Modification of existing antibiotics in the form of precursor prodrugs that can be subsequently activated by nitroreductases of the target pathogen

Article abstract of DOI:10.1016/j.bmcl.2016.06.081

The use of existing antibiotics in the form of prodrug followed by activation using enzymes of pathogenic origin could be a useful approach for antimicrobial therapy. To investigate this idea, a common antibiotic, sulfamethoxazole has been redesigned in t

Full text of DOI:10.1016/j.bmcl.2016.06.081

Accepted Manuscript
Modification of existing antibiotics in the form of precursor prodrugs that can
be subsequently activated by nitroreductases of the target pathogen
Ayhan Çelik, Gülden Yetiş, Mehmet Ay, Tuğba Güngör
PII:
S0960-894X(16)30692-8
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.06.081
BMCL 24034
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
9 April 2016
25 June 2016
27 June 2016
Please cite this article as: Çelik, A., Yetiş, G., Ay, M., Güngör, T., Modification of existing antibiotics in the form
of precursor prodrugs that can be subsequently activated by nitroreductases of the target pathogen, Bioorganic &
Medicinal Chemistry Letters (2016), doi: http://dx.doi.org/10.1016/j.bmcl.2016.06.081
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Modification of existing antibiotics in the form of precursor prodrugs that
can be subsequently activated by nitroreductases of the target pathogen
Ayhan Çelik,a * Gülden Yetişa Mehmet Ayb and Tuğba Güngörb
aDepartment of Chemistry, Gebze Technical University Cayirova Campus, Gebze-Kocaeli,
41400, Turkey
bDepartment of Chemistry, Canakkale 18 Mart University, Çanakkale, Turkey
* Corresponding author: E-mail:ayhancelik@gtu.edu.tr (Ayhan Çelik); Tel: +90 2626053087,
Fax: +90 262 605 3005
Abstract- The use of existing antibiotics in the form of prodrug followed by activation using
enzymes of pathogenic origin could be a useful approach for antimicrobial therapy. To
investigate this idea, a common antibiotic, sulfamethoxazole has been redesigned in the form
of a prodrug by simple functional group replacement. Upon reductive activation by a type I
nitroreductase from a pathogen, the drug displayed enhanced antimicrobial capacity. This
strategy could improve the efficacy and selectively of antibiotics and reduce the incidence of
resistance.
Keywords: nitroreductase, prodrugs, sulfamethoxazole
1

Since its original development in 1958, the prodrug strategy has been extensively
utilized to improve the physicochemical and pharmacokinetic properties of drugs in order to
lessen their adverse effects. In fact, approximately one third of drugs currently on the market
are in the form of prodrugs with improved ADMET properties.1 Moreover, this strategy has
evolved into several novel applications that have attracted increasing attention over the past
20 years and has recently entered clinical trials for cancer therapy.2 The prodrug concept is
based on the conversion of an inactive, or less active, derivative of a drug (prodrug) into an
active drug (or its metabolites), which can then stimulate its desired pharmacological effect in
the body. This conversion can either be chemically or enzymatically stimulated. Among the
enzymes that are traditionally used for activation are unspecific esterases,3 alkaline
phosphateses,4 peptidases5 and specific cytochrome P450s.6,7
Nitroreductases (NTRs) are a relatively new class of enzyme in the area of prodrug
activation.8,9 Unlike traditional enzymes, which activate mostly carrier-linked prodrugs (drug
with promoiety), NTRs are involved in the activation of bioprecursor prodrugs, which are
transformed by direct reduction or reduction followed by elimination to the active agent.10
The best known example is the combination of NfsB from E. coli with CB1954, which are
currently in clinical trials (phase I/II) for patients with late stage tumors.11,12,13,14 Given their
versatility, we decided to explore the potential use of NTR enzymes for antibiotic activation.
The work comprises the synthesis of novel ‘Pre-antibiotics’ of known antibiotics and their
reductive activation by a NTRs followed by metabolite analysis and characterization. By
adopting this strategy, it might then be possible to re-design existing drugs as precursor
prodrugs that could be activated by bacterial NTRs from pathogens of interest for enhanced
efficacy and possibly selective antimicrobial activity. As prodrug activating enzymes, NTRs
from pathogenic sources have a number of advantages. Specifically, (i) the absence, or low
catalytic activities, of mammalian equivalent enzymes to the bacterial NTRs, and (ii)
differences in substrate specificity between the bacterial NTRs and the corresponding
mammalian enzymes. These properties help prevent nonspecific prodrug activation at sites
other than the pathogen itself.
In order to test the principle, there are two main requirements: (i) a NTR from a
pathogen, and (ii) a drug molecule that is used for the treatment of diseases caused by the
pathogen of interest. Therefore, we have focused on Urinary Tract Infections (UTI) and
sulphonamide-based antibiotics that have been used to treat UTIs. Based on whole genome
sequence data, a novel NTR from Staphylococcus saprophyticus, has been cloned and the
2

recombinant enzyme purified in an active form (named SsapNtrB).15 S. saprophyticus is an
uropathogenic Staphylococcus usually isolated from young female patients presenting with
uncomplicated UTI.16 Sulfamethoxazole (SMX) is a sulphonamide bacteriostatic antibiotic,
which is used to treat UTI and RTI (Respiratory Tract Infections). The antibiotic is usually
given in combination with trimethoprim (in a 5:1 ratio in co-trimoxazole) to combat infections
caused by most Gram-positive and many Gram-negative organisms including Streptococcus,
Staphylococcus aureus (including MRSA), Escherichia coli, Haemophilus influenza, and oral
anaerobes.17 SMX is on the World Health Organization’s List of Essential Medicines-2015,
as one of the most important medications required in a basic healthcare system
(http://www.who.int/medicines).18 The prodrug form of sulfamethoxazole, namely N-(5-
methylisoxazol-3-yl)-4-nitrobenzenesulfonamide (3), was synthesized following a literature
procedure 19 using 4-nitrobenzenesulfonyl chloride (1) and 3-amino-5-methylisoxazol (2)
(Scheme 1A) in 53% yield. Detailed synthetic steps and characterisation of 3 has been
described in the ESI†. Briefly, 4-nitrobenzenesulfonyl chloride was added to a solution of 3-
amino-5-methylisoxazol in pyridine at 0˚C. After stirring at this temperature for few hours,
the mixture was allowed to warm to room temperature and stirred for a further 24 h. The
solution was then poured into cold water. The resulting yellow precipitate was filtered,
washed several times with water and dried at room temperature. The crude product was
purified by crystallization with hexane/ethyl acetate.
3

Scheme 1. (A) Synthesis of the prodrug of sulfamethoxazole (NO2-SMX, 3) and (B) its possible reduction
products (NO-SMX, 4; N4-OH-SMX, 5 and SMX, 6) catalysed by a nitroreductase enzyme (SsapNtrB) together
with a cofactor regeneration system (middle).
Direct reductive activation of 3 (NO2-SMX) catalysed by SsapNtrB in the presence of NADH
or NADPH (or with a cofactor regeneration system) in vitro, results in potentially three
metabolites: a nitroso (NO-SMX, 4), a N4-hydroxylamino (N-OH-SMX, 5) and an amine
derivative (SMX, 6) of the prodrug (Scheme 1B) depending on the relative stability of the
metabolites. It should be noted that the amine derivative (SMX) is the actual antibiotic, SMX.
Preliminary activity experiments with a fixed substrate concentration showed that the purified
enzyme has potential for prodrug activation. The rate of depletion of NADH was observed by
monitoring the absorbance at 340 nm. (Fig. 1A). Therefore, the reductive metabolism of 3 by
SsapNtrB was investigated in more detail. HPLC analysis together with LC-MS/MS analysis
revealed that the dominant metabolite of 3 is the N4-hydroxy-sulfamethoxazole, together with
a small quantity of 4. (Fig. 1B).
4

Figure 1. (A) Time dependent Uv-vis absorption spectra of the reaction in which the prodrug (NO2-SMX, 3)
reacted with SsapNtrB in the presence of NADH. The consumption of the cofactor is indicated by an arrow at
340 nm. (B) Corresponding HPLC analysis of the metabolite(s) of the prodrug. (C) Steady state kinetic data of
SsapNtrB catalysed reduction of 3 by NADH (Insert: Lineweaver–Burk plot).
Metabolite 4 is relatively unstable under the reaction conditions used in this study. The
formation of 4 was only observed using direct mass spectrometric analysis using ESI-MS in
negative mode with molecular ion [M-H]- at m/z=266 (with fragmentations, m/z=170, 122
and 92) indicating a molecular mass of 267 Da (Fig. S4, ESI†). The main metabolite of 3,
however, was the hydroxylamine species (5) with ESI-MS in negative mode and molecular
ion [M-H]- at m/z = 268 (with fragmentations, m/z= 251, 155, 107) indicating a molecular
mass of 269 Da (Fig. S5, ESI†). The amine derivative, 6 has not been observed in vitro under
the experimental setup described here. A cofactor regeneration system for NAD(P)H was
utilized using NAD- or NADP-dependent formate dehydrogenases in the presence of auxiliary
substrate sodium formate (FDHs)20,21 to ensure the complete conversion of 3 without cofactor
limitation and forcing the reaction to go to completion. Steady state kinetics of the reduction
of 3 obeyed Michaelis-Menten kinetics and a typical plot of rate versus substrate
concentration is shown in Fig. 1C. Steady state parameters based on NADH consumption, kcat
and KM were 2.95 ± 0.11 s-1 and 92.61±7.25 µM, respectively. At a fixed substrate
concentration, NADH usage vs product formation (done by HPLC) was estimated to be 2.5 to
5

1 (NADH used: product formed). Given that N4-hydroxylamino (N4-OH-SMX, 5) is the major
product and requires at least 4 electrons for its formation, the results are as expected.
To demonstrate the antimicrobial ability of the prodrug upon reduction, antimicrobial
susceptibility assays were conducted. For quantifying the ability of prodrugs of SMX (3) and
its N4-hydroxy-SMX derivative (5) by comparison to the unmodified antibiotic itself, we
conducted both disk diffusion22 and liquid assays23 for S. saprophyticus. The results are
shown in Figure 2A and 2B, respectively. In the agar diffusion assay, zones of inhibition with
a radius of 10-12 mm and 16-18 mm were observed for entries 5 and 6, respectively. By
contrast, no zone of inhibition was observed for the prodrug (3) along with the control
experiments, in which co-solvent (DMSO) alone was applied. This is rather surprising, but
not totally unexpected. Results are interpreted from this type of assay based on the assumption
that the antibiotic diffuses freely in the solid nutrient medium. However, sometimes this
assumption is incorrect. In such cases, significant deviation from the expected behaviour is
observed for the analysis of diffusion in solid agar. A number of antibiotics fall into this
category, in particular those with hydrophobic and amphipathic characteristics. Given that the
logD values at pH 6.8 for entries 3, 5 and 6 are 0.81, 0.53 and 0.24, respectively,24 it is clear
that the prodrug (3) has a low diffusion rate on the plate assay. It is known25 that certain
antibiotics can be problematic to test with the disk diffusion method because of the specific
physiochemical properties of the molecules. This is particularly the case for antibiotics with a
hydrophobic and amphipathic nature (e.g. subtilin). However, improved membrane
permeability of the prodrug was demonstrated in the liquid assay (Fig. 2B). Our results clearly
indicated that the prodrug (3) has enhanced inhibitory effects over that of SMX in liquid
medium. More details regarding minimum inhibitory concentration (MIC) values for entries
3, 5 and 6 and common antibiotics (nitrofurazone, nitrofurantoin) for S. saprophyticus ATCC
15305 were given in Table S1 (ESI†).
6

Figure 2. (A) Inhibitory activity of entries of 3, 5 and 6 against S. saprophyticus as determined by the disk
diffusion susceptibility test. (B) Inhibitory activity of entries of 3 and 6 against S. saprophyticus as determined
growth curves in liquid assay. SMX, NO2-SMX (3) and the enzymatic reduction metabolite (5) were prepared in
DMSO, which was used for all stock solutions and final concentration was 1%.
In conclusion, we have demonstrated that the use of existing FDA approved antibiotics
in the form of pre-antibiotics could be a useful strategy for combating certain infectious
diseases caused by bacterial pathogens. This was demonstrated using the example of the
sulphonamide-based antibiotic of sulfamethoxazole (SMX). The prodrug of SMS (NO2-SMX,
3) was synthesized and characterized. The reductive activation of 3 catalyzed by SsapNtrB
from the pathogen S. saprophyticus yielded N4-hydroxy-SMX (5) as the major metabolite.
The metabolite showed comparable antimicrobial activity with that of the antibiotic (SMX) in
the agar diffusion assay, whereas the prodrug showed no clear zone on the plate assay,
possibly due to its low solubility. In the liquid assay, however, the prodrug showed superior
antimicrobial activity over that of SMX, probably due to its increased membrane permeability
and hence absorption followed by reduction via pathogen specific stimuli (e.g.
nitroreductases). Moreover, combating pathogens by exploiting their own enzymes may
provide a useful alternative strategy not only for using lower amounts of drug with improved
selectively but also with the potential of limiting the development of antibiotic resistance.
SMX is metabolized in human by acetylation to generate N-acetyl and N4-acetyl derivatives,
which limits its availability for antimicrobial activity.26,27 Moreover, increasing the plasma
concentration of SMX to >400 mg/L leads to toxic adverse effects. The prodrug strategy,
which masks the biological activity of an active compound before it reaches its target, may
7

overcome these limitations. In addition, our strategy could improve antibiotic specificity for
the target pathogen and help optimize its pharmacokinetics. In silico analysis of the toxic
potential of entries 3-6 showed that the differences are insignificant (Table S2, ESI†). These
findings indicate that converting an amine to a nitro, nitroso and hydroxylamine moiety does
not create any undesired properties. In addition, our strategy is generic and can be used to
improve many antibiotics presently on the market.
Acknowledgements
This work was supported by The Scientific and Technological Research Council of Turkey
(TUBITAK, Grant No.110T754&113Z706). We would like to thank Dr. Gareth A. Roberts
(University of Edinburgh, UK) for his critical reading of this manuscript and his comments.
8

References
1
Rautio, J.; Kumpulainen, H.; Heimbach, T.; Oliyai, R.; Oh, D.; Jarvinen, T.; Savolainen, J. Nat
Rev Drug Discov 2008, 7, 255.
2
Williams, E. M.; Little, R. F.; Mowday, A. M.; Rich, M. H.; Chan-Hyams, J. V.; Copp, J. N.;
Smaill, J. B.; Patterson, A. V.; Ackerley, D. F. Biochem J 2015, 471, 131.
3
4
5
6
Redinbo, M. R.; Bencharit, S.; Potter, P. M. Biochem Soc T 2003, 31, 620.
Huttunen, K. M.; Raunio, H.; Rautio, J. Pharmacol Rev 2011, 63, 750.
Zhang, J.; Kale, V.; Chen, M. AAPS J 2015, 17, 102.
Huttunen, K. M.; Mahonen, N.; Leppanen, J.; Vepsalainen, J.; Juvonen, R. O.; Raunio, H.;
Kumpulainen, H.; Jarvinen, T.; Rautio, J. Pharm Res 2007, 24, 679.
7
8
9
Ortiz de Montellano, P. R. Future Med Chem 2013, 5, 213.
Denny, W. A. Curr Pharm Design 2002, 8, 1349.
Prosser, G. A.; Copp, J. N.; Syddall, S. P.; Williams, E. M.; Smaill, J. B.; Wilson, W. R.; Patterson,
A. V.; Ackerley, D. F. Biochem Pharmacol 2010, 79, 678.
10
11
Hu, L. In Prodrug Approaches to Drug deivery; John Wiley & Sons, Nerw Jersey, 2005. pp 125.
James, N. D.; Patel, P.; Mautner, V.; Young, J. G.; Hull, D.; Searle, P.; Leung, H. Y.; Ellis, J.;
Wallace, M.; Young, L. S. J Clin Oncol 2004, 22, 399s.
12
13
Rooseboom, M.; Commandeur, J. N.; Vermeulen, N. P. Pharmacol Rev 2004, 56, 53.
Searle, P. F.; Salman, A.; Viney, R.; Patel, P.; Onion, D.; James, N. D.; Porfiri, E.; Mautner, V.
Hum Gene Ther 2010, 21, 522.
14
Patel, P.; Young, J. G.; Mautner, V.; Ashdown, D.; Bonney, S.; Pineda, R. G.; Collins, S. I.;
Searle, P. F.; Hull, D.; Peers, E.; Chester, J.; Wallace, D. M.; Doherty, A.; Leung, H.; Young, L. S.; James,
N. D. Mol Ther 2009, 17, 1292.
15
16
Celik, A.; Yetis, G. Bioorgan Med Chem 2012, 20, 3540.
Kuroda, M.; Yamashita, A.; Hirakawa, H.; Kumano, M.; Morikawa, K.; Higashide, M.;
Maruyama, A.; Inose, Y.; Matoba, K.; Toh, H.; Kuhara, S.; Hattori, M.; Ohta, T. P Natl Acad Sci USA
2005, 102, 13272.
17
18
19
Goodyer, L. Int J Nurs Stud 2000, 37, 555.
Organization, W. H. In WHO Model List of Essential Medicines 2015. pp 13.
Dea-Ayuela, M. A.; Castillo, E.; Gonzalez-Alvarez, M.; Vega, C.; Rolon, M.; Bolas-Fernandez, F.;
Borras, J.; Gonzalez-Rosende, M. E. Bioorgan Med Chem 2009, 17, 7449.
20
21
22
Davis, B. G.; Celik, A.; Davies, G. J.; Ruane, K. M. In NOVEL ENZYME, United States, 2010.
Demir, A. S.; Talpur, F. N.; Sopaci, S. B.; Kohring, G. W.; Celik, A. J Biotechnol 2011, 152, 176.
Institute, C. a. L. S. In Performance Standards for Antimicrobial Susceptibility Testing; Twenty-
Fourth Informational Supplement; Clinical and Laboratory Standards Institute, USA, 2014. pp 68.
23
24
Sakinc, T.; Michalski, N.; Kleine, B.; Gatermann, S. G. FEMS microbiology letters 2009, 299, 60.
Majewsky, M.; Wagner, D.; Delay, M.; Brase, S.; Yargeau, V.; Horn, H. Chem Res Toxicol 2014,
27, 1821.
25
26
27
Bonev, B.; Hooper, J.; Parisot, J. J Antimicrob Chemother 2008, 61, 1295.
Sim, E.; Abuhammad, A.; Ryan, A. Br J Pharmacol 2014, 171, 2705.
Cribb, A. E.; Nakamura, H.; Grant, D. M.; Miller, M. A.; Spielberg, S. P. Biochem Pharmacol
1993, 45, 1277.
9

Graphical abstract
10