Synthesis, docking and antibacterial studies of more potent amine and hydrazone rifamycin congeners than rifampicin

Article abstract of DOI:10.1016/j.ejmech.2019.02.009

New rifamycin congeners (1-33) with incorporated amine and hydrazone substituents leading to lipophilic and/or basic nature and altered rigidity of modified C(3) arm were synthesized and structurally characterized in detail. NMR spectroscopic studies at d

Full text of DOI:10.1016/j.ejmech.2019.02.009

Accepted Manuscript
Synthesis, docking and antibacterial studies of more potent amine and hydrazone
rifamycin congeners than rifampicin
Krystian Pyta, Anna Janas, Monika Szukowska, Paulina Pecyna, Marcelina Jaworska,
Marzena Gajecka, Franz Bartl, Piotr Przybylski
PII:
S0223-5234(19)30119-9
DOI:
https://doi.org/10.1016/j.ejmech.2019.02.009
EJMECH 11101
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 27 October 2018
Revised Date: 18 January 2019
Accepted Date: 3 February 2019
Please cite this article as: K. Pyta, A. Janas, M. Szukowska, P. Pecyna, M. Jaworska, M. Gajecka, F.
Bartl, P. Przybylski, Synthesis, docking and antibacterial studies of more potent amine and hydrazone
rifamycin congeners than rifampicin, European Journal of Medicinal Chemistry (2019), doi: https://
doi.org/10.1016/j.ejmech.2019.02.009.
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.

ACCEPTED MANUSCRIPT
1
Graphical abstract
Synthesis, docking and antibacterial studies of more potent amine and hydrazone rifamycin
congeners than rifampicin
Krystian Pyta,a Anna Janas,a Monika Szukowska,a Paulina Pecyna,b Marcelina Jaworska,b
Marzena Gajecka,b, c Franz Bartl,d Piotr Przybylskia*
SAR of newly obtained rifamycin congeners of the two types: hydrazone (29) and amine (secondary and tertiary e.g.
15) has been discussed in view of their zwitterionic and non-ionic structures formed in solution. The most
antibacterially potent new rifamycin congeners (MICs ~ 8.5 nM), irrespectively whether they are of hydrazone or
amine type, were those having zwitterionic structure and rigid and protonated C(3) arms (15, 29), contributing to
improved docking mode to bacterial RNA polymerase and favorable water solubility.

ACCEPTED MANUSCRIPT
2
Highlights
•
•
•
•
•
33 new rifamycin congeners of hydrazone and amine types were synthesized
Zwitterionic and non-ionic rifamycins are detected in solution
Docking to RNAP, clogP, water solubility and antibacterial potency were studied
Rigidity and protonation of C(3) arm are crucial for rifamycin activity
Higher antibacterial potency of the congeners (~8.5 nM) than for rifampicin is noted

ACCEPTED MANUSCRIPT
3
Synthesis, docking and antibacterial studies of more potent amine
and hydrazone rifamycin congeners than rifampicin
Krystian Pyta,a Anna Janas,a Monika Szukowska,a Paulina Pecyna,b Marcelina Jaworska,b
Marzena Gajecka,b, c Franz Bartl,d Piotr Przybylskia*
aFaculty of Chemistry, A. Mickiewicz University, Umultowska 89b, 61-614 Poznan, Poland
bDepartment of Genetics and Pharmaceutical Microbiology, University of Medical Sciences,
Swiecickiego 4, 60-781 Poznan, Poland
cInstitute of Human Genetics, Polish Academy of Science, Strzeszynska 32, 60-479 Poznan, Poland
dHumboldt-Universität zu Berlin, Lebenswissenschaftliche Fakultät,, Institut für Biologie,
Biophysikalische Chemie, Invalidenstr. 42 10099 Berlin, Germany
*E-mail: piotrp@amu.edu.pl, tel. +48 61 8291693
ABSTRACT
New rifamycin congeners (1-33) with incorporated amine and hydrazone substituents leading to
lipophilic and/or basic nature and altered rigidity of modified C(3) arm were synthesized and
structurally characterized in detail. NMR spectroscopic studies at different temperatures indicate
two types of structures of rifamycin congeners that are realized in solution: zwitterionic and non-
ionic forms in dependence of the basicity of modified C(3) arm. The presence of rifamycin
congeners in these two possible forms has a significant impact on the physico-chemical
parameters such as lipophilicity (clogP) and water solubility and different binding mode of the
C(3) arm of antibiotic at RNAP binding pocket (molecular target) leading to different
antibacterial potency. The highest antibacterial potency against S. aureus (including MRSA and

ACCEPTED MANUSCRIPT
4
MLSB+ strains) and S. epidermidis strains, even higher than reference rifampicin (Rif) and
rifaximin (Rifx) antibiotics, was found for rifamycin congeners bearing at C(3) arm relatively
rigid and basic substituent (bipiperidine and guanidine groups). These modifications provide
favorable docking mode and excellent water solubility resulting in high potency (MICs 0.0078
µg/mL what gives ~ 8.5 nM), irrespective whether rifamycin congener is a tertiary amine (15) or
hydrazone (29). In turn, for a higher antibacterial potency of rifamycin congeners against E.
faecalis strain (MICs 0.5 µg/mL that is 0.6 µM) as compared to Rif and Rifx, the most crucial
factors are: bulkiness and the lipophilic character of the end of the C(3) rebuilt arm.
Keywords: Antibiotics; antibacterial; RNA polymerase inhibitors; drug-enzyme model, SAR.
1. Introduction
Rifamycins are ansamacrolide type antibiotics with medically important antibacterial potency
against Gram - positive and some Gram - negative bacteria strains, including especially
mycobacteria as e.g. M. bovis or M. tuberculosis [1–7]. This type of antibacterial ansamacrolides
such as rifampicin (Rif), rifapentine (Rifp), rifaximin (Rifx), rifalazil (Rifl) and rifabutin (Rifb)
have a common characteristic basket-like structure, built from “ansa” handle i.e. aliphatic ansa-
bridge joined to rigid aromatic core, which is crucial for their inhibitory activity of the
transcription process realized via bacterial DNA-dependent RNA polymerases (RNAP), at the
site located about 12 Å away from the active center of the enzyme [8
rifampicin (in US called rifampin, Rif), obtained from 3-formylrifamycin SV (Ral), belongs to
the most recognized and powerful representatives of this type of antibacterial agents [15 17].
Important problem concerning the use of rifampicin and related ansamacrolides in therapy is the
fast growing bacteria resistance, especially in treatment of many diseases [18 22]. The resistance
mechanism of bacteria towards rifampicin-like antibiotics is realized by several ways. This
–14]. Semisynthetic
–
–

ACCEPTED MANUSCRIPT
5
process comprises mutation of the RNAP’s enzyme sequence leading to substitutions of amino
acids at the binding pocket and nearby [23, 24], ribosylation, glucosylation or phosphorylation of
the ansa-bridge hydroxyls O(21)H and O(23)H thereby preventing effective antibiotic binding at
RNAP pocket [25, 26], as well as by hydrolysis of the C(3) arm and through monooxygenation of
nitrogens of amino-piperazine moiety which destabilizes the complex between the antibiotic and
RNAP [27, 28].
Bacteria resistance to rifamycin type antibiotics is also associated with a changed
structure of cell wall lipids in e.g. rifampicin-resistant strains that disturbs the transport of the
drug to the target [29]. Up to now, many active rifamycins have been designed and found to be
attractive for the treatment of bacterial infections. Synoradzki’s group earlier proposed an
interesting tandem approach via imino intermediates for the synthesis of rifamycins with α,β-
unsaturated ketones at C(3), however, their antibacterial potency have not been revealed so far
[30]. Recently, the same group described a new synthetic method for the preparation of stable
zwitterionic rifamycin derivatives with α,β-unsaturated pyrrolidinium substituents at C(3), which
exhibit antimycobacterial activities at MICs equal 12.5 µg/mL, which is in some cases even
better than that of Rif [31]. 1,8-Dibenzoyl-rifampicin is effective by its inhibitory effect on the P-
glycoprotein, which facilitates to overcome the blood-brain barrier for new drugs [32]. New
benzoxazinorifamycins, structurally similar to Rifx or Rifl, shown high in vitro activity against
M. tuberculosis (H37RV) at low MIC90’s level in the range from 0.08 to 0.004 µM [33].
Simultaneously they effectively inhibit the transcription process, catalyzed by RNA polymerases
of E. coli (wild-type and Rif-resistant mutants), even at IC50 below 0.01 µM [34]. Interestingly,
rifamycin analogs modified at C(3) bind to the β-forkloop of RNAP, whereas Rifl and Rifb
analogs bind between the β-forkloop and the σ-finger subunit of RNAP, which influences

ACCEPTED MANUSCRIPT
6
different profiles of antibacterial potency. The former group is more active against Gram-positive
whereas the latter one was more active against Gram-negative bacteria strains. In turn, new
rifamorpholines isolated from Amycolatopsis sp. Hca4, with a more rigid ansa-chain due to the
presence of cyclopropane ring, show lower antibacterial potency against panel of representative
bacteria S. aureus (MRSA), S. pyogenes, B. subtilis and M. luteus at MICs in the range 0.5 – 32
µM, that indicates the importance of the ansa-bridge flexibility for the biological effectivity of
rifamycins [35, 36]. Recently, we postulated that rifamycins with incorporated C(3) basic
substituent interacts mainly with N448 and E445 amino acids of RNAP (number in sequence
according to RNAP from T. aquaticus [8]), which is in line with the observed SAR for primary
amine analogs of rifampicin [7, 17, 37]. However, for the most promising C(3) benzyl-piperidyl-
primary amine Rif’s analog used in the form of salt we found antibacterial activities only close or
equal with those of Rif against S. aureus and S. epidermidis strains [37]. We demonstrated also
that for Rif and its active analogs, a conformational change of piperazine/piperidine rings
(rotation around C-N bond) was necessary in order to adopt an arrangement favorable for
interaction with E445 of RNAP. In aim to enhance antibacterial potency via improving e.g.
binding mode of the C(3) arm of rifamycins at RNAPs, we synthesized two series of new
rifamycin amine and hydrazone type congeners with mainly basic centres at C(3). Taking into
account recent results i.e. the discovering of the favorable formation of zwitterionic structures of
this type antibiotics in solution [7, 17, 37] and a rifampicin charge transfer polymorphism in solid
[38], we expected that rigid and basic substituents at C(3) arm should be desirable partners for
very a strong interaction with the highly conserved E445 of RNAPs. We supposed that this, on
the one hand, could contribute to better stability of the antibiotic-RNAP complex and, on the
other hand, should improve water solubility both enhancing the biological potency. The twenty

ACCEPTED MANUSCRIPT
7
six of our rifamycin congeners are new compounds, whereas derivatives 10, 12, 25, 27 and 30-33
were synthesized previously by other research teams [39 - 45].
2. Results and discussion
2.1. Chemistry
Reductive amination of 3-formyl rifamycin SV (Ral) with different primary and secondary
amines, performed with catalytic amounts of HCl in EtOH and NaBH3CN as reductant, results in
two series of new rifamycin congeners: one group with basic centres terminated with
hydrophobic or hydrophilic moieties (7, 8, 10 - 21) and the second group without additional basic
centres at C(3) arm (1 - 6, 9, 22, 23) (yields 31 – 73%, Scheme 1a). To compare the biological
potency of rifamycin amine (1 - 23) and hydrazone analogs of rifampicin, reactions between Ral
and respective hydrazines in CH2Cl2 were performed (hydrazones 24 - 33, yields 52 – 93%,
Scheme 1b). All congeners of hydrazone and amine types were characterized in detail by HPLC,
spectroscopic (1H, 13C NMR, FT-IR) and spectrometric (ESI MS) methods (see Supporting
Informations Tables 1S-8S) as well as by experimentally determined physico-chemical
parameters as lipophilicity (clogP) and water solubility (Table 9S).
In our recent studies it was reported that proton transfer process within rifampicin analogs
influenced their biological activity [37, 46]. Generally, for reaction products of primary amines
and Ral (compounds 1 - 9, Scheme 1a) we observed in respective 1H NMR spectra characteristic
signals ~ 9 ppm from diastereotopic –N+H2 protons at N(38) (Table 1S, Fig. 1 – 1 and 7 as
examples). These signals were assigned to the two –N+H2 protons of which one is involved in

ACCEPTED MANUSCRIPT
8
31
20
31
20
32
32
1
NH -
R
2
23
21
23
21
HO
HO
a)
or
36
36
HO
19
19
22
33
22
33
35
35
24
24
OH
OH
18
O
O
R1
OH
O
17
16
17
16
18
25
25
N
O
O
H
26
27
26
27
O
34
34
O
O
O
O
O
O
or
30
30
OH
OH
OH
OH
37
37
NaBH3CN
EtOH, HCl
N1H5
N1H5
28
28
OH OH
8
8
9
1
1
2
9
2
7
7
NH
O
29
29
14
14
H
3
3
R1
6
6
N
N
38
R1
10
10
O
O
O4H
O4H
5
5
O
38
12
12
O
O
OH
O
O
H
11
11
O
O
13
13
Ral
for: 1 - 9
for: 10 - 23
R1 =
O
O
O
O
O
O
O
2
4
5
1
3
6
N
NH
NH2
N
N
8
9
11
12
7
10
O
N
N
N
N
N
S
N
OH
O
N
F
F
H
N
14
15
13
16
17
N
N
N
N
O
OH
F
18
19
20
21
22
23
O
N
N
N
N
N
N
N
N
NH
O S O
S
N
O
N
F
F
F
31
20
32
23
21
HO
HO
36
b)
19
22
33
35
24
OH
OH
OH
O
O
17
16
18
25
O
O
NH2 -R2
26
27
34
O
O
O
O
30
OH
OH
37
N1H5
28
OH
8
1
CH2Cl2
NH
9
2
7
29
14
3
6
N
O
R2
10
O
O
O4H
5
38
12
OH
O
O
H
11
O
O
13
Ral
24 - 33
R2 =
N
27
24
25
26
28
N
H
N
N
N
N
H
H
H
H
Cl
HN
33
29
N
31
32
30
N
N
N
N
N
H
O
Scheme 1 Synthesis and structure together with the atom numbering of rifampicin inspired new:
a) secondary and tertiary amine and b) hydrazone rifamycin congeners.

ACCEPTED MANUSCRIPT
9
intramolecular hydrogen bond with the oxygen of lactam group and the second is exposed to
interactions with solvent molecules. Deprotonation of the O(8)H phenol within the 1 - 9
congeners is reflected in two resonance signals above 180 ppms, assigned to C(8) and C(11)
carbons (13C NMR – Table 2S). Similar values of these chemical shifts are related to the strong
resonance effect between phenolate at C(8) and p-ketone moiety at C(11) in result of the proton
1
transfer in solution. The recorded H NMR spectral image for 1 – 9 above 7 ppms, assigned on
the basis of 1H-1H COSY, 1H-13C HSCQC and 1H-13C HMBC two-dimensional NMR
experiments, provides clear evidence of the proton transfer from O(8) phenolic group to
secondary N(38) atom. The second group are tertiary amine congeners of rifamycins at C(3) arm
(compounds 10 - 23, Scheme 1a) for which some exemplary 1H NMR spectra are given in Fig. 1a
(for compounds 11, 15 and 17). For this type congeners the signal of the transferred proton from
phenol O(8)H to tertiary amine is well visible in the range 8-9.5 ppms. Additionally, for
derivative 15 in solution, the proton from the carboxyl terminal group is transferred to the
nitrogen N(41) of the piperidine ring and exchanged probably with water molecules because its
signal was detected above 3 ppm, similarly as in the case of 7 (Tables 1S and 3S). Thus, both 7
and 15 rifamycin congeners are present as double zwitterions of excellent water solubility. A
more complex situation was noted for derivative 10, where some equilibrium in solution was
found (Figs 1 and 2, Tables 3S and 4S). 1H NMR spectra recorded at different temperatures (Fig.
1b) clearly show double character of the signals, assigned to O(1)H, O(4)H, N(2)H and N+(38)H
of some A and B forms. At room temperature fast interconversion between A and B forms of 10
was observed, what was manifested by the broadened resonance signals. At -20º C the spectral
image of the A/B equilibrium has become more clarified. The mutual ratio of A and B forms is
strongly dependent on the water content in DMSO-d6. At the low temperature A form was
predominant when DMSO-d6 was dried over molecular sieves, whereas comparable amounts of

ACCEPTED MANUSCRIPT
10
Figure 1 Comparison of: a) 1H NMR spectra of 1, 7, 11, 15, 17, 28 and 32 in DMSO-d6 [c 0.005
M, 600 MHz] in the range δ=7-17 ppm, b) 1H NMR temperature-dependent spectra of 10 in 1:1
mixture of DMSO-d6:CD3CN with or without presence of water.

ACCEPTED MANUSCRIPT
11
A and B forms were detected in DMSO-d6 solution containing traces of water. Theoretical
calculations of the assumed models of 10A and 10B forms, together with the NMR data, suggest
that water molecule mediates the proton transfer between two tertiary nitrogens N(38) and N(40)
of equivalent basicity (Fig. 2) in this way contributing for the appearance of the doubled 1H NMR
signals and limited availability of the transferred proton for the other external proton-acceptors.
For structurally analogous congeners 12-14, comparable equilibrium was not observed due to the
bulkiness of N(40) substituents preventing stabilization of water molecule close to the piperazine
ring. A comparison of the 1H NMR spectra of hydrazone congeners 24 – 33 (examples 28 and 32,
Fig. 1b, Tables 5S and 7S) indicate that for 24 – 27 and 30 – 32 no proton transfer occurs because
the proton signal of the O(8)H phenol is well visible in the spectral range above 11 ppm. This
conclusion is true for the congeners 24 – 27 and 30 – 32, irrespective whether an aprotic or
aprotic including traces of water was used (Fig. 1S). In contrast to the hydrazone congeners
Figure 2 Water-dependent equilibrium between A and B forms of 10 (indicated experimentally
by 1H NMR and calculated by DFT method - DGauss using the B88-LYP GGA energy functional
with the DZVP basis sets), based on fluctuation of the transferred proton between two nitrogen
centers of equivalent basicity and mediated by water [the transferred proton between N(38) and
N(40) is marked by yellow].

ACCEPTED MANUSCRIPT
12
described above, for the compounds with chloroquinoline (28) and guanidine (29) bases at the
hydrazone moiety, proton transfer was observed due to the presence of an additional basic proton
acceptor centres, as proved by averaged position of C(8) and C(11) resonances in 13C NMR
spectra (Table 8S, Fig. 1S) and the lack of the O(8)H proton signal in the 1H NMR spectra (Fig.
1S, Table 7S).
2.2. Biological tests and docking studies
The new secondary amine 1 – 9, tertiary amine 10 – 23 and hydrazone 24 – 33 congeners
were tested against Gram-(+) and Gram-(-) bacteria including S. aureus (ATCC, MRSA and
MLSB+), S. epidermidis, E. faecalis and E. coli strains (Tables 1 and 2) in the range of MIC
0.0078 – 16 µg/mL. Antibacterial test results of rifamycin congeners 1- 33 were compared to
those of reference antibiotics such as rifampicin (Rif) and rifaximin (Rifx) as well as to the
synthetic precursor of the analyzed congeners - 3-formylrifamycin SV (rifaldehyde, Ral, Scheme
1).
Analysis of data given in Tables 1 and 2 show that all obtained rifamycin congeners,
irrespective of whether they are of amine or hydrazone types, together with reference compounds,
do not show any potency against E. coli strain. A comparison of the antibacterial potency of the
secondary amine rifamycin congeners (1 – 9) reveals that those with rather hydrophobic C(3)
substituents (1 and 8) and containing lactam/amide/sulfonamide groups (2 – 6 and 9) are of weak
or good activities against most tested strains (MICs are 4 – 0.125 µg/mL), which is, however,
significantly reduced when compared to Rif and Rifx. In turn, the rifamycin congener 7, with a
piperidine arm terminated with carboxylate (see structural discussion above), showed an only
four-fold lower antibacterial potency than Rif and a two-fold lower activity than Rifx, albeit,

ACCEPTED MANUSCRIPT
13
exclusively against S. epidermidis strains (MICs 0.0625 µg/mL). Analysis of the data for the
tertiary amine congeners (10 – 23) given in Tables 1 and 2, reveal that 16 is of comparable
activity as 7, despite a clear differences in their water solubilities (Table 9S and 10S). It should be
underlined, that water solubilities for 16 and 7 are higher than for Rif, whereby especially
pronounced changes are between 7 and Rif. In turn, 7 and 15 containing terminal carboxylates
have an extremly high water solubility (>100 mg/mL), in comparison to that of Rif (1.4 mg/mL),
and both should be expected as the most potent congeners according to this criterion. Some
drawback of congeners 7 and 15 is concerned with their limited lipophilicity (-clogP values)
whereas Rif has clogP = 1.66. In result of biological studies we found that the teriary amine
congener 15 was the most potent of all synthesized rifamycins, having MICs against I-V bacteria
strains equal 0.0078 µg/mL. Moreover, the antibacterial potency of 15, tested against S. aureus
and S. epidermidis strains, was about 2-4 and 4-16 fold greater than the reference rifamycins Rif
and Rifx, respectively. If MICs are expressed in nM units we obtained the following results: MIC
= 18.9 nM for Rif and MIC = 8.46 nM for 15 rifamycin congener (differeces are even greater).
Unexpectedly different activities were observed for compounds 11 (clogP = 1.23, watersol = 1.87
mg/mL), 16 (clogP = 0.72, watersol = 1.83 mg/mL) and 22 (clogP = 1.41, watersol = 1.73 mg/mL)
as compared to those of Rif (clogP = 1.66, watersol = 1.40 mg/mL) taking into account only clogP
and water solubility parameters. Furthermore, congener 15 is significantly more potent than 7,
despite of a similar: clogP, water solubility and structure of the C(3) arm end (end carboxylate
group). Differences between potencies of 15 and 7 and the lack of comparable activity between
e.g. 11, 16 and Rif are better understandable when their docking models at RNAP are considered
(Fig. 3).
Structures of rifamycin congeners 7, 11, 15 and 16 in solution, assumed on the basis

ACCEPTED MANUSCRIPT
14
NMR studies and initially optimized by B88 LYP(GGA) DFT method, were docked at X-ray
structure of RNAP binding pocket of T. aquaticus [8] using MOG-PM6 MYOZYME algorithm
for huge molecules, where mutual fitting of antibiotic and the molecular target took place (Fig 3
and Fig. 2S). A comparison of the interactions between 7, 11, 15 and 16 with bacterial RNAP
indicates that the C(3) arms of 15 and 16 are stabilized by additional intermolecular H-bonds to
E445 or Q1020 of RNAP (Fig. 3b and c, Table 10S), whereas for the zwitterionic form of 11 here
a proton transfer from the O(8)H phenol to the tertary nitrogen N(38) occurs (Fig. 2S) and
exclusively H-bonds of the ansamycin core with Q393, F394, D396, R409, S411 of β - fork loop
of RNAP are observed (Table 10S). These differences in stabilization of the C(3) arms of
rifamycin congeners at the RNAP binding pocket are also reflected in the calculated binding
energies (∆Hºf) for the docking models, where we found for 11 higher value (less favorable) than
for 7, 15 and 16 (Table 10S). A similar less effective binding mode as for 11, resulting from the
localization of the proton at nitrogen N(38) of the secondary amine but not at the tertiary amine,
can be expected also for derivatives 8-14 and 17-23 based on structure in solution determined by
NMR spectroscopy. Therefore, in the most abundant structures of congeners 8-14 and 17-23
detected in solution, in the presence of water, the tertiary nitrogens (when exist) are non-
protonated and thus not ready to interact with a highly conserved E445 of RNAP, what can
explain the lower potency of these congeners. If we compare the structures of 15 and Rif we can
conclude that the distances between the C(38) and the N(41)+-H group for 15 and between the
C(38) and the N(40)+-H groups for Rif, both in zwitterionic structures, are comparable and in
each of cases strong H-bond between the N+-H group and the carboxylate of E445 from RNAP is
formed (see example Fig. 3b). Molecular docking at the β-forkloop of RNAP indicates that the
presence of the carboxylate group for 15 creates an additional opportunity for the stabilization of
the C(3) arm via an extra H-bond (15)COO¯ …H-N(N448) of RNAP [distance donor-acceptor =

ACCEPTED MANUSCRIPT
15
2.54 Å; angle 126º] (Fig. 3b, Table 10S). The excellent water solubility and the improved
docking mode of 15 relative to those of Rif, clearly explain the extremely high antibacterial
potency of congener 15. Furthermore, differences in antibacterial activities between 7 and 15, at
their comparable physico-chemical properties, can also be explained in view of their binding
mode to RNAP. In structure of congener 7, additional H-bond intramolecular stabilization of the
transferred proton from carboxylate to nitrogen N(41), limits availability of the transferred proton
for possible intermolecular interaction with E445 (Fig. 3a). Furthermore, in consequence of
repulsion between the two negatively charged carboxylates of 7 and of E445 (RNAP),
intermolecular stabilization of the C(3) arm at the binding pocket of RNAP becomes less
favorable (Fig. 3a, Table 10S).
Antibacterial studies of the third group of hydrazone congeners (24 - 33) against S. aureus
and S. epidermidis I-V strains (Table 2) seem to confirm the impact of the basicity of the C(3)
arm on the potency of this type antibiotics. Relatively well-balanced water solubility (4.25
mg/mL) and lipophilicity (clogP = 1.33) together with a favorable binding mode, including the
strong interaction between protonated guanidine group of the C(3) arm and carboxylate of E445
(RNAP) for congener 29 (Fig. 3d) explain its very attractive potency, even greater when
compared to that of Rif. It should be noted here that the antibacterial potency even against
resistant S. aureus strains (MRSA and MLSB+) of 29 and 15 are extremely high and comparable
to each other, suggesting that such a level of antimicrobial potency can be achieved alternatively
for rifamycins of different types as a compromise between different factors: water solubility,
lipophilicity and improved binding mode to RNAP. Biological results obtained for 31 and 32
hydrazone congeners showed that they are rather active against I, II and IV and V bacteria
strains, probably due to the isostructural construction (lack of NH group) and bulkiness of the

ACCEPTED MANUSCRIPT
16
(a)
(b)
(d)
(c)
Figure 3 Docking models for new rifamycin congeners (grey): 7 (a), 15 (b), 16 (c) and 29 (d) at
the binding site of RNAP from T. aquaticus [8] (RNAP-yellow), optimized via MO-G PM6 semi-
empirical method using MOZYME algorithm for huge molecules (Scigress package 3.1.8, 2008-
2015). Intermolecular interactions between new rifamycin congeners and the key binding amino
acids (pink) of RNAP are marked by green dots.

ACCEPTED MANUSCRIPT
17
C(3) arm, when compared to that of Rif (Tables 1 and 2). It can be seen, that one disadvantage of
our synthesized hydrazone congeners 24-28 and 30-33 is their limited water solubility reflected in
a lower antibacterial potency, in contrast to Rif and 29.
Interesting SAR was noted for antimicrobial potency of 1 – 33 congeners against E. faecalis
bacteria strain (Tables 1 and 2). Rif and Rifx showed limited activity against E. faecalis at MICs
= 8 µg/mL. Our most potent rifamycin congeners against I-V reference bacteria strains (15 and
29), against E. faecalis (VI - Tables 1 and 2), revealed also higher potency (of 8-fold) in
comparison to those of Rif and Rifx. However, the best activity against E. faecalis strain of all
studied congeners showed 31 and 32 (at MICs 0.5 µg/mL what gives ~ 0.6 µM), isostructural to
Rif and with hydrophobic C(3) arms. Interestingly the congeners 1 and 25, which are of low
activity against I-V strains, were selectively more potent against E. faecalis (at MICs 1 µg/mL)
than Rif and Rifx because of the lipophilic nature of the C(3) arm. Thus, the hydrophobicity of
the C(3) arm, within the congeners, irrespective of their type, seems to be reflected in a
selectively improved potency against E. faecalis strain (for 6, 8, 11, 13, 16, 17, 24, 28 and 30 –
32, MICs 0.5 - 8 µg/mL).
3. Conclusions
Newly synthesized rifamycin congeners (1-33) bearing modified the C(3) arms of rigid and
lipophilic as well as rigid and basic nature revealed two-types structures in solution: zwitterionic
and non-ionic ones in dependence on the construction of the C(3) arm. For 10 tertiary amine
rifamycin congener the proton-transfer equilibrium, supported by the presence of water in
solution, was noted. The lack of attractive potency of 10 underlined the significance of the
availability of the transferred proton (from the O(8) phenol to rebuilt the C(3) arm) for

ACCEPTED MANUSCRIPT
18
intermolecular interactions with the negatively charged and highly conserved E445 of RNAP
stabilizing antibiotic-RNA polymerase complex. The presence of rifamycin congeners as
zwitterions (e.g. 33) and especially double zwitterions (e.g. 15) with the intramolecularly
transferred protons, localized at the C(3) rigid arm, improves water solubility and binding mode
of the antibiotic to molecular target - RNAP. Furthermore, improvement of the two above-
mentioned factors for rifamycin congeners resulted in the presence of extremely high
antibacterial potency against S. aureus (including MRSA and MLSB+ strains) and S. epidermidis
strains (MICs (15, 33) 0.0078 µg/mL what gives ~ 8.5 nM), even higher (2-16 fold) when compared
to those of reference antibiotics: rifampicin (Rif) and rifaximin (Rifx). In turn, biological potency
of rifamycin congeners against E. faecalis strain are dependent on the other factors as: bulkiness
and lipophilic character of the end of the rebuilt arm at C(3) position. For the most active
piperidine (31) and azepane (32) hydrazone rifamycin congeners MICs against E. faecalis strain
are lower of 8-fold (MICs 0.6 µM) than those noted for standard antibiotics Rif and Rifx.
Supporting Informations
Experimental procedures including synthesis (yields) and characterization data for all 1-33
1
congeners (copy of H, 13C NMR and FT-IR spectra) as well as HPLC, ESI MS data together with
full assignment of NMR resonances (Tables 1S-8S), experimentally determined physico-chemical
parameters of congeners 1-33 (clogP, water solubility – Table 9S), binding energies (∆Hf˚) and
intermolecular interactions stabilizing congeners 7, 11, 15, 16 and 29 at RNAP binding pocket
(Table 10S) and microbiological tests. This material is available free of charge via the Internet at
http://
Acknowledgements
Financial assistance of National Science Centre (NCN, Poland)
–
Grant UMO
2015/19/B/ST5/00231 is gratefully acknowledged by Authors. P. Przybylski wishes also to thank

ACCEPTED MANUSCRIPT
19
DAAD (Deutscher Akademischer Austauschdienst – German Academic Exchange Service) for
financial support of a part of research performed at Humboldt University (HU), Berlin, Germany.
References and notes
[1]
[2]
[3]
H.G. Floss, T.-W. Yu, Rifamycin mode of action, resistance, and biosynthesis, Chem.
Rev. 105 (2005) 621 – 632.
P. Sensi, Rifampicin, in: J.S. Bindera, D. Lednicer (Eds.), Chronicles of Drug
Discovery, Wiley, New York, 1982, pp. 201 – 221.
C.-K. Tan, C.-C. Lai, C.-H. Liao, S.-H. Lin, Y.-T. Huang, P.-R. Hsueh, Increased
rifampicin resistance in blood isolates of meticillin-resistant Staphylococcus aureus
(MRSA) amongst patients exposed to rifampicin-containing antituberculous treatment,
Int. J. Antimicrob. Agent. 37 (2011) 550 – 553.
[4]
E. Ngaimisi, S. Mugusi, O. Minzi, P. Sasi, K.-D. Riedel, A. Suda, N. Ueda, M. Janabi, F.
Mugusi, W. E. Haefeli, L. Bertilsson, J. Burhenne, E. Aklillu, Effect of rifampicin and
CYP2B6 genotype on long-term efavirenz autoinduction and plasma exposure in HIV
patients with or without tuberculosis, Clin. Pharmacol. Ther., 90 (2011) 406 – 413.
I. H. Kim, K. D. Combrink, Z. Ma, K. Chapo, D. Yan, P. Renick, T. W. Morris, M. Pulse,
J.W. Simecka, C. Z. Ding, Synthesis and antibacterial evaluation of a novel series of
rifabutin-like spirorifamycins, Bioorg. Med. Chem. Lett. 17 (2007) 1181 – 1184.
K. Bujnowski, L. Synoradzki, R. C. Darłak, T. A. Zevaco, E. Dinjus, Semi-synthetic
zwitterionic rifamycins: A promising class of antibiotics; Survey of their chemistry and
biological activities. RSC Adv. 6 (2016) 114758 – 114772.
[5]
[6]
[7]
[8]
[9]
K. Pyta, P. Przybylski, K. Klich, J. Stefańska, A new model of binding of rifampicin and
its amino analogues as zwitterions to bacterial RNA polymerase. Org. Biomol. Chem. 10
(2012) 8283 – 8297.
E. A. Campbell, N. Korzheva, A. Mustaev, K. Murakami, S. Nair, A. Goldfarb, S.A.
Darst, Structural mechanism for rifampicin inhibition of bacterial RNA polymerase, Cell
104 (2001) 901 – 912.
E. A. Campbell, O. Pavlova, N. Zenkin, F. Leon, H. Irschik, R. Jansen, K. Severinov, S.
A. Darst, Structural, functional, and genetic analysis of sorangicin inhibition of bacterial
RNA polymerase, EMBO 24 (2005) 674 – 682;

ACCEPTED MANUSCRIPT
20
[10] A. Feklistov, V. Mekler, Q. Jiang, L. F. Westblade, H. Irschik, R. Jansen, A. Mustaev, S.
A. Darst, R. H. Ebright, Rifamycins do not function by allosteric modulation of binding of
Mg2+ to the RNA polymerase active center, Proceed. Nat. Acad. Sci. USA 105 (2008)
14820 – 14825.
[11] I. Artsimovitch, M. N. Vassylyeva, D. Svetlov, V. Svetlov, A. Perederina, N. Igarashi, N.
Matsugaki, S. Wakatsuki, T. H. Tahirov, D. G. Vassylyev, Allosteric modulation of the
RNA polymerase catalytic reaction is an essential component of transcription control by
rifamycins, Cell 122 (2005) 351 – 363.
[12] L. Cella, H. Heumann, G. Baer, W. Werel, Mechanism of inhibition of DNA-dependent
RNA polymerase of Escherichia coli by rifamycins, Eur. J. Med. Chem. 24 (1989) 105 –
110.
[13] K. S. Murakami, X-ray Crystal Structure of Escherichia coli RNA Polymerase σ70
Holoenzyme, J. Biol. Chem. 288 (2013) 9126 – 9134.
[14] R. Mariani, S. I. Maffioli, Bacterial RNA Polymerase Inhibitors: An Organized Overview
of their Structure, Derivatives, Biological Activity and Current Clinical Development
Status. Curr. Med. Chem. 16 (2009) 430 – 454.
[15] Y.-F. Hu, C.-P. Liu, N.-Y. Wang, S.-C. Shih, In vitro antibacterial activity
of rifampicin in combination with imipenem, meropenem and doripenem against
multidrug-resistant clinical isolates of Pseudomonas aeruginosa. BMC Inf. Dis. 16 (2016)
art. No. 444.
[16] R. Figueiredo, D. F. Ramos, C. Moiteiro, M. A. Medeiros, M. J. Marcelo Curto, J.
Cardoso De Menezes, R. H. Pando, P. E. A. Silva, M. D. C. Costa, Pharmacophore
insights into rpoB gene mutations in Mycobacterium tuberculosis rifampicin resistant
isolates. Eur. J. Med. Chem. 47 (2012) 186 – 193.
[17] K. Pyta, P. Przybylski, B. Wicher, M. Gdaniec, J. Stefańska, Intramolecular proton
transfer impact on antibacterial properties of ansamycin antibiotic rifampicin and its new
amino analogues. Org. Biomol. Chem. 10 (2012) 2385 – 2388.
[18] U. T. Dang, I. Zamora, K. E. Hevener, S. Adhikari, X. Wu, J. G. Hurdle, Rifamycin
resistance in Clostridium difficile is generally associated with a low fitness burden.
Antimicrob. Ag. Chemother. 60 (2016) 5604 – 5607.

ACCEPTED MANUSCRIPT
21
[19] W. El Maraachli, M. Slater, Z. L. Berrada, S-Y. G. Lin, A. Catanzaro, E. Desmond, C.
Rodrigues, T. C. Victor, V. Crudu, M. T. Gler, T. C. Rodwell, Predicting differential
rifamycin resistance in clinical Mycobacterium tuberculosis isolates by specific rpoB
mutations. Int. J. Tuberc. Lung Dis. 19 (2015) 1222 – 1226.
[20] B. P. Goldstein, Resistance to rifampicin: A review. J. Antibiot. 67 (2014) 625 – 630.
[21] M. J. Pons, L. Mensa, J. Gascón, J. Ruiz, Fitness and molecular mechanisms of resistance
to rifaximin in in vitro selected Escherichia coli mutants. Microb. Drug Resist. 18 (2012)
376 – 379.
[22] V. Molodtsov, N. T. Scharf, M. A. Stefan, G. A. Garcia, K. S. Murakami, Structural basis
for rifamycin resistance of bacterial RNA polymerase by the three most clinically
important RpoB mutations found in Mycobacterium tuberculosis. Mol. Microbiol. 103
(2017) 1034-1045.
[23] Y.-T. Horng, W.-Y. Jeng, Y.-Y. Chen, C.-H. Liu, H.-Y. Dou, J.-J. Leef, K.-C. Chang, C.-
C. Chien, P.-C. Soo, Molecular analysis of codon 548 in the rpoB gene involved in
Mycobacterium tuberculosis resistance to rifampin. Antimicrob. Ag. Chemother. 59
(2015) 1542 – 1548.
[24] M. P. Cummings, M. R. Segal, Few amino acid positions in rpoB are associated with most
of the rifampin resistance in Mycobacterium tuberculosis. BMC Bioinform. 5 (2004) art.
no. 137.
[25] A. D. Giddey, E. De Kock, K. C.Nakedi, S. Garnett, A. J. M. Nel, N. C. Soares, J. M.
Blackburn, A temporal proteome dynamics study reveals the molecular basis of induced
phenotypic resistance in
Mycobacterium smegmatis at
sub-lethal
rifampicin
concentrations. Sci. Rep. 7 (2017) art. no. 43858.
[26] P. J. Stogios, G. Cox, P. Spanogiannopoulos, M. C. Pillon, N. Waglechner, T. Skarina, K.
Koteva, A. Guarné, A. Savchenko, G. D. Wright, Rifampin phosphotransferase is an
unusual antibiotic resistance kinase. Nat. Commun. 7 (2016) art. no. 11343.
[27] K. Yazawa, Y. Mikami, A. Maeda, M. Akao, N. Morisaki, S. Iwasaki, Inactivation of
rifampin by Nocardia brasiliensis. Antimicrob. Ag. Chemother. 37 (1993) 1313 – 1317.
[28] Y. Hoshino, S. Fujii, H. Shinonaga, K. Arai, F. Saito, T. Fukai, H. Satoh, Y. Miyazaki, J.
Ishikawa, Monooxygenation of rifampicin catalyzed by the rox gene product of Nocardia

ACCEPTED MANUSCRIPT
22
farcinica: structure elucidation, gene identification and role in drug resistance. J. Antibiot.
63 (2010) 23 – 28.
[29] N. Lahiri, R. R. Shah, E. Layre, D. Young, C. Ford, M. B. Murray, S. M. Fortune, D.
Branch Moody, Rifampin resistance mutations are associated with broad chemical
remodeling of mycobacterium tuberculosis. J. Biol. Chem. 291 (2016) 14248 – 14256.
[30] K. Bujnowski, L. Synoradzki, T. Zevaco, E. Dinjus, Rifamycin antibiotics - new
compounds and synthetic methods. Part 3: Study of the reaction of 3-formylrifamycin SV
with primary amines and ketones. Tetrahedron 68 (2012) 5925 – 5934.
[31] K. Bujnowski, L. Synoradzki, T. A. Zevaco, E. Dinjus, E. Augustynowicz-Kopec, A.
Napiorkowska, Rifamycin antibiotics - new compounds and synthetic methods. Part 4:
Study of the reaction of 3-formylrifamycin SV with secondary amines and ketones.
Tetrahedron 71 (2015) 158 – 169.
[32] V. Vilas-Boas, R. Silva, C. Nunes, S. Reis, L. Ferreira, C. Vieira, F. Carvalho, M. D. L.
Bastos, F. Remião, Mechanisms of P-gp inhibition and effects on membrane fluidity of a
new rifampicin derivative, 1,8-dibenzoyl-rifampicin.Toxicol. Lett. 220 (2013) 259 – 266.
[33] S. K. Gill, H. Xu, P. D. Kirchhoff, T. Cierpicki, A. J. Turbiak, B. Wan, N. Zhang, K. W.
Peng, S. G. Franzblau, G. A. Garcia, H. D. Showalter, Structure-based design of novel
benzoxazinorifamycins with potent binding affinity to wild-type and rifampin-resistant
mutant Mycobacterium tuberculosis RNA polymerases. J. Med. Chem. 55 (2012) 3814 –
3826.
[34] V. Molodtsov, I. N. Nawarathne, N. T. Scharf, P. D. Kirchhoff, H. D. H. Showalter, G. A.
Garcia, K. S. Murakami, X-ray crystal structures of the Escherichia coli RNA polymerase
in complex with benzoxazinorifamycins. J. Med. Chem. 56 (2013) 4758 – 4763.
[35] Y. S. Xiao, B. Zhang, M. Zhang, Z. K. Guo, X. Z. Deng, J. Shi, W. Li, R. H. Jiao, R. X.
Tan, H. M. Ge, Rifamorpholines A–E, potential antibiotics from locust-associated
actinobacteria Amycolatopsis sp. Hca4. Org. Biomol. Chem. 15 (2017) 3909 – 3916.
[36] G. Poce, M. Cocozza, S. Consalvi, M. Biava, SAR analysis of new anti-TB drugs
currently in pre-clinical and clinical development. Eur. J. Med. Chem. 86 (2014) 335 –
351.
[37] K. Pyta, K. Klich, J. Domagalska, P. Przybylski, Structure and evaluation of antibacterial
and antitubercular properties of new basic and heterocyclic 3-formylrifamycin SV

ACCEPTED MANUSCRIPT
23
derivatives obtained via ‘click chemistry’ approach. Eur. J. Med. Chem. 84 (2014) 651 –
676.
[38] L. de Pinho Pessoa Nogueira, Y. S. de Oliveira, J. de C. Fonseca, W. S. Costa, F. N.
Raffin, J. Ellena, A. P. Ayala, Crystalline structure of the marketed form of Rifampicin: a
case of conformational and charge transfer polymorphism. J. Mol. Struct. 1155 (2018)
260 – 266.
[39] N. Maggi, V. Arioli, P. Sensi, Rifamycins. XLI. A new class of active semisynthetic
rifamycins. N-substituted aminomethyl derivatives of rifamycin SV. J. Med. Chem. 8
(1965) 790 – 793.
[40] R. Cricchio, G. Carniti, G. Cietto, G. Tamborini, V. Arioli, Hydrazones of 3
formylrifamycin SV. Part III. N(mono and di)substituted hydrazone derivatives: synthesis,
antibacterial activity and other biological properties. Farmaco, Ed. Sci. 30 (1975) 704 –
720.
[41] N. Maggi, S. Furesz, P. Sensi, The influence of the carboxyl group upon the antibacterial
activity of rifamycins. J. Med. Chem. 11 (1968) 368-369.
[42] J. R. McCarthy, J. L. Moore, D. V. Wysong, C. D. Aldrich, 3-N-substituted derivatives of
rifamycin SV. A convenient method of synthesis, cyclization of certain derivatives, and
anticellular and antiviral activities of several derivatives. J. Med. Chem. 20 (1977) 1272 –
1276.
[43] A. N. Tischler, F. M. Thompson, L. J. Libertini, M. Calvin, Rifamycin derivatives as
inhibitors of a ribonucleic acid instructed deoxyribonucleic acid polymerase function.
Effect of lipophilicity. J. Med. Chem. 17 (1974) 948-352.
[44] C. Bartolucci, L. Cellai, P. Di Filippo, A. Segre, M. Brufani, L. Filocamo, A. D. Bianco,
M. Guiso, V. Brizzi, A. Benedetto, A. Di Caro, G. Elia, Rifamycins as inhibitors of
retroviral reverse transcriptase from M-MuLV, RAV-2, and HIV-1. Farmaco 47 (1992)
1367-1383.
[45] P. Traxler, W. Kump, K. Mueller, W. Tosch, Hypolipidemic activityof rifamycin
derivatives. J. Med. Chem. 33 (1990) 552-360.
[46] K. Pyta, P. Przybylski, F. Bartl, Regioselective Long-Range Proton Transfer in New
Rifamycin Antibiotics: A Process in which Crown Ethers Act as Stronger Brønsted Bases
than Amines. ChemPhysChem 16 (2015) 938 – 942.

ACCEPTED MANUSCRIPT
24
Table 1. MIC values [µg/mL] of 3-formylrifamycin SV (Ral), rifampicin (Rif) and rifaximin (Rifx) and the amine rifamycin
congeners 1-15, evaluated in the concentration range 0.0078-8 µg/mL.
MIC (Minimal inhibitory concentration)
Strain
rifamycin congener
Rif
Rifx
Ral
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
0.0156
0.0625
2
0.5
0.5
0.5
1
2
0.5
0.25
0.25
1
0.5
0.25
4
1
4
0.0078
I
0.0156
0.0625
4
0.5
0.5
4
0.25
0.5
0.5
1
2
0.5
0.5
0.25
0.25
4
0.25
0.5
0.25
1
0.5
1
0.25
0.125
0.125
0.125
8
2
1
4
0.0078
0.0078
0.0078
0.0078
1
II
0.03125 0.125
2
0.25
0.25
0.25
>8
1
4
0.125 0.5
2
0.25
0.25
0.25
8
8
III
IV
V
0.0156
0.0156
8
0.03125
2
0.25
0.25
>8
0.5
0.25
>8
>8
2
0.0625 0.125 0.5
0.0625 0.125 0.25
0.25
0.25
>8
0.5
1
4
0.03125
2
2
1
2
8
>8
>8
1
>8
>8
>8
>8
4
>8
>8
>8
>8
>8
>8
VI
VII
>8
>8
>8
>8
>8
>8
>8
>8
>8
>8
>8
SDMIC = 0
Strains:
I S. aureus ATCC 6538
II S. aureus MRSA
III S. aureus MLSB +
IV S. epidermidis ATCC 12228
V S. epidermidis ATCC 49134
VI E. faecalis ATCC 29212
VII E. coli ATCC 25922

ACCEPTED MANUSCRIPT
25
Table 2. MIC values [µg/mL] of amine 16-23 and hydrazone 24-33 rifamycin congeners, evaluated in the concentration range 0.0078-
8 µg/mL.
MIC µg/mL (Minimal Inhibitory Concentration)
Strain
rifamycin congener
16
17
8
18
19
2
20
1
21
2
22
23
4
24
1
25
26
0.5
1
27
0.5
0.5
1
28
29
30
31
32
33
0.25
2
0.5
0.5
0.5
0.5
0.25
0.25
0.0078
0.0156
0.25
0.25
0.5
0.0156
0.0156
0.5
0.0625
0.0625
0.5
1
I
0.125
0.125
0.0625
0.0625
8
1
2
1
4
2
1
0.5
0.5
1
II
8
0.5
0.25
0.25
2
2
2
2
2
0.5
0.5
0.125 0.0078
0.125 0.0078
III
IV
V
2
0.5
0.5
0.25
0.25
0.5
0.5
0.25 0.5
0.25 0.5
1
0.25
0.25
0.25 0.5
0.25
0.25
0.0156
0.03125
0.03125
0.03125
0.25
0.25
2
0.5
0.25 0.25 0.125 0.0078
2
4
>8
>8
>8
>8
>8
>8
>8
>8
>8
>8
>8
>8
8
1
>8
>8
>8
>8
8
1
4
0.5
>8
0.5
>8
>8
>8
VI
VII
>8
>8
>8
>8
>8
>8
>8
SDMIC = 0
Strains:
I S. aureus ATCC 6538
II S. aureus MRSA
III S. aureus MLSB +
IV S. epidermidis ATCC 12228
V S. epidermidis ATCC 49134
VI E. faecalis ATCC 29212
VII E. coli ATCC 25922