Encased Gold Nanoparticle Synthesis as a Probe for Oleuropein Self-Assembled Structure Formation

materials Article Encased Gold Nanoparticle Synthesis as a Probe for Oleuropein Self-Assembled Structure Formation Aila Jimenez-Ruiz 1,*, Rafael Prado-Gotor 1,* , JoséG. Fernández-Bolaños 2, Alejandro González-Benjumea 2 and JoséMaría Carnerero 1   Citation: Jimenez-Ruiz, A.; Prado-Gotor, R.; Fernández-Bolaños, J.G.; González-Benjumea, A.; Carnerero, J.M. Encased Gold Nanoparticle Synthesis as a Probe for Oleuropein Self-Assembled Structure Formation. Materials 2021,14, 50. https://dx.doi.org/10.3390/ ma14010050 Received: 4 November 2020 Accepted: 15 December 2020 Published: 24 December 2020 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/ licenses/by/4.0/). 1Department of Physical Chemistry, University of Seville, c/Profesor García González 1, 41012 Seville, Spain; jcarnerer[email protected] 2Department of Organic Chemistry, University of Seville, c/Profesor García González 1, 41012 Seville, Spain; [email protected] (J.G.F.-B.); [email protected] (A.G.-B.) *Correspondence: [email protected] (A.J.-R.); [email protected] (R.P.-G.) Abstract: Stable oleuropein-coated gold nanoparticles in aqueous media were synthesized for the first time. Oleuropein (OLE) concentration in the reaction medium was found to greatly influence the outcome and stability of the resulting nanocolloid, with a marked decrease in particle size being found for the more concentrated oleuropein solutions. The protection mechanisms involved in the stabilized nanosystems were analyzed. Oleuropein self-assembled structures were found to be formed at a concentration threshold of [OLE] > 5 × 10 −5 M, and observed through the use of CryoSEM imaging. Those structures were responsible for both the increased stability and the decrease in size observed at the more concentrated solutions. Keywords: oleuropein; gold nanoparticles; nanosystem characterization; self-assembly 1. Introduction Oleuropein, a natural phenolic compound (Scheme 1), has been found to be one of the major active components that can be isolated from olive leaves, although it is also present in olives and in the bark of the olive tree. It was described for the first time by Shasha and Leibowitz in 1959 [ 1 ], who later termed it “the bitter principle of olives” [ 2 ] due to its strong taste, which causes it to need partial or total degradation during olive processing [ 3 ]. Oleuropein is a non-toxic compound that can be isolated, among other sources, from olive processing waste [ 4 , 5 ], reducing the amount of residue that need to be eliminated. It has been reported to present a strong antioxidant effect [ 6 – 8 ], as well as antibacterial and antiviral properties [ 9 – 11 ], owing to its role as a natural defense mechanism. Most importantly, it possesses important anticarcinogenic properties that have contributed to placing it in the spotlight [ 12 – 16 ]. As an example, the anti-inflammatory properties of oleuropein and its potential role for the prevention and management of rheumatoid arthritis have been recently described [17]. Materials 2021, 14, x FOR PEER REVIEW 2 of 16 UV–Vis (ultraviolet-visible) absorbance, allows for them to be used as a probe for the presence of suprastructures at high oleuropein concentrations. Scheme 1. Oleuropein chemical structure. Gold nanoparticles (AuNPs) present interesting properties in terms of their use as colorimetric sensors, mainly due to their optical properties—the color of different dispersions depends on the size and shape of the particles. The surrounding medium (characterized by its refractive index) also plays a crucial role in the solution color [22–24]. In this sense, the use of UV–VIS spectra in order to monitor changes on the surface plasmon band (SPB) of non-functionalized gold nanoparticles becomes a subject of experimental interest, allowing us to obtain information about a possible red-shift of the SPB due to an aggregation process of the AuNPs [25]. 2. Materials and Methods 2.1. Materials HAuCl4 (Sigma-Aldrich, Darmstadt, Germany, 520918) and sodium borohydride were of analytical grade and used as received. Oleuropein was extracted using 70:30 (v/v) ethanol–water from olive leaves (Olea europaea L., Picual cultivar), dried at room temperature. After removing the solvent at reduced pressure, we purified the solid extract by silica gel column chromatography using a gradient elution with CH2Cl2–MeOH (10:1 → 5:1) to give oleuropein as a yellowish powder. 1H-NMR spectroscopic data were in agreement with those previously reported [26]. All working solutions were prepared with deionized water from a Milli-Q system, with a conductivity less than 10−6 S m−1. 2.2. Synthesis Procedure Colloidal systems were obtained by direct room temperature reduction of HAuCl4 (at a fixed final concentration of 2.5 × 10−4 M) by sodium borohydride (2.5 × 10−4 M) in the presence of varying concentrations of oleuropein (OLE). In order to carry out the procedure, we placed 20 mL of a 5 × 10−4 M HAuCl4 solution in water at room temperature into a beaker, and the mixture was then magnetically stirred. Then, in rapid succession, 20 mL of OLE at double the final concentration on each case, and 0.1 mL of sodium borohydride (0.1 M) were added. Since OLE was shown to be able to start the reduction procedure by itself, the addition of sodium borohydride after reactant mixing needed to be fast in order to minimize reduction of the gold salts by OLE. Solutions turned purple to dark red, depending on OLE concentration, after mixing. The color slowly turned brown to light red in the following hours. In all cases, stirring was maintained for at least 24 h to allow for the total reduction of the gold salts. Aliquots of some of the as-prepared colloids were dialyzed on a cellulose ester dialysis membrane (MWCO = 10000) overnight. The resulting samples were labelled as S1 to S7, and the final [OLE] in the reaction media can be found in Table 1. For control purposes, as borohydride by itself was also found to be able to stabilize the gold colloids, we obtained an additional synthesis in the absence of OLE, which was labelled as S0. Scheme 1. Oleuropein chemical structure. Despite its chemical importance, the physical properties of oleuropein have received comparatively little attention. In the fields of nanotechnology and metal conjugation, although some reports exist of gold and silver [ 18 , 19 ] as well as non-noble metal [ 20 , 21 ] Materials 2021,14, 50. https://dx.doi.org/10.3390/ma14010050 https://www.mdpi.com/journal/materials Materials 2021,14, 50 2 of 16 nanoparticle synthesis being carried out by using olive leaf extracts, no data have been reported of any colloidal synthesis using pure oleuropein as the main stabilizing agent. In the present work, we carried out a series of gold nanoparticle syntheses in the presence of varying concentrations of oleuropein. Its influence in the outcome of the procedure was analyzed, and the colloids were characterized in-depth. Results showed that some degree of oleuropein self-assembly is to be expected at the highest concentrations, and this fact dramatically affects nanoparticle morphology and stability. The analysis of those parameters for in situ synthesized nanoparticles, via simple techniques such as UV–Vis (ultraviolet-visible) absorbance, allows for them to be used as a probe for the presence of suprastructures at high oleuropein concentrations. Gold nanoparticles (AuNPs) present interesting properties in terms of their use as colorimetric sensors, mainly due to their optical properties—the color of different dispersions depends on the size and shape of the particles. The surrounding medium (characterized by its refractive index) also plays a crucial role in the solution color [ 22 – 24 ]. In this sense, the use of UV–VIS spectra in order to monitor changes on the surface plasmon band (SPB) of non-functionalized gold nanoparticles becomes a subject of experimental interest, allowing us to obtain information about a possible red-shift of the SPB due to an aggregation process of the AuNPs [25]. 2. Materials and Methods 2.1. Materials HAuCl 4 (Sigma-Aldrich, Darmstadt, Germany, 520918) and sodium borohydride were of analytical grade and used as received. Oleuropein was extracted using 70:30 (v/v) ethanol–water from olive leaves (Olea europaea L., Picual cultivar), dried at room temperature. After removing the solvent at reduced pressure, we purified the solid extract by silica gel column chromatography using a gradient elution with CH 2 Cl 2 –MeOH (10:1 → 5:1) to give oleuropein as a yellowish powder. 1 H-NMR spectroscopic data were in agreement with those previously reported [ 26 ]. All working solutions were prepared with deionized water from a Milli-Q system, with a conductivity less than 10−6S m−1. 2.2. Synthesis Procedure Colloidal systems were obtained by direct room temperature reduction of HAuCl 4 (at a fixed final concentration of 2.5 × 10 −4 M) by sodium borohydride (2.5 × 10 −4 M) in the presence of varying concentrations of oleuropein (OLE). In order to carry out the procedure, we placed 20 mL of a 5 × 10 −4 M HAuCl 4 solution in water at room temperature into a beaker, and the mixture was then magnetically stirred. Then, in rapid succession, 20 mL of OLE at double the final concentration on each case, and 0.1 mL of sodium borohydride (0.1 M) were added. Since OLE was shown to be able to start the reduction procedure by itself, the addition of sodium borohydride after reactant mixing needed to be fast in order to minimize reduction of the gold salts by OLE. Solutions turned purple to dark red, depending on OLE concentration, after mixing. The color slowly turned brown to light red in the following hours. In all cases, stirring was maintained for at least 24 h to allow for the total reduction of the gold salts. Aliquots of some of the as-prepared colloids were dialyzed on a cellulose ester dialysis membrane (MWCO = 10000) overnight. The resulting samples were labelled as S1 to S7, and the final [OLE] in the reaction media can be found in Table 1. For control purposes, as borohydride by itself was also found to be able to stabilize the gold colloids, we obtained an additional synthesis in the absence of OLE, which was labelled as S0. Materials 2021,14, 50 3 of 16 Table 1. Sample nomenclature and oleuropein concentration. Samples [OLE]/M S0 0 S1 5×10−7 S2 10−6 S3 5×10−6 S4 10−5 S5 5×10−5 S6 10−4 S7 3.3 ×10−4 2.3. Absorbance Measurements UV–VIS spectra were obtained in a Cary-500 spectrophotometer (Agilent, Santa Clara, CA 95051 USA), with a wavelength accuracy of ± 0.3 nm and a spectral bandwidth of 0.5 nm. For sample characterization, spectra were obtained in the visible region of the spectrum, from 400 to 800 nm, and before and after dialysis, we registered spectra between 200 and 800 nm in order to monitor the UV absorption bands of the oleuropein. In all cases, temperature was kept at 298.3 K and the solvent was ultrapure Milli-Q water, with a conductivity less than 10−6S m−1. RGB Color Conversion XYZ color space measurements were derived from transmittance values according to the CIE (International Commission on Illumination) standards for a D65 illuminant and 2 º standard observer [ 27 ]. Equations for RGB color conversion from the XYZ data were obtained from https://www.easyrgb.com/en/math.php and implemented as described therein. 2.4. DLS and Zeta Potential Measurements Both dynamic light scattering (DLS) measurements of colloids and oleuropein solutions and zeta potentials of the as-synthesized and dialyzed colloids were registered in a Zetasizer Nano ZS-90 (Malvern Panalytical, Spectris plc, Surrey, England) spectrophotometer equipped with a red laser (633 nm). A DTS1060 polycarbonate capillary cell was employed for all measurements. For DLS measurements, the detection was carried out at a fixed angle of 173 ◦ from the light source, and the fluctuations in scattered light intensity were analyzed. Zeta potentials were derived from the electrophoretic mobility of the sample particles. In both cases, calculations were made with the help of the Zetasizer (Malvern Panalytical, Spectris plc, Surrey, England) software. 2.5. TEM Measurements Transmission electron microscopy (TEM) images of the as-synthesized colloids were obtained in a Philips CM200 (Philips Ibérica, S.A., Madrid, Spain) microscope working at 200 kV. Sample preparation was carried out by deposition of a single drop of nanoparticle suspension in a carbon-coated copper grid, which was then left to dry overnight at room temperature. 2.6. CryoSEM Measurements In order to obtain cryo-scanning electron microscopy (CrioSEM) images of oleuropein solutions, we carried out sample freezing out by quick immersion in liquid nitrogen (77 K), followed by an etching process in which temperature was raised in 5 K/min steps to a final temperature of 183 K, in order to minimize ice crystal formation. A Leica EM VCT100 (Leica Camera Iberia, S.L., Madrid, Spain) contamination-free cryo transfer system was employed in order to keep the sample cooled during the whole measurement process. A 9 nm gold coating was deposited through sputtering before imaging by using a Leica ACE600 (Leica Camera Iberia, S.L., Madrid, Spain) sputter coater. Images were obtained in Materials 2021,14, 50 4 of 16 a Zeiss EVO L315 (Carl Zeiss Iberia, S.L, 28760 Tres Cantos, Madrid, Spain) SEM microscope at acceleration voltages of 5 kV and 8 kV. 3. Results A photograph and a digital simulation (obtained from the conversion of their absorption spectra into RGB—see the Supplementary Materials) of the final color of all colloids resulting from the experimental procedure described is shown in Figure 1. From simple color observation from the naked eye, we could establish a clear division between the syntheses that were carried out at [OLE] > 5 × 10 −5 M (S7 to S5) and those that contained lower concentrations (S4 to S1). It is important to note that, although oleuropein possessed some reducing power and was able to start gold salt reduction by itself, the resulting colloids were neither stable nor concentrated enough for analysis. The slow reduction speed observed when pure oleuropein was employed as the sole reducing agent resulted in uncontrolled particle growth (likely due to a low rate of nucleation in relation to particle growth) that caused the suspensions to turn unstable and precipitate after a short period of time. To avoid this, sodium borohydride was employed as a reducing agent. NaBH 4 is a fast-reducing agent commonly employed for nanoparticle growth in the presence of organic capping agents, from biopolymers to proteins [ 28 – 31 ]. The fast reaction time linked to borohydride reactions ensured the formation of small and monodisperse particles. However, there was also a possibility of borohydride subproducts interfering with the result, as they possessed some residual stabilizing effect (as seen for synthesis S0) and could adsorb on the surface of the synthesized nanoparticles, which conferred them a negative charge and stabilized the colloid. This was an undesirable effect, and to minimize it, residual borohydride by-products should be present as minimally as possible in the final reaction product. Materials 2021, 14, x FOR PEER REVIEW 4 of 16 VCT100 (Leica Camera Iberia, S.L., Madrid, Spain) contamination-free cryo transfer system was employed in order to keep the sample cooled during the whole measurement process. A 9 nm gold coating was deposited through sputtering before imaging by using a Leica ACE600 (Leica Camera Iberia, S.L., Madrid, Spain) sputter coater. Images were obtained in a Zeiss EVO L315 (Carl Zeiss Iberia, S.L, 28760 Tres Cantos, Madrid, Spain) SEM microscope at acceleration voltages of 5 kV and 8 kV. 3. Results A photograph and a digital simulation (obtained from the conversion of their absorption spectra into RGB—see the Supplementary Materials) of the final color of all colloids resulting from the experimental procedure described is shown in Figure 1. From simple color observation from the naked eye, we could establish a clear division between the syntheses that were carried out at [OLE] > 5 × 10−5M (S7 to S5) and those that contained lower concentrations (S4 to S1). It is important to note that, although oleuropein possessed some reducing power and was able to start gold salt reduction by itself, the resulting colloids were neither stable nor concentrated enough for analysis. The slow reduction speed observed when pure oleuropein was employed as the sole reducing agent resulted in uncontrolled particle growth (likely due to a low rate of nucleation in relation to particle growth) that caused the suspensions to turn unstable and precipitate after a short period of time. To avoid this, sodium borohydride was employed as a reducing agent. NaBH4is a fast-reducing agent commonly employed for nanoparticle growth in the presence of organic capping agents, from biopolymers to proteins [28–31]. The fast reaction time linked to borohydride reactions ensured the formation of small and monodisperse particles. However, there was also a possibility of borohydride subproducts interfering with the result, as they possessed some residual stabilizing effect (as seen for synthesis S0) and could adsorb on the surface of the synthesized nanoparticles, which conferred them a negative charge and stabilized the colloid. This was an undesirable effect, and to minimize it, residual borohydride by-products should be present as minimally as possible in the final reaction product. The observed color differences between samples were clearly reflected in their respective absorbance spectra (see Figure 2). S7-5 plasmon bands were wider and slightly blue-shifted compared to that of the control sample S0, while S4-1 presented a more defined, narrower absorbance band whose peak intensity wavelength matched that of sample S0. Figure 1. Photograph (upper) and RGB color simulation derived from absorbance spectra (lower) of all gold colloids obtained. Figure 1. Photograph (upper) and RGB color simulation derived from absorbance spectra (lower) of all gold colloids obtained. The observed color differences between samples were clearly reflected in their respective absorbance spectra (see Figure 2). S7-5 plasmon bands were wider and slightly blue-shifted compared to that of the control sample S0, while S4-1 presented a more defined, narrower absorbance band whose peak intensity wavelength matched that of sample S0. Materials 2021,14, 50 5 of 16 Materials 2021, 14, x FOR PEER REVIEW 5 of 16 λ/nm 400 450 500 550 600 650 700 750 800 Absorbance intensity 0.00 0.02 0.04 0.06 0.08 0.10 S7 S6 S5 S4 S3 S2 S1 S0 Figure 2. Normalized absorbance spectra, taken between 24 h and 96 h after synthesis, for 1:10 dilutions of samples S7 to S1 and S0. Vertical line marks the position for the absorption maxima of sample S0 at 531 nm. The as-prepared samples were stored at room temperature and protected from light. Their stability was then monitored via absorbance spectroscopy over a four-week span time (see Figure 3). Colloid stability was found to be good, with most of them being stable (no precipitation and no notable aggregation phenomena) over a 15- or even 30-day period. Once a nanocolloid was found to be stable for a measurable amount of time, the next step was to ascertain the exact character of the protective coating that stabilized it. Generally speaking, stabilization can be achieved through adsorption of either a charged or a neutral but voluminous ligand over the particle surface; the former procedure gives rise to positively or negatively charged nanoparticles, while the latter produces neutral colloids. A way to test for the type of stabilization of a finished colloid is to add moderate concentrations of a neutral electrolyte, such as NaCl. Electrostatically protected nanoparticles will become neutralized by the added ions and aggregate, quite often changing color in the process. Sterically protected colloids will be unaffected. On the other hand, dialysis is one of the most common ways of eliminating undesired by-products—in this case, borohydride derivatives. This process breaks the ion exchange effect that stabilizes electrostatically protected colloids (such as commonly used citrate-capped nanoparticles) and, if they are the only source of stabilization, will result in precipitation of the colloid. Oleuropein stabilization, coming from a bulky, uncharged ligand, is expected to be due to steric effects, and therefore unaffected by dialysis. In order to ascertain both the nature of the nanoparticle-capping agent interaction and the ionic character of the colloids, we designed two experiments. First, samples S6, S5, and S0 were dialyzed in Milli-Q ultrapure water over a 72 h period; then, the stability of all samples to 0.150 M NaCl was tested. Figure 2. Normalized absorbance spectra, taken between 24 h and 96 h after synthesis, for 1:10 dilutions of samples S7 to S1 and S0. Vertical line marks the position for the absorption maxima of sample S0 at 531 nm. The as-prepared samples were stored at room temperature and protected from light. Their stability was then monitored via absorbance spectroscopy over a four-week span time (see Figure 3). Colloid stability was found to be good, with most of them being stable (no precipitation and no notable aggregation phenomena) over a 15- or even 30-day period. Materials 2021, 14, x FOR PEER REVIEW 6 of 16 λ/nm 400 450 500 550 600 650 700 750 800 Absorbance 0.00 0.02 0.04 0.06 0.08 0.10 S0 initial S0 15 days S0 30 days λ/nm 400 450 500 550 600 650 700 750 800 Absorbance 0.00 0.01 0.02 0.03 0.04 S2 initial S2 15 days S2 30 days λ/nm 400 450 500 550 600 650 700 750 800 Absorbance 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 S4 initial S4 15 days S4 30 days λ/nm 400 450 500 550 600 650 700 750 800 Absorbance 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 S5 initial S5 15 days S5 30 days Figure 3. Normalized absorption spectra over time of 1:10 dilutions of samples (a) S0, (b) S2, (c) S4, and (d) S5. As seen in Figure 4, both S5 and S6 were dialyzable with a minimum loss of stability, while S0 precipitated upon dialysis. Therefore, we can infer that those two syntheses presented a markedly different character to that of pure borohydride nanoparticles, which can be attributed to the oleuropein stabilization, making them desirable for further study. λ/nm 200 300 400 500 600 700 800 Absorbance 0.0 0.1 0.2 0.3 0.4 S5 S5 (dialyzed) S6 S6 (dialyzed) Dialysis water Figure 4. Before and after dialysis spectra for samples S5 and S6. NaCl acts as a good agglomerant agent for those colloids whose protection is charge-based, while having no effect over steric-protected colloids. In this way, testing for NaCl aggregation helps determine the nature of the protective coating. For gold nanoparticles, this aggregation visually manifests in the shape of a red-shifted wide absorption band (corresponding to the anisotropic multi-particle aggregates) and a shift (a) (b) (c) (d) Figure 3. Normalized absorption spectra over time of 1:10 dilutions of samples (a) S0, (b) S2, (c) S4, and (d) S5. Once a nanocolloid was found to be stable for a measurable amount of time, the next step was to ascertain the exact character of the protective coating that stabilized it. Generally speaking, stabilization can be achieved through adsorption of either a charged Materials 2021,14, 50 6 of 16 or a neutral but voluminous ligand over the particle surface; the former procedure gives rise to positively or negatively charged nanoparticles, while the latter produces neutral colloids. A way to test for the type of stabilization of a finished colloid is to add moderate concentrations of a neutral electrolyte, such as NaCl. Electrostatically protected nanoparticles will become neutralized by the added ions and aggregate, quite often changing color in the process. Sterically protected colloids will be unaffected. On the other hand, dialysis is one of the most common ways of eliminating undesired by-products—in this case, borohydride derivatives. This process breaks the ion exchange effect that stabilizes electrostatically protected colloids (such as commonly used citrate-capped nanoparticles) and, if they are the only source of stabilization, will result in precipitation of the colloid. Oleuropein stabilization, coming from a bulky, uncharged ligand, is expected to be due to steric effects, and therefore unaffected by dialysis. In order to ascertain both the nature of the nanoparticle-capping agent interaction and the ionic character of the colloids, we designed two experiments. First, samples S6, S5, and S0 were dialyzed in Milli-Q ultrapure water over a 72 h period; then, the stability of all samples to 0.150 M NaCl was tested. As seen in Figure 4, both S5 and S6 were dialyzable with a minimum loss of stability, while S0 precipitated upon dialysis. Therefore, we can infer that those two syntheses presented a markedly different character to that of pure borohydride nanoparticles, which can be attributed to the oleuropein stabilization, making them desirable for further study. Materials 2021, 14, x FOR PEER REVIEW 6 of 16 λ/nm 400 450 500 550 600 650 700 750 800 Absorbance 0.00 0.02 0.04 0.06 0.08 0.10 S0 initial S0 15 days S0 30 days λ/nm 400 450 500 550 600 650 700 750 800 Absorbance 0.00 0.01 0.02 0.03 0.04 S2 initial S2 15 days S2 30 days λ/nm 400 450 500 550 600 650 700 750 800 Absorbance 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 S4 initial S4 15 days S4 30 days λ/nm 400 450 500 550 600 650 700 750 800 Absorbance 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 S5 initial S5 15 days S5 30 days Figure 3. Normalized absorption spectra over time of 1:10 dilutions of samples (a) S0, (b) S2, (c) S4, and (d) S5. As seen in Figure 4, both S5 and S6 were dialyzable with a minimum loss of stability, while S0 precipitated upon dialysis. Therefore, we can infer that those two syntheses presented a markedly different character to that of pure borohydride nanoparticles, which can be attributed to the oleuropein stabilization, making them desirable for further study. λ/nm 200 300 400 500 600 700 800 Absorbance 0.0 0.1 0.2 0.3 0.4 S5 S5 (dialyzed) S6 S6 (dialyzed) Dialysis water Figure 4. Before and after dialysis spectra for samples S5 and S6. NaCl acts as a good agglomerant agent for those colloids whose protection is charge-based, while having no effect over steric-protected colloids. In this way, testing for NaCl aggregation helps determine the nature of the protective coating. For gold nanoparticles, this aggregation visually manifests in the shape of a red-shifted wide absorption band (corresponding to the anisotropic multi-particle aggregates) and a shift (a) (b) (c) (d) Figure 4. Before and after dialysis spectra for samples S5 and S6. NaCl acts as a good agglomerant agent for those colloids whose protection is chargebased, while having no effect over steric-protected colloids. In this way, testing for NaCl aggregation helps determine the nature of the protective coating. For gold nanoparticles, this aggregation visually manifests in the shape of a red-shifted wide absorption band (corresponding to the anisotropic multi-particle aggregates) and a shift from red to a purple or blue tint in the nanoparticle suspension; suspensions can also turn colorless when the particle size becomes big enough for precipitation. Figure 5shows the results of the NaCl test as a RGB color simulation, and for illustrative purposes, the original absorption spectra of OLE-protected S6 (before and after dialysis) and borohydride-protected S3 are also shown. Notice the blue tint appearing for S0-S4 in the presence of NaCl, as well as the wide band over 750 nm on the S3+NaCl spectrum. As reflected in Figure 5, samples obtained in the presence of [OLE] < 5 × 10 −5 M were markedly unstable to NaCl addition, as was the blank. Almost complete precipitation of the system occurred, with the formation of an aggregation band over 750 nm and a marked color change from red to blue; the color intensity of the latter was observed to decrease with lower [OLE] as the aggregation band shifted to higher wavelengths in the infrared region of the spectrum. Undialyzed samples containing [OLE] > 5 × 10 −5 M presented aggregation with a shoulder centered around 650 nm, and their color remained close to that of the original sample, although a slight darkening was observed. As expected, S5 to Materials 2021,14, 50 7 of 16 S7 remained stable in the presence of NaCl, with this stabilization being even clearer for dialyzed samples. Materials 2021, 14, x FOR PEER REVIEW 7 of 16 from red to a purple or blue tint in the nanoparticle suspension; suspensions can also turn colorless when the particle size becomes big enough for precipitation. Figure 5 shows the results of the NaCl test as a RGB color simulation, and for illustrative purposes, the original absorption spectra of OLE-protected S6 (before and after dialysis) and borohydride-protected S3 are also shown. Notice the blue tint appearing for S0-S4 in the presence of NaCl, as well as the wide band over 750 nm on the S3+NaCl spectrum. λ/nm 400 450 500 550 600 650 700 750 800 Absorbance 0.00 0.01 0.02 0.03 0.04 0.05 0.06 S6 + NaCl S6 S6 (dialyzed) + NaCl S6 (dialyzed) λ/nm 400 450 500 550 600 650 700 750 800 Absorbance 0.00 0.01 0.02 0.03 0.04 0.05 0.06 S3 + NaCl S3 (a) (b) (c) Figure 5. Normalized absorption spectra for 1:10 dilutions of samples S6 (a) and S3 (b) and RGB color simulation for all samples (c); upper row corresponds to diluted samples before NaCl addition and lower row to the same samples after an incubation period of 10 min with NaCl. Color intensity was enhanced for better perception. As reflected in Figure 5, samples obtained in the presence of [OLE] < 5 × 10−5 M were markedly unstable to NaCl addition, as was the blank. Almost complete precipitation of the system occurred, with the formation of an aggregation band over 750 nm and a marked color change from red to blue; the color intensity of the latter was observed to decrease with lower [OLE] as the aggregation band shifted to higher wavelengths in the infrared region of the spectrum. Undialyzed samples containing [OLE] > 5 × 10−5 M presented aggregation with a shoulder centered around 650 nm, and their color remained close to that of the original sample, although a slight darkening was observed. As expected, S5 to S7 remained stable in the presence of NaCl, with this stabilization being even clearer for dialyzed samples. Surface zeta potential measurements were carried out for samples S6 and S5 (both dialyzed and undyalized), and S3 and S0. The results are shown in Table 2. As can be observed, no significant difference was found between surface potential of the higher [OLE] samples (S6, S5) and the one obtained in absence of oleuropein (S0), while the sample containing the lower [OLE] (S3) showed a shift towards a more neutral mean Figure 5. Normalized absorption spectra for 1:10 dilutions of samples S6 ( a ) and S3 ( b ) and RGB color simulation for all samples ( c ); upper row corresponds to diluted samples before NaCl addition and lower row to the same samples after an incubation period of 10 min with NaCl. Color intensity was enhanced for better perception. Surface zeta potential measurements were carried out for samples S6 and S5 (both dialyzed and undyalized), and S3 and S0. The results are shown in Table 2. As can be observed, no significant difference was found between surface potential of the higher [OLE] samples (S6, S5) and the one obtained in absence of oleuropein (S0), while the sample containing the lower [OLE] (S3) showed a shift towards a more neutral mean value, although it remained in the negative zone. Despite a clear diminution of the salt concentration upon dialysis, the procedure did not significantly alter the surface potential of the sample. This was a notable finding, highlighting the fact that even oleuropein-protected nanoparticles retain a slight anionic character even after dialysis. The enhanced stability to ionic aggregation of those colloids can be attributed to them not being reliant on borohydride absorption–desorption equilibria for stability, even though they are negatively charged. Materials 2021,14, 50 8 of 16 Table 2. Mean zeta potential values for all analyzed samples. Samples Zeta Potential (mV) S6 −28.8 S6 (dialyzed) −29.6 S5 −34.6 S5 (dialyzed) −29.8 S3 −20.6 S0 −30.7 3.1. TEM Imaging For the final step of characterization, and in order to obtain accurate information about particle size and shape, we obtained TEM images for samples S7 to S3 and S0 ( Figures 6and 7 ). TEM particle analysis by using ImageJ software also allowed us to determine mean particle size for all cases (Table 3). Again, clear differences in size and morphology can be appreciated between those samples with oleuropein concentrations over and under 5 × 10 −5 M. Nanoparticles obtained in the absence of oleuropein had a mean diameter of 32.8 nm, and showed a small degree of sphericity with the presence of clear multi-faceted shapes. The same situation was observed for those particles obtained in the presence of [OLE] = 5 × 10 −6 M and 10 −5 M (S3 and S4), which showed a similar morphology and slightly bigger diameters (35.0 nm in both cases), the latter having a markedly greater polydispersity. S4 was also found to be the less stable of all colloids obtained (see Figure 3). Materials 2021, 14, x FOR PEER REVIEW 9 of 17 (a) (b) (c) (d) (e) (f) Figure 6. From up to down and left to right: TEM images of samples (a) S7, (b) S6, (c) S5, (d) S4, (e) S3, and (f) S0. All images were taken at the same magnification. Scale bar corresponds to 100 nm. For concentrations of oleuropein of and over 5 × 10−5 M, mean particle diameter decreased dramatically and clear shape changes were observed (Figure 7). S5 nanoparticles (with [OLE] = 5 × 10−5 M) were mostly spherical, with a medium diameter of 5.6 nm, no clear faceted growth, and a small population of bigger particles (around 14 nm), alongside elongated and triangular shapes (with a mean 9.9 nm side length). S6 ([OLE] = 10−4 M) nanoparticles were slightly smaller (4.6 nm) and did not show any non-spherical shapes, although some coalesced or twinned particles were observed. Polydispersity sharply decreased from that of the more dilute oleuropein solutions. Finally, for synthesis S7 ([OLE] = 3.3 × 10−4 M), the sample presented a mean diameter and polydispersity similar to that of S6, but a sizable population of multi-twinned particles was also found alongside single spherical ones. Polydispersity was found to be around 20% for all samples, except those closer to the size transition zone, S4 (40%) and S5 (36%). (a) (c) (e) (b) (a) Figure 6. Cont. Materials 2021,14, 50 9 of 16 Materials 2021, 14, x FOR PEER REVIEW 9 of 17 (a) (b) (c) (d) (e) (f) Figure 6. From up to down and left to right: TEM images of samples (a) S7, (b) S6, (c) S5, (d) S4, (e) S3, and (f) S0. All images were taken at the same magnification. Scale bar corresponds to 100 nm. For concentrations of oleuropein of and over 5 × 10−5 M, mean particle diameter decreased dramatically and clear shape changes were observed (Figure 7). S5 nanoparticles (with [OLE] = 5 × 10−5 M) were mostly spherical, with a medium diameter of 5.6 nm, no clear faceted growth, and a small population of bigger particles (around 14 nm), alongside elongated and triangular shapes (with a mean 9.9 nm side length). S6 ([OLE] = 10−4 M) nanoparticles were slightly smaller (4.6 nm) and did not show any non-spherical shapes, although some coalesced or twinned particles were observed. Polydispersity sharply decreased from that of the more dilute oleuropein solutions. Finally, for synthesis S7 ([OLE] = 3.3 × 10−4 M), the sample presented a mean diameter and polydispersity similar to that of S6, but a sizable population of multi-twinned particles was also found alongside single spherical ones. Polydispersity was found to be around 20% for all samples, except those closer to the size transition zone, S4 (40%) and S5 (36%). (a) (c) (e) (b) (a) Figure 6. From up to down and left to right: TEM images of samples ( a ) S7, ( b ) S6, ( c ) S5, ( d ) S4, ( e ) S3, and ( f ) S0. All images were taken at the same magnification. Scale bar corresponds to 100 nm. Materials 2021, 14, x FOR PEER REVIEW 9 of 16 (c) (d) (e) (f) Figure 6. From up to down and left to right: TEM images of samples (a) S7, (b) S6, (c) S5, (d) S4, (e) S3, and (f) S0. All images were taken at the same magnification. Scale bar corresponds to 100 nm. For concentrations of oleuropein of and over 5 × 10−5 M, mean particle diameter decreased dramatically and clear shape changes were observed (Figure 7). S5 nanoparticles (with [OLE] = 5 × 10−5 M) were mostly spherical, with a medium diameter of 5.6 nm, no clear faceted growth, and a small population of bigger particles (around 14 nm), alongside elongated and triangular shapes (with a mean 9.9 nm side length). S6 ([OLE] = 10−4 M) nanoparticles were slightly smaller (4.6 nm) and did not show any non-spherical shapes, although some coalesced or twinned particles were observed. Polydispersity sharply decreased from that of the more dilute oleuropein solutions. Finally, for synthesis S7 ([OLE] = 3.3 × 10−4 M), the sample presented a mean diameter and polydispersity similar to that of S6, but a sizable population of multi-twinned particles was also found alongside single spherical ones. Polydispersity was found to be around 20% for all samples, except those closer to the size transition zone, S4 (40%) and S5 (36%). (a) (b) (c) (c) (b) (a) Figure 7. Higher magnification images of samples ( a ) S7, ( b ) S6, and ( c ) S5, showing morphology details. Scale bar corresponds to 20 nm in all cases. Table 3. Mean particle size of samples S0 and S3 to S7. Samples [OLE]/M Mean Diameter/nm Polydispersity S0 0 32.8 ±7.3 22% S3 5×10−635.0 ±6.6 19% S4 10−535.0 ±14.1 40% S5 5×10−55.6 ±2.0 36% S6 10−44.6 ±1.0 22% S7 3.3 ×10−44.8 ±0.9 18% For concentrations of oleuropein of and over 5 × 10 −5 M, mean particle diameter decreased dramatically and clear shape changes were observed (Figure 7). S5 nanoparticles Materials 2021,14, 50 16 of 16 40. Bravo-Anaya, L.M.; Soltero, J.F.A.; Rinaudo, M. DNA/chitosan electrostatic complex. Int. J. Biol. Macromol. 2016 ,88, 345–353. [CrossRef] 41. Huo, Y.; Qi, L.; Lv, X.-J.; Lai, T.; Zhang, J.; Zhang, Z.-Q. A sensitive aptasensor for colorimetric detection of adenosine triphosphate based on the protective effect of ATP-aptamer complexes on unmodified gold nanoparticles. Biosens. Bioelectron. 2016 ,78, 315–320. [CrossRef] 42. Guerra, J.; Rodrigo, A.C.; Merino, S.; Tejeda, J.; García-Martínez, J.C.; Sánchez-Verdú, P.; Ceña, V.; Rodríguez-López, J. PPVPAMAM hybrid dendrimers: Self-assembly and stabilization of gold nanoparticles. Macromolecules 2013 ,46, 7316–7324. [CrossRef] 43. Zhang, L.; Sun, X.; Song, Y.; Jiang, X.; Dong, S.; Wang, E. Didodecyldimethylammonium bromide lipid bilayer-protected gold nanoparticles: Synthesis, characterization, and self-assembly. Langmuir 2006,22, 2838–2843. [CrossRef] [PubMed] 44. Yang, T.; Li, Z.; Wang, L.; Guo, C.; Sun, Y. Synthesis, characterization, and self-assembly of protein lysozyme monolayer-stabilized gold nanoparticles. Langmuir 2007,23, 10533–10538. [CrossRef] [PubMed] 45. Kazakova, J.; García-Povea, A.; Fernández-Palacios, M.; Villar-Navarro, M.; Carnerero, J.M.; Jimenez-Ruiz, A.; Prado-Gotor, R. A colorimetric study of the interaction of cationic and anionic surfactants with anionic gold nanoparticles. Colloid Polym. Sci. 2017 . [CrossRef] 46. Mandal, T.; Dasgupta, C.; Maiti, P.K. Engineering gold nanoparticle interaction by PAMAM dendrimer. J. Phys. Chem. C 2013 ,117, 13627–13636. [CrossRef] 47. Regev, O.; Backov, R.; Faure, C. Gold nanoparticles spontaneously generated in onion-type multilamellar vesicles. Bilayers-particle coupling imaged by Cryo-TEM. Chem. Mater. 2004,16, 5280–5285. [CrossRef] 48. Studer, D.; Graber, W.; Al-Amoudi, A.; Eggli, P. A new approach for cryofixation by high-pressure freezing. J. Microsc. 2001 ,203, 285–294. [CrossRef] 49. Umrath, W. Cooling bath for rapid freezing in electron microscopy. J. Microsc. 1974,101, 103–105. [CrossRef] 50. Rubino, S.; Akhtar, S.; Melin, P.; Searle, A.; Spellward, P.; Leifer, K. A site-specific focused-ion-beam lift-out method for cryo Transmission Electron Microscopy. J. Struct. Biol. 2012,180, 572–576. [CrossRef] 51. Walther, P. Recent progress in freeze-fracturing of high-pressure frozen samples. J. Microsc. 2003,212, 34–43. [CrossRef]