A class of selective antibacterials derived froma protein kinase inhibitor pharmacophoreJ. Richard Millera,1,2,3, Steve Dunhama,1,4, Igor Mochalkina,1,2, Craig Banotaia, Matthew Bowmana, Susan Buista,Bill Dunklea,4, Debra Hannaa,2, H. James Harwoodb, Michael D. Hubanda,2, Alla Karnovskya, Michael Kuhna,2,Chris Limberakisa,2, Jia Y. Liua, Shawn Mehrensa,2, W. Thomas Muellera,5, Lakshmi Narasimhana,6, Adam Ogdena,2,Jeff Ohrena,b, J. V. N. Vara Prasada, John A. Shellya,4, Laura Skerlosa, Mark Sulavika, V. Hayden Thomasa,2,Steve VanderRoesta, LiAnn Wangc, Zhigang Wanga,7, Amy Whittona, Tong Zhua,2, and C. Kendall Stovera,4aPfizer, Inc., Ann Arbor, MI 48105;bPfizer, Inc., Groton, CT 06340; andcPfizer, Inc., Cambridge, MA 02139Edited by Michael A. Marletta, University of California, Berkeley, CA, and approved December 5, 2008 (received for review November 10, 2008)As the need for novel antibiotic classes to combat bacterial drugresistance increases, the paucity of leads resulting from target-based antibacterial screening of pharmaceutical compound librar-ies is of major concern. One explanation for this lack of success isthat antibacterial screening efforts have not leveraged the eukary-otic bias resulting from more extensive chemistry efforts targetingeukaryotic gene families such as G protein-coupled receptors andprotein kinases. Consistent with a focus on antibacterial targetspace resembling these eukaryotic targets, we used whole-cellscreening to identify a series of antibacterial pyridopyrimidinesderived from a protein kinase inhibitor pharmacophore. In bacte-ria, the pyridopyrimidines target the ATP-binding site of biotincarboxylase (BC), which catalyzes the first enzymatic step of fattyacid biosynthesis. These inhibitors are effective in vitro and in vivoagainst fastidious Gram-negative pathogens including Haemophi-lus influenzae. Although the BC active site has architectural simi-larity to those of eukaryotic protein kinases, inhibitor binding tothe BC ATP-binding site is distinct from the protein kinase-bindingmode, such that the inhibitors are selective for bacterial BC. Insummary, we have discovered a promising class of potent anti-bacterials with a previously undescribed mechanism of action. Inconsideration of the eukaryotic bias of pharmaceutical libraries,our findings also suggest that pursuit of a novel inhibitor leads forantibacterial targets with active-site structural similarity to knownhuman targets will likely be more fruitful than the traditional focuson unique bacterial target space, particularly when structure-basedand computational methodologies are applied to ensure bacterialselectivity.acetylcoenzyme A carboxylase 兩 biotin carboxylase 兩 crystal structure 兩high-throughput screening 兩 fatty acid biosynthesisThe well-documented increase in antibacterial re sistance over thepast few decades has led to intensive efforts to discover novelantibacterial agents. Literally thousands of new essential bacterialtargets from human pathogens were identified as a result of thegenomics revolution (1–3). This necessitated a means to ‘‘triage’’these targets to identify the most promising ones to pursue by highthroughput screening of pharmaceutical compound files. In thesetarget analyses, one desirable attribute was little or no sequenceand/or structural homology to human gene products to improve thechances of finding bacterial-selective hits. However, this selectioncriterion could also have a deleterious effect on the number ofpotent hits identified because the medicinal chemistry efforts thatbuilt these compound files were largely focused on human thera-peutic targets such as kinases, G protein-coupled receptors, pro-teases, etc. (1–4). Therefore, pursuit of bacterial targets with thegreatest sequence and/or structural relatedne ss to proven eukary-otic drug targets could be more fruitful. Consistent with thishypothesis, we used unbiased whole-bacterial cell screening of thePfizer compound library to discover a series of antibacterial pyri-dopyrimidines (1, 2, and 3;Fig.1A) that emerged from a structure-based drug design program targeting eukaryotic tyrosine proteinkinase s (5). By using genetic and biochemical tools, the bacterialtarget of these compounds was identified as biotin carboxylase(BC); one portion of the acetyl-CoA carboxylase (ACCase) mul-tienzyme complex responsible for the first step of fatty acidbiosynthesis (6). BC has an active site with considerable structuralsimilarity to eukaryotic protein kinases, yet it is sufficiently differentto allow for selectivity against both human kinases and the eukary-otic ACCase.The identification of selective antibacterials containing akinase inhibitor pharmacophore has intriguing implications forantibacterial drug discovery, particularly given that the targets,biotin carboxylase and eukaryotic protein kinases, have structurallyrelated ATP-binding sites. It remains to be seen if the huge arrayof eukaryotic inhibitors present in pharmaceutical libraries can bemined for their activity against structurally related bacterial targetssuch as the bacterial histidine kinases involved in cell–cell signaling,lipopolysac charide sugar kinases involved in Gram-negative cellwall formation, antibiotic kinases that deactivate specific anti-bacterial agents, or less obvious targets, such as biotin carboxy-lase. Our results argue that there may be value in reassessingantibacterial t arget space for prev iously unexplored (or under-ex plored) targets amenable to an approach based on repurpos-ing eukaryotic pharmacophores.ResultsIdentification of Antibacterial Pyridopyrimidines. As part of ourantibacterial drug discovery effort, a library of ⬇1.6 millionAuthor contributions: J.R.M., S.D., I.M., S.B., D.H., L.N., A.O., J.V.N.V.P., M.S., V.H.T., Z.W.,T.Z., and C.K.S. designed research; J.R.M., S.D., I.M., C.B., M.B., S.B., B.D., M.D.H., M.K., C.L.,J.Y.L., S.M., W.T.M., L.N., A.O., J.O., J.V.N.V.P., J.A.S., L.S., S.V., L.W., Z.W., and A.W.performed research; J.R.M., S.D., I.M., C.B., M.B., B.D., H.J.H., A.K., M.K., C.L., J.Y.L., S.M.,W.T.M., L.N., J.V.N.V.P., J.A.S., V.H.T., S.V., and Z.W. contributed new reagents/analytictools; J.R.M., S.D., I.M., M.B., S.B., D.H., H.J.H., M.D.H., A.K., C.L., J.Y.L., L.N., A.O., J.O.,J.V.N.V.P., L.S., M.S., L.W., Z.W., A.W., T.Z., and C.K.S. analyzed data; and J.R.M., S.D., I.M.,J.O., and C.K.S. wrote the paper.The authors declare no conflict of interest.This article is a PNAS Direct