Directional Neurite Outgrowth on Superaligned Carbon Nanotube

Jun 13, 2012 - silicon wafer, which consists of vertically superaligned multi- ... Figure 1. Procedure of making superaligned CNT yarn patterned subst...
0 downloads 0 Views 402KB Size
Download PDF
Letter pubs.acs.org/NanoLett

Directional Neurite Outgrowth on Superaligned Carbon Nanotube Yarn Patterned Substrate Li Fan,*,† Chen Feng,‡ Wenmei Zhao,‡ Li Qian,‡ Yuquan Wang,‡ and Yadong Li† †

Department of Chemistry, Tsinghua University, Beijing 100084, China Tsinghua-Foxconn Nanotechnology Center, Tsinghua University, Beijing 100084, China

‡

ABSTRACT: Superaligned carbon nanotube (CNT) yarn patterned substrates were developed as the topographic scaffold for guiding the neurite outgrowth. As-prepared patterned substrates were used for culturing rat hippocampal neurons, without purifying and functionalizing processes on the CNTs. The neurite outgrowth on the patterned substrate exhibited a strong tendency to being aligned along the CNT yarns long axes. The neurite grown along the CNT yarns had much less branching than the one on a uniform planar substrate typically used for neuron culture. These results indicate that the pure CNT yarns possess the main characteristics of a guidance scaffold for neurite outgrowth. Furthermore, the CNT yarns can be mass produced and be easily weaved into desired structures, which may make them attractive for neuronal regeneration and tissue engineering. KEYWORDS: carbon nanotube, scaffold, directional neurite outgrowth, minimizing branching, neural regeneration

T

main function of scaffold is guiding neurits to their proper target sites, and excessive neurite branching might negatively influence the accuracy of target reinnervation and should be minimized.4,18−22 Our results indicate the CNT yarn is of the main characteristics of the scaffold for neural regeneration. Moreover, less pretreating need to be used on the CNT yarn patterned substrates, because the CNT materials we used are free of metal catalyst remainders. In our experiment, the superaligned CNT yarn patterned substrates used for neuron growth were made of the superaligned CNT film, developed by Jiang et al.23,24 As shown in Figure 1a, the suspended pure superaligned CNT film is directly drawn out from the carbon nanotube arrays on silicon wafer, which consists of vertically superaligned multiwalled carbon nanotubes with the diameters around 10 nm.24 SEM picture (Insert of Figure 1a) shows that the superaligned CNT film consists of the parallel-aligned thin carbon nanotube wires along the drawing direction. By spraying ethanol upon the suspended superaligned CNT film, the carbon nanotube wires in the film were further shrunk and self-assembled into the superaligned CNT yarns with the diameters of 2−5 μm and spaced at intervals of 30−50 μm, as shown in Figure 1b. The suspended superaligned CNT yarns can be directly laid down to a polystyrene sheet to form superaligned CNT yarns/ polystyrene substrate. Figure 1b shows the procedure to make two different patterns of superaligned CNT yarns on polystyrene: the parallel-aligned CNT yarn patterned substrate (PAP) and the cross-linked CNT yarn patterned substrate

he regeneration of nervous systems is a complex pathological and physiological process. Neurons of the central nervous system (CNS) are much more difficult to regenerate than neurons of the peripheral nervous system (PNS). Accordingly, current researches on CNS regeneration are mainly focused on two approaches. One is to study how to improve the intrinsic neurite growth ability, the other is to study how to change the micro environment to help the regeneration of CNS. Some researches have shown that scaffolds by some materials are one of the critical factors to change the micro environment in PNS and CNS, and affect the ability of injured peripheral and central neurons to generate an effective growth response.1 It has been reported that aligned polymer based nanofiber substrate could provide better topographic guidance effects toward directional neurite outgrowth.2−6 Since the discovery of the carbon nanotubes (CNTs),7 there have been growing interests in using CNTs as the scaffold for neuronal growth, due to their unique structural and physical properties, such as high aspect ratio, strong yet flexibility, high conductivity, and chemical stability.8−17 However, purifying carbon nanotubes to increase biocompatibility and aligning carbon nanotubes to enhance the topography guidance effect on the neurite outgrowth still remain challenges. In this report, we developed a unique and controllable method to make superaligned CNT yarn patterned substrates as the topographic scaffold for guiding the neurite outgrowth. The developed superaligned CNT yarn patterned substrates were used to grow rat hippocampal neurons in vitro. It has been observed that the superaligned CNT yarns presented strong guidance effects on neurite outgrowth of the rat hippocampal neurons and can also minimize branching of neurites growing along the yarns. In neural regeneration, the © 2012 American Chemical Society

Received: April 16, 2012 Revised: May 28, 2012 Published: June 13, 2012 3668

dx.doi.org/10.1021/nl301428w | Nano Lett. 2012, 12, 3668−3673

Nano Letters

Letter

Figure 1. Procedure of making superaligned CNT yarn patterned substrates: (a) The CNT film is drawn out from the vertical CNT arrays on the silicon wafer. The insert SEM image shows the film consists of aligned CNT wires. (b) Superaligned CNT yarns, which were self-assembled from the CNT film by spraying ethanol on the film, the diameters of the yarn are 2−5 μm and spaced at intervals of 30−50 μm. (c) Procedure to make patterned CNT yarns/polystyrene substrate. Parallel black solid lines represent carbon nanotube yarns. Gray sheet represents polystyrene film.

were soaked in 2 mL 20 μg/mL poly D-lysine solution for 2−8 h, and then washed by deionized water before experiment. No further functionalization was performed on the culture substrates. The hippocampal neurons were obtained from the embryonic rats (E18) by using the method similar to one previously described by Banker and Cowan.26 Hippocampal tissues from 18 days fetal rats were dissected and treated with 0.25% trypsin (GIBCO, USA) at 37 °C for 15−20 min. They were then dissociated by trituration using a glass Pasteur pipet. Approximately 100 000 cells were planted into each culture dish with culture DMEM (GIBCO, USA) containing 5% fetal bovine serum and 5% horse serum (HYCLONE, USA). The total culture medium in each dish is about 2 mL. Then the culture dishes were put into a CO2 incubator maintained at 37 °C. On the second day after planting, the culture medium was replaced by a serum-free Neurobasal medium (GIBCO, USA) containing B27 supplement (GIBCO, USA) and 500 μM Glutamax (GIBCO, USA) for reducing glial growth. Afterward, one-half of the volume of the Neurobasal medium was replaced with fresh Neurobasal medium every 2 days. The confocal microscopy (Nikon A1RSi+90i) was performed to visualize the cultured neurons. In order to label the cultured hippocampal neurons, 200 μL Calcein-AM (30 μg/mL) (DojinDO, Japan) solution was added into the culture medium in each culture dish, which then was incubated at 37 °C in 5% CO2 for 25 min. Afterward, staining neurons were washed by 2 mL 1 × phosphate buffered saline (PBS, GIBCO, USA) twice. After each washing, the culture dish was incubated at 37 °C in

(CLP), which would be used as the substrate for rat hippocampal neuron growth in the following experiments. The diameters of the yarns are similar with the diameters of typical neurites. This size similarity may be of benefit to enhance the topographic guidance effect of the carbon nanotubes. The intrinsic cytotoxicity of carbon nanotubes is still debatable. However, the harm of residual metal catalyst has highly been noticed.12 Carbon nanotubes are usually made by using the powder catalyst with metals. The raw product is a mixture of carbon nanotubes and catalyst particles, so purification steps for removing residual metal catalysts before biomedical uses are critical and also challenges. As described above, the CNT yarns we used were drawn out from the carbon nanotube arrays on silicon substrate, where the catalyst is a thin metal film physically deposited and firmly bonded to the silicon surface. There is almost no metal particles get in to CNT yarns during the drawing processing. A TGA test has been conducted on the CNTs removed from the CNT arrays. The results show the remainder is only 0.66% in mass after full oxidation of the CNTs. However the remainder is around 20% for a comparison CNT sample made by powder catalyst.25 Based on this result, we will deliberately omit the commonly used purification steps on carbon nanotubes in the following experiments, in order to test the biocompatibility of the pure carbon nanotube materials we used. The rat hippocampal neuron cultures were planted in 35 mm plastic dishes with two types of the patterned CNT yarns/ polystyrene substrates, whose diameter is 28 mm. The plastic culture dishes with the plain polystyrene substrates (o.d. 28 mm) were also used as a control group. The CNT substrates 3669

dx.doi.org/10.1021/nl301428w | Nano Lett. 2012, 12, 3668−3673

Nano Letters

Letter

Figure 2. High-magnification confocal microscopy images taken from the cultured neurons on (a) the parallel-aligned CNT yarn patterned substrate, (b) the cross-linked CNT yarn patterned substrate, and (c) the plain polystyrene substrates, respectively, after a 5-day culture. The solid black lines in parts a and b are the CNT yarns.

of the neurites grow along the direction of nanotube yarns (Figure 3). In particular, almost all long neurites grow exactly

5% CO2 for 3 min. Finally, 2 mL maintaining culture (culture media) was added for visualizing. Figure 2 shows high-magnification confocal microscopy images taken from the cultured neurons on the parallel-aligned CNT yarn patterned polystyrene substrate, the cross-linked CNT yarn patterned polystyrene substrate, and the plain polystyrene substrate respectively, which had been cultured for 5 days in the same conditions. Figure 2c shows the rat hippocampal neurons on the plain polystyrene substrate. It can be seen that the growth directions of the neurites are random, and no preferring direction can be observed. However, the cultured rat hippocampal neurons on the CNT yarn patterned substrates behave total differently, as shown in Figure 2, parts a and b. Figure 2a shows the neurons grown on the substrate with the parallel-aligned CNT yarn pattern, where the black parallel lines are CNT yarns. It is clearly shown that almost all neurites grow along the CNT yarns. This topographic guidance effect of the CNT yarns on the neurite outgrowth has been further demonstrated on the substrate with the cross-linked CNT yarn pattern, as shown in Figure 2b. On the cross-linked CNT yarn patterned substrate, it can be seen that a neurite can grow along one CNT yarn and then takes a turn into another cross-linked yarns, which means neurites might be able to grow along any rationally designed CNT yarn path, or network. We performed a statistics analysis on the directionality of the CNT yarns to the neurite outgrowth. The statistics was conducted based on 700 images randomly selected from 100 cross-linked CNT yarn patterned substrates (o.d. 28 mm) after 3-, 5-, and 7-day cultures. Statistics results show that nearly 80%

Figure 3. Proportion of directional or unidirectional neurites to the total neurites on the cross-linked CNT yarn patterned substrate.

along the direction of CNT yarns. This statistics result further indicates that the CNT yarns have strong guidance effects on neurite outgrowth. Parts a and b of Figure 4 show high-magnification confocal microscopy images taken from the cultured neurons on the parallel-aligned and cross-linked patterned substrate after 7-day culture. Again, it is clear that almost all of neurites outgrowth is along the CNT yarns, the same as in Figure 2. Moreover, from the confocal microscopy images we notice that the long neurite growing along the CNT yarns has little, even no branching. However, neurites growing on the plain substrate surface tend to branch, as shown in Figure 4c. SEM images (Figure 4d and 3670

dx.doi.org/10.1021/nl301428w | Nano Lett. 2012, 12, 3668−3673

Nano Letters

Letter

Figure 4. High-magnification confocal microscopy images taken from the cultured neurons on (a) the parallel-aligned CNT yarn patterned substrate, (b) the cross-linked CNT yarn patterned substrate, and (c) the plain polystyrene substrate, respectively, after 7 days in the same condition. SEM images of hippocampal neurons growing on (d) the plain polystyrene substrate, and (e) the parallel-aligned CNT yarn patterned substrate, after culturing for 3 days in the same experimental conditions.

shows that the proportion of the neurites with no branch is up to 75%. And this proportion almost maintains the same with the increase of culture days. The statistics clearly indicates the branching of the neurites growing along CNT yarns has been minimized. This phenomenon and its significance have been discussed by others.4,18−22 Patel et al. have noticed that neurites following the guidance of nanofibers exhibited little branching on aligned nanofibers, and pointed out that branching of neurites is detrimental to nerve regeneration and should be minimized.4 Guntinas-Lichius et al. pointed out that the reduced axonal branching meant a better axonal pathfinding, which in turn provided an excellent recovery of function.20 We have estimated the viable cell density on the CNT yarn patterned substrates based on the images used for the statistics

4e) revealed more detailed neurite branching behavior of a single cell. The SEM image were taken from the hippocampal neurons growing on the plain polystyrene substrate (Figure 4d) and the parallel-aligned CNT yarn patterned substrate (Figure 4e) respectively after culturing for 5 days in the same experimental conditions. Figure 4d shows the neurites growing on the plain substrate tended to have more branching. However, if a neurite meets a CNT yarn, as shown in Figure 4e, it tends to grow all way along the yarns with no more branching. Figure 5 shows the statistics results of the branching behavior of neuritis growing along the CNT yarns on parallelaligned CNT yarn patterned substrates after 3-, 5-, and 7-day cultures. To choose the parallel-aligned is because that the long neurite along the CNT yarns are easy to be traced. Figure 5 3671

dx.doi.org/10.1021/nl301428w | Nano Lett. 2012, 12, 3668−3673

Nano Letters

Letter

even in vivo. In addition, the superaligned CNT film we used can be produced on large scale and be shrunk (or spun) into aligned CNT yarns in our laboratory. Both CNT films and CNT yarns can be used to fabricate the desired scaffold structures, which may have a wide range of applications in tissue engineering. Overall, a more complete understanding of how CNT films and yarns affect neuronal regeneration in vitro will provide clues and basis for developing improved clinical treatments for nervous system injury in the future.

■

AUTHOR INFORMATION

Corresponding Author

Figure 5. The proportion of branching neurites growing along the CNT yarns on parallel-aligned CNT yarn patterned substrates after 3-, 5- and 7-day cultures.

*E-mail: [email protected]. Telephone: 86-1062772070. Fax: 86-10-62789851. Notes

The authors declare no competing financial interest.

■

of Figure 3. The average viable cell density is around 140/mm2, which is close to the average cell density of 162/mm2 planted into each dish (100 000 cells on o.d. 28 mm substrate). This result may indicate that the neuron cells have high viability on the CNT yarns patterned substrate. We have planted more cells, such as 250 000 and 350 000 cells into each dish. The high cell viability was still maintained, however, too dense tangled neurites made the directivity observation and analysis more difficult. Therefore, the lower cell density samples have been used for the purpose of our present work. A neurite length statistics was conducted on the CNT yarn patterned substrates after 7 day culture, the average neurite length is about 152 μm, and the longest one is up to 1100 μm. Some 20 day cultured samples had been observed, where the neuron cells still remained alive well. All these results indicate that the CNT materials used in our experiment have a good biocompatibility, at least, with the neuronal cells. Comparing to previous commonly used CNTs for cell cultures, the main difference is that CNT yarns we used are free of metal catalysts, as discussed in previous paragraph, and the complicated chemically purifying and pretreating have been avoided. This result may be an additional evidence for the cytotoxicity of metal remainder in common used CNTs.12 Also, less chemical pretreating on the CNT scaffold is expected, because chemically pretreating may damage the CNTs and leave some chemical remainders on the CNT scaffold.27 So far, the CNT yarn is nonbiodegradable. However, in vitro hippocampus neuron culture results shown the CNT yarns have very low cytotoxicity and high biocompatibility, which may be used in vivo as nondegradable implant materials in some special cases. On the other hand, there is an urgent need for developing neuron interface nanomaterials for neural disorder therapies involving drug delivery, tissue repair, and electrical implants.28 The unique properties of the CNT yarns, such as strong, flexible, conductive, easily manufacturable, and high biocompatibility, make them excellent candidates for developing neural interface materials and devices. In conclusion, the superaligned CNT yarn patterned substrates we developed present strong guidance effects on neurite outgrowth of the rat hippocampal neurons. Besides of the guidance effect, the CNT yarns can also minimize branching of neurites growing along the yarns, which is important to neuronal regeneration. Less pretreating being used on the CNT substrate means that CNT yarns have high biocompatibility. All these results indicate that the aligned CNT film and yarns are the good candidates as the synthetic guidance scaffold for neuronal regeneration and repair in vitro,

REFERENCES

(1) Schmidt, C. E.; Leach, J. B. Annu. Rev. Biomed. Eng. 2003, 5, 293− 347. (2) Yang, F.; Murugan, R.; Wang, S.; Ramakrishna, S. Biomaterials 2005, 26, 2603−2610. (3) Schnell, E.; Klinkhammer, K.; Balzer, S.; Brook, G.; Kleeb, D.; Dalton, P.; Mey, J. Biomaterials 2007, 28, 3012−3025. (4) Patel, S.; Kurpinski, K.; Quigley, R.; Gao, H.; Hsiao, B. S.; Poo, M. M.; Li, S. Nano Lett. 2007, 7, 2122−2128. (5) Ghasemi-Mobarakeh, L.; Prabhakaran, M. P.; Morshed, M.; NasrEsfahani, M.; Ramakrishna, S. Biomaterials 2008, 29, 4532−4539. (6) Dahlin, R. L.; Kasper, F. K.; Mikos, A. G. Tissue Engineering Part B: Reviews 2011, 17, 349-364 (7) Ijima, S. Nature 1991, 3, 54−56. (8) Mattson, M. P.; Haddon, R. C.; Rao, A. M. J. Mol. Neurosci. 2000, 14, 175−182. (9) Hu, H.; Ni, Y.; Montana, V.; Haddon, R. C.; Parpura, V. Nano Lett. 2004, 4, 507−511. (10) Lovat, V.; Pantarotto, D.; Lagostena, L.; Cacciar, B.; Grandolfo, M.; Righi, M.; Spalluto, G.; Prato, M.; Ballerini, L. Nano Lett. 2005, 5, 1107−1110. (11) Gabay, T.; Jakobs, E.; Ben-Jacob, E.; Hanein, Y. Physica A 2005, 250, 611−621. (12) Harrison, B. S.; Atala, A. Biomaterials 2007, 28, 344−353. (13) Matsumoto, K.; Sato, C.; Naka, Y.; Whitby, R.; Shimizu, N. Nanotechnology 2010, 21, 115101. (14) Malarkey, E. B.; Parpura, V. Acta. Neurochir. Suppl. 2010, 106, 337−341. (15) Cellot, G.; Toma, F. M.; Varley, Z. K.; Laishram, J.; Villari, A.; Quintana, M.; Cipollone., S.; Prota, M.; Ballerini, L. J. Neurosci. 2011, 31, 12945−12953. (16) Jang, M. J.; Namgung, S.; Hong, S.; Nam, Y. Nanotechnology 2010, 21, 235102−235107. (17) Galvan-Garcia, P.; Keefer, E. W.; Yang, F.; Zhang, M.; Fang, S.; Zakhidov, A. A.; Baughman, R. H.; Romero, M. I. J. Biomater. Sci. Polym. Edn. 2007, 18, 1245−1261. (18) Norman, L. L.; Stroka, K.; Aranda-Espinoza, H. Tissue Eng.: Part B 2009, 15, 291−305. (19) Tomita, K.; Kubo, T.; Matsuda, K.; Yano, K.; Tohyama, M.; Hosokawa, K. GLIA 2007, 55, 1498−1507. (20) Guntinas-Lichius, O.; Wewetzer, K.; Tomov, T. L.; Azzolin, N.; Kazemi, S.; Streppel, M.; Neiss, W. F.; Angelov, D. N. J. Neurosci. 2002, 22, 7121−7131. (21) Streppel, M.; Azzolin, N.; Dohm, S.; Guntinas-Lichius, O.; Haas, C.; Grothe, C.; Wevers, A.; Neiss, W. F.; Angelov, D. N. Eur. J. Neurosci. 2002, 15, 1327−1342. (22) Ozsoy, U.; Demirel, B. M.; Hizay, A.; Ozsoy, O.; Ankerne, J.; Angelova, S.; Sarikcioglu, L.; Ucar, Y.; Angelov, D. N. Restor. Neurol. Neurosci. 2011, 29, 227−242. (23) Jiang, K. L.; Li, Q. Q.; Fan, S. S. Nature 2002, 419, 801. 3672

dx.doi.org/10.1021/nl301428w | Nano Lett. 2012, 12, 3668−3673

Nano Letters

Letter

(24) Zhang, X. B.; Jiang, K. L.; Feng, C.; Liu, P.; Zhang, L. N.; Kong, J.; Zhang, T. H.; Li, Q. Q.; Fan, S. S. Adv. Mater. 2006, 18, 1505−1510. (25) Luo, S.; Wang, K.; Wang, J. P.; Jiang, K. L.; Li, Q. Q.; Fan, S. S. Adv. Mater. 2012, 24, 2294−2298 See also the Supporting Information for Adv. Mater., DOI: DOI: 10.1002/adma. 201104720. (26) Banker, G. A.; Cowan, W. M. Brain Res. 1977, 126, 397−425. (27) Hou, P. X.; Bai, S.; Yang, Q. H.; Liu, C.; Cheng, H. M. Carbon 2002, 40, 81−85. (28) Kotov, N. A.; Winter, J. O.; Clements, I. P.; Jan, E.; Timko, B. P.; Campidelli, S.; Pathak, S.; Mazzatenta, A.; Lieber, C. M.; Prato, M.; Bellamkonda, R. V.; Silva, G. A.; Kam, N. W. S.; Patolsky, F.; Ballerini, L. Adv. Mater. 2009, 21, 3970−4004.

3673

dx.doi.org/10.1021/nl301428w | Nano Lett. 2012, 12, 3668−3673