Ellen Berg, Ph.D. – General Manager – BioSeek Inc.
Ellen Berg was a cofounder of BioSeek, Inc. and served as its Chief Scientific Officer. She was appointed as General Manager, Bio Seek LLC after the merger of BioSeek and Asterand in 2010. Dr. Berg has more than 20 years of research experience in pathophysiologic mechanisms of inflammation and immunity. Her expertise in complex cell-based biological assays led to the development of the company’s proprietary BioMAP® technology for target validation and drug characterization using primary human cell systems. Prior to founding BioSeek, Inc., Dr. Berg was at Protein Design Labs, Inc. where she was involved in the discovery and development of therapeutic antibodies for treatment of inflammatory diseases. Dr. Berg holds a Ph.D. from Northwestern University and was a postdoctoral fellow at Stanford University where she was supported by fellowships from the American Cancer Society, the Stanford Cancer Biology Program, and the Leukemia Society of America. Dr. Berg holds a number of patents in the fields of inflammation biology and cell adhesion, and has over 60 publications in peer-reviewed journals.
“I was always interested in sort of the bench-to-bedside areas; how do we do a better job of finding new drugs that are actually going to work in people safely.”
Interviewed by Barry Bunin, PhD, CEO, Collaborative Drug Discovery, Inc.
The CDD Spotlight gives a chance for us to profile some of the scientists that we interact with, and so I was just going to start with a little bit of your background and the research that you’ve done in the past and that you’re doing currently, obviously the focus on BioSeek and Asterand.
The research we do here at BioSeek is focused on building cell-based assays that we could use to screen compounds and new materials and be better positioned to predict their activities in people, so that’s a tough hurdle.
Our approach has been to put together cell cultures, co-cultures in complex environments that do a better job of modeling or mimicking disease environments, so they’re not simple assays. They tend to be biologically fairly complex, and what we’ve discovered is that by putting together the right combinations of environmental factors and cell types, once you’re able to do a better job of modeling the in- vivo biology, it’s actually more reproducible than if you are looking at, for example, a pathway in isolation or a signaling pathway.
We have taken that discovery of complexity is actually more reproducible and developed a platform of assays where we have built a database of reference compounds- biologically interesting agents, and profiled them through our panels and suite of assays. Over the years now we have developed this database that we can use to compare new compounds and learn a whole lot more about the mechanisms of action of drugs that are currently on the market that we don’t really understand why they work, which has led us to new ways of looking for new agents.
We do screening, we do mechanism of action studies, and we help folks translate molecular information, biochemical level information, assay data into information that they can use to go into animal models or actually predict how their compound is going to work in the clinic, whether it’s going to be active at all, is it metabolically stable, reasonably stable, and what sort of activities can one look at in the clinic.
We cover a lot of different areas of biology, all kinds of targets: GPCRs, kinases, nuclear hormone receptors, et cetera, and a wide range of biology as well, auto immune, inflammation mechanisms, tissue remodeling mechanisms, and development and differentiation.
That’s really the work that we’ve been doing here at BioSeek – for now over ten years. Before that, I was in therapeutic antibody discovery and development, so I was always interested in sort of the bench-to-bedside areas; how do we do a better job of finding new drugs that are actually going to work in people safely.
So that’s a big, bold, audacious goal of trying to mimic people better and something that people have been trying to crack into drug discovery for a long time and so I’m curious about some more details in terms of the means to the end – which techniques or approaches have you leveraged or found interesting as you’ve made progress against this tough goal.
The key for us was the realization early on that you had to have high throughput data because there are so many variables to look at, being able to do cell-based assays at scale made this possible.
There’s a lot of variation between people, so we had to have a technology where you could compare a lot of different donors – all at the same time. You’d want to look at a lot of different biological activities in each assay and do it multiple times, 8, 10, 12 replicates, so you understand: where is the reproducibility, where’s the variability?
High throughput technologies were key and advances in the medias and the culture hardware or culture materials to grow various different human cell types. There’s been a big advance over the last 15, 20 years in that. Those were key and all the tools that we use for data analysis. It’s just been absolutely wonderful.
Based on what you said about cell-based assays from lots of donors, I was wondering where you fall in terms of the debate between patient stratification of drugs versus understanding the system well enough to have one drug that hits a larger percentage of the patient population, as you’ve been looking at more data than other people typically do.
Yes and what we see, especially in our systems, is we now have a better appreciation for idiosyncratic types of outcomes or activities. Why are only some patients – why are 30% of patients improved for many therapies, right? They only work on certain patients and there are so many factors – in addition to genetics, there’s just co-morbidity whether or not they have another underlying disease, whether they’re suffering from the flu, all these kinds of variables will impact the outcome, whether it’s an efficacy or a safety outcome.
It’s the combination of effects that has made it so difficult to stratify patients. It’s not just the genetics and it’s rarely just the genetics. We’d like to believe that one can use genes to stratify patients, but I think one of the things we’ve discovered in sort of the investigative toxicology area is that many of the toxicities that are observed are idiosyncratic, and that if you could push the dose up for all patients, they’d all eventually be affected by the toxicity. Also when patients are having some sort of ongoing inflammation, that will completely change their response – whether or not they’ll have an adverse event.
The more data we have on patients, the more I think will help us figure these things out and be able to do a better job at the prescribing, so if you have this or this, don’t take this drug in addition to the genetic considerations.
The one thing I will say about research with patient stratification is it’s really expensive, so we need to do a better job of funding that type of work. It just gets very, very expensive to get approval for use of patient materials and to just setup the experiments – all the paperwork that has to be done – and you have to protect the patients and make sure that they’re informed, and that’s expensive. Somehow we have to make that easier to do.
Sounds like that’s a potential future challenge for us at CDD to consider, which we haven’t yet… It’s interesting to think about it. I recently stopped by your labs, so I can say with confidence that you’ve been through a successful merger and acquisition of BioSeek and that everyone’s working really well and pleasantly together there. I’m just curious about how you work collaboratively within your group, between groups, as well as some of your collaborations with others. I know you’ve announced collaborations with the EPA. You have a number of drug discovery collaborations – what’s worked well, what could work better, what are some of the sort of challenges and opportunities associated with “playing well with others” and collaborating?
For us, we’re committed to scientific excellence, and I think that’s helped us a lot. We’ve tried not to over-promise what we can do for our collaborators, be open about the science. This is always, always the best approach. In terms of collaborators, we’re very big on teamwork and it comes from the fact that we’re biologists here and we really understand that no one person can know everything, so we really depend on good working relationships with our clients. They get the most out of what we do by working closely with us and telling us as much as they can about their problem, we find that it’s sort of helps us put all the pieces together for investigating a mechanism of action. Often you just have to have all the information to be able to nail that down, and close work relationships help us be successful as well and for our clients, so do the best science is always our goal.
I wanted to talk about an interesting “ah-hah” moment in science. It could be in the industry or academia or where there’s just something outside the ordinary day-to-day work that was interesting and worth sharing, that made your head spin a little.
Actually for us, it’s very frequently because not very many people look so broadly in drug mechanisms of action, and we learn something new almost everyday. Well, I wouldn’t say everyday, but frequently someone is running out to the middle of the room and saying, “Oh, oh, such and such does this, and we didn’t know it and it explains everything,” so those events, those discoveries are really exciting for us. I mean that’s what I love about what we’re doing, is that it’s important to medicine- understanding drug mechanisms of action.
Outside your own research, we’re all going to talks and reading the literature, what’s been a fascinating development or study you’ve seen, ideally something that would be of interest to other curious scientific brains out there?
I certainly follow all the stem cell work and I’m just absolutely fascinated by some of the advances, so if any one thing that stands out, just that whole ability to use drugs, if you will, or small molecules to induce differentiation into various cell lineages, because in the past, we always were looking for the natural factors that drove differentiation, right?
Vitamin, different vitamins or retinoic acid, things like that, instead of actually looking for new small molecules that might do the same thing – because that has been something of a Holy Grail to be able to have in cell culture, every cell type; (to mimic) a patient in a test tube and just the ability to control that – that was probably one of the more exciting things that I’ve been following in the last year or so.
Interesting. I remember reading a review by Peter Schultz and others about the different small molecules that can affect stem cell differentiation, and at CDD, we have this growing balloon of public data, with known drugs, gene-family wide, compounds right off the shelf, data that would generally be of interest to researchers to look at for the first time securely with their own data. And one of the things that I’ve been debating in the back of my mind, especially after the California initiatives for stem cells, is if it would be worth aggregating the information on all of the different small molecules and how they impact stem cell reproduction and differentiation? So that all those molecules could be in one place, it’s not a huge set – it’s pretty finite actually.
Oh, absolutely. It’s certain pathways and combination of pathways that you’re activating or whatever you’re doing and, yes, I think that would be a great resource.
Most of my questions have been general and focused on you, but we are CDD and we’re a company as well, so I’m just curious – share why you initially chose CDD, where the need was and the interest for collaborating?
And still today, I don’t see any competitors. I just don’t see anyone doing what CDD does, do you? Do you have competitors, to do exactly what you need (to securely collaborate while keeping other data private in one system)?
Our exposure to CDD came about when we were working on a project with the EPA where we were screening a library of compounds and generating data as part of a contract. It occurred to us, as we were shipping or emailing Excel files to them that it would be really great to have one (place to have aggregated information over time) – to be able to share this type of data more easily through the Web, on a Web-based interface, that was private because of course we were aware of PubChem, which is the government sponsored, publicly available website with chemical and assay data, but it was very cumbersome to use and cumbersome to sort through.
It was, in the case of our contract with the EPA, they wanted to do the quality control on the data and look at it internally before they actually published it to the public because they wanted to make sure it was good, that everything was carefully checked, et cetera, and so I saw an obvious place for the kind of software tool that CDD has for sharing privately and that uploading all the data, getting it and then flicking a switch to make it public. And that was my kind of “ah-hah” moment about where we got involved with CDD, so we sort of worked on some ways of importing our data and then sharing it with the EPA.
I still think today there aren’t good ways to share data, but, of course, we as a small company, we’re innovators. We like to be on the cutting edge. We’re still waiting for more folks to jump on the bandwagon with CDD and it certainly useful for small companies and expanding what CDD does with the government makes sense. I think we’re still at the tip of the iceberg for the need for this.
This blog is authored by members of the CDD Vault community. CDD Vault is a hosted drug discovery informatics platform that securely manages both private and external biological and chemical data. It provides core functionality including chemical registration, structure activity relationship, chemical inventory, and electronic lab notebook capabilities!
CDD Vault: Drug Discovery Informatics your whole project team will embrace!