New and potent quinuclidine-based antimicrobial agents

Article abstract of DOI:10.3390/molecules24142675

Developing new antibiotics is currently very important since antibiotic resistance is one of the biggest problems of global health today. In the search for a new class of potential antimicrobial agents, ten new compounds were designed and synthesized based on the quinuclidinium heterocyclic core and the oxime functional group. The antimicrobial activity was assessed against a panel of representative gram-positive and gram-negative bacteria. All compounds demonstrated potent activity against the tested microorganisms, with the minimum inhibitory concentration (MIC) values ranging from 0.25 to 256.00 μg/mL. Among the tested compounds, two quaternary compounds, para-N-chlorobenzyl and meta-N-bromobenzyl quinuclidinium oximes, displayed the most potent and broad-spectrum activity against both gram-positive and gram-negative bacterial strains (MIC values from 0.25 to 4.00 μg/mL), with the lowest value for the important multidrug resistant gram-negative pathogen Pseudomonas aeruginosa. In the case of Klebsiella pneumoniae, activity of those compounds are 256-fold and 16-fold better than gentamicin, respectively. MTT assays showed that compounds are nontoxic for human cell lines. Multi-way analysis was used to separately reduce dimensionality of quantum chemical data and biological activity data to obtain a regression model and the required parameters for the enhancement of biological activity.

Full text of DOI:10.3390/molecules24142675

molecules
Article
New and Potent Quinuclidine-Based
Antimicrobial Agents
1
2
3
1
1
Andreja Radman Kastelic , Renata Odžak , Iskra Pezdirc , Karlo Sovi c´ , Tomica Hrenar ,
ˇ
4
5,
1,
Ana Cipak Gašparovi c´ , Mirjana Sko cˇ ibuši c´
*
and Ines Primoži cˇ *
1
Department of Chemistry, Faculty of Science, University of Zagreb, Horvatovac 102a,
HR-10000 Zagreb, Croatia
2
3
4
5
Department of Chemistry, Faculty of Science, University of Split, R. Boškovi c´ a 33, HR-21000 Split, Croatia
KB “Sestre milosrdnice”, Vinogradska cesta 29, HR-10000 Zagreb, Croatia
Institut Ruđer Boškovi c´ , Bijeni cˇ ka cesta 54, HR-10000 Zagreb, Croatia
Department of Biology, Faculty of Science, University of Split, R. Boškovi c´ a 33, HR-21000 Split, Croatia
Correspondence: mirjana.skocibusic@pmfst.hr (M.S.); ines.primozic@chem.pmf.hr (I.P.);
Tel.: +385-21-619-292 (M.S.); +385-1-4606-408 (I.P.)
*
ꢀ
ꢁꢂꢀꢃꢄꢅꢆꢇ
ꢀ
ꢁꢂꢃꢄꢅꢆ
Received: 9 June 2019; Accepted: 18 July 2019; Published: 23 July 2019
Abstract: Developing new antibiotics is currently very important since antibiotic resistance is one of
the biggest problems of global health today. In the search for a new class of potential antimicrobial
agents, ten new compounds were designed and synthesized based on the quinuclidinium heterocyclic
core and the oxime functional group. The antimicrobial activity was assessed against a panel of
representative gram-positive and gram-negative bacteria. All compounds demonstrated potent
activity against the tested microorganisms, with the minimum inhibitory concentration (MIC) values
ranging from 0.25 to 256.00 µg/mL. Among the tested compounds, two quaternary compounds,
para-N-chlorobenzyl and meta-N-bromobenzyl quinuclidinium oximes, displayed the most potent and
broad-spectrum activity against both gram-positive and gram-negative bacterial strains (MIC values
from 0.25 to 4.00
pathogen Pseudomonas aeruginosa. In the case of Klebsiella pneumoniae, activity of those compounds are
56-fold and 16-fold better than gentamicin, respectively. MTT assays showed that compounds are
µg/mL), with the lowest value for the important multidrug resistant gram-negative
2
nontoxic for human cell lines. Multi-way analysis was used to separately reduce dimensionality of
quantum chemical data and biological activity data to obtain a regression model and the required
parameters for the enhancement of biological activity.
Keywords: antimicrobial potency; quinuclidinium oximes; gram-positive and gram-negative bacteria;
multi-way analysis
1
. Introduction
Infectious diseases are the second leading cause of death worldwide and cause significant morbidity,
having a profound effect on global health [1]. The emergence of multidrug-resistant bacteria has become
a serious threat to public health and is considered one of the greatest challenges for contemporary
medicine. The ESKAPE group of bacterial pathogens (Enterococcus faecium, Staphylococcus aureus,
Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species) is
responsible for a majority of hospital-acquired infections and presents critical threats in nosocomial
pathogenesis, transmission, and resistance [2,3]. Various health organizations have spoken about
the urgent need to develop new antibiotics, especially against drug-resistant gram-negative ESKAPE
bacilli. For instance, the World Health Organization (WHO) has raised its utmost concern about the
post-antibiotic era where common infections and minor injuries may result in significant morbidity
Molecules 2019, 24, 2675; doi:10.3390/molecules24142675
www.mdpi.com/journal/molecules

Molecules 2019, 24, 2675
2 of 17
and mortality [4]. An important potential strategy to help combat the resistance problem involves the
discovery and development of new active agents capable of partly or completely suppressing bacterial
resistance mechanisms.
Heterocyclic compounds are the key motifs in a wide range of natural products such as vitamins,
carbohydrates, and nucleic and amino acids, and medicinal chemistry often engages in stimulating
such natural motifs. Heterocyclic compounds constitute a major portion of the currently approved
drugs by the Food and Drug Administration (FDA), and over 80% of top small molecule drugs by US
retail sales in 2010 contain at least one heterocyclic fragment in their structure [5]. Synthetic heterocycles
have widespread therapeutic uses such as antibacterial, antifungal, antimycobacterial, trypanocidal,
anti-HIV activity, antileishmanial agents, genotoxic, antitubercular, antimalarial, herbicidal, analgesic,
anti-inflammatory, muscle relaxants, anticonvulsant, anticancer and lipid peroxidation inhibitor,
hypnotics, antidepressant, antitumoral, and anthelmintic and insecticidal agents [6–10].
Quinuclidine, a bicyclic heterocyclic compound, is found in many natural products such as
the bark of Cinchona trees and is isolated by alkaloid extraction [11]. It has also been used in the
synthesis of catalysts which are useful in asymmetric reactions such as aldol reactions, Henry and
Aza–Henry reactions, Diels–Alder reactions, etc. [12,13]. Quinuclidine moiety is used to prepare
various therapeutically potent derivatives, such as FDA-approved drugs Azasetron, Benzoclidine,
Palonosetron, Solifenacin, and Quinupramine, which contain Quinuclidine as its main component
(Figure 1) [14].
CH3
N
O
Cl
O
N
N
O
HN
O
O
N
N
N
a)
b)
c)
H
N
O
O
N
H C
3
O
S
N
O
N
d)
e)
f)
) Solifenacin; (c) Quinupramine;
Figure 1. Quinuclidine-3-yl-based drugs:
(
a
) Azasetron; (
b
(
d) Palonosetron; (e) Cevimeline; and (f) Benzoclidine.
Avibactam, the first non-β-lactam
β-lactamase inhibitor, is also a quinuclidine-based drug, which
showed excellent inhibitory activity against most class A and class C and some class D
β-lactamases
and is proven to inhibit penicillin-binding proteins. Thus, quinuclidines are recognized as a desirable
structure and a useful template for design of novel compounds with potential pharmacological interest.
In our recent work, we reported synthesis of novel quaternary 3-hydroxy [15
,16] and
3-hydroxyimino [17] quinuclidinium bromides with different alkyl chains lengths (3–16 carbon
atoms) and showed that bicyclic quinuclidinium compunds have excellent and promising activity
against most gram-positive and clinically relevant gram-negative bacterial strains compared to
conventional antimicrobial agents. One of the expected antimicrobial mechanisms of cationic biocides
involves the destructive interaction with the outer membrane [18]. In this work, ten different

Molecules 2019, 24, 2675
3 of 17
quaternary N-alkyl/alkylaryl derivatives of 3-hydroxyiminoquinuclidine were prepared; see Figure 2.
All compounds are new and not previously described in the literature. For all compounds, antimicrobial
activity toward multidrug-resistant gram-negative and gram-positive bacteria was estimated.
NOH
NOH
Br-
N
qox
N
R
Comp.
R
Comp.
R
1
2
3
4
5
But-3-enyl
Benzyl
6
7
8
9
10
meta-ClC H CH –
6 5 2
para-CH C H CH –
3
6
5
2
para-NO C H CH –
meta-CH C H CH –
2
6
5
2
3 6 5 2
meta-NO C H CH –
para-BrC H CH –
6 5 2
2
6
5
2
para-ClC H CH –
meta-BrC H CH –
6
5
2
6 5 2
Figure 2. Synthesized compounds: 3-hydroxyiminoquinuclidine (qox) and quaternary derivatives 1–10.
2
. Results and Discussion
2
.1. Synthesis of Compounds
The initial compound for the oxime synthesis was quinuclidin-3-one prepared from commercially
available quinuclidin-3-one hydrochloride by a previously described procedure with potassium
carbonate as a base in a high yield [19]. 3-Hydroxyiminoquinuclidine was prepared by the reaction
of quinuclidin-3-one with hydroxylamine hydrochloride and sodium hydroxide [20] (Figure 3).
NMR analysis showed that prepared oxime has an (E)-configuration which is expected to be
thermodynamically favored as determined previously [17]. To prepare compounds 1–10, a Menshutkin
reaction was employed to convert tertiary amine to quaternary ammonium salt by the reaction of
appropriate alkyl/aryl halide. As a rule, the optimal conditions were obtained with 1 eq. of alkyl/aryl
bromide in dry tetrahydrofurane at reflux temperature under nitrogen atmosphere. The products
were obtained in 60–90% yield, and all of them were solids that precipitated from the reaction mixture
using diethyl ether. The pK
of qox were already determined experimentally: 10.805
As expected, a positive charge on the quinuclidinium nitrogen atom upon quarternization stabilizes
a
values for the oxime hydroxyl group and the quinuclidine nitrogen atom
±
0.030 and 8.072 0.034, respectively [17].
±
the conjugated base and the acidity of the oxime group of compound 10 is increased to 10.05
Compounds 1–10 are new compounds and are not previously described in the literature.
±
0.03.
O
O
NOH
a
b
N
Cl-
N
N
qox
H
NOH
c
N
Br-
R
1-10
Figure 3. The synthetic route to target compounds. Reagents and conditions: (
a
) K CO , H O, Et O,
2 3 2 2
◦
◦
HCl, NaOH, MeOH, 25 C, 1 h; and (c) alkyl/aryl bromide, THF, Et O,
2
2
6
5 C, 30 min; (
6 C, 5 h.
b
) NH OH
×
2
◦

Molecules 2019, 24, 2675
4 of 17
2
.2. Antimicrobial Activity
In this study, all synthesized novel N-alkyl derivatives of quinuclidinium oximes
1–
10 were
preliminary screened against a diverse panel of selected antibiotic susceptible gram-positive
bacteria purchased from the American Type Culture Collection (ATCC), Rockville, MD, USA:
Bacillus cereus ATCC 11778, Enterococcus faecalis ATCC 29212, Staphylococcus aureus ATCC 25923,
and Clostridium perfringens; and antibiotic resistant gram-negative bacteria obtained from Microbiology
Laboratory, Department of Biology, Faculty of Science, University of Split, Croatia (FNSST):
Escherichia coli FNSST 982, Klebsiella pneumoniae FNSST 011, and Pseudomonas aeruginosa FNSST
9
82 by disc diffusion assay. Cefotaxime and gentamicin, which are well-known relevant antimicrobial
agents [21 22] were included in the assays, serving as positive controls and as a basis for comparison.
,
The compounds showed to be active against antibiotic susceptible gram-positive bacteria but also
exhibited significant antibacterial effects against gram-negative gentamicin-resistant isolates. As shown
in Table 1, the results of the disc diffusion assay pointed out that the compounds exhibit moderate
to high potency and broad spectrum activity against the bacterial strains tested. The mean zones
of inhibition of the target compounds against all bacterial strains tested were found in the range of
6
.2
nonquaternary 3-hydroxyiminoquinuclidine (QOX) showed very poor activity against all strains tested
7.6 1.1 to 10.8 1.4 mm). The antimicrobial activity trend did not change significantly for quaternary
compound with butenyl group as a substituent, but the compounds containing benzyl moiety showed
±
1.3 to 29.8
±
1.3 mm. It is noteworthy that, among the synthesized compounds, unsubstituted
(
±
±
1
appreciably better antimicrobial effects; see Table 1. As expected, the strongest and most promising
antimicrobial potential against a panel of tested strains was observed with compounds which have a
halogen atom as a substituent (
substituent and and 10 with bromine atoms). Among all tested derivatives, 10 showed the maximum
inhibition against all tested bacterial strains (21.0 0.8 to 29.8 0.1.3 mm). The highest antibacterial
5 and 6 with chlorine atoms at the meta and para positions of the benzyl
9
±
±
activity was recorded against Pseudomonas aeruginosa, followed by Klebsiella pneumoniae and Bacillus
cereus.
Table 1. Antimicrobial activity of 3-hydroxyiminoquinuclidine (qox) and its quaternary salts (
1–10) by
disc diffusion assay.
Diameters of the Inhibition Zone (mm)
Comp.
Gram-Positive Bacteria
Gram-Negative Bacteria
Bacillus
cereus
Enterococcus
faecalis
Staphylococcus
Escherichia
coli
Klebsiella
pneumoniae
Pseudomonas
aeruginosa
aureus
qox
1
8.2 ± 0.9
9.9 ± 1.4
7.7 ± 1.4
6.2 ± 1.3
14.2 ± 2.1
12.7 ± 1.4
15.4 ± 1.2
19.4 ± 1.8
17.4 ± 1.7
6.4 ± 2.1
9.6 ± 1.2
17.3 ± 2.4
23.0 ± 1.4
13.2 ± 0.6
10.4 ± 1.7
9.3 ± 0.8
8.6 ± 1.9
9.6 ± 1.1
14.8 ± 1.2
15.1 ± 1.2
17.7 ± 0.9
20.8 ± 2.1
19.1 ± 1.5
8.1 ± 1.2
6.6 ± 1.1
22.7 ± 0.8
22.4 ± 0.9
11.5 ± 0.9
7.6 ± 1.1
8.4 ± 1.1
16.1 ± 0.9
13.7 ± 1.8
14.7 ± 0.4
19.8 ± 0.4
18.1 ± 1.8
7.5 ± 1.4
8.6 ± 1.1
24.7 ± 1.3
25.1 ± 0.6
16.9 ± 0.3
10.8 ± 1.4
12.1 ± 1.8
19.2 ± 1.4
16.3 ± 1.5
18.7 ± 0.9
24.8 ± 1.7
20.1 ± 2.1
11.0 ± 0.8
10.6 ± 1.1
27.6 ± 1.7
29.8 ± 1.3
9.4 ± 0.6
2
12.1 ± 1.8
10.4 ± 1.5
16.1 ± 0.4
18.2 ± 1.2
16.8 ± 0.8
10.8 ± 1.7
7.8 ± 0.3
23.5 ± 1.3
24.0 ± 1.2
18.2 ± 0.7
15.7 ± 1.7
13.5 ± 1.7
18.1 ± 1.1
20.8 ± 1.1
17.2 ± 0.3
9.5 ± 1.3
11.4 ± 0.7
27.2 ± 1.3
21.0 ± 0.8
23.9 ± 0.4
3
4
5
6
7
8
9
1
0
Gen
Antimicrobial efficacy was also evaluated by the broth microdilution method to determine the
minimum inhibitory concentrations (MICs) of a series of novel derivatives of quinuclidinium oxime
derivatives necessary to inhibit bacterial growth. Cefotaxime and gentamicin were also included in the

Molecules 2019, 24, 2675
5 of 17
assays as positive controls. The results presented in Table 2 show moderate to excellent antibacterial
activity against tested strains with MIC values from 0.25 to 256.00 g/mL. Nonquaternary derivative
qox showed very poor activity against all strains tested (MIC = 256 g/mL except for S. aureus and
P. aeruginosa 128 g/mL). On the other hand, evaluations pointed out compound 10 as the most
active compound in the series, especially against gram-negative bacteria (MIC values from 0.25 to
.00 g/mL). Compound 10 exhibited excellent in vitro activity against a broad spectrum of clinically
important resistant gram-negative pathogens such as Pseudomonas aeruginosa (MIC = 0.25 g/mL) and
Klebsiella pneumoniae (MIC = 0.50 g/mL). These MICs are 256-fold and 16-fold better than gentamicin,
respectively. Compound also showed very strong activity against all tested gram-negative bacteria,
with MIC values from 0.25 to 1.00 g/mL. Compounds and showed very weak antimicrobial
µ
µ
µ
1
µ
µ
µ
5
µ
7
8
properties among quaternary derivatives. These compounds have a methyl group as a substituent
on an aromatic ring that donates electrons, unlike the electronegative nitro group and halogen atom
present in other substituted compounds. The data indicate that, in general, the quaternary N-benzyl
derivatives of quinuclidine oximes
2
–10 are more potent and have broader antimicrobial activity
spectrums compared to their uncharged analog qox and butyl derivative
1. A comparison of meta- and
para-substituted derivatives reveals that the position of the substituent at the benzyl moiety does not
correlate with antimicrobial efficacy.
Table 2. In vitro antibacterial activities for compounds 1–10 expressed as minimum inhibitory
a,b
concentrations (MIC; µg/mL)
.
MIC (µg/mL)
Comp.
Gram-Positive Bacteria
Gram-Negative Bacteria
B. cereus
E. faecalis
256.00
64.00
32.00
16.00
2.00
S. aureus
E. coli
128.00
16.00
16.00
32.00
1.00
K. pneumoniae
256.00
16.00
P. aeruginosa
1
2
3
4
5
6
7
8
9
128.00
64.00
64.00
32.00
4.00
128.00
32.00
32.00
16.00
1.00
128.00
8.00
64.00
32.00
16.00
0.25
32.00
2.00
8.00
4.00
4.00
4.00
8.00
2.00
64.00
128.00
1.00
128.00
64.00
8.00
64.00
64.00
12.50
2.00
128.00
256.00
25.00
1.00
256.00
128.00
50.00
32.00
128.00
0.50
1
0
1.00
4.00
0.50
0.25
Gentamicin
Cefotaxime
4.00
4.00
1.00
32.00
0.50
8.00
64.00
16.00
0.25
0.50
0.50
0.50
a
Minimum inhibitory concentrations were determined by the micro broth dilution method for microdilution plates.
Gentamicin and cefotaxime were used as the positive controls.
b
2
.3. MTT Assay
The results of MTT assay are shown in Figure 4. Compared to normal physiological cell growth,
the compounds did not cause significant changes in HaCat or HMEC cell growth; see Figure 4a.
In addition, testing a wider concentration range of compounds with confirmed antibacterial activity
showed no changes in cell viability; see Figure 4b.

Molecules 2019, 24, 2675
6 of 17
Figure 4. Effects of all synthesized compounds on normal human cell lines viability: Concentrations up
to (a) 100 µM; (b) 200 µM; and (c) assay controls H O and kanamycin.
2
2
2
.4. Effects of the Compounds on Cellular Reactive Oxygen Species and Antioxidative Defence
The effects of the compounds with antibacterial activity, , and 10, were further tested for the
5, 6, 9
ability to change intracellular reactive oxygen species (ROS) levels and on the cellular antioxidative
defense system glutathione (GSH) and catalase (Figure 5). The HMEC cell line showed higher basal
levels of ROS when compared to the HaCaT cell line (p < 0.05) (Figure 5a). The compounds had
different effects on ROS level depending on the cell line. In HaCaT cells,
at the lowest concentration applied and the effect diminished with increasing concentration of the
compound. In HMEC cells, the compounds decreased ROS with the exception of 1 M of (p < 0.05).
These results are in accordance with catalase activity (Figure 5b). In HaCaT cells, catalase activity is
5, 6, 9, and 10 increased ROS
µ
5

Molecules 2019, 24, 2675
7 of 17
changed neither by DMSO nor the compounds. In contrast, in HMEC cells, the activity increased with
addition of DMSO and the compounds; therefore, the effects of the compounds themselves cannot be
resolved, as their effect on catalase activity is the same as that of DMSO. Unlike catalase, GSH levels are
not changed by the addition of DMSO,
Figure 5c). These results indicate that DMSO could cause a slight shift in the cellular redox balance
but without influencing cell viability. The potential mechanism is to be investigated.
5, 6, 9, or 10, with the exception of 200 µM of 6 in HMEC cells
(
Figure 5. Tests of the ability of compounds
1–10 to change (a) intracellular ROS levels with assay
controls H O and kanamycin; (b) cellular antioxidative defense system GSH; and (c) catalase.
2
2

Molecules 2019, 24, 2675
8 of 17
2
.5. Multi-Way Analysis
To additionally improve the relationship between structure and antimicrobial activity, tensor
decomposition methods were utilized on the potential energy surfaces (PES) and biological activity
data. First, a full conformational analysis for all compounds was performed. Conformers were
determined by using the recently developed approach consisting of ab initio molecular dynamics
and probability search in reduced space [23]. The example of conducted conformational analysis is
presented on Figure 6.
Figure 6. Total number of strict local maxima in the probability distribution for geometries of
compound
2 in the reduced space calculated by the principal component analysis of the molecular
dynamics trajectory.
A plateau of the local maxima in probability distribution is achieved with 4 principal components
and approx. 1,000,000 points in the molecular dynamics trajectory.
The lowest energy conformer for each compound was subjected to quantum chemical calculation
of two-dimensional PES, which was calculated in dependence on two torsional coordinates (Scheme 1).
NOH
N
Scheme 1. Torsional coordinates used in 2-D potential energy surface calculation.
Equivalent torsional coordinates were used for all other compounds. After the 2-D PES scan,
a multi-way analysis of energy data was performed for compounds 2–10. Tensor of PES data was
decomposed using the Tucker3 model, and the first principal component described 91% of the total
variance among the data, providing an excellent description of molecular energetics. On the other
hand, biological activity data were decomposed using principal component analysis (PCA) and the first
principal component described 82% of the total variance among the data. Again, this one component
was sufficient to describe all important trends in biological activity data. A polynomial regression
model between the biological activity and potential energy surfaces was established and is presented
on Figure 7.

Molecules 2019, 24, 2675
9 of 17
Figure 7. Multivariate regression model of the first principal component of biological activity in
dependence on the first principal component of potential energy surfaces.
2
The R value for the established polynomial model of degree 3 was 0.9623, and the model very
well describes the biological activity of investigated compounds in relation to their potential energy
surfaces. It can be seen that the two most active compounds
the first principal component describing biological activity. In a similar matter, the two least active
derivatives and present extrema points on the activity axis. A regression model between calculated
5 and 10 have almost the same value along
7
8
potential energy surfaces and biological activity enables in silico experiments with compounds having
slightly different variations in the structure. In this way, the structure of compounds can be adjusted
and optimized in search of biological activity enhancement within the tested bacteria set. To confirm
the validity of the established model, a leave-one-out cross validation procedure was performed.
The mean squared error of prediction in dependence on the polynomial degree is presented in Figure 8.
Figure 8. Mean squared error of model prediction obtained by leave-one-out cross validation on the set
of the first principal component of biological activity in dependence on the first principal component of
2
potential energy surfaces. (R values for each model are placed above the bars.)
The presented polynomial model of degree 3 has the lowest mean squared error with high value
2
for the coefficient of determination R , thus ensuring good prediction properties in the domain ranging
within the values of the first principal component of the decomposed potential energy surfaces.

Molecules 2019, 24, 2675
10 of 17
3
. Materials and Methods
3
.1. Synthesis of Compounds
All reagents and solvents were used as purchased from commercial suppliers without further
purification. Thin-layer chromatography (TLC) was performed on Aluminum oxide 60 F254 plates
Sigma-Aldrich, St. Louis, MO, USA) and visualized under UV light (254 nm) or by iodine fumes.
(
Melting points were determined on a Melting Point B-540 apparatus (Büchi Labortechnik GmbH, Essen,
Germany) and are uncorrected. Elemental analysis (CHN) was performed on a Perkin Elmer 2400 Series
II CHNS analyzer (PerkinElmer, Inc., Waltham, MA, USA), and the purity of all compounds was
≥
99%.
FTIR spectra were recorded as KBr pellets on a Perkin-Elmer Spectrum Two (PerkinElmer, Inc., Waltham,
−
1
MA, USA), and the signals were reported in cm . NMR spectra were recorded on a Bruker Avance III
◦
HD 400 MHz/54 mm Ascend spectrometer (Bruker Corporation, Billerica, MA, USA) at 22 C in CDCl
3
and DMSO with tetramethylsilane as the internal standard. Chemical shifts are given in ppm downfield
from tetramethylsilane as the internal standard, and coupling constants are given (J) in Hz. Splitting
patterns are designated as s (singlet), d (doublet), dd (doublet of doublets), t (triplet), q (quartet), or m
(
multiplet). Anhydrous reactions were carried out in dried glassware under a nitrogen atmosphere.
The compound qox was synthesized from quinuclidin-3-one hydrochloride ( 98.0%, Fluka, Honeywell
Research Chemicals, Charlotte, NC USA) following published procedure [20]. Compound was
prepared in the reaction of qox with butenyl bromide, was prepared with benzyl bromide, and 10
≥
1
2
3–
were prepared with appropriate meta- and para-substituted benzyl bromides. All halides were obtained
from Sigma–Aldrich Co., St. Louis, MO, USA and used without further purification.
General Procedure for the Synthesis of N-quaternary 3-hydroxyiminoquinuclidinium bromides 1–10
To the solution of 3-hydroxyquinuclidine (1 mmol) in minimal volume of dry tetrahydrofuran
equimolar amount of the appropriate quaternization reagent (butenyl bromide benzyl bromide, para-
and meta-nitrobenzyl bromide, para- and meta-chlorobenzyl bromide, para- and meta-methylbenzyl
bromide, and para- and meta-bromobenzyl bromide each) was added at room temperature. The reaction
mixture was refluxed overnight under nitrogen atmosphere to obtain a solid product which was
washed several times with dry diethyl ether. Digestion gave appropriate quaternary compounds as
white solids.
∼
◦
−1
:
N-but-3-enyl-3-hydroxyquinuclidinium bromide (
319–2686, 1643, 1463, 1423, 1085, 838, 700. H NMR (400 MHz, DMSO-d )
1
). Yield: 62%; mp: 211.5–212.1 C; IR (KBr)
ν
/cm
1
3
δ
/ppm: 1.89–2.14 (m, 4H,
6
H8, H5), 2.48–2.56 (m, 2H, H10), 2.79–2.82 (m, 1H, H4), 3.37–3.66 (m, 8H, H6, H7, H9), 4.36 (s, 2H, H2),
5
δ
.13–5.26 (m, 2H, H12), 5.72–5.83 (m, 1H, H11), 11.15 (s, 1H, C=N–OH). 13C NMR (101 MHz, DMSO-d6)
/ppm: 23.53 (C5, C8), 26.48 (C10), 26.95 (C4), 54.73 (C6, C7), 57.73 (C2), 62.48 (C9), 118.86 (C12), 133.62
(
C11), 151.62 (C3). CHN analysis/%: Anal. calcd. for C H BrN O: C 48.01; H 6.96; Br 29.04; N 10.18;
11 19 2
O 5.81. Found: C 48.14; H 6.95; N 10.20.
◦
N-benzyl-3-hydroxyiminoquinuclidinium bromide (
2
). Yield: 85%; mp: 219.7–222.6 C; IR (KBr)
∼
−1
1
ν
/cm : 3390–2689, 1673, 1460, 1419, 1063, 944, 770, 713. H NMR (400 MHz, DMSO-d6)
δ/ppm:
1
.88–2.15 (m, 4H, H5, H8), 2.76–2.81 (m, 1H, H4), 3.49–3.66 (m, 4H, H6, H7), 4.32 (m, 2H, H2), 4.66–4.69
13
(
m, 2H, Ar-CH ), 7.54–7.57 (m, 5H, Ar), 11.14 (s, 1H C=N–OH). C NMR (101 MHz, DMSO-d6)
δ
/ppm:
3.49 (C5, C8), 27.47 (C4), 54.55 (C6, C7), 57.31 (C2), 66.64 (C9), 127.81 (C10), 129.55; 130.79; 133.54;
C11-C15), 151.41 (C3). CHN analysis/%: Anal. calcd. for C H BrN O: C 54.03, H 6.15, Br 25.67, N
2
2
(
14
19
2
9
.00, O 5.14. Found: C 53.88, H 6.16, N 9.02.
◦
N-(para-nitrobenzyl)-3-hydroxyiminoquinuclidinium bromide (
3
). Yield: 91%; mp: 232.5–233.2 C;
∼
−1
1
IR (KBr)
ν
/cm : 3214–2846, 1608, 1524, 1353, 958, 856, 706. H NMR (400 MHz, DMSO-d ) /ppm:
δ
6
1
2
.88–2.15 (m, 4H, H5, H8), 2.76–2.81 (m, 1H H4), 3.49–3.66 (m, 4H, H6, H7), 4.37 (m, 2H, H2), 4.83 (s,
13
H, Ar-CH ), 7.84–7.89 (m, 2H, H11, H15), 8.36–8.41 (m, 2H, H12, H14), 11.16 (s, 1H C=N–OH).
C
2
NMR (101 MHz, DMSO-d6) δ/ppm: 23.52 (C5, C8), 27.33 (C4), 54.87 (C6, C7), 57.51 (C2), 65.19 (C9),

Molecules 2019, 24, 2675
11 of 17
134.89 (C10), 124.40 (C11, C15), 135.14 (C12,C14), 149.12 (C13), 151.21 (C3). CHN analysis/%: Anal.
calcd. for C H BrN O : C 47.20, H 5.09, Br 22.43, N 11.80, O 13.47. Found: C 47.10, H 5.08, N 11.81.
1
4
18
3
3
◦
N-(meta-nitrobenzyl)-3-hydroxyiminoquinuclidinium bromide (
4
). Yield: 81%; mp: 261–262.6 C; IR
∼
−1
1
(
KBr)
ν
/cm : 3377–2846, 1670, 1618, 1531, 1353, 945, 707. H NMR (400 MHz, DMSO-d6) /ppm:
δ
1.88–2.15 (m, 4H, H5, H8), 2.76–2.81 (m, 1H H4), 3.49–3.66 (m, 4H, H6, H7), 4.32 (m, 2H, H2), 4.66–4.69
(
m, 2H, Ar–CH ), 7.80–7.89 (m, 1H, H14), 8.00–8.06 (m, 1H, H15), 8.38–8.42 (m, 1H, H13), 8.45–8.47
2
13
(
m, 1H, H11), 11.16 (s, 1H C=N–OH). C NMR (101 MHz, DMSO-d6)
C4), 54.76 (C6, C7), 57.45 (C2), 65.21 (C9), 129.78 (C10), 125.60, 125.25, 131.16, 140.10 (C11–C15), 148.49
C12), 151.23 (C3). CHN analysis/%: Anal. calcd. for C H BrN O : C 47.20, H 5.09, Br 22.43, N 11.80,
δ/ppm: 23.52 (C5, C8), 27.41
(
(
14
18
3
3
O 13.47. Found: C 47.32, H 5.10, N 11.81.
◦
N-(para-chlorobenzyl)-3-hydroxyiminoquinuclidinium bromide (
5
). Yield: 92%; mp: 239.7–240.3 C; IR
∼
−1
1
(KBr)
ν
/cm : 3184–2781, 1652, 1598, 1493, 1437, 1384, 951, 883, 626. H NMR (400 MHz, DMSO-d )
6
δ
/ppm: 1.88–2.15 (m, 4H, H5, H8), 2.76–2.81 (m, 1H H4), 3.49–3.63 (m, 4H, H6, H7), 4.32 (m, 2H, H2),
13
4
.68 (s, 2H, Ar–CH ), 7.56–7.65 (m, 4H, H11, H12, H14, H15), 11.14 (s, 1H C=N–OH). C NMR (101
2
MHz, DMSO-d6)
δ/ppm: 23.49 (C5, C8), 27.44 (C4), 54.54 (C6, C7), 57.28 (C2), 65.64 (C9), 126.80 (C10),
129.59 (C11, C15), 135.65 (C12,C14), 135.80 (C13), 151.35 (C3). CHN analysis/%: Anal. calcd. for
C H BrClN O: C 48.65, H 5.25, Br 23.12, Cl 10.26, N 8.10, O 4.63. Found: C 48.52, H 5.26, N 8.12.
14
18
2
◦
N-(meta-chlorobenzyl)-3-hydroxyiminoquinuclidinium bromide (
6
). Yield: 89%; mp: 239.2–241.0 C;
∼
−1
1
IR (KBr)
ν
/cm : 3445–2825, 1675, 1574, 1467, 1378, 934, 796, 715. H NMR (400 MHz, DMSO-d6)
δ
/ppm: 1.88–2.15 (m, 4H, H5, H8), 2.76–2.81 (m, 1H H4), 3.49–3.66 (m, 4H, H6, H7), 4.34 (m, 2H, H2),
.68 (s, 2H, Ar–CH ), 7.52–7.71 (m, 4H, H11, H13, H14, H15), 11.15 (s, 1H C=N–OH). C NMR (101
2
13
4
MHz, DMSO-d6)
δ/ppm: 23.50 (C5,C8), 27.39 (C4), 54.74 (C6, C7), 57.44 (C2), 65.69 (C9), 130.11 (C10),
1
30.81, 131.40, 132.33, 133.13 (C11-C15), 134.01 (C12), 151.32 (C3). CHN analysis/%: Anal. calcd. for
C H BrClN O: C 48.65; H 5.25; Br 23.12; Cl 10.26; N 8.10; O 4.63. Found: C 48.53, H 5.26, N 8.08.
14
18
2
◦
N-(para-methylbenzyl)-3-hydroxyiminoquinuclidinium bromide (
7
). Yield: 94%; mp: 233.4–233.9 C;
∼
−1
1
IR (KBr)
ν
/cm : 2992–3228, 1674, 1611, 1513, 1460, 1376, 1064, 1029, 956, 829, 689. H NMR (400 MHz,
DMSO-d6)
δ/ppm: 1.87–2.15 (m, 4H, H5, H8), 2.36 (s, 3H, CH ), 2.75–2.80 (m, 1H H4), 3.46–3.62 (m, 4H,
3
H6, H7), 4.30 (m, 2H, H2), 4.62 (s, 2H, Ar–CH ), 7.31–7.39 (m, 2H, H11, H15), 7.40–7.46 (m, 2H, H12,
2
H14), 11.13 (s, 1H C=N–OH), 13C NMR (101 MHz, DMSO-d6)
δ/ppm: 21.34 (CH ), 23.48 (C5, C8), 27.50
3
(C4), 54.41 (C6, C7), 57.22 (C2), 66.49 (C9), 124.79 (C10), 130.09 (C11, C15), 133.41 (C12,C14), 140.50
(
C13), 151.44 (C3). CHN analysis/%: Anal. calcd. for C H BrN O: C 55.39; H 6.51, Br 24.57, N 8.61, O
15
21
2
4
.92. Found: C 55.49, H 6.52, N 8.60.
◦
N-(meta-methylbenzyl)-3-hydroxyiminoquinuclidinium bromide (
8
). Yield: 91%; mp: 218.6–219.4 C;
∼
−1
1
IR (KBr)
ν
/cm : 3511, 3433–2758, 1671, 1659, 1466, 1418, 1381, 940, 887, 797, 711. H NMR (400 MHz,
DMSO-d6) δ/ppm: 1.88–2.15 (m, 4H, H5, H8), 2.76–2.81 (m, 1H H4), 3.50–3.59 (m, 4H, H6, H7), 4.32 (m,
13
2
H, H2), 4.61 (s, 2H, Ar–CH ), 7.32–7.45 (m, 4H, H11, H13, H14, H15), 11.14 (s, 1H C=N–OH), C NMR
2
(
(
101 MHz, DMSO-d6)
δ
/ppm: 21.40 (CH ), 23.49 (C5,C8), 27.46 (C4), 54.53 (C6, C7), 57.43 (C2), 66.74
3
C9), 127.70 (C10), 129.42, 130.64, 131.40, 133.95 (C11-C15), 138.85 (C12), 151.44 (C3). CHN analysis/%:
Anal. calcd. for C H BrN O: C 55.39, H 6.51, Br 24.57, N 8.61, O 4.92. Found: C 55.43, H 6.50, N 8.59.
1
5
21
2
◦
N-(para-bromobenzyl)-3-hydroxyiminoquinuclidinium bromide (
9
). Yield: 95%; mp: 234.9–235.7 C; IR
∼
−
1
1
(
KBr)
ν
/cm : 3346–2840, 1656, 1630, 1588, 1460, 1376, 1072, 945, 842. H NMR (400 MHz, DMSO-d )
6
δ
/ppm: 1.89–2.14 (m, 4H, H5, H8), 2.76–2.81 (m, 1H H4), 3.49–3.65 (m, 4H, H6, H7), 4.31 (m, 2H, H2),
.66 (s, 2H, Ar–CH ), 7.48–7.53 (m, 2H, H11, H15), 7.73–8.79 (m, 2H, H12, H14), 11.14 (s, 1H C=N–OH),
4
2
13
C NMR (101 MHz, DMSO-d6)
δ/ppm: 23.49 (C5, C8), 27.43 (C4), 54.55 (C6, C7), 57.29 (C2), 65.74 (C9),
124.67 (C10), 127.13 (C13), 132.54 (C11,C15), 135.60 (C12,C14), 151.34 (C3). CHN analysis/%: Anal.
calcd. for C H Br N O: C 43.10, H 4.65, Br 40.96, N 7.18, O 4.10. Found: C 43.99, H 4.64, N 7.20.
14
18
2
2
◦
N-(meta-bromobenzyl)-3-hydroxyiminoquinuclidinium bromide (10). Yield: 89%; mp: 221.7–222.4 C;
∼
−1
1
IR (KBr)
ν
/cm : 3481, 3360–2801, 1673, 1569, 1431, 1377, 1215, 1064, 943, 719, 711. H NMR (400 MHz,
/ppm: 1.89–2.15 (m, 4H, H5, H8), 2.78–2.81 (m, 1H H4), 3.50–3.63 (m, 4H, H6, H7), 4.34
s, 2H, H2), 4.67 (s, 2H, Ar–CH ), 7.48–7.54 (m, 1H, H14), 7.56–7.59 (m, 1H, H15), 7.75–7.78 (m, 1H,
DMSO-d6)
δ
(
2

Molecules 2019, 24, 2675
12 of 17
H13), 7.80–7.82 (m, 1H, H11), 11.15 (s, 1H C=N–OH), 13C NMR (101 MHz, DMSO-d6)
δ
/ppm: 23.51
(C5,C8), 27.40 (C4), 54.72 (C6, C7), 57.44 (C2), 65.64 (C9), 122.57 (C10), 130.38 (C12), 131.64, 132.69,
133.68, 135.95, (C11-C15), 151.32 (C3). CHN analysis/%: Anal. calcd. for C H Br N O: C 43.10, H
14 18 2 2
4
.65, Br 40.96, N 7.18, O 4.10. Found: C 43.21, H 4.65, N 7.16.
3
.2. Antimicrobial Evaluation
The tested microorganisms were obtained from the culture collection at the American
Type Culture Collection (ATCC) (Rockville, MD, USA) and at the Microbiology laboratory,
Department of Biology, Faculty of Natural Science, University of Split, Croatia (FNSST). The assayed
collection included gram-positive bacteria Bacillus cereus (ATCC 11778), Enterococcus faecalis (ATCC
29212), and Staphylococcus aureus (ATCC 25923) and three gram-negative ampicillin-resistant
environmental bacterial strains Escherichia coli (FNSST 111), Klebsiella pneumoniae (FNSST 014),
and Pseudomonas aeruginosa (FNSST 982). All gram-negative strains were resistant to first- and
second-generation cephalosporins, penicillins, quinolones, tetracyclines, gentamicin, and trimethoprim
◦
as determined previously [24]. Bacterial strains were cultured overnight at 37 C in tryptic soy broth
(
6
TSB) to achieve optical densities corresponding to 10 colony forming units (cfu/mL) for bacteria.
All strains were grown in lysogenic broth (LB) media while being shaken at 150 rpm for 12 h at 37
◦
C. Part of the bacterial suspension (100
µ
L) was transferred to a vial containing 5 mL of LB media
◦
and incubated for 2 h at 37 C to get mid logarithmic phase growth of bacteria. Finally, the bacterial
suspension was diluted using a phosphate buffer saline (PBS, pH 7.2) to achieve 10 bacteria/mL.
5
3
.2.1. Disc Diffusion Assay
In order to investigate the antimicrobial activities of the synthesized N-alkyl derivatives of
quinucline-3-oximes, a disc diffusion assay was employed according to the Clinical and Laboratory
5
Standards Institute (CLSI) guidelines [25]. Briefly, 100
µ
L of suspension containing 10 colony-forming
units (cfu/mL) of bacterial cells was spread on a Mueller Hinton agar (Becton Dickinson, Sparks, MD).
The stock solutions of quaternary salts were prepared by dissolving in sterile distilled water to a final
concentration of 10 mg/mL. The sterile filter discs (6 mm) were individually loaded with 25
µL of the
stock solution, equivalent to a final concentration of 250 g/disc of synthesized compounds, and then
µ
placed on the nutrient agar that had been previously inoculated with the target microbial strains.
◦
The plates were incubated for 24 h at 37 C for the bacterial strain. Antibacterial activity was assessed
by measuring the diameter of the inhibition zone in millimetres, including disc diameter for the test
isolates. Samples were assayed in triplicate for each condition, and the diameters of inhibition zones
were presented as mean values.
3
.2.2. Minimum Inhibitory Concentration Assay
In addition, antimicrobial activities of the synthesized compounds were tested by a broth
microdilution assay in 96-well microtitre plates. The standard two-fold serial microdilution assay
described by the Clinical and Laboratory Standards Institute [24] was performed for the assessment of
the minimum inhibitory concentrations (MICs). Bacteria were grown overnight in Mueller–Hinton
◦
broth (MHB) at 37 C to the stationary phase. The microbial cultures were diluted in fresh MHB to
5
a final concentration of 10 cfu/mL for bacteria. The tested compounds dissolved in sterile distilled
water to the highest concentration of 10 mg/mL. The stock solution and reference drugs were then
serially two-fold diluted to obtain concentrations ranging from 0.06–1024
µg/mL in sterile plates
containing Mueller Hinton broth (MHB) for bacterial. Serial dilutions of the compounds were added
to the microtiter plates in a volume of 100
of inoculums of the target microorganism and incubated at 37 C for 18–24 h. The MIC value was
determined as the lowest concentration of the sample at which the tested microorganisms did not
µ
L. Each well was additionally inoculated with 100
µ
L
◦
demonstrate any visible growth after incubation. As an indicator of bacterial growth, 50
µL of
0.2 mg/mL p-iodonitrotetrazolium chloride (INT; Sigma-Aldrich Co. Ltd., Poole, UK) was added to

Molecules 2019, 24, 2675
13 of 17
◦
the wells and incubated at 37 C for 30 min. Following addition of INT and incubation, the MIC was
determined as the lowest sample concentration at which no pink color appeared. Cefotaxime and
gentamicin were used as positive controls.
3
.3. MTT Assay
In order to investigate the effect of the compounds on normal human cells, two normal human
cell lines were used, HMEC-1, human dermal microvascular endothelium cell line, and HaCaT,
human keratinocyte cell line. Both cell lines were cultured in Dulbecco’s modified Eagle medium
◦
(
DMEM) supplemented with 10% fetal calf serum (FCS). Cells were cultivated on 37 C in 5% CO
2
humidified atmosphere.
For the cell viability screening assay, all compounds were used at concentrations 1
µM, 10
µM,
and 100 µM. Then, the viability assay was repeated with the compounds that showed antimicrobial
activity with a wider concentration range.
4
Cells were seeded at 10 cells/well left for 24 h to attach. The next day cells were treated, and all
compounds were used at concentrations of 1
µM, 10 µM, and 100 µM. For the compounds that showed
antimicrobial activity, a wider range of drug concentrations was used to calculate IC .
50
In order to perform proper controls, the compounds were dived into two groups, water and DMSO
soluble. DMSO controls were used at the equivalent concentration present in compound dilutions.
Additional controls of the assay were performed with the same experimental scheme with H O and
2
2
kanamycin as cytotoxic and known antibiotics with serial double dilutions starting from 400
600
µ
M and
1
µ
g/mL, respectively (Figure 4). After 24 h, cell viability was determined by an MTT-based assay,
EZ4U, following the manufacturer’s recommendations (Biomedica, Vienna, Austria). The assay is
based on the reduction of the colorless dye in mitochondria of the living cell to the yellow, water-soluble
derivative. Briefly, cells were incubated with the MTT dye for an hour, and the absorbance was
measured on a plate reader at 450 nm with reference wavelength at 620 nm (Easy-Reader 400 FW;
SLT-Lab Instruments, GmbH, Grödig, Austria).
ROS
4
Cells were seeded at 10 cells/well left for 24 h to attach. The next day, cells were incubated with
2
,7-dichlorodihydrofluorescein diacetate (DCFH-DA, Fluka, Honeywell Research Chemicals, Charlotte,
NC, USA) for 60 minutes. This nonfluorescent dye is deacetylized inside the cells and, after oxidation
by intracellular ROS, turns fluorescent. After removing the excess dye, cells were washed once and
were treated with the compounds in a range of concentrations. Additional controls of the assay were
performed with the same experimental scheme with H O and kanamycin as representative of ROS
2
2
and known antibiotics with serial double dilutions starting from 400
(
(
µ
M and 1600
Figure 5). Fluorescence intensity was measured after 1 h on spectrofluorimeter Varian Cary Eclipse
λex = 500 nm, λem = 529 nm).
µg/mL, respectively
3
.4. Antioxidant Measurements
To assess the effects of the compounds on the cellular antioxidative system, cells were seeded
6
at 0.5 × 10 cell/well and treated with compounds qox
, 6, 9, and 10 at concentrations of 5 µM and
2
00 µM. As controls, cells were grown only in DMEM with 10% FCS and with the addition of
DMSO in equivalent concentrations to ones present in compound dilutions. Cells were harvested by
◦
trypsinization 24 h after treatment, and dry pellets were stored at
−
80 C until analysis. For analysis,
cells were lysed in PBS by four freeze/thaw cycles, and the debris was removed by centrifugation
at 14,000 g for 15 min. The catalase activity was measured by a modified method of Goth [26 27].
For catalase activity assay, 40 L of supernatant was mixed with 65-mM H O for the start of the
reaction. The reaction was stopped after 5 minutes by the addition of 100
molybdate and colour development was measured spectrophotometrically in a plate reader at 405 nm
Shimadzu UV-1601, Kyoto, Japan). One unit of catalase activity is defined as the amount of enzyme
,
µ
2
2
µL of 200-mM ammonium
(

Molecules 2019, 24, 2675
14 of 17
◦
needed for the degradation of 1 µmol of H O /min at 25 C. Catalase activity was expressed as units
2 2
−1
per milligram of proteins in cell lysate (U mg ).
The intracellular GSH content was measured by modification of the protocol described by
Tietze [28]. Briefly, 150
which started by the addition of freshly prepared reaction mix: 1.8-mM 5,5-dithio-bis-2-nitrobenzoic
acid, 0.4 Units of GSH reductase, and 0.6-mM NADPH in phosphate buffer (100-mM NaH PO , 5-mM
µL of samples diluted to 0.03 mg/mL protein was used for the the reaction,
2
4
EDTA pH 7.4). The formation of 2-nitro-5-thiobenzoic acid was monitored spectrophotometrically in a
plate reader at 405 nm (Shimadzu UV-1601, Japan). GSH concentration in cell lysates was expressed as
−
1
µM of GSH per milligram of total protein (nmol mg ).
3
.5. Statistical Analysis
All experiments were performed in biological and technical triplicates. The inhibitory
concentrations of 50% (IC ) were calculated using nonlinear regression curve fitting log(inhibitor)
50
vs. response and variable slope with a least squares (ordinary) fit, using GraphPadPrism Version 5
software (grAphPad Corporation, San Diego, CA, USA). Statistical analyses were performed using
two-way analysis of variance (ANOVA) with Bonferroni post hoc test and Student’s t-test using the
same software. Values of P < 0.05 were considered significant.
3
.6. Computational Methods
Conformational analysis for all compounds was performed by using the tensor decomposition of
ab initio molecular dynamics trajectories using the previously published procedure [23]. Ab initio
molecular dynamics were performed using on-the-fly calculations of forces in each point of the
simulation and integration with the velocity Verlet algorithm. The forces were calculated using the
PM7 method [29] implemented in MOPAC2016 (Stewart Computational Chemistry, Colorado Springs,
CO, USA) [30]. All molecular dynamics simulations were performed using our own program qcc (T.
Hrenar, Zagreb, Croatia) [31,32]. To ensure that most of the phase space relevant to the conformational
space of the investigated compounds was adequately sampled, the initial temperature for the Maxwell
distribution of velocities was set at 673.15 K, and this temperature was kept constant throughout
simulations using a velocity scaling algorithm. The step size was 0.5 fs, and the total length of
simulation was 2.5 ns (a total of 5,000,000 steps).
Tensor Decomposition: Molecular dynamics trajectories were decomposed using the principal
component analysis implemented in program moonee (T. Hrenar, Zagreb, Croatia) [33–35].
Local maxima from corresponding probability distribution were determined and used as an initial
guess for geometry optimization of structures at the B3LYP-D3/6-311++G(d,p) level of the theory.
After geometry optimizations, clustering of obtained geometries produced the full conformational
space of each investigated compound. All quantum-chemical calculations were performed using the
Gaussian 16 program (Gaussian, Inc., Wallingford, CT, USA) [36]. Figures were plotted using graphing
utility within free and open-source program Gnuplot 5.0. [37].
Potential energy surfaces: Potential energy surfaces for conformers of the lowest energy were
calculated by systematic variation of torsional angles for the chemical groups attached to the nitrogen
atom. For each point, an energy calculation for the structure was performed using the Gaussian
16 program package [36]. Energy values were arranged in data matrix or 2nd-order data tensors
and combined for all investigated compounds, making a 3rd-order data tensor. This tensor was
decomposed by using Tucker3 tensor decomposition method [38].
4
. Conclusions
The antimicrobial profiles of presented novel N-alkyl derivatives of quinuclinium oximes
were determined. Compounds and 10 (MIC values 0.25 g/mL) demonstrated a very high potential
1–10
5
µ
in combating multidrug-resistant gram-negative bacterial strains, especially the emerging pathogen
Pseudomonas aeruginosa. Both compounds showed broad spectrum antimicrobial activity against

Molecules 2019, 24, 2675
15 of 17
gram-positive and gram-negative bacteria. Compounds have no inhibitory effects on normal human
cells growth and are suitable for further development. Continued studies on these molecules, including
combinatorial therapies with conventional antibiotics and animal modeling, will enable the elucidation
of the mechanisms by which these derivatives provide a potent antimicrobial effect. Regression model
between the biological activity and potential energy surfaces was established using the tensor reduction
2
techniques. The high R value of the established model confirms the capabilities of the model to
describe the biological activity of investigated compounds in relation to their potential energy surfaces.
ˇ
Author Contributions: Conceptualization, I.P. and T.H.; methodology, I.P., A.C.G., and M.S.; software, T.H.;
validation, A.R.K., M.S., I.P., and K.S.; formal analysis, T.H.; writing—original draft preparation, M.S., I.P., R.O.,
ˇ
A.C.G., and T.H.; writing—review and editing, I.P.; visualization, T.H.; funding acquisition, T.H.
Funding: This research was funded by the Croatian Science Foundation, grant number IP-2016-06-3775 ”ADESIRE“
ˇ
Acknowledgments: I.P. and A.C.G. would like to acknowledge networking contributions by the COST Action
CM1407 “Challenging organic syntheses inspired by nature—from natural products chemistry to drug discovery.”
Conflicts of Interest: The authors declare no conflict of interest.
References
1
.
Fauci, A.S. Infectious Diseases: Considerations for the 21st Century. Clin. Infect. Dis. 2001, 32, 675–685.
CrossRef] [PubMed]
[
2
.
Boucher, H.W.; Talbot, G.H.; Bradley, J.S.; Edwards, J.E.; Gilbert, D.; Rice, L.B.; Scheld, M.; Spellberg, B.;
Bartlett, J. Bad bugs, no drugs: No ESKAPE! An update from the Infectious Diseases Society of America.
Clin. Infect. Dis. 2009, 48, 1–12. [CrossRef] [PubMed]
3
.
.
Pendleton, J.N.; Gorman, S.P.; Gilmore, B.F. Clinical relevance of the ESKAPE pathogens. Expert. Rev. Anti.
Infect. Ther. 2013, 11, 297–308. [CrossRef] [PubMed]
World Health Organization. Antimicrobial Resistance: Global Report on Surveillance; World Health Organization:
Geneva, Switzerland, 2014.
4
5
6
.
.
Gomtsyan, A. Heterocycles in drugs and drug discovery. Chem. Heterocycl. Compd. 2012, 48, 7–10. [CrossRef]
Mittal, A. Synthetic Nitroimidazoles: Biological Activities and Mutagenicity Relationships. Sci. Pharm. 2009
7, 497–520. [CrossRef]
,
7
7
.
Nagalakshmi, G. Synthesis, antimicrobial and antiinflammatory activity of 2,5-disubstituted-1,3,4-oxadiazoles.
Indian J. Pharm. Sci. 2008, 70, 49–55. [CrossRef] [PubMed]
8
9
.
.
Joule, J.A.; Mills, K. Heterocyclic Chemistry, 4th ed.; Blackwell Publishing: Hoboken, NJ, USA, 2000; p. 369.
Nekrasov, D.D. Biological Activity of 5-and 6-Membered Azaheterocycles and Their Synthesis from
5
-Aryl-2,3-Dihydrofuran-2,3-diones. Chem. Heterocycl. Compd. 2001, 37, 263–275. [CrossRef]
1
1
1
0. Sperry, J.B.; Wright, D.L. Furans, thiophenes and related heterocycles in drug discovery. Curr. Opin. Drug
Discovery Dev. 2005, 8, 723–740.
1. Polshettiwar, V.; Varma, R.S. Greener and expeditious synthesis of bioactive heterocycles using microwave
irradiation. Pure Appl. Chem. 2008, 80, 777–790. [CrossRef]
2. Katritzky, A.R. Heterocyclic chemistry: An academic subject of immense industrial importance.
Chem. Heterocycl. Compd. 1992, 28, 241–259. [CrossRef]
1
1
3. Song, C.E. An Overview of Cinchona Alkaloids in Chemistry; Wiley: Hoboken, NJ, USA, 2009.
4. Gralla, R.; Lichinitser, M.; Van der Vegt, S.; Sleeboom, H.; Mezger, J.; Peschel, C.; Tonini, G.; Labianca, R.;
Macciocchi, A.; Aapro, M. Palonosetron improves prevention of chemotherapy-induced nausea and vomiting
following moderately emetogenic chemotherapy: Results of a double-blind randomized phase III trial
comparing single doses of palonosetron with ondansetron. Ann. Oncol. 2003, 14, 1570–1577. [CrossRef]
[PubMed]
1
5. Bazina, L.; Maravi c´ , A.; Krce, L.; Soldo, B.; Odžak, R.; Bu cˇ evi c´ Popovi c´ , V.; Aviani, I.; Primoži cˇ , I.;
Šprung, M. Discovery of novel quaternary ammonium compounds based on quinuclidin-3-ol as new
potential antimicrobial candidates. Eur. J. Med. Chem. 2019, 163, 626–635. [CrossRef] [PubMed]
1
6. Odžak, R.; Šprung, M.; Soldo, B.; Sko cˇ ibuši c´ , M.; Gudelj, M.; Mui c´ , A.; Primoži cˇ , I. Quaternary salts derived
from 3-substituted quinuclidine as potential antioxidative and antimicrobial agents. Open Chem. 2017, 15,
3
20–331. [CrossRef]

Molecules 2019, 24, 2675
16 of 17
1
7. Sko cˇ ibuši c´ , M.; Odžak, R.; Štefani c´ , Z.; Križi c´ , I.; Krišto, L.; Jovi c´ , O.; Hrenar, T.; Primoži cˇ , I.; Jurašin, D.
Structure-Property Relationship of Quinuclidinium Surfactants—Towards Multifunctional Biologically
Active Molecules. Colloids Surf. B Biointerfaces 2016, 140, 548–559. [CrossRef]
1
8. Gerba, C.P. Quaternary Ammonium Biocides: Efficacy in Application. Appl. Environ. Microbiol. 2015, 81,
4
64–469. [CrossRef]
1
9. Grob, C.A.; Renk, E. 3-Chinuclidincarbonsaure. Helv. Chim. Acta 1954, 37, 1689–1698. [CrossRef]
2
0. Sternbach, L.H.; Kaiser, S. Antispasmodics. I. Bicyclic Basic Alcohols. J. Am. Chem. Soc. 1952, 74, 2215–2218.
[CrossRef]
2
2
2
2
2
1. World Health Organization. Critically Important Antimicrobials for Human Medicine: Ranking of Antimicrobial
Agents for Risk Management of Antimicrobial Resistance Due to Non-Human Use; World Health Organization:
Geneva, Switzerland, 2017.
2. Tam, V.H.; Kabbara, S.; Vo, G.; Schilling, A.N.; Coyle, E.A. Comparative pharmacodynamics of gentamicin
against Staphylococcus aureus and Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 2006, 50,
2
626–2631. [CrossRef]
3. Hrenar, T.; Primoži cˇ , I.; Fijan, D.; Majeri c´ Elenkov, M. Conformational analysis of spiro-epoxides by principal
component analysis of molecular dynamics trajectories. Phys. Chem. Chem. Phys. 2017, 19, 31706–31713.
[
CrossRef]
4. Maravi c´ , A.; Sko cˇ ibuši c´ , M.; Cvjetan, S.; Šamani c´ , I.; Fredotovi c´ , Ž.; Puizina, J. Prevalence and diversity of
extended-spectrum- -lactamase-producing Enterobacteriaceae from marine beach waters. Mar. Pollut. Bull.
015, 90, 60–67. [CrossRef]
β
2
5. Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing;
eighteenth informational supplement, CLSI document M100-18; Clinical and Laboratory Standards Institute:
Wayne, PA, USA, 2008.
2
2
2
6.
G
ó
th, L.A. Simple method for determination of serum catalase activity and revision of reference range.
Clin. Chim. Acta 1991, 196, 143–151. [CrossRef]
7. Campbell, M.K.; Farrell, S.O. Biochemistry; Brooks/Cole: Belmont, CA, USA; Cengage Learning: Boston,
MA, USA, 2012.
8. Tietze, F. Enzymic method for quantitative determination of nanogram amounts of total and oxidized
glutathione: Applications to mammalian blood and other tissues. Anal. Biochem. 1969, 27, 502–522.
[CrossRef]
2
9. Stewart, J.J.P. Optimization of parameters for semiempirical methods VI: More modifications to the NDDO
approximations and re-optimization of parameters. J. Mol. Model. 2013, 19, 1–32. [CrossRef] [PubMed]
0. Stewart, J.J.P. Stewart Computational Chemistry; MOPAC2016: Colorado Springs, CO, USA, 2016.
1. Hrenar, T. QCC, Quantum Chemistry Code; Revision 0.6826; Zagreb, Croatia, 2019; Available online:
https://masongroup.lab.uiowa.edu/qcc-everyone (accessed on 1 June 2019).
3
3
3
3
3
3
2. Primoži cˇ , I.; Hrenar, T.; Baumann, K.; Krišto, L.; Križi c´ , I.; Tomi c´ , S. Mechanochemical and Conformational
Study of N-heterocyclic Carbonyl-Oxime Transformations. Croat. Chem. Acta 2014, 87, 155–162. [CrossRef]
3. Hrenar, T. moonee, Code for Manipulation and Analysis of Multi- and Univariate Data; Revision 0.6826;
University of Zagreb: Zagreb, Croatia, 2019.
4. Jovi c´ , O.; Smoli c´ , T.; Juriši c´ , Z.; Mei c´ , Z.; Hrenar, T. Chemometric Analysis of Croatian Extra Virgin Olive Oils
from Central Dalmatia Region. Croat. Chem. Acta 2013, 86, 335–344. [CrossRef]
5. Novak, P.; Kiši c´ , A.; Hrenar, T.; Jedna cˇ ak, T.; Miljani c´ , S.; Verbanec, G. In-line reaction monitoring of
entacapone synthesis by Raman spectroscopy and multivariate analysis. J. Pharm. Biomed. Anal. 2011, 54,
6
60–666. [CrossRef] [PubMed]
3
6. Frisch, M.J.; Trucks, G.W.; Schlegel, H.B.; Scuseria, G.E.; Robb, M.A.; Cheeseman, J.R.; Scalmani, G.; Barone, V.;
Petersson, G.A.; Nakatsuji, H.; et al. Gaussian 16, Revision A.03; Gaussian, Inc.: Wallingford, CT, USA, 2016.

Molecules 2019, 24, 2675
17 of 17
3
3
7. Gnuplot 5.0. Available online: http://gnuplot.info/ (accessed on 1 June 2019).
8. Tucker, L. Some mathematical notes on three-mode factor analysis. Psychometrika 1966, 31, 279–311. [CrossRef]
Sample Availability: Samples of the compounds 1–10 are available from the authors.
©
2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).