Penicillin-bound polyacrylate nanoparticles: Restoring the activity of β-lactam antibiotics against MRSA

Article abstract of DOI:10.1016/j.bmcl.2007.03.077

This report describes the preparation of antibacterially active emulsified polyacrylate nanoparticles in which a penicillin antibiotic is covalently conjugated onto the polymeric framework. These nanoparticles were prepared in water by emulsion polymeriza

Full text of DOI:10.1016/j.bmcl.2007.03.077

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3468–3472
Penicillin-bound polyacrylate nanoparticles: Restoring the
activity of b-lactam antibiotics against MRSA
Edward Turos,a,* G. Suresh Kumar Reddy,a Kerriann Greenhalgh,a Praveen Ramaraju,a
Sampath C. Abeylath,a Seyoung Jang,b Sonja Dickeyc and Daniel V. Limc
aCenter for Molecular Diversity in Drug Design, Discovery, and Delivery, Department of Chemistry, 4202 East Fowler Avenue,
CHE 207, University of South Florida, Tampa, FL 33620, USA
bNanopharma Technologies, Inc., 3803 Spectrum Boulevard, Suite 150, Tampa, FL 33612, USA
cDepartment of Biology, 4202 East Fowler Avenue, University of South Florida, Tampa, FL 33620, USA
Received 28 February 2007; revised 22 March 2007; accepted 23 March 2007
Available online 27 March 2007
Abstract—This report describes the preparation of antibacterially active emulsified polyacrylate nanoparticles in which a penicillin
antibiotic is covalently conjugated onto the polymeric framework. These nanoparticles were prepared in water by emulsion poly-
merization of an acrylated penicillin analogue pre-dissolved in a 7:3 (w:w) mixture of butyl acrylate and styrene in the presence
of sodium dodecyl sulfate (surfactant) and potassium persulfate (radical initiator). Dynamic light scattering analysis and atomic
force microscopy images show that the emulsions contain nanoparticles of approximately 40 nm in diameter. The nanoparticles have
equipotent in vitro antibacterial properties against methicillin-susceptible and methicillin-resistant forms of Staphylococcus aureus
and indefinite stability toward b-lactamase.
Ó 2007 Elsevier Ltd. All rights reserved.
The continuing rise in microbial drug resistance has led
to widespread problems in the treatment of bacterial
infections.1 Of particular concern are those illnesses
caused by methicillin-resistant Staphylococcus aureus
(MRSA), which are responsible for a majority of hospi-
tal-acquired infections, clinical complications, and
nearly 100,000 deaths each year in the United States
alone.2,3 The loss of effectiveness of commonly used
antibacterial antibiotics such as penicillin and other
b-lactam drugs further adds to the dilemma, calling
for the immediate need for improvements in drug
design, discovery, and delivery. One of the major chal-
lenges in treating antibiotic-resistant bacterial infections
is the need to develop agents that can stop the infection
at the site of initiation, which frequently occurs in re-
gions of the body where water-soluble drugs typically
have poor access. However, the application of lipophilic
agents to combat such infections likewise has limited
effectiveness due to uptake and delivery issues resulting
from low water solubility and biodistribution.4 The abil-
ity to deliver antibacterial drugs to infections in fatty tis-
sue or on the surface of implanted medical devices, for
example, where microbial biofilms often develop, ulti-
mately determines if the infection can be cleared without
surgical intervention.5 New drug delivery vehicles such
as liposomes and nanoparticles offer a promising way
to improve bioavailability, efficacy, and specificity of
pharmaceutical compounds in general. Several groups
have reported previously on the preparation and anti-
bacterial testing of various penicillin- or ampicillin-en-
trapped polycyanoacrylates formed by anionic
emulsion polymerization in water.6–15 These emulsified
suspensions consisted of drug-containing particles con-
siderably larger than 100 nm in diameter, and reportedly
provided enhancement of the drug’s performance
against some of the microbes tested. The use of antibi-
otic-laden nanoparticles such as those for drug-resistant
bacteria such as MRSA has not been investigated, nor
has there been an attempt to evaluate the effect of hav-
ing the drug covalently bound to the nanoparticle versus
non-covalently encapsulated within its matrix. In this re-
port, we describe the results of our studies on the synthe-
sis and evaluation of penicillin-bound polyacrylate
nanoparticles against MRSA as a function of the
Keywords: Penicillin-bound nanoparticles; Bacterial drug resistance;
Emulsions; Polyacrylate nanoparticles; Antibiotic-conjugated nano-
particles; Antibiotics; MRSA; Staphylococcus aureus; Drug delivery.
*
Corresponding author. E-mail: eturos@shell.cas.usf.edu
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.03.077

E. Turos et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3468–3472
3469
O
N
H
N
S
S
O
O
O
O
O
S
emulsion
polymerization
O
Drug
H3C
OC4H9
O
O
N
N
CH3
Drug
O
O
+
COOH
O-Drug
CO2H
water, 70 oC
2
1
O
O
O
O
H
N
O
H
N
nanoparticles in water
O
S
N
N
O
O
O
O
Scheme 1. Emulsion polymerization to make antibiotic-conjugated
polyacrylate nanoparticles in water.
O
O
O
COOH
O
3
4
O
O
penicillin structure, linkage type, and location (for cova-
lently bound systems), and covalently conjugated versus
drug-encapsulated nanoparticle emulsions.
OAc
O
O
H
N
S
O
O
O
O
N
Recently we published15 a procedure for preparing anti-
bacterially active polyacrylate nanoparticles, which en-
hances the water-solubility and antibacterial properties
of highly lipophilic N-thiolated b-lactam antibacteri-
als.16,17 In this method, the water-insoluble antibiotic
was first converted to an acrylated form, then dissolved
in a mixture of liquid monomers (butyl acrylate and sty-
rene) before being emulsified with a surfactant and poly-
merized in the presence of a water-soluble radical
initiator (Scheme 1). This procedure enhances the per-
formance of water-insoluble antibiotics that act on tar-
gets within the bacterial cell.
O
COOH
5
Figure 1. Acrylated penicillin monomers 1–5 used to prepare drug-
conjugated nanoparticles NP1–NP5, respectively.
H
N
H
N
1) Dil. HCl, 0oC
Water/ EtOAc
S
S
CH3
CH3
CH3
CH3
O
N
O
N
O
O
N,O-Bis(trimethylsilyl)
acetamide, CH2Cl2
2)
COOK
COOTMS
Cl
NaH, CH2Cl2
0oC to RT
O
The aim of the current study was to try to advance this
nanoparticle methodology to penicillins that are used
clinically in a water-soluble salt form and whose targets
reside along the outer periphery of bacterial cell mem-
branes. b-Lactam antibiotics such as the penicillins have
been mainstays of clinical treatment for many types of
bacterial infections, however, their effectiveness is signif-
icantly compromised in bacteria that produce b-lacta-
mase enzymes which hydrolyze the b-lactam ring to an
inactive ring-opened product. The hope was that penicil-
lins could be incorporated into the nanoparticle frame-
work during emulsion polymerization, despite their
ready water-solubility, and be sheltered inside the matrix
of the nanoparticle from bacterial penicillinase degrada-
tion. If this could be achieved, the b-lactam antibiotics
could potentially be rendered highly effective against
MRSA. As prototypes to explore this, five acrylated
penicillins 1–5 were employed as monomeric substrates
as a means to produce antibiotic-containing nanoparti-
cles in aqueous media (Fig. 1). Our primary objective
in selecting these particular penicillin derivatives was
to evaluate the effect of the location, length, and polarity
of the acrylate moiety linking the drug to the polymer
framework (NP1–NP5), versus having the drug mole-
cule non-covalently contained within the nanoparticle
matrix (NP6 and NP7).
O
N
S
CH3
CH3
O
N
O
COOH
1
Scheme 2. Synthesis of penicillin acrylamide monomer 1.
tively on the amido nitrogen rather than on the meso-
meric oxygen center, to give the imide 1 as shown.
Penicillin acrylamide monomers 2 and 3 were synthe-
sized from 6-aminopenicillanic acid using either acryloyl
chloride or acryloyloxyacetyl chloride in the presence of
triethylamine in methylene chloride. Similarly, glycosyl-
ated penicillin monomer 5 was synthesized from 6-amin-
openicillanic acid and the relevant carbohydrate acyl
chloride. Acrylated penicillin G monomer 4 was pre-
pared by stirring penicillin G free acid with triethyl-
amine and an equivalent of ethyl chloroformate at
0 °C, then further with 2-hydroxyethyl acrylate
(2-HEA) as a means to produce the esterified penicillin
G (Scheme 3).
Two penicillin G analogues (6 and 7) were also prepared
that lack an acrylated side chain in order to construct
nanoparticles (NP6 and NP7) where the drug is incorpo-
rated non-covalently (Fig. 2). These two analogues were
chosen because while penicillin G (6) possesses the free
carboxyl acid needed for binding to the transpeptidase
target, analogue 7 does not and would presumably
Penicillin acrylimide 1 was synthesized from commer-
cially available penicillin G potassium salt via its
trimethylsilyl ester, which was reacted with acryloyl
chloride followed by an aqueous work up (Scheme 2).
It is our assumption that the acylation occurred selec-

3470
E. Turos et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3468–3472
H
N
H
N
H
N
H
N
In vitro experiments were conducted to determine if the
penicillin-containing nanoparticles displayed antibacte-
rial activity against a methicillin-susceptible form of
S. aureus (MSSA), and if so, whether this activity could
be retained against a methicillin-resistant form (MRSA).
Minimum inhibitory concentration values (determined
by broth dilution assays) for each of the samples are
provided in Table 1. These assays showed that penicil-
lin-bound nanoparticles (NP1–NP4) and the penicillin-
encapsulated nanoparticles (NP6 and NP7) displayed
significant antibacterial activity against both MSSA
and MRSA, while the antibiotic-free nanoparticles
(NP0) were completely inactive. Penicillin G (free drug
form) is about 1500 times less active against MRSA
compared to MSSA, indicative of the presence of peni-
cillinase in MRSA.
S
S
Ph
Ph
ClCO2Et
O
O
O
O
Et3N, 0o
CH2Cl2
C
O
CO2H
O
O
O
Penicillin G
Et3N
2-HEA
CH2Cl2
H
N
H
N
S
Ph
O
O
O
O
O
O
4
Scheme 3. Synthesis of penicillin monomer 4.
While none of the penicillin nanoparticles had the anti-
bacterial activity of penicillin G alone against S. aureus,
the nanoparticles NP1 did display much stronger activity
than penicillin G against MRSA. In fact, the penicillin-
containing nanoparticles displayed equal antibacterial
activities for S. aureus and MRSA, indicating that the
b-lactamase enzyme exuded by the bacterium does not
at all affect the activity of the drug-bound nanoparticles.
This was further demonstrated with a control experiment
on agar plates in which the nanoparticle emulsion NP1
was screened against S. aureus (MSSA) in the absence
versus the presence of added b-lactamase enzyme
(Fig. 3). The image on the left shows three different
amounts of NP1 (20, 50, and 100 lg of antibiotic) as well
as penicillin G (20 lg) as a positive control. The penicillin
G free drug produces a large clearing zone due to inhibi-
tion of MSSA growth. However, when 100 lg of com-
mercial penicillinase (Type I from Bacillus cereus) is
added to the media (right image), penicillin G loses all
of its antibiotic activity, while the nanoparticles NP1 re-
tain their original inhibition capabilities. This confirms
the ability of the nanoparticle to protect the antibiotic
from enzymatic degradation by penicillinase (even at
unusually high concentrations of protein) without cur-
tailing its antibacterial properties. While this agar diffu-
sion experiment is illustrative, the broth MIC
H
H
N
N
S
S
O
N
N
O
O
O
O
O
O
CO2H
O
6
7
Figure 2. Penicillin G (6) and penicillin ester 7 used for preparing
drug-encapsulated nanoparticles NP6 and NP7, respectively.
require cleavage of the ester moiety to be effective as an
antibiotic.
Commercially available penicillin G potassium salt was
first converted to the more highly organic-soluble free
acid 6 using dilute hydrochloric acid, which was then
subjected to EDCI coupling with 2-hydroxyethyl propi-
onate to produce ester derivative 7. Penicillin analogues
1–7 were rigorously purified by column chromatography
prior to use in the emulsion polymerization.
Penicillin nanoparticles were prepared by dissolving one
of the seven antibiotics (1% w/w) in a 7:3 (w/w) mixture
of butyl acrylate and styrene (Scheme 4).8 This mixture
was then pre-emulsified by adding sodium dodecyl sul-
fate (3% w/w) as a surfactant with stirring in nano-puri-
fied water. The resulting micelles were then heated in a
70 °C oil bath under an atmosphere of nitrogen and
treated with potassium persulfate (0.5% w/w), a water-
soluble radical initiator, to induce free radical polymer-
ization. The reactions were allowed to run at 70 °C for
6 h with rapid stirring. After cooling to room tempera-
ture, the samples were characterized to determine their
antibacterial capabilities, average size, shape, and stabil-
ity in solution.
Table 1. Antibacterial activity of penicillin-containing nanoparticles
NP1–NP7 against Staphylococcus aureus (ATCC 25923) and MRSA
(ATCC 43300), as determined by broth dilution MIC assaysa
Sample
Microbe
Staphylococcus aureus (lg/mL) MRSA (lg/mL)
NP0 (control) >256
>256
2
64
NP1
NP2
2
64
NP3
NP4
16
16
16
16
NP5
NP6
>256
64
64
>256
64
64
O
O
O
O
O
N
O
S
O
Butyl acrylate-Styrene
N
NP7
Penicillin G
O
N
potassium persulfate
sodium dodecyl sulfate
H2O, 70oC, 6-8h
0.012
16
COOH
S
N
a MIC values refer to the lowest concentration of the bound drug
(lg/mL) required to completely inhibit bacterial growth for 24 h in
culture. MIC determinations were run in triplicate and the values
obtained for each compound were found to be invariant.
1
HOOC
Scheme 4. Representative example of an emulsion polymerization
using acrylated penicillin monomer 1.

E. Turos et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3468–3472
3471
carboxylic ester-linked system, NP4, is active against
both MSSA and MRSA is notable, in that it implies that
the active drug is cleaved from the nanoparticle frame-
work, perhaps by esterases in the bacterial cell or along
the bacterial membrane, since the free carboxylic acid
form of penicillin is required for interaction with the tar-
get transpeptidases of S. aureus. These data also indicate
that antibacterial activity is somewhat greater when the
drug is covalently bound to the polymeric nanoparticle,
as opposed to being non-covalently encapsulated within
the nanoparticle, as in NP6 and NP7. This suggests that
the nanoparticle may enhance delivery of the bound
antibiotic to the bacterial cell. Additional studies are
underway in our laboratory to understand this interac-
tion and to obtain detailed insight as to how the drug
is being released into the cell. The MIC data obtained
for all of the penicillin nanoparticles against MSSA
and MRSA infer that drug release must occur after
interaction of the particle with the cell, since release of
the b-lactam drug outside the cell would likely result
in rapid hydrolytic cleavage of the antibiotic by extracel-
lular penicillinases exuded by the MRSA, and thus loss
of bioactivity. This expected loss of activity is clearly not
what we observe for the nanoparticles. Our attempts to
measure the rate of cleavage of the drug molecules from
the nanoparticle matrix using various commercial ester-
ases failed to provide any useful information thus far.
We suspect that the cleavage event may require not only
the presence of the esterase but also the interaction be-
tween the nanoparticle and the cellular membrane.
Figure 3. Kirby–Bauer studies of NP1 using Staphylococcus aureus
(ATCC 25923). Assays were performed first in the absence of added
penicillinase protein (left image) and then in the presence of 100 lg of
penicillinase added to the agar (right image). The control, penicillin G,
lost all of its activity in the presence of the enzyme, while NP1 retained
its original activity at all three drug amounts, as noted. The cloudy
white spots appearing in the inhibition zones in the penicillinase-
treated plate (right) appear to be due to uneven diffusion of the
polymer through the agar in the presence of the added protein, which
we have observed occasionally in Kirby–Bauer experiments (and seen
around the edges of the wells), and not from surviving bacterial
colonies. These spots were sampled and cultured on agar to confirm
that no surviving bacteria were present.
measurements in Table 1 are more meaningful for quan-
titatively assessing in vitro activity of these nanoparticle-
bound antibiotics. NP1 also retained its full antibacterial
activity against MRSA after being exposed to 10% fetal
bovine serum (FBS) for 24 h, indicating that the nano-
particles are stable to the various enzymes and other deg-
radative components of serum. MIC assays performed
by substituting the commonly used Mueller–Hinton
broth with 10% FBS solution confirmed that the MIC
values of NP1 remained at 2 lg/mL, respectively, against
both the MSSA and MRSA strains.
Particle size analysis of the emulsions was performed
using transmission electron microscopy (TEM), atomic
force microscopy (AFM), and dynamic light scattering
(DLS) analysis, which all indicated that the nanoparti-
cles possess uniformly spherical morphology with diam-
eters of 25–40 nm (Fig. 4). Zeta potential analysis
showed that the particles’ surface charge in the emul-
sions ranged from À40 to À80 mV, suggesting high par-
ticle stability. The thermal stability of the nanoparticle
emulsions was also examined over a range of storage
conditions. Experiments indicated that the original
emulsions (containing 20% solid polymer content) are
stable from pH 1 through pH 10, and to temperatures
from 5 to 40 °C without precipitation or changes in par-
ticle size or morphology, as determined by DLS. More-
over, nanoparticle samples NP0 (no drug), NP1, and
NP4 were analyzed for cytotoxicity against human
The microbiological data in Table 1 suggest that the
penicillin G-conjugated nanoparticles (NP1 and NP4)
are significantly more active than the 6-aminopenicillan-
ic acid-containing nanoparticles (NP2 and NP3), and
that the type of linkage used to covalently attach the
drug moiety to the polymer backbone significantly alters
the activity of the drug-conjugated nanoparticle. Both
the type and the location of the linkage were varied in
order to optimize the antibacterial activity of the penicil-
lin nanoparticle system. The data suggest that the bioac-
tivity follows the relative order of hydrolytic sensitivity
of the linkage, with the imide > ester > amide
(NP1 > NP3 = NP4 > NP2 ꢀ NP5). The fact that the
Figure 4. AFM images show the particles from the NP1 and NP4 emulsions are uniformly in the range of 25–40 nm in diameter.

3472
E. Turos et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3468–3472
dermal fibroblast cells using a CyQUANT DNA bind-
ing assay. These experiments showed that all three nano-
particle formulations are non-toxic to the fibroblast cells
at 1 mg/mL drug concentrations, which are more than
10 times the bacterial MIC levels.
Department of Homeland Security for graduate fellow-
ships to K.G.
References and notes
In summary, these studies have established that penicil-
lin-containing polyacrylate nanoparticles can be easily
prepared by free radical emulsion polymerization in
water and are effective at helping to rejuvenate the
in vitro antibacterial activity of penicillin drugs against
beta-lactamase-producing MRSA. The favorable
features of this methodology include simple, one-step
preparation of penicillin-containing nanoparticles in
aqueous media, rigid control of nanoparticle size
in the 25–40 nm range, absence of cytotoxic effects in
healthy human cell culture, and ability to incorporate
drugs covalently onto the polymer framework or non-
covalently (encapsulated) without the need for post-syn-
thetic modification of the nanoparticle. Ongoing studies
in our laboratory are attempting to elucidate the mech-
anism of bioactivity and drug release of the nanoparti-
cles, and examining their in vivo properties in animal
infection models to determine their efficacy and poten-
tial for therapeutic applications.
1. Zafeiris, S. Nature 2004, 431, 892.
2. Norrby, S. N.; Nord, C. E.; Finch, R. Lancet Infect. Dis.
2005, 5, 115.
3. Walsh, C. Nat. Rev. Microbiol. 2005, 1, 65.
4. Patel, V. R.; Amiji, M. M. Pharm. Res. 1996, 13, 588.
5. Costerton, J. W.; Stewart, P. S.; Greenberg, E. P. Science
1999, 284, 1318.
6. Michelland, S. H.; Alonso, M. J.; Andremont, A.;
Maincen, P.; Sauzieres, J.; Couvreur, P. Int. J. Pharm.
1987, 35, 121.
7. Fattal, E.; Youssef, M.; Couvreur, P.; Andremont, A.
Antimicrob. Agents Chemother. 1989, 33, 1540.
8. Couvreur, P.; Fattal, E.; Alphandary, H.; Puisieux, F.;
Andremont, A. J. Controlled Release 1992, 19, 259.
9. Couvreur, P.; Dubernet, C.; Puisieux, F. Eur. J. Pharm.
Biopharm. 1995, 41, 2.
10. Alle’mann, E.; Gurny, R.; Doelker, E. Eur. J. Pharm.
Biopharm. 1993, 39, 173.
11. Cavallaro, G.; Fresta, M.; Giammona, G.; Puglisi, G.;
Villari, A. Int. J. Pharm. 1994, 111, 31.
12. Fontana, G.; Pitarresi, G.; Tomarchio, V.; Carlisi, B.; San
Biagio, P. L. Biomaterials 1998, 19, 1009.
13. Fontana, G.; Licciardi, M.; Mansueto, S.; Schillaci, D.;
Giammona, G. Biomaterials 2001, 22, 2857.
14. Santos-Magalhaes, N. S.; Pontes, A.; Pereira, V. M. W.;
Caetano, M. N. P. Int. J. Pharm. 2000, 208, 71.
15. Turos, E.; Shim, J.-Y.; Wang, Y.; Greenhalgh, K.; Reddy,
G. S. K.; Dickey, S.; Lim, D. V. Bioorg. Med. Chem. Lett.
2007, 17, 5392.
16. Turos, E.; Long, T. E.; Konaklieva, M. I.; Coates, C.;
Shim, J.-Y.; Dickey, S.; Lim, D. V.; Cannons, A. Bioorg.
Med. Chem. Lett. 2002, 12, 2229.
Acknowledgments
We thank Dr. Thomas Koob (Shriners Hospital, Tam-
pa) for assistance with performing the cytotoxicity
experiments, Gil Brubaker (University of Florida Parti-
cle Engineering Research Center) for helping with parti-
cle analyses, and August Heim (University of South
Florida) for assistance with AFM imaging experiments.
Financial support from the National Institutes of Health
(R01 AI051351) and National Science Foundation
(0620572) is gratefully acknowledged. We sincerely
thank the NSF-IGERT (DGE 0221681) and U.S.
17. Turos, E.; Coates, C.; Shim, J.-Y.; Wang, Y.; Leslie, J. M.;
Long, T. E.; Reddy, G. S. K.; Ortiz, A.; Culbreath, M.;
Dickey, S.; Lim, D. V.; Alonso, E.; Gonzalez, J. Bioorg.
Med. Chem. 2005, 13, 6289.