Correlating network structure with functional properties of capillary alginate gels for muscle fiber formation

Food Hydrocolloids 72 (2017) 210e218

Contents lists available at ScienceDirect

Food Hydrocolloids journal homepage: www.elsevier.com/locate/foodhyd

Correlating network structure with functional properties of capillary alginate gels for muscle fiber formation €m c, d, * E. Schuster a, P. Wallin b, F.P. Klose b, J. Gold b, A. Stro a

Soft Material Science, SP - Food and Bioscience, Gothenburg, Sweden Biological Physics, Department of Physics, Chalmers University of Technology, Gothenburg, Sweden c Applied Chemistry, Department of Chemistry and Chemical Engineering, Chalmers University of Technology, Gothenburg, Sweden d SuMo Biomaterials, VINN Excellence Center, Chalmers University of Technology, Gothenburg, Sweden b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 27 February 2017 Received in revised form 23 May 2017 Accepted 28 May 2017 Available online 31 May 2017

Capillary alginate gels have the potential to be used as scaffold for the growth of muscle cells for cultured meat owing to the formation of aligned skeletal muscle cells along the length of self-assembled microcapillaries within the calcium alginate gel. The functional properties (mechanical and permeability) of the gels were determined and correlated to the nano-lengthscale of the gel network using small-angle Xray scattering. Calcium ions were let to diffuse into the alginate solution in order to obtain spontaneously formed capillaries. We show that the resulting calcium alginate network is isotropic in the plane perpendicular to the inflow of cross linking ions while anisotropic in the parallel plane. The structural anisotropicity is reflected in the mechanical properties (measured via uniaxial stress relaxation) of the gel, where a larger force is required to compress the gel in the isotropic plane than in the anisotropic plane. The findings suggest that the network is layered, or composed of “sheets” with denser regions of alginate, sheets that are weakly attached to each other, similar to the structure of bacterial cellulose. Such structure would further explain the increased permeability of labeled dextran (as determined using fluorescence recovery after photo-bleaching) that we observed in the alginate gels used in this study, as compared to internally set calcium alginate gel. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Cultured meat Anisotropic gels Ionotropic gelation SAXS Rheology

1. Introduction Cultured meat, or the cultivation of skeletal muscle cells ex vivo for food purposes (Edelman, McFarland, Mironov, & Matheny, 2005), has the potential to reduce the environmental impact of meat products compared with conventional meat production (Tuomisto & Teixeira de Mattos, 2011; Mattick, Landis, Allenby, & Genovese, 2015). Cultured meat has received growing interest, and different ways of culturing muscles cells for food purposes have been proposed (Datar & Betti, 2010; Kadim, Mahgoub, Baqir, Faye, & Purchas, 2015; Verbeke, Sans, & Van Loo, 2015). However, there is a lack of scalable cell culture substrates, also known as scaffolds, and of cell culture approaches suitable for the large quantities required in food production (Post, 2012). Skeletal muscle tissue consists of long muscle fibers that are

* Corresponding author. Department of Chemistry and Chemical Engineering, Chalmers University of Technology, Kemiv€ agen 10, 412 96, Gothenburg, Sweden. € m). E-mail address: [email protected] (A. Stro http://dx.doi.org/10.1016/j.foodhyd.2017.05.036 0268-005X/© 2017 Elsevier Ltd. All rights reserved.

aligned parallel to each other, and one important aspect is the ability of the scaffold to promote such structural organization (Zhao, Zeng, Nam, & Agarwal, 2009). Therefore, alginate capillary gels, which contains self-assembled and parallel capillary channels interspersed within the gel (Thiele & Hallich, 1958; Schuster et al., 2014), offer an interesting alternative to other scalable scaffold systems for cultured meat applications for example, those based on beads (Wallin, Hglund, Wildt-Persson, & Gold, 2012; Bhat & Fayaz, 2011). Furthermore, calcium alginate gels form under mild conditions, such as room temperature, neutral pH, and in the absence of any toxic substances (Draget & Taylor, 2011), they are edible and available from non-animal sources (Strom et al., 2009). Alginate gels can be set in various ways resulting in different gel microstructures. Internally set gels are made from slow dissolution of a sparingly soluble calcium salt dispersed within the alginate €m & solution and in the presence of acid (Draget et al., 2001; Stro Williams, 2003). Externally set gels are prepared by dripping an alginate solution into calcium solution or alternatively by pouring the alginate solution into a dialysis membrane, followed by immersion of the dialysis membrane in calcium chloride solution

E. Schuster et al. / Food Hydrocolloids 72 (2017) 210e218

(Draget et al., 2001). In both cases, crosslinking ions are let to diffuse into the alginate solution, upon which a gel is formed. Capillary alginate gels are formed via the external route but have the added step of the alginate being anchored to the beaker wall. Again, calcium solution is let to diffuse into the alginate solution giving rise to spontaneous creation of capillaries within the formed gel (Thiele & Hallich, 1958; Schuster et al., 2014), provided the € m, concentration of the ion solution is high enough (Caccavo, Stro Larsson, & Lamberti, 2016). In this paper, we refer to the gels produced by the latter method (i.e., externally set gels where the alginate has been anchored to the beaker wall) as directed externally set alginate gels. The network of internally set calcium alginate gels contain junctions zones that are composed of multiple chain segments as revealed using small-angle X-ray scattering (SAXS) (Stokke et al., 2000). The size of the junction zones (or the number chains segments joined together) is controlled by the ratio of [Ca2þ] to Cp, where [Ca2þ] refers to calcium concentration and Cp to alginate polymer concentration. The authors found however no clear correlation between the network structure and the rheological properties of the gels. Externally set alginate gels (calcium was let to diffuse across a dialysis membrane, in which the alginate was contained) were shown to be composed of alginate molecules that were aligned perpendicular to the direction of the inflow of Ca2þ (Maki et al., 2011). Kratky plots of the scattering profiles of both internally (Stokke et al., 2000; Hermansson, Schuster, Lindgren, €r, & Stro €m, 2016) and externally (Maki et al., 2011) set gels Altska have revealed the formation of cylindrical rod structures composed of several alginate chains. In the case of externally set gels, such structures were observed closest to the center of the gel Maki et al., 2011. As far as we are aware, mechanical and/or rheological properties of externally set gels have not been correlated to the gel network structure revealed by SAXS. It could be argued that directed externally set alginate gels should have some similarity in structure to the externally set gels, which were formed within a dialyse tube immersed into CaCl2 solution. In both cases, crosslinking ions are set to diffuse into the gel from an outer compartment; however, the directed externally set gels have the extra complexity that the alginate is anchored, adding an internal stress vector within the gel. The rheological and molecular transport properties of gel for tissue engineering scaffolds are key determinants of cell viability, migration, and differentiation (Engler et al., 2004), and need to be considered in the evaluation of potential scaffold materials (Fitzgerald, Bootsma, Berberich, & Sparks, 2015; Chaudhuri et al., 2016). Suitable mechanical properties of the gel are also an important requirement for the practical handling, delivery, and use of the scaffolds. The rheological (moduli) and mechanical (stress and strain at break) properties of internally set gels correlate with both the alginate concentration and the calcium concentration, and are generally considered to be related to the number and strength of load-bearing junction zones (Mitchell & Blanshard, 1976; Zhang, Daubert, & Foegeding, 2005; Draget (2009, pp. 807e828)). Such correlation was not found for capillary alginate gels, and microstructural investigations of capillary alginate gels using TEM revealed the presence of two distinct network structures. When visualized in one plane from above the capillaries, an anisotropic area was observed in close vicinity to the capillaries, while a seemingly isotropic area was observed in the bulk between capillaries (Schuster et al., 2016). The presence of the markedly different features of the network, together with the additional difficulty of open capillaries interspersed within the gel, makes the correlation of functionality with microstructure, so far challenging or not possible for the capillary alginate gels (Schuster et al., 2014). In this study, we set out to (a) determine the potential of

211

capillary alginate gels to act as guidance cues for muscle cell alignment and (b) improve the current understanding of the microstructure of capillary alginate gels. The latter being of high importance in order to facilitate controllable mechanical and transport properties of the gel. Stress relaxation was used to study the mechanical properties of the gels, since both the elastic response of tissue-engineered scaffolds (Fitzgerald et al., 2015) and the reversibility of crosslinks, exhibited as relaxation, are of importance for cell migration (Chaudhuri et al., 2016). Fluorescence recovery after photobleaching (FRAP) was used to study the molecular transport of dextran within the gel. The functional properties of gels were determined and correlated to the nano-lengthscale of the gels using SAXS. 2. Materials and methods 2.1. Materials The alginate, Protanal RC 6650, was provided by FMC BioPolymer, UK. The alginate contains 70% guluronate, according to the supplier. The intrinsic viscosity of the alginate (dialyzed and dissolved in 50 mM Na2SO4) was determined to be 0.54 ml/mg, from which the molecular weight was estimated as 159 kDa using the Mark-Houwink equation, [h] ¼ KMa, where K ¼ 4.85106 and a ¼ 0.97 (Hermansson et al., 2016). RGD pre-coupled MVG GRGDSP peptide-coupled alginate was purchased from NovaMatrix, Norway. CaCO3 with an average particle size of 10 mm was obtained from Provencale SA, France. Glucono-d-lactone (GDL) was obtained from SigmaAldrich, USA, as was NaCl, CaCl2, HCl and NaOH. The fluorescent probes used were FITC-dextran with 10 kDa, 70 kDa and 500 kDa molecular weight obtained from Invitrogen Molecular Probes, USA. The secure-seal spacers used were 120 mm thick and 9 mm in diameter (Invitrogen, USA). C2C12 skeletal muscle progenitor cells from mice (91031101) were obtained from ECACC, UK. 2.1.1. C2C12 growth medium (GM) Dulbecco's Modified Eagle's Medium high glucose 4,5 g/l with sodium pyruvate without L-glutamine was obtained from PAA Laboratories, Austria, and supplemented with 1% L-glutamine (LG) and 1% penicillin/streptomycin (PS), both from Invitrogen, USA, and with 10% fetal bovine serum (FBS), which was obtained from PAA Laboratories, Austria. 2.1.2. C2C12 differentiation medium (DM) This medium is similar to C2C12 GM but with the FBS being replaced with horse serum (HS), and it was obtained from PAA Laboratories, Austria. Phosphate-buffered saline (P4417-100TAB) was purchased from Sigma-Aldrich, USA. 2.2. Methods Alginate solutions (1.8 and 1.5% w/w) were prepared by careful addition of alginate powder to deionized water at room temperature under vigorous stirring. The dispersion was thereafter heated to 353 K in a water bath and kept at this temperature for 30 min or until dissolution was obtained. The pH of the polymeric solution was adjusted from pH 7.3 to pH 7 using 0.1 M HCl. 2.2.1. Gel preparation Internally set alginate gels were prepared by controlled release of calcium. CaCO3 and GDL were rapidly dispersed in water and immediately added to the alginate solution to yield a final alginate concentration of 1.5%. The dispersions were poured into cylindrical Teflon molds (h ¼ 12.5 mm; d ¼ 12.5 mm). The molds were sealed and the samples were allowed to equilibrate and set at room

212

E. Schuster et al. / Food Hydrocolloids 72 (2017) 210e218

temperature for 48 h prior to use. It is important to note that the GDL was always used in stoichiometric equivalence to the CaCO3 (e.g., 15 mM CaCO3 and 30 mM GDL), to keep the pH constant during network formation. Directed externally set alginate gels were prepared by using a brush to coat the internal wall of a glass beaker (V ¼ 50 ml and d ¼ 40 mm) with alginate by brushing the internal wall with alginate solution, which then was allowed to dry in an oven set at 383 K for 30 min. The procedure was repeated three times. A volume of 20 ml alginate solution (1.5% w/w) at 293 K was poured into the glass beaker. The surface of the solution was sprayed with CaCl2 solution until a gel membrane was formed. The gel membrane was left to set for 30 min, after which a CaCl2 solution at 293 K was carefully poured on top of the membrane. The CaCl2 solution was left to diffuse through the membrane and into the alginate solution for 48 h prior to use of the gel. For RGD-functionalized gels, the RGD-modified alginate was mixed in a 1:5 ratio with regular alginate prior to the gelation process. The internally and externally set gels were compared at similar R values, i.e., at a similar ratio between [Ca2þ] and [guluronate] according to;

solution, and the cells were incubated for an additional hour at room temperature. Thorough washing with PBS was performed at the end. After the myosin heavy chain staining, cell nuclei were stained with 46-diimidazolin-2-phenylindole (DAPI 1:1000, Invitrogen, USA), actin filaments were stained with rhodamine/phalloidin (1:400, Invitrogen, USA), and cell membranes were permeabilized with 0.1% Triton-X100 solution all diluted in PBS. Samples were incubated for 30 min at room temperature and then washed with PBS.

  2 Ca2þ R¼ ½guluronate

2.2.5. The confocal laser scanning microscopy-FRAP protocol The confocal laser scanning microscope system used consists of a Leica SP2 AOBS microscope (Germany) with a 20, 0.5 NA water objective, with the following settings: 256 256 pixels, zoom factor 4 (with a zoom-in during bleaching), and 800 Hz, yielding a pixel size of 0.73 mm and an image acquisition rate of two images per second. The FRAP images were stored as 12-bit TIFF-images. The 488 nm line of an argon laser was utilized to excite the fluorescent probes. The beam expander was set to 1, which lowered the effective NA to approximately 0.35 and yielded a slightly better bleaching and a more cylindrical bleaching profile. In this paper, we refer to the bleached areas as the region of interest, and the areas were 30 mm large disks (nominal radius rn ¼ 15 mm) at 50 mm into the sample. The measurement routine consisted of 20 pre-bleach images, 1 bleach image-gaining an initial bleaching depth of z 35% of the pre-bleach intensity in the region of interest and 50 postbleach frames, recording the recovery. The FRAP data were normalized by the pre-bleach fluorescence intensity. The free diffusion coefficients D0 of the probes in the absence of alginate were determined at 298 K. The probes were dissolved in CaCl2 or CaCO3 and GDL solutions corresponding to the respective R-values. Then 7 ml of the probe solutions were placed into secureseal spacer grids between two cover glass slides, and the FRAP measurements were carried out on the locked samples. To conduct FRAP measurements, corresponding amounts of FITC-dextran of molecular weights of 10, 70, and 500 kDa were introduced in the alginate solution before the addition of calcium, to yield a probe concentration of 100 ppm in the gelled system. The probe concentration was chosen to be in the linear regime of the fluorescence dependence on the concentration (Jonasson, Loren, Olofsson, Nyden, & Rudemo, 2008). As described above, the capillary gels were set during 2 days. The membrane above the formed gel was removed, and a 2 cm  2 cm x 2 cm sample was cut out centrally and directly under the membrane - on which to perform the FRAP measurement. The surface of the cube, cut perpendicular to the direction of the capillary growth and calcium diffusion, was placed on a cover glass slide and then loaded on the microscope stage and FRAP measurements were carried out in the upright mode of the microscope at constant 298 K. At least six FRAP measurements were performed on different spatial coordinates per sample. The FRAP model termed the “most likelihood estimation method for FRAP data with a Gaussian starting profile” (Jonasson et al., 2008) was utilized to analyze the data with MATLAB software (Mathworks, USA).

(1)

Internally set gels were studied with R-values of 1 and 3. The same R values were obtained for the externally set gels, but for these gels the concentration of the Ca2þ solution (0.025, 0.04, 0.1 and 0.2 M) and volume were varied. 2.2.2. Gel preparation for cell experiments After the cross-linking of the alginate hydrogels, the gels were cut open and small disks (7 mm in diameter and 2 mm thick) were cut parallel to the direction of the capillaries. The disks were carefully transferred into petri dishes with the cut-open capillaries exposed at the top. Prior to the cell culture, the disks were sterilized with ethanol, in accordance with Stoppel et al.’s procedure (Stoppel et al., 2014). In brief, disks were immersed in 70% ethanol for 20 min, washed in sterile MilliQ water for 5 min twice, and then underwent one more extensive washing step in sterile MilliQ water for at least 10 min to remove all remaining ethanol from the gel. 2.2.3. Cell culture C2C12 cells were cultured in conventional T-25 flasks (Nunc, Denmark) in preparation for the cell experiments and were always passaged prior to reaching confluence, passages used for experiments: 17e30. All cell handling was performed under sterile conditions with an aseptic technique and cells were cultured at 310 K with 5% CO2 and 96% humidity. Cells were seeded on the prepared alginate capillary disks at a density of 1300 cells/mm2 and cultured in GM for 24 h. After this initial attachment phase, the medium was exchanged for DM and cells were cultured for an additional 7 days. Samples were then fixated with ice-cold 70% ethanol for 20 min and washed with PBS. The first staining performed was for myosin heavy chains using two-stage antibody staining. Directly after fixation the cultures were incubated in a blocking solution consisting of 3% skimmed milk powder (Scharlab, Spain), 0.1% Triton X-100 solution (Sigma-Aldrich, USA), and PBS for 10 min at room temperature. The solution was exchanged for the blocking solution now containing the primary monoclonal mouse antibody anti-myosin MY-32 (1:1000 M4276 Sigma-Aldrich, USA) and incubated for 1 h at room temperature. The samples were washed with the pure blocking solution three times before applying a fluorescent secondary antibody against mouse (1:2000 AlexaFluor488, Invitrogen, USA). The secondary antibody was also diluted in the blocking

2.2.4. Fluorescence microscopy Fluorescence microscopy images were taken with an upright fluorescent microscope (Axioplan 2, Zeiss, Germany) with a black and white digital camera module (AxioCam, Zeiss, Germany). Image acquisition and microscope control were carried out with AxioVision software 4.2 (Zeiss, Germany). The image spots on the samples were chosen manually.

E. Schuster et al. / Food Hydrocolloids 72 (2017) 210e218

2.2.6. Small-angle X-ray scattering Experiments were carried out using the I911-4 beamline at the MAX IV Laboratory in Lund, Sweden. An X-ray beam with a wavelength of 0.91 Å was selected. The SAXS patterns were collected using a PILATUS 1 M detector (DECTRIS, Switzerland) with a pixel size of 172 mm, which was located 1887 mm from the sample position, yielding a range of q ¼ 0.07e0.397 nm1. Scattering patterns were acquired at room temperature using exposure times of 30 s for solutions and 60 s for gelled samples. The data processing was carried out using Bli911-4 software. The alginate solutions were placed in a multiple-position sample holder (7.9  4  1.7 mm); the sample holder was then sealed with Kapton tape on both sides. The alginate gels were prepared as follows: directly before the measurement, they were cut into 4  4  1.7 mm blocks, placed in the multiple-position sample holder, and sealed with Kapton tape. The contribution from scattering on the Kapton tape was subtracted from all data. To test anisotropy the SAXS patterns were analyzed with Matlab software, the azimuthal angle SAXS pattern was analyzed for different azimuthal scattering angles f. 2.2.7. Mechanical testing Uniaxial compression tests were performed on all gels using an Instron mechanical test frame (model 5565A). At least three repeats were done for each sample. Each gel was carefully removed from the beaker, after which the membrane was removed using a razor blade and cylinders were stamped out from the top part of the gel. Each gel was carefully examined for any cracks or deformation resulting from handling prior to testing. The gels were aligned in the center of stainless steel compression plates. The gels were slippery, allowing for free expansion of the gels when compressed. Stress relaxation of the samples was studied at compressions of 5, 10, and 20% strain using an initial crosshead speed of 4% strain/ second. The stress response upon relaxation of the gel was studied for up to 300 s. The stress was calculated from the force curve by s ¼ AF0 , with F and A0 being the force used to compress the sample and the initial area of the sample. The modulus of the gels was calculated via GðtÞ ¼ sðtÞ g . 3. Results and discussion

213

alignment of the actin fibers, as well as nuclei elongation, along the direction of the visible capillary. This means that the cells have sufficient contact guidance cues from the capillaries to be able to elongate and potentially self-organize into longer muscle fibers at later stages, similar to observations made by Zhao and co-workers (Zhao et al., 2009) on microfabricated surfaces. However, even though the cells are elongated, are present at high density, and are cultured with differentiation media, they do not stain positive for myosin heavy chains, a typical marker for skeletal muscle cell differentiation (Gang et al., 2004). 3.2. Microstructural anisotropy is reflected in functional properties of the gels Capillary alginate gels exhibit mechanical and mass transport properties that differ from the properties of their internally set counterparts (Schuster et al., 2014). A microstructural investigation of the network nano-structure via TEM did not provide insights into whether those differences in functionality are linked to the differences in the bulk network structure or to the presence of capillaries and the highly ordered network structure observed in the close vicinity of the capillaries (Schuster et al., 2016). To address the structure-function relationship of the capillary alginate gels, we prepared directed externally set alginate gels without capillaries, the only difference being that the calcium concentration used was too low to give rise to capillary formation (Caccavo et al., 2016). In such way, we reduced the complexity of the system, allowing for easier structure-function correlation. SAXS, probe diffusion and stress relaxation measurements were performed on such gels and compared with the measurements for internally set gels. 3.2.1. Characterisation of the alginate network by SAXS Scattering data were obtained as described from internally and

Table 1 Nomenclature, composition, resulting R value and gelation method for the samples studied using SAXS, FRAP and mechanical testing where cP refers to polymer concentration (%w/w).

3.1. Elongation of muscle cells on open capillary alginate gels

Sample

cP

R

[CaCl2]

gelling method

capillaries

The suitability of RGD-functionalized capillary alginate gels as scaffolds for C2C12 skeletal muscles cells was explored by cutting the alginate gels in thin disks along the capillary axis to expose open capillaries to the cells, as illustrated in Fig. 1a. Fig. 1b shows a large number of cells (actin filaments in red and nuclei in blue) within an open capillary after 8 days of culture. There is a weak

IR1 IR3 E0:04 E0:1 E0:2 sample frap sample frap sample stress rel

1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5

1 3 1 1 3 1 1 1

e e 0.04 0.1 0.2 e 0.04 0.04

internal internal external external external internal external external

no no no no yes no no no

Fig. 1. a) Schematic of the simplified cell culture model setup with the cut open alginate capillaries and the cells on top (not drawn to scale). b) Micrograph of the cells (red ¼ Actin filaments and blue ¼ cell nuclei) cultured for 1 day in GM and 7 days in DM. The cells are located and aligned in an open capillary (the capillary is not cut in the center, thus the diameter appears smaller). c) Schematic of how the capillary alginate can used as a scaffold material for cultured meat purposes.

214

E. Schuster et al. / Food Hydrocolloids 72 (2017) 210e218

directed externally gelled samples (Table 1). To test anisotropy in the scattering experiments an X-ray beam is aligned to hit crosssections of the gel cut either perpendicular or parallel to the inflow direction of the calcium as illustrated in Fig. 2a. In a second step, the scattering pattern is evaluated at different azimuthal scattering angles f, as depicted in Fig. 2b. Fig. 3 shows the scattering profiles of internally and directed externally set alginate gels. The insets in (a and b) depict the crosssectional Guinier plots and the cross-sectional radius of gyration, Rc , of the scattering entities, which is accessible via the fitting of the following equation into the Guinier regime q < 1=Rc (0:2 < q2 < 0:5) (Kratky & Glatter, 1982):

  1 qIðqÞyIð0Þexp  R2c q2 2

(2)

The Rc reveals information about the average network bundle sizes, assuming rigid cylinders whose radius is much less than the cylinder length. For the internally set samples IR1 and IR3 , this yielded a radius Rc ¼ 1:62±0:01 nm and Rc ¼ 2:14±0:01 nm, respectively as shown in Table 2. The cross-sectional radii obtained for the two samples indicates that increasing the calcium concentration increases the network strand thickness, in agreement with data for internally set alginate gel (Stokke et al., 2000). The externally set gels shows the same trend, i.e., increasing R from 1 to 3 leads to an increase in Rc (Table 2). Using a higher calcium concentration of the ion solution that was used to gel the alginate solution (but where R remained at 1) did not change the Rc as demonstrated for sample E0:04 and E0:1 . Furthermore, all external gels showed a higher Rc compared to the internally set gels, e.g., 1.62 ± 0.01 nm for internally set gel and 2.10 ± 0.01 nm for external gels, where both types of gels were of R ¼ 1. The same is seen for gels of R ¼ 3 where Rc obtained for internally set gel was 2.14 ± 0.01 nm against 2.23 ± 0.01 nm for the externally set gel. An alternative representation of the data is shown in Fig. 3c, via the so-called Kratky plot. It confirms the same trend of increasing bundle sizes with increasing calcium concentration. The scattering profile can be calculated from a molecular model by arranging the polymer chains in egg-box junction zones of varying degrees of aggregation (Stokke et al., 2000). The model predicts a distinct peak in the Kratky plot at the q-value, which corresponds to the size of the dominating strand thickness. For increasing calcium

concentration, a shift of the peak towards lower q-values, i.e., larger bundles, is observed. Assuming that a sugar ring is approximately 0.5 nm in diameter, we can further estimate that all investigated samples are found to consist of network bundles consisting of at least six aggregated network strands. The comparison of the scattering profiles/cross sectional radii of gyration of internally and externally set alginate gels revealed differences in the bundle size, but no striking differences in the overall SAXS pattern, when recorded with X-rays on a perpendicular cut cross-section (Fig. 4b and c), was observed, which corresponds to the very similar microstructures observed via TEM (Schuster et al., 2014). However, a quick visual inspection of the SAXS pattern of an externally set gel with X-rays on a parallel-cut cross-section (Fig. 4a) reveals anisotropy, as reflected by the ellipsoidal shape of the recorded SAXS pattern. Radial integration of the SAXS pattern in order to gain the classical scattering profile IðqÞ will not suffice in this case. The artifacts of this integration for the externally set gels are illustrated in Fig. 5, where the case of “X-rays parallel” shows a non-smooth Kratky plot at around q ¼ 0:75nm1 . We therefore investigate the scattering intensities at different azimuthal scattering angles, as introduced in Fig. 2. Fig. 6 displays the anisotropy clearly, where the scattering profile for the gel E0:2 “X-rays parallel” for three azimuthal scattering angles is plotted, and in particular the decay for one specific q-value is visualized. The highest magnitude peaks in the Kratky plots are found for 0 , and the lowest magnitude for 90 azimuthal angle. This angle dependence indicates that gel structures investigated at 0 show the most complex/compact structure, a branched mass fractal. Gel structures investigated at 90 still follow the pattern of a mass fractal with a distinct peak in the Kratky plot that is, however, less pronounced, indicating a larger spread in the alginate bundle sizes. This effect of anisotropy in the bundle sizes is additionally represented when looking at Rc for the azimuthal angles, as described above. Guinier analysis yields Rc ¼ 2:14±0:01 nm for f ¼ 0 , Rc ¼ 2:19±0:01 nm for f ¼ 45 , and Rc ¼ 2:27±0:01 for f ¼ 90 . It is worth to mention that capillary alginate gels prepared for muscle cell growth, i.e., RGD - functionalized alginate, were also analyzed using SAXS and showed the same anisotropic feature as discussed above.

Fig. 2. Schematic drawing of the alginate gel and of the SAXS evaluation. (a) Sketch of the external gelation in a cylindrical beaker, indicating the gelation front and the resulting shape and direction of aligned, parallel capillaries. SAXS experiments are carried out on two cross sections of the gel: either perpendicular or parallel to the growth direction of the capillaries. (b) SAXS pattern of a typical measurement; introducing the azimuthal scattering angle f. The scattering intensity is recorded on several charge-coupled device (CCD) blocks; the distinct stripes indicate the boundaries between neighboring CCD blocks, and the square in the middle reflects the beam stopper.

E. Schuster et al. / Food Hydrocolloids 72 (2017) 210e218

215

Fig. 3. SAXS scattering profiles IðqÞ, cross-sectional Guinier plots (insets), and Kratky plots of internally and externally set gels. (a) IðqÞ of the internally set alginate gels IR1 (,) and IR3 (B). (b) IðqÞ of externally set alginate gels E0:025 (△), E0:04 (▽), E0:1 (>), E0:2 (+). The scattering experiments were carried out on gels cut perpendicular to the growth direction of the capillaries (label perpendicular in Fig. 2). (c) Kratky plots of the data on the externally and internally set gels presented in (a) and (b).

Table 2 Rc for internally and directed externally set calcium alginate gels at different R values. Sample

R

internally set

IR1 IR3 E0:04 E0:1 E0:2

1 3 1 1 3

yes yes

externally set

Rc /nm

yes yes yes

1.62 2.14 2.10 2.10 2.23

± ± ± ± ±

0.01 0.01 0.01 0.01 0.01

3.2.2. Determination of dextran permeability in directed externally set gels using FRAP The normalized diffusion coefficient of differently sized (10 and 70 kDa) fluorescently labeled dextran is less hindered in the

capillary alginate gels than in the internally set alginate gels (Schuster et al., 2014). Directed externally set alginate gels, as studied here, show the same behavior as observed for the capillary alginate gels. Fluorescently labeled dextran of 10 and 70 kDa is less hindered to diffuse in directed externally set alginate gels than in internally set gels (Fig. 7). However, the 500 kDa dextran is hindered in both types of gel, as was the case also for the capillary alginate gels. The data suggests that it is neither the capillaries nor the highly directed microstructure, as found in the close vicinity of the capillaries (Schuster et al., 2016), that reduces the ability of the alginate gels to hinder small and medium-sized dextran rather, it is a feature of the bulk gels, something that could not be concluded in previous study performed by Schuster and co-workers (Schuster et al., 2014).

Fig. 4. Anisotropic and isotropic SAXS pattern of (a) externally set gel E0:2 with the gel cut to investigate with X-rays parallel, (b) externally set gel E0:2 , gel cut to investigate with Xrays perpendicular and (c) internally set gel IR3 . The same anisotropic pattern for externally set gels shown in (a) and (b) is also found for samples E004 and E01.

Fig. 5. (a) Scattering profiles IðqÞ and (b) Kratky plots of internally set gel IR3 (B), externally set E0:2 perpendicular cut (open stars), and E0:2 parallel cut (full stars).

216

E. Schuster et al. / Food Hydrocolloids 72 (2017) 210e218

Fig. 6. (a) Scattering profiles IðqÞ, (b) anisotropic scattering intensity at q ¼ 0:25nm1 and (c) Kratky plots of E0:2 parallel-cut externally set alginate gel. Displayed are profiles for three azimuthal angles f: (open) 0 , (half) 45 , (full) 90 .

Fig. 7. FRAP experiments on diffusion probes with FITC dextran of 10, 70 and 500 kDa molecular weight in alginate gels. The normalized diffusion coefficient D=D0 is plotted for externally set gels (diamond) using 0.04 M CaCl2 and for an internally set gel (sphere), both gel types having an R-value of 1.

3.2.3. Stress relaxation behavior The stress relaxation curves of the internally set gel (Fig. 8a) show similar relaxation behavior as to that reported by Mancini and co-workers (Mancini, Moresi, & Rancini, 1999) and by Schuster and co-workers (Schuster et al., 2014). The figure further shows that directed externally set gels have faster relaxation than the internally set gels (both gels tested exhibit R-values of 1) and decay at a rapid rate to a compression force of zero, thus exhibiting pronounced plastic behavior. The rapid decay of force requiring to keep the gel at a specific strain is similar to the behavior of the capillary gels (Schuster et al., 2014), why we attribute the plasticity observed in the types of gels studied here (externally directed alginate gels) as well as capillary alginate gels to the bulk network, and not, to the presence of capillaries or the presence of the highly oriented polymer network observed in the vicinity of each capillary. The stress required to deform the directed externally set alginate gels in the perpendicular plane is substantially higher than the stress observed for internally set gels (Fig. 8). Once deformed, the axial force required to retain the gel at 10% deformation is reduced by 80% within the first 10 min, indicating large reshaping of the network. When the parallel plane is compressed, less force is required to deform the gel than for the perpendicular

Fig. 8. Stress relaxation at 10% strain of internally set alginate gel (black line) and externally set directed alginate gel cut perpendicular (red line) and parallel (grey line). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

plane and even faster relaxation is observed. The corresponding moduli of the gels (GðtÞ ¼ sðtÞ g ) range from 10 to 90 kPa in its unrelaxed and relaxed state. The different response to compression exerted by the parallel- or perpendicular-cut gel cylinder correlates well with the different network structures revealed by SAXS and discussed above. The structural investigation done in this study clearly shows that the functional behavior (mechanical and ability of the network to reduce molecular transport) of the capillary alginate gel is related to its bulk network, and that this network is anisotropic in the parallel plane while being isotropic and similar as to internally set gels, in the plane perpendicular to the inflow of calcium ions. Furthermore, this means that the functional properties of the gel is not related to the presence of capillaries per se nor the presence of the aligned network structures observed in the close vicinity of the capillaries. Analysis of the stress relaxation data of the alginate gels, using a poroviscoelastic model, as done for bacterial cellulose gels (LopezSanchez et al., 2014), can provide further information on the mechanisms governing the mechanical properties, especially as the permeability of the externally set gels differs, as shown by the

E. Schuster et al. / Food Hydrocolloids 72 (2017) 210e218

increased permeability of dextran from that of the internally set ones. Furthermore, it is worthwhile to note that when compressed to a large extent, the directed externally set gels do not reform. The mechanical behavior of the directed externally set gels resembles the behavior reported of bacterial cellulose gels, as these show little or no expansion in a radial direction upon compression, a Poisson ratio of close to 0.01 or negative, and they do not recover their initial thickness when compressed (Lopez-Sanchez et al., 2014). Bacterial cellulose gels are built up from different sheets owing to their production using bacteria and the described mechanical properties of bacterial cellulose is believed to be related to weakness or lack of vertical support between the sheets (Lopez-Sanchez et al., 2014). A sheet-like structure of the gel would further explain the increased diffusion of labeled dextran up to a size of 70 kDa in the externally set alginate gels. It could be speculated that the area between the sheets is too narrow to allow increased diffusion of the largest dextran (500 kDa) tested here. Routes towards controlling the mechanical properties of the externally set alginate gels should thus involve strengthening of the forces interlocking the stacked alginate sheets, however, more studies are required to prove such hypothesis. The study shows that capillary alginate gels have enough contact cues to elongate muscle cells, and has via its feature of spontaneously formed capillaries ability to act as scalable scaffolds for muscle cell production. However, we observe lack of differentiation of the cells despite the modulus of the gel being in the same order of magnitude as muscle tissues (Discher, Janmey, & Wang, 2005). The reasons for the lack of differentiation of the cells need future investigation and includes rheological characterisation of the gels over time and in the nutrient solution used for the cells as well as improved characterisation of the surface chemistry of the capillary wall. Overcoming the lack of muscle cell differentiation, it is foreseen that a system using capillary alginate gels could be employed in cultured meat production, the capillary gels can be placed in a bioreactor that allows for cell-cultured media to be slowly pumped through the capillaries to constantly replenish nutrients and remove waste products. The cells are grown on the walls of the capillaries, where they can align and differentiate into long muscle fibers while allowing media to flow through the center of the channels. For cultured meat applications, there are two potential ways how the final cell culture product could be used. First, the gel could be degraded, using a calcium chelating agent, to harvest the muscle fibers and process them further. The second alternative is to use the final product as it is and process the muscle fibers together with the alginate gel scaffold. Alginate is edible and available from non-animal sources (Strom et al., 2009), it provides therefore an interesting opportunity to produce food products that combine cultured meat with plant-based materials in a synergistic process. 4. Conclusions Capillary alginate gels are an interesting candidate for cultured meat production systems, as they are able to provide contact guidance to the cells. The atypical behavior of capillary alginate gels (plasticity and reduced hindrance to the transport of dextran molecules) is governed by the properties of the bulk structure and not the presence of capillaries or the network structure in the close vicinity of the capillaries per se. The network obtained in the slice referred to as perpendicular (perpendicular to the inflow of CaCl2) are similar in nano-structure to the internally set alginate gels, with alginate bundles of at least six alginate chains and with an isotropic scattering profile. Analysing the network of slices referred to as parallel (parallel to the inflow of CaCl2), show anisotropic scattering profiles where the

217

radius of strand aggregation depends on the scattering angle. The mechanical properties of the gel show strong correlation to the network structure observed by SAXS where a perpendicular-cut gel has a higher non-relaxed modulus than a parallel-cut gel has. Furthermore, the relaxation behavior of samples compressed in either direction is rapid, and substantial rearrangement of the network or fluid dispersion occurs upon constant pressure. Further analysis of the relaxation behavior using the poroviscoelastic model may shed light on the importance of the network versus the fluid dispersion for the reported mechanical properties the gels. Acknowledgement The financial contribution from the VINN Excellence Centers SuMo Biomaterials and from Vinnova's Vinnmer program is acknowledged. We also thank the MAX IV Laboratory for the use of the MAX II SAXS beamline I911-4 and Ana Labrador at the MAX IV Laboratory for her excellent support. References Bhat, Z. F., & Fayaz, H. (2011). Prospectus of cultured meatdadvancing meat alternatives. Journal of Food Science and Technology, 48(2), 125e140. € m, A., Larsson, A., & Lamberti, G. (2016). Modeling capillary forCaccavo, D., Stro mation in calcium and copper alginate gels. Materials Science and Engineering: C, 58, 442e449. Chaudhuri, O., Gu, L., Klumpers, D., Darnell, M., Bencherif, S. A., Weaver, J. C., et al. (2016). Hydrogels with tunable stress relaxation regulate stem cell fate and activity. Nature Materials, 15(3), 326e334. Datar, I., & Betti, M. (2010). Possibilities for an in vitro meat production system. Innovative Food Science & Emerging Technologies, 11(1), 13e22. Discher, D. E., Janmey, P., & Wang, Y.-l. (2005). Tissue cells feel and respond to the stiffness of their substrate. Science, 310(5751), 1139e1143. Draget, K. (2009). Alginates. Woodhead Publishing Ltd. Draget, K. I., Gåserød, O., Aune, I., Andersen, P. O., Storbakken, B., Stokke, B. T., et al. (2001). Effects of molecular weight and elastic segment flexibility on syneresis in ca-alginate gels. Food Hydrocolloids, 15(4), 485e490. Draget, K., & Taylor, C. (2011). Chemical, physical and biological properties of alginates and their biomedical implications. Food Hydrocolloids, 25(2), 251e256. cited By 87. Edelman, P., McFarland, D., Mironov, V., & Matheny, J. (2005). Commentary: In vitrocultured meat production. Tissue Engineering, 11(5e6), 659e662. €nnemann, C. G., Sweeney, H. L., & Discher, D. E. Engler, A. J., Griffin, M. A., Sen, S., Bo (2004). Myotubes differentiate optimally on substrates with tissue-like stiffness pathological implications for soft or stiff microenvironments. The Journal of Cell Biology, 166(6), 877e887. Fitzgerald, M. M., Bootsma, K., Berberich, J. A., & Sparks, J. L. (2015). Tunable stress relaxation behavior of an alginate-polyacrylamide hydrogel: Comparison with muscle tissue. Biomacromolecules, 16(5), 1497e1505. pMID: 25812913. Gang, E. J., Jeong, J. A., Hong, S. H., Hwang, S. H., Kim, S. W., Yang, I. H., et al. (2004). Skeletal myogenic differentiation of mesenchymal stem cells isolated from human umbilical cord blood. Stem Cells, 22(4), 617e624. €m, A. (2016). Impact of Hermansson, E., Schuster, E., Lindgren, L., Altsk€ ar, A., & Stro solvent quality on the network strength and structure of alginate gels. Carbohydrate Polymers, 144, 289e296. Jonasson, J. K., Loren, N., Olofsson, P., Nyden, M., & Rudemo, M. (Nov. 2008). A pixelbased likelihood framework for analysis of fluorescence recovery after photobleaching data. Journal of Microscopie, 232(2), 260e269. Kadim, I. T., Mahgoub, O., Baqir, S., Faye, B., & Purchas, R. (2015). Cultured meat from muscle stem cells: A review of challenges and prospects. Journal of Integrative Agriculture, 14(2), 222e233. Kratky, O., & Glatter, O. (1982). Small angle x-ray scattering. London; New York: Academic Press. Lopez-Sanchez, P., Rincon, M., Wang, D., Brulhart, S., Stokes, J., & Gidley, M. (2014). Micromechanics and poroelasticity of hydrated cellulose networks. Biomacromolecules, 15(6), 2274e2284. Maki, Y., Ito, K., Hosoya, N., Yoneyama, C., Furusawa, K., Yamamoto, T., et al. (2011). Anisotropic structure of calcium-induced alginate gels by optical and smallangle x-ray scattering measurements. Biomacromolecules, 12(6), 2145e2152. pMID: 21504159. Mancini, M., Moresi, M., & Rancini, R. (1999). Uniaxial compression and stress relaxation tests on alginate gels. Journal of Texture Studies, 30(6), 639e657. Mattick, C. S., Landis, A. E., Allenby, B. R., & Genovese, N. J. (2015). Anticipatory life cycle analysis of in vitro biomass cultivation for cultured meat production in the United States. Environmental Science & Technology, 49(19), 11941e11949. pMID: 26383898. Mitchell, J., & Blanshard, J. (1976). Rheological properties of alginate gels. Journal of Texture Studies, 7(2), 219e234.

218

E. Schuster et al. / Food Hydrocolloids 72 (2017) 210e218

Post, M. J. (2012). Cultured meat from stem cells: Challenges and prospects. Meat Science, 92(3), 297e301. n, N., Altsk€ Schuster, E., Eckardt, J., Hermansson, A.-M., Larsson, A., Lore ar, A., et al. (2014). Microstructural, mechanical and mass transport properties of isotropic and capillary alginate gels. Soft Matter, 10(2), 357e366. € m, A., Altska €r, A., Smisdom, N., Geba €ck, T., et al. (2016). Schuster, E., Sott, K., Stro Interplay between flow and diffusion in capillary alginate hydrogels. Soft Matter, 12(17), 3897e3907. Stokke, B. T., Draget, K. I., Smidsrød, O., Yuguchi, Y., Urakawa, H., & Kajiwara, K. (2000). Small-angle x-ray scattering and rheological characterization of alginate gels. 1. ca-alginate gels. Macromolecules, 33(5), 1853e1863. Stoppel, W. L., White, J. C., Horava, S. D., Henry, A. C., Roberts, S. C., & Bhatia, S. R. (2014). Terminal sterilization of alginate hydrogels: Efficacy and impact on mechanical properties. Journal of Biomedical Materials Research Part B: Applied Biomaterials, 102(4), 877e884. Strom, A., Boers, H. M., Koppert, R., Melnikov, S. M., Wiseman, S., & Peters, H. P. F. (2009). Physico-chemical properties of hydrocolloids determine their appetite effects. In Gums and stabilisers for the food industry 15 (pp. 341e355). The Royal Society of Chemistry.

€m, A., & Williams, M. (2003). Controlled calcium release in the absence and Stro presence of an ion-binding polymer. Journal of Physical Chemistry B, 107(40), 10995e10999. cited By 21. Thiele, H., & Hallich, K. (1958). Uber ionotrope gele mit kapillarstruktur. Zeitschrift Fur Naturforschung Part B-chemie Biochemie Biophysik Biologie Und Verwandten Gebiete, 13(9), 580e588. Tuomisto, H. L., & Teixeira de Mattos, M. J. (2011). Environmental impacts of cultured meat production. Environmental Science & Technology, 45(14), 6117e6123. Verbeke, W., Sans, P., & Van Loo, E. J. (2015). Challenges and prospects for consumer acceptance of cultured meat. Journal of Integrative Agriculture, 14(2), 285e294. Wallin, P., Hglund, K., Wildt-Persson, K., & Gold, J. (2012). Skeletal myoblast differentiation on starch microspheres for the development of cultured meat. Journal of Tissue Engineering and Regenerative Medicine, 378 (abstract 62.02). Zhang, J., Daubert, C. R., & Foegeding, E. A. (2005). Fracture analysis of alginate gels. Journal of Food Science, 70(7), e425ee431. Zhao, Y., Zeng, H., Nam, J., & Agarwal, S. (2009). Fabrication of skeletal muscle constructs by topographic activation of cell alignment. Biotechnology and Bioengineering, 102(2), 624e631.