In vitro characterization of 6-Coumarin loaded solid lipid nanoparticles and their uptake by immunocompetent fish cells

Page 1

In

APa

b

c

3d

a

ARR2AA

KSMFCF

1

aWredtofw

Bf

h0

Colloids and Surfaces B: Biointerfaces 127 (2015) 79–88

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces

j o ur nal ho me pa ge: www.elsev ier .com/ locate /co lsur fb

n vitro characterization of 6-Coumarin loaded solid lipidanoparticles and their uptake by immunocompetent fish cells

driana Trapania, Delia Mandracchiaa, Cinzia Di Francob, Héctor Corderoc,atricia Morcilloc, Roberto Comparelli d, Alberto Cuestac, Maria Angeles Estebanc,∗

Department of Pharmacy-Drug Sciences, University of Bari “Aldo Moro”, Via Orabona 4, 70125 Bari, ItalyCNR-IFN Bari, Bari via Amendola 173, 70126 Bari, ItalyDepartment of Cell Biology and Histology, Faculty of Biology, Regional Campus of International Excellence “Campus Mare Nostrum”, University of Murcia,0100 Murcia, SpainCNR-IPCF, Istituto per i Processi Chimici e Fisici, UOS Bari, c/o Dipartimento di Chimica, University of Bari “Aldo Moro”, Via Orabona 4, 70126 Bari, Italy

r t i c l e i n f o

rticle history:eceived 1 November 2014eceived in revised form2 December 2014ccepted 13 January 2015vailable online 21 January 2015

eywords:olid lipid nanoparticlesicroparticles

low cytometryonfocal microscopy

a b s t r a c t

The primary aim of the present work was to evaluate the in vitro uptake of 6-Coumarin (6COUM) loadedsolid lipid nanoparticles (SLN) by two gilthead seabream (Sparus aurata L.) cell types: an established cellline (SAF-1 cells) and the primary cultures of head-kidney (HK)—the main haemopoietic organ in fish,equivalent to mammalian bone marrow—leucocytes. For this purpose, after the physicochemical charac-terization of SLN, the uptake by those immunocompetent fish cells was evaluated using flow cytometryand confocal microscopy. Concomitantly, the uptake of 6-COUM loaded SLN was compared with thatachieved with 6-COUM loaded pectin microparticles (MPs), which were selected as a competitor of thedelivery carriers. After SLN and MP physicochemical characterization, the results demonstrated that SAF-1 cells were able to internalize high percentages of 6-COUM SLNs when incubated for 4, 8 and 24 h, withthe highest SLN concentration tested (10 �g/ml). The ability of HK leucocytes to internalize SLN wasalso found to vary depending on both incubation time and SLN concentration. The highest values of HK

ish leucocytes internalizing SLN particles (around 16%) were detected at the maximum SLN concentration(20 �g/ml) at incubation times of 4 or 8 h. Conversely, HK leucocytes were unable to internalize MPs atany tested concentration and incubation time. A possible mechanism explaining the uptake into cellsis proposed. The present work constitutes the first approximation to consider SLN as nanocarriers fordelivering biologically active substances to fish.

. Introduction

Fish are exposed to several environmental insults, includingerobic and anaerobic bacteria, viruses, parasites and pollutants.hile the mucosal surfaces of fish are protected by a mucus bar-

ier, in many circumstances, this protection may be ineffective,ven leading to significant fish diseases. Furthermore, for females,ue to unsatisfactory oocyte maturation and ovulation, reproduc-ive failure takes place and, hence, successful spawning does not

ccur. The ultimate consequence of all these environmental riskactors is an extensive loss of useful components to humans’ diets,ith related economic damage. Hence, there is a need to develop

∗ Corresponding author at: Department of Cell Biology and Histology, Faculty ofiology, University of Murcia, 30100 Murcia, Spain. Tel.: +34 868887665;

ax: +34 868883963.E-mail address: [email protected] (M.A. Esteban).

ttp://dx.doi.org/10.1016/j.colsurfb.2015.01.022927-7765/© 2015 Elsevier B.V. All rights reserved.

© 2015 Elsevier B.V. All rights reserved.

methods to protect fish from pathogenic microorganisms or fromstressful situations and to control their reproductive processes,thus limiting the mentioned drawbacks. To address these prob-lems, several strategies are emerging, including those involvingenhancing the innate immune system of fish, which is the arm ofthe immune system most well developed in this group of verte-brates [1–3], as well as those involving exploiting pharmaceuticalnanocarriers capable to incorporate biologically active substances[3–6]. Due to their size, nanocarriers can be taken up by fish cells,offering controlled administration of substances such as immunos-timulants, antibiotics, vaccines and hormones, which can improvewell-being, health status or spawning problems. Besides nanomet-ric size, these systems should possess further requirements such asbiocompatibility, sufficient drug loading, protection from degrada-

tion of the encapsulated drug and good mucoadhesive properties,which allow for prolonged contact with the mucus layer [5].

In the last decade, it has been widely demonstrated that solidlipid nanoparticles (SLN) have most of these favourable features

Page 2

8 faces B

apcpdtkr

wbCaswlfimmSmopa

2

2

(tayc(ma

2

lGtCg8ta(rtbra

2

dtast

0 A. Trapani et al. / Colloids and Sur

nd represent a valuable alternative to other nanocarriers, such asolymeric nanoparticles and liposomes [7]. In fact, SLN have a bio-ompatible lipid matrix which is solid at body temperature, showrotection from degradation, sustained release of the encapsulatedrug, reduced cytotoxicity as well as cell uptake [8–10]. Amonghe SLN applications already reported [7,11–15], to the best of ournowledge, SLN have not previously been considered as nanocar-iers for delivering biologically active substances to fish.

Taken together all these considerations, the aim of the presentork was to evaluate in vitro the potential of SLN to be internalized

y immunocompetent fish cells. For this purpose, the uptake of 6-oumarin (6-COUM) loaded SLN was studied using flow cytometrynd confocal microscopy. Two cell types obtained from giltheadeabream (Sparus aurata, the second-most cultured marine fishorldwide, after salmon) were evaluated: cells from an estab-

ished cell line (SAF-1 cells, fibroblast-like cells derived from then of S. aurata) and primary cultures of head-kidney (HK)-theain haemopoietic organ in fish equivalent to mammalian bonearrow–leucocytes. Concomitantly, the uptake of 6-COUM loaded

LN was compared with that arising from 6-COUM loaded pectinicroparticles (MPs), which were selected as control carriers in

rder to highlight the role of particle size in the internalizationrocess. The results concerning the performances of SLN and MPsre herein reported and discussed.

. Materials and methods

.1. Materials

Gelucire® 50/13 and pectin were kindly donated by GattefossèMilan, Italy) and Farmalabor (Canosa di Puglia, Italy), respec-ively. 3-(2′-Benzothiazolyl)-7-diethylaminocoumarin (6-COUM)nd Tween 85 were purchased from Sigma–Aldrich (Italy). Dial-sis tubing regenerated cellulose membranes (molecular weightut-off 12,000–14,000 g/mol) were purchased from Spectra-PoreMilan, Italy). RPMI-1640 culture medium, L-15 Leibovitz culture

edium and heparin were provided by Sigma–Aldrich (St. Louis),nd antibiotics by Flow (Spain).

.2. Preparation of 6-Coumarin loaded SLN

All the procedures involving 6-COUM were carried out underight protection. Based on the melt-emulsification method [16–18],elucire® 50/13 (60 mg) was melted at 70 ◦C in a glass vial, and

hen the appropriate amount of 6-COUM (6 mg or 12 mg of 6-OUM) was added to the melted Gelucire® 50/13. In a separatelass vial, an aqueous solution (1.37 mL) of the surfactant (Tween5, 60 mg) was heated at the same temperature and then added tohe melted phase at 70 ◦C, forming an emulsion by homogenizationt 12,300 rpm for 2 min with an UltraTurrax model T25 apparatusJanke and Kunkel, Germany). The nanosuspension was cooled atoom temperature and the resulting SLN were collected by cen-rifugation (16,000 × g, 45 min, Eppendorf 5415D, Germany). Thus,esides the unloaded SLN, two formulations containing the fluo-escent probe were prepared and are denoted as 6 mg 6-COUM SLNnd 12 mg 6-COUM SLN, respectively.

.3. Preparation of 6-Coumarin loaded pectin MPs

The unloaded MPs were prepared according to the methodescribed in the literature with few modifications [19]. To adsorb

he fluorescent probe onto the surface of unloaded pectin MP, theppropriate amount of 6-COUM (6 or 12 mg of 6-COUM) was dis-olved in ethanol (3 mL) and then the resulting solution was addedo the freeze dried unloaded pectin MPs (25 mg).

: Biointerfaces 127 (2015) 79–88

2.4. Physicochemical characterization of SLN

The particle size and polydispersion index (PI) of all preparedSLN were determined in double distilled water by photon cor-relation spectroscopy (PCS) using a Zetasizer NanoZS (ZEN 3600,Malvern, UK). The determination of the �-potential was performedusing laser Doppler anemometry (Zetasizer NanoZS, ZEN 3600,Malvern, UK) after dilution in KCl (1 mM, pH 7) [20].

The morphology of unloaded and 6-COUM-loaded SLN wasexamined by transmission electron microscopy (TEM, model JEM-1011, JEOL, Tokyo, Japan) operated at 80 kV. The SLN samples werecast on a carbon coated copper grid and negatively stained witha 2% (v/v) phosphotungstic acid solution for 30 s. The solvent wasallowed to dry overnight at room temperature prior to TEM visual-ization.

2.5. Physicochemical characterization of pectin MPs

The particle size of the freeze-dried pectin MPs was determinedusing a light stereomicroscope (Leica Galen III) equipped with aPanasonic (WV CP 230) camera and Leica Qwin v. 2.4 software. Thearithmetic mean diameter of MPs was determined by averaging theindividual values of 150 MPs from each sample.

The morphological characterization of the unloaded- and 6-COUM-loaded-MP powders was performed by a field emissionscanning electron microscope (FE-SEM) (Sigma Zeiss). Prior to theanalysis of the powders, the samples were stuck on stubs with adouble-face carbon adhesive disc and coated with a 2 nm palladiumlayer by an electron beam evaporator. The samples were examinedat 5–10 keV, 10 �m aperture, using the in-lens detector.

2.6. Determination of 6-COUM content in SLN and MPs

The amount of entrapped or adsorbed 6-COUM was expressedas association efficiency (AE):

AE = Total weight of 6-COUM − Weight of 6-COUM in the supernatantTotal weight of 6-COUM

× 100

The mass added (namely, the total weight of 6-COUM) and theamount of 6-COUM associated with the SLN supernatant (or to MPsupernatant) were determined by spectrophotometric analysis ata wavelength of 300 nm using a Perkin-Elmer Lambda Bio 20 spec-trophotometer. The measurements were performed in triplicate at25 ◦C, and the concentrations of the tracer were determined from astandard curve of 6-COUM in ethanol (concentrations ranging from1 to 100 �g/mL).

2.7. Stability studies of SLN

Freshly prepared 6-COUM loaded SLN at the two dosesemployed (i.e., 6 and 12 mg) were evaluated for their stability upto 4 weeks under two different temperature conditions (4 ◦C and22 ◦C [21]) without any stirring. The particle size of the sampleswas determined at scheduled time intervals using a PCS apparatus.Each experiment was performed in triplicate.

2.8. In vitro release studies

These studies were carried out up to 48 h using a mixture(PBS, pH 7.4)/EtOH, 99/1 (v/v) according to the dialysis method[22]. The fluorescent tracer concentrations were determined, aspreviously reported [23], from a standard curve of 6-COUM in ace-

tonitrile (concentrations ranging from 31 to 1000 ng/mL), usinga fluorometer instrument (Fluostar Omega, Excitation: 495 nm;Emission 525 nm, BMG Labtech, U.K.) and conventional 96-wellplates (MicrotestTM, OptiluxTM, Becton Dickinson, U.K.).

Page 3

faces B

2

I6a4

opepsflt

2

poslSkit

2

2

p2L(asStctt

2

vostsswmacww

2p

ad(

A. Trapani et al. / Colloids and Sur

.9. Solid state studies

The FT-IR spectra were acquired using a PerkinElmer 1600 FT-R spectrometer (PerkinElmer, Italy). Freeze-dried unloaded- and-COUM loaded SLN as well as pectin MPs were mixed with anppropriate amount of KBr. The scanning range examined was00–4000 cm−1 with a resolution of 1 cm−1.

Differential scanning calorimetry (DSC) thermograms werebtained on a Mettler Toledo DSC 822e apparatus using aluminiumans of 40 �L capacity, and a sealed empty pan was used as a ref-rence. Aliquots of about 5 mg of each lyophilized sample werelaced in the pans. Thermograms were measured by heating theample from 25 to 250 ◦C at a rate of 5 ◦C/min, under nitrogenow of 50 cm3/min. Indium was used as standard for calibratinghe temperature.

.10. Animals

Nine specimens (100 g mean weight) of the hermaphroditerotandrous seawater teleost gilthead seabream (S. aurata L.)btained from a local farm (Murcia, Spain) were kept in runningeawater tanks (flow rate 900 L/h) at 20 ◦C with a 12 h dark:12 hight photoperiod and fed with a commercial pellet diet (Skretting,pain). Fish were allowed to acclimatize for 15 days and they wereept 24 h without feeding before sampling. The studies presentedn this manuscript were approved by the Bioethical Committee ofhe University of Murcia.

.11. Cells

.11.1. SAF-1 cell cultureThe established cell line SAF-1 [24] was seeded in 75-cm2

lastic tissue culture flasks (Nunc, Denmark) and cultured at5 ◦C in an atmosphere with 85% relative humidity using L-15eibowitz medium supplemented with 10% foetal bovine serumFBS, Sigma–Aldrich), 2 mM L-glutamine, 100 �g/mL streptomycinnd 100 U/mL penicillin. The subculture was done according totandard trypsinization methods (0.25% trypsin/0.53 mM EDTA,igma–Aldrich). The cells were centrifuged (200 × g, 10 min) andhe viability determined by the trypan blue exclusion test. SAF-1ells were plated in 48-well plates at 2.5 × 105 cell/well and cul-ured overnight at 25 ◦C with 85% relative humidity and 5% CO2 inhe incubator chamber.

.11.2. Head kidney leucocyte isolationAfter gilthead seabream bleeding, the HK was removed by a

entral incision, cut into small fragments and transferred to 8 mLf sRPMI [RPMI-1640 culture medium supplemented with 0.35%odium chloride to adjust the medium’s osmolarity to the gil-head seabream plasma osmolarity (of 353.33 mOsm), 2% foetal calferum (Sigma–Aldrich), 100 U/mL penicillin (Flow) and 100 �g/mLtreptomycin (Flow)] in sterile conditions [25]. Cell suspensionsere obtained by forcing fragments of the organ through a nylonesh (mesh size: 100 �m). Cell suspensions were collected with

Pasteur pipette, washed twice, counted and adjusted to 107

ells/mL in sRPMI. Aliquots of HK leucocytes were plated in 48-ell plates at 2.5 × 105 cell/well and cultured overnight at 25 ◦Cith 85% relative humidity and 5% CO2 in the incubator chamber.

.11.3. In vitro incubation of SAF-1 cells and HK-leucocytes witharticles

To study the uptake of the particles on seabream SAF-1 cellsnd HK leucocytes, cells were incubated without (control) or withifferent concentrations of 6-COUM loaded particles (SLN or MP)1, 10 and 20 �g/mL of culture medium) at 25 ◦C with 85% relative

: Biointerfaces 127 (2015) 79–88 81

humidity and 5% CO2 atmosphere for 4, 8 and 24 h for SAF-1 cellsand 2, 4 and 8 h for HK leucocytes.

2.11.4. Cell viabilityTo determine whether particles affect SAF-1 cells and HK leu-

cocyte viability, aliquots of 50 �L of cells previously incubatedwithout (control) or with particles were placed in 5 mL tubes(Falcon, Becton Dickinson) and 30 �L of propidium iodide (PI,400 �g/mL, Sigma–Aldrich) were added to each sample. The tubeswere gently mixed before analysis in a FACScan (Becton Dickinson,Madrid, Spain) flow cytometer with an argon-ion laser adjustedto 488 nm. Analyses were performed on 5000 cells, which wereacquired at a rate of 300 cells/s. Data were collected in the form oftwo-parameter side scatter (granularity) (SSC) and forward scatter(size) (FSC), and red fluorescence (FL2, PI) dot plots or histogramswere made on a computerized system. Dead cells were estimatedas the percentage of cells with PI (red-PI fluorescent cells).

2.11.5. Internalization of particles by fish cellsTo determine whether the particles were internalized by the

fish cells, after incubation with particles, the cells were studied byflow cytometry using to the method of Esteban et al. [25]. Samplesof 100 �L of cell suspensions (SAF-1 cells or HK leucocytes) wereplaced in 5 mL tubes (Falcon, Becton Dickinson) and the fluores-cence of the extracellular particles (i.e., free particles and particlesadhered to cells but not ingested) was quenched by adding 40 �Lice-cold trypan blue (0.4% in PBS) per sample. Standard samples ofparticles containing 6-COUM or SAF-1 cells and HK leucocytes wereincluded in each assay. Samples with particles without 6-COUMwere used as negative controls. The cytometer was set to analyzethe fluorescent cells (FL1, green-6-COUM fluorescent cells) gatedfrom all the cells present in the samples. Cells internalizing parti-cles were defined as the percentage of green-6-COUM fluorescentcells within the cell population.

Furthermore, samples of SAF-1 cells and HK leucocytes incu-bated with 6-COUM loaded-SLN were studied alive on the plateby confocal scanner microscopy (Leica TCS SP2). Micrographs weretaken with a Nikon 90i camera.

2.12. Statistical assays

Data from different experimental groups were compared by aone-way analysis of variance (ANOVA) using GraphPad Prism v.5.00 computer program (GraphPad Software, Inc. CA, USA) and dif-ferences were considered significant at a confidence level of 95%(p < 0.05). For biological evaluation, all assays were performed intriplicate and data were represented as mean ± standard error (SE)for each experimental group (n = 9). The data from the flow cyto-metric assays were studied by using the statistical option of theLysis Software Package (Becton Dickinson). Data were statisticallyanalyzed by one-way ANOVA and Tukey’s test as a post hoc compari-son. In all cases, differences were considered statistically significantwhen p < 0.05.

3. Results

3.1. Physicochemical characterization of SLN and MPs

The mean particle size, polydispersion index (PI), zeta potentialand association efficiency (AE) for unloaded- and 6-COUM-loadedSLN are shown in Table 1. The particle size of the prepared SLNwas in the range of 141–335 nm, with the lowest size observed for

unloaded SLN and the largest for 12 mg 6-COUM SLN. An interme-diate size was found with 6 mg 6-COUM SLN (235 nm). It shouldbe mentioned that when using one-way ANOVA, these differenceswere statistically significant (p < 0.05) compared to unloaded SLN.

Page 4

82 A. Trapani et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 79–88

Table 1Physicochemical properties of 6-COUM loaded SLN and pectin MPs. Mean ± S.D. are reported, n = 6. Unloaded Gelucire SLN and pectin MPs were used as control.

SLN sample Mean diameter (nm) Polydispersion index Zeta potential (mV) A.E.%

Unloaded Gelucire SLN 141 ± 11 0.34 ± 0.06 −9.7 ± 0.8 –6 mg 6-COUM Gelucire SLN 235 ± 14* 0.55 ± 0.04 −16.3 ± 1.9* 79.5 ± 8.812 mg 6-COUM Gelucire SLN 335 ± 12* 0.45 ± 0.12 −12.5 ± 1.9 88.2 ± 1.2

MP sample Mean diameter (�m) A.E.%

Pectin MPs 143 ± 406 mg 6-COUM pectin MP 170 ± 34 68 ± 1

Celi(Apilost

12 mg 6-COUM pectin MP 295 ± 31*

* p < 0.05

oncerning PI, except for the unloaded SLN, which can be consid-red as a mono-dispersed system, the 6-COUM-loaded SLN formu-ations were characterized by a wide distribution. The zeta potentialncreased in 6 mg 6-COUM SLN (–16.3 mV) and 12 mg 6-COUM SLN–12.5 mV) when compared to unloaded SLN (–9.7 mV). Regarding.E., in both cases, values in the range 80–88% were found. The meanarticle size for unloaded- and 6-COUM-loaded MPs is also shown

n Table 1 and it was in the range of 143–295 �m. Again, for MPs the

owest value was observed for the unloaded ones, while the biggestne was found for 12 mg 6-COUM MPs. An intermediate particleize value (170 �m) was seen for 6 mg 6-COUM MPs. Moreover,he differences were statistically significant (p < 0.05) compared to

Fig. 1. (A) TEM micrograph of 6 mg 6-COUM SLN. Insets: high magnification i

78 ± 5

unloaded MPs only in the case of 12 mg 6-COUM loaded ones. Theassociation efficiencies resulted slightly higher with the biggestMPs but lower than the corresponding values observed for SLN.

The stability on storage at 22 ◦C and 4 ◦C of both 6-COUM loadedSLN was also evaluated by monitoring their particle size by PCS upto 1 month. The obtained results show that the formulation withhigher 6-COUM content (12 mg) provided a clear increase in parti-cle size during the period of observation (Fig. 1, Appendix A), unlike

the 6 mg 6-COUM loaded SLN.

According to the TEM observations (Fig. 1A), it appears that allSLN were composed of nearly spherical particles, with a core–shellstructure. In addition, in the inset of Fig. 1A, the 4 nm wide dark

mages of the same sample. (B) SEM micrograph of 12 mg 6-COUM MPs.

Page 5

A. Trapani et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 79–88 83

F re Gel UM (a6

cf

bntt

3

m(ppiolw

(Cpopi

ig. 2. (A) Differential scanning calorimetric thermograms for pure 6-COUM (a), puoaded SLN (e). (B) Differential scanning calorimetric thermograms for pure 6-CO-COUM-loaded MPs (e).

ircular line around the SLN can be reasonably ascribed to the sur-actant monolayer surrounding the SLN.

The morphology of loaded and unloaded MPs was investigatedy SEM (Fig. 1B). All the 6-COUM loaded pectin MPs displayed aon-spherical shape (Fig. 1B) [26]. Several needle-shaped struc-ures were found and, additionally, the higher the additive content,he more densely packed needles were observed (Fig. 1B).

.2. Solid state studies of SLN and pectin MPs

To gain information on the solid state features, the DSC ther-ograms for different lyophilized SLN formulations were recorded

Fig. 2A). A sharp endothermic peak at 209 ◦C was observed withure 6-COUM due to the melting point of the fluorescent probe. Thiseak was absent in the 6-COUM containing formulations, suggest-

ng a notable decrease in the crystallinity of the fluorescent probence encapsulated in SLN. The pure Gelucire® 50/13 showed twoarge and small endothermic peaks at 49 ◦C and 55 ◦C, respectively,

hich were practically absent in all the formulations.Concerning the DSC thermograms of the 6-COUM in pectin MPs

Fig. 2B), aside from the endothermic peak at 209 ◦C of the pure 6-OUM and the melting endotherm at 156 ◦C observed in the pure

ectin, clear differences were noted between the thermogramsf unloaded and loaded MPs. Thus, while the melting endothermeak of pectin for unloaded MPs was still detected, even though

t was slightly lower at 150 ◦C, such a peak was absent in the

lucire® 50/13 (b), 6 mg 6-COUM-loaded SLN (c), unloaded SLN (d), 12 mg 6-COUM-), pure pectin (b), unloaded pectin MPs (c), 6 mg 6-COUM-loaded MPs (d), 12 mg

thermograms of both 6-COUM-loaded MPs. In these cases, how-ever, the sharp endothermic peak at 207 ◦C, attributable to themelting of the 6-COUM, was still present. These results suggest thatthe notable decrease in crystallinity observed for the fluorescentprobe when encapsulated in SLN does not occur when formulatingpectin MPs.

The FT-IR spectra of pure 6-COUM and 6-COUM loaded SLN areshown in Fig. 3A. In the spectrum of the pure 6-COUM, an absorp-tion band at 1712 cm−1 was clearly detected and it can be attributedto the C O group of the laser dye. This band was present in boththe FT-IR spectra of 6-COUM loaded SLN, even with reduced inten-sity, and it supports previous DSC studies showing a decrease ofcrystallinity of a probe when it is encapsulated in SLN. Interest-ingly, in the case of FT-IR analysis of MPs, the spectrum of 12 mg6-COUM loaded MPs shows that the absorption band at 1712 cm−1

together with other absorption bands of the fluorescent probe arestill present (Fig. 3B).

3.3. In vitro release studies

No 6-COUM from either SLN and pectin MPs was released overthe course of 48 h at 22 ◦C using a mixture of PBS/EtOH (99/1 v/v,

20 mL), which was selected as a release medium (Fig. 2, AppendixA). In vitro release studies essentially indicated that around 1% flu-orescent probe is released even after 48 h from both fluorescentSLN and MPs. Thus, due to the negligible amount of dye released, it

Page 6

84 A. Trapani et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 79–88

F oadedp Ps (d)

cdeiaH

3

wv8

vTwiNir(

ig. 3. (A) Fourier transform infrared spectra for pure 6-COUM (a), 6 mg 6-COUM-lure 6-COUM (a), pure pectin (b), unloaded pectin MPs (c), 6 mg 6-COUM-loadedM

an be stated that, under the conditions we used, the fluorescenceetected in the cells (see below) must be only due to the tracerncapsulated in the carrier rather than free 6-COUM. This is usefuln view of the application of 6-COUM loaded-SLN and pectin MPss fluorescent probes for visualizing the uptake by SAF-1 cells andK leucocytes of S. aurata.

.4. Internalization of SLN and MPs by gilthead seabream cells

The incubation of SAF-1 cells or HK leucocytes with SLN or MPsith or without 6-COUM (1, 10 or 20 �g/mL) did not affect cell

iability after 4, 8 or 24 h of incubation (for SAF-1 cells) or 2, 4 or h of incubation (for HK leucocytes) (Fig. 3, Appendix A).

For SAF-1 cells, the number of cells able to internalize the SLNaried depending on the incubation time and 6-COUM content.he number of SAF-1 cells internalizing SLN particles was very lowhen SLN particle concentration was at 0.2 and 1 �g/mL, as well as

n the control samples (SAF-1 cells incubated without SLN) (Fig. 4A).

evertheless, around 48%, 28% and 9% of SAF-1 cells, were able to

nternalize 6-COUM SLN particles when incubated for 4, 8 and 24 h,espectively, with the highest SLN concentration tested (10 �g/mL)Fig. 4A). On the contrary, less than 1% of SAF-1 cells were detected

SLN (b), 12 mg 6-COUM-loaded SLN (c). (B) Fourier transform infrared spectra for, 12 mg 6-COUM-loaded MPs (e).

after incubation for 4, 8 or 24 h with MPs at different concentrations(0.2, 1 and 10 �g/mL) (Fig. 4B).

Whether HK leucocytes were able to internalize the SLN alsovaried depending on both incubation time (2, 4 and 8 h) and SLNconcentration (1, 10 and 20 �g/mL). No fluorescence was detectedin the HK leucocytes after incubation with 1 �g/mL SLN particlesfor any assayed time. Around 13% and 11% of HK leucocytes inter-nalized SLN particles when incubated with SLN at 10 �g/mL for 4and 8 h, respectively. The highest values of HK leucocytes inter-nalizing SLN particles (around 16%) were detected for the highestassayed SLN concentration (20 �g/mL) and incubation times of 4 or8 h (Fig. 5). Seabream HK leucocytes were unable to internalize MPsat any concentration and incubation time tested (data not shown).

Aliquots of SAF-1 cells and HK leucocytes, previously incu-bated for 4 h with 6-COUM loaded SLN (10 �g/mL of 6-Coumarin),were studied by confocal scanner microscopy (Fig. 6A). Green flu-orescence (due to 6-COUM) was clearly visualized only insidethe cytoplasm of the cells. The fluorescence was seen irregularly

distributed and seemed to be concentrated inside the cytoplas-mic granules around the cell nuclei (Fig. 6A) and appeared tobe homogeneously distributed inside the HK leucocyte cytoplasm(Fig. 6B).

Page 7

A. Trapani et al. / Colloids and Surfaces B

F(A

4

srarenwaRtant

Fmd

ig. 4. Percentage of SAF-1 cells internalizing 6-COUM SLN (A) or 6-COUM MPsB) was determined by flow cytometry. Data are represented as mean ± SD (n = 3).sterisks denote significant differences (p < 0.05).

. Discussion

Although world aquaculture production continues to grow at alowing rate, the global trend of its development for fish supply hasemained uninterrupted. Among marine fish, gilthead seabream (S.urata L.) is one of the most interesting fish species in Mediter-anean marine aquaculture due to its nutritive value. In order tostablish a more sustainable aquatic food production, there is aeed for new microbial management strategies [27,28], amonghich dietary administration of immunostimulants and pro-, pre-

nd synbiotics have, at present, been widely explored [27,29].egarding on-going novel treatment strategies, the so-called “par-

iculate based approach”, consisting of micro- and nanoparticless a drug delivery system, has not been exploited in fish untilow, although there have been some promising results relatedo several human diseases [30]. For this reason, the aim of the

ig. 5. Percentage of head-kidney (HK) leucocytes internalizing 6-COUM SLN deter-ined by flow cytometry. Data are represented as mean ± SD (n = 3). Asterisks

enote significant differences (p < 0.05).

: Biointerfaces 127 (2015) 79–88 85

present work was to test the potential of SLN to be internalized bythe cells of immunocompetent system of fish, such as fibroblast-like SAF-1 cells and HK leucocytes, taking into account that suchnanocarriers have not previously been investigated for this pur-pose and, moreover, that these systems can adopt biocompatibleand biodegradable materials [31]. Thus, 6-COUM was inserted inthe particles because its green fluorescence makes its identificationvery easy.

We first determined that SLN did not alter the cell viability ofSAF-1 cells or the HK leucocytes because, to the best of our knowl-edge, there are no previous studies on this. Therefore, 6-COUMloaded SLN were prepared and their uptake by fibroblast-like SAF-1cells and gilthead seabream HK leucocytes was studied. This uptakehas been compared with that of the polysaccharidic micro-sizedparticles (i.e., pectin MPs). It should be noted that the SLN car-rier was prepared following a slightly modified melt-emulsificationmethod and, precisely, instead of melt-emulsification coupled withultrasonication homogenization, melt-emulsification coupled withUltraTurrax homogenizer was chosen. Such a method enabledthe preparation of SLN with (i) Tween 85 as the surfactant andGelucire® 50/13 as the lipid matrix, which is a mixture of stearoylpolyoxyl-32 glycerides, a small glyceride fraction and free PEG ableto self-emulsify in aqueous media giving rise to fine dispersions;the presence of PEG makes more hydrophilic the matrix and itmay allow also the delivery to fish of compounds more hydrophilicthan 6-COUM; (ii) a weight ratio of surfactant-to-lipid 1:1, whichis lower than that used in many other studies [16]. The particlesize of 6-COUM-loaded SLN is clearly dependent on the amountof employed dye (Table 1). Hence, it is not surprising that thesize of our SLN was slightly greater than those prepared using themicroemulsion method [10] since in the latter case the content ofthe fluorescent probe was lower. Moreover, in this study, the lipidphase was not homogenized by a high shear dispersing device or byan ultrasonic probe, and a PI greater than 0.3 occurred for 6-COUM-loaded SLN [32]. TEM analysis evidenced the spherical shape and alipid core/surfactant shell structure that has already been observedfor these nanocarriers [15], with a particle size slightly lower thanthat determined by PCS [33].

In addition, the zeta potentials of the SLN were in the range −9.7to 16.3 mV (Table 1), with the highest value measured for 6 mg 6-COUM-loaded SLN; thus, the higher zeta value can be invoked toexplain the higher stability resulting after four weeks of storage.

The assessment of the physical state of the components in theSLN by DSC and FT-IR is important because it greatly influencesdrug incorporation and release rates [32]. As for the decrease inthe crystallinity of 6-COUM when encapsulated in SLN, it shouldbe noted that the absence in the DSC thermogram of the meltingendotherm has been interpreted by some authors as complete sol-ubilization into the lipid matrix and conversion into the amorphousstate [16]. However, the data from the DSC analysis combined withthose from FT-IR spectroscopy ruled out a complete conversionto an amorphous state but indicated a marked reduction in crys-tallinity because in the FT-IR spectra of the 6-COUM-loaded SLNformulation (Fig. 3A), the characteristic absorption band of the dyeat 1712 cm−1 was still present.

From in vitro release profiles of SLN, it seems likely that 6-COUMis located in the core of the nanocarrier, according to a “drug-enriched core model” [32].

For the sake of comparison, we also prepared 6-COUM pectinMPs, taking into account that pectin mucoadhesive properties wereexpected to provide cell adhesion [34,35]. Surprisingly, comparedto other polysaccharides, such as alginate and chitosan, which

have already been investigated for fish immunization [5,36], thepotential of pectin MPs for this purpose has never been previ-ously explored. Hence, pectin MPs were used as competitor of theSLN essentially for the following two reasons: (i) to explore the

Page 8

86 A. Trapani et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 79–88

F ls (A)

6

cepettuptcaawt([

ig. 6. Confocal laser scanning microscopy microphotographs (63×) of SAF-1 cel-COUM). Bar: 50 �m; N, nucleus; L, lymphocyte; G, granulocyte.

onsequence on the fish cell uptake of lipid based colloidal carri-rs in comparison with micro-carriers made of an mucoadhesiveolymer and, thus, to gain information on the size effect; (ii) tovaluate a mucoadhesive polysaccharide as pectin whose poten-ial for the purpose has never been previously explored. Given thathe hydrophilic microenvironment created by the pectin MPs isnfavourable for 6-COUM encapsulation, the required MPs wererepared by 6-COUM adsorption onto the unloaded MPs. Also, forhese carriers, both the particle size and the association efficien-ies of 6-COUM-loaded pectin MPs were clearly dependent on themount of fluorescent probe employed. SEM analysis showed thatll pectin MPs displayed non-spherical and complex shapes. Indeed,

e observed that the surface structure of MPs was demonstrated

o be was considerably influenced by adding the 6-COUM contentdata not shown), in agreement with previous literature reports37,38].

and HK leucocytes (B) incubated for 4 h with 6-COUM loaded-SLN (10 �g/mL of

For biological evaluation, two different cell types from giltheadseabream were incubated with the 6-COUM SLN and MPs. TheSAF-1 cell line was established from seabream fibroblasts [36],mimicking the general behaviour of fibroblasts in wound healing,their involvement in part of the gilthead sea bream inflammatoryresponse [39], the coordination of the movement of immune cellsthroughout the tissue and the modulation of T lymphocytes activ-ity [40]. SAF-1 cells constitute an in vitro model for studies on fishimmunology and pathology, both for their ability to produce keydefence molecules and their susceptibility to bacteria and viruses[41,42].

The teleost HK is considered a haemopoietic organ similar to

the bone marrow of higher vertebrates [43]. We aimed to incu-bate the SLN and MPs with the fish cells for a range of timefrom 2 h up to 8 h or 24 h, which appeared a suitable time-frame according to previous in vitro assays. In our experimental

Page 9

faces B

aaAcbop

tattitmparsmoatnScwupecmsttIb(aitaceCint

5

GlflamtpcaItwi

[

[[

[

[

[[[[

[

[

[

[

[[[[[[[

[

[

A. Trapani et al. / Colloids and Sur

ssays, cells were individually treated with SLN and MPs and,fter incubation, the green fluorescence was detected inside them.s demonstrated by flow cytometry, the internalized fluores-ence of the cells was due to the cellular uptake of the particles,eing Trypan blue excluded by viable cells, and the residual flu-rescence in the cells implied the existence of intracellular SLNarticles.

Concerning the uptake mechanism of 6-COUM-loaded SLN intohe fish cells, several literature reports have suggested that fishre able to take up nanostructured materials by endocytosis ratherhan through other mechanisms [44]. Moreover, it is also knownhat vesicular transport in fish can occur at different target organs,ncluding the gills, gut and liver. In fish, endocytosis is a mul-istep transport mechanism quite similar to that occurring in

ammalian cells. Briefly, it proceeds through attachment of thearticles to cell surfaces, invagination of the plasma membranend their internalization to form endosomes, which, ultimately,elease their content, interacting with lysosomes [45]. A crucialtep is the attachment of the particles to the cell surface thatay occur by electrostatic interaction or by “receptor-mediated”

r “clathrin-mediated” pathways. Thus, particle size and shapere key factors in the endocytosis process. Indeed, it is acceptedhat endocytosis works well with colloidal sized particles butot with bigger ones. Therefore, the observed cellular uptake ofLN but not of MPs may suggest that the former colloidal parti-les are internalized through endocytosis by SAF-1 and HK cells,hereas MPs should be excluded for their larger size and irreg-lar shape. In addition, the negative charge of SLN ruled out theossibility that their interaction with the cell surface proceeds forlectrostatic causes, being that the plasma membrane negativelyharged too. It is likely that the endocytosis process is “clathrin-ediated” since, to our knowledge, specific receptors on the cell

urface of SLN components have not been reported. Concerninghe role of SNL particle shape, Shi et al. [46] clarified that par-icles with a higher aspect ratio are internalized less efficiently.t seems that 6-COUM-loaded pectin MPs cannot be internalizedecause of their size and their shape at the higher aspect ratioFig. 1B). A further outcome is that the uptake is concentration-nd time-dependent for both SAF-1 cells and HK leucocytes. Whilet is not surprising that the uptake is concentration-dependent,he fact that the highest values of SLN internalized were observedfter 4 or 8 h of incubation time and decreased at 24 h implies thatell turnover degrades the ingested particles. Overall, this worklucidates in vitro that fish immune cells absorbed the intact 6-OUM-SLN, which then remained in the cell cytoplasm. Further

n vivo studies will help to determine whether SLN are also inter-alized by immune cells when they are administered directly intohe feed.

. Conclusions

We have developed a novel lipid nanocarrier consisting ofelucire® 50/13 that is internalized by fibroblast-like SAF-1 and HK

eucocytes immunocompetent fish cells in vitro as demonstrated byow cytometry and confocal microscopy analyses using 6-COUMs the fluorescent probe. The finding that 6-COUM-loaded pectinicroparticles are unable to undergo a similar uptake process by

he same fish cells may suggest that the internalization of the SLN isroceeded by endocytosis transport mechanism. The present workonstitutes a starting point for considering such colloidal carrierss useful delivery platforms of biologically active substances to fish.

n addition, further studies are required to focus on the advan-age of using these formulations in aquaculture, replacing 6-COUMith substances like antioxidants or immunostimulants, aiming at

mproving fish health.

[[

[

: Biointerfaces 127 (2015) 79–88 87

Acknowledgements

A.T. and M.A.E. would like to acknowledge Dr. Nunzia Rella andDr. Michele Tedone (Erasmus Academic Exchange Service Scholar-ships). H. Cordero wishes to thank for a FPI fellowship. Thanks arealso due to Dr. Gulay Buyukkoroglu, for her help with lipid nanopar-ticles. TEM, SLN particle size and fish cells analyses were partiallysupported by Sens&Micro LAB Project (2007–2013) funded byApulia Region (Italy), PONa300369 “Laboratorio per lo SviluppoIntegrato delle Scienze e delle Tecnologie dei Materiali Avanzatie per dispositive innovativi-LABORATORIO SISTEMA” financed bythe Italian (Ministry of Education, University and Research) andSpanish Ministerio de Economía y Competitividad (grant num-ber AGL2011-30381-C03-01) and Fundación Séneca de la Regiónde Murcia (Grupo de Excelencia grant number 04538/GERM/06),respectively.

Appendix A. Supplementary data

Supplementary data associated with this article can befound, in the online version, at http://dx.doi.org/10.1016/j.colsurfb.2015.01.022.

References

[1] S. Benhamed, F.A. Guardiola, M. Mars, M.A. Esteban, Vet. Microbiol. 171 (2014)1.

[2] F.A. Guardiola, A. Cuesta, M. Arizcun, J. Meseguer, M.A. Esteban, Fish ShellfishImmunol. 36 (2014) 545.

[3] A. Ruyra, M. Cano-Sarabia, P. García-Valtanen, D. Yero, I. Gibert, S.A. Mackenzie,A. Estepa, D. Maspoch, N. Roher, Vaccine 32 (2014) 3955.

[4] M.A. Rather, R. Sharma, S. Gupta, S. Ferosekhan, V.L. Ramya, S.B. Jadhao, PLoSOne 8 (2013) e57094.

[5] A. Rivas-Aravena, A.M. Sandino, E. Spencer, Biol. Res. 46 (2013) 407.[6] L. Li, S.L. Lin, L. Deng, Z.G. Liu, J. Fish Dis. 36 (2013) 987.[7] S. Cai, Q. Zhang, T. Bagby, M.L. Forrest, Adv. Drug Deliv. Rev. 63 (2011) 901.[8] S.A. Wissing, O. Kayser, R.H. Müller, Adv. Drug Deliv. Rev. 56 (2004) 1257.[9] W. Mehnert, K. Mäder, Adv. Drug Deliv. Rev. 47 (2001) 165.10] I. Rivolta, A. Panariti, B. Lettiero, S. Sesana, P. Gasco, M.R. Gasco, M. Masserini,

G. Miserocchi, J. Physiol. Pharmacol. 62 (2011) 45.11] S. Lim, C. Kim, Int. J. Pharm. 243 (2002) 135.12] L. Montenegro, A. Trapani, C. Carbone, A. Latrofa, G. Puglisi, J. Nanosci. Nan-

otechnol. 12 (2012) 330.13] J. Duan, F.G. Vogt, X. Li, D.J. Hayes, H.M. Mansour, Int. J. Nanomed. 8 (2013)

3489.14] A. Rodríguez-Gascón, A. del Pozo-Rodríguez, M.Á. Solinís, Int. J. Nanomed. 9

(2014) 1833.15] Z.H. Zhen, Y.-L. Zhang, J.-P. Zhou, H.-X. Lu, Int. J. Nanomed. 7 (2012) 3333.16] W.M. Ibrahim, A.H. AlOmrani, A.E. Yassin, Int. J. Nanomed. 9 (2014) 129.17] J.O. Woo, M. Misran, P.F. Lee, L.P. Tan, Sci. World J. (2014) 205703.18] Li. Su, J.M. Zhaoshuai, J. Zou, X. Nie, Y. Shi, G. Cheng, AAPS Pharm. Sci. Tech. 12

(2011) 1011.19] G.A. Soares, A.D. de Castro, B.S. Cury, R.C. Evangelista, Carbohydr. Polym. 91

(2013) 135.20] D. Mandracchia, G. Tripodo, A. Latrofa, R. Dorati, Carbohydr. Polym. 103 (2014)

46.21] R. Cerezuela, F.A. Guardiola, P. González, J. Meseguer, M.Á. Esteban, Fish Shell-

fish Immunol. 33 (2012) 342.22] A.R. Gardouh, S. Gad, H.M. Ghonaim, M.M. Ghorab, Br. J. Pharm. Res. 3 (2013)

326.23] A. Trapani, J. Sitterberg, U. Bakowsky, T. Kissel, Int. J. Pharm. 375 (2009) 97.24] J. Béjar, J. Porta, J.J. Borrego, M.C. Alvarez, Mar. Biotechnol. 7 (2005) 389.25] M.A. Esteban, V. Mulero, J. Munoz, J. Meseguer, Cell Tissue Res. 293 (1998) 133.26] X. Guo, S. Zhu, Food Chem. 162 (2014) 99.27] F.C. Cabello, Environ. Microbiol. 8 (2006) 1137.28] P. De Schryver, O. Vadstein, ISME J. (2014), http://dx.doi.org/10.1038/ismej.29] E. Ringø, R.E. Olsen, T.ø. Gifstad, R.A. Dalmo, H. AmLund, G.I. Hemre, et al., Aquac.

Nutr. 16 (2010) 117.30] E. De Giglio, A. Trapani, D. Cafagna, L. Sabbatini, S. Cometa, Anal. Bioanal. Chem.

400 (2011) 1997.31] M. Demirel, Y. Yazan, R.H. Müller, F. Kilic , B. Bozan, J. Microencapsul. 18 (2001)

359.

32] S. Das, A. Chaudhury, AAPS Pharm. Sci. Tech. 12 (2011) 62.33] J. Panyam, S.K. Sahoo, S. Prabha, T. Bargar, V. Labhasetwar, Int. J. Pharm. 262

(2003) 1.34] D. Pliszczak, C. Bordes, S. Bourgeois, P. Marote, H. Zahouani, S. Tupin, C.P. Mattei,

P. Lantéri, Colloids Surf. B: Biointerfaces 92 (2012) 168.

Page 10

8 faces B

[

[[[[[

[

[

[

[44] R.D. Handy, T.B. Henry, T.M. Scown, B.D. Johnston, C.R. Tyler, Ecotoxicology 17(2008) 396.

8 A. Trapani et al. / Colloids and Sur

35] G. Perera, J. Barthelmes, A. Bernkop-Schnürch, J. Control. Release 145 (2010)240.

36] T. Behera, P. Swain, Fish Shellfish Immunol. 5 (2013) 785.37] B.F. Sahin, M. Bayansal, N. Yuksel, H.A. Biyikli, C etinkara 40 (2014) 5237.38] M. Marquis, J. Davy, A. Fang, D. Renard, Biomacromolecules 15 (2014) 1568.39] J. Bejar, J.J. Borrego, M.C. Alvarez, Aquaculture 150 (1997) 143.

40] P. Castillo-Briceno, D. Bihan, M. Nilges, S. Hamaia, J. Meseguer, A. García-Ayala,

et al., Mol. Immunol. 48 (2011) 826.41] C. Daniels, S.H. Hoseinifar, Prebiotic applications in shellfish, in: E. Ringø, D.L.

Merrifield (Eds.), Aquaculture Nutrition: Gut Health, Probiotics and Prebiotics,Wiley-Blackwell Scientific Publication, London, 2014.

[[

: Biointerfaces 127 (2015) 79–88

42] S.I. Pérez-Prieto, S. Rodríguez-Saint-Jean, E. García-Rosado, D. Castro, M.C.Alvarez, J.J. Borrego, Dis. Aquat. Organ. 35 (1999) 149.

43] H.E. Jordan, Comparative haematology, in: H. Downey (Ed.), Handbook ofHematology, 1938, pp. 715–721.

45] H. Liu, Y. Liu, S. Liu, D.W. Pang, G. Xiao, J. Virol. 85 (2011) 6252.46] J. Shi, J.L. Choi, B. Chou, R.N. Johnson, J.G. Schellinger, S.H. Pun, ACS Nano 7

(2013) 10612.