Langmuir 2005, 21, 1679-1682
1679
Layer-by-Layer Assembly of Human Serum Albumin and Phospholipid Nanotubes Based on a Template Gang Lu, Sufen Ai, and Junbai Li* International Joint Lab, Key Lab of Colloid and Interface Science, The Center for Molecular Science, Institute of Chemistry, Chinese Academy of Sciences, Zhong Guan Cun, Beijing 100080, China Received September 7, 2004. In Final Form: January 13, 2005 The preparation of nanotubes from human serum albumin (HSA) and mixtures of L-R-dimyristoylphosphatidic acid (DMPA)/HSA is described. The nanotubes were prepared via alternate adsorption of HSA of different/opposite charges (by variation of the pH) or by sequential adsorption of DMPA and HSA, respectively, onto the inner surfaces of porous anodic alumina templates. This simple layer-by-layer assembly results in a monodisperse size distribution and a uniform orientation. The nanotubes allow the specific incorporation of lipophilic components such as channels or receptors and may thus serve as probes or sensors for biological systems.
Introduction Since the first discovery of carbon nanotubes in 1991,1 nanotubes have been made from a variety of materials including carbon, ceramics, metals, and organic polymers that have been summarized in several reviews.2,3 A very interesting application of nanotubes for biosensors or bioreactors has been their suggested use as templates for the immobilization or growth of biomolecules.4,5 In the following we will show how biogenic nanotubes of uniform size and distribution can be produced by the templatebased method6 from pure human serum albumin (HSA) as well as from complexes of HSA with L-R-dimyristoylphosphatidic acid (DMPA). Their sizes, shapes, and structural properties are controlled by the template that has been used. Many groups have prepared nanoscale materials with anodic porous alumina as a template.7-15 Recently, several groups have also reported a layer-bylayer (LbL) nanotube template synthesis.16-18 The presented experimental results are based on our previous * To whom correspondence should be addressed. Prof. Dr. Junbai Li, International Joint Lab, The Center for Molecular Science, Institute of Chemistry, Chinese Academy of Sciences, Zhong Guan Cun, Bei Yi Jie No. 2, Beijing 100080, China. Tel: +86 10 82614087. Fax: +86 10 82612629. E-mail:
[email protected]. (1) Iijima, S. Nature 1991, 354, 56. (2) Hu, J.; Odom, T. W.; Lieber, C. M. Acc. Chem. Res. 1999, 32, 435. (3) Edelmann, F. T. Angew. Chem., Int. Ed. 1999, 38, 1381. (4) Balavoine, F.; Schultz, P.; Richard, C.; Mallouh, V.; Ebbesen, T. W.; Mioskowski, C. Angew. Chem., Int. Ed. 1999, 38, 1912. (5) Tsang, S. C.; Guo, Z.; Chen, Y. K.; Green, M. L. H.; Hill, H. A. O.; Hembley, T. W.; Sadler, P. J. Angew. Chem., Int. Ed. 1997, 36, 2198. (6) Martin, C. R. Science 1994, 266, 1961. (7) Jirage, K. B.; Hulteen, J. C.; Martin, C. R. Science 1997, 278, 655. (8) Che, G.; Lakshmi, B. B.; Fisher, E. R.; Martin, C. R. Nature 1998, 393, 346. (9) Steinhart, M.; Wendorff, J. H.; Greiner, A.; Wehrspohn, R. B.; Nielsch, K.; Schilling, J.; Choi, J.; Go¨sele, U. Science 2002, 296, 1997. (10) Bao, J.; Tie, C.; Xu, Z.; Ma, Q.; Hong, J.; Sang, H.; Sheng, D. Adv. Mater. 2002, 14, 44. (11) Shelimov, K. B.; Mosskovits, M. Chem. Mater. 2000, 12, 250. (12) Prieto, A. L.; Sander, M. S.; Martı´n-Gonza´lez, M. S.; Gronsky, R.; Sands, T.; Stacy, A. M. J. Am. Chem. Soc. 2001, 123, 7160. (13) Liu, H.; Li, Y.; Jiang, L.; Luo, H.; Xiao, S.; Fang, H.; Li, H.; Zhu, D.; Yu, D.; Xu, J.; Xiang, B. J. Am. Chem. Soc. 2002, 124, 13370. (14) Rahman, S.; Yang, H. Nano Lett. 2003, 3, 439. (15) Lahav, M.; Sehayek, T.; Vaskevich, A.; Rubinstein, I. Angew. Chem., Int. Ed. 2003, 42, 5576. (16) Ai, S.; Lu, G.; He, Q.; Li, J. J. Am. Chem. Soc. 2003, 125, 11140. (17) Liang, Z.; Susha, A. S.; Yu, A.; Caruso, F. Adv. Mater. 2003, 15, 1849. (18) Hou, S.; Harrell, C. C.; Trofin, L.; Kohli, P.; Martin, C. R. J. Am. Chem. Soc. 2004, 126, 5674.
work, which demonstrates that HSA or a mixture of HSA and lipids can form stable multilayers at interfaces.19-23 This allows the specific incorporation of lipophilic components such as channels or receptors, which may serve as probes or sensors for biological systems. Experimental Section DMPA and HSA (essentially fatty acid free, g96% albumin) were purchased from Sigma and used without further purification. In the experiments, HSA was dissolved in a 10 mM phosphate buffer solution at pH 3.8 or pH 7.0. The DMPA was dissolved in aqueous solution or chloroform. To prepare the DMPA aqueous solution, a DMPA solution in 8:2 (v/v) CHCl3-CH3OH was rotary evaporated at 40 °C in a round-bottom flask to give a thin film, which was dried for 2 h at 25 °C, and then pure water was added and sonicated. The water used in all experiments was prepared in a three-stage Millipore Milli-Q Plus 185 purification system and had a resistivity higher than 18.2 MΩ cm-1. All experiments were performed at room temperature. The anodic porous alumina template membrane (Purchased from Whatman Corp.) inserted into two porous Millipore filters was fixed together onto a stainless steel filter (Millipore, Microsyringe). The detailed experimental procedure is similar to that described previously.16 HSA solution (5 mL) at pH 3.8 was injected into the filter first through the syringe and then passed through the porous template with a certain pressure. Afterward, pure water was employed to wash the pores in the template. The HSA solution at pH 7.0, the DMPA aqueous solution, or the DMPA chloroform solution was used to form a second layer, respectively. The alternative adsorption for five times leads to the formation of the multilayer nanotubes. After the nanotubes were formed in the template pores, the template membrane was immersed into 1 M NaOH solution to remove the alumina template at room temperature for about 2 h. To remove the rest of the salt, the nanotubes were washed several times with pure water. For the measurements of scanning electron microscopy, the nanotubes were glued on the silicon with epoxy resin and then sputtered with gold. Scanning electron microscope (SEM) images were obtained by an Hitachi S-4300 instrument. The HSA aqueous solution was measured by a U-3010 UV/Vis spectrophotometer (Hitachi, Japan) at different pH values. (19) Li, J. B.; Zhang, Y.; Yan, L. L. Angew. Chem., Int. Ed. 2001, 40, 891. (20) Lu, G.; Chen, H.; Li, J. B. Colloids Interfaces A 2003, 215, 25. (21) Lu, G.; An, Z.; Tao, C.; Li, J. Langmuir 2004, 20, 8401. (22) An, Z.; Lu, G.; Mo¨hwald, H.; Li, J. Chem.sEur. J. 2004, 10, 5848. (23) Wang, X.; He, Q.; Zheng, S.; Brezesinski, G.; Mo¨hwald, H.; Li, J. J. Phys. Chem. B 2004, 108, 14171.
10.1021/la047771r CCC: $30.25 © 2005 American Chemical Society Published on Web 01/29/2005
1680
Langmuir, Vol. 21, No. 5, 2005
Letters
Circular dichroism (CD) spectra of HSA were measured on a JASCO J-810 spectropolarimeter at room temperature over a wavelength range of 190-300 nm. All measurements in solution were recorded in a 5-mm path length quartz cell. Each spectrum is the average of four measurements.
Results and Discussion HSA, the most abundant plasma protein of the circulatory system, has long been used as a model protein and served numerous applications in both industrial processes and academic research areas. It contributes significantly to colloidal osmotic blood pressure and has a stable configuration in buffer solution. HSA consists of 585 amino acids with a molecular weight of 66.5 kD, stabilized by 17 disulfide bridges, and contains a single free sulfydryl. The location of the disulfide bridges appears to confer some rigidity within each subdomain, but the lack of interdomain disulfide bridges allows significant modifications in the shape of HSA in response to changes in pH.24,25 The strong binding site within subdomains in HSA contributes to the stabilization of the HSA structural framework at even very acidic or basic conditions.26 Through modifying the surface charge of HSA molecules at different pH conditions, which is beyond the isoelectric point (pH 4.8), the alternative adsorption of charged HSA (positive charge at pH 3.8 and negative charge at pH 7.0) at each deposition (neutral washing steps in between) results in the LbL deposition of a HSA multilayer on the inner walls of the membrane. By using a pressure-filter-template technique, the inner wall of the porous membrane will be covered by deposition of oppositely charged species through layerby-layer assembly.27,28 With the subsequent removal of the template, the wall controlled nanotubes are obtained. A HSA buffer solution was first passed through one layer filter before getting contact with the porous alumina template membrane. At pH 3.8 the framework of HSA is positive charge enriched. While the HSA solution is brought in contact with the porous templates, the HSA molecules will adsorb on the pore wall with the anionic site of the template to form a thin film through adhesive and electrostatic forces. Then HSA solution at pH 7.0 was used to form the next layer. After alternate adsorption for five times, the entire assembly is immersed in aqueous natrium hydroxide to dissolve the template membrane at room temperature and subsequently washed several times with Millipore water. The results presented here indicate the possibility to construct biogenic nanotubes, which may serve as bioreactors, biosensors, or delivery carriers in a living system. Figure 1 depicts an array of aligned HSA nanotubes, which have smooth and clean surfaces with a wall thickness of around 30 nm (Figure 1A). The highly ordered array of HSA nanotubes exhibits good stability (Figure 1B). The length of the resulting nanotubes is about 60 µm, the order of the template thickness (Figure 1C). The tube diameter can be varied by controlling the porous size of templates. The thickness of the walls enables the HSA nanotubes to bend with a high flexibility (Figure 1D). To prove the stability of the HSA system under drastic pH condition changes, one has to clarify the structural stability of HSA with the pH modification. Therefore, the CD technique was used to determine the helical state of (24) Olivieri, J. R.; Craievich, A. F. Eur. Biophys. J. 1995, 24, 77. (25) He, X. M.; Carter, D. C. Nature 1992, 358, 209. (26) Carter, D. C.; He, X. M.; Munson, S. H.; Twigg, P. D.; Gernert, K. M.; Broom, M. B.; Miller, T. Y. Science 1989, 244, 1195. (27) Decher, G. Science 1997, 277, 1232. (28) Vieira, E. P.; Hermel, H.; Mo¨hwald, H. Biochim. Biophys. Acta 2003, 1645, 6.
Figure 1. Scanning electron micrographs of human serum albumin nanotubes obtained by self-assembly. (A) HSA nanotube arrays after the removal of templates. (B) Highly ordered HSA nanotubes with a wall thickness of around 30 nm. (C) The length of the resulting nanotubes is about 60 µm. (D) Highly flexible HSA nanotubes.
Figure 2. CD (A) and UV (B) spectra of HSA solution at different pH values and room temperature.
HSA at a wide pH range. Figure 2 shows the CD (A) and UV (B) spectra of HSA solution from pH 0.7 to 12.3. The CD spectra of HSA exhibit a typical shape of an R-helix rich secondary structure (two minima at approximately 208 and 222 nm).25 The curves indicate that no obvious variation of R-helix structure at pH 3.8 and pH 7.0 is observed. This demonstrates that the secondary structure of HSA is not destroyed in such a wide pH range.
Letters
Langmuir, Vol. 21, No. 5, 2005 1681
Figure 3. Scanning electron micrographs of DMPA/HSA nanotubes obtained by template-based self-assembly.
Figure 4. Scanning electron micrographs of DMPA/HSA nanotubes obtained by filtering the HSA aqueous solution and DMPA chloroform solution alternatively.
The UV spectrograph of HSA solution shows that the absorption spectrum at 280 nm results from the sidechain groups of some amino acids (e.g., tryptophan). In the far-ultraviolet range, around 220 nm it shows the peptide-bond absorbance of the protein. The UV absorption experiments demonstrate that the composite of protein HSA is not changed within the pH range. At pH > 12, both CD and UV spectra exhibit a red shift, indicating that the HSA helix structure might be alternated or it caused a protein aggregation at a higher base condition.28-30 With a similar method, we switched to another biological pair system to construct the phospholipid/protein nanotubes. From a previous study, we have learned that the coadsorption of phospholipids and proteins can form a very stable and multilayer structure.19 In view of the charge interaction between DMPA and HSA based on the pressure-filter-template technique the alternative adsorption will result in the formation of DMPA and HSA nanotubes. We first filtered the pH 3.8 HSA solution to pass through the alumina template. HSA molecules will adsorb on the pore wall to form a HSA film. The DMPA is dissolved in aqueous solution or organic solvent instead of pH 7.0 HSA solution to form a second layer with a sequential adsorption. The alternative adsorption of HSA and DMPA for five times leads to the multilayer formation of DMPA/HSA in the pores of the template. Then DMPA/ HSA complex nanotubes were obtained after the removal of the alumina template (Figure 3). The DMPA/HSA nanotubes have a wall thickness similar to that of pure HSA nanotubes and can be also perfectly ordered arrays (Figure 3A,B). The success of the preparation of protein nanotubes was further confirmed by energy-depressive X-microanalysis. Chemical analysis indicated the presence (29) Schladitz, C.; Vieira, E. P.; Hemel, H.; Mo¨hwald, H. Biophys. J. 1999, 77, 3305. (30) Safar, J.; Roller, P. P.; Gajduesk, D. C.; Gibbs, C. J. J. Biol. Chem. 1993, 268, 20276.
of the main composites, C, N, and O in the HSA nanotubes (see the Supporting Information). Steinhart et al.9 have used the template technique to prepare polymer nanotubes by means of the wetting properties. However, for an aqueous system the higher surface tension of water will prevent the species from entering the pores of the templates. In most cases, the materials remain at the surface of templates. For such systems as shown above the pressure-filter pretreatment might be imperative. The charge interaction provides the possibility of the oppositely charged compound adsorption as a second layer in increasing the thickness of the tube wall. However, adhesive forces between the adsorbed species and template and between the species may also dominate the formation of HSA or DMPA/HSA tube structure. To prove this, we used DMPA chloroform solution instead of its aqueous solution to assembly the DMPA/HSA nanotubes. As a consequence, a highly ordered array of tubes is also obtained (Figure 4). These experimental results show that the wetting property should be suitable in a broad range to prepare nanotubes. Conclusions As demonstrated, with the alternate adsorption of HSA with different charges by varying the pH value at each deposition or mixing with DMPA, we fabricated HSA and DMPA/HSA nanotubes with a monodisperse size distribution and uniform orientation through layer-by-layer assembly. The modification of HSA structure or the lipid bilayer binding to HSA results in the high flexibility of both types of nanotubes. It is noted that the experimental conditions presented here are not physiological. It should be fairly considered that the structure of HSA might be changed during the process of assembly as well as the template removal although our CD spectra showed no dramatic change of the helix structure within a lower pH range. Even so, we can simply assume that HSA as a natural protein or cross-linked materials provide a support
1682
Langmuir, Vol. 21, No. 5, 2005
layer for lipids, which may incorporate membrane specific components such as channels and receptors for specific permeation and recognition. Thus HSA/DMPA nanotubes with a new feature may result. Acknowledgment. This work was financially supported by the National Nature Science Foundation of China (NNSFC29925307 and NNSFC90206035) as well
Letters
as the collaborated project of the German Max Planck Society. The authors acknowledge Hans Riegler for critically reading the manuscript. Supporting Information Available: Energy-depressive X-microanalysis of HSA nanotubes. This material is available free of charge via the Internet at http://pubs.acs.org. LA047771R
