Collagen coated tantalum substrate for cell proliferation

Collagen coated tantalum substrate for cell proliferation

Colloids and Surfaces B: Biointerfaces 95 (2012) 10–15 Contents lists available at SciVerse ScienceDirect Colloids and Surfaces B: Biointerfaces jou...

1MB Sizes 0 Downloads 11 Views

Colloids and Surfaces B: Biointerfaces 95 (2012) 10–15

Contents lists available at SciVerse ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage:

Collagen coated tantalum substrate for cell proliferation Yinli Li a,1 , Shuai Zhang b,1 , Lijun Guo a , Mingdong Dong b , Bo Liu a,∗ , Wael Mamdouh c,∗∗ a

Institute of Photo-biophysics, School of Physics and Electronics, Henan University, Kaifeng, 475004 Henan, PR China Interdisciplinary Nanoscience Center (iNANO), Aarhus University, DK-8000 Aarhus, Denmark c School of Sciences and Engineering, Department of Chemistry, the American University in Cairo, Cairo 11835, Egypt b

a r t i c l e

i n f o

Article history: Received 16 September 2011 Received in revised form 8 December 2011 Accepted 3 January 2012 Available online 16 January 2012 Keywords: Type I collagen fibrils Tantalum Hydrophobicity AFM Physical adsorption Cell proliferation

a b s t r a c t The extracellular matrix (ECM) plays a key role in cell culture in various physiological and pathological processes in the field of tissue engineering. Recently, the type I collagen ECM has been widely utilized in vitro model systems for the attachment of many different cell lines since it has multi-functions in human tissues. For example it accounts for 6% of the weight of strong, tendinous muscles. In this paper, we reported a new material by coating tantalum (Ta), one highly biocompatible metal, with type I collagen fibrils. The morphology of the new material was studied by high resolution atomic force microscope. It was shown that the adhesion force between type I collagen fibrils network and Ta was strong enough to overcome surface defects. A possible way to explain the phenomenon is that the longitudinal periodicity of collagen fibrils matches the grain size of the Ta domains, which results in increase of the physical adsorption contact area, thereby inducing the dramatic adhesion enhancement between collagen fibrils and Ta. The obtained material was then employed as a template for cell proliferation. Although the surface of this template is more hydrophobic by comparison with the bare Ta surface, the cells on this material were successfully incubated, indicating that the collagen coated Ta might be used as the buffer layer for proliferating cells in hydrophobic biomaterials. © 2012 Published by Elsevier B.V.

1. Introduction Developing novel strategies to prepare biocompatible implant materials for cell attachment is of crucial importance in the field of tissue engineering. In the early studies, the compacted titanium (Ti), tantalum (Ta), and their alloys with different surfaces [1] were explored to meet the needs of biocompatible materials in clinical treatment. Many efforts have been made to estimate the mechanical behavior [2], evaluate cytotoxic effect [1,3] and even investigate the application of these new biomaterials. However, these metallic scaffolds are bioinert, with little capacity to encourage cell culture and tissue regeneration. For example, the scaffolds made of Ti or porous Ti alloy would not facilitate bone formation [4]. An alternative method to solve this problem is to modify the implant surface property so as to support the adhesion, organization, differentiation, and matrix mineralization of cells [5]. Surface manipulation has become an interesting field in current biomaterial science [6], and has been proved as a powerful method in designing new

∗ Corresponding author. Tel.: +86 378 3881058; fax: +86 378 3881602. ∗∗ Corresponding author. Tel.: +20 2 2615 1000; fax: +20 2 27957565. E-mail addresses: [email protected] (B. Liu), wael [email protected] (W. Mamdouh). 1 These authors contributed equally to this work. 0927-7765/$ – see front matter © 2012 Published by Elsevier B.V. doi:10.1016/j.colsurfb.2012.01.009

biocompatible materials with similar topography, composition and properties to the natural extracellular matrix (ECM) [4] in vivo. The ECM is the extracellular part of animal tissue, which usually consists of an organic phase and a mineral phase. The ECM can exert major effect to cell attachment as mechanical scaffold, which helps shaping and formation of the tissue to be defined and maintained [7]. ECM-inspired schemes for biomaterials surface manipulation in preparing bioactive material include functionalizing implants with ECM proteins or ECM derived peptides or proteins [5]. Collagen, which is abundant in mammals [8] and exists in a variety of forms like filaments, sheets, and fibrils [9], can be synthesized by several different types of cells, such as fibroblasts, osteoblasts, chondrocytes and endothelial cells. As a natural protein and main component of the ECM in vivo, collagen has been studied extensively. Type I collagen is the most common type, which is up to 80–90% in the total body’s collagen. It provides the molecular scaffold for mineralization of bone and comprises 90% of the organic bone matrix. In combination with minerals, collagen enables bone some unique mechanical properties in elasticity and strength, which could be used in tissue repair and replacement as biocompatible bone substitute [10–12]. Recently, type I collagen has been employed to coat tissueculture surfaces in order to enhance cell attachment [13] and proliferation [14]. In the earlier studies, Mica was used as the solid substrate for the collagen ECM attaching and further cell culturing [15]. Obviously, mica is not a good candidate in real tissue

Y. Li et al. / Colloids and Surfaces B: Biointerfaces 95 (2012) 10–15

engineering for human body in the future because of its poor biocompatibility. Atomic force microscope (AFM) [16] is designed to exploit both conductors and insulators by measurement of the interaction between a sharp tip and a sample. Besides monitoring the interactions of samples like quartz crystal microbalance (QCM) [17], AFM can also image molecules on a substrate in a nanoscale. Recently, AFM has been used to monitor the self-assembly of protein molecules [18], the mechanical characterization of amyloid fibrils [19], single-molecule interactions [20] or other physical properties of biomolecules. In this paper, by the high resolution AFM, the surface properties of new prepared ECMs were studied. Tantalum, a highly biocompatible metal, was prepared and further employed as the solid substrate, instead of mica, for coating type I collagen. Then, the collagen coated Ta was tested for cell proliferation. It was found that the adhesion force between type I collagen fibrils network and Ta was extremely strong, and the strong adhesion could minimize the influence from substrate defects, demonstrating the advantage in fabricating complicated cell culture templates. Although the collagen coated Ta matrix became hydrophobic, cells could successfully be cultured on it, indicating that the new complex could be used as a potential bridge between cells and hydrophobic biomaterials. 2. Experimental 2.1. Preparation of tantalum substrate The Ta films were prepared on gold-coated substrates at room temperature by evaporating Ta flux with e-gun at an oblique angle between Ta flux and gold surface. Then Ta substrates were bathed in 2% Hellmanex solution for 2 h and 4% sodium dodecyl sulphate (SDS) solution for 12 h, respectively, and cleaned by UV and nitrogen gas in the end. 2.2. Materials Type I collagen was purified from rat tail, and dissolved to 25 ␮g/ml concentration in 10 mM acetic acid. 10 ␮l type I collagen solution was then to deposit onto Ta substrate for approximately 30 min. The sample was rinsed extensively with Milli-Q water (millipore) and dried by nitrogen gas before further measurement. Pre-osteoblastic cells were cultured on the collagen coated Ta substrate, and then removed by ultra-sonic method, which has been successfully used to remove bacteria or cells from substrate surface [21] or even harvest cells in solutions [22]. 2.3. Contact angle measurement The contact angle was measured by drop shape analysis method, which was used to determine the hydrophilicity of bare Ta and that of collagen coated Ta surface. One droplet of Milli-Q water was added onto the sample surface, and then the drop shape was captured and fitted with circle fitting by DSA100 Contact Angle Measuring Instrument (Kruss GmbH, Hamburg, Germany) to obtain the contact angle value. The average angle values of bare Ta surface and that of the collagen fibrils covered Ta surface were achieved after ten times measurements. 2.4. AFM imaging All AFM images were recorded under ambient conditions in the tapping mode using a Nanoscope IIIa microscopy with an E-scanner (Nanocope IIIa Veeco AFM, Digital Instruments, Santa Barbara, California, USA). The cantilevers were of 200 ␮m long and 40 ␮m wide silicon triangular (type NSCii/ALBS from MikroMascg, Spain),


with a nominal spring constant of 3 N/m and typical tip radius of 10 nm. The cantilever driving frequency was optimized in the range between 45 kHz and 75 kHz, while the imaging frequency was fixed to 1.5 Hz. The resolution of images was 256 × 256 pixels per image. The recorded images were flattened and analyzed using the commercial Scanning Probe Image Processor (SPIPTM ) software (image metrology ApS, version 5.13, Lyngby, Denmark).

3. Results and discussion 3.1. Surface properties Among medical implants, Ta is a very interesting substrate for its biocompatibility and capability to resist the body fluids’ attack efficiently. It is also known that the hydrophilicity of substrate is critical to cell attachment. In our experiments, the contact angle of the bare Ta surface was measured to be 10 ± 1◦ (Fig. 1(A)), and after the deposition of the collagen fibrils, the contact angle was changed to 94 ± 2◦ (Fig. 1(B)), which means the surface of Ta is strong hydrophilic and while the collagen coated Ta surface is hydrophobic. These results clearly demonstrate that the property of the Ta surface is changed from hydrophilicity to hydrophobicity after the deposition of the collagen fibrils. In order to understand the mechanism for the different hydrophilicity, detailed surface information of Ta and collagen coated Ta was investigated with high resolution AFM. The topography image of bare Ta surface shows particles with some defects as shown in Fig. 1(C) (pointed by an arrow (i) in Fig. 1(C)), however, the image of collagen coated Ta surface represents a closely packed arrangement of collagen fibrils in Fig. 1(D). The difference could be seen much clearly in the zoom-in image as shown in Fig. 1(E), which reveals a periodic structure in longitudinal direction of the fibrils. It is known that the contact angle would be influenced by the asperities of rough surface. By introducing nanopatterns on the surface, the surface property could be tailored to meet a specific requirement [23]. In other words, the increased hydrophobicity of collagen coated Ta surface could be explained by the formation of the nanopatterns of collagen. Similar phenomena could be seen in nature, for example, the nanopattern composing of wax and bumps makes lotus leaf hydrophobic [23] with a contact angle of 162◦ . Fig. 1(F) shows the cross section of a–a in Fig. 1(E), and the peaks on the curve clearly illustrate the periodicity of the collagen fibrils. The Fourier transform result of the cross-section (Fig. 1(G)) indicates that there are two periodicities in the collagen fibrils (represented by peak I and peak II, respectively). Peak I and Peak II locates at 5.0 ± 1.3 and 15 ± 3.2 respectively, implying the smallest periodicity of the collagen fibrils was 66 ± 11 nm. This result is in the same order as the data reported in previous literatures [10,24,25]. In order to investigate the adhesion force between the collagen fibril and Ta surface, detailed information of the collagen coated Ta substrate was explored. As described in the experimental sections, the Ta covered with collagen matrix template was rinsed adequately with Milli-Q water and dried by nitrogen gas before further measurements. Interestingly, Fig. 1(D) shows that the collagen fibrils were not washed away from the Ta substrate in the process of rinsing, which clearly indicates that the collagen fibrils were tightly attached on the Ta surface. For the closely packed multilayer fibrils, the detailed structural information is probably hidden behind due to the blurring as shown in Fig. 1(D) and (E). So the focus was switched to the loosely packed network and individual collagen fibrils. In Fig. 2(A), it can be seen that the loosely aggregated fibrils form large bundles and highly packed network, and these large bundles appear to be cross-linked in some regions. In zoom-in area indicated by a rectangle in Fig. 2(A), several collagen


Y. Li et al. / Colloids and Surfaces B: Biointerfaces 95 (2012) 10–15

Fig. 1. (A) The contact angle of bare Ta surface. (B) The contact angle of the collagen fibril covered Ta surface. (C) The AFM morphology image of bare Ta surface. (D) The AFM morphology image of compressed collagen fibril net work on Ta surface. (E) The high resolution AFM morphology image of compressed collagen fibril network. (F) The line profile of the solid line a–a in Fig. 1(E), showing the periodicity of the collagen fibrils. (G) The Fourier transformation of line profile in Fig. 1(F): the row number of peak I is 5.0 ± 1.3 and that of peak II is 15 ± 3.2, the peak II represents the smallest periodicity of the collagen fibrils, which is 66 ± 11 nm.

Fig. 2. (A) The AFM morphology image of the collagen fibrils with relatively low coverage, (B) and (C) the high resolution morphology images of the collagen fibrils with relatively low coverage. The arrow (i) represents the buckle structure and the arrow (ii) represents the kink structures of the collagen fibrils.

Y. Li et al. / Colloids and Surfaces B: Biointerfaces 95 (2012) 10–15


Fig. 3. (A) The high resolution morphology image of the individual collagen fibrils and Ta surface, (i) and (ii) are the zoomed in images. (B) The line profile of the solid line a–a in Fig. 3(ii). (C) The line profile of the solid line b–b in Fig. 3(ii). (D) The line profile of the solid line c–c in Fig. 3(A). (E) The line profile of the solid line d–d in Fig. 3(A).

fibrils form buckles instead of loops, which bent back onto themselves (pointed by arrow (i) in Fig. 2(B)). Moreover, some fibres have kinks (pointed by arrow (ii) in Fig. 2(B)). These observations suggest that the fibrils behave more like hollow cylinders or tubes rather than solid rods, which is in agreement with previous work [24]. The

arrangement of the collagen fibres inside a network or in a parallel array adsorbed on surfaces has previously been proved to depend on the preparation, such as the buffer condition and hydrodynamic flow, etc. [19–21] In the present study, the collagen fibres were incubated onto Ta surface using the same buffer as the one used in


Y. Li et al. / Colloids and Surfaces B: Biointerfaces 95 (2012) 10–15

Fig. 4. (A) The AFM morphology of the compressed collagen fibrils matrix after removing the attached cells; (B) and (C) the zoomed in morphology images of Fig. 4(A). The solid line divided each image into three regions. The regions I and II represent the position where cells once attached and the region III represents the one where no cells attached.

previous literature, but without hydrodynamic flow. Therefore one possibility for the formation of the collagen network, instead of a parallel array of fibres, might be the lack of hydrodynamic flow in the sample preparation procedure. The periodic structure of collagen fibrils could be clearly seen in high resolution image (Fig. 2(C)), and more details could be obtained from Fig. 3. Panel (A) in Fig. 3 is the topography image of a selected zone, showing only a few collagen fibres on Ta substrate. Fig. 3(i) is a part of zoom-in image in Fig. 3(A), which shows Ta particles with the diameter of 65 ± 12 nm calculated from the cross section curve of a–a in Fig. 3(B). Not like the sheet-like structures such as mica and Highly Ordered Pyrolytic Graphite (HOPG) with atomic flat, the Ta substrate consists of dot-like structures (Fig. 3(i)) with high surface roughness. Fig. 3(ii) is another part of zoom-in image in Fig. 3(A), which shows several collagen fibrils with clear periodic structures on Ta substrate. The periodicity of collagen fibrils was 60 ± 8 nm calculated from the cross section curve of b–b in Fig. 3(C), which is consistent with the size of Ta particles. The periodic structure of collagen fibrils is also supported by Fig. 1(F) in the zone where there were closely packed collagen fibres with the smallest periodicity of 66 ± 11 nm. The consistency between longitudinal periodicity of collagen fibrils and the grain size of Ta particles increased the physical adsorption contact area dramatically. Therefore, from the detailed imaging study, the strong adhesion force between collagen fibrils and Ta could be caused by two reasons: first, the large contact area indicates that it is more possible to form non-bond interactions (such as hydrogen bond and van der Waals force, etc.) between Ta and collagen fibrils, which strengthens the adsorption of collagen fibrils network onto Ta substrate; second, higher roughness of Ta surface is helpful to the adhesion compared to the mica surface. The strong adhesion force between collagen fibrils and Ta is fully demonstrated in Fig. 3(A), which shows several collagen fibres elongated from the bottom of a defect in Ta substrate to the outside, pointing forward to different directions. This phenomenon was also observed in Fig. 2(B). The c–c and d–d are the cross sections of two adjective regions at the edge of the defect without and with collagen fibre in Fig. 3(A), respectively. The sharp change in height of the Ta

defect edge was 27 ± 2 nm calculated from the cross section curve in Fig. 3(D), which was nearly equal to the height difference of the collagen (25 ± 2 nm) across the edge of the defect measured from Fig. 3(E). This phenomenon also indicated that the type I collagen fibres could be tightly adsorbed onto Ta surface, and the adhesion force was strong enough to overwhelm the defects in substrate. 3.2. Template for cell culture The new collagen coated Ta substrate was used as a template for cell proliferation in vitro. After cell culture and further removal by ultra-sonic method, the topography image of the substrate was recorded with AFM as shown in Fig. 4(A). Compared to Fig. 1(D), clearly the morphology of the collagen matrix has changed dramatically. Before cell adhesion, the collagen network covered the whole area and the height difference across the surface was small. However, after the adhesion and further cell removal, it is possible to distinguish different regions of collagen fibril network depending on the morphological difference, mainly the height contrast. In Fig. 4(A), three regions have been outlined by solid curved lines (regions I, II, and III). The region I and II with relatively higher topography correspond to the areas where cells once attached. The reasons for that are two folds. First, during the removal of the cells, cytosol is expected to be released which could form aggregates on the template in the following drying process, which leads to the height increases in the region I and II; second, cells attachment onto the collagen fibrils network could deform the matrix asymmetrically and bundle individual fibrils into larger ones [15], which then contributes to increase the height. On the other hand, region III, between regions I and II, appears to be lower, and it should be the area where cells were not attached on. From the zoom-in images (Fig. 4(B) and (C)), it could be observed that the detailed morphology of region III in the two areas was totally different. Unfortunately, it is hard to characterize the accurate structure of collagen fibrils network, except some looming individual fibrils in Fig. 4(B). However, obviously there is a layer of collagen fibrils in the region III of Fig. 4(C), though it is much lower than that in regions I and II. The phenomenon could be caused

Y. Li et al. / Colloids and Surfaces B: Biointerfaces 95 (2012) 10–15

by the distortion of collagen fibrils much more in the region III in Fig. 4(B) than that in Fig. 4(C). Comparing the region III, the straight distance between regions I and II in Fig. 4(B) is smaller than that in Fig. 4(C). It means that the deforming force applied by cells [15] to the collagen fibrils matrix in Fig. 4(B) should be much higher than that in Fig. 4(C). In the end, the collagen fibrils in region III in Fig. 4(B) were pushed away by the higher deforming force and even nearly disappeared, but some collagen matrix was still left in region III in Fig. 4(C) due to the lower deforming force. The redistribution of the collagen fibrils before and after the cells proliferation shows that cells could be tightly attached to the collagen coated Ta ECM. It is supposed that hydrophilic surface is crucial to cell proliferation [23–25], and the hydrophobic surfaces were hard for cell attachment, growth and differentiation [24], even though recent studies showed that some cell actually could be cultured on the hydrophobic surface [26–28]. In our study, it seems that the hydrophobicity of collagen coated Ta substrate could be overcome by the strong interaction between collagen ECM and cells, which indicates that the hydrophobic surface is also suitable for cells to attach, grow and differentiate. 4. Conclusions A new biocompatible ECM material, collagen coated Ta substrate, was successfully prepared. With the assistance of contact angle measurement and the ambient AFM imaging, the surface properties of the ECM have been studied. The contact angle measurements indicated that the substrate surface was changed from hydrophilic to hydrophobic after the deposition of collagen on Ta surface, and it was proposed that the formation of the nanopatterned structures of collagen fibrils results in the increased hydrophobicity of collagen coated Ta surface. From the detailed imaging, it was found that the physical adhesion force between collagen and Ta surfaces was strong enough to overcome surface defects besides adsorption along 2D plane. The cooperation between the longitudinal periodicity of collagen fibrils and the diameter of Ta clusters increased the contact area dramatically, and exerted as positive effect to the physical bonding between Ta and collagen fibrils. Moreover, the collagen coated Ta substrate was applied as a substrate for cell proliferation. The results showed the success of cell culture on the hydrophobic surface of the new material, indicating that the new material could be employed as a potential template for cells proliferation by taking advantage of strong cell attachment onto its surface. The success of cell cultured on the hydrophobic surface makes it applicable to employ such


template as the bridge between cell and some kinds of hydrophobic biomaterials. Acknowledgments The authors acknowledge financial support from the National Natural Science Foundation of China (no. 30900280), and Henan Natural Science Research Office of Education project (no. 2008A14003 and no. 2010A140001). References [1] J. Uggeri, S. Guizzardi, R. Scandroglio, R. Gatti, Micron 41 (2010) 210–219. [2] J. Vivanco, Z.B. Fang, D. Levine, H.L. Ploeg, J. Appl. Biomater. Biomech. 7 (2009) 34–42. [3] Y. Li, C. Wong, J. Xiong, P. Hodgson, C. Wen, J. Dent. Res. 89 (2010) 493–497. [4] Y. Han, J.H. Zhou, L. Zhang, K.W. Xu, Nanotechnology 22 (2011). [5] A. Shekaran, A.J. Garcia, J. Biomed. Mater. Res. Part A 96A (2011) 261–272. [6] R. Beutner, J. Michael, B. Schwenzer, D. Scharnweber, J. R. Soc. Interface 7 (2010) S93–S105. [7] K.E. Kadler, D.F. Holmes, J.A. Trotter, J.A. Chapman, Biochem. J. 316 (Pt 1) (1996) 1–11. [8] F.Z. Cui, Y. Li, J. Ge, Mater. Sci. Eng., R Rep. 57 (2007) 1–27. [9] M. van der Rest, P. Bruckner, Curr. Opin. Struct. Biol. 3 (1993) 430–436. [10] H.X. Zhao, H. Jin, J.Y. Cai, S. Ding, Ultramicroscopy 110 (2010) 1306–1311. [11] H.S. Gupta, W. Wagermaier, G.A. Zickler, D.R.B. Aroush, S.S. Funari, P. Roschger, H.D. Wagner, P. Fratzl, Nano Lett. 5 (2005) 2108–2111. [12] O. Akkus, J. Biomech. Eng.-Trans. Asme 127 (2005) 383–390. [13] R.J. Klebe, Nature 250 (1974) 248–251. [14] S.L. Schor, J. Cell Sci. 41 (1980) 159–175. [15] J. Friedrichs, A. Taubenberger, C.M. Franz, D.J. Muller, J. Mol. Biol. 372 (2007) 594–607. [16] G. Binnig, C.F. Quate, C. Gerber, Phys. Rev. Lett. 56 (1986) 930–933. [17] O. Hayden, R. Bindeus, F.L. Dickert, Meas. Sci. Technol. 14 (2003) 1876. [18] M. Dong, S. Xu, M.H. Bünger, H. Birkedal, F. Besenbacher, Adv. Eng. Mater. 9 (2007) 1129–1133. [19] M. Dong, M.B. Hovgaard, W. Mamdouh, S. Xu, D.E. Otzen, F. Besenbacher, Nanotechnology 19 (2008) pp. 384013. [20] M. Dong, O. Sahin, Nat. Commun. 2 (2011) 247. [21] D. Bagge, M. Hjelm, C. Johansen, I. Huber, L. Gram, Appl. Environ. Microbiol. 67 (2001) 2319. [22] R. Bosma, W.A. van Spronsen, J. Tramper, R.H. Wijffels, J. Appl. Phycol. 15 (2003) 143–153. [23] C. Neinhuis, W. Barthlott, Ann. Bot. 79 (1997) 667–677. [24] T. Gutsmann, G.E. Fantner, M. Venturoni, A. Ekani-Nkodo, J.B. Thompson, J.H. Kindt, D.E. Morse, D.K. Fygenson, P.K. Hansma, Biophys. J. 84 (2003) 2593– 2598. [25] S. Strasser, A. Zink, M. Janko, W.M. Heckl, S. Thalhammer, Biochem. Biophys. Res. Commun. 354 (2007) 27–32. [26] F.Z. Jiang, H. Horber, J. Howard, D.J. Muller, J. Struct. Biol. 148 (2004) 268–278. [27] D.A. Cisneros, J. Friedrichs, A. Taubenberger, C.M. Franz, D.J. Muller, Small 3 (2007) 956–963. [28] F.Z. Jiang, K. Khairy, K. Poole, J. Howard, D.J. Muller, Microsc. Res. Tech. 64 (2004) 435–440.