- Email: [email protected]
Swelling behavior of a genetically engineered silk-elastinlike protein polymer hydrogel
Biomaterials 23 (2002) 4203–4210
Swelling behavior of a genetically engineered silk-elastinlike protein polymer hydrogel Adam A. Dinermana, Joseph Cappellob, Hamidreza Ghandeharia,*, Stephen W. Hoaga,* a
Department of Pharmaceutical Sciences, University of Maryland School of Pharmacy, 20 N. Pine Street, Baltimore, MD 21201, USA b Protein Polymer Technologies, Inc, 10655 Sorrento Valley Road, San Diego, CA 92121, USA Received 28 November 2001; accepted 6 May 2002
Abstract The influence of environmental conditions such as pH, temperature, and ionic strength on the equilibrium swelling ratio of physically crosslinked networks of a genetically engineered silk-elastinlike protein-based copolymer (SELP) with an amino acid repeat sequence of [(GVGVP)4GKGVP(GVGVP)3(GAGAGS)4]12 was investigated. The effects of gelation cure time and initial polymer concentration on the equilibrium swelling ratio and soluble fraction of the hydrogels were also studied. It was found that the soluble fraction linearly correlated with the initial polymer concentration at higher gelation times. Soluble fraction results suggest that final hydrogel water content may be controlled by both initial polymer concentration and gelation time. Equilibrium swelling studies demonstrated that these hydrogels are relatively insensitive to environmental changes such as pH, temperature, and ionic strength. Over the concentration range studied, it was found that an increase in gelation time at 371C resulted in lower hydrogel weight equilibrium swelling ratios, which corresponds to less soluble polymer released post-gelation. Together, these results have implications for the controlled delivery of bioactive agents from silk-elastinlike hydrogels. r 2002 Elsevier Science Ltd. All rights reserved. Keywords: Genetically engineered polymers; Drug delivery; Silk-elastinlike; Hydrogels
1. Introduction In recent years there has been considerable interest in developing genetically engineered polymers for use as biomaterials that self-assemble into semi-solid matrices under physiological conditions [1–3]. Potential applications for genetically engineered protein polymers range from tissue engineering and drug delivery to surgical applications such as soft tissue augmentation and bone repair. A major advantage of genetically engineered polymers over their conventional synthetic counterparts is the ability to produce biopolymers with uniform composition, molecular weight, and amino acid sequence [4]. Since polymer structure dictates physicochemical properties, genetically engineered polymers offer the unique ability to control the microstructure, function, and fate of biomaterials. In addition, since most self-assembling protein polymers are water soluble *Corresponding coauthors. Tel.: +1-410-706-8650; fax: +1-410706-0346. E-mail address: [email protected] (H. Ghandehari).
at room temperature, aqueous-based vehicles may be used to produce resorbable, in situ gel forming implants for minimally invasive biomedical applications such as drug delivery. Silk-elastinlike protein-based polymers (SELP) are genetically engineered biopolymers composed of repeated sequences of amino acid blocks derived from silk (Gly Ala–Gly Ala–Gly Ser) and elastin (Gly Val–Gly Val–Pro) [5]. SELP copolymers irreversibly self-assemble under physiological conditions depending on the number of silklike blocks in the repeat unit allowing them to be potentially administered via minimally invasive procedures. Previous research has demonstrated that SELP copolymers self-assemble to form gels through an irreversible exothermic event, which is consistent with the crystallization of the silklike domains via hydrogen bond-mediated physical crosslinks [1]. The irreversible sol-to-gel assembly of SELP copolymers under physiological conditions has been exploited for the controlled delivery of bioactive agents [1]; however, much remains to be investigated regarding the physiochemical characterization of silk-elastinlike hydrogels as
0142-9612/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 2 - 9 6 1 2 ( 0 2 ) 0 0 1 6 4 - 3
4204
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
they relate to controlled drug delivery. For example, the effect of gelation time and environmental conditions on equilibrium swelling and soluble fraction in physically crosslinked protein polymer systems ‘‘post-gelation’’ would likely impact the use of these systems in vivo and has not been evaluated. This study examines the network swelling behavior of a physically crosslinked hydrogel composed of a SELP copolymer with an amino acid repeat sequence of a [(GVGVP)4GKGVP(GVGVP)3(GAGAGS)4]12. In addition, the effect of gelation time, polymer concentration, and post-gelation environmental conditions on the weight equilibrium swelling ratio of the hydrogels are examined and the extent of polymer soluble fraction in the network post-gelation are investigated.
2. Materials and methods 2.1. Materials Frozen Polymer 47K solutions (Polymer 47 K UBA, 12 wt%) and lyophilized Polymer 47K were generously provided by Protein Polymer Technologies, Inc. (San Diego, CA). Dulbecco’s 1X phosphate buffered saline (PBS) purchased from Life Technologies, Inc. (Rockville, MD) and filtered Milli-Q deionized water were used throughout the experiments. Chlorotrimethylsilane and dichloromethane were purchased from Aldrich (Milwaukee, WI) and used to prepare the silanating solution. For protein detection, micro-Lowry reagents were obtained from Sigma Chemical Co. (St. Louis, MO). All other chemicals were reagent grade and purchased from Sigma Chemical Co. (St. Louis, MO). 2.2. Composition of the SELP copolymer Polymer 47 K (Protein Polymer Technologies, Inc., San Diego, CA) is a block co-polymer composed of both silk and elastinlike amino acid sequence motifs, with a total of 884 amino acids and a molecular weight of 69,814. The complete predicted amino acid sequence of Polymer 47K is given in Table 1.
2.3. Preparation of hydrogels Frozen 12 wt% Polymer 47K stock solutions were thawed by immersing polymer-containing syringes into a room-temperature water bath for 5 min. These solutions were gently mixed by inverting the syringes manually. Lower Polymer 47K concentrations (4, 8, and 10 wt%) were obtained by diluting the thawed 12 wt% solution with a chilled PBS solution (1:1.41 dilution of Dulbecco’s 1X PBS in Milli-Q water). All protein polymer solutions were slowly injected into gel molds consisting of a silicone gasket (ca. 1.6 mm thickness) sandwiched between two silanated glass slides (pretreated with 5 v/v% chlorotrimethylsilane in dichloromethane) secured by binder clips. Filled gel molds were placed into a temperature controlled convection incubator (Precision, Winchester, VA) at 371C for either 1 or 24 h. Following incubation, gel disks (ca. thickness=1.6 mm, ca. diameter=6.0 mm) were obtained using a cork borer. To remove soluble polymer fractions remaining in the hydrogels post-gelation, disks were extensively washed for 1 week in 1X PBS (10 mm PBS, pH 7.4, m ¼ 0:15) with 0.01 w/v% sodium azide under mild agitation (Speed=120 RPM) in a temperature controlled convection incubator (VWR, Model 1575, Bridgeport, NJ) set at 371C. Gel washing was performed at 371C to simulate physiological conditions and to ensure that the gelation process was complete prior to storage and use. Fresh buffer was replaced daily throughout the washing period. Prior to use, washed hydrogel disks were stored at 2–81C in 1X PBS with 0.01 w/v% sodium azide. 2.4. Determination of polymer-soluble fraction The Polymer 47K soluble fraction was determined by placing a known weight of polymer solution into polypropylene vials and incubating the sealed vial along with each batch of filled gel molds. Post-gelation, 1 ml of a 1X PBS with 0.01 w/v% sodium azide was placed in the polypropylene vial and incubated at 371C along with the hydrogel disks for 1 week. After the washing period, polypropylene vials were sampled and diluted in 1X PBS with 0.01 w/v% sodium azide. Diluted samples were
Table 1 The amino acid sequence and molecular weight of the SELP copolymer under study Amino acid sequencea
Total number of amino acids
Molecular weight (g/mol)
MDVVLQRRDWENPGVTQLNRLAAHPP FASDPMGAGSGAGAGS[(GVGVP)4 GKGVP(GVGVP)3(GAGAGS)4]12(GVGVP)4 GKGVP(GVGVP)3(GAGAGS)2 GAGAMDPGRYQDLRSHHHHHH
884
69,814
a Single amino acid abbreviations used: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; L, Leu; M, Met; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W,Trp; and Y,Tyr.
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
4205
analyzed using the micro-Lowry protein determination method [6] at a wavelength of 595 nm. A standard curve was prepared by serial dilution of a 1 mg/ml stock solution of lyophilized Polymer 47K (Protein Polymer Technologies, Inc., batch 980402) in 1X PBS with 0.01 w/v% sodium azide. SDS-PAGE analysis was used to confirm that no change in polymer molecular weight occurred after 1 week incubation of a 1 mg/ml solution at 371C. The polymer soluble fraction was expressed as milligrams of soluble polymer post-gelation as determined by Lowry assay per milligrams of initial polymer (label claim).
daily basis. The initial and post-24 h pH values were monitored. The effect of ionic strength on q was determined by placing swollen hydrogel disks into closed vials containing 10 mm phosphate buffer solution (pH 7.4) with 0.01 w/v% sodium azide incubated at 371C. The ionic strength of each solution was adjusted with NaCl and q was monitored over the range 0.01–1.0 m. For comparison, hydrogel disks were also equilibrated in deionized water (mB0).
2.5. Equilibrium swelling studies
Single factor analysis of variance (ANOVA) at an a ¼ 0:05 was used to determine statistically significant differences between test samples. All studies were performed in triplicate or quadruplicate unless otherwise noted.
The weight equilibrium swelling ratio (q) was experimentally determined using Eq. (1), where Ws is the swollen hydrogel weight at specified environmental conditions and Wd is the dry hydrogel weight: q¼
Ws : Wd
ð1Þ
The equilibrium water content or equilibrium hydration (H) of the hydrogel was determined using Eq. (2): 1 W s Wd H ¼ 1 ¼ : ð2Þ q Ws To measure Ws ; swollen hydrogels were removed from test conditions then gently blotted with a lint-free wipe prior to weighing. At study completion, swollen hydrogels were extensively washed with deionized water to remove buffer salts. Dry hydrogels were obtained by placing swollen hydrogels in a dessicator containing Drierites for not less than 5 days. Gel disks were then weighed and placed into a vacuum oven at 301C for 24 h and then re-weighed. Virtually no change in dry hydrogel weight was found following vacuum drying. The effect of temperature on q was determined by placing hydrogel disks that had been cured at 371C for specific times in closed vials containing 1X PBS with 0.01 w/v% sodium azide. The vials were placed in a temperature controlled water bath (Neslab RTE-100, Newington, NH) and equilibrated for at least 24 h at each temperature prior to weight determination. The temperature of the water bath was controlled over the range 6–501C. The effect of pH on q was determined by placing swollen hydrogel disks into closed vials containing buffer solutions (pH 2–pH 9) incubated at 371C. Buffer solutions were prepared using pHydrion (Micro-Essential Laboratory, Brooklyn, NY) capsules and the ionic strength of each solution was adjusted to 0.15 m with NaCl. Hydrogel disks were equilibrated for at least 24 h in each buffer solution prior to weight determination. Each vial was replaced with fresh buffer solution on a
2.6. Statistical analysis
3. Results 3.1. Polymer soluble fraction analysis Incubation of aqueous polymer solutions at conditions shown in Table 2 produced gels that were translucent. The hydrogel specimens were easily cut from the gel molds without fracture; however, specimens were not very pliable. As the incubation period increased, the hydrogels became progressively less
Table 2 Initial polymer concentrations and incubation conditions for the hydrogels Hydrogel codea
Gelation time @ 371C (h)
Initial polymer content (wt%)
1.1.2 1.1.4 2.1.4 3.1.2 3.1.3 4.1.4 5.1.4 5.1.2 4.2.1 6.2.1 6.2.2 6.2.4 7.2.1 7.2.2 8.2.4 9.2.4 10.2.4
1 1 1 1 1 1 1 1 24 24 24 24 24 24 24 24 24
8.0 12.0 12.0 8.1 10.0 12.0 12.0 8.0 4.0 4.0 8.0 12.0 4.0 8.0 12.0 12.0 12.0
a Hydrogel code designation x; y; z: where x corresponds to the syringe from which hydrogels were prepared, y refers to the gelation time (1=1 h, 2=24 h), and z refers to the initial polymer content (1=4 wt%, 2=8 wt%, 3=10 wt%, and 4=12 wt%).
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
4206
Table 3 Soluble fraction of 12 wt% hydrogels as a function of gelation time Hydrogel codea
Gelation time @ 371C (h)
Soluble fraction (wt%)
Mean7SD (wt%)
1.1.4 2.1.4 4.1.4 5.1.4
1
27.6 29.7 16.8 31.9
26.576.7
6.2.4 8.2.4 9.2.4 10.2.4
24
7.7 10.0 7.6 6.6
8.071.4
a
See Table 2 for an explanation of hydrogel code designations.
pliable indicating a further increase in mechanical stiffness of the network. The fractions of polymer chains that did not participate in physical crosslinking (soluble fraction) are shown in Table 3 as a function of gelation time for 12 wt% hydrogels. The soluble fraction represents the percent of polymer chains in the initial aqueous solution that was not involved in network formation. Overall, the soluble fraction for the 12 wt% hydrogels after 1 h of incubation at 371C was greater than those after 24 h of incubation. The average soluble fraction remaining in a 12 wt% hydrogel after 24 h of incubation was 8.071.4 wt% of protein compared to 26.576.7 wt% of protein after a 1 h incubation. To examine the influence of initial polymer concentration on the soluble fraction present in the hydrogels, the final polymer content of the hydrogels was plotted as a function of the initial polymer content for hydrogels cured between 1 and 24 h (Fig. 1a and b). In Fig. 1a, each data point represents the results for a separate hydrogel specimen prepared for each designated initial polymer content. For a given polymer concentration, less inter-specimen variation was observed with hydrogels incubated for 24 h compared to hydrogels incubated for 1 h. Similar to the results observed for the 12 wt% hydrogels (Table 3), the results suggest that regardless of initial polymer content more polymer chain entanglement and physical interactions between polymer chains occurs at longer incubation times. A linear correlation was observed between the final polymer content and the initial polymer content (R2 ¼ 0:994) for the 24 h incubation period (Fig. 1a). Fig. 1b illustrates the effect of the initial polymer concentration on the soluble fraction post-gelation as a function of incubation time. Data for hydrogels prepared at both high and low concentrations for each respective incubation time are represented. At both 1 and 24 h incubation times, there was no significant difference in soluble fraction between hydrogels prepared at high and low polymer concentrations. These results suggest that over the concentrations studied the
Fig. 1. Influence of soluble fraction on final Polymer 47K content in hydrogel: (a) effect of gelation/cure time. (K) gelation time=24 h, (J) gelation time=1 h, (—) best-fit line. Each symbol represents a single determination for each hydrogel specimen; (b) effect of Polymer 47K concentration, (&) 4 wt%, ( ) 8 wt%, ( ) 12 wt%. Bars represent mean7one standard deviation (n ¼ 3).
initial polymer concentration does not have a significant influence on the soluble fraction present in the hydrogels post-gelation. Based on these results, it appears that the extent of physical crosslinking primarily depends on the gelation time for a given polymer concentration. 3.2. Influence of temperature, pH, ionic strength, and polymer concentration on the hydrogel equilibrium swelling ratio The effect of temperature on the weight equilibrium swelling ratio for 12 wt% hydrogels cured for 1 h is reported in Fig. 2a. The equilibrium swelling ratio for 12 wt% hydrogels is not significantly different over the temperature range investigated. Fig. 2b illustrates the effect of pH on the weight equilibrium swelling ratio for 12 wt% hydrogels cured for 1 h. Results demonstrate that these hydrogels do not display any pH sensitivity in the range of pH between 2 and 9 (p ¼ 0:171).
20
20
18
18
16
16
14
14
12
12
q
q
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
10
10
8
8
6
6
4
4 0
10
20
30
40
50
60
Temperature (˚C)
(a)
0
1 2
3 4
(b)
5
6 7
8 9 10
pH
20
32
18
28
16
24
14
20 q
q
4207
12
16
10 12
8
8
6
4
4 0 (c)
0.01
0.15
0
1
Ionic Strength (M)
(d)
10
20
30
40
50
60
Temperature (˚C)
Fig. 2. Influence of environmental conditions on the weight equilibrium swelling ratio (q) for hydrogels cured for 1 h at 371C: (a) effect of temperature on q for 12 wt% hydrogels in PBS (pH 7.4, m ¼ 0:15); (b) effect of pH on q for 12 wt% hydrogels at 371C; (c) effect of ionic strength on q for 12 wt% hydrogels at 371C in phosphate buffer solution (pH 7.4); and (d) effect of polymer concentration on q for hydrogels at 371C in PBS (pH 7.4, m ¼ 0:15), (K) 8 wt%, (&) 10 wt%, (m) 12 wt%. Symbols represent mean value7one standard deviation (n ¼ 3). Some error bars are smaller than symbols.
To investigate the effect of ionic strength on Polymer 47K hydrogel swelling, the ionic strength of a phosphate buffer solution (pH 7.4) was adjusted from 0.01 to 1 m with NaCl. Fig. 2c shows that the equilibrium swelling ratio for 12 wt% hydrogel cured for 1 h at 371C does not change significantly despite a two log increase in ionic strength. Fig. 2d displays the effect of temperature on the equilibrium swelling ratio for hydrogels cured for 1 h at 371C as a function of polymer concentration. Similar swelling profiles were observed for 10 and 12 wt% hydrogels demonstrating a lack of temperature sensitivity over the temperature range investigated. When the aqueous polymer solution was further diluted to 8 wt%, a statistically significant increase in the equilibrium swelling ratio was found between 12 and 8 wt% hydrogels. For the 8 wt% gel, a slight decrease in q was observed at temperatures >371C. However, these differences were not statistically significant at an a ¼ 0:05: Further dilution of Polymer 47K did not yield physically robust hydrogel disks after 1 h of incubation; therefore, concentrations lower than 8 wt% were not evaluated.
3.3. Effect of gelation time on hydrogel equilibrium swelling ratio Since a 1 h incubation time did not yield physically robust hydrogel disks at polymer concentrations below 8 wt%, the incubation period was extended to 24 h to investigate the swelling behavior at lower polymer concentrations. Fig. 3 illustrates the effect of temperature on the equilibrium swelling ratio for Polymer 47K solutions incubated for 24 h at 371C as a function of polymer concentration. After the 24 h incubation, temperature sensitivity was not observed between 8 and 12 wt% hydrogels. The 4 wt% hydrogel displayed a decreasing trend in the equilibrium swelling ratio with an increase in temperature though the values were not statistically significant at an a ¼ 0:05: A similar, but less pronounced, trend in q was observed for 8 wt% polymer solutions incubated for 1 h (see Fig. 2d). It is possible that gravimetric analysis is not sensitive enough to distinguish a transition of the elastinlike domains in Polymer 47K hydrogels. For this reason, we also examined the temperature sensitivity of hydrogels prepared at 4, 8, and 12 wt% concentrations using
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
4208
32
1 Hydration (g water/g gel)
28 24
q
20 16 12 8
0.95 0.9 0.85 0.8
4 0
10
20
30
40
50
0
60
Temperature (˚C) Fig. 3. Effect of temperature on q for SELP hydrogels cured for 24 h. All hydrogels were stored in 1X PBS. (’) 4 wt%, (K) 8 wt%, (m) 12 wt%. Symbols represent mean value7one standard deviation (n ¼ 4). Some error bars are smaller than symbols.
30
q
20
10
0 1
2
4
6
8
10
12
14
Final Polymer Concentration (wt%)
24 Gelation Time (hours)
Fig. 4. Effect of gelation (cure) time on q for SELP hydrogels. All hydrogels were stored in 1X PBS at 201C. ( ) 8 wt%, (&) 12 wt%. Bars represent mean value7one standard deviation (n ¼ 3).
differential scanning calorimetry (DSC). DSC results for hydrogels cured for 24 h (not shown) did not display any thermal transitions in the temperature range 6–701C using a scanning rate of 21C/min. As a consequence any conformational changes resulting from an increase in temperature cannot be detected using our experimental methods. Fig. 4 displays the effect of the gelation time at 371C on the hydrogel equilibrium swelling ratio in 1X PBS (pH 7.4) at 201C. For a given polymer concentration, the equilibrium swelling ratio decreased with an increased incubation period indicating a higher crosslinking density. Swelling ratios between 8 and 12 wt%
Fig. 5. Comparison of degree of hydration and final Polymer 47K content in hydrogels as a function of gelation (cure) time. (K) gelation time=24 h, (J) gelation time=1 h, (—) H (calc), (- - -) H (calc,72%). Each symbol represents the mean value for a different batch of hydrogels (n ¼ 3).
hydrogels after 1 or 24 h incubation were significantly different (p{0:05). It is worthwhile to note that the change in swelling ratio with incubation time was more pronounced for the 8 wt% hydrogels compared with the 12 wt% hydrogels, which confirms that lower polymer concentrations need sufficient time for polymer–polymer interactions to occur. Higher variability in q was observed at lower polymer concentrations for each incubation period. This is most likely due to a fewer number of polymer chains that are able to interact with one another at lower polymer concentrations. A comparison of the final polymer concentration (adjusted for soluble fraction) and the equilibrium hydration of the hydrogels is shown in Fig. 5. The solid line in Fig. 5 represents the hydrogel hydration (H) calculated from the final polymer content. The calculated hydration is the expected amount of water in the hydrogels if the polymer content remains the same after storage and testing in the swelling studies. The experimental data fall within 72% of the line, though H and the soluble fraction were determined independently. Based on these results we can conclude that no significant hydrogel degradation or dissolution took place during storage or during the swelling studies, and any observed changes in the equilibrium swelling ratio at higher temperatures should only be related to a network de-swelling phenomena.
4. Discussion The extent of soluble fraction present in a gel-formed implant as well as its equilibrium swelling ratio may have a profound effect on the ability of the hydrogel to reproducibly and effectively control the release of
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
bioactive agents. In regards to the polymer soluble fraction, the presence of uncrosslinked polymer may result in complex formation with certain bioactive agents, such as oligonulceotides, genes, and therapeutic proteins, resulting in either precipitation of the complex in the gel or in the case of soluble complexes decreased release rates from implant devices followed by attenuated bioactivity. In addition, localized high concentrations of uncrosslinked polymer may diffuse from a gelformed implant, creating water filled channels or pores that may sufficiently increase the equilibrium swelling ratio and impact solute release behavior. The analysis of soluble fraction for Polymer 47K hydrogels demonstrate an increase in gelation time from 1 to 24 h resulting in increased polymer–polymer interactions (Fig. 1a). As a consequence, the number of polymer chains involved in the hydrogel network is increased, which leads to a decreased soluble fraction. It was found that regardless of the initial polymer content, hydrogels produced after a 24 h incubation period had a lower degree of swelling than those produced after a 1 h incubation period (Fig. 1b), suggesting more polymer chain entanglement and physical interactions between polymer chains at a longer incubation time. A linear correlation between the final polymer content and the initial polymer content for the 24 h incubation period (Fig. 1a) suggests that the soluble fraction can be controlled by adjusting the polymer gelation time. Similarly, it was found that the equilibrium swelling ratio may be controlled, within certain constraints, by adjusting the gelation time of polymer solutions (Fig. 4). Overall, these results suggest that the extent of physical crosslinking primarily depends on the gelation time for a given concentration of the SELP copolymer under study. Environmental sensitivity of crosslinked polymer networks may cause changes in the equilibrium swelling ratio that can have an effect on their ability to be used as controlled release devices for a variety of biomedical applications. Our results indicate that physically crosslinked networks of the SELP copolymer under study are insensitive to temperature over a range 6–501C. Watersoluble SELPs containing elastin repeat units exhibit an inverse temperature solubility transition [7]. In addition, previous research with covalently crosslinked elastinmimetic hydrogels has shown a decrease in the equilibrium weight swelling ratio with an increase in temperature [8–10]. Our results indicate that a reduction in polymer concentration does not significantly amplify SELP hydrogel response to temperature even though there is a significant decrease in crosslinking density (indicated by an increase in the swelling ratio, q). The temperature sensitivity effect is absent possibly due to the formation of rigid, physical crosslinks as a result of the increased number of silk units in the repeat structure (Table 1).
4209
Polymer 47K hydrogels did not display pH sensitivity (no change in q over a pH range 2–9) despite the fact that the polymer carries a net positive charge at pH 7.4 due to the presence of lysine groups (pKa 10.5) in the repeat units of the polymer backbone. Past research has demonstrated hydrophobicity induced pKa shifts in elastin protein-based polymers [3] suggesting the actual pKa of lysine in the Polymer 47K backbone may not be 10.5. No significant change in the equilibrium weight swelling ratio was observed over the range of pH studied most likely due to the nature and the extent of physical crosslinking between the silk units. Hydrogel swelling–deswelling behavior may also be influenced by the ionic strength because of factors such as the presence of ions in the network, counterions shielding ionizible groups in the polymer backbone, polymer–polymer repulsion, and the balance between the polymer–solvent interaction and the nature of chain packing in the hydrogel. For protein-based polymers, hydrophobic self-assembly is due to the competition between apolar (hydrophobic) and polar residues that are restrained by sequence in the polymer backbone such that their proximity allows them to compete for the same water molecules for hydration [11]. Thus, for water-soluble polymers, the effect of ionic strength typically manifests itself in driving hydrophobic selfassembly by counterion shielding of polar moieties in the polymer backbone, which leads to decreased competition for water molecules of hydration with its more apolar neighbors. The SELP hydrogels in this study did not display sensitivity to ionic strength over the range 0.01–1 m at pH 7.4, further highlighting how rigid crosslinks formed between the silk units overcome the potential interactions of the polymer chain with ions in the solution. Based on previous research with related genetically engineered polymers [12], we believe that Polymer 47K hydrogel crosslinks consist of localized assemblies of silklike blocks forming hydrogen-bonded b-sheets. The experimental results suggest a physically constrained, rigid network that may prevent elastinlike domains from reversibly folding in SELP hydrogels. This is supported by the fact that the presence of NaCl did not significantly change the equilibrium swelling ratio of SELP hydrogels. It is well known that with the addition of salt to aqueous protein solutions, ions compete for water of hydrophobic hydration as well as screen charges on the protein; thus, making them more prone to hydrophobic aggregation. Based on this work, it is possible that the lack of temperature sensitivity displayed by Polymer 47K hydrogels may be primarily due to crystallites prohibiting reversible transitions of elastinlike blocks as a result of the nature of the crosslinks (chain packing). Furthermore, a significant increase in the equilibrium swelling ratio was not found at temperatures o251C, where elastinlike domains
4210
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
would most likely unfold if they were in their folded state in the hydrogel. This clearly suggests that the crystallites may restrain the folding/unfolding process. This is further supported by the fact that hydrogels prepared at lower polymer concentrations displayed minimal temperature effects despite a significant decrease in crosslinking density as demonstrated by high equilibrium swelling ratios. Even though a decreasing trend in the equilibrium swelling ratio was more evident at higher temperatures for hydrogels prepared at lower polymer concentrations these changes were not statistically significant.
dates as in situ gel forming implants for localized drug delivery where stimuli-sensitivity is not required.
Acknowledgements This work was partially funded by an American Foundation for Pharmaceutical Education (AFPE) Fellowship for AAD.
References 5. Conclusions Silk-elastinlike hydrogels made from a genetically engineered copolymer with an amino acid repeat sequence of [(GVGVP)4GKGVP(GVGVP)3(GAGAGS)4]12, were evaluated for their soluble fraction content as a function of gelation time and initial polymer concentration. It was found that the hydrogels contained a fraction of uncrosslinked polymer (soluble fraction) post-gelation. The extent of soluble fraction remaining in the gel was primarily dependent on the gelation time and varied from 14 to 35 wt% after 1 h incubation and from 8 to 17 wt% after 24 h incubation. Larger variations in the soluble fraction were observed for gels produced after 1 h incubation. These hydrogels were evaluated for sensitivity to environmental stimuli such as pH, temperature, and ionic strength. Unlike other crosslinked elastin-mimetic hydrogels that undergo temperature-dependent swelling and deswelling, the SELP hydrogels under study did not display significant environmental sensitivity post-gelation. A decrease in the polymer concentration resulting in higher equilibrium swelling ratios (decreasing the apparent crosslinking density) did not significantly influence the environmental sensitivity of the hydrogels. These results suggest that irreversible crystallization of silklike blocks at physiological conditions prevents elastinlike blocks from reversibly folding in the crosslinked state. The lack of temperature sensitivity in the hydrogels is probably due to the nature of polymer chain packing in the hydrogen-bonded crystallites formed from the interaction of the silk units. These results demonstrate that the SELP hydrogels with the structure above will not undergo significant changes in swelling post-gelation at physiological conditions making them promising candi-
[1] Cappello J, Crissman JW, Crissman M, Ferrari FA, Textor G, Wallis O, Whitledge JR, Zhou X, Burman D, Aukerman L, Stedronsky ER. In-situ self-assembling protein polymer gel systems for administration, delivery, and release of drugs. J Controlled Release 1998;53:105–17. [2] Petka WA, Harden JL, McGrath KP, Wirtz D, Tirrell DA. Reversible hydrogels from self-assembling artificial proteins. Science 1998;281:389–92. [3] Urry DW, Harris CM, Luan CX, Luan C-H, Gowda DC, Parker TM, Peng SQ, Xu J. Transductional protein-based polymers as new controlled-release vehicles. In: Park K, editor. Controlled drug delivery-strategies and challenges. Washington, DC: American Chemical Society, 1997. p. 405–36. [4] Nagarsekar A, Ghandehari H. Genetically engineered polymers for drug delivery. J Drug Targeting 1999;7:11–32. [5] Cappello J. Synthetically designed protein polymer biomaterials. In: Park K, editor. Controlled drug delivery: Strategies and challenges. Washington, DC: American Chemical Society, 1997. p. 439–53. [6] Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the folin-phenol reagents. J Biol Chem 1951;193:265–75. [7] Nagarsekar A, Crissman M, Crissman J, Ferrari F, Cappello J, Ghandehari H. Genetic synthesis and characterization of pH and temperature-sensitive silk-elastinlike block copolymers. J Biomed Mater Res, in press. [8] Lee J, Macosko CW, Urry DW. Elastomeric polypentapeptides cross-linked into matrixes and fibers. Biomacromolecules 2001;2:170–9. [9] Lee J, Macosko CW, Urry DW. Swelling behavior of g-irradiated cross-linked elastomeric polypentapeptide-based hydrogels. Macromolecules 2001;34:4114–23. [10] McMillan RA, Conticello VP. Synthesis and characterization of elastin-mimetic protein gels derived from a well-defined polypeptide precursor. Macromolecules 2000;33:4809–21. [11] Urry DW. Physical chemistry of biological free energy transduction as demonstrated by elastic protein-based polymers. J Phys Chem B 1997;101:11007–28. [12] Cappello J, Crissman J, Dorman M, Mikolajczak M, Textor G, Marquet M, Ferrari F. Genetic engineering of structural protein polymers. Biotechnol Prog 1990;6:198–202.
Swelling behavior of a genetically engineered silk-elastinlike protein polymer hydrogel Adam A. Dinermana, Joseph Cappellob, Hamidreza Ghandeharia,*, Stephen W. Hoaga,* a
Department of Pharmaceutical Sciences, University of Maryland School of Pharmacy, 20 N. Pine Street, Baltimore, MD 21201, USA b Protein Polymer Technologies, Inc, 10655 Sorrento Valley Road, San Diego, CA 92121, USA Received 28 November 2001; accepted 6 May 2002
Abstract The influence of environmental conditions such as pH, temperature, and ionic strength on the equilibrium swelling ratio of physically crosslinked networks of a genetically engineered silk-elastinlike protein-based copolymer (SELP) with an amino acid repeat sequence of [(GVGVP)4GKGVP(GVGVP)3(GAGAGS)4]12 was investigated. The effects of gelation cure time and initial polymer concentration on the equilibrium swelling ratio and soluble fraction of the hydrogels were also studied. It was found that the soluble fraction linearly correlated with the initial polymer concentration at higher gelation times. Soluble fraction results suggest that final hydrogel water content may be controlled by both initial polymer concentration and gelation time. Equilibrium swelling studies demonstrated that these hydrogels are relatively insensitive to environmental changes such as pH, temperature, and ionic strength. Over the concentration range studied, it was found that an increase in gelation time at 371C resulted in lower hydrogel weight equilibrium swelling ratios, which corresponds to less soluble polymer released post-gelation. Together, these results have implications for the controlled delivery of bioactive agents from silk-elastinlike hydrogels. r 2002 Elsevier Science Ltd. All rights reserved. Keywords: Genetically engineered polymers; Drug delivery; Silk-elastinlike; Hydrogels
1. Introduction In recent years there has been considerable interest in developing genetically engineered polymers for use as biomaterials that self-assemble into semi-solid matrices under physiological conditions [1–3]. Potential applications for genetically engineered protein polymers range from tissue engineering and drug delivery to surgical applications such as soft tissue augmentation and bone repair. A major advantage of genetically engineered polymers over their conventional synthetic counterparts is the ability to produce biopolymers with uniform composition, molecular weight, and amino acid sequence [4]. Since polymer structure dictates physicochemical properties, genetically engineered polymers offer the unique ability to control the microstructure, function, and fate of biomaterials. In addition, since most self-assembling protein polymers are water soluble *Corresponding coauthors. Tel.: +1-410-706-8650; fax: +1-410706-0346. E-mail address: [email protected] (H. Ghandehari).
at room temperature, aqueous-based vehicles may be used to produce resorbable, in situ gel forming implants for minimally invasive biomedical applications such as drug delivery. Silk-elastinlike protein-based polymers (SELP) are genetically engineered biopolymers composed of repeated sequences of amino acid blocks derived from silk (Gly Ala–Gly Ala–Gly Ser) and elastin (Gly Val–Gly Val–Pro) [5]. SELP copolymers irreversibly self-assemble under physiological conditions depending on the number of silklike blocks in the repeat unit allowing them to be potentially administered via minimally invasive procedures. Previous research has demonstrated that SELP copolymers self-assemble to form gels through an irreversible exothermic event, which is consistent with the crystallization of the silklike domains via hydrogen bond-mediated physical crosslinks [1]. The irreversible sol-to-gel assembly of SELP copolymers under physiological conditions has been exploited for the controlled delivery of bioactive agents [1]; however, much remains to be investigated regarding the physiochemical characterization of silk-elastinlike hydrogels as
0142-9612/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 2 - 9 6 1 2 ( 0 2 ) 0 0 1 6 4 - 3
4204
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
they relate to controlled drug delivery. For example, the effect of gelation time and environmental conditions on equilibrium swelling and soluble fraction in physically crosslinked protein polymer systems ‘‘post-gelation’’ would likely impact the use of these systems in vivo and has not been evaluated. This study examines the network swelling behavior of a physically crosslinked hydrogel composed of a SELP copolymer with an amino acid repeat sequence of a [(GVGVP)4GKGVP(GVGVP)3(GAGAGS)4]12. In addition, the effect of gelation time, polymer concentration, and post-gelation environmental conditions on the weight equilibrium swelling ratio of the hydrogels are examined and the extent of polymer soluble fraction in the network post-gelation are investigated.
2. Materials and methods 2.1. Materials Frozen Polymer 47K solutions (Polymer 47 K UBA, 12 wt%) and lyophilized Polymer 47K were generously provided by Protein Polymer Technologies, Inc. (San Diego, CA). Dulbecco’s 1X phosphate buffered saline (PBS) purchased from Life Technologies, Inc. (Rockville, MD) and filtered Milli-Q deionized water were used throughout the experiments. Chlorotrimethylsilane and dichloromethane were purchased from Aldrich (Milwaukee, WI) and used to prepare the silanating solution. For protein detection, micro-Lowry reagents were obtained from Sigma Chemical Co. (St. Louis, MO). All other chemicals were reagent grade and purchased from Sigma Chemical Co. (St. Louis, MO). 2.2. Composition of the SELP copolymer Polymer 47 K (Protein Polymer Technologies, Inc., San Diego, CA) is a block co-polymer composed of both silk and elastinlike amino acid sequence motifs, with a total of 884 amino acids and a molecular weight of 69,814. The complete predicted amino acid sequence of Polymer 47K is given in Table 1.
2.3. Preparation of hydrogels Frozen 12 wt% Polymer 47K stock solutions were thawed by immersing polymer-containing syringes into a room-temperature water bath for 5 min. These solutions were gently mixed by inverting the syringes manually. Lower Polymer 47K concentrations (4, 8, and 10 wt%) were obtained by diluting the thawed 12 wt% solution with a chilled PBS solution (1:1.41 dilution of Dulbecco’s 1X PBS in Milli-Q water). All protein polymer solutions were slowly injected into gel molds consisting of a silicone gasket (ca. 1.6 mm thickness) sandwiched between two silanated glass slides (pretreated with 5 v/v% chlorotrimethylsilane in dichloromethane) secured by binder clips. Filled gel molds were placed into a temperature controlled convection incubator (Precision, Winchester, VA) at 371C for either 1 or 24 h. Following incubation, gel disks (ca. thickness=1.6 mm, ca. diameter=6.0 mm) were obtained using a cork borer. To remove soluble polymer fractions remaining in the hydrogels post-gelation, disks were extensively washed for 1 week in 1X PBS (10 mm PBS, pH 7.4, m ¼ 0:15) with 0.01 w/v% sodium azide under mild agitation (Speed=120 RPM) in a temperature controlled convection incubator (VWR, Model 1575, Bridgeport, NJ) set at 371C. Gel washing was performed at 371C to simulate physiological conditions and to ensure that the gelation process was complete prior to storage and use. Fresh buffer was replaced daily throughout the washing period. Prior to use, washed hydrogel disks were stored at 2–81C in 1X PBS with 0.01 w/v% sodium azide. 2.4. Determination of polymer-soluble fraction The Polymer 47K soluble fraction was determined by placing a known weight of polymer solution into polypropylene vials and incubating the sealed vial along with each batch of filled gel molds. Post-gelation, 1 ml of a 1X PBS with 0.01 w/v% sodium azide was placed in the polypropylene vial and incubated at 371C along with the hydrogel disks for 1 week. After the washing period, polypropylene vials were sampled and diluted in 1X PBS with 0.01 w/v% sodium azide. Diluted samples were
Table 1 The amino acid sequence and molecular weight of the SELP copolymer under study Amino acid sequencea
Total number of amino acids
Molecular weight (g/mol)
MDVVLQRRDWENPGVTQLNRLAAHPP FASDPMGAGSGAGAGS[(GVGVP)4 GKGVP(GVGVP)3(GAGAGS)4]12(GVGVP)4 GKGVP(GVGVP)3(GAGAGS)2 GAGAMDPGRYQDLRSHHHHHH
884
69,814
a Single amino acid abbreviations used: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; L, Leu; M, Met; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W,Trp; and Y,Tyr.
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
4205
analyzed using the micro-Lowry protein determination method [6] at a wavelength of 595 nm. A standard curve was prepared by serial dilution of a 1 mg/ml stock solution of lyophilized Polymer 47K (Protein Polymer Technologies, Inc., batch 980402) in 1X PBS with 0.01 w/v% sodium azide. SDS-PAGE analysis was used to confirm that no change in polymer molecular weight occurred after 1 week incubation of a 1 mg/ml solution at 371C. The polymer soluble fraction was expressed as milligrams of soluble polymer post-gelation as determined by Lowry assay per milligrams of initial polymer (label claim).
daily basis. The initial and post-24 h pH values were monitored. The effect of ionic strength on q was determined by placing swollen hydrogel disks into closed vials containing 10 mm phosphate buffer solution (pH 7.4) with 0.01 w/v% sodium azide incubated at 371C. The ionic strength of each solution was adjusted with NaCl and q was monitored over the range 0.01–1.0 m. For comparison, hydrogel disks were also equilibrated in deionized water (mB0).
2.5. Equilibrium swelling studies
Single factor analysis of variance (ANOVA) at an a ¼ 0:05 was used to determine statistically significant differences between test samples. All studies were performed in triplicate or quadruplicate unless otherwise noted.
The weight equilibrium swelling ratio (q) was experimentally determined using Eq. (1), where Ws is the swollen hydrogel weight at specified environmental conditions and Wd is the dry hydrogel weight: q¼
Ws : Wd
ð1Þ
The equilibrium water content or equilibrium hydration (H) of the hydrogel was determined using Eq. (2): 1 W s Wd H ¼ 1 ¼ : ð2Þ q Ws To measure Ws ; swollen hydrogels were removed from test conditions then gently blotted with a lint-free wipe prior to weighing. At study completion, swollen hydrogels were extensively washed with deionized water to remove buffer salts. Dry hydrogels were obtained by placing swollen hydrogels in a dessicator containing Drierites for not less than 5 days. Gel disks were then weighed and placed into a vacuum oven at 301C for 24 h and then re-weighed. Virtually no change in dry hydrogel weight was found following vacuum drying. The effect of temperature on q was determined by placing hydrogel disks that had been cured at 371C for specific times in closed vials containing 1X PBS with 0.01 w/v% sodium azide. The vials were placed in a temperature controlled water bath (Neslab RTE-100, Newington, NH) and equilibrated for at least 24 h at each temperature prior to weight determination. The temperature of the water bath was controlled over the range 6–501C. The effect of pH on q was determined by placing swollen hydrogel disks into closed vials containing buffer solutions (pH 2–pH 9) incubated at 371C. Buffer solutions were prepared using pHydrion (Micro-Essential Laboratory, Brooklyn, NY) capsules and the ionic strength of each solution was adjusted to 0.15 m with NaCl. Hydrogel disks were equilibrated for at least 24 h in each buffer solution prior to weight determination. Each vial was replaced with fresh buffer solution on a
2.6. Statistical analysis
3. Results 3.1. Polymer soluble fraction analysis Incubation of aqueous polymer solutions at conditions shown in Table 2 produced gels that were translucent. The hydrogel specimens were easily cut from the gel molds without fracture; however, specimens were not very pliable. As the incubation period increased, the hydrogels became progressively less
Table 2 Initial polymer concentrations and incubation conditions for the hydrogels Hydrogel codea
Gelation time @ 371C (h)
Initial polymer content (wt%)
1.1.2 1.1.4 2.1.4 3.1.2 3.1.3 4.1.4 5.1.4 5.1.2 4.2.1 6.2.1 6.2.2 6.2.4 7.2.1 7.2.2 8.2.4 9.2.4 10.2.4
1 1 1 1 1 1 1 1 24 24 24 24 24 24 24 24 24
8.0 12.0 12.0 8.1 10.0 12.0 12.0 8.0 4.0 4.0 8.0 12.0 4.0 8.0 12.0 12.0 12.0
a Hydrogel code designation x; y; z: where x corresponds to the syringe from which hydrogels were prepared, y refers to the gelation time (1=1 h, 2=24 h), and z refers to the initial polymer content (1=4 wt%, 2=8 wt%, 3=10 wt%, and 4=12 wt%).
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
4206
Table 3 Soluble fraction of 12 wt% hydrogels as a function of gelation time Hydrogel codea
Gelation time @ 371C (h)
Soluble fraction (wt%)
Mean7SD (wt%)
1.1.4 2.1.4 4.1.4 5.1.4
1
27.6 29.7 16.8 31.9
26.576.7
6.2.4 8.2.4 9.2.4 10.2.4
24
7.7 10.0 7.6 6.6
8.071.4
a
See Table 2 for an explanation of hydrogel code designations.
pliable indicating a further increase in mechanical stiffness of the network. The fractions of polymer chains that did not participate in physical crosslinking (soluble fraction) are shown in Table 3 as a function of gelation time for 12 wt% hydrogels. The soluble fraction represents the percent of polymer chains in the initial aqueous solution that was not involved in network formation. Overall, the soluble fraction for the 12 wt% hydrogels after 1 h of incubation at 371C was greater than those after 24 h of incubation. The average soluble fraction remaining in a 12 wt% hydrogel after 24 h of incubation was 8.071.4 wt% of protein compared to 26.576.7 wt% of protein after a 1 h incubation. To examine the influence of initial polymer concentration on the soluble fraction present in the hydrogels, the final polymer content of the hydrogels was plotted as a function of the initial polymer content for hydrogels cured between 1 and 24 h (Fig. 1a and b). In Fig. 1a, each data point represents the results for a separate hydrogel specimen prepared for each designated initial polymer content. For a given polymer concentration, less inter-specimen variation was observed with hydrogels incubated for 24 h compared to hydrogels incubated for 1 h. Similar to the results observed for the 12 wt% hydrogels (Table 3), the results suggest that regardless of initial polymer content more polymer chain entanglement and physical interactions between polymer chains occurs at longer incubation times. A linear correlation was observed between the final polymer content and the initial polymer content (R2 ¼ 0:994) for the 24 h incubation period (Fig. 1a). Fig. 1b illustrates the effect of the initial polymer concentration on the soluble fraction post-gelation as a function of incubation time. Data for hydrogels prepared at both high and low concentrations for each respective incubation time are represented. At both 1 and 24 h incubation times, there was no significant difference in soluble fraction between hydrogels prepared at high and low polymer concentrations. These results suggest that over the concentrations studied the
Fig. 1. Influence of soluble fraction on final Polymer 47K content in hydrogel: (a) effect of gelation/cure time. (K) gelation time=24 h, (J) gelation time=1 h, (—) best-fit line. Each symbol represents a single determination for each hydrogel specimen; (b) effect of Polymer 47K concentration, (&) 4 wt%, ( ) 8 wt%, ( ) 12 wt%. Bars represent mean7one standard deviation (n ¼ 3).
initial polymer concentration does not have a significant influence on the soluble fraction present in the hydrogels post-gelation. Based on these results, it appears that the extent of physical crosslinking primarily depends on the gelation time for a given polymer concentration. 3.2. Influence of temperature, pH, ionic strength, and polymer concentration on the hydrogel equilibrium swelling ratio The effect of temperature on the weight equilibrium swelling ratio for 12 wt% hydrogels cured for 1 h is reported in Fig. 2a. The equilibrium swelling ratio for 12 wt% hydrogels is not significantly different over the temperature range investigated. Fig. 2b illustrates the effect of pH on the weight equilibrium swelling ratio for 12 wt% hydrogels cured for 1 h. Results demonstrate that these hydrogels do not display any pH sensitivity in the range of pH between 2 and 9 (p ¼ 0:171).
20
20
18
18
16
16
14
14
12
12
q
q
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
10
10
8
8
6
6
4
4 0
10
20
30
40
50
60
Temperature (˚C)
(a)
0
1 2
3 4
(b)
5
6 7
8 9 10
pH
20
32
18
28
16
24
14
20 q
q
4207
12
16
10 12
8
8
6
4
4 0 (c)
0.01
0.15
0
1
Ionic Strength (M)
(d)
10
20
30
40
50
60
Temperature (˚C)
Fig. 2. Influence of environmental conditions on the weight equilibrium swelling ratio (q) for hydrogels cured for 1 h at 371C: (a) effect of temperature on q for 12 wt% hydrogels in PBS (pH 7.4, m ¼ 0:15); (b) effect of pH on q for 12 wt% hydrogels at 371C; (c) effect of ionic strength on q for 12 wt% hydrogels at 371C in phosphate buffer solution (pH 7.4); and (d) effect of polymer concentration on q for hydrogels at 371C in PBS (pH 7.4, m ¼ 0:15), (K) 8 wt%, (&) 10 wt%, (m) 12 wt%. Symbols represent mean value7one standard deviation (n ¼ 3). Some error bars are smaller than symbols.
To investigate the effect of ionic strength on Polymer 47K hydrogel swelling, the ionic strength of a phosphate buffer solution (pH 7.4) was adjusted from 0.01 to 1 m with NaCl. Fig. 2c shows that the equilibrium swelling ratio for 12 wt% hydrogel cured for 1 h at 371C does not change significantly despite a two log increase in ionic strength. Fig. 2d displays the effect of temperature on the equilibrium swelling ratio for hydrogels cured for 1 h at 371C as a function of polymer concentration. Similar swelling profiles were observed for 10 and 12 wt% hydrogels demonstrating a lack of temperature sensitivity over the temperature range investigated. When the aqueous polymer solution was further diluted to 8 wt%, a statistically significant increase in the equilibrium swelling ratio was found between 12 and 8 wt% hydrogels. For the 8 wt% gel, a slight decrease in q was observed at temperatures >371C. However, these differences were not statistically significant at an a ¼ 0:05: Further dilution of Polymer 47K did not yield physically robust hydrogel disks after 1 h of incubation; therefore, concentrations lower than 8 wt% were not evaluated.
3.3. Effect of gelation time on hydrogel equilibrium swelling ratio Since a 1 h incubation time did not yield physically robust hydrogel disks at polymer concentrations below 8 wt%, the incubation period was extended to 24 h to investigate the swelling behavior at lower polymer concentrations. Fig. 3 illustrates the effect of temperature on the equilibrium swelling ratio for Polymer 47K solutions incubated for 24 h at 371C as a function of polymer concentration. After the 24 h incubation, temperature sensitivity was not observed between 8 and 12 wt% hydrogels. The 4 wt% hydrogel displayed a decreasing trend in the equilibrium swelling ratio with an increase in temperature though the values were not statistically significant at an a ¼ 0:05: A similar, but less pronounced, trend in q was observed for 8 wt% polymer solutions incubated for 1 h (see Fig. 2d). It is possible that gravimetric analysis is not sensitive enough to distinguish a transition of the elastinlike domains in Polymer 47K hydrogels. For this reason, we also examined the temperature sensitivity of hydrogels prepared at 4, 8, and 12 wt% concentrations using
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
4208
32
1 Hydration (g water/g gel)
28 24
q
20 16 12 8
0.95 0.9 0.85 0.8
4 0
10
20
30
40
50
0
60
Temperature (˚C) Fig. 3. Effect of temperature on q for SELP hydrogels cured for 24 h. All hydrogels were stored in 1X PBS. (’) 4 wt%, (K) 8 wt%, (m) 12 wt%. Symbols represent mean value7one standard deviation (n ¼ 4). Some error bars are smaller than symbols.
30
q
20
10
0 1
2
4
6
8
10
12
14
Final Polymer Concentration (wt%)
24 Gelation Time (hours)
Fig. 4. Effect of gelation (cure) time on q for SELP hydrogels. All hydrogels were stored in 1X PBS at 201C. ( ) 8 wt%, (&) 12 wt%. Bars represent mean value7one standard deviation (n ¼ 3).
differential scanning calorimetry (DSC). DSC results for hydrogels cured for 24 h (not shown) did not display any thermal transitions in the temperature range 6–701C using a scanning rate of 21C/min. As a consequence any conformational changes resulting from an increase in temperature cannot be detected using our experimental methods. Fig. 4 displays the effect of the gelation time at 371C on the hydrogel equilibrium swelling ratio in 1X PBS (pH 7.4) at 201C. For a given polymer concentration, the equilibrium swelling ratio decreased with an increased incubation period indicating a higher crosslinking density. Swelling ratios between 8 and 12 wt%
Fig. 5. Comparison of degree of hydration and final Polymer 47K content in hydrogels as a function of gelation (cure) time. (K) gelation time=24 h, (J) gelation time=1 h, (—) H (calc), (- - -) H (calc,72%). Each symbol represents the mean value for a different batch of hydrogels (n ¼ 3).
hydrogels after 1 or 24 h incubation were significantly different (p{0:05). It is worthwhile to note that the change in swelling ratio with incubation time was more pronounced for the 8 wt% hydrogels compared with the 12 wt% hydrogels, which confirms that lower polymer concentrations need sufficient time for polymer–polymer interactions to occur. Higher variability in q was observed at lower polymer concentrations for each incubation period. This is most likely due to a fewer number of polymer chains that are able to interact with one another at lower polymer concentrations. A comparison of the final polymer concentration (adjusted for soluble fraction) and the equilibrium hydration of the hydrogels is shown in Fig. 5. The solid line in Fig. 5 represents the hydrogel hydration (H) calculated from the final polymer content. The calculated hydration is the expected amount of water in the hydrogels if the polymer content remains the same after storage and testing in the swelling studies. The experimental data fall within 72% of the line, though H and the soluble fraction were determined independently. Based on these results we can conclude that no significant hydrogel degradation or dissolution took place during storage or during the swelling studies, and any observed changes in the equilibrium swelling ratio at higher temperatures should only be related to a network de-swelling phenomena.
4. Discussion The extent of soluble fraction present in a gel-formed implant as well as its equilibrium swelling ratio may have a profound effect on the ability of the hydrogel to reproducibly and effectively control the release of
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
bioactive agents. In regards to the polymer soluble fraction, the presence of uncrosslinked polymer may result in complex formation with certain bioactive agents, such as oligonulceotides, genes, and therapeutic proteins, resulting in either precipitation of the complex in the gel or in the case of soluble complexes decreased release rates from implant devices followed by attenuated bioactivity. In addition, localized high concentrations of uncrosslinked polymer may diffuse from a gelformed implant, creating water filled channels or pores that may sufficiently increase the equilibrium swelling ratio and impact solute release behavior. The analysis of soluble fraction for Polymer 47K hydrogels demonstrate an increase in gelation time from 1 to 24 h resulting in increased polymer–polymer interactions (Fig. 1a). As a consequence, the number of polymer chains involved in the hydrogel network is increased, which leads to a decreased soluble fraction. It was found that regardless of the initial polymer content, hydrogels produced after a 24 h incubation period had a lower degree of swelling than those produced after a 1 h incubation period (Fig. 1b), suggesting more polymer chain entanglement and physical interactions between polymer chains at a longer incubation time. A linear correlation between the final polymer content and the initial polymer content for the 24 h incubation period (Fig. 1a) suggests that the soluble fraction can be controlled by adjusting the polymer gelation time. Similarly, it was found that the equilibrium swelling ratio may be controlled, within certain constraints, by adjusting the gelation time of polymer solutions (Fig. 4). Overall, these results suggest that the extent of physical crosslinking primarily depends on the gelation time for a given concentration of the SELP copolymer under study. Environmental sensitivity of crosslinked polymer networks may cause changes in the equilibrium swelling ratio that can have an effect on their ability to be used as controlled release devices for a variety of biomedical applications. Our results indicate that physically crosslinked networks of the SELP copolymer under study are insensitive to temperature over a range 6–501C. Watersoluble SELPs containing elastin repeat units exhibit an inverse temperature solubility transition [7]. In addition, previous research with covalently crosslinked elastinmimetic hydrogels has shown a decrease in the equilibrium weight swelling ratio with an increase in temperature [8–10]. Our results indicate that a reduction in polymer concentration does not significantly amplify SELP hydrogel response to temperature even though there is a significant decrease in crosslinking density (indicated by an increase in the swelling ratio, q). The temperature sensitivity effect is absent possibly due to the formation of rigid, physical crosslinks as a result of the increased number of silk units in the repeat structure (Table 1).
4209
Polymer 47K hydrogels did not display pH sensitivity (no change in q over a pH range 2–9) despite the fact that the polymer carries a net positive charge at pH 7.4 due to the presence of lysine groups (pKa 10.5) in the repeat units of the polymer backbone. Past research has demonstrated hydrophobicity induced pKa shifts in elastin protein-based polymers [3] suggesting the actual pKa of lysine in the Polymer 47K backbone may not be 10.5. No significant change in the equilibrium weight swelling ratio was observed over the range of pH studied most likely due to the nature and the extent of physical crosslinking between the silk units. Hydrogel swelling–deswelling behavior may also be influenced by the ionic strength because of factors such as the presence of ions in the network, counterions shielding ionizible groups in the polymer backbone, polymer–polymer repulsion, and the balance between the polymer–solvent interaction and the nature of chain packing in the hydrogel. For protein-based polymers, hydrophobic self-assembly is due to the competition between apolar (hydrophobic) and polar residues that are restrained by sequence in the polymer backbone such that their proximity allows them to compete for the same water molecules for hydration [11]. Thus, for water-soluble polymers, the effect of ionic strength typically manifests itself in driving hydrophobic selfassembly by counterion shielding of polar moieties in the polymer backbone, which leads to decreased competition for water molecules of hydration with its more apolar neighbors. The SELP hydrogels in this study did not display sensitivity to ionic strength over the range 0.01–1 m at pH 7.4, further highlighting how rigid crosslinks formed between the silk units overcome the potential interactions of the polymer chain with ions in the solution. Based on previous research with related genetically engineered polymers [12], we believe that Polymer 47K hydrogel crosslinks consist of localized assemblies of silklike blocks forming hydrogen-bonded b-sheets. The experimental results suggest a physically constrained, rigid network that may prevent elastinlike domains from reversibly folding in SELP hydrogels. This is supported by the fact that the presence of NaCl did not significantly change the equilibrium swelling ratio of SELP hydrogels. It is well known that with the addition of salt to aqueous protein solutions, ions compete for water of hydrophobic hydration as well as screen charges on the protein; thus, making them more prone to hydrophobic aggregation. Based on this work, it is possible that the lack of temperature sensitivity displayed by Polymer 47K hydrogels may be primarily due to crystallites prohibiting reversible transitions of elastinlike blocks as a result of the nature of the crosslinks (chain packing). Furthermore, a significant increase in the equilibrium swelling ratio was not found at temperatures o251C, where elastinlike domains
4210
A.A. Dinerman et al. / Biomaterials 23 (2002) 4203–4210
would most likely unfold if they were in their folded state in the hydrogel. This clearly suggests that the crystallites may restrain the folding/unfolding process. This is further supported by the fact that hydrogels prepared at lower polymer concentrations displayed minimal temperature effects despite a significant decrease in crosslinking density as demonstrated by high equilibrium swelling ratios. Even though a decreasing trend in the equilibrium swelling ratio was more evident at higher temperatures for hydrogels prepared at lower polymer concentrations these changes were not statistically significant.
dates as in situ gel forming implants for localized drug delivery where stimuli-sensitivity is not required.
Acknowledgements This work was partially funded by an American Foundation for Pharmaceutical Education (AFPE) Fellowship for AAD.
References 5. Conclusions Silk-elastinlike hydrogels made from a genetically engineered copolymer with an amino acid repeat sequence of [(GVGVP)4GKGVP(GVGVP)3(GAGAGS)4]12, were evaluated for their soluble fraction content as a function of gelation time and initial polymer concentration. It was found that the hydrogels contained a fraction of uncrosslinked polymer (soluble fraction) post-gelation. The extent of soluble fraction remaining in the gel was primarily dependent on the gelation time and varied from 14 to 35 wt% after 1 h incubation and from 8 to 17 wt% after 24 h incubation. Larger variations in the soluble fraction were observed for gels produced after 1 h incubation. These hydrogels were evaluated for sensitivity to environmental stimuli such as pH, temperature, and ionic strength. Unlike other crosslinked elastin-mimetic hydrogels that undergo temperature-dependent swelling and deswelling, the SELP hydrogels under study did not display significant environmental sensitivity post-gelation. A decrease in the polymer concentration resulting in higher equilibrium swelling ratios (decreasing the apparent crosslinking density) did not significantly influence the environmental sensitivity of the hydrogels. These results suggest that irreversible crystallization of silklike blocks at physiological conditions prevents elastinlike blocks from reversibly folding in the crosslinked state. The lack of temperature sensitivity in the hydrogels is probably due to the nature of polymer chain packing in the hydrogen-bonded crystallites formed from the interaction of the silk units. These results demonstrate that the SELP hydrogels with the structure above will not undergo significant changes in swelling post-gelation at physiological conditions making them promising candi-
[1] Cappello J, Crissman JW, Crissman M, Ferrari FA, Textor G, Wallis O, Whitledge JR, Zhou X, Burman D, Aukerman L, Stedronsky ER. In-situ self-assembling protein polymer gel systems for administration, delivery, and release of drugs. J Controlled Release 1998;53:105–17. [2] Petka WA, Harden JL, McGrath KP, Wirtz D, Tirrell DA. Reversible hydrogels from self-assembling artificial proteins. Science 1998;281:389–92. [3] Urry DW, Harris CM, Luan CX, Luan C-H, Gowda DC, Parker TM, Peng SQ, Xu J. Transductional protein-based polymers as new controlled-release vehicles. In: Park K, editor. Controlled drug delivery-strategies and challenges. Washington, DC: American Chemical Society, 1997. p. 405–36. [4] Nagarsekar A, Ghandehari H. Genetically engineered polymers for drug delivery. J Drug Targeting 1999;7:11–32. [5] Cappello J. Synthetically designed protein polymer biomaterials. In: Park K, editor. Controlled drug delivery: Strategies and challenges. Washington, DC: American Chemical Society, 1997. p. 439–53. [6] Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the folin-phenol reagents. J Biol Chem 1951;193:265–75. [7] Nagarsekar A, Crissman M, Crissman J, Ferrari F, Cappello J, Ghandehari H. Genetic synthesis and characterization of pH and temperature-sensitive silk-elastinlike block copolymers. J Biomed Mater Res, in press. [8] Lee J, Macosko CW, Urry DW. Elastomeric polypentapeptides cross-linked into matrixes and fibers. Biomacromolecules 2001;2:170–9. [9] Lee J, Macosko CW, Urry DW. Swelling behavior of g-irradiated cross-linked elastomeric polypentapeptide-based hydrogels. Macromolecules 2001;34:4114–23. [10] McMillan RA, Conticello VP. Synthesis and characterization of elastin-mimetic protein gels derived from a well-defined polypeptide precursor. Macromolecules 2000;33:4809–21. [11] Urry DW. Physical chemistry of biological free energy transduction as demonstrated by elastic protein-based polymers. J Phys Chem B 1997;101:11007–28. [12] Cappello J, Crissman J, Dorman M, Mikolajczak M, Textor G, Marquet M, Ferrari F. Genetic engineering of structural protein polymers. Biotechnol Prog 1990;6:198–202.