food and bioproducts processing 1 0 4 ( 2 0 1 7 ) 137–146

Contents lists available at ScienceDirect

Food and Bioproducts Processing

journal homepage: www.elsevier.com/locate/fbp

Nanoencapsulation of passion fruit by-products

extracts for enhanced antimicrobial activity

Daniela A. Oliveira a,b , Mariana Angonese a , Sandra R.S. Ferreira a , Carmen

L. Gomes b,∗
a Chemical and Food Engineering Department, Federal University of Santa Catarina, EQA/UFSC, C.P. 476, CEP
88040-900 Florianópolis, SC, Brazil
b Department of Biological & Agricultural Engineering, Texas A&M University, College Station, TX 77843-2117, USA

a r t i c l e i n f o a b s t r a c t

Article history: Large amounts of passion fruit residues are underutilized by juice industries that can poten-
Received 8 October 2016 tially be a source of bioactive compounds including antimicrobials. Co-precipitation with
Received in revised form 25 May biodegradable polymers, including poly (dl-lactide-co-glycolide) (PLGA), may be used to
2017 enhance these compounds bioactivities and provide controlled release. This study aimed to
Accepted 29 May 2017 produce PLGA particles of passion fruit by-products (seed and cake) extracts with different
PLGA lactide to glycolide (50:50 and 65:35) ratios using the emulsion/solvent evapora-
tion method (ESE). Characterization analyses indicated extracts encapsulation, controlled
Keywords: release and spherical shape for most treatments. Particle sizes ranged from 355 to 470 nm
Passion fruit seed and entrapment efficiencies (EE) from 23.8 to 79%. The Gompertz model for bacterial growth
Passion fruit residue fitted well the extracts release data. Results suggest that PLGA 65:35 is more suitable to
Controlled release both extracts encapsulation. Although, for the cake extract, EE was significantly lower than
Co-precipitation the seed extract. The ESE encapsulation using both PLGAs notably enhanced both extracts
Natural antimicrobial antimicrobial activities.
PLGA nanoparticles © 2017 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Escherichia coli
Listeria innocua

1. Introduction ious applications in the food, pharmaceutical and cosmetic industries,

Brazil is the world’s largest producer and consumer of fresh and pro-
cessed yellow passion fruit (Passilfora edulis f. flavicarpa), accounting for
50–60% of the total world production, which was estimated to be in 1.4
metric tons in 2013 (USAID, 2014). The production of concentrated juice
has the highest economic impact for the yellow passion fruit market as
its demand is growing worldwide (USAID, 2014, 2011; von der Linden,
2007). The great volume of juice produced generates a large amount of
by-products such as seeds and rind, but few studies can be found in the
literature concerning the reuse of seeds (Ferreira et al., 2011; Malacrida
and Jorge, 2012; Piombo et al., 2006). Seeds represent 4–12% of passion
fruits composition and contain around 30% oil (Malacrida and Jorge,
2012), and great part of this by-product is still wasted. Part of it has been
used to produce seed oil by conventional mechanical press finding var-
due to its high content of unsaturated fatty acids, especially linoleic
acid (up to 70%) (Ferreira et al., 2011; Malacrida and Jorge, 2012; Piombo
et al., 2006). The seed cake, a by-product derived from the cold press
of the seeds from oil production, can be a potential source of bioac-
tive compounds (anthocyanins, flavonoids, carotenoids, and phenolics)
including antioxidants, antimicrobials and antitumor compounds as
they are originally present in the seeds (Ferreira et al., 2011; Malacrida
and Jorge, 2012; Sano et al., 2011; Silva et al., 2014a,b).
In a previous study, we demonstrated the antioxidant and antimi-
crobial activities of passion fruit seed and seed cake extracts obtained
by different methods and solvents (Oliveira et al., 2016). Natural antiox-
idants and antimicrobials are an excellent way to improve food quality
and shelf-life without introducing undesirable chemical preservatives.
These materials can also be useful for cosmetics and pharmaceuti-
cal products; however, exposure to oxygen, heat and light can cause

∗
Corresponding author at: Biological & Agricultural Engineering Department, Texas A&M University, College Station, 303B Scoates Hall,
TX 77843-2117, USA. Tel.: +1 979 845 2455.
E-mail address: carmen@tamu.edu (C.L. Gomes).
http://dx.doi.org/10.1016/j.fbp.2017.05.009
0960-3085/© 2017 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
138 food and bioproducts processing 1 0 4 ( 2 0 1 7 ) 137–146
loss or reduction of extracts’s bioactivities. Thus, co-precipitation with
polymers may be used to preserve active compounds.
Co-precipitation or encapsulation with polymer have been largely
applied to protect sensitive substances from adverse effects of the
surrounding environment, mask odor or taste, control and target the
delivery of active compound, improve aqueous solubility, among oth-
ers. The polymer used plays an important role on the final product
quality as, besides the protection, it determines how the encapsulated
compounds will be released (Weiss et al., 2006). Poly(dl-lactide-co-
glycolide) (PLGA) is a biocompatible and biodegradable polymer, FDA
approved, that degrades by hydrolysis of the ester backbone into the
homopolymers of lactic acid and glycolic acid, known products of cel-
lular intermediary metabolism, meaning non-harmful and non-toxic
compounds (Anderson and Shive, 1997; Astete and Sabliov, 2006; Park,
1995; Stevanović and Uskoković, 2009). Its application could be useful
in the food industry to disperse hydrophobic compounds in hydrophilic
compositions (Hill et al., 2013).
Extensive research led to a full range of PLGA polymers currently
commercially available. The ratio of monomers glycolide to lactide at
different compositions determines the degree of crystallinity and the
rate of degradation of the polymers. The polymerization of crystalline
PGA with PLA reduces the degree of crystallinity leading to increases in
rates of hydration and hydrolysis. In general, the higher the content of
glycolide, the quicker the rate of degradation (Cohn and Younes, 1987;
Park, 1995).
Considering the potential bioactivity of the passion fruit by-
products extracts and their potential applications in food and
pharmaceuticals, the aim of this study was to produce PLGA parti-
cles of passion fruit by-products (seed and cake) extracts with two
different lactide to glycolide (50:50 and 65:35) ratios of PLGA composi-
tion using the emulsion/solvent evaporation method (ESE) and evaluate
their physico-chemical properties (size, morphology, entrapment effi-
ciency, in vitro release) and their antimicrobial activities.
2. Methods
2.1. Raw materials and reagents
The raw materials were supplied by the company Extrair Óleos
Naturais located in the State of Rio de Janeiro, Brazil. This
company collects and processes the by-products from passion
fruit juice producers. Cleaning (process under patent registra-
tion) and drying (60 ◦ C in rotational drier) are performed on
the same day. The dried seeds are used by the company to
produce passion fruit seed oil by cold pressing. After the oil
production, the remaining by-product is a brown dry powder
called seed cake.
Seeds and seed cake were shipped to Federal University of
Santa Catarina, SC, Brazil, where they were stored at −18 ◦ C
until their use. Upon arrival, the passion fruit seeds and
seed cake presented a moisture content of 8.50 ± 0.09% (w/w)
and 6.26 ± 0.06% (w/w), respectively, determined according to
AOAC method 940.26 (AOAC International, 2005). Extracts were
selected from a previous study of Oliveira et al. (2016), aiming
to have one sample of each raw material that showed a good
combination of extraction yield and biological activity, mainly
antimicrobial. Briefly, the seeds were ground in a domestic
blender (Black & Decker, SP, Brazil) prior to extraction with
supercritical carbon dioxide (SC-CO2 ) at 150 bar, 40 ◦ C and con-
stant solvent flow rate of 0.5 kg CO2 /h. While the seed cake did
not require any pre-treatment to be macerated with a mixture
of ethanol and water (1/1, v/v) (EtOH-H2 O).
PLGA, with copolymer ratios of dl-lactide to glycol-
ide of 50:50 (Mw 30,000–38,000 g/mol, Resomer RG 503
ester terminated, Evonik, Essen, Germany) and 65:35 (Mw
40,000–75,000 g/mol, Sigma–Aldrich, St. Louis, MO) were used
in this study. The surfactant used in the emulsification
process was polyvinyl alcohol (PVA) (98%–99% hydrolysis
degree and average Mw 30,000–50,000 g/mol, Sigma–Aldrich).
Dichloromethane, 99% acetonitrile (HPLC grade), d(+)trehalose
98%, tryptic soy broth (TSB), tryptic soy agar (TSA), and pep-
tone water were obtained from VWR International (West
Chester, PA). All other reagents were of analytical grade and
were used as obtained from the supplier without further pro-
cessing.
2.2. Particle synthesis
Particles were produced using the emulsion solvent evapora-
tion (ESE) method similar to the method outlined by Gomes
et al. (2011). First, the organic phase was prepared by dis-
solving 50 mg of PLGA into 2 mL dichloromethane along with
16% (w/w relative to PLGA) of each extract. The organic phase
was added drop-wise to 20 mL of aqueous 0.5% (w/w) PVA
solution in 0.2-␮m filtered (Nalgene Filtration Products) water.
This mixture was emulsified for 2 min at 9500 rpm using
an Ultra-Turrax T25 basic Ika (Works, Inc., Wilmington, NC).
The emulsion was sonicated in an ice bath for 30 min at
70 W (Cole Parmer sonicator 8890, Vernon Hill, IL), before
removing the dichloromethane using a rotoevaporator (Buchi
R-210 Rotavapor, Buchi Co., New Castle, DE) under vacuum
(0.97 psi). Unloaded (control) particles were produced by the
same method without adding extract in the organic phase.
Dichloromethane was selected to prepare the particles since
it is the most volatile solvent among the ones that solubilize
PLGA, and it is a commonly used solvent for PLGA nanoen-
capsulation process (Astete and Sabliov, 2006). Then, the
particles were purified by ultrafiltration to remove excess PVA
and non-encapsulated extracts using a Millipore-labscaleTM
TFF system fitted with a 50 g/mol molecular weight cutoff
Pellicon XL-Millipore (Millipore Co., Kankakee, IL). The par-
ticles were ultrafiltered with 200 mL of water, inlet pressure
of 30 psi and outlet pressure of 10 psi, collecting the reten-
tate when it reached 50 mL. After ultrafiltration, d(+)-trehalose
was added to the particle solution at a 1:1 ratio relative
to the PLGA to act as a cryoprotectant. The particles solu-
tion was kept at −20 ◦ C until being lyophilized at −50 ◦ C and
9.67 × 10−5 psi vacuum in a Labconco Freeze Dry-5 unit (Lab-
conco, Kansas City, MO). Dried particles were collected and
stored in a desiccator at −20 ◦ C for further analysis. It is rele-
vant to consider dichloromethane residue in the particles after
their synthesis prior to any approval for food or pharmaceu-
tical applications. Furthermore, most of the dichloromethane
should be removed during the rotoevaporation step, and even
if some dichloromethane traces would remain, they should be
removed on the subsequent steps including ultrafiltration and
freeze drying.
2.3. Particle size analysis and morphology
Particle size measurements were obtained by dynamic light
scattering (DLS) in a Nano–ZS90 ZetaSizer (633 nm He-Ne 200,
Malvern, UK). Particles were dispersed in 0.2 ␮m-filtered dis-
tilled water at a concentration of 1 mg/mL and agitated until
solubilized before analysis using 1-cm path length disposable
polystyrene cuvettes at scattering angle of 90◦ , refractive index
of 1.590 and temperature of 25 ◦ C.
Aqueous suspensions of particles were examined using
a FEI Morgagni Transmission Electron Microscope (TEM) (FEI
Co., Hillsboro, OR) at the School of Veterinary Medicine and
 food and bioproducts processing 1 0 4 ( 2 0 1 7 ) 137–146
Biomedical Sciences of Texas A&M University (College Station,
TX). Aqueous suspensions of particles at a concentration of
1 mg/mL were placed on 200 mesh carbon filmed copper grids
and stained with a 2% (w/v) phosphotungstic acid aqueous
stain (Electron Microscopy Sciences, Hatfield, PA) to provide
contrast under magnification. Excess liquid on the mesh was
removed with filter paper and the grid was allowed to dry
before viewing up to 110,000 times magnification at 80 kV.
2.4. Entrapment efficiency
The amount of each extract entrapped in the PLGA nanoparti-
cles was determined spectrophotometrically (model Genesys
10S UV/vis, Thermo Scientific, Madison, WI) at 294 nm (deter-
mined as the maximum absorbance in preliminary tests
by scanning the UV–vis spectra) based on the procedure
described by Silva et al. (2014a). The particle samples were dis-
solved in 95% (w/v) acetonitrile at a concentration of 1 mg/mL
and left in solution for 12 h with periodic mixing to allow
enough time for all entrapped extracts to be in solution.
The solutions were passed through 0.2-␮m polypropylene
membrane syringe filters (Acrodisc, Pall Life Sciences, Port
Washington, NY) to remove all PLGA and PVA prior to measure-
ments in a 1-cm path length quartz cuvette. A standard curve
of each extract was prepared under the same conditions to
correlate the entrapped extract quantity. The entrapment effi-
ciency was expressed as the percentage of extract entrapped
in the nanoparticles reported to the initial amount used for
nanoparticles preparation.
2.5. Differential scanning calorimetry (DSC)
DSC studies were performed using a Perkin-Elmer DSC 6
with ISOTEMP 1028P (Fischer Scientific, Pittsburg, PA) with
an integrated cooling system (Pyris 5 Software, Boston, MA).
Samples of free extracts, both copolymers ratio PLGAs and
PLGA nanoparticles with entrapped extracts were accurately
weighed (∼5 mg) and placed into hermetically sealed and
crimped 20-␮L aluminum pans with one hole in the lid. The
instrument was calibrated using zinc and indium metals
before sample testing. The samples were cooled from room
temperature to 2 ◦ C under nitrogen atmosphere at 2 ◦ C/min,
remained at 2 ◦ C for 1 min and then heated up to 442 ◦ C at
20 ◦ C/min.
2.6. Controlled release
Controlled release experiments were conducted to observe
the rate of release of extracts from the PLGA particles sim-
ilarly as reported by Zigoneanu et al. (2008). The release
medium consisted of phosphate buffered saline (PBS, 0.15 M,
pH 7.2) and after addition of the particles at a concentration
of 1 mg/mL, the solution was divided into 1.5 mL aliquots in
Eppendorf tubes. The tubes were placed in a shaking water
bath (VWR Intl., West Chester, PA) set at 100 rpm and 35 ◦ C,
from which samples were removed and analyzed for extract
content at predetermined time points up to 72 h. Samples
were filtered using 0.2-␮m polypropylene membrane syringe
filters (Acrodisc, Pall Life Sciences, Port Washington, NY) prior
to spectrophotometric analysis at 294 nm. The temperature
(35 ◦ C) and buffer pH (7.2) were chosen as an ideal case for
microbial growth.
In this study, we are proposing an adaptation of the Gom-
pertz model, commonly used bacterial growth model (Baty
139
and Delignette-muller, 2004), to describe the release of the
passion fruit by-product extracts from PLGA nanoparticles:
   
( + t)
y(t) = yo + (ymax − y0 ) exp − exp 1+ max (e1 ) (1)
(ymax − yo )
where y0 is the initial percentage of extract released for the
first steady release period (called here as lag phase by associa-
tion with the microbial growth); ␭ is the lag phase duration
(h); ␮max is the maximum release rate (slope) (% extract
released/h) and ymax is the maximum percentage of extract
released. Microsoft Excel 2016 (Microsoft Office 365 ProPlus,
7870.2031) was used to fit the data.
2.7. Antimicrobial activity
2.7.1. Bacterial cultures
Escherichia coli (ATCC 35218) and Listeria innocua (NRRL/USDA
B33076) were obtained from Texas A&M University Food
Microbiology Laboratory culture collection (College Station,
TX). These strain selections are representative of typical
Gram-negative and Gram-positive pathogenic microorgan-
isms, respectively; commonly occurring in various food
products. E. coli and L. inoccua were resuscitated in TSB (tryptic
soy broth) and TPB (tryptose phosphate broth); respectively,
by identical duplicate transfers and incubated aerobically for
24 h at 35 ◦ C. The bacterial cultures were maintained on TSA
(tryptic soy agar) and TSAYE (tryptic soy agar enriched with
0.6% yeast extract) slants at 4 ◦ C. Transfers from slants were
conducted similarly to the resuscitation method to prepare
microorganisms for analysis.
2.7.2. Minimum inhibitory and bactericidal concentration
Minimum inhibitory concentrations (MICs) for the extracts
were determined using the microdilution method in cul-
ture broth (Brandt et al., 2010). E. coli and L. innocua cultures
were incubated 24 h and then prepared by serial dilution in
double-strength TSB (2xTSB) and TPB (2xTPB), respectively,
for an initial inoculum of approximately 5.0 log10 CFU/mL. Ini-
tial inocula were enumerated via spread plating on TSA and
TSAYE; respectively, and incubated for 24 h at 35 ◦ C.
MIC analyses were carried out in a microdilution plate
with 96 wells (sterilized 300 ␮L-capacity, MicroWell, NUNC,
Thermo-Fisher Scientific, Waltham, MA). The particles were
dissolved in 0.2-␮m filtered and sterile water. The initial
solutions of the particles were prepared and, further, serial
dilutions of the extracts were performed to reach a final
extract concentration range between 261 to 41 ␮g/mL for the
EtOH-H2 O maceration seed cake extract particle and 105 to
746 ␮g/mL for the supercritical CO2 seed extract particles.
These concentrations were calculated based on the initial
extracts quantity used to form the particles as well as the
encapsulation efficiency results obtained. Unloaded particles
were also analyzed to assure other materials such as PLGA,
PVA, etc. had no inhibitory effect on bacterial growth. Positive
controls (growth control) were prepared containing inocu-
lum and sterile water (without extracts). Negative controls
were also prepared with antimicrobial solutions and sterile
2× broth to ensure they were not contaminated. Once plates
were prepared, they were covered with Mylar plate sealer
(Thermo-Fisher Scientific) and OD at 630 nm was read (0 h)
in an Epoch BioTek UV/vis range spectrophotometer (VERSA
Max, Molecular Devices, Sunnyvale, CA). Any antimicrobial
test well that showed ≤0.05 change in OD at 630 nm were
considered “inhibited” by the antimicrobial (after appropriate
140 food and bioproducts processing 1 0 4 ( 2 0 1 7 ) 137–146
baseline adjustments). The MIC was determined as the low-
est concentration of an antimicrobial that inhibited the visible
growth of a microorganism after 24 h incubation.
All wells that exhibited inhibition of the test microorgan-
ism were then tested for bactericidal capability by spreading
0.1 mL of their content onto TSA or TSAYE plates and incu-
bated for 24 h at 35 ◦ C. If no colonies were observed on the
plate surfaces following incubation, the treatment concen-
tration was considered bactericidal. The lowest concentration
of antimicrobial demonstrating bactericidal activity across all
replicates was considered the minimum bactericidal concen-
tration (MBC) (Hill et al., 2013).
2.8. Statistical analysis
For all analyses, determinations were made at least in dupli-
cate as independent experiments based on a completely
randomized design with equal replications. The particles size
and entrapment efficiency results were statistically evaluated
by using the MEANS and ANOVA procedures of SAS version
9.3 (SAS Ins. Inc., Cary, NC) with Tukey’s Honestly Significant
Differences (HSD) Test when significant differences at level of
5% were observed.
3. Results and discussion
3.1. Particle size and morphology
Particles size and morphology play a key role in the target
component release dynamics (rate and concentration) and
its uptake (adhesion and interaction with cells) (Stevanović
and Uskoković, 2009). The results of the transmission elec-
tron microscopy (TEM) performed to the precipitated particles
obtained with the passion fruit by-products extracts and PLGA
50:50 and PLGA 65:35 are shown in Fig. 1. As the shape of all
particle samples is very similar in all images obtained, dif-
ferent amplifications were selected to allow the visualization
of both morphology and tendency to form clusters. Table 1
presents the average particle size of both passion fruit by-
product extracts encapsulated with both polymers.
Particles size ranged from 269 to 470 nm and almost all
treatments provided similar (p < 0.05) results, except for the
unloaded PLGA 50:50 that showed no significant difference
(p < 0.05) only to the seed SC-CO2 150 bar/40 ◦ C/PLGA 50:50.
Silva et al. (2014a) also observed little influence of different
extracts on the particle sizes. Also the difference detected in
nanoparticles size was minimal when different co-polymer
ratios were used. Konan et al. (2003) obtained nanoparticles
with a mean size of 93 nm for 50:50 PLGA and 95 nm for 75:25
PLGA. On the other hand, Silva et al. (2014a) showed that
the particles formed with PLGA 65:35 were significantly larger
(p < 0.05) than the ones produced with PLGA 50:50, which was
due to the higher molecular weight of PLGA 65:35, between
40 to 75 kDa compared to 30 to 38 kDa for PLGA 50:50, as it is
usually reported in the literature that higher molecular weight
polymers yield larger nanoparticles (Astete and Sabliov, 2006;
Stevanović and Uskoković, 2009).
The TEM images (Fig. 1) show that all particles synthetized
by ESE method have similar spherical shape, smooth surface
and tendency to form clusters. In general, regardless of the
preparation method, the aggregation of PLGA particles is a
notable problem. In order to prevent this, polymer stabiliz-
ers, which are amphiphilic molecules with both hydrophilic
and hydrophobic parts are often used. However, size and
shape of the particles can also be influenced by the stabi-
lizer used (Stevanović and Uskoković, 2009). For instance, PVA
is a large molecule leading to a large nanoparticle surface;
nevertheless, the sizes observed in this study are in accor-
dance with the literature as most of studies covering this
method present spherical particles with size between 130 and
600 nm (Astete and Sabliov, 2006). In Fig. 1b and d, the parti-
cles bright color surrounded by a dark contour is associated
with the PVA and indicate encapsulation. The PVA chain is
formed from alternating hydrophilic and hydrophobic seg-
ments. The latter interconnects with the PLGA chains to create
a matrix, whereas the affinity of PVA with water pulls its
hydrophilic regions to the surface of micelles, being exposed
to the water phase. At the same time, the hydrophobic seg-
ments are directed into the inner core of the micelles (Gomes
et al., 2011; Zigoneanu et al., 2008).
In the ESE method, particles size varies due to different
factors such as organic solution viscosity; homogenizer speed
and sonication power; type and concentration of solvents,
interest compound to be encapsulated and emulsifiers (Astete
and Sabliov, 2006; Mainardes and Evangelista, 2005). Therefore,
depending on the application and release profile desired, the
process can be adjusted to reach a suitable particle size.
3.2. Entrapment efficiency
The entrapment efficiencies of passion fruit by-products (seed
and cake) extracts in PLGA by ESE method are presented
in Table 1. The entrapment efficiency depends on the affin-
ity and interaction between the materials involved (extracts,
polymers and emulsifier) (Astete and Sabliov, 2006). The cake
extract was less (p < 0.05) efficiently entrapped than the seed
extract, 23.8 ± 0.7% and 27 ± 2% against 61 ± 1% and 79 ± 2% for
the PLGA 50:50 and 65:35, respectively. This is probably due to
the higher hydrophilicity of the cake extract (extracted with
polar solvent), meaning it has a higher affinity to the solution
aqueous phase, consequently hampering its interaction with
the hydrophobic PLGA chain. The difference in hydrophobic-
ity between PLGA ratios also play a role on the entrapment
efficiency. PLGA 65:35 is more hydrophobic than PLGA 50:50
due to its higher lactide composition, and presented higher
entrapment efficiency for both extracts, significant in the case
of the seed extract (79 ± 2%). Silva et al. (2014a) reported sim-
ilar behavior of higher entrapment efficiencies for PLGA 65:35
than PLGA 50:50 with passion fruit and guava extracts. How-
ever, other factors might interfere on entrapment efficiency
that are specific to each extract, such as compounds steric
conformation. This was observed by Gomes et al. (2011) when
studying two phenolic compounds with similar hydropho-
bicity and molecular weight (cinnamaldehyde and eugenol)
entrapped in PLGA 65:35 showing different (p < 0.05) values for
entrapment efficiencies. Since the particle size and morphol-
ogy of both polymers were similar, the entrapment efficiency
results can indicate PLGA 65:35 to be more suitable for the
encapsulation of passion fruit seed extracts. In the case of
cake extract, encapsulation was not significantly influenced
by the type of polymer used in this study.
3.3. Calorimetric profiles of particles
Fig. 2 presents the calorimetric profile of free passion fruit by-
products (seed and cake) extracts and loaded and unloaded
PLGAs particles obtained by ESE process. The DSC results
give an indication about changes on the polymer and active
 food and bioproducts processing 1 0 4 ( 2 0 1 7 ) 137–146 141

Fig. 1 – Micrographs obtained by transmission electron microscopy (TEM) of the seed and seed cake extracts particles
synthetized with poly-dl-lactide-co-glycolide (PLGA) with copolymer ratios of 50:50 and 65:35: a) supercritical fluid extract
of passion fruit seed and PLGA 50:50; b) supercritical fluid extract of passion fruit seed and PLGA 65:35; c) maceration
passion fruit seed cake extract and PLGA 50:50; d) maceration passion fruit seed cake extract and PLGA 65:35.

Table 1 – Average particle size and entrapment efficiency of poly-d,l-lactide-co-glycolide (PLGA) with copolymer ratios of
50:50 and 65:35 particles loaded with passion fruit by-products extracts (seed and cake).
Extract/treatment Size (nm) Entrapment efficiency (%)

PLGA 50:50 PLGA 65:35 PLGA 50:50 PLGA 65:35

Unloaded 269 ± 71
392 ± 412,3
– –
Seed SC-CO2 150 bar/40 ◦ C 355 ± 101,2 417 ± 102,3 61 ± 1b 79 ± 2a
Cake MAC EtOH-H2 O 381 ± 332,3 470 ± 153 23.8 ± 0.7c 27 ± 2c

Same superscript numbers or letters indicate no significant difference results (p < 0.05) between size or entrapment efficiency, respectively.
Tests were carried out in triplicate.
SC-CO2 : supercritical carbon dioxide; MAC: maceration; EtOH-H2 O: ethanol and water solution (1:1, v/v).

compound crystallinity as they can undergo modifications
during the particle formation process (Cocero et al., 2009;
Kiran et al., 2008). It can also be used to indicate the co-
precipitation performance, suggesting the effectiveness of the
target compound (extract) encapsulation inside the carrier
matrix (polymer) when the crystalline and/or the melting
peaks of the extract cannot be observed on the particle curve.
Otherwise, when the compound is not encapsulated, the
product is simply a mixture of segregated particles of target
compound and carrier and their peaks are normally observed
on the particle DSC as a superposition of the diagrams of each
substance, as discussed by Cocero et al. (2009). A peak near
45 ◦ C was observed for the PLGA 65:35, while the PLGA 50:50
peak occurred at 52 ◦ C. For the unloaded particles, synthetized
with all other materials on the process but the extracts, the
peaks dislocated to around 97 ◦ C, which can be associate with
some change on the molecular structure of the polymer during
the recrystallization process after the organic solvent evapo-
ration. According to Kiran et al. (2008), polymers can have their
crystallinity modified during the precipitation process, and
their changes can be detected by DSC analysis. This change
can also be related to the trehalose (cryoprotector added at the
142 food and bioproducts processing 1 0 4 ( 2 0 1 7 ) 137–146

4 pure seed extract

pure cake extract
PLGA 50:50
3.5 PLGA 65:35
PLGA 50:50 unloaded parcle
PLGA 65:35 unloaded parcle
3
seed extract + PLGA 50:50 parcles
cake extract + PLGA 50:50 parcles
Heat flux (mW/g)

2.5 seed extract + PLGA 65:35 parcle

cake extract + PLGA 65:35 parcle

1.5

0.5
0 50 100 150 200 250

Temperature (°C)

Fig. 2 – Calorimetric profile of free passion fruit by-product extracts (seed and cake extracts), poly-dl-lactide-co-glycolide
(PLGA) with copolymer ratios of 50:50 and 65:35 and particles obtained by the emulsion/solvent evaporation (ESE) process.

end of the process) with its peak reported in the literature at

Table 2 – Gompertz parameters values for the passion
106 ◦ C (Roe and Labuza, 2005). Furthermore, the oscillation on fruit by-products (cake and seed) extracts release from
the heat flux around 180 ◦ C for all particles can be associated poly-dl-lactide to glycolide (PLGA) particles with
with PVA (surfactant used on the process), as Patel et al. (2014) copolymer ratios of 50:50 and 65:35 determined by
reported its peak at 185 ◦ C. The extracts exhibited very differ- fitting the Gompertz model equation (Eq. (1)) (Baty and
ent behaviors from each other, evincing their compositional Delignette-muller, 2004) to the experimental data.
differences. The supercritical seed extract presented only a Particle y0 ymax ␮max ␭ R2
small change on the heat flux close to 17 ◦ C, while the mac-
Seed extract + PLGA 50:50 30.43 92.59 7.67 12.80 0.995
eration cake extract showed a larger heat flux oscillation in Seed extract + PLGA 65:35 41.41 92.14 6.31 16.03 0.989
a broad range of temperatures; thus for the cake extract two Cake extract + PLGA 50:50 39.68 92.50 5.05 12.84 0.994
peaks were considered for the particles comparison, close to Cake extract + PLGA 65:35 35.98 77.90 13.12 9.60 0.994
85.5 ◦ C and at 122.5 ◦ C. In the case of the cake extract + PLGA
y0 : initial percentage of extract released for the first steady release
65:35 particles, there is a proximity to the free extract peak at
period (lag phase) (%); ymax : maximum percentage of extract
85.5 ◦ C, consequently it is insufficient to confirm the complete
released (%); ␭: lag phase duration (h); ␮max : maximum release
extract encapsulation in this treatment. On the other hand, rate (slope) (% extract released/h); R2 : coefficient of determination.
the calorimetric profiles of the other loaded particle samples Parameters were fitted using the average experimental data from
were similar to the unloaded particles ones suggesting encap- three repetitions.
sulation.

3.4. Controlled release
The release profiles of passion fruit by-products extracts (seed
and cake) from PLGA nanoparticles were evaluated in vitro as
a function of time. The results over 72 h releasing time are
shown in Fig. 3. The release profiles for both passion fruit by-
product extracts from PLGA nanoparticles, both PLGA 50:50
and PLGA 65:35, were characterized by an initial burst release
in the first hour followed by a steady release plateau at a
lower level where the percentage of extract dispersed in the
medium remain nearly constant; thereafter, an accelerated
release between 12 and 24 h followed by another steady release
plateau at an upper level. The burst effect can be explained by
the fast release of active compound found close to or attached
to the surface of the nanoparticles or by the existence of super-
ficial pores initially at the polymeric matrix (Zigoneanu et al.,
2008). The following lower steady release plateau may be due
to a spontaneous reduction of size and number of pores at the
surface together with the low diffusion of the extract inside
the polymeric matrix (Wang et al., 2002; Zigoneanu et al., 2008).
After that, the accelerated release up to 24 h is probably related
to the erosion of the particles (Makino et al., 2000; Mu and
Feng, 2003). In 24 h, between 72 and 88% of the extract from its
respective nanoparticles was released from both PLGA 50:50
and PLGA 65:35 remaining about this level up to 72 h. These
percentages are comparable to other studies using PLGA and
ESE method such as Gomes et al. (2011) that reported release
up to 90% for one single compound.
All the release profiles were similar to a general bacteria
growth curve. Consequently, in general, the adapted Gompertz
model Eq. (1) fitted well to the experimental release data for
all nanoparticles. Table 2 presents the Gompertz parameters
values for the passion fruit by-products extracts release from
the PLGA nanoparticles and the coefficients of determination,
 food and bioproducts processing 1 0 4 ( 2 0 1 7 ) 137–146 143

a) 100

90

80
Released extract (%)

70

60

50

40

30
PLGA 50 seed extract
20 PLGA 65 seed extract
Gompertz model PLGA50/seed extract
10
Gompertz model PLGA65/seed extract
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75

Time (h)

b) 100

90

80

70
Released extract (%)

60

50

40

30
PLGA 50 cake extract
20 PLGA 65 cake extract
Gompertz model PLGA50/cake extract
10
Gompertz model PLGA65/cake extract
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75

Time (h)
Fig. 3 – Experimental and modeled release kinetics of passion fruit by-products extracts from poly-dl-lactide-co-glycolide
(PLGA) particles in phosphate buffer at 35 ◦ C: a) seed extract and b) cake extract using PLGA nanoparticles with copolymer
ratio of 50:50 (PLGA 50) and 65:35 (PLGA 65). Experimental curves represent the average ± standard deviation of three
repetitions.

R2 , which were all higher than 0.98. Thus, this model could
predict with good fitness the release behavior, the maximum
released extract percentage (ymax ) and the maximum release
rate (␮max ) of passion fruit by-products extracts from PLGA
polymers.
The release profile of the active compound is a combina-
tion of different effects, including penetration of the release
medium into the particle matrix, diffusion of the active com-
pound through the matrix, capsule wall thickness, affinity
of the active compounds with polymer, diffusion of active
compound through the polymer matrix, polymer swelling
and degradation (Sansdrap and Moes, 1997). Chemical struc-
ture, molecular weight, composition of the polymer, as well
as the synthesis conditions, are parameters which influence
the final morphology and porosity of the particle (Stevanović
and Uskoković, 2009). Biodegradation of hydrolysable poly-
mers proceeds in a diffuse manner, with amorphous regions
degrading prior to the complete split of crystalline and cross-
linked regions (Zhang et al., 2006). Generally, PLGA degradation
is slow, so active compound release is more dependent on dif-
fusion, PLGA swelling, and surface or bulk PLGA erosion (Mu
and Feng, 2003). The literature reports that copolymer with
144 food and bioproducts processing 1 0 4 ( 2 0 1 7 ) 137–146
Table 3 – Minimum inhibitory and bactericidal concentration (MIC and MBC) (␮g/mL) values of passion fruit by-products
(seed and cake) extracts and their respective particles with poly-dl-lactide to glycolide (PLGA) with copolymer ratios of
50:50 and 65:35 determined by the microdilution broth method against bacteria.
Extract Treatment
Seed SC-CO2 a
Free extract
150 bar/40 ◦ C PLGA 50:50 particles
PLGA 65:35 particles
Cake MAC EtOH-H2 O Free extracta
PLGA 50:50 particles
PLGA 65:35 particles
Extract Treatment
◦
Seed SC-CO2 150 bar/40 C PLGA 65:35 particles
Tests were carried out in triplicate. SC-CO2 : supercritical carbon dioxide; MAC: maceration; EtOH-H2 O: ethanol and water solution (1:1, v/v).
a
Oliveira et al. (2016).
b
Equivalent extract concentration in the PLGA particles.
50:50 ratio of monomers undergoes faster degradation in both
in vitro and in vivo conditions (Miller et al., 1977; Stevanović
and Uskoković, 2009); however, in the present study it was not
possible to observe this behavior. The maximum release rates
(Table 2) for the PLGA 50:50 were 7.67 and 5.05 and for the
PLGA 65:35 were 6.31 and 13.12% extract released/h to seed
and cake extract, respectively. Miller et al. (1977) have shown
that PLGA 50:50 is the fastest degrading composition, with
the degradation rate being decreased when either lactide or
glycolide content of the copolymer was increased. According
to Park (1995), the release kinetics from the microspheres are
hard to predict, but often they exhibited an initial burst release
followed by very slow release kinetics.
3.5. Antimicrobial activity
The minimum inhibitory concentrations (MICs) of passion
fruit by-products free extracts and their respective particles
with PLGA 50:50 and PLGA 65:35 against E. coli and L. innocua
growth are presented in Table 3. The concentrations were cal-
culated based on the initial extract quantities used to form
the particles as well as the encapsulation efficiency results
obtained. The concentrations that showed inhibition of the
tested organism were subsequently tested for bactericidal
effect to determine the minimum bactericidal concentrations
(MBCs). Unloaded particles were tested at the same concen-
trations as control to evaluate the antimicrobial effect of all
the other materials used on the process and did not exhibit
any antimicrobial activity at any concentration tested against
E. coli and L. innocua (data not shown).
In a previous study by Oliveira et al. (2016), it was reported
that the free extracts of the passion fruit seeds showed
inhibitory activity against L. innocua at a concentration of
8000 ␮g/mL, while the free cake extract inhibited the growth
of both microorganisms, E. coli and L. innocua, at 4000 ␮g/mL;
and none of them were bactericidal at the concentrations
tested. The high MIC values observed for free extracts are
a result of their low water solubility, since it is difficult for
them to interact with microorganisms in aqueous suspensions
(Ravichandran et al., 2011).
The passion fruit seed extract particles made with PLGA
50:50 were effective to inhibit the bacteria growth at equivalent
MIC (␮gextract /mL)b
Escherichia coli Listeria innocua
– 8000
585 421
– 373
4000 4000
226 226
261 188
MBC (␮gextract /mL)b
Escherichia coli Listeria innocua
– 537
extract concentrations of 421 and 585 ␮g/mL for L. innocua and
E. coli, respectively. The particles produced with the same seed
extract and PLGA 65:35 did not prevent E. coli growth at the con-
centrations tested but inhibited L. innocua growth at 373 ␮g/mL
and, even more relevant, presented a bactericidal effect at
537 ␮g/mL equivalent extract concentration. Comparing with
the antimicrobial result of the free extract, the encapsulation
with PLGA 50:50 and PLGA 65:35 allowed a reduction of 19
and 21 times, respectively; on the passion fruit seed extract
concentration necessary to inhibit L. innocua growth. For the
cake extract, the equivalent extract concentration necessary
to inhibit the growth of both bacteria tested was 226 ␮g/mL
when encapsulated in PLGA 50:50, a reduction of 18 times with
respect to the free extract; while the samples produced with
PLGA 65:35 inhibit E. coli and L. innocua growth at equivalent
extract concentrations of 261 and 188 ␮g/mL, respectively; cor-
responding to 15 and 21 times less than the free passion fruit
seed cake extract. These results indicate the benefits of encap-
sulation with PLGA since it reduces the required antimicrobial
concentration needed for inhibition; which is corroborated by
other studies on encapsulation of essential oils and phenolic
compounds with PLGA against foodborne pathogens (Gomes
et al., 2011; Hill et al., 2013; Ravichandran et al., 2011).
The MIC results show that the particles were slightly more
effective against the Gram-positive bacteria, L. innocua; com-
pared to the results for the Gram-negative one, E. coli. The
Gram-negative bacteria have more sophisticated permeabil-
ity barrier compared to the simpler cellular membrane of the
Gram-positive bacteria, which facilitates the penetration of
compounds (Smith-Palmer et al., 1998). This also explains the
reason for the seed supercritical extract, composed mostly of
fatty acids, being active only against L. innocua.
Since hydrophobic compounds and extracts are difficult to
access bacteria cells in aqueous environment, once they are
encapsulated in PLGA, they can more effectively be distributed
throughout the aqueous environment and delivered to the
bacteria cells (Ravichandran et al., 2011). Moreover, another
advantage of using PLGA is that it not only assists with dis-
tribution and delivery, but also as PLGA begins to degrade it
produces acid by-products (lactic and glycolic acids) (Wischke
and Schwendeman, 2008) that reduce the surrounding pH,
increasing the extracts hydrophobic nature and allowing them
 food and bioproducts processing 1 0 4 ( 2 0 1 7 ) 137–146
to more readily partition into the lipids of the cell membrane
or bind to hydrophobic regions of membrane proteins (Burt,
2004; Hill et al., 2013). Moreover, the presence of PVA on the
particles coating, a bioadhesive material, may improve their
adhesion and absorption into cells (Stevanović and Uskoković,
2009).
The release profile observed certainly had a positive impact
on the antimicrobial activity of the passion fruit by-product
extracts. The consistent release of compounds and the sim-
ilarity of the extracts release behavior to the usual bacteria
growth curve were effective in preventing microorganism
growth maintaining the antimicrobial compounds concentra-
tion at a level that allowed an inhibitory activity over a longer
period of time. This is relevant since the main goal of the
nanoencapsulation of the passion fruit by-products extracts
is their application in pharmaceuticals and food systems. Fur-
thermore, as temperatures lower than 35 ◦ C are typically used
for food storage, likely the release rates will be slower as well as
the microbial growth rate, consequently, the product’s shelf-
life is expected to be extended.
4. Conclusions
Considering the great amount of passion fruit juice produced
in Brazil and the scarce number of studies concerning its
by-products reuse, this work demonstrates opportunities to
add value to these agro industrial wastes. Applying the emul-
sion/solvent evaporation method, both PLGA applied (50:50
and 65:35) provided nanoparticles with spherical shape and
release behavior of the antimicrobial extracts suitable to
prevent bacterial growth. The adapted Gompertz model fit-
ted well the experimental release data and could be used
to predict the release behavior of passion fruit by-products
extracts from PLGA polymers. Results suggest the PLGA 65:35
is more suitable to the encapsulation of the passion fruit seed
extract. The hydrophobicity of seed cake extract reduces its
interaction with PLGA, and hence its entrapment efficiency,
but still enables great antimicrobial results. Most impor-
tantly, the encapsulation with both PLGA greatly enhanced
the antimicrobial activity of both extracts as it decreased
up to 95% the concentration of extracts needed to inhibit
microorganisms improving the delivery of antimicrobials to
the microorganisms in aqueous environment and could be
used as alternatives to chemical disinfectants to better con-
trol pathogens. In summary, results in this study show that
the polymeric nanoparticles could be used as a strategy to
enhance fruit by-products extracts bioactivities and present
potential for further investigations in food systems and phar-
maceutical applications.
Funding
CAPES, Brazilian funding agency.
Conflict of interest
The authors declare no competing financial interest.
Acknowledgement
The authors wish to thank Extrair Óleos Naturais for the raw
material supply.
145
References
Anderson, J.M., Shive, M.S., 1997. Biodegradation and
biocompatibility of PLA and PLGA microspheres. Adv. Drug
Deliv. Rev. 28, 5–24,
http://dx.doi.org/10.1016/j.addr.2012.09.004.
AOAC International, 2005. AOAC Official Methods of Analysis of
AOAC International.
Astete, C.E., Sabliov, C.M., 2006. Synthesis and characterization of
PLGA nanoparticles. J. Biomater. Sci. Polym. Ed. 17, 247–289,
http://dx.doi.org/10.1163/156856206775997322.
Baty, F., Delignette-muller, M., 2004. Estimating the bacterial lag
time: which model, which precision? Int. J. Food Microbiol. 91,
261–277, http://dx.doi.org/10.1016/j.ijfoodmicro.2003.07.002.
Brandt, A.L., Castillo, A., Harris, K.B., Keeton, J.T., Hardin, M.D.,
Taylor, T.M., 2010. Inhibition of Listeria monocytogenes by food
antimicrobials applied singly and in combination. J. Food Sci.
75, M557–M563,
http://dx.doi.org/10.1111/j.1750-3841.2010.01843.x.
Burt, S., 2004. Essential oils: their antibacterial properties and
potential applications in foods—a review. Int. J. Food
Microbiol. 94, 223–253,
http://dx.doi.org/10.1016/j.ijfoodmicro.2004.03.022.
Cocero, M.J., Martín, Á., Mattea, F., Varona, S., 2009. Encapsulation
and co-precipitation processes with supercritical fluids:
fundamentals and applications. J. Supercrit. Fluids 47,
546–555, http://dx.doi.org/10.1016/j.supflu.2008.08.015.
Cohn, D., Younes, H., 1987. Amorphous and crystalline
morphologies in glycolic acid and lactic acid polymers.
Polymer (Guildf.) 28, 2018–2022.
Ferreira, B.S., De Almeida, C.G., Faza, L.P., De Almeida, A., Diniz,
C.G., Da Silva, V.L., Grazul, R.M., Le Hyaric, M., 2011.
Comparative properties of amazonian oils obtained by
different extraction methods. Molecules 16, 5874–5885,
http://dx.doi.org/10.3390/molecules16075875.
Gomes, C., Moreira, R.G., Castell-Perez, E., 2011.
Poly(dl-lactide-co-glycolide) (PLGA) nanoparticles with
entrapped trans-cinnamaldehyde and eugenol for
antimicrobial delivery applications. J. Food Sci. 76, 16–24,
http://dx.doi.org/10.1111/j.1750-3841.2010.01985.x.
Hill, L.E., Taylor, T.M., Gomes, C., 2013. Antimicrobial efficacy of
poly(dl-lactide-co-glycolide) (PLGA) nanoparticles with
entrapped cinnamon bark extract against Listeria
monocytogenes and Salmonella typhimurium. J. Food Sci., 78,
http://dx.doi.org/10.1111/1750-3841.12069.
Kiran, E., Liu, K., Ramsdell, K., 2008. Morphological changes in
poly(e-caprolactone) in dense carbon dioxide. Polymer
(Guildf.) 49, 1853–1859,
http://dx.doi.org/10.1016/j.polymer.2008.02.017.
Konan, Y.N., Cerny, R., Favet, J., Berton, M., Gurny, R., Allemann, E.,
2003. Preparation and characterization of sterile sub-200 nm
for photodynamic therapy. Eur. J. Pharm. Biopharm. 55,
115–124, http://dx.doi.org/10.1016/S0939-6411(02)00128-5.
Mainardes, R.M., Evangelista, R.C., 2005. PLGA nanoparticles
containing praziquantel: effect of formulation variables on
size distribution. Int. J. Pharm. 290, 137–144,
http://dx.doi.org/10.1016/j.ijpharm.2004.11.027.
Makino, K., Mogi, T., Ohtake, N., Yoshida, M., Ando, S., Nakajima,
T., Ohshima, H., 2000. Pulsatile drug release from
poly(lactide-co-glycolide) microspheres: how does the
composition of the polymer matrices affect the time interval
between the initial burst and the pulsatile release of drugs?
Colloids Surf. B Biointerfaces 19, 173–179.
Malacrida, C.R., Jorge, N., 2012. Yellow passion fruit seed oil
(Passiflora edulis f. flavicarpa): physical and chemical
characteristics. Braz. Arch. Biol. Technol. 55, 127–134,
http://dx.doi.org/10.1590/S1516-89132012000100016.
Miller, R.A., Brady, J.M., Cut-Right, D.E., 1977. Degradation rates of
oral resorbable implants (polylactates and polyglycolates):
rate modification with changes in PLA/PGA copolymer ratios.
J. Biomed. Mater. Res. 11, 711–719.
146 food and bioproducts processing 1 0 4 ( 2 0 1 7 ) 137–146
Mu, L., Feng, S., 2003. PLGA/TPGS nanoparticles for controlled
release of paclitaxel: effects of the emulsifier and drug loading
ratio. Pharm. Res. 20, 1864–1872.
Oliveira, D.A., Angonese, M., Gomes, C., Ferreira, S.R.S., 2016.
Valorization of passion fruit (Passiflora edulis sp.) by-products:
sustainable recovery and biological activities. J. Supercrit.
Fluids 111, 55–62,
http://dx.doi.org/10.1016/j.supflu.2016.01.010.
Park, T.G., 1995. Degradation of poly(lactic-co-glycolic acid)
microspheres: effect of copolymer composition. Biomaterials
16, 1123–1130.
Patel, A.K., Bajpai, R., Keller, J.M., 2014. On the crystallinity of
PVA/palm leaf biocomposite using DSC and XRD techniques.
Microsyst. Technol. 20, 41–49,
http://dx.doi.org/10.1007/s00542-013-1882-0.
Piombo, G., Barouh, N., Barea, B., Boulanger, R., Brat, P., Pina, M.,
Villeneuve, P., 2006. Characterization of the seed oils from kiwi
(Actinia chinensis), passion fruit (Passiflora edulis) and guava
(Psidium guajava). Oléagineux Corps Gras Lipides 13, 195–199.
Ravichandran, M., Hettiarachchy, N.S., Ganesh, V., Ricke, S.C.,
Singh, S., 2011. Enhancement of antimicrobial activities of
naturally occurring phenolic compounds by nanoscale
delivery against Listeria monocytogenes, Escherichia coli o157:H7
and Salmonella typhimurium in broth and chicken meat system.
J. Food Saf. 31, 462–471,
http://dx.doi.org/10.1111/j.1745-4565.2011.00322.x.
Roe, K.D., Labuza, T.P., 2005. Glass transition and crystallization of
amorphous trehalose–sucrose mixtures. Int. J. Food Prop. 8,
559–574, http://dx.doi.org/10.1080/10942910500269824.
Sano, S., Sugiyama, K., Ito, T., Katano, Y., Ishihata, A., 2011.
Identification of the strong vasorelaxing substance scirpusin
B, a dimer of piceatannol, from passion fruit (Passiflora edulis)
seed. J. Agric. Food. Chem., 6209–6213.
Sansdrap, P., Moes, A.J., 1997. In vitro evaluation of the hydrolytic
degradation of dispersed and aggregated
poly(dl-lactide-co-glycolide) microspheres. J. Control. Release
43, 47–58.
Silva, L.M., Hill, L.E., Figueiredo, E., Gomes, C.L., 2014a. Delivery of
phytochemicals of tropical fruit by-products using
poly(dl-lactide-co-glycolide) (PLGA) nanoparticles: synthesis,
characterization, and antimicrobial activity. Food Chem. 165,
362–370, http://dx.doi.org/10.1016/j.foodchem.2014.05.118.
Silva, L.M.R., Figueiredo, E.A.T., Ricardo, N.M.P.S., Vieira, I.G.P.,
Figueiredo, R.W., Brasil, I.M., Gomes, C.L., 2014b.
Quantification of bioactive compounds in pulps and
by-products of tropical fruits from Brazil. Food Chem. 143,
398–404, http://dx.doi.org/10.1016/j.foodchem.2013.08.001.
Smith-Palmer, A., Stewart, J., Fyfe, L., 1998. Antimicrobial
properties of plant essential oils and essences against five
important food-borne pathogens. Lett. Appl. Microbiol. 26,
118–122, http://dx.doi.org/10.1046/j.1472-765X.1998.00303.x.
Stevanović, M.M., Uskoković, D.P., 2009.
Poly(lactide-co-glycolide)-based micro and nanoparticles for
the controlled drug delivery of vitamins. Curr. Nanosci. 5,
1–14.
USAID, 2014. The US market for fresh passion fruit, Acceso
Market Brief 16.
USAID, 2011. The EU market for passion fruit. Acceso Mark. Br. 5
1–6.
von der Linden, U.M., 2007. The passion fruit market—is it
controllable? Fruit Process., 6–9.
Wang, J., Wang, B.M., Schwendeman, S.P., 2002. Characterization
of the initial burst release of a model peptide from
poly(d,l-lactide-co-glycolide) microspheres. J. Control. Release
82, 289–307.
Weiss, J., Takhistov, P., McClements, D.J., 2006. Functional
materials in food nanotechnology. J. Food Sci. 71, 107–116,
http://dx.doi.org/10.1111/j.1750-3841.2006.00195.x.
Wischke, C., Schwendeman, S.P., 2008. Principles of encapsulating
hydrophobic drugs in PLA/PLGA microparticles. Int. J. Pharm.
364, 298–327, http://dx.doi.org/10.1016/j.ijpharm.2008.04.042.
Zhang, H., Cui, W., Bei, J., Wang, S., 2006. Preparation of
poly(lactide-co-glycolide-co-caprolactone)nanoparticles and
their degradation behaviour in aqueous solution. Polym.
Degrad. Stab. 91, 1929–1936,
http://dx.doi.org/10.1016/j.polymdegradstab.2006.03.004.
Zigoneanu, I.G., Astete, C.E., Sabliov, C.M., 2008. Nanoparticles
with entrapped ␣-tocopherol: synthesis, characterization, and
controlled release. Nanotechnology 19, 105606,
http://dx.doi.org/10.1088/0957-4484/19/10/105606.