Control of Integrin Affinity by Confining RGD ... - ACS Publications

Apr 30, 2018 - Institute of Physical Chemistry, Göttingen University, Tammanstrasse ... The University of Texas at El Paso, El Paso, Texas 79968, Uni...
1 downloads 0 Views 2MB Size
Subscriber access provided by Kaohsiung Medical University

Biological and Medical Applications of Materials and Interfaces

Control of integrin affinity by confining RGD peptides on fluorescent carbon nanotubes Elena Polo, Tadeusz T. Nitka, Elsa Neubert, Luise Erpenbeck, Lela Vukovic, and Sebastian Kruss ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b04373 • Publication Date (Web): 30 Apr 2018 Downloaded from http://pubs.acs.org on May 2, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Control of integrin affinity by confining RGD peptides on fluorescent carbon nanotubes

Elena Polo 1, Tadeusz T. Nitka2, Elsa Neubert1,3, Luise Erpenbeck3, Lela Vuković², Sebastian Kruss1,4* 1

Institute of Physical Chemistry, Göttingen University, Tammanstrasse 6, 37077 Göttingen,

Germany 2

Department of Chemistry and Biochemistry, The University of Texas at El Paso, El Paso,

Texas, 79968, USA 3

University Medical Center, Department of Dermatology, Göttingen University, Germany.

4

Center for Nanoscale Microscopy and Molecular Physiology of the Brain (CNMPB), 37073

Göttingen, Germany *Corresponding author: [email protected]

KEYWORDS: Carbon nanotubes, surface functionalization, integrins, RGD, cell adhesion, near infrared fluorescence, fluorescent probes

1 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 35

ABSTRACT

Integrins are transmembrane receptors that mediate cell-adhesion, signaling cascades and platelet-mediated blood clotting. Most integrins bind to the common short peptide Arg-Gly-Asp (RGD). The conformational freedom of the RGD motif determines how strong and to which integrins it binds. Here, we present a novel approach to tune binding constants by confining RGD peptide motifs via non-covalent adsorption of single stranded DNA (ssDNA) anchors onto singlewalled carbon nanotubes (SWCNTs). Semiconducting SWCNTs display fluorescence in the near infrared (nIR) region and are versatile fluorescent building blocks for imaging and biosensing. The basic idea of this approach is that the DNA adsorbed on the SWCNT surface determines the conformational freedom of the RGD motif and affects binding affinities. The RGD motif was conjugated to different ssDNA sequences in both linear ssDNA-RGD and bridged ssDNA-RGDssDNA geometries. Molecular dynamics (MD) simulations show that the RGD motif in all the synthesized systems is mostly exposed to solvent and thus available for binding, but its flexibility depends on the exact geometry. The affinity for the human platelet integrin αIIbβ3 could be modulated up to 15-fold by changing the ssDNA sequence. IC50 values varied from 309 nM for (C)20-RGD/SWCNT hybrids to 29 nM for (GT)15-RGD/SWCNT hybrids. When immobilized onto surfaces adhesion of epithelial cells increased 6-fold for (GT)15-RGD/SWCNTs. (GT)15RGD/SWCNTs also increased the number of adhering human platelets by a factor of 4.8. Additionally, αIIbβ3 integrins on human platelets were labeled in the nIR by incubating them with these ssDNA-peptide/SWCNT hybrids.

In summary, we show that ssDNA-peptide hybrid structures non-covalently adsorb onto SWCNTs and serve as recognition units for cell surface receptors such as integrins. The DNA sequence affects the overall RGD affinity, which is a versatile and straightforward approach to 2 ACS Paragon Plus Environment

Page 3 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

tune binding affinities. In combination with the nIR fluorescence properties of SWCNTs these new hybrid materials promise many applications in integrin targeting and bioimaging.

3 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 35

Introduction Nanomaterials are important building blocks for a variety of functional materials and enable completely new applications1-4. Among them carbon nanotubes are very attractive due to their unique photophysical properties and 1D geometry5-9. Since their discovery they have been explored as functional building blocks in biosensing and bioimaging10-14. Single-walled carbon nanotubes (SWCNTs) are rolled-up monolayers of carbon. Their structure is described by the chirality or chiral index (n,m), which determines their photophysical properties. Semiconducting SWCNT have a bandgap that results in near infrared (nIR) fluorescence (850 – 1700 nm)15. This spectral range is promising for biomedical application because it falls into the optical transparency window of biological materials16. Additionally, SWCNTs display ultra-low bleaching and a large Stokes-shift of several hundred nanometers17. To use SWCNTs as building blocks for functional materials their surface typically needs to be chemically modified. However, covalent modification such as oxidation distorts the conjugated sp2 system and in most cases destroys the nIR fluorescence even though certain defects create red-shifted emission peaks18-22. In contrast, non-covalent functionalization preserves their nIR fluorescence and renders the intrinsically hydrophobic SWCNTs water-soluble23. Due to their 1D nature SWCNTs are sensitive to their environment and are useful building blocks for sensors24. So far, SWCNTs non-covalently functionalized with different polymers and biomacromolecules such as single-stranded DNA (ssDNA) were used to create fluorescent biosensors for reactive oxygen species, small molecules, neurotransmitters, nucleic acids, sugars and proteins24-28. One important example are SWCNTs functionalized with specific DNA sequences such as (GT)15-ssDNA that were responsive to the neurotransmitter dopamine24. These sensors were used to image and monitor dopamine release from living neuronal cells13. In the presence of dopamine 4 ACS Paragon Plus Environment

Page 5 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

the fluorescence intensity of these SWCNT-based sensors increased. The mechanism was attributed to a conformational change of the ssDNA upon interaction with dopamine causing a change of exciton decay routes13,23. In this particular case interactions between charged moieties of the DNA backbone (phosphate groups) and dopamine played an important role13. Hence, the ssDNA served as a conformational quantum yield switch and a recognition motif for dopamine. By changing the DNA sequence the selectivity and dissociations constants for similar molecules such as epinephrine and norepinephrine could be modified over several orders of magnitude29. For imaging of fast processes such as chemical signaling sensor kinetics play a crucial role. Kinetic Monte-Carlo simulations showed that rate constants, which are directly related to the surface functionalization, are the key parameter for fast fluorescent sensing and have to be in a certain range for optimal sensor performance30. Although ssDNA/SWCNTs sensors are highly sensitive, selectivity still remains a challenge. Proteins and peptides are well-known for their specific molecular interactions in nature. So far, peptides were much less used to functionalize SWCNTS than ssDNA31-34. The reason for that is that not every peptide functionalization guarantees colloidal stability and the higher tendency of aggregation. On the other hand, an organic phase based on peptides opens new possibilities for incorporating specific receptors for sensing. Recently, a novel approach to encapsulate SWCNT in peptide barrels was introduced but in general new concepts to bring peptide moieties into the organic phase around SWCNTs are needed35. Powerful functionalization concepts are necessary to ensure specific binding to targets such as cell surface receptors. One important class of cell surface receptors are integrins, which mediate cell-adhesion, cell motility and transduce signals36. In addition to its role in cell adhesion, integrins also participate in signaling events that govern epithelial differentiating and play a 5 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 35

pivotal role in platelet-mediated blood clotting37-38. Integrins are formed through non-covalent association of different α and β subunits39. For example, αIIbβ3 integrins are prevalent on platelets and implicated in hemostasis and thrombosis40. This integrin has received a lot of attention in drug discovery efforts41. Interestingly, several integrins recognize the peptide sequence Arg-GlyAsp (RGD), a motif found in their target proteins42. The presence and chemical nature of flanking residues as well the conformational geometry of the peptide sequence govern integrin selectivity. The group of Kessler was able to identify structural and geometrical factors of RGD-motifs responsible for selective binding to various integrin types43. Nanomaterials have been extensively modified with RGD motifs including gold and silver nanoparticles44-47. SWCNTs have also been modified by attaching RGD modified phospholipid-polyethyleneglycols (PL-PEG)

48-49

. They

were used for nIR in vivo labeling of the integrin αvβ3 and to investigate the distribution and dissemination of αvβ3-positive cancer cells in mice. Here, we present a novel strategy to conjugate peptide motifs such as the integrin binding RGD sequence to near infrared fluorescent carbon nanotubes and tune their binding affinities. The approach is based on non-covalent adsorption of a DNA backbone, which determines the conformational freedom of a linked RGD peptide motif and affects its binding affinities to integrins. Consequently, binding constants are affected by DNA sequence, length or the geometry of the DNA-peptide hybrid. We demonstrate how such DNA-peptide/SWCNTs hybrids increase cell adhesion and employ them as nIR labels for integrins on tissue and blood cells.

6 ACS Paragon Plus Environment

Page 7 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Results

Figure 1. Modular ssDNA-peptide/SWCNT hybrids. a) Schematic of the hybrid materials. The RGD peptide motif is anchored between two ssDNA sequences (‘bridge’) or attached to one side of a ssDNA sequence (‘linear’). The RGD motif serves as (integrin) recognition unit, while the ssDNA adsorbs on the SWCNT surface and affects the exact RGD conformation and flexibility. b) Schematic of the possible interactions between integrins on cell surfaces and DNA-RGD/SWCNT hybrids. c) Typical absorption spectrum (here the (C)30-RGD/SWCNT hybrid is shown). d) Typical nIR fluorescence emission spectrum (here the (C)30-RGD/SWCNT hybrids shown). e) Conjugation strategy for ssDNA-peptide-ssDNA (‘bridge’) and ssDNA-peptide (‘linear’) motifs. In order to bring the RGD motif into the vicinity of SWCNTs the RGD motif was conjugated in different ways to single stranded (ss)DNA sequences: a) the RGD-sequence was conjugated to one end of a ssDNA strands (linear, ssDNA-RGD) and b) the RGD-sequence was placed in between two ssDNA strands (bridge, ssDNA-RGD-ssDNA). Fig. 1a shows schematics of both approaches and the possible interaction of ssDNA-RGD/SWCNT hybrids with integrins. The hybrids were prepared by tip sonication of (6,5) chirality enriched SWCNTs in a solution of the respective ssDNA-peptide molecules. Absorption spectra of these hybrids show in the nIR typical 7 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 35

but redshifted S11 peaks and in the visible range typical S22 peaks of (6,5)-SWCNTs (Fig. 1c, (C)30-RGD/SWCNT, see Fig. S1 for spectra of all hybrids). Similarly, the hybrids fluoresce in the nIR (Fig. 1d and Fig. S1), indicating that the chemical approach leads to stable suspensions and that the nIR fluorescence is preserved. Two different general strategies (linear vs. bridged) should have implications on the binding behavior. Linear peptide hybrids should have more conformational freedom, facilitating easier binding into the integrin’s binding pocket. On the other side conformational promiscuity might lead to low selectivity differences for the individual integrin types43. This drawback can be significantly improved by cyclization or any other confinement or reduction of the conformational space. Our approach restricts the conformational freedom of the RGD motif by coupling it to a single ssDNA sequence or by placing it between two ssDNA units. These macromolecules are then adsorbed to the surface of a SWCNT to fix the conformation in space. Fig. 1e shows the exact synthesis routes for both bridged (ssDNA-RGD-ssDNA) and linear approach (ssDNA-RGD) as well as the linker chemistry (see Materials and Methods section for more details).

8 ACS Paragon Plus Environment

Page 9 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 2. Inhibition of integrin/fibrinogen binding by ssDNA-RGD/SWCNTs. a) Schematic of the inhibition assay developed to quantify affinity between ssDNA-RGD/SWCNTs and αIIbβ3 integrin: If integrins bind to fibrinogen, horseradish peroxidase (HRP) binds via two antibodies and a colored product forms. RGD motifs compete for the free integrin, which changes the signal depending on affinity. b) Inhibition curves for (GT)15-RGD/SWCNTs (black squares) and C20RGE-C20/SWCNTs (red dots). Intensities are normalized to the maximal value without ssDNApeptide/SWCNTs and background signal of well-plates without integrin. Error bars are standard errors, n = 2. c) Inhibition curves for linear C20-RGD/SWCNTs (black squares) and bridged C20RGD-C20/SWCNTs (red dots). Four parameter logistic function fits are shown as red line. IC50 values mark the 50% drop in initial intensity (dashed orange line). Error bars are standard errors, n = 2.

9 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 35

In the next step, we investigated how different ssDNA-peptide/SWCNT hybrids affect/inhibit the binding of integrins to its natural binding partners. For this purpose, an enzyme-linked immunosorbent binding assay (ELISA) was developed to detect free αIIbβ3 integrin and assess binding affinity of ssDNA-RGD/SWCNT hybrids to this integrin type (Fig. 2a). First, the glass surface was coated with fibrinogen from human plasma. After that soluble human αIIbβ3 integrin (from blood platelets), anti-αIIbβ3 antibody and a secondary IgG antibody conjugated to horseradish peroxidase (HRP) was added. HRP catalyzes the reaction of a substrate into a product that absorbs light at 492 nm. This assay is able to detect and quantify the presence of the integrin. ssDNA-RGD/SWCNTs compete with fibrinogen for the integrin and the ELISA signal decreases. After successful binding, the ssDNA-RGD/SWCNT-integrin compound is washed away, which decreases the intensity. Fig. 2b shows representative normalized inhibition curves for (GT)15RGD/SWCNTs (black squares) and C20-RGE-C20/SWCNTs (red dots). The difference between RGD and RGE is a simple amino acid (aspartic acid (D) instead of glutaminc acid (E)). However, the RGE peptide binds with much smaller affinity to integrin binding pockets50. Therefore, ssDNA-RGE/SWCNTs hybrids can serve as excellent control to exclude any unspecific binding between ssDNA-peptide/SWCNTs and αIIbβ3 integrin (Fig. 2b). Logistic function fits (Fig. 2c) are shown as dashed red lines and are used to derive the 50% inhibition (IC50) values for different ssDNA-peptide/SWCNTs (see also Materials and Methods).

10 ACS Paragon Plus Environment

Page 11 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 3. IC50 values of ssDNA-peptide/SWCNT hybrids. a) IC50 values of ssDNApeptide/SWCNTs binding to αIIbβ3 integrin. The values were calculated by logistic function fits from ELISA assays (color code: black = low IC50, white = high IC50). b) Number of ssDNApeptide strands per nanotube. c) IC50 values (effective, arbitrary units) for binding of ssDNApeptide/SWCNTs to αIIbβ3 integrin normalized to the number of RGD-motifs per nanotube (IC50 value x number of RGD motifs per nanotube). Fig. 3a shows IC50 values of ssDNA-peptide/SWCNT hybrids binding to αIIbβ3 integrin. Each IC50 value corresponds to n=2 inhibition curves fitted to a logistic function as shown in Fig. 2c 11 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 35

(see also Materials and Methods) and represents the concentration that is necessary to prevent 50% integrin from binding to fibrinogen. Interestingly, all three parameters of the DNA, namely sequence, length and confinement (linear vs. bridged), affect the affinity of RGD to the αIIbβ3 integrin. The lowest IC 50 values were observed for bridged C20-RGD-C20SWCNT (20 nM). In contrast, linear (GT)15-RGD/SWCNT had lower IC50 values (29 nM) compared to bridged (GT)15-RGD-(GT)15/SWCNT. These results suggest that there is a complex relationship between binding affinities and the exact structure of the macromolecule. Different ssDNA sequences exhibit different affinities to SWCNTs. Iliafar et al. measured this affinity by pulling single-stranded DNA sequences from the SWCNTs surface and determining the applied force51. Based on their calculation we expect the (GT)-sequences to bind stronger to SWCNTs than the C-sequences. Different affinities could cause different conformations of the ssDNA around nanotube and affect the scaffolding and confinement of the peptide sequence. Another possible explanation for the different affinities is the number of adsorbed ssDNApeptide macromolecules and thus the number of RGD motifs per SWCNT. It is therefore important to distinguish between affinity/avidity differences due to different numbers of RGD motifs. Fig. 3b shows the number of different ssDNA-peptide units per single SWCNT. They were calculated by determining the free DNA after tip sonication and filtration (see materials and methods). The results indicate up to a 4-fold difference in the number of RGD motifs per nanotube. As expected, there is a tendency that longer ssDNA-peptide sequences occupy more space and thus fewer RGD motifs are present on the SWCNTs. These numbers were used to ‘normalize’ the IC50 values by multiplying them with the IC50 values of Fig. 3a. Consequently, the inhibition pattern changes but the candidates with the lowest IC50 values stay the same indicating

12 ACS Paragon Plus Environment

Page 13 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

that most of the affinity can be contributed to conformational aspects and not the number of RGD motifs.

Figure 4. Molecular view of ssDNA-RGD/SWCNTs. a) A representative view of a single (GT)5RGD molecule adsorbed on a SWCNT. b) A representative view of a single (GT)5-RGD-(GT)5 molecule adsorbed on a SWCNT. c) A fraction of the exposed surface area of RGD motif on SWCNT surfaces, calculated in three independent simulations, for (GT)5-RGD (red) and (GT)5RGD-(GT)5 (blue) systems. d) A crystal structure of the RGD motif binding to the integrin αIIbβ3 (pdbID: 2vdr). The RGD motif binds into a shallow surface pocket of the protein complex. To determine the molecular origin of differences in IC50 values of the studied systems (Fig. 3 a,c), we performed atomistic molecular dynamics (MD) simulations of linear and bridged hybrids. Representative linear (GT)5-RGD and bridge (GT)5-RGD-(GT)5 conjugates were adsorbed and simulated on the SWCNT surface. 10 nucleotides long ssDNA sequences were selected to better track the dynamics of the conjugates on the 100 ns timescale (longer ssDNA would have slower dynamics). ssDNA molecules initially wrapped SWNTs in helix-like conformations52-54, where

13 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 35

(GT)5 formed nearly one full turn of the helix. In 100 ns simulations, ssDNA molecules remained wrapped and stacked on SWCNTs, as shown in Fig. 4 a,b and Fig. S2.

Figure 5. Dynamics of RGD motifs adsorbed on SWCNTs. Dynamics are shown in terms of cylindrical coordinates (a) r; (b) θ; and (c) z. θ and z motions of RGDs are calculated with respect to the ssDNA parts of molecules. Translation and rotation of the whole (GT)5-RGD (right) and (GT)5-RGD-(GT)5 (left) molecules were subtracted out, in order to get pure dynamics of RGDs. 14 ACS Paragon Plus Environment

Page 15 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

The solvent exposure of the RGD motif was analyzed for the two selected systems in three independent simulations (Fig. 4c). The RGD motif was mostly (70-80%) exposed to the solvent in all the cases, and thus available to bind to the integrin. The other 20-30% of the surface of the RGD motif interacted with SWCNT and ssDNA. This exposure of RGD should be sufficient for integrin binding, since RGD binds into a shallow surface pocket of the integrin, as shown for the integrin αIIbβ3 in Fig. 4d. The observed availability of RGD to bind to the integrin is in agreement with the binding observed in experiments (Fig. 2,3). According to Fig. 3c, bridged hybrids always have low IC50 values (