Nanostructured films by the self-assembly of bioactive copolymer

Abstract

The synthesis of multifunctional macromolecules has become an important topic, as advances are made in biological sensing technology. Various templates that may be quantified with the presentation of bioactive molecules and the control over their density and orientation are required. The paper shows that thin films based on a bioactive copolymer, acrylamide/N-(2-dibenzylamino-ethyl)-acrylamide (1), with arrays of submicro and nanoscale cavities can be formed under ambient conditions, using a self-organization method, such as a dewetting process. We showed that the concentration of 1 in its water solution, as well as the hydrophobicity of a film support, are key parameters, which allow the control over nanocavity formation on the surfaces of the thin films of 1; the surface morphologies were studied using Atomic Force Microscopy (AFM). The simple procedure used for engineering DNA trapping nanocavities on the surface of the bioactive thin films of 1 makes these films very promising for applications where the nanoscale detection of biomacromolecules is required.

---

Introduction

Bioactive nanostructured surfaces are of strong interest for basic research as well as for future biomedical applications. For example, progress has been done in transplantation of “encapsulate” beta cells using both natural and synthetic membranes. On the one hand, such membranes have pores large enough to allow insulin and glucose to pass freely, but these pores are small enough to permit antibodies to reach and target the transplanted foreign beta cells.^1,2 On the other hand, nanopores of specific characteristics are very useful for the successful detection of biomolecule complexes.^3–6 O. A. Saleh and L. L. Sohn successfully demonstrated the capabilities of a nanopore-based device to sense electronically single DNA molecules.⁷ Porous nanocapsules and polymer hydrogels are extremely interesting for cell and drug delivery.^8–11 In this context, the development of thin films based on multifunctional macromolecules that are able to recognize different biomacromolecules has become an important topic. Recently, we described the synthesis of the copolymer acrylamide/N-(2-dibenzylamino-ethyl)-acrylamide (1, Fig. 1a and b) and showed that 1, united with double-stranded DNA, is able to form a set of various nanostructures.^12,13

Fig. 1 Skeletal formula (a) and schematic image (b) of copolymer 1.

The electrostatic trapping of DNA molecules by bioactive copolymer 1 results in the formation of nanostructures reported by some of the authors of this article.¹²

To extend the capability of 1 to be used as a biomacromolecule sensing component, a method to prepare thin 1-based films with controlled nanocavity surfaces has to be developed. Engineering such sensing films is related to nanotechnology that is of great importance to the industry striving for novel miniature sensing devices. Taking into account that dewetting of polymeric films may be successfully used for engineering nanostructured polymer surfaces,¹⁴ we used it for building nanostructured 1-based films, which was the goal of this paper. AFM was applied for the nanoscale study of the surfaces of the developed films.

Here, we present the preparation of nanopatterns of thin 1-based films using a dewetting process. The effect of the concentration of 1 in its water solution on the nanopattern of the film surfaces is demonstrated. We also describe the important role of the hydrophobicity of a supporting substrate in 1-based film stability.

Experimental

Copolymer synthesis

1 was prepared by free-radical polymerization in an aqueous medium using a red-ox system of ammonium persulfate–tetramethyl-ethylenediamine as it was reported previously.¹²

Support preparation

Natural mica crystals and highly oriented pyrolytic graphite (HOPG) were chosen as hydrophilic and hydrophobic supports, respectively. They were cleaved just before the experiments using Scotch tape. HOPG was produced by Atmograph-crystal.

Film preparation

Water solutions of different concentrations of 1 (3 mg ml⁻¹, 0.03 mg ml⁻¹, and ≅0.003 mg ml⁻¹) were prepared using bidistilled water; films were cast from the water solutions of 1 over mica- and graphite-based supports (2.5 μl) and dried under ambient conditions. The names of the films and their characteristics are summarized in Table 1. For comparative purpose, two films of each type were prepared.

Table 1 The effect of the concentration (C) of 1 and the nature of the supporting substrates on the surface morphologies of the bioactive 1-based films

Film type	C, mg ml^−1	Supporting substrate	Cavity morphology		Cavity density
			Diameter, nm	Depth, nm	Cavities/μm^2
A	3.00	Mica crystal	218 ± 71	1.9 ± 1.2	10/100
B	0.03	Mica crystal	300 ± 230	9.4 ± 3.7	150–160/100
C	≅0.003	Mica crystal	51 ± 16	1.5 ± 0.2	4500/100
D	0.03	Graphite	111 ± 80	13 ± 10	100–150/100

---

DNA trapping experiment

A 0.01 mg ml⁻¹ water solution of λ-phage DNA (Fermentas) was used in the DNA trapping experiments. Drops of the DNA solution (1 μl) were applied to the surfaces of the B films. Then, the samples were dried under ambient conditions.

Surface morphology analysis

The textures of 1-based films were investigated using a multifunctional scanning probe microscope “FemtoScan”, produced by Advanced Technologies Center. Scanning was performed in contact and tapping modes of AFM; fpC11 cantilevers, produced by the Institute of Physical Problems named after F. V. Lukin, were used in the contact mode and MikroMasch cantilevers of the 15 series were used in the tapping mode. The experimental AFM data were processed and analyzed using the “FemtoScan Online” software (Advanced Technologies Center).

Measurements of the film thickness

The thickness of film A was estimated by micrometer “MK25” (Kalibr). The thickness of films B, C, and D was measured by AFM. The copolymer was removed in a small area using the AFM tip in contact mode, and then the area was imaged in tapping mode to determine the height difference between the copolymer and the substrate surfaces (Fig. S1, ESI†).

Results and discussion

To generate a patterned surface of a 1-based film at the nanometric level, we used a dewetting process. Assuming that the viscosity of the casting solutions might be one of the key parameters, which determines a film formation process, the surface properties of the 1-based films were tailored by controlling the initial concentration of 1 in its water solutions. In line with the research aim, we prepared 1-based films (A, B, and C) over the surface of a mica crystal from solutions that significantly differ in concentration (see Table 1). The fourth sample, film D, was formed over a graphite surface from the same water solution that was used for film B casting; all films were prepared using a drop casting method under ambient conditions. The morphologies of the surfaces of the developed films were studied using AFM. It is necessary to note that it was not possible to observe any difference between the surface morphologies of initial films A, B, C, and D and the textures of their duplicates. Table 1 indicates the manner in which the morphology of the 1-based films varies with casting conditions.

Surface morphology of film A prepared from a concentrated water solution

The AFM image of film A (Fig. 2a and b) shows a smooth surface with some very shallow cavities, which are few in number: 10 cavities per 100 μm². The average cavity diameter is 218 ± 71 nm and the average cavity depth is very small as 1.9 ± 1.2 nm. The maximum height difference was 30 nm in defect-free areas of 4 μm² and the root-mean-square roughness was 0.8 nm. Therefore, with the concentration of 1 being 3 mg ml⁻¹, the viscosity of its initial water solution is enough to yield films, which are too stable for hole formation. In this case one may conclude that film A is thick enough to prevent the hole formation process, which is conducted by the surface tension force: there is a lot of film mass to transfer to enable the growth of holes. The thickness of film A was 10 ± 5 μm.

Fig. 2 A typical AFM image of the surface of film A cast over a mica crystal, with a frame size of 10.3 × 10.3 μm² (a), and a 3D AFM image of one of the cavities (b).

Morphology of film B prepared from a diluted water solution

In contrast to film A, the surface of film B has a regular nanostructure, which is shown in Fig. 3a and b and S2, ESI.† The average cavity diameter is 300 nm (the standard deviation of 230 nm). The average cavity depth is about ten nanometers (Table 1). The data were calculated for 60 cavities. The film was partially removed from the mica surface using the AFM tip in contact mode that allowed us to measure the film thickness, which was about 10 nm.

Fig. 3 A typical AFM image of the surface of film B over a mica crystal, with a frame size of 10.1 × 10.1 μm² (a), and a 3D AFM image of one of the cavities (b).

Some wide cavities, which have diameters of 1000–2500 nm and a depth of 11.8–11.9 nm, may be also observed (Fig. S2, ESI†). Below, we will focus on some details of the hole morphologies (Fig. 3). To properly compare textures inside and outside the cavities, the filter “highlight” was applied to the AFM image of the surface of film B (Fig. 4).

Fig. 4 AFM image of the cavity formed on the surface of film B cast over a mica crystal. The frame size is 1.24 × 1.24 μm²; the filter “highlight” was applied to the image.

As may be seen from Fig. 4, it is not possible to find any difference between the surface structure inside and outside the cavity. But the root-mean-square roughness was slightly higher for the inner region. The roughness was calculated for regions with an area of 0.11 μm². The roughness parameters are listed in Table S1, ESI.†

Since the mica surface is atomically smooth, its roughness is determined mainly by the noise of the microscope. The noise level doesn’t exceed 0.1 nm. As the surface roughness inside the pores is an order of magnitude larger than the roughness of the pure mica, we suggest that a very thin wetting layer of 1 remains on the mica surface.

The stability of a thin film is governed by the dependence of the effective interface potential on the film thickness.¹⁵ The presence of the wetting layer on the mica surface indicates the existence of a global minimum of the effective interface potential at critical thickness equal to the thickness of the layer inside the cavity. The root-mean-square roughness of the wetting layer of 1.7 nm may be considered as a lower-bound estimation for the critical thickness.

The lateral force distribution (Fig. S3b, ESI†) shows that the areas inside the large cavities are characterized by a higher value of the lateral forces. One of the possible reasons of the high lateral forces may be the increase of the surface roughness inside the cavities. Moreover, mica becomes negatively charged in water solutions. The smaller the thickness of the copolymer film, the weaker it screens the charge. The additional electrostatic interaction between the AFM probe and the surface also can lead to stronger lateral forces.

Below, we will clarify some details of the hole growth in very thin film B (h ∼ 10 nm). It is known that thin films are unstable to hole formation.¹⁶ As the high regularity of cavities is observed in the AFM image of film B (Fig. 3a), the nucleation of holes is initiated by a spontaneous (spinodal) process rather than by the process of hole nucleation at some defects. In this case, dispersion forces are at play that results in the unstable form of a thin film with a very high surface-to-volume ratio, which in turn causes the hole to grow.¹⁶ Here, it is pertinent to mention that 1 contains N-(2-dibenzylamino-ethyl)-acrylamide, which has hydrophobic benzyl moieties. A scheme of the cavity formation is shown in Fig. S4, ESI.†

After water evaporation, the mobility of the macromolecules of 1 becomes significantly smaller, which prevents further changes in the morphology of the film. The film demonstrated good long-term stability: the AFM studies were not able to reveal any alterations of its morphology during the storage under ambient conditions (see Fig. S5, ESI†). Additionally, it should be noted that 1 also retained its bioactivity manifested in the capability to form filamentous complexes with DNA (see Fig. S6a and b, ESI†).

Trapping DNA using nanostructured film B

To clarify the ability of nanocavities, formed at the surface of film B, to trap DNA molecules, we dripped a drop of DNA water solution on the surface of film B. As Fig. 5a and b show, convex particles with a height of about 5.8 ± 1.5 nm are clearly visible instead of the cavities. The diameter of the particles (318 ± 39 nm) and their surface density (150/100 μm²) are just the same as for the cavities. Maps of lateral force distribution were obtained simultaneously with topographic images (see Fig. S7a and b, ESI†). Their analysis revealed a decrease of lateral forces in the areas of the particle location. Thus, the trapped particles demonstrate a different nanotribological response in comparison to those of the 1-based film.

Fig. 5 A typical AFM image of the surface of film B over a mica crystal with trapped DNA molecules, with a frame size of 10.1 × 10.1 μm² (a), and a 3D AFM image of one of the particles (b).

The observed changes of both film morphology and lateral force distribution may be explained in terms of electrostatic trapping of DNA molecules by the film cavities. The film-forming molecules of 1 contain tertiary amine groups that are usually protonated in water solutions.¹⁷ Thus, the surface of a 1-based film is positively charged when a DNA water solution comes into contact with it. However, DNA molecules are negatively charged in water solutions.¹⁸ Therefore, in the presence of water molecules, the positively charged surface of the 1-based film is able to trap the negatively charged DNA molecules using electrostatic interactions. The cavities are the most preferable sites for DNA binding because their surface geometry is able to provide the greatest number of electrostatic interactions (Fig. 6). By our assumption, the water evaporation may lead to the formation of compact DNA-based particles, which are mainly located at the film cavities.

Fig. 6 Scheme of the DNA trapping by the cavities of film B.

Surface morphology of film C prepared from the most diluted solution

To further decrease the film thickness and increase the surface-to-volume ratio, 1 was deposited from the most diluted solution (≅0.003 mg ml⁻¹) over a mica crystal (film C). Its AFM investigation revealed the formation of cavities with an average diameter of about 50 nm (Fig. 7a and b). The details of the cavity morphology are listed in Table 1. Their shape deviates significantly from the form of a circle. By our assumption, this effect is related to the partial coalescence of the cavities. It is interesting to note that their surface density is significantly higher for film C in comparison to that for film B, which indicates a strong increase in the amount of cavity nucleation centers in the thinner film. The thickness of film C is about 1.5 nm.

Fig. 7 A typical AFM image of the surface of film C cast over mica, with a frame size of 1.5 × 1.5 μm² (a), and a 3D AFM image of one of the cavities (b).

Our attempts to prepare a film over the mica surface from more diluted solutions of 1 resulted in its self-assembly as separated “islands” (see S8, ESI†).

As the mica surface is hydrophilic,^19,20 the hydrophobic components of 1 might significantly reduce the stability of the film due to increasing the interface potential between the film and the mica surface. If this is the case, there are bound to be substrates which, thanks to their hydrophobic properties, will be able to decrease the interface potential between the 1-based film and a substrate surface that in turn may stabilize the thin 1-based films.

Surface morphology of film D cast over a hydrophobic support

To clarify the additional ability to control over nanostructured surfaces of bioactive 1-based films, we prepared film D over a graphite surface from the same water solution that was used for film B casting.

Film D on the graphite surface (Fig. 8a and b) has also cavities, but the mean pore diameter is 111 nm (the standard deviation of the average diameter is 80 nm), which is three times smaller than that of the mica substrate (film B, Fig. 3). Moreover, as Fig. 8 shows, the cavity density is slightly lower in comparison to that observed for the film prepared over a hydrophilic mica surface.

Fig. 8 A typical AFM image of the surface of film D over graphite, with a frame size of 11.1 × 11.1 μm² (a), and a 3D AFM image of one of the cavities (b).

There is a difference between the values of the pore depth in the trace and the retrace (forward and backward direction) for a line in the AFM images of film D. According to our assumption, the effect is caused by the twist motion of the cantilever beam as it passes the surface regions with sharp changes in topography. Taking into account the twist motion of the cantilever beam, the pore depth is calculated as half the sum of the pore depths measured in the trace and the retrace images. The depth, averaged over all found pores, is 13 nm; the standard deviation of the depth is 10 nm.

Cavities in the film were created for additional control of the film thickness. For this purpose, a preliminary scan was carried out in resonant mode. Then, a small area was selected and scanned in contact mode with a large force (∼50 nN). As a result, the upper surface layers were destroyed and a cavity was formed. The surface structure of graphite was preserved inside a cavity with a depth of ∼8 nm, whereas the graphite structure was changed inside a cavity with a depth of ∼20 nm, which indicated the damage of the substrate. Thus, we concluded that the film thickness of the copolymer lies in the range of 8–20 nm. The thickness of film B lies in the same range.

Cleavage steps were not observed in the images of the 1-based film, which was cast from a solution with a higher copolymer concentration, indicating the large thickness of the copolymer film. Cleavage steps with a height of 1 nm are visible for a 1-based film, prepared from a solution with a concentration of 0.03 mg ml⁻¹.

Conclusions

Our study shows that a dewetting process can be successfully utilized as a simple procedure for engineering surface morphologies of bioactive films on a nanoscale, in which the copolymer acrylamide/N-(2-dibenzylamino-ethyl)-acrylamide is at play.

A protocol of the film casting, which permits to reproducibly prepare 1-based films with various regular cavity-like nanostructured surfaces, was developed.

The proof-of-concept experiments demonstrate the possibility of DNA trapping by the nanocavities formed at the surfaces of the 1-based films.

Thus, the control over the morphology of the 1-based films paves the way for their use in biotechnological applications. Furthermore, the sensitivity of 1 to biomacromolecules makes it a useful system for fundamental studies.

Acknowledgements

(1) The work was supported by the Agreement with Ministry of Education and Science of Russian Federation. (2) E.L. thanks CIBER-BBN, an initiative funded by the VI National R&D&i Plan 2008–2011, Iniciativa Ingenio 2010, Consolider Program, CIBER Actions and financed by the Instituto de Salud Carlos III with assistance from the European Regional Development Fund.

References

1. http://www.diabetesnet.com/beta-cell-transplants.
2. G. Orive, R. M. Hernández, A. R. Gascón, R. Calafiore, T. M. S. Chang, P. de Vos, G. Hortelano, D. Hunkeler, I. Lacík, A. M. J. Shapiro and J. L. Pedraz, Nat. Med., 2003, 9(1), 104–107.
3. S. Howorka and Z. Siwy, Chem. Soc. Rev., 2009, 38(8), 2360–2384.
4. J. Shim, G. I. Humphreys, B. M. Venkatesan, J. M. Munz, X. Zou, C. Sathe, K. Schulten, F. Kosari, A. M. Nardulli, G. Vasmatzis and R. Bashir, Sci. Rep., 2013, 3, 1389.
5. R. M. M. Smeets, S. W. Kowalczyk, A. R. Hall, N. H. Dekker and C. Dekker, Nano Lett., 2009, 9, 3089–3095.
6. J. Shim, J. A. Rivera and R. Bashir, Nanoscale, 2013, 5, 10887–10893.
7. O. A. Saleh and L. L. Sohn, Nano Lett., 2003, 3(1), 37–38.
8. P. Petrov, E. Petrova and C. B. Tsvetanov, Polymer, 2009, 50, 1118–1123.
9. P. Petrov, D. Momekova, B. Kostova, G. Momekov, N. Toncheva-Moncheva, C. B. Tsvetanov and N. Lambov, J. Controlled Release, 2010, 148(1), e81–e82.
10. P. Petrov and C. B. Tsvetanov, Polymeric Cryogels: Macroporous Gels with Remarkable Properties, Series: Advances in Polymer Science, 2014, vol. 263, pp. 199–222.
11. B. Daglar, E. Ozgur, M. E. Corman, L. Uzun and G. B. Demirel, RSC Adv., 2014, 4, 48639–48659.
12. N. K. Davydova, O. V. Sinitsyna, I. V. Yaminsky, E. V. Kalinina and K. E. Zinoviev, Macromol. Symp., 2012, 321–322(1), 84–89.
13. N. K. Davydova, O. V. Sinitsyna and K. E. Zinoviev, AIP Conf. Proc., 2012, 1459, 220–222.
14. P. Müller-Buschbaum, J. Phys.: Condens. Matter, 2003, 15, R1549.
15. O. Bäumchen and K. Jacobs, J. Phys.: Condens. Matter, 2010, 22, 033102.
16. K. Dalnoki-Veress, J. R. Dutcher and J. A. Forrest, Phys. Can., 2003, 59(2), 75–84.
17. M. B. Smith and J. March, March’s Advanced Organic Chemistry: Reactions, Mechanisms, and Structure, Wiley, New Jersey, 6th edn, 2007.
18. M. J. Tiera, Q. Shi, F. M. Winnik and J. C. Fernandes, Curr. Gene Ther., 2011, 11(4), 288–306.
19. D. H. Bangham and Z. Saweris, Trans. Faraday Soc., 1938, 34, 554–570.
20. J. Drelich, E. Chibowski, D. D. Meng and K. Terpilowski, Soft Matter, 2011, 7, 9804–9828.

---

Footnote

† Electronic supplementary information (ESI) available: Roughness parameters for the surfaces inside and outside the typical cavity. Film B thickness measurement. Typical AFM image of the surface of the film B cast over a mica crystal showing large cavities. Images of topography and lateral force distribution on film B. Scheme of the cavity formation in film B. Typical AFM image of the surface of film B stored six months under ambient conditions. Typical AFM image of the complex between λ-phage DNA and copolymer 1 formed on the mica surface (the sample of copolymer 1 was synthesized six months before the DNA trapping experiment) and schematic view of the complex formation. Images of topography and lateral force distribution on film B with trapped DNA molecules. Typical AFM image of “islands”, which are formed on a mica crystal from a very diluted water solution of copolymer 1. See DOI: 10.1039/c4ra11748d

---