Yin and Yang in medicinal chemistry: what does drug-likeness mean?

Future Medicinal ChemistryVol. 3, No. 5 EditorialFree AccessYin and Yang in medicinal chemistry: what does drug-likeness mean?Ricardo Macarrón & Juan I LuengoRicardo Macarrón† Author for correspondencePlatform Technology and Science, GlaxoSmithKline R&D, 1250 S Collegeville Rd UP 12-200, P.O. Box 5089 Collegeville, PA 19426-0989, USA. Search for more papers by this authorEmail the corresponding author at ricardo.macarron@gsk.com & Juan I LuengoOncology R&D, GlaxoSmithKline R&D, Collegeville, PA, USASearch for more papers by this authorPublished Online:28 Apr 2011https://doi.org/10.4155/fmc.11.19AboutSectionsPDF/EPUB ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinkedInRedditEmail Keywords: drug-likelead-likenatural productsoral drugsrule of fiveFigure 1. Examples of drugs with natural origin that do not comply with Lipinski’s rule of five.“All things are poison and nothing is without poison, only the dose permits something not to be poisonous.” (Paracelsus, 1493–1541).So what distinguishes a drug from a poison? It is the dose, stupid! Medicinal chemistry is the art of discovering molecules that display a therapeutic window, that is, a separation between the efficacious and the toxic doses, which allows a safe administration to cure a disease. In this quest, medicinal chemists need to combine the knowledge learnt from biological and pharmacological models with those from their own experiences and intuition to make decisions on how to convert hits to leads and leads to drugs. Given the complexity, costs and length of the drug-discovery process, it is crucial to recognize what a drug looks like and, as important, to know what can make a candidate molecule fail in the process. This editorial focuses on small molecules (chemicals of molecular weight typically lower than 800 Da) intended to be administered orally. Notice that drugs can be also administered by non-oral routes, and that biological drugs, mostly proteins, are an important component of today’s pharmacopeia.In the last 70 years, approximately 1400 small-molecule oral drugs have been approved by the US FDA for human use. In 1997 a seminal paper by Lipinski and colleagues at Pfizer evaluated the chemical properties of oral drugs previously approved, or at the late stages of clinical trials and enunciated the now famous ‘rule of five’ (Ro5) [1]. This was the first modern definition of drug-like features and had a profound impact on the field, especially at a time in which combinatorial chemistry and high-throughput screening were being recognized as the new drug-discovery paradigms.Further examination of approved drugs and failed clinical candidates has re-enforced the general principles of the Ro5; although there is increased evidence to suggest that the main physicochemical property that leads to clinical failure is high lipophilicity and that size alone is not a discriminating factor between drugs and failed candidates [2].Despite the wide acceptance of the Ro5 throughout the pharmaceutical industry, it is becoming clear that the ‘rules’ should be used just as general guidelines during the lead-optimization process. Indeed, there are numerous examples of successful orally active drugs whose properties fall outside the ‘Ro5’, and these exceptions to the rules provide rich veins for exploitation. This is especially true in the area of natural product therapeutics.Natural products and semi-synthetic compounds remain a great source of drugs, with approximately 30% of new chemical entities in the last 30 years being derived from them [3]. Even though drugs from natural origin do not always adhere to the Ro5, they must reside in a drug-like chemical space. Lipinski and colleagues recognized this fact in their original publication, capturing these natural product outliers under a fifth rule: compound classes that are substrates for biological transporters are exception to the Ro5 [1]. However, the reason for a number of the more complex natural products circumventing the Ro5 is probably a result of their particular molecular architecture and arrangement of functional groups, rather than their facilitated transport into the cell [4]. This is very likely the case with orally active natural products, such as cyclosporine A, rapamycin, ivermectin, vinorelbine (Figure 1), paclitaxel, FK506, erythromycin or rifampicin, which are known to penetrate cells via passive diffusion despite their high molecular weight (>750 Da). A common feature of most of these compounds is a dense substitution and unsaturated functionality packed within a large macrocyclic ring. A macrocyclic scaffold is a clever solution to the problem of bioavalability, since it serves not only to limit the molecular flexibility (i.e., increases conformational restriction due to fewer rotable bonds), but also to override the Ro5’s third and fourth rules, by canceling out hydrogen bonds intramolecularly. With cyclosporine A, nature’s solution turns out to be even more ingenious. Depending on the solvent, the molecule can adopt two distinct conformations: either with the NH’s intramolecularly folded in (in hydrophobic solvents) or with the NH’s all exposed to the solvent (in water). This is presumably the way cyclosporine A can partition in both lipid and aqueous environments. Solubility in both lipids and water is the key to cell permeability, and it would be interesting to find out if the other macrocyclic natural products also use similar ‘tricks’ to that of cyclosporine A. A more complete understanding of how nature manages to bypass the restraints imposed by the Ro5 will be of great value to drug-discovery researchers. Indeed, several groups are exploring this area, applying the principles of diversity-oriented synthesis to scaffolds that are inspired by natural products [5].There is another caveat of the current definitions of drug-likeness (Ro5 and variants thereof). They are focused on the output from historical drug discovery and leave no room to the exploration of unchartered areas of chemical space. This is especially important in the new era of drug discovery, in which novel agents are strongly needed to address a number of unmet medical needs. Can truly novel therapeutics be found by relying exclusively on the properties of old drugs? If Christopher Columbus had based his quest for a new (better) maritime route to reach India solely on existing maps, his famous trip in 1492 would have never happened. Owing to economical constraints, there is little room for wild rides in drug discovery. Nevertheless, exploration of new paradigms might deserve a bit more attention from the biomedical research community, which after years of applying drug-like concepts is still suffering very high attrition in the clinic (∼95%), with a significant proportion of molecule-related failure (mostly toxicity).Lead optimization tends to augment the Ro5 parameters, therefore, lead-likeness has been defined within more stringent boundaries [6]. In reflecting on the high attrition rate in the last decade, it is imperative to question certain dogmas. One such dogma has been that drugs should exhibit nanomolar affinity against the biological target of therapeutic interest to enlarge its therapeutic window. In pursuit of potency, medicinal chemists have been victims of a lipophilic trap: hydrophobic interactions lead to increased potency on the target of interest but also augment the chances of off-target activities and other undesirable properties [7]. Working in this dangerous zone seems to be a recipe for failure. Specificity driven by precise polar interactions has been largely an unexplored avenue in oral drug discovery. Recent advances in carbohydrate chemistry [8] and the few examples of sugar-related drugs [9] give hope in this area.As the number of molecular targets keeps increasing every year, it will be important to consider multiple approaches to explore the novel chemical space revealed by these new findings. This seems to be the case for some of the more recent targets, in which approaches as varied as computer-assisted drug discovery, fragment-based or diversity-oriented synthesis are being pursued in parallel by many groups. A more conservative approach should be taken in areas where the existing evidence is very strong (e.g., avoidance of insoluble and extremely lipophilic compounds), but a grain of adventure can be a key ingredient where the current mantra is not fully supported by facts (e.g., exploration of bigger molecules that mimic successful exceptions to the Ro5).One last important aspect to consider is the effect that the Ro5 (or any other rule) might have on the creative thinking and innovation of the scientists working in the area, especially in those new to the field. Changes in the drug-discovery process happen at an ever-accelerating pace. New developments on drug formulation and delivery, along with a better understanding of safety and metabolism could move the goalposts of ‘druggability’ parameters and, in time, make some of the current rules obsolete.In closing, there is no absolute definition of drug-likeness and a judicious balance of knowledge and exploration is required to successfully discover new medicines.DisclaimerThis work is the opinion of the authors and does not represent the views of GlaxoSmithKline.Financial & competing interests disclosureThe authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.No writing assistance was utilized in the production of this manuscript.Bibliography1 Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Deliver. Rev.23(1–3),3–25 (1997).Crossref, CAS, Google Scholar2 Leeson PD, Empfield JR. Reducing the risk of drug attrition associated with physicochemical properties. Annu. Rep. Med. Chem.45(C),393–407 (2010).CAS, Google Scholar3 Newman DJ, Cragg GM. Natural products as sources of new drugs over the last 25 years. J. Nat. Prod.70(3),461–477 (2007).Crossref, Medline, CAS, Google Scholar4 Sugano K, Kansy M, Artursson P et al. Coexistence of passive and carrier-mediated processes in drug transport. Nat. Rev. Drug Discov.9(8),597–614 (2010).Crossref, Medline, CAS, Google Scholar5 Cordier C, Morton D, Murrison S, Nelson A, O’Leary-Steele C. Natural products as an inspiration in the diversity-oriented synthesis of bioactive compound libraries. Nat. Prod. Rep.25,719–737 (2008).Crossref, Medline, CAS, Google Scholar6 Rishton GM. Molecular diversity in the context of leadlikeness: compound properties that enable effective biochemical screening. Curr. Opin. Chem. Biol.12(3),340–351 (2008).Crossref, Medline, CAS, Google Scholar7 Hann MM. Molecular obesity, potency and other addictions in drug discovery. Med. Chem. Commun. DOI: 10.1039/C1MD00017A (2011) (Epub ahead of print).Google Scholar8 Kiessling LL, Splain RA. Chemical approaches to glycobiology. Ann. Rev. Biochem.79,619–653 (2010).Crossref, Medline, CAS, Google Scholar9 Ernst B, Magnani JL. From carbohydrate leads to glycomimetic drugs. Nat. Rev. Drug Discov.8(8),661–677 (2009).Crossref, Medline, CAS, Google ScholarFiguresReferencesRelatedDetailsCited ByPharmaceutical profilingExploring new targets and chemical space with affinity selection-mass spectrometry21 October 2020 | Nature Reviews Chemistry, Vol. 5, No. 1ADMET-score – a comprehensive scoring function for evaluation of chemical drug-likeness1 January 2019 | MedChemComm, Vol. 10, No. 1Selective Inhibition of the Mitochondrial Permeability Transition Pore Protects against Neurodegeneration in Experimental Multiple SclerosisJournal of Biological Chemistry, Vol. 291, No. 9Analysis of Physicochemical Properties for Drugs of Natural Origin3 June 2015 | Journal of Natural Products, Vol. 78, No. 6The application of in silico drug-likeness predictions in pharmaceutical researchAdvanced Drug Delivery Reviews, Vol. 86How drug-like are ‘ugly’ drugs: do drug-likeness metrics predict ADME behaviour in humans?Drug Discovery Today, Vol. 19, No. 4Guiding principles for natural product drug discoveryDavid Camp, Rohan A Davis, Elizabeth A Evans-Illidge & Ronald J Quinn18 June 2012 | Future Medicinal Chemistry, Vol. 4, No. 9Deployment of in silico and in vitro safety assays in early-stage drug discoveryYvonne Will & Thomas Schroeter16 July 2012 | Future Medicinal Chemistry, Vol. 4, No. 10Established and Emerging Trends in Computational Drug Discovery in the Structural Genomics EraChemistry & Biology, Vol. 19, No. 1 Vol. 3, No. 5 Follow us on social media for the latest updates Metrics History Published online 28 April 2011 Published in print April 2011 Information© Future Science LtdKeywordsdrug-likelead-likenatural productsoral drugsrule of fiveDisclaimerThis work is the opinion of the authors and does not represent the views of GlaxoSmithKline.Financial & competing interests disclosureThe authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.No writing assistance was utilized in the production of this manuscript.PDF download