Lap

NDa

b

c

a

ARRAA

KPANP

1

(esoefgtotc

0h

Colloids and Surfaces B: Biointerfaces 103 (2013) 298– 303

Contents lists available at SciVerse ScienceDirect

Colloids and Surfaces B: Biointerfaces

jou rna l h om epa g e: www.elsev ier .com/ locate /co lsur fb

ysine-tagged peptide coupling onto polylactide nanoparticles coated withctivated ester-based amphiphilic copolymer: A route to highlyeptide-functionalized biodegradable carriers

adège Handkéa, Damien Ficheuxb, Marion Rolleta, Thierry Delair c, Kamel Mabrouka,enis Bertina, Didier Gigmesa, Bernard Verrierb, Thomas Trimaillea,∗

Aix-Marseille Université, CNRS, UMR 7273, Institut de Chimie Radicalaire, Avenue Escadrille Normandie-Niemen, 13397 Marseille Cedex 20, FranceUniversité Lyon 1, Univ Lyon, CNRS, FRE 3310, Dysfonctionnement de l’Homéostasie Tissulaire et Ingénierie Thérapeutique, IBCP, 7 passage du Vercors, 69367 Lyon Cedex 07, FranceUniversité Lyon 1, Univ Lyon, CNRS, UMR 5223, Ingénierie des Matériaux Polymères, 15 Boulevard Latarjet, 69622 Villeurbanne, France

r t i c l e i n f o

rticle history:eceived 1 May 2012eceived in revised form 19 October 2012ccepted 22 October 2012vailable online xxx

eywords:olylactide nanoparticlesmphiphilic block copolymer-succinimidyl estereptide conjugation

a b s t r a c t

Efficient biomolecule conjugation to the surface of biodegradable colloidal carriers is crucial for their tar-geting efficiency in drug/vaccine delivery applications. We here propose a potent strategy to drasticallyimprove peptide immobilization on biodegradable polylactide (PLA) nanoparticles (NPs). Our approachparticularly relies on the use of an amphiphilic block copolymer PLA-b-poly(N-acryloxysuccinimide-co-N-vinylpyrrolidone) (PLA-b-P(NAS-co-NVP)) as NP surface modifier, whose the N-succinimidyl (NS)ester functions of the NAS units along the polymer chain ensure N-terminal amine peptide coupling.The well-known immunostimulatory peptide sequence derived from the human interleukin 1� (IL-1�),VQGEESNDK, was coupled on the NPs of 169 nm mean diameter in phosphate buffer (pH 8, 10 mM). A max-imum amount of 2 mg immobilized per gram of NPs (i.e. 0.042 peptide nm−2) was obtained. Introductionof a three lysine tag at the peptide N-terminus (KKKVQGEESNDK) resulted in a dramatic improvementof the immobilized peptide amounts (27.5 mg/g NP, i.e. 0.417 peptide nm−2). As a comparison, the den-

sity of tagged peptide achievable on surfactant free PLA NPs of similar size (140 nm), through classicalEDC or EDC/NHS activation of the surface PLA carboxylic end-groups, was found to be 6 mg/g NP (i.e.0.075 peptide nm−2), showing the decisive impact of the P(NAS-co-NVP)-based hairy corona for highpeptide coupling. These results demonstrate that combined use of lysine tag and PLA-b-P(NAS-co-NVP)surfactant represents a valuable platform to tune and optimize surface bio-functionalization of PLA-basedbiodegradable carriers.

. Introduction

Biodegradable polylactide (PLA) and poly(lactide-co-glycolide)PLGA) based nanoparticles (NPs) are attractive candidate carri-rs for drug [1,2] and vaccine [3–5] delivery. In this field, efficienturface conjugation of selective biomolecules, particularly peptidicnes [6–8], is of crucial importance for the capacity of the carrier tofficiently target cells of interest. However, the ability of PLA NPsor surface modification is rather poor due to the lack of functionalroups along the PLA backbone. By far, such conjugation has beenypically achieved by two kinds of strategies: (i) the post-coupling

f the peptides on the prepared NPs, which requires activation ofhe PLA carboxylic end-groups of the NP surface through classi-al EDC/NHS activation, and generally leads to low biomolecule

∗ Corresponding author.E-mail address: thomas.trimaille@univ-amu.fr (T. Trimaille).

927-7765/$ – see front matter © 2012 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.colsurfb.2012.10.032

© 2012 Elsevier B.V. All rights reserved.

densities [9–11]; (ii) chain-end modification of polyethylene glycol(PEG)-based compounds (PEG-b-PLA, PEG-b-PCL, Pluronics) used assurfactant [6,12–15]. Regarding this strategy, several teams haveshowed that (peptidic) ligand density has a significant impact onNP cell uptake, by varying the ratio of peptide modified PEG-based surfactant to non-functionalized one in the NP preparationprocess [6,14]. However maximal ligand density achievable onthe NPs is still limited due to the non-functionalizable polyethernature of PEG backbone (i.e. 1–2 ligands max. per copolymerchain). Moreover, the chemistry for conjugation of the ligandon the PEG chain end can be laborious, often requiring severalsteps [12]. There is thus a need for versatile alternative strate-gies affording improved surface bio-functionalization of PLA NPs.In a recent work [16], we described the synthesis of a novel

amphiphilic block copolymer PLA-b-poly(N-acryloxysuccinimide-co-N-vinylpyrrolidone) (PLA-b-P(NAS-co-NVP)), whose NAS andNVP units have a strong alternating tendency. Its use as a surfacemodifier in the PLA nanoprecipitation process (Fig. 1) leads to NPs

N. Handké et al. / Colloids and Surfaces B:

Fig. 1. Strategy of preparation of the PLA-b-P(NAS-co-NVP) (NS-cop) coatedn

pe(upicNetiwTtPbpo

2

2

pAwF15NatK1(b(Hf

(27.5 mg/g NP, 10 mL, 5 mg mL−1 NP) was washed from NHS and

anoparticles and further peptide functionalized NPs.

resenting at their surface a high density of N-succinimidyl (NS)ster moieties potentially available for coupling of biomolecules2.4 functions nm−2, omitting the hydrolyzed fraction) while NVPnits offer a convenient alternative to PEG [17]. In a vaccine deliveryerspective, we further interested to introduce at NP surface the

mmunostimulatory sequence VQGEESNDK of the IL-1� [18] (so-alled iIL-1�). Coupling of such sequence on the negatively chargedPs [16] was expected to be difficult through its �-amine (consid-ring the presence of three carboxylate lateral functions from thewo E and D residues, and a C-terminal lysine). We thus stud-ed, as a comparison, a peptide analog in which a three lysine tag

as introduced at the peptide N-terminus (i.e. KKKVQGEESNDK).o our knowledge, the lysine tag strategy [19] for favoring pep-ide coupling has never been applied to polymeric NPs, particularlyLA/PLGA-based carrier systems. We show hereafter that the com-ined use of lysine tag and P(NAS-co-NVP) block represents aromising platform to drastically enhance peptide immobilizationn PLA NPs.

. Materials and methods

.1. Materials

N-Hydroxysuccinimide (NHS, 98%) and N-(3-dimethylamino-ropyl)-N′-ethylcarbodiimide (EDC, >97%) were purchased fromldrich. Poly(d,l-lactide) (PLA50, Mn = 30,000 g mol−1, PDI = 1.7)ith a carboxylic end group was purchased from Phusis (Grenoble,

rance). PLA-b-P(NAS-co-NVP) copolymer (referred as “NS-cop”,9,000-b-13,000 g mol−1, PDI = 1.5, NAS and NVP molar ratios of3% and 47%) and PLA nanoparticles coated with NS-cop (NS-copPs) or surfactant-free (NP0) were prepared and characterizeds previously described [16]. VQGEESNDK peptide sequence ofhe IL-1� (referred as iIL-1� and “non-tagged” peptide) andKKVQGEESNDK lysine tagged peptide analog (referred as KKK-iIL-� and “tagged” peptide) were synthesized by solid phase methodFmoc amide resin) using Fmoc/tBu chemistry, and characterizedy mass spectrometry ([M+H]+ = 1004.6 for non-tagged peptideM = 1003.5); [M+H]+ = 1388.8 for tagged peptide (M = 1387.7)) and

PLC. Peptide powder used for experiments was under TFA salt

orm (Mnon-tagged peptide = 1232; Mtagged peptide = 1959).

Biointerfaces 103 (2013) 298– 303 299

2.2. Peptide coupling on the particles

Typically, 200 �L of peptide solution at a given concentration(ranging from 0.12 to 0.42 mg mL−1) in phosphate buffer pH 8,20 mM were briefly added to 200 �L of NS-cop NPs dispersion at10 mg mL−1 in milli-Q water, and the coupling medium stirredfor 2 or 24 h (final coupling medium: NPs: 5 mg mL−1, peptide:0.06–0.21 mg mL−1, phosphate buffer, 10 mM, pH 8). Then, thedispersion was centrifuged at 15,000 × g for 10 min and the super-natant analyzed by HPLC for quantifying non-coupled peptide.Peptide coupling on naked PLA NPs (NP0) was performed followingthe same procedure, except that the carboxylic groups at the sur-face of the PLA NPs dispersion (200 �L, 10 mg mL−1) were activatedby addition of 7 �L of EDC solution (8.77 mg mL−1 in water, freshlyprepared) and (eventually) 4.7 �L of NHS solution (7.9 mg mL−1 inwater) prior to addition of the 200 �L peptide solution. Couplingwas studied at pH 8 and 6.1 for EDC/NHS activation, and pH 6.1 forsingle EDC activation.

The number of moles of the surface COOH groups per gram ofNP0 activable by EDC was determined by HPLC on supernatants bythe relation N = NEDC,0 × (PAEDC,0 − PAEDC)/PAEDC,0, where NEDC,0 isthe initially introduced number of moles of EDC per gram of NP0,PAEDC,0 is the EDC peak area (at 8.1 min) corresponding the initiallyintroduced concentration, and PAEDC is the EDC peak area corre-sponding to the remaining concentration in the supernatant after2 h. The number of COOH functions per nm2 was calculated usingthe specific surface of the NP0 nanoparticles.

2.3. Coupling analysis by HPLC

Peptide coupling yields on the NPs and immobilized amountswere obtained through determination of the amounts of non-coupled peptide in the supernatants (recovered as mentionedabove) by reverse phase (RP) HPLC. The HPLC analysis was per-formed on a Agilent 1100 series instrument (column: Jupiter,Phenomenex, C18, 5 �m, 250 × 4.6 mm; injected volume: 20 �L)using a linear water/acetonitrile gradient (0.8 mL min−1, eluent A is0.1% TFA in water, eluent B is 0.09% TFA in acetonitrile/water 70/30by vol., 0–40% B in 15 min, then 40–0% B in 2 min, UV detectionat 215 nm, injected volume: 20 �L). The non-tagged peptide wasdetected at 12.1 min retention time and the tagged one at 12.6 min.The NHS was also detected at 6.2 min. Peptide and NHS calibrationcurves (peak area (PA) vs concentration) were established in thesame conditions than for coupling, in the absence of the NPs (phos-phate buffer, 10 mM, pH 8). The coupling yield (CY) was obtainedby the relation CY = 100 × (PA0 − PA)/PA0, where PA0 is the peakarea corresponding to the initially introduced peptide concentra-tion and PA is the peak area relative to the peptide concentrationin the supernatant of the dispersions after coupling. The immo-bilized peptide amount per gram of NP (Nmg/g) was obtained bythe relation: Nmg/g = CY × N0,mg/g/100 where N0,mg/g is the initiallyintroduced peptide amount per gram of NPs. The number of immo-bilized peptide per nm2 was calculated using the specific surface ofthe nanoparticles.

2.4. Peptide-polymer conjugate analysis by RP-HPLC and GPCafter NP hydrolysis

NVP-based block with grafted peptide (or without peptide, asa reference) was analyzed by HPLC and GPC after destroying NPsthrough PLA hydrolysis. The procedure was as follows: the copoly-mer coated NPs functionalized (or not) with the tagged peptide

non-coupled peptide by centrifugation (15,000 × g, 10 min) andredispersion in water. After one more centrifugation (15,000 × g,10 min) and supernatant removal, the pellet was washed one last

3 ces B: Biointerfaces 103 (2013) 298– 303

thHwei0laudLoPNNs

2

(UwsTapum

2

awtpNewbc

3

3i

cna[ttlbuofPfPo1

introduced peptide amounts provided valuable information. Asshown in Fig. 3, the NHS peak area corresponding to increasingamounts of introduced non-tagged peptide remained quasiunchanged, since the immobilized peptide amounts were nearly

3200

3700

4200

4700

5200

5700

0

5

10

15

20

25

30

0 10 20 30 40 50

NHS

pea

k ar

ea

g/gm(tnuo

maeditpep

dezilibom

mIN

P)

Introdu ced pep�de amoun t (m g/g NP)

00 N. Handké et al. / Colloids and Surfa

ime with water and an NaOH solution was added for PLA esterydrolysis. After neutralization, the samples were analyzed byPLC, on the same apparatus as mentioned above using a linearater/acetonitrile gradient (eluent A is 0.1% TFA in water, elu-

nt B is 0.08% TFA in acetonitrile/water 95/5 by vol., 0–100% Bn 20 min, then 100% B during 3 min, and 100–0% in 2 min, flow:.8 mL min−1, UV detection at 215 nm). Samples were also ana-

yzed by GPC on a system composed of a Waters 515 HPLC pump, rheodyne valve for manual injection (100 �L loop), a Waters col-mn oven thermostated at 40 ◦C, a Waters Refractive Index 2414etector and a Brookhaven Instrument Company Multi-angle Laseright Scattering detector. The stationary phase was a combinationf PSS Suprema 30A column, PSS Suprema 1000A column and aSS Suprema Precolumn. Eluent was a mixture of water and 0.1 MaNO3 at a flow rate of 1 mL min−1. Eluent was filtered on a 0.2 �mylon membrane filter (Alltech). Samples were filtered on a 0.2 �m

yringe filter (Sodipro) prior to injection.

.5. Size and zeta potential measurements

Size of the NPs was determined by dynamic light scatteringDLS) at 25 ◦C, using a Zeta Sizer NanoZS (Malvern instruments,K). Highly diluted colloidal dispersions in 1 mM NaCl solutionere used, and each value is at least the average of three mea-

urements. Zeta potentials were measured with Zeta Sizer NanoZS.he measurements of the electrophoretic mobility were carried outt 25 ◦C on dispersion samples highly diluted in phosphate bufferH 6.1, 10 mM, and the data were converted to the zeta potentialssing Smoluchowski equation. Each value was the average of fiveeasurements.

.6. Cytotoxicity studies on dendritic cells (DCs)

The cells were obtained after the various stages of separationnd culture corresponding to the Lanzavecchia method [20]. Cellsere issued from peripheral human blood, obtained via a conven-

ion signed with EFS (Etablissement Franc ais du Sang). Cells wereut in contact during 48 h with the naked or copolymer coated PLAPs in PBS (8100 NPs per DC). Cells alone were also studied as a ref-rence (for natural cell death). 2 �L of propidium iodide (2 �g/mL)ere added in the cell dispersion (100 �L) and cells were analyzed

y fluorescence activated cell sorting (FACS). The experiments werearried out on three independent donors.

. Results and discussion

.1. PLA-b-P(NAS-co-NVP) copolymer coated NPs (NS-cop NPs)nvestigated

We have reported recently the synthesis of an amphiphilic blockopolymer PLA-b-P(NAS-co-NVP) (NS-cop), showing strong alter-ating tendency for NAS and NVP units, and its further use as

surface modifier in the PLA nanoprecipitation process (Fig. 1)16]. The procedure relies on the solubilization of homo-PLA andhe copolymer (10 wt.% compared to the homo-PLA) in ace-one/acetonitrile, followed by its addition into water non-solvent,eading to the formation of the nanoparticles stabilized with thelock copolymer (NS-cop NPs). The NS activated esters of the NASnits were present in a quite high density (∼2.4 functions nm−2,mitting the hydrolyzed ester fraction) and therefore, available forurther biomolecule coupling at the surface of the NPs. The PLA-b-(NAS-co-NVP) copolymer used for the preparation of the NS ester

unctionalized PLA NPs studied in the present work exhibited aDI of 1.5, a PLA Mn of 19,000 g mol−1 and a P(NAS-co-NVP) Mn

f 13,000 g mol−1. The NS-cop NPs presented a mean diameter of69 nm and a polydispersity index (PI) of 0.12, as determined by

Fig. 2. Typical HPLC chromatogram of the supernatant of the coupling medium forKKK-iIL-1� lysine tagged peptide (12.6 min); NHS peak at 6.2 min.

photon correlation spectroscopy (PCS), and 31% hydrolyzed esterfraction (determined as previously described [16]).

3.2. Influence of lysine tag on peptide coupling efficiency onNS-cop NPs

The coupling of the peptides was performed in 10 mM sodiumphosphate buffer pH 8, for 24 h, at a NS-cop NP concentration of5 mg mL−1. The coupling efficiency was determined from the assayof the remaining free peptide in the supernatant of the couplingmedium by HPLC (Fig. 2, for typical HPLC chromatogram of thesupernatant). The influence of the lysine tag on the immobilizedpeptide amounts can be clearly evidenced in Fig. 3. The amountsof immobilized tagged peptide increased with increasing amountsof peptide added in the reaction mixture, whereas, for the non-tagged peptide, the amount of grafted peptide remained nearlyconstant at a value of about 2 mg/g NP (i.e. 0.042 peptide nm−2)over the peptide concentration range investigated. By plottingthe coupling isotherm, which represents the immobilized peptideamounts as a function of the residual peptide concentration in thesupernatant (Fig. 4), it appeared that the plateau of maximum pep-tide coupled was nearly reached at a value close to 27.5 mg/g, i.e.0.417 peptide nm−2, corresponding to a 10-fold molar increase incomparison with the non-tagged peptide.

The HPLC monitoring of the released NHS moieties in the super-natants after coupling (NHS peak at 6.2 min, Fig. 2) vs the different

Fig. 3. Immobilized amounts of KKK-iIL-1� tagged peptide (�) and iIL-1� non-tagged peptide (©) on NS-cop NPs, and NHS peak areas in HPLC (for tagged (�)and non-tagged (�) peptides); 24 h coupling reaction, phosphate buffer 10 mM,pH 8, NS-cop NP: 5 mg mL−1). Error bars represent standard deviation (SD) of n = 3independent experiments.

N. Handké et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 298– 303 301

Table 1Comparison of the concentrations of immobilized tagged peptide KKK-iIL-1� on NS-cop NPs and NHS released upon coupling (24 h coupling reaction, phosphate buffer10 mM, pH 8, NS-cop NP concentration: 5 mg mL−1). Results are given as the mean ± standard deviation (SD) of n = 3 independent experiments.

Introduced peptideamount (mg/g NP)

Immobilized peptideamount (mg/g NP)

Immobilized peptideconcentration (nmol mL−1)

NHS concentration(nmol mL−1)

ichccttwNNrt(htit

(dw2otatnoacaN

Rt

FNabr

12 11.7 ± 0.3

24 21.5 ± 0.6

36 27.5 ± 0.6

dentical. Conversely, for the tagged one, the NHS release signifi-antly increased, in parallel with the grafting of the tagged peptide,ence confirming the covalent character of the immobilization pro-ess. The value of NHS peak area obtained by extrapolation of theurve NHS peak area = f(introduced peptide amount) at zero pep-ide amount (i.e. 3650 peak area value) can be representative ofhe part of NHS released upon hydrolysis, which is in competitionith peptide coupling. Therefore, subtracting this value from theHS peak area values plotted on Fig. 3 provided access, throughHS calibration curve, to the concentration of the NHS actually

eleased during the peptide coupling step. These NHS concentra-ions were relatively close to those of immobilized tagged peptideTable 1), particularly at low peptide amounts. Interestingly, for theigher amounts of grafted peptides, the NHS release was superioro the expected 1:1 stoichiometry suggesting the grafted peptidesnduced a chain spacing in the hydrophilic corona, provoking fur-her access to until then “hidden” NS esters, for hydrolysis.

As a control experiment, coupling of the tagged peptideintroduced at 12 mg/g NP) was also performed in the same con-itions (10 mM phosphate buffer pH 8) on copolymer coated NPshose NS ester functions were hydrolyzed at more than 90%. After

h reaction, the coupling was 5%, compared to 56% for couplingf tagged peptide on the NS-cop NPs of 31% hydrolyzed ester frac-ion, again indicating that ester functions were well involved in

covalent coupling with the peptide, excluding thus purely elec-rostatic interactions between the lysine-tagged peptides and theegatively charged particles. At a lower pH of 6.1, though, bindingf the peptide via electrostatic interactions occurred up to 18%. Thisdsorption of the tagged peptide on the 90% ester hydrolyzed NPsould reasonably be attributed to electrostatically attractive inter-ctions between the fully protonated amines (lysine residues and-terminus) of the peptide and the negatively charged NPs.

To further confirm the peptide coupling, we performed polymerP-HPLC analysis after hydrolysis of the NPs (loaded with 27.5 mg/gagged peptide, and purified from NHS and non-coupled peptide)

0

5

10

15

20

25

30

0 0.02 0.04 0.06 0.08

Imm

obili

zed

pep�

de a

mou

nt (m

g/g

NP)

Residual pep�de concentra�on (m g/mL)

ig. 4. Immobilized amounts of tagged (�) and non-tagged (©) peptides on NS-copPs, as well as those of tagged (�) and non-tagged (�) peptides on naked NP0 NPss a function of residual peptide concentration (24 h coupling reaction, phosphateuffer 10 mM pH 8, NP concentration: 5 mg mL−1, EDC activation for NP0). Error barsepresent standard deviation (SD), n = 3.

42 ± 1 45 ± 977 ± 2 93 ± 1299 ± 2 147 ± 11

in alkaline conditions (NaOH). Through this procedure, esters ofboth the homo-PLA and the copolymer were converted into lac-tic acid (or small oligomers), giving raise to the hydrophilic NVPbased block with grafted peptide as the unique remaining poly-mer species (with presumably acrylic acid units left due to nontotal peptide coupling on the NAS units). As a reference, copoly-mer coated NPs in absence of peptide were also hydrolyzed instrictly the same conditions and subject to HPLC to analyze the NVPbased block without peptide (i.e. presenting only acrylic acid unitsfollowing hydrolysis of the NAS units). Free peptide was also ana-lyzed as a reference (elution time: ∼8 min). As shown in Fig. 5a,the reference copolymer was eluted under the form of the twopeaks (11.4 and 13.3 min elution times, respectively, probably dueto the two different chain ends on the polymer, i.e. presumablywith or without SG1 nitroxide), which were rather broad due tothe polydispersity of the polymer. As a comparison, interestingly,the peptide-polymer conjugate was eluted at lower elution times(again under the form of two broad peaks, 10.7 and 12.3 min) asa result of an increased hydrophilicity due to the presence of thepeptide coupled on the polymer. The peaks were broader thanthose of the reference polymer probably due to the various peptideconjugation degrees. Furthermore, since the introduced amount ofPLA/copolymer was the same for both NP samples, the peaks areasof peptide-polymer and of reference polymer (obtained by calibrat-ing at the same value the peak area of the formed lactic acid at5 min; not shown for sake of clarity) are interesting to compare.The total peaks area for the conjugate was about two-fold higherthan that of the reference copolymer, again indicating coupling ofthe peptide, consistent with the results obtained by indirect super-natant based HPLC method (27.5 mg of coupled peptide for 1 g of NP,i.e. for ∼40 mg NVP-based hydrophilic block). The presence of freepeptide (8 min) was detected in negligible amounts on the chro-matogram of the hydrolyzed peptide-NP sample, confirming thequasi-exclusive covalent nature of the binding. Finally, GPC analysiswas also performed for both peptide-polymer and reference poly-mer obtained after NP hydrolysis. As shown in Fig. 5b, the elutiontime of the peptide-polymer remained quasi unchanged comparedto the reference polymer, probably due to non significant changes inthe random coil size induced by the grafting of the peptide, whichis rather small. However, again, the peak of the peptide graftedpolymer was broader compared to the non functionalized polymerdue to the various peptide conjugation degrees, and the peak areahigher (normalizing the lactic acid peak at 21.3 min), due to graftedpeptide.

3.3. Influence of the P(NAS-co-NVP) corona on the immobilizedpeptide amounts

We further investigated the impact of the hydrophilic P(NAS-co-NVP) segment of the copolymer present at the particle surfaceon the immobilized peptide amounts. For this, we examinedthe coupling of tagged peptide on naked nanoparticles (without

copolymer) prepared in the same conditions following thenanoprecipitation process (NP0, 140 nm mean diameter,PI = 0.095). The latter particles indeed present carboxylic acidgroups at their surface that can be activated for peptide coupling

302 N. Handké et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 298– 303

Table 2Immobilized KKK-iIL-1� tagged peptide amounts on naked NP0 nanoparticles (phosphate buffer, 10 mM, NP0 concentration: 5 mg mL−1) in various coupling conditions.Results are described as the mean ± standard deviation (SD), n = 3.

Entry pH Couplingconditions

Reaction time (h) Introduced peptideamount (mg/g NP)

Couplingyield (%)

Immobilizedpeptide amount(mg/g NP)

1 8 – 2 12 8 0.96 ± 0.32 6.1 – 2 12 15 1.8 ± 0.43 6.1 EDC 2 12 54 6.48 ± 1.14 6.1 EDC 24 12 55 6.6 ± 1.05 6.1 EDC/NHS 2 12 41 4.92 ± 0.96 6.1 EDC/NHS 24 12 50 6.00 ± 0.87 8 EDC/NHS 2 12 40 4.80 ± 0.7

vcpoFtbettiabn(eaa6

cetTsa

Fwr

8 8 EDC/NHS 24

ia a carbodiimide/NHS procedure. We particularly studied theoupling efficiency of the tagged peptide on NP0 for an introducedeptide amount of 12 mg/g NP (Table 2), for comparing it to thatbtained on the NS-cop NPs (i.e. 97%, 11.7 mg/g immobilized,ig. 3). As expected, in the same conditions than those used withhe NS-cop NP particles, i.e. without coupling agent (phosphateuffer 10 mM, pH 8), the immobilization was not significant (8%,ntry 1) and very likely originated from electrostatic interac-ions between the lysine tag and the carboxylate functions ofhe particles. At pH 6.1, this non-covalent immobilization wasmproved due to fully protonated lysine amines and N-terminalmine of the peptide (15%, entry 2). For enabling covalent immo-ilization of the peptide on NP0, classical conditions were tested,amely EDC at pH 6.1 (entries 3 and 4), EDC/NHS at pH 6.1entries 5 and 6) and EDC/NHS at pH 8 (entries 7 and 8). What-ver the coupling conditions used, the coupling efficiency waslways about 50% (with no significant coupling improvementfter 2 h reaction), leading to immobilized amounts of about

mg/g NP.Finally, the influence of the lysine tag was again very signifi-

ant for coupling on the naked particles since the immobilizationxperiment with EDC activation (pH 6.1) performed with the non-

agged peptide led to very low coupling yield (8%, i.e. 1.1 mg/g NP).his coupling yield could be clearly attributed to covalent bindingince the same experiment performed in absence of EDC couplinggent yielded to negligible immobilization (1%).

ig. 5. (a) Reverse phase HPLC and (b) GPC chromatograms of NVP based hydrophilic blith 27.5 mg/g of tagged peptide (continuous line) and non-functionalized (short dotte

eference for HPLC.

12 47 5.64 ± 0.5

As shown in Fig. 4, the immobilized tagged peptide amountsat the plateau remained typically close to 6 mg/g for the nakedNP0 NPs, i.e. 0.075 peptide nm−2, while they could reach 27.5 mg/gfor the NS-cop NPs, i.e. 0.417 peptide nm−2. It is to mention thatnaked NP0 particles present however a similar density of functionalgroups (COOH) for coupling than the NS-cop NPs (NS ester func-tions), i.e. about 2 functions nm−2, highlighting the key role of theaccessibility allowed by the expanded polymer corona at the parti-cle surface for enabling such high peptide amounts immobilized.To our knowledge, such densities have never been achieved forsuch colloidal systems in the literature, the typical peptide densitiesbeing generally lower [6,9–11,14,21].

3.4. Peptide functionalized NPs colloidal characterization andcytotoxicity

The copolymer coated nanoparticles functionalized with thenon-tagged and tagged peptides (iIL-1� cop NP and KKK-iIL-1�cop NP) after washing by centrifugation/redispersion in water,were characterized in terms of size and zeta potential. As shownin Table 3, the size of the NPs after coupling of the peptideswas significantly increased to values of about 220 nm, as com-

pared to the initial particle size (169 nm), whatever the natureand amounts of peptide (Table 3, no significant increase of thesize distribution). In fact, this increase in size was essentiallyattributed to the swelling of the polymer corona upon competitive

ock obtained after alkaline hydrolysis of the copolymer coated NPs functionalizedd line); long dotted line represents the chromatogram of the tagged peptide as a

N. Handké et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 298– 303 303

Table 3Size and zeta potential (mean ± SD, n = 3) of the copolymer-peptide functionalized NPs after washing (centrifugation/redispersion in water).

NP code Immobilized peptideamount (mg/g NP)

Mean diameter (nm) PI Zeta potential (mV)

NS-cop NP – 169 ± 5 0.125 ± 0.025 –NS-cop NP-coup. mediuma – 225 ± 6 0.094 ± 0.014 −46 ± 0.5iIL-1� cop NP 1.73 226 ± 8 0.135 ± 0.035 −44.8 ± 0.4

1.60 222 ± 6 0.161 ± 0.028 −44.1 ± 1.11.95 228 ± 5 0.122 ± 0.020 −44.5 ± 0.6

KKK-iIL-1� cop NP 11.7 215 ± 4 0.165 ± 0.022 −43.4 ± 0.721.5 213 ± 3 0.164 ± 0.015 −42.2 ± 1.0

218

edium

hfswptbptcmsouptaua

4

pNmfmisaocttsa

[[[

[

[

[

[

[

[

27.5

a NS-cop NP-coup. medium: blank NS-cop NP that were placed in the coupling m

ydrolysis of the NS esters, since reference NS-cop NPs placedor 24 h in the coupling medium, but in absence of peptide,howed a mean size of 225 nm. An increase in zeta potentialas observed with increasing amounts of surface coupled taggedeptide, indicating the influence of the cationic lysine tags athe particle surface. However, this peptide presents several car-oxylic acids, which explains the rather moderate changes in zetaotential. It is to point out that the peptide-functionalized par-icles remained quite stable upon storage. In a further step, wehecked that no cytotoxicity was induced by these novel copoly-er based PLA NPs. We chose human dendritic cells as a model

ince they are known as a major target involved in the inductionf immune responses. Cell death was evaluated by FACS with these of propidium iodide. Cell death in presence of the copolymer-eptide functionalized NPs was very low (1.7 ± 0.9%) and similaro that obtained in presence of surfactant free PLA NPs (1.4 ± 0.8%)nd in absence of NPs (2 ± 1%, natural cell death), validating these of the copolymer at colloid interfaces for promoting peptidettachment.

. Conclusion

We reported in this paper a simple and efficient strategy forreparation of highly peptide functionalized biodegradable PLAPs. Our approach relies on the design of NS ester based copoly-er (PLA-b-P(NAS-co-NVP) surfactant and lysine-tagged peptide to

avor its approach toward the negatively charged NPs. Both copoly-er coating and lysine tag were shown to contribute to a dramatic

mprovement of the immobilized peptide amounts to reach den-ities never achieved until then for similar sized peptides. Thispproach can be applied to other peptides of interest for the designf highly peptide functionalized carriers in tailored drug or vac-ine delivery applications. Further studies are particularly ongoing

o evaluate the impact of the increased density of this immunos-imulatory peptide and its favored deployment at the particleurface on the vaccine potential of such novel PLA based particulatedjuvants.

[

[[

± 6 0.177 ± 0.010 −40.4 ± 0.5

(phosphate buffer 10 mM, pH 8, for 24 h), in absence of peptide.

Acknowledgements

The authors thank Aix-Marseille Université and CNRS forfinancial support, and E. Diesis from IBCP-UMS BiosciencesGerland–Lyon Sud for peptide synthesis.

References

[1] N. Kolishetti, S. Dhar, P.M. Valencia, L.Q. Lin, R. Karnik, S.J. Lippard, R. Langer,O.C. Farokhzad, Proc. Natl. Acad. Sci. U.S.A. 107 (2010) 17939.

[2] R.C. Mundargi, V.R. Babu, V. Rangaswamy, P. Patel, T.M. Aminabhavi, J. Control.Release 125 (2008) 193.

[3] A. Bershteyn, M.C. Hanson, M.P. Crespo, J.J. Moon, A.V. Li, H. Suh, D.J. Irvine, J.Control. Release 157 (2012) 354.

[4] V.W. Bramwell, Y. Perrie, Crit. Rev. Ther. Drug Carrier Syst. 22 (2005) 151.[5] T.M. Fahmy, S.L. Demento, M.J. Caplan, I. Mellman, W.M. Saltzman,

Nanomedicine 3 (2008) 343.[6] P.M. Valencia, M.H. Hanewich-Hollatz, W. Gao, F. Karim, R. Langer, R. Karnik,

O.C. Farokhzad, Biomaterials 32 (2011) 6226.[7] X. Jiang, X. Sha, H. Xin, L. Chen, X. Gao, X. Wang, K. Law, J. Gu, Y. Chen, Y. Jiang,

X. Ren, Q. Ren, X. Fang, Biomaterials 32 (2011) 9457.[8] J. Park, T. Mattessich, S.M. Jay, A. Agawu, W.M. Saltzman, T.M. Fahmy, J. Control.

Release 156 (2011) 109.[9] N. Brandhonneur, F. Chevanne, V. Vié, B. Frisch, R. Primault, M.F. Le Potier, P. Le

Corre, Eur. J. Pharm. Sci. 36 (2009) 474.10] Z. Wang, W.K. Chui, P.C. Ho, Pharm. Res. 28 (2011) 585.11] D.H. Yu, Q. Lu, J. Xie, C. Fang, H.Z. Chen, Biomaterials 31 (2010) 2278.12] V. Fievez, L. Plapied, A. des Rieux, V. Pourcelle, H. Freichels, V. Wascotte,

M.L. Vanderhaeghen, C. Jerôme, A. Vanderplasschen, J. Marchand-Brynaert, Y.J.Schneider, V. Préat, Eur. J. Pharm. Biopharm. 73 (2009) 16.

13] J. Rieger, H. Freichels, A. Imberty, J.L. Putaux, T. Delair, C. Jerome, R. Auzely-Velty,Biomacromolecules 10 (2009) 651.

14] A. Fakhari, A. Baoum, T.J. Siahaan, K.B. Le, C. Berkland, J. Pharm. Sci. 100 (2011)1045.

15] N. Nasongkla, X. Shuai, H. Ai, B.D. Weinberg, J. Pink, D.A. Boothman, J. Gao,Angew. Chem. 116 (2004) 6483.

16] N. Handké, T. Trimaille, E. Luciani, M. Rollet, T. Delair, B. Verrier, D. Bertin, D.Gigmes, J. Polym. Sci. A Polym. Chem. 49 (2011) 1341.

17] K. Knop, R. Hoogenboom, D. Fischer, U.S. Schubert, Angew. Chem. Int. Ed. 49(2010) 6288.

18] D. Boraschi, A. Tagliabue, Methods 19 (1999) 108.

19] C. Ladavière, C. Lorenzo, A. Elaïssari, B. Mandrand, T. Delair, Bioconjug. Chem.

11 (2000) 146.20] A. Lanzavecchia, F. Sallusto, Science 290 (2000) 92.21] N. Zhang, C. Chittapuso, C. Duangrat, T.J. Siahaan, C. Berkland, Bioconjug. Chem.

19 (2008) 145.