Biotemplated synthesis of stimuli-responsive nanopatterned polymer brushes on microtubules

Abstract

We report the synthesis of nanostructured stimuli-responsive polymer brushes using atom transfer radical polymerisation initiated on protein filaments of the cytoskeleton. In particular, we used microtubules and prepared thermoresponsive poly-(N-isopropylacrylamide) brushes with incorporated fluorescent groups. The use of microtubules as templates for fabrication of “soft” polymeric nanostructures on surfaces opens new possibilities for the design of functional materials.

---

Nanostructured surfaces are of growing interest for information storage,¹ microfluidics,² design of smart materials,³ biotechnology,⁴ microelectronics,⁵ photonic applications⁶etc. Thereby, the use of biological objects as templates for the design of the nanostructures offers a range of advantages.⁷ In particular, selective molecular recognition and proof-reading during the build-up of biomolecules can provide a high degree of uniformity in the possible structures. Different biomacromolecules and their assemblies including DNA, viral capsids, cytoskeleton filaments and protein crystals were successfully applied as templates for the fabrication of inorganic nanostructured materials (see recent review⁷). On the other hand, design of functional organic nanostructures^8–10 using biotemplating is still an almost unexplored field.

Microtubules, which are cylindrical protein filaments with outer diameters of about 24 nm and lengths up to tens of micrometres, deserve a particular interests as templates.^11–14 First, microtubules can be easily formed by self-assembly of tubulin dimmers. Second, microtubules are able to form segmented structures with different functionality in a controlled manner.¹⁵ Third, a number of approaches to control position as well as orientation of microtubules on artificial substrates are available.^16–20 This makes microtubules highly promising for templated synthesis of complex materials.

Here, we demonstrate the fabrication of stimuli-responsive nanopatterned polymer brushes using microtubules as templates. In particular, we report on the “grafting from” synthesis of thermoresponsive poly-(N-isopropylacrylamide)—PNIPAM—brushes by atom transfer radical polymerisation (ATRP) initiated on microtubules. The outer surfaces of microtubules contain reactive amino groups providing the possibility for chemical modification with fluorescent and other functional groups. In our approach, we substituted the amino groups with reactive groups capable of initiating ATRP and grew polymer chains from surface-adsorbed microtubules (Fig. 1).

Fig. 1 Schematic diagram of microtubule modification and polymer grafting. The polymer chains were grown from microtubules that were adsorbed on a hydrophobic glass substrate and sequentially crosslinked with glutaraldehyde, amino-functionalised using 2,2′-(ethylenedioxy)bis(ethylamine) and modified by 2-bromo-2-methylpropanoyl bromide.

Rhodamine-labelled microtubules (fluorescent emission in the red wavelength range) were prepared by self-assembly of α,β-tubulin dimers (see Experimental part). The microtubules were then adsorbed on DDS-coated glass surfaces by perfusing them in buffer solution through a narrow channel between two glass cover slips. The adsorbed microtubules were crosslinked by glutaraldehyde, which reacted with lysine residues and formed chemical links between neighbouring amino groups (Fig. 2a). Because part of the glutaraldehyde molecules did not react bifunctionally with the lysine residues, the outer surface of the crosslinked microtubules then contained aldehyde groups. Initiator groups were then immobilized onto the crosslinked microtubules by sequentially treating them with 2,2′-(ethylenedioxy)bis(ethylamine) and bromo-2-methylpropanoyl bromide (see Experimental part for details). The initiator-modified microtubules possessed a rod-like shape, which remained undeformed after multiple rinsings in different organic solvents (Fig. 2b). Using the simplified ATRP developed by Matyjaszewski et al.,²¹ we demonstrated the polymerisation of N-isopropylacrylamide with small addition (2 wt%) of fluorescein o-acrylate (Fig. 2c). Most strikingly, the diameters of the polymer-modified microtubules were substantially increased and their shape was often deformed.

Fig. 2 Morphology of microtubules at different stages of the modification procedure. AFM images (middle panel) of microtubules (a) after crosslinking with glutaraldehyde, (b) after immobiliztion of initiator, and (c) after grafting of poly-(N-isopropylacrylamide–fluorescein o-acrylate) (M_n = 69 000 g mol⁻¹, M_w = 152 000 g mol⁻¹). Cross-sections (bottom panel) are given along the paths indicated with the corresponding colour on the AFM images. No tip-deconvolution was performed.

The role of fluorescein o-acrylate (fluorescent emission in the green wavelength range) was to make the synthesized polymer brushes detectable by fluorescence microscopy. Verifying the successful synthesis of the polymer chains on the microtubule we found that the originally “red-fluorescent” microtubules became additionally “green-fluorescent” after polymerisation (Fig. 3).

Fig. 3 Epi-fluporescent images of surface-adsorbed microtubules after immobiliztion of initiator (a: microtubule signal, b: polymer signal) and after grafting of poly-(N-isopropylacrylamide–fluorescein o-acrylate) (M_n = 59 000 g mol⁻¹, M_w = 136 000 g mol⁻¹) (c: microtubule signal, d: polymer signal).

In order to investigate the structural changes of the microtubule in detail, we imaged the same area of interest by AFM before and after the polymerisation (Fig. 4). These measurements revealed that the contour length of the microtubules increased by up to 100%. On the other hand, the end-to-end distance of the surface-attached microtubules remained almost unchanged. We attribute the increase of the contour length to stretching within and between the tubulin dimers caused by steric repulsions between the grafted polymer chains rather than to high temperature and interactions with solvents.²²

Fig. 4 AFM images of microtubules in the same area of interest before (a, b) and after (c, d) grafting of poly-(N-isopropylacrylamide–fluorescein o-acrylate) (M_n = 47  000 g mol⁻¹, M_w = 122 000 g mol⁻¹). No tip-deconvolution of AFM images was performed.

We estimated the grafting densities of the polymer chains per microtubule length (Γ_length) and area (Γ_area) using the following equations:²³

Γ_area = Γ_length/C_SEC (2)

where S_SEC is the cross-sectional area of the polymer-modified microtubule, ρ_PNIPAM is the polymer mass density, M_w is the weight averaged molecular weight, N_A is the Avogadro constant, and C_SEC is the arc-length over the surface of a given section cut perpendicularly to the microtubule. Thereby S_SEC is derived experimentally as the integrated area of the cross-section of a polymer-modified microtubule as shown in Fig. 2c (lowest panel). Here, the contribution from the microtubule before the polymerisation is neglected. C_SEC is determined as the circumference of the initiator-modified microtubule as shown in Fig. 2b (lowest panel). For the estimation of S_SEC and C_SEC the AFM curves have been corrected for the tip radius. We derived grafting densities of Γ_length ≈ 40 chains nm⁻¹ (Γ_area ≈ 0.3 chains nm⁻²) and Γ_length ≈ 42 chains nm⁻¹ (Γ_area ≈ 0.32 chains nm⁻²) for the data presented in Fig. 2 and 4, respectively. The average distance D between grafting sites could then be calculated by D = Γ_area^−1/2 ≈ 1.8 nm. This value is substantially smaller than the gyration radius of PNIPAM chains with similar molecular weights in the collapsed state (R_g ≈ 20 nm).²⁹ Consequently, the polymer-grafted shell can be considered as a brush-like one.

We explored the correlation between the grafted amount of polymer chains and the number of amino acid residues, which were able to react with α-bromoisobutyryl bromide initiator for the polymerisation. Microtubules are composed of about 13 protofilaments made up of α,β-tubulin dimers (repeat length 8 nm). Each α,β-tubulin dimer contains 34 lysine, 42 arginine and 50 serine amino acid residues, corresponding to N ≈ 200 of potential grafting sites per nanometre of microtubule length. On the other hand, since the contour length of microtubules increased after polymerisation, the apparent density of potential grafting sites decreased by a factor of L_C-POL/L_C, where L_C-POL and L_C are the contour lengths of the microtubules after and before polymerisation. The apparent density of potential grafting sites N_POL was found to be N_POL = 130 nm⁻¹ (data from Fig. 2) and N_POL = 100 nm⁻¹ (data from Fig. 4) meaning that about Γ_length/N_POL = 30%–40% reactive amino acid residues initiated growth of polymer chains²³

Finally, we investigated the switching behaviour of the grafted poly-(N-isopropylacrylamide) brushes in an aqueous environment. We found that the fluorescence intensity of the polymer chains (obtained from images as in Fig 3d) gradually decreased with increasing temperature (Fig. 5), reaching almost zero at the low critical solution temperature of PNIPAM (T = 33 °C). The reason for this effect was, most probably, fluorophore quenching by the collapsed polymer chains^24–26 indicating that the polymer chains were capable of switching. Given the optical resolution of our imaging system, we did, on the other hand, not detect any morphological changes of the PNIPAM-decorated microtubules as a result of the polymer swelling and collapse.

Fig. 5 Switching of PNIPAM grafted on microtubules. Dependence of fluorescence intensity of poly-(N-isopropylacrylamide–fluorescein o-acrylate) brushes on temperature upon heating (open squares, solid line) and cooling (open circles, dashed line).

In conclusion, we presented a novel approach for the design of nanopatterned polymer brushes based on the use of biological templates. Using microtubules as templates, we fabricated thermoresponsive nanopatterned poly-(N-isopropylacrylamide) brushes which can be further used for design of responsive material systems²⁷ and biomolecular switches.²⁸ We believe that microtubule-based polymeric materials will be of interest for a variety of nanotechnological and microelectronic applications.

Experimental part

Materials

2,2′-(Ethylenedioxy)bis(ethylamine) (Aldrich), glutaraldehyde (Sigma-Aldrich), N-isopropylacrylamide (NIPAM, Aldrich), acetone (Aldrich), N,N,N′,N″,N″-pentamethyldiethylenetriamine (PMDTA, Aldrich), ethyl-2-bromoisobutyrate (EbiB, Aldrich), ethylenediamine (ED, Fluka), dimethyldichlorosilane (DDS, Fluka), anhydrous dichloromethane (Aldrich), 2-bromo-2-methylpropanoyl bromide (BMPB, Aldrich), triethylamine (Fluka), L-ascorbic acid (Sigma), fluorescein o-acrylate (FA, Aldrich), copper (II) bromide (Aldrich) were used as received.

Preparation of DDS-coated glasses

Piranha-cleaned glass substrates were treated by DDS (125 µL) solution in 250 mL of trichloroethylene over a period of 60 min. The DDS-coated glasses were rinsed several times in methanol and water. DDS was used to prevent the adsorption of ethylenediamine and immobilization of initiator on the glass substrate.

Preparation of microtubules

Rhodamine-labelled microtubules were grown from 10 µL of bovine brain tubulin (4 mg mL⁻¹, mixture of 1 rhodamine-labelled/3 unlabelled tubulin units) in BRB80 buffer (80 mM potassium PIPES, pH 6.9, 1 mM EGTA, 1 mM MgCl₂) with 4 mM MgCl₂, 1 mM Mg-guanosine 5′-triphosphate sodium salt, and 5% DMSO at 37 °C. After 30 min, the microtubules were stabilized and 100-fold diluted in room-temperature BRB80 containing 10 µM taxol.

Modification of microtubules

Grown microtubules were purified from free tubulin by centrifuging at 40 000 rpm for 10 min. The supernatant was substituted with 10 µM taxol solution in BRB80 buffer. The microtubules were perfused and adsorbed in 2 mm wide flow cells self-built from two DDS-coated coverslips (Corning, 22 × 22 µm² and 18 × 18 mm²) and two pieces of double-sided sticky tape (Scotch 3M, thickness 0.1 mm). The non-adsorbed microtubules were washed out by perfusing 10 µM taxol in BRB80. The 1% and 3% solution of glutaraldehyde in 10 µM taxol in BRB80 were subsequently perfused for 5 min in the flow cell to crosslink the microtubules. Final crosslinking was performed by 25% water solution of glutaraldehyde over a period of 20 min. The crosslinked microtubules were subsequently treated with 100% 2,2′-(ethylenedioxy)bis(ethylamine) for 20 min. The initiator was immobilized on microtubules from a solution of BMPB (0.96 g, 4.2 × 10⁻³ mol) and triethylamine (0.72 g, 7 × 10⁻³ mol) in anhydrous dichloromethane (100 mL) over a period of 2 h. The modified microtubules were rinsed in dichloromethane 2×, ethanol 2× and dried with nitrogen flux.

Grafting of polymer on microtubules

Polymer was grafted on initiator-modified microtubules as follows: NIPAM (2 g, 1.75 × 10⁻² mol), FA (40 mg, 1 × 10⁻⁴ mol), EBIB (0.65 mg, 3.3 × 10⁻⁶ mol), CuBr₂ (0.35 mg, 1.6 × 10⁻⁶ mol), PMDTA (1.3 mg, 8 × 10⁻⁶ mol) were dissolved in 2 ml of acetone. The reaction solution was added to the tube containing the DDS-glass substrate with adsorbed initiator-modified microtubules. After the tube was sealed with a rubber septum, a solution of L-ascorbic acid (18 mg, 1 × 10⁻⁴ mol) in water (0.06 mL) was injected. The vial was placed in a 70 °C oil bath. The polymerisation was stopped after 1 h. The polymer obtained in solution was collected by precipitation in diethyl ether and used for determination of molecular weight. It is believed that these values should be close to those of the grafted polymer brushes. The substrates with polymer-modified microtubules were rinsed several times in hot chloroform and acetone. DMAC–2% H₂O–LiCl (3 g L⁻¹) was used as eluent for GPC.

Epi-fluorescent microscopy

Fluorescence images were obtained using an Axiovert 200M inverted microscope with a 40× objective (Zeiss, Oberkochen, Germany) equipped with FluoArc lamp. For data acquisition standard TRITC (excitation: HQ 535/50; dichroic: Q 565 LP; emission: HQ 610/75, Chroma Technology) and FITC (excitation: HQ 480/40; dichroic: Q 505 LP; emission: HQ 535/50, Chroma Technology) filter sets in conjunction with a Micromax 512 BFT camera (Photometrics, Tucson, AZ) and a MetaMorph imaging system (Universal Imaging, Downingtown, PA) were used.

Atomic force microscopy

AFM studies were performed with a Dimension 3100 (Digital Instruments, Inc., Santa Barbara, CA) microscope. Tapping mode was used to map the film morphology at ambient conditions.

Acknowledgements

M. Stamm, A. Synytska (IPF Dresden) and T. Korten are acknowledged for fruitful discussions and careful reading of the manuscript. Petra Treppe (IPF Dresden) is acknowledged for technical support. This work was supported by the BMBF (Grant 03N8712), the Volkswagen Foundation and the Max-Planck-Society.

References

1. H. Sato and T. Homma, J. Nanosci. Nanotechnol., 2007, 7, 225–231.
2. L. Ionov, N. Houbenov, A. Sidorenko, M. Stamm and S. Minko, Adv. Funct. Mater., 2006, 16, 1153–1160.
3. A. Sidorenko, T. Krupenkin, A. Taylor, P. Fratzl and J. Aizenberg, Science, 2007, 315, 487–490.
4. M. J. Dalby, N. Gadegaard, R. Tare, A. Andar, M. O. Riehle, P. Herzyk, C. D. W. Wilkinson and R. O. C. Oreffo, Nat. Mater., 2007, 6, 997–1003.
5. E. Menard, M. A. Meitl, Y. Sun, J. U. Park, D. J. L. Shir, Y. S. Nam, S. Jeon and J. A. Rogers, Chem. Rev., 2007, 107, 1117–1160.
6. A. Boker, J. He, T. Emrick and T. P. Russell, Soft Matter, 2007, 3, 1231–1248.
7. S. Sotiropoulou, Y. Sierra-Sastre, S. S. Mark and C. A. Batt, Chem. Mater., 2008, 20, 821–834.
8. C. Reuther, L. Hajdo, R. Tucker, A. A. Kasprzak and S. Diez, Nano Lett., 2006, 6, 2177–2183.
9. Z. Niu, J. Liu, L. A. Lee, M. A. Bruckman, D. Zhao, G. Koley and Q. Wang, Nano Lett., 2007, 7, 3729–3733.
10. M. Young, D. Willits, M. Uchida and T. Douglas, Annual Review of Phytopathology, 2008, 46, 361–384.
11. C. Brunner, C. Wahnes and V. Vogel, Lab Chip, 2007, 7, 1263–1271.
12. S. Behrens, K. Rahn, W. Habicht, K. J. Bohm, H. Rosner, E. Dinjus and E. Unger, Adv. Mater., 2002, 14, 1621–1625.
13. S. Hiyama, T. Inoue, T. Shima, Y. Moritani, T. Suda and K. Sutoh, Small, 2008, 4, 410–415.
14. M. Platt, G. Muthukrishnan, W. O. Hancock and M. E. Williams, J. Am. Chem. Soc., 2005, 127, 15686–15687.
15. G. D. Bachand, S. B. Rivera, A. K. Boal, J. Gaudioso, J. Liu and B. C. Bunker, Nano Lett., 2004, 4, 817–821.
16. L. Ionov, M. Stamm and S. Diez, Nano Lett., 2006, 6, 1982–1987.
17. F.-U. Gast, P. S. Dittrich, P. Schwille, M. Weigel, M. Mertig, J. Opitz, U. Queitsch, S. Diez, B. Lincoln, F. Wottawah, S. Schinkinger, J. Guck, J. Käs, J. Smolinski, K. Salchert, C. Werner, C. Duschl, M. S. Jäger, K. Uhlig, P. Geggier and S. Howitz, Microfluid. Nanofluid., 2006, 2, 21–36.
18. M. G. L. van den Heuvel, M. P. De Graaff and C. Dekker, Science, 2006, 312, 910–914.
19. B. M. Hutchins, M. Platt, W. O. Hancock and M. E. Williams, Small, 2007, 3, 126–131.
20. R. Tucker, P. Katira and H. Hess, Nano Lett., 2008, 8, 221–226.
21. K. Matyjaszewski, H. C. Dong, W. Jakubowski, J. Pietrasik and A. Kusumo, Langmuir, 2007, 23, 4528–4531.
22. We found that the morphology of the glutamate-crosslinked microtubules with immobilized initiator did not change after boiling in acetone.
23. The estimation is based on the assumption that the mass density of grafted polymer is identical to that of the bulk polymer. Moreover, precise evaluation of the grafting density is hampered by the polydispersity of the polymer chains. Use of M_n instead of M_w reduces the calculated grafting density by factor of M_w/M_n ≈2.2.
24. S. Morita, F. Tsunomori and H. Ushiki, Eur. Polym. J., 2002, 38, 1863.
25. Intensity of fluorescein acrylate slightly increases with temperature.
26. The measured decrease in fluorescence intensity was not related to a direct temperature dependence of the fluorescence as we did not find such dependence in control experiments (data not shown).
27. M. Kaholek, W. K. Lee, S. J. Ahn, H. W. Ma, K. C. Caster, B. LaMattina and S. Zauscher, Chem. Mater., 2004, 16, 3688–3696.
28. J. Hyun, W. K. Lee, N. Nath, A. Chilkoti and S. Zauscher, J. Am. Chem. Soc., 2004, 126, 7330–7335.
29. X. H. Wang and C. Wu, Macromolecules, 1999, 32, 4299.

---