The molecular biology revolution of the 1960s and 70s was built on the backs of microscopic bacteria such as E. coli and the tiny bacteriophage viruses that infect them. As it became clear that the same DNA and protein code was at work in everything from bacteria to elephants, plants to pandas, scientists thought that the underlying processes controlling genes would be the same too.
Frustratingly, this turned out not to be the case.
Whenever a gene is switched on, its DNA sequence is ‘read’, or transcribed, to make a kind of molecular photocopy known as RNA. And it’s this RNA that acts as the instructions that tell the machinery inside a cell to make a particular protein.
So, you might think that any given piece of RNA would be an exact match for the gene it was transcribed from, right? Wrong.
By the 1970s, researchers knew exactly how big every gene in E.coli was, and could match each one up to an RNA message of identical length. But when they looked at genes in larger, more complex organisms like humans, something didn't add up.
The RNA messages in the nuclei of human cells (where the DNA is) were far too long, each one many times bigger than it needed to be to encode the recipe for a protein. But, weirdly, the RNAs in the rest of the cell, where the protein-making-machinery was found, were a much more sensible length.
So what was happening?
One of the people who solved the mystery was Rich Roberts, winner of a share of the 1993 Nobel Prize in Physiology or Medicine and the more dubious honour of being designated Dr December in the 1997 Studmuffins of Science calendar.
His discovery started from an experiment trying to find out whether genes in adenovirus – a tiny virus that causes the common cold – worked in the same way as genes in bacteria. And, by deduction, because the cold virus genes get switched on by human gene-reading machinery when it infects our cells, it would suggest that human cells must work in the same way too.
Roberts and a new recruit to his lab at Cold Spring Harbor, Richard Gelinas, set about coming up with a clever way of grabbing the RNA messages from adenoviruses as they were made inside human cells and matching them up to their corresponding DNA-based genes.
But something was wrong.
Instead of seeing a neat match between the 12 viral genes and 12 viral RNAs, their experiments showed there was just one RNA, which appeared to match all the genes. Weird, huh?
After double-, triple- and quadruple-checking that Gelinas hadn’t messed up the experiment, Roberts realised that the only way to figure out what was going on would be to get hold of a high-powered electron microscope to have a look at the DNA and its corresponding RNAs, to see how they matched up.
When I spoke to Roberts for my book about how genes work, Herding Hemingway’s Cats, he told me the story of how this experiment came to be.
“Neither Richard nor I were electron microscopists, but we had a couple of people just down the hall who were really good. On a Saturday morning we went down to them and said, ‘we've got this idea for an experiment, could you do it?’ And they said, ‘well it's never quite been done that way before, but we'll give it a go.’”
Those two talented electron microscopists were Louise Chow and her husband Thomas Broker, who ended up having to invent a completely new technique to look at how DNA and RNA were matching up together in order to solve the mystery.
Their images were grainy and confusing, looking more like strands of spaghetti on a dirty plate than microscopic strands of RNA and DNA, but they told an incredible story. Each strand of DNA was matched up to a long string of RNA, stuck together where the sequences matched. But in certain places the DNA had nothing to pair with and was looping out, as if there were bits of the RNA message missing.
It looked like the virus was making one long string of RNA covering all its genes, then cutting it up and pasting bits together to make twelve separate messages encoding proteins – something that had never been seen before.
According to Roberts, these pictures were the crucial evidence that convinced sceptical journal editors and reviewers that what they had found was real.
“If there really were millions of molecules in your initial sample and you're just looking at two or three, how do you know you're not looking at the weird ones?” he told me. “But we already had all the biochemical evidence to back it up, and as soon as people saw the pictures everybody believed it at once.”
In 1977, Roberts published a paper in the journal Cell with the title, “An amazing sequence arrangement at the 5' ends of adenovirus 2 messenger RNA.” It was the first description of ‘split genes’, or what we now call RNA splicing, and it was well worth the superlative headline. In keeping with this excessive trend, their follow-up paper described the peculiar cutting and pasting arrangement as ‘baroque nuclear architecture’.
Rather than being a concise, neat recipe to make a protein, every gene in complex organisms is a rambling, almost-incoherent mess of exons – those are the bits that actually encode for proteins – interspersed with a bunch of unwanted nonsense, known as introns. Once an RNA message is transcribed, the introns get chopped out, the exons get glued together, and there you have it – a short, coherent recipe to make a protein.
Thanks to Roberts and Gelinas’ biochemistry and Chow and Broker’s electron microscopy skills, a whole bunch of weird observations from labs around the world started to make sense. RNA splicing was the perfect explanation, and the idea quickly became established dogma.
Importantly, the discovery of splicing also explained another biological mystery: if human cells have around 20,000 genes, how do we manage to make around 200,000 different proteins?
The answer lies in alternative splicing, where a selection of specific exons gets glued together to make one version of a protein, and a different set is stuck together to make a different version.
Finally, there are another couple of splice girls involved in this story, and that’s Sue Berget – a postdoc in the lab of Phil Sharp at MIT - and Claire Moore, who was their technician at the time. Around the same time that Roberts made his discovery and started talking about it at meetings and conferences, Berget was noticing something similarly strange in her own research on genes.
Roberts told me, “She had spotted something odd and knew that something funny was going on, which Phil didn't believe. And I think it was only later that he came to realise that what she'd been telling him all along was correct.”
Once Sharp heard about splicing, everything slotted in to place and he and his team raced to publish their own paper. While Sharp got a share of the 1993 Nobel prize for the co-discovery of splicing along with Roberts, Berget and Moore were left out, although they were given a nod in MIT’s press release about the happy occasion.
On the other side, Gelinas, Chow and Broker could only bask in Roberts’ reflected glory. I don’t know whether they were also invited to use the croquet lawn that he bought with his prize money.
As Chow pointed out at the 2017 AAAS meeting, “As a reserved foreign woman scientist, I was not accorded recognition commensurate with these revolutionary discoveries. Proper credit remains a touchy issue for budding scientists.”
As we’ve heard in previous podcasts, these omissions happen because of the rules of the Nobel prize - with only three living scientists able to take a share, Nobels tend to go to lab heads and lead investigators. Is this a fair system? As science becomes ever more of a team effort, I’m increasingly feeling that it isn’t.
What do you think?
References and further reading
Louise Chow - AAAS meeting 2017 https://aaas.confex.com/aaas/2017/webprogram/Paper19519.html
MIT announcement of Phil Sharp’s Nobel prize https://news.mit.edu/1993/sharp-1020
Adenovirus-2 messengers--an example of baroque molecular architecture. Broker TR, et al. Cold Spring Harb Symp Quant Biol. 1978;42 Pt 1:531-53. https://www.ncbi.nlm.nih.gov/pubmed/277362
An amazing sequence arrangement at the 5' ends of adenovirus 2 messenger RNA. Chow LT, Gelinas RE, Broker TR, Roberts RJ. Cell. 1977 Sep;12(1):1-8. https://www.ncbi.nlm.nih.gov/pubmed/902310
Louise Chow presentation slides, Cols Spring Harbor Meeting 2017 http://library.cshl.edu/Meetings/mRNA-Splicing/slide-pages/Chow.php
1993 Nobel Prize in physiology or medicine https://www.nobelprize.org/prizes/medicine/1993/press-release/