Skip to main content
Premium Trial:

Request an Annual Quote

Aptamers Aren t So Easy to Turn into Drugs -- Hmm, What About Using Them in Arrays? Wonders SomaLogic s CEO Larry Gold

Premium

AT A GLANCE

NAME: Larry Gold

AGE: 60

POSITION: CEO, SomaLogic

PRIOR EXPERIENCE: Professor of Molecular, Cellular, and Developmental Biology, University of Colorado, Boulder; co-founded Synergen, and founded NeXagen, which later became NeXstart Pharmaceuticals

  In the first of a two-part interview, Larry Gold discusses his entry into the field of protein capture arrays. Next week: the evolution and future of SomaLogic.

How did you get involved in proteomics?

[It] started out with the discovery of how to find these aptamers that Craig Tuerk and I did when I was back at the University of Colorado in the late ‘80s. This led to the formation of NeXagen around ‘91-’92, and we were going to make aptamers into pharmaceuticals. We had chosen an incredibly gifted scientist at NeXagen/NeXstart named Nebojsa Janjic, who was responsible for pushing us toward VEGF antagonists for the treatment of macular degeneration. That was going to be NeXstart’s first drug from the technology. At the same time, we finally understood that even though this aptamer for therapeutics was likely to be good, maybe there was another use of aptamers worth thinking about, and that was to use the aptamers in chips for proteomics. Sometime around ‘96 we got really serious at NeXagen about using big arrays of aptamers to do proteomics. We had a little group of 20 to 25 people from the research group working on this, and then in July ‘99 we sold the company to Gilead [Sciences]. I was chairman of the board, and I spent a lot of time explaining to the very nice people from Gilead all the stuff that they had just bought.

After the sale was completed, they sold SomaLogic everything we’d been doing for the last couple years so that we could become the proteomics array company. If I were to think about it, I had this epiphany: There was a moment where I thought something like, ‘Oh damn, therapeutics are really hard; hmmm, arrays, lots of measurements, hmmm, aptamers, gee, they’re pretty good.’ I figured there ought to be a way to turn the value of aptamers into something useful for both pharma companies as well as clinical diagnostics and not face some of the issues that you have to face with aptamers as therapeutics. Gilead wasn’t going to pursue this [so] when we had a chance to pursue this, we jumped at it.

Did SomaLogic already exist when Gilead sold it the protein-chip project?

After the sale I worked honestly with all the people from Gilead to tell them what my dreams were about proteomic chips and what you could use them for. But Gilead is an extremely focused therapeutics company and they didn’t want to be in that busines. After that deal was completed, they said, ‘Larry, if you’re really serious, why don’t you make us an offer?’ So we formed SomaLogic and made them an offer and they accepted. Or they made us an offer and we accepted. I can’t remember. It wasn’t complicated because they didn’t want to do it, and they knew I did.

 How did you come up with the photoaptamer production process initially?

Craig Tuerk was this wonderful grad student and incredibly smart guy. He was working on this basic problem and there was a moment when he became really interested in a little RNA structure, a little hairpin with eight nucleotides in a loop. He said, ‘I’m going to make lots and lots of changes in this loop because that loop seems to be involved in binding to some target protein. I’m going to do this expensive thing; I’m going to make three changes in every one of those eight positions, from an A to a C, G, or T,’—that sort of thing. Because I was a professor and he was a grad student I got to be a [jerk], and I said, ‘Craig, my friend, what about all the double [mers]?’ This led to my saying, with great smugness, ‘You’ll actually need to make 65,000 [different oligonucleotides] because you have to make 48 [different variations].’ So he made the pool of 65,000 sequences, and challenged that pool with a protein that we knew bound to at least one of those 65,000 sequences. He found to his astonishment — and mine — that there was a second sequence that bound exactly as well. So there were two sequences out of 65,000 loops that bound the same — the wild-type and one that had four changes in those eight nucleotides. If you were trained as a classic geneticist, the chance that you can get four changes in eight nucleotides over 4 billion years is zero. That just doesn’t happen. So this is probably a sequence that never got tried in nature, but was just as good as the one that he had been studying. This was just mind-boggling. As it says in an article I wrote a few years ago, the moment that he got this data, we stood in front of a white board in my lab at the university, fighting for the magic marker, filling up the board with every single idea we had. Out of that came NeXagen, NeXstar, and a patent, because we basically had [concluded] that everything you could do with antibodies, you could [also] do with single-chain oligos. It was wonderful because whoever had the magic marker got to write things on the board. And eventually we got disgusted and went off and got our own magic markers!

You’ve said that Pat O’Farrell worked in your lab. How did that happen?

His ex-wife Patricia Zambryski O’Farrell was my first grad student at the University of Colorado in about 1971. [Pat O’Farrell] is unbelievably smart. I was running tons and tons of 1D slab SDS gels, and [Patricia] was running them [as well]. Pat came down one day and said something like, ‘Well, they just threw my advisor out.’ The department had just decided not to give his advisor tenure, but he was so smart nobody was going to be his new advisor. They were all afraid of him. He essentially did a PhD thesis without an advisor, and he did half of it sitting in my lab with [Patricia] and me. He just started running these 2D gels — he’d figured it out. It was unbelievable. He ran at that time 2D gels that were as good as the best 2D gels you’d ever see. Nothing’s changed. They get bigger, they get smaller, they get wider, they get shorter, but you know, they still look terrible!

But you weren’t doing large scale protein analysis at the time?

No, I was actually looking at bacteriophage-infected E. coli on a 1D SDS gel. It was exactly what [people are] doing now [with 2D gels], except that you were looking at 1D gels because 40 or 50 bands of proteins were good enough to get a handle on what was happening in that system with time. What we had was lots of genetics and we knocked various proteins and ran gels that were missing one band. So it was exactly the same game as today but studying a different problem. So actually my life hasn’t changed at all in 25, 30 years! I think I’ll go play golf!

The story of course is that for every problem there’s an amount of resolution you need to understand a problem. For a T4-infected E. coli, a 1D SDS gel is just fine. For [looking at E. coli with or without their lac operon expressed as a system], Pat O’Farrell said, “Gee, I think that’s not enough resolution, I’d better make a second dimension.’ And then for the human proteome, it’s clear you’ve got to go ultimately from 1,000 spots to whatever you think the proteome is: 10,000, 50,000, or 100,000. And since you’re not going to do a 3D gel [and study] a big cube of acrylamide, you’re going to have to do something like an array.

The Scan

Could Cost Billions

NBC News reports that the new Alzheimer's disease drug from Biogen could cost Medicare in the US billions of dollars.

Not Quite Sent

The Biden Administration likely won't meet its goal of sending 80 million SARS-CoV-2 vaccine doses abroad by the end of the month, according to the Washington Post.

DTC Regulation Proposals

A new report calls on UK policymakers to review direct-to-consumer genetic testing regulations, the Independent reports.

PNAS Papers on Mosquito MicroRNAs, Acute Kidney Injury, Trichothiodystrophy

In PNAS this week: microRNAs involved in Aedes aegypti reproduction, proximal tubule cell response to kidney injury, and more.