MATERIALS FOR LIFE SCIENCES Synthesis, characterization and antibacterial activity of thymol-loaded polylactic acid microparticles entrapped with essential oils of varying viscosity Agni Kumar Biswal 1 , Isha Vashisht 1 , Aamir Khan 1 , Shivangi Sharma 1 , and Sampa Saha 1, * 1 Department of Materials Science and Engineering, Indian Institute of Technology, Delhi, India Received: 9 December 2018 Accepted: 1 April 2019 Ó Springer Science+Business Media, LLC, part of Springer Nature 2019 ABSTRACT Double emulsion technique (W 1 /O/W 2 ) was used to achieve PLA (polylactic acid)-based microparticles loaded with thymol (hydrophilic antibacterial active) with high encapsulation efficiency ( [ 90%) that was later released in water in controlled fashion. In order to enhance the release of thymol from semicrys- talline slow degrading PLA matrix, pores were introduced into the particles by incorporating minute quantity of essential oils (EOs) such as castor, mustard, olive and coconut oils of varying viscosity (from 128 to 13 mPa S). Strikingly, it was found that the pore size of microparticles (pore size 1.2–0.53 lm and specific surface area 7.5–2.1 m 2 /g) has been influenced by the viscosity of EOs. The plausible mechanism of formation of pores using EOs was also explored. As expected, the release rate of thymol did get accelerate with the increase in pore density and its release mechanism can well be explained with the help of power law model. In addition to that, these porous particles were demonstrated to act as an effective carrier for continuous delivery of thymol with enhanced inhibi- tory activity probably due to synergistic action of EOs and thymol. Therefore, EOs employed here served dual purposes. These PLA-based microparticles composed of completely natural ingredients can be exploited as active food packaging material to prolong the shelf life of food. Introduction Traditionally, active compounds were directly added to food formulation for protecting food from micro- bial spoilage, thereby enhancing food shelf life. But these methods suffer from several limitations such as more amount of additives are needed to protect food materials from deterioration, raw/fresh food may lose their taste/flavor, and also there may be some complex reaction between food and actives [1–3]. Consequently, there is a demand in food industry for polymer packaging from which active compounds are diffused onto the food surface in order to prevent Address correspondence to E-mail: [email protected]https://doi.org/10.1007/s10853-019-03593-7 J Mater Sci Materials for life sciences
15
Embed
Synthesis, characterization and antibacterial activity of thymol …download.xuebalib.com/5sa9FavC8SrI.pdf · MATERIALS FOR LIFE SCIENCES Synthesis, characterization and antibacterial
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
MATERIALS FOR LIFE SCIENCES
Synthesis, characterization and antibacterial activity
of thymol-loaded polylactic acid microparticles
entrapped with essential oils of varying viscosity
aViscosity of oils was measured by using parallel-plate rheometerbParticle size, pore size and pore density were measured from SEM images using ImageJ software; average pore number was obtained by
counting only the visible side of the microparticles in SEM images; so, the pore number in the whole particle should be approximately
double of the reported valuescSpecific surface area was measured by BET method, and the values are average of three runsdEncapsulation efficiency was determined from UV/Vis spectrophotometer, and the values are average of three runs
J Mater Sci
from polymer phase [23] (see arrows in Fig S1). Ini-
tially, polymer (PLA) and EOs were well soluble in
dichloromethane (DCM). As evaporation of dichlor-
omethane was progressed, both polymer and EO
concentration got increased that resulted the phase
separation of EO and polymer due to their immisci-
bility [29]. Then, initially formed tiny EO droplets
within emulsion droplet may coalesce into small EO
droplets. It was postulated that the porous skin layer
of particles resulted from the inclusion of inner
aqueous droplets stabilized by these small EO dro-
plets along with PVA (surfactant) within the polymer
skin layer which was solidified at the outer interface
[26]. This kinetic arrest traps inner water phase to
yield pores/voids after evaporation of water. To
further understand the water droplet stabilization by
EO, the viscosity of organic phase and interfacial
tension between organic and water phases (which
contain 0.1 wt% and 0.25 wt% PVA solution as inner
water and outer water phase, respectively) were
measured for all the samples (samples A–E, Table SI).
It is clear from Table SI that the value of interfacial
tension between water and organic phase is sub-
stantially higher in the presence of EOs for 0.1 wt%
PVA solution than the interfacial tension between
organic phase with EOs and water phase comprising
0.25 wt% PVA due to higher surfactant concentration
at the interface. This may drive the EO droplets
toward the outer water phase from the inner water
phase [30, 31]. But, there is insignificant change in
interfacial tension among the various EOs or no oil at
that interface. In this scenario, viscosity of the organic
phase plays the pivotal role for arresting the inner
water droplet. As expected, viscosity of organic
phase increases with the increase in viscosity of EOs.
Hence, it can be assumed that the EO droplets will be
kinetically trapped at the skin for stabilizing inner
water droplet because of high viscosity exerted by the
oil phase leading to formation of greater pore size
and pore density in the case of castor oil having the
highest viscosity. This is clearly evident in Fig S1b
(see the red arrows). For the similar reason, coconut
oil with least viscosity would favor the formation of
particle skin with minimum pore size and density.
Similar observation was also reported by Kim et al.
[32]. They have observed the arresting of 2-methyl
pentane droplet onto the surface of oil (PLGA in
DCM) droplet similar to particle-stabilized emulsion
owing to the fast extraction of DCM. In our system
also, amount of DCM was kept low (3.3 vol%) which
was marginally higher than its solubility limit (2
vol%) [33] in water so that quick evaporation of DCM
can take place, because, after diffusing through water
only, DCM will be evaporated at water/air interface.
This accelerated DCM evaporation in turn decreased
the time allowed for coalescence of inner water dro-
plets (W1) with the external water phase (W2) and
hence increased the chance of pore formation by
retention of inner aqueous phase (W1). Furthermore,
in another report, Shi et al. [34] also concluded that
the rate of solvent removal plays a key role in
resulting porous structure of microspheres. Finally,
during freeze-drying process under high-vacuum
inner water droplets stabilized by EO droplets along
with PVA and PLA were evaporated and converted
into holes/pores (Scheme 1). However, when these
inner EO droplets were not included into the skin
layer (interfacial region), particles with non-porous
surfaces were formed (sample A). Hence, thymol-
loaded particles with no addition of EO rendered
non-porous surface. But, double emulsion was also
necessary to make the porous shell layer. In our
Scheme 1 Schematic
representation of the formation
of porous microparticles.
J Mater Sci
previous work, it was found that without inclusion of
inner aqueous droplet, no pores can be found onto
the surface of microspheres [23]. Therefore, it was
essential to add EO in the oil phase of double emul-
sion to make particles with porous surface. After
cross-sectioning the particles, it was found that inside
the microspheres, pores were uniformly distributed
throughout the particles even in the case of particles
having no EO (Fig S2(i)). Because of fast removal of
solvent, polymer precipitation will proceed from
surface to the core of microspheres. Hence, inner
aqueous droplet will be trapped inside the micro-
spheres by high-viscous polymer layer as discussed
above. This will finally develop into small pores
inside the microspheres after evaporation of water
[25] even in the case of sample A. It was also noted
that for high-viscous castor oil, pore density and size
were quite bigger compared to the rest of EOs. This
could be attributed to the high-viscous oil phase
(PLA along with EO in DCM) which might have
stabilized the inner water (primary emulsion) droplet
against coalescence by providing strong interfacial
tension [25] (see green arrow in Fig S1b). Therefore,
coconut oil has least viscosity among the four EOs
and displayed minimal porosity (Fig S1e) due to ease
of migration of low-viscous EO toward outer water
phase leading to reduced number of trapped inner
water droplets at the surface
Encapsulation efficiency of thymoland in vitro release study
The encapsulation efficiency of thymol was quite
high and found to vary from* 94 to 99% for samples
B–E (Table 1). A marginal decrease in encapsulation
efficiency was found for castor oil-loaded particles
probably due to its larger pore size compared to
others. However, the low encapsulation efficiency of
thymol for non-porous sample (sample A) implied
the role of hydrophobic interaction of EO with thy-
mol resulting in higher retention of it even with
porous surfaces. In order to predict particles’
behavior as active packaging materials, the in vitro
release study of thymol was conducted for samples
A–E at 37 �C in neutral PBS buffer (pH 7); the same
conditions were used for the bacterial growth during
antibacterial study [35]. As shown in Fig. 2, all sam-
ples displayed a high initial burst release of[ 35%
within 24 h because of adsorption of thymol onto the
surface of particles. Castor oil-loaded particles
showed highest burst release since the pores were
largest among all, and hence the diffusion rate of
thymol was high, especially in comparison with no-
oil-loaded particles having non-porous surface mor-
phology. A large initial release was definitely a plus
for antibacterial packaging applications to provide
immediate protection from bacterial contamination
[36]. After the initial burst, a slow and continuous
release was obtained from internally trapped thymol
irrespective of the nature of samples. As expected,
castor oil-loaded particles with the highest porosity
showed a relatively faster and maximum % of
release (* 65%) compared to its analogs. However,
complete release was not attainable for any samples,
probably due to strong interaction between thymol
and PLA matrix (discussed later). Similar trend was
observed in releasing thymol in water from other
polymeric films [18, 37]. It was noteworthy that
release of thymol from non-porous surface was not
hugely different from the porous particles, partly due
to achievement of considerable porosity inside the
particles and also being a small molecule, diffusion of
thymol remained largely unaffected by the pore size.
Another important observation was that the encap-
sulation of thymol (volatile) in these particles pre-
vented its evaporation to air in its dried form and
hence increased its storage stability [36]. There was
no change in weight of particles over a long period of
time (15 days) when the dry particles were stored in
open air.
Several approaches were reported in the literature
to understand the release kinetics of entrapped agent
Figure 2 In vitro release study of samples A, B, C, D and E in
PBS (pH 7) at 37 �C. All experiments were run in triplicates, and
the average values are shown with error bars.
J Mater Sci
from polymeric matrix. In this respect, many
researchers utilized a classic semiempirical power
law model (Eq. 1) that can be fitted well with the
experimental results to understand the release phe-
nomena predominantly occurred by diffusion,
degradation or a combination of both [38–40]:
Mt
M1¼ ktn; ð1Þ
where Mt and M? are cumulative release of thymol
at time t and infinite, n was a release exponent which
mainly defines the release mechanism, and k was the
rate constant. Such model was applied in our system
too and found to be well-fitted by the experimental
data (Fig S4, Table 2). It was interesting to note that
the release of thymol was characterized by low
release exponent value (n\ 0.2) that proportionally
increased with the porosity of the particles. Same
trend was observed for the value of rate constant
(k) which was least for non-porous particles. As per
the model, for spherical particles, when
n value\ 0.20, release kinetics would follow combi-
nation of both diffusion and erosion path. These
observations suggest that after initial burst release of
surface-adsorbed thymol, water ingression through
the pores was increased marginally for the porous
particles which favored its release slightly over the
non-porous particles. This might be due to strong
aggregation of thymol in PLA matrix that needed to
be solubilized before being released in pure diffusive
mode. Hence, it did not follow pure Fickian release
mode. This indirectly implied the existence of strong
attractive forces between thymol and polymer matrix.
Moreover, according to our hypothesis of pore for-
mation, pores would be surrounded by hydrophobic
EO layer which may significantly inhibit water
ingression through the pores resulting marginal
increase in release rate. In addition, the low n and
k values implied that the PLA degradation did not
happen significantly within the span of testing per-
iod, otherwise release would have been much accel-
erated resulting in an increased n and k values [41].
Active–polymer interactions: analysisof thermal transitions by DSCand spectroscopic investigations by Ramanmicroscopy
It was evident from the release study that there exists
a strong interaction between thymol and PLA matrix.
In order to investigate the type of interactions
between active and polymer, the following investi-
gations were executed. Figure 3 displays the DSC
heating traces of samples A–E. The second heating
traces were recorded after removing the thermal
history of all the samples at 200 �C. All the experi-
mentally determined values of phase transition tem-
peratures such as Tg (glass transition temperature),
Tcc (cold crystallization temperature) and Tm (melting
point) are displayed in Table SII. Introduction of
thymol (melting point 49 �C) into PLA matrix sig-
nificantly affects the crystallization temperature (Tcc).
However, Tg of all samples remains almost unaf-
fected even after the addition of essential oils,
implying the absence of any plasticization/anti-
plasticization effect induced by thymol and essential
oils [11, 42]. Probably, molten thymol acts as sticky
additive which sticks to polymer melt and decreases
the chain mobility to inhibit crystallization and hence
Table 2 Release kinetic parameters (n, k) and correlation
coefficient (R2) calculated by fitting experimental results (Fig
S2) to Eq. 1
Sample Name n k R2
Sample A 0.174 0.34 0.976
Sample B 0.151 0.437 0.965
Sample C 0.145 0.399 0.965
Sample D 0.149 0.387 0.963
Sample E 0.153 0.373 0.968
Figure 3 DSC thermograms of pure PLA, samples A, B, C, D
and E.
J Mater Sci
significantly lowers the crystallinity due to strong
interaction between additive and polymer chain [43].
Furthermore, the absence of melting peak of crystal-
lites of thymol indicates the inhibitory effect on thy-
mol crystallization induced by PLA matrix due to
appreciable interaction between the two [44, 45]. In
addition to that, thymol- and EO-loaded PLA parti-
cles exhibit one broad melting peaks instead of two
peaks observed for pure PLA samples (because of
polymorphic crystalline transition [46, 47]). This
indirectly implies the inhibition of PLA crystalliza-
tion due to significant attractive forces between active
(primarily thymol) and polymer matrix. To further
investigate about the possible interaction between the
thymol and PLA matrix, Raman microscopy was
carried out for all the samples (Fig. 4). After analyz-
ing all the Raman spectra obtained from samples A–
E, noticeable shift of peaks was observed for both
PLA and thymol portions. For example, carbonyl
stretching frequency of PLA has been shifted from
1772 cm-1 (pure PLA) to * 1767 cm-1 for thymol-
and EO-loaded PLA particles presumably due to
H-bonding between ester group of PLA and hydroxyl
group of thymol. Similarly, shifting of characteristic
peaks from thymol portions such as 957 cm-1 (aro-
matic =C–H stretching from thymol) to 961 cm-1,
1262 cm-1 (–C–O stretching from thymol) to
1273 cm-1 and 1624 cm-1 (aromatic –C=C stretching
of thymol) to 1620 cm-1, respectively, for active loa-
ded particles is clearly evident in Fig. 4. Shifting of
these peak positions definitely indicated the preva-
lence of hydrophobic interactions between the aro-
matic ring of thymol and polymer matrix in addition
to H-bonding between them [48, 49].
Evaluation of antibacterial activity
As per literature information, thymol shows antibac-
terial activity against various Gram-positive and
Gram-negative bacteria [36, 50–52]. Herein, E. coli
(Gram-negative bacteria) was chosen to demonstrate
thymol’s antibacterial activity and MIC (minimum
inhibitory concentration) was determined and found
to be 350 lg/mL. This value was in close agreement
with the literature value reported by Cosentino and
group [36] who found a range from 225 to 450 lg/mL
as MIC for E. coli. In order to demonstrate the
antibacterial activity of thymol-loaded PLA particles,
appropriate concentration of thymol-loaded particles
(samples A–E) was first found out at which complete
inhibition of bacterial growth (E. coli) takes place using
plate spreading method (Fig S5) and summarized in
Table SIII. It was evident from these values that con-
centration of thymol-loaded particles was enormously
high compared to the concentration of free thymol due
to the slow release of thymol in water from PLA par-
ticles within the span of testing period (24 h). Similar
observation was made by Esposti et al. [37], who
demonstrated antibacterial activity of thymol released
from acrylic resin. Interestingly, among all the sam-
ples, thymol along with EO-loaded particles showed
lower particle concentration than that of particles with
no oil (sample A). This definitely implied the syner-
gistic antibacterial action of EOs along with thymol
[53]. Among all the EO employed, olive oil displayed
theminimumparticle concentration and hence sample
E can be considered as the most potent particles in
terms of antibacterial action. As the amount of thymol
released was almost equal in all cases within the span
of testing time (24 h), it can be assumed that the con-
siderable difference of reduction in particle concen-
tration among the samples was primarily originated
from essential oils which are well-known natural
antibacterial even at very low concentration. TheFigure 4 Raman spectra for pure thymol, pure PLA, samples A,
B, C, D and E starting from bottom to top.
J Mater Sci
strong antimicrobial activities in olive oil were mainly
due to the presence of long-chain a,b-unsaturatedaldehydes [54], and the reported MIC value of this oil
ranges from 10 to 60 lg/mL [55, 56] for complete
inhibition of bacterial growth. Similarly, the antimi-
crobial activity of mustard oil was attributed to the
presence of allyl isothiocyanate (AITC) [57], MIC of
which ranges from 50 to 1000 lg/ml for complete
inhibition of bacteria [57, 58]. Furthermore, it was also
reported that castor and coconut oil too have the
antimicrobial property because of the presence of
several fatty acids like Ricinoleic acid [59], lauric acids
[60], etc., and their MIC values were found to be in the
range of 50–250 lg/mL [61] and 60–128 lg/mL [62],
respectively. Therefore, it can be assumed that the
antibacterial effect of these oil-loaded microparticles
was due to synergistic combinations of thymol along
with the small amount of EOs [49]. The prolonged
antibacterial activity was also measured for all the
samples against E. coli over a period of[ 1 week. The
data accumulated from the antibacterial tests were
calculated by the ANOVA and student t tests, keeping
the significant level\ 0.05, 0.01 and 0.001 on various
samples (Fig. 5, Fig S6). Significant reduction in bac-
terial growth was observed for all the samples (except
sample A) till 11 days due to appreciable release of
thymol from porous particles (Fig. 5, Fig S6). More-
over, particles having no EO showed bacterial growth
inhibition only till 5 days due to low release of thymol
(especially after initial burst) in addition to no-oil effect
from non-porous particles (Fig S6). Apart from this,
E. coli was also incubated with the samples having no
thymol (Fig S7) and it was found that the bacterial
growthwas slightly less than the control. These results
implied that the enhanced antibacterial activity was
only shown by those particles which contain both
essential oil and thymol probably due to synergistic
action of the two actives. Overall, it can be inferred that
continuous and most efficient bacterial growth inhi-
bition effect was observed over a week from porous
PLA particles having olive/mustard oil (with low
particle concentration for complete inhibition) along
with thymol loaded into it.
Conclusions
A facile method based on W/O/W double emulsion
technique was developed for the fabrication of thy-
mol-encapsulated PLA-based microparticles. In order
to accelerate the release of thymol inwater, pores were
generated onto the surface of microparticles by
incorporating minute quantity of essential oils (castor,
olive, mustard and coconut) of varying viscosity.
Interestingly, it was realized that the pore density and
pore size can be easily varied by altering the viscosity
of added essential oils, e.g., being themost viscous EO,
castor oil containing particles showed maximum pore
density and pore size. Presumably, stabilization of
inner aqueous phase by essential oil droplet (phase
separated) along with PVA and PLA was considered
to be responsible for the surface pore formation. The
release mechanism of thymol from these variably
porous particles canwell be explainedwith the help of
power lawmodel. Furthermore, the present study also
demonstrated the effectiveness of thymol-loaded
particles entrapped with essential oils as antibacterial
carrier with enhanced inhibitory activity probably due
to synergistic action of thymol and essential oils. Since
the particles made of completely natural ingredients
displayed sustainable antibacterial activity over
10 days, these can be attractive as active food pack-
aging material for prolonging the food shelf life.
Figure 5 Antimicrobial activity of samples A, B, C, D and E
against E. coli expressed in bacterial growth inhibition %, and the
data are processed using ANOVA one-way statistical analysis and
are significant at p value\ 0.001.
J Mater Sci
Acknowledgements
The research leading to these results has received
funding from the Department of Science and Tech-
nology (DST), India, under Extramural Research
Grant: SB/S3/CE/068/2015, and IIT Delhi, New
Delhi, India.
Electronic supplementary material: The online
version of this article (https://doi.org/10.1007/s108
53-019-03593-7) contains supplementary material,
which is available to authorized users.
Compliance with ethical standards
Conflict of interest Authors have no conflict of
interest to declare.
References
[1] Tajkarimi MM, Ibrahim SA, Cliver DO (2010) Antimicrobial
herb and spice compounds in food. Food Control
21(9):1199–1218
[2] Cammack R, Joannou CL, Cui X-Y, Torres Martinez C,
Maraj SR, Hughes MN (1999) Nitrite and nitrosyl com-
pounds in food preservation. Biochim Biophys Acta
Bioenerg 1411(2):475–488
[3] Abdulmumeen HA, Risikat AN, Sururah AR (2012) Food:
its preservatives, additives and applications. Int J Chem
Biochem Sci 1:36–47
[4] LaCoste A, Schaich KM, Zumbrunnen D, Yam KL (2005)
Advancing controlled release packaging through smart
blending. Packag Technol Sci 18(2):77–87
[5] Appendini P, Hotchkiss JH (2002) Review of antimicrobial
food packaging. Innov Food Sci Emerg Technol
3(2):113–126
[6] Wyrwa J, Barska A (2017) Innovations in the food pack-
aging market: active packaging. Eur Food Res Technol
243(10):1681–1692
[7] Chung Donghwan, Chikindas Michael L, Yam KL (2001)
Inhibition of Saccharomyces cerevisiae by slow release of
propyl paraben from a polymer coating. J Food Protect
64(9):1420–1424
[8] Ha JU, Kim YM, Lee DS (2001) Multilayered antimicrobial
polyethylene films applied to the packaging of ground beef.
Packag Technol Sci 14(2):55–62
[9] Sharma GK, Madhura CV, Arya SS (1990) Interaction of
plastic films with foods. II. Effect of polyethylene and
polypropylene films on the stability of vegetable Oils. J Food
Sci Technol 27:327–331
[10] Scarfato P, Avallone E, Galdi MR, Di Maio L, Incarnato L
(2017) Preparation, characterization, and oxygen scavenging
capacity of biodegradable a-tocopherol/PLA microparticles
for active food packaging applications. Polym Compos
38(5):981–986
[11] Biswal AK, Usmani M, Ahammad SZ, Saha S (2018)
Unveiling the slow release behavior of hollow particles with