Self-assembled nanoparticles as multifunctional drugs for anti-microbial therapies

Article abstract of DOI:10.1039/c4cc00349g

A self-assembled nanoparticle containing a photosensitizer and a Trojan-horse moiety (cholesterol), binds an anti-TB pro-drug and increases 1000-fold its activity against mycobacteria. These minimalist constructs will allow development of economically viable, efficient drug preparations for the treatment of drug-resistant TB infections.

Full text of DOI:10.1039/c4cc00349g

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Self-assembled nanoparticles as multifunctional
drugs for anti-microbial therapies†
Manpreet Bhatti,*a Timothy D. McHugh,b Lilia Milanesic and Salvador Tomas*a
Cite this: Chem. Commun., 2014,
50, 7649
Received 15th January 2014,
Accepted 22nd May 2014
DOI: 10.1039/c4cc00349g
www.rsc.org/chemcomm
A self-assembled nanoparticle containing a photosensitizer and a release of the drug.9,10 The level of sophistication that can potentially
Trojan-horse moiety (cholesterol), binds an anti-TB pro-drug and be reached is illustrated by the ‘‘artificial organelle’’ approach.11
increases 1000-fold its activity against mycobacteria. These minimalist These approaches, although potentially very efficient, are expensive
constructs will allow development of economically viable, efficient to develop and produce. On the other hand, a nanoparticle can be
drug preparations for the treatment of drug-resistant TB infections.
produced that is formed exclusively of small, uncomplicated mole-
cules, each one with different functionality, but with an overall
The recent UK Chief Medical Officers report has emphasized the therapeutic effect that is larger than the sum of the individual parts.
increasing threat of antimicrobial resistance and the need for new Here we develop a multifunctional nanoparticle that can be poten-
strategies for control.1,2 This is particularly true for tuberculosis tially applied to the treatment of TB. The nanoparticle will contain
(TB). TB accounts for around 9 million new cases a year globally bacterial nutrient, a photosensitizer (PS) for photodynamic therapy
and 1.4 million deaths.3 This fact is compounded by over half a (PDT) and an anti-TB drug. In PDT, a PS is activated by light of
million cases of multi drug resistant (MDR) and extreme drug an appropriate wavelength producing reactive oxygen species that
resistant (XDR) TB per year. The options for treating TB are cause cell death. PDT is being used for cancer treatment,12 but its
limited, the drugs in the approved regimens were introduced in application to bacterial infections has been so far limited.13,14
the 1960’s and new moieties are still in early phase testing. The
Recently, we described the synthesis and molecular recogni-
cause for concern is that the new compounds are focused on tion properties of very stable (CMC 11 nM), hydrophobically self-
limited targets and so cross-resistance is a real risk.4 Clearly, new assembled nanoparticles based on a Zn-metalloporphyrin amphiphile
paradigms for antimicrobial delivery are required. One such that contained also cholesterol.15 We reasoned that nanoparticles
paradigm may rest in the development of soft nanoparticles as based on this design can be used for the targeted delivery of
vehicles for targeted and controlled delivery.
drugs that bind to the nanoparticle. Additionally, the chemistry of
In the last years the development of antimicrobials based on the amphiphile can be tailored in order to introduce targeting
nanoparticles has attracted strong interest.5,6 For the treatment and enhance transport and towards photodynamic therapy (PDT)
of Mycobacterium tuberculosis specifically, studies have so far applications.16 For the development of an agent for the treatment of
focused on the sustained drug delivery properties of relatively mycobacterium infections, we used the same amphiphilic building
simple nanoparticle preparations (mostly solid lipid and polymer block as earlier described, but introduced Co instead of Zn as the
based), using traditional drugs.7,8 However nanoparticle assemblies metal centre, leading to amphiphile 1 (Fig. 1 and ESI,† Fig. S1).
offer the possibility of introducing functionality beyond the passive
It is known that mycobacteria are capable of using cholesterol as
an important carbon source during infections and during growth in
culture.17 Therefore, the presence of the cholesterol in the molecule
is expected to provide some degree of targeting and facilitate the
transport inside the bacterium. Co was chosen as the porphyrin
metal centre for two reasons. First, Co metalloporphyrins are likely
to be more efficient photosensitizers than Zn metalloporphyrins as
they quench oxygen at a slower rate than Zn metalloporphyrins,16
rendering a Co-porphyrin derivative a (potentially) better PDT agent.
Second, Co-metalloporphyrins have a stronger affinity for N basic-
bearing ligands than the Zn counterparts, which will make it a more
efficient molecular receptor to our drug model and potentially
a Institute of Structural and Molecular Biology and Department of Biological
Sciences, School of Science, Birkbeck, University of London, Malet Street,
London WC1E 7HX, UK. E-mail: s.tomas@bbk.ac.uk, m.bhatti@ucl.ac.uk
b Centre for Clinical Microbiology, Royal Free Campus, University College London,
Rowland Hill Street, London NW3 2QG, UK
c School of Biological and Chemical Sciences, Queen Mary, University of London,
Mile End Road, London E1 4NS, UK
† Electronic supplementary information (ESI) available: General experimental
methods, synthesis and characterization of 1 and 2, binding constants measure-
ments, TEM experimental details and detailed microbiology experiments. See
DOI: 10.1039/c4cc00349g
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 7649--7651 | 7649

View Article Online
Communication
ChemComm
Fig. 1 Chemical structure of water soluble porphyrin 1R, amphiphilic
porphyrin 1 and pyrazinoic acid ester 2, together with their cartoon
representations.
Fig. 2 (A) Cartoon representation of the nanoparticle assembly and the
formation of the complex nanoparticle-ester 2. (B) TEM picture (negative
stain) of 1 nanoparticles in the working buffer. The scale bar represents
20 nm. (C) Variation of the extinction coefficient of a solution of nano-
particles upon addition of ester 2 (empty circles). The fitting to a binding
model for the sequential binding of two 2 molecules per molecule of 1 is
shown as a line for changes at 428 nm (upper curve) and at 415 nm (lower
curve). See ESI,† Table S1 for Kapp values and ESI,† Fig. S4B for limiting
values of extinction coefficients. (D) TEM picture (negative stain) of the
nanoparticles in presence of 2. The scale bar represents 20 nm.
enhance its transport. A version of 1, unable to form nanoparticles,
1R, was also synthesized for comparison purposes (Fig. 1 and ESI,†
Fig. S2). As a drug model we synthesized a hydrophobic pyrazinoic
acid derivative, pyrazinoic acid hexylester 2, by treatment of
pyrazinoic acid with oxalyl chloride followed by addition of
hexanol (Fig. 1 and ESI,† Fig. S3)
While pyrazinoic acid is the active form of the drug,18 pyrazinoic easily released. TEM pictures of 1 show that the ligand increases
acid ester 2 was selected because its hydrophobicity should make it the size of the nanoparticles from an average 9.5 to 12.3 nm
bind more strongly to the nanoparticle than the free acid, and diameter in presence of 2 (Fig. 2B and D), consistent with ligand
because ester moieties hydrolyze slowly but spontaneously in water, binding to the nanoparticle.
while amides typically require enzymatic assistance for the hydro-
Mycobacterium fortuitum was chosen as a model for in vitro
lysis to take place in measurable timescales.19 This may lead to bacteria killing experiments. M. fortuitum is a fast growing myco-
bacterial resistance as demonstrated by mycobacterium resistance bacterium that can be manipulated under containment level 2
to the prodrug pyrazinamide.20
conditions making it a useful surrogate for M. tuberculosis.22
Amphiphilic porphyrin 1 assembles into a nanoparticle of M. fortuitum was treated either with nanoparticle 1, ester 2 or
similar size and appearance to that of the previously described nanoparticle 1 in presence of 2, at a molar ratio 2:1 drug 2/receptor
Zn derivative15 (Fig. 2A and B). The apparent binding affinity of 1 (see ESI,† for details). For M. fortuitum treated with 2 alone, a
pyrazinoic acid and the ester 2 with 1 was determined by means concentration up to 50 mM (8.6 mg mLÀ1) is required for bacterial
of UV titration experiments (Fig. 2C and ESI,† Fig. S4). The binding cell death under the conditions used in this study. Below these
isotherm is consistent with the sequential binding of two molecules concentrations the drug does not show any effect. For these experi-
of 2 for each molecule of receptor 1. This result is in agreement ments, the addition of water soluble porphyrin 1R up to 280 mM did
with literature data on complexation of Co metalloporphyrins with not have any effect on the cell survival. For nanoparticle 1 alone, the
amines, which show that the metal centre can bind two ligands.21 bacterium survival is around 40–50% when the concentration of the
The apparent binding constants are 3.4 Â 104 MÀ1 and 4.5 Â nanoparticle 35 mM, going up to 60% when the concentration is
103 MÀ1 for the first and second binding event respectively (see further lowered to 17.5 mM (Fig. 3A and ESI,† Table S2).
Materials and methods section for details and ESI,† Table S1).
Irradiation of the samples with light at 430 nm (the maximum
In contrast, the binding of 2 to a cobalt porphyrin receptor absorption of the nanoparticle) results on a decrease in survival of
that does not form nanoparticles show binding constants of 10–15% in relation to not irradiated samples for all the concentra-
4.5 Â 103 MÀ1 for the first binding event and 50 MÀ1 for the tions tested (Fig. 3A and ESI,† Table S2). This result is attributed
second (see ESI,† Table S1). Moreover, for the nanoparticle, but to the photosensitizer action of the metalloporphyrin moiety within
not for 1R, the ligand may also bind in a non-specific manner the bacteria upon irradiation. When treated with nanoparticle 1 in
(yielding complexes with stoichiometry larger than 1 : 2) via the presence of 2, quantitative killing takes place for samples
insertion of the hydrophobic tail. The implications are that, 280 mM in 1 (with 370 mM of 2 bound to nanoparticle), while the
upon de-assembly of the particle (for example, as the cholesterol survival of bacterium increases in a typical dose-dependent fashion
moiety is metabolized into the bacteria), the drug will be more as the concentrations tested decrease, up to 50% survival for
7650 | Chem. Commun., 2014, 50, 7649--7651
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
that currently require drastic treatment, including pulmonary
surgery.23 A combined therapy based on the use of a system such
as the one described here will offer a much more convenient
alternative, both from the patient and the economic point of
view. The use of UV to enhance drug activity will not be practical
in most patients. However, in those with MDR and XDR
TB, where treatment options are limited, use of an adapted
bronchoscope could provide targeted UV delivery to the site of
disease. The modular nature of the nanoparticles and the
relative simplicity of the building blocks should allow us to
develop related systems with different properties. For example, a
version bearing ergosterol may target fungi24 and can be tested
against post-transplant fungal infections. The potential toxicity
of our porphyrin derivatives may be mitigated by developing a
nanoparticle version bearing more biocompatible (e.g. heme
derivative) moieties. The potential of these systems is currently
being analysed in our labs.
Fig. 3 (A) Percentage survival of M. fortuitum 4 days after treatment with
pure 1. Red columns represent the survival for non-irradiated samples and
the blue columns the survival for samples that where irradiated 24 hours
after exposure to the nanoparticle (75 J cmÀ2 energy dose). (B) Idem for
samples treated with 1 and 2.The ratio of concentrations [2] : [1] is 2 in all
samples. The concentration of 2 bound to the nanoparticle is shown in
parenthesis and has been calculated from the binding constants. See ESI,†
Tables S2 and S3 for numerical data. In all cases, the error bars represent
twice the standard deviation of three measurements.
Notes and references
1 S. C. Davies, Annual Report of the Chief Medical Officer, 2011, 2.
2 Infections and the rise of antimicrobial resistance, London, Depart-
ment of Health, 2013.
3 World Health Organisation World tuberculosis report, 2012. http://
apps.who.int/iris/bitstream/10665/91355/1/9789241564656_eng.pdf.
4 A. S. Ginsburg, J. H. Grosset and W. R. Bishai, Lancet Infect. Dis.,
2003, 3, 432.
5 A. J. Huh and Y. J. Kwon, J. Controlled Release, 2011, 156, 128.
6 A. C. Engler, N. Wiradharma, Z. Y. Ong, D. J. Coady, J. L. Hedrick
and Y. Y. Yang, Nano Today, 2012, 7, 201.
7 J. C. Sung, D. J. Padilla, L. Garcia-Contreras, J. L. VerBerkmoes,
D. Durbin, C. A. Peloquin, K. J. Elbert, A. J. Hickey and D. A. Edwards,
Pharm. Res., 2009, 26, 1847.
8 A. Sosnik, A. M. Carcaboso, R. J. Glisoni, M. A. Moretton and
D. A. Chiappetta, Adv. Drug Delivery Rev., 2010, 62, 547.
9 M. H. Xiong, Y. Bao, X. Z. Yang, Y. C. Wang, B. Sun and J. Wang,
J. Am. Chem. Soc., 2012, 134, 4355.
10 A. F. Radovic-Moreno, T. K. Lu, V. A. Puscasu, C. J. Yoon, R. Langer
and O. C. Farokhzad, ACS Nano, 2012, 6, 4279.
11 P. Tanner, P. Baumann, R. Enea, O. Onaca, C. Palivan and W. Meier,
Acc. Chem. Res., 2011, 44, 1039.
12 D. Fayter, M. Corbett, M. Heirs, D. Fox and A. Eastwood, Health
Technol. Assess., 2010, 14, 1.
samples with 17.5 mM in 1 (with 9.5 mM of 2 bound to nanoparticle)
(Fig. 3B and ESI,† Table S3). Finally, when samples treated with 1 in
presence of 2 are also irradiated at 430 nm quantitative killing takes
place down to concentrations of 1 35 mM (with 26 mM of 2 bound
to nanoparticle), while a survival of 40% approx. is observed
when the concentration is further lowered to 17.5 mM in 1 (with
9.5 mM of 2 bound to nanoparticle). This result is consistent with
the nanoparticle facilitating the transport of the pyrazinoic acid
ester 2, together with the expected photosensitizer action of the
porphyrin moiety.
In summary, this work shows that the combination of a small
molecule based nanoparticle (1) with the appropriate derivative
of a known drug (2) results in quantitative bacterial killing for
M. fortuitum, a widely used model for M. tuberculosis down to
15 mg mLÀ1 from 10 mg mLÀ1 for the drug 2 in absence of the
nanoparticle (Fig. 3 and ESI,† Tables S2 and S3). The enhancing
of the drug activity in the presence of the nanoparticle is
attributed to the ability of the nanoparticle to facilitate the
transport of the drug into the bacteria and is supported by the
fact that the nanoparticle on its own leads to bacterial killing
upon irradiation. The transport activity can be attributed to
13 G. B. Kharkwal, S. K. Sharma, Y. Y. Huang, D. Tianhong and
M. R. Hamblin, Lasers Surg. Med., 2011, 43, 755.
14 K. O’Riordan, D. S. Sharlin, J. Gross, S. Chang, D. Errabelli,
O. E. Akilov, S. Kosaka, G. J. Nau and T. Hasan, Antimicrob. Agents
Chemother., 2006, 50, 1828.
15 S. Tomas and L. Milanesi, J. Am. Chem. Soc., 2009, 131, 6618.
16 R. Bonnet and G. Martinez, Tetrahedron, 2001, 57, 9513.
the presence of a cholesterol moiety. Cholesterol is used by 17 A. Brzostek, J. Pawelczyk and J. Dziadek, J. Bacteriol., 2009,
191, 6584.
M. fortuitum as nutrient enabling the transport of the other
moieties associated to it, such as drug model 2 and the
18 Y. Zhang and D. Mitchison, Int. J. Tuberc. Lung. Dis., 2003, 7, 6.
19 M. F. Simoes, E. Valente, M. J. Gomez, E. Anes and L. Constanito,
covalently linked Co metalloporphyrin moiety. It is also possible
that the nanoparticle components de-assemble when in contact with
the bacterial cell wall and incorporate as individual molecules.
Eur. J. Pharm. Sci., 2009, 37, 257.
20 J. K. McClatchy, A. Y. Tsang and M. S. Cernich, Antimicrob. Agents
Chemother., 1982, 20, 556.
21 U. Michelsen and C. A. Hunter, Angew. Chem., Int. Ed., 2000, 39, 764.
Irradiation may then lead to bacterial wall damage resulting in 22 A. Gupta and S. J. Bhakta, J. Antimicrob. Chemother., 2012, 67, 1380.
23 L. Bertolaccini, A. Viti, D. P. Giovanni and A. J. Terzi, J. Thorac. Dis.,
enhanced drug absorption. Whatever the precise mechanism of
action, these kinds of nanoparticles are a strong candidate to be
2013, 5, 198.
24 R. J. Kieber, W. J. Payne and G. S. Appleton, J. Appl. Microbiol., 1955,
developed into therapeutic agents for resilient strains of tuberculosis
3, 247.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 7649--7651 | 7651