- Email: [email protected]
Solvent and temperature effects on folding of electrospun collagen nanofibers
Materials Letters 130 (2014) 223–226
Contents lists available at ScienceDirect
Materials Letters journal homepage: www.elsevier.com/locate/matlet
Solvent and temperature effects on folding of electrospun collagen nanofibers Murat Kazanci n Department of Biophysics, School of Medicine, Bahcesehir University, Sahrayıcedid Mah. Batman Sok. No. 66-68, Yenisahra, Kadıkoy, 34734 Istanbul, Turkey
art ic l e i nf o
a b s t r a c t
Article history: Received 25 March 2014 Accepted 17 May 2014 Available online 27 May 2014
Electrospinning is a well-known method to obtain nanofibers and collagen is a very attractive material for biological applications due to its bioactive nature. However, the preservation of the natural structure of collagen at the end products due to the different production parameters is an open question. In this study the importance of the preparation temperature for CD measurements and the solvent effect that employed in the electrospinning process was discussed. The decrease of the preparation temperature by 101 increased the measured PP-II (folded) fraction ratio from 37% to 52.5% and the nanofibers that were obtained from acidic solvent scored 59% of PP-II fraction ratio. & 2014 Elsevier B.V. All rights reserved.
Keywords: Biomaterials Biomimetic Electrospinning Collagen Nanofibers CD spectra
1. Introduction Collagen type I makes up 70–90% of the collagen in human body. It is present in the form of elongated fibers in various tissues. These building blocks are rod-like triple helices that are stabilized by intramolecular hydrogen bonds between Gly and Hyp in adjacent chains [1]. This 3D folded structure makes collagen a bioactive material due to its special topography and certain order of aminoacids in the 3D structure by sending the necessary signals for the cell activities. Electrospinning is a suitable way to produce fibers with diameters smaller than 1 mm and has a number of advantages [2]. In order to regenerate materials in forms of nanofibers, it is necessary to dissolve them in suitable solvents and many other production parameters could affect the final structures. Extracted type I collagen is favored for biomedical applications; under appropriate conditions it will spontaneously self-assemble to form biodegradable and biocompatible insoluble fibrils [3]. Architecture, topography, biochemical and mechanical features of the nanofibrous scaffolds have been shown to significantly improve cellular events and in vivo cellular regenerations [4]. Liu et al. [5] and Yang et al. [6] investigated the mechanical properties of single electrospun collagen type I fibers. The bending moduli of the cross-linked collagen fibers range from 1.3 to 7.8 GPa. The electrospun fibers showed anisotropic mechanical properties with two orders of magnitude lower shear modulus compared to the bending modulus. Dos Santos et al. [7] improved
n
Tel.: þ 90 216 5798229. E-mail address: [email protected]
http://dx.doi.org/10.1016/j.matlet.2014.05.114 0167-577X/& 2014 Elsevier B.V. All rights reserved.
the electrospinning technique to impart the same morphological and mechanical properties to each point of the produced structure. Cai et al. [8] correlated the process parameters as well as relative humidity and evaporation rate to fiber diameter. Whited et al. [9] showed the correlation of fiber orientation and cell alignment. Oh et al. [10] enhanced the stability of stem cells by the nanofibers via an electrospinning technique. We demonstrated that the electrospun collagen nanofibers have been mostly unfolded in our previous work and speculated that it is mostly due to the acting high shear forces during the electrospinning process [11,12]. Collagen has better preserved its native structure in acidic solvents; however there has been no real success to produce electrospun nanofibers from acidic solutions till now. In this work, we were able to produce electrospun nanofibers from acidic solutions, by increasing the collagen concentration to 40% and the results were compared with the electrospun nanofibers that were obtained from fluorinated solutions.
2. Experimental The collagen-type I used in these experiments was a waterinsoluble lyophilized foam powder consisting of tropocollagen extracted from bovine dermis (generously donated by the Kensey Nash Corporation, USA) and was used without further purification. Acetic acid (HAc, 99%) was purchased from Merck KGaA, Darmstadt. 1,1,1,3,3,3-hexafluoro-2-propanol (HFP, Z99.8%) was purchased from Sigma-Aldrich and used as received. Two different samples were prepared by dissolving collagen in different solvent systems:
224
M. Kazanci / Materials Letters 130 (2014) 223–226
Fig. 1. SEM images of electrospun nanofibers from (a) HFP and (b) 40% HAc, and TEM images of electrospun nanofibers from (c) HFP and (d) 40% HAc.
(1) 10% w/v collagen solution was prepared by dissolving collagen in HFP; (2) 40% w/v collagen solution was prepared by dissolving collagen in 40% v/v HAc. The prepared electrospinning solutions were loaded into a 2 mL syringe (Omnifix, B. Braun, Melsungen, Germany) with a blunt end nozzle, controlled by a syringe pump (Pump 33 Harvard Apparatus, Holliston, USA). The solution was pushed through a capillary blunt steel needle (21 gauge, 0.7 mm i.d. 50 mm length) at a constant speed (0.5 μL/min). The steel needle was coupled to a high voltage source (Spellman Bertan Series 205B, NY, USA). The electric potential was needed to start the spinning process and thus form a jet. The applied DC voltage was held at 16 kV. A Cu collector was placed 15 cm from the needle tip to collect the electrospun collagen nanofibers. The nanofiber meshes were collected on cover-glasses placed onto the collector.
3. Results and discussion Fig. 1 (a) and (b) shows the SEM images, and (c) and (d) shows the TEM images of the electrospun nanofibers obtained from two different solvents. Electrospinning yielded randomly oriented and
Fig. 2. Raman shift of (a) native collagen, (b) electrospun collagen nanofibers obtained from HAc, and (c) electrospun collagen nanofibers obtained from HFP.
M. Kazanci / Materials Letters 130 (2014) 223–226
225
Fig. 3. CD spectra of electrospun nanofiber meshes a) dissolved at different temperatures, and b) obtained from different solvents.
interconnected fibrous meshes with the fiber diameter in the nanometer range. The electrospun nanofiber diameters were mainly determined by the collagen concentrations and electrospinning solvents. 10% collagen solution in HFP produced electrospun nanofibrous meshes with the fiber diameter. ranging between 150 and 200 nm (Fig. 1 (a)), and 40% collagen in 40% HAc produced electrospun nanofibrous meshes in a wide range of fiber diameters and with some bead formations (Fig. 1b). Collagen can be easily dissolved under acidic conditions to produce an electrospinning solution. However, the slow evaporation rate of the acid and its strong affinity with collagen leads to the deposition of wet fibers on the target, which can partially weld together and lessen the porosity of the mat. There was no observation of triple helix folding or the characteristic 67 nm D-spacing at TEM images shown in Fig. 1(c) and (d). The Raman shift of native collagen displays the peak position at 1670 cm 1 which is the finger print of type I collagens [13] with a shoulder at 1645 cm 1 at Amide I region [14]. Amide I band arises from CQO stretching of the peptidic bond in the Gly-X-Y tripeptide sequence. We observed broader peaks and peak shifts to higher wavenumbers for both type of collagen nanofibers in Fig. 2. The Raman peak positions at 1677 cm 1 are mainly indications of β-sheet formations for the electrospun nanofibers [15]. The frequency shift of the Amide I band suggests a different proportion of amino acid residues. CD utilizes the differential absorption of left and right handed circular polarized light in an asymmetric environment to assess secondary structure. The left handed polyproline II (PPII) helix gives rise to a circular dichroism spectrum that is remarkably similar to that of folded/unfolded proteins [16]. Single wavelength estimation of collagen folding was employed to extract the fraction of triple helix, mainly relying on the equilibrium state of collagen spectra. The fraction of folded collagen (θ) can be defined as θ ¼ ðθobs –θu Þ=ðθt –θu Þ
ð1Þ
where θobs, θt, and θu represent the ellipticities of observed, triple helix, and unordered state, respectively. θobs, θt and θu were measured at a wavelength of 221.5 nm for the sample in interest, and for the collagen solutions at 10 1C and 90 1C, respectively, in 0.05 M HAc [11]. Fig. 3 (a) shows the CD spectra of electrospun collagen nanofibrous mesh, obtained by using the same solvent (in 0.05 M HAc) and at the same collagen concentration (10%). However, the one which was solubilized at room temperature (25 1C) displayed relatively lower PP-II fraction ratio (θ¼0.37) than the one which was solubilized at 15 1C (θ¼ 0.52). Fig. 3 (b) shows the native collagen spectra and two electrospun collagen nanofibers spectra that were obtained from different solvents at different
concentrations. We observed slight increase in the PP-II ratio of collagen that was obtained from acidic solvent (θ¼ 0.59). It is known that collagen better preserved its native structure in acidic solvents [11]. The solubilizing temperature of the collagen nanofibers for the structural characterization methods is extremely important and it will definitely influence the experimental results.
4. Conclusion In the previous study [11], we stated that the spinning process causes refolding of nanofibers, but PP-II fraction did not exceed 42%. In this work, solubilizing the electrospun collagen nanofibers at lower temperatures (15 1C) for CD measurements, it was possible to increase the folded fraction ratio from 37% to 52.5%. The preparation temperature is extremely important for collagen samples. They are extremely sensitive to the temperature and this could be also one of the important factors that cause discrepancy in experimental results in the literature. It was possible to increase the folded fraction of collagen by almost 1.5 fold, just decreasing the temperature by 101. Another important result is that it is well known that collagen best preserves its native structure in acidic solvents. However, there was no real success obtaining electrospun collagen nanofibers from acidic solvents, due to their relatively low evaporation temperatures that cause the binding of fibers and deforms the fiber morphology. In this study, the electrospun collagen nanofibers were successfully obtained from 40% HAc, by increasing the collagen concentration to 40%. Otherwise the bead formation is more pronounced than the fiber formation at low concentrations. There is a small increase in the folded ratio, that is almost negligible, compared to the ones obtained from fluorinated solvents (θ¼0.59 vs. θ¼0.52). The morphology of the nanofibers from acidic solvent displays 3D porous, interconnected network rather than the fibrous structures, due to the slow evaporation rate of the acid and its strong affinity with collagen.
Acknowledgments This study is supported by TUBITAK's Co-Funded Brain Circulation Program (2236, Project number 112C025). The author would like to thank Dr. Jochen Bürck and Bianca Posselt for the CD measurements (IBG2, Karlsruhe Institute of Technology, Germany). The author appreciates Onur Aras' help in the preparation of electrospun collagen nanofibrous meshes (School of Medicine, Bahcesehir University, Istanbul, Turkey).
226
M. Kazanci / Materials Letters 130 (2014) 223–226
References [1] Zeugolis DI, Khew ST, Yew ESY, Ekaputra AK, Tong YW, Yung L-YL, et al. Biomaterial 2008;29:2293–305. [2] Teo W,E, Ramakrishna S. Nanotechnology 2006;17:89–106. [3] Lynn AK, Yannas I, Bonfield V, Biomed. WJ. Mater Res B: Appl Biomater 2004;71B:343–54. [4] Cheng YX, Ramos D, Lee P, Liang DN, Yu XJ, Kumbar SGJ. Biomed Nanotechnol 2014;10:287–98. [5] Liu J, Argenta L, Morykwas M, Wagner WD. J Biomed Appl 2014;28:729–38. [6] Yang L, Fitie CFC, Werf KO, Van der; Bennink ML, Dijkstra PJ, Feijen J. Biomaterials 2008;29:955–62. [7] dos Santos AM, Dierck J, Troch M, Podevijn M, Schacht E. Macro Mater Eng 2011;296:637–44.
[8] [9] [10] [11] [12] [13] [14] [15] [16]
Cai YS, Gevelber M. J Mater Sci 2013;48:7812–26. Whited BM, Rylander MN. Biotechnol Bioeng 2014;111:184–95. Oh B, Lee CH. Expert Opin Drug Deliv 2013;10:1565–82. Bürck J, Heissler S, Geckle U, Ardakani MF, Schneider R, Ulrich AS, et al. Langmuir 2013;29:1562–72. Diesner MO, Welle A, Kazanci M, Kaiser P, Spatz J, Koelsch P. Biointerphaces 2011;6:171–9. Nguyen TT, Gobinet C, Feru J, Brassart -Pasco S, Manfait M, Piot O. Spectroscopy 2012;27:421–7. Kazanci M, Roschger P, Paschalis EP, Klaushofer K, Fratzl P. J Struct Biol 2006;156:489–96. Gullekson C, Leanne Lucas L, Hewitt K, Kreplak L. Biophys J 2011;100:1837–45. Rucker AL, Creamer TP. Protein Sci 2002;11:980–5.
Contents lists available at ScienceDirect
Materials Letters journal homepage: www.elsevier.com/locate/matlet
Solvent and temperature effects on folding of electrospun collagen nanofibers Murat Kazanci n Department of Biophysics, School of Medicine, Bahcesehir University, Sahrayıcedid Mah. Batman Sok. No. 66-68, Yenisahra, Kadıkoy, 34734 Istanbul, Turkey
art ic l e i nf o
a b s t r a c t
Article history: Received 25 March 2014 Accepted 17 May 2014 Available online 27 May 2014
Electrospinning is a well-known method to obtain nanofibers and collagen is a very attractive material for biological applications due to its bioactive nature. However, the preservation of the natural structure of collagen at the end products due to the different production parameters is an open question. In this study the importance of the preparation temperature for CD measurements and the solvent effect that employed in the electrospinning process was discussed. The decrease of the preparation temperature by 101 increased the measured PP-II (folded) fraction ratio from 37% to 52.5% and the nanofibers that were obtained from acidic solvent scored 59% of PP-II fraction ratio. & 2014 Elsevier B.V. All rights reserved.
Keywords: Biomaterials Biomimetic Electrospinning Collagen Nanofibers CD spectra
1. Introduction Collagen type I makes up 70–90% of the collagen in human body. It is present in the form of elongated fibers in various tissues. These building blocks are rod-like triple helices that are stabilized by intramolecular hydrogen bonds between Gly and Hyp in adjacent chains [1]. This 3D folded structure makes collagen a bioactive material due to its special topography and certain order of aminoacids in the 3D structure by sending the necessary signals for the cell activities. Electrospinning is a suitable way to produce fibers with diameters smaller than 1 mm and has a number of advantages [2]. In order to regenerate materials in forms of nanofibers, it is necessary to dissolve them in suitable solvents and many other production parameters could affect the final structures. Extracted type I collagen is favored for biomedical applications; under appropriate conditions it will spontaneously self-assemble to form biodegradable and biocompatible insoluble fibrils [3]. Architecture, topography, biochemical and mechanical features of the nanofibrous scaffolds have been shown to significantly improve cellular events and in vivo cellular regenerations [4]. Liu et al. [5] and Yang et al. [6] investigated the mechanical properties of single electrospun collagen type I fibers. The bending moduli of the cross-linked collagen fibers range from 1.3 to 7.8 GPa. The electrospun fibers showed anisotropic mechanical properties with two orders of magnitude lower shear modulus compared to the bending modulus. Dos Santos et al. [7] improved
n
Tel.: þ 90 216 5798229. E-mail address: [email protected]
http://dx.doi.org/10.1016/j.matlet.2014.05.114 0167-577X/& 2014 Elsevier B.V. All rights reserved.
the electrospinning technique to impart the same morphological and mechanical properties to each point of the produced structure. Cai et al. [8] correlated the process parameters as well as relative humidity and evaporation rate to fiber diameter. Whited et al. [9] showed the correlation of fiber orientation and cell alignment. Oh et al. [10] enhanced the stability of stem cells by the nanofibers via an electrospinning technique. We demonstrated that the electrospun collagen nanofibers have been mostly unfolded in our previous work and speculated that it is mostly due to the acting high shear forces during the electrospinning process [11,12]. Collagen has better preserved its native structure in acidic solvents; however there has been no real success to produce electrospun nanofibers from acidic solutions till now. In this work, we were able to produce electrospun nanofibers from acidic solutions, by increasing the collagen concentration to 40% and the results were compared with the electrospun nanofibers that were obtained from fluorinated solutions.
2. Experimental The collagen-type I used in these experiments was a waterinsoluble lyophilized foam powder consisting of tropocollagen extracted from bovine dermis (generously donated by the Kensey Nash Corporation, USA) and was used without further purification. Acetic acid (HAc, 99%) was purchased from Merck KGaA, Darmstadt. 1,1,1,3,3,3-hexafluoro-2-propanol (HFP, Z99.8%) was purchased from Sigma-Aldrich and used as received. Two different samples were prepared by dissolving collagen in different solvent systems:
224
M. Kazanci / Materials Letters 130 (2014) 223–226
Fig. 1. SEM images of electrospun nanofibers from (a) HFP and (b) 40% HAc, and TEM images of electrospun nanofibers from (c) HFP and (d) 40% HAc.
(1) 10% w/v collagen solution was prepared by dissolving collagen in HFP; (2) 40% w/v collagen solution was prepared by dissolving collagen in 40% v/v HAc. The prepared electrospinning solutions were loaded into a 2 mL syringe (Omnifix, B. Braun, Melsungen, Germany) with a blunt end nozzle, controlled by a syringe pump (Pump 33 Harvard Apparatus, Holliston, USA). The solution was pushed through a capillary blunt steel needle (21 gauge, 0.7 mm i.d. 50 mm length) at a constant speed (0.5 μL/min). The steel needle was coupled to a high voltage source (Spellman Bertan Series 205B, NY, USA). The electric potential was needed to start the spinning process and thus form a jet. The applied DC voltage was held at 16 kV. A Cu collector was placed 15 cm from the needle tip to collect the electrospun collagen nanofibers. The nanofiber meshes were collected on cover-glasses placed onto the collector.
3. Results and discussion Fig. 1 (a) and (b) shows the SEM images, and (c) and (d) shows the TEM images of the electrospun nanofibers obtained from two different solvents. Electrospinning yielded randomly oriented and
Fig. 2. Raman shift of (a) native collagen, (b) electrospun collagen nanofibers obtained from HAc, and (c) electrospun collagen nanofibers obtained from HFP.
M. Kazanci / Materials Letters 130 (2014) 223–226
225
Fig. 3. CD spectra of electrospun nanofiber meshes a) dissolved at different temperatures, and b) obtained from different solvents.
interconnected fibrous meshes with the fiber diameter in the nanometer range. The electrospun nanofiber diameters were mainly determined by the collagen concentrations and electrospinning solvents. 10% collagen solution in HFP produced electrospun nanofibrous meshes with the fiber diameter. ranging between 150 and 200 nm (Fig. 1 (a)), and 40% collagen in 40% HAc produced electrospun nanofibrous meshes in a wide range of fiber diameters and with some bead formations (Fig. 1b). Collagen can be easily dissolved under acidic conditions to produce an electrospinning solution. However, the slow evaporation rate of the acid and its strong affinity with collagen leads to the deposition of wet fibers on the target, which can partially weld together and lessen the porosity of the mat. There was no observation of triple helix folding or the characteristic 67 nm D-spacing at TEM images shown in Fig. 1(c) and (d). The Raman shift of native collagen displays the peak position at 1670 cm 1 which is the finger print of type I collagens [13] with a shoulder at 1645 cm 1 at Amide I region [14]. Amide I band arises from CQO stretching of the peptidic bond in the Gly-X-Y tripeptide sequence. We observed broader peaks and peak shifts to higher wavenumbers for both type of collagen nanofibers in Fig. 2. The Raman peak positions at 1677 cm 1 are mainly indications of β-sheet formations for the electrospun nanofibers [15]. The frequency shift of the Amide I band suggests a different proportion of amino acid residues. CD utilizes the differential absorption of left and right handed circular polarized light in an asymmetric environment to assess secondary structure. The left handed polyproline II (PPII) helix gives rise to a circular dichroism spectrum that is remarkably similar to that of folded/unfolded proteins [16]. Single wavelength estimation of collagen folding was employed to extract the fraction of triple helix, mainly relying on the equilibrium state of collagen spectra. The fraction of folded collagen (θ) can be defined as θ ¼ ðθobs –θu Þ=ðθt –θu Þ
ð1Þ
where θobs, θt, and θu represent the ellipticities of observed, triple helix, and unordered state, respectively. θobs, θt and θu were measured at a wavelength of 221.5 nm for the sample in interest, and for the collagen solutions at 10 1C and 90 1C, respectively, in 0.05 M HAc [11]. Fig. 3 (a) shows the CD spectra of electrospun collagen nanofibrous mesh, obtained by using the same solvent (in 0.05 M HAc) and at the same collagen concentration (10%). However, the one which was solubilized at room temperature (25 1C) displayed relatively lower PP-II fraction ratio (θ¼0.37) than the one which was solubilized at 15 1C (θ¼ 0.52). Fig. 3 (b) shows the native collagen spectra and two electrospun collagen nanofibers spectra that were obtained from different solvents at different
concentrations. We observed slight increase in the PP-II ratio of collagen that was obtained from acidic solvent (θ¼ 0.59). It is known that collagen better preserved its native structure in acidic solvents [11]. The solubilizing temperature of the collagen nanofibers for the structural characterization methods is extremely important and it will definitely influence the experimental results.
4. Conclusion In the previous study [11], we stated that the spinning process causes refolding of nanofibers, but PP-II fraction did not exceed 42%. In this work, solubilizing the electrospun collagen nanofibers at lower temperatures (15 1C) for CD measurements, it was possible to increase the folded fraction ratio from 37% to 52.5%. The preparation temperature is extremely important for collagen samples. They are extremely sensitive to the temperature and this could be also one of the important factors that cause discrepancy in experimental results in the literature. It was possible to increase the folded fraction of collagen by almost 1.5 fold, just decreasing the temperature by 101. Another important result is that it is well known that collagen best preserves its native structure in acidic solvents. However, there was no real success obtaining electrospun collagen nanofibers from acidic solvents, due to their relatively low evaporation temperatures that cause the binding of fibers and deforms the fiber morphology. In this study, the electrospun collagen nanofibers were successfully obtained from 40% HAc, by increasing the collagen concentration to 40%. Otherwise the bead formation is more pronounced than the fiber formation at low concentrations. There is a small increase in the folded ratio, that is almost negligible, compared to the ones obtained from fluorinated solvents (θ¼0.59 vs. θ¼0.52). The morphology of the nanofibers from acidic solvent displays 3D porous, interconnected network rather than the fibrous structures, due to the slow evaporation rate of the acid and its strong affinity with collagen.
Acknowledgments This study is supported by TUBITAK's Co-Funded Brain Circulation Program (2236, Project number 112C025). The author would like to thank Dr. Jochen Bürck and Bianca Posselt for the CD measurements (IBG2, Karlsruhe Institute of Technology, Germany). The author appreciates Onur Aras' help in the preparation of electrospun collagen nanofibrous meshes (School of Medicine, Bahcesehir University, Istanbul, Turkey).
226
M. Kazanci / Materials Letters 130 (2014) 223–226
References [1] Zeugolis DI, Khew ST, Yew ESY, Ekaputra AK, Tong YW, Yung L-YL, et al. Biomaterial 2008;29:2293–305. [2] Teo W,E, Ramakrishna S. Nanotechnology 2006;17:89–106. [3] Lynn AK, Yannas I, Bonfield V, Biomed. WJ. Mater Res B: Appl Biomater 2004;71B:343–54. [4] Cheng YX, Ramos D, Lee P, Liang DN, Yu XJ, Kumbar SGJ. Biomed Nanotechnol 2014;10:287–98. [5] Liu J, Argenta L, Morykwas M, Wagner WD. J Biomed Appl 2014;28:729–38. [6] Yang L, Fitie CFC, Werf KO, Van der; Bennink ML, Dijkstra PJ, Feijen J. Biomaterials 2008;29:955–62. [7] dos Santos AM, Dierck J, Troch M, Podevijn M, Schacht E. Macro Mater Eng 2011;296:637–44.
[8] [9] [10] [11] [12] [13] [14] [15] [16]
Cai YS, Gevelber M. J Mater Sci 2013;48:7812–26. Whited BM, Rylander MN. Biotechnol Bioeng 2014;111:184–95. Oh B, Lee CH. Expert Opin Drug Deliv 2013;10:1565–82. Bürck J, Heissler S, Geckle U, Ardakani MF, Schneider R, Ulrich AS, et al. Langmuir 2013;29:1562–72. Diesner MO, Welle A, Kazanci M, Kaiser P, Spatz J, Koelsch P. Biointerphaces 2011;6:171–9. Nguyen TT, Gobinet C, Feru J, Brassart -Pasco S, Manfait M, Piot O. Spectroscopy 2012;27:421–7. Kazanci M, Roschger P, Paschalis EP, Klaushofer K, Fratzl P. J Struct Biol 2006;156:489–96. Gullekson C, Leanne Lucas L, Hewitt K, Kreplak L. Biophys J 2011;100:1837–45. Rucker AL, Creamer TP. Protein Sci 2002;11:980–5.