Discovery of Novel Antibiotics Via Streptomyces Sporulation by Scott

Home , Bacitracin, Cephamycin, Gepotidacin, List of antibiotics, Mervyn Bibb, Streptomyces venezuelae

Discovery of novel antibiotics via Streptomyces sporulation

Scott McAuley

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Graduate Department of Biochemistry University of Toronto

Discovery of novel antibiotics via Streptomyces sporulation

Scott McAuley Doctor of Philosophy Graduate Department of Biochemistry University of Toronto 2019

Streptomyces are filamentous bacteria known for producing a wide range of bioactive molecules. In addition to being nature’s chemists they have a unique multicellular lifecycle that ends with the sporulation of aerial hyphae. I used this unique Streptomyces development cycle, specifically sporulation, as a screening platform to discover novel biologically active molecules and characterized their mechanisms of action. Min-1 inhibits the growth of a range of Gram-positive bacteria, is active against the cell envelope, and induces a short cell phenotype in B. subtilis. Another molecule, EN-7, inhibits bacterial gyrase and is active against extensively resistant Staphylococcus aureus in addition to other Gram-positive pathogens. In addition to investigating these specific molecules, I developed a high-throughput screen to identify molecules that disrupt the Gram-positive bacterial membrane and found that Streptomyces venezuelae sporulation is highly sensitive to multiple forms of DNA damage.

ii Contents

1 Introduction 1 1.1 The challenges of antibiotic discovery and development ...... 1 1.2 Identifying novel antimicrobials and their mechanism of action ...... 3 1.2.1 Identifying bioactive molecules through growth inhibition . . . . . 4 1.2.2 Identifying mechanism of action through reporter screens and assays 7 1.2.3 Target-based screening ...... 13 1.2.4 Novel screening approaches ...... 14 1.2.5 Conclusions ...... 16 1.3 Small molecule impact on bacterial morphology ...... 17 1.3.1 Activators of the SOS response disrupt the cell cycle ...... 18 1.3.2 Cell envelope determines cell shape ...... 21 1.3.3 Direct disruption of FtsZ disrupts the cell cycle ...... 23 1.3.4 Nutrient availability impacts both growth and the cell cycle . . . 25 1.3.5 Cytological proﬁling identiﬁes antibiotic mechanisms of action . . 27 1.3.6 Conclusion ...... 27 1.4 Thesis objectives and outline ...... 28

2 Membrane activity proﬁling of small molecule B. subtilis growth inhibitors utilizing novel duel-dye ﬂuorescence assay 29 2.1 Abstract ...... 29 2.2 Introduction ...... 30 2.3 Results and Discussion ...... 32 2.3.1 High-throughput assay for determining impact on membrane potential and permeability ...... 32 2.3.2 Screen for biologically active small molecules against B. subtilis . 34 2.3.3 Membrane disruption by biologically active small molecules . . . . 36 2.4 Conclusion ...... 40

iii 3 A chemical inhibitor of cell growth reduces cell size in Bacillus subtilis 46 3.1 Abstract ...... 46 3.2 Introduction ...... 47 3.3 Results and Discussion ...... 48 3.3.1 Min-1 inhibits Streptomyces sporulation at sub-inhibitory concentrations ...... 48 3.3.2 Min-1 inhibits growth of Gram-positive bacteria and reduces cell length of B. subtilis ...... 49 3.3.3 Structural analogs of Min-1 have altered eﬀects on growth rate and cell length ...... 51 3.3.4 Min-1 disrupts coordination between growth and FtsZ ring assembly 55 3.3.5 Min-1 targets the cell envelope ...... 56 3.3.6 Min-1 is a novel inhibitor of bacterial growth ...... 60

4 Discovery of a novel DNA gyrase-targeting antibiotic through the chemical perturbation of Streptomyces venezuelae sporulation 62 4.1 Abstract ...... 62 4.2 Introduction ...... 63 4.3 Results ...... 64 4.3.1 The Streptomyces venezuelae sporulation program is sensitive to DNA damage ...... 64 4.3.2 Novel small molecule sporulation inhibitors ...... 66 4.3.3 EN-7 targets DNA gyrase ...... 68 4.4 Discussion ...... 75

5 Concluding Remarks 79 5.1 Thesis Summary ...... 79 5.2 Future Directions ...... 80 5.2.1 Investigating mechanism of Streptomyces sporulation inhibition via DNA damage ...... 80 5.2.2 Additional approaches for elucidating Min-1’s target and mechanism of action ...... 80 5.2.3 Identify mechanism of gyrase inhibition for EN-7 ...... 81 5.2.4 Creating high-thoughput method of screening Streptomyces sporulation inhibition ...... 81 5.2.5 Additional small molecule screens against bacterial development . 81 5.3 Conclusion ...... 82

iv 6 Materials and Methods 83 6.1 General Experimental Procedures ...... 83 6.1.1 Strains and plasmids ...... 83 6.2 Growth inhibition assays ...... 83 6.2.1 Broth dilution assay ...... 83 6.2.2 Streptomyces sporulation inhibition ...... 83 6.2.3 Superﬁcial S. aureus skin infection model ...... 85 6.3 PAINs assays ...... 85 6.3.1 Duel-dye membrane disruption screen ...... 85 6.3.2 Dynamic Light Scattering ...... 85 6.4 Miscroscopy Methods ...... 86 6.4.1 Scanning electron miscroscopy ...... 86 6.4.2 B. subtilis cell length measurement and FtsZ immunoﬂuorescence labeling ...... 86 6.5 Reporter assays ...... 87 6.6 Isolation and characterization of resistant mutants ...... 87 6.6.1 Resistant mutant generation ...... 87 6.6.2 Sequencing resistant mutants ...... 87 6.6.3 S. aureus allelic exchange ...... 88 6.7 Gyrase inhibition ...... 88 6.7.1 E. coli gyrase supercoiling ...... 88 6.7.2 S. aureus gyrase supercoiling ...... 89 6.7.3 E. coli topoisomerase IV decatenation ...... 89 6.7.4 S.aureus topoisomerase IV decatenation ...... 89 6.7.5 S.aureus gyrase cleavage ...... 89

7 Appendix 1: A chemical inhibitor of cell growth reduces cell size in Bacillus subtilis 90 7.1 Results and Discussion ...... 90 7.1.1 Min-1 does not inhibit B. subtilis sporulation ...... 90

7.1.2 Min-1 activates B. subtilis PywaC-lux reporter ...... 91 7.1.3 Unable to generate Min-1 resistant mutants ...... 91 7.1.4 Min-1 potentiates some known antibiotics ...... 92 7.1.5 Identifying sensitive strains using a B. subtilis knockdown library 93 7.1.6 Impact of Min-1 on HEK293 viability ...... 96

v 8 Appendix 2: Discovery of a novel DNA gyrase-targeting antibiotic through the chemical perturbation of Streptomyces venezuelae sporulation 98 8.1 Results and Discussion ...... 98 8.1.1 Growth inhibition and reporter activation of S. venezuelae sporulation screen hits ...... 98 8.1.2 Aggregation activity of S. venezuelae sporulation screen hits . . . 101 8.1.3 Gyrase S. aureus antisense strains show resistance to EN-7 . . . . 101 8.1.4 EN-7 does not reduce S. aureus bacterial load in a skin mouse infection model ...... 104

Bibliography 106

vi List of Tables

1.1 Reporter strains used to identify antibiotic MOA ...... 10 1.2 Cell morphologies induced by cell wall inhibiting antibiotics ...... 24

3.1 Minimum inhibitory concentration of Min-1 ...... 50

4.1 Table of gyrase mutations found in EN-7 resistant mutants ...... 68

6.1 Table of strains used in studies ...... 84

7.1 Impact of Min-1 on activity of known antibiotics ...... 94 7.2 B. subtilis CRISPRi hits on xylose induction with Min-1 treatment . . . 96

8.1 Growth inhibition of S. venezuelae sporulation screen hits ...... 100 8.2 Growth inhibition of S. venezuelae sporulation screen hits ...... 102

vii List of Figures

1.1 Simpliﬁed screening processes for discovering new antibiotics and uncovering their mechanism of action ...... 4 1.2 Examples of bioactive molecules containing PAIN substructures . . . . . 5 1.3 Whole-cell reporter assays for determining MOA ...... 9 1.4 Chemical-genetic assays for determining MOA ...... 12 1.5 Streptomyces lifecycle ...... 16 1.6 Small molecules activate the SOS response and disrupts the cell cycle . . 19

2.1 Chemical and optical properties of TO-PRO-3 iodide and DiOC2(3) . . . 33 2.2 Eﬀect of known nisin, CCCP, and vancomycin on TO-PRO-3 iodide and

DiOC2(3) ﬂuorescence ...... 35 2.3 Screen of bioactive synthetic molecules for membrane activity ...... 37 2.4 Membrane activity screen summary ...... 39 2.5 Concentration dependent impact of bioactive molecules on TO-PRO-3 io-

dide and DiOC2(3) ﬂuorescence ...... 40 2.6 Membrane permeability hits ...... 42 2.7 Membrane potential hits ...... 43 2.8 Membrane potential hits, cont...... 44 2.9 Calculated logP values for bioactive compound library ...... 45

3.1 Effect of Min-1 on S. venezuelae development ...... 49 3.2 Cell length and growth rate effects of Min-1 on B. subtilis ...... 52 3.3 Structure and activity of Min-1 analogs against B. subtilis and S. aureus 53 3.4 Effect of Min-1 analogs on B. subtilis growth and cell length ...... 54 3.5 Min-1 disrupts coordination between growth and FtsZ ring assembly independent of known cell size mechanisms ...... 57 3.6 Min-1 targets the cell envelope ...... 59 3.7 Impact of supplemental magnesium on Min-1 activity ...... 60

viii 4.1 Streptomyces venezuelae sporulation is sensitive to DNA damage . . . . . 65 4.2 Chemical structures and induced S. venezuelae phenotypes ...... 66 4.3 Identiﬁcation of EN-7 as an inhibitor of S. venezuale and Gram-positive pathogens ...... 67 4.4 EN-7 inhibits growth of extensively resistant S. aureus ...... 69 4.5 Allelic Exchange of EN-7 resistance mutations into S. aureus ...... 71 4.6 in-vitro inhibition of S. aureus gyrase by EN-7 ...... 72 4.7 Cross-resistance of EN-7 resistant strains with other gyrase inhibitors . . 73 4.8 Chemical structures of EN-7 and other investigational gyrase inhibitors . 74 4.9 Analysis of EN-7 resistance gyrase mutations ...... 76

7.1 Min-1 does not disrupt B. subtilis sporulation ...... 91

7.2 Min-1 activates B. subtilis PywaC-lux reporter strain ...... 92 7.3 Min-1 potentiates known antibiotics ...... 93 7.4 Impact of Min-1 on growth of a B. subtilis CRISPRi essential knock-down library ...... 95 7.5 Impact of Min-1 on HEK293 viability ...... 97

8.1 Aggregation activity of S. venezuelae sporulation screen hits ...... 101 8.2 Gyrase S. aureus antisense strains show resistance to EN-7 ...... 103 8.3 EN-7 is not able to reduce the bacterial load in a S. aureus superﬁcial skin mouse infection model ...... 105

ix Chapter 1

Introduction

1.1 The challenges of antibiotic discovery and development

Antibiotics are the cornerstone of modern medicine. They not only offer a cure for life- threatening infections but enable other vital therapies and procedures, from chemotherapy to complex organ transplants. With increasing rates of antibiotic resistance, we not only face the possibility of increasing mortality and morbidity from previously treatable infections, but that the numerous treatments that are enabled by antibiotics will not longer be safe or effective. Patients facing cancer, organ failure, childbirth complications, weakened immune systems, and more will face significantly increased risk. However, the challenges of developing new treatments for bacterial infectious diseases span science, society, regulations, and business. A 2016 report funded by the United Kingdom estimates that by 2050, 10 million lives a year and a cumulative $100 trillion USD of economic activity are at risk due to the rise of drug-resistant infections [198]. Even today, more than 700,000 people around the world die from drug resistant infections every year. Antibiotic resistance itself is natural, inevitable, and ancient [66], and has been acknowledged as a challenge for as long as we have been using antibiotics. Alexander Fleming acknowledged the potential for antibiotic resistance in his Nobel prize speech, “The time may come when penicillin can be bought by anyone in the shops. Then there is the danger that the ignorant man may easily underdose himself and by exposing his microbes to non-lethal quantities of the drug make them resistant” [97]. The success of these inexpensive, effective, drugs has resulted in ongoing challenges of overuse and new product development. One of the challenges leading to increased resistance is the overuse of antibiotics in

1 Chapter 1. Introduction 2 human health. The European Centre for Disease Prevention and Control (ECDC) has estimated that 30–50% of all antimicrobials prescribed to human patients are unneces- sary [165]. Many countries around the world do not require a prescription to purchase antibiotics and even in countries where a doctors visit is required, many patients will de- mand antibiotics from their family doctor even for what may be a viral infection. While new and rapid antimicrobial susceptibility testing technologies are being developed, there is no single major breakthrough technology that is leading the way to clearly identify the causative agent of an infection and profile its antimicrobial resistance profile [272]. Challenges for such systems include the complex process of validating new technology against reference methods beyond the proof-of-concept phase, the legal and regulatory landscapes, costs, the uptake of new tools, optimization of target product profiles, and difficulties conducting clinical trials. However, such innovation will be key to ensuring proper prescription of antibiotics and slowing the development of resistance. Unlike any other class of drugs, the inevitability of resistance requires constant development of new therapies as older drugs lose their activity. Despite the challenging and expensive process of developing these curative drugs, there is an expectation in the medical community that antibiotics are inexpensive. However, beyond antibiotics, medicine does not use the word ‘cure’ lightly. Patients facing cancer can go into remission, diabetes can be managed, and HIV viral loads can be reduced to undetectable levels. Recently, modern genetic engineering has the potential to cure otherwise untreatable genetic dis- orders, and patients suffering from hepatitis C can now expect to be cured of the virus. These new treatments are incredibly expensive, with Gilead’s Sovaldi hepatitis C treatment costing $84,000 when it was launched in 2013 or Spark Therapeutics’ Luxturna for genetic congenital blindness costing $800,000. By comparison, antibiotics offer a cure for life-threatening diseases over a short treatment course for incredibly low prices. From the perspective of a company developing new drugs, an inexpensive and short-course product is not one that has much potential to provide much of a return on investment. If you are able to develop and launch a new antibiotic, you then would like patients to use it. Antibiotic stewardship programs control the use of newer or stronger antibiotics in the attempt to reduce the spread of resistance [248]. As a developer of a new antibiotic, this means that use of your new product will be curtailed until absolutely necessary. Not a good position if you are a company looking to provide a return on your discovery and development investment. To overcome this economic challenge, new reimbursement models and incentive programs are being proposed to uncouple reimbursement from antibiotic sales [221]. Specifically, there are proposals for an insurance model where institutions pay a flat fee in addition to a per-unit fee for the antibiotics used [263] Chapter 1. Introduction 3 as well as transferable vouchers that would allow companies to extend the exclusivity lifetime and revenue from other, more profitable, products. In order to overcome challenges in raising early-stage funding for antibiotic development programs, a public-private partnership called CARB-X was created in 2016 to invest $350 million over 5 years to accelerate progression of innovative antimicrobial products into the clinic [203]. As of November 2018, the organization has invested in 27 pre-clinical and Phase I programs as well as 5 antibacterial devices and diagnostic products that range from new chemical classes to novel targets. Looking at later stage programs, as of July 2018 there are 30 new chemical entity (NCE) antibacterial drugs, ten biologics, ten NCEs against Mycobacterium tuberculosis, and four NCEs against C. difficile in development [261][91]. While many of these later-stage programs have activity against priority pathogens, they are dominated by derivatives of established classes and activity against extensively resistant Gram-negative pathogens remains a challenge. An ideal antibiotic needs to be both selective and safe, for while it may be relatively simple to kill bacteria with numerous toxic chemicals, it is challenging to do so without harming the patient. Other requirements include a useful spectrum of activity against clinically relevant organisms, a lack of cross-resistance with existing therapeutics, low propensity for rapid resistance selection, and pharmacological properties that allow effective systemic dosing. In this next section I will review some of the methods by which we identify new antimicrobial leads and uncover their mechanisms of action.

1.2 Identifying novel antimicrobials and their mechanism of action

The journey of identifying new antibiotics can take researchers down a number of different paths (Figure 1.1). Initially, researchers must balance the goals of target identification with ensuring suitable chemical properties, such as the ability to reach the intended target in a whole-cell environment and reducing toxic and off-target effects. In light of these goals, early-stage screening programs can take a variety of approaches. The simpliest method involves screening a collection of molecules for growth inhibition. While this initial screening can be done quickly, researchers must then use a variety of tools, generally with lower throughput, to narrow in on particular mechanisms of action and molecular targets for the initial hits. On the other hand, target-based approaches use purified proteins in a variety of in vitro assays to determine the binding and inhibition of the target. While this allows for Chapter 1. Introduction 4 precise target identification, the hit molecules may not have whole-cell activity due to an inability to bypass the cell membrane, be easily removed from the cell by efflux, or be inactivated by other resistance mechanisms. It has been suggested that these challenges are the cause of low success rate from target-based programs for new antimicrobials [242]. Increasingly, whole-cell reporters, chemical-genetic methods, or induction of particular phenotypes are being used. The increased availability of high-throughput screening tools as well as the advantage of selecting for molecules with favorable chemical and physical properties for whole-cell activity combined with some indication of the molecular target has led to increased interest in this type of screening in academic and industrial environments [293]. In the following sections I will review some examples of traditional mechanisms of lead identification, from growth inhibition, target-based screening, and reporter assays, as well as some more novel approaches that are at the forefront of antimicrobial screening. For additional information on the challenges of antimicrobial discovery, determining mechanism of action, and clinical development, I point the reader towards a number of excellent reviews [205][40][144][88].

Molecules Primary Screen PAINs Screen Secondary Assays Additional Assays

Growth inhibition Reporter or in vitro assay Eukaryote toxicity Defines bioactivity Membrane Target identification Compound Library activity Synthetic molecules in vivo activity and purified natural Solubility products Resistance mechanisms Extract Library Growth inhibition in vitro assay Redox Crude natural Confirm whole-cell activity SAR product extracts Defines target potential affinity Crystallography Fluorescence

Figure 1.1: Simpliﬁed screening processes for discovering new antibiotics and uncovering their mechanism of action

1.2.1 Identifying bioactive molecules through growth inhibition

Researchers have been screening bacterial and fungal natural product extracts for antimicrobial activity since the discovery that a Penicillium notatum colony could inhibit growth of nearby bacteria [96][162] [144]. Following the production of penicillin, systematic screening of natural product extracts from soil actinomycetes led to the discovery of actinomycin, streptothricin, and streptomycin [277]. Subsequent work led to the Chapter 1. Introduction 5 discovery of many of the antibiotics and other drugs that are still in use today, including macrolides (tylosin and spiramycin), aminoglycosides (neomycin and kanamycin), β-lactams (cephamycin and carbapenems), tetracyclines (tetracycline, chlortetracycline, and oxytetracycline), chloramphenicol, and many more. Despite, or perhaps because of, the simplicity of these techniques, growth inhibition screens are still widely used today. Teixobactin was recently discovered by screening extracts of previously unculturable soil organisms against S. aureus [163]. In addition, our lab has recently screened Strepto- myces extracts containing an engineered constitutively active pleiotropic regulator AfsQ1 against B. subtilis and M. luteus [63]. Therefore, screening for growth inhibition of indi- cator organisms by natural product extracts has been one of the most fruitful platforms for drug discovery over the past 80 years and researchers continue to use it to discover novel bioactive molecules. There have also been a significant number of antimicrobial screens using synthetic compound libraries. One example is a screen performed by GlaxoSmithKline of about 500,000 synthetic compounds against growth of E. coli and S. aureus [205]. The E. coli screen encountered many nuisance compounds and did not yield exploitable hits. The S. aureus screen identified thousands of molecules, approximately 300 of which possessed antibacterial activity against S. aureus as well as one other Gram-positive or Gram-negative pathogens and which were judged to be chemically tractable. The great majority of these, however, were subsequently ruled out for follow-up due to non-specific disruption of the membrane potential or permeability.

Figure 1.2: Examples of bioactive molecules containing PAIN substructures Chemical structures of drugs with their pan-assay interference (PAIN) sub-structures highlighted in red, including mitomycin C (quinone and aziridine), rifampicin (hydrox- yphenylhydrazone), apomorphine (catechol), and cephalexin (β-lactam).

This challenge of non-specific activity is a common issue in screens for growth inhibition. Molecules can inhibit growth through a variety of promiscuous and non-specific Chapter 1. Introduction 6 mechanisms, including aggregation, membrane disruption, and redox activity [13]. These pan-assay interference compounds (PAINs) function as reactive chemicals rather than discriminating drugs. It has been estimated that 5-12% of the molecules in a typical academic screening library contains PAINs [15]. Some examples of PAIN substructures that can be found in chemical libraries and reported as specific inhibitors include the redox cyclers quinones and catechols, the unspecific covalent modifier rhodanine, and the membrane disruptor curcumin [13]. While there are structure-based PAINs filters that can be used to identify common troublesome sub-structures, indiscriminate use of these tools can be overly simplistic [16]. For example, multiple natural products contain PAINs sub-structures, such as the redox-active quinone found in the DNA cross-linker mitomycin C (Figure 1.2) [14]. This highlights the need to carefully curate the molecules used as part of a screen. For example, while corporate compound collections are heavily biased towards compounds that follow Lipinski’s ’rule of five’ [164] [205], which suggests that an orally active drugs should have no more than 5 hydrogen bond donors, 10 hydrogen bond acceptors, a molecular mass of less than 500, and a logP that is less than 5, known antibacterials do not generally follow these rules. Therefore, care must be taken when evaluating high-throughput screening libraries and be on the lookout for hits with promiscuous activity. Assays such as dynamic light scattering can be used to identify a compound’s solubility while numerous different assays can be used to identify membrane-active molecules. The best way to identify target specificity is to use multiple, orthogonal assays that use different readouts and are not susceptible to the same forms of interference. If you identify an antimicrobial hit that does not have non-specific activity, the next challenge is to identify the compound’s mechanism of action. These mechanism of action studies can include radiolabelled macromolecular synthesis [59], mutation analysis in resistant strains [127], in vitro assays against individual protein targets [95], sensitivity or resistance of specific strains in knock-down or deletion libraries [207], reporter strains for specific bacteria stress responses [60], or identification of morphological perturbation induced by the molecule [191]. Simplifying and accelerating mechanism of action analysis is a robust area of research. For example, one method based only on growth-inhibition was recently reported where the authors discovered that kinetically monitoring bacterial growth at sub-inhibitory concentrations of antibiotics with different targets generates unique dose response profiles. These profiles can be used to identify the mechanism of action for new molecules [170]. The benefit of using growth inhibition as a readout for a screen is that you will catch the largest possible number of hits covering a wide range of biology. The challenge Chapter 1. Introduction 7 lies in developing effective and efficient secondary and tertiary assays to ensure proper prioritization of these hits based on physical and chemical properties as well as their mechanism of action. One method that gains the benefit of selecting for whole-cell active molecules as well as leading more directly to possible mechanisms of action are the use of whole-cell reporter assays.

1.2.2 Identifying mechanism of action through reporter screens and assays

Whole-cell reporter assays help identify biologically active molecules while also providing initial insights into their mechanisms of action. These assays rely on an antibiotic activating stress response or other pathways that result in changes to gene expression and can utilize a variety of readouts, including fluorescence (green fluorescent protein (GFP) or yellow fluorescent protein (YFP)), luminescence (luciferase), β-galactosidase synthesis (lacZ ), morphological changes, growth sensitivity, or resistance. Depending on the throughput and fidelity of a particular reporter assay, it may be used either during primary screening or as a secondary screen to narrow down possible mechanisms of action of a molecule identified through a previous screen. While not as old as screening for simple growth inhibition, the use of reporter outputs has been used since at least the late-60’s. An early example of a morphological-based reporter screen was performed by Merck to identify inhibitors of cell wall synthesis. By screening molecules for the generation of spheroplasts, they discovered a number of antibiotics, including fosfomycin, cephamycin C, thienamycin and several carbapenems [241][120][249]. In the following decades, more sophisticated methods have been used to correlate induced cell morphology to a molecule’s mechanism of action, an approach called cytological profiling. This technique uses fluorescent microscopy to quantify changes to a cell’s length, width, chromosome condensation, and other measures following treatment with a molecule of interest [206][191][158]. The method has recently been used to identify the mechanism of action of novel lipoprotein trafficking inhibitors in Gram-negative bacteria [178], the antimicrobial peptides tyrocidine and gramicidin S [282], and the B. subtilis natural products bacillaene and bacillaene B [192]. I will discuss more about the impact of antibiotics on bacterial cell size and morphology in a subsequent section. A number of reporter systems have been developed that output light, either through luminescence or fluorescence, in response to the activation of a stress response pathway (Figure 1.3A and 1.3B). These strains generally fuse the promoter for a stress response, such as cell envelope stress or the SOS response, to a luciferase operon such as luxABCDE Chapter 1. Introduction 8

[181] or to a gene for a fluorescent protein such as gfp or yfp [79]. When the cell is treated with a molecule that activates the pathway, the luciferase or fluorescent protein is also expressed and can be quantified. For example, a PywaC-lux strain of B. subtilis was con- structed to respond to increased expression of the (p)ppGpp synthase YwaC in response to inhibition of the cell envelope [71]. This strain was used to screen 26,000 molecules and led to the discovery of nine novel cell wall-active compounds [60]. Another example is the use of an E. coli PompC -yfp reporter strain in the discovery of a cyclic peptide that inhibits σE, an extracytoplasmic function sigma factor that is critical for maintenance of the cell envelope and virulence in a number of Gram-negative pathogens [79]. In addition to the use of specific gene promoters, Davis et al. used random cloning to create a series of antibiotic responsive promoter fragments. Testing them against a range of antibiotics with diverse targets created a series of response profiles that could be used to identify the mechanism of novel molecules [179]. This collection of 14 strains has recently been used to investigate the mechanism of action for the novel S. aureus active synthetic molecule propyl-5-hydroxy-3-methyl-1-phenyl-1H-pyrazole-4-carbodithioate (HMPC) [142]. Another frequent tool used in reporter assays is the synthesis of β-galactosidase, or lactase, from the gene lacZ (Figure 1.3C and 1.3D). This glycoside hydrolase enzyme can be combined with X-gal (5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside), such that expression of β-galactosidase results in breaking the glycosidic bond in X-gal and formation of an intense blue product that is easy to identify and quantify. This system is frequently used for blue-white screening or for visualizing particular bacterial stress path-

ways. For example, a PdinC-lacZ reporter strain created to investigate the SOS response [166] has been used to identify molecules that induce DNA damage [139], and a strain of B. subtilis was created to constitutively synthesize intracellular β-galactosidase, which is released and detected on inhibition of cell wall synthesis [156] or membrane disruption [86]. Additional examples of various reporter strains used to understand the impact of antibiotics on the cell and characterize the mechanism of action of new antibiotics are given in Table 1.1. The explosion of genetic data and tools to quickly manipulate and sequence bacterial DNA has accelerated the development of chemical-genetic screening platforms for the discovery of new antimicrobials [100] [230]. One example of this is the use of inducible antisense libraries to identify strains with speciﬁc gene knockdowns that are particularly sensitive or resistant to a molecule or library of molecules. For example, S. aureus strains with antisense knockdowns of the condensing enzyme (FabH/FabF) in fatty acid biosynthesis were used to identify natural product extracts for speciﬁc fatty acid synthesis inhibitors, leading to the discovery of the novel antibiotic platensimycin [244]. Also, a Chapter 1. Introduction 9

A B luxB luxC luxA Fosfomycin P Cell wall inhibition Positive ywaC

luxD e c

+ n e c

luxE s e n i

P m ywaC P u

erm ywaC L

Negative

[Concentration] C D Bacitracin lacI Lipid II inhibition Positive

bla lacZ + erm

P liaI P liaI

P liaI Negative

Figure 1.3: Whole-cell reporter assays for determining a molecule’s mechanism of action A. Sample of a lux reporter plasmid using the PywaC promoter. B. Sample output of a lux reporter system. C. Sample of a β-galactosidase reporter plasmid using the PliaI promoter. D. Sample output of a lacZ reporter system. Chapter 1. Introduction 10

Promoter Output Target References liaI lux Cell wall [149] liaI lacZ Cell wall [172] σW lacZ Cell wall [45] vanH lacZ Cell wall [68] thiM lacZ Riboswitches [167] dinC lacZ DNA damage [166] [139] recA lux DNA damage [180] ompC yfp σE/Outer membrane [79] ywaC lux Cell wall [60] ampC nitroceﬁn Cell wall [190] fabD lux Fatty acid [278] cpxP lacZ Virulence [273] sulA lacZ DNA [233] sulA rfp DNA [202] cda gfp DNA [2] recA gfp DNA [87] cspA, sulA lux Translation, DNA [234] fabHB lux Fatty acid [93] yorB, yvgS, yheI, ypuA lux DNA, transcription, trans- [268] lation, and cell wall cspA, ibp, P3rpoH lacZ Cold shock, cytoplasmic [27] stress, and protein misfold- ing pbp2, tcaA, vraSR, sgtB, lacZ Cell wall [250] and lytR dinB, yneA, yorB, fabHB, lux Everything [134] glpD, ytrA, ywoB, yrzI, ypbG, ydeK, yvgS, expZ spa, hla, saeP1, saeP3, lux, gfp Virulence [101] agrA, RNAIII, sarA, sarS, map, rot, lukFS, eap, fnbA, fnbB, srtA, clfA, ami

Table 1.1: Reporter strains used to identify antibiotic MOA Chapter 1. Introduction 11

pathway-specific synthetic lethal approach has been used by screening otherwise viable S. aureus strains containing deletions in the wall teichoic acid synthesis pathway (∆tarO, ∆dltA) against a small molecule library to identify that the molecule amsacrine, used as an anticancer agent, also targets DltB [204]. In addition to the pathway specific screens, there are also strain libraries containing gene deletions or knockdowns in S. aureus [92] [74] [287], B. subtilis [207] [150], E. coli [12] [73], Salmonella Typhimurium [212], Acinetobacter baylyi [67], and Pseudomonas aeruginosa [117]. For example, a S. aureus antisense library was used to determine that the structurally related molecules DMPI and CDFI target the cell wall through an uncharacterized gene SAV1754, which the authors hypothesize is a peptidoglycan flippase [130].

In addition to strain libraries, deep sequencing technology such as RNA-Seq or whole genome sequencing of resistant strains is providing new insights into antibiotic targets. RNA-Seq not only uses wild-type strains without the genetic perturbations of genomic libraries but can capture the response of both coding and non-coding RNAs. For example, RNA-Seq of Pseudomonas putida [185] and Acinetobacter oleivorans [121] with sub- inhibitory concentrations of known antibiotics was used to create expression profiles for molecules with distinct molecular targets. And the generation, isolation, and sequencing of resistant strains can help identify direct targets, such as with the riboflavin riboswitch inhibiting molecule ribocil [127]. This technique is most useful when resistance is aquired from a mutation in the target, since resistance may also arise through the up-regulation of multi-drug efflux pumps or pleiotropic resistance elements, which may be less useful in pointing towards the antibiotic’s molecular target. As the technologies to enable these high-throughput and data-intensive methods continue to become cheaper, easier, and even open-source, such as 3D printed fluorescence imaging boxes [99], these techniques will continue to develop.

These reporter assays combine the best of simple growth inhibition assays that help identify a wide range of bioactive molecules in a whole-cell context with some added specificity of possible mechanisms of action. They can vary in specificity, from patterns of global transcription disruption, activation of specific stress response pathways, to individual deletions, mutations, or knockdowns that confer resistance. While new strains, assays, and tools are continually being developed, many of the molecules identified in these screens do not progress into further pre-clinical studies and many of the assays are not tested with molecule or extract libraries beyond the initial screen. Therefore, it is a challenge to sort through the libraries of forgotten molecules and screening assays to identify molecules with promising lead characteristics. Chapter 1. Introduction 12

A Deletion Library Overexpression Library Antisense Library P P Kan Chromosomal gene xyz Promoter-driven xyz Promoter-driven deletions ORF expression antisense expression

Kan Kan

CRISPRi Library Tn-Seq Resistant Mutations

dcas9 Promoter-driven Tn Chromosomal transposon Random chromosomal sgRNA Cas9 and guide RNA insertion mutations expression

Tn Tn

B C Growth Suppression Growth Rescue

Sequencing \ Strain ID Pinned library grown on media containing antibiotic

Figure 1.4: Chemical-genetic assays for determining MOA A. Examples of genetic strain libraries. B. Libraries can be pinned to agar slabs containing the molecule of interest to identify genes capable of growth suppression or rescue and potential targets. C. Libraries or cultures can also be grown in liquid culture containing the molecule of interest. Any growth rescue mutation can then be identiﬁed through DNA sequencing. Chapter 1. Introduction 13

1.2.3 Target-based screening

If reporter and chemical-genetic screens provide initial information on possible mechanisms of action of new antibiotics, target-based approaches narrow the spectrum of possible activity even further. These screens and assays test molecules against a specific molecular target by identifying molecule-target binding or inhibition of target activity in vitro. However, while numerous target-based screens have been performed, the technique has not led to significant number of molecule leads. In the late 90’s to early 2000’s, GlaxoSmithKline ran 67 high-throughput in vitro campaigns on diverse antibacterial targets, from fatty-acid synthesis, DNA replication, translation, cell-wall synthesis, and more [205]. These screens utilized the company’s in-house compound collection of up to 530,000 compounds and represented a significant effort at the time for a single therapy area. However, only 16 of the target screens gave rise to hits with only 5 resulting in leads. A number of these hits turned out to be non-specifically toxic to both mammalian and bacterial cells, usually as a result of indiscriminate cell-membrane disruption. This example demonstrates the challenges in identifying and developing in vitro screening hits [242]. This non-specific activity caused by PAIN molecules is a challenge shared with assays investigating whole cell growth inhibition. First, while a molecule may demonstrate activity in an in vitro assay, that activity may not lead to growth inhibition in a whole-cell environment. This could be due to an inability to enter the cell, being removed from the cell by efflux, or degradation pathways that prevent the molecule from accessing its intended target. For example, the activity of the essential cell-division protein FtsZ has been a target of significant interest for new antibiotic development [226]. The GTPase protein localizes to the site of septum formation, polymerizes into the Z-ring and provides a scaffold for other divisome proteins [3]. There are dozens of reported synthetic and natural product inhibitors of FtsZ, many of which are identified using in vitro assays for FtsZ polymerization or GTPase inhibition [111]. However, a review of nine published FtsZ inhibitors found that four of these molecules likely derive their activity through non- specific mechanisms, such as aggregation of the molecules, and the reported activity of four other molecules could not be reproduced [9]. This suggests that many of the reported FtsZ inhibitors are promiscuous molecules that would not be good development candidates. Therefore, target-based assays are most useful not as screening platforms, but as assays to confirm possible mechanisms of action based on orthogonal assays and to better understand the exact mechanism of inhibition. For example, in vitro assays have been Chapter 1. Introduction 14 successfully used in understanding the mechanism of action for topoisomerase inhibitors, such as simocyclinone D8 [95] [39]. Topoisomerases, including gyrase and topoisomerase IV, are essential enzymes that play a vital role in controlling DNA topology during replication and transcription, including the inverconversion between relaxed–supercoiled, knotted–unknotted, and catenated–decatenated. They act by binding two sections of double stranded DNA, forming a transient double stranded break, passaging the intact strand through the cleaved strand, and re-ligating the break [37] [276]. In vitro assays have been developed to measure inhibition of each of these steps and have been used in my studies to understand the mechanism of action of a gyrase targeting molecule, EN-7. Even beyond the field of antibiotics, target-based approaches to drug discovery have not led to significant innovations. A review of the discovery strategy and molecular mechanism of action of new molecular entities and biologics approved by the FDA between 1999 and 2008 found that 28 of first-in-class-molecular drugs were discovered through phenotypic screening while 17 came from target-based screens [255]. However, of the 164 follower drugs approved in this time frame, 83 were discovered with target-based approaches while 30 came from phenotypic screens. Therefore, phenotype-based assays were more productive in novel therapeutic areas whereas target-based approaches fared well in areas with more established biology. With a need for new targets and chemistry, this suggests that the use of new phenotype-based methods provide a better opportunity for uncovering antibiotics to help stem rising rates of antibiotic resistance.

1.2.4 Novel screening approaches

In many ways the search for new antibiotic leads has not changed significantly over the past few years. Many researchers use growth inhibition to identify biologically active molecules followed by a collection of assays to identify the molecule’s target, toxicity, and chemical properties. However, there is exciting work being done to develop new and unconventional screening approaches to identify molecules with novel mechanisms of action. Four examples include a platform for discovering antibiotic adjuvants that disrupt resistance mechanisms to existing antibiotics [57], assays to find molecules that are active under different nutrient and growth conditions [81], the use of thermal proteome profiling (TPP) to directly identify antibiotic targets [173], and the use of Streptomyces sporulation, a developmentally regulated form of cell division, as a platform for identifying new bioactive molecules [139]. One successful strategy that has emerged for overcoming rising rates of antibiotic resistance is the use of adjuvants. These molecules are combined with antibiotics and Chapter 1. Introduction 15

block resistance elements, freeing the antibiotic to its original effect [78] [22] [75]. The first example of this approach was the β-lactamase inhibitor clavulanic acid, which was discovered in the 1970s [220]. Now, numerous co-formulations of β-lactam antibiotics and β-lactamase inhibitors have been brought to market or are in clinical trials [36]. A new approach towards the discovery of novel antibiotic adjuvants is the antibiotic resistance platform (ARP) [57]. This E. coli library consists of strains containing individual resistance genes, the expression of which is controlled through two promoters of differing strength. This system has been used to discover aspergillomarasmine A, a novel metallo- β-lactamase (MBL) inhibitor, that is a rapid and potent inhibitor of the β-lactamase enzymes NDM-1 and VIM-2 [147]. Another unique screening method has been to identify molecules that target bacteria under particular metabolic stress [294] [62] [81] [80]. When E. coli is grown under nutrient limiting conditions an additional 119 genes become essential due to the expanded need for biosynthetic capacity of amino acids, vitamins, and more [12]. Therefore, screening under such nutrient limiting conditions opens a number of new potential target pathways that would be masked when grown in nutrient rich media. This metabolic suppression screening approach then uses an array of metabolites and metabolite pools that can be screened for supressors of inhibitory molecules, leading the way to potential mechanism of action and targets. This strategy has led to the discovery and characterization of three new antibacterial compounds, MAC168425, MAC173979 and MAC13772, which target glycine metabolism, p-aminobenzoic acid biosynthesis, and biotin biosynthesis respectively [294]. Thermal proteome profiling (TPP) is a new in situ, proteome-wide approach for determining the targets of bioactive molecules [174]. The technique combines the principle of a cellular thermal shift assay with high-throughput mass spectrometry by measuring the soluble proteome of treated cultures that have been heated to a range of temperatures. This effectively determines the melting behavior of thousands of proteins simultaneously. Since protein–drug interactions typically increase the thermal tolerance of proteins, differences in the melting behaviour between treated and untreated cultures allows for the identification of molecular targets. While this protocol was originally used for kinase inhibitors [227], it has since been adapted for use with antibiotics against E. coli [173]. Finally, our lab has turned to the unconventional bacteria Streptomyces as a screening tool for novel antibiotic molecules [139]. Streptomyces undergo a unique multi-cellular bacterial lifecycle where cell division takes place during a developmentally controlled sporulation program (Figure 1.5) [177]. Since sporulation is accompanied by the synthesis of a coloured pigment [65], small molecule sporulation inhibitors can be identified by eye Chapter 1. Introduction 16

from a disk diﬀusion assay where a white halo surrounds the zone of inhibition. This method has been used by a previous student to identify three novel cell division inhibitors Fil-1, Fil-2, and Fil-3 [139], a molecule that disrupts the coordination between cell growth and division in B. subtilis called Min-1, and my work on a novel gyrase inhibitor EN-7. I will discuss each of these molecules in later chapters.

A B

Germination

Aerial Hyphae Development

Sporulation

Streptomyces pristinaespiralis

Figure 1.5: Streptomyces lifecycle Streptomyces undergo a unique multi-cellular bacterial lifecyle, beginning with the germination of a spore and ending with a developmentally controlled cell division event that leads to the formation of new spores. A. Graphical representation of the lifecycle. B. Photos of agar plates with drug disks demonstrating the changing phenotypes of the Streptomyces lifecycle.

1.2.5 Conclusions

The discovery and development of new antibiotics is a pressing medical need. While there has been signiﬁcant work to create new screening methods, discover new molecular targets, and unlock new natural product chemistry, we still lack a robust candidate pipeline. While small companies and organizations like the Pew Trust, the Wellcome Trust, and CARB-X are working to support researchers and companies developing novel antibiotics, challenging reimbursement environments and clinical trial requirements are leading larger companies to pull out of the space. There is a need for new ideas across the science, business, clinical, political, and regulatory spheres in order to combat this challenge. Chapter 1. Introduction 17

1.3 Small molecule impact on bacterial morphology

Bacteria do not go quietly when treated with antibiotics. They respond in complex and interesting ways by activating various stress response systems that can strongly impact metabolism and reproduction. This makes antibiotics marvelous tools to study bacterial biology. One area of active research that has successfully used antibiotic treatment as a method to further understand bacterial biology is the regulation of cell size and shape. It has been over 50 years since the observation that Salmonella enterica cells could increase in size by increasing their growth rate [228]. Combined with additional experiments on Bacillus subtilis[225], these early results led to the formation of the ‘growth law,’ which has influenced thinking on cell size for decades. Since then there has been significant interest in the mechanisms by which cells control and change their size and morphology. Recent results from detailed single cell, population, and modelling-based experiments have replaced this early ‘growth law’ with a more nuanced view of how bacteria regulate their size. This includes observations that support the ‘adder’ principal in C. crescentus, E. coli, B. subtilis, and P. aeruginosa where cells grown under steady-state growth conditions add a constant volume to their size before division, irrespective of their size at birth [44][257][69]. Other work has focused on nutrient availability and the idea that cell size is coordinated with growth rate through nutrient-dependent changes in both cell cycle progression and biosynthetic capacity [270]. Finally, a combination of modeling and single-cell experiments have shown that E. coli cell size for any steady-state condition can be predicted by projecting all biosynthesis into three variables representing replication initiation, replication-division cycle, and the global biosynthesis rate [236]. However, what about cells that have been disturbed from a steady-state through changes in their environment or by inhibition of key biosynthetic pathways by antibiotics? This question is actually even older than the observation of the impact of nutrients on cell size. In the early 1950’s, Pulvertaft investigated how E. coli cell morphology changed following treatment with a series of five antibiotics: penicillin, streptomycin, aureomycin (chlortetracycline), chloromycetin (chloramphenicol) and terramycin (oxytetracycline) [216]. He observed that at a critical concentration all five antibiotics resulted in bacterial enlargement to a ‘fantastic size and form.’ Also, he observed that the resulting morphologies were dependent on the concentration of the antibiotics. Over the next few decades, various groups would observe the filamentation of E. coli when treated with the DNA crosslinker mitomycin C [252] as well as the impact of penicillin G, cephalothin, and carbenicillin on the surface morphology of S. aureus, E. coli, and P. aeruginosa [148]. Chapter 1. Introduction 18

The decades since this early work has improved our understanding of some of these mechanisms while others remain illusive. Some alterations are the direct results of concerted action by a cell to respond to chemical pertubation while others are caused by the antibiotic directly impacting cellular structure. In this section, I review how activation of the bacterial SOS response, disruption of the cell envelope, and inhibition of cell division by antibiotics disrupt cell morphology. In addition, while not antibiotics, nutrients are vitally important small molecules that signiﬁcantly impact cell morphology. Therefore, I will review known systems that alter cell morphology in response to nutrient conditions. Finally, in reference to a previous section, I will touch on how these small-molecule pertubations of cell morphology are being leveraged to discover novel antibiotics.

1.3.1 Activators of the SOS response disrupt the cell cycle

Induction of cell ﬁlamentation as a result of DNA damage has been observed since the late 1960’s [252]. This elongation is caused by inhibition of the cell cycle via the SOS response, a powerful adaptation system designed to allow cells to recover from genotoxic stress [83][19]. While the SOS response can be induced by a variety of external factors, including UV radiation, pressure [4], reactive-oxygen species [140], and R-loop (transcriptional DNA/RNA hybrids) [284], I will focus on induction by antibiotics. Brieﬂy, the SOS response is activated when RecA is recruited to ssDNA by either RecBCD, which recognizes double-strand breaks (DSB), or RecFOR, which recognizes DNA nicks and gaps (Figure 1.6). RecA binds ssDNA and catalyzes the auto-proteolysis of the repressor LexA. Under normal conditions LexA represses the SOS regulon by binding its cognate LexA box sequence on their promoters. Therefore, LexA proteolysis leads to derepression of this regulon, comprised of about 40 genes in E. coli [56], 33 genes in B. subtilis [10], 16 genes in S. aureus [54], and 15 genes in P. aeruginosa [55]. While many of the SOS regulon are involved in homologous recombination and other genome repair pathways it also induces proteins that directly interfere with cell division. SulA is one of the proteins induced by the SOS response in E. coli and related Gram- negative bacteria. The protein, along with its Gram-positive homolog YneA, binds FtsZ and inhibits cell division by increasing the apparent critical concentration required for FtsZ-ring assembly, resulting in cell elongation [131][168][132][25][145][52]. There are numerous antibiotics capable of activating the SOS response, including those that directly damage DNA, those that target DNA gyrase, as well as trimethoprim, aminoglycosides, and β-lactam antibiotics. As mentioned previously, one of the earliest examples of a DNA damaging agent Chapter 1. Introduction 19

Mitomycin C Ciprofloxacin Trimethoprim Ampicillin DNA crosslinks GyrA inhibition Folate biosynthesis Cell wall synthesis

dsDNA

RecA

Gap Double stranded break

Activated RecA

LexA LexA

SOS Box SulA SOS Box SulA

FtsZ-ring SulA inhibits FtsZ polymerization and results in filamentation

Figure 1.6: Small molecules activate the SOS response and disrupt the cell cycle Small molecules, including mitomycin C, ciproﬂoxacin, trimethoprim, and ampicillin, activate the SOS response. RecA is recruited to sites of DNA damage, becomes activated, and cleaves the LexA transcription inhibitor. This activates expression of numerous genes, including those that inhibit cell division and result in cell ﬁlamentation. Chapter 1. Introduction 20

disrupting cell morphology is the use of mitomycin C to induce cell filamentation in E. coli [252]. Mitomycin C acts by reducing its quinone functional group and covalently crosslinking the complementary strands of the DNA double helix, resulting in DNA break formation [262]. Daunorubicin and its derivative doxorubicin create DNA adducts by intercalating between the stacked bases of DNA [254]. This intercalation results in lengthening and unwinding of dsDNA [8] [114], which could result in activation of the SOS response. The bleomycins, discovered from cultures of Streptomyces verticillus [266], directly induce double strand breaks in the presence of oxygen and metal cofactors [169]. These molecules, all of which have been used in the clinic as anticancer agents, generate DNA lesions which activate the SOS response and result in inhibition of cell division and cell elongation. One of the most clinically relevant antibiotic targets are the TypeIIA topoisomerases gyrase and topoisomerase IV. These enzymes are composed of a heterotetramer of either GyrA and GyrB for gyrase or ParC and ParE for topoisomerase IV. They are responsible for controlling DNA topology during DNA replication and transcription by inducing or relaxing supercoiling and unlinking catenated DNA strands through ATP-dependent DNA cleavage, passage, and ligation [276]. They are the targets of numerous synthetic small molecules, natural products, and protein toxins. However, activation of the SOS response differs depending on the specific mechanism of gyrase inhibition. Fluoroquinolones such as ciprofloxacin poison gyrase by intercalating within the cleavage DNA site and inhibiting ligation, resulting in numerous double stranded breaks across the genome [159]. Even at sub-inhibitory concentrations, ciprofloxacin is able to induce the SOS response in Salmonella enterica [289] and S. aureus [180]. Aminocoumarins such as novobiocin and coumermycin A1 inhibit the ATPase activity of the enzyme by competitively binding the active site, leading to positive supercoiling and stalling of the replication machinery [161]. Stalling of the replication machinery can result in collisions between the replication and transcription machinery as well as R-loops, which can induce the SOS response [119]. While it has been shown that novobiocin does not induce the SOS response in S. aureus [232] [180] and E. coli [72], treatment does lead to cell elongation in E. coli [191]. Other gyrase inhibitors include a series of structurally related novel bacterial topoisomerase inhibitors (NBTIs), including gepotidacin [94], GSK299423 [21], and NXL101 [31] that act by binding a pre-cleavage DNA-gyrase complex and inhibiting DNA cleavage. However, the ability of these molecules to activate the SOS response and their impact on cell morphology have not been investigated. Finally, simocyclinone D8, a natural product of Streptomyces antibioticus, binds to the N-terminus of the GyrA subunit and inhibits the binding of gyrase with DNA [39] but does not activate the SOS Chapter 1. Introduction 21

response [200] and its impact on cell morphology has not been investigated. In addition to molecules that either directly target DNA or interfere with topoisomerases, there are a number of other antibiotics that activate the SOS response through more indirect methods. Trimethoprim, which prevents the incorporation of thymine into bacterial DNA by inhibiting dihydrofolate reductase [107], resulting in DNA breakage [1], activates the SOS response in E. coli [160], S. aureus [180], and Vibrio cholerae [18] as well as inducing filamentation in clinical Enterobacteriaceae strains [136] and other bacteria. While protein synthesis inhibitors such as aminoglycosides (kanamycin, to- bramycin, gentamicin, and neomycin), tetracyline, and chloramphenicol, do not activate the SOS response in E. coli [235], they do in Vibrio cholerae [18] [17]. Finally, treatment with ampicillin, a β-lactam antibiotic, has been shown to induce the SOS response and activate SulA expression in E. coli through induction of the DpiBA two-component system [182] [183]. However, there are conflicting results in S. aureus, with some results showing no impact of penicillin G treatment on SOS gene expression [180] with other results showing activation of SOS-dependent phage replication [171]. The SOS response is a powerful mechanism for cells to recover from a variety of DNA-targeting stressors. Given that cells would not want to pass damaged DNA to daughter cells, it makes sense that they would delay cell cycle progression until the cell can repair the damaged DNA. Cell growth continues while division is inhibited, resulting in filamentation of rod-shaped cells. Since the early observation of this effect decades ago, this phenotype remains in use today to identify the mechanism of action of new molecules.

1.3.2 Cell envelope determines cell shape

The connection between cell morphology and cell envelope inhibition has been observed since early E. coli experiments in 1952 [216]. The cell envelope is a complex and vital component of bacterial physiology. Gram-positive bacteria are enclosed by the cell membrane and a thick peptidoglycan layer that is decorated with wall teichoic acids (WTA) and lipoteichoic acids (LTA) while the Gram-negative envelope consists of an inner membrane, a thin peptidoglycan layer and an outer asymmetric membrane that is decorated with lipopolysaccharides. Each of these systems are the targets of numerous antibiotics that result in an array of interesting morphologies. The bacterial membrane is a complex target for antibiotics. There are numerous existing and experimental antibacterial treatments that act through disruptions of the bacterial membrane. Antimicrobial peptides, including nisin [283], daptomycin [209], Chapter 1. Introduction 22 polymyxin B [30], and colistin [189] have long been used to treat various bacterial infections. Reports have shown that nisin treatment, which interacts with lipid II and forms pores through the membrane that disrupt both membrane permeability and the electrochemical potential, results in cell shortening and rounding in B. subtilis [135] and E. coli [191]. Other molecules that specifically target the membrane electrochemical potential while leaving the membrane structually intact, including the ionophores nigericin and CCCP, also result in cell shortening in E. coli [191]. In addition to direct disruption of the membrane potential or permeability, membrane integrity can also be disrupted by upstream inhibition of fatty acid synthesis. To highlight the role of fatty acids in determining cell size, an E. coli mutant lacking FabH was found to be 70% smaller than wild-type in nutrient rich conditions [288]. Inhibiting fatty acid synthesis by cerulenin treatment has been shown to reduce cell length in both E. coli and B. subtilis by approximately 25% [271]. Also, in Chapter 3, I report a membrane active molecule called Min-1 that disrupts the coordination between cell growth and cell cycle progression in B. subtilis, resulting in a short cell phenotype. These results suggest that fatty acid synthesis and membrane integrity play an important role in setting the capacity for cell envelope expansion that is required for increasing cell size as the cell grows. Rod-shaped bacteria such as B. subtilis and E. coli maintain their shape with the help of the cytoskeletcal protein MreB [154]. This bacterial actin-like protein forms membrane-associated dynamic filaments that rotate around the cell circumference along with cell wall synthesis enzymes [274][102]. A feedback loop between cell geometry and MreB localization maintains elongated cell shape by targeting cell wall growth to regions of negative cell wall curvature located along the cell length versus the positive curvature located at the poles [269]. The molecule A22 and its analogs bind MreB at the nucleotide- binding pocket and inhibits ATP binding, altering both the kinetics and extent of MreB polymerization [23][38]. Inhibiting MreB disrupts cell elongation and results in a loss of the rod-shape and increase in cell width [264][269]. The cell wall is a vital bacterial structure that provides important protective and structural support for bacteria [11]. This structure, especially peptidoglycan, is an incredibly important historial target for antibiotics, including multiple classes of β-lactams, vancomycin, and bacitracin. It also remains a target of interest for novel antibiotics, including the recently discovered teixobactin that targets lipid II, a peptidoglycan precursor [163]. Disruption of the cell wall results in significant morphological changes across both Gram-positive and Gram-negative bacteria [211]. For example, it has been known since Chapter 1. Introduction 23

1975 that disruption of different penicillin-binding proteins (PBPs), which cross-link peptidoglycan, results in distinct morphological defects in E. coli [246]. Inhibition of the cell elongation directing PBP2 by mecillinam results in rounded cells while inhibition of the septation-focused PBP3 by ampicillin or benzylpenicillin prevents cell separation and results in cell filamentation [246]. Inhibiting teichoic acid synthesis also has a significant impact on morphology. B. subtilis tagO mutants lose their rod shape and swell in size [70]. Chemical inhibition of both of these systems results in a wide variety of morphologies, which can be dependent on the molecule’s specific target, the concentration used, and the treatment time. In addition to peptidoglycan, recent studies have implicated LTAs and WTAs in cell growth, division, and morphogenesis. Across a variety of Gram-positive organisms, preventing WTA expression results in the production of round, severely defective progeny, while preventing LTA biosynthesis causes major defects in septum formation and cell separation [256] [210] [229]. Table 1.2 lists a variety of observed morphological effects caused by inhibition or disruption of the cell envelope components of both Gram-positive and Gram-negative bacteria. The cell wall is an incredibly important part of bacterial physiology and a prime target of numerous antibiotic classes. Both its direct effect on cell shape through physical structural support as well as its connection to numerous stress responses systems, including the SOS response, influence the wide array of morphological pertubations induced by cell wall targeting antibiotics.

1.3.3 Direct disruption of FtsZ disrupts the cell cycle

The highly conserved FtsZ protein is the principle driver of cell division [201]. Its polymerization from a monomeric form to the FtsZ-ring polymer is a vital step in the formation of the divisome, which contains the proteins responsible for formation of the division septum and eventual separation of the daughter cells[26] [224]. This places it as a key regulator in the bacterial cell cycle. While there are numerous systems that use FtsZ inhibition to control the cell cycle, including the DNA damage-sensing SOS response and the nutrient-sensing UgtP, there are also small molecules that directly inhibit FtsZ function and result in cell ﬁlamentation. Its central role in cell division has resulted in signiﬁcant interest as a target for new antibiotics, with over 22 proposed natural products and many more synthetic FtsZ inhibitors [111]. One of the most extensively studied FtsZ inhibitors is the synthetic molecule PC190723, which is a derivate of 3-methoxybenzamide (3-MBA). 3-MBA, while Chapter 1. Introduction 24

Molecule Primary Target Bacteria Phenotype References A22 MreB E. coli Cell rounding [264] Ampicillin PBP3 E. coli Cell elongation [246] [223] Benzylpenicillin PBP3 E. coli Cell elongation [246] Bacitracin UppP S. pyogenes Cell shortening [251] Carbenicillin PBP3 E. coli Cell elongation [223] Cephalexin PBP3 E. coli Cell elongation [223] [35] [269] Cephalothin PBP1b S. aureus Cell blebbing [148] Cephalothin PBP1b E. coli Cell blebbing [148] and rounding Cephalothin PBP1b P. aeruginosa Cell blebbing [148] and elongation Clomiphene UppS B. subtilis Cell width in- [89] crease d-Cycloserine Ddl S. pyogenes Cell enlargement [251] Daptomycin Cell membrane E. faecalis Cell elongation [50] Flavomycin PBP1a S. pyogenes Inhibition of cell [251] separation Fosfomycin MurA E. coli Cell elongation [112] Furazlocillin PBP3 E. coli Cell elongation [35] Mecillinam PBP2 E. coli Cell rounding [247] [246] [264] Oxacillin PBP3 S. pyogenes Cell enlargement [251] Ramoplanin MurG S. pyogenes Cell shortening [251] Penicillin G PBP4 S. aureus Cell blebbing [148] Penicillin G PBP4 E. coli Cell blebbing [148] and rounding Penicillin G PBP4 P. aeruginosa Cell blebbing [148] and elongation Piperacillin PBP3 E. coli Cell elongation [35] Targocil TarG S. aureus Inhibition of cell [43] separation and cell enlargement Tunicamycin TarO/MraY Listeria mono- Cell rounding [292] cytogenes Tunicamycin TarO/MraY S. aureus Inhibition of cell [42] separation and cell enlargement Vancomycin D-Ala-D-Ala E. faecalis Cell elongation [50] Vancomycin D-Ala-D-Ala S. aurues Inhibition of cell [237] separation Vancomycin D-Ala-D-Ala E. coli Cell blebbing [128]

Table 1.2: Cell morphologies induced by cell wall inhibiting antibiotics Chapter 1. Introduction 25

initially known as an inhibitor of ADP-ribosyltransferase, was discovered to cause cell ﬁl- amentation in B. subtilis [195]. A genetic analysis of 3-MBA-resistant strains discovered FtsZ mutations A47P and V173F, leading to the hypothesis that the molecule inhibits GTPase activity of the essential divisome protein. While the molecule is able to enter the cell and target an essential and unique cellular target, the molecule has a very high minimum inhibitory concentration. Therefore, PC190723 was discovered through a medicinal chemistry program designed to determine the structure-activity relationships of the benzamide [115]. The potency of PC190723 is over 2000 times greater than 3-MBA against S. aureus and drug resistant S. aureus, however has little activity against enterococci and Gram-negative pathogens. This demonstrates that FtsZ can be directly inhibited by small molecules in addition to being inhibited by stress response activated inhibitors.

1.3.4 Nutrient availability impacts both growth and the cell cycle

One of the most significant determinants of cell size is nutrient availability [245][270]. While nutrients are not antibiotics, which have been the focus of this discussion to this point, cells have complex mechanisms to interact with these vital molecules that significantly impact their growth, cell cycle, morphology. Examples of this interaction include the UgtP system that detects UDP-glucose, induction of (p)ppGpp and the remodeling of gene expression under amino acid starvation, and even interactions between pyruvate metabolism and cell division. The most direct examples of the connection with nutrient availability with cell size are the enzymes UgtP in B. subtilis [279] and OpgH in E. coli [123]. UgtP is a glucosyltransferase responsible for synthesis of the diglucosyl diacylglycerol (DiGlcDAG) anchor for lipoteichoic acid (LTA) from the precursor UDP-glucose. An increase in UDP-glucose availability under high nutrient conditions results in increased UgtP expression. Increased expression of the membrane-associated protein results in localization at the division site and interaction with FtsZ, inhibiting cell division [53]. While being functionally analo- gous to UgtP, OpgH shares no homology, has distinct enzymatic activities, and appears to inhibit FtsZ assembly through different mechanism than its B. subtilis counterpart. An inner-membrane glucosyltransferase that synthesizes osmoregulated periplasmic glucans (OPGs) from UDP-glucose, OpgH localizes to the division ring in nutrient-rich conditions and blocks FtsZ-ring formation by sequestering FtsZ monomers and preventing them from assembling into single-stranded polymers [123]. In both of these systems it is increased nutrient content, specifically UDP-glucose, that leads directly to inhibition of Chapter 1. Introduction 26 cell division and elongated cells. The stringent response is a significant modulator of bacterial gene expression [113]. It is controlled by members of the RelA–SpoT homologue (RSH) protein family in response to stress and is mediated by two related alarmone nucleotides, guanosine tetraphosphate and guanosine pentaphosphate, that are collectively referred to as (p)ppGpp. The regulatory role of (p)ppGpp in general metabolism is exerted by direct binding to the β’- and ω-subunits of RNA polymerase (RNAP) in E. coli or indirect changes in the concentrations of GTP and ATP. In both E. coli and B. subtilis, activation of the stringent response results in reduction in growth rate and cell size [231] [259] [288] [215]. In Mycobacterium smegmatis, low (p)ppGpp levels results in slowed growth in nutrient poor conditions and elongated cells while high levels reduce cell length [109]. In Caulobacter crescentus, an increase in the intracellular concentration of (p)ppGpp slows cell growth, swarmer-to-stalked cell differ- entiation, and delays the initiation of chromosome replication [108]. A (p)ppGpp null strain of Streptomyces coelicolor was unable to complete the development of aerial hyphae into spores [122]. It has been suggested that the physiological impact of (p)ppGpp synthesis is to coordinate cell envelope biogenesis with synthesis of cytoplasmic material to ensure cytoplasmic volume does not exceed the ability of the membrane to expand and compromise membrane integrity [271]. Finally, a connection has been made between cell division and glycolysis in B. subtilis [186] [187]. Mutant strains with disrupted pyruvate synthesis exhibit major defects in FtsZ assembly and cell division, including mislocalised Z rings and aberrant division events at the cell poles. Importantly, these defects can be completely alleviated by the addition of exogenous pyruvate, implicating pyruvate itself as a key metabolite in regulating division. These results highlight the complex interactions between bacteria and the molecules in their environment, both those that are designed to kill them and the molecules they need in order to survive. While nutrients clearly play a vital role in regulating bacterial morphology, many of the specific mechanisms of growth and cell cycle control are still not fully understood and are an active area of exciting research.

1.3.5 Cytological proﬁling identiﬁes antibiotic mechanisms of action

Technical advances in automated ﬂuorescence microscopy, high-throughput liquid han- dling, and new computer vision analysis algorithms have created new opportunities in Chapter 1. Introduction 27

drug discovery [151] [41] [152]. As demonstrated by the preceeding sections, the link between the morphological effects induced by an antibiotic and its mechanism of action goes back decades. Combining these induced morphological phenotypes with new image analysis technology has led to the development of cytological profiling as a new assay for investigating mechanism of action [206]. This technique uses fluorescent stains to label cellular components, such as the cell membrane and the chromosome. Fluorescent microscopy is used to identify unique cytological profiles, consisting of effects on cell area, cell length, cell width, and chromosome condensation, following antibiotic treatment. These are then analyzed by principle component analysis to identify clustering of the morphological effects based on the molecule’s target. This can be used to not only distinguish between different molecular targets, such as the ribosome or bacterial gyrase, but different mechanisms of inhibition for the same molecular target. For example, molecules such as ciprofloxacin and nalidixic acid, which target the GyrA subunit of bacteria gyrase, induce a distinct profile from novobiocin, which targets the GyrB subunit [191]. This technique has been used to identify that spirohexenolide A collapses the proton motive force across the bacterial membrane of S. aureus [191] and that the B. subtilis natural product bacillaene inhibits translation [192]. In addition to its use as a mechanism of action assay, cytological profiling has also been proposed as a diagnostic tool for rapid phenotypic antimicrobial suseptibility testing against pathogenic S. aureus [217]. Changes in cell morpholoy of clinical isolates following antibiotic treatment allowed the strains to be correctly characterized as MRSA and MSSA within 1 hour. Likewise, daptomycin susceptible and daptomycin non-susceptible S. aureus strains were correctly classified after 30 min of antibiotic treatment.

1.3.6 Conclusion

Bacteria are complex organisms. From antibiotics to nutrients, they are impacted by and react to numerous changes to their chemical environment. One vital aspect of these responses is changes to cell morphology. Growth in different media conditions results in cells of different size, and treatment with different antibiotics results in a whole array of different morpholgoies. While we have understood these changes on a macro level for decades, and even used them to discover new antibiotics, we are still unraveling the underlying mechanisms that cause these morphological changes. I hope that this review and the subsequent chapters highlight the opportunities to leverage these morphological features and other tools to uncover new chemistry and biology that will hopefully, one day, lead to new therapies and a better understanding of these small, complex organisms. Chapter 1. Introduction 28

1.4 Thesis objectives and outline

The object of my Ph.D. research is to use the unique development cycle of Streptomyces to discover novel biologically active molecules and to characterize the mechanisms of action of these new molecules. This work has been divided into three separate chapters. In Chapter 2, I highlight a new assay for uncovering membrane active molecules and use this assay on a synthetic library of molecules that inhibit growth of B. subtilis.I found that of 3,520 molecules that were screened, 138 disrupted the membrane potential while 151 disrupt membrane permeability. Considering the challenges of discovering molecules with specific activity, I was expecting more of these syntheic molecule to be membrane active. In Chapter 3, I describe a new bioactive molecule, Min-1, that inhibits Streptomyces sporulation, is active against a range of Gram-positive bacteria and induces a unique short cell phenotype in B. subtilis. I found that Min-1 is active against the cell envelope and that disruption of the coordination between cell growth and the cell cycle is the cause of the short cell phenotype. Finally, in Chapter 4, I describe a novel gyrase- targeting molecule called EN-7 and I find that Streptomyces venezuelae sporulation is very sensitive to DNA damage and gyrase inhibition, with a variety of different known DNA and gyrase targeting antibiotics strongly inhibiting sporulation. EN-7 itself is highly active against Gram-positive pathogens, including extensively resistant S. aureus, and EN-7R S. aureus mutants contain mutations in either GyrA or GyrB gyrase subunits. The molecule strongly inhibits gyrase induced supercoiling in a mechanism that differs significantly from fluoroquinolones and aminocoumarins. These results not only introduce a series of new biologically active molecules, but highlight the opportunities for using Streptomyces sporulation to discover new and interesting molecules that disrupt the biology of numerous bacteria. Chapter 2

Membrane activity proﬁling of small molecule B. subtilis growth inhibitors utilizing novel duel-dye ﬂuorescence assay

The data reported in this chapter have been published in the following peer-reviewed publication: Scott McAuley, Alan Huynh, Tomasz Czarny, Eric D. Brown and Justin R. Nod- well (2018) Membrane activity profiling of small molecule B. subtilis growth inhibitors utilizing novel duel-dye fluorescence assay. Med. Chem. Communic., 9, 554-561 The initial screen against B. subtilis was performed by Tomasz Czarny in the lab of Eric Brown. I performed initial testing of the fluorescent dyes and testing the assay against vancomycin, CCCP, and nisin. Screening the bioactive library was performed by Alan Huynh, while I performed the data analysis and interpretation.

2.1 Abstract

Small molecule disruption of the bacterial membrane is both a challenge and of interest for drug development. While some avoid membrane activity due to toxicity issues, others are interested in leveraging the eﬀects for new treatments. Existing assays are available for measuring disruption of membrane potential or membrane permeability, two key characteristics of the bacterial membrane, however they are limited in their ability to distinguish between these properties. Here, we demonstrate a high throughput assay for detection

29 Chapter 2. B. subtilis membrane activity profiling 30 and characterization of membrane active compounds. The assay distinguishes the eﬀect of small molecules on either the membrane potential or membrane permeability using the

ﬂuorescent dyes TO-PRO-3 iodide and DiOC2(3) without the need for secondary assays. I then applied this assay to a library of 3,520 synthetic molecules previously shown to inhibit growth of B. subtilis in order to determine the frequency of membrane activity within such a biologically active library. From the library, we found 249 compounds that demonstrated signiﬁcant membrane activity, suggesting that synthetic libraries of this kind do not contain a plurality of membrane active molecules.

2.2 Introduction

The bacterial membrane is a complex target for antibiotics. There are numerous existing and experimental antibacterial treatments that act through disruptions of the bacterial membrane. Antimicrobial peptides, including nisin [283], daptomycin [209], polymyxin B [30], and colistin [189], have long been used to treat various bacterial infections. There are also a number of small molecules under development that are designed to mimic the membrane effects of antimicrobial peptides [106] as well as inhibit the growth of biofilms [125]. In addition to using membrane active molecules as stand-alone treatment, these membrane active molecules are being investigated for use as adjuvants to potentiate known molecules against both Gram-negative bacteria such as E. coli [267] and Gram- positive bacteria such as S. aureus [90]. The bacterial membrane is a complex structure. Although the composition can vary from organism to organism, the single phospholipid bilayer of Gram-positive organisms includes large amounts of phosphatidylglycerol and cardiolipin [219][239]. In Bacillus species, phosphatidylethanolamine is also abundant. In addition to the lipid component, the cell membrane consists of the lipid anchor component of lipoteichoic acid as well as numerous transmembrane and lipoproteins. Along with the peptidoglycan layer, these membranes play a role in maintaining cell shape [271] and also provide a surface for cellular respiration, the transport of various ions, toxins, and other solutes in and out of the cell, and cell-cell communication [133]. Molecules that target the membrane generally act through two distinct yet related mechanisms, permeabilization and depolarization. Permeabilization, as induced by compounds such as nisin and daptomycin, is the result of pores or other detrimental disruptions to the structure of the membrane. Depolarization, which is caused by ionophores that move ions against the concentrations gradients im- posed by the membrane, targets the proton motive force (PMF). The PMF is composed of the transmembrane electric potential and proton gradient that drive ATP synthesis Chapter 2. B. subtilis membrane activity profiling 31 and the specific transport of various solutes across the cell membrane [184].

While some are specifically interested in identifying membrane active molecules as potential antibiotics, there are others who are interested in removing membrane active compounds from their hit follow-up. This could be due to concerns over off-target effects, if the intended target is not the membrane, or due to the toxicity challenges of membrane active molecules [197] [240]. To this end, a number of techniques have been created to test compounds for membrane activity. These methods generally use fluorescent dyes to determine the effect of a molecule of interest on membrane permeability or membrane potential. However, testing only for permeability disruption would miss compounds that depolarize without impacting the membrane permeability, and testing for depolarization would be unable to distinguish those that only depolarize from those that both depolarize and impact permeability. A combination of two dyes, TO-PRO-3 iodide to measure cell permeability and DiOC2(3) to measure cell polarity, has been used previously to investigate the mechanism of daptomycin [243], but not in a high-throughput assay capable of screening large numbers of molecules.

Therefore, I set out to establish and utilize a simple and high-throughput method to quickly characterize the activity of novel bioactive compounds against the bacterial membrane of B. subtilis, a model Gram-positive bacterium. The assay utilizes two existing dyes previously used for measuring membrane disruption, TO-PRO-3 iodide and

DiOC2(3), and combines them in a single high-throughput assay. The specificity of the assay is demonstrated using three well-characterized molecules: nisin, a lantibiotic that binds lipid II and creates pores in the cell membrane, CCCP, a proton ionophore, and vancomycin, an antibiotic that inhibits cell wall synthesis while leaving the membrane intact. I also use this assay to screen a collection of 3,520 biologically active small molecules to determine the prevalence of membrane activity in a collection of synthetic molecules known to inhibit growth of B. subtilis. In performing this screen, I found 448 molecules, or 12.7% of the library, that interfere with the fluorescence of the dyes in the absence of B. subtilis; these were removed from further analysis. I found 7.1% of compounds induced a membrane effect in B. subtilis at 20 µM while 5.2% induced an effect at 5 µM. This suggests that membrane activity may not be as prevalent in synthetic compound libraries, and highlights that care must be taken to ensure no interference between the molecules and the method used to determine membrane activity. Chapter 2. B. subtilis membrane activity profiling 32

2.3 Results and Discussion

2.3.1 High-throughput assay for determining impact on membrane potential and permeability

Fluorescent dyes have been used to determine the impact of physical and chemical stresses on cell permeability and the membrane potential for over 30 years. These include ﬂu- orescein and ethidium bromide [5], DiSC3(5) [90][285], DiSC2(5) [82], DiOC2(3) [104],

SYTO9 and propidium iodide [214], TO-PRO-3 iodide [286], and DiOC2(3) [85]. In order to measure the impact on both membrane potential and membrane permeability simultaneously and in a high-throughput manner, I needed to identify dyes having non-over lapping excitation and emission spectra so that the signal from one would be independent of the other.

By measuring the absorption and ﬂuorescence emission spectra of four dyes, DiOC2(3),

DiSC2(5), TO-PRO-3 iodide, and propidium iodide, I selected TO-PRO-3 iodide to measure membrane permeability and DiOC2(3) to measure changes to the membrane potential. The chemical structures of these molecules are shown in Figure 2.1A and the optical separation of their absorbance spectra shown in Figure 2.1B. DiOC2(3) absorbs strongly at 450 – 500 nm and is strongly fluorescent at 500 – 550 nm, as shown in Figure 2.1C. TO-PRO-3 iodide absorbs strongly at 575 – 650 nm and is fluorescent around 650 – 700 nm, as seen in Figure 2.1D. In order to achieve good optical separation on our plate reader, I used an λex of 600 nm and λem of 650 nm for TO-PRO-3 iodide and λex of 450 nm and λem of 510 nm for DiOC2(3). TO-PRO-3 iodide is a cell impermeant carbocyanine monomer nucleic acid stain that is frequently used in fluorescence and laser confocal microscopy [253]. In the presence of an intact membrane the dye is unable to enter the cytoplasm, however when the cell membrane is damaged the dye is able to cross into the cell and bind DNA, increasing its

fluorescence emission 20- to 30- fold. DiOC2(3) is a cell permeant dye that aggregates in response to a bacterial membrane potential, quenching its fluorescence output. Disrup- tion of the membrane potential reduces this aggregation and significantly increases the fluorescence emission intensity [193]. I used a number of antibiotic molecules with well-characterized mechanisms of action to validate the use of these dyes. Nisin is a lantibiotic known to bind the lipid II component of the peptidoglycan synthesis machinery. It then inserts itself into the bacterial membrane, creating a pore that disrupts the membrane permeability [283]. This activity not only increases the permeability of the cell membrane but also disrupts the membrane Chapter 2. B. subtilis membrane activity profiling 33

Figure 2.1: Chemical and optical properties of TO-PRO-3 iodide and DiOC2(3) A. Chemical structures of TO-PRO-3 iodide and DiOC2(3). B. Absorbance spectra of TO-PRO-3 iodide () and DiOC2(3) (•) showing the distinct absorbance spectra of the two dyes. C. Absorbance spectrum of DiOC2(3) () with corresponding emission spectrum at λex = 450nm (•). D. Absorbance spectrum of TO-PRO-3 iodide () with corresponding emission spectrum at λex = 600nm (•). Chapter 2. B. subtilis membrane activity profiling 34

potential. This is because the membrane potential relies on the physical separation of an ion gradient, and disruption to the permeability dissipates this gradient. Treating B. subtilis with increasing concentrations of nisin resulted in an increase in the ﬂuores-

cence emission of both DiOC2(3) and TO-PRO-3 iodide relative to an untreated control, shown in Figure 2.2A. This change in signal took place beginning below 0.1 µg/mL and continued until the minimal inhibitory concentration (MIC) of 4 µg/mL. This indicates that both the membrane permeability and membrane potential have been compromised. Treatment with CCCP, a proton ionophore that disrupts the membrane potential by

shuttling protons across the bacterial membrane, increased DiOC2(3) fluorescence until it plateaued at 2 µM, near the MIC of 8 µM, as seen in Figure 2.2B. CCCP had no effect on TO-PRO-3 iodide fluorescence. This demonstrated that while the membrane potential was disrupted, there was no impact on the membrane permeability. Finally, vancomycin, which binds the pentapeptide component of peptidoglycan and inhibits cell wall synthesis [20], does not disrupt the cell membrane and thus showed no change in the fluorescence output of either dye. This effect can be seen in Figure 2.2C where the fluorescence output was flat even significantly above the concentration required to inhibit

growth (0.5 µg/mL). Therefore, the use of DiOC2(3) and TO-PRO-3 iodide allows me to determine the impact of a molecule on the bacterial cell membrane and distinguish between those that disrupt the membrane potential and the membrane permeability.

2.3.2 Screen for biologically active small molecules against B. subtilis

Collaborators in Prof. Eric Brown’s lab at McMaster University had previously identiﬁed 3,705 biologically active compounds able to inhibit growth of B. subtilis by screening a collection of 141,899 molecules [61]. In this screen, 10 µM of each compound in the library was added to cultures of B. subtilis 168 and incubated overnight. Growth inhibition

was determined by measuring the OD600 following overnight incubation and hits were selected using a statistical cutoﬀ of 2 standard deviations below the mean of the full data set. Therefore, molecules that inhibited growth by more than 30% were selected as hits, resulting in a hit rate of 2.61% and a total of 3,705 B. subtilis active molecules. I took advantage of this collection of prioritized bioactives to test the two-dye assay and investigate the frequency of membrane active molecules within the collection. Chapter 2. B. subtilis membrane activity profiling 35

Figure 2.2: Effect of known nisin, CCCP, and vancomycin on TO-PRO-3 iodide and DiOC2(3) fluorescence Concentration dependent impact of known antibiotics, nisin (A), CCCP (B) and vancomycin (C), on the relative fluorescence of TO-PRO-3 iodide () and DiOC2(3) (N) in B. subtilis cultures. Absorbance spectra of overnight cultures of treated B. subtilis was used to compare the fluorescence changes to the molecule’s MIC (•). Error bars represent the standard error of three replicate experiments. Chapter 2. B. subtilis membrane activity profiling 36

2.3.3 Membrane disruption by biologically active small molecules

One challenge of using fluorescence as an output for screening is the potential for the screening compounds themselves to directly interfere with the fluorescence emission independent of the effect on a cell. Such inference could include absorption at the excitation frequency, fluorescence quenching by absorbing the emission output, fluorescence of the test compound, or chemical reactivity between the dye and the molecule being tested. In order to remove any such molecules from our dataset, I first screened the sub-library for molecules that alter the dye fluorescence independent of membranes to identify com-

pounds that interfere with TO-PRO-3 and DiOC2(3) in the absence of B. subtilis. In order to account for plate-to-plate variability in fluorescence emissions, the readings were normalized by taking the log2 ratio of the emission of the treated dye solution relative to the untreated dyes. The results of this experiment are shown in Figure 2.3A and Figure 2.3B, with the black dots representing treatment at 20 µM and the grey dots representing treatment at 5 µM. To capture all of the interfering compounds, I used the fluorescence effects of the higher treatment concentration to filter out reactive or optically active compounds. Out of the 3,520 compounds tested, 448 had a significant impact on dye fluorescence at 20 µM in the absence of B. subtilis. I defined significant as values greater than or less than two standard deviations from the mean of the collection. This cut-off is shown in Figure 2.3A and Figure 2.3B as a solid horizontal line for 5 µM and the dashed line for 20 µM. Out of interfering molecules, 302 disrupted DiOC2(3) fluorescence while 196 disrupted TO-PRO-3. Therefore, 12.7% of the compounds in the bioactive collection were omitted from further analysis. To determine the membrane effects of the compounds, I screened the collection at two concentrations, 5 µM and 20 µM, against B. subtilis in the presence of both TO-PRO-3

iodide and DiOC2(3). The ﬂuorescence output was processed using the same protocol as

the compound controls, with the log2 ratio of the treated versus the untreated emission

intensities used to correct for plate to plate variability (Figure 2.3C for DiOC2(3) and Figure 2.3D for TO-PRO-3 iodide). I calculated an average Z’ value for the screening

assay as 0.54 for membrane depolarization through DiOC2(3) fluorescence using 10 µM CCCP as a positive control and 0.71 for membrane permeabilization through TO-PRO-3 iodide fluorescence using 10 µg/mL nisin as a positive control [291]. Using corrected fluorescence output values of greater than two standard deviations from the population mean at 20 µM as my cut-off, I identified 142 compounds as active against membrane potential (cut-off ratio of 0.64) and 151 as active against membrane permeability (cut-off ratio of 0.58). Using the same definition for the lower concentration of 5 µM, I observed 128 hits for membrane potential (cut-off ratio of 0.47) and 88 for membrane permeability Chapter 2. B. subtilis membrane activity profiling 37

(cut-oﬀ ratio of 0.28). This gives an overall rate of active molecules of 7.1% at 20 µM and 5.1% at 5 µM.

Figure 2.3: Screen of bioactive synthetic molecules for membrane activity Log2 ratio of the fluorescence readings from a screen of biologically active compounds at 5 µM (grey) and 20 µM (black) against DiOC2(3) control (A), TO-PRO-3 iodide control (B), membrane potential (C), and membrane permeability (D). The controls consist of the dye in buffer without any B. subtilis. Data are represented by a scatter plot (left) as well as a density distribution (right). Solid lines refer to the significance cut-offs for the 5 µM (solid) and 20 µM (dashed) concentrations.

I expected that compounds that permeabilized the cells would also depolarize the membrane on the grounds that loss of membrane integrity would allow ions to pass into the cytoplasm, compromising the electrochemical gradient. This effect is evident in the nisin treatment in Figure 2.2A where both the TO-PRO-3 iodide and DiOC2(3) fluorescence increase with increasing concentration of the molecule. To determine whether this was the case, I plotted the data so as to reveal such cross-activity (Figure 2.4A). In the 20 µM data: of 249 total hits, 98 acted exclusively on the electrochemical gradient, 44 compromised the electrochemical gradient and permeablized the cells while 107 appeared to permeablize the cells but had no effect on the electrochemical gradient. Similarly, in the 5 µM data: of 183 total hits, 95 compounds compromised the electrochemical gradient exclusively, 33 compromised the electrochemical gradient and permeablized cells while Chapter 2. B. subtilis membrane activity profiling 38

55 appeared to permeabilize cells but do not inﬂuence the electrochemical gradient. I reasoned that this apparent paradox of molecules disrupting membrane permeability without impacting potential may be due to the cut-oﬀ values used to select the relative activity. To investigate this further, I plotted the data as shown in Figure 2.4B and

Figure 2.4C showing scatter plots of the relative activity values of DiOC2(3) against TO- PRO-3 iodide at 20 µM and 5 µM treatment concentrations. The vertical line shows the cut-off value selected for membrane permeabilization while the horizontal line indicates the selected cut-off for activity against membrane potential. Although there are some compounds defined as membrane permeability hits that are just below the membrane potential cut-off, these are not the majority, suggesting that there are compounds that impact TO-PRO-3 iodide fluorescence while not affecting DiOC2(3) fluorescence. This may be caused by specific uptake of the TO-PRO-3 iodide dye or temporary permeabilization that allows the dye to enter the cell while not resulting in complete depolarization. Additionally, the compounds may be inducing secretion of extracellcular DNA, either by induction of biofilm formation[196] or natural competence systems [290] where the extracellular TO-PRO-3 iodide is able to bind and increase its fluorescence emission. As another means of assessing this data, I compared the effects at the two concentrations for each compound to determine whether the fluorescence gave a dose-response with each compound. By plotting the log2 ratio for the dye fluorescence against the test compound concentration, I found that most of the active compounds induced an increased fluorescence output at higher concentration, indicating a dose response, and that the effect of some molecules decreased at higher concentration. This effect is seen in Figure 2.5 with the blue solid lines showing compounds with increased effect at higher concentrations while the red dashed lines show decreasing effect. The compounds with reduced effect at higher concentrations were likely due to chemical instability or aggregation at higher concentrations, resulting in lower activity. This highlights the need to assess chemical effects at various concentrations, for compounds that appear to have no effect at high concentrations may be masking an effect visible at lower concentrations. Finally, I wished to assess the biological relevance of these induced fluorescence changes. Looking at the impact of nisin and CCCP on the dye fluorescence (Figures

2.1A and 2.1B), both increased the ﬂuorescence output by a log2 between 1 to 1.5, equivalent to a 2 to 3-fold increase in ﬂuorescence. For the 3,705 molecules in the bioactives collection, there are only 17 compounds that induced this magnitude of change in TO-

PRO-3 iodide and 37 that induce this magnitude of change in DiOC2(3). The chemical structures for the compounds are found in Figure 2.6, Figure 2.7, and Figure 2.8. I calculated the logP octanol/water partition coeﬃcient [275] for each of the molecules in Chapter 2. B. subtilis membrane activity profiling 39

Figure 2.4: Membrane activity screen summary A. Results summary showing the initial screen for inhibitors of B. subtilis growth, the compounds that interfere with the dye fluorescence output, as well as the hits that were determined from the screen. Of the 448 compounds removed from the screen due to fluorescence interference, 302 were for interference with DiOC2(3) and 196 for interfering with TO-PRO-3 iodide, with 50 interfering with both. B. Comparing the relative results DiOC2(3) fluorescence compared to TO-PRO-3 iodide fluorescence at 20 µM and (C) at 5 µM. The vertical line denotes the statistic cut-off used for determining activity. Chapter 2. B. subtilis membrane activity profiling 40

Figure 2.5: Concentration dependent impact of bioactive molecules on TO- PRO-3 iodide and DiOC2(3) fluorescence Concentration dependent effect on of fluorescence activity for TO-PRO-3 iodide (A) and DiOC2(3) (B). Lines are shown for compounds that show significant concentration effects, with blue indicating an increase in fluorescence emission with increasing concentration and red indicating a decrease in fluorescence emission.

order to compare these highly active compounds relative to the whole sub-library. In Figure 2.9, I show a density plot of the logP values for the whole sub-library, the membrane potential hits, and the membrane permeability hits. I observe a slight increase in the logP values for the membrane potential hits relative to the full set of molecules and an additional slight increase for the membrane permeability hits. Therefore, the hits show slightly higher lipid solubility than the library as a whole. These data indicate that molecules with signiﬁcant membrane eﬀects comparable with known membrane disrupting antibiotics are not common within this synthetic compound library.

2.4 Conclusion

I demonstrated the use of a two-dye assay for rapidly identifying and characterizing molecules that permeablize cells and/or compromise the electrochemical gradient in B. subtilis. This assay can be used to test antibacterial candidate compounds for membrane activity to remove them from further consideration or to speciﬁcally identify membrane active compounds, as desired. I found that in a collection of 3,520 synthetic bioactive compounds, membrane activity was a relatively rare molecular property. With 249 membrane active molecules in the whole collection, and only 54 inducing an eﬀect comparable Chapter 2. B. subtilis membrane activity profiling 41 with known membrane active molecules such as nisin, this suggests that synthetic compound libraries do not contain a plurality of membrane active molecules. Chapter 2. B. subtilis membrane activity profiling 42

N N O O- N+ O N H H S N N N O NH O O N F OH H Cl F OICR-CP24-C05* F OICR-CP1-H08 Br F OH OICR-CP36-C11

N N+ HN Cl N H N O N N N N N S O OICR-CP3-G11 OICR-CP25-F02* N

O HN N O OICR-CP37-F05 F N HO OH HN O O O

O O NH OICR-CP5-G04 OICR-CP27-B06 N F F S OH F

H2N O OICR-CP38-D03 H N OH N N O S N NH2 O OICR-CP29-F04 OICR-CP6-E11

O H H H N N N O O Cl HN N Cl O Cl O OICR-CP30-A06 OICR-CP10-E07 N F NH N H O N HN HN N

F N OICR-CP10-F10 OICR-CP31-G05

H N N N N O H N N N N S Cl O N O O F O OICR-CP14-F04 OICR-CP33-A08

Figure 2.6: Membrane permeability hits Chemical structures of the significant membrane permeability hits with the corresponding library identifiers. Those with * indicate compounds with both significant activity against membrane potential and membrane permeability. Chapter 2. B. subtilis membrane activity profiling 43

-O O OH O H H H N+ N N N N NH N N N S H O - - O O + +O + Cl N N N -O N O O O F F F

OICR-CP1-C03 OICR-CP7-B05 OICR-CP19-H09

N O H H O N N HN HN N NH N - S Br O ON+ Cl N N O

OICR-CP2-D10 OICR-CP7-E04 OICR-CP20-B09 O O N+O- O H H O N N F S N N H F N HN S - O NH F N+O O F F F N OICR-CP20-E08 OICR-CP3-E02 OICR-CP7-E11 Cl HO H H H N N O N N O N S - S O +O F F N -O N+ O O F F F OICR-CP20-E11 OICR-CP3-F06 OICR-CP16-F09

O H H O H O H H Cl + N N N N -ON N F N - O F S +O H S N F N Br O OICR-CP3-H10 OICR-CP19-B07 OICR-CP20-F09 Cl H H OH O H H Cl N N N N N N S S H S NH Cl + F -ONO F F OICR-CP19-C09 OICR-CP21-B05 OICR-CP6-D02

Cl H H H F F N N N OH Cl NH F S O O N F F F N Cl F O Cl

OICR-CP6-E02 OICR-CP19-F07 OICR-CP21-D02

Figure 2.7: Membrane potential hits Chemical structures of the significant membrane potential hits with the corresponding library identifiers. Those with * indicate compounds with both significant activity against membrane potential and membrane permeability. Chapter 2. B. subtilis membrane activity profiling 44

O + F N - H F F Br O - N OH O +O N H N O H F N N F F O OH F O F F F Cl OICR-CP21-E03 OICR-CP23-H02 OICR-CP40-F08 Cl F O O H H H N N N N H O Cl N N O N O S H + F O - N F Cl O O OICR-CP22-D02 F F OICR-CP24-C05* OICR-CP40-H11 Br N H N N N H S N N O S O S F F OICR-CP22-F06 F OICR-CP24-D06

Br H H HN OH H N N N N O O O ON+ - O F F F OICR-CP23-C02 OICR-CP25-B06

Br H H N N S O ON+ Cl N+ O- HN OICR-CP23-D02 O OICR-CP25-F02*

N NH - O + H Cl H O O N N Cl N N+ -O O O O F O F F Cl OICR-CP23-E02 OICR-CP25-F04 F N F O H N Br H F N O F S O OH O O OICR-CP25-G05

OICR-CP23-F05

Figure 2.8: Membrane potential hits, cont. Chemical structures of the significant membrane potential hits with the corresponding library identifiers. Those with * indicate compounds with both significant activity against membrane potential and membrane permeability. Chapter 2. B. subtilis membrane activity profiling 45

0.6

0.4 density

0.2

0.0 −2 0 2 4 6 logP

Figure 2.9: Calculated logP values for bioactive compound library Density plot of the calculated logP values for the whole sub-library (black), the highly active membrane potential hits (red) and the highly active membrane permeability hits (blue). Chapter 3

A chemical inhibitor of cell growth reduces cell size in Bacillus subtilis

The data reported in this chapter have been published as part of the following peer- reviewed publication: Scott McAuley, Stephen Vadia, Charul Jani, Alan Huynh, Zhizhou Yang, Petra Anne Levin, Justin R. Nodwell (2019) A chemical inhibitor of cell growth reduces cell size in Bacillus subtilis, ACS Chem. Bio., In Press, doi: 10.1021/acschembio.8b01066 Charul Jani initially observed Min-1’s short cell phenotype in B. subtilis. Additional microscopy, cell length measurements, and FtsZ-ring counting was performed by Stephen Vadia and Zhizhou Yang in the lab of Petra Anne Levin. Testing the bioactivity of Min-1 analogs was performed by Alan Huynh. I performed all other experiments and analysis, including bioactivity of the parent molecule, activation of B. subtilis reporter strains, and membrane activity.

3.1 Abstract

Bacteria exhibit complex responses to biologically active small molecules. These responses include reductions in transcriptional and translational efficiency, alterations in metabolic flux, and in some cases, dramatic changes in growth and morphology. Here, I describe Min-1, a novel small molecule that inhibits growth of Gram-positive bacteria by targeting the cell envelope. Sub-inhibitory levels of Min-1 inhibits sporulation in Streptomyces venezuelae and reduces growth rate and cell length in Bacillus subtilis. The effect of Min-1 on B. subtilis cell length is significant at high growth rates sustained by nutrient rich media but drops off when growth rate is reduced during growth on less energy rich carbon sources. In each medium, Min-1 has no impact on the proportion

46 Chapter 3. Min-1 47 of cells containing FtsZ-rings, suggesting that Min-1 reduces the mass at which FtsZ assembly is initiated. The eﬀect of Min-1 on size is independent of UDP-glucose, which couples cell division to carbon availability, and the alarmone ppGpp, which reduces cell size via its impact on fatty acid synthesis. Min-1 activates the LiaRS stress response, which is sensitive to disruptions in the lipid II cycle and the cell membrane, and also compromises cell membrane integrity. Therefore, this novel synthetic molecule inhibits growth at high concentrations and induces a short-cell phenotype at sub-inhibitory concentrations that is independent of known systems that inﬂuence cell length, highlighting the complex interactions between small molecules and cell morphology.

3.2 Introduction

The impact of antibiotics on cell morphology has been an area of investigation since 1952 [216]. Since then, we have learned that treating different bacteria with antibiotics of varying targets and concentrations can induce different cell morphologies. These can be used as a tool to understand the underlying molecular mechanisms that govern cell growth and development as well as aid in identifying the targets of novel antibiotics. Small molecule perturbation of cell morphology can provide a means of understanding the underlying molecular mechanisms that govern cell growth and development and aid in identifying the targets of novel antibiotics. This cytological profiling approach of characterizing the effects of antibiotics on cell morphology and subcellular organization has been harnessed to identify the target pathways for both known antibiotics and novel natural products [158][192]. Specifically, the nature of morphological changes induced by antimicrobial exposure are often suggestive of the compound’s general macromolecular target. Activation of the SOS response by DNA damaging agents such as mitomycin C, bleomycin, and ciprofloxacin result in cell filamentation through the direct binding of FtsZ by SulA in E. coli [265] or YneA in B. subtilis [145]. Inhibition of fatty acid synthesis by cerulenin slows the rate of cell envelope synthesis and results in reduced cell size in E. coli, B. subtilis, and Saccharomyces cerevisiae [271]. The small molecule A22 binds the ATP-binding pocket of MreB, an actin-like protein that serves as a treadmilling platform for peptidoglycan synthesis, resulting in significant changes in cell width in rod-shaped bacteria [23][264]. Previous members of our lab have identified compounds, including Fil-1, Fil-2 and Fil-3, that inhibit division in several Gram-positive bacteria, resulting in cell filamentation, and strongly impact spore development in the actinomycete Streptomyces coelicolor [58][139]. While some of these connections between antibiotic target and induced cell morphology are well-characterized, our understanding of how Chapter 3. Min-1 48 cells maintain their size and how it is influenced by external factors such as antibiotics is still an ongoing line of research. Here I report another Streptomyces sporulation inhibitor and novel molecule, Min-1, which inhibits the growth of Gram-positive bacteria including B. subtilis and S. aureus, by targeting the cell envelope. Min-1 mediated growth inhibition is concentration- dependent and accompanied by a distinct short cell phenotype in B. subtilis at sub- inhibitory concentrations. Analysis of Min-1 treated cells suggests that the short cell phenotype is a result of disruption of the cell envelope and is independent of UgtP and (p)ppGpp. The effect of Min-1 on cell size is significant in nutrient rich media supporting rapid growth, but its effect on cell size is modest in defined minimal media. Combining the short cell phenotype with the observation that Min-1 treatment does not alter the proportion of cells containing FtsZ-rings suggests that the compound reduces the mass at which FtsZ assembly is initiated. Min-1 activates a cell stress response associated with disruption of the lipid II cycle, has no impact on the cell-wall damage sensing σW extracytoplasmic function sigma factor, increases membrane permeability, and disrupts membrane potential. These effects are related to those demonstrated for the antibacterial peptide nisin, however, unlike nisin, supplementing the media with magnesium does not block the effects of Min-1. These findings reinforce the importance of the cell envelope in determining cell size and describe the bioactivity of a novel molecule discovered by screening against Streptomyces sporulation.

3.3 Results and Discussion

3.3.1 Min-1 inhibits Streptomyces sporulation at sub-inhibitory concentrations

Min-1 is a synthetic molecule (Figure 3.1A) identified in a screen of 30,569 small molecules for compounds that blocked sporulation of S. coelicolor [58]. It was of interest because, while it blocked spore formation it appeared to have relatively little effect on formation of vegetative substrate hyphae or sporogenic aerial hyphae. To confirm the original screening result I tested Min-1 against the distantly related species Streptomyces venezuelae. Normally, S. venezuelae colonies go through a cycle of growth in which they first produce a beige mycelium of filamentous substrate hyphae. Later, they produce a white layer of sporogenic aerial hyphae, which grow upwards, conferring a white, fuzzy appearance to the colony surface. When the aerial hyphae sporulate, they turn green due to the activation of the whiE spore pigment genes. Chapter 3. Min-1 49

I spotted 1 µL of 25 mM Min-1 on a lawn of S. venezuelae spores and, following growth, observed a zone of inhibition surrounded by a white halo. Further from the source of Min-1, the lawn exhibited the characteristic green pigmentation of S. venezuelae (Figure 3.1B). The white halo indicates a region where cells failed to express the whiE genes, a classic indication of a sporulation block in streptomycetes [49]. By contrast, compounds like vancomycin, which inhibit growth but not sporulation, conferred a typical zone of inhibition with no white halo. To confirm this phenotype, I used scanning electron microscopy (SEM) to image the surface of the bacterial lawn. Cells sampled from the green zone grew in chains of septated hyphae, indicating normal division into spores (Figure 3.1D). By contrast, cells in the white halo region lacked septated hyphae (Figure 3.1C). This confirms that Min-1 blocks sporulation in S. venezuelae indicating that this is a general effect of the drug against this genus. Min-1 had no other observable effects on the filamentous Streptomyces cells.

Figure 3.1: Eﬀect of Min-1 on S. venezuelae development A. Chemical structure of Min-1. B. Spot diﬀusion assay of S. venezuelae lawns treated with vancomycin and Min-1. C. SEM image of the S. venezuelae lawn in the Min-1 sub-MIC region D. SEM image of untreated S. venezuelae lawn with arrows indicating septation of the hyphae.

3.3.2 Min-1 inhibits growth of Gram-positive bacteria and reduces cell length of B. subtilis

Streptomyces cells employ molecular mechanisms for cell growth and division that are similar to those of rod shaped and coccoid bacteria though they deploy them diﬀerently. I therefore predicted that Min-1 might have antibiotic activity against other bacteria. Indeed, when I treated B. subtilis and S. aureus with Min-1 I found that it inhibited growth at a minimum inhibitory concentration (MIC) of 31 µM when cultured in LB medium (Table 3.1). The MIC of Min-1 against B. subtilis declined with the medium-

supported growth rate, decreasing 2-fold in both S750-glucose and S750-glycerol indicating Chapter 3. Min-1 50

Strain Min-1 MIC (µM) B. subtilis JH642 (LB) 31 B. subtilis JH642 (S750-glucose) 16 B. subtilis JH642 (S750-glycerol) 16 S. aureus ATCC 29213 30 S. aureus ATCC BAA-44 22 S. aureus ATCC BAA-41 33 E. coli K-12 >250 E. coli MC1061 61

Table 3.1: Minimum inhibitory concentration of Min-1

greater antimicrobial potency under these slower growth conditions. The compound had no inhibitory effect against E. coli at concentrations up to 100 µM though it was able to inhibit the growth of the hyper-permeable strain MC1061 [46] at 61 µM. This suggests that the lipopolysaccharide (lps) layer of the Gram-negative outer membrane blocks access of Min-1 to potential targets. Dr. Stephen Vadia, a post-doctoral fellow in the lab of Prof. Petra Levin explored the phenotypic effects of Min-1 on B. subtilis cultured in several different media (Figure 3.2A) supporting a range of doubling times. In LB-glucose, B. subtilis grew with a doubling time of 21 min and reached an average length of 5.59 µm. Addition of 3 µM Min-1 reduced cell size to 4.78 µm but had no effect on growth rate. Higher concentrations of the drug further decreased size and growth rate, with 12 µM Min-1 reducing the average cell size by 40% to 3.42 µm and increasing the doubling time to 29 min. While Min-1 exposure reduced cell length, Dr. Vadia did not observe any other significant morphological changes upon Min-1 treatment (Figure 3.2B).

Despite having greater antibiotic potency in deﬁned S750 media (Table 3.1), Min-1’s eﬀect on cell length was greatly reduced when B. subtilis was cultured in minimal media.

Untreated cells cultured in S750-glucose [137] grew slower and were significantly smaller than cells cultured in LB-glucose, reaching an average length of 3.17 µm. Treatment with 12 µM Min-1 reduced cell length to 2.94 µm. Similarly, untreated cells cultured in S750-glycerol reached an average length of 2.8 µm, which was reduced to 2.51 µm by treatment with 12 µM Min-1. Therefore, the effect of Min-1 on cell length was influenced by growth conditions: it was greatest when cells were growing rapidly in rich, undefined medium and did not have a significant effect in defined media supporting slower growth rates. While a reduction in cell size is accompanied by slower growth rate under nutrient- Chapter 3. Min-1 51 limiting conditions, as we observed in the untreated cultures, the two factors are not strictly linked. Alterations in temperature [228], transcription inhibition [271], and thymine limitation [236] can result in changes to either growth rate or cell size without impacting the other, demonstrating that the cell cycle can be decoupled from growth. Also, while various molecules and mechanisms induce cell filamentation (ciprofloxacin, ampicillin) or rounding (A22, mecillinam), small molecule induced cell shortening is a less frequently observed and understood phenotype [191]. Disruption of some cellular mechanisms, specifically UgtP [279] and fatty acid synthesis [279], have resulted in short cells and cytological profiling studies have shown that small molecule membrane disruption at high concentrations can lead to short cells [191]. I was therefore interested in further understanding the bioactivity of this molecule and how it disrupts cell growth and the cell cycle.

3.3.3 Structural analogs of Min-1 have altered eﬀects on growth rate and cell length

To identify chemical features responsible for growth inhibition and the short cell phenotype I compared the activities of Min-1 with 13 structural analogues. The results indicate that minor changes in structure have profound eﬀects on the molecule’s activity (Figure 3.3). For example, when compared with Min-1, the analogues Min-1-5, Min-1-8, Min-1-9, Min-1-10, Min-1-11, Min-1-12 and Min-1-14 had greatly reduced potency (MIC greater than 100 µM) or in some cases no ability to block growth (MIC greater than 250 µM). Min-1-3, Min-1-13, Min-1-15 were active but exhibited higher MICs than the parent molecule, indicating lower potency. The potency of Min-1-6 and Min-1-9 were similar to Min-1’s in this assay and Min-1-7 exhibited improved potency with an MIC of 10 µM. I note that a dynamic light scattering experiment indicated that, unlike the rest of the compounds, M1-3, M1-4, M1-7 and M1-15 were prone to aggregation.

Changes to Min-1 structure also altered the eﬀects of the molecules on cell length in B. subtilis (Figure 3.4). Testing the analogs against B. subtilis at 6 µM I observed that the more active molecules also shortened cells to a greater degree (Figure 3.4A). Indeed, I observed a striking correlation between the reduction in growth rate and the short cell phenotype (Figure 3.4B). This suggests that the eﬀect of Min-1 on cell size is directly coupled to its antimicrobial potency when cultured in nutrient rich media. Chapter 3. Min-1 52

Figure 3.2: Cell length and growth rate eﬀects of Min-1 on B. subtilis A. Impact of increasing concentrations of Min-1 on B. subtilis cell length cultured in LB-glucose, S750-glucose, and S750-glycerol. Larger and darker points represent higher Min-1 concentrations, from 0 µM to 12 µM. Error bars represent the standard error of the mean from three biological replicates with at least 100 cell lengths measured per replicate. B. Microscopy images of untreated (left) and Min-1 treated at 6 µM (right) B. subtilis cultured in LB-glucose and cell wall labelled with agglutinin-tetramethylrhodamine. Chapter 3. Min-1 53

Figure 3.3: Structure and activity of Min-1 analogs against B. subtilis and S. aureus Chapter 3. Min-1 54

Figure 3.4: Eﬀect of Min-1 analogs on B. subtilis growth and cell length A. Min-1 analogs demonstrate a link between cell length phenotype and inhibitory activity when cultured in LB-glucose, with the more active analogs inducing a shorter cell phenotype. B. The Min-1 analogs that reduce growth rate, as shown by cell doublings per hour, also have a greater impact on cell length. Error bars for each panel represent the standard error of the mean from three biological replicates with at least 100 cell lengths measured per replicate. Chapter 3. Min-1 55

3.3.4 Min-1 disrupts coordination between growth and FtsZ ring assembly

Polymerization of the tubulin homologue FtsZ into a ring-like structure establishes the location of the bacterial division site. The FtsZ ring serves as a framework for assembly of the division machinery and is thus an excellent marker for cell cycle progression. The proportion of cells with FtsZ rings localized at midcell in a population reflects the relative time at which rings first form following the previous cell division. This frequency remains constant if the time at which rings first form relative to the time from cell birth to division is unchanged. In B. subtilis, the percentage of cells with FtsZ rings is high ( 90%) in long, fast growing bacteria cultured in nutrient rich medium, consistent with their rapid mass doubling time. In contrast, the percentage of cells with FtsZ rings is lower in short, slow growing bacteria cultured in nutrient poor medium, consistent with their longer cell cycle time [280]. To assess the influence of Min-1 on the cell cycle, Dr. Vadia first determined the effect of Min-1 on the fraction of cells with FtsZ rings during growth in LB-glucose, a nutrient rich medium. Consistent with previous findings [280], 95.4% of cells grown in LB-glucose contained a visible FtsZ-ring. This number was reduced to 79.1% and 63.7% in S750-glucose and S750-glycerol respectively, demonstrating that cells cultured in media supporting slower growth rates display a delay in FtsZ ring formation. Based on the reduction in growth rate and cell size induced by Min-1, we expected Min-1 to also reduce the proportion of cells containing FtsZ rings relative to the untreated population. Strikingly, cells cultured with up to 12 µM Min-1 exhibited no significant change in the proportion of cells with FtsZ-rings in all three media (Figure 3.5). In B. subtilis cultures treated with 12 µM Min-1 Dr. Vadia observed an average of 91.8% of cells cultured in LB-glucose, 83.9% in S750-glucose, and 71.7% in S750-glycerol contained FtsZ-rings. Changes in the length-to-ring ratio of a population of cells can help to identify instances in which a treatment disrupts coordination between growth and the cell cycle. Quantifying the total length of cells in the population and the number of FtsZ-rings present when cultured in LB-glucose revealed an average cell length-to-ring ratio of 7.3 µm/FtsZ ring (Figure 3.5B). Addition of up to 12 µM Min-1 reduced this ratio to 4.7. The degree to which Min-1 affected the length-to-ring ratio varied with the growth medium, decreasing from 4.8 in untreated culture to 4.2 with 12 µM Min-1 in cells cultured in

S750-glucose, and from 5.5 to 4.4 in cells cultured in S750-glycerol. Together these data suggest that Min-1 reduces the mass at which FtsZ assembly is initiated, disrupting Chapter 3. Min-1 56

the coordination between growth and FtsZ-ring assembly, and that this eﬀect is most signiﬁcant when cells are grown in nutrient rich media.

To evaluate possible mechanisms through which Min-1 could impact cell length, I ﬁrst investigated the UDP-glucose dependent division-inhibitor and glycosyltransferase UgtP [279]. UgtP coordinates carbon availability with cell size in B. subtilis via changes in FtsZ assembly dynamics. During growth in carbon rich media, high concentrations of its substrate UDP-glucose stimulates interaction between UgtP and FtsZ, delaying assembly of the division machinery and linking nutrient availability to cell length. Because UgtP impacts the cell cycle, defects in the three genes in the UDP-glucose biosynthesis pathway, ugtP, gtaB and pgcA reduce cell length under nutrient rich conditions but do not substantially impact growth rate.

To evaluate whether Min-1 acts through UgtP to induce the short cell phenotype, Dr. Vadia compared the effect of Min-1 on wild type cells with its effects on strains bearing ugtP::cat, gtaB::spc, and pgcA::Tn10 mutations. As expected, the lengths of the untreated mutants were 20% shorter than wild-type (Figure 3.5C). Application of Min-1 to each mutant strain cultured in LB-glucose reduced cell length by an additional 10–15%, an effect equivalent to that on wild type cells. Furthermore, the MIC of Min-1 against the mutants was 31 µM, identical to that in the wild type. This demonstrates that Min-1 action is independent of the UgtP pathway.

(p)ppGpp is a pleiotropic inhibitor of macromolecular synthesis in bacteria that reduces cell size, largely due to its impact on lipid synthesis [215][288]. To determine whether the cell length effect of Min-1 is mediated through (p)ppGpp, Dr. Vadia tested Min-1 against a (p)ppGpp-deficient strain of B. subtilis (JH642 ∆yjbM ∆ywaC::kan ∆relA::mls) [153]. If the cell shortening and growth inhibition phenotypes were mediated through Min-1 induction of (p)ppGpp we would expect the (p)ppGpp-deficient strain to be resistant to the Min-1 induced shortening and growth inhibition. Consis- tent with Min-1 impacting growth independently of (p)ppGpp, the MIC of Min-1 for the (p)ppGpp-deficient strain is the same as wild-type B. subtilis (31 µM). Dr. Vadia also found that treatment with 6 µM of Min-1 reduced the length of (p)ppGpp-deficient B. subtilis by more than 20%, the same magnitude as observed following treatment of the wild-type strain (Figure 3.5D). This demonstrates that Min-1 acts independently of (p)ppGpp. Chapter 3. Min-1 57

Figure 3.5: Min-1 disrupts coordination between growth and FtsZ ring assembly independent of known cell size mechanisms A. Proportion of treated B. subtilis cells with FtsZ rings is reduced with decreasing nutrient availability, but is not impacted over changing growth rates induced by increasing Min-1 concentrations. Min-1 treatment concentrations range from 0 µM (light) to a maximum of 12 µM (dark). B. The cell length to Z-ring ratio decreases as the growth rate slows at higher Min-1 concentrations. C Cell length of knock-out mutants in the B. subtilis UgtP pathway, including ugtP, gtaB, and pgcA treated with 6 µM Min-1. D. Impact of 6 µM Min-1 on the cell length of a (p)ppGpp null B. subtilis strain. Error bars for each panel represent the standard error of the mean from three biological replicates with at least 100 cells measured per replicate. Chapter 3. Min-1 58

3.3.5 Min-1 targets the cell envelope

The cell envelope plays a vital role in determining and maintaining cell shape. To assess the possible impact of Min-1 on cell wall synthesis, I used two B. subtilis reporter

strains, PliaI-lacZ [143] and PσW-lacZ [129]. These reporters are activated upon inhibition of diﬀerent components of the cell wall synthesis machinery. PliaI is the promoter of the LiaRS two-component system that is strongly activated by direct interference with the lipid II cycle or by daptomycin, and, to a lesser extent, certain organic solvents and alkaline shock [172][218]. Lipid II transports precursors for the biosynthesis of peptidoglycan across the bacterial membrane and this pathway is the target of many diﬀerent antibiotics, including bacitracin [238], nisin [283], teixobactin [163], and malacidin A [126]. σW is an extracytoplasmic function (ECF) sigma factor that activates a cell wall stress response to antibiotics that interfere with the assembly of peptidoglycan, including vancomycin and the β-lactams, but less so for those that directly interfere with the lipid II cycle [45][118].

When I applied Min-1 to the PliaI-lacZ reporter strain I observed a clear blue halo around the zone of inhibition (Figure 3.6A), indicating activation of the reporter. By contrast, when I applied Min-1 to a PσW-lacZ B. subtilis reporter strain I did not observe any reporter activation (Figure 3.6A). This suggests that Min-1 targets the cell envelope at the membrane, either through the lipid II cycle or by directly disrupting membrane integrity. Interestingly, this activity profile resembles that of nisin, an antimicrobial peptide that elicits a positive response from PliaI -lacZ [129][143] but not PσW-lacZ [45]. It has also been reported that sub-MIC concentrations of nisin reduce cell length in B. subtilis [135]. Nisin action involves direct binding of lipid II with the concomitant inhibition of peptidoglycan synthesis; moreover, nisin also forms depolarizing pores across the cell membrane. This membrane activity is considered to be its principle mechanism of antimicrobial action [283]. To determine whether Min-1 has an effect on membrane potential and permeability, I employed the DNA-binding fluorescent dyes, TO-PRO-3 iodide and DiOC2(3). TO-PRO- 3 iodide strongly fluoresces when bound to DNA, however, it is unable to cross intact bacterial membranes and can therefore be used to monitor membrane integrity [286][176].

DiOC2(3) is a cell permeant dye that, in response to a bacterial membrane potential, quenches its fluorescence output. Disruption of the membrane potential reduces this quenching and significantly increases the fluorescence emission intensity. Consistent with a disruption of both the cell membrane permeability and depolarization, Min-1 cells exhibited a significant increase in TO-PRO-3 and DiOC2(3) fluorescence beginning at the Chapter 3. Min-1 59

sub-inhibitory concentration of 10 µM and increasing through the MIC (Figure 3.6B). By contrast, ﬂuorescence output was unchanged in untreated controls.

Figure 3.6: Min-1 targets the cell envelope A. Impact of Min-1 and control antibiotics on B. subtilis reporter strains for cell wall W assembly (PliaI-lacZ and Pσ -lacZ ). B. TO-PRO-3 iodide ﬂuorescence (• - left axis), DiOC2(3) ﬂuorescence (green - N), and B. subtilis optical density at 600 nm ( - right axis) with increasing Min-1 treatment concentration. Error bars represent the standard error of the mean from three replicates.

Finally, to determine whether protecting the membrane against depolarization compromises the antibacterial effect of Min-1 I performed a checkerboard assay in which varying concentrations of Min-1 were incubated with cells in the presence of varying concentrations of supplemental Mg2+. I observed no effect of added Mg2+ on the MIC of Min-1, indicating that the action of the compound is more complex than simple disruption of the membrane permeability. By contrast, the addition of supplemental Mg2+ had a strong protective effect against nisin: the MIC was doubled in the presence of 5 mM Mg2+ and increased more than 8-fold by 40 mM Mg2+ (Figure 3.7B). In summary, these data suggest that Min-1 compromises the lipid II cycle, thereby activating LiaRS. This could involve a direct interaction with lipid II or one of the enzymes that binds to it. Alternatively, Min-1 might interact with the membrane, compromising lipid II indirectly. Chapter 3. Min-1 60

Figure 3.7: Impact of supplemental magnesium on Min-1 activity Checkerboards of A. Min-1 and supplemental magnesium and B. nisin and supplemental magnesium against B. subtilis grown in LB.

3.3.6 Min-1 is a novel inhibitor of bacterial growth

I have demonstrated that Min-1, discovered initially for its capacity to inhibit sporulation in streptomycetes, also has antibacterial activity against B. subtilis and S. aureus, conferring a distinctive short cell phenotype at sub-MIC concentrations on the former. The reduction in B. subtilis cell size is accompanied by an increase in mass doubling time but does not change the proportion of cells exhibiting FtsZ-rings. Therefore, Min- 1’s effect does not involve compromising cell cycle progression while reducing cell length. Instead, I propose that the reduced cell size at sub-MIC concentrations is the result of FtsZ-rings forming at reduced cell mass relative to an untreated population. This effect is independent of two known cell size regulators that result in short cell phenotypes: the UgtP pathway and (p)ppGpp. Rather, the mechanism by which this occurs involves depolarization of the membrane and a defect in the lipid II cycle, resulting in the activation of the LiaRS stress response. The morphological impact of numerous antibiotics has been described previously and, to our knowledge, the spectrum of effects exhibited by Min-1 is not common. For example, of 43 known antibiotics that have undergone cytological profiling, 11 resulted in a reduction in cell length [191]. These include ionophores or other molecules that desta- bilize the bacterial membrane, such as nisin, CCCP, nigericin, and spirohexenolide A, or target fatty acid synthesis, such as cerulenin or triclosan. Therefore, either direct or indirect disruption of the membrane inhibits membrane expansion and results in reduc- Chapter 3. Min-1 61 tions in cell length. This could be caused by reduction in nutrient uptake or disruption of ATP synthesis via membrane depolarization. However, Min-1 also induces the LiaRS stress response, which is also activated by the lipid II targeting nisin, bacitracin, and vancomycin, and the membrane disrupting daptomycin, but not the membrane disrupting molecules polymyxin B or friulimicin B [281]. This level of specificity suggests that the action of Min-1 is more complex than simple membrane disruption. Chapter 4

Discovery of a novel DNA gyrase-targeting antibiotic through the chemical perturbation of Streptomyces venezuelae sporulation

The data reported in this chapter is under review as part of the following peer-reviewed publication: Scott McAuley, Alan Huynh, Alison Howells, Tony Maxwell, Justin R. Nodwell (2019) Discovery of a novel DNA gyrase-targeting antibiotic through the chemical perturbation of Streptomyces venezuelae sporulation, Cell Chem. Bio, Under Review. I performed all experiments and analysis in this chapter with the exception of testing EN-7 for growth inhibition of lab strains, which was performed by Alan Huynh, and the topoisomerase in vitro assays, which were performed by Alison Howells in the lab of Tony Maxwell.

4.1 Abstract

Researchers have a wide variety of techniques for identifying new antibiotics. Screening for growth inhibition and numerous whole-cell reporter assays have all uncovered interesting chemistry in the attempt to overcome growing rates of antibiotic resistance. Here, we utilize Streptomyces venezuelae sporulation as an unconventional screening platform

62 Chapter 4. EN-7 63 to uncover a novel gyrase inhibitor, EN-7, capable of overcoming ciprofloxacin resistance and inhibiting growth of extensively resistant Gram-positive pathogens. Using a panel of known antibiotics, we identify that S. venezuelae sporulation is exquisitely sensitive to a wide array of DNA damaging agents. Screening 3,705 bioactive synthetic molecules against S. venezuelae sporulation identified 10 molecules, including EN-7. Utilizing both genetic and in vitro methods, we demonstrate that EN-7 directly and specifically targets DNA gyrase. This work highlights the utility of leveraging unconventional organisms as platforms for novel antibiotic screening programs and introduces the activity of EN-7, a novel gyrase inhibitor.

4.2 Introduction

It is widely acknowledged that there is a significant and unmet need for new antibacterial compounds due to the growing and global problem of antibiotic resistance (National Academies of Sciences and Medicine, Engineering, 2018). At present, significant discovery efforts in this area are focused on the chemical evolution of existing antibiotic classes to circumvent resistance [91]. Alternatively, searches for new compounds are routinely initiated with high throughout screens of vast compound libraries to identify chemical matter that specifically inhibits the growth of target pathogens [293] or directly inhibit a particular resistance mechanism [260]. Screens against specific molecular targets, reconstituted in vitro, have also served to provide lead compounds for antibiotic development [77]. All these approaches have merit; however, given the paucity of new antibacterial scaffolds in clinical trials it is clear that new thinking is also needed. For example, compounds that interfere with specific metabolic states [103], previously untargeted molecular processes [115], or bacterial morphology [23] also have significant potential for antibiotic discovery. To this end, we have focused our efforts on the bacterial genus Streptomyces. The streptomycetes are filamentous bacteria that exhibit a complex multicellular life cycle. This life cycle starts with spore germination, continues with the growth of vegetative “substrate hyphae” followed by the erection of a fuzzy layer of “aerial hyphae” on the colony surface. The aerial hyphae then undergo a concerted round of cell division that divides each filament into a chain of unigenomic compartments that then mature into spores [177]. By contrast, the substrate hyphae are sites of significant “secondary metabolism” that generates antibacterial compounds that, in nature, are believed to protect the developing colony from competing microorganisms [28]. Each stage of this process is accompanied by a visual phenotype: defects in spore maturation manifest as “white” mutants because the colonies are unable to synthesize the whiE-associated spore Chapter 4. EN-7 64 pigment that is believed to be the final step in this life cycle [65]. In previous research, our lab has screened compound libraries for molecules that target secondary metabolism in Streptomyces coelicolor [58][208]. They noted in this and subsequent work [6] that some of the compounds appeared to target other aspects of the streptomycetes lifecycle, including sporulation. Intriguingly, some of the compounds that target sporulation turned out to have antibacterial effects against rod shaped and coccoid bacteria such as Bacillus subtilis and Staphylococcus aureus. These molecules targeted cells in unusual ways that are clearly outside the most common suite of antibiotic targets: one compound blocked septation [139] and another interferes with the relationship between cell growth and cell division. Having discovered that some inhibitors of the Streptomyces sporulation pathway have antibiotic activity against rod shaped and coccoid bacteria, we decided to determine whether the opposite was also true. By screening antibiotics with known targets I now report that while compounds that target the ribosome, RNA polymerase, the cell wall and the cell’s electrochemical gradient have a simple antibacterial effect, any compound that targets the integrity of the DNA confers a sporulation defect at concentrations below the MIC. I then leveraged this discovery by screening a small library of 3,705 synthetic compounds, previously demonstrated to have biological activity against Gram positive bacteria [61] [176] for compounds having the capacity to inhibit sporulation in Streptomyces venezuelae. I report here that the most potent of these is a novel inhibitor of DNA gyrase identified as EN-7.

4.3 Results

4.3.1 The Streptomyces venezuelae sporulation program is sensitive to DNA damage

To determine the effect of chemical inhibitors on morphogenesis in S. venezuelae I applied antibiotics having diverse molecular targets at concentrations above their MIC to filter disks on lawns of spores and observed how growth and development responded (Figure 4.1A). Antibiotics such as rifampicin, triclosan, vancomycin and kanamycin conferred a zone of inhibited growth as expected. Further from the filter disk, at concentrations below the MIC, growth and sporulation were normal as demonstrated by the green pigmentation of the lawn surface. This was also the effect of the ionophores nigericin and CCCP. By contrast, compounds that target DNA blocked growth at high concentrations but, at concentrations below the MIC, conferred a white phenotype consistent with a failure Chapter 4. EN-7 65

to produce the green WhiE spore pigment that is associated with the completion of S. venezuelae sporulation cycle. To confirm that DNA damage blocked sporulation I carried out scanning electron microscopy (SEM) on a section of the S. venezuelae lawn immediately adjacent to the zone of inhibition formed by treatment with mitomycin C and ciprofloxacin (DNA damage) as well as kanamycin (translation inhibitor) and rifampicin (transcription inhibitor) (Figure 4.1B). I found that while the kanamycin and rifampicin images show septated aerial hyphae, indicating sporulation, the samples treated with the DNA damage agents do not show any septation. This confirms that the white phenotype induced by these molecules is caused by an inhibition of S. venezuelae sporulation.

Figure 4.1: Streptomyces venezuelae sporulation is sensitive to DNA damage A. S. venezuelae was treated with antibiotics of various targets to determine the impact on growth development. Following 48 hours of incubation sporulation inhibition was visualized by the appearance of a white halo (whi phenotype) in the sub-MIC region surrounding the zone of inhibition. Each disk is treated with 10 µL of the following antibiotics, 100 µg/mL rifampicin, 1 mg/mL triclosan, 100 µg/mL vancomycin, 250 µg/mL kanamycin, 3 mM CCCP, 100 µg/mL novobiocin, 250 µg/mL ciproﬂoxacin, 20 µg/mL mitomycin C, and 100 µg/mL bleomycin. The concentrations were selected in order to get a relatively consistent zone of inhibition between the various treatment molecules. B. Scanning electron microscopy images near the zone of inhibition of S. venezuelae treated with kanamycin, rifampicin, mitomycin C, and ciproﬂoxacin. Chapter 4. EN-7 66

4.3.2 Novel small molecule sporulation inhibitors

To determine whether this phenomenon could be used to identify novel antimicrobial agents, I screened 3,705 synthetic small molecules that had previously been screened for B. subtilis inhibition [61] for the capacity to block sporulation in S. venezuelae at sub- MIC concentrations. In this way I identiﬁed 10 compounds, all of which were antibacterial at higher concentrations but that conferred a reproducible white phenotype below the MIC (Figure 4.2). Four of the molecules, CB-6, EN-1, EN-4, and EN-7, contain a quinoline group, a common feature in synthetic compounds. Aside from this, the 10 molecules shared no signiﬁcant structural features.

Figure 4.2: Chemical structures and induced S. venezuelae phenotypes 10 µL of each molecule was added to the drug disk, placed on a lawn of S. venezuelae spores and incubated at 30 C for 48 hours before imaging. The EN-1 and AS-5 stocks were 10 mM while the remaining were at 2.5 mM.

Out of these molecules, EN-7 displayed both significant growth inhibition and sporulation inhibition (Figure 4.3A). Treatment with a filter disk containing 10 µL of 2.5 mM EN-7 resulted in a significant zone of inhibition as well as a striking band of inhibited sporulation below the MIC for growth (Figure 4.3B). To confirm that the white phenotype in this zone was due to inhibited sporulation, I imaged both green and white sections of the lawn by SEM. Consistent with my screening criteria, the green region Chapter 4. EN-7 67

showed a healthy lawn containing aerial hyphae septated into mature spores while the white region showed aerial hyphae without any septation. In spite of the exceptional EN-7 sensitivity of the S. venezuelae sporulation program, the compound conferred no other morphological defects at these concentrations. I tested EN-7 for antibiotic activity against laboratory strains of two Gram-positive pathogens (Figure 4.3C) and found that the molecule had an MIC of 0.6-1.0 µM against laboratory strains of S. aureus (ATCC 29213, TCH1516, and ATCC 33591) and E. faecalis (ATCC 51299). The compound was bacteriostatic. I also tested EN-7 for activity against three laboratory strains of the Gram-negative species E. coli. I observed no growth inhibition in either MC1061 [46] and BW25113 [64] strains at concentrations as high as 64 µM. I did, however, observe antibiotic activity against E. coli BW25113 ∆tolC ∆bamB [147] which is deficient in an outer membrane assembly factor (bamB) [222] and a major component of several efflux systems (tolC ) [32]. In this outer-membrane and efflux deficient E. coli I found that EN-7 had a MIC of 1 µM, the same as the Gram-positive pathogens. This demonstrates that EN-7 is a potent antibiotic against Gram-positive cells and suggests that its target might be conserved in Gram-negative cells though its activity is impeded by their outer membrane.

Figure 4.3: Identiﬁcation of EN-7 as an inhibitor of S. venezuale and Gram- positive pathogens A. Chemical structure of EN-7. B. Sporulation inhibition phenotype induced by EN-7. A section of the lawn was cut from the agar and both the treated (left) and untreated (right) sections imaged with scanning electron microscopy (SEM). C. EN-7 inhibits growth of Gram-positive pathogens.

In addition to the laboratory strains I also tested EN-7 for activity against a panel of eight clinical strains of S. aureus that had accompanying clinical antibiogram and full genome sequencing data (Figure 4.4). I found that EN-7 had an MIC of 1 µM for four of Chapter 4. EN-7 68

Strain Mutation mut-1 GyrB K417E mut-2 GyrB K417E mut-3 GyrA A34T mut-4 GyrB K417E mut-5 GyrB K417E mut-6 GyrA P219Q mut-7 GyrB D437N mut-8 GyrB D437N mut-9 GyrA A34T

Table 4.1: Table of gyrase mutations found in EN-7 resistant mutants

these strains (C0017, C0024, C0023, and C0019), a slightly increased MIC of 4 µM against one of them (C0112) and was inactive against the other three (C0018, C0032, and C0117) at concentrations up to 128 µM. The clinical antibiograms for all of these strains indicated extensive antibiotic resistance to ciprofloxacin, clindamycin, erythromycin, levofloxacin, and oxacillin. Using the Comprehensive Antibiotic Resistance Database (CARD) [141], I was able to identify many of the antibiotic resistance genes or modified targets within the genomes of these strains. While I could not find a clear explanation for the activity difference observed for these strains, it shows that EN-7 is active against clinically relevant strains and able to overcome a diverse range of resistance mechanisms.

4.3.3 EN-7 targets DNA gyrase

To gain insight into EN-7’s mechanism of action, I raised nine EN-7R mutants of S. aureus strain ATCC 29213. I isolated three such mutants using a serial passaging approach (mut- 1, mut-2, and mut-3) and six more by direct plating on agar containing 10 µM, or 10x MIC, of EN-7 (mut-4 through mut-9). I isolated chromosomal DNA and determined full-length genome sequences of the nine EN-7 resistant strains and found that, among a number of different mutations, each of them had acquired a point mutation in the gyrAB operon for bacterial gyrase resulting in amino acid substitutions. Three of the mutants had an altered GyrA sequence: A34T (mut-3 and mut-9) and P219Q (mut-6). The other six mutants had an altered GyrB sequence: K417E (mut-1, mut-2, mut-4, and mut-5) and D437N (mut-7 and mut-8) (Table 2). I confirmed the EN-7 resistance of these cultures using a broth dilution MIC assay, showing no activity up to 16 µM (16x MIC), the highest concentration tested. To confirm that the mutations are the cause of the observed EN-7 resistance I selected Chapter 4. EN-7 69

Figure 4.4: EN-7 inhibits growth of extensively resistant S. aureus Strains containing a particular resistance elements in their genome are labelled with blue boxes while those that do not contain the element are labelled with white. Chapter 4. EN-7 70

the most common mutation, GyrB K417E, and reconstituted it in wild-type S. aureus ATCC 29213 through allelic exchange [188]. Due to the essential nature of gyrase, the allelic exchange was accomplished in a single step by spreading plasmid containing cells as well as wild-type S. aureus ATCC 29213 cells to agar plates containing 10 µM EN-7 and 1 µg/mL anhydrotetracycline and incubated overnight at the plasmid non-replicating temperature of 37C. Following overnight incubation, I observed approximately 50 colonies on each of the wild type plates and over 2,000 on each of the plates containing the GyrB K417E allelic exchange strain (Figure 4.5A). I confirmed the loss of the plasmid by replicate plating with 10 µM EN-7 or 10 µM EN-7 with 25 µg/mL chloramphenicol (Figure 4.5B) and confirmed the EN-7 resistance for three of the resulting strains through a broth dilution assay up to 128 µM (Figure 4.5C). Finally, I confirmed the K417E mutation by amplifying and sequencing gyrA and gyrB from eight independent allelic exchange cultures (Figure 4.5D). Therefore, the EN-7 resistance observed in these strains is due to the single point mutation induced by allelic exchange. To directly test EN-7 activity, Tony Maxwell and his team at Inspiralis Inc. utilized a series of in vitro assays against purified S. aureus gyrase and topoisomerase IV (Fig- ure 4.6). First, they found that EN-7 inhibits S. aureus gyrase supercoiling activity at an IC50 of 85 nM (Figure 4.6A) while inhibiting E. coli gyrase supercoiling at an IC50 of 1.75 µM (Figure 4.6B). However, the activity against topoisomerase IV induced decatenation was reversed, with no activity, up to 100 µM observed against the S. aureus enzyme (Figure 4.6D) while the molecule inhibited E. coli topoisomerase IV induced decatenation at an IC50 of 0.89 µM (Figure 4.6C). To further investigate possible specific mechanisms of gyrase activity, they found that EN-7 does not stabilize the gyrase-DNA cleavage complex in a manner similar to fluoroquinolones (Figure 4.6E). Co-treatment with both ciprofloxacin and EN-7 does not prevent ciprofloxacin-induced formation of the double-stranded cleavage complex (Figure 4.6F), suggesting that EN-7 does not inhibit gyrase-DNA binding and acts in a manner that does not prevent the formation of the double-stranded break. In addition, EN-7 inhibits ATP-independent S. aureus gyrase- induced relaxation at an IC50 of approximately 2.5 µM (Figure 4.6G), confirming that the molecule does not act as an ATPase inhibitor. Finally, we found that S. aureus gyrase containing the fluoroquinolone resistance mutation GyrA S84L has an IC50 of 9.96 µM, an over 100-fold increase relative to the wild-type enzyme (Figure 4.6H). This collection of in vitro assays demonstrates that EN-7 is a highly potent topoisomerase inhibitor that acts independently of conventional mechanisms while being less active, in vitro, against gyrase containing the GyrA S84L mutation. Since all four mutations were in gyrAB, I determined whether they conferred resistance Chapter 4. EN-7 71

Figure 4.5: Allelic Exchange of EN-7 resistance mutations into S. aureus A. Image of three replicate plates of the allelic exchange of the EN-7 resistance GyrB K417E mutation into S. aureus versus wild type S. aureus following incubation in the presence of 1 µg/mL anhydrotetracycline and 10 µM EN-7 and incubated at the pIMAY non-replicating temperature of 37C. B. Replicate plating of 24 colonies from the allelic exchange plates in A to agar containing 10 µM EN-7 (left) and 10 µM EN-7 and 25 µg/mL chloramphenicol (right) demonstrating the loss of the pIMAY plasmid following allelic exchange. C. Determining MIC of three of the replicate plated EN-7 resistant strains from B showing no EN-7 activity up to 128 µM relative to the MIC of 1 µM in the wild type S. aureus ATCC 29213 strain. D. Sequencing of the gyrB amplicon showing the point mutation in each of the tested allelic exchange strains (top) and the translation of the corresponding region showing the K417E amino acid substitution (bottom). Chapter 4. EN-7 72

A S. aureus gyrase supercoiling E S. aureus gyrase cleavage CFX EN-7 μM CFX EN-7 μM 1 5 0 5 0 1 5 0 0 0 0 0 0 5 1 0 .0 .0 .0 .0 .1 .5 .0 .0 0 5 0 0 50 100 0 5 0 0 0 . . . . . - + 25 50 0 0 0 0 0 0 1 5 1 2 5 1 - + 5 2 1 5 1 5 1 0 0 0 R R L

SC SC

B E. coli gyrase supercoiling F S. aureus gyrase cleavage CFX EN-7 μM 50 μM CFX + EN-7 μM 1 5 + 0 0 CFX O 0 0 1 5 0 S . . 0 0 1 5 0 0 0 5 0 0 0 5 - + 0.1 0.5 0 0 ...... 0 5 0 0 0 M 0 5 0 0 0 .0 .0 .5 .1 .0 0 0 0 0 1 5 1 2 5 1 - + 5 D 5 2 1 5 1 5 1 0 0 0 R R L SC SC

C E. coli topoisomerase IV decatenation G S. aureus gyrase relaxation CFX EN-7 μM 1 5 MOX EN-7 μM 0 0 1 5 0 - + 5 10 .0 .0 .0 .0 .1 .5 .0 .0 0 5 0 0 1 5 1 5 0 0 0 0 0 0 1 5 1 2 5 1 .0 .0 .1 .5 .0 0 .0 .0 .1 .5 .0 0 - + 0 0 0 0 1 5 1 0 0 0 0 1 5 1

C R

SC D

D S. aureus topoisomerase IV decatenation H S. aureus mutant gyrase GyrA S84L supercoiling CFX EN-7 μM MOX EN-7 μM 1 5 0 0 0 0 .0 .0 1 5 0 0 0 0 5 0 0 0 0 0 - + 0 0 .0 .0 .1 .5 .0 .0 0 5 0 0 0 0 . . . 0 5 0 0 5 0 0 5 10 0 0 0 0 1 5 1 2 5 1 - + 5 1 1 2 5 1 2 5 1 2 5 1

R C

D SC

Figure 4.6: in-vitro inhibition of S. aureus gyrase by EN-7 A series of DNA gels showing the in vitro EN-7, ciprofloxacin (CFX), and moxifloxacin (MOX) activity against DNA gyrase and topoisomerase IV alterations to DNA topology. R - relaxed, SC - supercoiled, L - linear, C - catenated, D - decatenated. (+) is a control showing the effect of untreated enzyme while (-) is a control showing the impact of no added enzyme. Chapter 4. EN-7 73

with the classical DNA gyrase targeting antibiotics ciprofloxacin and novobiocin as well as the novel bacterial topoisomerase inhibitor (NBTI) gepotidacin (Figure 4.7). I observed no significant cross-resistance with ciprofloxacin or novobiocin. The MICs of ciprofloxacin and novobiocin against the mutants were the same as those against the parent strain, 0.2 µg/mL and 0.06 µg/mL respectively. Additionally, all of the clinical strains that I tested (Figure 4.4) contained fluoroquinolone resistant GyrA mutations. Along with the in vitro data, this confirms that the mechanism of EN-7 resistance is distinct from that of the other two antibiotics. However, I observed partial cross resistance with gepotidacin. The NBTI molecule inhibited growth of the wild-type strain at 0.5 µM. Strains containing GyrA A34T (mut-3 and mut-9), GyrA P219Q (mut-6), and GyrB K417E (mut-1, mut-2, mut-4, and mut-5) exhibited moderate resistance with MICs of either 4 or 8 µM while the strains containing GyrB D437N (mut-7 and mut-8) displayed a higher level of resistance with an MIC of 32 µM. This suggests that EN-7 may share a resistance mechanism with gepotidacin. While EN-7 is structurally unique, it shares some superficial structural features with NBTIs such as gepotidacin, including a quinoline group connected to a heterocycle via a flexible linker region (Figure 4.8).

Figure 4.7: Cross-resistance of EN-7 resistant strains with other gyrase inhibitors No observed cross-resistance between EN-7 and other gyrase targeting antibiotics, however, there is some cross resistance to the NBTI gepotidacin. Plots of optical densify of wild-type and resistant mutant strains treated with EN-7 as well as ciproﬂoxacin and novobiocin.

I took two approaches to determine whether the EN-7R mutations fall in previously identiﬁed functional regions of DNA gyrase. First, I mapped the mutations onto an alignment of published GyrA and GyrB sequences from S. aureus, E. faecalis, S. coelicolor, S. venezuale, B. subtilis, and E. coli (Figure 4.9A). All four of the residues altered in EN-7R strains are fully conserved across these six diverse bacteria. Second, I mapped my Chapter 4. EN-7 74

Figure 4.8: Chemical structures of EN-7 and other investigational gyrase inhibitors Chapter 4. EN-7 75

EN-7 resistant mutations to published gyrase structures in complex with ciprofloxacin (2XCT) and GSK299423 (2XCS), an NBTI molecule (Figure 4.9B and 4.9C) [21]. Con- sistent with their resistance profile, all four of the mutations are distinct from common fluoroquinolone resistance mutations at GyrA residues S84, S85 and E88 [7], previously reported NBTI NXL101 resistance mutation GyrA M121K [31] and NBTI-gyrase interaction pocket GyrA A68, G72, M75, and M121 [21]. Both the GyrB D437N and GyrB K417E mutations are located in the DNA binding TOPRIM domain and directly con- tact the DNA backbone. The GyrA A34T mutation is found in the winged-helix DNA binding domain and interacts with the DNA backbone far from the NBTI binding site while P219Q is located further from the direct DNA binding site.

4.4 Discussion

In this study, I describe the discovery of a novel gyrase inhibitor, EN-7, from a screen against Streptomyces sporulation septation. Our lab has previously used Streptomyces as a small molecule screening platform for compounds that impact secondary metabolism [58] and sporulation [139], which is a form of developmentally controlled cell division. Streptomyces sporulation employs specific genes that drive changes in cell morphology and fate. However, these sporulation mechanisms also co-opt core macromolecular machinery traditionally employed for vegetative growth and division, exposing unique points of vulnerability. We have found that these sporulation mechanisms can be specifically targeted by novel antimicrobials at sub-inhibitory concentrations, selecting for molecules with more specific modes of action than screening for general growth inhibition. Here, I identify that S. venezuelae sporulation is highly sensitive to known antibiotics that directly target DNA, such as mitomycin C and bleomycin, as well as those that interfere with DNA gyrase, such as ciprofloxacin and novobiocin. Using the classic white (whi) phenotype to identify sporulation inhibition, I observed that sporulation was inhibited by several types of DNA damage: mitomycin C is a DNA crosslinker [262], bleomycin causes double strand breaks [116], ciprofloxacin blocks DNA gyrase action after it confers a double strand break [76] and novobiocin blocks DNA gyrase ATPase activity, leading to positive supercoiling and stalling of the replication machinery [175]. This is the first report that such diverse forms of DNA damage block sporulation in a streptomycete. While DNA damage induced cell division inhibition has been extensively described in other bacteria, such a system is not known within Streptomyces. Across diverse bacteria, the SOS-response induces the expression of specific proteins that delay cell division. This includes YneA in B. subtilis [145], SulA in E. coli [25], and SosA in S. aureus [34], Chapter 4. EN-7 76

Figure 4.9: Analysis of EN-7 resistance gyrase mutations A. Amino acid alignment of the two GyrA and two GyrB mutations found in the EN-7 resistant S. aureus strains against published S. aureus (SAURE), parent strain (JN-44), En- terococcus faecalis (EFAEC), Streptomyces coelicolor (SCOEL), Streptomyces venezuale (SVENE), Bacillus subtilis (BSUBT), and Escherichia coli (ECOLI) sequences. B. Lo- cation of EN-7 resistant mutations on linear GyrA and GyrB domain structure. C. Location of EN-7 resistant mutations on previously published gyrase crystal structure. The structural information was created by overlaying a S. aureus gyrase structure with GSK299423 (2XCS) and a S. aureus gyrase structure with ciproﬂoxacin (2XCT). Gyrase is depicted in the structure as red sticks while GSK299423 is green. Resistance determinants for EN-7 (blue), GyrA A34T, GyrA P219Q, GyrB K417E and GyrB D437N, ciproﬂoxacin (yellow), GyrA S84, GyrA S85 and GyrA E88 and GSK299423 (cyan) GyrA M121K are shown on the structure. Chapter 4. EN-7 77

as well as Rv2719c and DivS in the Actinobacteria Mycobacterium [51] and Corynebac- terium [194], respectively. While S. venezuelae contains homologs to the SOS-response modulators LexA and RecA, no equivalent division inhibitor has yet been identified. However, my findings demonstrate that S. venezuelae is very sensitive to a variety of forms of DNA damage and responds by delaying sporulation, possibly to prevent passing damaged chromosomes to its progeny. I then used this effect as a platform for screening small molecules for bioactivity and identified 10 active molecules. The most potent of these is EN-7, a novel bacterial topoisomerase inhibitor. EN-7 is highly active against a variety of Gram-positive pathogens, including extensively resistant clinical S. aureus strains containing resistant elements to fluoroquinolone and other antibiotic classes. While I did not observe activity against wild-type E. coli, an efflux and outer-membrane deficient strain shows comparable activity to S. aureus, suggesting either the outer-membrane is inhibiting entry of the molecule to the cell or the compound is being removed from the cell by efflux. Both genetic and biochemical methods were used to identify DNA gyrase as the target of EN-7. Bacteria contain two principle Type IIA topoisomerases, DNA gyrase and topoisomerase IV. DNA gyrase cleaves and reseals DNA to induce negative supercoiling and relieve torsional stress induced by transcription or translation [276]. It is a

tetramer composed of GyrA2GyrB2, is highly conserved across bacteria, and its function can be inhibited through multiple mechanisms and by several different antibiotic classes. Quinolones (naladixic acid) [110], fluoroquinolones (ciprofloxacin) [76], and the toxin CcdB [24] inhibit gyrase by stabilizing the gyrase-DNA cleavage complex while aminocoumarins (novobiocin) [175] inhibits the ATPase activity of the enzyme. Ad- ditionally, there are a growing number of investigational gyrase inhibitors, including simocylonone D8 [39], thiophenes [47], imidazopyrazinones [105], and the NBTIs (novel bacterial type II topoisomerase inhibitors) [21] such as gepotidacin [29][94][258]. These novel inhibitors act by inhibiting DNA-gyrase binding, stabilizing a pre-cleave complex, or stabilizing DNA-cleavage complexes, highlighting the diverse gyrase inhibition mechanisms. Topoisomerase IV, while also able to cleave and reseal double stranded DNA and

composed of a heterotetramer of ParC2ParE2, has a sightly different function than gyrase. It is responsible for decatenating DNA following replication and relax positive DNA supercoils [33] [124]. In many instances molecules that target gyrase also inhibit topoisomerase IV; however, specific molecules can preferentially inhibit either enzyme and there are also differences across different organisms. For example, in S. aureus, topoisomerase IV is more sensitive than gyrase to ciprofloxacin inhibition [33] and norfloxacin Chapter 4. EN-7 78 preferentially targets topoisomerase IV while nalidixic acid targets gyrase [98]. In E. coli, ciprofloxacin preferentially inhibits gyrase over topoisomerase IV [146]. Tony Maxwell observed that for S. aureus enzymes, EN-7 strongly inhibits gyrase supercoiling while having no impact on topoisomerase IV decatenation activity. However, in E. coli the molecule inhibits both enzymes while maintaining greater activity against gyrase. Ad- ditionally, he saw that EN-7 does not stabilize the DNA-gyrase cleavage complex, does not inhibit ciprofloxacin-induced DNA cleavage complex, and does not interfere with the ATPase function of gyrase. While we do not know the exact mode of action, these results show that EN-7 is not acting in a manner similar to classic gyrase inhibitors by inhibiting DNA-gyrase binding, stabilizing the cleavage complex, or inhibiting ATPase activity. In addition to the biochemical results, I observe four mutations in S. aureus that give rise to EN-7 resistance, GyrA A34T, GyrA P219Q, GyrB K417E, and GyrB D437N. While the mutations do not cluster around a possible binding site, two of the mutations have been previously described. Studies have found that a GyrB D437N mutation results in resistance to non-NBTI gyrase inhibitors QPT-1, etoposide, and AZD0914 [48] while both GyrB D437N and GyrB K417E mutations result in resistance to NBTI’s AZ6142, AZ0217, and MCHEM18 [157]. In addition, a nearby A32V mutation has also been shown to alter NBTI sensitivity [157] and a study in multi-drug resistant Neisseria gonorrhoeae showed that GyrA D82N and ParC D79N mutations resulted in gepotidacin resistance [138]. Testing our EN-7 resistant strains against other gyrase inhibitors, we found no cross- resistance with ciprofloxacin or novobiocin. However, we do see both intermediate resistance in GyrA A34T, GyrA P219Q, and GyrB K417E strains and significant resistance in GyrB D437N strains against the NBTI gepotidacin. This is the first observation that resistance to gepotidacin can arise from GyrA A34T, GyrA P219Q, GyrB K417E, and GyrB D437N mutations. It also highlights previous reports that NBTI activity is susceptible to mutations that are far from the specific binding site [157]. Finally, I suggest that there are many pathways that could be subjected to similar screening methods, including endospore formation in Bacillus subtilis, myxospore formation in Myxococcus xanthus and even sporulation programs in eukaryotic organisms such as Saccharomyces cereviciae. Chemical screens against these morphogenetic programs could be a valuable means of identifying new chemical matter of potential clinical interest. Chapter 5

Concluding Remarks

5.1 Thesis Summary

Streptomyces are fantastic organisms. Not only are they natural chemists able to synthesize complex bioactive molecules, their development program repurposes many conserved elements of other bacteria. Therefore, these organisms are of signiﬁcant interest to identify new natural products, but also to better understand their biology and use this biology as a screening platform for new molecules.

I have demonstrated that Streptomyces sporulation can be used as a screening platform to identify molecules with novel bioactivities and have worked to understand the mechanisms of action for these molecules. The possibility of non-specific mechanisms of action is a significant challenge when deciding what screening hits to prioritize. There- fore, I developed a high-throughput screen using two fluorescent dyes, TO-PRO-3 iodide

and DiOC2(3), and applied this method to a collection of 3,705 bioactive molecules and to one Streptomyces sporulation inhibitor in particular, Min-1. While I was not able to identify a specific molecular target for Min-1, I was able to find that it disrupts the cell membrane while activating the LiaRS stress response. Most interestingly, working with Prof. Petra Levin, we found that it induces a concentration dependent short cell phenotype when cultured in nutrient-rich media. Another new molecule that strongly inhibits Streptomyces sporulation is EN-7. I identified that it targets bacterial gyrase and, working in collaboration with Prof. Tony Maxwell, showed it inhibits DNA supercoiling at concentrations significantly lower than existing gyrase targeting antibiotics. Finally, using a panel of known antibiotics, I showed that S. venezuelae sporulation is very sensitive to DNA damage.

79 Chapter 5. Concluding Remarks 80

5.2 Future Directions

I believe that my results open a number of further avenues for research.

5.2.1 Investigating mechanism of Streptomyces sporulation inhibition via DNA damage

While S. venezuelae contains homologs to well-described DNA-damage response modulators LexA and RecA, no specific division inhibitor has yet been identified. However, my findings demonstrate that S. venezuelae is very sensitive to a variety of forms of DNA damage and responds by delaying sporulation, possibly to prevent passing damaged chromosomes to its progeny. Previous RNA-seq results from our lab suggests that SsgB, a positive regulator of FtsZ-ring formation in Streptomyces, may play a role in this DNA-damage response. Another student in the lab is following up with this work by placing ssgB under the control of an inducible promoter to determine if increasing expression can rescue the DNA damage-induced sporulation inhibition. Another exciting opportunity would be to work with Dr. Susan Schlimpert at the John Innes Centre to use her Streptomyces microfluidic fluorescence time-lapse imaging system to identify protein mislocalization and expression changes under DNA damage conditions during sporulation.

5.2.2 Additional approaches for elucidating Min-1’s target and mechanism of action

Despite using a number of techniques, I was unable to identify a speciﬁc molecular target for Min-1 and the mechanism behind its short-cell phenotype. However, there are still a number of methods and tools that I have not used that could help further narrow in on possible mechanisms. First is using the B. subtilis non-essential deletion library [150]. While I was unable to reproduce results from the a CRISPRi essential knockdown library, it is possible that the expanded set of possible targets provided by the other library may give new insight. Second, RNA-seq can provide an extensive picture of how a bacteria responds to antibiotic treatment. Identifying changes in gene expression on Min-1 exposure could help provide new insight into its mechanism of action and its impact on cell length. Chapter 5. Concluding Remarks 81

5.2.3 Identify mechanism of gyrase inhibition for EN-7

With the help of Prof. Tony Maxwell, we have been able to dive fairly deep into possible mechanisms of gyrase inhibition by EN-7. We have ruled out inhibition of DNA-gyrase binding, disruption of ATPase activity, and stabilization of the cleavage complex; however, questions around the speciﬁc mechanism of action remain. One method that would help answer these questions is to determine a co-structure between EN-7 and gyrase. Fortunately, due to the number of clinical and investigational gyrase inhibitors in the lit- erature, there is an abundance of structural work done on the enzyme. The RCSB Protein Data Bank contains 179 structures of gyrase or gyrase subunits, with 42 of them from S. aureus and many of these in complex with small molecule inhibitors. Not only would solving a structure help identify potential inhibition mechanisms, it could help guide in creating more potent EN-7 analogs with improved physical and chemical properties.

5.2.4 Creating high-thoughput method of screening Strepto- myces sporulation inhibition

This work has shown that inhibiting Streptomyces sporulation can be a useful tool in uncovering novel bioactive molecule; however, the methodology is relatively low throughput and manual. Therefore, creating a verion of the same assay that can be performed in an automated, high-throughput method could help make future screening programs more expansive. Possible methods for accomplishing this include creating lux or gfp fusions with both a housekeeping gene like hrdB and a late-stage sporulation gene like whiE. Screening a dilution series of unknown molecules could identify molecules that, below an inhibitory concentration, inhibit sporulation while not impacting overall growth. These hits could then be screened for DNA targeting and other mechanisms.

5.2.5 Additional small molecule screens against bacterial development

In a similar method to screening against Streptomyces sporulation, other bacteria with unique development programs could be used as screening platforms to identify new bioactive molecules. These include endospore formation in B. subtilis, myxospore formation in Myxococcus xanthus and even sporulation programs in eukaryotic organisms such as Saccharomyces cereviciae. Chapter 5. Concluding Remarks 82

5.3 Conclusion

Although Streptomyces have been heavily used in the discovery of new natural products, my work has shown that their unique biology can also be used to screen for identifying new bioactive molecules. In this thesis I report on the results of this screening method as well as deep investigations into the mechanism of action of two novel molecules, the cell envelope active Min-1 and gyrase inhibiting EN-7. In addition to opening new avenues for identifying molecules, I have contributed to our evolving understanding of how small molecules impact cell size and morphology as well as inhibition of bacterial gyrase. Chapter 6

Materials and Methods

6.1 General Experimental Procedures

6.1.1 Strains and plasmids

Descriptions and references for the bacterial strains used in this study are provided in Table 6.1

6.2 Growth inhibition assays

6.2.1 Broth dilution assay

An overnight culture was diluted 1/100 in fresh LB media and incubated at 37 C on a rotating incubator set to 225 rpm until reaching an OD600 of 0.4. The cells were diluted 1:1000 and 198 µL added to a 96-well plate. 2 µL of treatment compound was added and the plate incubated for 18 hours in a rotating incubator at 37 C. After overnight growth, the optical density at 600 nm was read and the MIC determined by the lowest concentration showing less than 10% growth.

6.2.2 Streptomyces sporulation inhibition

S. venezuelae spore stock was diluted in sterile saline, spread on MYM agar and allowed to dry. For spot diffusion, either 1 or 2 µL of molecule was spotted on the agar while for disk diffusion 10 µL of molecule was added to a drug disk and allowed to dry in air before placing on MYM agar. Following treatment, the plate was incubated at 30 C for 48 hours and photographed. Identification of sporulation inhibition was performed by eye as the

83 Chapter 6. Materials and Methods 84

Strain Note Reference S. venezuelae NRRL B-65442 B. subtilis 168 B. subtilis JH642 B. subtilis HB0050 PσW-lacZ [129] B. subtilis BFS2470 CU1065 PliaI -pMUTIN [172] B. subtilis AHL35 gtaB::spc [279] B. subtilis PL1310 pgcA::Tn10 cat [279] B. subtilis BW421 ugtp::cat [279] B. subtilis (p)ppGpp0 JH642 ∆yjbM ywaC::kan [153] relA::mls S. aureus ATCC 29213 S. aureus TCH 1516 Used in mouse model experiment S. aureus ATCC 33519 S. aureus ATCC BAA-44 S. aureus ATCC BAA-41 S. aureus C0017 Wright clinical strains database S. aureus C0024 Wright clinical strains database S. aureus C0023 Wright clinical strains database S. aureus C0019 Wright clinical strains database S. aureus C0112 Wright clinical strains database S. aureus C0018 Wright clinical strains database S. aureus C0032 Wright clinical strains database S. aureus C0117 Wright clinical strains database E. faecalis ATCC29212 E. coli K-12 E. coli MC1061 [46] E. coli BW25113 [64] E. coli BW25113 ∆tolC ∆bamB [147]

Table 6.1: Table of strains used in studies Chapter 6. Materials and Methods 85

formation of a white ring beyond the zone of inhibition. Speciﬁcally, the screen of B. subtilis active molecules for S. venezuelae sporulation inhibition was performed with 2 µL of 2.5 mM molecules spotted directly to the MYM plate and incubated as above.

6.2.3 Superﬁcial S. aureus skin infection model

Superﬁcial skin lesions were created (tape stripping) on the scruﬀ of BALB/c female mice. S. aureus TCH1516 at 107 cfu per 5 µL was used to infect the tape striped area. Four hours after infection 20 µL of treatment was applied to the skin with additional treatments ever 12 hours over four days. Mice were harvested 12 hours after the last treatment, tissues collected, homogenized, diluted, plated on TSA supplemented with 4 µg/mL oxacillin and 50 µg/mL kanamycin and incubated overnight at 37 C. Colonies were counted the following morning.

6.3 PAINs assays

6.3.1 Duel-dye membrane disruption screen

An overnight culture of B. subtilis 168 was diluted 1:100 in LB media and incubated at 37C on a rotating incubator set to 225 rpm until reaching an OD600 of 0.4. Cells were pelleted at 3000g for 10 minutes, the supernatant decanted, and washed twice in PBS+ (0.14 M NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4 at pH 7.4 supplemented with 10 mM glucose and 0.5 mM MgCl2). The pellet was then resuspended in PBS+ and diluted to an OD600 of 0.1. Concurrently, a blank PBS+ sample was also prepared without B. subtilis. Dyes were added into both the sample and control to a

ﬁnal concentration of 625 nM TO-PRO-3 iodide (λex = 600nm, λem = 650nm) and 10 µM

DiOC2(3) (λex = 450nm, λem = 510nm). Both the culture and dye control solutions were added to 96-well plates and the test compounded were added to a ﬁnal concentration of 5 µM and 20 µM. Each plate also contained treatment controls wells for DiOC2(3) (10 µM CCCP) and TO-PRO-3 iodide (10 µg/mL nisin). Samples were incubated at room temperature in the dark for 5 minutes, then treated with the test compound. Fluorescence was read immediately following treatment using a BioTek Synergy H1 plate reader.

6.3.2 Dynamic Light Scattering

Molecules of interest were diluted 1:100 in LB media in a 96-well plate. Light scattering was then measured using a Malvern Analytical Zetasizer. Chapter 6. Materials and Methods 86

6.4 Miscroscopy Methods

6.4.1 Scanning electron miscroscopy

Cultures of S. venezuelae were treated with molecules in a disk diffusion assay, as described above. Samples were preared for SEM by cuting at the edge of the zone of inhibition and fixed with 2% glutaraldehyde in 0.1% sodium cacodylate buffer followed by sputter coating with gold. Images were obtained using an FEI XL30 ESEM at the Nanoscale Biomedical Imaging Facility at Sick Kids Hospital.

6.4.2 B. subtilis cell length measurement and FtsZ immunoﬂu- orescence labeling

B. subtilis JH642 cultures were incubated on a roller drum at 37 C from a single colony in LB + 0.2% glucose to OD600 = 0.2, then diluted 1:100 in LB-glucose with sub-inhibitory concentrations of Min-1 and grown to early exponential phase (OD600 = 0.15-0.3). For samples cultured in minimal media, cells where ﬁrst cultured from a single colony in

S750-glucose or S750-glycerol at 37 C overnight. The overnight cultures were diluted into minimal medium to an OD600 = 0.005, then cultured to OD600 = 0.2 before being diluted 1:100 +/- Min-1 as indicated for the LB-glucose cultures. Early exponential phase samples were fixed with paraformaldehyde/glutaraldehyde (2.6%/0.008%) for 15 min at room temperature followed by a 30-min incubation on ice. Fixed cells were then adhered to an 18 well slide with 1% poly-l-lysine. Adherent cells were briefly incubated with 1 mg/mL lysozyme, then washed with PBS, followed by incubation with wheat germ agglutinin-tetramethylrhodamine to label the cell wall. FtsZ rings were labeled by immunofluorescence, using affinity-purified polyclonal rabbit anti-FtsZ antibody [280] followed by a goat anti-rabbit secondary antibody conjugated to Alexa-488. Images were acquired with an Olympus BX51 microscope with an OrcaERG camera and captured using Nikon Elements. Cell lengths were determined by manually measuring the distance between adjacent septa using Nikon Elements analysis software. Cells were scored as positive for FtsZ-ring formation if they contained a fluorescent band across the width of the cell, two fluorescent foci across from one another at midcell, or a single focus visible at the midpoint of an invaginating septum. Chapter 6. Materials and Methods 87

6.5 Reporter assays

Screening for activation of PliaI and PσW was performed by disc diﬀusion. A single colony of the reporter strain was incubated in LB broth overnight at 37 C with the corresponding antibiotics. The overnight culture was diluted 20-fold in liquid LB and 500 µL of the diluted culture was added to an LB agar plate containing 500 µL of a 2% X-Gal solution. The antibiotics were either spotted directly to the LB agar plate or absorbed into a drug disk that was placed on the agar prior to incubating overnight at 37 C. After incubation, the plates were scored for appearance of the blue rings at or near the edge of the zone of inhibition.

6.6 Isolation and characterization of resistant mutants

6.6.1 Resistant mutant generation

Resistant mutants were generated using both serial passaging and spontaneous mutant methods. For serial passaging, a single colony of S. aureus ATCC 29213 was incubated overnight in TSB liquid media at 37 C. Following overnight growth, 2 µL of the overnight culture was diluted into 5 mL of fresh media containing 0.25X MIC of EN-7 and incubated overnight at 37 C. The culture was then diluted into 0.5X MIC EN-7 and the procedure repeated with each subsequent dilution increasing the EN-7 concentration by 2X until reaching 16X MIC of EN-7. This culture was then plated to solid TSB agar and a single colony selected as the mutant strain. For spontaneous mutants, approximately 108 S. aureus cells from an overnight culture were plated to BHI agar plate containing 10 µM (10X MIC) EN-7 and incubated overnight at 37 oC. The resistance mutation frequency was determined by counting the number of colonies on the EN-7 plates versus the number of colonies of the untreated plates.

6.6.2 Sequencing resistant mutants

Genomic DNA was isolated from overnight cultures of the mutant S. aureus strains using a GeneJET Genomic DNA Purification Kit from Thermo Scientific. Briefly, 2 mL of overnight culture of S. aureus was harvested by centrifuging at 5000 g for 10 minutes. The pellet was resuspended in lysis buffer (20 mM Tris-HCl, pH 8.0, 2 mM EDTA, 1.2% Triton X-100) with 2 µg/mL lysostaphin and incubated in a water bath at 37 C for 30 Chapter 6. Materials and Methods 88

minutes. Following incubation, 200 µL of lysis solution with 20 µL Proteinase K was added to each sample and incubated at 56 C for another 30 minutes. Following lysis, 20 µL RNaseA solution was added and allowed to incubate at room temperature for 10 minutes. Addition of 400 µL of 50% ethanol was then followed by DNA isolation and puriﬁcation on a GeneJET Genomic DNA Puriﬁcation Column. Sequencing was performed using an Illumina MiSeq with Nextera XT library prepa- ration and analyzed with Geneious software to map the reads to the S. aureus ATCC 29213 assembly (RefSeq GCF 001879295.1) and identify SNPs.

6.6.3 S. aureus allelic exchange

The gyrB sequence containing the K417E mutation was amplified from EN-7 resistant S. aureus mut-1 genomic DNA from the using primers gyrB K417E F (ATATGGTAC- CACTCAGGATATGCCACAAATCT) and gyrB K417E R (ATATGCGGCCGCATCGT- GCAATAGACCATTTTGG). The amplicon was digested using Fast Digest KpnI and NotI and ligated into similarly cut pIMAY. The pIMAY gyrB K417E plasmid was pas- saged through S. aureus RN4220 at 30 C before being electroporated into ATCC 29213 and also grown at 30 C. A single colony from the transformation plate as well as a colony of wild-type S. aureus ATCC 29213 were selected and grown overnight in 5 mL TSB broth. 100 µL of the overnight cultures were plated to 10 mL BHI agar plates containing 1 µg/mL anhydrotetracycline and 10 µM EN-7 and incubated overnight at 37 C. To confirm that the pIMAY plasmid had been lost, 24 colonies from the transformation plates were replicate plated to a BHI plate containing 10 µM EN-7 and another plate containing both 10 µM EN-7 and 25 µg/mL chloramphenicol. The confirm the allelic exchange, colony PCR was performed on eight of the 24 colonies using both gyrB K417E F and gyrB K417E R primers as well as gyrA A34T F (ATATGGTACCTGCACAGCCACCGTTGTATA) and gyrA P219Q R (ATATGCGGCCGCTGTGGGCACGATCTTTAGCT) primers. These amplicons were sequenced using an ABI 3730XL instrument.

6.7 Gyrase inhibition

6.7.1 E. coli gyrase supercoiling

1 U (pre-determined cleavage activity) of DNA gyrase was incubated with 0.5 µg of relaxed pBR322 DNA in a 30 µL reaction at 37 C for 30 minutes under the following

conditions: 35 mM Tris.HCl (pH 7.5), 24 mM KCl, 4 mM MgCl2, 2 mM DTT, 1.8 mM Spermidine, 1 mM ATP, 6.5% (w/v) glycerol and 0.1 mg/ml BSA. Each reaction was Chapter 6. Materials and Methods 89

stopped by the addition of 30 µl chloroform/iso-amyl alcohol (24:1) and 20 µl Stop Dye (40% sucrose, 100 mM Tris-HCl (pH 7.5), 10 mM EDTA, 0.5 µg/ml bromophenol blue), before being loaded on a 1.0% TAE agarose gel. Gels run at 90V for 2 hours.

6.7.2 S. aureus gyrase supercoiling

As for E. coli gyrase except assay conditions were: 40 mM HEPES. KOH (pH 7.6), 10 mM magnesium acetate, 10 mM DTT, 2 mM ATP, 500 mM potassium glutamate, and 0.05 mg/ml BSA. Gels run at 80V for 3 hours.

6.7.3 E. coli topoisomerase IV decatenation

1 U of topoisomerase IV was incubated with 200 ng kinetoplast catenated DNA in a 30 µL reaction at 37 C for 30 minutes under the following conditions: 50 mM HEPES-KOH (pH 7.6), 100 mM potassium glutamate, 10 mM magnesium acetate, 10 mM dithiothreitol, 1 mM ATP and 50 µg/ml BSA. Each reaction was stopped by the addition of 30 µL chloroform/iso-amyl alcohol (24:1) and 20 µl Stop Dye, before being loaded on a 1.0% TAE agarose gel. Gels run at 90V for 2 hours.

6.7.4 S.aureus topoisomerase IV decatenation

As for E coli topoisomerase IV except assay conditions were: 50 mM Tris-HCl (7.5), 5

mM MgCl2 5 mM DTT, 1.5 mM ATP, 350 mM potassium glutamate and 0.05 mg/ml BSA. Gels run at 80V for 2 hours.

6.7.5 S.aureus gyrase cleavage

1 U of gyrase was incubated with 0.5 µg of supercoiled pBR322 DNA in a 30 µL reaction at 37 C for 30 minutes under the following conditions: 40 mM HEPES. KOH (pH 7.6), 10 mM magnesium acetate, 10 mM DTT, 100 mM potassium glutamate and 0.05 mg/ml albumin. 0.2% SDS and 0.1 mg/mL Proteinase K were added before a further incubation at 37 C for 30 minutes. Each reaction was stopped by the addition of 30 µl chloroform/iso- amyl alcohol (24:1) and 20 µL Stop Dye, before being loaded on a 1.0% TAE agarose gel. Gels run at 80V for 2 hours. Chapter 7

Appendix 1: A chemical inhibitor of cell growth reduces cell size in Bacillus subtilis

The results and discussion presented in this chapter are supplemental to Chapter 3: A chemical inhibitor of cell growth reduces cell size in Bacillus subtilis

7.1 Results and Discussion

7.1.1 Min-1 does not inhibit B. subtilis sporulation

Min-1 was initially discovered from a screen against Streptomyces coelicolour sporulation. One hypothesis for this activity was that the molecule inhibited the divisome. The divisome is highly conserved across bacteria and even used in non-binary ﬁssion forms of reproduction, including in Streptomyces sporulation and B. subtilis sporulation. B. subtilis sporulation is an asymmetric form of cell division that is induced by nutrient limitation that generates endospores that are resistant to antibiotics, high heat, UV radiation, and other extreme environments [84]. This resistance makes them a fantastic model for study, since any defects in the pathway results in heat-sensitive endospores. Therefore, we could use the formation of heat-sensitive endospores as a method to screen for divisome inhibition by Min-1 or other molecules. Treating B. subtilis cells with up to 250 µM Min-1 had no impact on endospore formation (Figure 7.1). This demonstrated that Min-1 is not directly inhibiting the Streptomyces or Bacillus divisome.

90 Chapter 7. Appendix 1 91

107 106 105 104 103 102 1 Number of spores/mL 10 100 0 50 100 150 200 250 Min-1 concentration (µM)

Figure 7.1: Min-1 does not disrupt B. subtilis sporulation

7.1.2 Min-1 activates B. subtilis PywaC-lux reporter

In addition to the PliaI -lacZ and PσW-lacZ reporter used in studying Min-1 we also used a PywaC -lux reporter strain to conﬁrm the cell envelope activity. The reporter had been designed in the lab of Prof. Eric Brown at McMaster University and was shown to be highly sensitive to antibiotics that disrupt peptidoglycan, WTA, and undecaprenol synthesis [71]. The strain has also been used by Prof. Brown’s former student Tomasz Czarney to screen for cell wall active molecules [60]. To test Min-1, we sent the molecule to Prof. Brown’s lab and Tomasz Czarney applied it to the PywaC -lux strain. He observed that Min-1 activation of the reporter peaks at 18 µM, which then decreases until 64 µM

(Figure 7.2). This supports the PliaI -lacZ activation result, demonstrating that Min-1 disrupts the cell envelope.

7.1.3 Unable to generate Min-1 resistant mutants

As seen in my study of EN-7, generating resistant mutants against a molecule can be an effective way to narrow in on possible mechanisms of action. However, despite attempting with both B. subtilis and S. aureus using both serial dilution and plating methods, I was unable to generate resistant mutants against Min-1. This was unfortunate in that it was not able to help in determining the molecule’s mechanism of action; however, it suggests that its target cannot be easily modified or that simple alterations of the cell wall or efflux capability are not able to inhibit the molecule’s activity. The inability to generate resistant mutants against teixobactin was used as a demonstration of its clinical utility [163]. Chapter 7. Appendix 1 92

0 Fold Luminescense Increase 1 10 100 1000 Min-1 concentration (µM)

Figure 7.2: Min-1 activates B. subtilis PywaC-lux reporter strain

7.1.4 Min-1 potentiates some known antibiotics

Identifying synergy, additive eﬀects, or antagonism with known antibiotics can be a useful method to help understand the mechanism of action of a new molecule [90]. I used checkerboard assays to test the ability of Min-1 to impact the activity of a range of diﬀerent known antibiotics (Figure 7.3). Any interaction between the two molecules used in these assays is determined by determining an FIC value. This is calculated by determining the MIC of the molecule alone along with the lowest MIC observed in the drug combination (Equation 7.1).

MICofAincombination MICofBincombination FIC = + (7.1) AB MICofAalone MICofBalone I have summarized the MIC and FIC results in Table 7.2 and show sample checkerboard plates in Figure 7.3. Min-1 has an additive effect on a variety of different antibiotics and no effect on others. It shows significant effect on the aminoglycoside kanamycin and closely related spectinomycin with an FIC of 0.25 and 0.19 respectively while having no effect on the macrolide erythromycin with an FIC of 2 as well as the 50S ribosome-binding antibiotic chloremphenicol with an FIC of 2. At the cell envelope, the molecule is additive with the lipid II targetting vancomycin (0.75), bacitracin (0.37), and nisin (0.59), while having no effect on ampicillin (2). Finally, the molecule has no impact on trimethoprim (2) and gyrase (2) while being additive with rifampicin (0.25). This shows that Min-1 is not simply disrupting the membrane and allowing molecules that act intracellularly to more easily access their target while also being able to improve the activity of molecules that act outside of the cell, such as vancomycin and nisin. I cannot fully explain this Chapter 7. Appendix 1 93 collection of activities, but this shows that sub-inhibitory concentrations of Min-1 has a multitude of effects on various antibiotics.

Figure 7.3: Min-1 potentiates known antibiotics Sample of four checkerboard assays performed to determine the impact of Min-1 on the action of known antibiotics

7.1.5 Identifying sensitive strains using a B. subtilis knockdown library

As high-throughput equipment and genetic manipulation becomes more available, chemical- genetic screening of new bioactive molecules is becoming increasingly common [100] [230]. One such tool is the B. subtilis CRISPRi essential knock-down library, each strain of which contains a CRISPRi element for speciﬁc genes that is inducible by xylose [207]. Prof. Eric Brown has a copy of this collection and his lab tested the eﬀect of Min-1 on the growth of these strains. First, they found that none of the strains showed growth when treated with Min-1 concentrations greater than the MIC and induced with either 0.01% and 0.05% xylose. This demonstrates that reducing the expression of an individual Chapter 7. Appendix 1 94

Antibiotic Target FIC Spectinomycin 30S ribosome 0.19 Kanamycin 30S ribosome 0.25 Chloremphenicol 50S ribosome 2 Erythromycin 50S ribosome 2 Nisin Lipid II 0.59 Trimethoprim Folate biosynthesis 2 Rifampicin RNA polymerase 0.25 Ciproﬂoxacin Gyrase 2 Doxorubicin DNA intercalation 1 Vancomycin Lipid II 0.75 Bacitracin Lipid II 0.37 Ampicillin PBP 2

Table 7.1: Impact of Min-1 on activity of known antibiotics

target does not rescue growth against Min-1. In addition, they treated the library with two subinhibitory Min-1 concentration, 1/4x and 1/8x MIC, at xylose concentrations of 0.01% and 0.05%. To determine the growth eﬀects of the knockdown combined with Min-1 treatment,

I calculated a log2 ratio of the treated and induced growth relative to the untreated and induced growth (Figure 7.4). In this calculation, a value of 0 represents no change in growth between the treated and untreated condition, a positive number represents an increase in growth in the presence of Min-1, while a negative value demonstrates growth suppression. I observe that both the xylose and Min-1 concentrations have significant impact on the growth of the strains in the library. Increasing the xylose concentration results in more changes in growth, as seen by a widening of the growh distribution between Figure 7.4A and Figure 7.4B. In addition, under both xylose concentrations, increasing the Min-1 concentration from 1/8x to 1/4x MIC resulted in growth impacts, again observed by the widening of the distribution between the two concentrations. I selected 14 strains in the collection that were most sensitive to Min-1 treatment and ordered them from the Bacillus Genetic Stock Centre (BGSC) to test and confirm the sensitivity results (Table 7.2). In order to confirm the results and determine if any of these were specific targets of Min-1 and the cause of the short cell phenotype, I performed checkerboard assays with increasing Min-1 and xylose concentrations and sent the strains to Prof. Petra Levin’s lab to test for impacts on cell length. Despite the significant impact seen in the library experiment when pinned to agar containing Min-1, I could not reproduce the observed sensitivity for any of the strains in my liquid Chapter 7. Appendix 1 95

Figure 7.4: Impact of Min-1 on growth of a B. subtilis CRISPRi essential knock-down library Density plot showing the growth impact of Min-1 treatment versus untreated cultures at 0.01% (A.) and 0.05% (B.) xylose induction. C. Cell length eﬀects of CRISPRi at 0.01% (A.) and 0.05% (B.) xylose induction. Chapter 7. Appendix 1 96

Gene BGSC ID Product folC BEC28080 Dihydrofolate synthase ftsL BEC15150 Cell division protein FtsL gatC BEC06670 Glutamyl-tRNA(Gln) amidotransferase subunit C hisS BEC27560 Histidine–tRNA ligase holA BEC25560 DNA polymerase III subunit delta ileS BEC15430 Isoleucine–tRNA ligase mreD BEC28010 Rod shape-determining protein MreD murE BEC15180 UDP-N-acetylmuramoyl-L-alanyl-D- glutamate–2,6-diaminopimelate ligase prs BEC00510 Ribose-phosphate pyrophosphokinase rnpA BEC41050 Ribonuclease P protein component rplE BEC01280 50S ribosomal protein L5 rplJ BEC01040 50S ribosomal protein L10 rpsR BEC40890 30S ribosomal protein S18 yerQ BEC06720 Diacylglycerol kinase

Table 7.2: B. subtilis CRISPRi hits on xylose induction with Min-1 treatment checkerboard assays. In addition, Stephen Vadia, a post-doctoral fellow working with Prof. Petra Levin, did not observe any significant cell length reduction when the strains were induced with either 0.01% or 0.05% xylose (Figure 7.4C). This inability to reproduce the growth sensitivity or observe any significant morphology effects led me to drop this methodology for investigating Min-1’s mechanism of action.

7.1.6 Impact of Min-1 on HEK293 viability

To investigate the impact of Min-1 on eukaryotic cell viability I gave some of the molecule to Lauren Whyte, a post-doctoral fellow in the lab of Prof. Ruth Ross. Lauren tested the molecule on HEK293 cells using resazurin ﬂuorescence as a proxy for cell viability (Figure 7.5). She treated HEK293 cells with Min-1 up to 50 µM and calculated percent cell viability at 24, 48, and 72 hours. She observed that after 24 hours, the cell population remained stable with 80% of cells remainng viable. After 48 hours, this was reduced to around 50% at 50 µM Min-1 and that after 72 hours viability dropped to almost 0 at 50 µM Min-1. This shows that Min-1 signiﬁcantly reduces HEK293 viability at concentrations greater than the MIC after 72 hours. Chapter 7. Appendix 1 97

120

100

80 24 hrs 60 48 hrs 40 72 hrs

20 Cell Viability (% control) 0 0 1 5 10 15 20 25 30 40 50 Min-1 Concentraon (µM)

Figure 7.5: Impact of Min-1 on HEK293 viability Chapter 8

Appendix 2: Discovery of a novel DNA gyrase-targeting antibiotic through the chemical perturbation of Streptomyces venezuelae sporulation

The results and discussion presented in this chapter are supplemental to Chapter 4: Discovery of a novel DNA gyrase-targeting antibiotic through the chemical perturbation of Streptomyces venezuelae sporulation

8.1 Results and Discussion

8.1.1 Growth inhibition and reporter activation of S. venezuelae sporulation screen hits

Following the screen against S. venezuelae, I wanted to determine the activity of the hits against other model organisms and pathogens. The results of these initial MIC assays are given in Table 8.2. First, CB-1 and EN-1 were the only molecules with activity against the Gram-negative E. coli, while AS-5 has no activity against any of the bacteria that were tested. A number of the molecules do show signiﬁcant activity, such as EN-4, EN-7, AS-3, and AS-4, against a wide range of bacteria including antibiotic resistant S. aureus and VRE.

98 Chapter 8. Appendix 2 99

As with Min-1, I tested the S. venezuelae sporulation screen hits against a number of diﬀerent B. subtilis reporter strains. Speciﬁcally, I used the DNA damage sensing

PdinC -lacZ and the cell wall activity sensing PliaI -lacZ and PσW-lacZ. Testing each of the molecules against these strains resulted in only one hit. EN-7 was active against PdinC - lacZ as observed by a coloured halo around the zone of inhibition. The EN-7 response was quite weak due to its relative inactivity against B. subtilis; however, this was the ﬁrst indication that EN-7 acted against a DNA-related target. Chapter 8. Appendix 2 100 lacZ - W σ P lacZ - liaI P lacZ - dinC P 3232 ------32 - - - 32 - - - 32 - - - 3232 - + - - - - 32 - - - > > > > > > > > E. coli sporulation screen hits 100 > VRE ATCC 51299 S. venezuelae 100 > E. faecalis 100 > ATCC BAA-41 Growth inhibition of Table 8.1: 100 50 100 0.6 1.3 2.5 1.25 > > B. subtilis S. aureus S. aureus AS-5 AS-4 2.5 2.5 2.5 5 3.1 AS-3 0.9 1.25 10 10 15 CB-6 3.0 5 2.5 5 2.5 EN-7 CB-1CB-4 2.7 20 5.0 12.5 10 20 100 100 100 100 16 - - - EN-4 1.0 1.3 1.3 3.1 2.5 Molecule EN-1EN-3 8.2 31 5 20 5 31 50 80 50 80 16 - - - Chapter 8. Appendix 2 101

8.1.2 Aggregation activity of S. venezuelae sporulation screen hits

One common issue with synthetic screening hits is their solubility. Small molecule aggregation can cause non-speciﬁc growth inhibition and I wanted to determine if this was the cause of the bioactivity seen in the previous section. In order to determine the aggregation activity of the molecules I used dynamic light scattering, which measures the amount of forward scattering induced by a molecule in solution. Increased scattering suggests an increase in molecule aggregation. Screening each hit molecule to a maximum of 20 µM uncovered a few aggregating molecules. Speciﬁcally, EN-4, CB-6, and AS-4 (Figure 8.1A), resulted in the strongest aggregation activity at the concentrations tested whereas the other screening hits did not (Figure 8.1B).

Figure 8.1: Aggregation activity of S. venezuelae sporulation screen hits A. Screening hits that demonstrate aggregation activity. B. Screening hits that do not demonstrate aggregation activity

8.1.3 Gyrase S. aureus antisense strains show resistance to EN- 7

To follow-up on EN-7 activating the PdinC -lacZ reporter, I tested the molecule against a series of nine S. aureus strains with xylose inducible antisense expression [287]. While our lab has the entire library of 236 strains I focused on strains containing antisense Chapter 8. Appendix 2 102

Strain Gene Description Xylose induction (mM) AS-007 gyrA DNA gyrase subunit B 6.6 AS-009 parC Topoisomerase IV subunit A 7.5 AS-026 parE Topoisomerase IV subunit B 17 AS-045 dnaA Chromosomal replication initiator protein 5.8 AS-046 dnaC Replicative DNA helicase 6.7 AS-307 gyrB DNA gyrase B subunit 17 AS-311 topA DNA topoisomerase I 6.6 AS-314 dnaG DNA primase 4.8 AS-320 polA DNA polymerase I 8.8

Table 8.2: Growth inhibition of S. venezuelae sporulation screen hits

expression for DNA-related genes (Table 8.2). If EN-7 was activating the SOS response and targeting a DNA-related gene, I hoped to see either growth sensitivity or resistance with reduced expression of the antisense target gene.

First, I confirmed that the antisense expression was functional by culturing the strains in increasing xylose concentrations (Figure 8.2A). I observed that each of the nine strains saw reduced growth with increased xylose concentration, suggesting that the antisense sequence expression was functional and impacting the growth of the strains. I then tested the impact of EN-7 on the growth of these strains with xylose induction at the concentrations reported in the initial library construction (Figure 8.2B)) [287]. Both the gyrA and gyrB strains showed a 2x resistance to EN-7 with an MIC of 1.6 µM versus the uninduced MIC of 0.8 µM, while the other strains showed no significant change in sensitivity to the molecule. To confirm this effect I performed checkerboard assays with increasing concentrations of both xylose and EN-7 and observed an MIC of 6.4 µM at 15 mM xylose for the gyrA strain and an MIC of 3.2 µM for the gyrB at 12.5 mM xylose. This shows that reduced gyrase expression results in EN-7 resistance.

This was the ﬁrst indication that EN-7 directly targets gyrase. Since gyrase is a

heterotetramer of GyrA2GyrB2, it was exciting to see that both subunits of this enzyme showed similar activity. The observed resistance suggests that EN-7’s activity creates a toxic gyrase that inhibits cell growth. However, since we do not yet have a full understanding of EN-7’s activity against gyrase, I am not certain how this is happening. However, this result helped signiﬁcantly in identifying gyrase as the principle target for EN-7. Chapter 8. Appendix 2 103

Figure 8.2: Gyrase S. aureus antisense strains show resistance to EN-7 A. Xylose induced antisense expression inhibits growth of selected strains. B. Reduced expression of gyrA and gyrB results in resistance to EN-7. C. Checkerboard assay with EN-7 and xylose conﬁrms reduced gyrB expression confers resistance to EN-7. D. Checkerboard assay with EN-7 and xylose conﬁrms reduced gyrA expression confers resistance to EN-7. Chapter 8. Appendix 2 104

8.1.4 EN-7 does not reduce S. aureus bacterial load in a skin mouse infection model

While inhibiting bacterial growth in a 96-well plate or an petri dish is an important first step, any molecule with aspirations to become an antibiotic needs to be able to reduce the bacterial load in an in vivo infection model. To test EN-7 in such an environment we collaboated with Prof. Brian Coombes and his research technician Aline Comyn. They used a superficial skin infection model that involves exposing and disupting the skin with repeated application and removal of tape to test three concentrations of EN-7 against S. aureus TCH1516 [213] [199] [155]. In this assay, EN-7 was dissolved in PBS and directly added to the infected skin. The mice were sacrificed following four days where the mouse recieved treatment every 12 hours and bacterial colony counts counted for various organs in addition to skin and blood. They observed that EN-7 treatment at the concentrations tested did not reduce the bacterial load relative to PBS control on the skin, spleed, liver, or blood (Figure 8.3A). They also found that treatment did not impact mouse weight relative to the PBS control (Figure 8.3B). This lack of activity could be caused by a number of factors. For example, the molecule may not be stable or may require a different formulation to function in an in vivo system or the concentration required in this system may be higher than the in vitro assays. Chapter 8. Appendix 2 105

Figure 8.3: EN-7 is not able to reduce the bacterial load in a S. aureus super- ﬁcial skin mouse infection model A. Colony counts from various mouse organs following treatment with buﬀer, 0.5X, 1X, and 10X MIC S. aureus TCH1516. B. Measurement of mouse weight during treatment relative to pre-infection weight. Bibliography

[1] A Abou-Eisha. Evaluation of cytogenetic and dna damage induced by the antibacterial drug, trimethoprim. Toxicology in Vitro, 20:601–607, August 2006.

[2] Sam Abraham, James Chin, Huub J M Brouwers, Bernadette Turner, Ren Zhang, and Toni A Chapman. Green ﬂuorescent protein-based biosensor to detect and quantify stress responses induced by dna-degrading colicins. Applied and environmental microbiology, 77:6691–6693, September 2011.

[3] David W Adams and Jeﬀ Errington. Bacterial cell division: assembly, maintenance and disassembly of the z ring. Nature Reviews Microbiology, 7:642–653, September 2009.

[4] Abram Aertsen and Chris W Michiels. Mrr instigates the sos response after high pressure stress in escherichia coli. Molecular microbiology, 58:1381–1391, December 2005.

[5] M Aeschbacher, C A Reinhardt, and G Zbinden. A rapid cell membrane permeability test using ﬂuorescent dyes and ﬂow cytometry. Cell Biology and Toxicology, 2:247–255, June 1986.

[6] Salman Ahmed, Arryn Craney, Sheila M Pimentel-Elardo, and Justin R Nodwell. A synthetic, species-speciﬁc activator of secondary metabolism and sporulation in streptomyces coelicolor. Chembiochem, 14:83–91, January 2013.

[7] Katie J Aldred, Sylvia A McPherson, Charles L Turnbough, Robert J Kerns, and Neil Osheroﬀ. Topoisomerase iv-quinolone interactions are mediated through a water-metal ion bridge: mechanistic basis of quinolone resistance. Nucleic Acids Research, 41:4628–4639, April 2013.

[8] Ali A Almaqwashi, Thayaparan Paramanathan, Ioulia Rouzina, and Mark C Williams. Mechanisms of small molecule-dna interactions probed by single-molecule force spectroscopy. Nucleic Acids Research, 44:3971–3988, May 2016.

[9] David E Anderson, Michelle B Kim, Jared T Moore, Terrence E O’Brien, No- hemy A Sorto, Charles I Grove, Laura L Lackner, James B Ames, and Jared T Shaw. Comparison of small molecule inhibitors of the bacterial cell division protein ftsz and identiﬁcation of a reliable cross-species inhibitor. ACS Chemical Biology, 7:1918–1928, November 2012.

106 Bibliography 107

[10] Nora Au, Elke Kuester-Schoeck, Veena Mandava, Laura E Bothwell, Susan P Canny, Karen Chachu, Sierra A Colavito, Shakierah N Fuller, Eli S Groban, Laura A Hensley, Theresa C O’Brien, Amish Shah, Jessica T Tierney, Louise L Tomm, Thomas M O’Gara, Alexi I Goranov, Alan D Grossman, and Charles M Lovett. Genetic composition of the bacillus subtilis sos system. Journal of Bacte- riology, 187:7655–7666, November 2005.

[11] George K Auer and Douglas B Weibel. Bacterial cell mechanics. Biochemistry, 56:3710–3724, July 2017.

[12] Tomoya Baba, Takeshi Ara, Miki Hasegawa, Yuki Takai, Yoshiko Okumura, Miki Baba, Kirill A Datsenko, Masaru Tomita, Barry L Wanner, and Hirotada Mori. Construction of escherichia coli k-12 in-frame, single-gene knockout mutants: the keio collection. Molecular Systems Biology, 2:2006.0008, 2006.

[13] Jonathan Baell and Michael A Walters. Chemistry: Chemical con artists foil drug discovery. Nature, 513:481–483, September 2014.

[14] Jonathan B Baell. Feeling nature’s pains: Natural products, natural product drugs, and pan assay interference compounds (pains). Journal of Natural Prod- ucts, 79:616–628, March 2016.

[15] Jonathan B Baell and Georgina A Holloway. New substructure ﬁlters for removal of pan assay interference compounds (pains) from screening libraries and for their exclusion in bioassays. Journal of Medicinal Chemistry, 53:2719–2740, April 2010.

[16] Jonathan B Baell and J Willem M Nissink. Seven year itch: Pan-assay interference compounds (pains) in 2017-utility and limitations. ACS Chemical Biology, 13:36– 44, January 2018.

[17] Zeynep Baharoglu, Evelyne Krin, and Didier Mazel. Rpos plays a central role in the sos induction by sub-lethal aminoglycoside concentrations in vibrio cholerae. PLOS Genetics, 9:e1003421, 2013.

[18] Zeynep Baharoglu and Didier Mazel. Vibrio cholerae triggers sos and mutagen- esis in response to a wide range of antibiotics: a route towards multiresistance. Antimicrobial Agents and Chemotherapy, 55:2438–2441, May 2011.

[19] Zeynep Baharoglu and Didier Mazel. Sos, the formidable strategy of bacteria against aggressions. FEMS Microbiology Reviews, 38:1126–1145, November 2014.

[20] J C Barna and D H Williams. The structure and mode of action of glycopeptide antibiotics of the vancomycin group. Annual Review of Microbiology, 38:339–357, 1984.

[21] Benjamin D Bax, Pan F Chan, Drake S Eggleston, Andrew Fosberry, Daniel R Gentry, Fabrice Gorrec, Ilaria Giordano, Michael M Hann, Alan Hennessy, Martin Hibbs, Jianzhong Huang, Emma Jones, Jo Jones, Kristin Koretke Brown, Ceri J Bibliography 108

Lewis, Earl W May, Martin R Saunders, Onkar Singh, Claus E Spitzfaden, Carol Shen, Anthony Shillings, Andrew J Theobald, Alexandre Wohlkonig, Neil D Pear- son, and Michael N Gwynn. Type iia topoisomerase inhibition by a new class of antibacterial agents. Nature, 466:935–940, August 2010.

[22] Michael Baym, Laura K Stone, and Roy Kishony. Multidrug evolutionary strategies to reverse antibiotic resistance. Science, 351:aad3292, January 2016.

[23] G. J. Bean, S. T. Flickinger, W. M. Westler, M. E. McCully, D. Sept, D. B. Weibel, and K. J. Amann. A22 disrupts the bacterial actin cytoskeleton by directly binding and inducing a low-aﬃnity state in MreB. Biochemistry, 48(22):4852–4857, jun 2009.

[24] P Bernard and M Couturier. Cell killing by the f plasmid ccdb protein involves poisoning of dna-topoisomerase ii complexes. Journal of Molecular Biology, 226:735– 745, August 1992.

[25] E Bi and J Lutkenhaus. Cell division inhibitors sula and mincd prevent formation of the ftsz ring. Journal of Bacteriology, 175:1118–1125, February 1993.

[26] E F Bi and J Lutkenhaus. Ftsz ring structure associated with division in escherichia coli. Nature, 354:161–164, November 1991.

[27] A A Bianchi and F Baneyx. Stress responses as a tool to detect and characterize the mode of action of antibacterial agents. Applied and Environmental Microbiology, 65:5023–5027, November 1999.

[28] M Bibb. 1995 colworth prize lecture. the regulation of antibiotic production in streptomyces coelicolor a3(2). Microbiology, 142 ( Pt 6):1335–1344, June 1996.

[29] D J Biedenbach, S K Bouchillon, M Hackel, L A Miller, N E Scangarella-Oman, C Jakielaszek, and D F Sahm. In vitro activity of gepotidacin, a novel triazaacenaphthylene bacterial topoisomerase inhibitor, against a broad spectrum of bacterial pathogens. Antimicrobial Agents and Chemotherapy, 60:1918–1923, January 2016.

[30] K Biswas and S Mukerjee. Studies on the mechanism of action of polymyxin b on cholera vibrios. Proceedings of the Society for Experimental Biology and Medicine, 126:103–108, October 1967.

[31] Michael T Black, ThérèseStachyra, Denis Platel, Anne-Marie Girard, Monique Claudon, Jean-Michel Bruneau, and Christine Miossec. Mechanism of action of the antibiotic nxl101, a novel nonfluoroquinolone inhibitor of bacterial type ii topoisomerases. Antimicrobial Agents and Chemotherapy, 52:3339–3349, September 2008.

[32] Jessica M A Blair and Laura J V Piddock. Structure, function and inhibition of rnd eﬄux pumps in gram-negative bacteria: an update. Current Opinion in Microbiology, 12:512–519, October 2009. Bibliography 109

[33] F Blanche, B Cameron, F X Bernard, L Maton, B Manse, L Ferrero, N Ratet, C Lecoq, A Goniot, D Bisch, and J Crouzet. Diﬀerential behaviors of staphylococcus aureus and escherichia coli type ii dna topoisomerases. Antimicrobial Agents and Chemotherapy, 40:2714–2720, December 1996.

[34] Martin S Bojer, Katarzyna Wacnik, Peter Kjelgaard, Clement Gallay, Amy L Bot- tomley, Marianne T Cohn, Gunnar Lindahl, Dorte Frees, Jan-Willem Veening, Si- mon J Foster, et al. Sosa inhibits cell division in staphylococcus aureus in response to dna damage. bioRxiv, page 364299, 2018.

[35] G A Botta and J T Park. Evidence for involvement of penicillin-binding protein 3 in murein synthesis during septation but not during cell elongation. Journal of Bacteriology, 145:333–340, January 1981.

[36] Karen Bush. A resurgence of -lactamase inhibitor combinations eﬀective against multidrug-resistant gram-negative pathogens. International journal of antimicrobial agents, 46:483–493, November 2015.

[37] Natassja G Bush, Katherine Evans-Roberts, and Anthony Maxwell. Dna topoisomerases. EcoSal Plus, 6, 2015.

[38] Jackson A Buss, Vadim Baidin, Michael A Welsh, Josu´eFlores-Kim, Hongbaek Cho, B McKay Wood, Tsuyoshi Uehara, Suzanne Walker, Daniel Kahne, and Thomas G Bernhardt. A pathway-directed screen for inhibitors of the bacterial cell elongation machinery. Antimicrobial Agents and Chemotherapy, October 2018.

[39] Mark J Buttner, Martin Sch¨afer,David M Lawson, and Anthony Maxwell. Struc- tural insights into simocyclinone as an antibiotic, eﬀector ligand and substrate. FEMS Microbiology Reviews, 42, January 2018.

[40] J´anosB´erdy. Thoughts and facts about antibiotics: where we are now and where we are heading. The Journal of Antibiotics, 65:385–395, August 2012.

[41] Juan C Caicedo, Sam Cooper, Florian Heigwer, Scott Warchal, Peng Qiu, Csaba Molnar, Aliaksei S Vasilevich, Joseph D Barry, Harmanjit Singh Bansal, Oren Kraus, Mathias Wawer, Lassi Paavolainen, Markus D Herrmann, Mohammad Ro- hban, Jane Hung, Holger Hennig, John Concannon, Ian Smith, Paul A Clemons, Shantanu Singh, Paul Rees, Peter Horvath, Roger G Linington, and Anne E Car- penter. Data-analysis strategies for image-based cell proﬁling. Nature Methods, 14:849–863, August 2017.

[42] Jennifer Campbell, Atul K Singh, John P Santa Maria, Younghoon Kim, Stephanie Brown, Jonathan G Swoboda, Eleftherios Mylonakis, Brian J Wilkinson, and Suzanne Walker. Synthetic lethal compound combinations reveal a fundamental connection between wall teichoic acid and peptidoglycan biosyntheses in staphylococcus aureus. ACS Chemical Biology, 6:106–116, January 2011. Bibliography 110

[43] Jennifer Campbell, Atul K Singh, Jonathan G Swoboda, Michael S Gilmore, Brian J Wilkinson, and Suzanne Walker. An antibiotic that inhibits a late step in wall teichoic acid biosynthesis induces the cell wall stress stimulon in staphylococcus aureus. Antimicrobial Agents and Chemotherapy, 56:1810–1820, April 2012.

[44] Manuel Campos, Ivan V Surovtsev, Setsu Kato, Ahmad Paintdakhi, Bruno Beltran, Sarah E Ebmeier, and Christine Jacobs-Wagner. A constant size extension drives bacterial cell size homeostasis. Cell, 159:1433–1446, December 2014.

[45] Min Cao, Tao Wang, Rick Ye, and John D Helmann. Antibiotics that inhibit cell wall biosynthesis induce expression of the bacillus subtilis sigma(w) and sigma(m) regulons. Molecular Microbiology, 45:1267–1276, September 2002.

[46] M J Casadaban and S N Cohen. Analysis of gene control signals by dna fusion and cloning in escherichia coli. Journal of Molecular Biology, 138:179–207, April 1980.

[47] Pan F Chan, Thomas Germe, Benjamin D Bax, Jianzhong Huang, Reema K Thalji, Eric Bacqu´e,Anna Checchia, Dongzhao Chen, Haifeng Cui, Xiao Ding, Karen Ingraham, Lynn McCloskey, Kaushik Raha, Velupillai Srikannathasan, Anthony Maxwell, and Robert A Stavenger. Thiophene antibacterials that allosterically stabilize dna-cleavage complexes with dna gyrase. Proceedings of the National Academy of Sciences of the United States of America, 114:E4492–E4500, May 2017.

[48] Pan F Chan, Velupillai Srikannathasan, Jianzhong Huang, Haifeng Cui, Andrew P Fosberry, Minghua Gu, Michael M Hann, Martin Hibbs, Paul Homes, Karen Ingra- ham, Jason Pizzollo, Carol Shen, Anthony J Shillings, Claus E Spitzfaden, Robert Tanner, Andrew J Theobald, Robert A Stavenger, Benjamin D Bax, and Michael N Gwynn. Structural basis of dna gyrase inhibition by antibacterial qpt-1, anticancer drug etoposide and moxiﬂoxacin. Nature Communications, 6:10048, December 2015.

[49] Keith F Chater. Recent advances in understanding streptomyces. F1000Research, 5:2795, 2016.

[50] Fran¸coiseChau, AgnèsLefort, Samira Benadda, Vincent Dubée,and Bruno Fantin. Flow cytometry as a tool to determine the effects of cell wall-active antibiotics on vancomycin-susceptible and -resistant enterococcus faecalis strains. Antimicrobial Agents and Chemotherapy, 55:395–398, January 2011.

[51] Ashwini Chauhan, Hava Lofton, Erin Maloney, Jacob Moore, Marek Fol, Murty V V S Madiraju, and Malini Rajagopalan. Interference of mycobacterium tuberculosis cell division by rv2719c, a cell wall hydrolase. Molecular Microbiology, 62:132–147, October 2006.

[52] Yaodong Chen, Sara L Milam, and Harold P Erickson. Sula inhibits assembly of ftsz by a simple sequestration mechanism. Biochemistry, 51:3100–3109, April 2012. Bibliography 111

[53] An-Chun Chien, Shannon Kian Gharabiklou Zareh, Yan Mei Wang, and Petra Anne Levin. Changes in the oligomerization potential of the division inhibitor ugtp coordinate bacillus subtilis cell size with nutrient availability. Molecular Microbiology, 86:594–610, November 2012.

[54] Ryan T Cirz, Marcus B Jones, Neill A Gingles, Timothy D Minogue, Behnam Jarrahi, Scott N Peterson, and Floyd E Romesberg. Complete and sos-mediated response of staphylococcus aureus to the antibiotic ciproﬂoxacin. Journal of Bac- teriology, 189:531–539, January 2007.

[55] Ryan T Cirz, Bryan M O’Neill, Jennifer A Hammond, Steven R Head, and Floyd E Romesberg. Deﬁning the pseudomonas aeruginosa sos response and its role in the global response to the antibiotic ciproﬂoxacin. Journal of Bacteriology, 188:7101– 7110, October 2006.

[56] J Courcelle, A Khodursky, B Peter, P O Brown, and P C Hanawalt. Compara- tive gene expression proﬁles following uv exposure in wild-type and sos-deﬁcient escherichia coli. Genetics, 158:41–64, May 2001.

[57] Georgina Cox, Arthur Sieron, Andrew M King, Gianfranco De Pascale, Andrew C Pawlowski, Kalinka Koteva, and Gerard D Wright. A common platform for antibiotic dereplication and adjuvant discovery. Cell Chemical Biology, 24:98–109, January 2017.

[58] Arryn Craney, Cory Ozimok, Sheila Marie Pimentel-Elardo, Alfredo Capretta, and Justin R Nodwell. Chemical perturbation of secondary metabolism demonstrates important links to primary metabolism. Chemistry & Biology, 19:1020–1027, Au- gust 2012.

[59] Mark L Cunningham, Bryan P Kwan, Kirk J Nelson, Daniel C Bensen, and Karen J Shaw. Distinguishing on-target versus oﬀ-target activity in early antibacterial drug discovery using a macromolecular synthesis assay. Journal of Biomolecular Screen- ing, 18:1018–1026, October 2013.

[60] T L Czarny, A L Perri, S French, and E D Brown. Discovery of novel cell wall- active compounds using p ywac, a sensitive reporter of cell wall stress, in the model gram-positive bacterium bacillus subtilis. Antimicrobial Agents and Chemotherapy, 58:3261–3269, June 2014.

[61] Tomasz L Czarny and Eric D Brown. A small-molecule screening platform for the discovery of inhibitors of undecaprenyl diphosphate synthase. ACS Infectious Diseases, 2:489–499, July 2016.

[62] Jean-Philippe Cˆot´e,Shawn French, Sebastian S Gehrke, Craig R MacNair, Chand S Mangat, Amrita Bharat, and Eric D Brown. The genome-wide interaction network of nutrient stress genes in escherichia coli. mBio, 7, November 2016. Bibliography 112

[63] Martin Daniel-Ivad, Nabeela Hameed, Stephanie Tan, Rachna Dhanjal, Daniel Socko, Patricia Pak, Tomas Gverzdys, Marie A Elliot, and Justin R Nodwell. An engineered allele of afsq1 facilitates the discovery and investigation of cryptic natural products. ACS Chemical Biology, 12:628–634, March 2017.

[64] K A Datsenko and B L Wanner. One-step inactivation of chromosomal genes in escherichia coli k-12 using pcr products. Proceedings of the National Academy of Sciences of the United States of America, 97:6640–6645, June 2000.

[65] N K Davis and K F Chater. Spore colour in streptomyces coelicolor a3(2) involves the developmentally regulated synthesis of a compound biosynthetically related to polyketide antibiotics. Molecular Microbiology, 4:1679–1691, October 1990.

[66] Vanessa M D’Costa, Christine E King, Lindsay Kalan, Mariya Morar, Wilson W L Sung, Carsten Schwarz, Duane Froese, Grant Zazula, Fabrice Calmels, Regis De- bruyne, G Brian Golding, Hendrik N Poinar, and Gerard D Wright. Antibiotic resistance is ancient. Nature, 477:457–461, August 2011.

[67] Véroniquede Berardinis, David Vallenet, Vanina Castelli, Marielle Besnard, Agnès Pinet, Corinne Cruaud, Sumitta Samair, Christophe Lechaplais, Gabor Gya- pay, CélineRichez, Maxime Durot, Annett Kreimeyer, Fran¸coisLe Fèvre,Vin- cent Schächter, ValériePezo, Volker Döring,Claude Scarpelli, Claudine Médigue, Georges N Cohen, Philippe Marlière,Marcel Salanoubat, and Jean Weissenbach. A complete collection of single-gene deletion mutants of acinetobacter baylyi adp1. Molecular Systems Biology, 4:174, 2008.

[68] G De Pascale, C Grigoriadou, D Losi, I Ciciliato, M Sosio, and S Donadio. Valida- tion for high-throughput screening of a vanrs-based reporter gene assay for bacterial cell wall inhibitors. Journal of Applied Microbiology, 103:133–140, July 2007.

[69] Maxime Deforet, Dave van Ditmarsch, and Jo˜aoB Xavier. Cell-size homeostasis and the incremental rule in a bacterial pathogen. Biophysical Journal, 109:521–528, August 2015.

[70] Michael A D’Elia, Kathryn E Millar, Terry J Beveridge, and Eric D Brown. Wall teichoic acid polymers are dispensable for cell viability in bacillus subtilis. Journal of Bacteriology, 188:8313–8316, December 2006.

[71] Michael A D’Elia, Kathryn E Millar, Amit P Bhavsar, Ana M Tomljenovic, Bernd Hutter, Christoph Schaab, Gabriel Moreno-Hagelsieb, and Eric D Brown. Probing teichoic acid genetics with bioactive molecules reveals new interactions among diverse processes in bacterial cell wall biogenesis. Chemistry & Biology, 16:548–556, May 2009.

[72] D M DeMarini and B K Lawrence. Prophage induction by dna topoisomerase ii poisons and reactive-oxygen species: role of dna breaks. Mutation Research, 267:1– 17, May 1992. Bibliography 113

[73] Joseph A DeVito, Jonathan A Mills, Veronica G Liu, Anjana Agarwal, Christine F Sizemore, Zhongjie Yao, Daniel M Stoughton, Maria Grazia Cappiello, Maria D F S Barbosa, Lorie A Foster, and David L Pompliano. An array of target-speciﬁc screening strains for antibacterial discovery. Nature Biotechnology, 20:478–483, May 2002.

[74] Robert G K Donald, Stephen Skwish, R Allyn Forsyth, Jennifer W Anderson, Tanya Zhong, Colleen Burns, Suzy Lee, Xin Meng, Lynn LoCastro, Lisa Wang Jarantow, Jesus Martin, Sang Ho Lee, Ian Taylor, David Robbins, Cheryl Malone, Liangsu Wang, Carlos S Zamudio, Philip J Youngman, and John W Phillips. A staphylococcus aureus ﬁtness test platform for mechanism-based proﬁling of antibacterial compounds. Chemistry & Biology, 16:826–836, August 2009.

[75] Sarah M Drawz and Robert A Bonomo. Three decades of beta-lactamase inhibitors. Clinical Microbiology Reviews, 23:160–201, January 2010.

[76] K Drlica and X Zhao. Dna gyrase, topoisomerase iv, and the 4-quinolones. Micro- biology and Molecular Biology Reviews, 61:377–392, September 1997.

[77] Ann E Eakin, Oluyinka Green, Neil Hales, Grant K Walkup, Shanta Bist, Alok Singh, George Mullen, Joanna Bryant, Kevin Embrey, Ning Gao, Alex Breeze, Dave Timms, Beth Andrews, Maria Uria-Nickelsen, Julie Demeritt, James T Loch, Ken Hull, April Blodgett, Ruth N Illingworth, Bryan Prince, P Ann Boriack-Sjodin, Sheila Hauck, Lawrence J MacPherson, Haihong Ni, and Brian Sherer. Pyrrolamide dna gyrase inhibitors: fragment-based nuclear magnetic resonance screening to identify antibacterial agents. Antimicrobial Agents and Chemotherapy, 56:1240– 1246, March 2012.

[78] Linda Ejim, Maya A Farha, Shannon B Falconer, Jan Wildenhain, Brian K Coombes, Mike Tyers, Eric D Brown, and Gerard D Wright. Combinations of antibiotics and nonantibiotic drugs enhance antimicrobial eﬃcacy. Nature Chemi- cal Biology, 7:348–350, June 2011.

[79] S A El-Mowafi, E Sineva, J N Alumasa, H Nicoloff, J W Tomsho, S E Ades, and K C Keiler. Identification of inhibitors of a bacterial sigma factor using a new high- throughput screening assay. Antimicrobial Agents and Chemotherapy, 59:193–205, January 2015.

[80] Sara S El Zahed and Eric D Brown. Chemical-chemical combinations map un- charted interactions in escherichia coli under nutrient stress. iScience, 2:168–181, April 2018.

[81] Sara S El Zahed, Garima Kumar, Madeline Tong, and Eric D Brown. Nutrient stress small-molecule screening platform for escherichia coli. Methods in Molecular Biology, 1787:1–18, 2018. Bibliography 114

[82] Raquel F Epand, Jake E Pollard, Jonathan O Wright, Paul B Savage, and Richard M Epand. Depolarization, bacterial membrane composition, and the antimicrobial action of ceragenins. Antimicrobial Agents and Chemotherapy, 54:3708– 3713, September 2010.

[83] Ivan Erill, Susana Campoy, and Jordi Barb´e. Aeons of distress: an evolutionary perspective on the bacterial sos response. FEMS Microbiology Reviews, 31:637–656, November 2007.

[84] Jeﬀ Errington. From spores to antibiotics via the cell cycle. Microbiology, 156:1–13, January 2010.

[85] Ye-Jin Eun, Marie H Foss, Daniela Kiekebusch, Daniel A Pauw, William M West- ler, Martin Thanbichler, and Douglas B Weibel. Dcap: a broad-spectrum antibiotic that targets the cytoplasmic membrane of bacteria. Journal of the American Chem- ical Society, 134:11322–11325, July 2012.

[86] Shaun P Falk, James W Noah, and Bernard Weisblum. Screen for inducers of autolysis in bacillus subtilis. Antimicrobial Agents and Chemotherapy, 54:3723– 3729, September 2010.

[87] Jun Fan, Boudewijn L M de Jonge, Kathy MacCormack, Shubha Sriram, Robert E McLaughlin, Helen Plant, Marian Preston, Paul R Fleming, Robert Albert, Melinda Foulk, and Scott D Mills. A novel high-throughput cell-based assay aimed at identifying inhibitors of dna metabolism in bacteria. Antimicrobial Agents and Chemotherapy, 58:7264–7272, December 2014.

[88] Maya A Farha and Eric D Brown. Strategies for target identiﬁcation of antimicrobial natural products. Natural Product Reports, 33:668–680, May 2016.

[89] Maya A Farha, Tomasz L Czarny, Cullen L Myers, Liam J Worrall, Shawn French, Deborah G Conrady, Yang Wang, Eric Oldﬁeld, Natalie C J Strynadka, and Eric D Brown. Antagonism screen for inhibitors of bacterial cell wall biogenesis uncovers an inhibitor of undecaprenyl diphosphate synthase. Proceedings of the National Academy of Sciences of the United States of America, 112:11048–11053, September 2015.

[90] Maya A Farha, Chris P Verschoor, Dawn Bowdish, and Eric D Brown. Collapsing the proton motive force to identify synergistic combinations against staphylococcus aureus. Chemistry & Biology, 20:1168–1178, September 2013.

[91] Prabhavathi Fernandes and Evan Martens. Antibiotics in late clinical development. Biochemical Pharmacology, 133:152–163, June 2017.

[92] Paul D Fey, Jennifer L Endres, Vijaya Kumar Yajjala, Todd J Widhelm, Robert J Boissy, Jeﬀrey L Bose, and Kenneth W Bayles. A genetic resource for rapid and comprehensive phenotype screening of nonessential staphylococcus aureus genes. mBio, 4:e00537–e00512, February 2013. Bibliography 115

[93] Hans Peter Fischer, Nina A Brunner, Bernd Wieland, Jesse Paquette, Ludwig Macko, Karl Ziegelbauer, and Christoph Freiberg. Identiﬁcation of antibiotic stress- inducible promoters: a systematic approach to novel pathway-speciﬁc reporter assays for antibacterial drug discovery. Genome research, 14:90–98, January 2004.

[94] R K Flamm, D J Farrell, P R Rhomberg, N E Scangarella-Oman, and H S Sader. Gepotidacin (gsk2140944) activity against gram-positive and gram-negative bacteria. Antimicrobial Agents and Chemotherapy, 61, July 2017.

[95] Ruth H Flatman, Alison J Howells, Lutz Heide, Hans-Peter Fiedler, and Anthony Maxwell. Simocyclinone d8, an inhibitor of dna gyrase with a novel mode of action. Antimicrobial Agents and Chemotherapy, 49:1093–1100, March 2005.

[96] A Fleming. Classics in infectious diseases: on the antibacterial action of cultures of a penicillium, with special reference to their use in the isolation of b. inﬂuenzae by alexander ﬂeming, reprinted from the british journal of experimental pathology 10:226-236, 1929. Reviews of Infectious Diseases, 2:129–139, 1980.

[97] Alexander Fleming. Nobel lecture. 1945.

[98] B Fournier, X Zhao, T Lu, K Drlica, and D C Hooper. Selective targeting of topoisomerase iv and dna gyrase in staphylococcus aureus: diﬀerent patterns of quinolone- induced inhibition of dna synthesis. Antimicrobial Agents and Chemotherapy, 44:2160–2165, August 2000.

[99] Shawn French, Brittney E Coutts, and Eric D Brown. Open-source high-throughput phenomics of bacterial promoter-reporter strains. Cell Systems, 7:339–346.e3, September 2018.

[100] Shawn French, Michael J Ellis, Brittney E Coutts, and Eric D Brown. Chemical genomics reveals mechanistic hypotheses for uncharacterized bioactive molecules in bacteria. Current Opinion in Microbiology, 39:42–47, October 2017.

[101] Peng Gao, Yanli Wang, Iv´anVillanueva, Pak Leung Ho, Julian Davies, and Richard Yi Tsun Kao. Construction of a multiplex promoter reporter platform to monitor staphylococcus aureus virulence gene expression and the identiﬁcation of usnic acid as a potent suppressor of psm gene expression. Frontiers in Microbiology, 7:1344, 2016.

[102] Ethan C Garner, Remi Bernard, Wenqin Wang, Xiaowei Zhuang, David Z Rudner, and Tim Mitchison. Coupled, circumferential motions of the cell wall synthesis machinery and mreb ﬁlaments in b. subtilis. Science, 333:222–225, July 2011.

[103] Sebastian S Gehrke, Garima Kumar, Nicole A Yokubynas, Jean-Philippe Cˆot´e, Wenliang Wang, Shawn French, Craig R MacNair, Gerard D Wright, and Eric D Brown. Exploiting the sensitivity of nutrient transporter deletion strains in discovery of natural product antimetabolites. ACS Infectious Diseases, 3:955–965, December 2017. Bibliography 116

[104] Daniel R Gentry, Imogen Wilding, John M Johnson, Dongzhao Chen, Katja Rem- linger, Cindy Richards, Susan Neill, Magdalena Zalacain, Stephen F Rittenhouse, and Michael N Gwynn. A rapid microtiter plate assay for measuring the eﬀect of compounds on staphylococcus aureus membrane potential. Journal of Microbio- logical Methods, 83:254–256, November 2010.

[105] Thomas Germe, Judit Vörös, Frederic Jeannot, Thomas Taillier, Robert A Stavenger, Eric Bacqué,Anthony Maxwell, and Benjamin D Bax. A new class of antibacterials, the imidazopyrazinones, reveal structural transitions involved in dna gyrase poisoning and mechanisms of resistance. Nucleic Acids Research, 46:4114– 4128, May 2018.

[106] Chandradhish Ghosh and Jayanta Haldar. Membrane-active small molecules: De- signs inspired by antimicrobial peptides. ChemMedChem, 10:1606–1624, October 2015.

[107] R Gleckman, N Blagg, and D W Joubert. Trimethoprim: mechanisms of action, antimicrobial activity, bacterial resistance, pharmacokinetics, adverse reactions, and therapeutic indications. Pharmacotherapy, 1:14–20, 1981.

[108] Diego Gonzalez and Justine Collier. Eﬀects of (p)ppgpp on the progression of the cell cycle of caulobacter crescentus. Journal of Bacteriology, 196:2514–2525, July 2014.

[109] Kuldeepkumar Ramnaresh Gupta, Priyanka Baloni, Shantinath S Indi, and Di- pankar Chatterji. Regulation of growth, cell shape, cell division, and gene expression by second messengers (p)ppgpp and cyclic di-gmp in mycobacterium smegmatis. Journal of Bacteriology, 198:1414–1422, May 2016.

[110] Arnaud Gutierrez, Jonathan M Stokes, and Ivan Matic. Our evolving understanding of the mechanism of quinolones. Antibiotics, 7, April 2018.

[111] Krupanandan Haranahalli, Simon Tong, and Iwao Ojima. Recent advances in the discovery and development of antibacterial agents targeting the cell-division protein ftsz. Bioorganic & Medicinal Chemistry, 24:6354–6369, December 2016.

[112] Leigh K Harris and Julie A Theriot. Relative rates of surface and volume synthesis set bacterial cell size. Cell, 165:1479–1492, June 2016.

[113] Vasili Hauryliuk, Gemma C Atkinson, Katsuhiko S Murakami, Tanel Tenson, and Kenn Gerdes. Recent functional insights into the role of (p)ppgpp in bacterial physiology. Nature Reviews Microbiology, 13:298–309, May 2015.

[114] Masahito Hayashi and Yoshie Harada. Direct observation of the reversible unwinding of a single dna molecule caused by the intercalation of ethidium bromide. Nucleic Acids Research, 35:e125, 2007. Bibliography 117

[115] David J Haydon, Neil R Stokes, Rebecca Ure, Greta Galbraith, James M Bennett, David R Brown, Patrick J Baker, Vladimir V Barynin, David W Rice, Sveta E Sedelnikova, Jonathan R Heal, Joseph M Sheridan, Sachin T Aiwale, Pramod K Chauhan, Anil Srivastava, Amit Taneja, Ian Collins, Jeff Errington, and Lloyd G Czaplewski. An inhibitor of ftsz with potent and selective anti-staphylococcal activity. Science, 321:1673–1675, September 2008. [116] S M Hecht. Bleomycin: new perspectives on the mechanism of action. Journal of Natural Products, 63:158–168, January 2000. [117] Kiara Held, Elizabeth Ramage, Michael Jacobs, Larry Gallagher, and Colin Manoil. Sequence-verified two-allele transposon mutant library for pseudomonas aeruginosa pao1. Journal of Bacteriology, 194:6387–6389, December 2012. [118] John D Helmann. Bacillus subtilis extracytoplasmic function (ecf) sigma factors and defense of the cell envelope. Current Opinion in Microbiology, 30:122–132, April 2016. [119] Anne Helmrich, Monica Ballarino, Evgeny Nudler, and Laszlo Tora. Transcription- replication encounters, consequences and genomic instability. Nature Structural & Molecular Biology, 20:412–418, April 2013. [120] D Hendlin, E O Stapley, M Jackson, H Wallick, A K Miller, F J Wolf, T W Miller, L Chaiet, F M Kahan, E L Foltz, H B Woodruff, J M Mata, S Hernandez, and S Mochales. Phosphonomycin, a new antibiotic produced by strains of streptomyces. Science, 166:122–123, October 1969. [121] Aram Heo, Hyun-Jin Jang, Jung-Suk Sung, and Woojun Park. Global transcrip- tome and physiological responses of acinetobacter oleivorans dr1 exposed to distinct classes of antibiotics. PLOS One, 9:e110215, 2014. [122] Andrew Hesketh, Wenqiong Joan Chen, Jamie Ryding, Sherman Chang, and Mervyn Bibb. The global role of ppgpp synthesis in morphological differentia- tion and antibiotic production in streptomyces coelicolor a3(2). Genome Biology, 8:R161, 2007. [123] Norbert S Hill, Paul J Buske, Yue Shi, and Petra Anne Levin. A moonlighting enzyme links escherichia coli cell size with central metabolism. PLOS Genetics, 9:e1003663, 2013. [124] David C Hooper and George A Jacoby. Topoisomerase inhibitors: Fluoroquinolone mechanisms of action and resistance. Cold Spring Harbor perspectives in medicine, 6, September 2016. [125] Jiaul Hoque, Mohini M Konai, Spandhana Gonuguntla, Goutham B Manjunath, Sandip Samaddar, Venkateswarlu Yarlagadda, and Jayanta Haldar. Membrane active small molecules show selective broad spectrum antibacterial activity with no detectable resistance and eradicate biofilms. Journal of Medicinal Chemistry, 58:5486–5500, July 2015. Bibliography 118

[126] Bradley M Hover, Seong-Hwan Kim, Micah Katz, Zachary Charlop-Powers, Jeremy G Owen, Melinda A Ternei, Jeffrey Maniko, Andreia B Estrela, Henrik Molina, Steven Park, David S Perlin, and Sean F Brady. Culture-independent discovery of the malacidins as calcium-dependent antibiotics with activity against multidrug-resistant gram-positive pathogens. Nature Microbiology, 3:415–422, April 2018. [127] John A Howe, Hao Wang, Thierry O Fischmann, Carl J Balibar, Li Xiao, An- drew M Galgoci, Juliana C Malinverni, Todd Mayhood, Artjohn Villafania, Ali Nahvi, Nicholas Murgolo, Christopher M Barbieri, Paul A Mann, Donna Carr, Ellen Xia, Paul Zuck, Dan Riley, Ronald E Painter, Scott S Walker, Brad Sher- borne, Reynalda de Jesus, Weidong Pan, Michael A Plotkin, Jin Wu, Diane Rind- gen, John Cummings, Charles G Garlisi, Rumin Zhang, Payal R Sheth, Charles J Gill, Haifeng Tang, and Terry Roemer. Selective small-molecule inhibition of an rna structural element. Nature, 526:672–677, October 2015. [128] Kerwyn Casey Huang, Ranjan Mukhopadhyay, Bingni Wen, Zemer Gitai, and Ned S Wingreen. Cell shape and cell-wall organization in gram-negative bacteria. Proceedings of the National Academy of Sciences of the United States of America, 105:19282–19287, December 2008. [129] X Huang, K L Fredrick, and J D Helmann. Promoter recognition by bacillus subtilis sigmaw: autoregulation and partial overlap with the sigmax regulon. Journal of Bacteriology, 180:3765–3770, August 1998. [130] Joann Huber, Robert G K Donald, Sang Ho Lee, Lisa Wang Jarantow, Michael J Salvatore, Xin Meng, Ronald Painter, Russell H Onishi, James Occi, Karen Dorso, Katherine Young, Young Whan Park, Stephen Skwish, Michael J Szymonifka, Tim S Waddell, Lynn Miesel, John W Phillips, and Terry Roemer. Chemical genetic identification of peptidoglycan inhibitors potentiating carbapenem activity against methicillin-resistant staphylococcus aureus. Chemistry & Biology, 16:837– 848, August 2009. [131] O Huisman and R D’Ari. An inducible dna replication-cell division coupling mechanism in e. coli. Nature, 290:797–799, April 1981. [132] O Huisman, R D’Ari, and S Gottesman. Cell-division control in escherichia coli: specific induction of the sos function sfia protein is sufficient to block septation. Proceedings of the National Academy of Sciences of the United States of America, 81:4490–4494, July 1984. [133] Julian G Hurdle, Alex J O’Neill, Ian Chopra, and Richard E Lee. Targeting bacterial membrane function: an underexploited mechanism for treating persistent infections. Nature Reviews Microbiology, 9:62–75, January 2011. [134] Bernd Hutter, Christina Fischer, Alexander Jacobi, Christoph Schaab, and Hannes Loferer. Panel of bacillus subtilis reporter strains indicative of various modes of action. Antimicrobial Agents and Chemotherapy, 48:2588–2594, July 2004. Bibliography 119

[135] Alexander J Hyde, Judica¨elParisot, Adam McNichol, and Boyan B Bonev. Nisin- induced changes in bacillus morphology suggest a paradigm of antibiotic action. Proceedings of the National Academy of Sciences of the United States of America, 103:19896–19901, December 2006.

[136] Colin J Ingham, Maaike van den Ende, Peter C Wever, and Peter M Schneeberger. Rapid antibiotic sensitivity testing and trimethoprim-mediated ﬁlamentation of clinical isolates of the enterobacteriaceae assayed on a novel porous culture support. Journal of Medical Microbiology, 55:1511–1519, November 2006.

[137] K J Jaacks, J Healy, R Losick, and A D Grossman. Identiﬁcation and characterization of genes controlled by the sporulation-regulatory gene spo0h in bacillus subtilis. Journal of Bacteriology, 171:4121–4129, August 1989.

[138] Susanne Jacobsson, Daniel Golparian, Nicole Scangarella-Oman, and Magnus Un- emo. In vitro activity of the novel triazaacenaphthylene gepotidacin (gsk2140944) against mdr neisseria gonorrhoeae. The Journal of Antimicrobial Chemotherapy, 73:2072–2077, August 2018.

[139] Charul Jani, Elitza I Tocheva, Scott McAuley, Arryn Craney, Grant J Jensen, and Justin Nodwell. Streptomyces: a screening tool for bacterial cell division inhibitors. Journal of Biomolecular Screening, 20:275–284, February 2015.

[140] N R Jena. Dna damage by reactive species: Mechanisms, mutation and repair. Journal of Biosciences, 37:503–517, July 2012.

[141] Baofeng Jia, Amogelang R Raphenya, Brian Alcock, Nicholas Waglechner, Peiyao Guo, Kara K Tsang, Briony A Lago, Biren M Dave, Sheldon Pereira, Ar- jun N Sharma, Sachin Doshi, M´elanieCourtot, Raymond Lo, Laura E Williams, Jonathan G Frye, Tariq Elsayegh, Daim Sardar, Erin L Westman, Andrew C Pawlowski, Timothy A Johnson, Fiona S L Brinkman, Gerard D Wright, and Andrew G McArthur. Card 2017: expansion and model-centric curation of the comprehensive antibiotic resistance database. Nucleic Acids Research, 45:D566– D573, January 2017.

[142] Tatiana Johnston, Daria Van Tyne, Roy F Chen, Nicolas L Fawzi, Bumsup Kwon, Michael J Kelso, Michael S Gilmore, and Eleftherios Mylonakis. Propyl-5- hydroxy-3-methyl-1-phenyl-1h-pyrazole-4-carbodithioate (hmpc): a new bacteriostatic agent against methicillin-resistant staphylococcus aureus. Scientiﬁc Reports, 8:7062, May 2018.

[143] Sina Jordan, Anja Junker, John D Helmann, and Thorsten Mascher. Regulation of liars-dependent gene expression in bacillus subtilis: identiﬁcation of inhibitor proteins, regulator binding sites, and target genes of a conserved cell envelope stress-sensing two-component system. Journal of Bacteriology, 188:5153–5166, July 2006. Bibliography 120

[144] Leonard Katz and Richard H Baltz. Natural product discovery: past, present, and future. Journal of Industrial Microbiology & Biotechnology, 43:155–176, March 2016.

[145] Yoshikazu Kawai, Shigeki Moriya, and Naotake Ogasawara. Identiﬁcation of a protein, ynea, responsible for cell division suppression during the sos response in bacillus subtilis. Molecular Microbiology, 47:1113–1122, February 2003.

[146] A B Khodursky, E L Zechiedrich, and N R Cozzarelli. Topoisomerase iv is a target of quinolones in escherichia coli. Proceedings of the National Academy of Sciences of the United States of America, 92:11801–11805, December 1995.

[147] Andrew M King, Sarah A Reid-Yu, Wenliang Wang, Dustin T King, Gianfranco De Pascale, Natalie C Strynadka, Timothy R Walsh, Brian K Coombes, and Ger- ard D Wright. Aspergillomarasmine a overcomes metallo--lactamase antibiotic resistance. Nature, 510:503–506, June 2014.

[148] A S Klainer and R L Perkins. Antibiotic-induced alterations in the surface morphology of bacterial cells: a scanning-beam electron miscroscopy study. The Journal of Infectious Diseases, 122:323–328, October 1970.

[149] Carolin Martina Kobras, Thorsten Mascher, and Susanne Gebhard. Application of a bacillus subtilis whole-cell biosensor (pliai-lux) for the identiﬁcation of cell wall active antibacterial compounds. Methods in Molecular Biology, 1520:121–131, 2017.

[150] Byoung-Mo Koo, George Kritikos, Jeremiah D Farelli, Horia Todor, Kenneth Tong, Harvey Kimsey, Ilan Wapinski, Marco Galardini, Angelo Cabal, Jason M Peters, Anna-Barbara Hachmann, David Z Rudner, Karen N Allen, Athanasios Typas, and Carol A Gross. Construction and analysis of two genome-scale deletion libraries for bacillus subtilis. Cell Systems, 4:291–305.e7, March 2017.

[151] Oren Z Kraus and Brendan J Frey. Computer vision for high content screening. Critical Reviews in Biochemistry and Molecular Biology, 51:102–109, 2016.

[152] Oren Z Kraus, Ben T Grys, Jimmy Ba, Yolanda Chong, Brendan J Frey, Charles Boone, and Brenda J Andrews. Automated analysis of high-content microscopy data with deep learning. Molecular Systems Biology, 13:924, April 2017.

[153] Allison Kriel, Alycia N Bittner, Sok Ho Kim, Kuanqing Liu, Ashley K Tehranchi, Winnie Y Zou, Samantha Rendon, Rui Chen, Benjamin P Tu, and Jue D Wang. Direct regulation of gtp homeostasis by (p)ppgpp: a critical component of viability and stress resistance. Molecular Cell, 48:231–241, October 2012.

[154] Thomas Kruse, Jette Bork-Jensen, and Kenn Gerdes. The morphogenetic mrebcd proteins of escherichia coli form an essential membrane-bound complex. Molecular Microbiology, 55(1):78–89, 2005. Bibliography 121

[155] Elisabeth Kugelberg, Tobias Norstr¨om,Thomas K Petersen, Tore Duvold, Dan I Andersson, and Diarmaid Hughes. Establishment of a superﬁcial skin infection model in mice by using staphylococcus aureus and streptococcus pyogenes. An- timicrobial Agents and Chemotherapy, 49:3435–3441, August 2005.

[156] Christopher J Lacriola, Shaun P Falk, and Bernard Weisblum. Screen for agents that induce autolysis in bacillus subtilis. Antimicrobial Agents and Chemotherapy, 57:229–234, January 2013.

[157] Sushmita D Lahiri, Amy Kutschke, Kathy McCormack, and Richard A Alm. In- sights into the mechanism of inhibition of novel bacterial topoisomerase inhibitors from characterization of resistant mutants of staphylococcus aureus. Antimicrobial Agents and Chemotherapy, 59:5278–5287, September 2015.

[158] Anne Lamsa, Javier Lopez-Garrido, Diana Quach, Eammon P Riley, Joe Pogliano, and Kit Pogliano. Rapid inhibition proﬁling in bacillus subtilis to identify the mechanism of action of new antimicrobials. ACS Chemical Biology, 11:2222–2231, August 2016.

[159] Ivan Laponogov, Xiao-Su Pan, Dennis A Veselkov, Katherine E McAuley, L Mark Fisher, and Mark R Sanderson. Structural basis of gate-dna breakage and resealing by type ii topoisomerases. PLOS One, 5:e11338, June 2010.

[160] C S Lewin and S G Amyes. The role of the sos response in bacteria exposed to zidovudine or trimethoprim. Journal of Medical Microbiology, 34:329–332, June 1991.

[161] R J Lewis, O M Singh, C V Smith, T Skarzynski, A Maxwell, A J Wonacott, and D B Wigley. The nature of inhibition of dna gyrase by the coumarins and the cyclothialidines revealed by x-ray crystallography. The EMBO Journal, 15:1412– 1420, March 1996.

[162] B Lee Ligon. Penicillin: its discovery and early development. Seminars in Pediatric Infectious Siseases, 15:52–57, January 2004.

[163] Losee L Ling, Tanja Schneider, Aaron J Peoples, Amy L Spoering, Ina Engels, Brian P Conlon, Anna Mueller, Till F Sch¨aberle, Dallas E Hughes, Slava Epstein, Michael Jones, Linos Lazarides, Victoria A Steadman, Douglas R Cohen, Cintia R Felix, K Ashley Fetterman, William P Millett, Anthony G Nitti, Ashley M Zullo, Chao Chen, and Kim Lewis. A new antibiotic kills pathogens without detectable resistance. Nature, 517:455–459, January 2015.

[164] C A Lipinski, F Lombardo, B W Dominy, and P J Feeney. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Advanced Drug Delivery Reviews, 46:3–26, March 2001.

[165] Carl Llor, Ana Moragas, Carolina Bayona, Josep M Cots, JoséM Molero, Joana Ribas, Julio Francisco Fóthy, Isabel Gutiérrez,Coro Sánchez, JesúsOrtega, Javier Bibliography 122

Arranz, Jenifer Botanes, and PurificaciónRobles. The stop-ab trial protocol: ef- ficacy and safety of discontinuing patient antibiotic treatment when physicians no longer consider it necessary. BMJ Open, 7:e015814, June 2017.

[166] P E Love, M J Lyle, and R E Yasbin. Dna-damage-inducible (din) loci are tran- scriptionally activated in competent bacillus subtilis. Proceedings of the National Academy of Sciences of the United States of America, 82:6201–6205, September 1985.

[167] Christina E. L¨unseand G¨unter Mayer. Reporter Gene-Based Screening for TPP Riboswitch Activators, pages 227–235. Springer New York, New York, NY, 2017.

[168] J F Lutkenhaus. Coupling of dna replication and cell division: sulb is an allele of ftsz. Journal of Bacteriology, 154:1339–1346, June 1983.

[169] Qian Ma, Yoshitsugu Akiyama, Zhidong Xu, Kazuhide Konishi, and Sidney M Hecht. Identiﬁcation and cleavage site analysis of dna sequences bound strongly by bleomycin. Journal of the American Chemical Society, 131:2013–2022, February 2009.

[170] Craig R MacNair, Jonathan M Stokes, Shawn French, Cullen L Myers, Kali R Iyer, and Eric D Brown. A cell-based approach to characterize antimicrobial compounds through kinetic dose response. Bioorganic & Medicinal Chemistry, 24:6315–6319, December 2016.

[171] Elisa Maiques, Carles Ubeda, Susana Campoy, Noelia Salvador, Iñigo Lasa, Richard P Novick, Jordi Barbé,and JoséR Penadés. beta-lactam antibiotics induce the sos response and horizontal transfer of virulence factors in staphylococcus aureus. Journal of Bacteriology, 188:2726–2729, April 2006.

[172] Thorsten Mascher, Sara L Zimmer, Terry-Ann Smith, and John D Helmann. Antibiotic-inducible promoter regulated by the cell envelope stress-sensing two- component system liars of bacillus subtilis. Antimicrobial Agents and Chemother- apy, 48:2888–2896, August 2004.

[173] Andr´eMateus, Jacob Bobonis, Nils Kurzawa, Frank Stein, Dominic Helm, Jo- hannes Hevler, Athanasios Typas, and Mikhail M Savitski. Thermal proteome proﬁling in bacteria: probing protein state. Molecular Systems Biology, 14:e8242, July 2018.

[174] AndréMateus, Tomi A Määttä,and Mikhail M Savitski. Thermal proteome profiling: unbiased assessment of protein state through heat-induced stability changes. Proteome Science, 15:13, 2016.

[175] A Maxwell. The interaction between coumarin drugs and dna gyrase. Molecular Microbiology, 9:681–686, August 1993. Bibliography 123

[176] S McAuley, A Huynh, T L Czarny, E D Brown, and J R Nodwell. Membrane activity proﬁling of small molecule , javax.xml.bind.jaxbelement@7c5f723e, growth inhibitors utilizing novel duel-dye ﬂuorescence assay. MedChemComm, 9:554–561, March 2018.

[177] Joseph R McCormick and Klas Fl¨ardh.Signals and regulators that govern streptomyces development. FEMS Microbiology Reviews, 36:206–231, January 2012.

[178] Sarah M McLeod, Paul R Fleming, Kathleen MacCormack, Robert E McLaughlin, James D Whiteaker, Shin-Ichiro Narita, Makiko Mori, Hajime Tokuda, and Alita A Miller. Small-molecule inhibitors of gram-negative lipoprotein traﬃcking discovered by phenotypic screening. Journal of Bacteriology, 197:1075–1082, March 2015.

[179] Lili R Mesak, Shuhua Qi, Ivan Villanueva, Vivian Miao, and Julian Davies. Staphy- lococcus aureus promoter-lux reporters for drug discovery. The Journal of Antibi- otics, 63:492–498, August 2010.

[180] Lili Rosana Mesak, Vivian Miao, and Julian Davies. Eﬀects of subinhibitory concentrations of antibiotics on sos and dna repair gene expression in staphylococcus aureus. Antimicrobial Agents and Chemotherapy, 52:3394–3397, September 2008.

[181] Lili Rosana Mesak, Grace Yim, and Julian Davies. Improved lux reporters for use in staphylococcus aureus. Plasmid, 61:182–187, May 2009.

[182] Christine Miller, Hanne Ingmer, Line Elnif Thomsen, Kirsten Skarstad, and Stan- ley N Cohen. Dpia binding to the replication origin of escherichia coli plasmids and chromosomes destabilizes plasmid inheritance and induces the bacterial sos response. Journal of Bacteriology, 185:6025–6031, October 2003.

[183] Christine Miller, Line Elnif Thomsen, Carina Gaggero, Ronen Mosseri, Hanne In- gmer, and Stanley N Cohen. Sos response induction by beta-lactams and bacterial defense against antibiotic lethality. Science, 305:1629–1631, September 2004.

[184] P Mitchell. Chemiosmotic coupling in oxidative and photosynthetic phosphoryla- tion. Biological Reviews of the Cambridge Philosophical Society, 41:445–502, Au- gust 1966.

[185] Carlos Molina-Santiago, Abdelali Daddaoua, Mar´ıa G´omez-Lozano, Zulema Udaondo, Søren Molin, and Juan-Luis Ramos. Diﬀerential transcriptional response to antibiotics by pseudomonas putidadot-t1e. Environmental Microbiology, 17:3251–3262, September 2015.

[186] Leigh G Monahan, Isabella V Hajduk, Sinead P Blaber, Ian G Charles, and Eliza- beth J Harry. Coordinating bacterial cell division with nutrient availability: a role for glycolysis. mBio, 5:e00935–e00914, May 2014.

[187] Leigh G Monahan and Elizabeth J Harry. You are what you eat: Metabolic control of bacterial division. Trends in Microbiology, 24:181–189, March 2016. Bibliography 124

[188] Ian R Monk, Ishita M Shah, Min Xu, Man-Wah Tan, and Timothy J Foster. Trans- forming the untransformable: application of direct transformation to manipulate genetically staphylococcus aureus and staphylococcus epidermidis. mBio, 3, 2012.

[189] K Nakajima and J Kawamata. Studies on the mechanism of action of colistin. ii. alteration of permeability of escherichia coli by colistin. Biken Journal, 8:233–239, December 1965.

[190] Asha S Nayar, Thomas J Dougherty, Keith E Ferguson, Brett A Granger, Lisa McWilliams, Clare Stacey, Lindsey J Leach, Shin-Ichiro Narita, Hajime Tokuda, Alita A Miller, Dean G Brown, and Sarah M McLeod. Novel antibacterial targets and compounds revealed by a high-throughput cell wall reporter assay. Journal of Bacteriology, 197:1726–1734, May 2015.

[191] Poochit Nonejuie, Michael Burkart, Kit Pogliano, and Joe Pogliano. Bacterial cytological proﬁling rapidly identiﬁes the cellular pathways targeted by antibacterial molecules. Proceedings of the National Academy of Sciences of the United States of America, 110:16169–16174, October 2013.

[192] Poochit Nonejuie, Rachelle M Trial, Gerald L Newton, Anne Lamsa, Varahenage Ranmali Perera, Julieta Aguilar, Wei-Ting Liu, Pieter C Dorrestein, Joe Pogliano, and Kit Pogliano. Application of bacterial cytological proﬁling to crude natural product extracts reveals the antibacterial arsenal of bacillus subtilis. The Journal of Antibiotics, 69:353–361, May 2016.

[193] D Novo, N G Perlmutter, R H Hunt, and H M Shapiro. Accurate ﬂow cytometric membrane potential measurement in bacteria using diethyloxacarbocyanine and a ratiometric technique. Cytometry, 35:55–63, January 1999.

[194] Hidetaka Ogino, Haruhiko Teramoto, Masayuki Inui, and Hideaki Yukawa. Divs, a novel sos-inducible cell-division suppressor in corynebacterium glutamicum. Molec- ular Microbiology, 67:597–608, February 2008.

[195] Y Ohashi, Y Chijiiwa, K Suzuki, K Takahashi, H Nanamiya, T Sato, Y Hosoya, K Ochi, and F Kawamura. The lethal eﬀect of a benzamide derivative, 3- methoxybenzamide, can be suppressed by mutations within a cell division gene, ftsz, in bacillus subtilis. Journal of Bacteriology, 181:1348–1351, February 1999.

[196] Mira Okshevsky and Rikke Louise Meyer. The role of extracellular dna in the establishment, maintenance and perpetuation of bacterial bioﬁlms. Critical Reviews in Microbiology, 41:341–352, 2015.

[197] Alex J O’Neill and Ian Chopra. Preclinical evaluation of novel antibacterial agents by microbiological and molecular techniques. Expert Opinion on Investigational Drugs, 13:1045–1063, August 2004.

[198] Jim O’Neill. Tackling drug-resistant infections globally: Final report and recom- mendations. the review on antimicrobial resistance. 2016. Bibliography 125

[199] Charles C Onunkwo, Beth L Hahn, and Peter G Sohnle. Clearance of experimental cutaneous staphylococcus aureus infections in mice. Archives of Dermatological Research, 302:375–382, July 2010.

[200] Lisa M Oppegard, Bree L Hamann, Kathryn R Streck, Keith C Ellis, Hans-Peter Fiedler, Arkady B Khodursky, and Hiroshi Hiasa. In vivo and in vitro patterns of the activity of simocyclinone d8, an angucyclinone antibiotic from streptomyces antibioticus. Antimicrobial Agents and Chemotherapy, 53:2110–2119, May 2009.

[201] Cristina Ortiz, Paolo Natale, Laura Cueto, and Miguel Vicente. The keepers of the ring: regulators of ftsz assembly. FEMS Microbiology Reviews, 40:57–67, January 2016.

[202] Ilya A Osterman, Ekaterina S Komarova, Dmitry I Shiryaev, Ilya A Korniltsev, Irina M Khven, Dmitry A Lukyanov, Vadim N Tashlitsky, Marina V Serebryakova, Olga V Efremenkova, Yan A Ivanenkov, Alexey A Bogdanov, Petr V Sergiev, and Olga A Dontsova. Sorting out antibiotics’ mechanisms of action: a double ﬂuores- cent protein reporter for high-throughput screening of ribosome and dna biosynthesis inhibitors. Antimicrobial Agents and Chemotherapy, 60:7481–7489, December 2016.

[203] Kevin Outterson, John H Rex, Tim Jinks, Peter Jackson, John Hallinan, Steve Karp, Deborah T Hung, Francois Franceschi, Tyler Merkeley, Christopher Houchens, Dennis M Dixon, Michael G Kurilla, Rosemarie Aurigemma, and Joseph Larsen. Accelerating global innovation to address antibacterial resistance: intro- ducing carb-x. Nature Reviews Drug Discovery, 15:589–590, September 2016.

[204] Lincoln Pasquina, John P Santa Maria, B McKay Wood, Samir H Moussa, Leigh M Matano, Marina Santiago, Sara E S Martin, Wonsik Lee, Timothy C Meredith, and Suzanne Walker. A synthetic lethal approach for compound and target identiﬁca- tion in staphylococcus aureus. Nature Chemical Biology, 12:40–45, January 2016.

[205] David J Payne, Michael N Gwynn, David J Holmes, and David L Pompliano. Drugs for bad bugs: confronting the challenges of antibacterial discovery. Nature Reviews Drug Discovery, 6:29–40, January 2007.

[206] Kelly C Peach, Walter M Bray, Dustin Winslow, Peter F Linington, and Roger G Linington. Mechanism of action-based classiﬁcation of antibiotics using high- content bacterial image analysis. Molecular BioSystems, 9:1837–1848, July 2013.

[207] Jason M Peters, Alexandre Colavin, Handuo Shi, Tomasz L Czarny, Matthew H Larson, Spencer Wong, John S Hawkins, Candy H S Lu, Byoung-Mo Koo, Eliza- beth Marta, Anthony L Shiver, Evan H Whitehead, Jonathan S Weissman, Eric D Brown, Lei S Qi, Kerwyn Casey Huang, and Carol A Gross. A comprehensive, crispr-based functional analysis of essential genes in bacteria. Cell, 165:1493–1506, June 2016. Bibliography 126

[208] Sheila M Pimentel-Elardo, Dan Sørensen, Louis Ho, Mikaela Ziko, Stephanie A Bueler, Stella Lu, Joe Tao, Arvin Moser, Richard Lee, David Agard, Greg Fairn, John L Rubinstein, Brian K Shoichet, and Justin R Nodwell. Activity-independent discovery of secondary metabolites using chemical elicitation and cheminformatic inference. ACS Chemical Biology, 10:2616–2623, November 2015.

[209] Joe Pogliano, Nicolas Pogliano, and Jared A Silverman. Daptomycin-mediated reorganization of membrane architecture causes mislocalization of essential cell division proteins. Journal of Bacteriology, 194:4494–4504, September 2012.

[210] J H Pollack and F C Neuhaus. Changes in wall teichoic acid during the rod- sphere transition of bacillus subtilis 168. Journal of Bacteriology, 176:7252–7259, December 1994.

[211] David L Popham and Kevin D Young. Role of penicillin-binding proteins in bacterial cell morphogenesis. Current Opinion in Microbiology, 6:594–599, December 2003.

[212] Steﬀen Porwollik, Carlos A Santiviago, Pui Cheng, Fred Long, Prerak Desai, Jen- nifer Fredlund, Shabarinath Srikumar, Cecilia A Silva, Weiping Chu, Xin Chen, Roc´ıoCanals, M Megan Reynolds, Lydia Bogomolnaya, Christine Shields, Ping Cui, Jinbai Guo, Yi Zheng, Tiana Endicott-Yazdani, Hee-Jeong Yang, Aimee Maple, Yury Ragoza, Carlos J Blondel, Camila Valenzuela, Helene Andrews- Polymenis, and Michael McClelland. Deﬁned single-gene and multi-gene deletion mutant collections in salmonella enterica sv typhimurium. PLOS One, 9:e99820, 2014.

[213] Ranjani Prabhakara, Oded Foreman, Roberto De Pascalis, Gloria M Lee, Roger D Plaut, Stanley Y Kim, Scott Stibitz, Karen L Elkins, and Tod J Merkel. Epicuta- neous model of community-acquired staphylococcus aureus skin infections. Infec- tion and immunity, 81:1306–1315, April 2013.

[214] Maya Prakash Singh. Rapid test for distinguishing membrane-active antibacterial agents. Journal of Microbiological Methods, 67:125–130, October 2006.

[215] Andr´e A Pulschen, Diego E Sastre, Federico Machinandiarena, Agostina Crotta Asis, Daniela Albanesi, Diego de Mendoza, and Frederico J Gueiros-Filho. The stringent response plays a key role in bacillus subtilis survival of fatty acid starvation. Molecular Microbiology, 103:698–712, February 2017.

[216] R J V Pulvertaft. The eﬀect of antibiotics on growing cultures of bacterium coli. The Journal of Pathology and Bacteriology, 64:75–89, January 1952.

[217] D T Quach, G Sakoulas, V Nizet, J Pogliano, and K Pogliano. Bacterial cytological proﬁling (bcp) as a rapid and accurate antimicrobial susceptibility testing method for staphylococcus aureus. EBioMedicine, 4:95–103, February 2016. Bibliography 127

[218] Jara Radeck, Georg Fritz, and Thorsten Mascher. The cell envelope stress response of bacillus subtilis: from static signaling devices to dynamic regulatory network. Current Genetics, 63:79–90, February 2017.

[219] Mithila Rajagopal and Suzanne Walker. Envelope structures of gram-positive bacteria. Current Topics in Microbiology and Immunology, 404:1–44, 2017.

[220] C Reading and M Cole. Clavulanic acid: a beta-lactamase-inhiting beta-lactam from streptomyces clavuligerus. Antimicrobial Agents and Chemotherapy, 11:852– 857, May 1977.

[221] John H Rex and Kevin Outterson. Antibiotic reimbursement in a model delinked from sales: a benchmark-based worldwide approach. Lancet Infectious Diseases, 16:500–505, April 2016.

[222] Dante P Ricci and Thomas J Silhavy. The bam machine: a molecular cooper. Biochimica et Biophysica Acta, 1818:1067–1084, April 2012.

[223] G N Rolinson. Eﬀect of beta-lactam antibiotics on bacterial cell growth rate. Journal of General Microbiology, 120:317–323, October 1980.

[224] Veronica W Rowlett and William Margolin. The bacterial divisome: ready for its close-up. Philosophical Transactions of the Royal Society of London, 370, October 2015.

[225] M G Sargent. Control of cell length in bacillus subtilis. Journal of Bacteriology, 123:7–19, July 1975.

[226] Peter Sass and Heike Br¨otz-Oesterhelt. Bacterial cell division as a target for new antibiotics. Current Opinion in Microbiology, 16:522–530, October 2013.

[227] Mikhail M Savitski, Friedrich B M Reinhard, Holger Franken, Thilo Werner, Maria FälthSavitski, Dirk Eberhard, Daniel Martinez Molina, Rozbeh Jafari, Re- becca Bakszt Dovega, Susan Klaeger, Bernhard Kuster, PärNordlund, Marcus Bantscheff, and Gerard Drewes. Tracking cancer drugs in living cells by thermal profiling of the proteome. Science, 346:1255784, October 2014.

[228] M Schaechter, O Maaloe, and N O Kjeldgaard. Dependency on medium and temperature of cell size and chemical composition during balanced grown of salmonella typhimurium. Journal of General Microbiology, 19:592–606, December 1958.

[229] Kathrin Schirner, Jon Marles-Wright, Richard J Lewis, and Jeﬀ Errington. Distinct and essential morphogenic functions for wall- and lipo-teichoic acids in bacillus subtilis. EMBO Journal, 28:830–842, April 2009.

[230] Dirk Schnappinger. Genetic approaches to facilitate antibacterial drug development. Cold Spring Harbor Perspectives in Medicine, 5:a021139, February 2015. Bibliography 128

[231] G Schreiber, E Z Ron, and G Glaser. ppgpp-mediated regulation of dna replication and cell division in escherichia coli. Current Microbiology, 30:27–32, January 1995. [232] Wiebke Schröder,Christiane Goerke, and Christiane Wolz. Opposing effects of aminocoumarins and fluoroquinolones on the sos response and adaptability in staphylococcus aureus. The Journal of Antimicrobial Chemotherapy, 68:529–538, March 2013. [233] Elyse Shapiro and Fran¸coisBaneyx. Stress-based identification and classification of antibacterial agents: second-generation escherichia coli reporter strains and optimization of detection. Antimicrobial Agents and Chemotherapy, 46:2490–2497, August 2002. [234] Elyse Shapiro and Fran¸coisBaneyx. Stress-activated bioluminescent escherichia coli sensors for antimicrobial agents detection. Journal of Biotechnology, 132:487–493, December 2007. [235] Karen Joy Shaw, Nancy Miller, Xuejun Liu, Dmitri Lerner, Jackson Wan, Anton Bittner, and Brian J Morrow. Comparison of the changes in global gene expression of escherichia coli induced by four bactericidal agents. Journal of Molecular Microbiology and Biotechnology, 5:105–122, 2003. [236] Fangwei Si, Dongyang Li, Sarah E Cox, John T Sauls, Omid Azizi, Cindy Sou, Amy B Schwartz, Michael J Erickstad, Yonggun Jun, Xintian Li, and Suckjoon Jun. Invariance of initiation mass and predictability of cell size in escherichia coli. Current Biology, 27:1278–1287, May 2017. [237] Krzysztof Sieradzki and Alexander Tomasz. Inhibition of the autolytic system by vancomycin causes mimicry of vancomycin-intermediate staphylococcus aureus- type resistance, cell concentration dependence of the mic, and antibiotic tolerance in vancomycin-susceptible s. aureus. Antimicrobial Agents and Chemotherapy, 50:527– 533, February 2006. [238] G Siewert and J L Strominger. Bacitracin: an inhibitor of the dephosphorylation of lipid pyrophosphate, an intermediate in the biosynthesis of the peptidoglycan of bacterial cell walls. Proceedings of the National Academy of Sciences of the United States of America, 57:767–773, March 1967. [239] Thomas J Silhavy, Daniel Kahne, and Suzanne Walker. The bacterial cell envelope. Cold Spring Harbor Perspectives in Biology, 2:a000414, May 2010. [240] Lynn L Silver. Challenges of antibacterial discovery. Clinical Microbiology Reviews, 24:71–109, January 2011. [241] Lynn L. Silver. Rational Approaches to Antibacterial Discovery: Pre-Genomic Directed and Phenotypic Screening, pages 33–75. Springer US, Boston, MA, 2012. [242] Lynn L Silver. Appropriate targets for antibacterial drugs. Cold Spring Harbor Perspectives in Medicine, 6, December 2016. Bibliography 129

[243] Jared A Silverman, Nancy G Perlmutter, and Howard M Shapiro. Correlation of daptomycin bactericidal activity and membrane depolarization in staphylococcus aureus. Antimicrobial Agents and Chemotherapy, 47:2538–2544, August 2003.

[244] Sheo B Singh, John W Phillips, and Jun Wang. Highly sensitive target-based whole- cell antibacterial discovery strategy by antisense rna silencing. Current Opinion in Drug Discovery & Development, 10:160–166, March 2007.

[245] Anthony M Sperber and Jennifer K Herman. Metabolism shapes the cell. Journal of Bacteriology, 199, June 2017.

[246] B G Spratt. Distinct penicillin binding proteins involved in the division, elongation, and shape of escherichia coli k12. Proceedings of the National Academy of Sciences of the United States of America, 72:2999–3003, August 1975.

[247] B G Spratt. The mechanism of action of mecillinam. The Journal of Antimicrobial Chemotherapy, 3 Suppl B:13–19, July 1977.

[248] Arjun Srinivasan. Antibiotic stewardship: Why we must, how we can. Cleveland Clinic Journal of Medicine, 84:673–679, September 2017.

[249] E O Stapley, D Hendlin, J M Mata, M Jackson, H Wallick, S Hernandez, S Mochales, S A Currie, and R M Miller. Phosphonomycin. i. discovery and in vitro biological characterization. Antimicrobial Agents and Chemotherapy, 9:284– 290, 1969.

[250] Rebecca Steidl, Stacy Pearson, Robert E Stephenson, Nagender Ledala, Sutthirat Sitthisak, Brian J Wilkinson, and Radheshyam K Jayaswal. Staphylococcus aureus cell wall stress stimulon gene-lacz fusion strains: potential for use in screening for cell wall-active antimicrobials. Antimicrobial Agents and Chemotherapy, 52:2923– 2925, August 2008.

[251] Atsushi Sugimoto, Asuka Maeda, Kaori Itto, and Hirokazu Arimoto. Deciphering the mode of action of cell wall-inhibiting antibiotics using metabolic labeling of growing peptidoglycan in streptococcus pyogenes. Scientiﬁc Reports, 7:1129, April 2017.

[252] H. Suzuki, J. Pangborn, and W W Kilgore. Filamentous cells of escherichia coli formed in the presence of mitomycin. Journal of Bacteriology, 93:683–688, February 1967.

[253] T Suzuki, K Fujikura, T Higashiyama, and K Takata. Dna staining for ﬂuorescence and laser confocal microscopy. The Journal of Histochemistry and Cytochemistry, 45:49–53, January 1997.

[254] Lonnie P Swift, Ada Rephaeli, Abraham Nudelman, Don R Phillips, and Suzanne M Cutts. Doxorubicin-dna adducts induce a non-topoisomerase ii- mediated form of cell death. Cancer Research, 66:4863–4871, May 2006. Bibliography 130

[255] David C Swinney and Jason Anthony. How were new medicines discovered? Nature Reviews Drug Discovery, 10:507–519, June 2011.

[256] Jonathan G Swoboda, Jennifer Campbell, Timothy C Meredith, and Suzanne Walker. Wall teichoic acid function, biosynthesis, and inhibition. Chembiochem, 11:35–45, January 2010.

[257] Sattar Taheri-Araghi, Serena Bradde, John T Sauls, Norbert S Hill, Petra A Levin, Johan Paulsson, Massimo Vergassola, and Suckjoon Jun. Cell-size control and homeostasis in bacteria. Current Biology, 25:385–391, February 2015.

[258] Stephanie N Taylor, David H Morris, Ann K Avery, Kimberly A Workowski, By- ron E Batteiger, Courtney A Tiﬀany, Caroline R Perry, Aparna Raychaudhuri, Nicole E Scangarella-Oman, Mohammad Hossain, and Etienne F Dumont. Gepoti- dacin for the treatment of uncomplicated urogenital gonorrhea: A phase 2, ran- domized, dose-ranging, single-oral dose evaluation. Clinical Infectious Diseases, 67:504–512, August 2018.

[259] Ashley K Tehranchi, Matthew D Blankschien, Yan Zhang, Jennifer A Halliday, An- jana Srivatsan, Jia Peng, Christophe Herman, and Jue D Wang. The transcription factor dksa prevents conﬂicts between dna replication and transcription machinery. Cell, 141:595–605, May 2010.

[260] Kamaleddin H M E Tehrani and Nathaniel I Martin. -lactam/-lactamase inhibitor combinations: an update. MedChemComm, 9:1439–1456, September 2018.

[261] Ursula Theuretzbacher, Simon Gottwalt, Peter Beyer, Mark Butler, Lloyd Czaplewski, Christian Lienhardt, Lorenzo Moja, Mical Paul, Sarah Paulin, John H Rex, Lynn L Silver, Melvin Spigelman, Guy E Thwaites, Jean-Pierre Paccaud, and Stephan Harbarth. Analysis of the clinical antibacterial and antituberculosis pipeline. The Lancet Infectious Diseases, October 2018.

[262] M Tomasz. Mitomycin c: small, fast and deadly (but very selective). Chemistry & Biology, 2:575–579, September 1995.

[263] Adrian Towse, Christopher K Hoyle, Jonathan Goodall, Mark Hirsch, Jorge Mestre- Ferrandiz, and John H Rex. Time for a change in how new antibiotics are reim- bursed: Development of an insurance framework for funding new antibiotics based on a policy of risk mitigation. Health Policy, 121:1025–1030, October 2017.

[264] Carolina Tropini, Timothy K Lee, Jen Hsin, Samantha M Desmarais, Tristan Ursell, Russell D Monds, and Kerwyn Casey Huang. Principles of bacterial cell-size deter- mination revealed by cell-wall synthesis perturbations. Cell Reports, 9:1520–1527, November 2014.

[265] D Trusca, S Scott, C Thompson, and D Bramhill. Bacterial sos checkpoint protein sula inhibits polymerization of puriﬁed ftsz cell division protein. Journal of Bacteriology, 180:3946–3953, August 1998. Bibliography 131

[266] H Umezawa, K Maeda, T Takeuchi, and Y Okami. New antibiotics, bleomycin a and b. The Journal of Antibiotics, 19:200–209, September 1966.

[267] Divakara S S M Uppu, Mohini M Konai, Paramita Sarkar, Sandip Samaddar, Isabel C M Fensterseifer, Celio Farias-Junior, Paramanandam Krishnamoorthy, Bibek R Shome, Oct´avioL Franco, and Jayanta Haldar. Membrane-active macromolecules kill antibiotic-tolerant bacteria and potentiate antibiotics towards gram-negative bacteria. PLOS One, 12:e0183263, 2017.

[268] Andreas Urban, Stefan Eckermann, Beate Fast, Susanne Metzger, Matthias Gehling, Karl Ziegelbauer, Helga R¨ubsamen-Waigmann, and Christoph Freiberg. Novel whole-cell antibiotic biosensors for compound discovery. Applied and environmental microbiology, 73:6436–6443, October 2007.

[269] Tristan S Ursell, Jeﬀrey Nguyen, Russell D Monds, Alexandre Colavin, Gabriel Billings, Nikolay Ouzounov, Zemer Gitai, Joshua W Shaevitz, and Kerwyn Casey Huang. Rod-like bacterial shape is maintained by feedback between cell curvature and cytoskeletal localization. Proceedings of the National Academy of Sciences of the United States of America, 111:E1025–E1034, March 2014.

[270] Stephen Vadia and Petra Anne Levin. Growth rate and cell size: a re-examination of the growth law. Current Opinion in Microbiology, 24:96–103, April 2015.

[271] Stephen Vadia, Jessica L Tse, Rafael Lucena, Zhizhou Yang, Douglas R Kellogg, Jue D Wang, and Petra Anne Levin. Fatty acid availability sets cell envelope capacity and dictates microbial cell size. Current Biology, 27:1757–1767.e5, June 2017.

[272] Alex van Belkum, Till T Bachmann, Gerd L¨udke, Jan Gorm Lisby, Gunnar Kahl- meter, Allan Mohess, Karsten Becker, John P Hays, Neil Woodford, Konstantinos Mitsakakis, Jacob Moran-Gilad, Jordi Vila, Harald Peter, John H Rex, Wm Michael Dunne, JPIAMR AMR-RDT Working Group on Antimicrobial Resistance, and Rapid Diagnostic Testing. Developmental roadmap for antimicrobial susceptibility testing systems. Nature Reviews Microbiology, 17:51–62, January 2019.

[273] Julia J van Rensburg, Kate R Fortney, Lan Chen, Andrew J Krieger, Bruno P Lima, Alan J Wolfe, Barry P Katz, Zhong-Yin Zhang, and Stanley M Spinola. Development and validation of a high-throughput cell-based screen to identify activators of a bacterial two-component signal transduction system. Antimicrobial Agents and Chemotherapy, 59:3789–3799, July 2015.

[274] Sven van Teeﬀelen, Siyuan Wang, Leon Furchtgott, Kerwyn Casey Huang, Ned S Wingreen, Joshua W Shaevitz, and Zemer Gitai. The bacterial actin mreb rotates, and rotation depends on cell-wall assembly. Proceedings of the National Academy of Sciences of the United States of America, 108:15822–15827, September 2011. Bibliography 132

[275] Vellarkad N. Viswanadhan, Arup K. Ghose, Ganapathi R. Revankar, and Roland K. Robins. Atomic physicochemical parameters for three dimensional structure directed quantitative structure-activity relationships. 4. additional parameters for hydrophobic and dispersive interactions and their application for an automated su- perposition of certain naturally occurring nucleoside antibiotics. Journal of Chem- ical Information and Computer Sciences, 29(3):163–172, 1989.

[276] Seychelle M Vos, Elsa M Tretter, Bryan H Schmidt, and James M Berger. All tangled up: how cells direct, manage and exploit topoisomerase function. Nature Reviews Molecular Cell Biology, 12:827–841, November 2011.

[277] S A Waksman. Streptomycin: Background, isolation, properties, and utilization. Science, 118:259–266, September 1953.

[278] Joselynn Wallace, Nicholas O Bowlin, Debra M Mills, Panatda Saenkham, Steven M Kwasny, Timothy J Opperman, John D Williams, Charles O Rock, Terry L Bowlin, and Donald T Moir. Discovery of bacterial fatty acid synthase type ii inhibitors using a novel cellular bioluminescent reporter assay. Antimicro- bial Agents and Chemotherapy, 59:5775–5787, September 2015.

[279] Richard B Weart, Amy H Lee, An-Chun Chien, Daniel P Haeusser, Norbert S Hill, and Petra Anne Levin. A metabolic sensor governing cell size in bacteria. Cell, 130:335–347, July 2007.

[280] Richard B Weart and Petra Anne Levin. Growth rate-dependent regulation of medial ftsz ring formation. Journal of Bacteriology, 185:2826–2834, May 2003.

[281] Tina Wecke, Daniela Zühlke, Ulrike Mäder, Sina Jordan, Birgit Voigt, Ste- fan Pelzer, Harald Labischinski, Georg Homuth, Michael Hecker, and Thorsten Mascher. Daptomycin versus friulimicin b: in-depth profiling of bacillus subtilis cell envelope stress responses. Antimicrobial Agents and Chemotherapy, 53:1619– 1623, April 2009.

[282] Michaela Wenzel, Marina Rautenbach, J Arnold Vosloo, Tjalling Siersma, Christo- pher H M Aisenbrey, Ekaterina Zaitseva, Wikus E Laubscher, Wilma van Rensburg, Jan C Behrends, Burkhard Bechinger, and Leendert W Hamoen. The multifaceted antibacterial mechanisms of the pioneering peptide antibiotics tyrocidine and gramicidin s. mBio, 9, October 2018.

[283] I Wiedemann, E Breukink, C van Kraaij, O P Kuipers, G Bierbaum, B de Kruijﬀ, and H G Sahl. Speciﬁc binding of nisin to the peptidoglycan precursor lipid ii combines pore formation and inhibition of cell wall biosynthesis for potent antibiotic activity. The Journal of Biological Chemistry, 276:1772–1779, January 2001.

[284] Hallie Wimberly, Chandan Shee, P C Thornton, Priya Sivaramakrishnan, Susan M Rosenberg, and P J Hastings. R-loops and nicks initiate dna breakage and genome instability in non-growing escherichia coli. Nature Communications, 4:2115, 2013. Bibliography 133

[285] M Wu, E Maier, R Benz, and R E Hancock. Mechanism of interaction of different classes of cationic antimicrobial peptides with planar bilayers and with the cytoplasmic membrane of escherichia coli. Biochemistry, 38:7235–7242, June 1999.

[286] Yu-hua Xing, Wei Wang, Su-qin Dai, Ti-yan Liu, Jun-jie Tan, Guo-long Qu, Yu- xia Li, Yan Ling, Gang Liu, Xue-qi Fu, and Hui-peng Chen. Daptomycin exerts rapid bactericidal activity against bacillus anthracis without disrupting membrane integrity. Acta Pharmacologica Sinica, 35:211–218, February 2014.

[287] H Howard Xu, John D Trawick, Robert J Haselbeck, R Allyn Forsyth, Robert T Yamamoto, Rich Archer, Joe Patterson, Molly Allen, Jamie M Froelich, Ian Taylor, Danny Nakaji, Randy Maile, G C Kedar, Marshall Pilcher, Vickie Brown-Driver, Melissa McCarthy, Amy Files, David Robbins, Paula King, Susan Sillaots, Cheryl Malone, Carlos S Zamudio, Terry Roemer, Liangsu Wang, Philip J Youngman, and Daniel Wall. Staphylococcus aureus targetarray: comprehensive diﬀerential essential gene expression as a mechanistic tool to proﬁle antibacterials. Antimicrobial Agents and Chemotherapy, 54:3659–3670, September 2010.

[288] Zhizhong Yao, Rebecca M Davis, Roy Kishony, Daniel Kahne, and Natividad Ruiz. Regulation of cell size in response to nutrient availability by fatty acid biosynthesis in escherichia coli. Proceedings of the National Academy of Sciences of the United States of America, 109:E2561–E2568, September 2012.

[289] Grace Yim, JoAnn McClure, Michael G Surette, and Julian E Davies. Modulation of salmonella gene expression by subinhibitory concentrations of quinolones. The Journal of Antibiotics, 64:73–78, January 2011.

[290] Olga Zafra, Mar´ıa Lamprecht-Grand´ıo, Carolina González de Figueras, and JoséEduardo González-Pastor. Extracellular dna release by undomesticated bacillus subtilis is regulated by early competence. PLOS One, 7:e48716, 2012.

[291] Zhang, Chung, and Oldenburg. A simple statistical parameter for use in evaluation and validation of high throughput screening assays. Journal of Biomolecular Screening, 4:67–73, 1999.

[292] Xingyue Zhu, Dongqi Liu, Atul K Singh, Rishi Drolia, Xingjian Bai, Shivendra Ten- guria, and Arun K Bhunia. Tunicamycin mediated inhibition of wall teichoic acid aﬀects staphylococcus aureus and listeria monocytogenes cell morphology, bioﬁlm formation and virulence. Frontiers in Microbiology, 9:1352, 2018.

[293] Soumaya Zlitni, Jan E. Blanchard, and Eric D. Brown. High-Throughput Screening of Model Bacteria, pages 13–27. Humana Press, Totowa, NJ, 2009.

[294] Soumaya Zlitni, Lauren F Ferruccio, and Eric D Brown. Metabolic suppression identiﬁes new antibacterial inhibitors under nutrient limitation. Nature Chemical Biology, 9:796–804, December 2013.