The “Century of Biology” and the Evolving Role of Medicinal Chemists

Viewpoint pubs.acs.org/chemneuro

The “Century of Biology” and the Evolving Role of Medicinal Chemists in Neuroscience Dario Doller* Alcyoneus/ScienceWorks, Sparta, New Jersey 07871, United States ABSTRACT: Society expects that the wave of contemporary new discoveries in biological sciences will soon lead to novel treatments for human diseases, including many devastating brain disorders. Historically, medicinal chemists have contributed to drug discovery teams in ways that synergize with those from their partner sciences, and help transform new knowledge into the ultimate tangible asset: a new drug. The optimal balance of resources and the right strategy to minimize the risk of late clinical failure may differ for different therapeutic indications. Recent progress in the oncology and neuroscience therapeutic areas is compared and contrasted, in particular looking at the biological target space and functional attributes of recently FDA-approved drugs and those in the late clinical pipeline. Medicinal chemists are poised to have major influence in neuroscience drug research, and examples of areas of potential impact are presented, together with a discussion of the soft skills they bring to their project teams and why they have been so impactful. KEYWORDS: Neuroscience drug discovery strategy, medicinal chemistry skills and opportunities, allosteric modulation, biological targets activation or inhibition

Probably the best definition of pharmacology is that which describes it as the study of the action of chemical substances upon living things. Pharmacological problems, therefore, fall into three groups, namely, those which relate (1) to the chemical substances (usually drugs or poisons), (2) to the living things (which may be anything from simple cells to highly complex organisms), and (3) to the reaction between the one and the other. Evidently, then a pharmacological problem without a chemical bearing must indeed be superficial.

■

drug. Therefore, there is a chemistry viewpoint that genuinely belongs in this effortone that goes beyond making the compounds for testing and qualifying them as having the right physicochemical, biochemical, pharmacokinetic, and toxicological properties. The identification and derisking of novel disease targets has become a major focus of life science research. These are biology-centered activities using the most sophisticated and constantly improving technologies, and aimed at establishing confidence that the right modulation of the right target will lead to an efficacious therapy. However, sometimes these efforts tend to defer the integration of the effects of the drug molecule in the biological system under study. Indeed, oftentimes the (unqualified) assumption is made that when a target eventually becomes derisked by biology standards, the medicinal chemists will somehow provide the tool compound in the desired modality, that would close the loop and deliver preclinical proof of concept. In a rather simplistic way, compounds are grouped based on equivalent “class effects”, such as receptor antagonists, enzyme inhibitors, etc., and each class would lead to equivalent clinical pharmacology. The narrative goes that when sufficient unbound drug concentrations at the appropriate biophase ensure target engagement, the compound acts as a light switch, enabling a change in the target’s functional activity that is hypothesized to be linked to the dysfunction. Thus, under this paradigm, medicinal chemistry is somewhat of a stand-alone component of the project drug discovery strategy.

Henry G. Barbour, 1923

THE CHEMISTRY CHAPTER IN THE “TEXTBOOK OF HUMAN BIOLOGY” While the 1800s witnessed major progress in chemical sciences, and the 1900s were the physics era, we are now in the “Century of Biology”. The promise is that of “a complete description of life at the most fundamental level of the genetic code”.1 The inference is that the understanding in human biology gained from such description will produce a road map to treat disease. If anything, recent work has shown us that human biology is as amazing as it is complex. Findings from genomics, epigenetics, proteomics, metabolomics, to name a few areas, suggest that developing a blueprint to enable the treatment of human disease necessitates major progress in f undamental sciences well beyond those listed aboveincluding some not yet known. The massive scope of these tasks often requires establishing consortia. This information would then be integrated using system-based approaches to create novel testable hypotheses, possibly moving away from simplistic linear models that dominate current thinking. Macromolecules such as proteins, nucleic acids, and complex carbohydrates are components of the systems under interrogationand so are lower molecular weight chemicals such as endogenous amino acids, lipids, ions, and most importantly the (small-molecule) © XXXX American Chemical Society

Received: November 19, 2016 Accepted: November 22, 2016

A

DOI: 10.1021/acschemneuro.6b00394 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Viewpoint

ACS Chemical Neuroscience

■

SMALL MOLECULE DRUG DESIGN IN NEUROSCIENCE: IDENTIFYING THE OPPORTUNITIES Every single new drug discovered, in every indication, is a major accomplishment reflecting the ingenuity and dedication of a large number of scientists working together. However, at present times, not all therapeutic areas are equally favored, as high failure rates in late clinical trials have led to reduced investment in brain disorders.2 A recent analysis showed that 68% of FDA-approved drugs for neurological disorders target either G protein-coupled receptors (GPCRs) or ion channels.3 For psychiatric disorders, GPCRs, ion channels, and transporters constitute 90% of the targets for drugs approved.4 These membrane-bound biological targets are key in enabling the discovery of drug treatments for CNS diseases, albeit at a relatively low average rate of 2 new chemical entities (NCEs) per year for psychiatric and neurological disorders combined. The oncology field is having a much higher rate of FDA-approved new molecular entities (NMEs), at around 10/year. Small molecules targeting kinases in general, and tyrosine kinases, in particular, constitute around 50% of these NMEs.5 Thus, at the present time most CNS small molecule drugs lie in a dif ferent biological target space than those in oncology. In addition, the vast majority of FDA-approved oncology drugs targeting kinases are f unctional inhibitors targeting the ATP binding site. A strong arsenal of knowledge and technologies has developed specific to this area, and methodologies like structure-based drug design (SBDD) and fragment based drug design (FBDD) are powerful enablers of successful oncology drug design. Inhibitors that are substrate-competitive or bind at allosteric sites are only beginning to be studied, and represent a promising strategy in tackling kinase inhibitor selectivity. In contrast, for CNS targets, the ratio of FDA-approved inhibitors to activators is about 1:1 for GPCRs and ion channels, whereas for transporters only inhibitors have been FDA-approved. The design of allosteric modulators for these targets is on the rise, with an increasing number of these reaching clinical testing.6 Are these differences in the type and functionality of the drugs targeting oncology vs CNS disorders relevant to medicinal chemists, as often put forward? We suggest that is indeed the case, and they constitute a major opportunity to impact the field of neuroscience. Indeed, a recent PubMed search of papers published in The Journal of Medicinal Chemistry showed that the percentage of articles dealing with inhibition or antagonism has been consistently growing for decades, and approaching 70% and 50%, respectively (Figure 1). In contrast, reported research on agonists has been flat (or even decreasing) for decades, and work on allosteric modulators is clearly in its early days. Similar results were obtained with publications such as Bioorganic & Medicinal Chemistry Letters and European Journal of Medicinal Chemistry. Therefore, the design of allosteric drugs represents a major opportunity for medicinal chemists in neuroscience. This difference between drug discovery in oncology and neurotherapeutics is not only of theoretical interest. An examination of recent drug approvals and late clinical pipeline for CNS drugs shows that, while novel mechanisms of action are indeed pursued (e.g., BACE-1 inhibitors, LRKK2 inhibitors, PDE10 inhibitors), a substantial portion of the late stage CNS

Figure 1. Time course for papers published in The Journal of Medicinal Chemistry as a percentage of total papers since its inception in 1964 and ending in 2015 for the generic (*) terms shown in the legend. PubMed search conducted on June 2016. PAM = positive allosteric modulator; NAM = negative allosteric modulator.

drug pipeline is based on compounds known for years, if not decades (e.g., ketamine, amantadine, levodopa/carbidopa, apomorphine, allopregnanolone, cannabidiol, buprenorphine, tetrabenazine, dextromethorphan/quinidine, aripiprazole). Therefore, in the short term, new drug therapies for CNS indications will most likely come from improvements to known drugs made possible by a deeper understanding of how these drugs work. One might speculate on the reasons for these differences. Oncology drugs tend to be used for shorter periods of time than in CNS, where up to decades of treatment adherence is often required. The safety demands associated with such differences in length of administration are also relevant. While few diseases are fully monogenic, CNS diseases are thought to be highly multifactorial in their etiology, and finding singleorigin causative target-to-disease relationships is unlikely. On the contrary, in oncology, precision medicine is enabled by the identification of a small number of actionable molecular targets for which drugs have been developed. Thus, the strategies to accelerate the discovery of medicines in these two therapeutic areas need not be identical, and may follow different trajectories. What is the role of medicinal chemists to help define the better paths forward in neuroscience drug discovery?

■

THE ROLE OF MEDICINAL CHEMISTRYPERCEPTIONS AND REALITY Without a doubt, drug discovery is a formidably complex undertaking, as demonstrated by the broad and ever expanding range of scientific disciplines represented in project teams. This dynamic constantly redefines the role of medicinal chemists. Among some members of the life science community there is a process-like view of drug discovery, in which medicinal chemists’ main role is “making compounds”in other words, the chemical synthesis piece. These compounds are submitted for biological testing. The process goes on until a compound meets a previously defined target product profile (TPP) of appropriate activity, selectivity, and DMPK properties. The structured nature of this process may give the appearance of a mechanical exercise. On the contrary, the task of engineering in a molecule the TPP requires the testing of a number of creative design hypotheses laying at the interface between medicinal chemistry and other areas (e.g., DMPK, molecular pharmacology, enzymology, toxicology, etc.). In order to formulate such hypotheses, medicinal chemists educate themselves in new B

DOI: 10.1021/acschemneuro.6b00394 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Viewpoint

C

imaging agents

system chemistry smarter tool compounds

drugs as ensemble modulators

biased ligands

Table 1

area

BONUS: SOFT SKILLS OF MEDICINAL CHEMISTS Drug discovery organizations used to hire well-trained synthetic chemists. Why? Actually, their up-to-date organic synthesis knowledge was only one of the ways to impact drug research. There are a number of “soft” traits that medicinal chemists bring to their teams, and were learned during their academic training. The stepwise nature of organic synthesis and the logical thinking of mechanistic organic chemistry teach planning, multitasking, and prioritizing skills, incorporation of unexpected findings into nonlinear thinking, creativity to generate and test novel hypotheses, and the tenacity to deliver under high pressure in the face of adversity. Drug discovery is a risky undertaking, and medicinal chemists thrive in taking risks. They inject into their teams an indispensable dose of enthusiasm that the next hypothesis will work out and deliver the progress needed. Medicinal chemists are good at solving problems in team environmentsboth intra- and interdisciplinary. In fact, part of

from receptors to ensembles

■

class effect

impact

ALLOSTERISM AND DRUG ACTION Formally presented in the late 1800s by the late E. Fischer, the “lock & key” model is still extensively used by medicinal chemists. This model uses a simple “1 molecule + 1 receptor” relationship, where the drug acts at the same binding site as the target’s endogenous ligand and impacts the balance between two well-defined states, one “active” and one “inactive”. A different model emerged based on drugs modulating receptor function by acting at allosteric sites. This model uses a more complex “1 endogenous orthosteric ligand + 1 xenobiotic allosteric ligand + 1 receptor” relationship. In theory, this strategy could deliver drugs in areas where the orthosteric approach failed or could be improved. After a decade of allosteric drug design, a number of challenges have been encountered and overcome. In retrospect, many of these “bumps in the road” originated in extrapolating in a linear way to this new paradigm, the lessons learned designing drugs that are competitive inhibitors. The collaboration between medicinal chemists and molecular biologists has now shown some new attributes of membrane-bound targets, often at the core of various CNS diseases. These learnings (Table 1) create major opportunities for medicinal chemists in neuroscience, with the potential to increase our confidence in novel mechanisms, as well as re-examine old drugs.

compound selectivity

■

structure-pathways relationships

areasnot to the level of an expert, obviously, but to enable asking good questions. The knowledge developed through decades inhibiting biological targets (Figure 1) sustains the success of this paradigm for drugs that are functional inhibitors (e.g., enzyme inhibitors, ion channels, inhibitors, GPCR antagonists). However, CNS targets often require activation of a biological target, whose activity became compromised by neurodegenerative processes. It is not unreasonable to argue that new scientif ic knowledge must be developed to confidently design drugs that activate targets, and how to translate their effects to the clinical setting. This represent an example where CNS medicinal chemists can help better understand brain physiology and create new pharmacological ways to modulate targetsa necessary stepping stone on the way toward fulfilling the goals of the century of biology. Early work on allosteric modulators as potential CNS drugs suggests this area is fertile ground for such aims.

GPCRs can activate parallel signaling pathways to initiate distinct signal transduction cascades independent of G protein pathways. When a link between a pathway and a pathophysiology exists, biased ligands can activate select intracellular responses. The clinical validation of this concept may come from clinical candidates such as TRV130. Developing “structure−activity relationships” (SAR) between a chemical series and a biological target has expanded to include selective pathway effects. This additional level of detail may impact the type and magnitude of the efficacy seen in the clinic, and therefore may enhance the rate of clinical success. Historically established using binding assays on broad panels of key receptors, with well-known orthosteric and competitive radioligands. Recently, a functional version of this broad panel became available to qualify selectivity of allosteric ligands. These screens enable ruling out allosteric effects of a drug candidate with targets tested, but the potential for allosteric effects at untested biological targets still exists. Different compounds, whether having small or large structural differences, may bind a target with exquisite selectivity, yet direct different cellular pathways and phenotypes. This suggests the concept of “class effect” (two drugs with similar attributes will elicit similar effects) may be misleading and not reflect reality. Receptors are not merely the physical host where drugs elicit an effect, but a molecular hub where relevant ligands interact. Functional activity is the result of complex interactions among ensemble components. The concentrations of these ligands may reflect the pathophysiology, enabling a molecular target-to-disease link. Multiple conformational states may coexist for an ensemble, leading to a wide range of functional efficacy and signaling pathways. The ensemble determines the extent and the quality of the functional response. The drug acts by perturbing such balance. Metabolomics information may be used to study the impact of a certain catabolite on the drug functional effects. A way to shed light on complex biologies is using fit-for-purpose chemical tools to interrogate specific aspects of the drug−receptor interactions. For example, functional receptor activation can occur by enhancing either the affinity or the efficacy of an endogenous agonist, or by an allosteric agonist. A negative allosteric modulator may produce full inhibition, partial inhibition, and be biased toward different signal transduction pathways. These key differences are seen in mechanistic models reflecting the safety and efficacy of different compounds. Given the impossibility to access brain tissue from patients, complementary imaging technologies reporting binding and functional states of brain regions may be the best available option. Allosteric PET agents may help decipher the binding-function disconnect in some targets.

ACS Chemical Neuroscience

DOI: 10.1021/acschemneuro.6b00394 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Viewpoint

ACS Chemical Neuroscience

(6) Lindsley, C. W., Emmitte, K. A., Hopkins, C. R., Bridges, T. M., Gregory, K. J., Niswender, C. M., and Conn, P. J. (2016) Practical Strategies and Concepts in GPCR Allosteric Modulator Discovery: Recent Advances with Metabotropic Glutamate Receptors. Chem. Rev. 116 (11), 6707−41.

the credit for the improvement in success rates of early clinical trials is due to collaborations with colleagues in the DMPK and toxicology lines. Medicinal chemists learn to ask incisive, critical questions and then seek answers to them. Indeed, there is recognition of the ability of medicinal chemists to develop outside their comfort zone to generate a distinct set of questions to interrogate pharmacology, which is orthogonal to that derived from a biology perspective. This, in turn, leads to better definition of the target product profile sought in the drugs they are creating. All this, and the acquired ability to think in molecular bites, is what makes the contributions of medicinal chemists unique. One last skill medicinal chemists were forced to develop is adjusting to geopolitical and market changes that have characterized the pharmaceutical industry in the past decade. These resulted in the offshoring of many of the tasks previously conducted my medicinal chemists, and sharpened their entrepreneurial skills and instincts to keep themselves relevant to the drug discovery enterprise. In summary, historically medicinal chemists have contributed to drug discovery by a combination of “hard” and “soft” skills, many of which are learned during academic training and provide a different perspective, key to neuroscience research. The close collaboration between medicinal chemists and molecular pharmacologists engaged in the discovery of allosteric drugs has resulted in a number of new concepts to enhance the qualification and differentiation of compounds (including marketed drugs) previously seen as equivalent. This knowledge enables the mechanistic interrogation of CNS biology with an unprecedented level of detail at the molecular level. Clinical testing of so-characterized compounds will likely provide novel insight of translational value, including clues on molecular aspects of disease, otherwise unattainable. When this new molecular level knowledge is integrated with ongoing excellent biology-centered efforts to decipher human biology, the goals set forth for the “Century of Biology” will become actionable and reinvigorate industry investments in small molecule neuroscience research. These chemistry-driven concepts should impact team membership and be a key part of the strategy pursued to deliver urgently needed drugs therapies for brain disorders.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Dario Doller: 0000-0002-5944-383X Notes

The author declares no competing financial interest.

■

REFERENCES

(1) Venter, C., and Cohen, D. (2004) The Century of Biology. New Perspectives Quarterly 21, 73−77. (2) Hyman, S. E. (2016) Back to basics: luring industry back into neuroscience. Nat. Neurosci. 19, 1383−4. (3) Kinch, M. S. (2015) An analysis of FDA-approved drugs for neurological disorders. Drug Discovery Today 20 (9), 1040−3. (4) Kinch, M. S., and Patridge, E. (2015) An analysis of FDAapproved drugs for psychiatric disorders. Drug Discovery Today 20 (9), 292−5. (5) Kinch, M. S. (2014) An analysis of FDA-approved drugs for oncology. Drug Discovery Today 19 (12), 1831−5. D

DOI: 10.1021/acschemneuro.6b00394 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX