Arch Pharm Res Vol 34, No 10, 1663-1678, 2011

DOI 10.1007/s12272-011-1011-5

Moxifloxacin-Gelrite In Situ Ophthalmic Gelling System against Pho-

todynamic Therapy for Treatment of Bacterial Corneal Inflammation
Hanan M. El-Laithy1, Demiana I. Nesseem2, Amira A. El-Adly3, and Meriana Shoukry2
1
Department of Pharmaceutics and Industrial Pharmacy, Pharmacy College, Cairo University, Cairo, Egypt, 2Department
of Pharmaceutics, National Organization for Drug Control and Research, Cairo, Egypt, and 3Laser Microbiological Lab.,
National Institute of Laser Enhanced Science (NILES), Cairo University, Cairo, Egypt

(Received April 3, 2011/Revised May 16, 2011/Accepted May 29, 2011)

In this study, six in situ gelling formulations based on Gelrite were prepared and evaluated for
the retained ophthalmic delivery of Moxifloxacin (Mox). The effectiveness of the best developed
formula G5 was compared with photodynamic therapy (PDT), the recent expanding approach
for the treatment of ophthalmologic disorders after the assessment of optimum photodynamic
inactivation parameters that permit efficient pathogens eradication. It was found that, Staphy-
lococcus aureus (S. aureus) (Gram-positive) was more susceptible to effective lethal photosensi-
tization that reaches 93.5% reduction in viable count than Escherichia coli (E. coli) (Gram-
negative) of 76.1% using 3 mg/mL Hematoporphyrin (HP), illuminated by 630 nm Light Emit-
ting Diode (LED) at 9 J/cm2 and incubated for 15 min. Following topical instillation of G5 to
rabbits corneas, higher amount of Mox was retained in the aqueous humor up to 24 h with sig-
nificant 6-fold increase in the Cmax and AUC(0-∞) compared to vigamox® commercial eye drops.
After post corneal infection with S. aureus, both approaches were effectively treating the infec-
tion without causing ocular irritation or collateral damage to corneal tissue where G5 showed
remarkable improvement after four days compared to seven days of PDT treatment.
Key words: Moxifloxacin, Ocular delivery, In situ gelling systems, Gelrite®, Photodynamic
therapy

logical constraints imposed by the protective mecha-

Selected by Editors nisms of the eye such as lachrymal drainage, reflex
blinking and drug dilution by tears resulting in rapid
INTRODUCTION and extensive precorneal loss of drugs (Ali and
Lehmussaari, 2006; Liu et al., 2006; Nanjawade et al.,
The eye is a unique organ, which presents many chal- 2007) and short ocular residence time (Nanjawade et
lenges to develop effective ophthalmic dosage forms al., 2007; Anumolu et al., 2009). This poor ocular bio-
(Qi et al., 2006). Topical application of drug solutions availability implies the necessity of frequent instilla-
(eye drops) is the most popular route of administra- tions to achieve the therapeutic effect, a situation that
tion for the treatment of various eye disorders (Ludwig, is frequently associated with poor patient compliance
2005). These conventional dosage forms account for and risk of undesirable side effects caused by systemic
nearly 90% of the currently available marketed formu- drug absorption (Balasubramaniam and Pandit,
lations owing to their simplicity and good acceptance 2003; Weyenberg et al., 2004; Qi et al., 2007; Kalam et
by patients (Gan et al., 2009). However, the physio- al., 2008). Various ophthalmic vehicles have been
developed to increase the precorneal residence time
but have not been widely accepted such as gels, oint-
Correspondence to: Demiana I. Nesseem, Department of Pharma-
ceutics, National Organization for Drug Control and Research
ments and polymeric inserts due to their drawbacks,
(NODCAR), Cairo, Egypt such as blurred vision, sticky sensation, reflex blink-
Tel: 202-5851299 / 202-5851278, Fax: 202-35855582 ing, patient discomfort and lack of compliance (Ludwig,
E-mail: [email protected] 2005; Liu et al., 2006; Kalam et al., 2008; Ma et al.,

1663
1664 H. M. El-Laithy et al.

2008). formulation Blocarden® Depot (Timoptic®) by Merck

From the point of view of patient acceptability, a (Nanjawade et al., 2007).
dosage form that: (1) can be administered as a liquid On the other hand, photodynamic therapy (PDT) is
dosage form, (2) creates no irritation or vision pro- a rapidly expanding approach for the treatment of
blems, (3) has suitable strength to endure tear fluid several diseases such as cancer, dermatologic and
dilution without rapid precorneal elimination, (4) has ophthalmologic disorders (Demidova et al., 2005; Dai
suitable mucoadhesive force to improve the drug re- et al., 2010). To date the most successful application of
tention in the precorneal area, (5) dosed once or twice PDT that recently received FDA approval has been in
daily, is an ideal ophthalmic formulation (Kao et al., ophthalmology for the treatment of age-related macu-
2006; Qi et al., 2007). This can be achieved by in situ lar degeneration (O’Riordan et al., 2005). Very Recently,
gel-forming (phase transition) systems which can be PDT approach has shown effectiveness as antimicro-
instilled as a liquid dosage form and upon exposure to bial agent in the treatment of localized infections as
physiological conditions (pH of the tear fluid, tempera- wound infections (Wong et al., 2005; Zolfaghari et al.,
ture at the eye surface or electrolytes present in the 2009), burn infections (Lambrechts et al., 2005), per-
tear film) shifts to the gel phase (Srividya et al., 2001; iodontal infections (Kömerik et al., 2003; de Almeida
Sultana et al., 2006). et al., 2008; Fernandes et al., 2009), Helicobacter pylori
Gelrite® (deacetylated gellan gum) is an ion-sensi- gastric infection (Ganz et al., 2005; Lembo et al.,
tive, linear, anionic heteropolysaccharide polymer pro- 2009), acne vulgaris (Hongcharu et al., 2000; Wiegell
duced by the bacterium Sphingomonas elodea under and Wulf, 2006) and oral candidiasis (Teichert et al.,
aerobic fermentation. The polymer composed of tetra- 2002).
saccharide repeating unit linked together which con- The PDT mechanism is based on the use of light
sists of glucose, glucuronic acid, and rhamnose in the sensitive non-toxic dyes called photosensitizers (PS),
molar ratio 2:1:1 (Bajaj et al., 2007; Nanjawade et al., of which anionic Hematoporphyrin (HP) is best known
2007). example, in combination with low intensity visible
Ophthalmic formulations of Gelrite® can be admin- light of appropriate wavelength to excite the PS
istered as a liquid form and transformed to clear gel (Castano et al., 2004; Hamblin and Hasan, 2004). Ab-
on contact with the ions in tear fluid (Kubo et al., sorption of light elevates the PS from the ground
2003). In addition Gelrite® exhibits interesting physico- (singlet) to the first excited (singlet) state followed by
chemical characteristics as bioadhesion, non-irritant, conversion to the triplet state via intersystem crossing.
excellent ocular tolerance, residence time improvement, The longer lifetime of the triplet state enables the
increase ocular bioavailability, absence of abnormal interaction of the excited PS with the surrounding
clinical signs to the cornea, iris, or conjunctivae (Liu et molecules, by one or both of two pathways. (Sharman
al., 2010). Previous studies have shown that Gelrite® et al., 1999; Demidova et al., 2005). Type I reaction
formulations were better retained in human eyes for leads to generation of radicals through interaction with
20 h (Carlfors et al., 1998) with favorable rheological a solvent or substrate via electron or proton transfer
behavior where, its solutions have shown pseudoplas- that can damage the target cells either directly or
tic and viscoelastic properties (Deasy and Quigley, after interaction with molecular oxygen to produce
1991; Nanjawade et al., 2007). This behavior is parti- highly reactive oxygen species. While type II generates
cularly important in ophthalmic formulations since it cytotoxic singlet oxygen by direct energy transfer to
facilitates the retention while it permit the easy molecular oxygen. These reactive species from both
spreading of the formulation due to the blinking of the mechanisms can lead to efficient killing of target cells
eye (Badawi et al., 2008). through interaction with biological molecules as pro-
Gelrite® was granted regulatory approval as pharma- teins, lipids and nucleic acids (Demidova et al., 2005;
ceutical excipient to improve the bioavailability of many Donnelly et al., 2008; Ishikawa et al., 2010).
drugs such as methylprednisolone, timolol and carbonic Various light sources have been researched in order
anhydrase inhibitors (Rozier et al., 1989; Bourlais et to be used as alternative to the high cost laser sources,
al., 1998; Ludwig, 2005; Sultana et al., 2006). The the most interesting of them is the light emitting
commercial product Timoptol XE® containing Gelrite® diode (LED). LED has several advantages of low price,
showed a sufficient residence time increase after a capability for illuminating large areas, less hazardous,
once-daily topical instillation for the reduction in the thermally non-destructive, readily available, and easy
intraocular pressure when compared to conventional to operate and occupy minimum space. Moreover, it is
timolol maleate eye drops (Ludwig, 2005). Moreover, possible to find different colors of LED light in the
Gelrite® is marketed in a controlled release glaucoma market, with radiations covering almost all of the
Moxifloxacin-Gelrite In Situ Ophthalmic Gelling System 1665

visible electromagnetic spectrum, including red light the effect of the developed formula on the healing rate
(630-650 nm region) (Peloi et al., 2008; Neupane et al., of rabbit eyes induced S. aureus keratitis relative to
2010). PDT.
PDT has the advantage over other therapies of dual
selectivity: not only is the PS targeted to its destin- MATERIALS AND METHODS
ation cell or tissue, but the light can also be spatially
delivered to the affected tissue. Moreover, bacteria Materials
will not be easily able to develop resistance (Hamblin Moxifloxacin hydrochloride (Mox), was kindly supplied
and Hasan, 2004; Demidova et al., 2005). by EVA pharma company. Gelrite® was purchased
Moxifloxacin (Mox) a fourth generation fluoroqui- from Sigma Chemical Co. Methylcellulose E461 (MC)
nolone with diazabicyclononyl ring in the C7 position was supplied by Carl Roth GmbH +. Mucin from
and a methoxy group in the C8 position was used in porcine stomach, Type III (bound sialic acid 0.5-1.5%,
the present study. It is commercially available as a partially purified powder) was purchased from Sigma
0.5% solution for topical ocular use (Vigamox®) with Chemical Co. Acetonitrile HPLC grade was supplied
high potency against both Gram-positive and Gram- by S.D. Fine Chem.-Limited. Vigamox® eye drops (con-
negative bacterial pathogens and lower risk for de- tains: active: Moxifloxacin hydrochloride 0.5% and
veloping resistance (Kim et al., 2005; Stroman et al., inactives: sodium chloride, boric acid and purified
2005). Its bactericidal activity is through inhibi- water) (Alcon Laboratories, Inc.). Spectra/Por® 3 dialy-
tion of bacterial topoisomerase II (DNA gyrase) and sis membrane (Cellophane membrane of MWCO: 3500
topoisomerase IV enzymes which are critical in the Daltons) was obtained from Spectrum Laboratories
maintenance, synthesis, and replication of DNA (Levine Inc. Escherichia coli (E. coli) ATCC® 10536™ and Sta-
et al., 2004; Torkildsen and O'Brien, 2008). Mox has a phylococcus aureus (S. aureus) ATCC® 29737™ were
unique structure, its amphoteric nature (with pKa1 of supplied by LyfoCults® Plus, PML Microbiologicals, Inc.,
6.25 for the carboxyl group in the C3 position and BioMerieux, Inc. Hematoporphyrin (HP) was purchas-
pKa2 of 9.29 for diazabicyclononyl in the C7 position) ed from porphyrin products, Fluka. Hematoxylin and
permits the drug to overcome both the hydrophilic/ Eosin (H & E) stain was provided by Newcomer supply
lipophilic corneal structures (Kaur et al., 2004; Langlois Inc. All other chemicals and solvents were of reagent
et al., 2005; Robertson et al., 2005). At a neutral tears grade and were obtained from standard commercial
pH, Mox exists in an unionized form resulting in an suppliers.
increased corneal permeation and higher concentra-
tion in the aqueous humor at least 2 to 3 times that of Preparation of Mox in situ gelling systems
other fluoroquinolones (Kim et al., 2005; Solomon et The Composition of Gelrite based formulations pre-
al., 2005; Pawar and Majumdar, 2006; Rathore and pared in this study is shown in (Table I). The drug was
Majumdar, 2006). dissolved in distilled water to yield a final concentra-
Based on these considerations and keeping in mind tion of 0.5% w/v. The calculated amount of MC was
that enhancement of the precorneal processes such as dispersed in the drug solution and stirred until dis-
the slow removal of Mox from the absorption site and solved. Different concentration of Gelrite (0.4, 0.6, 0.8
enduring the dilution by tear fluid would be of great % w/v) were dispersed in drug solution and heated to
benefit to reduce the dose frequent instillation and 90oC under stirring for 20 min and allowed to cool at
consequently improve patient compliance. Thus the room temperature while stirring. The pH of all poly-
principle aim of this work was i) to formulate and mer solutions was adjusted to 7.4 using 0.1 N NaOH.
evaluate different Mox based Gelrite in situ gelling All formulations were stored in a refrigerator (4-8oC)
system benefiting from the commercial success of until further use.
marketed Timoptic® and Timoptol XE® to increase
residence time of Mox in the precorneal area; ii) to Evaluation of formulations
assess the potential of the developed system to pro- In order to identify the compositions suitable for use
mote corneal penetration and hence therapeutic index as in situ gelling drug delivery vehicles, the developed
by measuring Mox levels in rabbits aqueous humor systems were evaluated for drug content uniformity,
relative to commercial Mox eye drops; iii) to investi- clarity by visual observation against a black and white
gate the optimum photodynamic inactivation para- background in a well lit cabinet, gelling capacity, flow
meters (HP concentration, light dose and incubation behavior, rheological study, in vitro drug release and
period) that permit highest reduction in bacterial mucoadhesion evaluation.
species (S. aureus and E. coli). Finally, iv) to compare
1666 H. M. El-Laithy et al.

Drug content uniformity molecular weight cut off soaked overnight in freshly
100 µL of each system was diluted in a test tube to prepared STF) from one end and attached to the shafts
10 mL with simulated tear fluid (STF) composed of: of the USP dissolution tester apparatus, instead of the
sodium chloride, 0.670 g; sodium bicarbonate, 0.200 g; baskets, from the other end. The shafts were then
calcium chloride·2H2O, 0.008 g and purified water up lowered to the vessels of the dissolution apparatus
to 100 g (Lin and Sung, 2000). The tube was then containing 100 mL of STF. The release study was
covered with a Parafilm® and sonicated for 2 min. A carried out at 35 ± 1oC, and the stirring shafts were
clear solution was obtained, from which 1 mL was rotated at a speed of 50 rpm. At predetermined time
taken and suitably diluted. The concentration of Mox intervals, samples (3 mL) were withdrawn and analyzed
was determined spectrophotometrically (Shimadzu for Mox content spectrophotometrically at λ 288 nm
UV-1601 Double Beam) at λmax 288 nm (Anumolu et against the sample withdrawn at respective time in-
al., 2009) against the sample withdrawn from plain Mox terval from plain Mox free system treated in a similar
free system treated in a similar manner. The method manner. Every withdrawal was followed by replace-
was validated, the accuracy, repeatability (intra-day ment with fresh medium to maintain a constant volume.
and intermediate precision (inter-day) and reliability The results were the mean value of 3 runs each repre-
were ensured. The recovery% was > 98%. The drug senting one batch.
content was determined relative to the original drug
concentration according to the following equation: Mucoadhesion measurement
The mucoadhesive behavior was evaluated accord-
Drug content uniformity %
ing to the method described by (Hassan and Gallo,
= (Practical drug content/Theoretical drug content)
1990) based on the idea that the chemical interaction
×100
and entanglements between the polymer and gly-
The results were the mean value of 3 replicates. coproteins in mucus causes a rheological synergism.
Dried mucin was hydrated with STF by stirring for 3
Gelling capacity h at room temperature to yield a dispersion of 20% (w/
The gelling capacity was determined by placing 100 w). Six grams of the latter mucin dispersion were
µL of the prepared formulations into a vial containing mixed for 15 min with 2 g of G5 formula such that the
2 mL of STF freshly prepared. Gelation was assessed final concentration of mucin was 15% (w/w). The visco-
visually and noting the time for gelation and the time sity of mucin (15% w/w) alone (ηm), G5 formula (diluted
taken for the gel formed to dissolve (Srividya et al., with STF in the ratio 40:7) (ηp) and their mixture (ηt)
2001; Qi et al., 2007). was measured in order to evaluate the mucoadhesion
property of the tested formula. Viscosity was measured
Viscosity and rheological studies at 35 ± 1oC at the shear rates D of 10, 20, 50, and 100
The viscosity and rheological behavior of the pre- sec−1. The measurements were performed for 1 min
pared systems in centipoise (cp) was measured at after 3 min of applying the shear force to allow the
various shear rates at 35 ± 1oC using Brookfield DV shear force to be homogeneously distributed throughout
III viscometer fitted with CP-52 cone and plate the sample. The viscosity component of bioadhesion (ηb)
spindle (Brookfield Engineering Inc.). In order to was calculated from the following equation,
evaluate the change in viscosity after administration
ηb = η t − (ηm + ηp) (1)
and mixing with the tear fluid, the measurements
were repeated after diluting each system with STF in The mucoadhesion index M (cp) was calculated using
a ratio of 40:7 based on conventional commercial the shear rate D (1/sec) and the viscosity component
eyedropper delivers an average volume of about 40 µL due to bioadhesion (ηb) (cp) according to the equation:
while available tear fluid is 7 µL (Wei et al., 2002; Qi
M = ηb*D (2)
et al., 2007). The results were the average of three
experiments ± S.D. Where D is the shear rate per second. (ηb) was cal-
culated from Eq. (1). Since (ηb) may decrease with the
In vitro drug release increase in the applied shear rate D, it was decided to
This study was carried out using a USP Dissolution use a high value of D to eliminate weakly bioadhesive
tester (Apparatus I, Hanson SR6). A 1-mL volume of materials (Hassan and Gallo, 1990).
the formulation was accurately placed in glass cylin-
drical tubes (2.5 cm in diameter and 10 cm in length). Measurement of osmotic pressure
Each tube is tightly covered with a Spectra por© (3500 The osmotic pressure of G5 formula as well as HP
Moxifloxacin-Gelrite In Situ Ophthalmic Gelling System 1667

dye solution was determined by the freezing point p.s.i. for 20 min. The animals were randomly divided
depression method using 3D3 Single-Sample Osmom- into six groups, each of six rabbits. The study was
eter (Advanced Instruments) on a 0.25-mL aliquot of conducted on two phases. In phase I, two groups of
the sample. The apparatus was calibrated using stand- animals were used to evaluate the concentration of
ard reference solutions (100-3000 mOsmol). Mox in aqueous humor following instillation of the
selected in situ gelling formula G5 compared to the
Photodynamic inactivation (PDI) commercial Vigamox® eye drops which used as refer-
Bacterial species were grown aerobically overnight ence.
in 10 mL of brain heart infusion broth (Oxoid Ltd) at The rabbits received a single dose (40 µL) of the
37oC. After 18 h the stationary phase microorganisms tested preparation applied in the cul-de-sac of the
were harvested by centrifugation at 3000 rpm for 10 right eyes. During the experiment, the rabbits were
min, washed twice with phosphate buffered saline maintained in an upright position by placing in
(PBS; pH 7.4) and suspended in 2 mL PBS. Stock restraining boxes where they could move their heads
solution of HP was prepared by dissolving in minimal and eyes freely. At different times post-instillation
amount of 0.1 N NaOH then complete the volume (0.25, 0.5, 1, 4, 8, 12, 24 h) the animals were anesthetiz-
with saline and kept at 2-4oC in the dark. Different ed with intramuscular injections of ketamine hydrochlo-
concentrations (0.5, 1, 2, and 3 mg/mL) were prepared ride 50 mg/kg, Xylazine 10 mg/kg (Yoon et al., 2006),
for further experiments. and ~200 µL aqueous humor was withdrawn with the
Aliquots (100 µL) of bacterial suspension in saline help of 26-gauge needle attached to 1 mL disposable
were transferred into wells of a 96 well flat-bottomed syringe inserted through the corneal-scleral junction
microtitier plate (Sterilin Ltd.) and dark incubated for and slightly upwards into the anterior chamber (El-
various time (5-45 min) with 100 µL HP of different Kamel, 2002). The samples were collected in glass
concentrations. At the end of each incubation time, tubes and stored at −80oC until the spectrofluorimetry
each well was exposed to (0.9-9 J/cm2) red light emitted assay was carried out. The degree of drug penetration
from LED system manufactured by photon scientific is expressed as the maximum Mox ocular concentra-
with spectrum emission ranging from 630 nm. tion measured in µg per mL aqueous humor.
Control group of cells were prepared to determine In phase II, bacterial corneal inflammation experi-
the number of viable bacteria prior to treatment with ment was studied on the other four groups of rabbits
neither light nor the sensitizer (L−P−). Besides, other using S. aureus, the most common pathogen causing
cell groups were prepared to observe the effect of: light eye infection (Constantinou et al., 2007). First group
alone, no sensitizer being added (L+P−) and photosen- was served as negative control (not infected). The
sitizer alone with no exposure to light (L−P+). At the other three groups were infected with S. aureus where
end of the treatments, 100 µL aliquots of the cell second group served as positive control (infected not
suspensions were plated from each well onto nutrient treated), third group was treated with G5 formula and
agar plate and allowed to grow for 24 h at 37oC. The fourth group was treated with PDT. S. aureus was
number of colonies were determined by direct plate grown aerobically overnight in 10 mL of brain heart
enumeration on the plates and expressed as colony infusion broth (Oxoid Ltd) at 37oC. After 18 h, the
forming units/mL (CFU/mL). The data collected were stationary phase microorganisms were harvested by
means of 3 duplicate experiment for each studied centrifugation at 3000 rpm for 10 min, washed twice
parameter (n = 6). with phosphate buffered saline (PBS; pH 7.4) and
suspended in 2 mL PBS. 100 µL of the S. aureus sus-
In vivo studies pension (containing approximately 107-108 CFU/mL)
36 male healthy New Zeland Albino rabbits weighing was inoculated in rabbit eye just after the corneal
between 2.0-2.5 kg were used in this study because epithelium has been scratched with a 26-gauge needle.
rabbit eyes simulate adult human eyes with respect to Treatment was initiated 24 h later. For third group,
size, shape, physiology and composition of tears (Charoo one drop of G5 formula was instilled once daily. For
et al., 2003). The study performed was approved by fourth group, treatment was done by instillation of
the university protection for animals care and use 100 µL of 3 mg/mL HP then after 15 min incubation
committee and the protocol complied with “the Prin- period, light was irradiated perpendicular to the cor-
ciples of Laboratory Animal Care” [NIH publication # nea for 10 min at 9 J/cm2. The cornea was protected
85-23, revised 1985]. The candidate formula G5 used by topical application of physiological saline during
was freshly prepared without any preservative addi- light irradiation.
tion and was sterilized by autoclaving at 121oC and 15 Eye examination was performed every day for signs
1668 H. M. El-Laithy et al.

of infection (mucoid discharge) and inflammation (red- batch precision and accuracy of the analytical pro-
ness and lacrimation). Moreover, fluorescein sodium cedure were evaluated after replicate analysis (n = 9)
(0.2%) solution was used to ascertain inflammation of control aqueous humor samples spiked at three
visually. Since, fluorescein does not stain tissue unless concentration levels for each standard calibration
the epithelium is disrupted; the inflammatory areas curve. The lower limit of quantification was 12.23 ng/
develop fluorescence. The absence of color was thus mL and 324.65 ng/mL for high and low sensitivity
taken as a criterion of healing. Thereafter, the eyes standard curves, respectively. With a linear response
were photographed using digital camera in order to across the full range of concentrations from 17 to 250
find out signs of improvement. ng/mL (R2 = 0.9998) and from 333 to 1666 ng/mL (R2
In order to confirm the evaluation of corneal healing = 0.9999).
performed by visual inspection, histological examin-
ation was done. The histological changes were evalu- Aqueous humor analysis
ated for the treated eyes in third and fourth groups, The samples were thawed at room temperature and
where almost no fluorescence could be detected. 150 µL of each sample was extracted with acetonitrile
Histological examination for negative and positive in a ratio 1:5 and centrifuged at 5000 rpm for 30 min
control (first and second groups) was also performed at 4oC. 300 µL from the supernatant was taken and
for comparison and evaluation as well. The test was evaporated to dryness. Residues were reconstituted
done under deep anesthesia as previously mentioned. with 1 mL 0.1 M H2SO4 and its fluorescence intensity
The eyes were enucleated. The corneas were dissected measured against plain Mox free aqueous humor
along the limbus with scissors. The corneas were treated in a similar manner.
placed in 10% formalin. Following fixation they were
embedded in paraffin and stained with Hematoxylin Pharmacokinetic analysis
and Eosin for routine histological examination. The area under the curve, AUC(0-∞) (µg·h/mL) of Mox
Specimens showed different patterns of cellular acti- concentration in the aqueous humor were calculated
vity through the examination by light microscopy. using the trapezoidal rule. The maximum Mox con-
centration Cmax (µg/mL) in the aqueous humor and the
Spectrofluorimetric analysis time at which Cmax is achieved Tmax (h) were determined
The Mox contents in aqueous humor were measured from actual data points.
using spectrofluorimetric method depending on the
native fluorescence of fluoroquinolones due to the high Statistical analysis
degree of conjugation found in the structure (Salem, The data obtained from different formulations were
2005). Fluorescence measurements were performed analyzed for statistical significance by one-way analy-
with a Shimadzu spectrofluorimeter Model RF-1501 sis of variance (ANOVA) adopting SPSS statistics
equipped with a Xenon lamp. All the measurements program (version 16, SPSS Inc.) followed by post hoc
took place in a standard 10 mm path length quartz multiple comparisons using the least square differ-
cell. The fluorescence intensity of Mox was measured ence (LSD). Differences between series were considered
at 520 nm using an excitation wavelength of 293 nm to be significant at p ≤ 0.05.
in the presence of 0.1 M H2SO4 because at basic pH
the fluorescence was inhibited whereas, at acidic pH RESULTS AND DISCUSSION
the fluorescence was enhanced (Ocaña et al., 2000;
Razek et al., 2008). Evaluation of formulations
The composition of various Gelrite® based formula-
Standard solutions tions is shown in Table I. The drug content as well as
Blank aqueous humor samples were spiked with the clarity was found to be satisfactory for all formu-
Mox stock solution (1 mg/mL) to give the range of 17- lations.
1666 ng/mL. To cover the fluorescence intensities of The feasibility of the in situ gelling system as an
all samples measured, two calibration curves were ocular drug delivery should be a free flowing liquid
constructed by plotting the fluorescence intensity with low viscosity at non-physiological condition to
versus Mox concentrations in aqueous humor. The allow reproducible administration into the eye as
standard solutions were in the range of 17-250 ng/mL drops; it should also undergo in situ sol-to-gel phase
and 333-1666 ng/mL measured at high and low sensi- transition at physiological condition to form gel
tivity instrument condition respectively. capable of enduring shear forces expected in the eye
During the assay of the samples, the intra and inter- during and between blinking and facilitate sustained
Moxifloxacin-Gelrite In Situ Ophthalmic Gelling System 1669

Table I. Composition of the prepared Gelrite formulations

Formulation Concentration (% w/v) Drug content Gelling
Clarity Viscosity (cp)
code Gelrite MC (% w/v)a capacity
G1 0.4 101.5 ± 2.53 Clear +
G2 0.6 199.9 ± 2.64 Clear ++
G3 0.8 100.6 ± 1.81 Clear +++
G4 0.6 0.5 199.4 ± 2.87 Clear 144.6*
G5 0.6 1 100.3 ± 2.62 Clear 158.8**
G6 0.6 2 198.6 ± 1.52 Clear 1426***
a
Each value represents the mean ± S.D. of three experiments.
+: gel formed after few minutes and dissolves rapidly, ++: immediate gelation which remains for few hours, +++:
immediate gelation that remains for extended period.
*liquid, flow easy, **liquid gel like, flow easy, ***gel formed, flow difficult.

drug release (Lin and Sung, 2000; Kalam et al., 2008; preparations were either liquid with very low viscosity
Al-Kassas and El-Khatib, 2009; Liu et al., 2010). or formed gel at non-physiological condition (during
Therefore, the gelling capacity, viscosity, in vitro re- preparation), which causes blurred vision and was not
lease study and the mucoadhesive properties of the suitable for dropping. Therefore, formula G5 consists
prepared Mox ophthalmic formulations were evaluated. of 0.6% Gelrite and 1% MC presents a satisfactory
attributes of gelling capacity and consistency.
Gelling capacity
Aqueous solution of Gelrite® gels in the presence of Viscosity studies
monovalent or divalent cations. The mechanism of The viscosity of Gelrite® formulations before and
gelation involves the formation of double helical after dilution with STF was depicted in Fig. 1. All
junction zones followed by aggregation of double-
helical segments to form a three dimensional network
by complexation with cations and hydrogen bonding
with water (Kubo et al., 2003; Liu et al., 2010).
As the two main prerequisites of a phase transition
system are viscosity and gelling capacity (speed and
extent of gelation), from visual and manual inspection
we found that all formulae coded in Table I formed gel
that commences immediately upon contact with STF,
where the ionic interaction takes place. However, it is
clear that the nature of the gel formed depended upon
the polymer concentration (Balasubramaniam and
Pandit, 2003). As shown, G1 containing the minimal
amount of Gelrite® (0.4%) showed the weakest gelation
under physiological condition and the gel formed can
be easily destroyed by shearing. On the other hand,
increasing Gelrite® concentration to 0.8% (w/v) in G2,
resulting in a formation of a stiff gel under physio-
logical condition. Intermediate Gelrite® concentration
of 0.6% (w/v) showed the preferred gelling capacity
(Rozier et al., 1989; Sanzgiri et al., 1993; Sultana et
al., 2006).
In a trial to improve the flow behavior of the formu-
lation and to minimize its leakage from the eye during
instillation, a combination systems (G4 - G6) of 0.6% Fig. 1. Rheology profiles of different formulations (A) before
Gelrite® with various concentration of MC were tested. and (B) after dilution with simulated tear fluid (STF).
As depicted from results in Table I, neither 0.5% nor Results represent the mean of three experiments. S.D. was
2% MC added were suitable concentration as their always less than 10%.
1670 H. M. El-Laithy et al.

formulations transformed into gels with high viscosity would be beneficial as it would help achieve higher
after dilution with STF. This confirms the sol-gel drug concentration in a minimum time, and the slow
phase transition process. It was clear also that the steady state drug release later on would provide a
viscosity was directly dependent on the polymeric con- sustained release of the drug (Sultana et al., 2006).
tent of the formula (G3 > G2 > G1) which was further The incorporation of 1 and 2% MC to Gelrite® modi-
increased upon combining with MC (G6 > G5 > G4). fied the drug liberation profile. Significant decrease (p
Moreover, all formulations exhibited pseudoplastic < 0.05) in the percentage drug released were achieved
property as evidenced by shear thinning and the de- after 1 and 48 h. Similar results were obtained by (El-
crease in viscosity with the increase in angular velocity. Kamel, 2002), who reported that the addition of cellu-
This property is in favor of sustaining drainage of lose additives to polymer solution significantly pro-
drugs from the conjunctival sac of the eye as without longed Timolol maleate release.
blinking there will be a difficulty for undergoing shear The release data was kinetically analyzed using the
thinning (Wu et al., 2007; Liu et al., 2010). In addition empirical equation: Log Q = Log k + n Log t (Ritger
a desirable characteristic of the selected system to be and Peppas, 1987) where, Q is the fraction of drug
used in ophthalmic preparation was addressed by released in time t, k is a constant characteristic of the
(Srividya et al., 2001; Balasubramaniam and Pandit, drug polymer interaction and n is an empirical para-
2003; Liu et al., 2006) in that since the range of ocular meter characterizing the release mechanism. Based
shear rates associated with normal blinking is ex- on the diffusional exponent, the Gelrite® based formu-
tremely wide, ranging from 0.03 S−1 during inter- lations revealed n-values of (1 > n > 0.5), meaning
blinking periods to 4250-28500 S−1 during blinking, non-Fickian or anomalous behavior where the rate of
therefore, viscoelastic fluids with a viscosity that is diffusion of drug and polymer relaxation are compar-
high under the conditions of low shear rate and low able. Drug release was dependent on two simultan-
under the conditions of high shear rate are preferred. eous rate processes, water migration into the hydrogel
and drug diffusion through continuously swelling
In vitro release hydrogel (Singh et al., 2007; Badawi et al., 2008;
The release profile of a drug predicts how a delivery Anumolu et al., 2009).
system might give valuable insight into its in vivo It was worthy that the obtained in vitro release
behaviour. The general feature of Mox release from results correlated well with the results of the gelling
Gelrite-based formulations revealed the ability of the capacity and the rheological studies. Therefore, based
developed systems to retain the drug in its matrix on the previous results, the combination that better fit
network (Fig. 2). It was apparent that the release the requirements for an acceptable ophthalmic delivery
profiles obtained exhibited an initial burst release due system was G5. For these reasons, G5 was chosen for
to migration of the drug toward the surface of the in vitro mucoadhesion determination.
matrix and amounted to be 28.76%, 25.13%, 24.92%,
and 21.32% from G2, G4, G5, and G6, respectively Mucoadhesive studies
within the first hour followed by slow steady state The mucoadhesion property has a dominant role as
drug release reaching 76.88%, 71.08%, 67.84%, and the retention time of the formulation in the ocular
60.53% in 48 h, respectively. This initial rapid release area may be improved. The viscosity as a function of
shear rate of mucin dispersion, G5 formula and their

Fig. 2. Release profiles of Moxifloxacin from Gelrite® in situ- Fig. 3. Flow curves of G5 formula, M (15% w/w mucin
gelling system at 35 ± 1oC (mean ± S.D., n = 3) dispersion), and their mixture (G5 + M) (mean ± S.D., n = 3)
Moxifloxacin-Gelrite In Situ Ophthalmic Gelling System 1671

Table II. The viscosity (ηt, ηp, ηm and ηb) at various rate et al., 2007). The above finding was substantiated by
of shear the high molecular weight of Gelrite® and the presence
Rate of shear Viscosity (cp) of carboxyl and hydroxyl functional groups that
(1/sec) ηt ηp ηm ηb enable the formation of hydrogen bonds with mucin
molecules. Moreover, the presence of MC as a neutral
110 1703.11 1130 211.67 361.43
polymer allows polymer swelling in water and helps
120 1197.33 1704 17211. 321.33
entanglement of the polymer chains with the mucin
150 1776.53 1422 144.21 210.33
molecules. All these contribute to the formation of a
100 1553.47 1280 109.11 164.37
strengthened network (Andrews et al., 2009; Roy et
Results represent the mean of three experiments. S.D. was al., 2009; Sensoy et al., 2009).
always less than 10%.
ηt: the viscosity of G5 formula and mucin mixture; ηp: the
viscosity of G5 formula (diluted with STF in the ratio 40:7); Osmolality
ηm: the viscosity of mucin (15% w/w); ηb: the viscosity com- The osmolalities of G5 formula and HP photosensi-
ponent of bioadhesion. tizer, were 280 and 292 mOsmol/kg, respectively,
which were very close to the physiological range of
mixture is shown in Fig. 3. It was revealed from Table the lachrymal fluid osmolality (280-293 mOsmol/kg)
II that, the presence of mucin in G5 strongly affected (Vandamme and Brobeck, 2005). Since hypoosmolar
its flow behaviour and the viscosity response of G5/ and hyperosmolar can possibly cause eye pain, inflam-
mucin mixture is higher than the sum of the contri- mation, pathologic changes and harm (Li et al., 2004;
bution from separate components at all shear rates Schrage et al., 2004; Luo et al., 2005), therefore, this
investigated. The mucoadhesion index, (M), calculated parameter is considered important as it is directly
at D = 100 was equal to 16.44 (Pa.s). These results related to ocular comfort (Stahl et al., 2009) and
suggest positive interaction (synergism) between G5 consequently, the two preparations would not induce
and mucin which insure prolong ocular residence time a feeling of irritation or discomfort.
of the selected formula as a result of its binding to the
mucus layer coating the corneal and conjunctival epi- Photodynamic inactivation (PDI)
thelium (Vandamme and Brobeck, 2005; Nanjawade The present study demonstrated the photodynamic

Table III. The photodynamic effect on the percent reduction of S. aureus and E. coli at variable radiant exposure using
four different HP doses employing different incubation times (mean ± S.D., n = 6).
S. aureus E. coli
Incubat
HP Light fluence (J/cm²) Light fluence (J/cm²)
ion
(mg/
time 0.9 2.7 4.5 9 0.9 2.7 4.5 9
mL)
(min) a a
% Reduction (CFU/mL) % Reduction (CFU/mL)
05 02.98 ± 2.13 07.29 ± 0.54 11.23 ± 0.80 14.04 ± 3.04 01.53 ± 1.05 08.15 ± 1.10 11.39 ± 3.04 15.27 ± 5.74
15 07.54 ± 2.37 10.94 ± 0.56 12.40 ± 2.16 24.74 ± 4.31 03.36 ± 1.88 10.07 ± 1.45 10.90 ± 1.16 19.01 ± 7.53
0.5
30 12.98 ± 1.85 16.53 ± 1.65 18.07 ± 1.61 30.70 ± 1.52 08.50 ± 2.54 13.71 ± 3.68 15.05 ± 5.33 23.90 ± 1.23
45 16.32 ± 2.11 20.67 ± 2.15 28.95 ± 2.63 38.42 ± 3.20 14.48 ± 7.36 20.22 ± 0.89 25.87 ± 1.68 32.34 ± 2.89
05 31.58 ± 1.05 36.14 ± 1.22 46.67 ± 1.69 58.95 ± 1.39 26.92 ± 5.94 28.45 ± 4.98 33.34 ± 3.32 36.87 ± 4.39
15 38.77 ± 4.09 39.82 ± 1.61 55.10 ± 1.62 60.70 ± 5.00 29.73 ± 6.69 31.86 ± 4.69 35.20 ± 1.94 44.98 ± 1.18
1
30 43.86 ± 1.52 55.44 ± 2.37 66.32 ± 0.91 71.40 ± 2.13 33.44 ± 8.02 35.95 ± 8.11 38.87 ± 5.20 46.84 ± 2.22
45 46.32 ± 3.68 60.18 ± 1.10 71.23 ± 2.37 76.32 ± 2.41 38.14 ± 6.20 42.65 ± 7.34 44.92 ± 6.33 47.51 ± 3.27
05 64.91 ± 7.6 72.46 ± 1.32 75.09 ± 5.8 78.25 ± 3.99 45.67 ± 8.47 47.06 ± 6.33 48.63 ± 6.82 55.24 ± 4.81
15 67.72 ± 3.73 74.74 ± 1.39 79.5 ± 1.08 80.70 ± 3.04 46.10 ± 2.32 55.60 ± 6.01 57.80 ± 5.04 64.30 ± 8.52
2
30 72.63 ± 1.39 77.19 ± 1.61 89.30 ± 2.19 90.18 ± 5.27 52.48 ± 9.60 53.12 ± 7.24 58.21 ± 7.34 65.90 ± 7.26
45 80.00 ± 3.97 81.58 ± 2.41 91.05 ± 1.39 92.28 ± 4.72 55.83 ± 7.46 60.74 ± 7.79 63.28 ± 7.01 67.37 ± 6.11
05 80.18 ± 3.38 82.98 ± 1.32 87.37 ± 0.91 89.20 ± 1.62 59.36 ± 7.93 62.86 ± 8.83 66.12 ± 5.52 69.72 ± 2.02
15 84.74 ± 2.79 87.19 ± 2.60 89.20 ± 2.16 93.50 ± 0.54 62.47 ± 8.52 63.30 ± 9.74 71.30 ± 2.33 76.10 ± 0.59
3
30 89.82 ± 4.86 91.93 ± 4.48 92.63 ± 0.53 9400. ± 1.08 63.90 ± 7.73 64.86 ± 9.03 73.14 ± 3.40 76.20 ± 0.22
45 91.75 ± 2.13 92.11 ± 2.63 92.81 ± 0.80 93.80 ± 0.83 64.41 ± 9.17 67.85 ± 8.57 74.82 ± 3.83 76.70 ± 0.39
a
colony forming units/mL (CFU/mL)
1672 H. M. El-Laithy et al.

inactivation (PDI) on S. aureus as an example of positive and Gram-negative bacteria to the photoinac-
Gram-positive bacteria and E. coli as an example of tivation could be explained on the basis of the differ-
Gram-negative bacteria by a combination of HP and ences in their cell wall structures (Omar et al., 2008).
LED red light. The photodynamic effect on the percent Gram-positive bacteria possesses a cytoplasmic mem-
reduction of S. aureus and E. coli at variable radiant brane coated with a relatively porous layer of pepti-
exposure using four different dye doses employing doglycan and lipoteichoic acid; therefore the most
different incubation times have been studied and commonly used PSs can bind and penetrate its cell
summarized in Table III. The results demonstrated wall. While, in contrast, the outer wall of Gram-negative
that, PDI of bacterial species increased as a function bacteria consists of an outer membrane and an inner
of PS concentration, light dosimetry and incubation cytoplasmic membrane that is separated by a layer of
time. The percentage reduction in viability of bacterial peptidoglycan. This structure renders its surface highly
species exposed to light in presence of photosensitizer negatively charged and more effective permeability
(L+P+) was higher compared to the control group barrier that tends to restrict the penetration of neutral
treated with neither light nor photosensitizer (L−P−), or anionic PS (Jori et al, 2006; El-Adly, 2008; George
where S. aureus was more sensitive to PDI than E. et al., 2009; Huang et al., 2010).
coli. A significant increase (p < 0.05) in the percentage
reduction reaching 89.2% and 71.3% for S. aureus and Effect of dark photosensitizer cytotoxicity
E. coli respectively was induced upon increasing the It is clear from Table IV, that there is an evident of
concentration of HP to 3 mg/mL after 15 min incuba- non-illuminated haematoporphyrin (L−P+) cytotoxity
tion and exposure to 4.5 J/cm2 (Fig. 4). Interestingly, against both microorganisms studied with more effici-
increasing fluence dose to 9 J/cm2 was accompanied by ent bactericidal activity against Gram-positive S.
a relative decrease in the viable count percentage aureus (2.8-21.4%) than Gram-negative E. coli (0.95-
reaching 93.5% for S. aureus and 76.1% for E. coli. 12.6%) under the same conditions. This dark toxicity
The same held true for increasing incubation time up was previously attributed to the ability of PS to form
to 15 min where the uptake of photosensitizer in- additional dimmers within a bacterial cell by the bio-
creased into the cells, leading to increase the photo- polymeric microenvironment of the cell, where the dye
toxicity. However, further increase in incubation time dimmers in the cell may be the predominant participants
beyond 15 min resulted in transition and viable count in cellular damage (Lasocki et al., 1999; Usacheva et
leveling off with no significant difference (p < 0.05) al., 2001; Demidova et al., 2005; Nisnevitch et al.,
between 30 and 45 min for both species (94%, 93.8% 2010).
for S. aureus, and 76.2%, 76.7% for E. coli). Similar
results were previously reported (Bertoloni et al., Effect of light cytotoxicity in absence of pho-
2000; Zeina et al., 2001; Hamblin et al., 2002; Giusti tosensitizer
et al., 2008; Peloi et al., 2008). Small reduction (13.1% and 12.5% in case of S.
The differences in the susceptibilities of the Gram- aureus and E.coli respectively) was observed in the
viable count in absence of photosensitizer after illu-
mination (L+P−) at fluence dose 4.5 J/cm2 as shown in
Table V. This reduction of colony forming units was
increased to 25.5% and 25.1% respectively at exposure
to 9 J/cm2. The killing effect of light illuminated from
LED source in absence of exogenous PS might be
supported by photostimulation of endogenous por-
phyrin molecules which resulted in the generation of
reactive species predominantely the singlet oxygen,

Table IV. Effect of HP concentrations (0.5-3mg/mL) on

the %reduction of bacterial species (mean ± S.D., n = 6)
Different HP concentrations (mg/mL)
Fig. 4. The percentage reduction of the viable count of Bacterial
0.5 1 2 3
bacterial species (Staphylococcus aureus “S. aureus” and species
Escherichia coli “E. coli”) as a function of Hematoporphyrin % Reduction (CFU/mL)
(HP) concentration (0.5-3 mg/mL), samples were incubated S. aureus 2.80 ± 200. 4.7 ± 1.80 11.8 ± 2.60 21.4 ± 2.83
for 15 min and then irradiated at 4.5 J/cm−2 (mean ± S.D., n
E. coli 0.95 ± 0.62 1.6 ± 1.22 05.8 ± 2.46 12.6 ± 2.27
= 6).
Moxifloxacin-Gelrite In Situ Ophthalmic Gelling System 1673

Table V. Effect of light fluence (0.9-9 J/cm2) on the % fically, at 8 h post-instillation of vigamox®, the Mox
reduction of bacterial species (mean ± S.D., n = 6) concentration level had decreased to 0.051 µg/mL
Light fluence (J/cm2) which was below the minimum inhibitory concentra-
Bacterial tion (MIC) of S. aureus ~0.12 µg/mL (Levine et al.,
0.9 2.7 4.5 9
species 2004). Whereas formula G5 provided significant high
% Reduction (CFU/mL)
Mox conc levels of 0.35 µg/mL at 8 h and maintained
S. aureus 5.2 ± 1.22 7.8 ± 1.84 13.1 ± 3.79 25.5 ± 5.8 above MIC up to 24 h (0.13 µg/mL).
E. coli 6.5 ± 2.42 9.4 ± 2.57 12.5 ± 2.09 25.1 ± 1.67 The ocular bioavailability of Mox was illustrated by
the area under the curve (AUC), maximum Mox con-
centration (Cmax) and the time at which the Cmax is
causing cell death (Nitzan et al., 2004; Lipovsky et al., achieved (Tmax). The AUC and Cmax values revealed
2009; Maclean et al., 2009). approximately 6-fold increase for G5 relative to
Therefore, on the basis of overall results obtained, vigamox® eye drops. The reason for such improved
both organisms tested S. aureus (Gram-positive) and ocular bioavailability was attributed mainly to the
E. coli (Gram-negative) were susceptible to effective mucoadhesive property of Gelrite® and its in situ-
lethal photosensitization using HP as a photosen- gelling effect which would diminish drainage after
sitizer in a concentration of 3 mg/mL, illuminated by application (decreased clearance by tears). Moreover,
630 nm LED at 9 J/cm2 and incubated for 15 min with the ability of Gelrite to interact with calcium ions
S. aureus being more sensitive (divalent cations) might have permeation enhancing
effect comparable to that of EDTA (Hartmann and
In vivo studies Keipert, 2000; Ludwig, 2005). Since the cellular calcium
Concentration-time profiles of Mox in aqueous humor levels probably play an important role in the intactness
after topical instillation of G5 formula and commercial of the tight junctions between the corneal epithelial
eye drops vigamox® are shown in Fig. 5. Animals cells that limit the drug permeation (Jarvinen et al.,
treated with G5 had significantly higher drug level (p 1995; Urtti, 2006). Decreasing the extracellular calcium
< 0.05) than those treated with vigamox® at all time concentration (by calcium chelating agents as EDTA)
points. It was interesting to note that, the aqueous leads to open and increases the permeability of tight
humor level showed a maximum at 1 h post admin- junctions (Borchard et al., 1996; Kaur et al., 2000;
istration which decreased gradually afterwards. These Nanjawade et al., 2007).
results were correlated with the in vitro release study Consequently, the in vivo results coupled with the
of the tested in situ gelling system where a high in vitro ones suggested that the G5 in situ gelling
amount of the drug was released (burst effect) during formula significantly prolonged the Mox residence
the first hours followed by gradual drug release on the time, which makes it a superior formula for sustained
following duration of the study. It is important to note ocular Mox delivery compared to conventional eye
that the higher concentration of the antibiotic is drops. Therefore, G5 formula was given once daily in
always desirable at early time of infection. More speci- phase II study of in vivo treatment of corneal infection
by S. aureus.
On the other hand, the ability of the optimized PDT
to effectively eradicate S. aureus of the post fourth
animal infected group using HP photosensitizer and
LED 630 nm was illustrated in Fig. 6D.
It is obvious from visual inspection (Fig. 6) that, the
rabbits cornea of positive control group (infected not
treated) showed redness, lacrimation, mucoid discharge
and fluorescein staining. Also the histological exam-
ination revealed desquamation of the outermost su-
perficial cells in corneal epithelium, stromal edema,
prominent blood capillaries with numerous inflamma-
tory cells (lymphocytes) aggregated in perivascular
area in the stroma and reduced number of keratocytes
Fig. 5. Moxifloxacin concentrations (µg/mL) attained in in the stroma, compared with normal rabbit cornea as
aqueous humor after application of G5 and vigamox® (mean shown in Fig. 7. Treatment with formula G5 or PDT
± S.D., n = 6) showed good response where, the signs of infection
1674 H. M. El-Laithy et al.

Fig. 6. Growth appearance of rabbit cornea demonstrate (A) normal, (B) infected with S. aureus, (C) treated with G5 after
4 days and (D) treated with PDT after 7 days.

and inflammation were disappeared with no fluores- represent a signal of a new interest encouraging
cence detection. It is worthy noting that, group of further in vivo studies to explore the synergy of using
animals treated with formula G5 showed remarkable both techniques in the treatment of severe corneal
improvement of stromal edema and increase in the infections. PDT might provide a valuable adjuvant
number of keratocytes with faster epithelium healing antibacterial effect when combined with the antibiotic
after four days of G5 application compared to seven therapy in the treatment of intractable bacterial infec-
days of PDT treatment (Fig. 7). tions. Hopefully this recently emerging field will be
In conclusion, this study investigated the potential worthy for more study.
of ophthalmic delivery of moxifloxacin from Gelrite in
situ gelling system and PDT for treatment of bacterial ACKNOWLEDGEMENTS
corneal infection. Both approaches were effectively
treating the infection without causing ocular irritation The authors are grateful to Dr. M. Nebsen (Depart-
or collateral damage to corneal tissue. This study may ment of Analytical Chemistry, Pharmacy College,
Cairo University, Cairo, Egypt) for carrying out the
spectrofluorimetric analysis to determine the ocular
residence time. The authors thank Dr. Sahar K.
Darwish, Ph.D. Department of Histology (NODCAR)
and Dr. Yousef El-Kommos Samuel, MD for doing and
reviewing the interpretation data of histopathological
study.

REFERENCES

Ali, Y. and Lehmussaari, K., Industrial perspective in ocular

drug delivery. Adv. Drug Deliv. Rev., 58, 1258-1268 (2006).
Al-Kassas, R. S. and El-Khatib, M. M., Ophthalmic controlled
release in situ gelling systems for ciprofloxacin based on
polymeric carriers. Drug Deliv., 16, 145-152 (2009).
Andrews, G. P., Laverty, T. P., and Jones, D. S., Mucoadhesive
polymeric platforms for controlled drug delivery. Eur. J.
Pharm. Biopharm., 71, 505-518 (2009).
Anumolu, S. S., Singh, Y., Gao, D., Stein, S., and Sinko, P. J.,
Fig. 7. Histology of (A) normal rabbit corneal tissue (control Design and evaluation of novel fast forming pilocarpine-
negative), showing normal corneal epithelium, stroma with loaded ocular hydrogels for sustained pharmacological
keratocytes. (B) Infected non-treated rabbit corneal tissue response. J. Control. Release, 137, 152-159 (2009).
(control positive), showing desquamation of corneal epithe- Badawi, A. A., El-Laithy, H. M., El Qidra, R. K., El Mofty,
lium, blood capillaries, edematous stroma with inflamma- H., and El dally, M., Chitosan based nanocarriers for
tory cells. (C) Treated rabbits corneal tissue with G5 formula
indomethacin ocular delivery. Arch. Pharm. Res., 31,
and (D) treated rabbits with PDT showing intact corneal
epithelium and stroma with keratocytes. H&E, X: 100. (S) = 1040-1049 (2008).
stroma, (arrow) = corneal epithelium, (arrow head) = Bajaj, I. B., Survase, S. A., Saudagar, P. S., and Singhal, R.
keratocytes, (thin arrow) = blood capillaries, (double arrow S., Gellan gum: fermentative production, downstream
head) = inflammatory cells. processing and applications. Food Technol. Biotechnol.,
Moxifloxacin-Gelrite In Situ Ophthalmic Gelling System 1675

45, 341-354 (2007). Int. J. Pharm., 241, 47-55 (2002).

Balasubramaniam, J. and Pandit, J. K., Ion-activated in situ Fernandes, L. A., de Almeida, J. M., Theodoro, L. H., Bosco,
gelling systems for sustained ophthalmic delivery of cipro- A. F., Nagata, M. J., Martins, T. M., Okamoto, T., and
floxacin hydrochloride. Drug Deliv., 10, 185-191 (2003). Garcia, V. G., Treatment of experimental periodontal
Bertoloni, G., Lauro, F. M., Cortella, G., and Merchat, M., disease by photodynamic therapy in immunosuppressed
Photosensitizing activity of hematoporphyrin on Staphyl- rats. J. Clin. Periodontol., 36, 219-228 (2009).
ococcus aureus cells. Biochim. Biophys. Acta, 1475, 169- Gan, L., Gan, Y., Zhu, C., Zhang, X., and Zhu, J., Novel micro-
174 (2000). emulsion in situ electrolyte-triggered gelling system for
Borchard, G., Lueβen, H. L., de Boer, A. G., Verhoef, J. C., ophthalmic delivery of lipophilic cyclosporine A: in vitro
Lehr, C.-M., and Junginger, H. E., The potential of mu- and in vivo results. Int. J. Pharm., 365, 143-149 (2009).
coadhesive polymers in enhancing intestinal peptide drug Ganz, R. A., Viveiros, J., Ahmad, A., Ahmadi, A., Khalil, A.,
absorption. III: Effects of chitosan-glutamate and carbomer Tolkoff, M. J., Nishioka, N. S., and Hamblin, M. R., Heli-
on epithelial tight junctions in vitro. J. Control. Release, cobacter pylori in patients can be killed by visible light.
39, 131-138 (1996). Lasers Surg. Med., 36, 260-265 (2005).
Bourlais, C. L., Acar, L., Zia, H., Sado, P. A., Needham, T., George, S., Hamblin, M. R., and Kishen, A., Uptake path-
and Leverge, R., Ophthalmic drug delivery systems-recent ways of anionic and cationic photosensitizers into bac-
advances. Prog. Retin. Eye Res., 17, 33-58 (1998). teria. Photochem. Photobiol. Sci., 8, 788-795 (2009).
Carlfors, J., Edsman, K., Petersson, R., and Jörnving, K., Giusti, J. S., Santos-Pinto, L., Pizzolito, A. C., Helmerson,
Rheological evaluation of Gelrite in situ gels for ophthalmic K., Carvalho-Filho, E., Kurachi, C., and Bagnato, V. S.,
use. Eur. J. Pharm. Sci., 6, 113-119 (1998). Antimicrobial photodynamic action on dentin using a light-
Castano, A. P., Demidova, T. N., and Hamblin, M. R., Mech- emitting diode light source. Photomed. Laser Surg., 26,
anisms in photodynamic therapy: part one-photosensitizers, 281-287 (2008).
photochemistry and cellular localization. Photodiagnosis Hamblin, M. R., O'Donnell, D. A., Murthy, N., Rajagopalan,
Photodyn. Ther., 1, 279-293 (2004). K., Michaud, N., Sherwood, M. E., and Hasan, T., Poly-
Charoo, N. A., Kohli, K., and Ali, A., Preparation of in situ- cationic photosensitizer conjugates: effects of chain length
forming ophthalmic gels of ciprofloxacin hydrochloride for and Gram classification on the photodynamic inactivation
the treatment of bacterial conjunctivitis: in vitro and in of bacteria. J. Antimicrob. Chemother., 49, 941-951 (2002).
vivo studies. J. Pharm. Sci., 92, 407-413 (2003). Hamblin, M. R. and Hasan, T., Photodynamic therapy: a
Constantinou, M., Daniell, M., Snibson, G. R., Vu, H. T., and new antimicrobial approach to infectious disease? Photo-
Taylor, H. R., Clinical efficacy of moxifloxacin in the treat- chem. Photobiol. Sci., 3, 436-450 (2004).
ment of bacterial keratitis: a randomized clinical trial. Hartmann, V. and Keipert, S., Physico-chemical, in vitro
Ophthalmology, 114, 1622-1629 (2007). and in vivo characterization of polymers for ocular use.
Dai, T., Tegos, G. P., Zhiyentayev, T., Mylonakis, E., and Pharmazie, 55, 440-443 (2000).
Hamblin, M. R., Photodynamic therapy for methicillin- Hassan, E. E. and Gallo, J. M., A simple rheological method
resistant Staphylococcus aureus infection in a mouse skin for the in vitro assessment of mucin-polymer bioadhesive
abrasion model. Lasers Surg. Med., 42, 38-44 (2010). bond strength. Pharm. Res., 7, 491-495 (1990).
de Almeida, J. M., Theodoro, L. H., Bosco, A. F., Nagata, M. Hongcharu, W., Taylor, C. R., Chang, Y., Aghassi, D.,
J., Bonfante, S., and Garcia, V. G., Treatment of experi- Suthamjariya, K., and Anderson, R. R., Topical ALA-
mental periodontal disease by photodynamic therapy in photodynamic therapy for the treatment of acne vulgaris.
rats with diabetes. J. Periodontol., 79, 2156-2165 (2008). J. Invest. Dermatol., 115, 183-192 (2000).
Deasy, P. B. and Quigley, K. J., Rheological evaluation of Huang, L., Terakawa, M., Zhiyentayev, T., Huang, Y. Y.,
deacetylated gellan gum (Gelrite) for pharmaceutical use. Sawayama, Y., Jahnke, A., Tegos, G. P., Wharton, T., and
Int. J. Pharm., 73, 117-123 (1991). Hamblin, M. R., Innovative cationic fullerenes as broad-
Demidova, T. N., Gad, F., Zahra, T., Francis, K. P., and spectrum light-activated antimicrobials. Nanomedicine,
Hamblin, M. R., Monitoring photodynamic therapy of 6, 442-452 (2010).
localized infections by bioluminescence imaging of gene- Ishikawa, S., Suzuki, K., Fukuda, E., Arihara, K., Yamamoto,
tically engineered bacteria. J. Photochem. Photobiol. B, Y., Mukai, T., and Itoh, M., Photodynamic antimicrobial
81, 15-25 (2005). activity of avian eggshell pigments. FEBS Lett., 584, 770-
Donnelly, R. F., McCarron, P. A., and Tunney, M. M., Anti- 774 (2010).
fungal photodynamic therapy. Microbiol. Res., 163, 1-12 Jarvinen, K., Jarvinen, T., and Urtti, A., Ocular absorption
(2008). following topical delivery. Adv. Drug Deliv. Rev., 16, 3-19
El-Adly, A. A., Photoactive anionic porphyrin derivative (1995).
against Gram-positive and Gram-negative bacteria. J. Appl. Jori, G., Fabris, C., Soncin, M., Ferro, S., Coppellotti, O., Dei,
Sci. Res., 4, 1817-1821 (2008). D., Fantetti, L., Chiti, G., and Roncucci, G., Photodynamic
El-Kamel, A. H., In vitro and in vivo evaluation of Pluronic therapy in the treatment of microbial infections: basic
F127-based ocular delivery system for timolol maleate. principles and perspective applications. Lasers Surg.
1676 H. M. El-Laithy et al.

Med., 38, 468-481 (2006). solutions for ophthalmic drug delivery. J. Control. Release,
Kalam, M. A., Sultana, Y., Samad, A., Ali, A., Aqil, M., 69, 379-388 (2000).
Sharma, M., and Mishra, A. K., Gelrite-based in vitro Lipovsky, A., Nitzan, Y., Friedmann, H., and Lubart, R.,
gelation ophthalmic drug delivery system of gatifloxacin. Sensitivity of Staphylococcus aureus strains to broadband
J. Dispers. Sci. Technol., 29, 89-96 (2008). visible light. Photochem. Photobiol., 85, 255-260 (2009).
Kao, H.-J., Lo, Y.-L., Vong, W.-J., Lin, Y.-J., and Lin, H.-R., Liu, Y., Liu, J., Zhang, X., Zhang, R., Huang, Y., and Wu, C.,
Treating allergic conjunctivitis using in situ polyelec- In situ gelling gelrite/alginate formulations as vehicles for
trolyte gelling systems. J. Biomater. Sci. Polym. Ed., 17, ophthalmic drug delivery. AAPS PharmSciTech, 11, 610-
1191-1205 (2006). 620 (2010).
Kaur, I. P., Singh, M., and Kanwar, M., Formulation and Liu, Z., Li, J., Nie, S., Liu, H., Ding, P., and Pan, W., Study
evaluation of ophthalmic preparations of acetazolamide. of an alginate/HPMC-based in situ gelling ophthalmic
Int. J. Pharm., 199, 119-127 (2000). delivery system for gatifloxacin. Int. J. Pharm., 315, 12-17
Kaur, I. P., Garg, A., Singla, A. K., and Aggarwal, D., Vesi- (2006).
cular systems in ocular drug delivery: an overview. Int. J. Ludwig, A., The use of mucoadhesive polymers in ocular
Pharm., 269, 1-14 (2004). drug delivery. Adv. Drug Deliv. Rev., 57, 1595- 1639 (2005).
Kim, D. H., Stark, W. J., O’Brien, T. P., and Dick, J. D., Luo, L., Li, D. Q., Corrales, R. M., and Pflugfelder, S. C.,
Aqueous penetration and biological activity of moxifloxa- Hyperosmolar saline is a proinflammatory stress on the
cin 0.5% ophthalmic solution and gatifloxacin 0.3% solu- mouse ocular surface. Eye Contact Lens, 31, 186-193
tion in cataract surgery patients. Ophthalmology, 112, (2005).
1992-1996 (2005). Ma, W. D., Xu, H., Wang, C., Nie, S. F., and Pan, W. S.,
Kömerik, N., Nakanishi, H., MacRobert, A. J., Henderson, Pluronic F127-g-poly(acrylic acid) copolymers as in situ
B., Speight, P., and Wilson, M., In vivo killing of Porphyr- gelling vehicle for ophthalmic drug delivery system. Int. J.
omonas gingivalis by toluidine blue-mediated photosensi- Pharm., 350, 247-256 (2008).
tization in an animal model. Antimicrob. Agents Chemother., Maclean, M., MacGregor, S. J., Anderson, J. G., and
47, 932-940 (2003). Woolsey, G., Inactivation of bacterial pathogens following
Kubo, W., Miyazaki, S., and Attwood, D., Oral sustained exposure to light from a 405-nanometer light-emitting
delivery of paracetamol from in situ-gelling gellan and diode array. Appl. Environ. Microbiol., 75, 1932-1937
sodium alginate formulations. Int. J. Pharm., 258, 55-64 (2009).
(2003). Nanjawade, B. K., Manvi, F. V., and Manjappa, A. S., In
Lambrechts, S. A., Demidova, T. N., Aalders, M. C., Hasan, situ-forming hydrogels for sustained ophthalmic drug
T., and Hamblin, M. R., Photodynamic therapy for Sta- delivery. J. Control. Release, 122, 119-134 (2007).
phylococcus aureus infected burn wounds in mice. Neupane, J., Ghimire, S., Shakya, S., Chaudhary, L., and
Photochem. Photobiol. Sci., 4, 503-509 (2005). Shrivastava, V. P., Effect of light emitting diodes in the
Langlois, M. H., Montagut, M., Dubost, J. P., Grellet, J., and photodynamic therapy of rheumatoid arthritis. Photodia-
Saux, M. C., Protonation equilibrium and lipophilicity of gnosis Photodyn. Ther., 7, 44-49 (2010).
moxifloxacin. J. Pharm. Biomed. Anal., 37, 389-393 (2005). Nisnevitch, M., Nakonechny, F., and Nitzan, Y., Photo-
Lasocki, K., Szpakowska, M., Grzybowski, J., and Graczyk, dynamic antimicrobial chemotherapy by liposome-encap-
A., Examination of antibacterial activity of the photoac- sulated water-soluble photosensitizers. Russ. J. Bioorganic
tivated arginine haematoporphyrin derivative. Pharmacol. Chem., 36, 363-369 (2010).
Res., 39, 181-184 (1999). Nitzan, Y., Salmon-Divon, M., Shporen, E., and Malik, Z.,
Lembo, A. J., Ganz, R. A., Sheth, S., Cave, D., Kelly, C., ALA induced photodynamic effects on gram positive and
Levin, P., Kazlas, P. T., Baldwin, P. C., 3rd, Lindmark, W. negative bacteria. Photochem. Photobiol. Sci., 3, 430-435
R., McGrath, J. R., and Hamblin M. R., Treatment of (2004).
Helicobacter pylori infection with intra-gastric violet light Ocaña, J. A., Barragán, F. J., and Callejón, M., Spectrofluo-
phototherapy: a pilot clinical trial. Lasers Surg. Med., 41, rimetric determination of moxifloxacin in tablets, human
337-344 (2009). urine and serum. Analyst, 125, 2322-2325 (2000).
Levine, J. M., Noecker, R. J., Lane, L. C., Herrygers, L., Nix, Omar, G. S., Wilson, M., and Nair, S. P., Lethal photosensi-
D., and Snyder, R. W., Comparative penetration of moxi- tization of wound-associated microbes using indocyanine
floxacin and gatifloxacin in rabbit aqueous humor after green and near-infrared light. BMC Microbiol., 8, 111-120
topical dosing. J. Cataract Refract. Surg., 30, 2177-2182 (2008).
(2004). O’Riordan, K., Akilov, O. E., and Hasan, T., The potential for
Li, D. Q., Chen, Z., Song, X. J., Luo, L., and Pflugfelder, S. C., photodynamic therapy in the treatment of localized infec-
Stimulation of matrix metalloproteinases by hypero- tions. Photodiagnosis Photodyn. Ther., 2, 247-262 (2005).
smolarity via a JNK pathway in human corneal epithelial Pawar, P. K. and Majumdar, D. K., Effect of formulation
cells. Invest. Ophthalmol. Vis. Sci., 45, 4302-4311 (2004). factors on in vitro permeation of moxifloxacin from aqueous
Lin, H. R. and Sung, K. C., Carbopol/pluronic phase change drops through excised goat, sheep, and buffalo corneas.
Moxifloxacin-Gelrite In Situ Ophthalmic Gelling System 1677

AAPS PharmSciTech, 7, E13 (2006). The release dynamics of salicylic acid and tetracycline
Peloi, L. S., Soares, R. R., Biondo, C. E., Souza, V. R., Hioka, hydrochloride from the psyllium and polyacrylamide
N., and Kimura, E., Photodynamic effect of light-emitting based hydrogels (II). Carbohydr. Polym., 67, 559-565
diode light on cell growth inhibition induced by methylene (2007).
blue. J. Biosci., 33, 231-237 (2008). Solomon, R., Donnenfeld, E. D., Perry, H. D., Snyder, R. W.,
Qi, H., Li, L., Huang, C., Li, W., and Wu, C., Optimization Nedrud, C., Stein, J., and Bloom, A., Penetration of topi-
and physicochemical characterization of thermosensitive cally applied gatifloxacin 0.3%, moxifloxacin 0.5%, and
poloxamer gel containing puerarin for ophthalmic use. ciprofloxacin 0.3% into the aqueous humor. Ophthalmology,
Chem. Pharm. Bull. (Tokyo), 54, 1500-1507 (2006). 112, 466-469 (2005).
Qi, H., Chen, W., Huang, C., Li, L., Chen, C., Li, W., and Wu, Srividya, B., Cardoza, R. M., and Amin, P. D., Sustained
C., Development of a poloxamer analogs/carbopol-based ophthalmic delivery of ofloxacin from a pH triggered in
in situ gelling and mucoadhesive ophthalmic delivery situ gelling system. J. Control. Release, 73, 205-211 (2001).
system for puerarin. Int. J. Pharm., 337, 178-187 (2007). Stahl, U., Willcox, M. D., Naduvilath, T., and Stapleton, F.,
Rathore, M. S. and Majumdar, D. K., Effect of formulation Influence of tear film and contact lens osmolality on
factors on in vitro transcorneal permeation of gatifloxacin ocular comfort in contact lens wear. Optom. Vis. Sci., 86,
from aqueous drops. AAPS PharmSciTech, 7, 57 (2006). 857-867 (2009).
Razek, T. M. A., El-Baqary, R. I., and Ramadan, A. E., Fluo- Stroman, D. W., Dajcs, J. J., Cupp, G. A., and Schlech, B. A.,
rimetric determination of gatifloxacin in aqueous, pure and In vitro and in vivo potency of moxifloxacin and moxi-
pharmaceutical formulations. Anal. Lett., 41, 417-423 (2008). floxacin ophthalmic solution 0.5%, a new topical fluoro-
Ritger, P. L. and Peppas, N. A., A simple equation for quinolone. Surv. Ophthalmol., 50 Suppl 1, S16-S31
description of solute release II. Fickian and anomalous (2005).
release from swellable devices. J. Control. Release, 5, 37- Sultana, Y., Aqil, M., and Ali, A., Ion-activated, Gelrite-
42 (1987). based in situ ophthalmic gels of pefloxacin mesylate: com-
Robertson, S. M., Curtis, M. A., Schlech, B. A., Rusinko, A., parison with conventional eye drops. Drug Deliv., 13, 215-
Owen, G. R., Dembinska, O., Liao, J., and Dahlin, D. C., 219 (2006).
Ocular pharmacokinetics of moxifloxacin after topical Teichert, M. C., Jones, J. W., Usacheva, M. N., and Biel, M.
treatment of animals and humans. Surv. Ophthalmol., 50 A., Treatment of oral candidiasis with methylene blue-
Suppl 1, S32-S45 (2005). mediated photodynamic therapy in an immunodeficient
Roy, S., Pal, K., Anis, A., Pramanik, K., and Prabhakar, B., murine model. Oral Surg. Oral Med. Oral Pathol. Oral
Polymers in mucoadhesive drug delivery system: a brief Radiol. Endod., 93, 155-160 (2002).
note. Des. Monomers Polym., 12, 483-495 (2009). Torkildsen, G. and O'Brien, T. P., Conjunctival tissue
Rozier, A., Mazuel, C., Grove, J., and Plazonnet, B., Gelrite®: pharmacokinetic properties of topical azithromycin 1%
a novel, ion-activated in situ gelling polymer for oph- and moxifloxacin 0.5% ophthalmic solutions: a single-
thalmic vehicles. Effect on bioavailability of timolol. Int. J. dose, randomized, open-label, active-controlled trial in
Pharm., 57, 163-168 (1989). healthy adult volunteers. Clin. Ther., 30, 2005-2014 (2008).
Salem, H., Spectrofluorimetric, atomic absorption spectro- Urtti, A., Challenges and obstacles of ocular pharmacokine-
metric and spectrophotometric determination of some tics and drug delivery. Adv. Drug Deliv. Rev., 58, 1131-
fluoroquinolones. Am. J. Appl. Sci., 2, 719-729 (2005). 1135 (2006).
Sanzgiri, Y. D., Maschi, S., Crescenzi, V., Calligaro, L., Topp, Usacheva, M. N., Teichert, M. C., and Biel, M. A., Compari-
E. M., and Stella, V. J., Gellan-based systems for oph- son of the methylene blue and toluidine blue photobacteri-
thalmic sustained delivery of methylprednisolone (MP). J. cidal efficacy against gram-positive and gram-negative
Control. Release, 26, 195-201 (1993). microorganisms. Lasers Surg. Med., 29, 165-173 (2001).
Schrage, N., Wuestemeyer, H., and Langefeld, S., Do differ- Vandamme, T. F. and Brobeck, L., Poly(amidoamine) den-
ent osmolar solutions change the epithelial surface of the drimers as ophthalmic vehicles for ocular delivery of pilo-
healthy rabbit cornea? Graefes Arch. Clin. Exp. Ophthalmol., carpine nitrate and tropicamide. J. Control. Release, 102,
242, 668-673 (2004). 23-38 (2005).
Sensoy, D., Cevher, E., Sarici, A., Yilmaz, M., Ozdamar, A., Wei, G., Xu, H., Ding, P. T., Li, S. M., and Zheng, J. M.,
and Bergişadi, N., Bioadhesive sulfacetamide sodium Thermosetting gels with modulated gelation temperature
microspheres: evaluation of their effectiveness in the for ophthalmic use: the rheological and gamma scinti-
treatment of bacterial keratitis caused by Staphylococcus graphic studies. J. Control. Release, 83, 65-74 (2002).
aureus and Pseudomonas aeruginosa in a rabbit model. Weyenberg, W., Vermeire, A., Dhondt, M. M., Adriaens, E.,
Eur. J. Pharm. Biopharm., 72, 487-495 (2009). Kestelyn, P., Remon, J. P., and Ludwig, A., Ocular bioero-
Sharman, W. M., Allen, C. M., and van Lier, J. E., Photo- dible minitablets as strategy for the management of
dynamic therapeutics: basic principles and clinical appli- microbial keratitis. Invest. Ophthalmol. Vis. Sci., 45,
cations. Drug Discov. Today, 4, 507-517 (1999). 3229-3233 (2004).
Singh, B., Chauhan, G. S., Sharma, D. K., and Chauhan, N., Wiegell, S. R. and Wulf, H. C., Photodynamic therapy of acne
1678 H. M. El-Laithy et al.

vulgaris using methyl aminolaevulinate: a blinded, rando- H. J., Jeong, I. Y., Park, S. W., Park, Y. G., Nah, H. J., and
mized, controlled trial. Br. J. Dermatol., 154, 969-976 Im, W. B., Experimental inhibition of corneal neova-
(2006). scularization by photodynamic therapy with verteporfin.
Wong, T. W, Wang, Y. Y., Sheu, H. M., and Chuang, Y. C., Curr. Eye Res., 31, 215-224 (2006).
Bactericidal effects of toluidine blue-mediated photo- Zeina, B., Greenman, J., Purcell, W. M., and Das, B., Killing
dynamic action on Vibrio vulnificus. Antimicrob. Agents of cutaneous microbial species by photodynamic therapy.
Chemother., 49, 895-902 (2005). Br. J. Dermatol., 144, 274-278 (2001).
Wu, C., Qi, H., Chen, W., Huang, C., Su, C., Li, W., and Hou, Zolfaghari, P. S., Packer, S., Singer, M., Nair, S. P., Bennett,
S., Preparation and evaluation of a Carbopol/HPMC- J., Street, C., and Wilson, M., In vivo killing of Staphylo-
based in situ gelling ophthalmic system for puerarin. coccus aureus using a light-activated antimicrobial agent.
Yakugaku Zasshi, 127, 183-191 (2007). BMC Microbiol., 9, 27 (2009).
Yoon, K. C., Ahn, K. Y., Lee, S. E., Kim, K. K., Im, S. K., Oh,