From humble beginnings
16 May 2012
This is the story of a quirky little peptide sequence whose discovery led to the development of a crucial tool for today's biotechnology industry, leading to cancer treatments and advances in regenerative medicine, and where ongoing research is helping to explain the modern-day evolution of pathogens.
FMDV infecting bovine epithelium. Copyright: Philippa Hawes, Institute for Animal Health
It all started in the mid 1980s when researchers at the Institute for Animal Health were dissecting the foot-and-mouth disease virus (FMDV) with the aim of developing new vaccines from its constituent parts. Postdoctoral researcher Martin Ryan was developing a method to produce empty virus capsids (outer shell) when he noticed something rather unusual.
"We knew that a peptide sequence called 2A played an important role in processing the FMDV polyprotein into separate protein segments, but it did not behave as we had expected," says Ryan, now Professor of translational virology at the University of St Andrews.
Like other members of the picornavirus family, which includes poliovirus and rhinoviruses (the usual suspects behind the common cold), the FMDV genome encodes a single protein precursor molecule (the polyprotein) that is separated into three regions – P1, P2 and P3 - by 2A and another viral protein (3C).
In poliovirus, 2A behaves as an enzyme, cutting the polyprotein at a precise point between P1 and P2 . That's not the case in FMDV. Ryan, together with Professor Andrew King and Dr Graham Belsham, showed that cleavage in FMDV occurs at the end of the 2A region but how 2A triggered this processing event remained a mystery.
Uncovering the mechanism
To solve the riddle, Ryan created an artificial 'self-processing' polyprotein. Importantly, this system allowed him to demonstrate that 2A triggered cleavage of the polyprotein independently of any other FMDV protein sequence .
Cells co-expressing two different fluorescent proteins (linked via 2A) each of which is targeted to a different sub-cellular location.
Copyright: University of St Andrews
Now at the University of St Andrews, Ryan continued to refine his artificial system. By placing genes for green fluorescent protein (GFP) and β-glucuronidase (GUS) either side of 2A to create one long, single gene, he showed that cleavage led to an unusual imbalance in gene products, with twice as much GFP produced compared to GUS. He investigated 'flipping' the order of the components so that GUS was 'upstream' of 2A and GFP 'downstream'. Curiously, the imbalance remained.
"After ruling out degradation of the downstream products, we concluded that we were observing a novel translational 'recoding' event," says Ryan.
Experiments in yeast by Dr Jeremy Brown at Newcastle University, added further weight to this hypothesis. They showed that translation of the polypeptide from the RNA code pauses as the ribosome (the catalyst of protein formation) reaches the end of the 2A sequence, generating one product, before continuing to generate a separate downstream product .
"The discovery of this unusual self-cleavage mechanism was really exciting. And it raised the possibility of engineering multiple, different proteins, encoded by many genes, into a single gene, which was of huge significance to the biotechnology industry."
Real life applications
Since the introduction of synthetic human insulin 30 years ago, the production of high value, therapeutic proteins such as antibodies, enzymes and hormones has been one of the driving forces behind the biotechnology industry.
However, ensuring that such 'recombinant' proteins are biologically active when produced on a large scale, vital for medical use, has been hampered by the lack of suitable 'vectors' that are able to produce multiple proteins in their correct configuration.
Take interleukin 12 (IL-12) for example, a signal molecule which plays a key role in the control of cellular immune responses and, along with its ability to block the formation of new blood vessels, has potential use in vaccine development and cancer treatment. The IL-12 molecule is composed of two protein chains, which are encoded on two separate genes that need to be 'co-expressed' within the same cell in order to form an active molecule, an arrangement that was not possible before the discovery of 2A.
Hemera, copyright Thinkstock 2012
In 1999, Ryan, together with colleagues at IAH, constructed an IL-12 self-processing 2A polypeptide. At around the same time, Ryan was working on a project in collaboration with AstraZeneca to create 2A polyproteins in the tobacco plant.
"Together these studies provided the first real proof that 2A could be used in any biological system," says Ryan. "Since then, more than 500 research papers and many patents have been published citing the use of this sequence to co-express multiple proteins from a single open reading frame. In fact, it's now almost impossible to keep track of them all ."
But it was in 2006 that 2A really became an industry standard for the co-expression of proteins after scientists in the USA recreated a complete T-cell complex (key to generating a targeted immune response) in a transgenic mouse .
Since then, success has been notable in the fields of cancer treatment and regenerative medicine. 2A has been critical for the development of treatments for metastatic melanoma, colorectal cancer, synovial cell cancer and is entering trials for renal cancer. The sequence is also being used in the production of transgenic plants and tailor-made pluripotent stem cells.
It's an achievement which Ryan is rightly proud of. "It's been a buzz, an absolute thrill," he says. "When I saw the applications coming through in gene therapy I had a big cigar."
Not one for resting on his laurels though, Ryan is investigating the occurrence of 2A in other organisms: "We have found 2A in a very wide range of viruses. Our objective is to look at the relative merits of different sequences for different applications.
"For example, the 18 amino acid long 2A sequence from Thosea asigna virus, which infects butterflies, offers superior cleavage to its FMDV relative."
And by probing wider genome databases, Ryan and his St Andrews colleagues have found 2A-like sequences in the form of non-LTR retrotransposons (jumping genes) and proteins involved in the innate immune systems of eukaryotic cells including trypanosomes, purple sea urchins, desmosponges, and other primitive marine organisms.
"We did think that this approach would show us how 2A has evolved in viruses and cellular organisms," says Ryan. "But surprisingly, we now think we are seeing that organisms are picking up 2A sequences all the time as a means of bolting on extra functions."
"For example, in Rotaviruses – the most common cause of severe diarrhoea in young children, we see 2A in type C, but not in type A or B . The type C viruses have acquired a new function – an RNA binding domain added via a 2A sequence – which is absent from the same genome segment in type A and B viruses. Interestingly, this same RNA binding domain is bolted onto the beginning of the shrimp totivirus genome, again via 2A, and we think it helps protect these viruses from the innate immune system.
"It's amazing to think that basic research into something so small, supported by BBSRC over many years, can have such a major impact across a whole industry."
- Sommergruber W., et al. Polypeptide 2A of human rhinovirus type 2: identification as a protease and characterization by mutational analysis. (1989) Virology., Mar;169(1):68-77
- Ryan M.D., and J. Drew. Foot-and-mouth disease virus 2A oligopeptide mediated cleavage of an artificial polyprotein. (1994) The EMBO Journal, 13, 928-933
- Co-translational, intraribosomal cleavage of polypeptides by the foot-and-mouth disease virus 2A peptide. Journal of Biological Chemistry, DOI: 10.1074/jbc.M211644200
- Generation of T-cell receptor retrogenic mice. Nature Protocols, DOI:10.1038/nprot.2006.61