Rational design of nanoparticles towards targeting antigen-presenting cells and improved T cell priming

Accepted Manuscript Rational design of nanoparticles towards targeting antigenpresenting cells and improved T cell priming

Eva Zupančič, Caterina Curato, Maria Paisana, Catarina Rodrigues, Ziv Porat, Ana S. Viana, Carlos A.M. Afonso, João Pinto, Rogério Gaspar, João N. Moreira, Ronit Satchi-Fainaro, Steffen Jung, Helena F. Florindo PII: DOI: Reference:

S0168-3659(17)30589-8 doi: 10.1016/j.jconrel.2017.05.014 COREL 8802

To appear in:

Journal of Controlled Release

Received date: Revised date: Accepted date:

4 March 2017 9 May 2017 12 May 2017

Please cite this article as: Eva Zupančič, Caterina Curato, Maria Paisana, Catarina Rodrigues, Ziv Porat, Ana S. Viana, Carlos A.M. Afonso, João Pinto, Rogério Gaspar, João N. Moreira, Ronit Satchi-Fainaro, Steffen Jung, Helena F. Florindo , Rational design of nanoparticles towards targeting antigen-presenting cells and improved T cell priming. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Corel(2017), doi: 10.1016/j.jconrel.2017.05.014

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Rational design of nanoparticles towards targeting antigen-presenting cells and improved T cell priming Eva Zupančič

a,b,c

c

a

a

d

e

, Caterina Curato , Maria Paisana , Catarina Rodrigues , Ziv Porat , Ana S Viana , a

a

a

b,f

g

Carlos A M Afonso , João Pinto , Rogério Gaspar , João N Moreira , Ronit Satchi-Fainaro , Steffen c

Jung , Helena F Florindo

a*

a

Research Institute for Medicines (iMed.ULisboa), Faculty of Pharmacy, Universidade de Lisboa, b

CNC - Center for Neuroscience and Cell Biology University of Coimbra, Portugal; c

PT

Portugal;

d

Departments of Immunology and Biological Services, The Weizmann Institute of Science, Rehovot, Israel;

e

Chemistry and Biochemistry Center, Sciences Faculty, Universidade de Lisboa, 1749-016

RI

f

Lisbon, Portugal; FFUC - Faculty of Pharmacy, University of Coimbra, Portugal;

SC

Physiology and Pharmacology, Sack ler School of Medicine, Tel Aviv University, Israel

Prof. Helena F Florindo, PhD Research Institute for Medicines (iMed.ULisboa),

MA

Faculty of Pharmacy, Universidade de Lisboa, Portugal Tel: + 351 217946400 Fax: +351 217946470;

EP T

ED

e-mail: [email protected] (H. F. Florindo)

AC C

NU

*Corresponding author:

g

Department of

ACCEPTED MANUSCRIPT Abstract Vaccination is a promising strategy to trigger and boost immune responses against cancer or infectious disease. We have designed, synthesized and characterized aliphatic-polyester (poly(lacticco-glycolic acid) (PLGA) nanoparticles (NP) to investigate how the nature of protein association (adsorbed versus entrapped) and polymer/surfactant concentrations impact on the generation and modulation of antigen-specific immune responses. The ability of the NP formulations to target dendritic cells (DC), be internalized and activate the T cells was characterized and optimized in vitro and in vivo using markers of DC activation and co+

+

PT

stimulatory molecules. Ovalbumin (OVA) was used as a model antigen in combination with the engraftment of CD4 and CD8 T cells, carrying a transgenic OVA-responding T cell receptor (TCR), +

+

RI

to trace and characterize the activation of antigen-specific CD4 and CD8 lymph node T cells upon NP vaccination. Accordingly, the phenotype and frequency of immune cell stimulation induced by the

SC

NP loaded with OVA, isolated or in combination with synthetic unmethylated cytosine-phosphateguanine (CpG) oligodeoxynucleotide (ODN) motifs, were characterized. DC-NP interactions increased with incubation time, presenting internalization values between

NU

50-60% and 30-40%, in vitro and in vivo, respectively. Interestingly, animal immunization with antigenadsorbed NP up-regulated major histocompatibility complex (MHC) class II (MHCII), while NP entrapping the antigen up-regulated MHCI, suggesting a more efficient cross-presentation. On the

MA

other hand, rather surprisingly, the surfactant used in the NP formulation had a major impact on the activation of antigen presenting cells (APC). In fact, DC collected from lymph nodes of animals immunized with NP prepared using poly(vinil alcohol) (PVA), as a surfactant, expressed significantly

ED

higher levels of CD86, MHCI and MHCII. In addition, those NP prepared with PVA and co-entrapping OVA and the toll-like receptor (TLR) ligand CpG, induced the most profound antigen-specific T cell +

+

response, by both CD4 and CD8 T cells, in vivo.

EP T

Overall, our data reveal the impact of NP composition and surface properties on the type and extension of induced immune responses. Deeper understanding on the NP-immune cell crosstalk can guide the rational development of nano-immunotherapeutic systems with improved and specific

Keywords

AC C

therapeutic efficacy and avoiding off-target effects.

alpha-lactalbumin; dendritic cells; ovalbumin; PLGA-PEG; polymeric nanoparticles; vaccine delivery

ACCEPTED MANUSCRIPT

1.

Introduction The induction of effective immune responses results from a complex interplay between the

innate (antigen-nonspecific) and adaptive (antigen-specific) immune cell compartments. Dendritic cells (DC) are the most potent antigen presenting cells (APC), which makes them one of the key players in the crosstalk between innate and adaptive immunity. Given their capacity to capture and process internalized antigens, DC efficiently prime naive T cells against foreign antigens and polarize them towards distinct effector fates [1-3]. Moreover, the activation mode of specific DC subsets through +

PT

antigen engulfment determines whether clinically relevant responses are dominated by the effect of +

cytotoxic CD8 T cells or helper CD4 T cells [4].

RI

Vaccination protocols that target DC are promising strategies for the ini tiation and enhancement of immune responses for cancer treatment [5, 6], as well as viral and microbial

SC

infections [7-9]. Specifically, the concept of immunotherapy aims to harness the patient's own immune system to selectively target malignant or infected cells by activating in vivo immune effectors against tumor-associated antigens (TAA) or microbes [5, 10].

NU

Nanoparticle (NP)-based delivery systems hold great promise as immunotherapeutic platforms towards an effective in vivo targeting of immune cells. The use of nanotechnology in immunotherapy has indeed distinct advantages over the systemic delivery of the immunotherapeutic agents in

MA

solution, especially due to NP larger size and consequent altered biodistribution profile. NP can improve antigen stability, besides enabling a high loading capacity, as the result of their high surface to volume ratios. Polymeric NP can be biocompatible and biodegradable, with tunable sub-cellular size

ED

and controlled release properties [11, 12]. These carriers can therefore extend antigen exposure to immune cells, facilitate antigen uptake by DC and modulate antigen presentation pathways within these cells, thus enhancing T cell-mediated effector immune responses. However, there is a limited

EP T

knowledge about the specific mechanisms of interaction between the NP and immune cells. Herein we describe the rational development of different poly(lactic-co-glycolic acid) (PLGA)based nanosystems, characterized by specific physico-chemical parameters suitable for optimal

AC C

delivery of antigens and adjuvants. The composition of these nanoparticulate carriers was optimized by using Chitosan (CS) and distinct surfactants, such as Poly(vinil alcohol) (PVA) and Pluronic® F127 (PF127) in order to attain the successful entrapment of multiple antigens and the most adequate parameters for their recognition and capture by DC. These NP were deeply characterized in order to assess their size and size distribution, surface charge, nature of protein-polymer interaction and physical state of polymers. The developed NP successfully entrapped multiple bioactive molecules (antigens and adjuvants) and showed physicochemical properties adequate for DC targeting, with an average size below 200 nm and a neutral surface charge. DC extensively captured these polymeric NP, both in vitro and in vivo, successfully inducing the proliferation of T cells. Our data provide important insights into the correlation between the composition, method of antigen association and physico-chemical properties of NP with immune cell modulation, namely DC activation and T cell priming.

ACCEPTED MANUSCRIPT

2.

Materials and Methods 2.1 Materials Poly(lactic-co-glycolic acid) (PLGA) conjugated with polyethylene glycol (PEG), i.e. PLGA ®

PEG (5050 DLG mPEG 5000, 10 wt % PEG, molecular weight (MW) = 17 kDa), Resomer RG 502 Poly(lactic-co-glycolic acid) PLGA (MW = 7–17 kDa), Alpha-lactalbumin (LALBA, average MW = 14 kDa), Ovalbumin (OVA, average MW = 45 kDa), Poly (vinil alcohol) (PVA, MW = 13-23 kDa) and hydrolyzation 87-89%, Pluronic

®

F127 (PF127, MW = 12.6 kDa), Glycol Chitosan (CS), Sucrose and

PT

Dichloromethane (DCM) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Double distilled ®

water was used after filtration in a Millipore (Millipore, Billerica, MA) system. All other chemicals and

RI

reagents used in the study were of analytical grade.

MW markers for sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) TM

Prestained Protein Ladder 10-180 kDa #SM0671) were

purchased from Thermo Fisher Scientific (MA, USA).

SC

(Thermo Fisher Scientific, PageRuler

The Roswell Park Memorial Institute medium (RPMI 1640), heat inactivated fetal bovine s erum

NU

(FBS), L-glutamine 200 mM, 0.25 % (w/v) Trypsin/ 0.53 mM ethylenediamine tetracetic acid (TrypsinEDTA) solution Penicillin/Streptomycin (Pen-Strep) 10000 Unit/mL/10000 μg/mL and AlamarBlue

®

were purchased from Life Technologies (Carlsbad, CA, USA).

MA

Granulocyte-macrophage colony stimulating factor (GM-CSF) and interleukine-4 (IL-4) were purchased from PeproTech Inc. (Rocky Hill, NJ, USA).

Fixation/Permeabilization Kit (BD Cytofix/CytopermTM Cat: 554714) was purchased from BD

ED

Bioscience, San Diego, CA, USA.

The CpG ODN 1826 Vaccigrade InvivoGen (San Diego, CA, USA).

TM

(TCCATGACGTTCCTGACGTT) was purchased from the

EP T

Antibodies anti-mouse CD11c (clone: N418), CD11b (clone: M1/70), CD4 (clone: CK1.5), b

CD8α (clone: 53-6.7), I-Ab (MHCII; clone: AF6-120.1), H-2K (MHCI; clone: AF6-88.5), CD45.1 (clone: A20),

TCRβ (clone: H57-597), TCRVα2 (clone: B20.1), CD45R/B220 (clone: RA3-6B2), and

Carboxyfluorescein diacetate succinimidyl ester (CFSE) were purchased from BioLegend (San Diego,

AC C

CA, USA). Antibodies anti-mouse IL-12p40 (clone: C17.8) and IL-1β (clone: NJTEN3) were purchased from eBioscience.

2.2 Mice

Female OT-I (C57BL/6) TCR transgenic mice harbouring OVA-specific CD8 T cells [13] and OT-II (C57BL/6) TCR transgenic mice harbouring OVA-specific CD4 T cells [14] were in-house bred at the Weizmann Institute of Science (Rehovot, Israel), under conventional, specific pathogen-free (SPF) conditions. Female C57BL/6 mice (6-8 weeks old) were purchased from the Envigo RMS LTD (Jerusalem, Israel). All animals were handled according to protocols approved by the Weizmann Institute Animal Care Committee (IACUC) as per international guidelines. 2.3 Preparation of nanoparticles

ACCEPTED MANUSCRIPT Polymeric NP were aseptically prepared by the double emulsion (w/o/w) solvent evaporation method [15]. Briefly, the single emulsion (o/w) was formed by dispersing an organic phase, composed by the polymer dissolved in dichloromethane (DCM), into an aqueous solution using an ultrasonic processor (Sonifier Vibracell VC 375, Sonics & Materials Inc, USA), under continuous conditions for 15 s at 70 W, in an ice bath. Fluorescent NP were formulated by replacing one tenth of the polymer mass by Rhodamine-PLGA, synthesized in-house as previously described [16], or by replacing 50 μL of the organic polymer solution with (2 mg/mL) Rhodamine G6 (Sigma-Aldrich) solution in DCM. A 2% (w/v) PVA solution was added to the o/w primary emulsion and a second sonication was performed

PT

under the previously described conditions. The double emulsion (w/o/w) was added dropwise to an external aqueous phase, composed by a 0.3% (w/v) PVA or PF127 surfactant solution. The solvent

RI

evaporation was carried out by a gentle magnetic stirring at 37ºC during 1 h, thus enabling the formation of NP. The polymeric NP were harvested by centrifugation (22 000 x g, 45 min, 4ºC; ®

SC

Beckman Coulter, lnc, Avanti J-E Centrifuge JA-20, USA) and rinsed with ultrapure water, three times to remove excess of surfactant and non-associated antigen. NP were re-suspended in PBS pH 7.4 and kept at 4ºC, until further analysis. Two different strategies were followed to achieve the

NU

association of antigens to NP: i) antigen dissolution in the internal aqueous phase, leading to its dispersion within NP matrix (Entrapped (Entrap) protein NP); and ii) adsorption of protein through the incubation of 10 mg/mL protein solution (100 μL) with plain NP for 1 h, at room temperature (Adsorbed

MA

(Ads) protein NP). Protein-adsorbed NP were washed with ultrapure water and centrifuged at 22 000 x g for 20 min at 4ºC. When CpG ODN was used as adjuvant, it was dissolved in the internal phase, despite the method used for antigen association to NP. It is important to note that AdsNP thus

ED

presented the antigen adsorbed onto the surface of NP, while the CpG was dispersed within the carrier matrix. On the other hand, EntrapNP presented both antigen and adjuvant co-entrapped within

EP T

the polymeric matrix.

2.4 Size and surface charge analysis of PLGA-PEG-based NP NP size (Z-Average) and polydispersity index (PdI) were determined by Dynamic Light

AC C

Scattering (DLS), using the Malvern Nano S (Malvern Instruments, UK). Briefly, the final 20 mg/mL NP suspension was re-suspended in PBS pH 7.4 to prepare 1 mg/mL NP suspensions. The conditions were kept constant to have comparable results. Readings were obtained based on the scattering of laser light within the equipment by the Brownian motion of NP. Each analysis was carried out in triplicates, at 25°C. NP surface charge was inferred by their zeta potential, assessed by Laser Doppler Velocimetry (LDV) with Malvern Nano Z (Malvern Instruments, UK). Since zeta potential is highly dependent on measurement conditions, NP concentration and pH/ionic strength of the dispersant were maintained constant. All measurements were performed at 25°C in triplicates.

2.5 Study of PLGA-PEG NP morphology by atomic force microscopy The morphology of NP was investigated by Tapping mode atomic force microscopy (AFM) (Nanoscope IIIa Mulimode AFM, Digital Instruments, Veeco), using silicon tips (circa 300 kHz) at a

ACCEPTED MANUSCRIPT scan rate of ca. 1.6 Hz. Samples were diluted in purified water (10 mg/mL), a drop was placed onto freshly cleaved mica for 20 min and dried with N2 prior to analysis.

2.6 Quantification of protein loading The entrapment efficiency (EE, %) and the loading capacity (LC, μg/mg) of proteins were quantified indirectly by a reverse-phase high-performance liquid chromatography (HPLC) (RP-HPLC), using Beckman System Gold, with a UV-Vis detector (Beckman 166) at 220 nm, Beckman 126 solvent module and a Midas autosampler. Samples (20 µL) of the supernatants obtained from NP

PT

centrifugation were injected onto a Shodex Protein KW -803 series column (8.0 mm ID x 300 mm, 5 µm particle size, 300 Å pore size) and eluted with 50 mM sodium phosphate buffer (pH 7.0) plus 0.3 M

RI

NaCl at 1 mL/min, for 16 min at room temperature. Plain NP were used as controls and all measurements were performed in triplicates. The amount of protein present in each sample was 2

SC

calculated using a linear standard curve generated in the range of 0.5 – 5.0 mg/mL, r = 0.9987. The EE (%) and the LC (μg/mg) of protein were calculated as indicated in Equations 1 and 2.

NU

EE % = (total amount of protein – amount of protein in supernatant)/ (total amount of protein) × 100 (1),

MA

LC µg/mg = (total amount of protein – amount of protein in supernatant)/ (total amount of polymers) (2)

ED

2.7 PVA assay

To determine the residual amount of PVA at NP, this surfactant was indirectly quantified by HPLC using Beckman System Gold: UV-Vis detector (Beckman 166), Beckman 126 solvent module,

EP T

Midas autosampler, and Shodex Protein KW -803 column (8.0 mm ID x 300 mm, 5 µm particle size, 300 Å pore size). The surfactant washing efficiency was calculated as the percentage of PVA

AC C

quantified in the supernatants compared to the total amount used for NP formulation.

2.8 Freeze-drying of NP NP were freeze-dried in the presence of the 0.05% (w/v) sucrose, which was used as a cryoprotectant. Prior to lyophilization, all samples were placed into glass vials and froz en at -80ºC. On the following day, samples were freeze-dried for 24 h at -45°C, 0.2 mbar in a VirTis BenchTop L freeze-drier (SP Scientific, NY, US). Lyophilized NP were only used for Fourier transform infrared spectroscopy (FTIR) and thermal characterization.

2.9 Integrity of the NP-entrapped protein The assessment of protein integrity was performed by SDS-PAGE. LALBA standard solution (250 μg/mL) and NP suspension (20 mg/mL) were mixed with SDS-containing loading buffer and digested at 95ºC for 20 min. Samples were loaded at room temperature onto a 10% (w/w) polyacrilamide mini-gel (20 μL for samples and 5 μL for MW markers (Thermo Fisher Scientific, 10-

ACCEPTED MANUSCRIPT 180 kDa)) and electrophoresis was performed at a constant voltage of 170 V for 50 min, using a BioRad 300 power pack (Bio-Rad, Hercules, CA, USA). Gels were further stained with 0.025% (w/v) Coomassie Brilliant Blue G-250 (Thermo Fisher Scientific, MA, USA) to reveal protein bands [17].

2.10 FTIR characterization of NP Each sample was mixed with KBr at a concentration of 0.5% w/w. The mixtures (200 mg) were compressed into a tablet of 12 mm diameter. The infrared spectra were measured at the absorption -1

PT

mode (IR Affinity-1 Shimadzu spectrophotometer, Japan) with 64 scans, with a resolution of 2 cm .

2.11 Thermal characterization of NP

RI

The thermal properties of the samples were determined by differential scanning calorimetry (DSC; TA instruments, Q200, USA) after calibration with indium (TA instruments, USA; Tf us =156.55ºC, Prior to the

SC

Δf us H = 28. 51 J/g). Dry N2(g) was used as the purge gas (Air Liquide, 50 mL/min).

experiment, all samples were kept in dry atmosphere, with 11% of humidity (Lithium chloride supersaturated solution), for one week. The samples (2-3 mg) were placed into hermetic crucibles and

NU

heated up at 5ºC/min, from 0ºC to 100ºC. An empty crucible was used as reference.

2.12 Stability of NP

MA

Aseptically prepared polymeric NP were dispersed in PBS pH 7.4 and kept at 4ºC. After predetermined periods of time, samples were collected over 3 months and the particle size, P dI and

ED

zeta potential were evaluated.

2.13 In vitro evaluation of bone marrow derived dendritic cell (BMDC) viability in the presence of NP

EP T

DC were derived from bone marrow cells isolated from femura and tibiae of C57BL/6 mice, and cultured in RPMI medium 1640 supplemented with GM-CSF (50 ng/mL) and IL-4 (50 ng/mL) at 37ºC, 5% CO2. After 7 days, immature BMDC were harvested, washed and used for the in vitro experiments. Cell viability of BMDC was quantitatively evaluated by measuring the metabolic activity ®

AC C

of mitochondrial dehydrogenase by AlamarBlue . DC (5000 cells/100 μL/well) were incubated with increasing concentrations of different NP-based systems (100, 125, 150, 200, 250, 500, 1000 μL/mL), ®

in 96-well plates (Nunc , Roshilda, Denmark), for 21 h at 37ºC with 5% CO2. Cells in culture medium were used as a negative control, while the positive control was obtained after the treatment of cells ®

with a 0.5% (v/v) Triton X-100 (Gibco ) solution. After NP incubation, 10 μL of AlamarBlue

®

reagent

were added to each well and incubated for additional 3 h. After this total incubation period of 24 h, the absorbance of AlamarBlue

®

was determined with a Fluostar Omega microplate reader (BMG LAbtech,

Ortenberg, Germany) at 570 nm, using 600 nm as a reference wavelength.

2.14 Internalization of antigen-loaded NP by BMDC BMDC were seeded in 24-well plate (1.5 × 10

5

cells/ well), 24 h prior administration of

fluorescent NP (250 μg/mL) to cells. After incubation with OVA-loaded NP for 3, 6, 16 or 24 h, cells

ACCEPTED MANUSCRIPT were harvested, rinsed three times with PBS and stained with anti-mouse CD11c and CD11b antibodies, as well as, with DAPI for dead cells exclusion. Cells were analyzed by multispectral imaging flow cytometry (ImageStreamX markII flow cytometer; Amnis Corp, part of EMD millipore, Seattle, WA). A 60x magnification was used for all samples. At least 30,000 cells were collected for each sample and data were analyzed using a dedicated image analysis software (IDEAS 6.2; Amnis Corp). Images were compensated for fluorescent dye overlap by using single-stain controls. Cells were gated for single cells using the area and aspect ratio features, and for focused cells using the Gradient RMS feature [18]. Cells were

PT

further gated using a bivariate plot for circularity (the degree of the mask’s deviation from a circle) based on the object mask (a segmentation mask that creates a tight fit on the cell morphology) and

RI

the intensity of the side scatter channel (illuminated by the 785 nm laser and collected in channel 12). Particle internalization was calculated by the internalization feature, i.e. the ratio of the

SC

intensity inside the cell to the intensity of the entire cell, mapped to a log scale. To define the internal mask for the cell, the object mask of the bright field image was eroded by 5 pixels .

NU

2.15 In vivo study of activation markers and co-stimulatory molecules expressed on DC in draining lymph nodes

Female C57BL/6 mice, 8-weeks-old (n = 3/ group) were injected into the right flank by

MA

subcutaneous (s.c.) hock immunization with 50 μL of one of the following fluorescent Rhodamine G6labeled NP formulations (20 mg/mL): EntrapOVA(PVA), EntrapOVA(PF127), AdsOVA(PVA), and AdsOVA(PF127). Non-injected left flank served as a negative control. After 16 h post-immunization

ED

(p.i.), popliteal and inguinal lymph nodes were harvested and homogenized in a si ngle cell suspension. Cells were stained with fluorescent-labeled anti-mouse antibodies against CD11c, MHCII b

b

(I-A ), MHCI (H-2K ), CD80 and CD86, for 20 min at 4ºC. Samples were acquired with an LSR II

EP T

Fortessa flow cytometer (BD Bioscience) and analyzed with FlowJo software (Treestar).

2.16 Cytokines profile of DC upon NP immunization Female C57BL/6 mice, 8-weeks-old (n = 4/ group) were immunized by subcutaneous (s.c.)

AC C

hock immunization with 50 μL of NP suspension (20 mg/mL): EntrapOVA(PVA), EntrapOVA CpG(PVA), AdsOVA(PVA) and AdsOVA-CpG(PVA). Non-immunized mice served as a negative control. At 16 h p.i. the inguinal and popliteal lymph nodes were harvested, homogenized in a single cell suspension and intracellular cytokine staining was performed using the Cytofix/Cytoperm plus kit (BD Bioscience). Briefly, single cell suspensions of lymph nodes were washed with PBS. Extracellular antigens

(CD11c,

Ab-I,

TCRβ

and

CD45R/B220)

were

stained prior to cell

fixation and

permeabilization with 250 μL Cytofix/Cytoperm solution for 20 min. The staining for intracellular cytokine was performed with anti-cytokine fluorescent antibody conjugates to detect accumulated IL12p40 and IL-1β cytokines. Cells were washed with the permealization/washing solution and resuspended in 400 μL of FACS buffer prior to flow cytometry analysis (LSR II Fortessa, BD Bisociences). All samples were analyzed by FlowJo software (TreeStar, San Carlos, CA, USA).

ACCEPTED MANUSCRIPT 2.17 Analysis of in vivo T cell activation after NP immunization Eight-weeks-old female C57BL/6 mice (n = 6 mice/ group) were intravenously (i.v.) engrafted +

with CD8

+

OT-I and CD4

OT-II cells isolated by MACS from spleens and lymph nodes of the

respective TCR transgenic mice. Prior to the transfer, both T cell populations were labeled with 5 mM CFSE, for tracking in vivo proliferation [19]. On the following day, animals were injected s.c. (hock immunization) into the right flank with 50 μL of NP suspension (20 mg/mL): EntrapOVA(PVA), EntrapOVA-CpG(PVA), AdsOVA(PVA) and AdsOVA-CpG(PVA). Non-injected left flank served as a negative control. Three days post-immunization with NP, inguinal and popliteal lymph nodes were

PT

isolated. Single cell suspension was stained with CD4, CD8α, TCRβ, TCRVα2, and CD45.1 antibodies

RI

and analyzed by flow cytometry.

2.18 Statistics

SC

All results are presented as mean ± standard deviation (SD). One-way ANOVA analysis, following by a Tukey-Kramer test was applied to demonstrate statistical differences (p ˂ 0.001 (***), p ˂ 0.01 (**), p ˂ 0.05 (*)). For direct comparison between two groups of data, t -student test was ®

NU

applied. All tests were performed in GraphPad Prism 5 (GraphPad Software, Inc., La Jolla, CA) with a

3.

MA

statistical significance level of 0.05 and 0.001 considered very significant.

Results

ED

3.1 Physicochemical characterization of NP

In this study, we focused on the development of stable colloidal suspensions for the delivery of protein antigen to DC, the key APC that trigger responses of naive T cells. NP composition, size

immune response.

EP T

distribution, surface charge, shape and hydrophobicity are key parameters for successful targeting of

Four different NP delivery systems (Table 1) with distinct chemical compositions and different

AC C

types of Alpha-lactalbumin (LALBA) protein association (entrapped versus adsorbed) were developed and optimized. The biodegradable aliphatic polyester copolymers PLGA-PEG or PLGA, together with CS, were the main components of the NP matrix, and two different non-ionic surfactants (PVA or PF127) were used to ensure NP size below 200 nm with narrow distribution, PdI lower than 0.2, zeta potential close to neutrality, high loading efficiency and a minimum surfactant residue, particularly important to guarantee NP safety and biocompatibility . The three independent batches of each NP revealed to be highly consistent and reproducible (Table 2). EntrapLALBA(PVA) and EntrapLALBA(PF127) NP displayed an average size of 171 and 185 nm, respectively. On the other hand, the size of NP with adsorbed LALBA protein, AdsLALBA(PVA) and AdsLALBA(PF127), was 188 nm and 195 nm, respectively. All formulated NP batches presented surface charges close to neutrality, independently of the surfactant used in the external phase (PVA or PF127). This can be explained by the presence of the PEG–chains on the surface of the EntrapLALBA(PVA)

and

EntrapLALBA(PF127)

NP.

The

best

EE

was

obtained

for

the

ACCEPTED MANUSCRIPT EntrapLALBA(PVA) formulation, presenting nearly 80% of the initial protein solution entrapped within PLGA-PEG NP. When PF127 was used at external phase (EntrapLALBA(PF127)), the EE was close to 70%. However, it is important to highlight that despite the surfactant used to ensure NP stability, overall EE were considerably higher than those reported by other recently published studies [16, 20]. The same trend of results was obtained for the LC. Moreover, the adsorption efficiency of AdsLALBA(PVA) and AdsLALBA(PF127) NP was considerably close to 40% or 50%, respectively. The branched chemical structure of PF127 enabled a more efficient and stable protein adsorption by AdsLALBA(PF127) NP than NP prepared with PVA polymer. In all preparations, the percentage of

Table 1: Composition of polymeric NP formulations.

RI

Surfactant internal Phase

Matrix polym er

Aqueous phase com position

EntrapPROTEIN( PVA) EntrapPROTEIN( PF127)

PLGA-PEG

10 % (w /v) PVA, CS, protein

PLGA

10 % (w /v) PVA, CS

2% (w /v) PVA

Surfactant External Phase

Protein adsorption

0.3 % (w /v) PVA 0.3 % (w /v) PF127

-

0.3 % (w /v) PVA 0.3 % (w /v) PF127

+

NU

AdsPROTEIN(PF127)

SC

Form ulation

AdsPROTEIN(PVA)

PT

surfactant washing efficiency was higher than 96%, promoting NP safety and biocompatibility.

Table 2: Physico-chemical characterization of polymeric NP by DLS and LDV. Zeta potential (m V)

Protein loading (μg/m g) Loading Surface capacity adsorption

Size (nm )

PdI

EntrapLALAB(PVA) EntrapLALBA(PF127)

171 ± 3 185 ± 4

0.15 ± 0.02 0.17 ± 0.05

0.51 ± 0.1 0.37 ± 0.4

19.56 16.96

-

AdsLALBA(PVA) AdsLALBA(PF127)

188 ± 11 195 ± 10

0.18 ± 0.08 0.20 ± 0.09

-1.89 ± 0.5 -1.42 ± 0.2

-

19.72 25.10

ED

MA

Form ulation

3.2 NP morphology by AFM

EP T

The size and surface morphology of NP were also assessed by AFM (Figure 1). All NP presented smooth surfaces and spherical topography. The section analyses of PLGA -PEG NP with entrapped antigen showed uniform size distributions, which correlated with the data obtained by DLS

AC C

analyzer. The average NP diameter was 30 nm smaller than the hydrodynamic diameter obtained by DLS. As the NP were dried prior to AFM analysis, PEG chains could not form a corona around the NP, which can explain smaller NP sizes. However, there were no considerable differences between the formulations EntrapLALBA(PVA) (Figure 1A) and EntrapLALBA(PF127) (Figure 1B). Hence, the different surfactants present in the external phase did not greatly influence size or morphology of NP. On the other side, PLGA NP with adsorbed antigen demonstrated a wider particle size distribution, with an average diameter closer to the one determine by DLS. Also, no differences were found between the AdsLALBA(PVA) (Figure 1C) and AdsLALBA(PF127) NP (Figure 1D), despite the use of different surfactants in the external phase. However, AFM images showed a clear distinction between PEGylated and non-PEGylated PLGA NP. Formulations containing PLGA-PEG (EntrapLALBA(PVA) and EntrapLALBA(PF127)) showed regular spherical NP forming clusters due to the presence of hydrophilic PEG chains at NP surface, which tend to accommodate water molecules, precluding a sharp imaging of particles boundaries. This latter effect was not observed when PLGA polymer

ACCEPTED MANUSCRIPT (AdsLALBA(PVA) and AdsLALBA(PF127)) was used for NP formulation. These PLGA NP presented

SC

RI

PT

spherical shaped NP with well-defined contours.

Figure 1: Size and morphology characterization of formulated NP by AFM. Section analysis of NP EntrapLALBA(PVA) (A) and

NU

EntrapLALBA(PF127) (B) w ith the matrix polymer PLGA -PEG and w ith entrapped antigen; AdsLALBA(PVA) (C) and AdsLALBA(PF127) (D) w ith PLGA polymer and adsorbed LALBA protein.

MA

3.3 FTIR and DSC studies on lyophilized NP

FTIR spectroscopy allows the rapid and efficient identification of protein and polymer integrity in lyophilized NP. Differences related with the presence of the LALBA antigen entrapped into the

ED

polymeric matrix or adsorbed onto the surface of NP were also evaluated by FTIR (Figure 2A). All NP and both polymer (PLGA, PLGA-PEG) powders presented a strong stretching absorption at 1720 cm

−1

(§) which corresponds to C=O bond, indicating the presence of a carboxylic acid. The band at 2950 −1

(ф) indicated the presence of C-H bonds (ethylene glycol) [21] that was clearly more pronounced blank

in Entrap

EP T

cm

(PVA) and Entrap

blank

(PF127) NP, when PLGA-PEG was used as a matrix polymer. From -1

the LALBA spectrum, one can see two specific absorption bands at approximately 1600 cm . One at 1662 cm

-1

(*), specific for primary amides, while the peak at 1562 cm

-1

(Ϫ) was distinctive for -NH2

AC C

groups. Both of these bands were also identified on the FTIR of the AdsLALBA(PVA) and AdsLALBA(PF127) samples, where LALBA was adsorbed onto the NP surface. The slight shift to the higher wave numbers indicated changes on the hydrogen bonding strength between the protein and the polymer. The broad and strong band at 3400 cm

-1

(‡) was related to the overlapping of hydroxyl

and amine groups [22]. Also these bands were more pronounced for NP with adsorbed LALBA. The absence of these bands in the spectrum of EntrapLALBA(PVA) and EntrapLALBA(PF127) suggest ed that, when entrapped, the protein was well protected within the matrix of the polymer. DSC studies were performed with lyophilized NP at a rate of 5ºC/min from 0ºC to 100ºC. This thermo-analytical technique provided information on the thermal transformations of biodegradable polymers and the LALBA protein in NP (Figure 2B). Analysis of PLGA and PLGA-PEG demonstrated the presence of an endothermic event between 40ºC-50ºC. This endothermic band resulted from the glass transition of the polymers followed by an endothermic relaxation peak. The glass temperature (Tg) of polymers depends strongly on their composition and MW. Results show that the Tg values of

ACCEPTED MANUSCRIPT biodegradable PLGA copolymer were lower than those obtained for PLGA-PEG, due to the higher MW of PLGA–PEG. This finding was already previously reported [23]. When empty and LALBA-entrapped NP were evaluated, Tg were in the same range as Tg of PLGA and PLGA -PEG, revealing that the formulation process did not affect the polymer structure. Moreover, no specific interactions between the polymers and proteins were detected. From the LALBA thermogram there was no melting peak visible between 50ºC-70°C, in contrast to what has been previously reported in the literature [24, 25]. This suggests that the protein was in an amorphous phase (which was confirmed by XRPD analysis, data not shown). Also, no melting peak of native LALBA was detected in all NP with entrapped protein,

EP T

ED

MA

NU

SC

RI

PT

suggesting that the LALBA protein is in a disordered phase within the polymer matrix of NP [23].

Figure 2: FTIR and DSC studies on lyophilized NP. (A) FTIR spectrographs of NP and the components used for their

AC C

formulations. The symbols on the top of the graph correspond to the specific band. Peak at 1720 cm−1 (§) is a specific band for C=O bond; 1662 cm-1 (*) is specific for primary amides; peak at 1562 cm-1 (Ϫ) corresponds to vibration of -NH2 groups; 2950 cm−1 (Ф) indicates the presence of C-H bond; 3400 cm-1 (‡) correlates to the hydroxyl and amine groups overlapping. (B) DSC specters of formulated NP and the components used for their formulations .

3.4 Formulation procedures did not affect protein integrity A SDS-PAGE was performed to verify if the protein MW was altered during the NP formulation process (Figure 3). Protein band patterns of migration obtained for LALBA standard solution and those extracted

from

loaded

NP

(EntrapLALBA(PVA),

EntrapLALBA(PF127),

AdsLALBA(PVA),

and

AdsLALBA(PF127)) were identical to the ones obtained for LALBA standard solutions (14 kDa), suggesting that the MW of the protein was not affected by the NP formulation process. These observations are corroborated by the absence of those bands next to 14 kDa of the gel , in lanes 6, 8,

ACCEPTED MANUSCRIPT 11 and 13. Moreover, no other bands were detected at different MW in the solution extracted from

NU

SC

RI

PT

loaded NP, supporting the absence of protein fragmentation or aggregation.

MA

Figure 3: SDS–PAGE (12 % separating gel) of protein LALBA before and after entrapment in NP. Lanes: (1) standard molecular w eight (MW) markers, w ide range 10 – 180 kDa; (2) LALBA standard solution at 2 μg; (3) 4 μg; (4) 8 μg; (5) 16 μg all LALBA’s bands may be seen at the 14 kDa; (6) Entrap blank(PVA); (7) EntrapLALBA(PVA); (8) Entrap blank(PF127); (9) EntrapLALBA(PF127); (10) Ads blank(PVA); (11) AdsLALBA(PVA); (12) Adsblank(PF127). (13) AdsLALBA(PF127); A representative

ED

image of 3 independent experiments is presented.

3.5 NP stability over 3 months

EP T

Average size, PdI and zeta potential of polymeric NP were measured over a 3 months period, when dispersed in PBS pH 7.4 and stored at 4ºC. The main challenge of colloidal suspensions is to ensure their stability over time.

As it can be seen in Figure 4, the EntrapLALBA(PVA) and EntrapLALBA(PF127) formulations

AC C

showed high stability over a period of 60 days. Thus, there were no significant changes and deviations in size or PdI of NP over that period. This stability in suspension can be explained by steric hindrance and repulsion of hydrophilic PEG chains [26, 27]. The size of the NP was decreased slightly from 182 to 170 nm for the formulation EntrapLALBA(PVA), and 192 to 175 nm for EntrapLALBA(PF127), respectively. The zeta potential was kept close to neutrality, with a shift from slightly positive (0.3 mV) to negative (-1.2 mV). However, following 2 months, a drastic increase in size and PdI was detected, while zeta potential was decreased. These data suggest a possible degradation of polymeric PLGAPEG matrix through solvent penetration and bulk erosion due to hydrolytic cleavage of polymer backbone, which is in accordance with the characteristics of these aliphatic polyesters [28]. In fact, NP polymeric fragments are released from the NP matrix and not only from its surface, as this polyester is a bulk eroding polymer [29, 30], and therefore even the steric and hydrated barrier enabled by PEG could not avoid the aggregation of EntrapLALBA(PVA) and EntrapLALBA(PF127) NP.

ACCEPTED MANUSCRIPT On the other hand, the formulations AdsLALBA(PVA) and AdsLALBA(PF127) with adsorbed protein onto the surface of PLGA matrix, showed an increase in size of 80 and 40 nm, respectively, after day 30, suggesting a faster polymer degradation. This was further supported by an increased P dI (> 0.35) that occurred 20 days after NP storage in suspension at 4°C. It appears that the PF127 amphiphilic

character and branched chemical

structure led to the NP (AdsLALBA(PF127))

EP T

ED

MA

NU

SC

RI

PT

suspensions with higher stability over time and increased protein association affinity.

Figure 4: Variation of the average size (A and B), polydispersity index (PdI) (C), and zeta potential (D) of polymeric NP measured over a 3 month period w hen dispersed in PBS pH 7.4 and stored at 4˚C. Circles (●) present formulation EntrapLALBA(PVA), squares (■) EntrapLALBA(PF127), triangles (▲) AdsLALBA(PVA) and diamonds (▼) AdsLALBA(PF127)

AC C

F4.All results are expressed as mean ± SD (n > 3).

3.6 NP did not change the viability of immature BMDC and were extensively internalized by these APC

We focused on the development of a NP-based vaccine for antigen and adjuvant delivery to DC, and therefore, it was crucial to guarantee that these carriers did not impair the viability of functional APC. The viability of BMDC was not affected by any of the developed NP, in a concentration up to 1000 μg/mL (Figure 5).

PT

ACCEPTED MANUSCRIPT

Figure 5: Effect of NP on the viability of BMDC w as studied in vitro using the AlamarBlue® assay. We incubated bone marrow derived DC w ith increasing concentrations (100–1000 μg/mL) of NP for 24 h. Culture medium (0) and triton 0.5% (w /v) w ere

RI

used as negative and positive controls, respectively (Mean ± SD; n = 6).

TM

SC

Subsequently, NP-exposed BMDC were analyzed with high throughput and spatial resolution using imaging stream flow cytometry (ImageStream ) [31] to study the interaction and uptake of

AC C

EP T

ED

MA

NU

OVA-loaded carriers by these APC (Figure 6, Suppl. Figure S1).

Figure 6: In vitro NP internalization by BMDC. ImageStream analysis w as performed to detect PLGA -Rhodamine labeled NP in BMDC, 3, 6, 16 and 24 h after the incubation in vitro. Distribution of at least 30,000 cells w as calculated by Amnis IDEAS softw are and presented w ith the histograms show ing the percentages of CD11c + cells interacting w ith NP (A) and of CD11c +

ACCEPTED MANUSCRIPT cells w ith internalized NP (B). Examples of representative images captured by the Amnis ImageStreamX Flow Cytometer of DC incubated for 16 h at 37°C w ith AdsOVA(PVA) (C). First column show s brightfield (BF) images of the cells, second column show s images of PLGA-Rhodamine labeled NP (NP), third column show s APC-labeled CD11c + DC (CD11c), fourth column show s merged images of the precedent fields (BF/NP/CD11c) and fifth column show s DAPI signal for dead cells. (C.a) CD11c + w ithout NP. (C.b) NP onto the surface of the CD11c + cell. (C.c) Various NP entering the cytoplasm of CD11c + cell. (C.d) DC w ith various NP. (C.e) DC w ith star morphology, indicator of the DC maturation process. Results are mean ± SD, normalized to that of control non-labeled cells, representative of three independent experiments (n = 3). Scale bar, 7 μm.

These in vitro studies

demonstrated that

the

DC-NP

interactions, in particular for

PT

EntrapOVA(PVA) and AdsOVA(PVA) NP (Figure 6A), increased significantly after 16 h, thus presenting higher total fluorescence, max pixel and median pixel intensities.

Percentages of DC with internalized NP ranged from 50%-60%, and NP persisted for 1 day

RI

inside these DC (Figure 6B). The interaction between BMDC and NP containing the adsorbed antigen (AdsOVA(PVA) and AdsOVA(PF127)) was lower than the one obtained for protein-entrapped NP

SC

(Figure 6B).

Moreover, the distinct chemical composition of developed NP influenced their internalization

NU

profile by BMDC. NP formulations prepared with PF127 emulsifier in the external phase demonstrated a constant DC-NP interaction of ̴ 60% for EntrapOVA(PF127) and ̴ 40% for AdsOVA(PF127) (Figure 6A & B).

MA

Figure 6C shows different internalization states of green-fluorescent AdsOVA(PVA) NP by red+

+

labeled CD11c cells following 16 h of incubation. These images show CD11c cells in the absence of NP (Figure 6C.a), NP onto the surface of the CD11c+ cells (Figure 6C.b) and CD11c+ with various NP +

ED

entering the cytoplasm (Figure 6C.c & d). Of note, the star-shape of CD11c cells in the Figure 6C.e indicates the characteristic morphology change observed upon DC maturation.

EP T

3.7 In vivo impact of NP on DC in draining lymph nodes of challenged animals The in vitro experiments showed that all NP formulations were efficiently internalized by BMDC in culture. To test if this also holds for a physiological setting, we immunized C57BL/6 mice into a flank with NP and used the non-immunized contralateral flank as control. Rhodamine-labeled NP

AC C

allowed the distinction and comparison of subpopulations of DC with and without internalized NP. First, we tested the expression of MHCI, MHCII, and the co-stimulatory molecules CD80 and CD86 on +

+

all DC (gated as CD11c , MHCII ; gating strategy presented on Figure 7A). As shown in Figure 7B, the lymph nodes collected from mice immunized with antigen-loaded +

+

+

NP did not present a significant increase on the percentage of CD80 CD11c MHCII cells. However, there was a significant increase on the median fluorescence intensity (MFI; Figure 7C) of those cells obtained from mice treated with EntrapOVA(PVA) and AdsOVA(PVA) NP (p ˂ 0.001 (***)), indicating an increased expression of these co-stimulatory molecules on the cell surface. Upon immunization with EntrapOVA(PVA) and AdsOVA(PVA) NP, DC also expressed significantly higher levels of the co stimulatory molecule CD86 (Figure 7B &C), when compared with those exposed to NP prepared with PF127(p ˂ 0.001 (***)). MHCI was significantly increased after immunization with EntrapOVA(PVA), EntrapOVA(PF127) and AdsOVA(PVA) (p ˂ 0.001(**)) (Figure 7B). Surface MHCII expression was increased only in animals immunized with AdsOVA(PVA), compared to control (p ˂ 0.05 (*)). When

ACCEPTED MANUSCRIPT presented as MFI (Figure 7C) the same trend was observed without reaching a statistical significance. On the other side, vehicles (empty NP) served as a control and inherently did not induce an effect on DC (Figure 7A & D), as lymph nodes of animals immunized with empty NP (without antigen) were

AC C

EP T

ED

MA

NU

SC

RI

PT

similar to the ones obtained from non-immunized mice.

Figure 7: In vivo activation of DC in immunized and non-immunized female C57BL/6 mice 16 h upon immunization. (A) Gating strategies for CD11c +/MHCII+/NP+ or NP-DC. Histograms of expressed activation markers on DC, w ith and w ithout NP, after hock immunization. (B) Percentages and (C) median florescence intensity (MFI) of activated DC 16 h after immunization w ith different formulations of NP. (D) Expression of surface activation markers on DC, w ith and w ithout internalized NP, in the same lymph nodes, 16 h after immunization w ith different NP formulations. * p ˂ 0.05; ** p ˂ 0.01; *** p ˂ 0.001

ACCEPTED MANUSCRIPT

Next, we wanted to distinguish the differences between DC with and without internalized Rhodamine-labeled NP in a single lymph node (Figure 7D). Statistical analysis showed a significantly higher percent of surface expression for CD86, MHCI, MHCII (p ˂ 0.001 (***)) and CD80 (p ˂ 0.01 (**)) by DC with internalized NP. DC expressed significantly higher CD86 and MHCII levels after vaccination with AdsOVA(PVA) NP, than those entrapping AdsOVA(PF127) (p ˂ 0.01 (**)). Moreover, MHCI (p ˂ 0.001 (***)) and MHCII (p ˂ 0.01 (**)) expression, following immunization with EntrapOVA(PVA) NP was higher than the one induced by EntrapOVA(PF127) NP. Furthermore, a

EntrapOVA(PVA) and AdsOVA(PVA) formulations.

In fact,

PT

significant difference in the MHCI and MHCII activation markers (p ˂ 0.05 (*)) was observed for the internalized EntrapOVA(PVA) NP

preferentially led to the up-regulation of MHCI on DC (Figure 7D), driving the antigen processing and

RI

+

presentation pathway towards the activation of a cytotoxic phenotype of CD8 T cells [32-34], which is

SC

essential for the control of tumor and infectious diseases.

3.8 In vivo study of DC cytokine profile upon NP immunization

NU

During DC–T cell interaction, the secretion of cytokines is considered the third signal of the +

synapsis, which activates and imprints naïve CD4 T cells towards a specific T helper (Th) lineage [35]. Therefore, we evaluated the effect of NP immunization on secretion of cy tokines by DC in vivo.

MA

EntrapOVA(PVA) and AdsOVA(PVA) NP, which drove the most efficient up-regulation and activation of DC in vitro (Figure 6) and in vivo (Figure 7), were tested. The adjuvant CpG ODN, a TLR9 agonist, was entrapped within the matrix of NP to co-deliver antigen and adjuvant, therefore mimicking the

ED

complete signal for T cell priming. Specifically, we developed the following NP: EntrapOVA(PVA), EntrapOVA-CpG(PVA), AdsOVA(PVA), and AdsOVA-CpG(PVA). Figure 8 evidences a slight increase on the expression of IL-12 and IL-1β by DC in the lymph nodes of animals treated with NP loaded with

EP T

both the antigen and CpG. The most significant up-regulation of IL-12 was detected for EntrapOVACpG(PVA) (p ˂ 0.001 (***)), while IL-1β presented the highest values for AdsOVA-CpG(PVA) (p ˂ 0.001 (***)). Formulations without incorporated adjuvant, EntrapOVA(PVA) and AdsOVA(PVA), did not

AC C

significantly expressed those cytokines, compared to non-immunized control.

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 8: Flow cytometry analysis of IL-12- and IL-1β-secreting DC in vivo. Immunized and non-immunized (NI) female C57BL/6 mice w ere analysed 16 h upon immunization. (A) Histograms of cytokine profiles after hock immunization w ith (red)

EP T

and w ithout (grey) AdsOVA, AdsOVA-CpG, EntrapOVA or EntrapOVA-CpG NP. (B) Percentages of cytokine-secreting DC (gated: B220-, TCRβ-, CD11c +, MHCII+) 16 h p.i. w ith different NP formulations. Results are representative of tw o independent experiments (n = 4/ group). * p ˂ 0.05; ** p ˂ 0.01; *** p ˂ 0.001

3.9 Characterization of T cell proliferation induced in vivo by NP

CD4

AC C

Efficient adaptive immune responses are characterized by the stimulation of antigen-specific +

+

and CD8 T lymphocytes that proliferate and differentiate into effector T cells. It has been

demonstrated that the NP uptake by DC, antigen processing and presentation are essential parameters for the activation and differentiation of T cells [36]. Previous studies suggest that the expansion of T cells is also dictated by the affinity of the TCR for MHC-peptide ligand at DC [37]. +

We analyzed the proliferation of antigen-specific CD4

+

and CD8

T cells in response to

EntrapOVA(PVA), EntrapOVA-CpG(PVA), AdsOVA(PVA), and AdsOVA-CpG(PVA). Proliferative responses were assessed based on the dilution of the CFSE label of adoptively transferred OVA-specific T cells, in response to the challenge with the different NP formulations +

+

(Suppl. Figure S3). CD4 OT-II and CD8 OT-I T cells extensively proliferated after immunization with EntrapOVA(PVA), as peaks of the divided daughter cells were detected up to eight generations (Figure 9). The immunization with EntrapOVA(PVA) resulted, in fact, in a balanced expansion and +

proliferation of those CD4

+

OT-II and CD8

OT-I T cells. The EntrapOVA-CpG(PVA) formulation

ACCEPTED MANUSCRIPT showed the most profound and fastest antigen-specific response, presenting percentages of 81,4% +

and 85.2%, when CD4

+

OT-II and CD8

OT-I T cells, respectively, were approaching auto-

fluorescence of unlabelled lymphocytes in the recipient animal. The percentage of proliferating CFSElabeled OT-II CD4+ (16.8%) and OT-I CD8+ (14.7%) T cells was indeed lower for EntrapOVACpG(PVA) NP, when compared to those obtained for animals immunized with the other NP

MA

NU

SC

RI

PT

formulations (EntrapOVA(PVA), AdsOVA(PVA), and AdsOVA-CpG(PVA)).

ED

Figure 9: T cell proliferation w as studied using dilution of CFSE fluorescence of OVA -specific TCR-transgenic CD4+ OT-II and CD8+ OT-I T cells in vivo, 3 days post-immunization w ith NP. The histogram plots show the CFSE profiles of viable CD4 + OT-II (A) and CD8+ OT-I T cells (B) after the immunization w ith different NP (dark gray histogram) relative to the generation zero

EP T

CFSE profile of viable transferred CD4+ OT-II and CD8+ OT-I T cells, in the absence of OVA-loaded NP (light gray histogram). Representative histogram plots w ere selected from tree independent experiments (n = 6).

4. Discussion

AC C

Designing safe and efficacious NP vaccines requires a thorough understanding of the physical and chemical features of the carrier and their interaction with biological systems. We focused on the reproducibility and stability of four nanoparticulate delivery systems employing different methods for antigen association (entrapped versus adsorbed) for the systemic delivery of protein antigens to DC, the key APC that trigger T cell immunity. The aim of this study was to develop a colloidal suspension of NP smaller than 200 nm with narrow size distribution, PdI below 0.2 and zeta potential close to neutrality, in order to potentiate NP-APC interaction. Nanosystems were developed by the doubleemulsion solvent evaporation technique, using LALBA as a model protein, which emerged as a relevant antigen with 67% overexpression in primary breast carcinomas and 62% in metastatic cells [38-40]. This work particularly explored the impact of surfactants on NP physical and chemical properties, stability, and NP-DC interaction in vitro and in vivo. The PLGA-based NP are promising tools for the modulation of immune cells’ activity. The properties of this aliphatic polyester polymer can be modified in order to control antigen release

ACCEPTED MANUSCRIPT towards its extensive capture and prolonged presentation, promoting t he expansion of effector T cells [12, 41]. The PEG was used to modify PLGA hydrophobic surface with hydrophilic shielding properties to increase polymer water-solubility, and thus biocompatibility, and greatly reduce immunogenicity and toxicity [42]. CS was used in the internal phase of all formulations, due to its positive charge (+30 mV) [43, 44] to establish electrostatic interactions with the negatively charged groups of biodegradable polymers, thus increasing the zeta potential of NP. This was especially important for PLGA NP (AdsLALBA(PVA) and AdsLALBA(PF127) to augment the electrostatic bonds and therefore affinity of

PT

the adsorbed antigen onto the surface of the polymeric matrix. With an isoelectric point (pI) at 4.2, LALBA protein exhibits a negative charge under physiological pH (7.4) and thus may be adsorbed efficiently

onto

the

positive

surface

charge

of

PLGA-CS

NP.

In

the

formulations

RI

more

EntrapLALBA(PVA) and EntrapLALBA(PF127), CS not only contributed to the increase of particle

SC

surface charge, promoting the electrostatic interactions with the negatively charged protein, but also raised the viscosity of the internal phase [45, 46]. Therefore, both physical and chemical stabilities of the internal phase were increased with CS, which can explain the higher NP loading capacities .

NU

The stability of the NP was followed for 3 months in PBS at 4°C. This study did not show significant changes and deviation in size or PdI of NP, over a period of 2 months. Such robust stability of EntrapLALBA(PVA) and EntrapLALBA(PF127) NP could be explained by steric hindrance and

MA

repulsion of hydrophilic PEG chains, also known as a stealth effect [26, 27]. However, following 2 months a drastic increase in size and PdI was detected, while zeta potential was decreased. These data demonstrate the PLGA-PEG bulk erosion, swelling and degradation of NP, which is in

ED

accordance with the physico-chemical properties of these aliphatic polyesters [28]. In case of AdsLALBA(PVA) and AdsLALBA(PF127) NP, PLGA polymer was used and therefore these NP lack the PEG stealth effect [15]. However, adsorbed protein on the hydrophobic polymeric matrix may have

EP T

formed a protein corona that probably prevented NP aggregation, thus contributing to their stabilization via steric repulsion. Given the fact that the adsorption of antigens onto NP is based simply on charge and hydrophobic interactions, the interaction between NP and antigen is relatively weak. A

AC C

size increase was obviously due to the long incubation time in PBS that resulted in a higher electrostatic repulsion and therefore influenced the protein affinity and redistribution. However, clustering and aggregation of AdsLALBA(PVA) and AdsLALBA(PF127) NP detected at day 14 can however perfectly fit the period relevant for their application as vaccine delivery systems. In fact, the final product most probably would become available as an extemporaneous vaccine, containing the NP in the lyophilized form and an adequate buffer for NP dispersion previously to administration. FTIR, DCS and SDS-PAGE were used to verify the stability of polymers and proteins in the NP formulation. These techniques are complementary and evidenced that the integrity of LALBA protein and all the polymers were maintained after NP formulation. Thus, those antigen-loaded NP were adequate to pursue further in vitro and in vivo biological studies. A wide range of concentrations was screened to assess the effect of NP with different chemical compositions on the viability of BMDC. No morphological differences or reduction in cell viability were observed on the targeted cells as a function of NP concentration. This corroborates the

ACCEPTED MANUSCRIPT polymer biocompatibility and the safety of the residual amount of surfactants and organic solvent present in the final NP-based products [47]. There is a limited understanding of how the physico-chemical properties of NP affect their interactions with the biological system. In addition, a number of challenges still remain also due to the lack of in vitro and in vivo models available for a reliable characterization of antigen-specific immune responses particularly induced by an immune-adjuvant candidate. Even though, presently in the area of immunotherapy, the OVA-specific TCR transgenic mouse model [13, 14] constitutes a robust tool to characterize the intrinsic adjuvant activity of novel vaccine formulations, entrapping the highly

PT

immunogenic OVA as a model antigen. Therefore, OVA was loaded into the four types of NP developed in this study, to ensure reliable in vitro and in vivo results related to their ability to induce

RI

the activation and maturation of DC in vivo, with subsequent T cell proliferation [48, 49]. All OVAloaded NP formulations presented a uniform size distribution, below 200 nm, and zeta potential close

SC

to neutrality. There was no significant difference between the chemical and physical properties of formulated NP, when LALBA was replaced with OVA. Despite their difference in MW, both proteins possess similar chemical characteristics, such as pI, solubility (10 mg/mL) and expected amphiphilic

NU

properties [50].

The influence of surface chemistry and charge of NP on BMDC uptake was then investigated in living cells by image flow cytometry. Overall, we would expect a faster internalization for antigen-

MA

adsorbed NP, due to its similarity to the pathogens regarding the presence of antigens on the surface [12]. However, the slightly negative charge of protein covering the NP surface may have influenced the endocytosis of NP [51]. Our in vitro studies demonstrated that the interaction between the BMDC

ED

with NP containing the adsorbed antigen (AdsOVA(PVA) and AdsOVA(PF127)) was slower, especially for the NP formulation AdsOVA(PVA). After 3 h of incubation, only 55% of CD11c

+

cells were

interacting with the AdsOVA(PVA), while this amount almost doubled after 24 h (~90%) of incubation.

EP T

The uptake of AdsOVA(PF127) NP did not significantly change with the incubation time. These data suggested that NP with antigen adsorbed on the surface, do not necessarily emerge with higher interaction and phagocytosis by APC. On the other hand, NP with entrapped antigen were coated with

AC C

the hydrophilic PEG, but exhibited a higher phagocytic uptake by BMDC after 24h incubation. This could be explained by the slightly positive surface charge of those NP containing the antigen dispersed within the polymeric matrix EntrapOVA(PVA) and EntrapOVA(PF127), which may have facilitated the interaction with the negatively charged cell membrane. In addition, the chemical composition of NP influenced their internalization by BMDC. All of the formulations

developed using the PF127 emulsifier in the

external phase presented lower

internalization values, which can be explained not only by the influence of a more negative surface charge of those NP, but also by the steric repulsion effect of PF127 triblock copolymer (PEO-b-PPO-bPEO). PPO polymer present the hydrophobic chain that is normally oriented to the matrix of the NP, while PEO, the two edges of the polymer, is hydrophilic and thus similar to PEG, being thus usually oriented away from the core of NP [52, 53]. This phenomenon was not observed in NP formulations EntrapOVA(PVA) and AdsOVA(PVA), when PVA was used in the external phase. The interaction and internalization of these NP increased by extending the incubation time. However, despite their different

ACCEPTED MANUSCRIPT surface properties and their divergent tendency to interact with BMDC, all of the formulated NP were efficiently internalized by these APC. Physiological activation of DC was examined upon a single immunization with NP injection. DC with internalized NP expressed significantly higher levels of the co-stimulatory molecule CD86, especially after the immunization with the EntrapOVA(PVA) and AdsOVA(PVA) formulations. Furthermore, there was no significant difference between the NP with entrapped OVA and those containing the adsorbed antigen. The formulation of AdsOVA(PVA) NP using PVA in the external phase induced higher expression of both the co-stimulatory molecule CD86 and the activation marker

PT

MHCII, than AdsOVA(PF127) NP. On the other side, internalized EntrapOVA(PVA) NP preferentially led to the up-regulation of MHCI on the DC, driving the antigen processing and presentation pathway +

RI

towards the activation of a cytotoxic phenotype of CD8 T cells, which is essential for the control of tumor and infectious diseases. However, even if a certain NP formulation demonstrates to drive a

SC

preferential antigen-presentation pathway, it does not mean the exclusion of the other alternative pathways for antigen processing. Among all of the developed types of NP, those without PF127 in their composition were far more efficient, thus showing the highest activation levels of co-stimulatory

NU

molecules and activation markers. This could be explained by the constant levels of those DC surface markers obtained for the interaction of both NP EntrapOVA(PF127) and AdsOVA(PF127) with BMDC across different time points. This low up-regulation of DC activation markers may have resulted from a

MA

slow rate of internalization and degradation of these NP in BMDC, and therefore from a subsequent lower antigen release.

Our in vitro study and the in vivo experiment on DC activation were conclusive in showing that

ED

minor differences in the chemical composition and protein affinity (entrapped or adsorbed) in the developed NP had an extensive impact on their uptake and activation of BMDC. EntrapOVA(PVA) NP and AdsOVA(PVA) NP, with and without the immune potentiator CpG,

EP T

were selected to further study their in vivo effect on the secretion of cytokines by DC, as these NP showed the best uptake profiles and the most efficient up-regulation of the tested DC activation markers. Inclusion of CpG in the nanoparticulate systems led to the up-regulation of both IL-12p40

AC C

and IL-1β, particularly IL-12 by the EntrapOVA-CpG(PVA) NP. IL-12 is one of the major players of the DC-T cell cross-talk, driving the polarization of T cell towards a Th1 phenotype. Our data support the evidence that EntrapOVA-CpG(PVA) NP enhanced both the uptake of antigen and the immune-potentiating effect of adjuvants on DC and augmented vigorously the nascent +

adaptive immune response. Therefore, the incorporation of CpG adjuvant boosted both CD4 OT-II +

and CD8 OT-I T cells proliferation and expansion.

5. Conclusions In this study, our major goal was to design, synthesize and optimize a simple, highly reproducible and efficient well-characterized delivery system to target DC. Developing safe and efficacious NP-based vaccines requires an extensive understanding on NP composition and physicochemical properties. The combination of all of these parameters determines the fate and overall

ACCEPTED MANUSCRIPT efficiency of NP on promoting antigen recognition, capture and processing by DC, thus having a critical role in the differentiation and expansion of T cell subsets. The developed NP presented the desirable physicochemical properties, with good loading capacity and colloidal stability in biological media. All NP showed prompt and efficient DC activation +

+

and maturation, followed by CD4 T OT-II and CD8 OT-I cell proliferation and activation through the antigen-specific immune response, especially when antigen and adjuvant were combined within NP matrix. These are pivotal steps in the T cell immunity that leads to a complete and effective immune response. the developed NP-based system constitutes a promising platform for the

PT

Therefore,

development of prophylactic and therapeutic cancer vaccines, as well as for the induction of specific

SC

RI

immunity against infectious and viral diseases.

Acknowledgements

This research work was funded by Fundação para a Ciência e a Technologia (FCT), Ministério da

NU

Educação e Ciência, Portugal for the PhD Grant SFRH/BD/78480/2011 and Research projects UTAP ICDT/DTP-FTO/0016/2014 and ENMed/0003/2015; iMed.ULisboa grant UID/DTP/04138/2013. The Jung laboratory is supported by the European Research Council (340345). The authors are grateful to

MA

the EMBO for the Short Term Fellowship (ASTF 277-2015). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest or financial conflict with the subject matter or materials discussed in the manuscript.

ED

The authors would like to thank Sigalit Boura-Halfon, PhD, and Ana Salgado for technical assistance.

Declaration of interest

writing of this article.

EP T

The authors report no conflicts of interest. The authors alone are responsible for the content and

AC C

Ethical conduct of research

Animals were maintained and treated according to the Weizmann Institute of Science animal facility and National Institutes of Health guidelines, Israel. Procedures with animal subjects have been approved by Institutional Animal Care and Use Committee (IACUC) at the Weizmann Institute of Science, Israel.

ACCEPTED MANUSCRIPT References [1] J. Banchereau, R. Steinman, Dendritic cells and the control of immunity, Nature 19(1998) 245-52. [2] C.J. How ard, B. Charleston, S.A. Stephens, P. Sopp, J.C. Hope, The role of dendritic cells in shaping the immune response, Anim. Health Res. Rev. 5(2004) 191-5. [3] I. Mellman, Dendritic Cells: Master Regulators of the Immune Response, Cancer Immunol. Res. 1(2013) 145-9. [4] S.S. Saluja, D.J. Hanlon, F.A. Sharp, E. Hong, D. Khalil, E. Robinson, et al. Targeting human dendritic cells via DEC-205 using PLGA nanoparticles leads to enhanced cross-presentation of a melanoma-associated antigen, Int. J. Nanomedicine 9(2014) 5231-46. [5] J. M. Silva, M. Videira, R. Gaspar, V. Preat, H.F. Florindo, Immune system targeting by biodegradable nanoparticles for cancer vaccines, J. Control. Release 168(2013) 179-99.

PT

[6] K. Palucka, J. Banchereau, Cancer immunotherapy via dendritic cells, Nat. Rev. Cancer 12(2012) 265-77. [7] Y. Zhou, Y. Zhang, Y. Zhiqiang, J.P. Moorman, J. Zhansheng, Dendritic cell-based immunity and vaccination against hepatitis C virus infection, Immunology 136(2012) 385–96.

RI

[8] B.B. Finlay, G. McFadden, Anti-Immunology: Evasion of the Host Immune System by Bacterial and Viral Pathogens, Cell 124(2006) 767–82.

SC

[9] H. Kumar, T. Kaw ai, S. Akira, Pathogen recognition in the innate immune response, Biochem. J. 420(2009) 1-16. [10] J.M. Silva, E. Zupancic, C. Perez, L.C. Silva, R. Gaspar, V. Preat, H.F. Florindo, Immunomodulation against microbal pathogens through dendritic cells, in: A. Mendez-Vilas (Fourth Ed.), Microbal pathogens and strategies for combating them: science, technology and education, Formatex Research Center, Spain, 2013, pp.1686-1705.

NU

[11] S. Prasad, V. Cody, J. K. Saucier-Saw yer, W.M. Saltzman, C.T. Sasaki, R.L. Edelson, et al. Polymer nanoparticles containing tumor lysates as antigen delivery vehicles for dendritic cell-based antitumor immunotherapy, Nanomedicine 7(2011) 1-10. immunotherapy, Immune Netw . 13(2013) 177-83.

MA

[12] Y.M. Park, S.J. Lee, Y.S. Kim, M.H. Lee, G.S. Cha, I.D. Jung, et al. Nanoparticle-based vaccine delivery for cancer [13] K.A. Hogquist, S.C. Jameson, W.R. Heath, J.L. How ard, M.J. Bevan, F.R. Carbone, T cell receptor antagonist peptides induce positive selection, Cell 76(1994) 17–27. J. Immunol. 164(2000) 4706-12.

ED

[14] J.M. Robertson, P.E. Jensen, B.D. Evavold, DO11.10 and OT-II T cells recognize a C-terminal ovalbumin 323-339 epitope, [15] M. Garinot, V. Fiévez, V. Pourcelle, F. Stoffelbach, A. des Rieux, L. Plapied, et al. PEGylated PLGA -based nanoparticles targeting M cells for oral vaccination, J. Control. Release 120(2007) 195-204.

EP T

[16] J.M. Silva, G. Vandermeulen, V.G. Oliveira, S.N. Pinto, C. Rodrigues, A. Salgado, et al. Development of functionalized nanoparticles for vaccine delivery to dendritic cells: a mechanistic approach, Nanomedicine 9(2014) 2639-56. [17] F. Florindo, S. Pandit, L. Lacerda, L.M. Goncalves, H.O. Alpar, A.J. Almeida, The enhancement of the immune response against S. equi antigens through the intranasal administration of poly -epsilon-caprolactone-based nanoparticles, Biomaterials

AC C

30(2009) 879-91.

[18] T.C. George, S.L. Fanning, P. Fitzgerald-Bocarsly, R.B. Medeiros, S. Highfill, Y. Shimizu, et al. Quantitative measurement of nuclear translocation events using similarity analysis of multispectral cellular images obtained in flow , J. Immunol. Methods 311(2006) 117-29.

[19] B.J.C. Quah, H.S. Warren, C.R. Parish, Monitoring lymphocyte proliferation in vitro and in vivo w ith the intracellular fluorescent dye carboxyfluorescein diacetate succinimidyl ester, Nat. Protoc. 2. 2(2007) 2049-56. [20] B. Mukherjee, K. Santra, G. Pattnaik, S. Ghosh, Preparation, characterization and in-vitro evaluation of sustained release protein-loaded nanoparticles based on biodegradable polymers, Int. J. Nanomedicine 3(2008) 487-96. [21] S. Alimohammadi, R. Salehi, N. Amini, S. Davaran, Synthesis and Physicochemical Characterization of Biodegradable PLGA-based Magnetic Nanoparticles Containing Amoxicilin, Bull. Korean Chem. Soc. 33(2012) 3225-32. [22] I. Amjadi, M. Rabiee, M.S. Hosseini, Anticancer activity of NPs based on PLGA and co-polymer in vitro evaluation, Iran. J. Pharm. Res. 12(2013) 623-34. [23] Z. Kulinski, E. Piorkow ska, K. Gadzinow ska, M. Stasiak, Plasticization of poly(l-lactide) w ith poly(propylene glycol), Biomacromolecules 7(2006) 2128-35. [24] M.S. Atri, A.A. Saboury, A.A. Moosavi-Movahedi, K. Kavousi, S. Ariaeenejad, Effects of zinc binding on the structure and thermal stability of camel alpha-lactalbumin, J. Therm. Anal. Calorim. 120(2015) 481-8.

ACCEPTED MANUSCRIPT [25] T. Hendrix, Y. Griko, P. Privalov, A calorimetric study of the influence of calcium on the stability of bovine a-lactalbumin, Biophys. Chem. 84(2000) 27-34. [26] N.J. Butcher, G.M. Mortimer, R.F. Minchin, Drug delivery: Unravelling the stealth effect, Nat. Nanotechnol. 11(2016) 310-1. [27] Z. Amoozgar, Y. Yeo, Recent advances in stealth coating of nanoparticle drug delivery systems, Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 4(2012) 219-33. [28] F. Mollica, M. Biondi, S. Muzzi, F. Ungaro, F. Quaglia, M.I. La Rotonda, et al. Mathematical modelling of the evolution of protein distribution w ithin single PLGA microspheres: prediction of local concentration profiles and release kinetics, J. Mat er. Sci. Mater. Med. 19(2008) 1587-93. [29] H.K. Makadia, S.J. Siegel, Poly Lactic-co-Glycolic Acid (PLGA) as Biodegradable Controlled Drug Delivery Carrier, Polymers 3(2011) 1377-97.

PT

[30] M. Tobio, R. Gref, A. Sanchez, R. Langer, M.J. Alonso, Stealth PLA-PEG nanoparticles as protein carriers for nasal administration, Pharm. Res. 15(1998) 270-5.

[31] L. Rodriguez-Lorenzo, K. Fytianos, F. Blank, C. von Garnier, B. Rothen-Rutishauser, A. Petri-Fink, Fluorescence-Encoded

RI

Gold Nanoparticles: Library Design and Modulation of Cellular Uptake into Dendritic Cells, Small, 10(2014) 1341–50. [32] F. Danhier, E. Ansorena, J.M. Silva, R. Coco, A. Le Breton, V. Preat, PLGA -based nanoparticles: an overview of biomedical applications, J. Control. Release 161(2012) 505-22.

SC

[33] R. Arens, S.P. Schoenberger, Plasticity in programming of effector and memory CD8+ T-cell formation, Immunol. Rev. 235(2012) 190–205.

[34] H. Shen, A.L. Ackerman, V. Cody, A. Giodini, E.R. Hinson, P. Cressw ell, et al. Enhanced and prolonged cross -presentation

NU

follow ing endosomal escape of exogenous antigens encapsulated in biodegradable nanoparticles, Immunology 117(2005) 78– 88.

[35] O'Shea JJ, Paul WE. Mechanisms underlying lineage commitment and plasticity of helper CD4+ T cells. Science.

MA

2010;327:1098-102.

[36] A.L. Silva, R.A. Rosalia, E. Varypataki, S. Sibuea, F. Ossendorp, W. Jiskoot, Poly-(lactic-co-glycolic-acid)-based particulate vaccines: Particle uptake by dendritic cells is a key parameter for immune activation, Vaccine 33(2015) 847–54. [37] D. Zehn, S.Y. Lee, M.J. Bevan, Complete but curtailed T-cell response to very low -affinity antigen, Nature 458(2009) 211-4. [38] A.K. Lee, P.P. Rosen, R.A. DeLellis, P.E. Saigo, M.D. Gangi, S. Groshen, et al. Tumor marker expression in breast

ED

carcinomas and relationship to prognosis. An immunohistochemical study, Am. J. Clin. Pathol. 84(1985) 687–96. [39] R. Jaini, P. Kesaraju, J.M. Johnson, C.Z. Altuntas, D. Jane-Wit, V.K. Tuohy, An autoimmune-mediated strategy for prophylactic breast cancer vaccination, Nat.Med. 16(2010) 799-803.

EP T

[40] C. Cohen, F. Sharkey, G. Shulman, E. Uthman, L. Budgeon, Tumor-associated antigens in breast carcinomas. Prognostic significance, Cancer 60(1987) 1294-8.

[41] Y. Krishnamachari, S. Geary, C. Lemke, A. Salem, Nanoparticle Delivery Systems in Cancer Vaccines, Pharm. Res. 28(2011) 215-36.

AC C

[42] J. Jeong, D. Lim, D. Han, T. Park, Synthesis, characterization and protein absorption behaviours of PLGA -PEG di-block copolymer blend film, Colloids Surf., B 8(2000) 371-9. [43] P. Pereira, S.S. Pedrosa, A. Correia, C.F. Lima, M.P. Olmedo, A. González-Fernández, et al. Biocompatibility of a self assembled glycol chitosan nanogel, Toxicol. in Vitro 29(2015) 638-46. [44] P. Pereira, D. Morgado, A. Crepet, L. David, F.M. Gama, Glycol chitosan-based nanogel as a potential targetable carrier for siRNA, Macromol. Biosci. 13(2013) 1369-78. [45] H. Hosseinzadeh, F. Atyabi, R. Dinarvand, S.N. Ostad, Chitosan-Pluronic nanoparticles as oral delivery of anticancer gemcitabine: preparation and in vitro study, Int. J. Nanomedicine 7(2012) 1851-63. [46] M.H. Park, J.S. Baek, C.A. Lee, C.W. Cho, Effect of chitosan on physicochemical properties of exenatide-loaded PLGA nanoparticles, J. Pharm. Investig. 43(2013) 489-97. [47] K.A. Athanasiou, G.G. Niederauer, C.M. Agrawal, Sterilization, toxicity, biocompatibility and clinical applications of polylactic acid / polyglycolic acid copolymers, Biomaterials 17(1996) 93-102. [48] V. Schijns, D. O'Hagan, Immunopotentiators in Modern Vaccines, First Ed., Elsevier Academic Press, London, 2006. [49] Q.A. Acton, Advances in Active Immunotherapy Research and Application, ScholarlyEditions, Atlanta, Georgia, 2011. [50] R.M. Kramer, V.R. Shende, N. Motl, C.N. Pace, J.M. Scholtz, Tow ard a Molecular Understanding of Protein Solubility: Increased Negative Surface Charge Correlates w ith Increased Solubility, Biophys. J. 102(2012) 1907-15. [51] N. Oh, J.H. Park, Endocytosis and exocytosis of nanoparticles in mammalian cells, Int. J. Nanomedicine 9(2014) 51-63.

ACCEPTED MANUSCRIPT [52] S. Miyazaki, T. Tobiyama, M. Takada, D. Attw ood, Percutaneous Absorption of Indomethacin from Pluronic F127 Gels in Rats, J. Pharm. Pharmacol. 47(1995) 455-7. [53] J.J. Escobar-Chávez, M. López-Cervantes, A. Naïk, Y.N. Kalia, D. Quintanar-Guerrero, A. Ganem-Quintanar, Applications

AC C

EP T

ED

MA

NU

SC

RI

PT

of Thermo-reversible Pluronic F-127 Gels in Pharmaceutical Formulations, J. Pharm. Pharmaceut. Sci. 9(2006) 339-58.

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT Rational design of nanoparticles towards targeting antigen-presenting cells and improved T cell priming Eva Zupančič

a,b,c

c

a

a

d

e

, Caterina Curato , Maria Paisana , Catarina Rodrigues , Ziv Porat , Ana S Viana , a

a

a

b,f

g

Carlos A M Afonso , João Pinto , Rogério Gaspar , João N Moreira , Ronit Satchi-Fainaro , Steffen c

Jung , Helena F Florindo

a*

PT

Figure captions Figure 1: Size and morphology characterization of formulated NP by AFM. Section analysis of NP EntrapLALBA(PVA) (A) and EntrapLALBA(PF127) (B) w ith the matrix polymer PLGA -PEG and w ith entrapped antigen; AdsLALBA(PVA) (C) and

RI

AdsLALBA(PF127) (D) w ith PLGA polymer and adsorbed LALBA protein.

SC

Figure 2: FTIR and DSC studies on lyophilized NP. (A) FTIR spectrographs of NP and the components used for their formulations. The symbols on the top of the graph correspond to the specific band. Peak at 1720 cm−1 (§) is a specific band for C=O bond; 1662 cm-1 (*) is specific for primary amides; peak at 1562 cm-1 (Ϫ) corresponds to vibration of -NH2 groups; 2950

NU

cm−1 (Ф) indicates the presence of C-H bond; 3400 cm-1 (‡) correlates to the hydroxyl and amine groups overlapping. (B) DSC specters of formulated NP and the components used for their formulations .

MA

Figure 3: SDS–PAGE (12 % separating gel) of protein LALBA before and after entrapment in NP. Lanes: (1) standard molecular w eight (MW) markers, w ide range 10 – 180 kDa; (2) LALBA standard solution at 2 μg; (3) 4 μg; (4) 8 μg; (5) 16 μg all LALBA’s bands may be seen at the 14 kDa; (6) Entrap blank(PVA); (7) EntrapLALBA(PVA); (8) Entrap blank(PF127); (9) EntrapLALBA(PF127); (10) Ads blank(PVA); (11) AdsLALBA(PVA); (12) Adsblank(PF127). (13) AdsLALBA(PF127); A representative

ED

image of 3 independent experiments is presented.

Figure 4: Variation of the average size (A and B), polydispersity index (PdI) (C), and zeta potential (D) of polymeric NP

EP T

measured over a 3 month period w hen dispersed in PBS pH 7.4 and stored at 4˚C. Circles (●) present formulation EntrapLALBA(PVA), squares (■) EntrapLALBA(PF127), triangles (▲) AdsLALBA(PVA) and diamonds (▼) AdsLALBA(PF127) F4.All results are expressed as mean ± SD (n > 3).

Figure 5: Effect of NP on the viability of BMDC w as studied in vitro using the AlamarBlue® assay. We incubated bone marrow

AC C

derived DC w ith increasing concentrations (100–1000 μg/mL) of NP for 24 h. Culture medium (0) and triton 0.5% (w /v) w ere used as negative and positive controls, respectively (Mean ± SD; n = 6).

Figure 6: In vitro NP internalization by BMDC. ImageStream analysis w as performed to detect PLGA -Rhodamine labeled NP in BMDC, 3, 6, 16 and 24 h after the incubation in vitro. Distribution of at least 30,000 cells w as calculated by Amnis IDEAS softw are and presented w ith the histograms show ing the percentages of CD11c + cells interacting w ith NP (A) and of CD11c + cells w ith internalized NP (B). Examples of representative images captured by the Amnis ImageStreamX Flow Cytometer of DC incubated for 16 h at 37°C w ith AdsOVA(PVA) (C). First column show s brightfield (BF) images of the cells, second column show s images of PLGA-Rhodamine labeled NP (NP), third column show s APC-labeled CD11c + DC (CD11c), fourth column show s merged images of the precedent fields (BF/NP/CD11c) and fifth column show s DAPI signal for dead cells. (C.a) CD11c + w ithout NP. (C.b) NP onto the surface of the CD11c + cell. (C.c) Various NP entering the cytoplasm of CD11c + cell. (C.d) DC w ith various NP. (C.e) DC w ith star morphology, indicator of the DC maturation process. Results are mean ± SD, normalized to that of control non-labeled cells, representative of three independent experiments (n = 3). Scale bar, 7 μm.

ACCEPTED MANUSCRIPT Figure 7: In vivo activation of DC in immunized and non-immunized female C57BL/6 mice 16 h upon immunization. (A) Gating strategies for CD11c +/MHCII+/NP+ or NP-DC. Histograms of expressed activation markers on DC, w ith and w ithout NP, after hock immunization. (B) Percentages and (C) median florescence intensity (MFI) of activated DC 16 h after immunization w ith different formulations of NP. (D) Expression of surface activation markers on DC, w ith and w ithout internalized NP, in the same lymph nodes, 16 h after immunization w ith different NP formulations. * p ˂ 0.05; ** p ˂ 0.01; *** p ˂ 0.001 Figure 8: Flow cytometry analysis of IL-12- and IL-1β-secreting DC in vivo. Immunized and non-immunized (NI) female C57BL/6 mice w ere analysed 16 h upon immunization. (A) Histograms of cytokine profiles after hock immunization w ith (red) and w ithout (grey) AdsOVA, AdsOVA -CpG, EntrapOVA or EntrapOVA-CpG NP. (B) Percentages of cytokine-secreting DC (gated: B220-, TCRβ-, CD11c +, MHCII+) 16 h p.i. w ith different NP formulations. Results are representative of tw o independent

PT

experiments (n = 4/ group). * p ˂ 0.05; ** p ˂ 0.01; *** p ˂ 0.001 Figure 9: T cell proliferation w as studied using dilution of CFSE fluorescence of OVA -specific TCR-transgenic CD4+ OT-II and CD8+ OT-I T cells in vivo, 3 days post-immunization w ith NP. The histogram plots show the CFSE profiles of viable CD4 + OT-II

RI

(A) and CD8+ OT-I T cells (B) after the immunization w ith different NP (dark gray histogram) relative to the generation zero CFSE profile of viable transferred CD4+ OT-II and CD8+ OT-I T cells, in the absence of OVA-loaded NP (light gray histogram).

AC C

EP T

ED

MA

NU

SC

Representative histogram plots w ere selected from tree independent experiments (n = 6).

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

NU

SC

RI

PT

Graphical abstract