Thread by Carolyn Bertozzi

twitter.com
1 Recommender
1 Mention
Recommend
Post
Save
Complete
Collect
Thread
People are asking me about the origins of and key milestones in the timeline of #bioorthogonal chemistry, so here is a quick summary. 1/n

As a postdoc @UCSF in Steve Rosen's lab, I was struck by how powerful live cell and in vivo imaging tools for proteins and nucleic acids had become, while cell-surface glycans remained in accessible. 2/n
Questions like, what kinds of sugars are expressed on different cell types and how do they change upon perturbation, could not be answered without destroying cells and much labor. The first project I worked on illustrates the point. 3/n
pubs.acs.org/doi/abs/10.1021/bi00182a010
Methods for imaging and profiling sugars on live cells/animals were lacking, a clear unmet need that I thought might be tackled with chemical approaches. But how to do it? 4/n
As a postdoc I had a serendipitous encounter with the great Prof. Dr. Werner Reutter at a boutique conference in Southampton, England, where he talked about sialic acid biosynthetic enzymes tolerating unnatural substrates: 5/n
www.jbc.org/article/S0021-9258(18)41874-1/pdf
The tolerance of that pathway for nonnative acyl groups got me thinking about whether metabolism could allow delivery of chemical handles to cell surface sialic acids for covalent labeling with imaging probes. But what type of chemistry would allow this in living systems? 6/n
The first idea was inspired by Keith Rose and coworkers, who had been publishing on complex peptide assemblies build with hydrazone and oxime chemistry: 7/n
pubs.acs.org/doi/pdf/10.1021/bc00024a014
These reactions are mild, selective, water compatible, conditions amenable to biomolecules and cells. So if we could get a ketone into cell surface sialic acids, we might be able to attach probes via this handle, a big step toward bioorthogonality. 8/n
Lara Mahal @glycocode @UAlberta and Kevin Yarema @JohnsHopkins pulled it off with this amazing study, the first big paper from my fledgling lab. 9/n
www.science.org/doi/10.1126/science.276.5315.1125?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr...
But the problem was not fully solved; ketone-hydrazone/oxime chemistry, while fine on cultured cells, was not ideal for in vivo imaging since other carbonyl metabolites could interfere and the pH optimum was too acidic. We needed something that was truly bioorthogonal. 10/n
I had a long-standing fascination with azides, having used them in the synthesis of amino sugars during grad school, and watching them undergo mild reduction with triphenyl phosphine in aqueous solvents as well as dipolar cycloadditions. 11/n
Here is a blast from the past wherein my 2-azido C-glycosyl allenes kept undergoing intramolecular cycloadditions during workup, what a pain! 12/n
www.sciencedirect.com/science/article/abs/pii/S0040403900798261?via%3Dihub
Also, azides undergo limited metabolism in humans, as evidenced by the HIV drug AZT. Could the azide replace the ketone as a truly bioorthogonal functional group? This would require a biocompatible reaction with a partner that led to formation of a stable covalent bond. 13/n
We thought the classic Staudinger reaction with triarylphosphines was close, but its immediate product, the iminophosphorane (aka "aza-ylide") is unstable in water. Solve that problem, and you might have a truly bioorthogonal reaction. 14/n
We figured out how to do just that with the "Staudinger Ligation" as we called it (and others added my name later on, "Bertozzi-Staudinger Ligation", 🥳). This was the next big story from our lab, published @ScienceMagazine in 2000: 15/n
www.science.org/doi/10.1126/science.287.5460.2007?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr...
Fun fact: The grad student who did this work, Eliana Saxon, was the third person I pitched the project to. The first two thought it wouldn't work and took a pass. Eliana got her first big results during the summer before grad school even started! 16/n
Amazingly, the Staudinger ligation could be performed in live animals, a feat accomplished by @jenn_prescher @UCIrvine and Danielle Dube @BowdoinCollege. A first-in-class achievement by these intrepid young chemists, published @Nature in 2004: 17/n
www.nature.com/articles/nature02791
But there was room for improvement. While incredibly clean and well-tolerated, the Staudinger ligation took many hours to proceed and this was too slow for imaging of dynamic processes that unfold more quickly. A faster bioorthogonal reaction was our next goal. 18/n
Meanwhile, my @NobelPrize partners Barry Sharpless and Morton Meldal had their own interests in highly selective "click reactions". 19/n
In 2002, they published two landmark papers on the fast cycloaddition of azides with terminal alkynes catalyzed by Cu:
Angew Chem Int Ed, 2002, 41, 2596.
J Org Chem, 2002, 67, 3057.
Could we use this miraculous process as a faster bioorthogonal reaction with azides ?
19/n
Not quite. The Cu catalyst was toxic to cells and animals. But perhaps there was a way to accelerate the azide-alkyne cycloaddition reaction without the use of a metal catalyst. 20/n
I was teaching first semester organic chem as I often do, and in the course of writing my annual lecture on ring strain a thought came to mind. Perhaps a strained alkyne in a ring system would react faster. 21/n
Grad student Nick Agard, now @genentech, found this precedent: Chem. Ber. 1961, 94, 3260. Cyclooctyne, the smallest of the stable cycloalkynes with about 19 kcal/mol strain energy, was found to react with phenylazide "like an explosion" (German translation). Bingo! 22/n
Sure enough, cyclooctyne probes reacted with cell-surface azidosialic acids under biocompatible conditions: 23/n
pubs.acs.org/doi/10.1021/ja044996f
Still room for improvement of reaction rates and several coworkers went at it. In particular, @basktastic @CornellChem put two fluorine atoms next to the alkyne to create DIFO, which had the kind of fast kinetics we needed to do live cell imaging: 24/n
www.pnas.org/doi/10.1073/pnas.0707090104?url_ver=Z39.88-2003&rfr_id=ori%3Arid%3Acrossref.org&rfr_dat=...
This form of "copper-free click chemistry" had all the right stuff for in vivo imaging. Scott Laughlin @stonybrooku and @basktasktic performed an elegant study of dynamic changes in cell-surface glycans during zebrafish development @ScienceMagazine: 25/n
www.science.org/doi/10.1126/science.1155106
More imaging studies and reaction improvements ensued, and the field took on a life of its own outside our lab: development of the amazing tetrazine ligation by Neal Devaraj and Joe Fox, more practical cyclooctyne syntheses by the van Hest, van Delft and Rutjes groups, etc. 26/n
And so many applications of bioorthogonal chemistry beyond glycan labeling, too many to recount. It has been an amazing ride of mechanistic and synthetic chemistry, cell biology and most recently clinical translation. 27/fin
Mentions
See All