Rb

SVa

b

c

d

e

h

�

�

�

�

a

ARRAA

KLTPCT

1

ptr

0h

Colloids and Surfaces A: Physicochem. Eng. Aspects 417 (2013) 83– 88

Contents lists available at SciVerse ScienceDirect

Colloids and Surfaces A: Physicochemical andEngineering Aspects

jo ur nal homep a ge: www.elsev ier .com/ locate /co lsur fa

elease kinetics of gold nanoparticles from collagen microcapsulesy total reflection X-ray fluorescence

vetlana Erokhinaa, Oleg Konovalovb, Paolo Bianchini c, Alberto Diasproc, Carmelina Ruggierod,ictor Erokhine, Laura Pastorinod,∗

Department of Biochemistry and Molecular Biology, University of Parma, Viale Usberti 23 A, 43100 Parma, ItalyEuropean Synchrotron Radiation Facility, 38043 Grenoble Cedex, FranceNanophysics, Fondazione Istituto Italiano di Tecnologia, 16163 Genoa, ItalyDepartment of Informatics, Bioengineering, Robotics and Systems Engineering, University of Genova, Via all’Opera Pia 13, 16145 Genova, ItalyCNR-IMEM, and Department of Physics, University of Parma, Viale Usberti 7A, 43124 Parma, Italy

i g h l i g h t s

Collagen based microcapsules forsmart drug delivery.Disease triggered release by over-expressed enzymes.Synchrotron radiation based totalreflection X-ray fluorescence.Real-time release kinetics of goldnanoparticles.

g r a p h i c a l a b s t r a c t

r t i c l e i n f o

rticle history:eceived 19 September 2012eceived in revised form 30 October 2012ccepted 2 November 2012vailable online 17 November 2012

a b s t r a c t

One of the main challenges in nanotechnology based drug delivery systems is the design of novel materialsfor patient specific therapeutic efficacy. This can be achieved taking into account changes in the expressionprofiles of related proteins. Here, we focus on collagen type I micro-capsules for releasing molecules usinga biological stimulus elicited by over-expressed collagenolytic matrix metalloproteinase I (MMPI). Wehave carried out label-free real time monitoring of the release kinetics of gold nanoparticles induced by

eywords:ayer-by-Layer capsulesriggered releaseermeability variationollagenotal reflection X-ray fluorescence

MMPI in order to define the pore opening process by synchrotron radiation based total reflection X-rayfluorescence. Scanning electron and confocal laser scanning microcopy investigations have been carriedout in order to further characterize the interaction between collagen micro-capsules and MMP1. Theresults enable to predict and to engineer the release time for drug molecules of different sizes. Moreover,the proposed label-free approach can be adopted for the visualization of pore formation and of releasekinetics for nano-engineered polymeric capsules.

. Introduction

Nanoengineered polymeric capsules have been shown to be

romising perspective candidates for use as smart containers forargeted drug release [1–7]. First, it allows decreasing the releaseate of the core substance [8]. Second, the possibility of pore

∗ Corresponding author. Tel.: +39 103532217; fax: +39 0103532154.E-mail address: laura.pastorino@unige.it (L. Pastorino).

927-7757/$ – see front matter © 2012 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.colsurfa.2012.11.012

© 2012 Elsevier B.V. All rights reserved.

opening in the shell of the capsules allows the controllable releaseof the encapsulated substance. Pore opening usually results frompH variation even though it is also possible to induce the release byother factors, such as light, ultrasound and magnetic field [9–16].It was reported also an utilization of ultrasonic treatment forthese reasons [17]. We proposed collagen based nanoengineered

capsules as smart drug containers for the therapy of diseases char-acterized by its up-regulated degradation, such as arthritis, cancerand cardiovascular diseases [18,19]. Such pathological states arecharacterized by the over expression of the collagenolytic enzymes

8 A: Physicochem. Eng. Aspects 417 (2013) 83– 88

motdmtccrHlcwnmasnGflhrra(

2

2

BioP6Pdc0tccta3sakcteachAshbp6iTwo

4 S. Erokhina et al. / Colloids and Surfaces

atrix metalloproteinases and by the down-regulated expressionf their inhibitors [20,21]. The drug release could be triggered byhe specific biochemistry of the pathological state under treatmentepending also on the individual levels of over-expression, whichay vary from patient to patient. We demonstrated the release of

he encapsulated dyes after the reaction of collagen type I micro-apsules with the collagenolytic enzyme MMP1 [18,19]. Usually,onfocal laser scanning microscopy is used to characterize theelease of fluorescent dyes encapsulated into polymeric capsules.owever, label free methods would be more suitable to study

ong term release kinetics [22–24]. Moreover, in order to properlyharacterize the pore opening process it is important to use objectsith well-defined and stable structure and size. To this aim, goldanoparticles (GNPs) can be very useful, since methods to fabricateonodisperse GNPs with precisely controlled sizes are available

nd GNPs have been used for improved drug delivery applicationsince they can be easily conjugated to drug molecules and inter-alized by cells [25]. We have focus on the use of two types ofNPs, namely 6 nm and 50 nm in diameter. We have used X-rayuorescence for real-time monitoring of the release, since goldas well detectable fluorescence lines. We have used synchrotronadiation for the excitation in order to have reliable signal-to-noiseatio. At the microscopic level, we have studied the capsules beforend after the exposition to MMP1 by scanning electron microscopySEM) on the samples used for X-ray fluorescence.

. Experimental section

.1. Synthesis and loading of collagen micro-capsules

The capsules have been prepared as previously described [18].riefly, 108 MnCO3 templates (PlasmaChem GmbH, Berlin), 6 �m

n diameter, have been covered by successively deposited layersf anionic polystyrene sulfonate (PSS) and cationic collagen Type I.SS and collagen solutions have been prepared in NaCl 0.05 M pH.5 respectively at a concentration of 2 mg/ml and 0.5 mg/ml. BothSS and collagen were from Sigma–Aldrich. Four bilayers have beeneposited onto the surface of the templates and the last one was ofollagen. Templates have been dissolved by three washings in NaCl.05 M at pH 2. GNPs have been synthesized with a diameter respec-ively of 6 nm with a concentration of 10 mg/ml and a negativeharge and of 50 nm with a concentration of 1 mg/ml and a positiveharge. Loading with GNPs [26] has been performed by exposinghe capsules to the solution of the GNPs at pH 3, when capsulesre in an open configuration, for 24 h with continuous shaking at00 rpm/min. Afterwards, the capsules have been precipitated, theupernatant has been removed and buffer solution (NaCl 50 mM)t pH 9 has been added in order to close the capsules. For releaseinetics characterization, MMP1 from Sigma has been used at aoncentration of 10 �g/ml in pure water. 8-Hydroxypirene-1,3,6-risulfonic acid trisodium salt (HPTS, fluorescence excitation andmission maxima of 460 and 510 nm, respectively; from Fluka) at

concentration of 0.025 mg/ml pH 3, has been used to load emptyapsules for confocal laser scanning microscopy. Loading with HPTSas been performed following the same procedure used for GNPs.

known amount of loaded capsules has been dispersed into aolution of MMP1 and incubated at 37 ◦C for 24 h. Investigationsave been performed on loaded untreated and treated capsulesy exposing them to a solution of the dye ATTO 647 (0.5 mg/mlH 7.5, fluorescence excitation and emission maxima of 645 and70 nm, respectively; Atto-Tec, Siegen, Germany) in order to put

n evidence the MMP1 mediated degradation of the capsule shell.he water, used in the experiments for solutions preparation andashing, has been purified by Milli-Q system and its resistance was

f 18.2 M � cm.

Fig. 1. (a) Fluorescence spectrum of gold. (b) Scheme of the experiment and photoof the cell on the 3D moving support.

2.2. Morphological characterization of collagen micro-capsules

SEM measurements have been carried out by Zeiss Supra micro-scope. Samples have been prepared by dropping the solutions onsilicon supports with successive drying. To this purpose, siliconsubstrates, having an area of 0.5 mm × 0.5 mm, have been cleanedwith H2SO4 at 150 ◦C for 20 min followed by extensive washing inpure water.

2.3. Real time release kinetics of gold nanoparticles

X-ray fluorescence measurements have been carried out atID10B beamline of European Synchrotron Radiation Facilities (ESRF,Grenoble) using 22 kV excitation energy. The fluorescence signalhas been detected by a silicon drift X-ray detector (Vortex-EMTM,SII NanoTechnology USA Inc.) located above the liquid surface at5 mm. Spectrum of the gold fluorescence, according to which thecalibration of the detector has been done, is shown in Fig. 1a.Two strong fluorescence lines have been taken for the furthermeasurements, namely 9.71 keV and 11.47 keV. Scheme of the mea-surements and photo of experimental cell are shown in Fig. 1b.Excitation was done with X-ray beam, oriented and collimatedparallel to the water surface in the cell. Fluorescence of gold wasregistered in the upper water volume (approximately 10 �m thick).Initially, we have supposed that it is necessary to separate the reac-tion area from the measuring one in order to avoid the presence ofthe capsules not yet reacted in the zone of measurements. There-fore, the filter with the pore size smaller than the capsule diameterwas inserted in order to separate the measuring area. However,during the experiments, we have realized that this action is not nec-essary and, moreover, provides additional side effects: the productsof the reaction were stopped at the surface of the filter preventingthe diffusion of GNPs. Thus, the final set-up, used for the experi-ment, was without the filter, assuming only on the precipitation

of the loaded capsules after a reasonable time (about 15 min wasfound to be enough to have no signal of the Au at the detectorfor a period of more than 10 h if no MMP1 was added). The mea-surements have been carried out as follows. The capsules loaded

: Physicochem. Eng. Aspects 417 (2013) 83– 88 85

wii

2

owtbim(T(lth1G6fi

3

cbafesiatctdAeobbGcGttcGaittcFtTttttt

S. Erokhina et al. / Colloids and Surfaces A

ith GNPs of 2 different sizes (6 and 50 nm) have been insertednto the measurement chamber. Then, after 15 min, MMP1 has beennjected into the chamber and the measurements started.

.4. Real time release kinetics of dye molecules

Confocal laser scanning microscopy (CLSM) has been carriedut as an additional experiment when the capsules were loadedith dye molecules (sizes are less than in the case of GNPs). Two

ypes of dye molecules with distinct emission wavelengths haveeen used for this purpose. Permeability variation has been stud-

ed by simultaneous analysis of the release of encapsulated dyeolecules (HPTS) from capsules and entrance of the dye molecules

ATTO) from the environmental solution into the capsules volume.he measurements have been carried out by a Leica TCS SPED-CWLeica Microsystems CMS, Mannheim, Germany) inverted confocalaser scanning microscope equipped with a set of lasers coveringhe 458-, 476-, 488-, 496-, 514-, 543-and 633 nm lines. The imagesave been collected using a Leica 100× HCX PL APO STEDorange NA.40 oil immersion objective (Leica Microsystems CMS, Mannheim,ermany). The images have been obtained using the 488 nm and33 nm laser lines. Under this imaging configuration, typical con-ocal resolution is of the order of 200 nm in the lateral and 500 nmn the axial direction.

. Results and discussion

Before starting the release investigating experiments, we haveompared the structure of the capsules before and after loadingy SEM imaging. The images of the 6 � dried capsules before (a)nd after (b) loading are shown in Fig. 2a and b. As can be seenrom Fig. 2a and b, a significant difference exists between thempty and loaded collagen micro-capsules. For the empty cap-ules (Fig. 2a), a flat shell can be observed, which is aggregatedn a not-homogeneous manner. This was already observed andttributed to the fact that collagen forms a stable configuration inhe balanced water-compensed environment, whereas in the dryonditions a self-aggregation can be observed, which originateshick zones where most of the collagen is aggregated. This con-ition is even more pronounced in the loaded capsules (Fig. 2b).s relates to the state of the shell after drying, similarly to thempty case (Fig. 2a) areas with increased thickness can be seen. Thisbservation can be associated to the differences that can be noticedetween the self-aggregation conditions of the molecular assem-ly in the isotropic (Fig. 2a) and anisotropic (Fig. 2b) conditions.NPs of 50 nm are clearly visible in the zoomed images of loadedapsules (Fig. 2c). However, it is not possible to state whether theNPs are in the core of the capsules, or they are mostly attached to

he shell. In the case of 6 nm particles, 3D features appear, similaro those shown in Fig. 1b. However, direct imaging of the parti-les themselves is very difficult. For real time release kinetics ofNPs, loaded capsules have been put into the measuring chambernd allowed to precipitate onto the bottom, then MMP1 has beennjected and the measurement started. The fluorescent signal, doo released GNPs, has been detected by a silicon drift X-ray detec-or located above the liquid surface. Real time release kinetic forapsules loaded with 6 nm GNPs in contact with MMP1 is shown inig. 3a. As can be noticed from Fig. 3a, initially GNPs were prac-ically not present in the upper part of the measuring solution.his is due to the fact that the capsules precipitated and concen-rated at the bottom of the measuring chamber. After about 1000 s,

he release of encapsulated GNPs started. This can be attributedo the formation of pores with sizes higher than 6 nm. We haveested this assumption by SEM imaging of the dried residual solu-ion, collected from the measuring chamber at the end of X-ray

Fig. 2. SEM images of 6 � before (a) and after (b) loading with 50 nm GNPs. (c)Zoomed SEM image of 50 nm GNPs loaded polymeric capsules.

fluorescence experiments. A typical image is shown in Fig. 3b. Thisimage demonstrates the presence of flat empty capsules with dis-turbed, non-spherical shape. The sizes of the objects in the image,however, are comparable with the initial capsule diameter (about6 �m). The comparison of the X-ray fluorescence data with thoseof SEM imaging supports the conclusion that the GNPs have beeneffectively released from the capsules through the pores, derivingfrom the collagenolytic action of MMP1. A further interesting fea-

ture of Fig. 3a is that the intensities of the Au fluorescence linesraise and subsequently decrease. This can be explained as follows.When pores of appropriate sizes are formed in the capsule shell,

86 S. Erokhina et al. / Colloids and Surfaces A: Phy

Fig. 3. (a) Temporal dependences of the intensities of Au fluorescence lines forccM

seacadtdtwfswOc

Fc

apsules, loaded with 6 nm GNPs. (b) SEM image of the dried solution of collagen-ontaining polymeric capsules, loaded with 6 nm GNPs, after the reaction withMP1.

ustained release of the GNPs occurs due to the significant differ-nce in their concentration between the part inside the capsulesnd the environment in which the capsules are dispersed. Afteromplete release of GNPs (maximum in the intensity), their aver-ge concentration in the volume in the solution is likely to decreaseue to 2 reasons: precipitation of the GNPs and their adsorption onhe walls of the measuring chamber. Thus, this brings about slowecrease of the fluorescence intensities of Au lines (upper layer ofhe liquid reaction medium). For capsules loaded with 50 nm GNPs,e observed a rather different situation. Real time release kinetic

or 50 nm GNPs is shown in Fig. 4. Comparing the kinetics for cap-

ules loaded with 6 nm GNPs (Fig. 3a) with those for capsules loadedith 50 nm (Fig. 4) reveals one similarity and several differences.ne obvious difference is the time when the release starts. In thease of 6 nm GNPs, this time was about 1000 s (Fig. 3a), while in

ig. 4. Kinetics of the Au fluorescence lines variation in the upper part of the workinghamber when the capsules are loaded with 50 nm GNPs in the presence of MMP1.

sicochem. Eng. Aspects 417 (2013) 83– 88

the case of 50 nm GNPs this time was about 8500 s (Fig. 4). Thisdifference is likely to be related to the difference in the particlesize. In fact, it can be noticed that the ratio between the releasestarting times (8.5) is close to the ratio between GNPs diameters(8.3). The presence of the fluorescence maximum in the kineticscurve is a similar feature of Figs. 3a and 4. Taking into account themaximum region in Fig. 3a and the first maximum of Fig. 4a, thismaximum can be related to the release of GNPs, previously encap-sulated in the core volume, and to their subsequent precipitationand adsorption onto the walls of the chamber. However, for timevalues greater than the values of the maximum region the behaviorof the kinetic curves for the 50 nm GNPs loaded capsules (Fig. 4) isdifferent from that of the 6 nm GNPs loaded ones (Fig. 3a). For 50 nmGNPs the initial decrease of the fluorescence can be explained asin the case of 6 nm GNPs. For greater time values, a non-regularvariation of the fluorescence intensity can be observed. In order tounderstand the origin of such behavior, SEM imaging of the driedresidual liquid after the end of the X-ray fluorescence measure-ments has been carried out. A wide area SEM image is shown inFig. 5a. The first important aspect of this image is that it does notcontain features similar to the ones of empty capsules. Instead, wecan see some light areas and some dark areas. Zooming of the lightareas (Fig. 5b and c) reveals the presence of the 50 nm GNPs used forthe capsule loading. It is worth noting that no such particles havebeen observed in the case of the analysis of dried capsule solutionsbefore their interaction with MMP1. This finding confirms that, atleast partially, particles have been encapsulated into the core vol-ume. The maximum of the fluorescence in Fig. 3a can be relatedto their release. Zooming of the dark areas (Fig. 5d) reveals thepresence of collapsed-like non structured objects and well-visible50 nm GNPs, used for the capsule loading. These objects can be asso-ciated with the fragments of the capsule shells with attached GNPs.On the whole, on the basis of the findings above we suggest the fol-lowing model of the interaction of collagen micro-capsules withMMP1. The destruction of the collagen (and, therefore, the wholeshell) is a rather slow process. The pores allowing the release of6 nm particles are formed only after about 1000 s. In the case ofsuch GNPs loading, the release kinetics experiment by X-ray fluo-rescence ended after about 3000 s, and the analysis of the residualdried liquid has shown that the capsules have maintained their size,but have completely lost their shape, demonstrating that all theencapsulated GNPs has been released. For 50 nm GNPs loaded cap-sules, a release similar to that found for 6 nm GNPs loaded capsules,occurring in the time interval of 8500–10,000 s, can be observed. Itis worth noting that the time necessary for the release is of thesame order as the difference in the encapsulated GNPs diameter. Itcan be therefore inferred that the interaction of MMP1 with colla-gen has rather linear character (at least for the particle size, used inthis study). The linear rate of the pore formation can be estimatedas about 6 × 10−3 nm/s. 11,000 s after the interaction with MMP1,the situation has drastically changed. An occasional variation of theintensity of Au fluorescence lines (Fig. 4) has been observed. Com-paring with SEM data, we can conclude that this stage correspondsto the destruction of the capsule structure. The subsequent randomvariation of the fluorescence intensity corresponds to the statisti-cal inflow of the fragments of the capsule with attached GNPs intothe upper part of the measuring chamber. For smaller GNPs, therelease has ended significantly earlier, therefore we have observedquasi-intact empty capsules, rather than shell fragments.

The other interesting conclusion can be done comparing theintensities, registered for GNPs of different dimensions. Consid-ering similar GNPs molar concentration of the initial loading

solutions, we can expect proximally equal amount of the loadedparticles in capsules. If such, in the case of saturation, molar con-centration of the GNPs in the volume of the solution after the releaseshould be the same after the release. Considering that the particle

S. Erokhina et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 417 (2013) 83– 88 87

F ing ch5 ) Zoo

dt

mo

a

ig. 5. (a) Wide area SEM image of the dried residual liquid, taken from the measur0 nm GNPs, in the presence of MMP1. (b and c) Zooming of the light areas of (a). (d

iameter ratio is about 10, we must have 3 orders of magnitude inhe difference of the registered fluorescence intensities.

As an additional test, we have carried out confocal fluorescence

easurements on capsules loaded with dye molecules. The results

f the experiments are shown in Fig. 6.Panel A and B are the transmission and the fluorescence aver-

ge projection image of the time sequence depicted in the graph

Fig. 6. Confocal fluorescence time lapse of HPTS loaded (PSS/collagen)4 capsule, perfu

amber after the end of X-ray fluorescence measurements on capsules, loaded withming of the dark areas the dark area of (a).

C. After the addition of MMP1 the fluorescence inside the capsuledecreases. After about 2000 s (practically complete release of theencapsulated dye molecules) red dye has been added to the solu-

tion. The following time sequence is shown in graph F, and panelD and E are the corresponding transmission and fluorescence aver-age projection image. The graph F shows that after the additionof a red shifted fluorophore the corresponding fluorescence inside

sed sequentially with a solution of MMP1 (A, B and C) and Atto 647 (D, E and F).

8 A: Phy

tBaeIpflb1tbsr

4

iiedOTwo

uolatp

A

f

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

8 S. Erokhina et al. / Colloids and Surfaces

he capsule increases, confirming that the capsules are open. In the fluorescence has been excited at 488 nm and collected in

500–550 nm spectral window. In E the fluorescence has beenxcited at 633 nm and collected in a 660–700 nm spectral window.n this case it is more difficult to estimate the time necessary forore formation allowing the dye release, since the decrease of theuorescence occurs immediately after the addition of MMP1. Suchehavior is due to the fact that the size of the molecule is less than.0 × 0.5 nm. Thus, taking the pore formation rate obtained fromhe experiments with GNPs, it can be estimated that the releaseegins after less than 300 s. In fact, the release can be even faster,ince molecules unlike GNPs, can vary their conformation for easierelease.

. Conclusions

Two main conclusions can be drawn from the present work.First, we have demonstrated the linear temporal effect of the

nteraction of MMP1 with collagen type I micro-capsules. The start-ng time of the release is directly proportional to the size of thencapsulated substance. On one hand, this makes it possible to pre-ict the time when the drug release occurs (knowing the drug sizes).n the other hand, this makes it possible to engineer the release.his can be achieved in several ways. Conjugating of drug moleculesith specific nanoparticles, such as GNPs or magnetic ones, allows

btaining a specific release profile.Second, the work also has a methodological value, since we have

sed label-free X-ray fluorescence for the real time monitoringf the pore formation and subsequent release of the encapsu-ated objects. Since monodispersed GNPs are easily available, thepproach that has been adopted can be a basis for the visualiza-ion of pore formation and of release kinetics for nano-engineeredolymeric capsules of different nature.

cknowledgment

We acknowledge the European Synchrotron Radiation Facilityor the use of the ID10 beamline.

eferences

[1] E. Donath, G.B. Sukhorukov, F. Caruso, S.A. Davis, H. Mohwald, Novel hol-low polymer shells by colloid-templated assembly of polyelectrolytes, Angew.Chem. Int. Ed. Engl. 37 (1998) 2202–2205.

[2] G.B. Sukhorukov, E. Donath, S. Davis, H. Lichtenfeld, F. Caruso, V.I. Popov,

H. Mohwald, Stepwise polyelectrolyte assembly on particle surfaces: a novelapproach to colloid design, Polym. Adv. Technol. 9 (1998) 759–767.

[3] A. Antipov, G.B. Sukhorukov, S. Leporatti, I.L. Radtchenko, E. Donath, H.Mohwald, Polyelectrolyte multilayer capsule permeability control, Colloid Surf.Physicochem. Eng. Asp. 198 (2002) 535–541.

[

[

sicochem. Eng. Aspects 417 (2013) 83– 88

[4] T. Mauser, C. Dejugnat, G.B. Sukhorukov, Reversible pH-dependent propertiesof multilayer microcapsules made of weak polyelectrolytes, Macromol. RapidCommun. 25 (2004) 1781–1785.

[5] K. Ariga, Y.M. Lvov, K. Kawakami, Q. Ji, J.P. Hill, Layer-by-layer self-assembledshells for drug delivery, Adv. Drug Deliv. Rev. 63 (2011) 762–771.

[6] N. Habibi, L. Pastorino, F.C. Soumetz, F. Sbrana, R. Raiteri, C. Ruggiero, Nano-engineered polymeric S-layers based capsules with targeting activity, ColloidsSurf. B: Biointerfaces 88 (2011) 366–372.

[7] N. Habibi, L. Pastorino, O.H. Soumetz, C. Ruggiero, Polyelectrolyte based molec-ular carriers: the role of self-assembled proteins in permeability properties, J.Biomater. Appl. (2012), http://dx.doi.org/10.1177/0885328212446358.

[8] Z. An, G. Lu, H. Mohwald, J. Li, Self-assembly of human serum albumin (HSA) andL-�-dimyristoylphosphatidic acid (DMPA) microcapsules for controlled drugrelease, Chem. Eur. J. 10 (2004) 5848–5852.

[9] B.G. De Geest, G.B. Sukhorukov, H. Möhwald, The pros and cons of polyelec-trolyte capsules in drug delivery, Expert Opin. Drug Deliv. 6 (2009) 613–624.

10] V. Vergaro, F. Scarlino, C. Bellomo, R. Rinaldi, D. Vergara, M. Maffia, F. Baldas-sarre, G. Giannelli, X. Zhang, Y.M. Lvov, S. Leporatti, Drug-loaded polyelectrolytemicrocapsules for sustained targeting of cancer cells, Adv. Drug Deliv. Rev. 63(2011) 847–864.

11] M. Delcea, H. Möhwald, A.G. Skirtach, Stimuli-responsive LbL capsules andnanoshells for drug delivery, Adv. Drug Deliv. Rev. 63 (2011) 730–747.

12] M.F. Bédard, B.G. De Geest, A.G. Skirtach, H. Möhwald, G.B. Sukhorukov, Poly-meric microcapsules with light responsive properties for encapsulation andrelease, Adv. Colloid Interface Sci. 158 (2010) 2–14.

13] S. Erokhina, L. Benassi, P. Bianchini, A. Diaspro, V. Erokhin, M.P. Fontana,Light-driven release from polymeric microcapsules functionalized with bac-teriorhodopsin, J. Am. Chem. Soc. 31 (2009) 9800–9804.

14] L. Pastorino, S. Erokhina, F. Caneva Soumetz, C. Ruggiero, Paclitaxel-containingnano-engineered polymeric capsules towards cancer therapy, J. Nanosci. Nan-otechnol. 9 (2009) 6753–6759.

15] D.G. Shchukin, D.A. Gorin, H. Mohwald, Ultrasonically induced opening of poly-electrolyte microcontainers, Langmuir 22 (2006) 7400–7404.

16] T.A. Kolesnikova, D.A. Gorin, P. Fernandes, S. Kessel, G.B. Khomutov, A. Fery, D.C.Shchukin, H. Mohwald, Nanocomposite microcontainers with high ultrasoundsensitivity, Adv. Funct. Mater. 20 (2010) 1189–1195.

17] W. Song, Q. He, H. Mohwald, Y. Yang, J. Li, J. Controlled Release 139 (2009)160–166.

18] L. Pastorino, S. Erokhina, F.C. Soumetz, P. Bianchini, O. Konovalov, A. Diaspro, C.Ruggiero, V. Erokhin, Collagen containing microcapsules: smart containers fordisease controlled therapy, J. Colloid Interface Sci. 357 (2011) 56–62.

19] L. Pastorino, S. Erokhina, O. Konovalov, P. Bianchini, A. Diaspro, C. Rug-giero, Permeability variation study in collagen-based polymeric capsules,BioNanoScience 1 (2011) 192–197.

20] H. Nagase, Matrix metalloproteinases, in: N.M. Hooper (Ed.), Zinc Met-alloproteases in Health and Disease, Taylor and Francis, London, 1996,pp. 153–205.

21] J.F. Woessner, The matrix metalloproteinase family, in: W.C. Parks, R.P. Mecham(Eds.), Matrix Metalloproteinases, Academic Press, San Diego, 1998, pp. 1–13.

22] H. Pool, D. Quintanar, J.D. Figueroa, J.E.H. Bechara, D.J. McClements, S. Men-doza, Polymeric nanoparticles as oral delivery systems for encapsulation andrelease of polyphenolic compounds: impact on quercetin antioxidant activityand bioaccessibility, Food Biophys. 7 (2012) 276–288.

23] S.E. Lee, L.P. Lee, Biomolecular plasmonics for quantitative biology andnanomedicine, Curr. Opin. Biotehnol. 21 (2010) 489–497.

24] G. Nakazawa, A.V. Finn, F.D. Kolodgie, R. Virmani, A review of current devicesand a look at new technology: drug-eluting stents, Expert Rev. Med. 6 (2009)33–42.

25] P. Ghosh, G. Han, M. De, C.K. Kim, V.M. Rotello, Gold nanoparticles in deliveryapplications, Adv. Drug Deliv. Rev. 60 (2008) 1307–1315.

26] T. Berzina, A. Pucci, G. Ruggieri, V. Erokhin, M.P. Fontana, Gold nanoparticles-polyaniline composite material: synthesis, structure and electrical properties,Synth. Met. 161 (2011) 1408–1413.