Acta BIOMATERIALIA

Acta Biomaterialia 2 (2006) 1928 www.actamat-journals.com

Mechanical properties of electrospun brinogen structures

Michael C. McManus a, Eugene D. Boland b, Harry P. Koo a, Catherine P. Barnes b, Kristin J. Pawlowski b, Gary E. Wnek c, David G. Simpson d, Gary L. Bowlin b,*
b a Department of Surgery, Virginia Commonwealth University, Richmond, VA 23298-0230, United States Department of Biomedical Engineering, Virginia Commonwealth University, P.O. Box 980694, Richmond, VA 23298-0694, United States c Department of Chemical Engineering, Case Western Reserve University, Cleveland, OH 44106, United States d Department of Anatomy and Neurobiology, Virginia Commonwealth University, Richmond, VA 23298-0709, United States

Received 8 August 2005; received in revised form 22 September 2005; accepted 22 September 2005

Abstract Fibrin and brinogen have a well-established track record in tissue engineering due to their innate ability to induce improved cellular interaction and subsequent scaold remodeling compared to synthetic scaolds. Use of brinogen as a primary scaold component, however, has been limited by traditional processing techniques that render scaolds with insucient mechanical properties. The goal of this study was to demonstrate, based on mechanical properties, that electrospun brinogen overcomes these limitations and can be successful as a tissue engineering scaold or wound dressing. Electrospun brinogen scaolds were characterized for ber diameter and pore area and subsequently tested for uniaxial mechanical properties while dry and hydrated. In addition, uniaxial mechanical testing was conducted on scaolds treated to regulate scaold degradation in serum-containing media by supplementing the media with aprotinin or cross-linking the scaolds with glutaraldehyde vapor. A linear relationship between electrospinning solution concentration and measured ber diameter was seen; ber diameters ranged from 120 to 610 nm over electrospinning concentrations of 80 to 140 mg/ml brinogen, respectively. Pore areas ranged from 1.3 lm2 to 13 lm2 over the same brinogen concentrations. Aprotinin in the culture media inhibited scaold degradation in a predictable fashion, but glutaraldehyde vapor xation produced less reliable results as evidenced by mechanical property testing. In conclusion, the mechanical characteristics of electrospun brinogen strongly support its potential use as a tissue engineering scaold or wound dressing. 2005 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
Keywords: Fibrinogen; Electrospinning; Tissue engineering; Scaold

1. Introduction Fibrinogen is a naturally occurring plasma protein (340 kDa, globular) that functions as a major element in the coagulation cascade, contributing to clot formation and wound healing [1,2]. Fibrinogen has also been dened as brin molecules coupled to charged peptides [3]. The brinogen molecule is composed of six chains two Aa, two Bb, and two c chainsthat are linked by

Corresponding author. Tel.: +1 804 828 2592; fax: +1 804 828 4454. E-mail address: glbowlin@vcu.edu (G.L. Bowlin).

29 disulde bonds [46]. This is often denoted by (AaBbc)2 since the molecule is composed of two identical halves (a dimer of AaBbc). When brinogen is exposed to thrombin, two peptides are cleaved to produce brin monomers. These monomers, in the presence of Ca2+ and factor XIII, lead to the assembly of stable brous clots (insoluble gels) and/or other brous structures. These stable structures function as natures provisional matrix, on which tissues rebuild and repair themselves, making this type of structure an attractive tissue engineering scaold [79]. Fibrinogen-based scaolds have previously been developed in the form of brin gels [7,8,1014] and wet extrusion

1742-7061/$ - see front matter 2005 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actbio.2005.09.008

20

M.C. McManus et al. / Acta Biomaterialia 2 (2006) 1928

bronectinbrinogen cables [15,16]. These studies demonstrated that brinogen-based scaolds are easily degradable and non-immunogenic [7,8] and promote increased cell migration [16]. The practical use of brin gels, however, has been limited by lack of structural integrity. Use of wet extrusion has been limited by large resulting ber size; the 200250 lm diameter ranges in which they are produced are several orders of magnitude larger than the native protein bers in the extracellular matrix, which typically have diameters of 50300 nm. A preliminary study demonstrated that the process of electrospinning could be utilized to produce brous, nonwoven brinogen structures composed of ber diameters as low as 80 nm [17]. These 80 nm bers also possessed the ultrastructure (22.5 nm banding) found in native, polymerized brinogen bers. In addition, this study showed that ber diameter was directly proportional to electrospinning solution concentration, as is typically seen for other natural and synthetic polymers [17]. Electrospinning is accomplished by inducing a large electric potential (typically 1530 kV direct current) in a polymer solution (or melt) that is separated by some distance from an oppositely charged target to create a static electric eld [18]. As electric eld potential increases, the electrostatic forces in the solution overcome the surface tension of the solution and a thin liquid jet composed of entangled polymer chains is ejected from the polymer reservoir. This jet then travels through space towards the oppositely charged target. Instabilities within the charged jet dene its orientation in space (condition previously described as whipping motion of the ber [19]). As the liquid jet travels through space, the solvent evaporates forming a ber that deposits as a dry, brous structure, and, eventually, a non-woven mat collects on the target. Over the last several years, both natural and synthetic polymers have been used in electrospun form for tissue engineering applications [17,18,2032]. While the feasibility of electrospinning non-woven brinogen scaolds of sub-micron diameter bers has previously been demonstrated, the material properties important to assess performance have not yet been evaluated. The corollary is that additional steps would be required to use brinogen in these applications since it can be rapidly degraded by the enzymatic activity of thrombin and other generic serum proteases present in any serumcontaining media or a wound site. Thus, the overriding hypothesis for this study is that brinogen can be electrospun into a non-woven structure (mat) that displays the biomimicking geometry of a native provisional matrix and subsequently be treated to maintain the necessary mechanical integrity for use as a tissue engineering scaold or wound dressing. For this study, various concentrations of brinogen were electrospun to form non-woven, brous structures, which were characterized in terms of ber diameter and approximate pore area followed by uniaxial mechanical testing of dry, hydrated and treated (to prevent serum degradation) samples.

2. Methods 2.1. Electrospinning For this study, brinogen (Bovine soluble fraction 1, Sigma Aldrich Chemical Co.) was dissolved in nine parts 1,1,1,3,3,3-hexauoro-2-propanol (HFP, Sigma Aldrich Chemical Co.) and 1 part 10 minimal essential medium (MEM) at concentrations of 80, 90, 100, 110, 120, 130, and 140 mg/ml. These concentrations were determined by a preliminary study to be the full range of solution concentrations (inherent viscosities) capable of producing electrospun bers under ambient conditions. Electrospinning was accomplished by loading the solutions into a 5.0-ml syringe, which was placed in a KD Scientic syringe pump (Model 100) for metered dispensing at 1.8 ml/h. The positive output lead of a high voltage power supply (Spellman CZE1000R; Spellman High Voltage Electronics Corp.), set to 22 kV, was attached to a blunt 18 gauge needle on the syringe as depicted in Fig. 1. A grounded target (2.5 cm wide 10 cm long 0.3 cm thick; 303 polished stainless steel) was placed 10 cm from the needle tip and rotated at 500 revolutions per minute and translated at 1.5 cm/s over a 7.5 cm travel distance to evenly coat the mandrel and create a mat of uniform thickness without imparting a large degree of alignment to the deposited bers. 2.2. Scaold characterization Following electrospinning, representative samples were taken from each mat for characterization of ber diameter and pore area. The samples were dry after the electrospinning process and required only sputter coating with gold (Electron Microscope Sciences Model 550) for scanning electron microscopy (SEM) (JEOL JSM-820 JE Electron Microscope) evaluation from 1000 to 4000 magnication. The SEM micrographs (Polaroids) were digitized with a at bed scanner (HewlettPackard Scanjet 6200C) and then analyzed by ImageTool 3.0 (Shareware provided by University of Texas Health Science Center in San Antonio) to determine the average ber diameters (average and

Fig. 1. A diagrammatic representation of the electrospinning process. A syringe is used as the reservoir and a blunt 18 gauge needle is used as the nozzle; a high voltage power supply charges the solution to 1530 kV.

M.C. McManus et al. / Acta Biomaterialia 2 (2006) 1928

21

standard deviation were calculated from 60 measurements per micrograph) in the various scaolds. All measurements were calibrated using the micrograph scale as a reference. Pore areas were also calculated by a subjective approximation of surface pores in the SEM micrographs (average and standard deviation calculated from 25 measurements per micrograph) that were at least 15 lm deep. Quantitative methods for determining porosity of electrospun scaolds are in development. Percentage open area and surface area to volume ratios were calculated using the known material density (1.38 g/cm3) [33], scaold mass, and scaold volume, as follows: Percent open area 1 calculated scaffold density 100% known material density measured mass of scaffold known material density p average fiber radius2 2 Surface area to volume ratio equivalent fiber length 2 p average fiber radius calculated scaffold volume 3 2.3. Mechanical evaluation As noted, scaolds were electrospun from brinogen solutions at concentrations ranging from 80 to 140 mg/ml to form non-woven mats for determining the resulting bulk mechanical properties. Both dry and hydrated electrospun mats were evaluated. Hydration was accomplished by placing the electrospun mats in phosphate buered saline (PBS) for 18 h at 37 C. Uniaxial material properties evaluation was performed on a MTS Bionix 200 (MTS Systems Corp.; Eden Prairie, MN) mechanical testing system incorporating a 100 N load cell with an extension rate of 10.0 mm/min to failure. A total of six test specimens (n = 6) were tested for each study group. Specimens were cut out of the scaffolds using a dog-bone (dumbbell) shaped template to assure uniformity and isolate the failure point away from the grips. The specimens had a width of 2.67 mm and a gauge length of 11.25 mm. Specimen thickness was measured on a Mitutoyo IP54 digital micrometer (Mitutoyo American Corp.; Aurora, IL) and ranged from 0.05 to 0.13 mm. Tangent modulus (tangential method automatically selected by the MTS TestWorks 4.0 software), peak stress (engineering stress based on the initial cross-sectional area), and strain to failure (also calculated automatically by the software) were determined. After evaluating the dierent electrospinning concentrations in terms of capacity to electrospin and physical characteristics, the brinogen concentration deemed to have the best mechanical and production properties was selected 1

Equivalent fiber length

for further evaluation of enzymatic scaold degradation. Mechanical properties of mats electrospun from solutions of this concentration were evaluated under simulated (no cells) cell culture conditions: (1) aprotinin treatment added to the cell culture media and (2) glutaraldehyde vapor cross-linking prior to cell culture, with the focus on scaold degradation in terms of mechanical properties. The rst group was incubated at 37 C for 24 h in complete culture medium containing Dulbeccos modied Eagles medium (DMEM)F12 nutrient mixture (F12) (1:1-DMEM:F12) supplemented with 10% fetal bovine serum (FBS) and 1.2% penicillinstreptomycin antibiotic (10,000 I.U. penicillin/mL, 10,000 lg/mL streptomycin, Mediatech, Inc.). Aprotinin (Fisher Scientic) was added in concentrations of 1, 10, 102, 103 and 104 kallikrein inhibitor units (KIU) per milliliter; this polypeptide inhibits the activity of several proteolytic enzymes such as kallikrein, plasmin, and trypsin and is present in blood and most tissues, with the highest concentration in the lungs. The second group was xed by vapor created by a 50% glutaraldehyde solution at ambient conditions for 15, 30, or 60 min at room temperature in an enclosed chamber system. These scaolds were then incubated at 37 C for 24 h in the same complete media without the aprotinin. A control group was established by incubating an untreated (no cross-linking) scaffold at 37 C for 24 h in complete media. 2.4. Statistics All statistical analyses were performed utilizing Sigma Stat (Version 2.03; SPSS, Inc.). For the data collected, normality and equal variance tests were set to reject for P 6 0.01. Samples that passed normality and equal variance tests were evaluated using a one-way analysis of variance (ANOVA) and then subjected to a pairwise multiple comparison procedure (Tukey test). Samples that failed the normality or equal variance tests were evaluated using KruskalWallis one-way ANOVA on ranks with pairwise comparisons by Dunns test and P values obtained by MannWhitney rank sum test. The a priori alpha value was set at 0.05 with signicance dened as P 6 0.05. 3. Results 3.1. Scaold characterization Fig. 2 illustrates a linear relationship (R2 = 0.97) between electrospinning solution concentration and measured ber diameter. These values closely compare with previous work by the authors [17]. Electrospun dry bers ranged in diameter from 120 to 610 nm at brinogen concentrations ranging from 80 to 140 mg/ml, respectively. SEM micrographs in Fig. 3 further illustrate the relationship between solution concentration and ber diameter. Bead formation in the 80 mg/ml brinogen scaold was caused by electrospinning near the transition concentration between electrospinning and electrospraying (insucient

22
1

M.C. McManus et al. / Acta Biomaterialia 2 (2006) 1928

Fiber Diameter (microns)

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 70 80 90 100

d= 0.0078c - 0.5121 R2 = 0.97

110

120

130

140

150

Electrospinning Fibrinogen Concentration (mg/ml)

25

tions, the average pore areas do not signicantly dier from one another. Thus, there is possibly a ber diameter threshold that must be achieved before an increase in pore area can be achieved. The scaolds produced by electrospinning were easily handled and manipulated, yet were also highly porous and had a high surface area to volume ratio. For example, calculated porosities of the scaolds were 59% and 54%, respectively, for the 110 mg/ml and 130 mg/ml concentrations. Surface area to volume ratios for these concentrations were 16,800 and 14,200 cm2/cm3, respectively, based on ber diameter. 3.2. Mechanical evaluation of electrospun scaolds Results of mechanical testing of the electrospun brinogen structures from various solution concentrations are presented in Figs. 46. Wet and dry scaolds electrospun from 80 mg/ml concentration solutions and hydrated scaolds electrospun from 90 mg/ml concentration solutions did not possess sucient structural integrity for uniaxial mechanical testing. Results presented here are based on structures created from 90 (dry only), 100, 110, 120, 130, and 140 mg/ml concentrations. In general, values for modulus and peak stress were higher for dry structures and increased for the dry structures as ber diameter increased. Values for strain at failure were higher for wet structures with a maximum at the 110 mg/ml concentration. Comparisons of dry structures to wet structures were only done on samples electrospun from the same concentration solutions; signicant dierences between the two types of structures were demonstrated in modulus of elasticity, peak stress, and strain at failure for all concentrations. More specically, analysis of modulus of elasticity data (Fig. 4) for the dry brous structures indicated that moduli produced from the 90110 mg/ml concentrations were signicantly dierent from 130 to 140 mg/ml structures (P < 0.05). Dry structures electrospun from 120 mg/ml

Pore Area (square microns)

20

R2 = 0.79
15 10

5 0 70 80 90 100 110 120 130 140 150

Electrospinning Fibrinogen Concentration (mg/ml)

Fig. 2. Line graphs depicting the relationships between brinogen electrospinning concentration, c, and ber diameter, d, (top graph) and brinogen electrospinning concentration and pore area (bottom graph). All values are statistically dierent (P < 0.05) from each other in pairwise testing, with the exception of pore areas for the 80100 mg/ml and 110 120 mg/ml concentrations.

chain entanglements resulting in the formation of discrete droplets). As seen in Fig. 2 and the micrographs presented in Fig. 3, the greatest variability in ber diameter (as a function of the mean) is expressed at a concentration of 140 mg/ml. Pore area, however, does not appear to follow the same linear trend as diameter. Pore areas ranged from 1.3 lm2 to 13 lm2 for the 80 and 140 mg/ml brinogen concentration, respectively. Pore area is aected by brinogen concentration; however, it appears that within a range of concentra-

Fig. 3. Scanning electron micrographs of electrospun brinogen at 4000 magnication (A) and at 2000 magnication (B, C). (A) 80 mg/ml with 0.12 0.06 lm bers (excluding beads) and 1.3 0.8 lm2 pores. (B) 110 mg/ml with 0.32 0.11 lm bers and 7.3 2.3 lm2 pores. (C) 140 mg/ml with 0.61 0.41 lm bers and 13.1 5.4 lm2 pores.

M.C. McManus et al. / Acta Biomaterialia 2 (2006) 1928

Modulus - Dry
Modulus of Elasticity (MPa)
90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 90 100 110 120 130 140 180.0% Dry 160.0% 140.0% Wet

23

Strain at Failure (%)

120.0% 100.0% 80.0% 60.0% 40.0% 20.0% 0.0% 90 100 110 120 130 140

Electrospinning Fibrinogen Concentration (mg/ml)

Modulus of Elasticity (MPa)

Modulus - Wet
0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
90 100 110 120 130 140

Electrospinning Fibrinogen Concentration (mg/ml)

Electrospinning Fibrinogen Concentration (mg/ml)

Fig. 6. Results of mechanical testing for dry (dark bars) and hydrated (white bars) electrospun brinogen structures are illustrated in terms of strain at failure versus the brinogen solution concentration from which structures were electrospun. Structures electrospun from the 90 mg/ml solution concentration did not maintain sucient structural integrity for material testing after being hydrated.

Fig. 4. Results of mechanical testing for dry (top) and hydrated (bottom) electrospun brinogen structures are illustrated in terms of elastic modulus versus the brinogen solution concentration from which structures were electrospun. Structures electrospun from the 90 mg/ml solution concentration did not maintain sucient structural integrity for material testing after being hydrated.

4.0 3.5
Dry Wet

Peak Stress (MPa)

3.0 2.5 2.0 1.5 1.0 0.5 0.0 90 100 110 120 130 140

from the dry 130140 mg/ml structures, and the dry 110 mg/ml structures were signicantly dierent from the 130 mg/ml structures. The only signicant dierence detected in the wet brous structure data was seen when comparing the 120130 mg/ml structures to the 140 mg/ml structures. Strain at failure data are presented in Fig. 6. Signicant dierences were only found when comparing the dry 110 mg/ml concentration brous structures to the dry 90 mg/ml and 140 mg/ml structures. 3.3. Mechanical evaluation of aprotinin-treated versus glutaraldehyde cross-linked scaolds Mechanical testing was done to compare the usefulness of both glutaraldehyde vapor cross-linking and aprotinin supplementation of culture media to control degradation of electrospun brinogen structures. The structures evaluated were electrospun from a 110 mg/ml brinogen concentration solution based on initial testing; a more complete rationale will be detailed in Section 4. Mechanical testing data are presented in Figs. 79. In general, the values for modulus of elasticity, peak stress, and strain at failure demonstrated an increasing trend up to and including 100 KIU/ml and then decreased at the higher concentration. Glutaraldehyde vapor xation for 60 min resulted in a signicant decrease in modulus but not in peak stress or strain at failure. Modulus of elasticity data are presented in Fig. 7. The moduli produced from 10 to 100 KIU/ml aprotinin concentrations were signicantly dierent from all other groups (P < 0.05) but not signicantly dierent from each other. Glutaraldehyde vapor xation for 60 min produced moduli signicantly smaller from all groups, except for the control. No other signicant dierences in moduli were observed.

Electrospinning Fibrinogen Concentration (mg/ml)

Fig. 5. Results of mechanical testing for dry (dark bars) and hydrated (white bars) electrospun brinogen structures are illustrated in terms of peak stress versus the brinogen solution concentration from which structures were electrospun. Structures electrospun from the 90 mg/ml solution concentration did not maintain sucient structural integrity for material testing after being hydrated.

concentration solutions were only signicantly dierent from the 130 mg/ml structures. As seen in Fig. 4, the wet structures demonstrated less variability in modulus between concentrations and only showed a signicant difference in modulus when comparing the 140 mg/ml structures to the 110 mg/ml and 130 mg/ml structures. Peak stress data are presented in Fig. 5. The dry 90 mg/ ml derived brous structures were signicantly dierent from the dry brous structures at all other concentrations. The dry 100 mg/ml structures were signicantly dierent

24
3.00

M.C. McManus et al. / Acta Biomaterialia 2 (2006) 1928

200.0% 180.0%

Modulus of Elasticity (MPa)

2.50

Strain at Failure (%)

C on 1 tr ol 10 /m l 10 IU /m l 10 K IU /m l 15 K IU /m l 30 in fix 60 fix

160.0% 140.0% 120.0% 100.0% 80.0% 60.0% 40.0% 20.0%

2.00 1.50 1.00 0.50 0.00

K IU K 0 00 m m in m in

0.0%
C on ol tr 1 K IU /m 10 K IU 10 0 K 10 00 K 15 m in 30 m in 60 m in

fix

IU

/m

IU

fix

fix

fix

/m

Fig. 7. Results of mechanical testing are illustrated in terms of elastic modulus versus treatment of brinogen structures electrospun from 110 mg/ml concentration. All groups were hydrated in serum-rich media. The control group received no treatment. The 1 KIU/ml through 1000 KIU/ml groups were hydrated in media supplemented with the corresponding aprotinin concentrations. The 15, 30 and 60 min x groups were placed in a glutaraldehyde vapor chamber for the corresponding time periods prior to hydration.

Fig. 9. Results of mechanical testing are illustrated in terms of strain at failure versus treatment of brinogen structures electrospun from 110 mg/ ml concentration. All groups were hydrated in serum-rich media. The control group received no treatment. The 1 KIU/ml through 1000 KIU/ml groups were hydrated in media supplemented with the corresponding aprotinin concentrations. The 15, 30 and 60 min x groups were placed in a glutaraldehyde vapor chamber for the corresponding time periods prior to hydration.

/m

2.50

4. Discussion 4.1. Scaold characterization The ability to produce bers that are more than an order of magnitude smaller than those produced by any traditional ber forming process remains one of the most significant advantages of electrospinning for biomedical applications. Fig. 2 illustrates a linear relationship (R2 = 0.97) between electrospinning solution concentration and measured ber diameter. These values closely compare with previous work by the authors [17]. As seen in Fig. 2, this linear relationship is described by the equation: d 0:0078c 0:5121 4 where d is the ber diameter and c is the solution concentration. The minimum solution concentration required to electrospin bers is predicted by this graph as the x-axis intercept or the point where ber size goes to zero. Conversion of the above equation to: d Ac c1 5 where A is a proportionality constant, c is the solution concentration and c1 is the minimum concentration required to electrospin bers allows c1 to be calculated as 65.7 mg/ml. The electrospinning threshold is dened as the observance of a phenomenon termed beads on a string which represents the combination of two distinct product morphologies: the presence of bers (electrospinning) and the presence of droplets resulting from electrospraying [27,32]. Preliminary studies demonstrated that brinogen structures electrospun from a solution concentration of 70 mg/ml produced the beads on a string morphology. Although the exact concentration at which

Peak Stress (MPa)

2.00

1.50

1.00

0.50

0.00
C on tr ol 1 K IU /m l 10 K IU /m l 10 0 K IU /m l 10 00 K IU /m l 15 m in fix 30 m in fix 60 m in fix

Fig. 8. Results of mechanical testing are illustrated in terms of peak stress versus treatment of brinogen structures electrospun from 110 mg/ml concentration. All groups were hydrated in serum-rich media. The control group received no treatment. The 1 KIU/ml through 1000 KIU/ml groups were hydrated in media supplemented with the corresponding aprotinin concentrations. The 15, 30 and 60 min x groups were placed in a glutaraldehyde vapor chamber for the corresponding time periods prior to hydration.

Peak stress data are presented in Fig. 8. The 10 100 KIU/ml aprotinin concentrations produced peak stresses that were signicantly dierent from both the control and 3060 min glutaraldehyde vapor xation. The 100 KIU/ml aprotinin concentration also produced peak stresses signicantly dierent from the 1 KIU/ml aprotinin concentration. No other signicant dierences in peak stress were observed. Strain at failure data are presented in Fig. 9. The strain at failure data showed no signicant dierences amongst the samples evaluated. Thus, no noticeable trends similar to modulus and peak stress were observed in strain at failure.

M.C. McManus et al. / Acta Biomaterialia 2 (2006) 1928

25

bers begin to form has not been fully elucidated experimentally, observational data correlates well with this predictive model. Alternatively, a power law relationship between ber size and solution concentration has previously been proposed [34,35]. Our observed data also t (R2 = 0.97) the power law correlation yielding the relationship: d 1:00E 06c
2:7

where c is the solution concentration. Interestingly, McGee [35] found poly(alkylmethacrylate) to form bers that scaled with concentration to the 2.7 power. Pore area did not vary linearly with ber diameter, but two other parameters become readily apparent when looking at the characterization data: large void volume and signicant surface area of the scaold. The large void volume within the scaold (>50%) provides space for cell inltration while also allowing three-dimensional support for the cells. The large surface area to volume ratio enables increased cellular interaction with the scaold. As previously hypothesized, small bers elicit a diminished immune response as compared to larger bers possibly through the sensing of membrane curvature [22,30]. This may be due to the fact that as cells attach to multiple thin bers, membrane curvature and focal adhesions may enhance signal transduction. Combined, these factors help explain the improvement in biocompatibility we have seen with electrospun scaolds. Pore area has been shown to be dependent on mandrel motion when electrospinning PGA by adjusting a combination of mandrel rotation and translational speeds [23]. However, it is believed that pore area may have a combined dependence on solution concentration (ber diameter) and mandrel motion. The electrospinning apparatus used in this study does not allow for the precise rotational and translational speed control that would be necessary to test this interdependence. These two controls are currently being implemented in a new electrospinning apparatus. Upon completion, the control of pore area will be addressed to further enhance the biomimicking capability of electrospun brinogen scaolds. 4.2. Mechanical evaluation The mechanical properties of a tissue engineered scaffolding material must be matched to those of the native tissue. Ideally, cells are contained in a microenvironment that achieves a distribution of forces which encourages native tissue regeneration and/or maintenance while avoiding a cellular hypotrophic or hypertrophic response. For performance analysis, modulus of elasticity, peak stress, and strain at failure were investigated since these values are commonly reported for soft tissues and other tissue engineering scaolding. The mechanical testing data illustrate the structural integrity of electrospun brinogen structures. Tangent modulus and peak stress of the dry samples were compara-

ble to those previously reported in preliminary studies on electrospun brinogen [17]. Hydration signicantly changed mechanical performance, as is often seen with natural materials. While strain at failure was signicantly increased, tangent modulus and peak stress were signicantly decreased. No single solution concentration produced scaolds with mechanical properties that were clearly superior. The 110 mg/ml solution concentration was chosen for further evaluation because it provided the smallest possible ber size without compromising the scaffolds mechanical properties. This was a subjective decision based upon our experience with electrospun scaolds. Further testing is required to conrm the ideal concentration for specic tissue engineering applications. Because brinogen is proteolytically degraded by plasmin and other metalloproteinases found in serum, we hypothesized that the mechanical properties of the scaold could be preserved in a cell culture environment by regulating brinogen degradation through physical cross-linking with glutaraldehyde vapor or enzymatic inhibition with aprotinin. Comparison of the control group (incubated at 37 C for 24 h in complete media) to the same concentration sample (110 mg/ml brinogen) hydrated with PBS showed similar values for modulus and percent strain. Peak stress in the control group was half that of the PBS hydrated group, indicating that some degree of degradation had occurred without scaold treatment. Glutaraldehyde has been used successfully to cross-link and stabilize electrospun collagen types I, II, and III in a cell culture environment [21,27]. Glutaraldehyde vapor cross-linking, however, did not demonstrate any signicant improvement in the mechanical properties of electrospun brinogen. Exposure to glutaraldehyde results in crosslinks being formed primarily between the R-group amines of lysine residues and the amino groups of other amino acids. Only those lysine residues which are on the exterior of the protein molecule can take part in this reaction [36,37]. Lysine residues are available in abundance on the surface of collagen molecules and thus, electrospun collagen scaolds can be stabilized by glutaraldehyde xation. It was theorized that glutaraldehyde xation would also increase brinogen scaold integrity because brinogen also has a signicant number of lysine residues. In the natural function of brin, these lysine residues stabilize brin clots by covalently cross-linking adjacent brin brils [38]. Fibrinogen is a soluble blood protein that does not clot because the lysine residues are hidden within the folded tertiary protein structure. After the conversion of brinogen to brin by thrombin, these lysine residues are exposed and brin clot formation ensues [39]. It must be concluded then that the process of electrospinning does not expose these lysine residues and therefore they are not available for glutaraldehyde cross-linking. Aprotinin is a broad spectrum reversible protease that inhibits the action of thrombin and a large variety of serum proteases. Addition of aprotinin to the culture media at concentrations of 10 KIU/ml and 100 KIU/ml led to a

26

M.C. McManus et al. / Acta Biomaterialia 2 (2006) 1928

signicant increase (P < 0.05) in modulus and peak stress compared to the control (Figs. 7 and 8). The addition of the high concentration of aprotinin (1000 KIU/ml) has a deleterious aect on the mechanical properties. Addition of aprotinin demonstrated no signicant improvement in strain at failure over the control (Fig. 9). While the treatment measures utilized in this study were ineective at modifying the elongation of the hydrated structures, aprotinin supplementation resulted in a signicant increase in modulus and peak stress. Fibrin clot formation, stabilization and degradation are a complex process involving multiple enzymes including thrombin, plasmin and dierent generic proteases. Aprotinin is a broad spectrum protease inhibitor that aects all of these enzymes to dierent degrees. It could be theorized that the improved material properties reported for the 100 KIU/ml and 1000 KIU/ml aprotinin groups is due to an ideal balance of scaold degradation and subsequent brin polymerization resulting in scaold welding or cross-linking. Further work will be required to fully dene the eects of aprotinin and this novel structure. The most notable result was the greater than 5-fold increase in percent strain at failure that was seen with hydration of electrospun brinogen scaolds. With the addition of aprotinin, values for percent strain at failure as high as 141% were maintained. This result is comparable to electrospun synthetic polymers [20,23] and indicates that electrospun brinogens mechanical properties make it an excellent candidate for soft tissue applications that require high extensibility. Many authors have recently reported using brinogen or brin as components of tissue engineering scaolds, but many have not reported on mechanical properties [4046]. Previously reported mechanical properties of other

brinogen-based and electrospun tissue engineering scaffolds are presented in Table 1 for comparison. Three comparisons are particularly noteworthy. First, although extruded bers have a much greater peak stress, the minimum ber size exceeds 400 lm in diameter [47], which is too large for proper cell interactions as previously discussed. Second, hydrated electrospun brinogen scaolds have comparable mechanical properties to brinogen hydrogels after the hydrogels have been remodeled by several weeks of cell culture with cultured cells producing native extracellular matrix within the hydrogel. Once treated with aprotinin, the electrospun scaolds demonstrated values that were approximately 2-fold greater for modulus and 2.5-fold greater for peak stress compared to hydrogels. It should also be noted that mechanical testing on the electrospun brinogen scaolds was done prior to cell seeding and culture. We hypothesize that scaold remodeling by seeded cells and the cells themselves will give the scaold increased tissue-like material properties. Studies are currently underway in our laboratory to address this issue. Third, the electrospun brinogen scaolds demonstrated strains at failure that were 2-fold greater than extruded bers. The exact impact that this will have on biomedical applications is unknown, but it can be hypothesized that this high degree of deformability may help facilitate cell migration through the scaold. One of the most attractive aspects of the electrospinning process is that scaolds of various shapes and sizes can be constructed with precise control of ber orientation, composition (blended bers), and dimension. Complex constructs can be fabricated to closely recapitulate the structural and chemical composition of the native tissue structures. For example, this process can fabricate at sheets for on-lay grafts, seamless tubes, or any three-dimensional

Table 1 Mechanical properties of other brinogen-based and electrospun tissue engineering scaolds compared to mechanical properties for electrospun brinogen Modulus (MPa) Dry scaolds Extruded brinogen (with bronectin) Electrospun PGA Electrospun PDS Electrospun brinogen 110 mg/ml Wet scaolds Fibrinogen hydrogel PEG-brinogen hydrogel Electrospun brinogen PBS hydrated 110 mg/ml in PBS 110 mg/ml-Media w/FBS Control Aprotinin Glutaraldehyde 2095 515 8.568.5 22.4 Peak stress (MPa) 82 1.728.1 25.7 0.703.34 2.35 Strain (%) Author Bak [36] Underwood [15] Boland [23] Boland [20]

18.652.4 80100 45170 1425 25

0.010.4 0.30.95 0.0001 0.300.58 0.38 0.34 0.371.81 0.190.55

0.0060.09 0.150.48 0.004 0.340.54 0.47 0.23 0.451.20 0.280.60

30 100130 130 103 110141 71105

Neidert [49] Ross [13] Almany [50]

All modulus and peak stress values reported in dierent units in the literature have been converted to MPa for straightforward comparison.

M.C. McManus et al. / Acta Biomaterialia 2 (2006) 1928

27

shape in which a mold can be fabricated. Also, the ability to co-spin polymers with various additives (e.g. growth factors) oers the possibility of tailoring the scaold to a specic site and application [48]. In short, electrospinning makes it possible to create ideal scaoldings for a wide array of biomedical and tissue engineering applications. 5. Conclusions In conclusion, it may now be possible to construct truly biomimicking brous scaold using the process of electrospinning to form a variety of composite structures specic to the application at hand. Specically, there is tremendous potential in the use of electrospun brinogen as a tissue engineering scaold or wound dressing. Its innate biological and electrospun scaolding mechanical properties make it an excellent candidate for soft tissue applications that require high extensibility. Disclosure: Several authors have United States and International patents pending concerning technology presented in this manuscript, and this technology has been licensed to NanoMatrix, Inc., of which several authors have a nancial interest. Acknowledgments The authors would like to thank the AD Williams Foundation and the Whitaker Foundation (TF-02-0013) for their support of this research. We would also like to thank Ms. Judy Williamson for her assistance in obtaining the SEM micrographs. Funded by: The Whitaker Foundation (TF-02-0013) and the A.D. Williams Trust Funds. References
[1] Weigel PH, Fuller GM, LeBoeuf RD. A model for the role of hyaluronic acid and brin in the early events during the inammatory response and wound healing. J Theor Biol 1986;119:21934. [2] Erban JK. P-selectin and wound healing. Behring Institute Mitteilungen 1993:24857. [3] Bettelheim F, Bailey K. The products of the action of thrombin on brinogen. Biochim Biophys Acta 1952;9:5789. [4] Blomback B, Blomback M, Edman P, Hessel B. Human brinopeptides. Isolation, characterization and structure. Biochim Biophys Acta 1966;115:37196. [5] Blomback M, Blomback B, Mammen E, Prasad A. Fibrinogen Detroita molecular defect in the N-terminal disulphide knot of human brinogen? Nature 1968;218:1347. [6] Mosesson M, Siebenlist K, Meh D. The structure and biological features of brinogen and brin. Ann NY Acad Sci 2001;936:1130. [7] Peretti G, Randolph M, Zaporojan V, Bonassar L, Xu J, Fellers J, et al. A biomechanical analysis of an engineered cell-scaold implant for cartilage repair. Ann Plast Surg 2001;46:5337. [8] Passaretti D, Silverman RP, Huang W, Kirchho CH, Ashiku S, Randolph MA, et al. Cultured chondrocytes produce injectable tissue-engineered cartilage in hydrogel polymer. Tissue Eng 2001;7:80515. [9] Clark R, Lanigan JM, DellaPelle P, Manseau E, Dvorak HF, Colvin RB. Fibronectin and brin provide a provisional matrix for epidermal

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17] [18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27] [28]

[29]

[30]

cell migration during wound reepithelialization. J Invest Dermatol 1982;79:2649. Mol A, van Lieshout MI, Dam-de Veen CG, Neuenschwander S, Hoerstrup SP, Baaijens FPT, et al. Fibrin as a cell carrier in cardiovascular tissue engineering applications. Biomaterials 2005;26:311321. Karp JM, Sarraf F, Shoichet MS, Davies JE. Fibrin-lled scaolds for bone-tissue engineering: an in vivo study. J Biomed Mater Res 2004;71A:162. Cummings CL, Gawlitta D, Nerem RM, Stegemann JP. Properties of engineered vascular constructs made from collagen, brin, and collagen-brin mixtures. Biomaterials 2004;25:3699706. Ross JJ, Tranquillo RT. ECM gene expression correlates with in vitro tissue growth and development in brin gel remodeled by neonatal smooth muscle cells. Matrix Biol 2003;22:47790. Shreiber DI, Enever PAJ, Tranquillo RT. Eects of PDGF-BB on rat dermal broblast behavior in mechanically stressed and unstressed collagen and brin gels. Exp Cell Res 2001;266:15566. Underwood S, Afoke A, Brown R, MacLeod A, Shamlou P, P D. Wet extrusion of bronectin-brinogen cables for application in tissue engineering. Biotechnol Bioeng 73;2001:295305. Ahmed Z, Underwood S, Brown R. Low concentrations of brinogen increase cell migration speed on bronectin/brinogen composite cables. Cell Motil Cytoskeleton 2000;46:616. Wnek GE, Carr M, Simpson DG, Bowlin GL. Electrospinning of nanober brinogen structures. Nano Lett 2003;3:2136. Huang ZM, Zhang YZ, Kotaki M, Ramakrishna S. A review on polymer nanobers by electrospinning and their applications in nanocomposites. Compos Sci Technol 2003;63:222353. Shin YM, Hohman MM, Brenner MP, Rutledge GC. Experimental characterization of electrospinning: the electrically forced jet and instabilities. Polymer 2001;42:995567. Boland ED, Coleman BD, Barnes CP, Simpson DG, Wnek GE, Bowlin GL. Electrospinning polydioxanone for biomedical applications. Acta Biomater 2005;1:11523. Boland ED, Matthews JA, Pawlowski KJ, Wnek GE, Simpson DG, Bowlin GL. Electrospinning collagen and elastin: preliminary vascular tissue engineering. Frontiers Biosci 2004;9:142232. Boland ED, Telemeco TA, Simpson DG, Wnek GE, Bowlin GL. Utilizing acid pretreatment and electrospinning to improve biocompatibility of poly(glycolic acid) for tissue engineering. J Biomed Mater Res 2004;71B:14452. Boland ED, Wnek GE, Simpson DG, Pawlowski KJ, Bowlin GL. Tailoring tissue engineering scaolds using electrostatic processing techniques: a study of poly(glycolic acid) electrospinning. J Macromol SciPure Appl Chem 2001;38:123143. Huang L, Nagapudi K, Apkarian RP, Chaikof EL. Engineered collagen-PEO nanobers and fabrics. J Biomater SciPolym Ed 2001;12:97993. Huang L, McMillan RA, Apkarian RP, Pourdeyhimi B, Conticello VP, Chaikof EL. Generation of synthetic elastin-mimetic small diameter bers and ber networks. Macromolecules 2000;33:298997. Matthews JA, Boland ED, Wnek GE, Simpson DG, Bowlin GL. Electrospinning of collagen type II: a feasibility study. J Bioactive Compatible Polym 2003;18:12534. Matthews JA, Wnek GE, Simpson DG, Bowlin GL. Electrospinning of collagen nanobers. Biomacromolecules 2002;3:2328. Min BM, Lee G, Kim SH, Nam YS, Lee TS, Park WH. Electrospinning of silk broin nanobers and its eect on the adhesion and spreading of normal human keratinocytes and broblasts in vitro. Biomaterials 2004;25:128997. Mo XM, Xu CY, Kotaki M, Ramakrishna S. Electrospun P(L-LA-CL) nanober: a biomimetic extracellular matrix for smooth muscle cell and endothelial cell proliferation. Biomaterials 2004;25: 188390. Sanders JE, Stiles CE, Hayes CL. Tissue response to single-polymer bers of varing diameters: evaluation of brous encapsulation and macrophage density. J Biomed Mater Res 2000;52:2317.

28

M.C. McManus et al. / Acta Biomaterialia 2 (2006) 1928 [41] Bensaid W, Tritt JT, Blanchat C, Oudina K, Sedel L, Petite H. A biodegradable brin scaold for mesenchymal stem cell transplantation. Biomaterials 2003;24:2497502. [42] Schoeller TN, Neumeister MW, Huemer GM, Russell RC, Lille S, Otto-Schoeller A, et al. Capsule induction technique in a rat model for bladder wall replacement: an overview. Biomaterials 2004;25:166373. [43] Ameer GA, Mahmood TA, Langer R. A biodegradable composite scaold for cell transplantation. J Orthop Res 2002;20:169. [44] Han B, Schwab IR, Madsen TK, Issero RR. A brin-based bioengineered ocular surface with human corneal epithelial stem cells. Cornea 2002;21:50510. [45] Long JL, Tranquillo RT. Elastic ber production in cardiovascular tissue-equivalents. Matrix Biol 2003;22:33950. [46] Zhang ZY, Shum P, Yates M, Messersmith PB, Thompson DH. Formation of brinogen-based hydrogels using phototriggerable diplasmalogen liposomes. Bioconjug Chem 2002;13:6406. [47] Bak H, Afoke A, McLeod AJ, Brown R, Ayazi Shamlou P, Dunnill P. The impact of rheology of human bronectin-brinogen solutions on bre extrusion for tissue engineering. Chem Eng Sci 2002;57:91320. [48] Kenawy ER, Bowlin GL, Manseld K, Layman J, Simpson DG, Sanders EH, et al. Release of tetracycline hydrochloride from electrospun poly(ethylene-co-vinylacetate), poly(lactic acid), and a blend. J Control Rel 2002;81:5764. [49] Neidert MR, Lee ES, Oegema TR, Tranquillo RT. Enhanced brin remodeling in vitro with TGF-beta 1, insulin and plasmin for improved tissue-equivalents. Biomaterials 2002;23:371731. [50] Almany L, Seliktar D. Biosynthetic hydrogel scaolds made from brinogen and polyethylene glycol for 3D cell cultures. Biomaterials 2005;26:246777.

[31] Shields K, Beckman M, Bowlin G, Wayne J. Mechanical properties and cellular proliferation of electrospun collagen type II. Tissue Eng 2004;10:15107. [32] Stitzel JD, Pawlowski KJ, Wnek GE, Simpson DG, L BG. Arterial smooth muscle cell proliferation on a novel biomimicking, biodegradable vascular graft scaold. J Biomater Appl 2001;15:112. [33] van der Drift ACM, Poppema A. Hydrodynamic characterization of human brinogen. In: Haverkate F, Henschen A, Nieuwenhuizen W, Straub PW, editors. Fibrinogen: structure, functional aspects, metabolism. New York, NY: Walter de Gruyter; 1983. p. 318. [34] Deitzel JM, Kleinmeyer J, Harris D, Tan NC. The eect of processing variables on the morphology of electrospun nanobers and textiles. Polymer 2001;42:26172. [35] McGee MG, Elkins CL, Long TE. Inuence of self-complementary hydrogen bonding on solution rheology/electrospinning relationships. Polymer 2004;45:870515. [36] Pearse A. Histochemistry, theory and applied. 4th ed. Edinburgh: Churchill-Livingstone; 1980. [37] Kiernan J. Formaldehyde, formalin, paraformaldehyde and glutaraldehyde: what they are and what they do? Microsc Today 2000;8:812. [38] Clark RA, Lanigan JM, DellaPelle P, Manseau E, Dvorak HF, Colvin RB. Fibronectin and brin provide a provisional matrix for epidermal cell migration during wound reepithelialization. J Invest Dermatol 1982;79:2649. [39] Ariens R, Lai T, Weisel J, Greenberg C, Grant P. Role of factor XIII in brin clot formation and eects of genetic polymorphisms. Blood 2002;100:74354. [40] Itabashi Y, Miyoshi S, Kawaguchi H, Yuasa S, Tanimoto K, Furuta A, et al. A new method for manufacturing cardiac cell sheets using brin-coated dishes and its electrophysiological studies by optical mapping. Artic Organs 2005;29:95103.