Identifying the best antibiotic-target genes in

essential biochemical pathways

Paraskevi Giachou

__________________________________________

Master Degree Project in Infection Biology, 45 credits. Spring 2018

Department of Medical Biochemistry and Microbiology (IMBIM)

Supervisor: Diarmaid Hughes

Co-Supervisors: Gerrit Brandis, Sha Cao

2

Table of Contents

Abstract ...................................................................................................................................... 3

Novel antibiotics: a solution to a post-antibiotic apocalypse ..................................................... 4

Introduction ................................................................................................................................ 5

The Non-Mevalonate (MEP) Pathway ................................................................................... 5

The Coenzyme A biosynthetic Pathway ................................................................................ 6

The Nicotinamide Adenine Dinucleotide (NAD) Biosynthetic Pathway .............................. 8

Aim of the project ...................................................................................................................... 9

Materials and Methods ............................................................................................................. 10

Strains and Growth conditions ............................................................................................. 10

Polymerase chain reaction .................................................................................................... 10

Purification of the PCR products ......................................................................................... 11

Plasmid extraction ................................................................................................................ 11

Plasmid transformation ........................................................................................................ 11

λ Red Recombineering ......................................................................................................... 11

Spot assay ............................................................................................................................. 12

Bioscreen .............................................................................................................................. 12

Minimal Inhibitory Concentration (MIC) assay................................................................... 13

MACS Flow Cytometry ....................................................................................................... 13

Sequencing ........................................................................................................................... 13

Results ...................................................................................................................................... 14

Construction of strains ......................................................................................................... 14

Spot Assay ............................................................................................................................ 15

Bioscreen .............................................................................................................................. 15

MIC Assay ............................................................................................................................ 17

Selection of ispG and ispH mutants ..................................................................................... 17

Yfp-kan cassette fusions ....................................................................................................... 18

Discussion ................................................................................................................................ 20

Acknowledgements .................................................................................................................. 22

References ................................................................................................................................ 23

Supplementary Material ........................................................................................................... 27

3

ABSTRACT

The rapid rise in the frequency of antibiotic resistance developed by many pathogens is an

urgent and concerning problem the world has to face nowadays. The discovery of novel

antibiotics is crucial to overcoming this matter and face multi-resistant microbes that could be

life threatening otherwise. In this project three essential pathways in bacteria (MEP, NAD,

CoA) were examined in order to identify the most potential antibiotic target genes. To mimic

the action of an antibiotic, strains in which expression of the genes of interest can be down

regulated (using the inducible araBAD promoter) were constructed. The constructed strains

were subsequently tested to determine if there is a correlation between gene expression and

bacterial growth. The construction of strains harboring the genes from the CoA has not been

finished yet but for the MEP pathway, genes dxs, ispG and ispH appeared to be the most

promising target genes, while the nadK gene was the only potential target of the NAD

pathway. Using Fosmidomycin, a commercially available drug targeting the MEP pathway

gene dxr, we were able to show that the constructed strains can be used to screen for novel

compounds that inhibit protein functions of genes within these pathways. Taken together, the

results of this study suggest that the products of the dxs, ispG, ispH and nadK genes of the

MEP and NAD pathway appear to be promising targets for novel antibiotics.

Key words E.coli, antibiotic resistance, MEP pathway, NAD pathway, Fosmidomycin

4

Novel antibiotics: a solution to a post-antibiotic apocalypse

Before the 20th

century bacterial infections used to be incurable and deadly. A simple

paper cut could cost someone his life, just because there was no way of fighting back the

invisible to the eye but yet vicious microbes. It all changed in 1928 when Alexander Fleming

discovered penicillin, the first true antibiotic. Ever since a flourishing antibiotic era began,

where more antimicrobial drugs that inhibit or kill bacteria were found and people started to

believe that infectious diseases have been defeated. Between 1930 and 1963 more than 20

novel classes of antibiotics were produced and reached the market but since then only two

novel ones are commercially available.

However, bacteria are versatile and antibiotic resistance did not take long to rise.

Bacteria started acquiring mechanisms that overcome the effect of the drug or mechanisms

that affect the drug itself. Infections that were once considered treatable are now resistant to

the usual drugs of choice and the appearance of multi-resistant bacteria like the Methicillin-

resistant Staphylococcus aureus (MRSA) pose a great threat to the public health. Drug–

resistant bacterial infections claim around 25,000 deaths per year in Europe and it has been

suggested that the number will continue to increase in the future if no measures are taken. The

effort to bring new antibiotics in the market has led to the production of more antimicrobial

drugs but no novel classes have been discovered.

During this study, three biochemical pathways (MEP, NAD, CoA) that are essential

for bacterial survival were examined to identify the most potential antibiotic target genes.

Each pathway has a certain number of genes and thus enzymes that are involved to carry out

the reactions. To mimic the action of an antibiotic that targets these essential enzymes, strains

were constructed for each pathway in a way that the amount of the essential proteins produced

was controllable. The construction of strains for the CoA pathway has not been finished yet

and therefore results for this pathway cannot be drawn. When the rest of the strains were

tested to produce lesser amounts of protein, bacterial growth was inhibited for the strains that

were producing less Dxs, IspG, IspH and NadK enzymes. It is therefore proposed that genes

dxs, ispG and ispH of the MEP pathway and gene nadK of the NAD pathway are attractive

target genes for future antibiotic development. Furthermore, it is shown that a combination

therapy targeting the first two enzymes of the MEP pathway (Dxs, Dxr) at the same time has a

better result than a mono-therapy.

Taken this together, the discovery of novel classes of antibiotics is possible and can

even defeat the multi-resistant pathogens that cause tens of thousands of deaths annually but

most importantly prevent the post-antibiotic apocalypse from becoming a real world

phenomenon.

5

INTRODUCTION

Antibiotics are drugs that are used to prevent and treat bacterial infections by inhibiting

essential mechanisms of the microorganisms. Misuse of these drugs has led to a dramatic

increase in antibiotic resistance that poses a major threat to global health. Serious infections

such as tuberculosis and pneumonia become challenging to treat because of multi-resistant

microbial agents that limit the effectiveness of the usual drugs of choice. The severity of this

issue is supported by the following statement made by the World Health Organization

(WHO): ’’A post-antibiotic era - in which common infections and minor injuries can kill - far

from being an apocalyptic fantasy, is instead a very real possibility for the 21st century’’ (1).

The need for discovery of novel antibacterial compounds is urgent in order to minimize the

risk for public health and metabolic pathways that are essential for bacterial viability represent

ideal targets for these compounds.

The Non-Mevalonate (MEP) Pathway

Isoprenoids are a wide range of metabolites that originate from the five-carbon

precursor isopentenyl phosphate (IPP) or its isomer methylallyl diphosphate (DMAPP). They

are synthesized in all organisms and compose the largest group of natural products with a

diversity of structures and biological functions (2). In bacteria, these compounds are involved

in essential tasks such as electron transport (ubiquinone, menaquinone), cell wall and

membrane biosynthesis (bactoprenol) or the conversion of light into chemical energy

(chlorophylls, carotenoids) (3). While it was believed for a long time that IPP was synthesized

exclusively via the mevalonate pathway (MVA) in all organisms, an alternative pathway

known as MEP was identified in 1993 (4). The newly discovered MEP pathway is the only

one present in the majority of bacteria, plant plastids and the apicoplasts of apicomplexan

protozoa. It is completely absent in animals, including humans, which makes the enzymes

involved in the MEP pathway promising targets for new drug development (5).

Figure 1: The MEP pathway present in bacteria, plant plastids and the apicoplasts of apicomplexan protozoa.

Seven enzymatic steps are required for the conversion of pyruvate (Pyr) and glyceraldehyde-3-phosphate (G3P)

to the isoprenoid precursors, IPP and DMAPP (DXP: 1-deoxy-D-xylulose 5-phosphate, MEP: 2-C-methyl-D-

erythritol 4-phosphate, CDP-ME: 4-diphosphocytidyl-2-C-methyl-D-erythritol, CDP-MEP: 4-diphosphocytidyl-

2-C-methyl-D-erythritol 2-phosphate, MEcPP: 2-C-methyl-D-erythritol 2,4-cyclopyrophosphate, HMB-PP: 4-

Hydroxy-3-methyl-but-2-enyl pyrophosphate). The enzymes that catalyze each reaction are shown in red

whereas the enzymatic products in black. The enzymes that when mutated can lead to the production of DXP are

depicted in green.

6

The MEP pathway requires in total seven enzymatic steps for the production of IPP

and DMAPP (Figure 1). Considering their essentiality and the fact that these enzymes have no

homologues in mammals (6), they all are appealing drug targets. However, drug designing

can be challenging for enzymes like the IspD that contains a highly polar active site or IspE

which is an ATP-binding enzyme, both features where selectivity could be an obstacle. In

addition, IspG and IspH are both 4Fe-4S proteins and potential inhibitors targeting the Fe-S

cluster also raise concerns about selectivity among mammalian and bacterial enzymes. Dxs,

Dxr and IspF are considered the most promising targets due to the fact that they contain

potential drug binding sites and lack features that raise concerns about selectivity (7). Dxr is

the second enzyme of the pathway and the one that has been studied the most thoroughly (8).

Dxr function is inhibited by Fosmidomycin (FSM), which results in the blockage of the

isoprenoid biosynthesis in P. falciparum and some multidrug-resistant strains of bacteria (9,

10). Nonetheless, bacteria can develop a number of mechanisms to avoid this inhibition such

as alterations of the FSM transporter molecule or mutations that modify the FSM binding site

of the Dxr enzyme (11, 12). In addition, bacteria can bypass the blockage of the first or the

second step of the pathway by mutations in enzymes, normally not related to isoprenoid

biosynthesis, to produce DXP or MEP. In particular, mutations in the aceA and ribB genes

that encode for the E1 subunit of the pyruvate dehydrogenase complex (PDH) and 3,4-

dihydroxy-2-butanone 4-phosphate (DHBPS) respectively were able to rescue the otherwise

lethal inactivation of the dxs gene. This could be a result of either production of DXP by the

mutant enzymes or production of an alternative substrate that can be used by Dxr (13, 14).

Mutants that rescue a dxr inactivation were observed but the mechanism of bypassing was not

identified (13). Consequently, other MEP enzymes should also be explored rigorously to

identify the most promising drug targets in this pathway.

The Coenzyme A biosynthetic Pathway

Coenzyme A (CoA) is an essential, universally distributed cofactor that is involved in

many metabolic reactions such as synthesis and degradation of fatty acids, synthesis of

phospholipids and the citric acid cycle. The pathway is catalyzed in five enzymatic steps

(Figure 2) starting with pantothenate as a substrate (15). Even though most organisms,

including animals, rely on the uptake of exogenous pantothenate, there are some exceptions.

Some bacteria (e.g. Escherichia coli, Salmonella typhimurium), plants and fungi can

synthesize pantothenate de novo by utilizing β-alanine (17, 18). Nevertheless, Gerdes et al.

(19) showed that disrupting the pantothenate biosynthesis genes in E. coli had no effect on the

growth of the microorganism. This led to the conclusion that none of the involved genes are

essential because the bacteria can import exogenous pantothenate (19). Therefore, focus has

been on the five-step pathway that starts with pantothenate and leads to the production of

CoA due to the fact that the genes involved are essential for growth.

The pantothenate kinase (PanK or CoaA) catalyzes the first ATP-dependent step of

the pathway and converts pantothenate to 4-phosphopantothenate (20, 21). PanK of E. coli

(Type I) shares homology with other PanK proteins identified in bacteria such as

Haemophilus influenzae, Mycobacterium tuberculosis and Streptococcus pneumoniae and has

a different structure from the eukaryotic enzyme (19, 22). However, there are bacteria e.g.

Staphylococcus aureus that produce another type of PanK (Type II) that differs from the Type

7

I and is more similar to a eukaryotic one (19). A third type of PanK (Type III) has recently

been identified in Bacillus subtilis and subsequently in other bacteria missing PanK I or PanK

II. The main difference with the other two is its low affinity for ATP (23). PanK is the

enzyme of the CoA pathway that research on antimicrobial targets has been focused on due to

the diversity in structures and properties, as well as the fact that it catalyzes the committing

and rate-limiting step of the pathway (24). Even though inhibitors have been identified in

vitro, especially for the M. tuberculosis PanK Type I, none of them is able to inhibit the cell

growth in in vivo studies conducted so far and therefore it was concluded that PanK has a

poor vulnerability (25). In most bacteria, including E. coli, the second and third steps of the

pathway are carried out by a bifunctional enzyme, PPCS/PPCDC, encoded by the coaBC gene

(previously known as dfp) (26). However, in humans two distinct enzymes are responsible for

these two enzymatic reactions. The human PPCDC shares homology with the bacterial one

while the PPCS is less conserved between prokaryotes and eukaryotes (27). Some inhibitors

have been identified for the PPCS domain but there are no reports about the effect on bacterial

growth so far (28). PPAT, the enzyme that catalyzes the fourth step of the pathway to form

dephospho-coA, is encoded by the coaD gene (29). In mammals, the PPAT enzyme is fused

with the dephospho-CoA kinase that catalyzes the last step of the pathway, forming the

bifunctional enzyme coA synthase (30). However, the prokaryotic PPAT differs significantly

from the eukaryotic PPAT domain and that is the reason this enzyme is considered a potential

target for antibacterial drugs (27). Cycloalkyl pyrimidines were identified as inhibitors against

PPAT by AstraZeneca and the results were quite promising. The inhibitors were able to

suppress the growth of Gram-positive bacteria in vivo and even in animal models when tested

in vitro, which shows the importance of this enzyme as a drug target. However, the

optimization of these compounds regarding pharmacokinetics and toxicity was not feasible

and therefore it did not proceed to the phase of clinical testing (31). The fifth and last step of

the pathway, the phosphorylation of dephospho-coA, is catalyzed by dephospho-coA kinase

(DPCK), which is encoded by the coaE gene (32). This enzyme shows high sequence

similarity to the eukaryotic DPCK domain of the coA synthase and consequently studies are

not focused on it as it does not appear to be an attractive drug target (27).

Figure 2: The universal Coenzyme A biosynthetic pathway in bacteria. Five enzymatic steps are required for the

conversion of pantothenate to Coenzyme A. PanK is the enzyme catalyzing the first step and dephospho-coA

kinase (DPCK) the last step of the pathway. Bacteria can either synthesize pantothenate de novo or utilize

mechanisms for the uptake of exogenous pantothenate. In any case, the coenzyme A biosynthetic pathway from

pantothenate remains the same. The enzymes that catalyze each step are depicted in red whereas the enzymatic

products in black.

8

The Nicotinamide Adenine Dinucleotide (NAD) Biosynthetic Pathway

The nicotinamide adenine dinucleotides (NAD, NADH, NADP and NADPH) are

important coenzymes that carry out redox reactions in all living cells. NAD and NADP act as

oxidizing agents, accepting electrons from other molecules while NADH and NADPH act as

reducing agents, donating electrons (33). Apart from redox reactions, nicotinamide adenine

dinucleotides are implicated in ADP ribosylation of proteins and can also act as a cosubstrate

for non-redox enzymes such as bacterial DNA ligases (34, 35). NAD can be synthesized de

novo by simple components or through salvage pathways dependent on the recycling of

normal cellular catabolism metabolites. Almost all living organisms can synthesize NAD de

novo with some notable differences between eukaryotes and prokaryotes. The initial step in

the NAD pathway is the production of quinolinic acid (QA) from an amino acid. In animals

and some bacteria, tryptophan is used as a substrate whereas most bacteria synthesize QA

from aspartic acid (36, 37). The prokaryotic NAD biosynthesis is a six step enzymatic process

(Figure 3) and starts with aspartic acid which is oxidized by the L-aspartate oxidase (NadB).

The first three genes of the pathway (nadA, nadB, nadC) have been shown to be non-essential

for the growth bacteria whereas inactivation of each of the last three genes of the pathway

(nadD, nadE, nadK) led to no bacterial growth (38). This observation, coupled with the fact

that these essential genes are present in almost all bacterial genomes, make them an attractive

antibacterial drug target (39).

Figure 3: The NAD pathway starting with aspartic acid in bacteria. Five enzymatic steps are required for the

conversation of L-aspartic acid to NAD. One additional phosphorylation step is needed for the production of

NADP. The non-essential enzymes of the pathway are depicted in green whereas the three essential ones are

depicted in red. The enzymatic products are shown in black.

NadD is the first essential enzyme of the pathway that catalyzes the conversion of

nicotinic acid mononucleotide (NaMN) to nicotinic acid adenine dinucleotide (NaAD), the

first shared intermediate product between most de novo and salvage pathways. Enzymes of

this family have been studied in pathogenic and model bacteria and it has been found that the

enzyme is catalyzing the reaction much faster by using NaMN as a substrate instead of

nicotinamide mononucleotide (NMN) which is also a NAD precursor (40, 41, 42). On the

other hand, the human enzyme that shares similar functions (hsNMNAT) has the same

substrate preference for NaMN and NMN (43). Furthermore, the 3D structures of human and

bacterial NadD show significant differences in their active site conformation, which opens a

window for using discriminating NadD inhibitors (44). Even though there are no known drugs

that inhibit NadD, some inhibitors that are able to act on both Gram-positive and Gram-

negative bacteria with a weak selectivity for the human enzymes have been identified.

9

However, the binding affinity of these inhibitors is not optimal but the results validate the

importance of NadD as a drug target (45). The second essential enzyme, NAD synthetase

(NadE), is responsible for the conversion of NaAD to NAD. The action of the enzyme can be

ammonia-dependent (most bacteria e.g. E. coli, B. subtilis) or glutamine-dependent

(eukaryotes and some prokaryotes e.g. M. tuberculosis) (46, 47, 48, 49). Both types of

enzymes contain a highly conserved C-terminal synthetase domain while the M. tuberculosis

synthetase shares 23% sequence similarity with the equivalent human enzyme (50). Based on

these observations studies focused on targeting NadE have been carried out and inhibitors

have been identified. Nevertheless, some of the inhibitors showed no correlation between

enzymatic inhibition and effect on bacterial growth whereas others had biochemical properties

not suitable for further drug development (51, 52, 53). The production of NADP is carried out

by the third enzyme of the pathway, NAD kinase, that phosphorylates NAD by using ATP or

in some cases a poly(P) polymer as a phosphodonor (54, 55). The activity of the NAD kinases

was known for a long time but their genes were identified recently, leading to the discovery of

NadK orthologs in all living organisms including humans (56, 57). Ever since, this enzyme

has received a lot of attention due to its importance in the NADP metabolism and the fact that

differences between the human and the prokaryotic NadK have been observed (57). Studies

have led to the identification of some NAD homologues that can specifically inhibit the

bacterial NadK but further improvement of these inhibitors is needed in order to achieve a

concordance between in vitro enzymatic activity and antibacterial effect (58, 59).

AIM OF THE PROJECT

Construct strains that mimic the action of antibiotics targeting the enzymes of the

three essential biochemical pathways (MEP, NAD, CoA) that were chosen for this

study.

Identify the most potential antibiotic target genes of these three pathways.

Test the antibiotic Fosmidomycin against all the constructed strains of the MEP

pathway to observe if targeting Dxr and another enzyme of the pathway has a better

outcome than a mono-therapy.

10

MATERIALS AND METHODS

Strains and Growth conditions

All strains are derived from E. coli K12 MG1655. Bacteria were routinely grown in LB broth

or on LB agar (LA) at 37°C (30°C for strains that carry the pSIM5-Tet plasmid) with shaking

at 200rpm for liquid cultures under normal atmospheric conditions. When required, arabinose

(ara) was added to a final concentration of 0.1% unless otherwise specified. Antibiotics were

added to the cultures at the following concentrations: Tetracycline: 15 mg/L;

Chloramphenicol: 30 mg/L and Kanamycin: 50 mg/L.

Polymerase chain reaction

PCR was used to amplify genetic regions that were inserted in the E. coli genome by the λ red

recombineering method, regions that were sent for sequencing and also to verify the

insertion/deletion of a gene. Phusion High-Fidelity 2X PCR Master Mix with HF Buffer (New

England Biolabs) was used for the amplification of the genes that were used for λ red

recombination whereas Taq 2X PCR MasterMix (Thermo Scientific) was used for sequencing

and verification of insertion/deletion. The PCR reactions and the cycling scheme were

performed as listed in the following table by using an S1000TM

Thermal Cycler ( Bio-Rad

Laboratories, California) (Tables 1 and 2). The primers that were used for the amplification of

the genes are listed in the supplementary table. Correct amplification was checked using a 1%

agarose gel.

PCR Reaction (25 μl)

Phusion/TaqMasterMix 12.5 μl

H2O 9.5 μl

Forward primer

(10pmol/μl)

1 μl

Reverse primer

(10pmol/μl)

1 μl

DNA 1 μl

Table 1: Components used for the PCR reactions and the respective volumes.

PCR Cycling Scheme Steps Phusion MasterMix Taq MasterMix

1. Initial denaturation 98°C for 5 min 95°C for 5 min

2. Denaturation 98°C for 30 sec 95°C for 30 sec

3. Annealing 55°C for 30 sec 55°C for 30 sec

4. Extension 72°C for 1 min/kb 72°C for 1 min/kb

5. Repeat Steps 2,3,4 for 34 times Steps 2,3,4 for 34 times

6. Final extension 72°C for 5 min 72°C for 5 min

7. Hold 4°C for ∞ 4°C for ∞

8. END END END

Table 2: The PCR cycling scheme that was used according to the type of polymerase in the MasterMix.

11

Purification of the PCR products

The PCR products were purified in order to be used for lambda red recombineering and sent

for sequencing. The purification was performed according to the instructions of the QIAquick

PCR Purification Kit from Qiagen.

Plasmid extraction

For the plasmid extraction a 2 ml overnight culture of the E. coli strain CH6997 carrying the

pSIM5-tet plasmid was used. The pelleted bacterial cells were collected after centrifugation of

the culture and the plasmid purification was performed based on the instructions of the

QIAprep Spin Miniprep kit (Qiagen).

Plasmid transformation

pSIM5-tet plasmids were inserted in the strains that had the desirable genes under the control

of the arabinose promoter but also contained the same genes in the original positions. For the

insertion of the plasmids overnight cultures of the strains were used. The cultures were

centrifuged for 5 min to collect the cells, the supernatant was then discarded and the cells

were resuspended in 1ml of water followed by another 5 min centrifugation. The supernatant

was discarded again and the procedure of washing was repeated for 4 more times. After the

last discarder, the cells were resuspended in the water that was left in the tube. In a clean 1.5

ml microcentrifuge tube 50 μl of the recipient cells were added along with 5 μl of the purified

plasmid. The mixture was then transferred in electroporation cuvettes and electroporation

occurred by using a Gene PulserTM

(Bio-Rad Laboratories, California) according to the

following protocol : 1.8KV, 25μF and 200Ω. The cells were then resuspended in 1 ml LB and

recovered at 30°C for at least an hour. After recovery, 100μl of the cells were plated on LA

plates containing 15 mg/ml tetracycline and incubated overnight at 30°C.

λ Red Recombineering

Recombineering was used to construct the strains by inserting the gene of interest behind the

araBAD promoter and deleting it from the original position. The same technique was used for

the yellow fluorescent protein (yfp)- kanamycin (kan) cassette fusions.

Day 1

An overnight culture of the recipient strain, carrying the pSIM5-tet plasmid with the λ red

genes essential for the recombination, was started in 2 ml LB containing 15 mg/L tetracycline.

The culture was grown overnight shaking at 30°C.

Day 2

The overnight culture was diluted (1:100) in 50 mL LB containing 15 mg/L

tetracycline.

12

The culture was incubated in a 30°C shaking water bath for around 2,5 hours to a

culture density of approximately OD600 = 0,2 - 0,3 measured by a spectrophotometer.

The culture was transferred to a 43°C shaking water bath for 15 minutes to induce the

production of the λ red proteins and transferred to ice water-bath for an additional 15

minutes.

The electrocompetent cells were subsequently collected by centrifugation at 4500 x g

for 8 min at 4°C, using a Heraeus Labofuge 16R (Thermo Scientific, Germany).

The pellet was resuspended in 15 ml of ice-cold H2O and washed three times in total

by repetitive steps of centrifugation and resuspension.

After the last washing, the pellet was resuspended in 150μl of ice-cold H2O.

50 μl of the cell suspension was mixed with 5 μl of purified PCR product (for DIRex:

5 µL of each PCR product) and transferred into an electroporation cuvette. A Gene

PulserTM

( Bio-Rad Laboratories, California) was used for the electroporation

according to the following protocol : 1.8KV, 25μF and 200Ω.

After electroporation, the cells were recovered in 1 mL LB and incubated overnight at

37°C. In case of sucrose counter-selection, cells were recovered in 15 ml of Salt-free

LB shaking at 37°C.

Day 3

100 μl of the overnight cultures were plated on the appropriate selective media. The

plates were incubated overnight at 37°C.

Spot assay

Cultures in LB containing 0.5% glucose were started for the constructed strains (strains

having the gene of interest behind the araBAD promoter and carrying the deletion of the gene

in the original position) and incubated overnight at 37°C shaking. The overnight cultures were

washed with 0.9% NaCl and a 10-fold serial dilution was performed by serially adding 100 μl

of cells in 900 μl of 0.9% NaCl. Square plates with different arabinose concentrations (0.1%,

0.01%, 0.001%, 0.0001%) and plates containing 0.5% glucose were used for the assay. For

each dilution a spot of 5 μl was added onto the plates. The plates were incubated overnight at

37°C.

Bioscreen

Cultures in LB containing 0.5% glucose were started for the constructed strains (strains

having the gene of interest behind the araBAD promoter and carrying the deletion of the gene

in the original position) and incubated overnight at 37°C shaking. The overnight cultures were

washed with 0.9% NaCl and diluted 10 fold. 100 well “honeycomb” plates and media

containing arabinose and 0.5% glucose were used for the assay. Each well was filled with 300

μl of media and 1 μl of the diluted cultures. The instrument used for the measurements was

BioscreenC (Oy Growth Curves Ab Ltd ), and the run was set for 18 hours.

13

Minimal Inhibitory Concentration (MIC) assay

A MIC assay for all the constructed strains was also performed against the antibiotic

Fosmidomycin. 96-well Microtiter plate and media containing arabinose and 0.5% glucose

were used for the test. Overnight cultures of the strains were washed with 0.9% NaCl and 50

μl were tranferred to 5 ml of 0.9% NaCl. The diluted cultures were measured by using a

SensititreTM

Nephelometer (ThermoFisher Scientific) and culture or 0.9% NaCl were added

until obtaining the desired bacterial concentration of 1x108 CFU/ml. The diluted cultures were

further diluted 2 fold in selective media (final bacterial cells: 1x106

CFU). The antibiotic

concentrations used for the assay were 128: 64: 32: 16: 8: 4: 2: 1: 0 μg/ml. 100 μl of the

diluted antibiotic (256 μg/ml) were transferred in column 1 and a series dilution of the

compound was performed until column 11. 50 μl of media containing cells were filled from

column 2 to 11 and 100 μl of media containing cells in column 12. The final volume in each

well was 100 μl. The plates were incubated for ~ 20 hours at 37°C without shaking.

MACS Flow Cytometry

Flow cytometry was used to measure the fluorescence levels in the constructed strains

harboring the yfp-kan cassette. Overnight cultures were started for the constructed strains in

selective media and subsequently 1μl of each overnight culture was mixed with 200μl of LB

in a 96-well Microtiter plate. Fluorescence levels were determined by using magnetic-

activated cell sorting (MACSQuant VYB, Miltenyi Biotec). For each sample the average

fluorescence of 20.000 single cells was determined. All values are averages ± standard

deviation of three independent cultures.

Sequencing

Sequencing was performed by Macrogen (The Netherlands) using purified PCR products.

14

RESULTS

Construction of strains

Since the essential genes of the three pathways (MEP, NAD, CoA) are considered to be

potential antibiotic targets, strains that mimic the action of an antibiotic were constructed. For

this, potential target genes were moved behind the inducible araBAD promoter in order to

assess if there is a correlation between the amount of the respective proteins and the bacterial

growth.

Strains for all the essential genes of the three pathways were constructed by λ red

recombineering (Figure 4). Insertion of the gene of interest behind the araBAD promoter was

performed with sucrose counter-selection whereas the deletion of the gene from the original

position was achieved with the DIRex method (60, 61). Seven strains were constructed for the

MEP pathway (strains: dxsara

, dxrara

, ispDara

, ispEara

, ispFara

, ispGara

, ispHara

) and three strains

for the NAD pathway (strains: nadDara

, nadEara

, nadKara

). During the project, it was

discovered that the ispEara

strain had a mutation that led to constitutive expression of the

araBAD promoter and for this reason the strain was excluded from the results. Strains that

harbor the gene of interest behind the araBAD promoter have been constructed for the CoA

pathway (strains: coaAwt+ara

, coaDwt+ara

, coaEwt+ara

) but the deletion of the original gene has to

be performed. The strain for the second gene of the CoA pathway (coaBC) has not been

constructed yet.

Figure 4: Construction of the strains for the essential genes of the three pathways exemplified with the dxs gene.

The gene of interest is first inserted behind the araBAD promoter by λ red recombineering with sucrose counter-

selection (A, B). In another step, the gene of interest is deleted from the original position by using the DIRex

method (C).

15

Spot Assay

The constructed strains of the MEP and NAD pathways were tested with the spot assay to

observe if the different expression levels of the gene of interest have an effect on bacterial

growth. A wild-type E. coli strain (strain CH1464) was used as a control. For the MEP

pathway, a reduction in the expression levels of the dxr, ispD and ispF genes had no visible

effect on bacterial growth compared to the wild type and the strains could still grow in all the

media used for the assay. However, the colonies were getting smaller as the arabinose

concentration was decreasing which shows that they could not grow as fast as in the media

containing the highest arabinose concentration. The other three genes exhibited an impact on

the bacterial growth. Gene dxs showed a correlation between the arabinose concentration and

the bacterial growth which was more clear on the plates containing 0.001%, 0.0001%

arabinose and 0.5% glucose where there were no colonies of the diluted cultures.

Surprisingly, the two last genes of the MEP pathway, ispG and ispH were the ones that

showed the most significant reduction in bacterial growth as the levels of arabinose and

consequently the levels of gene expression were decreasing. The effect of the ispG became

obvious on the plate containing 0.01% arabinose whereas the effect of the ispH was visible

even on the plate with the highest arabinose concentration, 0.1%. Both strains gave almost no

growth on the plate containing 0.5% glucose (Figure 5, Supplementary Figure 1).

For the NAD pathway, nadD and nadE did not show any effect on bacterial growth when

tested on the different arabinose and glucose concentrations compared to the wild type strain.

These strains could efficiently grow in all media and the size of the colonies in the lowest

arabinose concentration was the same as in the highest. Nevertheless, low levels of expression

of the last gene of the pathway, nadK, led to a poor bacterial growth, which was more

apparent on the plate with the lowest arabinose concentration (0.0001%) and on the plate

containing glucose.

Figure 5: a) Results of the spot assay for the dxrara

strain. The strain was not affected by the decreasing

concentrations of arabinose and could still grow even in the lowest concentration of arabinose and in media with

glucose. The size of the colonies thought was getting smaller as the arabinose concentration was decreasing. b)

Results of the spot assay for the ispHara

strain. There was a gradual decrease in bacterial growth and a strong

effect in media containing 0.0001% ara and 0.5% glu.

Bioscreen

a) b)

16

The Bioscreen was used as an alternative method to verify the results of the Spot assay by

measuring the bacterial growth over time.

The results for the MEP pathway match the ones from the Spot assay, as the dxrara

, ispDara

and

ispFara

strains did not show significant changes in growth. The bacteria were growing equally

good in all media even though a reduction in growth was observed in media containing

glucose. This correlates perfectly with the spot assay findings in which the strains could grow

in glucose but much slower compared to the higher arabinose concentrations. However, dxsara

,

ispGara

and ispHara

strains were the ones that had a distinguishable reduction in growth as the

arabinose concentration was decreasing. It was more obvious for the ispHara

strain that gave

no bacterial growth in media containing less than 0.01% arabinose (Figure 6).

The NAD pathway results were different from the Spot assay as the strain nadKara

did not

seem to be affected by the concentration of arabinose and the bacterial growth remained at the

same levels even though a reduction in media containing glucose was observed. Nevertheless,

the other two strains, nadDara

and nadEara

, showed no significant changes in growth as

expected from the Spot assay and the bacterial growth remained the same over time in all the

media used for this assay (data not shown).

Figure 6: a) The bioscreen results for the dxrara

strain which showed that the strain could grow at the

same levels in all the different concentrations of arabinose b) The results for the ispHara

strain indicated

that the strain could grow in media with the highest concentration of arabinose and less efficiently in

a)

b)

17

media containing 0.01% arabinose. However, no growth was observed in the other three media (0.001%

ara, 0.0001% ara, 0.5% glu).

MIC Assay

A MIC assay was performed for the dxr

ara strain against the antibiotic Fosmidomycin that is

known to inhibit the Dxr protein and has a MIC of about 128 mg/L on a wild-type E. coli

strain. When tested against the dxrara

strain there was a clear correlation between the arabinose

concentrations and the MIC values, verifying Dxr as the original target of FSM. As the

arabinose concentration was declining, the MIC was decreasing as well reaching a MIC of 8

mg/L in media containing 0.0001% arabinose and 0.5% glucose.

To test if there were synergistic effects with other genes of the pathway, the rest of the MEP

pathway strains were tested against the antibiotic in various concentrations of arabinose and

in glucose. No changes in MIC were observed for the strains ispDara

and ispFara

which had a

MIC of >128 mg/L in all the different media, indicating that low levels of the respective

proteins do not affect the action of the antibiotic. On the other hand, a reduction in the MIC

was observed when the strain dxsara

was tested against FSM, starting with a MIC of 128 mg/L

and reaching 4 mg/L in the lowest arabinose concentration. This suggests that when the levels

of the Dxs protein are decreasing, the FSM MIC is decreasing as well, demonstrating that

there could be synergistic effects between FSM and an antibiotic that could target Dxs.

Results for strains ispGara

and ispHara

can’t be drawn by the MIC assay as these strains can’t

grow in antibiotic-free media with low concentrations of arabinose and glucose (Table 3).

Media

MIC (mg/L) for strains:

dxs dxr ispD ispF ispG ispH

0.1% ara > 128 > 128 > 128 > 128 > 128 >128

0.01% ara > 128 32 > 128 > 128 128 -

0.001% ara 16 16 > 128 > 128 - -

0.0001% ara 4 8 > 128 > 128 - -

0.5% glucose 4 8 > 128 > 128 - -

Table 3: Results of the susceptibility test for the strains of the MEP pathway against the antibiotic

Fosmidomycin.

Selection of ispG and ispH mutants

Since ispG and ispH appeared to be the most potential antibiotic target genes of the MEP

pathway, as concluded from the spot assay and bioscreen results, mutant selection was

performed to determine the mechanism of resistance.

18

Mutants were selected on LA plates containing 0.5% glucose, as these two strains were not

able to grow in this media. 7 and 12 mutants were selected for the ispG and ispH strains

respectively and sent for sequencing of the ara operon. The sequencing results showed that all

mutants harbored mutations in the araC gene, which encodes for the regulatory protein of the

operon (Table 4). Most mutations are repeated, like the amino acid change L10Q (6x), P11L

(2x), S14L (2x), V20M (2x) and L156I (2x). Most of the mutation found in this project, have

been shown to modify the structure of the N-terminal domain of the regulatory araC protein

leading to a constitutive gene expression even in the absence of arabinose from the media

(62). The fact that all the mutations were found in the araC gene implies that there is no

easier bypass mechanism such as amplification of the ispG or ispH genes to overcome the

depletion of the proteins.

Strain Nucleotide change Amino acid change

ispG 1 C41T S14L

ispG 2 G35A G12E

ispG 3 C41T S14L

ispG 3 C455A A152E

ispG 5 G58A V20M

ispG 6 C32T P11L

ispG 7 T29A L10Q

ispH 1 T56C L19P

ispH 2 T26A L9Q

ispH 3 T29A L10Q

ispH 4 C32T P11L

ispH 5 T29A L10Q

ispH 6 A53C H18P

ispH 7 T29A L10Q

ispH 8 G58A V20M

ispH 9 C466A L156I

ispH 10 T29A L10Q

ispH 11 C466A L156I

ispH 12 T29A L10Q

Table 4: Sequencing results for the selected ispG and ispH mutants.

Yfp-kan cassette fusions

In order to verify that the araBAD promoter functions equally well in all the strains

constructed for the MEP and NAD pathways, a yfp-kan cassette was inserted behind the

araBAD promoter and the gene of interest (Figure 7). The strains containing the yfp-kan

fusions were tested in all media combinations and the levels of fluorescence were measured

by Magnetic-activated cell sorting (MACS).

Based on the results obtained from this assay, the fluorescence levels corresponding to the

activity of the araBAD promoter were comparable in all the strains (Table 5). There was a

reduction in the levels as the arabinose concentration was decreasing whereas in media

containing glucose the fluorescence levels were around zero. Both observations indicate that

the promoter and its activity depend on arabinose and the concentration of it in the media.

Furthermore, all the strains gave similar and comparable fluorescence levels, validating the

results of the spot assay and the bioscreen. In particular ispG and ispH genes that were found

to be promising antibiotic target genes, were amongst the strains with the highest fluorescence

levels. This means that the results of the spot assay and the bioscreen were only influenced by

19

the concentration of arabinose used and not by malfunction of the promoter. Even though

some strains, like nadEara

,dxrara

and especially ispDara

, did not give as high fluorescence as

the others, which could be interpreted as lower gene expression, it did not affect the assays as

the strains could still grow in all media combinations.

Figure 7: Construction of strains containing the yfp-kan cassette exemplified with strain dxs

ara . The cassette

was inserted behind the araBAD promoter and the gene of interest in each case by λ Red Recombineering.

Media dxs dxr ispD ispF ispG ispH nadD nadE nadK

0,1% ara 162,74 124,73 15,74 396,03 260,59 296,74 335,15 102,79 188,53

0,01% ara 107,44 95,54 11,18 300,96 187,60 235,17 222,88 78,14 130,94

0,001% ara 67,74 57,91 5,91 176,23 100,35 137,90 137,06 47,43 76,23

0,0001% ara 8,80 8,58 0,52 20,77 12,12 16,10 16,01 4,94 8,40

0,5% glu 0,75 1,53 -0,35 0,30 0,09 0,80 0,56 -0,31 -0,27

Table 5: The fluorescence levels for all the constructed strains of the yfp-kan fusions measured in arbitrary units

(AU). The strains were tested in different concentrations of arabinose and in glucose and the fluorescence was

measured by MACS.

20

DISCUSSION

Living in a time where antibiotic resistance becomes more and more frequent, the

discovery of new antibiotics is the key to dealing with this problem and even targeting multi-

resistant pathogens that cannot be treated by the usual drugs of choice. Addressing this

serious issue, the present study was focused on finding new promising antibiotic targets

involved in three essential pathways in bacteria, the MEP, the NAD and the CoA pathways.

The strains constructed in this study were designed to mimic the effects of a potential

antibiotic on a specific target protein. As antibiotics bind to their targets they inhibit the

protein’s function leading to a reduction in effective protein concentration. To mimic this

effect, we cloned potential target genes behind the inducible araBAD promoter and removed

the gene from its original chromosomal location. Thus, protein levels could be down

regulated as a result of decreased arabinose concentration in the media. Since the construction

of strains for the CoA pathway had not been completed, only the MEP and NAD were

explored as potential target pathways. Two methods (Spot assay, Bioscreen) were examined

in order to identify the one that gives more clear results and will be used to test the strains in

different arabinose concentrations. The Bioscreen showed no effect in growth for the strains

of the NAD pathway whereas there was some effect for some strains of the MEP pathway,

which was visible at the end of the assay as all strains had almost the same growth rate at the

beginning. This resemblance in the growth rates between all strains might be a result of liquid

nature of the LB which supplies the cells with nutrients for a longer period of time compared

to the solid media (LA) used in the Spot assay. The Spot assay gave more clear results

concerning the plating efficiency and the size of the bacterial colonies on the plate and for this

reason it was selected as the most appropriate method to test the strains. Growth of all strains

was affected when expression levels of the proteins were reduced. In most cases the effect

was not bactericidal (equal number of colonies on the plates containing various concentrations

of arabinose) but the size of the colonies became smaller as expression levels were reduced

(Fig. 5a, supplementary Fig. 1). However, low levels of Dxs, IspG, IspH and NadK proteins

additionally reduced the number of colony forming units by up to six orders of magnitude in

low concentrations of arabinose and in glucose (Fig. 5b, supplementary Fig. 1). Dxs has been

suggested to be the rate-limiting step of the pathway in Synechococcus leopoliensis (63) and

IspG in M. tuberculosis (64) indicating that both enzymes have low reaction rates. Both

findings suggest that the protein depletion, resulting from the low arabinose concentrations,

has the strongest effect on proteins that catalyze slow reaction of a pathway.

Overall, the effect on bacterial growth was the strongest for IspG and IspH proteins,

making those two the most potential antibiotic target genes in this study. Nevertheless,

previous studies have not been focused on IspG and IspH due to the lack of protein crystal

structures from relevant pathogenic organisms (65). Furthermore, both belong to the family of

iron-sulfur proteins that are highly distributed in the nature and which raises concerns with

regard to selectivity (66). Even though ispG and ispH are considered the most promising, this

does not exclude the potentiality of the rest of the tested genes. Fosmidomycin, a

commercially available antibiotic, acts as an alternative substrate for the Dxr enzyme and is

active against a wide range of bacteria that belong to the Enterobacteriaceae family (67). The

findings in this study showed that low levels of the Dxr enzyme in the bacterial cell did not

reduce the number of colony forming units within the culture but still had a significant effect

on the bacterial growth (reduction in colony size). This indicates that not only genes that

21

demonstrate antibacterial effects but instead display a lower growth rate could also be

considered as potential target genes. In addition, this observation further strengthens the

choice of dxs, ispG, ispH and nadK as the most promising target genes of the two pathways

since it can be assumed that a much lower dose of a potential antibiotic should be needed to

observe the same antibacterial effect like in the case of FSM. A lower dose would increase the

window between effective antibacterial concentration and toxicity effects of the drug and

therefore be beneficial for treatments.

Another question that was explored during this study was if the constructed strains

could be used to test for synergistic effects between antibiotics that target different enzymes

within the pathway. An example for such an effect is the combinatory use of Trimethoprim

and Sulfamethoxazole that target two enzymes in the folate synthesis pathway (68). To assess

this, a Fosmidomycin MIC assay in the presence of different arabinose concentrations was

performed for each strain. The basic idea behind this test is that Fosmidomycin directly

targets Dxr and reduced arabinose concentrations mimics a drug targeting another protein.

The expectation was that the FSM MIC should be independent from the arabinose

concentration for proteins that display no synergistic effect with Dxr inhibition and a

reduction in FSM MIC in lower arabinose concentrations would be an indication for

synergistic effects. The results validated Dxr as a target for FSM as when the expression of

the dxr gene was decreasing, the FSM MIC was going down as well showing that there is a

clear correlation between the levels of the Dxr protein and the effectiveness of the antibiotic

(Table 3). When FSM was tested against strains ispDara

and ispFara

there was no change in the

MIC compared to the one of a wild-type E. coli strain, indicating that that there is no

synergistic effect between targeting Dxr and these two proteins. Conclusions for strains

ispGara

and ispHara

cannot be drawn by this assay because the strains did not grow in low

levels of arabinose (due lack of IspG and IspH production). Surprisingly, there was one strain

(dxsara

) that gave lower FSM MIC values as the arabinose concentration was decreasing,

starting with a wild-type E. coli MIC of 128 mg/L and reaching a MIC of 4 mg/L in media

with the lowest arabinose concentration. Dxs is the first essential enzyme of the MEP

pathway, converting pyruvate and G3P to 1-deoxy-D-xylulose 5-phosphate (DXP), which is

then used as a substrate by Dxr. A reduction in the expression levels of the dxs gene causes

less production of the Dxs enzyme, which in turn generates less DXP. Lower levels of DXP

reduce the turnover rate of DXP to MEP by Dxr protein and thus a lower Fosmidomycin

dosage is required to further inhibit the enzyme’s activity. These results indicate that the

development of a combinatory therapy targeting Dxs and Dxr could be a promising strategy.

When antibiotics are used frequently, some of the bacteria can acquire resistance by

using three main mechanisms. One way to resist the effects of the drug is to inhibit its action,

for example by expressing efflux pumps, which transport the antibiotic outside of the cell

(69). Alternatively, bacteria can modify the drug target so that it is not recognizable by the

antibiotic anymore (70). A third way to gain resistance is by finding bypass mechanisms, for

instance amplification or activation of genes, that overcome the effect of the drug (71). Since

ispG and ispH genes were found to be promising drug targets during this study and

considering the fact that the strains were the only ones that could not grow in media

containing glucose, mutants were selected to evaluate the type of the acquired mutation. This

assay cannot select for mutations that inhibit the antibiotic or alter the drug target because

antibiotic effects are only mimicked. Two types of mutations were potentially expected: (i)

Mutations that bypass the negative effect on growth, and (ii) mutations that overcome the

gene repression by the araBAD promoter. All the ispG and ispH mutants that were selected on

glucose plates carried point mutations in the araC gene (Table 4). Some mutations were

22

repetitive showing that the nucleotide changes in this gene were not random but specific. The

araC gene encodes for a protein, which acts as a regulator of the araBAD operon. In the

absence of arabinose from the media, the AraC protein forms a dimer that binds to the distant

regulatory DNA sites I1 and O2 and causes the formation of a loop. This structure inhibits the

RNA polymerase from binding to the araBAD promoter and the genes are not transcribed.

When arabinose is added in the media, it binds to the AraC protein dimer, changing its

formation and causing the loop to move to the adjacent I1 and I2 sites. The RNA polymerase is

then free to bind to the promoter and transcribe the genes of the operon (72). The point

mutations that occurred in the araC gene have been shown to modify the structure of the

protein, leading to a constitutive araBAD promoter (62). With this way both strains could

survive and grow in the presence of glucose. A phenotype that is caused by a small number of

specific point mutations generally has a mutation rate of ~ 10-10

per cell/generation (73). It

can therefore be assumed that there is no other more frequent way of altering the growth

inhibition since araC mutations are present in all the sequenced mutants. However, the

existence of a bypassing mechanism with a frequency lower than 10-10

per cell/generation is

possible, but it cannot be demonstrated in this case as the constitutive mutations are preferred.

The urgent need for the discovery of novel antibiotics has become the center of

attention of the scientific world for many decades since antibiotic resistance is increasing

dramatically. Biochemical pathways that are essential for the bacteria are considered to be

attractive drug targets and both the MEP and the NAD pathways that were explored during

this study belong to this category. Especially the MEP pathway has been explored more

thoroughly by many studies since its unique for the bacteria and is absent from humans (6).

However, apart from Fosmidomycin, which targets the Dxr enzyme, no other antibiotic has

been discovered for the other genes of the pathway. During this study, three genes of the MEP

(dxs, ispG, ipsH) and one gene (nadK) of the NAD pathway were demonstrated to be critical

for the bacterial growth and survival. In addition, the findings suggest that a combination

therapy targeting Dxs and Dxr at the same time can have a better outcome than a mono-

therapy. It is therefore proposed that more research should be conducted to identify inhibitors

that block the action of the these four essential genes and this might be a step towards the

discovery of a new, life-saving drug.

ACKNOWLEDGEMENTS

I would like to express my gratitude to Prof. Diarmaid Hughes for letting me perform my

thesis in his laboratory (Department of Medical Biochemistry and Microbiology, Uppsala

University) and for the help I received throughout the project. I also owe a special thanks to

Dr. Gerrit Brandis for making this thesis possible and for all the guidance and support that he

gave me during these six months of the project. Finally, I would like to thank Dr. Sha Cao for

her help and ideas.

23

REFERENCES

1. World Health Organization. Antimicrobial resistance: global report on surveillance. 2014.

Available from:

http://www.who.int/drugresistance/documents/surveillancereport/en .

2. Bach T, Rohmer M. Isoprenoid Synthesis in Plants and Microorganisms. New York, NY:

Springer; 2013.

3. Pérez-Gil J, Rodríguez-Concepción M. Metabolic plasticity for isoprenoid biosynthesis in

bacteria. Biochemical journal. 2013;452(1):19-25.

4. Rohmer M, Knani M, Simonin P, Sutter B, Sahm H. Isoprenoid biosynthesis in bacteria: a

novel pathway for the early steps leading to isopentenyl diphosphate. Biochemical

Journal. 1993;295(2):517-524.

5. Heuston S, Begley M, Gahan C, Hill C. Isoprenoid biosynthesis in bacterial pathogens.

Microbiology. 2012;158(Pt_6):1389-1401.

6. Lange B, Rujan T, Martin W, Croteau R. Isoprenoid biosynthesis: The evolution of two

ancient and distinct pathways across genomes. Proceedings of the National Academy of

Sciences. 2000;97(24):13172-13177.

7. Wang X, Dowd C. The Methylerythritol Phosphate Pathway: Promising Drug Targets in

the Fight against Tuberculosis. ACS Infectious Diseases. 2018;.

8. Proteau P. 1-Deoxy-d-xylulose 5-phosphate reductoisomerase: an overview. Bioorganic

Chemistry. 2004;32(6):483-493.

9. Jomaa H. Inhibitors of the Nonmevalonate Pathway of Isoprenoid Biosynthesis as

Antimalarial Drugs. Science. 1999;285(5433):1573-1576

10. Singh N, Cheve G, Avery M, McCurdy C. Targeting the Methyl Erythritol Phosphate

(MEP) Pathway for Novel Antimalarial, Antibacterial and Herbicidal Drug Discovery:

Inhibition of 1-Deoxy-D-Xylulose-5-Phosphate Reductoisomerase (DXR) Enzyme.

Current Pharmaceutical Design. 2007;13(11):1161-1177

11. Brown A, Parish T. Dxr is essential in Mycobacterium tuberculosis and fosmidomycin

resistance is due to a lack of uptake. BMC Microbiology. 2008;8(1):78.

12. Armstrong C, Meyers D, Imlay L, Freel Meyers C, Odom A. Resistance to the

Antimicrobial Agent Fosmidomycin and an FR900098 Prodrug through Mutations in the

Deoxyxylulose Phosphate Reductoisomerase Gene (dxr). Antimicrobial Agents and

Chemotherapy. 2015;59(9):5511-5519.

13. Sauret-Güeto S, Urós E, Ibáñez E, Boronat A, Rodríguez-Concepción M. A mutant

pyruvate dehydrogenase E1 subunit allows survival of Escherichia coli strains defective in

1-deoxy-d-xylulose 5-phosphate synthase. FEBS Letters. 2006;580(3):736-740.

14. Perez-Gil J, Uros E, Sauret-Güeto S, Lois L, Kirby J, Nishimoto M et al. Mutations in

Escherichia coli aceE and ribB Genes Allow Survival of Strains Defective in the First

Step of the Isoprenoid Biosynthesis Pathway. PLoS ONE. 2012;7(8):e43775.

15. Spry C, Kirk K, Saliba K. Coenzyme A biosynthesis: an antimicrobial drug target. FEMS

Microbiology Reviews. 2008;32(1):56-106.

16. Brown GM. The metabolism of pantothenic acid. J Biol Chem. 1959;234:370–378

17. Jackowski S. Biosynthesis of pantothenic acid and coenzyme A. In: Neidhardt FC, Curtiss

R, Gross CA, Ingraham JL, Lin ECC, Low KB, Magasanik B, Reznikoff W, Riley M,

Schaechter M, Umbarger HE, editors. Escherichia coli and Salmonella typhimurium:

cellular and molecular biology.American Society for Microbiology; Washington, D.C.:

1996. pp. 687–694

18. Kurtov D, Kinghorn J. The Aspergillus nidulans panB gene encodes ketopantoate

hydroxymethyltransferase, required for biosynthesis of pantothenate and Coenzyme A.

Molecular and General Genetics MGG. 1999;262(1):115-120.

19. Gerdes S, Scholle M, D'Souza M, Bernal A, Baev M, Farrell M et al. From Genetic

Footprinting to Antimicrobial Drug Targets: Examples in Cofactor Biosynthetic

Pathways. Journal of Bacteriology. 2002;184(16):4555-4572.

24

20. Dunn S.D., Snell E.E. Isolation of temperature-sensitive pantothenate kinase mutants of

Salmonella typhimurium and mapping of the coaA gene. Journal of Bacteriology.

1979;140(3):805-808.

21. Vallari D, Rock C. Isolation and characterization of temperature-sensitive pantothenate

kinase (coaA) mutants of Escherichia coli. Journal of Bacteriology. 1987;169(12):5795-

5800.

22. Calder R. B., Williams R.S, Ramaswamy G., Rock C.O., Campbell E., Unkles S.E,

Kinghorn J.R, Jackowski S. Cloning and characterization of a eukaryotic pantothenate

kinase gene (panK) from Aspergillus nidulans. J. Biol. Chem. 1999;274(4):2014-2020

23. Hong B, Yun M, Zhang Y, Chohnan S, Rock C, White S et al. Prokaryotic Type II and

Type III Pantothenate Kinases: The Same Monomer Fold Creates Dimers with Distinct

Catalytic Properties. Structure. 2006;14(8):1251-1261.

24. Jackowski S, Rock CO. Regulation of coenzyme A biosynthesis. Journal of Bacteriology.

1981; 148(3):926-32.

25. Reddy B, Landge S, Ravishankar S, Patil V, Shinde V, Tantry S et al. Assessment of

Mycobacterium tuberculosis Pantothenate Kinase Vulnerability through Target

Knockdown and Mechanistically Diverse Inhibitors. Antimicrobial Agents and

Chemotherapy. 2014;58(6):3312-3326.

26. Strauss E, Kinsland C, Ge Y, McLafferty F, Begley T. Phosphopantothenoylcysteine

Synthetase from Escherichia coli. Identification and characterization of the last

unidentified coenzyme A biosynthetic enzyme in bacteria. Journal of Biological

Chemistry. 2001;276(17):13513-13516.

27. Daugherty M, Polanuyer B, Farrell M, Scholle M, Lykidis A, de Crécy-Lagard V et al.

Complete Reconstitution of the Human Coenzyme A Biosynthetic Pathway via

Comparative Genomics. Journal of Biological Chemistry. 2002;277(24):21431-21439.

28. Patrone J, Yao J, Scott N, Dotson G. Selective Inhibitors of Bacterial

Phosphopantothenoylcysteine Synthetase. Journal of the American Chemical Society.

2009;131(45):16340-16341.

29. Geerlof A, Lewendon A, Shaw W. Purification and Characterization of

Phosphopantetheine Adenylyltransferase from Escherichia coli. Journal of Biological

Chemistry. 1999;274(38):27105-27111.

30. Aghajanian S, Worrall DM. Identification and characterization of the gene encoding the

human phosphopantetheine adenylyltransferase and dephospho-CoA kinase bifunctional

enzyme (CoA synthase). Biochem J. 2002; 365:13–18

31. de Jonge B, Walkup G, Lahiri S, Huynh H, Neckermann G, Utley L et al. Discovery of

Inhibitors of 4′-Phosphopantetheine Adenylyltransferase (PPAT) To Validate PPAT as a

Target for Antibacterial Therapy. Antimicrobial Agents and Chemotherapy.

2013;57(12):6005-6015.

32. Mishra P, Park PK, Drueckhammer DG. Identification of yacE (coaE) as the structural

gene for dephosphocoenzyme A kinase in Escherichia coli K-12. J Bacteriol. 2001;

183:2774–2778.

33. Silverman Richard B. The Organic Chemistry of Enzyme Catalyzed Reactions. Analytical

Biochemistry. 2001; 291(1):172.

34. Honjo T, Nishizuka Y, Hayaishi O. Diphtheria toxin-dependent adenosine diphosphate

ribosylation of aminoacyl transferase II and inhibition of protein synthesis. J. Biol.

Chem. 1968;243(12):3553–3555.

35. Singleton M, Håkansson K, Timson D, Wigley D. Structure of the adenylation domain of

an NAD+-dependent DNA ligase. Structure. 1999;7(1):35-42.

36. Bender DA, Olufunwa R. Utilization of tryptophan, nicotinamide and nicotinic acid as

precursors for nicotinamide nucleotide synthesis in isolated rat liver cells. Br J Nutr

1988;59:279–287.

37. Seifert J, Kunz N, Flachmann R, Läufer A, Jany K, Gassen H. Expression of the E. coli

nadB Gene and Characterization of the Gene Product L-Aspartate Oxidase. Biological

Chemistry Hoppe-Seyler. 1990;371(1):239-248.

25

38. Baba T, Ara T, Hasegawa M, Takai Y, Okumura Y, Baba M et al. Construction of

Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection.

Molecular Systems Biology. 2006;2(1).

39. Osterman AL, Begley TP. A subsystems-based approach to the identification of drug

targets in bacterial pathogens. Prog Drug Res. 2007;64:133-170.

40. Han S, Forman M, Loulakis P, Rosner M, Xie Z, Wang H et al. Crystal Structure of

Nicotinic Acid Mononucleotide Adenylyltransferase from Staphyloccocus aureus:

Structural Basis for NaAD Interaction in Functional Dimer. Journal of Molecular Biology.

2006;360(4):814-825.

41. Sershon V, Santarsiero B, Mesecar A. Kinetic and X-Ray Structural Evidence for

Negative Cooperativity in Substrate Binding to Nicotinate Mononucleotide

Adenylyltransferase (NMAT) from Bacillus anthracis. Journal of Molecular Biology.

2009;385(3):867-888.

42. Zhang H, Zhou T, Kurnasov O, Cheek S, Grishin N, Osterman A. Crystal Structures of E.

coli Nicotinate Mononucleotide Adenylyltransferase and Its Complex with Deamido-

NAD. Structure. 2002;10(1):69-79.

43. Sorci L, Cimadamore F, Scotti S, Petrelli R, Cappellacci L, Franchetti P et al. Initial-Rate

Kinetics of Human NMN-Adenylyltransferases: Substrate and Metal Ion Specificity,

Inhibition by Products and Multisubstrate Analogues, and Isozyme Contributions to

NAD+Biosynthesis. Biochemistry. 2007;46(16):4912-4922.

44. Zhou T, Kurnasov O, Tomchick DR, Binns DD, Grishin NV, Marquez VE, Osterman AL,

Zhang H. Structure of human nicotinamide/nicotinic acid mononucleotide

adenylyltranferase. Basis for the dual substrate specificity and activation of the oncolytic

agent tiazofurin. J Biol Chem. 2002;277:13148-13154.

45. Sorci L, Pan Y, Eyobo Y, Rodionova I, Huang N, Kurnasov O et al. Targeting NAD

Biosynthesis in Bacterial Pathogens: Structure-Based Development of Inhibitors of

Nicotinate Mononucleotide Adenylyltransferase NadD. Chemistry & Biology.

2009;16(8):849-861.

46. Allibert P, Willison JC, Vignais PM. Complementation of nitrogen-regulatory (ntr-like)

mutations in Rhodobacter capsulatus by an Escherichia coli gene: Cloning and sequencing

of the gene and characterization of the gene product. J Bacteriol. 1987;169: 260–271.

47. Nessi C, Albertini AM, Speranza ML, Galizzi A. The outB gene of Bacillus subtilis codes

for NAD synthetase. J Biol Chem. 1995;270: 6181–6185.

48. Zerez CR, Wong MD, Tanaka KR. Partial purification and properties of nicotinamide

adenine dinucleotide synthetase from human erythrocytes: Evidence that enzyme activity

is a sensitive indicator of lead exposure. Blood. 1990;75: 1576–1582.

49. Bellinzoni M, Buroni S, Pasca MR, Guglierame P, Arcesi F, De Rossi E, Riccardi G.

Glutamine amidotransferase activity of NAD+ synthetase from Mycobacterium

tuberculosis depends on an amino-terminal nitrilase domain. Res Microbiol.

2005;156: 173–177.

50. Hara N, Yamada K, Terashima M, Osago H, Shimoyama M, Tsuchiya M. Molecular

Identification of Human Glutamine- and Ammonia-dependent NAD Synthetases. Journal

of Biological Chemistry. 2003;278(13):10914-10921.

51. Velu S, Cristofoli W, Garcia G, Brouillette C, Pierson M, Luan C et al. Tethered Dimers

as NAD Synthetase Inhibitors with Antibacterial Activity. Journal of Medicinal

Chemistry. 2003;46(15):3371-3381.

52. Moro W, Yang Z, Kane T, Brouillette C, Brouillette W. Virtual screening to identify lead

inhibitors for bacterial NAD synthetase (NADs). Bioorganic & Medicinal Chemistry

Letters. 2009;19(7):2001-2005.

53. Wang X, Ahn Y, Lentscher A, Lister J, Brothers R, Kneen M et al. Design, synthesis, and

evaluation of substituted nicotinamide adenine dinucleotide (NAD + ) synthetase

inhibitors as potential antitubercular agents. Bioorganic & Medicinal Chemistry Letters.

2017;27(18):4426-4430.

26

54. Garavaglia S, Raffaelli N, Finaurini L, Magni G, Rizzi M. A Novel Fold Revealed by

Mycobacterium tuberculosis NAD Kinase, a Key Allosteric Enzyme in NADP

Biosynthesis. Journal of Biological Chemistry. 2004;279(39):40980-40986.

55. Sakuraba H, Kawakami R, Ohshima T. First Archaeal Inorganic Polyphosphate/ATP-

Dependent NAD Kinase, from Hyperthermophilic Archaeon Pyrococcus horikoshii:

Cloning, Expression, and Characterization. Applied and Environmental Microbiology.

2005;71(8):4352-4358

56. Kawai S, Mori S, Mukai T, Suzuki S, Yamada T, Hashimoto W et al. Inorganic

Polyphosphate/ATP-NAD Kinase of Micrococcus flavus and Mycobacterium tuberculosis

H37Rv. Biochemical and Biophysical Research Communications. 2000;276(1):57-63.

57. Lerner F, Niere M, Ludwig A, Ziegler M. Structural and Functional Characterization of

Human NAD Kinase. Biochemical and Biophysical Research Communications.

2001;288(1):69-74.

58. Gelin M, Poncet-Montange G, Assairi L, Morellato L, Huteau V, Dugué L et al. Screening

and In Situ Synthesis Using Crystals of a NAD Kinase Lead to a Potent

Antistaphylococcal Compound. Structure. 2012;20(6):1107-1117

59. Paoletti J, Assairi L, Gelin M, Huteau V, Nahori M, Dussurget O et al. 8-Thioalkyl-

adenosine derivatives inhibit Listeria monocytogenes NAD kinase through a novel

binding mode. European Journal of Medicinal Chemistry. 2016;124:1041-1056.

60. Gay P, Le Coq D, Steinmetz M, Berkelman T, Kado CI. Positive selection procedure for

entrapment of insertion sequence elements in gram-negative bacteria. Journal of

bacteriology. 1985;164(2):918–21. pmid:2997137

61. Näsvall J. Direct and Inverted Repeat stimulated excision (DIRex): Simple, single-step,

and scar-free mutagenesis of bacterial genes. PLOS ONE. 2017;12(8):e0184126.

62. Dirla S., Chien J. and Schleif R. (2009). Constitutive Mutations in the Escherichia coli

AraC Protein. Journal of Bacteriology, 191(8), pp.2668-2674.

63. Miller B, Heuser T, Zimmer W. Functional involvement of a deoxy-D-xylulose 5-

phosphate reductoisomerase gene harboring locus of Synechococcus leopoliensis in

isoprenoid biosynthesis. FEBS Letters. 2000;481(3):221-226.

64. Brown A, Eberl M, Crick D, Jomaa H, Parish T. The Non mevalonate Pathway of

Isoprenoid Biosynthesis in Mycobacterium tuberculosis Is Essential and Transcriptionally

Regulated by Dxs. Journal of Bacteriology. 2010;192(9):2424-2433.

65. Masini, T. and Hirsch, A. (2014). Development of Inhibitors of the 2C-Methyl-d-

erythritol 4-Phosphate (MEP) Pathway Enzymes as Potential Anti-Infective

Agents. Journal of Medicinal Chemistry, 57(23), pp.9740-9763.

66. Masini, T., Kroezen, B. and Hirsch, A. (2013). Druggability of the enzymes of the non-

mevalonate-pathway. Drug Discovery Today, 18(23-24), pp.1256-1262.

67. Greenwood D, Fosfomycin and Fosmidomycin. In: Finch R, Greenwood D, Norrby S,

Whitley R. Antibiotic and Chemotherapy. 9th ed. 2012. p. 259-261.

68. Poe, M. (1976). Antibacterial synergism: a proposal for chemotherapeutic potentiation

between trimethoprim and sulfamethoxazole. Science, 194(4264), pp.533-535.

69. McMurry L, Petrucci R, Levy S. Active efflux of tetracycline encoded by four genetically

different tetracycline resistance determinants in Escherichia coli. Proceedings of the

National Academy of Sciences. 1980;77(7):3974-3977.

70. Haenni M, Galofaro L, Ythier M, Giddey M, Majcherczyk P, Moreillon P et al. Penicillin-

Binding Protein Gene Alterations in Streptococcus uberis Isolates Presenting Decreased

Susceptibility to Penicillin. Antimicrobial Agents and Chemotherapy. 2010;54(3):1140-

1145.

71. Sandegren L, Andersson D. Bacterial gene amplification: implications for the evolution of

antibiotic resistance. Nature Reviews Microbiology. 2009;7(8):578-588.

72. Hamilton E, Lee N. Three binding sites for AraC protein are required for autoregulation of

araC in Escherichia coli. Proceedings of the National Academy of Sciences.

1988;85(6):1749-1753.

73. Drake J. A constant rate of spontaneous mutation in DNA-based microbes. Proceedings of

the National Academy of Sciences. 1991;88(16):7160-7164.

27

Supplementary Material

Name Primer sequence (5’→3’) Type dfp_fw ACTGTTTCTCCATACCCGTTTTTTTGGATGGAGTGAAACGATGAGCCTGGCCGG

TAAAAA

A

dfp_rv AGCCTGGTTTCGTTTGATTGGCTGTGGTTTTATACAGTCATTAACGTCGATTTTT

TTCAT

A

coaD_fw ACTGTTTCTCCATACCCGTTTTTTTGGATGGAGTGAAACGATGCAAAAACGGGC

GATTTA

A

coaD_rv AGCCTGGTTTCGTTTGATTGGCTGTGGTTTTATACAGTCACTACGCTAACTTCGC

CATCA

A

coaE_fw ACTGTTTCTCCATACCCGTTTTTTTGGATGGAGTGAAACGATGAGGTATATAGTT

GCCTT

A

coaE_rv AGCCTGGTTTCGTTTGATTGGCTGTGGTTTTATACAGTCATTACGGTTTTTCCTG

TGAGA

A

nadD_fw ACTGTTTCTCCATACCCGTTTTTTTGGATGGAGTGAAACGATGAAATCTTTACAG

GCTCT

A

nadD_rv AGCCTGGTTTCGTTTGATTGGCTGTGGTTTTATACAGTCATCAGCGATACAAGCC

TTGTT

A

nadE_fw ACTGTTTCTCCATACCCGTTTTTTTGGATGGAGTGAAACGATGACATTGCAACA

ACAAAT

A

nadE_rv AGCCTGGTTTCGTTTGATTGGCTGTGGTTTTATACAGTCATTACTTTTTCCAGAA

ATCAT

A

nadK_fw ACTGTTTCTCCATACCCGTTTTTTTGGATGGAGTGAAACGATGAATAATCATTTC

AAGTG

A

nadK_rv AGCCTGGTTTCGTTTGATTGGCTGTGGTTTTATACAGTCATTAGAATAATTTTTT

TGACC

A

Ddfp_fw AAAAATCATCATGAGCCTGGCCGGTAAAAAAATCGTTCTCCCGCTCGATCCGGT

GCGTTAGTGTAGGCTGGAGCTGCTTC

B

Ddfp_rv CGGAGCTGTGATTAGAGATATAACGCACCGGATCGAGCGGGAGAACGATTTTT

TTACCGGGTGTAGGCTGGAGCTGCTTC

B

DcoaD_fw TGAGGTTGTTATGCAAAAACGGGCGATTTATCCGGGTACTCATCAGGCGCTGAT

GGCGAAGTGTAGGCTGGAGCTGCTTC

B

DcoaD_rv GGCATAAACGCTACGCTAACTTCGCCATCAGCGCCTGATGAGTACCCGGATAAA

TCGCCCGTGTAGGCTGGAGCTGCTTC

B

DcoaE_fw TGGGTCTGCATTACGGTTTTTCCTGTGAGACAAACTGCGAGCCTCCCGTTAAGG

CAACTAGTGTAGGCTGGAGCTGCTTC

B

DcoaE_rv AATATAAATTATGAGGTATATAGTTGCCTTAACGGGAGGCTCGCAGTTTGTCTC

ACAGGAGTGTAGGCTGGAGCTGCTTC

B

DnadD_fw ACGACAGGTATCAGCGATACAAGCCTTGTTGGTTAATGTAGCCGCCAAACAGA

GCCTGTAGTGTAGGCTGGAGCTGCTTC

B

DnadD_rv GACGGTTGATATGAAATCTTTACAGGCTCTGTTTGGCGGCTACATTAACCAACA

AGGCTTGTGTAGGCTGGAGCTGCTTC

B

DnadE_fw GAGGGGTTCAATGACATTGCAACAACAAATAATAAAGGCGACCGTTTTCGATG

ATTTCTGGTGTAGGCTGGAGCTGCTTC

B

DnadE_rv GTGCAAATTATTACTTTTTCCAGAAATCATCGAAAACGGTCGCCTTTATTATTTG

TTGTTGTGTAGGCTGGAGCTGCTTC

B

DnadK_fw CCTCGGAAAAATGAATAATCATTTCAAGTGTATTGGCATTAAGCTCGGCTGGTC

AAAAAAGTGTAGGCTGGAGCTGCTTC

B

DnadK_rv TGGCGTAAAATTAGAATAATTTTTTTGACCAGCCGAGCTTAATGCCAATACACT

TGAAATGTGTAGGCTGGAGCTGCTTC

B

PCR dfp_fw TTTTTCACCTTCACCTCCTT C

PCR dfp_rv AATATGAATCGCCAGCCC C

PCR coaD_fw GGGCCGCATACTTTTAACT C

PCR coaD_rv GTGTTGCTGCGTTCGATT C

PCR coaE_fw GCCGGTCAAGTTCTTTCA C

PCR coaE_rv ACGCTATCATACGCACTAAT C

PCR nadD_fw CCAGATCAAAGGAAACACCA C

PCR nadD_rv AACGGAACTCACCCTCAAA C

PCR nadE_fw TCTTGTTCATATTCCGTCCT C

PCR nadE_rv CTCCGAATGCTGGTTATCT C

PCR nadK_fw GGGGATCAGTTTCAGGGT C

PCR nadK_rv CCGGTCTCGCCAGTTATT C

HP580 CACGGCAGAAAAGTCCACATTG D,E

28

A: primers used to insert the gene after the ara promoter

B: primers used for the deletion of the gene in the original position (DIRex method)

C: primers used to verify the deletion of the original gene

D: universal primers for the amplification of the ara location

E: primers for the amplification and sequencing of the araC and the ara promoter

F: primers used for the yfp-kan cassette fusions

Supplementary Table 1: List of the primers used for insertions and deletions of genes as well as for

sequencing.

HP581 ATAAAACCTGCCTGCGCCAC D,E

araC_fw AAAAAGCGTCAGGTAGGA E

araC_rv CCAGAAAGGCCACCAACA E

dxs yfp-kan fw CGCTGGTATGGAAGCCAAAATCAAGGCCTGGCTGGCATAATACAGAAAGAGGA

GAAATACTAGATGGTTAGCAAGGGCGAAGAACT

F

dxr yfp-kan fw TGAAGTCGCCAGAAAAGAGGTGATGCGTCTCGCAAGCTGATACAGAAAGAGGA

GAAATACTAGATGGTTAGCAAGGGCGAAGAACT

F

ispD yfp-kan fw GTTTTACCTCACCCGAACCATCCATCAGGAGAATACATAATACAGAAAGAGGA

GAAATACTAGATGGTTAGCAAGGGCGAAGAACT

F

ispE yfp-kan fw AGGCGCTAATCTTTCCCCATTGCACAGAGCCATGCTTTAATACAGAAAGAGGAG

AAATACTAGATGGTTAGCAAGGGCGAAGAACT

F

ispF yfp-kan fw CTGTGAAGCGGTGGCGCTACTCATTAAGGCAACAAAATGATACAGAAAGAGGA

GAAATACTAGATGGTTAGCAAGGGCGAAGAACT

F

ispG yfp-kan fw CGAAGCGCGTCGAATTGACGTTCAGCAGGTTGAAAAATAATACAGAAAGAGGA

GAAATACTAGATGGTTAGCAAGGGCGAAGAACT

F

ispH yfp-kan fw GCCGAAAGAGCTGCGTGTCGATATTCGTGAAGTCGATTAATACAGAAAGAGGA

GAAATACTAGATGGTTAGCAAGGGCGAAGAACT

F

nadD yfp-kan fw GGTACTGACTTACATTAACCAACAAGGCTTGTATCGCTGATACAGAAAGAGGA

GAAATACTAGATGGTTAGCAAGGGCGAAGAACT

F

nadE yfp-kan fw TCCGCCAATTACCGTTTTCGATGATTTCTGGAAAAAGTAATACAGAAAGAGGAG

AAATACTAGATGGTTAGCAAGGGCGAAGAACT

F

nadK yfp-kan fw ATTAAGCACCAAGCTCGGCTGGTCAAAAAAATTATTCTAATACAGAAAGAGGA

GAAATACTAGATGGTTAGCAAGGGCGAAGAACT

F

yfp-kan rv AGCCTGGTTTCGTTTGATTGGCTGTGGTTTTATACAGTCACATATGAATATCCTC

CTTAG

F

29

Supplementary Figure 1: Results of the spot assay for strains of the MEP and NAD pathways. No

antibacterial effect was observed for strains ispDara

, ispFara

, nadDara

and nadEara

but smaller colonies

were obtained for the first two as the arabinose concentration was decreasing. A clear inhibition of

bacterial growth at lower arabinose concentrations was noticed for strains dxsara

, ispGara

and nadKara

.