Religion & Spirituality

A Cyclodextrin-Based Nanoassembly with Bimodal Photodynamic Action

Description
A Cyclodextrin-Based Nanoassembly with Bimodal Photodynamic Action
Published
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  DOI: 10.1002/chem.201101635 A Cyclodextrin-Based Nanoassembly with Bimodal Photodynamic Action Noufal Kandoth, [a] Elisa Vittorino, [a] Maria Teresa Sciortino, [b] Tiziana Parisi, [b] Ivana Colao, [b] Antonino Mazzaglia,* [c] and Salvatore Sortino* [a] Introduction Bimodal therapies aim to exploit either additive or synergis-tic effects arising from the generation of two active speciesin the same region of space, with the final goal of maximiz-ing the therapeutic efficacy. The site of action, timing, anddosage of the delivered species play key roles in determin-ing the therapeutic outcome. [1] Light is a highly orthogonaltrigger for the rapid introduction of therapeutic agents in adesired bio-environment with the potential for precise con-trol over all the above factors and the additional advantageof not affecting important physiological parameters such astemperature, pH, and ionic strength. [2] These propertiesmake phototherapeutic agents a powerful arsenal for treat-ing cancer diseases in a noninvasive way, [3] avoiding the pos-sible complications of surgery. [4] Nitric oxide (NO) is one of the most appealing and in-tensely studied molecules in the fascinating realm of the bio-medical sciences. [5] Besides its pivotal role in the mainte-nance and bioregulation of vital functions, [6] NO has recentlystimulated an upsurge of interest because of its promisinganticancer activity. [7] This exciting discovery has made thedevelopment of new strategies and methods for NO photo-delivery a hot topic with the intriguing prospect of tacklingcancer diseases. [8] In this regard, the use of NO in conjunc-tion with other anticancer species is highly desirable for ef-fective therapeutic action. [9] Singlet oxygen ( 1 O 2 ) is the best-known phototherapeuticagent and its role in photodynamic therapy (PDT) is well es-tablished. [10] In PDT, the cytotoxic  1 O 2  is generated by pho-toinduced energy transfer between the lowest excited tripletstate of a suitable photosensitizer (e.g., a porphyrin orphthalocyanine) and nearby molecular oxygen. [11] Unlikeother reactive oxygen species (that is, hydrogen peroxideand superoxide radical),  1 O 2  is not consumed by enzymessuch as catalase and superoxide dismutase produced bycancer cells. [12] Furthermore, PDT is not affected by multipledrug resistance, overcoming the major problems faced inchemo- and radiotherapy. [13] The combination of   1 O 2  withNO therefore in principle represents an ideal strategy for bi-modal treatments. On the basis of these considerations theobjective of this work was to develop a multifunctional mo-lecular ensemble capable of simultaneous photogeneration Abstract:  We have developed a supra-molecular nanoassembly capable of in-ducing remarkable levels of cancer cellmortality through a bimodal actionbased on the simultaneous photogener-ation of nitric oxide (NO) and singletoxygen ( 1 O 2 ). This was achievedthrough the appropriate incorporationof an anionic porphyrin (as  1 O 2  photo-sensitizer) and of a tailored NO photo-donor in different compartments of biocompatible nanoparticles based oncationic amphiphilic cyclodextrins. Thecombination of steady-state and time-resolved spectroscopic techniquesshowed the absence of significant intra-and interchromophoric interaction be-tween the two photoactive centers em-bedded in the nanoparticles, with con-sequent preservation of their photody-namic properties. Photodelivery of NOand  1 O 2  from the nanoassembly onvisible light excitation was unambigu-ously demonstrated by direct and real-time monitoring of these transient spe-cies through amperometric and time-resolved infrared luminescence meas-urements, respectively. The typical redfluorescence of the porphyrin units wasessentially unaffected in the bichromo-phoric nanoassembly, allowing its local-ization in living cells. The convergenceof the dual therapeutic action and theimaging capacities in one single struc-ture makes this supramolecular archi-tecture an appealing, multifunctionalcandidate for applications in biomedi-cal research. Keywords:  cyclodextrins  ·  drugdelivery  ·  imaging agents  ·  nitricoxide  ·  phototherapeutic agents  · singlet oxygen [a] N. Kandoth, Dr. E. Vittorino, Prof. S. SortinoLaboratory of PhotochemistryDepartment of Drug Sciences, Viale Andrea Doria 695125 Catania (Italy)Fax: (   39)095580138E-mail: ssortino@unict.it[b] M. T. Sciortino, T. Parisi, I. ColaoDipartimento di Scienze della VitaSezione di Scienze Microbiologiche Genetiche e MolecolariUniversit di Messina, Salita Sperone, 98166 Messina (Italy)[c] Dr. A. MazzagliaIstituto per lo Studio dei Materiali Nanostrutturati, ISMN-CNRc/o Dipartimento di Chimica Inorganica Analitica e Chimica FisicaUniversit di Messina, Salita Sperone, 98166 Messina (Italy)Fax: (   39)090393756E-mail: antonino.mazzaglia@ismn.cnr.it  2012 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim  Chem. Eur. J.  2012 ,  18 , 1684–1690 1684  of   1 O 2  and NO and offering imaging capacities in livingcells.Nanotechnology offers enormous opportunities for theconstruction of effective delivery vehicles. [14] In the case of photoactivated compounds, three of the most important cri-teria that a carrier system would meet are: 1) biocompatibil-ity, 2) cell-penetrating properties, and 3) preservation of thephotodynamic activity of the photosensitizer.Cyclodextrins (CDs) are water-soluble oligosaccharideswell-known for the formation of host–guest inclusion com-plexes with a range of substrates. [15] Appropriate functionali-zation of the primary, secondary, or both sides of CDs leadsto intriguing derivatives with amphiphilic character, able toself-organize in a variety of assemblies such as micelles, vesi-cles, and nanoparticles [16] with great potential in drug deliv-ery. [17] Previously, we prepared and characterized heterotop-ic colloidal nanoparticles (NPs) based on the cationic am-phiphilic CD  1  (Scheme 1), which is capable of trapping theoppositely charged porphyrin  2 , mainly through electrostaticinteractions. The NPs have hydrodynamic radii ranging from  140 to   1000 nm, depending on the porphyrin loading. [18] We also showed that such NPs are promising vehicles forphotodynamic therapy, because they combine low immuno-genic activity, high capability to convey photosensitizers intotumor cells, and excellent levels of cell mortality upon lightirradiation, due to effective photogeneration of   1 O 2 . [19] In ad-dition, we demonstrated that the porphyrin units are nothosted in the interiors or vicinities of the CD cavities, offer-ing the opportunity to exploit the empty cages in the CD-based NPs for the accommodation of additional guests. [20] Inthe wake of these promising results and motivated by ourongoing interest in developing NO photoreleasing sys-tems, [21] here we report a fluorescent supramolecular nano-assembly capable of photo-deactivating cancer cells througha bimodal action due to the simultaneous generation of   1 O 2 and NO under visible light excitation conditions (Scheme 1). Results and Discussion When CD  1  and the anionic porphyrin  2  are mixed togetherat a molar ratio of approximately 50:1, spontaneous forma-tion of amphiphilic NPs approximately 300 nm in diameteris observed, according to our previous results. [18] Underthese experimental conditions, the porphyrin is mainly en-tangled as the monomeric form, which is responsible for the 1 O 2  generation ( a  and  b  in Figure 1).To introduce an appropriate NO photoreleaser withinthese NPs, we designed and synthesized compound  3 Scheme 1. Idealized view of the photoactive CD-based bichromophoric nanoassembly.Figure 1. Absorption spectra (phosphate buffer, 10 m m , pH 7.4) of   2  a ) inthe absence and  b ) in the presence of CD  1 ,  c  ) of   3  in the presence of CD  1 ,  d  ) of the model compound  4  , and  e  ) of   2   3  in the presence of CD  1 . [ 1 ] = 40  m  m , [ 2 ] = 0.8  m  m , [ 3 ] = 10  m  m ,  T  = 25   C, cell path = 1 cm.The inset shows the size distribution of the bichromophoric nanoassem-bly obtained by dynamic light scattering measurements. Chem. Eur. J.  2012 ,  18 , 1684–1690  2012 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim  www.chemeurj.org  1685 FULL PAPER  (Scheme 1), in which a commercial nitroaniline derivativeand an adamantane appendage are joined together throughan alkyl spacer. We have discovered the nitroaniline deriva-tive to be a suitable NO photodonor [22] because it satisfiesseveral prerequisites for bio-applications. In this compound,the twisted conformation of the nitro group with respect tothe aromatic plane is crucial for the NO photorelease. How-ever, it has been shown that incorporation of this chromo-phore within constrained environments, such as  b -CD cavi-ties [23] or densely packed vesicles, [24] can lead to partial pla-narization of the nitro group with consequent dramaticmodification of the NO photoreleasing properties. With thisin mind, the rationale behind the design of compound  3  wasi) that the adamantane, a well-known and effective guest for b -CD, [25] would be expected to occupy the empty cavities inNPs based on CD  1 , [26] and ii) that use of an alkyl spacer of appropriate length should further encourage the binding of  3  with the NPs through cooperative hydrophobic interchaininteractions involving the alkyl branches of CD  1 . It was en-visaged that these effects should leave the nitroaniline chro-mophore mainly exposed to an aqueous environment, pre-serving the out-of-plane geometry and consequently the NOphotoreleasing capacity. The validity of our design is demon-strated by the absorption spectra shown in Figure 1. Com-pound  3  is insoluble in aqueous solution. On the other hand,it becomes fairly soluble in the presence of the NPs basedon CD  1 , as shown by the appearance of the characteristicabsorption of the nitroaniline chromophore in the visibleregion (spectrum  c  ). These absorption features are basicallythe same as those exhibited by the water-soluble modelcompound  4  in the absence of NPs (spectrum  d  ). Becausethe energy of the visible absorption band is very sensitiveboth to microenvironment polarity and to changes in the ge-ometry of the nitro group, [24] the absence of any significantspectral shift is highly consistent with the photoactive corebeing mainly exposed to an aqueous environment.The bichromophoric nanoassemblies were prepared byloading the CD-based NPs with  2  and  3  at the appropriatemolar ratio. The spectral characteristics observed ( e   inFigure 1) match the profile obtained by summing the spectraof the NPs loaded with the individual chromophores  2  or  3 ( b  and  c   in Figure 1) fairly well, with an experimental uncer-tainty of approximately 15% (in absorbance values). Thisconfirms their concurrent presence within the NPs and theabsence of relevant interactions with one another in theground state. Moreover, the unaltered position of the maxi-mum of the Soret band of   2  indicates that the presence of compound  3  does not induce any rearrangement (i.e., dis-placement/aggregation) of the porphyrins. This might be theconsequence of different affinities of the two componentsfor different binding sites in the NPs, consistently with theidealized view illustrated in Scheme 1.Dynamic light scattering measurements showed the meandiameter of the bichromophoric nanoassemblies to be ap-proximately 300 nm (Figure 1, inset), a value similar to thatalready reported for the NPs loaded with  2  alone at thesame molar ratio. [18] This is consistent with the unchargednature of compound  3 , which does not influence the chargebalance between the anionic  2  and the cationic CD  1 . Sucha balance has been demonstrated to be critical in drivingporphyrin reorganization and promoting the fusion of theamphiphilic NPs into large aggregates of several hundredsof nanometers. [18] The entangling of the monomeric porphyrin within theCD-based NPs results in a significant redshift of its typicaldual band fluorescence emission with negligible changes inthe fluorescence quantum yield (Figure 2A). [27] Interestingly,these emission properties were only slightly affected in thecase of the bichromophoric nanoassembly, ruling out any in-tramolecular quenching (i.e., photoinduced electron-trans-fer) of the excited porphyrin by the NO photodonor. It isalso worth noting that the positions of the two emissionbands were independent of the excitation energy (data notshown), suggesting the presence in the nanoassembly mainly Figure 2. A) Fluorescence emission spectra (  l exc = 440 nm) of   2  a ) in theabsence and  b ) in the presence of CD  1 , and  c  ) of   2   3  in the presence of CD  1 ; the inset shows a representative microscopy image of HeLa cancercells incubated with the bichromophoric nanoassembly for 1 h at 37   C.B) Transient absorption spectra observed upon laser excitation (532 nm)of Ar-saturated CD  1  solution loaded with  2  ( & ) and  2   3  ( * ), recorded0.1  m  s after the laser pulse. Each point was obtained by signal averagingof 10 traces. The inset shows the decay trace monitored at 450 nm.  E  532  12 mJ per pulse. Phosphate buffer (10 m m , pH 7.4), [ 1 ] = 40  m  m , [ 2 ] = 0.8  m  m , [ 3 ] = 10  m  m ,  T  = 25   C. www.chemeurj.org   2012 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim  Chem. Eur. J.  2012 ,  18 , 1684–1690 1686 S. Sortino, A. Mazzaglia et al.  of a single population of fluorophores (i.e., the monomericform). The preservation of the fluorescence properties of the porphyrins under these experimental conditions repre-sents a great advantage for mapping the nanoassembly inliving cells. The inset of Figure 2A shows a representativeoptical image recorded in fluorescence mode after incuba-tion of the bichromophoric NPs with HeLa tumor cells. Thisclearly represents the internalization of the supramolecularsystem in the cell compartment and shows that the localiza-tion takes place mainly at the cytoplasmatic level.The excited triplet state of the porphyrin is the key inter-mediate for the photosensitization of   1 O 2  and its effectivegeneration is thus crucial for the photodynamic action. [11] Figure 2B shows the characteristic triplet absorption of   2 when entangled within the NPs in the absence and in thepresence of the NO photodonor  3 . Because the two sampleswere optically matched at the excitation wavelength, the in-tensity of the transient absorption is directly related to thetriplet quantum yield. The comparable values obtained forthe two samples indicate that the triplet state of the porphy-rin is still populated efficiently in the bichromophoric nano-assembly. Furthermore, the monoexponential decay of thetriplet with a lifetime of approximately 1500  m  s (inset, Fig-ure 2B), which is the same as that observed in the absenceof   3 , [19] indicates that no quenching by the NO photodonoroccurs. Note that the observed lifetime is much longer thanthat of the free porphyrin in aqueous medium in the ab-sence of NPs (ca. 200  m  s), [19] ruling out any exit dynamics of the porphyrin triplet from the nanoassembly on this time-scale.The most convenient method for testing the suitability of the nanoassembly for photogeneration of NO and  1 O 2  is thereal-time monitoring of these transient species. To this end,an ultrasensitive NO electrode was used to detect NO con-centrations by an amperometric technique, whereas time-re-solved infrared luminescence was employed to monitor thetypical phosphorescence of   1 O 2  at 1270 nm. [28] Figure 3shows unambiguous evidence of the light-controlled genera-tion of NO and  1 O 2 . Indeed, we observed the linear photo-generation of NO, which promptly stopped when the lightwas turned off and restarted as the light was turned onagain. It is also noteworthy that the rate of NO release inthe nanoassembly is comparable to that obtained for theNO photodonor within the NPs in the absence of por   phy-   rins, indicating that the porphyrin units do not quench thephotoexcited NO donor. In addition, we observed the char-acteristic infrared luminescence of   1 O 2 , decaying by first-order kinetics with a lifetime of approximately 4  m  s.To validate the feasibility of using the bichromophoricnanoassembly for dual-function phototherapeutic activity,cancer cells were incubated for 1 h under different experi-mental conditions and either kept in the dark or irradiatedwith visible light for 30 min. The results illustrated inFigure 4 show that only irradiation of the photoactive com-ponents in the presence of the CD-based NPs leads to suc-cessful photomortality of the cells, thus confirming the pho-todynamic effects. [29] No significant cell death was detectedin the cells that were incubated in the dark or in the absenceof the photoactive compounds, indicating a good biocompat-ibility of the CD-based NPs. The almost complete photody- Figure 3. A) NO released upon light irradiation (400 nm) of the bichro-mophoric CD-based nanoassembly, and B) representative kinetic trace of  1 O 2  generated upon laser excitation (532 nm). Phosphate buffer (10 m m ,pH 7.4), [ 1 ] = 40  m  m , [ 2 ] = 0.8  m  m , [ 3 ] = 10  m  m ,  T  = 25   C.Figure 4. Cell viabilities of HeLa cells incubated with the NPs based onCD  1  (40  m  m ):  a ) without the photoactive components and loaded with b )  2  (0.8  m  m ),  c   )  3  (10  m  m ), and  d  )  2   3 . Chem. Eur. J.  2012 ,  18 , 1684–1690  2012 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim  www.chemeurj.org  1687 FULL PAPER A Cyclodextrin-Based Nanoassembly with Bimodal Photodynamic Action  namic inactivation induced by the bichromophoric nanoas-sembly—approximately 98%—in relation to the value ob-served with the NPs loaded with the single components  2  or 3  provides clear-cut evidence of the involvement of adouble-action photoinactivation mechanism in the celldeath, in which NO and  1 O 2  are believed to play a key role. Conclusion We have developed a multifunctional photoactive nanoas-sembly that exploits the different affinities of a  1 O 2  photo-sensitizer and a tailor-made NO photodonor towards differ-ent compartments in CD-based NPs. Both guests constituteindependent photoactive centers, as demonstrated by thepreservation of their photophysical and photochemical prop-erties after confinement within the NPs network. We wouldlike to stress that this finding, in contrast with the case of non-photoresponsive compounds, is not obvious. In mostcases, in fact, the photoresponse of single or multiple photo-active units located in a confined space can be considerablyaffected by the occurrence of competitive photoprocesses(e.g., photoinduced energy and/or electron transfer, hydro-gen abstraction, nonradiative deactivation, etc.), [30] whichpreclude the final goal.We have demonstrated that the fluorescence emission of the porphyrin units allows the localization of the nanoas-sembly in living cells and provides the bichromophoricsystem with the ability to generate  1 O 2  and NO effectivelyand concurrently, resulting in an amplified level of cancercell mortality. To the best of our knowledge this is the firstreport in which cancer cellular death due to the combinedaction of these two transient species has been shown. Theuniting of the dual photodynamic action and the imaging ca-pacities in one single nanostructure, together with its bio-compatibility, make this supramolecular architecture an ap-pealing candidate for applications in biomedical research.We finally envision that the extension of our results to a va-riety of   1 O 2  and NO photodispensers chosen ad hoc mightopen fascinating possibilities for novel classes of light-acti-vated, nanoscaled systems in the emerging field of nanome-dicine, for multimodal therapy. Experimental Section Materials : CD  1  and the model compound  4  were synthesized by the pre-viously reported procedures. [16b,31] The tetraanionic porphyrin  2  was pur-chased from Sigma–Aldrich and used as received. All other reagentswere of the highest commercial grade available and used without furtherpurification. All solvents used (Carlo Erba) were analytical grade. Syntheses : The synthesis of [7-(adamantan-1-yloxy)heptyl]-[4-nitro-3-(tri-fluoromethyl)phenyl]amin e  ( 3 ) was carried out in two steps. Syntheseswere carried out at low light intensity levels. 1-(7-Bromoheptyloxy)adamantane (   3a  ) : A THF solution of adamantanol(1 g, 6.56 mmol) was added at 0   C under argon to a suspension of sodium hydride (377 mg, 9.84 mmol) in dry THF (10 mL). After the mix-ture had been left at room temperature for 2 h, 1,7-dibromoheptane(3.92 mL, 22.96 mmol) was added. The reaction mixture was stirred over-night at room temperature and afterwards the excess of NaH wasquenched by addition of water. The mixture was concentrated under re-duced pressure and the aqueous phase was extracted with Et 2 O (310 mL). The organic phases were then collected, washed, dried withNa 2 SO 4 , filtered, concentrated under reduced pressure, and purified bycolumn chromatography (dichloromethane/cyclohexane 30:70) to give  3a (yield 90%).  1 H NMR (CDCl 3 , 500 MHz ):  d = 3.56 (d,  J  = 6.50 Hz, 2H),3.26 (t,  J  = 6.68 Hz, 2H), 1.90–1.84 (m, 3H), 1.72–1.35 ppm (m, 24H). [7-(Adamantan-1-yloxy)heptyl]-[4-nitro-3-(trifluoromethyl)phenyl]amin e  (   3  ) : A mixture of   3a  (1 g, 3.03 mmol) and 4-nitro-3-(trifluoromethyl)ani-line (200 mg, 1.01 mmoli) was heated at reflux in acetonitrile for 5 days.The organic mixture was dried under vacuum and purified by columnchromatography (dichloromethane/cyclohexane 70:30) to afford com-pound  3  as a yellowish powder (yield 60%).  1 H NMR (CDCl 3 ,500 MHz ):  d = 7.95 (d,  J  = 9.2 Hz, 1H), 6.80 (d,  J  = 2.4 Hz, 1H), 6.56 (dd,  J  1 = 9.2 Hz,  J  2 = 2.4 Hz, 1H), 4.47 (broad, 1H), 3.34 (dd,  J  1 = 9.5 Hz,  J  2 = 6.5 Hz, 2H), 3.15 (dd,  J  1 = 9.8 Hz,  J  2 = 6.9 Hz, 2H), 1.83–1.77 (m, 3H),1.63–1.57 (m, 6H) 1.44–1.18 ppm (m, 16H). Instrumentation : UV/Vis absorption and fluorescence spectra were re-corded with a Jasco V-560 spectrophotometer and a Fluorolog-2 (mod. F-111) spectrofluorimeter, respectively. Nanoparticle sizes were measuredwith a dynamic light scattering Horiba LS 550 apparatus fitted with adiode laser (wavelength 650 nm). Fluorescence images were taken with aBiomed fluorescence microscope (Leitz, Wetzlar, Germany). Sample preparation : NPs based on CD  1  were prepared from stock solu-tions (180  m  m ) in CHCl 3 , which were allowed to evaporate slowly to formthin films. The films were hydrated, sonicated for 20 min at 50   C, and al-lowed to equilibrate overnight. An aqueous solution of the porphyrin  2 was then added and each sample was adjusted to a final volume of 2 mLwith phosphate buffer. Compound  3  was dissolved in acetonitrile and al-lowed to evaporate slowly to form a thin film. This film was then hydrat-ed with the colloidal solutions of CD  1  either with or without the porphy-rin  2 . All the final solutions were allowed to equilibrate overnight at 4   C,sonicated for 15 min, and allowed to equilibrate at room temperature for20 min. Laser flash photolysis : All of the samples were excited with the secondharmonic of a Nd-YAG Continuum Surelite II-10 laser (532 nm, 6 nsFWHM), in quartz cells with a path length of 1.0 cm. The excited solu-tions were analyzed with a Luzchem Research mLFP-111 apparatus withan orthogonal pump/probe configuration. The probe source was a ceram-ic xenon lamp coupled to quartz fiber-optic cables. The laser pulse andthe mLFP-111 system were synchronized with a Tektronix TDS 3032 digi-tizer, operating in pre-trigger mode. The signals from a compact Hama-matsu photomultiplier were initially captured by the digitizer and thentransferred to a personal computer, controlled by Luzchem Researchsoftware operating in the National Instruments LabView 5.1 environ-ment. The solutions were deoxygenated by bubbling with a vigorous andconstant flux of pure argon (previously saturated with solvent). In all of these experiments, the solutions were renewed after each laser shot (in aflow cell of 1 cm optical path), to prevent probable autooxidation pro-cesses. The sample temperature was 295  2 K. The energy of the laserpulse was measured at each shot with a SPHD25 Scientech pyroelectricmeter. Singlet oxygen detection : Photogeneration of   1 O 2  upon laser excitation of the photosensitizer was monitored by luminescence measurements inoxygen-saturated solutions. The near-IR luminescence of singlet oxygenat 1.27  m  m (resulting from the forbidden transition  3 S g   ! 1 D g ) was probedorthogonally to the exciting beam with a pre-amplified (low impedance)Ge-photodiode (Hamamatsu EI-P, 300 ns resolution) maintained at  196   C and coupled to a long-pass silicon filter ( > 1.1  m  m) and an inter-ference filter (1.27  m  m). The pure signal of   1 O 2  was obtained as the differ-ence between signals in air- and Ar- saturated solutions. The temporalprofile of the luminescence was fitted to a single-exponential decay func-tion with exclusion of the initial portion of the plot, which was affectedby scattered excitation light, fluorescence, and the formation profile of singlet oxygen itself. www.chemeurj.org   2012 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim  Chem. Eur. J.  2012 ,  18 , 1684–1690 1688 S. Sortino, A. Mazzaglia et al.
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks