Antibacterial activities of mono-, di- and tri-substituted triphenylamine-based phosphonium ionic liquids

Article abstract of DOI:10.1016/j.bmcl.2018.01.057

We report the synthesis of new mono, di and tri phosphonium ionic liquids and the evaluation of their antibacterial activities on both Gram-positive and Gram-negative bacteria from the ESKAPE-group. Among the molecules synthesized some of them reveal a st

Full text of DOI:10.1016/j.bmcl.2018.01.057

Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Antibacterial activities of mono-, di- and tri-substituted triphenylamine-
based phosphonium ionic liquids
Frédéric Brunel a, Christelle Lautard b, Carole di Giorgio c, Frédéric Garzino a, Jean-Manuel Raimundo a,
,
⇑
Jean-Michel Bolla b, , Michel Camplo a,
⇑
⇑
a Centre Interdisciplinaire de Nanoscience de Marseille (CINaM), Aix-Marseille Université, CNRS UMR 7325, 163 Ave de Luminy case 913, 13288 Marseille Cedex 09, France
b Transporteurs Membranaires, Chimiorésistance et Drug Design, Aix-Marseille Université, IRBA, UMR-MD1, Facultés de Médecine et de Pharmacie, 13385 Marseille cedex 05, France
c Aix-Marseille Université, Laboratoire de mutagenèse environnementale, IMBE, IFR ECCOREV, Faculté de Pharmacie, Marseille cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history:
We report the synthesis of new mono, di and tri phosphonium ionic liquids and the evaluation of their
antibacterial activities on both Gram-positive and Gram-negative bacteria from the ESKAPE-group.
Among the molecules synthesized some of them reveal a strong bactericidal activity (MIC = 0.5 mg/L)
for Gram-positive bacteria (including resistant strains) comparable to that of standard antibiotics. A com-
parative Gram positive and Gram negative antibacterial activities shows that the nature of counter-ion
has no significant effects. Interestingly, the increase of phosphonium lateral chains (from 4 to 8 carbons)
results in a decrease of antibacterial activities. However, the increase of the spacer length has a positive
influence on the activity on both Gram-positive and Gram-negative bacteria except for E. aerogenes.
Finally, the increased charge density has no effect on the Gram-positive antibacterial activities (MIC
between 2 and 4 mg/L) but seems to attenuate (except for P. aeruginosa) the discrimination between
Gram-positive and Gram-negative. Overall these results suggest a unique mechanism of action of these
triphenylamine-phosphonium ionic liquid derivatives.
Received 29 November 2017
Revised 18 December 2017
Accepted 26 January 2018
Available online xxxx
Keywords:
Triphenylamine
Phosphonium ionic liquids
Antimicrobial
Microwave
Ó 2018 Elsevier Ltd. All rights reserved.
Nosocomial infections represent a major threat for public health
and litigation arising from these infections is increasing world-
wide. In EU, the number of deaths occurring as a direct conse-
quence of these infections is estimated to be at least 37 000 and
these infections are thought to contribute to an additional 110
000 deaths each year.1 Staphylococcus aureus, Escherichia coli and
Pseudomonas aeruginosa are commonly considered as responsible
of 80% of nosocomial infections. S. aureus, in addition to be respon-
sible for many hospital’s acquired infections,2 belongs to the
ESKAPE group that includes major human pathogens with multi-
drug resistance.3 There is still an urgent need to discover new
molecules to fight against these particularly adaptive pathogens
which are continuously developing mechanisms of resistance.4 In
a previous publication we have described the use of fluorescent
triphenylamine phosphonium (TPA-P) derivatives to tackle effi-
ciently S. aureus in particular.5 These molecules are belonging to
the class of ionic liquids for which there is a growing interest for
their use in biology.6 The aim of this continuing work is firstly to
synthesize new mono, di and tri TPA-P and secondly to evaluate
their antibacterial activities in order to identify the structural
parameters (degree of substitution, charge, spacer and phospho-
nium lateral chain length) essential for their antibacterial activity.
In order to investigate structure/activity relationship, we
designed different target molecules in which we varied the length
of phosphonium alkyl chains, those of the spacer between the
phosphonium and the TPA as well as the degree of substitution
(mono, di and tri phosphonium derivatives). Our strategy involves
a 6 steps sequence allowing a wide synthetic versatility (Scheme 1).
Firstly, mono-, di- and tri-formylated TPA can be synthesized fol-
lowing a Vilsmeier-Hack formylation from commercial TPA with
only slight modifications of experimental procedure depending
on the desired formylation degree. According this procedure
mono-1, di-2 and triformylated TPA 3 were respectively obtained
in 77%, 51% and 72% yield. Aldehyde functions were then success-
fully reduced in alcohol 4, 5 and 6 with NaBH4 in quantitative yield
in each case. Next, Williamson etherifications were carried out by
preliminary conversion of alcohol in corresponding sodium alkox-
ide using NaH in ethanol or ethanol/chloroform. The sodium salts
produced in situ reacted with previously prepared silylated com-
pounds Br-(CH2)n-OTBDMS (n = 6 or 10) to afford 7, 8, 9 and 10
respectively in 86%, 74%, 65% and 35% yield. Protecting group
was subsequently removed with TBAF and the corresponding alco-
hols were obtained in quantitative yield for 11, 13 and 14 and in
⇑
Corresponding authors.
E-mail address: michel.camplo@univ-amu.fr (M. Camplo).
https://doi.org/10.1016/j.bmcl.2018.01.057
0960-894X/Ó 2018 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Brunel F., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.01.057

2
F. Brunel et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
71% yield for long alkyl chain 12. The following halogenation of
resulting alcohols was performed using Appel reaction: chlorina-
tion could be successfully done using CCl4/PPh3 while bromination
required the NBS/PPh3 reagent combination affording 18 in 61%
yield and 19 in 11% yield. Lastly, in order to optimize yield and pur-
ity of final ionic liquids 20–27, we focused on a fast, clean and safe
solvent free microwave activation (200 W). Under these condi-
tions, chlorinated compounds 15, 16, 17 and brominated 18, 19
reacted with several trialkylphosphines P(alkyl)3 (alkyl = Bu, Hex
or Oct) affording, in most cases, ionic liquids in good yields.
The synthesis is depicted in Schemes 1 and 2.
The Minimal Inhibition Concentration (MIC) of representative
compounds was determined against a series of strains representa-
tive of the ESKAPE pathogens3 and are summarized in Table 1 (see
SI for experimental details). As already shown in our previous
paper,5 TPA and tetraalkylphosphonium did not show antimicro-
bial activity in a range of concentration up to 64 mg/L suggesting
that the combination of both is determinant for the antibacterial
activity.
The remaining tested compounds were found active on either
one or both Gram-positive and Gram-negative bacteria strains.
From these results, we can confirm that the counter-ion has no
effect on the activity. Also, the spacer (6 or 10 carbons) between
the positive phosphonium charge and the TPA seems to have a pos-
itive influence on the Gram-positive and Gram-negative bacteria
activities; for example, the activity is 4–8 times stronger on S. aur-
eus strains whereas it is 4 times stronger for Gram-negative A. bau-
mannii. This result is not surprising since it is likely that a longer
Scheme 1. i) POCl3, DMF; ii) NaBH4, EtOH/CHCl3 (1:1); iii) NaH, 18 crown-6, Br-
(CH2)n-OTBDMS (n = 6 or 10), 78 °C; iv) TBAF, THF, 25 °C; v) NBS, PPh3; vi) CCl4, PPh3
reflux; vii) PR3 (R = Bu, Hex or Oct), MW (200 W).
Scheme 2. i) POCl3, DMF; ii) NaBH4, EtOH/CHCl3 (1:1); iii) NaH, 18 crown-6, Br-(CH2)6-OTBDMS, 78 °C; iv) TBAF, THF, 25 °C; v) CCl4, PPh3 reflux; vi) PR’3 (R0 = Hex),
MW (200 W).
Please cite this article in press as: Brunel F., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.01.057

F. Brunel et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
3
Table 1
Minimal Inhibition Concentration (MIC) (mg/L) of compounds TPA, 20–22, 24–27 and X-+PR4 on ESKAPE bacteria.
Bacterial strain
Compounds
20
21
22
24
25
26
27
X-+PR4 TPA
20,477
1
4
16
4
8
2
4
>64
E. faecium
CIP 7625
S. aureus
1
2
8
2
2
2
2
>64
>64
>64
>64
>64
>64
>64
>64
CIP 82.91
K. pneumoniae
ATCC 19,606
A. baumannii
CIP 100,720
P. aeruginosa
ATCC 13,048
E. aerogenes
CIP 54.8
E. coli
SA1199
S. aureus
SA1199B
16
16
32
16
16
0.5
0.5
>64
>64
16
>64
>64
2
>64
>64
>64
>64
>64
8
64
64
64
>64
>64
2
32
16
64
>64
32
1
16
16
64
>64
32
2
16
16
>64
64
32
2
2
8
2
2
2
2
S. aureus
during a 24-h incubation period. As seen in Table 2 (see SI for
experimental details) it can be seen that among these compounds
tested, all were cytotoxic.
Table 2
In vitro cytotoxicity determination on human hepatocytes HepG2 of representative
compounds 21, 26 and 27. IC50: concentration of the test material causing a 50%
decrease of cell respiration as compared to the control culture.
By comparing the di- and tri-substituted compounds it
appeared that the mono-substituted derivative 21 was by far the
more cytotoxic. This might be due to its amphiphilic properties
and its possible nanoparticulate supramolecular organization as
it was shown on a similar compound in our previous paper.5 How-
ever, it is certain that relative to their high cytotoxicity the com-
pounds synthetized in this work are more likely to be used for
permanent surface/grafting deposition purposes.
Compounds
Mean-IC50
SD
21
26
27
0.09
4,16
4,28
0.002
0,04
0,14
spacer, by its intercalation between the lipophilic bilayer mem-
branes, can facilitate the crossing of the double Gram-negative
membranes in order to free the access of the cationic site. Also,
these results are consistent with the ones found in an SAR study
of imidazolium-type ionic liquid monomers reported by Zheng
et al.7 Concerning the number of positive charge (1–3, 21, 26, 27)
beared by the TPA derivatives, one can observe that if the density
of charge did not affect the activities on Gram-positive bacteria,
it has a significant impact on Gram-negative ones. Indeed, except
of E. aerogenes and P. aeruginosa the antibacterial activity tends
to increase with the degree of charge distribution with a result of
limiting the discrimination between Gram-positive and Gram-neg-
ative antibacterial activities. It is known that Gram-positive bacte-
ria are more sensitive to antibacterial than Gram-negative ones.8
Also, the overall charge of bacteria differs between these two cat-
egories of bacteria. Globally, the procaryotic cells have a negatively
charged membranes and the molecules responsible of that are of
different nature. For Gram-positive the anionic membrane compo-
nents are mainly glycopolymers like techoïc acid9 whereas for
Gram-negative there are lipopolysaccharides associated with
phospholipids. If the chemical structure and the density of these
are species specific, anionic and zwiterrionic components are gen-
erally more present in the membranes of Gram-negative bacteria.
This might explain the increasing affinity of polycationic com-
pounds for the Gram-negative cell wall and the observed decreased
discrimination with Gram-positive bacteria.10 Finally, we tested
the activity of compounds on a strain of S. aureus overproducing
the NorA efflux pump responsible for a multi-drug resistant pheno-
type.5 As observed previously, the same efficacy was observed on
the NorA overproducer SA1199B and its isogenic parent SA1199,
suggesting that the compounds were not efflux-pump substrates.2
The in vitro cytotoxicity of the three representative compounds 21,
26 and 27 was determined by measuring the relative reduction in
viability of human hepatocytes (HepG2) exposed to the substance
In summary, a structure activity relationship of TPA-P com-
pounds with different chains, spacer length and charges allowed
us to design some interesting criteria of selection. While most
compounds were active on Gram-positive bacteria these parame-
ters are more susceptible to discriminate the activities between
the two categories of bacteria, namely Gram-positive and Gram-
negative ones. Indeed, this discrimination was clearly observed in
favor of Gram-negative bacteria with the increase of the spacer
length as well as the charge density. In our opinion and despite
their cytotoxicity on HepG2 cells, these TPA-P derivatives might
constitute a useful biological tool since we have shown that they
are not efflux-pump substrates. Also, surface/grafting deposition
are still in progress in our laboratory.
Acknowledgement
This work was supported by the Centre National de la
Recherche Scientifique (CNRS) and the Ministère de l’Enseigne-
ment Supérieur et de la Recherche. We are indebted to Jean-Marie
Pagès, director of the UMR-MD1 for his support.
Notes
ESKAPE Enterococcus faecium, Staphylococcus aureus, Klebsiella
species, Acinetobacter baumannii, Pseudomonas aeruginosa, and
Enterobacter species; IL’s ionic liquids; MIC minimal inhibitory con-
centration, TBAF tetrabutylammonium fluoride; TPA-P tripheny-
lamine phosphonium.
A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.bmcl.2018.01.057.
Please cite this article in press as: Brunel F., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.01.057

4
F. Brunel et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
7. Zheng Z, Xu Q, Guo J, et al. ACS Appli. Mater. Interfaces. 2016;8:12684.
References
8. Silhavy TJ, Kahne D, Walker S. The bacterial cell envelope. Cold Spring Harb
Perspect Biol. 2010;2:123.
1. (a) <http://ecdc.europa.eu/en/healthtopics/healthcare-associated_infections/
pages/index.aspx#sthash.oB8HDypD.dpuf;
9. Brown S, Santa Maria JP, Walker S. Wall teichoic acids of gram-positive bacteria.
Annu Rev Microbiol. 2013;67:313.
10. Haldar J, Kondaiah P, Bhattacharya S. J Med Chem. 2005;48:3823.
(b) Tong SYC, Davis JS, Eichenberger E, Holland TL, Fowler VG. Clin Microbiol Rev.
2015;28(3):603.
2. Béthencourt MI, Srisombat L-O, Chinwangso PP, Lee TR. Langmuir.
2009;25:1265.
3. Rice LB. J Infect Dis. 2008;197:1079.
Further reading
4. Zilahi G, Artigas A, Martin-Loeches I. Annals of Intensive Care. 2016;6:96. https://
doi.org/10.1186/s13613-016-0199-4.
5. Brunel F, Lautard C, Garzino F, et al. Bioorg Med Chem Lett. 2016;26:3770.
6. (a) Egorova KS, Gordeev EG, Ananikov VP. Chem Rev. 2017;117:7132;
(b) Santos de Almeida T, Júlio A, Portugal Mota J, Rijo P, Pinto Reis C. Ther
Delivery. 2017;8:461.
11. Comite de l’Antibiogramme de la Societe Francaise de Microbiologie Report.
Int J Antimicrob Agents. 2003;2003:364.
Please cite this article in press as: Brunel F., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.01.057