Edited by: Alessandro Martucci, University of Padua, Italy
Reviewed by: Joshua A. Jackman, Sungkyunkwan University, South Korea; Rodolfo Ippoliti, University of L’Aquila, Italy
†These authors have contributed equally to this work
‡Present address: Giuseppe Trapani, Bioactive Materials Laboratory, Max Planck Institute for Molecular Biomedicine, Münster, Germany
This article was submitted to Nanobiotechnology, a section of the journal Frontiers in Bioengineering and Biotechnology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Engineered graphene-based derivatives are attractive and promising candidates for nanomedicine applications because of their versatility as 2D nanomaterials. However, the safe application of these materials needs to solve the still unanswered issue of graphene nanotoxicity. In this work, we investigated the self-assembly of dityrosine peptides driven by graphene oxide (GO) and/or copper ions in the comparison with the more hydrophobic diphenylalanine dipeptide. To scrutinize the peptide aggregation process, in the absence or presence of GO and/or Cu2+, we used atomic force microscopy, circular dichroism, UV–visible, fluorescence and electron paramagnetic resonance spectroscopies. The perturbative effect by the hybrid nanomaterials made of peptide-decorated GO nanosheets on model cell membranes of supported lipid bilayers was investigated. In particular, quartz crystal microbalance with dissipation monitoring and fluorescence recovery after photobleaching techniques were used to track the changes in the viscoelastic properties and fluidity of the cell membrane, respectively. Also, cellular experiments with two model tumour cell lines at a short time of incubation, evidenced the high potential of this approach to set up versatile nanoplatforms for nanomedicine and theranostic applications.
In the past decade, the use of engineered carbon-based nanomaterials for technological applications has attracted much interest in the field of nanomedicine. In particular, several salient features of graphene make it a potential candidate for biological and biomedical applications (
Graphene oxide (GO) has long been considered hydrophilic, owing to its excellent water dispersity (differently than hydrophobic graphene), with the nanosheets separated from each other due to the electrostatic repulsion between the negative charges present on the GO surface (
Peptide-based self-assembly offers new routes to fabricate various nanostructures including nanotubes, nanowires, nanospheres, nanofibrils, vesicles and hydrogels (
In the last few years, nanostructures based on the diphenyl peptide (FF) (
Unlike the most popular analogous of FF, there are only few reports on ordered assemblies of the dityrosine peptide (YY) (
We demonstrate here that the different surface decoration of the graphene-based 2D nanosheets by the two dipeptides allows for a modulated interaction with artificial cell membranes made of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) supported lipid bilayers (SLBs). Specifically, in this work we scrutinized the self-assembly of FF and YY dipeptides in the absence and/or presence of GO nanosheets as well as of copper ions (Cu2+). The morphology of the peptide assemblies was scrutinized by atomic force microscopy (AFM), while the aggregation process of the peptides in the presence of copper and/or GO was studied by circular dichroism (CD), UV–visible and fluorescence spectroscopies, to assess the conformational features, the copper complex formation and the energy transfer processes in the aggregates, respectively. Moreover, the coordination environment of the peptides-copper complexes grown in the presence of GO was investigated by means of electron paramagnetic resonance (EPR) spectroscopy. The perturbative effect by the hybrid nanomaterials on SLBs model cell membranes was studied by means of quartz crystal microbalance with dissipation monitoring (QCM-D) and fluorescence recovery after photobleaching (FRAP) techniques, to analyze, upon the formation of the hybrid SLB-dipeptide biointerface, the viscoelastic as well as the fluidity features of the artificial cell membrane, respectively.
The obtained results pointed out the different surface decoration of the graphene nanosheets, with the peptide molecules that selectively gather at their edges (FF) or basal planes (YY), with significant changes observed for the nanoassemblies grown in the presence of copper ions. Proof-of-work cellular experiments on human tumor (neuroblastoma and prostatic) cells highlighted the promising potential of these platforms for theranostic applications.
Phe-Phe (FF) and Tyr-Tyr (YY) dipeptides were obtained from CASLO ApS (Denmark). 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP), copper sulfate and Thioflavin-T (Th-T) were purchased from Sigma Aldrich (Italy). Ultrapure Millipore water (18.2 mΩ cm at 25°C) was used to prepare all aqueous solutions. GO was purchased from Graphenea Inc., (United States). Small unilamellar vesicles (SUVs) were prepared from 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and rhodamine-labeled 1,2-dihexadecanoyl-snglycero-3-(Rhod-DHPE) purchased from Avanti Polar Lipids (Alabaster, AL, United States). Phosphate buffer saline (PBS) solution (0.01 M phosphate buffer containing 0.003 M KCl and 0.14 M NaCl, pH 7.4) was prepared from tablets (Sigma). Dulbecco’s modified eagle medium (DMEM)-F12, RPMI 16-40 medium, fetal bovine serum (FBS), streptomycin and L-glutamine, dimethyl sulfoxide (DMSO), bovine serum albumin (BSA), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) were purchased from Sigma Aldrich.
As received GO was diluted in ultrapure Millipore water to a final concentration of 0.4 mg/mL. The GO sheets with different lateral sizes were prepared by sonication for 120 min using a titanium cup-horn sonicator (Hielscher UP200Ht) at 200 W and 24 KHz. The sonicated GO was centrifuged (13000 rpm, 20 min), to separate nanosheets (sub-micron lateral size) from bulky large sheets (up to several mm of lateral size), most likely assembled in the pellet. The collected supernatant typically contained approximately 0.3 mg/mL of GO, as determined by UV–visible spectroscopy (GO extinction coefficient, ε230nm, of 57.9 mg–1 mL cm–1). Fresh stock solutions of Phe-Phe and Tyr-Tyr were prepared by dissolving lyophilized form of the peptides in HFIP at a concentration of 100 mg/mL. The peptides stock solutions were diluted to a final concentration of 10–3 M in Millipore water, CuSO4 (10–3 M), GO dispersion (0.34 mg/mL) and GO/CuSO4 (0.34 mg/mL, 10–3 M). The pH of the solutions was adjusted to 7.4. The solutions were stored at room temperature.
Ten microliters of the 10–3 M solutions were put on freshly cleaved muscovite mica (Ted Pella, Inc.) and incubated at room temperature for 5 min. After that, the mica surface was washed with 1 mL of Millipore water, dried under a gentle nitrogen stream, and imaged. Scans were recorded using a Cypher AFM instrument (Asylum Research, Oxford Instruments, Santa Barbara, CA, United States), operating in tapping-mode and furnished with a scanner at an XY scan range of 30/40 μm (closed/open loop). Tetrahedral tips, made of silicon and mounted on rectangular 30 μm long cantilevers, were purchased from Olympus (AT240T S, Oxford Instruments). The probes had nominal spring constants of 2 N/m and driving frequencies of 70 kHz. Images with an area of 2 × 2 μm and 5 × 5 μm were scanned, and the sizes of particles were measured using a free tool in the MFP-3DTM offline section analysis software. AFM images were acquired at different aging time, namely 18, 42, 66, 120, and 240 h.
CD spectra were obtained at 25°C under a constant flow of nitrogen on a Jasco model 810 spectropolarimeter, calibrated with an aqueous solution of (1R)-(-)-10-camphorsulfonic acid. Measurements were carried using 1 cm path length quartz cuvettes. The CD spectra were obtained in the 190–350 nm wavelength region. CD spectra were acquired at different aging time, namely 18, 42, 66, 120, and 240 h. UV–visible spectroscopy was performed in quartz cuvettes with 1 cm optical path length on a Jasco spectrometer. All the solutions were freshly diluted using Millipore water. Fluorescence studies of all samples were carried out by a Horiba Jobin Yvon Fluoromax spectrometer with a path length of 1 cm. Emission spectra of the samples were recorded at the range of 235–425 nm with an excitation wavelength of 230 nm. All the solutions were freshly diluted using Millipore water. The used concentrations were of 5⋅10–5 M for dipeptide, 5⋅10–5 M for Cu2+ and 17 μg/mL for GO, respectively.
EPR measurements were carried out by using a Bruker Elexsys E500 CW-EPR spectrometer driven by a PC running the XEpr program on Linux and equipped with a Super-X microwave bridge, operating at 9.3–9.5 GHz, and a SHQE cavity were used throughout this work. All the EPR spectra of frozen solutions of Cu2+ complexes were recorded at 150 K by means of a ER4131VT variable temperature apparatus. EPR magnetic parameters were obtained directly from the experimental EPR spectra, calculating them from the 2nd and 3rd line to avoid second order effects. The instrument settings for the EPR spectra recordings of the copper2+-peptide complexes were as follows: number of scans 1–5; microwave frequency 9.344–9.376 GHz; modulation frequency 100 kHz; modulation amplitude 0.2–0.6 mT; time constant 164–327 ms; sweep time 2.8 min; microwave power 20–40 mW; receiver gain 50–60 dB. Copper2+-peptide complexes were prepared by adding the appropriate amount of isotopically pure copper, taken from a 0.05 M 63Cu(NO3)2 solution, to the peptide solution. The copper2+-peptide complex solutions were prepared at 1:1 metal-to-ligand ratio (10–3 M concentration), with the GO concentration fixed to 0.34 mg/mL.
A VarioSkan Flash fluorescence 96-well plate reader (Thermo Scientific) was used for the Th-T measurements. To minimize evaporation effects, the wells were sealed with a transparent heat-resistant plastic film. Readings were taken every 10 min, after gentle shaking for 10 s, during a total acquisition time of 2 days. The fluorescence excitation wavelength was set at 440 nm, and emission was detected at 480 nm. All Th-T experiments were carried out at pH 7.4, 37°C and in 10 mM phosphate buffer. The Th-T concentration was 60 μM. The Th-T curves represent the average of three independent experiments, which were each run in quadruplicate.
All the calculations were performed in the framework of the DFT theory, using the PBE0 functional (
A 5 mg/mL solution of POPC in chloroform, mixed with Rhod-DHPE (1 wt.%), was added in a round-bottom flask and the chloroform was evaporated under a flow of N2 while rotating the flask, in order to create a uniform film on the glass wall. The dried lipid film was emulsified by using 10 mM PBS, at room temperature, and vortexed to obtain the vesicles. Afterward, in order to obtain the SUVs dispersion, about 80–100 nm of diameter (
QCM-D measurements were carried out in flow mode (100 μL/min) on a Q-sense D 300 setup (Q-Sense AB, Gothenburg, Sweden). Prior to each measurement series, the crystals (SiO2-coated AT-cut quartz crystals with a fundamental resonance frequency of 5 MHz) were cleaned by immersion in 10 mM sodium dodecyl sulfate (SDS, >1 h), followed by rinsing with water, drying with nitrogen, and UV–O3 treatment (30 min). Frequency shifts were normalized by the overtone number. After stabilization of the baseline, 0.5 mL of a temperature-stabilized vesicle solution (100 μg/mL) was injected in the measurement chamber for the formation of the lipid bilayer on the sensor surface, by the spontaneous rupture/fusion processes following the adsorption of the vesicles onto the hydrophilic silica surfaces of the QCM-D sensor (
To carry out the FRAP experiments, glass bottom dishes (Willco Wells) were cleaned by two repeated cycles of 20 min UV–O3 treatment, with water rinsing and N2 flow drying in between and at the end of the treatment. Subsequently, the SUVs (100 μg/mL) were allowed to adsorb onto the glass surface of the dishes for 30 min, to form fluid SLBs by the spontaneous rupture/fusion processes following the adsorption of the vesicles onto the hydrophilic glass surfaces.(
SH-SY5Y neuroblastoma cells were cultured in Dulbecco’s modified Eagle medium (DMEM-F12) supplemented with 10% FBS and 2 mM glutamine; PC-3 prostate cancer cells were cultured in RPMI 16-40 medium supplemented with 10% FBS. 100 U penicillin and 0.1 mg/mL streptomycin were added in all the used cell media. Cells were grown in tissue-culture treated Corning® flasks (Sigma-Aldrich, St. Louis, MO, United States) under a humidified atmosphere of air/CO2 (95:5) at 37°C in an incubator (Heraeus Hera Cell 150C incubator).
To perform the cytotoxicity assay, cells were plated in three different 96-well plates in complete medium for 24 h, at a density of 15 × 103 cells per well for SH-SY5Y and 10 × 103 cells per well for PC-3, respectively. Afterword, cells were treated with FF (1⋅10–4 M, 5⋅10–5 M, and 2⋅10–5 M), Cu2+ (1⋅10–4 M, 5⋅10–5 M, and 2⋅10–5 M) and GO (28, 14, and 5.6 μg/mL for GO) for 1 h in medium supplemented with 1%. Cytotoxicity was determined at 37°C by using the tetrazolium dye 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). After 1 h of incubation, the enzymatic reduction of MTT to the insoluble purple formazan product was detected by dissolving the crystals with 100 μL of dimethyl sulfoxide and thus measuring the absorbance at 570 nm by Varioscan spectrophotometer. The experiments were performed in triplicate and the results are presented as the means ± SEM.
Since the growth process of self-assembling molecules follows kinetics and thermodynamics paths that depend on the used experimental conditions (
CD spectra of FF
For both FF and YY, the spectra of the self-assembled aggregates grown in the presence of GO are similar to those of the free dipeptides, with no significant differences found between the freshly dissolved peptides (
In the presence of divalent copper ions (1:1 molar ratio with the peptides), an intensity enhancement of the CD bands was observed, in parallel with the appearance (especially for FF) of a broad negative band peaked near 280 nm, due to the contribution by aromatic side chains (
In parallel to CD, the UV–visible and fluorescence spectra of the dipeptides were acquired at different times and up to 1 week since the sample dissolution in milliQ water (
The addition of Cu2+ quenches the fluorescence emission of both dipeptides due to the complex formation (
Similarly to what detected by CD spectra, the presence of copper during the growth of the peptide aggregates induces the appearance of the aromatic side-chains absorption, in the 230–300 nm range. No significant changes in the aromatic side-chains absorption (signature of differences in stacking interactions and hence in the supramolecular assembly) were observed during the aging time, but only a general increase of absorption for the signal relative to the peptide bond.
Both the enhancement of absorption with aging time as well as the quenching of the intrinsic amino acid fluorescence (
AFM analyses confirmed the strong effect of copper ions on the assembly process for FF and YY, as well as the different GO-driven ordered arrangement of the dipeptide nanostructures (
AFM topography images recorded in AC mode in air for dipeptides after 1-week self-assembly in:
According to a simple qualitative model, the most preferential interactions between GO and the aromatic dipeptides are hydrophobic forces (π–π stacking and hydrophobic effect) that occur on the graphitic zones of GO (unoxidized region) (
Theoretical simulations (
Noteworthy, the phase images shown in
AFM phase images (xy scale = 1 × 1 μm2) of:
To note, EPR analyses pointed to the formation of the planar structures of peptide-metal chelates that could prompt the supramolecular assembly in flat and extended domains (
The interaction of peptide-decorated GO nanosheets with POPC SLBs, used to test the hybrid nanobiointerface with model cell membranes (
QCM-D curves of frequency (black lines, left hand side axis) and dissipation (gray lines, right hand side axis) shifts for the SLB interaction with the aggregates of:
To be noted, this interpretation is also supported by CD results discussed above, since a β-sheet configuration is adapted for insertion into the lipid membrane (
where C is the mass sensitivity constant (∼−17.7 ng cm–2⋅Hz–1 for a 5 MHz quartz crystal) and
As to FF-GO, a minor decrease in
To note, for the addition of FF + Cu2+ (
Fluorescence recovery after photobleaching (FRAP) experiments confirmed the different interaction between the dipeptide nanoaggregates and SLB, as well as a modulating effect by the presence of GO and/or Cu2+, as measured in terms of the lateral diffusion coefficients of the lipids within the membrane (
Representative confocal micrographs (
Proof-of-work experiments on the different interaction of the peptide aggregates grown in the diverse experimental conditions with true cell membranes were performed. In particular, a dose-response experiment of MTT assay was carried out at a short time of incubation (
MTT results (
Cell viability of SH-SY5Y
As to the cellular treatments with FF, no cytotoxic effects were observed on both the cell lines instead, instead a significant increase of viability was detected on SH-SY5Y at the lowest concentration tested for FF and, in a dose-dependent manner, for the hybrid FF-GO. No significant difference with respect to the untreated cells were observed for PC-3 treated either with FF or FF-GO samples. Finally, as to the cellular treatments with FF + Cu2+ or FF-GO + Cu2+, roughly ∼ 20% of increased viability was exhibited by neuroblastoma cells (after the incubation with the FF + Cu2+ at the highest tested peptide concentration). On the other hand, a dose dependent cytotoxicity was still observed in PC-3 cells, with a similar but less pronounced trend as the treatment with Cu2+ alone. In summary, these results point to a cell-dependent response, with different trends of increase or decrease in cell viability by the various cellular treatments with dipeptide-GO, dipeptide+ Cu2+ or dipeptide-GO+ Cu2+ systems.
In this work we addressed the decoration of graphene oxide nanosheets by aromatic dipeptide nanostructures, self-assembled both in the absence and in the presence of copper ions. The spectroscopic and microscopic characterisation (CD, UV–visible, fluorescence, EPR, AFM) pointed to: (i) a strong interaction between the GO and the dipeptides, as monitored by the bands of π → π∗ transition, characteristic of peptide aggregates by the molecular stacking in beta-sheet conformation; (ii) a triggering effect by the copper ions on the peptide self-assembly process, by the formation of metal complexes, as well as on the preferential gathering of the peptide aggregates onto the edges and/or the basal planes of the GO nanosheets. Theoretical calculations confirmed the possibility to tune, by means of electrostatic vs. hydrophobic interaction forces, the growth of the peptide aggregates at the interface with GO.
QCM-D and FRAP experiments with SLBs and cellular experiments on two model cancer cells, pointed out the tuneability of the interaction between the peptide-decorated GO nanosheets and cellular membrane, as demonstrated in terms of viscoelastic properties and lateral diffusion of the lipids within the membrane as well as by proof-of-work
All datasets generated for this study are included in the article/
GTr and VC carried out the experiments and drafted the manuscript. LC, FA, and GTa carried out the experiments (CD, ThT, and EPR, respectively). GF helped with data analyses and modeling. DL and CS conceived the original idea, supervised the project, revised and wrote the final manuscript.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Authors acknowledge the Consorzio Interuniversitario di Ricerca in Chimica dei Metalli nei Sistemi Biologici (CIRCMSB), Bari, Italy.
The Supplementary Material for this article can be found online at: