"People sometimes refer to it as drug hunting. It starts with early discovery, finding interesting compound libraries and screening for hits, followed by a lot of intensive medicinal chemistry, microbiological and biochemical characterization and preclinical testing. Once you have an IND candidate, the focus expands to include process chemistry and toxicology testing. And that’s before you even consider human clinical trials. It’s a long process, but terrifically engaging. You have to balance optimism with realism. It takes great teamwork; to succeed, your team has to be smart, tenacious and lucky in about equal measures."
Thale Jarvis, Ph.D.
Vice President, Research & Development at Crestone, Inc.
Dr. Jarvis is co-Founder and Vice President of R & D at Crestone. She brings 22 years of experience in biopharmaceutical drug discovery and development, and is a co-author of 40 publications and over a dozen patents. Dr. Jarvis serves as Principal Investigator on several grants supporting preclinical drug discovery at Crestone, and as scientific liaison with NIAID for Phase I clinical studies of CRS3123 for treatment of Clostridium difficile infections. From 2009 to 2012, Dr. Jarvis worked at SomaLogic as Director of Technology Development (part-time), and led a structural biology program to characterize modified DNA aptamers with diverse therapeutic and diagnostic applications. Before joining SomaLogic, Dr. Jarvis was Senior Director of Biochemistry at Replidyne (2002-2008), where she played an integral role in multiple antibacterial drug discovery projects, and contributed to both IND and NDA regulatory filings. Her group was responsible for high throughput target-based screening for antibacterial drug leads, and structural and mechanistic characterization of enzyme inhibitors. In early 2000, Dr. Jarvis co-founded Impact Biosciences, where she served as VP R&D, focusing on target validation in mammalian cell culture systems. Prior to Impact, she served as Associate Director of Biology at Ribozyme Pharmaceuticals, Inc. (RPI) and its spin-off, Atugen USA. At RPI, Dr. Jarvis led discovery efforts for oligonucleotide-based therapeutics in the areas of oncology, arthritis, cardiovascular disease and virology, resulting in two programs that progressed into clinical development. She also served as project leader and scientific liaison for several projects with corporate partners. At Atugen, she pioneered the development of Atugen’s GeneBlocTM technology for the specific regulation of gene expression, and implemented high throughput screening strategies for validating oncology targets in a variety of human tumor cell lines. As a research scientist at Synergen (1992-1993), Dr. Jarvis worked on engineering protein-based therapeutics for inflammatory diseases. Dr. Jarvis currently serves as an advisor to the Seattle Structural Genomics Center for Infectious Disease. Dr. Jarvis received her BA degree with distinction in Chemistry from Carleton College in Northfield, MN. She received her PhD in Chemistry from the University of Oregon, where her doctoral work focused on mechanisms of bacteriophage DNA replication.
Interviewed by Collaborative Drug Discovery, Inc.
What makes C. difficile, excuse the pun but I can’t resist, so difficult?
It was discovered in the 1930s by a couple of microbiologists from Denver (Hall and O’Toole, Am J Dis Child. 1935;49(2):390-402) who were studying bacteria in the intestinal tracts of newborn babies. They named it “Bacillus difficilis” because it was so difficult to isolate and grow in the lab. The name was later changed to Clostridium difficile. It’s an apt name for a pathogen that is so difficult to treat.
Do we know anything about how C. difficile wins the race to populate the intestinal track following a broad spectrum antibiotic? Do we know anything about how compounds selectively hit C. difficile, or is it simply an observation of differential activity from phenotypic screening where perhaps later we will understand how it works?
C. difficile produces spores that survive antibiotic treatment. When a person is treated with a broad spectrum antibiotic, it wipes out all the normal bacteria in the gut and the C. difficile spores spring into action - germinating and spreading the infection. Without competition from normal (friendly) bacteria, C. difficile overgrows and can cause severe diarrhea and even death. Many antibiotics are very broad spectrum and therefore are not selective for C. difficile. This includes vancomycin, which is one of the drugs that is approved for treatment of C. difficile infections (CDI). We believe that a better approach to treatment of CDI is development of narrow spectrum antibiotics that are more selective for C. difficile.
The selectivity (spectrum) of an antibiotic is determined by several factors, including whether the drug target is present in the bacterial species, whether the drug can penetrate the bacterial cell wall, and whether the drug is subject to efflux (being pumped out by the bacterium). We are developing CRS3123 for treatment of CDI. CRS3123 stops C. difficile growth by blocking a particular enzyme, MetRS, which is essential for protein synthesis. It turns out that most normal intestinal bacteria have a different form of MetRS that isn’t susceptible to inhibition by CRS3123. Thus, the microbiological selectivity of CRS3123 is a direct result of differences in the target enzyme in different bacteria - so in a sense, we are taking advantage of this kind of serendipity.
At Crestone, you’ve managed to simultaneously advance both preclinical and clinical projects, how have you done it? How is the CDD Vault helping you to run these projects more cost-effectively?
Our clinical program has benefited from tremendous support from NIAID Division of Microbiology and Infectious Disease. They have sponsored the IND and managed the Phase I clinical studies, enabling this promising strategy for CDI treatment to progress into clinical development at a time when it was extremely difficult to raise investment money for an early-stage antibiotic drug development program.
(Editor's note: NIAID Division of Microbiology and Infectious Disease presented recently at the CDD 10th anniversary community meeting. Take a look at the presentation slides for more information on securing funding.)
Our other main program, which is a preclinical stage project, has also received critical support through the Small Business Innovative Research (SBIR) grant program. This program involves medicinal chemistry lead optimization of novel DNA polymerase inhibitors that are effective against Gram-positive bacteria.
The CDD Vault is a truly essential tool for late stage lead optimization, allowing us to store and retrieve chemical structures and associated data quickly and easily. It is so powerful and yet so user-friendly. We’ve used other database platforms in the past, but these would be cost-prohibitive for a small company. In addition, they were very complex to use, meaning that valuable time was spent on specialty training so that just a few researchers within the group could even use them. The CDD platform is very intuitive and allows us to keep a large body of data at our fingertips.
At CDD we’ve collaborated with a number of researchers working on TB via our collaborations with NIAID, BMGF, MM4TB and others in the antimicrobial community. You’ve recently begun working on a project targeting nontuberculous mycobacterium. How are they similar or different than TB? Why are they important?
Nontuberculous mycobacteria (NTM) are sometimes referred to as environmental mycobacteria because these bacteria are prevalent in wet soils and river systems; unlike tuberculosis (TB), which is transmitted from person to person, NTM infections come from exposure to pathogens that are present in the environment and result in lung or skin infections in immune-compromised patients. Incidence of NTM infections is rising. Although TB is more prevalent worldwide, in the US there are now more cases of NTM than TB. Unfortunately, therapeutic agents developed to treat TB infections often lack activity against NTM and treatment options for NTM are severely limited. While TB has been the subject of extensive drug discovery efforts, there are virtually no antimicrobial drug discovery programs specifically targeting NTM. Thus, NTM pose a unique and under-resourced therapeutic challenge.
You’ve been able to identify compounds active against all clinically-relevant Gram-positive pathogens? Can you share something interesting, whatever you can, about the discovery or mechanism of action of your compounds?
We have an active program geared at optimizing compounds that inhibit PolC, the replicative DNA polymerase in Gram-positive organisms. It is a novel target for an antibacterial agent, which means there is no pre-existing natural resistance. Consequently, our compounds have demonstrated activity against all clinically relevant Gram-positive pathogens, including methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistance enterococci (VRE), and penicillin-resistant Streptococcus pneunomiae (PRSP) as well as biodefense pathogens such as anthrax. Our compounds bind to PolC and DNA in a 3-way complex that prevents further DNA synthesis. A stalled replication complex triggers cell death, so the compounds are bactericidal.
It’s kind of ironic how we discovered this series. We initially screened a small library of compounds for inhibition of bacterial DNA polymerase holoenzyme and were able to identify a number of inhibitors, but none proved suitable for further development as therapeutics. Put simply, we came up empty-handed. We then performed whole cell screening on the same library, and found three related compounds that were bactericidal against Gram-positive bacteria. Target identification studies led us right back to DNA replication, and we were able to show unequivocally that PolC was the target. In retrospect, we realized that our enzyme screening assay wasn’t quite sensitive enough to pick up all relevant inhibitors. In particular, it wasn’t geared toward detecting inhibitors that are competitive with nucleotide triphosphates (as was the case with this series). It turns out that PolC is one of those targets that exhibit high “target vulnerability”. That is, even modest inhibition at the enzyme level translates into significant effects on cell viability.
What is the end goal (modest and more audacious) for each of the three projects you’ve mentioned? What are the means to the end?
Really the goal is the same for all three projects: we are trying to develop novel life-saving therapies to combat serious bacterial infections, including those caused by antibiotic-resistant bacteria.
People sometimes refer to it as drug hunting. It starts with early discovery, finding interesting compound libraries and screening for hits, followed by a lot of intensive medicinal chemistry, microbiological and biochemical characterization and preclinical testing. Once you have an IND candidate, the focus expands to include process chemistry and toxicology testing. And that’s before you even consider human clinical trials. It’s a long process, but terrifically engaging. You have to balance optimism with realism. It takes great teamwork; to succeed, your team has to be smart, tenacious and lucky in about equal measures.
What new techniques have you found particularly interesting?
I’m a big fan of crystallography, and have participated in several fascinating collaborations with Emerald BioStructures that led to drug-target co-crystal structures…a picture is worth a thousand words! Although crystallography itself is not new, there have been a number of recent technological and strategic advances that have greatly accelerated the process and expanded the range of tractable targets.
On a completely separate note, I’ve been intrigued with genome-wide association studies (GWAS). This is for sure not my field, but I’ve enjoyed following the emerging information about the relationships between genetics and disease susceptibility. It’s one of the first steps toward personalized medicine.
Before you used the CDD Vault, how did you manage your data or collaborate? Why did you choose CDD Vault and how do you use it currently?
We’ve used Activity Base at a previous company, but when we started at Crestone that was not an affordable option. Until we started working with CDD, we were basically storing information in Excel files, which have very limited search options. We chose CDD Vault based on the recommendation of a trusted colleague.
We use the CDD Vault as the primary database for storing compound structures and all of the associated data. We try to upload as much of the data as possible in order to maximize the potential for searching. We also attach relevant reports so that search results can lead to more in-depth information. As projects get bigger, it’s too easy to lose track of what’s been done unless you have a tool like CDD's Vault.
Could you share the top 2-3 things you like about the CDD Vault? Where would you like us to take the CDD Vault platform next?
I especially like being able to do substructure and related compound searches. I also frequently do multi-parameter searches, like searching for compounds that show a particular level of microbiological or enzyme inhibition, and combining that with another parameter like protein binding, metabolic stability or solubility. I also like the recent improvement to the IC50 curve fitting functions that make it easier to fix or float the upper and lower boundaries.
It would be great to have more graphing functionality - the ability to plot one experimental value vs another (maybe even in 3D?) and to be able to click on a data point and pull up the compound information.
Can you share a bit about your interactions with CDD as a company over the years?
When we first came to CDD, our main project had over a thousand compounds, with dozens of endpoints for each compound. Uploading this amount of data, and maintaining integrity, was a big task. The staff at CDD really knew their product, and took the time to walk us through the startup process. Since then, whenever we’ve encountered a question, CDD has always responded promptly with a solution.
Besides CDD Vault, what technologies do you use for your research?
We do a lot of plate-based assays for microbiological and enzymatic activity testing of compounds, often using fluorescent or colorimetric endpoints. It’s just an efficient way to generate a lot of valuable data in a controlled, consistent fashion…and activity data is the “A” in “SAR”, right? There other important assays that we can’t currently perform in house because we don’t have the necessary equipment or facilities – things like metabolic stability or in-vivo testing. We use specialty CROs for those studies.
What was a memorable interaction you've had w/ a brilliant scientist?
I’ve had the privilege of working with some fantastic scientists over the years. My Ph.D. thesis advisor, Dr. Peter von Hippel, was a wonderful mentor and inspiration. He is particularly adept at analyzing biological problems from a fundamental biophysical viewpoint, which often provides surprising mechanistic insights. Seemingly complex macromolecular interactions can be rationalized based on fairly simple underlying forces. Pete always advocated the use of “Occam’s Razor” (favor the simplest theory that explains the data), and liked to say that “biological systems are teetering on the brink of instability”, meaning that biological macromolecules would not be effective if they existed in deep potential energy wells (i.e., a state of high thermodynamic stability). In other words, dynamic, responsive biological systems require that small changes in input can lead to substantial consequences. An example would be the profound effect on signal transduction that can result from a single phosphorylation event.
How do you work collaboratively within your group? Talk about your collaborations with others - what works well, what could be better?
Team work is completely essential to the success of our projects. Because we’re a small company, everyone is used to pitching in wherever they’re needed, and that helps keep things running smoothly. We have regular team meetings, but to be honest a lot of the key communication just happens informally in the lab or over the lunch table. With outside collaborations, it really helps to start with face-to-face discussions. After that personal connection is made, we use email and teleconference calls to stay on top of project details.
How have you managed to successfully and reproducibly secure grant funding for your projects? What has been the key? What have you learned from the process (both from a writer and reviewers perspective)?
All of our projects have benefited from a strong set of supporting data which helps to address feasibility and likelihood of success. In grant applications, it important to articulate how your project addresses an unmet medical need. With bacterial infections, emergence of resistance and increased virulence present ongoing challenges to the medical community, so it’s easy to make a case for significance. Reviewers also like to see that you’ve thoroughly analyzed the potential pitfalls in your strategy, and that you have alternative strategies to overcome problems.
Talk about an ah-ha moment in science?I had one project where I was trying to identify one of the essential DNA replication proteins in Pseudomonas aeruginosa (J Biol Chem. 2005 280:40465). It wasn’t obvious based on homology to other bacterial species, so we isolated a native DNA polymerase complex through classical protein purification guided by functional assay. This gave us a hot prospect for the missing protein, which we identified by peptide mass fingerprinting as match to a “hypothetical protein” from the PA genome sequence. The only problem was that the native protein was much too big to be explained by the published ORF sequence. To make a long story short, it turned out that the mystery protein had a non-canonical translational start site, starting at an upstream UUG instead of AUG. The data were staring us in the face, but we had to think outside the box to understand it - that was definitely an “ah-ha” moment.
Outside of your own research, what is the most fascinating development or study you've recently seen?
When the human genome was first sequenced (not so very long ago), one of the surprising findings was that an enormous percentage of the genome did not code for proteins, and hence appeared to be “junk”. What has emerged, thanks in part to insights from scientists such as John Mattick (EMBO reports, 21:986, 2001) is a vast new area of study on a myriad of different non-coding RNAs. The significance of non-coding RNAs in biological systems cannot be understated, and the elucidation of this new field of biology has challenged dogma that was firmly accepted until quite recently. It’s a bit reminiscent of the revelation of thirty years earlier that mRNA expression was actually regulated, both temporally and between different cell types. It’s been fascinating to follow this nascent field of study, and to reflect on the immensity of the biological jigsaw puzzle that we’re collectively trying to solve.
Can you share a bit about your industry experience prior to Crestone and what you learned and how it has helped?
My work has always focused on developing therapeutics, but the modality and disease area has varied. I’ve done protein, oligonucleotide and small molecule drug discovery. My first position in industry was working as a protein chemist at Synergen, trying to modify cytokines to make receptor antagonists for the treatment of various inflammatory conditions. I then went to Ribozyme Pharmaceuticals, which was founded based on the discovery of catalytic RNA by Tom Cech. As a platform technology for down regulating RNA expression, it offered an opportunity to tackle a wide variety of targets in virtually any therapeutic area. I worked on projects in cardiovascular disease, arthritis, cancer and infectious diseases. In this process of optimizing ribozymes, we actually found ways to make better antisense oligonucleotides, and the company eventually changed focus (and name) to siRNA. Around the time the human genome sequence was published, I co-founded a start up functional genomics company called Impact Biosciences, which was geared at developing a database of phenotypic effects of antisense-based gene knockdowns in various mammalian cell culture models. That was a tough business model, however, so I eventually turned to other opportunities. This included work at SomaLogic, researching diagnostic applications of aptamers. In 2002, I took a position at Replidyne doing small molecule drug discovery of antibacterials. Both of Crestone’s programs grew out of the preclinical discovery efforts at Replidyne, and benefited from the wealth of experience from the team at Replidyne, ranging from early stage discovery through clinical and regulatory. We were fortunate to be able to license the programs after Replidyne discontinued its antibacterial discovery activities.
Similarly, can you share a bit about your formative educational experiences and your different experiences have influenced your working career?
Initially I wanted to become a veterinarian. I attended Carleton College as an undergraduate, and since there was no “pre-vet” major, the logical choice to satisfy a lot of vet school entrance requirements was to major in chemistry. Carleton had an outstanding chemistry department, and along the way I became so enamored with chemistry that I ended up going to graduate school instead of vet school. I attended graduate school at the University of Oregon and earned my Ph.D. in Chemistry with Dr. von Hippel, studying mechanisms of DNA replication in phage T4. Our lab was part of the Institute of Molecular Biology, which was a wonderful place to train. The stimulating yet collegial environment at Oregon confirmed the notion that it’s possible to do great science in an atmosphere that is collaborative, respectful and fun. I did my post-doctoral fellowship in Dr. Karla Kirkegaard’s lab at the University of Colorado, focusing on mechanisms of recombination in RNA viruses. Although my subsequent industry experiences (see above) have spanned many different therapeutic areas, it’s kind of fun that my current job at Crestone (studying DNA polymerase inhibitors as antibacterial agents) has brought me back to my scientific roots, both as a grad student studying DNA replication and as a postdoc studying infectious disease.
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!