A high-throughput screening for inhibitors of riboflavin synthase identifies novel antimicrobial compounds to treat brucellosis

Article abstract of DOI:10.1111/febs.14829

Brucella spp. are pathogenic intracellular Gram-negative bacteria adapted to life within cells of several mammals, including humans. These bacteria are the causative agent of brucellosis, one of the zoonotic infections with the highest incidence in the world and for which a human vaccine is still unavailable. Current therapeutic treatments against brucellosis are based on the combination of two or more antibiotics for prolonged periods, which may lead to antibiotic resistance in the population. Riboflavin (vitamin B2) is biosynthesized by microorganisms and plants but mammals, including humans, must obtain it from dietary sources. Owing to the absence of the riboflavin biosynthetic enzymes in animals, this pathway is nowadays regarded as a rich resource of targets for the development of new antimicrobial agents. In this work, we describe a high-throughput screening approach to identify inhibitors of the enzymatic activity of riboflavin synthase, the last enzyme in this pathway. We also provide evidence for their subsequent validation as potential drug candidates in an in?vitro brucellosis infection model. From an initial set of 44?000 highly diverse low molecular weight compounds with drug-like properties, we were able to identify ten molecules with 50% inhibitory concentrations in the low micromolar range. Further Brucella culture and intramacrophagic replication experiments showed that the most effective bactericidal compounds share a 2-Phenylamidazo[2,1-b][1,3]benzothiazole chemical scaffold. Altogether, these findings set up the basis for the subsequent lead optimization process and represent a promising advancement in the pursuit of novel and effective antimicrobial compounds against brucellosis.

Full text of DOI:10.1111/febs.14829

Received Date : 11-Jan-2019
Revised Date : 26-Feb-2019
Accepted Date : 29-Mar-2019
A high-throughput screening for inhibitors of riboflavin synthase identifies
novel antimicrobial compounds to treat brucellosis
María Inés Serer 1‡, Mariela del Carmen Carrica 1‡§, Jörg Trappe 2, Sandra López Romero 2,
Hernán Ruy Bonomi 1#, Sebastián Klinke 1,3, María Laura Cerutti 1,3, Fernando Alberto
Goldbaum 1,3*
1 Fundación Instituto Leloir, IIBBA-CONICET, Av. Patricias Argentinas 435 (C1405BWE)
Buenos Aires, Argentina.
2
Novartis Institutes for BioMedical Research, Novartis Campus, Fabrikst. 2 (CH-4056)
Basel, Switzerland.
3
Plataforma Argentina de Biología Estructural y Metabolómica PLABEM, Av. Patricias
Argentinas 435 (C1405BWE) Buenos Aires, Argentina.
§
Current address: Centro de Investigación y Desarrollo en Fermentaciones Industriales,
CINDEFI-CONICET, Facultad de Ciencias Exactas, Universidad Nacional de La Plata, Calle
50 Nº 227 (B1900AJL) La Plata, Argentina.
#
Current address: Centre de Biochimie Structurale, CNRS UMR5048, INSERM U1054,
Université de Montpellier, 29 rue de Navacelles (34090) Montpellier, France.
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/febs.14829
This article is protected by copyright. All rights reserved.

‡ These authors contributed equally to this work.
* To whom correspondence should be addressed: E-mail: fgoldbaum@leloir.org.ar.
Fundación Instituto Leloir, IIBBA-CONICET, Av. Patricias Argentinas 435 (C1405BWE)
Buenos Aires, Argentina. Tel.: 54-11-5238-7500; Fax: 54-11-5238-7501. Departamental
website: www.leloir.org.ar/goldbaum-en.
Running title
HTS targeting of Brucella riboflavin synthase
Article type : Original Article
Keywords
Brucellosis; Riboflavin synthase; Flavin synthesis; High-throughput screening; Drug
development.
Abbreviations
RS, Riboflavin synthase; LS, Lumazine synthase; HTS, High-throughput screening.
Conflicts of Interest
The authors declare no conflicts of interest.
This article is protected by copyright. All rights reserved.

Abstract
Brucella spp. are pathogenic intracellular Gram‐negative bacteria adapted to life within cells
of several mammals, including humans. These bacteria are the causative agent of brucellosis,
one of the zoonotic infections with highest incidence in the world and for which a human
vaccine is still unavailable. Current therapeutic treatments against brucellosis are based on
the combination of two or more antibiotics for prolonged periods, which may lead to
antibiotic resistance in the population. Riboflavin (vitamin B2) is biosynthesized by
microorganisms and plants but mammals, including humans, must obtain it from dietary
sources. Owing to the absence of the riboflavin biosynthetic enzymes in animals, this
pathway is nowadays regarded as a rich resource of targets for the development of new
antimicrobial agents. In this work, we describe a high-throughput screening approach to
identify inhibitors of the enzymatic activity of riboflavin synthase, the last enzyme in this
pathway. We also provide evidence for their subsequent validation as potential drug
candidates in an in vitro brucellosis infection model. From an initial set of 44,000 highly
diverse low molecular weight compounds with drug-like properties, we were able to identify
ten molecules with 50% inhibitory concentrations in the low micromolar range. Further
Brucella culture and intramacrophagic replication experiments showed that the most effective
bactericidal compounds share a 2-Phenylamidazo[2,1-b][1,3]benzothiazole chemical
scaffold. Altogether, these findings set up the basis for the subsequent lead optimization
process and represent a promising advancement in the pursuit of novel and effective
antimicrobial compounds against brucellosis.
Introduction
Brucellosis is one of the major zoonotic infections worldwide, causing devastating economic
losses to the livestock industry and deleterious effects in humans [1-3]. It is caused by the
pathogenic intracellular Gram-negative bacteria Brucella spp. and is transmitted to humans
This article is protected by copyright. All rights reserved.

through direct contact with infected animals or by consumption of infected, unpasteurized
animal-derived products. The disease is essentially endemic in marginalized populations in
low-income countries, reason why the World Health Organization (WHO) has recently
classified it as a neglected zoonotic disease [4].
With an estimated rate of 500,000 infections per year [5], there is still no available vaccine
against human brucellosis. Current recommended antibiotic treatments involve prolonged
administration regimens with two or more antimicrobial drugs to prevent relapses, which can
cause undesired side effects on the patients and lead to antibiotic resistance in the population
[2, 6]. In addition, brucellosis may be complicated by life-threatening conditions such as
endocarditis, neurobrucellosis or osteoarticular infections, which may require even longer or
more aggressive therapeutic treatments [7, 8]. Therefore, to deal with all these issues there is
a need for the development of new and effective antimicrobials against brucellosis.
Riboflavin (vitamin B2) is the precursor for the biosynthesis of flavin mononucleotide (FMN)
and flavin adenine dinucleotide (FAD) [9], two key redox cofactors involved in a variety of
cellular metabolic processes [10]. While bacteria, plants and fungi are able to biosynthesize
riboflavin, humans and animals must obtain it from their diet, positioning riboflavin
biosynthetic enzymes and regulatory factors as promising specific targets for antimicrobial
therapies [11]. In fact, riboflavin biosynthesis has been shown to be essential in many
pathogenic bacteria that lack riboflavin transporters, such as Mycobacterium tuberculosis,
Brucella abortus and some Salmonella and Escherichia species [12-16].
This article is protected by copyright. All rights reserved.

The biosynthetic riboflavin pathway has been described in detail [9, 17-19]. Lumazine
synthase (LS, EC database: 2.5.1.78) catalyzes the penultimate step, allowing the
condensation of 5-amino-6-ribitylamino-2,4(1H,3H)-pyrimidinedione with 3,4-dihydroxy-2-
butanone-4-phosphate to generate 6,7-dimethyl-8-ribityllumazine. Riboflavin synthase (RS,
EC database: 2.5.1.9) catalyzes the last step, which corresponds to the dismutation of two
molecules of 6,7-dimethyl-8-ribityllumazine (a and b) to yield one molecule of riboflavin c,
with the regeneration of one molecule of 5-amino-6-ribitylamino-2,4(1H,3H)-
pyrimidinedione d (Fig. 1A).
Our laboratory has studied in detail these enzymes from B. abortus. For the former, we have
identified two isoenzymes (RibH1 and RibH2/BLS) and characterized their tertiary and
quaternary structures as well as their biochemical properties [20, 21]. We have also assayed
the relevance of the RibH enzymes in bacterial infection and demonstrated that the presence
of at least one functional LS enzyme is essential for intracellular survival and that
RibH2/BLS is necessary for mice colonization. Moreover, RibH2/BLS constitutes a major
candidate for the development of a subunit vaccine against brucellosis [14, 22-24]. On the
other hand, RS has been proposed as another important immunogenic candidate protein for
the development of a brucellosis subunit vaccine [25]. More recently, our group has
described the crystallographic structure of RS, both as apo-protein and in complex with a few
ligands of interest, namely one of the products of the reaction (riboflavin) and two product
analogues [roseoflavin and 5-nitro-6-ribitylamino-2,4(1H,3H)-pyrimidinedione], which lead
to the description of its active site (Fig. 1B) [26].
This article is protected by copyright. All rights reserved.

The structure-based design and synthesis of compounds with inhibitory activity against LS
and RS have long been studied [27-36]. However, those compounds are very polar and thus
not suitable for drug development. Interestingly, a high-throughput screening (HTS) hit
compound targeting M. tuberculosis RS demonstrated antibiotic activity in vitro, although an
off-target effect could not be totally discarded [37]. We here describe the development of a
HTS assay for the identification of compounds with inhibitory activity against Brucella spp.
RS and validate them in an in vitro model of brucellosis infection. We have identified a series
of five hit compounds with drug-like properties sharing a common 2-Phenylamidazo[2,1-
b][1,3]benzothiazole (PABT) chemical scaffold. These compounds also displayed in vivo
bactericide activity against B. abortus and impaired bacterial replication in macrophagic
cells. These results constitute a promising starting point for the rational design of effective
drugs against brucellosis and other relevant infectious diseases.
Results
Development of a Mobility shift-based screening assay (MSA) for RS inhibitors
We developed a MSA based on the competitive binding of both riboflavin and inhibitor
compounds to the active site of the enzyme. The screening uses microfluidic technology and
is based on the differential migration of substrate and product in an electrophoretic
separation. Thus, the presence of compounds with RS inhibitory activity was detected as a
decrease in the fluorescence signal of riboflavin.
The MSA primary screening was performed over a highly diverse library of 44,000 low
molecular weight compounds with drug-like properties. This screening was run under initial
velocity conditions previously determined by both HPLC and the microfluidic system (Fig. 2
This article is protected by copyright. All rights reserved.

and 3). A fluorescent peptide (JAK3) was used as an internal standard to compensate for the
weak fluorescence of the substrate and to quantify product formation. A primary output of the
run is shown in Fig. 4 as an example. The assay was developed in an off-chip mode and at a
single compound concentration of 30 μM. As the hit calling criterion, the threshold value was
set to 30% inhibition, resulting in a total of 163 hits (Fig. 5). Subsequently, the primary
screening was validated by determining IC50 values by performing dose-response curves.
Table 1 shows the 10 most potent compounds found by this approach. The IC50 values range
between 2 and 32 μM, with IA-79-AF08 (inhibitor 1) being the most potent with an IC50 of
2.4 µM. The dose-response curves obtained for IA-79-AF08 and the other selected
compounds are shown in Fig. 6.
Structural analysis of the hits
The chemical structures of the 10 compounds with the highest RS inhibitory activities in the
dose-response secondary screening are shown in Fig. 7. According to their structure, these
molecules can be clustered into two groups. The first group, which is presented in the upper
row of Fig. 7 (inhibitors 4, 5, 6, 9 and 10), comprise compounds that share a common
chemical scaffold, namely a 2-Phenylamidazo[2,1-b][1,3]benzothiazole (PABT) core formed
by a heteroaromatic system of three rings linked to a phenyl group and two principal
substituents: an alkyl or ethoxide group at position R1 and an aliphatic group with a terminal
tertiary amine at position R2 (Fig. 8). On the other hand, the bottom row of Fig. 7 shows the
rest of the studied compounds (inhibitors 1, 2, 3, 7 and 8), without major chemical
similarities, and which includes the most potent of the in vitro series, namely IA-79-AF08.
This article is protected by copyright. All rights reserved.

Nowadays, the probability of a small molecule to become an effective drug is established by
the Lipinski “rules of five” and other criteria that evaluate the physicochemical properties of
the compounds according to: (i) molecular weight < 500 Da; (ii) lipophilicity index clogP <
5; (iii) number of hydrogen bond acceptors (HBA) < 10; (iv) number of hydrogen bond
donors (HBD) < 5; (v) number of rotatable bonds (nRot) < 10 and (vi) the bioavailability
descriptor topological polar surface area (tPSA) < 140 Å2 [38-41]. A comprehensive analysis
of the physicochemical properties of the selected hits revealed that, while compounds 1, 2, 7
and 8 presented slightly higher values in some of the properties, the PABT subset fulfilled the
criteria established for drug-like candidates (Table 2). In particular, the PABT compounds
presented lower tPSA values (< 90 Å2) suggesting that they might display higher membrane
passive diffusion rates in vivo [42].
Inhibition of B. abortus growth in culture
The inhibition of the RS activity in vivo is expected to abolish the capacity of Brucella to
grow in a regular medium that does not contain high amounts of riboflavin. Thus, we
evaluated whether the 10 selected RS inhibitors had an antibacterial effect on B. abortus on
liquid cultures. We observed that the five compounds of the PABT subset were effective in
decreasing the bacterial growth throughout the entire duration of the experiment (Fig. 9A).
Remarkably, cultures treated with inhibitor 4, 6 or 10 completely abolished bacterial growth.
On the other side 1, the compound that showed the highest in vitro inhibition activity on RS,
was unable to inhibit Brucella growth. This observation may be due to the fact that inhibitor
1, along with 2, 7 and 8, have higher tPSA values that could correlate with a reduced
permeability to cross the bacterial membrane.
This article is protected by copyright. All rights reserved.

In order to determine the inhibitory potency of each of the compounds of the PABT subset,
dose-response curves on B. abortus cultures were performed. Fig. 9B shows that the observed
inhibitory effect of the PABT compounds is concentration dependent. The IC50 values range
from 27.9 to 85.7 µM, being compounds 4 and 10 the ones with the lowest IC50 values (Table
1).
Considering that the OD600 measurements do not allow discerning whether these compounds
exert a bactericidal or a bacteriostatic effect, we subsequently evaluated whether the PABT
compounds can affect the ability of Brucella to generate colony forming units (CFU). For this
purpose, B. abortus cultures were treated with the three PABT compounds that showed the
highest potency in liquid cultures and the CFUs after 24 h of incubation were determined.
Bacteria treated with inhibitors 4, 6 and 10 showed a decrease of 4.5, 2.9 and 2.7 log units in
CFU per mL with respect to the initial culture, respectively (Fig. 9C). As expected for a 24 h
Brucella culture, the DMSO control showed an increment of approximately 4.0 log units in
CFU per mL relative to the inoculum (Fig. 9C). Altogether, these results demonstrate that the
PABT compounds can exert a bactericidal effect on B. abortus.
In vivo RS specific targeting by the PABT compounds
Although the experiments described above successfully proved the inhibitory activity of the
PABT compounds on B. abortus growth, they did not demonstrate whether this effect is in
fact associated with the inhibition of riboflavin biosynthesis. To rule out off-target effects, B.
abortus was grown in TSB medium supplemented with riboflavin or 6,7-dimethyl-8-
ribityllumazine, and treated with the indicated PABT compounds for 24 and 48 h. Indeed, the
addition of exogenous riboflavin counteracts the effect of the inhibitors on bacterial growth
after 48 h of treatment (Fig. 10A), while the supplementation with 6,7-dimethyl-8-
This article is protected by copyright. All rights reserved.

ribityllumazine could not revert the bactericidal effect of the PABT compounds (Fig. 10B).
Taking these results together, we conclude that the observed antibacterial effect of the PABT
inhibitors on B. abortus can be attributed to their specific targeting of the RS enzyme.
RS inhibitors prevent intracellular B. abortus replication
In the search for effective antimicrobial agents against brucellosis, it is not enough to achieve
bacterial inhibition in culture. Successful drugs should be able to penetrate the host cell to
reach the intracellular growth niche of the pathogen. Thus, we subsequently measured the
effect of the two most active compounds identified above (4 and 10) on the intracellular
replication of B. abortus. For this purpose, we first established the non-toxic concentration
ranges of inhibitors 4 and 10 on the J774A.1 cell line. In general, both compounds displayed
a similar concentration dependent cytotoxicity effect, with concentrations above 30 µM
producing death on most of the cultured cells, and no appreciable toxic effects at 10 µM (Fig.
11A). Thus, although these results indicate that there is a narrow working range, we still
proceeded to test the efficacy of the compounds in intracellular Brucella infection assays.
Remarkably, Fig. 11B shows that treatment of infected macrophages with 10 µM of
inhibitors 4 or 10 results in a significant reduction of viable intracellular bacteria (25 and 14-
fold reduction, respectively). Therefore, inhibition of bacterial growth both in liquid and
macrophage cultures indicate that the drug candidates 4 and 10 can efficiently cross the
macrophage membrane, the Brucella intracellular niche and the bacterial membranes, to
finally reach the cytoplasmic RS target.
This article is protected by copyright. All rights reserved.

Discussion
Targeting of metabolic pathways absent in mammals is an attractive strategy for the
development of antibacterial therapeutic compounds. Histidine and branched-chain essential
amino acid biosynthetic pathways have been targeted in Brucella spp. with some success [43,
44]. Another example of enzymes targeted for anti-Brucella drug development corresponds to
the carbonic anhydrases (CA). These enzymes are susceptible to selective inhibition over
human CA II by a wide range of classical aromatic and heteroaromatic sulfonamides as well
as carbohydrate-based compounds, affecting Brucella growth [45].
The riboflavin pathway is a very attractive target for antibiotic discovery since it is essential
for the survival, intracellular trafficking and persistence of B. abortus [44]. Cushman and
coworkers have reported the structure-based design and synthesis of several bacterial and
fungal LS and RS antimetabolites with strong in vitro enzymatic inhibitory activity [27, 29-
33, 36, 46-51]. However, although these studies generated insight about these targets, most of
the discovered compounds were hydrophilic and thus not suitable to penetrate cell
membranes and exert their antibiotic activity.
A compound able to inhibit the growth of intracellular pathogens should pass through several
membranes. Thus, is very important that the compounds assayed comply with the Lipinski
rules to sustain a successful development pathway [52]. In consequence, a HTS approach
using a library composed of compounds with drug-like properties is nowadays considered a
very powerful and direct approach to find effective drug candidates against intracellular
bacteria. In fact, a series of trifluoromethylated pyrazole compounds directed to RS with the
capacity to inhibit the growth of both replicating and non-replicating M. tuberculosis were
discovered using a commercial library HTS [37, 53]. However, off-target effects on other
This article is protected by copyright. All rights reserved.

bacterial enzymes such as M. tuberculosis CYP51, CYP121 and UDP-galactopyranose
mutase (UGM) could not be ruled out [54-56].
We have previously shown that riboflavin biosynthesis is essential for Brucella intracellular
survival and replication in its niche, the brucellosome [14]. Thus, an ideal antibiotic to treat
chronic brucellosis patients should attack bacteria in this reservoir, where Brucella can
survive for decades. The present work summarizes the results obtained in a HTS by designing
an automatable reaction to measure the RS inhibitory activity and using a large library with
drug-like properties. This strategy allowed us to identify a family of compounds sharing a 2-
Phenylamidazo[2,1-b][1,3]benzothiazole chemical scaffold, which showed inhibition of
Brucella RS with IC50 values ranging from 2 to 30 µM. These IC50 values are in a similar
range to the inhibition constants (Ki) described for some antimetabolites against RS and LS
reported by Cushman and coworkers [28, 31]. In addition, our RS inhibitors showed
antimicrobial activity against B. abortus growth in culture, which was reverted by
supplementing the medium with riboflavin, ruling out off-target effects. Most importantly,
these compounds were further tested for antimicrobial activity in an in vitro infection model,
demonstrating its antibiotic potential against mammalian intracellular pathogens.
These promising PABT hits can be further optimized. Based on our previous crystal
structures of RS as apo-enzyme and in complex with different substrates and product
analogues [26] and ongoing crystallization and NMR structural studies of RS:PABT
complexes, we expect to develop structure-activity relationships (SAR) in order to direct the
rational design of a lead compound. In particular, these studies may allow us to modify the
R1 and R2 groups and/or the scaffold to (i) improve their complementarity to the RS active
site, (ii) make the compounds more effective to cross cellular membranes and reach the
This article is protected by copyright. All rights reserved.

brucellosome, and (iii) avoid bacterial efflux systems. SAR also can be used to decrease the
toxicity of PABT hits for healthy cells.
An interesting point to discuss is the observation that adding riboflavin to the growth medium
alleviates the antibacterial activity of the PABT compounds (Fig. 10A). This might suggest
that these inhibitors could have reduced efficacy in individuals having an increased level of
riboflavin in their diets. In a hypothetical scenario where an antibacterial drug was developed
targeting the activity of Brucella RS, the extracellular riboflavin could in principle bypass
this inhibition if it enters the bacterial cytosol. However, on the one hand Brucella does not
code for riboflavin transporters and needs the vitamin to be in the extracellular medium at
millimolar concentration to diffuse into the intracellular space [14]. On the other hand, the
plasmatic concentration of riboflavin in humans lies at the nanomolar range [57] and the
intracellular concentration is even lower, in attomolar quantities [58]. Increasing the
riboflavin intake in the diet could eventually lead to an increase in its plasmatic level yet
remaining below the micromolar range [59]. Therefore, during a hypothetical anti-brucellosis
treatment with RS inhibitor drugs, the riboflavin concentration inside the host is not expected
to allow the bacteria to live neither in plasma nor inside the host cells.
In summary, we have performed a HTS and found several compounds that are effective
inhibitors of B. abortus RS. We have demonstrated that five of these compounds are
biologically active against Brucella by inhibiting its growth in culture, with two of them also
inhibiting bacterial growth within the host cell. Taken together, our data suggest that RS from
B. abortus constitutes a suitable target for alternative antibacterial therapy, notably against
strains resistant to conventional antibiotic treatments. Finally, as RS is highly conserved
across pathogenic microorganisms, the potential application of the compounds reported here
might also be effective for other infectious diseases.
This article is protected by copyright. All rights reserved.

Materials and Methods
Chemicals and biological reagents
Riboflavin, β-Glycerophosphate disodium salt hydrate, Bovine serum albumin fraction V
96% (BSA), Sodium orthovanadate (Na3VO4), HEPES and Dithiothreitol (DTT) were
purchased from Sigma–Aldrich (St. Louis, MO, USA). Coating reagents CR-3 and CR-8
were purchased from Caliper Life Sciences (Hopkinton, MA, USA). Tween 20 was
purchased from Bio-Rad (Hercules, CA, USA) and Trifluoroacetic Acid (TFA) from Pierce
(Dallas, TX, USA). 30% Aqueous Solution BRIJ 35 was purchased from Calbiochem (San
Diego, CA, USA). The 5-carboxyfluorescein labeled peptide JAK3 (122.6 kDa) was obtained
from Biosyntan GmbH (Berlin, Germany) and the 6,7-dimethyl-8-ribityllumazine substrate
(326 Da) was kindly provided by Prof. Markus Fischer from the University of Hamburg,
Germany.
RS expression and purification
RS from B. abortus (69 kDa) was expressed and purified as described previously [26]. In
short, the protein was expressed in E. coli BL21(DE3) cells (Stratagene, San Diego, CA,
USA) and purified by affinity chromatography with a HisTrap HP column followed by a size
exclusion chromatography step in a Superdex 200 column (all columns from GE Healthcare,
Chicago, IL, USA). Fractions corresponding to the protein were pooled, concentrated to 60
mg mL-1 in 10 mM Tris, 25 mM sodium chloride, pH 7.4, and stored at −80 °C until use.
Mobility shift-based screening assay (MSA)
The MSA was performed on a Caliper LabChip 3000 Drug Discovery instrument (Caliper
Life Sciences, Waltham, MA, USA) in an off-chip mode. The reaction buffer contained 50
mM HEPES, 0.02% Tween 20, 1 mM DTT, 0.02% BSA, 10 µM Na3VO4, and 10 mM β-
This article is protected by copyright. All rights reserved.

Glycerophosphate, pH 7.5. The NIBR Drug Discovery Incubator (NDDI) library for external
collaborations comprises 44,000 pure chemical compounds accessible from the Hit Finding
Unit of Novartis Pharma AG (Switzerland). The compounds were selected according to their
chemical diversity and drug-like properties, and their purity was assessed by LC-MS (data
not shown). 90 nL of the compound stock solutions (at 10 mM in 90% (v/v) DMSO/water)
were placed in a 384-well assay microtiter plate (Corning Inc., Corning, NY, USA) prior to
the addition of 4.5 μL of RS (16 nM) and the fluorescent peptide JAK3 (400 nM) in 1x
reaction buffer. Subsequently, 4.5 μL of the 6,7-dimethyl-8-ribityllumazine substrate at 40
μM in 1x reaction buffer were added. The final compound concentration was 30 μM and the
DMSO concentration was 0.9% (v/v). The microtiter plates with the assay solution were
incubated for 60 min at 30 °C and 95% humidity, prior to the addition of 1.2% (v/v)
TFA/buffer solution to stop the enzymatic reaction. The assay plate was subjected to the
Caliper LabChip 3000 reader, equipped with a 12-sipper chip. The reaction mixture was
sampled for 1 s for electrophoretic separation of the 6,7-dimethyl-8-ribityllumazine and
riboflavin peaks at a pressure of –2.1 psi and an upstream voltage of –750 V and a
downstream voltage of –2500 V.
The Caliper technology requires a standardization of the runtime and peak height. For this
purpose, the 5-carboxyfluorescein labeled peptide JAK3 (a reference dye successfully applied
in previous HTS experiments at Novartis) was used as an internal standard (λexcitation = 493
nm, λemission = 517 nm).
The conversion rate (r) was defined as the height of the product peak divided by the sum of
the peak heights from product and substrate. The effect of a test compound on the enzymatic
reaction was quantified using the Novartis proprietary software package HELIOS. The
inhibition percentage was calculated relative to the high and low controls as:
This article is protected by copyright. All rights reserved.

– – – ,
where the low controls corresponded to 100% inhibition in the presence of 5.5% TFA and the
high controls to the absence of inhibitor compounds.
The quality of the assay was assessed by calculating the Z’ factor value using the following
equation: –
[60].
–
Validation of screening hits by 50% inhibitory concentrations (IC50) values
IC50 values for selected primary screening hits were determined from dose-response curves.
Typically, reactions were performed in 384-well microtiter plates in reaction buffer at 0.008
µM RS and eight concentrations of the compounds (0.015 – 32 µM). The IC50 values were
calculated from the plot of percentage of inhibition relative to controls vs. inhibitor
concentration by a logistics fit using the HELIOS software.
Bacterial growth conditions
B. abortus 2308 strain cells were grown in tryptic soy broth (TSB) or on tryptic soy agar
(TSA) plates at 37 °C, on a rotary shaker (200 rpm) or incubator, respectively. All work with
viable B. abortus was performed on a biosafety level 3 laboratory.
Susceptibility of B. abortus to the RS inhibitor compounds
RS inhibitor compounds were dissolved in DMSO at a final concentration of 20 mM prior to
use. The determination of the susceptibility of B. abortus to these compounds was carried out
by measuring the optical density at 600 nm (OD600) of treated liquid cultures as a function of
time compared to a control (DMSO only treatment). Overnight cultures were diluted in TSB
at OD600 = 0.01, then the individual inhibitors were added to a final concentration of 150 µM
and the resulting OD600 were measured at 24 and 48 h post incubation. When indicated, the
This article is protected by copyright. All rights reserved.

cultures were diluted in TSB supplemented with 500 µM of riboflavin or 6,7-dimethyl-8-
ribityllumazine.
To perform the dose-response assays of the inhibitors 4, 5, 6, 9 and 10, B. abortus cultures
were diluted to OD600 = 0.01 and incubated with 2-fold serial dilutions of the indicated
compounds. Optical densities were measured 24 h later. Bacterial viability of cultures treated
with the compounds 4, 6 and 10 for 24 h was assessed by plating serial dilutions of B. abortus
cultures and counting the resulting colony forming units (CFU) 24 h later. Cultures incubated
with DMSO were used as control.
Macrophage toxicity assay
Murine macrophage-like cells J774A.1 (ATCC, Manassas, VA, USA) were seeded at a
density of 1.5 × 105 cells per well in 24-well plates (Corning) with Dulbecco’s modified
Eagle’s medium (DMEM) and 10% fetal bovine serum (Gibco, Waltham, MA, USA) at 37
°C under 5% CO2 24 h prior to treatment. After that, the macrophage cells were incubated
with 2, 10 and 30 μM of inhibitors 4 or 10 for another 24 h under similar conditions. Cell
viability was measured by counting the number of viable cells after Trypan blue staining
using a gridded hemocytometer. In all cases, at least 100 cells were counted to determine the
percentage of live cells. Assays were performed in duplicate wells. The results represent the
averages from at least three independent experiments.
Infection of J774A.1 cells
Murine macrophage-like J774A.1 cells were plated at a density of 1.5 × 105 cells per well in
24-well plates and grown in DMEM added with 10% fetal bovine serum for 24 h at 37 ºC and
5% CO2 prior to the infection. Then, cells were infected with B. abortus in triplicate wells at
a 100:1 multiplicity of infection. After 1 h, cells were washed with phosphate-buffered saline
This article is protected by copyright. All rights reserved.

(PBS) and incubated with DMEM added with 25 μg mL-1 gentamicin. After 1 h (time 0), the
monolayers were washed with DMEM and incubated for another 24 h with complete medium
containing 10 μg mL-1 gentamicin with the addition of 2, 6 or 10 M of inhibitors 4 or 10.
Then, infected macrophages were washed with PBS and lysed with 0.1% Triton X-100
(Sigma-Aldrich). Intracellular B. abortus CFUs were determined by plating serial dilutions of
the lysates and counting the bacterial colonies after 24 h of incubation. The results represent
the average ratio between the CFU per mL obtained from the treated wells and the control
wells, which were incubated only with DMSO.
Statistical analysis
Statistical analyses were performed with One-way ANOVA with a Bonferroni’s multiple
comparison post-hoc test using GraphPad Prism5. Data are presented as mean ± standard
deviation (SD). P-values of 0.05 or lower were considered as significant. Statistical
significance levels were defined as follows: *p < 0.05; **p < 0.01; ***p < 0.001.
Acknowledgements
This work was supported by the Argentinian Research Council (CONICET) and the
Argentinian Agency for Scientific and Technological Development (ANPCyT) through the
research grant PICT 2014-0959. We acknowledge Novartis for the participation of MIS in the
Next Generation Scientist Program in Basel, Switzerland.
Author contributions
JT and FAG conceived and supervised the study; MIS, MdCC, SLR and MLC performed
experiments; all authors analyzed data and wrote the manuscript.
This article is protected by copyright. All rights reserved.

Table 1: In vitro and in vivo effect of the 10 selected inhibitor compounds (sorted by IC50
value in RS inhibition in ascending order)
Compound
Compound IC50 (µM)
IC50 (µM)
commercial name
number
(RS inhibition)
(Bacterial growth inhibition) ‡
IA-79-AF08
OA-70-DS97
YC-52-NV98
QF-00-MV19
SC-03-QQ19
GF-00-MG19
IF-03-LB24
1
2
3
4
5
6
7
8
2.35
nd #
nd
19.12
19.15
25.64
26.38
27.80
29.64
30.53
nd
44.7
76.0
54.7
nd
UF-56-IL24
nd
WC-03-QM19
KA-06-JQ17
9
31.18
31.87
85.7
27.9
10
‡ calculated as percentage of inhibition relative to controls vs. inhibitor concentration
# nd, not detemined
This article is protected by copyright. All rights reserved.

Table 2: Physicochemical properties of the 10 compounds with the lowest IC50 values
against B. abortus RS enzymatic activity
Compound
Compound MW
clogP
HBA HBD
tPSA
nRot
PABT
commercial name number
subset
IA-79-AF08
OA-70-DS97
YC-52-NV98
QF-00-MV19
SC-03-QQ19
GF-00-MG19
IF-03-LB24
1
2
3
4
5
6
7
8
491.6
448.5
446.5
462.6
462.6
450.6
447.5
330.3
4.57
3.53
3.95
4.71
4.68
4.94
3.72
3.24
5
6
6
4
4
4
9
1
1
1
1
1
1
1
0
7
107.2
122.4
66.9
10
11
6
NO
NO
NO
87.1
8
YES
YES
YES
NO
87.1
7
87.1
10
11
6
124.3
95.3
UF-56-IL24
NO
WC-03-QM19
KA-06-JQ17
9
448.6
420.6
4.37
4.73
4
3
1
1
87.1
77.9
7
8
YES
YES
10
MW = Molecular weight (Da); clogP = lipophilicity; HBA = number of hydrogen bond acceptor groups
(sum of N and O atoms); HBD = number of hydrogen bond donor groups (sum of OH and NH groups);
tPSA= topological polar surface area (Å2); nRot = number of rotatable bonds.
This article is protected by copyright. All rights reserved.

References
1. Godfroid, J., Cloeckaert, A., Liautard, J. P., Kohler, S., Fretin, D., Walravens, K., Garin-
Bastuji, B. & Letesson, J. J. (2005) From the discovery of the Malta fever's agent to the
discovery of a marine mammal reservoir, brucellosis has continuously been a re-emerging
zoonosis, Vet Res. 36, 313-26.
2. Pappas, G., Akritidis, N., Bosilkovski, M. & Tsianos, E. (2005) Brucellosis, N Engl J
Med. 352, 2325-36.
3. Gorvel, J. P. (2008) Brucella: a Mr "Hide" converted into Dr Jekyll, Microbes Infect. 10,
1010-3.
4. Franc, K. A., Krecek, R. C., Hasler, B. N. & Arenas-Gamboa, A. M. (2018) Brucellosis
remains a neglected disease in the developing world: a call for interdisciplinary action, BMC
public health. 18, 125.
5. Pappas, G., Papadimitriou, P., Akritidis, N., Christou, L. & Tsianos, E. V. (2006) The new
global map of human brucellosis, Lancet Infect Dis. 6, 91-9.
6. Ariza, J., Bosilkovski, M., Cascio, A., Colmenero, J. D., Corbel, M. J., Falagas, M. E.,
Memish, Z. A., Roushan, M. R., Rubinstein, E., Sipsas, N. V., Solera, J., Young, E. J.,
Pappas, G., International Society of, C. & Institute of Continuing Medical Education of, I.
(2007) Perspectives for the treatment of brucellosis in the 21st century: the Ioannina
recommendations, PLoS Med. 4, e317.
7. Yousefi-Nooraie, R., Mortaz-Hejri, S., Mehrani, M. & Sadeghipour, P. (2012) Antibiotics
for treating human brucellosis, Cochrane Database Syst Rev. 10, CD007179.
8. Solis Garcia del Pozo, J. & Solera, J. (2012) Systematic review and meta-analysis of
randomized clinical trials in the treatment of human brucellosis, PloS one. 7, e32090.
This article is protected by copyright. All rights reserved.

9. Bacher, A., Eberhardt, S., Fischer, M., Kis, K. & Richter, G. (2000) Biosynthesis of
vitamin b2 (riboflavin), Annual review of nutrition. 20, 153-67.
10. Joosten, V. & van Berkel, W. J. (2007) Flavoenzymes, Curr Opin Chem Biol. 11, 195-
202.
11. Long, Q., Ji, L., Wang, H. & Xie, J. (2010) Riboflavin biosynthetic and regulatory
factors as potential novel anti-infective drug targets, Chem Biol Drug Des. 75, 339-47.
12. Morgunova, E., Illarionov, B., Sambaiah, T., Haase, I., Bacher, A., Cushman, M.,
Fischer, M. & Ladenstein, R. (2006) Structural and thermodynamic insights into the binding
mode of five novel inhibitors of lumazine synthase from Mycobacterium tuberculosis, FEBS
J. 273, 4790-804.
13. Rollenhagen, C. & Bumann, D. (2006) Salmonella enterica highly expressed genes are
disease specific, Infection and immunity. 74, 1649-60.
14. Bonomi, H. R., Marchesini, M. I., Klinke, S., Ugalde, J. E., Zylberman, V., Ugalde, R.
A., Comerci, D. J. & Goldbaum, F. A. (2010) An atypical riboflavin pathway is essential for
Brucella abortus virulence, PloS one. 5, e9435.
15. Garcia Angulo, V. A., Bonomi, H. R., Posadas, D. M., Serer, M. I., Torres, A. G.,
Zorreguieta, A. & Goldbaum, F. A. (2013) Identification and characterization of RibN, a
novel family of riboflavin transporters from Rhizobium leguminosarum and other
proteobacteria, Journal of bacteriology. 195, 4611-9.
16. Gutierrez-Preciado, A., Torres, A. G., Merino, E., Bonomi, H. R., Goldbaum, F. A. &
Garcia-Angulo, V. A. (2015) Extensive Identification of Bacterial Riboflavin Transporters
and Their Distribution across Bacterial Species, PloS one. 10, e0126124.
17. Fischer, M. & Bacher, A. (2008) Biosynthesis of vitamin B2: Structure and mechanism
of riboflavin synthase, Arch Biochem Biophys. 474, 252-65.
This article is protected by copyright. All rights reserved.

18. Fischer, M., Romisch, W., Illarionov, B., Eisenreich, W. & Bacher, A. (2005) Structures
and reaction mechanisms of riboflavin synthases of eubacterial and archaeal origin, Biochem
Soc Trans. 33, 780-4.
19. Fischer, M. & Bacher, A. (2005) Biosynthesis of flavocoenzymes, Natural product
reports. 22, 324-50.
20. Klinke, S., Zylberman, V., Vega, D. R., Guimaraes, B. G., Braden, B. C. & Goldbaum,
F. A. (2005) Crystallographic studies on decameric Brucella spp. Lumazine synthase: a novel
quaternary arrangement evolved for a new function?, Journal of molecular biology. 353, 124-
37.
21. Zylberman, V., Craig, P. O., Klinke, S., Braden, B. C., Cauerhff, A. & Goldbaum, F. A.
(2004) High order quaternary arrangement confers increased structural stability to Brucella
sp. lumazine synthase, The Journal of biological chemistry. 279, 8093-101.
22. Berguer, P. M., Mundinano, J., Piazzon, I. & Goldbaum, F. A. (2006) A polymeric
bacterial protein activates dendritic cells via TLR4, Journal of immunology. 176, 2366-72.
23. Craig, P. O., Alzogaray, V. & Goldbaum, F. A. (2012) Polymeric Display of Proteins
through High Affinity Leucine Zipper Peptide Adaptors, Biomacromolecules. 13, 1112-21.
24. Laplagne, D. A., Zylberman, V., Ainciart, N., Steward, M. W., Sciutto, E., Fossati, C. A.
& Goldbaum, F. A. (2004) Engineering of a polymeric bacterial protein as a scaffold for the
multiple display of peptides, Proteins. 57, 820-8.
25. Yang, Y., Wang, L., Yin, J., Wang, X., Cheng, S., Lang, X., Wang, X., Qu, H., Sun, C.,
Wang, J. & Zhang, R. (2011) Immunoproteomic analysis of Brucella melitensis and
identification of a new immunogenic candidate protein for the development of brucellosis
subunit vaccine, Mol Immunol. 49, 175-84.
26. Serer, M. I., Bonomi, H. R., Guimaraes, B. G., Rossi, R. C., Goldbaum, F. A. & Klinke,
S. (2014) Crystallographic and kinetic study of riboflavin synthase from Brucella abortus, a
This article is protected by copyright. All rights reserved.

chemotherapeutic target with an enhanced intrinsic flexibility, Acta Crystallogr D Biol
Crystallogr. 70, 1419-34.
27. Cushman, M., Mavandadi, F., Kugelbrey, K. & Bacher, A. (1998) Synthesis of 2,6-
dioxo-(1H,3H)-9-N-ribitylpurine and 2,6-dioxo-(1H,3H)-8-aza-9-N-ribitylpurine as inhibitors
of lumazine synthase and riboflavin synthase, Bioorg Med Chem. 6, 409-15.
28. Cushman, M., Yang, D., Kis, K. & Bacher, A. (2001) Design, synthesis, and evaluation
of 9-D-ribityl-1,3,7-trihydro-2,6,8-purinetrione, a potent inhibitor of riboflavin synthase and
lumazine synthase, J Org Chem. 66, 8320-7.
29. Cushman, M., Jin, G., Sambaiah, T., Illarionov, B., Fischer, M., Ladenstein, R. &
Bacher, A. (2005) Design, synthesis, and biochemical evaluation of 1,5,6,7-tetrahydro-6,7-
dioxo-9-D-ribitylaminolumazines bearing alkyl phosphate substituents as inhibitors of
lumazine synthase and riboflavin synthase, J Org Chem. 70, 8162-70.
30. Cushman, M., Sambaiah, T., Jin, G., Illarionov, B., Fischer, M. & Bacher, A. (2004)
Design, synthesis, and evaluation of 9-D-ribitylamino-1,3,7,9-tetrahydro-2,6,8-purinetriones
bearing alkyl phosphate and alpha,alpha-difluorophosphonate substituents as inhibitors of
tiboflavin synthase and lumazine synthase, J Org Chem. 69, 601-12.
31. Cushman, M., Yang, D., Gerhardt, S., Huber, R., Fischer, M., Kis, K. & Bacher, A.
(2002) Design, synthesis, and evaluation of 6-carboxyalkyl and 6-phosphonoxyalkyl
derivatives of 7-oxo-8-ribitylaminolumazines as inhibitors of riboflavin synthase and
lumazine synthase, J Org Chem. 67, 5807-16.
32. Scheuring, J., Fischer, M., Cushman, M., Lee, J., Bacher, A. & Oschkinat, H. (1996)
NMR analysis of site-specific ligand binding in oligomeric proteins. Dynamic studies on the
interaction of riboflavin synthase with trifluoromethyl-substituted intermediates,
Biochemistry. 35, 9637-46.
This article is protected by copyright. All rights reserved.

33. Scheuring, J., Lee, J., Cushman, M., Patel, H., Patrick, D. A. & Bacher, A. (1994)
(Trifluoromethyl)lumazine derivatives as 19F NMR probes for lumazine protein,
Biochemistry. 33, 7634-40.
34. Chen, J., Sambaiah, T., Illarionov, B., Fischer, M., Bacher, A. & Cushman, M. (2004)
Design, synthesis, and evaluation of acyclic C-nucleoside and N-methylated derivatives of
the ribitylaminopyrimidine substrate of lumazine synthase as potential enzyme inhibitors and
mechanistic probes, J Org Chem. 69, 6996-7003.
35. Zhang, Y., Illarionov, B., Bacher, A., Fischer, M., Georg, G. I., Ye, Q. Z., Vander Velde,
D., Fanwick, P. E., Song, Y. & Cushman, M. (2007) A novel lumazine synthase inhibitor
derived
from
oxidation
of
1,3,6,8-tetrahydroxy-2,7-naphthyridine
to
a
tetraazaperylenehexaone derivative, J Org Chem. 72, 2769-76.
36. Talukdar, A., Illarionov, B., Bacher, A., Fischer, M. & Cushman, M. (2007) Synthesis
and enzyme inhibitory activity of the s-nucleoside analogue of the ribitylaminopyrimidine
substrate of lumazine synthase and product of riboflavin synthase, J Org Chem. 72, 7167-75.
37. Zhao, Y., Bacher, A., Illarionov, B., Fischer, M., Georg, G., Ye, Q. Z., Fanwick, P. E.,
Franzblau, S. G., Wan, B. & Cushman, M. (2009) Discovery and development of the covalent
hydrates of trifluoromethylated pyrazoles as riboflavin synthase inhibitors with antibiotic
activity against Mycobacterium tuberculosis, J Org Chem. 74, 5297-303.
38. Lipinski, C. A., Lombardo, F., Dominy, B. W. & Feeney, P. J. (2001) Experimental and
computational approaches to estimate solubility and permeability in drug discovery and
development settings, Adv Drug Deliv Rev. 46, 3-26.
39. Veber, D. F., Johnson, S. R., Cheng, H. Y., Smith, B. R., Ward, K. W. & Kopple, K. D.
(2002) Molecular properties that influence the oral bioavailability of drug candidates, J Med
Chem. 45, 2615-23.
This article is protected by copyright. All rights reserved.

40. Camenisch, G., Folkers, G. & van de Waterbeemd, H. (1996) Review of theoretical
passive drug absorption models: historical background, recent developments and limitations,
Pharm Acta Helv. 71, 309-27.
41. Ertl, P., Rohde, B. & Selzer, P. (2000) Fast calculation of molecular polar surface area as
a sum of fragment-based contributions and its application to the prediction of drug transport
properties, J Med Chem. 43, 3714-7.
42. Artursson Per, B. C. A. S. (2004) Intestinal Absorption: The Role of Polar Surface Area
in Drug Bioavailability: Estimation of Solubility, Permeability, Absorption and
Bioavailability (Han van de Waterbeemd, H. L., Per Artursson ed).
43. Joseph, P., Abdo, M. R., Boigegrain, R. A., Montero, J. L., Winum, J. Y. & Kohler, S.
(2007) Targeting of the Brucella suis virulence factor histidinol dehydrogenase by histidinol
analogues results in inhibition of intramacrophagic multiplication of the pathogen,
Antimicrob Agents Chemother. 51, 3752-5.
44. Boigegrain, R. A., Liautard, J. P. & Kohler, S. (2005) Targeting of the virulence factor
acetohydroxyacid synthase by sulfonylureas results in inhibition of intramacrophagic
multiplication of Brucella suis, Antimicrob Agents Chemother. 49, 3922-5.
45. Kohler, S., Ouahrani-Bettache, S. & Winum, J. Y. (2017) Brucella suis carbonic
anhydrases and their inhibitors: Towards alternative antibiotics?, J Enzyme Inhib Med Chem.
32, 683-687.
46. Cushman, M., Mavandadi, F., Yang, D., Kugelbrey, K., Kis, K. & Bacher, A. (1999)
Synthesis and Biochemical Evaluation of Bis(6,7-dimethyl-8-D-ribityllumazines) as Potential
Bisubstrate Analogue Inhibitors of Riboflavin Synthase, J Org Chem. 64, 4635-4642.
47. Cushman, M., Mihalic, J. T., Kis, K. & Bacher, A. (1999) Design and synthesis of 6-(6-
D-ribitylamino-2,4-dihydroxypyrimidin-5-yl)-1-hexyl phosphonic acid, a potent inhibitor of
lumazine synthase, Bioorg Med Chem Lett. 9, 39-42.
This article is protected by copyright. All rights reserved.

48. Zhang, Y., Jin, G., Illarionov, B., Bacher, A., Fischer, M. & Cushman, M. (2007) A new
series of 3-alkyl phosphate derivatives of 4,5,6,7-tetrahydro-1-D-ribityl-1H-pyrazolo[3,4-
d]pyrimidinedione as inhibitors of lumazine synthase: design, synthesis, and evaluation, J
Org Chem. 72, 7176-84.
49. Illarionov, B., Kemter, K., Eberhardt, S., Richter, G., Cushman, M. & Bacher, A. (2001)
Riboflavin synthase of Escherichia coli. Effect of single amino acid substitutions on reaction
rate and ligand binding properties, The Journal of biological chemistry. 276, 11524-30.
50. Zhang, Y., Illarionov, B., Morgunova, E., Jin, G., Bacher, A., Fischer, M., Ladenstein, R.
& Cushman, M. (2008) A new series of N-[2,4-dioxo-6-d-ribitylamino-1,2,3,4-
tetrahydropyrimidin-5-yl]oxalamic acid derivatives as inhibitors of lumazine synthase and
riboflavin synthase: design, synthesis, biochemical evaluation, crystallography, and
mechanistic implications, J Org Chem. 73, 2715-24.
51. Harvey, R. A. & Plaut, G. W. (1966) Riboflavin synthetase from yeast. Properties of
complexes of the enzyme with lumazine derivatives and riboflavin, The Journal of biological
chemistry. 241, 2120-36.
52. Lipinski, C. A. (2004) Lead- and drug-like compounds: the rule-of-five revolution, Drug
Discov Today Technol. 1, 337-41.
53. Kaiser, J., Illarionov, B., Rohdich, F., Eisenreich, W., Saller, S., den Brulle, J. V.,
Cushman, M., Bacher, A. & Fischer, M. (2007) A high-throughput screening platform for
inhibitors of the riboflavin biosynthesis pathway, Anal Biochem. 365, 52-61.
54. McLean, K. J., Marshall, K. R., Richmond, A., Hunter, I. S., Fowler, K., Kieser, T.,
Gurcha, S. S., Besra, G. S. & Munro, A. W. (2002) Azole antifungals are potent inhibitors of
cytochrome P450 mono-oxygenases and bacterial growth in mycobacteria and
streptomycetes, Microbiology. 148, 2937-49.
This article is protected by copyright. All rights reserved.

55. Menozzi, G., Merello, L., Fossa, P., Schenone, S., Ranise, A., Mosti, L., Bondavalli, F.,
Loddo, R., Murgioni, C., Mascia, V., La Colla, P. & Tamburini, E. (2004) Synthesis,
antimicrobial activity and molecular modeling studies of halogenated 4-[1H-imidazol-1-
yl(phenyl)methyl]-1,5-diphenyl-1H-pyrazoles, Bioorg Med Chem. 12, 5465-83.
56. Carlson, E. E., May, J. F. & Kiessling, L. L. (2006) Chemical probes of UDP-
galactopyranose mutase, Chem Biol. 13, 825-37.
57. Eussen, S. J., Vollset, S. E., Hustad, S., Midttun, O., Meyer, K., Fredriksen, A., Ueland,
P. M., Jenab, M., Slimani, N., Boffetta, P., Overvad, K., Thorlacius-Ussing, O., Tjonneland,
A., Olsen, A., Clavel-Chapelon, F., Boutron-Ruault, M. C., Morois, S., Weikert, C., Pischon,
T., Linseisen, J., Kaaks, R., Trichopoulou, A., Zilis, D., Katsoulis, M., Palli, D., Pala, V.,
Vineis, P., Tumino, R., Panico, S., Peeters, P. H., Bueno-de-Mesquita, H. B., van
Duijnhoven, F. J., Skeie, G., Munoz, X., Martinez, C., Dorronsoro, M., Ardanaz, E., Navarro,
C., Rodriguez, L., VanGuelpen, B., Palmqvist, R., Manjer, J., Ericson, U., Bingham, S.,
Khaw, K. T., Norat, T. & Riboli, E. (2010) Plasma vitamins B2, B6, and B12, and related
genetic variants as predictors of colorectal cancer risk, Cancer epidemiology, biomarkers &
prevention : a publication of the American Association for Cancer Research, cosponsored by
the American Society of Preventive Oncology. 19, 2549-61.
58. Huhner, J., Ingles-Prieto, A., Neususs, C., Lammerhofer, M. & Janovjak, H. (2015)
Quantification of riboflavin, flavin mononucleotide, and flavin adenine dinucleotide in
mammalian model cells by CE with LED-induced fluorescence detection, Electrophoresis.
36, 518-25.
59. Zempleni, J., Galloway, J. R. & McCormick, D. B. (1996) Pharmacokinetics of orally
and intravenously administered riboflavin in healthy humans, The American journal of
clinical nutrition. 63, 54-66.
This article is protected by copyright. All rights reserved.

60. Zhang, J. H., Chung, T. D. & Oldenburg, K. R. (1999) A Simple Statistical Parameter for
Use in Evaluation and Validation of High Throughput Screening Assays, Journal of
biomolecular screening. 4, 67-73.
This article is protected by copyright. All rights reserved.

Figure 1: RS catalysis and structure. (A) Reaction catalyzed by RS. A donor 6,7-dimethyl-
8-ribityllumazine molecule a transfers a four carbon atom moiety to an acceptor 6,7-
dimethyl-8-ribityllumazine molecule b generating riboflavin c and the side product 5-amino-
6-ribitylamino-2,4(1H,3H)-pyrimidinedione d. (B) Crystal structure of B. abortus RS (as apo
enzyme) determined by our group (PDB: 4FXU) [26]. The inset shows a model of the
putative location of the substrate molecules in the active site.
This article is protected by copyright. All rights reserved.