Biomacromolecules 2011, 12, 75–81
75
Influence of Tyrosine-Derived Moieties and Drying Conditions on the Formation of Helices in Gelatin Alessandro Zaupa,†,‡ Axel T. Neffe,†,‡ Benjamin F. Pierce,† Ulrich No¨chel,† and Andreas Lendlein*,†,‡ Center for Biomaterial Development and Berlin-Brandenburg Center for Regenerative Therapies, Helmholtz-Zentrum Geesthacht, Kantstrasse 55, 14513 Teltow, Germany, and Institute of Chemistry, University of Potsdam, 14476 Potsdam-Golm, Germany Received September 1, 2010; Revised Manuscript Received October 29, 2010
The single and triple helical organization of protein chains strongly influences the mechanical properties of gelatinbased materials. A chemical method for obtaining different degrees of helical organization in gelatin is covalent functionalization, while a physical method for achieving the same goal is the variation of the drying conditions of gelatin solutions. Here we explored how the introduction of desaminotyrosine (DAT) and desaminotyrosyl tyrosine (DATT) linked to lysine residues of gelatin influenced the kinetics and thermodynamic equilibrium of the helicalization process of single and triple helices following different drying conditions. Drying at a temperature above the helix-to-coil transition temperature of gelatin (T > Tc, called Vshort) generally resulted in gelatins with relatively lower triple helical content (Xc,t ) 1-2%) than lower temperature drying (T < Tc, called Vlong) (Xc,t ) 8-10%), where the DAT(T) functional groups generally disrupted helix formation. While different helical contents affected the thermal transition temperatures only slightly, the mechanical properties were strongly affected for swollen hydrogels (E ) 4-13 kPa for samples treated by Vlong and E ) 120-700 kPa for samples treated by Vshort). This study shows that side group functionalization and different drying conditions are viable options to control the helicalization and macroscopic properties of gelatin-based materials.
Introduction Each biomedical application requires materials with tailored properties such as mechanical and thermal properties, water uptake and swelling behavior, as well as degradation behavior.1-4 Polymer systems, whose macroscopic properties can be adjusted in a wide range by only slightly changing their chemical structures, were developed to meet this demand.5-7 An attractive strategy for obtaining tunable polymer systems with multiple functions is to enable defined but reversible physical interactions along the polymer chains. In biopolymers, such physical interactions control their self-organization under physiological conditions,8 which among other factors depends on the electronic properties of the repeating units9 as well as the possibility to form hydrogen bonds in aqueous conditions.10,11 Collagen, the most abundant protein present in mammals, is a dynamic fibrillar system that undergoes self-organized helicalization processes during its biosynthesis.12 The single chains of collagen contain the highly repetitive triad (Gly-Xaa-Yaa)n, with Xaa being mostly proline (Pro) and Yaa predominantly hydroxyproline (Hyp), that gives rise to a left-handed helical conformation (“polyproline type II”), which is more elongated than typical R-helices as there are less hydrogen bonds because of the larger number of imino acids.16 Collagen naturally reorganizes such that three of these helices associate and intertwine to give a right-handed triple helix. The triple helix formation and stabilization relies on weak, cooperative forces and is dependent on the amino acid sequence. Specifically, hydroxyproline, hydroxylysine, lysine, aspartic acid, and glutamic acid tend to stabilize the triple helix, while aromatic amino * To whom correspondence should be addressed. Fax: +49-3328-352452. E-mail:
[email protected]. † Helmholtz-Zentrum Geesthacht. ‡ University of Potsdam.
acids destabilize the helix.17,18 Interestingly, the assembly of collagen fibrils during their biosynthesis is controlled by aromatic residues present at the nonhelical regions at both ends of a collagen molecule, known as telopeptides, which increase the kinetics of fibril self-association.19 The strong influence of the amino acid side chains on the formation of the triple helices is also evidenced by failed assembly of the collagen chains in hereditary, even lethal, diseases due to the exchange of single amino acids.20,21 The application of systems based on biopolymers derived from the extracellular matrix (ECM) is promising in various areas of regenerative medicine, such as bone regeneration,13 as the ECM plays an important role in healing processes.13 However, unfunctionalized collagen-based systems suffer from low or unstable mechanical properties, fast degradation behavior, and are generally unsuitable for long-term use.14 An alternative strategy involves using gelatin,15 which is derived by partially hydrolyzing collagen under deglycosylation, as a starting material to yield bone implants with a range of mechanical properties. For such material-induced autoregeneration applications, the mechanical properties of gelatin-based systems, which are at least partially influenced by the dynamic processes of internal helical (re)organization, need to be controlled.16-18 Gelatin chain organization is thermodynamically driven but kinetically controlled and the effective formation of stable triplehelical net points is affected by concentration, thermomechanical history, molecular weight distribution, and the amino acid sequence, which depend on the gelatin/collagen source.11,19-22 Gelatin can be dissolved in water and forms low viscosity solutions at temperatures above ca. 35 °C (helix-to-coil transition, Tc). On a molecular level, this dissolution occurs because the cooperative interactions between the gelatin chains are disrupted at T > Tc, forming random single coils. Cooling of
10.1021/bm101029k 2011 American Chemical Society Published on Web 12/08/2010
76
Biomacromolecules, Vol. 12, No. 1, 2011
Figure 1. (a) Structures of desaminotyrosine (DAT, 1) and desaminotyrosyl-tyrosine (DATT, 2). (b) Aim of this work was to establish if functionalization of gelatin with 1 or 2 influences the balance between triple helical and amorphous chain organization depending on the drying process.
such gelatin solutions to T < Tc results in physically cross-linked gels that, following certain drying conditions, possess at least partially reassociated collagen-like triple helices.20,21,23 Generally, low molecular weight gelatin or materials obtained by drying at T > Tc result in amorphous materials with lower triple helical contents and lower gel stability, which are less elastic17,23,24 than dried materials produced from higher molecular weight gelatins or those dried at T < Tc.25 However, gelatin is highly hygroscopic26 and may easily undergo structural changes during (re)wetting or during storage, which affect its helical content, thermal transitions, and mechanical properties.17,18 Therefore, chemical functionalization, which can be used to control the orientation of gelatin chains, can improve its performance in material-induced autoregeneration applications.14 The synthetic control of the helicalization can be achieved by covalently cross-linking gelatin in aqueous solution at T > Tc, which fixes the network permanently in a random-coil state so that the mechanical properties of the system solely depend on the density of the covalent cross-links.27 An alternative method involves enabling defined physical interactions within gelatin by mixing28 or chemically functionalizing29 gelatin chains with small molecule phenols, which stabilize gelatin gels via H-bonding between the phenolic hydroxyl function and the gelatin chains. In a previous communication,29 we used computational modeling to develop gelatin-based materials functionalized with the tyrosine-derived desaminotyrosine (DAT, 1) or desaminotyrosyl tyrosine (DATT, 2; Figure 1a). The computer models suggested specific interactions between the aromatic groups would lead to physical netpoints30 and indeed resulted in materials with increased Young’s moduli E, elongation at break εb, and maximum tensile strength σmax and reduced water uptake.29 These materials have recently been investigated as matrices for composite materials with hydroxyapatite.31 The aim of the work described here was to tune the extent of chain-chain interactions in gelatin, DAT-functionalized gelatin (Gel-DAT), and DATT-functionalized gelatin (Gel-DATT) by varying the drying conditions during film formation as well as their final water contents (Figure 1b).32,33 One treatment involved drying films below the gelation temperature (T < Tc), which would favor helix formation, while the other involved drying casted films above gelation temperature (T > Tc), which would hinder the formation of helices. Because the treatment procedures (V) achieve different time periods of chain mobility in the gelatin samples,34 we denote the lower temperature
Zaupa et al.
Figure 2. WAXS spectra of gelatin, Gel-DAT, and Gel-DATT treated by vlong drying procedure (top); vshort drying procedure (bottom) (light gray line) gelatin, (gray line) Gel-DAT, (black line) Gel-DATT.
treatment procedure as Vlong since longer times are required for film formation and the higher temperature treatment procedure as Vshort because shorter times are required for film formation. In addition, the absolute humidity (a.h.) of the drying chamber was varied during film formation of Vlong treated samples to obtain further control of helical (re)formation. The relative changes in helix content and transition temperatures were investigated by WAXS and temperature modulated (TM)-DSC. The influence of supramolecular organization on macroscopic properties was exemplarily analyzed by tensile tests on samples with 25 wt % water content and at equilibrium swelling (hydrogels), while rheological measurements were used to investigate gel strength and stability.
Experimental Section Materials. Gelatin type A, β-mercaptoethanol, desaminotyrosine (DAT, Figure 1, 1), N-hydroxysuccinimide (NHS), 1-ethyl-3-(3dimethyl-aminopropyl) carbodiimide (EDC), 2,4,6-trinitro-benzensulfonic acid (TNBS), and N,N-diisopropylethylamine (DIPEA) were purchased from Sigma Aldrich (Munich, Germany). Dimethyl sulfoxide (DMSO) was purchased from Merck (Darmstadt, Germany). All reagents and solvents were of analytical grade and used without further purification. Synthesis of Desaminotyrosyl-tyrosine (DATT, Figure 1, 2). The desaminotyrosyl-tyrosine was obtained by activation of desaminotyrosine (12.6 g, 76 mmol, 1.2 equiv) with EDC (17.4 g, 91 mmol, 1.4 equiv) and DIPEA (37 mL, 215 mmol, 3.4 equiv) in NMP (80 mL) at -5 °C and further reaction with H-L-Tyr-OtBu (63 mmol, 1 equiv) in NMP (30 mL) at room temperature for 17 h. A total of 1 L of water was added, the mixture extracted with ethyl acetate (3×), and the combined organic phases were washed with 0.1 M aq HCl, 0.1 M aq NaHCO3, and concd NaCl solution. The organic phases were dried over magnesium sulfate, filtered, and the solvent removed by evaporation under reduced pressure to give the protected DATT (14.35 g, 69%) as a white powder. C22H27NO5 m/z (ESI): Calcd, 385.19; Found, 386.19 (M + H+). mp 140.3 °C. 1H NMR (DMSO-d6) δ (ppm) 9.18, 9.10 (1H, s, OH), 8.09 (1H, d, NH), 6.95 (4H, mc, Ph), 6.63 (4H, mc, Ph), 4.27 (1H, mc, CRH), 2.79 (1H, mc, CßH), 2.73 (1H, mc, CßH), 2.63 (2H, mc, CßH), 2.31 (2H, mc, CRH), 1.32 (9H, s, CH3). IR (υmax/cm-1): 3342.85 (O-H), 2978.22 (C-H alkane), 2956.97 (C-H aliphatic), 1712.13 (s, CdO), 1645.3 (s, N-H), 1613.51 (N-H), 1600.01 (N-H), 1514.17 (vs, C-N), 1367.61 (C-O-), 1151.54 (C-O). Deprotected desaminotyrosyl-tyrosine (2) was obtained by reaction of tBu protected DATT (9 g, 23 mmol) with TFA (76 mL) in DCM (153 mL) at 0 °C (1 h) first and 12 h at room temperature. The solvents
Influencing the Helix Formation in Gelatin were evaporated under reduced pressure, and the residue was redissolved in 0.1 M HCl and extracted with EtOAc. The organic phase was washed with concd NaCl solution, dried over magnesium sulfate, filtered, and the solvent was removed by evaporation under reduced pressure to yield pure product 2 (7.1 g, 93%). C18H19NO5: m/z (ESI) Calcd, 329.13; Found, 330.13 (M + H+). mp 161.2; 1H NMR (DMSOd6) δ (ppm) 12.57 (1H, s, COOH), 9.18, 9.10 (1H, s, OH), 8.06 (1H, d, NH), 6.96 (4H, mc, Ph), 6.63 (4H, mc, Ph), 4.34 (1H, mc, CRH), 2.89 (1H, mc, CßH), 2.72 (1H, mc, CßH), 2.61 (2H, mc, CßH), 2.30 (2H, mc, CRH). IR (υmax/cm-1): 3247.3 (OH), 2957-2919.4 (C-H aliphatic), 1718.7- 1707 (vs, CdO), 1651.1 (s, N-C), 1613.51 (N-H), 1600.01 (N-H), 1514.17 (vs, C-N), 1549.6 (COO-), 1333.3 (COOH), 682 (COOH). Functionalization of Gelatin. DAT or DATT (29 mmol) were activated by reaction with EDC (32 mmol) and NHS (43 mmol) in 110 mL of DMSO at 37 °C. After 3 h, β-mercaptoethanol (43 mmol) was added. A gelatin solution (15 g in 150 mL of DMSO) was added, and the mixture was stirred at 37 °C for 5 h. The functionalized products (Gel-DAT and Gel-DATT, respectively) were then precipitated in ethanol, filtered, washed with ethanol and acetone, and dried under vacuum. The degree of substitution (90 mol %) was determined by 1H NMR spectroscopy and a TNBS colorimetric assay.35 Preparation of Films and Conditions of Hydration. Films were prepared by casting a 5 wt % aq gelatin solution into polystyrene Petri dishes. Drying conditions (temperature and absolute humidity (a.h.)) were varied using a climatic chamber KBT (Binder). The two processes were Vlong (10 °C and 2.81 g · m-3 a.h. until constant weight was achieved, which was obtained following 2 days of drying time) and Vshort (40 °C and 40.83 g · m-3 a.h. until constant weight was achieved, which was obtained within hours). In an experiment that focused on varying the water evaporation time during Vlong drying, 10 °C drying was performed at different a.h. of 0.92, 4.69, and 7.50 g · m-3. After 1, 3, and 10 days, respectively, constant weight was achieved. Shortly after drying, the films were cut into dogbone-shaped samples for tensile tests. The samples were later statically hydrated in a KBT climatic chamber (Binder, Tuttlingen, Germany) at 24.41 g · m-3 a.h. and 30 °C for 1 week to obtain films with 25 wt % water content.17 Temperature Modulated Differential Scanning Calorimetry (TM-DSC). A Pho¨nix DSC 204 F1 (Netzsch) apparatus was used to measure the melting point (Tm), enthalpy of melting (∆Hm) during the first heating run, the glass transition temperature (Tg), and the heat capacity change (∆cp) at the glass transition during the second heating run of polymer samples. The gelatin samples were hermetically sealed in a pan to prevent any loss of moisture during TM-DSC measurements. Two consecutive heating runs from -20 to 150 °C in the first heating run and up to 250 °C in the second heating run were performed with a modulated heating rate with a period of 60 s, amplitude of 0.5 °C, and an underlying heating rate of 5 °C · min-1. Between heating runs, the samples were rapidly cooled (10 °C · min-1). Wide-Angle X-ray Scattering Investigation (WAXS). Measurements were performed on a Bruker D8 Discover (Bruker AXS, Germany). The X-ray generator was operated at a voltage of 40 kV and a current of 40 mA. A copper anode and a graphite monochromator produced Cu KR radiation with a wavelength λ ) 0.154 nm. WAXS images were collected from gelatin films (thickness of about 0.3 mm, exposure times of 120 s per frame) in transmission geometry with a collimator-opening of 0.8 mm at a sample-to-detector distance of 15 cm. Samples were oscillated perpendicular to the primary beam direction (1 mm to average over a large region of the sample. Integration of the two-dimensional scattering data gave the intensity as a function of the scattering angle 2θ. A beamstop was placed immediately after the sample to avoid air scattering. Two procedures were tested for the extraction of the individual peak areas. A peakfitting program (TOPAS, from Bruker) inserted peaks at the desired positions and fit them to the total scattering curve, and then the peak areas were calculated. A second method consisted of cutting the helixpeaks from the curve and fit the residual data points with a spline
Biomacromolecules, Vol. 12, No. 1, 2011
77
function (leading in the pure amorphous scattering), after subtraction from the total scattering, the helix peaks were obtained, and the areas were calculated by integration. Mechanical Tests. Young’s modulus, E, maximum tensile strength, σmax, and elongation at break, εb, were measured using standard test specimen (ISO 527-2/1BB) punched from dried gelatin films for a minimum of five samples (sample thickness ) 380 ( 30 µm) using a Zwick Z005 (Zwick GmbH, Ulm, Germany), where the averages of the results from these samples are reported with standard deviation. Dry samples were statically hydrated in a climatic chamber KBT (Binder) at 24.41 g · m-3 a.h. and 30 °C for 1 week to obtain films with 25 wt % water content.17 For hydrogel measurements, the punched samples were swollen and the dimensions of hydrogels were measured with a micrometer screw gauge. A slow extension rate of 0.033 mm · s-1 was adopted to consider the elongation regime as isothermal. The tensile tests were performed at 25 °C for dry sample and samples swollen at 23 °C. Tensile properties were determined from stress-strain (σ ) f(ε)) diagrams. Degree of Swelling. The degree of swelling Q of gelatin films in water at 23 °C was calculated by36
Q ) 1 + Fg ·
(
msw 1 mdFH2O FH2O
)
(1)
where msw is the weight of the sample in the swollen state, md is the weight of the dry extracted sample, and FH2O (1 g · cm-1) and Fg are the specific densities of the water and of gelatin, respectively. Fg was determined using a Ultrapycnometer 1000 (Quantachrome GmbH & Col. KG, Germany) on gelatin powders dried for 48 h under vacuum. Rheology. Oscillatory experiments were performed with a plate-plate rheoviscosimeter (Thermo Electron, Karlsruhe, Germany) with a Peltierelement of 20 mm diameter. All dynamic measurements were performed in the linear viscosity region measured at 23 °C with a frequency of 1 Hz and constant stress of 5 Pa. A solvent trap was used to prevent evaporation of solvent. The viscoelastic properties of the associative networks were determined by measuring changes in the storage modulus G′ and loss modulus G′′ as a function of temperature by oscillatory shear stress at different temperatures ranging from 20 to 60 °C with a heating rate of 2.6 °C · min-1. The error for the Tc determination was (1-2 °C.
Results and Discussion Aqueous solutions (5 wt %) of unfunctionalized gelatin, GelDAT, and Gel-DATT were dried at T < Tc following the Vlong procedure or the Vshort procedure at T > Tc (see Experimental Section for details) giving films with thicknesses of 380 ( 30 µm. Drying at T < Tc (Vlong procedure) allowed the association and helicalization of chains given longer drying times, while drying at T > Tc (Vshort procedure) imparted higher chain mobility to the gelatin samples but for a shorter drying time, which inhibited helix formation and resulted in more amorphous films. Following the treatments Vlong and Vshort for aqueous solutions of gelatin, Gel-DAT, and Gel-DATT, the chain organization in the dried samples was investigated by wide-angle X-ray scattering (WAXS, Figure 2). By dividing the integrated intensity of the peaks that account for the lateral distances between single collagen chains in triple helices by the amorphous peak areas, it was possible to determine the degree of renaturation or relative amount of triple helices within the samples (Xc,t). Analogously, the single-helix peak area in relation to the amorphous area permitted the determination of the singlehelix ratio (Xc,s). This method monitored the changes in the scattered peak areas and therefore quantified the relative changes in the helix content of the samples (Xc,t and Xc,s, Table 1). The
78
Biomacromolecules, Vol. 12, No. 1, 2011
Zaupa et al.
Table 1. Evolution of Relative Single (Xc,s) and Triple Helix (Xc,t) Content in the WAXS Spectra for Gelatin Samples Dried Using vshort and vlong Procedures as a Function of Time vshort treated samples
vlong treated samples 1 day
2 days
3 days
Xc,t Xc,s Xc,t Xc,s Xc,t Xc,s sample ID [%] [%] [%] [%] [%] [%]
10 days Xc,t [%]
Xc,s [%]
Xc,t [%]
Gel 9.9 1.8 8.0 2.1 9.6 3.0 10.3 2.7 Gel-DAT 7.3 1.0 7.5 1.5 7.3 2.3 7.7 1.7 Gel-DATT 6.2 1.8 5.6 1.8 7.5 1.4 8.2 2.2
2.3 0.9 0.8
WAXS spectra for gelatin films (gelatin, Gel-DAT, and GelDATT) treated with Vlong showed three peaks. In addition to the broad scattering peak around 2θ ) 21°, which represents the average distance and distribution of the random coil chains (amorphous region), peaks at 2θ ) 7.5° (corresponding to the diameter of a triple-helix perpendicular to the chain direction) and at 2θ ) 32° (corresponding to the distance between prolines in a turn in the polyproline-II-helical single helices) were observed. These spectra indicated that the helicalization process was favored under the Vlong drying process (Xc,t ) 8-10%), which was in agreement with the literature.37,38 The presence of one (Gel-DAT) or two (Gel-DATT) aromatic groups per lysine caused a reduction in Xc,t following the Vlong treatment. When changing the drying temperature to 40 °C and the a.h. to 40.83 g · m-3 ()80% r.h.) (Vshort), a shorter time was required for constant weight to be achieved for the films than in Vlong,, and the higher temperature (T > Tc) hindered helix nucleation. Consequently, the WAXS spectra of Vshort treated gelatin showed only a small degree of single helices still present and low amounts of triple helical regions were observed (Xc,t ) 2.3%), while Vshort treated Gel-DAT and Gel-DATT contained small amounts of helical contents as calculated from their WAXS spectra (Xc,t ) 0.9% for Gel-DAT and 0.8% for Gel-DATT). Although longer chain mobility time periods generally promote formation of triple helices,39 the tendency for Gel-DAT and Gel-DATT to adopt typical collageneous features was reduced compared to gelatin by the introduction of the bulky aromatic groups. A kinetic drying study of the Vlong procedure at different humidities was performed for the three materials (Table 1) to identify whether the reduction in helical content observed in our study was a consequence of limited mobility of the chains due to the bulkiness of the introduced groups (kinetic effect). For Vshort treated films, drying at different relative humidities altered the time points at which constant water content and chain fixation were reached. On the molecular level, the X-ray patterns of the films showed in all cases the presence of a triple helical structure (2θ ) 8°) and the single and triple helical contents increased with the drying time (i.e., with the humidity during the drying). The maximum relative content Xc,t ) 10.3% of triple-helical structure was determined for the gelatin dried with the longest drying process. For the functionalized gelatins Gel-DAT and Gel-DATT, the same trend was observed although Xc,t was reduced compared to pure gelatin. When comparing the WAXS spectra for three different humidities during the drying, which corresponded to the drying time periods, the single helical content increased first followed by an increase in triple helical contents only when the drying time was elongated. This can be explained by a two-step renaturation process into triple helices, where first an organization of chains into single helical polyproline-II helices took place followed by an association process that formed triple helices.40,41 A separate report describing helix formation kinetics also divided
the process into multiple phases, which included the helix nucleation and helix growth.19,21 The helix nucleation process determines the number of triple helices and is temperature and concentration dependent. The helix growth process determines the length of helices and is temperature independent at T < Tc. The results from this WAXS study indicated that the introduction of bulky aromatic groups introduced by DAT(T) interfered more with the triple helix formation than with single helical formation during the Vlong treatments. Moreover, Gel-DAT and Gel-DATT showed lower Xc,t values following Vshort drying treatment compared with Vshort treated gelatin, although the differences observed between the samples for each treatment were relatively minor. The gelatin-based samples were characterized by TM-DSC to investigate whether the aromatic groups created a change in morphology within the gelatin-based materials (Table 2). The melting temperatures Tm correspond to the disaggregation of triple helices while the glass transition temperatures Tg correspond to chain mobility. Samples treated according to Vlong procedure showed a decrease of Tm from 142 to 126 °C when increasing the number of aromatic groups, which indicated a lower stability of the helical structures or a reduction of helix content when increasing the number of aromatic groups. The Tg of gelatin after Vlong treatment (116 °C) decreased to 99 °C for Gel-DAT and 106 °C for Gel-DATT. This decrease in Tg following functionalization with bulky side groups can be explained by the decrease in barriers of rotation in the random coils due to the disruption in helical structures. Furthermore, the mobility of chains in Gel-DAT and Gel-DATT was likely slightly higher than in gelatin as any residual water was not as tightly bound. This effect might be related to there being fewer interchain hydrogen bonds in Gel-DAT and Gel-DATT due to the introduced functional groups. The relative increase of Tg in Gel-DATT compared to that of Gel-DAT can be attributed to the fact that there were additional netpoints formed by aromatic clusters as predicted in modeling studies of this system.29,30 The predominantly amorphous gelatins obtained after Vshort treatment likewise showed a reduction of Tm and Tg comparing gelatin and the functionalized gelatins. The presence of Tm in the DSC thermograms for the Vshort treated samples was in agreement with the WAXS data, which indicated smaller yet still significant amounts of helical content. The Tgs from Vshort treated samples had the same tendency as observed above, where unfunctionalized gelatin was higher than DAT and DATT. The reduction of Tg in natural gelatin is potentially related to an increase in water bound to the gelatin backbone in case of amorphous samples leading to a stronger plasticization effect during TMDSC measurements.17,20,24,33,42,43 As no change in ∆cp was observed for this set of samples, the water content in the samples were similar, meaning that the presence of aromatic groups was not the only factor determining chain mobility. The decrease in Tm for the functionalized systems demonstrated that the aromatic side groups destabilized helices, which indicated that thermodynamic disruptions were present in addition to the kinetic differences in Vlong and Vshort drying processes. The aromatic functional groups had a strong influence on the swelling properties of the gelatin hydrogels. A decrease in the degree of swelling at equilibrium (Q) with increasing number of aromatic moieties introduced per lysine modification was observed. Gelatin films treated with Vlong showed a degree of swelling of 2080 ( 65 vol %, while the introduction of one aromatic moiety per lysine residue in Gel-DAT led to a decrease in volumetric swelling (1580 ( 440 vol %) and Gel-DATT showed drastically reduced degree of swelling (300 ( 5 vol
Influencing the Helix Formation in Gelatin
Biomacromolecules, Vol. 12, No. 1, 2011
79
Table 2. Thermal Transitions Determined by TM-DSC of Gelatin-Based Filmsa vlong treatment -1
vshort treatment -1
sample ID
Tm [°C] 1st run
∆Hm [J · g ] 1st run
Tg [°C] 2nd run
∆cp [J · (g · K) ] 2nd run
Tm [°C] 1st run
∆Hm [J · g-1] 1st run
Tg [°C] 2nd run
∆cp [J · (g · K)-1] 2nd run
Gel Gel-DAT Gel-DATT
142 131 126
18 26 19
116 99 106
0.35 0.33 0.29
144 123 127
22 17 14
110 103 105
0.35 0.35 0.36
a Tg: glass transition temperature; ∆cp: change of heat capacity at Tg; Tm: melting temperature, corresponding to the denaturation and ∆Hm: denaturation enthalpy.
Figure 3. Dynamic rheological measurements of swollen gelatin hydrogels Gel, Gel-DAT, and Gel-DATT. Temperature dependence of the storage modulus G′ (black square) and the loss modulus G′′ (gray square) for natural gelatin (a). Temperature dependence of the storage modulus G′ for vlong (b), and vshort (c) treated gelatin (light gray square), Gel-DAT (gray square), and Gel-DATT (black square). Measurements were performed at constant stress 4 Pa and a frequency of 1 Hz, in a temperature range form 10 to 50 °C with a temperature ramp of 2.6 °C · min-1.
%). For films treated by Vshort a similar trend was observed, however, the absolute values were slightly higher (Q varying from 2805 ( 330 for gelatin to 1730 ( 130 for Gel-DAT, and 265 ( 55 for Gel-DATT), which can be attributed in each case to the lower amounts of triple-helical domains. In addition to the drying processes Vshort and Vlong, the differences observed in these swelling experiments highlight the effects of the aromatic side groups in Gel-DAT and Gel-DATT with respect to gelatin. The hydrogels were furthermore subjected to dynamic mechanical analysis. Figure 3a shows the change of the storage modulus G′ and the loss modulus G′′ as a function of temperature for gelatin treated by Vlong and Figures 3b,c show the change of G′ comparing the differently functionalized
gelatins. Exact values are reported in Table 1 of the Supporting Information. G′ (T < Tc) values can be compared to estimate the differences in storage modulus within this temperature range. The temperature associated to the dissolution of the elastic networks (Tc) was identified here as the crossover point between G′ and G′′ during the rheological measurements,44 which were higher than values previously reported for bovine-hide gelatin and tuna-skin gelatin (Tc < 30 °C).22 At T < 30 °C, the gelatin systems showed storage moduli G′ with a pronounced plateau between 1 and 10 kPa and loss moduli G′′(ω) that ranged from 10 to 250 Pa, as determined in frequency sweep measurements. The G′ values for the different samples did not significantly differ at T < Tc; however, Tc was lower for the samples treated with Vlong than for the samples treated with Vshort for gelatin and Gel-DAT, while Tc was similar in both cases for Gel-DATT. The higher Tc values of Gel-DATT compared to Gel-DAT and gelatin in the Vlong samples might be related to the additional netpoints formed by the DATT groups, which were shown to stabilize the helices and network structures regardless of chain mobility time (WAXS). Data from tensile tests performed on materials with 25 wt % water content18 are shown in Table 3. The 25 wt % water content was selected for handling purposes as samples with lower water contents were too brittle and could not be characterized by tensile testing. All materials exhibited E ) 2-2.5 GPa under these conditions, which were similar values to previously reported for porcine gelatins.25 Gelatin films treated with Vlong, which possessed higher relative triple helix content than the samples produced with Vshort, showed elastic mechanical behavior with 25 wt % water content (Table 3). Arvanitoyannis et al. showed that mechanical properties of gelatin following different thermal treatments may become more disparate upon the introduction of a small molecule plasticizer, which can also increase the material elasticity.32 Vlong treated gelatin films and Gel-DAT/DATT films, which possessed a low triple-helical content, showed a brittle behavior of the material, which led to a reduction of maximum tensile strength σmax, elongation at break εb, and Young’s modulus E. These results were in agreement with previously observed mechanical properties for materials with differing chain mobility times.17,18,24 However, the mechanical properties of the functionalized gelatins with 25 wt % water content were not drastically affected by the different drying procedures, with greater differences being observed for the Gel-DAT samples than for the Gel-DATT samples. Overall, for materials containing 25 wt % water content, the elastic behavior did not correspond well with the samples’ respective renaturation degrees Xc,s or Xc,t. In addition, the mechanical properties of the hydrogels at equilibrium swelling were tested as these conditions were similar to potential applications under physiological conditions (Table 4). The change of mechanical properties according to the number of aromatic groups and helical content of samples at equilibrium swelling was remarkable. For example, tensile tests performed
80
Biomacromolecules, Vol. 12, No. 1, 2011
Zaupa et al.
Table 3. Mechanical Properties of vlong and vshort Treated Gelatin Films Containing 25 wt % Watera vlong treatment
a
vshort treatment
sample ID
E [GPa]
σmax [MPa]
εB [%]
E [GPa]
σmax [MPa]
εB [%]
Gel Gel-DAT Gel-DATT
2.38 ( 0.17 2.05 ( 0.47 2.05 ( 0.44
75 ( 11 69 ( 6 41 ( 12
14 ( 2.6 12 ( 1.4 2 ( 0.4
2.26 ( 0.27 2.03 ( 0.24 2.08 ( 0.46
69 ( 18 51 ( 27 39 ( 10
6 ( 1.2 3 ( 0.8 3 ( 0.9
E: Young’s modulus; σmax: maximum tensile strength; εb: elongation at break.
Table 4. Mechanical Properties of Polymer Samples at Equilibrium Swelling Measured by Tensile Test in Aqueous Environment at T ) 23 °Ca vlong treatment
a
vshort treatment
sample ID
E [kPa]
εb [%]
σmax [kPa]
E [kPa]
εb [%]
σmax [kPa]
Gel Gel-DAT Gel-DATT
10.4 ( 2.5 4.5 ( 0.5 8.6 ( 4.0
8.8 ( 1.3 21.2 ( 12 11 ( 8
1.9 ( 0.2 1.5 ( 0.6 2.1 ( 1.1
173 ( 59 190 ( 30 598 ( 92
15 ( 6 57 ( 23 80 ( 20
30 ( 10 60 ( 20 70 ( 40
E: Young’s modulus; σmax: ultimate tensile strength; εb: elongation at break.
in the swollen state showed a drastic increase in E and εB for Vshort treated samples compared to the Vlong treated samples (tensile curves: see Figure 2 in the Supporting Information). The difference in elastic behavior of the swollen samples can be correlated with their renaturation degree of triple helices Xc,t, where the Vshort treated samples containing Xc,t ) ∼1-2% showed higher E values than the Vlong treated samples containing Xc,t ) 8-10%. However, the trend observed here does not agree with a previous report from Djabourov that related elasticity and Xc,t in 5 wt % gelatin films,11 where the elastic moduli of gelatins were relatively low at Xc,t < 7% and further increased when Xc,t > 7% as tested between 24-28 °C. Rather, the data reported here showed that the swollen films containing Xc,t < 7% (Vshort treated samples) possessed much higher E values than the films containing Xc,t > 7% (Vlong treated samples). A possible explanation for this is that the Vlong treated samples, which possessed Xc,t ) 8-10%, agree with Djabourov’s conclusions because the amount of triple helical contents was sufficient to dictate the physical interactions at equilibrium swelling. Indeed, the previous models also showed that E ) 150 kPa for gelatin solutions (5 wt %) with Xc,t ) 15%, and the maximum E value obtained for the Vshort treated samples was 10 kPa for gelatin (Xc,t ) 10.3%), which are similar to the Vlong treated samples. Because the swollen Vshort treated samples were stronger and more elastic, yet possessed Xc,t ) 0.8-2.3%, their mechanical properties were ruled by other interactions, including those based on the aromatic functional groups for DAT- and DATTfunctionalized gelatins. The general increase in maximum tensile strength σmax, elongation at break εb, and E for Vshort treated samples can be explained by the swelling mechanism of gelatin, which causes the fibrils to align, expands the matrix, endows greater chain mobility, and ultimately strengthens the network.37,38 This also explains why the swollen unfunctionalized gelatin, which possessed no additional net points, also showed greater E values following Vshort treatment than following Vlong treatment, even though the Vlong gelatin had higher Xc,t. The E values increased for the Vshort treated gelatins with increasing aromatic functional groups because of the formation of additional net points as determined in molecular modeling studies.29,30 Overall, the differences between the Vshort and Vlong treated materials in the swollen state demonstrated how chemical functionalization combined with drying procedures may be used to achieve varied yet controlled properties in gelatin-based materials. The manipulation methods used here may be used to yield a new generation of gelatin-based implant materials for the induced autoregeneration of tissue, for example, bone, that may be
tailored according to the demands of a specific injury or particular patient.13,14
Conclusion Chemical and physical methods can be used to tailor the helical content of gelatin-based materials. Shorter drying times at elevated temperatures (Vshort drying procedure) created amorphous systems with lower helical contents (1-2%) when compared to samples dried at longer times at lower temperatures (Vlong drying procedure; 8-10%), as measured by WAXS. In addition, the functionalization of gelatin with DAT and DATT, which introduced one or two aromatic groups, respectively, per lysine residue of the gelatin chains, resulted in materials with lower tendencies to form helices when compared with pure gelatin. The reduction in the degree of renaturation was confirmed by TM-DSC analysis, which showed a reduction of helix stability in terms of reduced Tm. The degree of helicity furthermore had only little influence on the mechanical performance of the materials at 25 wt % water content (E ) 2.0-2.4 GPa), while at equilibrium swelling the amount of aromatic interactions showed to be a key factor (e.g., Q ) 2100 vol % for Vlong treated gelatin, 1600 vol % for Vlong treated Gel-DAT, and 300 vol % for Vlong treated Gel-DATT). The modification of the molecular architecture with aromatic moieties resulted in materials with reduced water uptake, increased gel strength, and increased elasticity at equilibrium swelling. The physical tunability within these networks provides different methods of attaining a broad range of consistent material properties that can be exploited in material-induced autoregeneration applications, for example, for bone. Potentially, these findings can be transferred to other synthetic and natural polymers, which would enable their controlled chain organization by side chain modification while suppressing inherent self-organization tendencies. Acknowledgment. We thank the Deutsche Forschungsgemeinschaft (DFG) for funding of this project through SFB 760 subproject B5, and Karolin Schma¨lzlin for the help with graphics. Supporting Information Available. WAXS spectra of gelatin, Gel-DAT and Gel-DATT after different drying processes, tensile curves, and values from the rheological measurements. This material is available free of charge via the Internet at http://pubs.acs.org.
Influencing the Helix Formation in Gelatin
References and Notes (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24)
Shastri, V.; Lendlein, A. MRS Bull. 2010, 35, 571–575. Shastri, V.; Lendlein, A. AdV. Mater. 2009, 21, 3231–3234. Middleton, J. C.; Tipton, A. J. Biomaterials 2000, 21, 2335–2346. Langer, R. Acc. Chem. Res. 1999, 33, 94–101. Lendlein, A.; Neuenschwander, P.; Suter, U. W. Macromol. Chem. Phys. 1998, 199, 2785–2796. Lendlein, A.; Colussi, M.; Neuenschwander, P.; Suter, U. W. Macromol. Chem. Phys. 2001, 202, 2702–2711. Lendlein, A.; Zotzmann, J.; Feng, Y.; Alteheld, A.; Kelch, S. Biomacromolecules 2009, 10, 975–982. Hong, H. D.; Park, S.; Jimenez, R. H. F.; Rinehart, D.; Tamm, L. K. J. Am. Chem. Soc. 2007, 129, 8320–8327. Burley, S. K.; Petsko, G. A. Science 1985, 229, 23–28. Park, S.; Radmer, R. J.; Klein, T. E.; Pande, V. S. J. Comput. Chem. 2005, 26, 1612–1616. Djabourov, M. Contemp. Phys. 1988, 29, 273–297. Speakman, P. T. Nature 1971, 229, 241–243. Place, E. S.; Evans, N. D.; Stevens, M. M. Nat. Mater. 2009, 8, 457– 470. Baroli, B. J. Pharm. Sci. 2009, 98, 1317–1375. Liu, X.; Smith, L. A.; Hu, J.; Ma, P. X. Biomaterials 2009, 30, 2252– 2258. Bigi, A.; Panzavolta, S.; Rubini, K. Biomaterials 2004, 25, 5675– 5680. Yakimets, I.; Wellner, N.; Smith, A. C.; Wilson, R. H.; Farhat, I.; Mitchell, J. Polymer 2005, 46, 12577–12585. Yakimets, I.; Paes, S. S.; Wellner, N.; Smith, A. C.; Wilson, R. H.; Mitchell, J. R. Biomacromolecules 2007, 8, 1710–1722. Normand, V.; Muller, S.; Ravey, J. C.; Parker, A. Macromolecules 2000, 33, 1063–1071. Gornall, J. L.; Terentjev, E. M. Soft Matter 2008, 4, 544–549. Guo, L.; Colby, R. H.; Lusignan, C. P.; Whitesides, T. H. Macromolecules 2003, 36, 9999–10008. Gomez-Estaca, J.; Montero, P.; Fernandez-Martin, F.; Gomez-Guillen, M. C. J. Food Eng. 2009, 90, 480–486. Benguigui, L.; Busnel, J.-P.; Durand, D. Polymer 1991, 32, 2680– 2685. Dai, C.-A.; Liu, M.-W. Mater. Sci. Eng. 2006, 423, 121–127.
Biomacromolecules, Vol. 12, No. 1, 2011
81
(25) Chiou, B.-S.; Avena-Bustillos, R. J.; Bechtel, P. J.; Jafri, H.; Narayan, R.; Imam, S. H.; Glenn, G. M.; Orts, W. J. Eur. Polym. J. 2008, 44, 3748–3753. (26) Choy, Y. B.; Cheng, F.; Choi, H.; Kim, K. Macromol. Biosci. 2008, 8, 758–765. (27) Tronci, G.; Neffe, A. T.; Pierce, B. F.; Lendlein, A. J. Mater. Chem. 2010, 20, 8875–8884. (28) Wu, J.; Chiu, S.-C.; Pearce, E. M.; Kwei, T. K. J. Polym. Sci., Part A: Polym. Chem. 2001, 39, 224–231. (29) Neffe, A. T.; Zaupa, A.; Pierce, B. F.; Hofmann, D.; Lendlein, A. Macromol. Rapid Commun. 2010, 31, 1534–1539. (30) Zaupa, A.; Neffe, A. T.; Pierce, B. F.; Lendlein, A.; Hofmann, D. Int. J. Artif. Organs 2010, E-pub ahead of print, DOI: 10.5301/ IJAO.2010.6083. (31) Neffe, A. T.; Lo¨bus, A.; Zaupa, A.; Sto¨tzel, C.; Mu¨ller, F. A.; Lendlein, A. Acta Biomater. 2010, DOI 10.1016/j.actbio.2010.11.025, published online 23.11.2010. (32) Arvanitoyannis, I.; Nakayama, A.; Aiba, S.-i. Carbohyd. Polym. 1998, 36, 105–119. (33) Arvanitoyannis, I. Formation and properties of collagen and gelatin films and coatings; CRC: New York, pp 275-304. (34) Lukasik, K. V.; Ludescher, R. D. Food Hydrocolloid 2006, 20, 96– 105. (35) Bubnis, W. A.; III, C. M. O. Anal. Biochem. 1992, 207, 120–133. (36) Lendlein, A.; Schmidt, A. M.; Langer, R. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 842–847. (37) Pinhas, M. F.; Blanshard, J. M. V.; Derbyshire, W.; Mitchell, J. R. Therm. Anal. Calorim. 1996, 47, 1499–1511. (38) Pezron, I.; Djabourov, M.; Bosio, L.; Leblond, J. J. Polym Sci., Part B: Polym. Phys. 1990, 28, 1823–1839. (39) Itoh, M.; Okawa, Y.; Kobayashi, H.; Ohno, T.; Okamoto, Y.; Katoh, T. J. Photogr. Sci. 1994, 42, 14–17. (40) Djabourov, M.; Leblond, J.; Papon, P. J. Phys. (Paris) 1988, 49, 319– 332. (41) Courty, S.; Gornall, J. L.; Terentjev, E. M. Biophys. J. 2006, 90, 1019– 1027. (42) Rivero, S.; Garcia, M. A.; Pinotti, A. InnoVatiVe Food Sci. Emerging Technol. 2010, 11, 369–375. (43) Sarti, B.; Scandola, M. Biomaterials 1995, 16, 785–792. (44) Loren, N.; Hermansson, A.-M. Int. J. Biol. Macromol. 2000, 27, 249–262.
BM101029K
