From the Ground Up

By Jason Socrates Bardi

Ten months ago, Professor Ed Roberts was sitting with one of his colleagues, Associate Professor Subhash Sinha, discussing possible future collaborations in Roberts' new office when Roberts' assistant interrupted them. She was amused to see that they had eschewed any temporary furniture to sit cross-legged Indian-style on the floor of his office.

"I really have been starting from absolute scratch," Roberts said, laughing.

As a newly appointed professor of Translational Chemistry and Medicine at The Scripps Research Institute in California, Roberts has been creating his laboratory from the ground up. Since February of '05, he has been writing grant applications and discussing possible collaborations with his new colleagues.

Partnering Chemistry and Biology

Roberts, who is a fellow of the Royal Society of Chemistry and of the Royal Society of Medicine, has experience in industry that runs deep and extends over a career that spans a few decades.

After completing his education in Great Britain, Roberts started out working in a small group at Addenbrookes Hospital in Cambridge, United Kingdom, set up by Warner Lambert. He worked his way up to head of Strategic and Medicinal Chemistry. Then, he left to begin a similar group with Astra (now AstraZeneca) in a facility located in Montreal that initially consisted of an old converted barn with a few fume hoods. "We built, in the end, a modern state-of-the art building and a very successful drug discovery unit," Roberts says.

From there, Roberts went to F. Hoffmann-La Roche in Basel, Switzerland, as senior vice president and head of Discovery Chemistry, overseeing a division of hundreds of chemists working in many disease areas and producing a prolific amount of research. It was the first time all the chemical research at Basel-based Roche had been under a single leader for over 25 years. During his tenure, the division moved into a newly built large glass structure dubbed the "The Crystal Palace," designed by the Pritzker Prize-winners Herzog & de Meuron. "The building, with a complete glass frontage was so bright at night, we had complaints from nearby residents that they could not sleep!" he recalls.

Roberts is known for his integrative approach to new medicine design and development.

At Roche, Roberts sought to forge an equal partnership between chemistry and biology. He saw that portfolio failure often occurred if there was an imbalance between the two disciplines when decisions were made. So Roberts promoted a partnership between chemists and biologists so they could work together on equal footing in the drug-design process, pairing biologically relevant targets with chemically feasible compounds. This led to marked increases in success rates and quality of output as clinical candidates entered development.

After Roche, Roberts came to La Jolla to start a company with his Scripps Research colleagues Tamas Bartfai and Julius Rebek. The company has been successful in a short time and its first orally active compound for inflammatory diseases such as rheumatoid arthritis is currently in clinical trials.

Tracking Down Elusive Targets

At Scripps Research, Roberts has been attending meetings and seminars in order to expose himself to the research programs that his new colleagues are pursuing.

Roberts' own research takes two approaches. He is particularly interested in small molecule modulation of interactions between human proteins and peptides involved in disease. Moreover, he is studying those targets that have eluded the pharmaceutical approaches to intervention. "This is not easy," says Roberts.

Targets that are under investigation include glucagon-like peptide -1 GLP-1 for diabetes and obesity; galanin for epilepsy and depression; vasopressin and oxytocin for pervasive developmental disorders (PDD) such as autism.

Thirteen years ago, the newly discovered neuropeptide galanin was shown to have anticonvulsant effects when applied into the ventricles. Galanin action has since been elucidated in cellular and molecular detail. The main elements of the anticonvulsant effects of galanin can be summarized as follows: a) Galanin acts at three GPCR-type receptors in the brain. These receptors are present in the hippocampal formation and in the Locus coreleus, DRN, and septum. The predominant galanin receptor types in the forebrain are GALR1 and GALR2. b) In the hippocampus, postsynaptic galanin receptors mediate hyperpolarizing actions in both in the dentate gyrus and CA1/CA3 regions. c) Galanin receptors localized presynaptically inhibit the release of glutamate but not that of GABA in the hippocampus and thus shift the excitation /inhibition dynamics towards inhibition.

To date, Roberts and his colleagues have shown:

• Galanin receptor agonists of peptide type inhibit both the initiation and maintenance phase of experimental seizures.
• Galanin-overexpressing animals have twice as high seizure threshold in the kindling model of epilepsy.
• When antisense oligonucleotides knock down GALR1 or GALR2 receptor subtypes, GALR1 inhibits initiation and GALR2 inhibits maintenance of seizures.
• Using GALR1 and GALR2 knockout animals, similar results were obtained regarding the role of GALR1 and GALR2 in control of seizures.
• Systemically active model compounds which are galanin receptor agonists—galnon-tripeptidomimetic (Barfai et al) and galmic, an oxazole backbone based peptidomimetic (Rebek et al)–both were active anticonvulsants in screening models for anticonvulsant drugs.

"Everybody has known about galanin for years," says Roberts. "Everybody has known what a great target it is, but nobody has produced drug-like small molecule agonists against it."

Why? Because although we know that galanin binds to its protein receptors on the surface of neurons, we don't know exactly how. Galanin binds as an alpha helix (Roberts et al.), but what part of the galanin helix binds to the receptor? Is it the end? Is it the face?

"Research needs to develop this area further," says Roberts.

Confronting Autism

Autism is a disorder Roberts is passionate about. He says, "I find it incredible that not more is being done by the pharmaceutical industry against a disorder that is approaching pandemic proportions. I believe we have an approach that is certainly not clinically proven but which gives hope to many. In this respect I am grateful to my TSRI colleagues Koob and Bartfai. We will work together on this program."

It is a fact that in the 10 years from 1993 to 2003 the cumulative growth rate of autism diagnosis in California was 1090%, compared to 32% for all other disabilities (ages 6 to 22). The figures for the United States suggest that 1 out of 160 births is autism-related. The neurochemical abnormalities in autism are largely unknown but animal studies have identified several neuropeptides in the basal forebrain, hypothalamus, and brainstem that have roles in social interaction and, as such, may be hypothesized to have a role in the social abnormalities associated with autism. These neuropeptides include vasopressin and oxytocin.

Roberts intends to tackle these problems chemically by drawing on information in the public domain to develop drug-like modulators of these proteins, which belong to a superfamily known as the G protein-coupled receptors (GPCRs) as well as screening for further intervening molecules. "We will also develop non-competitive ligands," says Roberts. "We will try to develop ligands that bind in an allosteric fashion, not to the site of the endogenous peptide ligand. In this way, we can take a newer approach to ligand identification, which would offer advantages over competitive inhibition or activation."

Two in One

Another area of interest for Roberts is blocking two targets within a signaling cascade (the various sequential biochemical reactions that underlie all of normal biology and disease). While scientists have known for years that it is often advantageous to block two different targets as a way of attacking a disease (Novartis's Gleevec, for example), two compounds means two separate drug development efforts, two sets of toxicities to deal with and so on. So Roberts is interested in a chimeric approach: finding a single compound that can hit two targets in a single cascade—like a key that opens two locks. Type 1 GPCRs often show affinity for a relatively small set of privileged structures. "We often see very close pharmacophores for very different receptors, and we can exploit this and our detailed biological knowledge to design new treatments that target two (or even more) proteins of interest," says Roberts.

With special focus on diseases like autism, depression, epilepsy, and pain, Roberts' goal for the next several years is to identify targets and determine whether or not they are relevant and amenable to chemistry. But a more immediate goal is just to get to know his new colleagues.

"I'm looking forward to establishing new friends and acquaintances and to working with people here in La Jolla as well as at Scripps Florida," Roberts says.

Send comments to: mikaono[at]scripps.edu

With special focus on diseases like autism, depression, epilepsy, and pain, Professor Ed Roberts is seeking to identify potential targets for intervention and to determine whether or not they are relevant and amenable to chemistry.