Formulation and evaluation of water-in-oil amoxicillin-loaded nanoemulsions
using for Helicobacter pylori eradication
Yu-Hsin Lin1＃†, Shu-Fen Chiou2†, Chih-Ho Lai3, Shih-Chang Tsai1,
Chih-Wei Chou4, Shu-Fen Peng1, Zih-Sian He1
1
Department of Biological Science and Technology, China Medical University, Taichung, Taiwan.
2
Institute of Molecular Medicine, National Tsing Hua University, Hsin Chu, Taiwan.
3
Department of Microbiology, School of Medicine, China Medical University, Taichung, Taiwan,
4
Department of Cosmeceutics, China Medical University, Taichung, Taiwan.
＃
Correspondence to:
Yu-Hsin Lin, PhD
Associate Professor
Department of Biological Science and Technology
China Medical University,
Taichung, Taiwan, 40402
Fax:
886-4-2207-1507
E-mail:
ylhsin@mail.cmu.edu.tw
†The first two authors (Yu-Hsin Lin and Shu-Fen Chiou) contributed equally to this work.
1
Abstract
Antibiotics have a short residence time and have low concentrations when absorbed through
the basolateral membrane in the stomach; this causes a failure to enhance drug concentrations at
Helicobacter pylori infection sites. This study developed a nanocarrier system with the ability to
carry amoxicillin to increase its efficacy against H. pylori. We used a water–in–oil emulsification
system to prepare a positively charged nanoemulsion particle composed of amoxicillin, chitosan,
and heparin. The particle size of the prepared nanoemulsion particle was controlled by the
constituted compositions. The morphology of the nanoemulsion particles was spherical. In vitro
analysis of amoxicillin released indicated that the nanocarrier system controlled amoxicillin release
in the gastrointestinal dissolution medium and amoxicillin-loaded nanoemulsion particles localized
to the site of H. pylori infection. Meanwhile, results of in vivo clearance assays indicated that the
prepared amoxicillin-loaded nanoemulsion particles had a significantly greater H. pylori clearance
effect in the gastric infection mouse model than the amoxicillin solution alone.
Keywords: Helicobacter pylori; amoxicillin; emulsification system; nanoemulsion particle
2
1. Introduction
Helicobacter pylori, a Gram-negative, microaerophilic spiral bacterium that colonizes the
mucosa of the human stomach, has been considered to be an important etiological factor in the
development of peptic ulcer disease and gastric cancer [1–3]. For effective H. pylori eradication, the
standard treatment in the case is a combination of drugs, including antibiotics and a proton pump
inhibitor [4]. The microorganism mainly produce virulence factors, including vacuolating cytotoxin
which causes epithelium cells degradation and colonized deeply within the gastric mucus layer [5,6].
The particulate system for translocation permeability through gastrointestinal mucin was found to
be increased as the particle size decreased to nanosize particles [7,8]. Encapsulation of low
molecular weight hydrophilic drugs such as antibiotics in these nanocarriers failed because such
small and hydrophilic molecules rapidly leaked from their, nanocarriers has resulted in poor
encapsulation efficiency and fast release upon dilution [9,10]. To overcome this issue, we
established the water-in-oil emulsification technique for preparing nanoemulsion particles with a
better encapsulated antibiotic ability. The emulsion technique is a heterogeneous system, that
consists of a water phase dispersed in an oil phase in the presence of blends of hydrophilic and
hydrophobic surfactants under homogenization to produce fine droplets [11]. The efficiency of
blends of hydrophilic and lipophilic surfactants between oil and water phase was primarily
determined by a hydrophilic-lipophilic balance (HLB) system. A surfactant with a higher HLB
value is usually more hydrophilic, and one with a lower value is more lipophilic. An optimal HLB
value of the surfactants was a key factor for formation of emulsion with minimal droplets [12,13].
In our emulsion method, we selected liquid paraffin as the oil phase and two surfactants,
sorbitan monolaurate (Span 20) and polyoxyethylene sorbitan monolaurate (Tween 20), to optimize
surfactant levels and HLB value of the antibiotic-loaded chitosan/heparin nanoemulsion particles.
Bayindir et al. previously developed niosomal formulations using nonionic surfactants (such as
Span 20 and Tween 20) to achieve gastrointestinal stability for paclitaxel oral delivery [14]. It has
also been reported that Span 20/Tween 20 mixtures increased the stability of the emulsification
procedure compared with pure Span or Tween systems [15]. We prepared nanoemulsion particles
composed of chitosan, heparin and an antibiotic (amoxicillin), with liquid paraffin and nonionic
3
surfactant mixture (Span 20/Tween 20) as the oil phase. The molecular formula of hydrophilic
amoxicillin is C16H19N3O5S-3H2O and the molecular weight (MW) is 419.45. The antibiotic
amoxicillin is a semisynthetic antibiotic that binds to penicillin-binding proteins and interferes with
bacterial cell wall synthesis, resulting in lysis of replicating bacteria [16,17]. It was reported that the
failure of antibiotic therapies is attributable to the poor stability of the drug in the gastric acid and
the poor permeation of the antibiotic across the mucus layer. This followed by resecretion into the
lumen, where a sufficient amount of the drug diffuses into the bacteria [18,19].
Thus, the hypothesis of our prepare nanoemulsion particles could encapsulate amoxicillin and
infiltrate into the mucus layer, subsequently, amoxicillin release from amoxicillin-loaded
nanoemulsion particles, then directly acts locally on H. pylori at a bactericidal concentration [Fig.
1(A)]. We examined their physicochemical characteristics using fourier-transformed infrared
spectroscopy (FT–IR), transmission electron microscopy (TEM), and dynamic light scattering. We
also investigated amoxicillin release characteristics from the prepared nanoemulsion particles and
examined the inhibition of H. pylori growth. In addition, the effect of the nanoemulsion particles
and their mechanism of interaction with H. pylori were investigated in the human gastric mucosal
AGS cell line (human gastric adenocarcinoma cell line) with confocal laser scanning microscopy
(CLSM) [20]. We also assessed the in vivo clearance effect of amoxicillin-loaded chitosan/heparin
nanoemulsion particles in H. pylori infected mice.
2. Materials and methods
2.1. Materials
Chitosan (MW 50 kDa) with approximately 85% deacetylation was obtained from Koyo
Chemical Co. Ltd. (Japan). Heparin (5000 IU/mL, MW 15 kDa, 179 IU/mg) was purchased from
Leo Chemical Factory (Ballerup, Denmark). Liquid paraffin was purchased from Wako Pure
Chemical Industries, Ltd. (Osaka, Japan). Tween 20, Span 20, amoxicillin, acetic acid,
3-(4,5-dimethyl-thiazol-yl)-2,5-diphenyltetrazolium bromide (MTT), 4',6-diamidino-2-phenylindole
(DAPI), fluoresceinamine isomer I (FA), phosphate buffered saline (PBS), sodium acetate,
paraformaldehyde, bismuth subnitrate, Hanks’ balanced salt solution (HBSS), and β-cyclodextrin
4
were purchased from Sigma–Aldrich (St Louis, MO). RPMI 1640, fetal bovine serum (FBS),
penicillin, streptomycin, and trypsin-ethylene diamine tetraacetic acid (trypsin-EDTA) were from
Gibco (Grand Island, NY). N-hydroxy-succinimide (NHS)-functionalized cyanine 3 (Cy3-NHS)
was
from
Amersham
Biosciences
(Piscataway,
1,1’-dioctadecyl-3,3,3‘,3’-tetramethylindodicarbocyanine-5,5’-disulfonic
acid
NJ).
(DilC18(5)-DS)
lipophilic dye was from Molecular Probes (Eugene, USA). All other chemicals and reagents were of
analytical grade.
2.2. Preparation of chitosan/heparin nanoemulsion particles by water-in-oil emulsification
The nanoemulsion particles were prepared by water-in-oil emulsification technique through
homogenization using a laboratory-type rotating blade homogenizer (IKA Labortechnik, Staufen,
Germany). Table 1 and Table 2 list the conditions for preparation of the nanoemulsion particles, as
well as the mean particle sizes and zeta potential values. First, the distinct Span 20:Tween 20 ratio
(50:50, 67:33, 75:25, and 80:20) of the Span 20/Tween 20 mixture surfactant (0.4, 0.6, 0.8, 1.0, and
1.2 mL) was added to liquid paraffin (80.0 mL) under continuous mixing. Second, the aqueous
heparin (0.2 mg/mL, 4 mL, pH 7.4) was adding to liquid paraffin containing the Span 20/Tween 20
surfactant and homogenization at 15,000 rpm at 4℃ for 2 min. Then, aqueous chitosan (0.6 mg/mL,
4 mL, pH 6.0) was slowly dropped into the resultant heparin emulsion and subject to
homogenization at 15,000 rpm at 4℃ for 2 min. The prepared chitosan/heparin nanoemulsion
particles were centrifuged twice at 32,000 rpm for 30 min, then the pellets were washed with
distilled water and 50% aqueous alcohol to remove traces of paraffin oil and surfactant in this way.
Finally, the pellets were collected and suspended in deionized water for further study.
2.3. Characterization of the prepared nanoemulsion particles
The size distribution and zeta potential value of the nanoemulsion particles at pH 1.2 (0.1 M
HCl, simulated gastric medium), pH 6.0, and pH 7.0 (10 mM PBS) simulating the gastric mucosa
and the H. pylori survival situation medium were measured using a dynamic light scattering
analyzer (Zetasizer ZS90, Malvern Instruments Ltd., UK) [21,22]. The peak variations of the
5
nanoemulsion particles at different pH values were characterized using FT–IR (Shimadzu Scientific
Instruments, USA). The morphology of the nanoemulsion particles was visualized under TEM at
different pH values. The TEM samples were prepared as follows. The particle suspension was
placed onto a 400 mesh copper grid coated with carbon. About 2 min after deposition, the grid was
tapped with a filter paper to remove surface water and positively stained with an alkaline bismuth
solution.
2.4. Encapsulation efficiency and release profiles of amoxicillin–loaded nanoemulsion particles
To study the release profiles of amoxicillin from amoxicillin-loaded chitosan/heparin
nanoemulsion particles, the amoxicillin–loaded nanoemulsion particles were prepared. In brief,
amoxicillin (0.2 mg/mL, 0.4 mg/mL, and 0.6 mg/mL, 2 mL, pH 7.4) was mixed with aqueous
heparin (0.4 mg/mL, 2 mL, pH 7.4) under continuous stirring for 12 hr at 4℃ and then added to
liquid paraffin containing the Span 20/Tween 20 mixture surfactant with homogenization, as
described above. Then, the aqueous chitosan (0.6 mg/mL, 4 mL, pH 6.0) was slowly dropped into
the resultant heparin/amoxicillin emulsion during homogenization at 15,000 rpm at 4℃ for 2 min.
Finally, the amoxicillin-loaded chitosan/heparin nanoemulsion particles were obtained after
centrifugation, washed, and then suspended in deionized water. To determine the loading efficiency
and loading content, the amoxicillin concentration was assayed by high–performance liquid
chromatography (HPLC). The release profiles of amoxicillin from test particles were investigated in
simulated dissolution medium (pH 1.2 for 120 min, pH 6.0 for 120 min, and pH 7.0 for 360 min) at
37℃. At set time intervals, samples were removed and centrifuged, and the supernatants were
subjected to HPLC. The percentage of cumulative amount of released amoxicillin was determined
using a standard calibration curve.
2.5. Viability of AGS cells treated with chitosan/heparin nanoemulsion particles
The AGS cell line (ATCC CRL 1739) was obtained from the American Type Culture
Collection (ATCC) and used between passages 40 and 60. The AGS cells were cultured in RPMI
1640 medium containing 10% FBS, penicillin (100 IU/mL), and streptomycin (100 mg/mL), and
6
were kept in an incubator at 37℃, 95% humidity, and 5% CO2 [23]. The cells were harvested for
subculture every 3 days with 0.25% trypsin plus 0.05% EDTA solution and were used for the
cytotoxicity experiments. The cytotoxicity of the test samples was evaluated in vitro using the MTT
assay. AGS cells were seeded at 5×104 cells/well in 96-well plates and allowed to adhere overnight.
The growth medium was replaced with HBSS solution (pH 6.0) that contained various
concentrations (0.02-0.20 mg/mL) of chitosan/heparin nanoemulsion particles and incubated for 2
hr. After 2 hr, the test samples were aspirated and the cells were washed twice with 100 µL of PBS.
The cells were then incubated in growth medium for an additional 22 hr. The cells were then
incubated in growth medium containing 1 mg/mL MTT for an additional 4 hr. Dimethyl sulfoxide
(100 µL) was added to each well to ensure solubilization of the formazan crystals. The optical
density was read with a Molecular Devices SpectraMax M2e microplate spectrofluorometer
(Sunnyvale, CA) at a wavelength of 570 nm. All experiments were performed six times with eight
replicate wells for every sample and control per assay.
2.6. In vitro cellular uptake and CLSM visualization
To track the cellular particles, fluorescent Cy3-chitosan and FA-heparin were prepared
according to the procedure described in our previous study [24]. The fluorescent
Cy3-chitosan/FA-heparin nanoemulsion particles were prepared according to the procedure
described in Section 2.2. The AGS cells were seeded onto 6-cm petri dishes at a density of 5 × 105
cells/cm2 and incubated for 2 days. After incubation, the medium was removed and cells were
treated with HBSS solution (pH 6.0) containing Cy3-chitosan/FA-heparin nanoemulsion particles at
a concentration of 0.1 mg/mL. After 2 hr, the test samples were aspirated and the cells were then
washed three times with PBS before being fixed in 3.7% paraformaldehyde. The cells were washed
three times with PBS and stained with DAPI, which specifically bind to the nuclei. The stained cells
were examined with excitation at 340, 488, and 543 nm using a CLSM. The images were
superimposed using LCS Lite software (version 2.0).
7
2.7. Evaluating the relationship between H. pylori and amoxicillin-loaded nanoemulsion particles
in co-culture with AGS cells
To observe the adhesion of H. pylori to cells, the fluorescent bacteria were labeled with
DilC18(5)-DS fluorescently labeled lipophilic dye according to the procedure described in our
previous study [25]. Additionally, synthesis of FA-amoxicillin was based on the reaction between
the amine group of FA and the carboxylic acid group of amoxicillin. Amoxicillin (60 mg) was
dissolved completely in 30 mL of deionized water and 2 mg of FA was dissolved completely in 1
mL of acetonitrile. The FA solution was added gradually to the amoxicillin solution, then 1 mg of
1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride was added with continuous stirring
at room temperature for 120 min and then freeze dried. To remove the unconjugated FA, the dry
FA-amoxicillin sample was precipitated by adding acetonitrile (30 mL), and the precipitate was
subjected to repeated cycles of washing and centrifugation until no fluorescence was detected in the
supernatant. The resultant FA-amoxicillin was dissolved in deionized water then freeze dried.
To examine the effects on H. pylori of amoxicillin–loaded nanoemulsion particles in
co-culture with AGS cells, the AGS cells were seeded on Costar Transwell six-well plates (Corning
Costar Corp., NY) at a seeding density of 5 ×105 cells/insert. The AGS cell culture medium was
added to both the donor and acceptor compartments. The medium was replaced every 48 hr for the
first 6 days and every 24 hr thereafter. The cultures were kept in an incubator and were used for the
H. pylori infection experiments 26–30 days after seeding [26,27]. DiIC18(5)-H. pylori was
incubated with the AGS cells for 2 hr. The medium (pH 6.0) containing the FA-amoxicillin–loaded
Cy3-chitosan/heparin nanoemulsion particles or fluorescent FA-amoxicillin solution were then
introduced into the donor compartment of the AGS cells for 2 hr at 37℃. After incubation, the test
samples were aspirated. The cells were washed twice with PBS and stained with DAPI for 15 min.
The stained cells were examined by CLSM with excitation at 340, 488, 543, and 633 nm at 0.2 μm
intervals, and 3D images were created using LCS Lite software.
2.8. In vitro and in vivo H. pylori growth inhibition study
We quantified the inhibitory effects of amoxicillin-loaded nanoemulsion particles on growth
8
of H. pylori strains. Briefly, the AGS cells were seeded in 24-well plates at 1×105 cells/well in 500
μL of medium and incubated overnight at 37℃. The following day, H. pylori at a final concentration
of 8×108 CFU/mL (colony-forming unit/mL) were added to the AGS cells in the well. After 2 hr,
the medium was removed and cells were treated with HBSS solution (pH 6.0) containing
amoxicillin solution and amoxicillin-loaded chitosan/heparin nanoemulsion particles with varying
amoxicillin concentrations (0.0 mg/L, 2.0 mg/L, 4.0 mg/L, 8.0 mg/L, 16.0 mg/L) for 2 hr. After 2 hr,
the treated AGS cells were washed and incubated for an additional 22 hr. The inhibition of AGS
cell growth by H. pylori was analyzed using the trypan blue exclusion [28].
An H. pylori infectious animal model was established according to Qian's method (China
Patent, CN 1304729A) with some modifications to determine the ability of amoxicillin or
amoxicillin–loaded nanoemulsion particles to clear H. pylori in vivo. Healthy and disease-free
6–week–old male C57BL/6J mice were used for the study. After an overnight fasting, mice were
inoculated with an equal amount of bacterial suspension (0.5 mL) by intragastric gavage using a
sterile oral feeding needle with each dose containing approximately 1×109 CFU/mL of H. pylori.
The dosing was repeated once daily for 7 consecutive days. After the development of infection after
1 week, the mice were randomly divided into different groups. Each group contained six mice and
received different amoxicillin formulations (15 mg/kg in the form of amoxicillin solution or
amoxicillin–loaded nanoemulsion particles, and deionized water as a control) once daily for 7
consecutive days. One day after administration of the final dose, the mice were sacrificed and their
stomachs removed and subjected to the following tests. Each stomach was homogenized with sterile
normal saline (3 mL/stomach) from which serial dilutions were plated on blood agar plates under
micro-aerophilic conditions for 5 days at 37°C. The viable bacterial count for each gastric wall was
calculated by counting the number of colonies on the agar plates.
2.9. Statistical analysis
Statistical analysis of the differences in the measured properties of the groups was performed
with one-way analysis of variance and the determination of confidence intervals, with the statistical
package Statistical Analysis System, version 6.08 (SAS Institute Inc., Cary, NC). All data are
9
presented as means and standard deviations, indicated as “mean ± SD”. Differences were
considered to be statistically significant when the P values were less than 0.05.
3. Results
3.1. Preparation of nanoemulsion particles
The Chitosan/heparin nanoemulsion particles were prepared by a water–in–oil emulsification
method using a mixture of Span 20 and Tween 20 nonionic surfactants in liquid paraffin. We
prepared emulsions with various Span 20:Tween 20 ratios, namely: 50:50, 67:33, 75:25, and 80:20
to determine the resulting HLB values. The 75:25 (w/w) Span 20:Tween 20 ratio produced small
particle sizes (270.1 ± 9.8 nm), with a small polydispersity index (PDI) 0.19 ± 0.06 and constant
HLB number of 10.6; this ratio was used throughout the remainder of the study (Table 1). We
investigated the effect of varying the concentration of the Span20-Tween20 surfactant mixture
(0.50% wt.%, 0.75% wt.%, 1.00% wt.%, 1.25% wt.%, and 1.50% wt.%) on the formation of
chitosan/heparin nanoemulsion particles. As shown in Table 2, as the amount of Span20/Tween20
mixture surfactant incorporated in emulsification phase was increased, the size of the resulting
particles decreased appreciably and to give a uniform matrix with a spherical shape [Fig. 1(B)].
From these results, optimal conditions for preparing a fine chitosan/heparin nanoemulsion particles
by water–in–oil emulsification were: an aqueous phase (4 mL) containing chitosan (0.6 mg/mL) and
heparin (0.2 mg/mL), and an oil phase of paraffin oil (80 mL) containing 1.2 mL of a
Span20:Tween20 in a 75:25 ratio. Nanoemulsion particles prepared to this specific composition
were used for the remainder of the study.
3.2. Characterization of the prepared nanoemulsion particles at specific pH values
The morphologies of the prepared chitosan/heparin nanoemulsion particles at various pH
values were examined by TEM, and pH related chemical changes were followed by FT–IR [Fig. 2].
At pH 6.0 (simulating the environment of the gastric mucosa, and ideal conditions for H. pylori
survival situation), the characteristic peaks observed at 1532 cm-1 and 1633 cm-1 were the
protonated amino group (–NH3+) on chitosan and the carboxylic ions (–COO–) on heparin,
10
respectively [Fig. 2(A)]. Thus ionized chitosan and heparin formed a polyelectrolyte complex with
a spherical matrix structure [Fig. 2(B)]. At pH 1.2 (simulating the pH of gastric acid), the
characteristic peak observed at 1239 cm-1 was attributed to the heparin sulfate ions (–SO4–), but
some of the –COO– groups on heparin also became protonated (–COOH, 1736 cm-1). Electrostatic
interactions between chitosan and heparin were therefore weaker at pH 1.2 than they were at pH 6.0,
leading to greater particle sizes (289.1 ± 5.6 nm). At pH 7.0, chitosan ammonium groups were
deprotonated [indicated by disappearance of the characteristic peak for –NH3+ on chitosan; Fig.
2(A)], leading to the collapse of the nanoparticles [Fig. 2(B)].
3.3. Encapsulation efficiency and release profile of amoxicillin-loaded nanoemulsion particles
We used chitosan/heparin nanoemulsion particles to encapsulate amoxicillin at various
concentrations. It was shown that for an amoxicillin concentration of 0.6 mg/mL, particle size is
296.5 ± 6.3 nm and zeta potential is 29.8 ± 3.1 mV, while the amoxicillin loading efficiency and
percentage loading content are 54.3 ± 2.8% and 19.2 ± 1.2%, respectively. Under optimal
conditions, these amoxicillin-loaded nanoemulsion particles remained spherical [Fig. 2(B)]. We
used the optimal amoxicillin concentration of 0.6 mg/mL for subsequent in vitro studies. Fig. 2(C)
shows the in vitro release of amoxicillin in solutions buffered pH 1.2, 6.0, and 7.0. At pH 1.2, the
nanoemulsions released 20.5 ± 1.2% amoxicillin over 120 min. At pH 6.0, the structure and
morphology of nanoparticle emulsion was stable but there was only a modest release of amoxicillin
from amoxicillin-loaded nanoemulsion particles. By contrast, at pH 7.0, the prepared emulsion
particles became unstable and collapsed, causing rapid release of the amoxicillin. These findings
suggest that chitosan/heparin nanoemulsion particles have potential for sustained drug release.
3.4. Viability and cellular uptake of cells treated with chitosan/heparin nanoemulsion particles
We evaluated the cytotoxicity of various concentrations of chitosan/heparin nanoemulsion
particles using AGS cells and MTT assay [Fig. 3(A)]. Cell viability was generally unaffected at
concentrations less than 0.14 mg/mL. Thereafter, we used emulsion particles at 0.1 mg/mL to track
cellular internalization of particles without causing damage to cultured cells. We used fluorescent
11
nanoemulsion particles [Cy3-chitosan (red)/FA-heparin (green) nanoemulsion particles] to
demonstrate cellular uptake. Fig. 3(B) shows the fluorescence intensity of those AGS cells that
internalized fluorescent particles. Fluorescent signals appeared in the intercellular spaces and cell
cytoplasm [superimposed red (Cy3-chitosan)/green (FA-heparin) spots; i.e., yellow spots, white
arrows; Fig. 3(B)] after 2 hr incubation with nanoemulsion particles at pH 6.0.
3.5. Relationship between H. pylori and amoxicillin–loaded nanoemulsion particles
To study the interaction of amoxicillin–loaded nanoemulsion particles with H. pylori attached
to the epithelia, we developed a process whereby fluorescent DiIC18 (5)-H. pylori infected
monolayers of AGS cells, and the infected cells were then treated with fluorescent FA-amoxicillin
alone, or FA-amoxicillin loaded in Cy3-chitosan/heparin nanoemulsion particles. As shown in Fig.
4, the fluorescent images (those after 3D reconstruction) of AGS cell monolayers were observed (the
XY plane) and appeared at a deep level (the XZ plane). The AGS–adapted DiIC18 (5)-labeled H.
pylori (purple spot) preferentially targeted cell-cell junctions and was present within the cells. The
AGS
cell
monolayers
were
incubated
with
fluorescent
FA-amoxicillin
loaded
in
Cy3-chitosan/heparin nanoemulsion particles (Cy3-chitosan: red spot, FA-amoxicillin: green spot)
and observed by CLSM to ascertain whether the nanoparticles and drug were attached in the same
way to the intercellular spaces and the cell cytoplasm. The superimposed images (3D reconstruction
of the XY and XZ plane), show that administered amoxicillin loaded chitosan/heparin nanoemulsion
particles co-localized and interacted at the same location of intercellular spaces and the cell
cytoplasm of H. pylori infectious sites (superimposed red spot: Cy3-chitosan, green spot:
FA-amoxicillin, and purple spot: DiIC18 (5)-H. pylori appearing as white spots, green arrows; Fig.
4). By contrast, cells incubated with FA-amoxicillin solution alone (green spot), showed less
obvious fluorescent signals in intercellular spaces than those seen after incubation with fluorescent
FA-amoxicillin loaded in Cy3-chitosan/heparin nanoemulsion particles. The amoxicillin-loaded
particles produced an intense fluorescence that emanated from deep within the cells, indicating that
the nanoemulsion particles were capable of carrying amoxicillin to the H. pylori infection site. This
result suggests that our nanoemulsion particles prepared had a specific interaction with AGS cell
12
monolayers infected with H. pylori.
3.6. H. pylori growth inhibition study
Fig. 5 shows the influence on gastric mucosal cell viability of H. pylori infection alone and
after pretreatment with amoxicillin-loaded chitosan/heparin nanoemulsion particles. The figure
shows a significant reduction in AGS cell viability after infection with H. pylori (8 108 CFU/mL)
alone; cell viability was 47.6 ± 2.9% that of uninfected controls. We determined the percentage of
H. pylori-infected AGS cells after 24 hr treatment using various amoxicillin concentrations (0.0, 2.0,
4.0, 8.0, and 16.0 by mg/L). Cell survival rates seen with amoxicillin solution alone was 52.3 ±
2.4% (amoxicillin concentration of 2.0 mg/L), 57.1 ± 2.2% (4.0 mg/mL), 63.9 ± 3.9% (8.0 mg/mL),
and 70.4 ± 3.5% (16.0 mg/L). Meanwhile, the cell survival rates with amoxicillin-loaded
chitosan/heparin nanoemulsion particles were 65.1 ± 2.9% (at amoxicillin concentration of 2.0
mg/L), 72.6 ± 3.2% (4.0 mg/mL), 81.4 ± 4.3% (8.0 mg/mL), and 90.5 ± 3.8% (16.0 mg/mL),
compared
with
control
(without
sample).
Therefore,
our
prepared
amoxicillin-loaded
chitosan/heparin nanoemulsion particles with a positive surface charge significantly increased the
inhibitory effects on H. pylori infected cells compared with amoxicillin solution alone (p < 0.05).
Fig. 6 shows the in vivo clearance data of H. pylori infection after administration of 15 mg/kg
amoxicillin alone or amoxicillin-loaded chitosan/heparin nanoemulsion particles for 7 consecutive
days. The mean bacterial count of the control group of mice that were given only sterile water was
137.3 ± 15.4 (CFU/stomach). Treatment with 15 mg/kg amoxicillin solution alone gave a mean
bacterial count of 45.2 ± 8.7 (CFU/stomach). Meanwhile, treatment with 15 mg/kg amoxicillin in
amoxicillin-loaded chitosan/heparin nanoemulsion particles gave a mean bacterial count of 11.3 ±
2.8 (CFU/stomach) with significantly increased the inhibitory effects on H. pylori infected mice
compared with amoxicillin solution alone. We used a rapid urease test, also known as the
Campylobacter-like organism test (CLO test), which was developed by Marahall and specifically
designed to detect H. pylori. In the presence of H. pylori urease, urea is converted into ammonium
hydroxide and is changed the color of the indicator from yellow to red. Fig. 6 shows the mice
infected with H. pylori and administered sterile water indicate a positive response as the media
13
changed color from yellow (healthy mice) to red (infected mice). Following treatment with 15
mg/kg amoxicillin the form of amoxicillin solution alone and amoxicillin-loaded chitosan/heparin
nanoemulsion particles, the media changed color to a reddish-orange and light orange, respectively.
Therefore, we concluded that prepared positive surface charge amoxicillin nanoemulsion particles
showed a more complete H. pylori clearance effect than amoxicillin alone.
4. Discussion
Chitosan-based nanocarriers produced by gelation are efficient drug delivery systems for
hydrophilic substances such as insulin and nucleic acids [29,30]. However, the gelation approach to
preparing nanocarriers to encapsulate hydrophilic amoxicillin introduces problems associated with
drug loading efficiency and drug content. We investigated use of water-in-oil emulsification with
various surfactants, in particular, Span 20 and Tween 20 at various ratios for the preparation of
chitosan/heparin nanoemulsion particles to encapsulate amoxicillin, so improving drug entrapment
efficiency and providing a slow drug release rate. Nanoemulsion particle sizes are affected by
interactions between surfactants that arise from surfactant mixtures with opposing HLB geometries
[14,31]. Tween 20 is a hydrophilic surfactant with an HLB of 16.7, while Span 20 is a hydrophobic
surfactant with an HLB of 8.6 [32,33]. Additionally, the mixed surfactant HLBmix value could be
calculated as follows: HLBmix = HLBT × T% + HLBS × S%, where HLBT and HLBS are the HLB
values of Tween 20 (16.7) and Span 20 (8.6) respectively, T% and S% are the respective weight
percentages of Tween 20 and Span 20 present in the surfactant mixtures [34]. We found that a ratio
of 75:25 (w/w) provided an HLB value of 10.6, and gave a particle size of 270.1 ± 9.8 nm, with
positive zeta potential, 32.6 ± 1.6 mV. Surfactant concentrations in the range 0.75-1.50 by wt.%
formed emulsion particles in the size range of 270-930 nm and the mean sizes of the prepared
nanoemulsion particles decreased with increasing surfactant concentration (Table 2). Izquierdo et al.
performed similar studies with nanoemulsions and found that the observed decrease in droplet size
with increasing surfactant concentration was due to an increase in the interfacial area and decrease
in interfacial tension [35].
H. pylori produce the enzyme urease, which hydrolyzes urea to ammonia and carbon dioxide;
14
these products in turn neutralize stomach acid and raise pH to allow successful colonization of the
gastric environment by H. pylori [36]. Chitosan is a weak base, and the amino group on chitosan
has a pKa value of about 6.5. Heparin is a polydispersed, highly sulfated polysaccharide, and the
sulfate monoesters and sulfamido groups have pKa values ranging from 0.5 to 1.5, whereas the
carboxylate group has a pKa value between 2 to 4. Chitosan and heparin ionize in the pH range
1.2-6.0, and may form polyelectrolyte complexes with a spherical matrix structure. When this
matrix encapsulates amoxicillin, the amoxicillin–loaded nanoemulsion particles release only a small
quantity of the drug. At pH 7.0, the nanoemulsion particles became unstable and broke apart,
allowing rapid release of the amoxicillin [Fig. 2(B) and Fig. 2(C)]. This occurs because at pH 7.0,
chitosan is deprotonated, causing collapse of the nanoemulsion particles [Fig. 2(A)]. Thus,
nanoparticles with pH-responsive characteristics that are stable at pH 1.2-6.0 can protect drugs from
gastric acid and control amoxicillin release into an H. pylori infection environment.
H. pylori infection is considered a primary risk factor for the development of gastric ulcer and
gastric cancer [37]. When H. pylori colonizes the human gastric mucus layer and adheres to surface
epithelial cells, it produces a vacuolating cytotoxin with the ability to modulate the integrity of the
epithelium by increasing tight junction permeability to small ions and molecules. H. pylori may
preferentially adhere not only to the gastric epithelium surface, but also close to the tight junctions
of gastric mucous cells [38,39]. Therefore, the ideal antibiotic dose for H. pylori eradication must
not only localize in the stomach, but must also interact locally with the bacterium. In our study of
the relationship between amoxicillin-loaded chitosan/heparin nanoemulsion particles and H. pylori
(Fig. 4), we used CLSM to show that prepared fluorescent FA-amoxicillin loaded in
Cy3-chitosan/heparin nanoemulsion particles (Cy3-chitosan: red spot, FA-amoxicillin: green spot)
co-localized and interacted locally at sites of H. pylori infection.
Amoxicillin is a semisynthetic, orally absorbed, broad-spectrum antibiotic, which has been
extensively used to provide resistance against H. pylori cultures [40,41]. However, amoxicillin
treatment results in incomplete eradication of H. pylori, likely due to the short residence time of the
drug in the stomach which prevents effective antimicrobial concentrations being achieved in the
gastric mucous layer or epithelial cell surfaces where H. pylori exists [42,43]. To improve the
15
efficacy of anti-H. pylori agents, antibiotics need to localize at the H. pylori infection site in the
gastric epithelium. Chitosan, a polycationic, nontoxic, and mucoadhesive polymer is safe, allows
prolonged interaction between the delivered drug and the membrane epithelia, and facilitates
efficient drug diffusion into the mucus and epithelium layer [44,45]. Fig. 5 shows the
amoxicillin–loaded chitosan/heparin nanoemulsion particles with various amoxicillin concentrations
significantly increased AGS cell growth compared with amoxicillin solution alone (p < 0.05).
Furthermore, clearance of H. pylori in vivo was observed with doses of 15 mg/kg amoxicillin in
amoxicillin–loaded chitosan/heparin nanoemulsion particles. The clearance rate showed a more
complete H. pylori clearance effect than with amoxicillin alone (Fig. 6).
5. Conclusions
We prepared chitosan/heparin nanoemulsion particles using a water–in–oil emulsification
system for delivery of amoxicillin to treat H. pylori infection. The nanoemulsion particles were
stable at pH 1.2. In vitro drug release analysis of the nanoemulsion particles indicated that the
system could control amoxicillin release in a simulated gastrointestinal dissolution medium. The
amoxicillin-loaded chitosan/heparin nanoemulsion particles could localize at intercellular spaces or
in the cell cytoplasm, the site of H. pylori infection, and significantly increased H. pylori growth
inhibition compared with amoxicillin alone. Results of in vivo clearance assays indicated that
amoxicillin-loaded chitosan/heparin nanoemulsion particles had a more complete H. pylori
clearance effect on induced gastric H. pylori infection mice compared with amoxicillin alone.
Acknowledgements
This
work
was
supported
by
grants
from
the
National
Science
Council
(NSC100-2628-E-039-003-MY3) and the China Medical University (CMU-100-S-23). The Leica
TSC SP2 confocal Spectral microscopy experiment and Malvern ZS90 Zetasizer Nano apparatus
supported by the Medical Research Core Facilities center, Office of Research & Development,
China Medical University were gratefully acknowledged.
16
References
[1] Vattem DA, Lin YT, Ghaedian R, Shetty K. Cranberry synergies for dietary management of
Helicobacter pylori infections. Process Biochem 2005;46:1583-92.
[2] Peterson WL. Helicobacter pylori and peptic ulcer disease. N Engl J Med 1991;324:1043-8.
[3] Hejazi R, Amiji M. Stomach-specific anti-H. pylori therapy. I: preparation and characterization
of tetracyline-loaded chitosan microspheres. Int J Pharm 2002;235:87-94.
[4] van der Hulst RW, Keller JJ, Rauws EA, Tytgat GN. Treatment of Helicobacter pylori infection:
a review of the world literature. Helicobacter 1996;1:6-19.
[5] van Amsterdam K, van Vliet AH, Kusters JG, van der Ende A. Of microbe and man:
determinants of Helicobacter pylori-related diseases. FEMS Microbiol Rev 2006;30:131-56.
[6] Cover TL, Blaser MJ. Purification and characterization of the vacuolating toxin from
Helicobacter pylori. J Biol Chem 1992;267:10570-5.
[7] Khanvilkar K, Donovan MD, Flanagan DR. Drug transfer through mucus. Adv Drug Deliv Rev
2001;48:173-93.
[8] Gonçalves C, Pereira P, Gama M. Self-assembled hydrogel nanoparticles for drug delivery
applications. Materials 2010;3:1420-60.
[9] Hillaireau H, Le Doan T, Appel M, Couvreur P. Hybrid polymer nanocapsules enhance in vitro
delivery of azidothymidine-triphosphate to macrophages. J Control Release 2006;116:346-52.
[10] Hillaireau H, Le Doan T, Besnard M, Chacun H, Janin J, Couvreur P. Encapsulation of
antiviral
nucleotide
analogues
azidothymidine-triphosphate
and
cidofovir
in
poly(iso-butylcyanoacrylate) nanocapsules. Int J Pharm 2006;324:37-42.
[11] Motohiro S, Yohei K, Yukitaka K, Shuji A, Ryuichi M. Effect of the hydrophilic surfactants
on the preparation and encapsulation efficiency in course and fine W/O/W type emulsions.
Colloids Surf A Physicochem Eng Asp 2004;238:83-90.
[12] Jina X, Ugine TA, Chena J, Streetta AD. Method for determining the best
hydrophilic-lipophilic balance (HLB) number for a compatible non-ionic surfactant in
formulation development for aerial conidia of metarhizium anisopliae (Hypocreales:
Clavicipitaceae). Biocontrol Sci Technol 2009;19:341-7.
17
[13] Boyd J, Parkinson C, Sherman P. Factors affecting emulsion stability and the HLB concept. J
Colloid Interface Sci 1972;41:359-70.
[14] Bayindir ZS, Yuksel N. Characterization of niosomes prepared with various nonionic
surfactants for paclitaxel oral delivery. J Pharm Sci 2010;99:2049-60.
[15] Sajjadi S, Zerfa M, Brooks BW. Phase inversion in p-xylene/water emulsions with the
non-ionic surfactant pair sorbitan monolaurate/polyoxyethylene sorbitan monolaurate (Span
20/Tween 20). Colloids Surf A Physicochem Eng Asp 2003;218:241-54.
[16] Shah S, Qaqish R, Patel V, Amiji M. Evaluation of the factors influencing stomach-specific
delivery of antibacterial agents for Helicobacter pylori infection. J Pharm Pharmacol 1999;51:
667-72.
[17] Kim JM, Kim JS, Jung HC, Kim N, Kim YJ, Song IS. Distribution of antibiotic MICs for
Helicobacter pylori strains over a 16-year period in patients from Seoul, South Korea.
Antimicrob Agents Chemother 2004;48:4843-7.
[18] Endo H, Yoshida H, Ohmi N, Ohta K, Higuchi S. Localization of [14C]amoxicillin in rat
gastric tissue when administered with lansoprazole and clarithromycin. J Antimicrob Chemother
2001;48:923–26.
[19] Endo H, Yoshida H, Ohmi N, Ohta K, Higuchi S, Suga T. Localization of [14C]clarithromycin
in rat gastric tissue when administered with lansoprazole and amoxicillin. J Antimicrob
Chemother 2002;50:285–88.
[20] Hall AJ, Tripp M, Howell T, Darland G, Bland JS, Babish JG. Gastric mucosal cell model for
estimating relative gastrointestinal toxicity of non-steroidal anti-inflammatory drugs.
Prostaglandins Leukot Essent Fatty Acids 2006;75:9-17.
[21] Li Y, Wang WP, Wang HY, Cho CH. Intragastric administration of heparin enhances gastric
ulcer healing through a nitric oxide-dependent mechanism in rats. Eur J Pharmacol
2000;399:205-14.
[22] Shu XZ, Zhu KJ, Song W. Novel pH-sensitive citrate cross-linked chitosan film for drug
controlled release. Int J Pharm 2001;212:19-28.
[23] Sheng H, Shah PK, Audus KL. Sucralfate effects on mucus synthesis and secretion by human
18
gastric epithelium in vitro. Int J Pharm 1996;131:159-69.
[24] Lin YH, Chang CH, Wu YS, Hsu YM, Chiu SF, Chen YJ. Development of pH-responsive
chitosan/heparin
nanoparticles
for
stomach-specific
anti-Helicobacter
pylori
therapy.
Biomaterials 2009;30:3332-42.
[25] Chang CH, Huang WY, Lai CH, Hsu YM, Yao YH, Chen TY, Wu JY, Peng SF, Lin YH.
Development of novel nanoparticles shelled with heparin for berberine delivery to treat
Helicobacter pylori. Acta Biomater 2011;7:593-603.
[26] Kim JS, Kim JM, Jung HC, Song IS. Expression of cyclooxygenase-2 in human neutrophils
activated by Helicobacter pylori water-soluble proteins: possible involvement of NF-kappaB
and MAP kinase signaling pathway. Dig Dis Sci 2001;46:2277-84.
[27] Hennig EE, Godlewski MM, Butruk E, Ostrowski J. Helicobacter pylori VacA cytotoxin
interacts with fibronectin and alters HeLa cell adhesion and cytoskeletal organization in vitro.
FEMS Immunol Med Microbiol 2005;44:143-50.
[28] Sieveking D, Leach ST, Mitchell HM, Day AS. Role of serum factors in epithelial cell
responses to Helicobacter pylori infection in vitro. J Gastroenterol Hepatol 2005;20:1610-5.
[29] Ma Z, Lim TM, Lim LY. Pharmacological activity of peroral chitosan-insulin nanoparticles in
diabetic rats. Int J Pharm 2005;293:271-80.
[30] Liu W, Sun S, Cao Z, Zhang X, Yao K, Lu WW, Luk KD. An investigation on the
physicochemical properties of chitosan/DNA polyelectrolyte complexes. Biomaterials
2005;26:2705-11.
[31] Leong TS, Wooster TJ, Kentish SE, Ashokkumar M. Minimising oil droplet size using
ultrasonic emulsification. Ultrason Sonochem 2009;16:721-7.
[32] Tsai HA, Huang DH, Ruaan RC, Lai JY. Mechanical properties of asymmetric polysulfone
membranes containing surfactant as additives. Ind Eng Chem Res 2001;40:5917-22.
[33] Barakat NS, Yassin AE. In vitro characterization of carbamazepine-loaded precifac lipospheres.
Drug Deliv 2006;13:95-104.
[34] Kunieda H, Shinoda K. Evaluation of the hydrophile-lipophile balance (HLB) of nonionic
surfactants. I. Multisurfactant systems. J Colloid Interface Sci 1985;107:107-21.
19
[35] Izquierdo P, Esquena J, Tadros ThF, Dederen C, Garcia MJ, Azemar N, Solans C. Formation
and stability of nano-emulsions prepared using the phase inversion temperature method.
Langmuir 2002;18:26-30.
[36] Scott DR, Weeks D, Hong C, Postius S, Melchers K, Sachs G. The role of internal urease in
acid resistance of Helicobacter pylori. Gastroenterology 1998;114:58-70.
[37] Beil W, Kilian P. EPs 7630, an extract from Pelargonium sidoides roots inhibits adherence of
Helicobacter pylori to gastric epithelial cells. Phytomedicine 2007;14:5-8.
[38] Bardonnet PL, Faivre V, Pugh WJ, Piffaretti JC, Falson F. Gastroretentive dosage forms:
overview and special case of Helicobacter pylori. J Control Release 2006;111:1-18.
[39] Pelicic V, Reyrat JM, Sartori L, Pagliaccia C, Rappuoli R, Telford JL, Montecucco C, Papini E.
Helicobacter pylori VacA cytotoxin associated with the bacteria increases epithelial
permeability independently of its vacuolating activity. Microbiology 1999;145:2043-50.
[40] Patel JK, Patel MM. Stomach specific anti-Helicobacter pylori therapy: preparation and
evaluation of amoxicillin-loaded chitosan mucoadhesive microspheres. Curr Drug Deliv 2007;4:
41-50.
[41] Lai CH, Kuo CH, Chen PY, Poon SK, Chang CS, Wang WC. Association of antibiotic
resistance and higher internalization activity in resistant Helicobacter pylori isolates. J
Antimicrob Chemother 2006;57:466-71.
[42] Cooreman MP, Krausgrill P, Hengels KJ. Local gastric and serum amoxicillin concentrations
after different oral application forms. Antimicrob Agents Chemother 1993;37:1506-9.
[43] Atherton JC, Cockayne A, Balsitis M, Kirk GE, Hawkey CJ, Spiller RC. Detection of the
intragastric sites at which Helicobacter pylori evades treatment with amoxicillin and cimetidine.
Gut 1995;36:670-4.
[44] Thanou M, Verhoef JC, Junginger HE. Chitosan and its derivatives as intestinal absorption
enhancers. Adv Drug Deliv Rev 2001;50:S91-101.
[45] Ko JA, Lim HJ, Park HJ. Effect of microencapsulated precipitants of Lactobacillus casei
ATCC 393 on Helicobacter pylori eradication. Process Biochem 2011;46:631-5.
20
Tables
Table 1. Mean particle sizes and zeta potential values for chitosan/heparin nanoemulsion particles
with different ratios of Span 20/Tween 20 mixture surfactant (1.50 wt%) and HLB values in
water–in–oil emulsification system (n = 5). HLB: hydrophilic-lipophilic balance.
Span20:Tween20 ratio HLB numbers
Mean particle size (nm)
Zeta potential values (mV)
50:50
12.7
▲
▲
67:33
11.3
638.9 ± 18.2
28.6 ± 4.1
75:25
10.6
270.1 ± 9.8
32.6 ± 1.6
80:20
10.2
305.3 ± 18.3
35.8 ± 3.1
▲ Precipitation of aggregates was observed.
Table 2. Mean particle sizes and zeta potential values for chitosan/heparin nanoemulsion particles in
with different concentrations of Span 20/Tween 20 mixture surfactant in the water–in–oil
emulsification system (n = 5).
Span 20/Tween 20 mixture
Mean particle size (nm)
Zeta potential values (mV)
0.50%
▲
▲
0.75%
932.6 ± 35.9
38.9 ± 9.8
1.00%
728.8 ± 5.2
36.6 ± 2.5
1.25%
324.2 ± 4.6
34.6 ± 3.6
1.50%
270.1 ± 9.8
32.6 ± 1.6
surfactant concentration (wt%)
▲ Precipitation of aggregates was observed.
21
Figure Captions
Fig. 1. (A) A representation for prepared amoxicillin loaded in chitosan/heparin nanoemulsion
particles and the strategy for eradicating H. pylori using the nanoemulsion particles. (B)
TEM micrographs of the nanoemulsion nanoparticles at different Span 20/Tween 20 mixture
surfactant compositions.
Fig. 2. (A) FT-IR spectra of chitosan/heparin nanoemulsion particles in specific pH environments.
(B) TEM micrographs of the prepared nanoemulsion nanoparticles in specific pH
environments. (C) In vitro release profiles of amoxicillin from amoxicillin-loaded
nanoemulsion particles at different pH values at 37℃ (n = 5).
Fig. 3. (A) Cell viability after treatment with different concentrations of chitosan/heparin
nanoemulsion particles (n = 6). (B) Confocal images of AGS cells incubated with
internalized Cy3-chitosan/FA-heparin nanoemulsion particles for 120 min.
Fig. 4. Fluorescent images (after 3D reconstruction) of AGS cell monolayers infected with
fluorescent H. pylori (DilC18(5)-labeled H. pylori) and incubated with FA-amoxicillin
solution alone or FA-amoxicillin-loaded fluorescent Cy3-chitosan/heparin nanoemulsion
particles for 2 hr.
Fig. 5. Changes in cell growth after AGS cells were pre-infected with H. pyloi for 2 hr and then
incubated with amoxicillin solution alone or amoxicillin-loaded chitosan/heparin
nanoemulsion particles.
Fig. 6. Effects of amoxicillin solution alone and amoxicillin-loaded chitosan/heparin nanoemulsion
particles in a H. pylori-induced gastric infection mouse model
22
23