't learn from history are doomed to repeat it" probably does more to confuse people about the subject than anything.  It instructs us to look at history as a moral lesson; this leads us to fit the facts to our biases, and that's the kind of 'bad history' I'd like to avoid.

I have no illusions that I'll dodge that misuse of history here, since want to highlight how it impacts science, but I'll try to be light on the moralizing.  Instead, I want to focus on how the way we tell our history impacts our view of the scientific enterprise.

Science is a peculiar enterprise, in part due to the whole 'quest for knowledge' angle.  It's strange because for most problems faced by ordinary people there is either already an answer somewhere, or at least an expert who knows how to solve the problem.  Science is the process of taking observations about phenomena - questions about the unknown - and applying them to a problem nobody knows how to solve yet.  Sometimes scientists identify a problem long before the solution.  For example the problem of producing energy through nuclear fusion, or curing cancer.  It might take years to find those solutions.  Or we may never find them.  But we don't know what the end will be from the beginning, including whether it will be a dead end.  There is no expert to consult who can solve the problem.  No cheats, no answers in the back of the book.

It would be nice if we could at least get a hint, though.  Perhaps we could look to past discoveries and get a clue about what will work for the next one.  That would help us focus on moving discovery in the right direction.  If we knew that most big questions like cancer or fusion are solved in specific ways we might take that information back to the lab when we tackle big problems.  For example, what if major discoveries are really just the cumulative work from dozens of scientists who didn't know they were working on the next big thing?  In that case we should just throw a bunch of money at all kinds of crazy research strategies.  A diversity of directions will produce unexpected results.  Perhaps the scientists working on jellyfish who won the Nobel Prize think this way.

What if, instead, it's more common that scientists aren't appreciated until they make big waves.  We ignore top talent because we're too focused on current models.  But groundbreaking science is about disrupting those models, so it's routinely ignored because we are unable to get our head around new science.  It's only when someone brazenly generates spectacular data that people turn and notice.  In that case we should be careful not to dismiss fringe scientists because they have unconventional ideas.  Barry Marshall might sign onto this, given the Nobel he got in part because he drank an infectious agent to prove it caused ulcers.  As the story goes, few people believed him until that daring act.

There are dozens of explanations like this, each with their own historical case study to support it.  My personal favorite answer is a combination of all and none of the narratives at the same time.  Of course, that answer doesn't help with the question: how do we target the next discovery?  But I'm less interested in 'getting an answer' than I am in making sure the answer is right.  A bad answer is worse than no answer, because no answer tells us we need to figure something out.  But a bad answer might convince us we're done searching such that we persist in our errors.  In the spirit of honest history, and of science, I'd like to embrace one good case study in how a narrative can shape the history - for good and ill.  History is not just a retelling of events, so it is often abused.

 Narratives in Science: Watson and Crick

At university, I first learned the Standard Narrative about the discovery of the structure of DNA - one of the defining moments in the field of molecular biology.  If you're not a biologist, it's hard to emphasize how critical this point was.  Once the structure of DNA was discovered, it became obvious to everyone how it would be able to encode all the information needed for life.  Although this point can be overstated, most modern work in biology would be impossible without this one discovery.  It's a major fulcrum point for biological knowledge, akin to Newton's laws in physics.

The two guys responsible for this discovery were James Watson and Francis Crick.  They were working in England, racing against an American group headed by the eminent chemist Linus Pauling.  If you've never heard of Pauling, the guy was pretty amazing.  He developed foundational ideas in biochemistry, including s-p orbital hybridization, and eventually won two Nobel prizes.

After Oswald Avery performed a famous experiment showing that DNA contains the genetic material, the race was on to discover the structure of DNA. Watson and Crick knew the stakes were high, and after Pauling published a paper with his own theory of the structure of DNA (which turned out to be wrong), the stakes grew higher.  Data from Erwin Chargaff showed that there were equal ratios of each pair of bases.  Watson and Crick put this together with some X-ray crystallography that showed DNA forms a helix.  By collecting all this data, building on the shoulders of various researchers, Watson and Crick  published their structure of DNA by showing that you could pair the bases to form a double-helix.  They made the discovery by diligently piecing together all the puzzle piece discoveries of other scientists.  Sure, they got the glory, but in the end it was a major team effort.

That's the story I heard in college.  It's a nice story that shows how cumulative discoveries in a field were collected by Watson and Crick, and this led to a groundbreaking understanding of how DNA carries the genetic code.  One of the reasons this narrative is so powerful for research is that a lot of science is exactly the opposite of groundbreaking.  The nature of research lends itself to this kind of investigation, where you end up specializing in some very specific sub-discipline, hoping to generalize your findings somehow.  Eventually.

Everyone is working in their own obscure corner of science, and everyone insists they are working on something that will one day be big - they hope.  The story of the Watson and Crick discovery appears to justify all research under the understanding that, after all the research is collected together it will eventually lead to great, massive breakthroughs.  Every grunt worker in the lab is driving toward a collective achievement!

Enter Rosalind Frankin

That's not the whole story, though.  Later on I learned that one of the crucial details glossed-over in the account above is the whole, "they got some X-ray crystallography" images part.  In every textbook I've seen that discusses the discovery of DNA there's a picture of the X-ray crystallography Watson and Crick used to discover DNA.
But it turns out neither Watson nor Crick was any good at X-ray crystallography.  In fact, neither of them was any good in the lab at all.  Crick's first experience at the bench came years after the DNA discovery, and he was this weird combination of lab novice and eminent professor peppering his assistant with questions that should have been obvious to someone of his stature.  It would be like a mathematician winning prestigious awards for solving intractable problems asking, "what's the order of operations again?  And why does it matter we do it in that order?"

Clearly the X-ray crystallography pictures couldn't have come from them.  Where did they get this pivotal data?

There was one very accomplished X-ray crystallographer just down the road from Watson and Crick at King's College: a young female researcher named Rosalind Frankin.  Her crystallography was top-notch; the result of careful technique.  She was also interested in the structure of DNA, and had produced beautiful crystallography images.  But Franklin was not well-liked among her colleagues.  She was a woman in a hostile, male-dominated research environment.  To survive, she kept her research close to the chest.  She had her own ambition of figuring out the structure of DNA and the notoriety that would come with it.

Watson and Crick knew about Frankin's data, and they wanted it.  They had been pestering Franklin to share her results with them, but she consistently refused.  She didn't want them swooping in and stealing all her hard work.  So while Franklin was working hard in the lab, Watson and Crick went to her superiors at King's college to get the images without her consent.  Franklin had two forms of DNA she was working on, an "A form" and a "B form".  When she wasn't occupied in the lab producing more images, Franklin was only studying the A form.  It seemed more interesting to her than the B form.  'Why not give the unused B form data to us?' W&C argued.  King's college agreed, and the images were taken from Franklin without her consent.

So sure Watson and Crick discovered the structure of DNA, but they did it on the back of Franklin's data.  They barely even acknowledged her in their publication, even though she did all the hard work that made it possible.  She got no credit despite being the one to all but make the final connection in the discovery!  This narrative shows how difficult it was for women in the 1950's and 1960's to get ahead in science, despite hard work and determination.  Systemic sexism was rampant, and Frankin was robbed of contemporary praise for her pivotal achievements.

Forget all that, here's how it really happened

Historical narratives all have an agenda, ideology, or motive attached to them.  That necessarily includes this one.  History is messy, and doesn't usually run smoothly from point to point in an easily-told path to victory, success, and discovery.  In other words, history looks a lot like real life.  The same could be said of how the structure of DNA was discovered.  First off, let's get some context for how people thought of DNA prior to the Watson and Crick publication.  So far, all the narratives look back and assume the inevitable future, where everyone knows DNA is going to be important, we just have to get this pesky structure business out of the way to make that happen.  But it wasn't a straight line of discovery.

The Standard Narrative says Avery showed DNA carried the genetic material, and after that the race was on to discover its structure.  But that's not right.  Sure, Avery did his experiments, but few people at the time thought the results were as convincing as later scholars make them out to be.  Most researchers at the time thought proteins carry the genetic code, not DNA.  This made a lot of sense, since proteins are complex, and DNA is very simple.  How could all that information be carried by DNA?  It's probably just a structural molecule holding things together in the cell.  Geneticists of the time were all about proteins, and weren't interested in this DNA nonsense.  Sure, Avery 'showed' DNA was needed for the genetic material, but if you go back and read Avery's original paper it's full of glaring holes - especially if you think protein is the answer, not DNA.  But a small community was inspired by the Avery paper anyway.  Let's call them the DNA nerds.

At that point, everyone already knew what they called the "subunit" structure of DNA.  In other words, they knew what each strand looks like, including the negatively-charged phosphate backbone (light blue circles below), the ribose sugar in the middle (green-gray pentagons), and the bases along the chain (colorful CGAT shapes).  That looks like this:


They didn't know how this thing came together inside the cell, or what it was doing there.  The DNA nerds figured it carried the genetic material, because of Avery's work, but they had a lot of uphill work to show how that was possible.  It's too simple a molecule.  Chargaff ... didn't really have anything to do with the final discovery.  That detail was added after-the-fact.  And Franklin's images were helpful in that they gave Watson and Crick some numbers like the distance between turns (10 nm), but they weren't the ultimate catalyst for discovery.

In fact, there's a reason Franklin didn't use the B form images Watson and Crick sneaked from her.  The difference between the A form and B form was that the A form had water in it - similar to how DNA would naturally be found inside a cell - and the B form was dehydrated.  Franklin wanted the actual structure, not some artifact.  That's good logic.

Watson and Crick liked the B form because it confirmed their suspicion that DNA is a helix.  Why did they think it was a helix?  Well, scientists working on protein structures had recently discovered lots of helices (like Linus Pauling), so they were trendy at the time, and Pauling had some X-ray crystallography suggesting a helix.  In fact, that's the form DNA took in Pauling's paper, largely because Pauling had crystallographers working for him, too.

If we look at that failed paper of Pauling's for a minute more, we might get a better grasp of where everyone was before the moment of discovery.  Pauling's structure was a train wreck.  First, he took all the phosphate backbones and shoved them in the center.  Remember these are all negatively charged. This would be like taking three powerful magnets and forcing all their S poles together almost to the point of touching.  Next, he put all the hydrophobic bases outside.  This is exactly like an oil-and-water situation.  If you tried to assemble this whole crazy structure, it should simultaneously fly apart, flip around, and tangle up.  It would violently turn into a mess.  Pauling justified his structure, saying maybe there were some support proteins or something holding it all together.  And besides, he had to do it this way because 1. his X-ray crystallography suggested a helical structure, and 2. the bases wouldn't pair, so he couldn't put them in the center.

When I first read Pauling's paper, this detail surprised me.  Watson and Crick's major discovery was supposed to be their pairing of the bases, which naturally leads to the rest of the double helix coming together.  But here is Pauling - an incredible chemist already at this time - saying, "I tried that and it didn't work."  What's going on

Here's where things get really specific.  Watson and Crick were also trying to get the bases to pair, but they hadn't been successful either.  Then one day Crick was at lunch.  He brought along some cardboard cutouts of the different bases of DNA, and was trying to work out some scheme to make them pair.  He used the cutouts to try stacking them, rotating them, bent angles, side-by-side, or end-to-end - but of course nothing worked.

A chemist happened by and stopped to chat.  He noticed these chemical structures Crick was working on, and Crick explained this was something called des-oxyribose nucleic acid (that's the old-timey way they said DNA it back then).  Although biology wasn't really the chemist's field, Crick was working on distinctly chemical structures, so he took a look at the cutouts.  He immediately asked, "is there a reason you're using the enol form instead of the keto form?"  I told you this would get really specific.

What Crick didn't know, and Pauling hadn't noticed, was that whoever first published the "subunit" structure (that's the individual strands) of DNA published it using enols.  Anyone who took organic chemistry will of course have forgotten the chapter about enol-keto tautomerization.  It's one of those really tiny, pesky details that almost never comes up but you have to memorize it for the class.  Basically, there's this chemical structure that's pretty rare and flips back and forth with another chemical structure in certain conditoins.  It can be one or the other in solution, but not both at the same time (it's not a resonance structure).  About 80% of the time it's the more stable ketone structure (the one on the right, below) and about 20% of the time it reacts with water to form the the enol structure (left below).



Crick, Pauling, and everyone else were using an unstable structure and didn't even know it.  After the chemist helped Crick correct the structure of his bases, they paired easily.  He went back to show Watson, and the two built their famous double-helix model of that very afternoon.

This is a narrative, so I guess I should have some take-home message, like the two before me.  That message isn't "we build on each others' shoulders to attain new heights", or "science has a history of suppressing the contributions of women".  That's not to say both those other messages are wrong.  Clearly we do build on others' shoulders, and what happened to Franklin was inexcusable.  But there was nothing inevitable or cumulative about correcting a near-imperceptible error.  And Franklin wouldn't have got there no matter how meticulously she worked on her technique.

Those two stories are not how the structure of DNA was discovered.  If there's a message here it's that sometimes while we're building on what others have created we randomly discover part of that foundation is wrong.  And new data might be helpful in some vague sense, but it can never overcome built-in errors that are holding us back and so small as to be invisible to everyone.

In the case of DNA, the wrong part was such a tiny detail it could easily have been overlooked for decades longer had Crick not been lucky enough to bump into the right person, while carrying around cardboard cutouts of DNA bases.  With nobody else interested in DNA, and a structure already proposed by the great Linus Pauling, it's easy to see how the foundations of a status quo had already been laid out.  If these two no-name upstarts hadn't come up with such an obviously-true model, how long would this "structural molecule" have been cast aside by serious researchers as uninteresting?  Might we still be in the dark about DNA to this day?

The limits of discovery

The case of DNA is not unique, but it's also not universal.  Some discoveries, such as the publication of the human genome, are grand achievements of science where a community of contributors comes together to brute-force a well-known problem (although that's also a messy story in its own right).  But other discoveries, such as the invention of knockout mice and the discovery of angiogenesis, are flukes of history that could easily have gone the other way.

This is where we run into questions we don't have any answers to.  How many discoveries are out there just waiting for some tiny error to be corrected, or for some fluke accident to bring all the right elements together?  Are we just lucky, and we've discovered all the low-hanging fruit?  Is everything else out there going to require cutting-edge technology?  Or have we left behind a long trail of missed opportunities from errors in tiny details that have been holding us back for decades - even centuries? If we focus on figuring those out will we see the floodgates of discovery burst through?  Maybe we'll never make progress on some problems until we fix those small errors.

These are all legitimate questions to which nobody knows the answers.  If we knew where to look for all the big discoveries, everyone would be out in that one place with a pick and a shovel furiously digging away to find them.  Instead disparate groups craft their own narratives and secretly think they've hit upon the solution to discovery.  They won't be like everyone else, just plugging along doing mundane science.  When I worked in the lab, we had our own ideas of where the frontiers of asthma discoveries were headed.  One of the post-docs scoffed at some of the home-brew geneticists who literally work out of their basement on their own dime trying to make new discoveries for science.  Given the budget differences between us, I remember at the time thinking how absurd it was for those small-time operations to think they could make any contributions at all, let alone fundamental discoveries.

Perhaps it takes all kinds.  If there's really no way to know where the next big discovery will come from we shouldn't discourage anyone.  But if that's true, how do we know who to encourage?  How do we direct funding?  I don't know.  Sometimes discoveries feel like Secrets of the Universe handed down from on High, and we just have to make sure we're ready to receive them.  The current narrative is wrong about something, whether it's how we got here or where we're going.  We just don't know how it's wrong.  And this is perhaps the most peculiar thing about science: you learn to forge ahead through the forest of everything you don't know.

Comments

Popular posts from this blog

Reverse Engineering Life

Cancer update: precision oncology comes into its own