Microstructure of Aggregated and Nonaggregated κ-Carrageenan

Owen G. Jones , Stephan Handschin , Jozef Adamcik , Ludger Harnau , Sreenath Bolisetty , and Raffaele ..... Owen Griffith Jones , David Julian McCleme...
0 downloads 0 Views 412KB Size
Download PDF
Biomacromolecules 2001, 2, 1331-1337

1331

Microstructure of Aggregated and Nonaggregated K-Carrageenan Helices Visualized by Atomic Force Microscopy Shinya Ikeda,*,† Victor J. Morris,‡ and Katsuyoshi Nishinari† Department of Food and Nutrition, Osaka City University, Sumiyoshi, Osaka 558-8585, Japan, and Institute of Food Research, Norwich Research Park, Colney, Norwich NR4 7UA, U.K. Received September 19, 2001

Gelation of κ-carrageenan is believed to involve two steps: helix formation on cooling and a further specific cation (salt) induced side-by-side aggregation of helices. Samples that should contain aggregated and also nonaggregated “helices” of κ-carrageenan were prepared in aqueous solutions, spread onto freshly cleaved mica surfaces, and visualized under butanol using atomic force microscopy. In the presence of an excessive amount of a gel-promoting salt, KCl, κ-carrageenan appeared to form rigid rodlike structures considered as large aggregates of double helices. Even when the side-by-side interhelical aggregation was suppressed by diluting random coiled solutions prior to cooling, by adding an aggregation-impeding salt, NaI, or by transforming κ-carrageenan into the tetramethylammonium (TMA) salt, branched rodlike structures were still evident, suggesting that the side-by-side aggregation of helices is not a prerequisite for κ-carrageenan to form a network structure, at least locally. Even in the absence of factors that promote side-by-side aggregation, κ-carrageenan helices appeared to be capable of associating and forming gel networks. Introduction κ-Carrageenan, composed ideally of a disaccharide repeating unit β(1f3)-D-galactose-4-sulfate and R(1f4)-3,6anhydro-D-galactose, is a member of a family of sulfated polysaccharides extracted from red marine algae. κ-Carrageenan is widely utilized in industry because it can form reasonably stiff and thermoreversible gels in the presence of so-called gel-promoting salts (e.g., KCl, RbCl, CsCl, etc.).1 In an aqueous medium, κ-carrageenan molecules are in the random coil form at sufficiently high temperature and convert into the double-helical conformation on cooling.1,2 These double helices are believed to undergo side-by-side aggregation into rigid rodlike structures at sufficiently low temperatures in the presence of gel-promoting salts, and a gel can be formed at a sufficiently high polymer concentration.1,3-5 Interhelical aggregation into bundles is believed to involve binding of specific cations and is usually evident from thermal hysteresis between the observed coil to helix and helix to coil transitions.1,6,7 Aggregated helices are more thermally stable so that they melt on heating (the helix-tocoil transition) at a temperature higher than the coil-to-helix transition temperature on cooling. Such thermal hysteresis between the coil-to-helix and aggregated helix-to-coil transitions is a common feature among gel-forming polysaccharides including κ-carrageenan, agarose, and gellan gum.8 In * To whom correspondence may be addressed at the Department of Food and Nutrition, Osaka City University, 3-3-138 Sugimoto, Sumiyoshi-ku, Osaka 558-8585, Japan. Phone: +81-6-6605-2862. Fax: +81-6-6605-3086. E-mail: [email protected]. † Osaka City University. ‡ Institute of Food Research.

the original domain model for carrageenan gelation,9,10 it has been considered that aggregated helices in these polysaccharide gels function as junction zones that are then linked by flexible “random coil” polymer chains. Thus the elasticity of the gels would arise from the entropic contribution of these flexible polymer chains on deformation of the gel.8 It is preferable to investigate the coil-to-helix transition and the side-by-side aggregation of helices independently in order to gain insight into the gelation mechanisms of polysaccharides. This approach, based on the use of nonaggregating bulky tetramethylammonium cations was used as a tool which helped in the development of the domain model of gelation.9,10 A second avenue for separating helix formation and aggregation was suggested by the observation that specific anions, such as iodide or thiocyanate, are known to prevent interhelical aggregation,11 thus allowing a stable dispersion of nonaggregating helices to be prepared in the presence of these specific anions. Viebke and co-workers have revealed that such a dispersion of nonaggregating κ-carrageenan helices can be reversibly transformed into a gel simply by dialyzing against a gel-promoting salt solution.3 On the basis of this observation, and other theoretical considerations, they have concluded that κ-carrageenan gel networks are formed by fibrously aggregated helices, not by double helices connected by flexible chains.3 In this model, branching within the gel network results from the aggregation process rather than the presence of helix-incompatible sugar residues that are proposed to promote intermolecular helix formation in the domain model. Early electron microscopy studies by Hermansson and co-workers4,5 also suggested a fibrous gel structure for carrageenan gels in the presence of

10.1021/bm015610l CCC: $20.00 © 2001 American Chemical Society Published on Web 10/30/2001

1332

Biomacromolecules, Vol. 2, No. 4, 2001

potassium cations. A similar fibrous gelation mechanism has been proposed for gellan gum based on atomic force microscopy (AFM) observations.12,13 AFM images of gellan gum networks and gels have failed to detect distinct junction zones or disordered flexible chains and have been taken to suggest that the entire gel network is composed of fibers of aggregated helices.12,13 However, in this model,12,13 the experimental data suggested that if side-by-side helix association is inhibited, then branching, and the formation of a fine network structure, can still arise simply due to helix formation and does not require side-by-side association of helices. A similar conclusion was reported4,5 for sodium κ-carrageenan networks, in which side-by-side association of helices should not occur, from electron microscopy studies. These fibrous models3-5,12,13 suggest that the elasticity of κ-carrageenan or gellan gum gels arises mainly from stretching and bending of fibrous strands. Some questions regarding the gelation mechanisms of helix-forming polysaccharides still remain unsolved. For example, the molecular weight of κ-carrageenan is not always exactly doubled in the event of the coil-to-helix transition.2,14,15 If only two molecules are involved in the formation of a double-helix, this should be always the case. Additionally, recent rheological studies have revealed that aqueous dispersions of nonaggregating κ-carrageenan helices exhibit weak-gel type rheological properties at a sufficiently high concentration (ca. >1% w/w).16-18 A weak-gel is a term used for designating a polymer dispersion that exhibits “gel-like” viscoelasticity against linearly small deformation but can steadily flow without breaking under large deformation.19 Several polysaccharide systems are known to possess weak-gel rheological properties, and those systems may be classified into three categories: polymer dispersions with a sufficiently long lifetime of entanglements,20,21 a three-dimensional weak network that can break under large deformation but can heal at rest,20,21 and dispersions of micrometer-sized insoluble particles that interact sterically with each other.22-26 Further experimental evidence is needed to determine into which category a κ-carrageenan weak gel falls because, for instance, if multiple polymer chains are involved in the formation of a doublehelix, branching could be realized without assuming “sideby-side” interhelical aggregation.12,13,15 Microscopy has an advantage over other physical characterization techniques such as light, small-angle X-ray or neutron scattering, osmotic pressure, or sedimentation methods in that microscopy can directly visualize heterogeneity in the sample, while other techniques give results spatially averaged over a number of polymers in the sample under examination.27,28 AFM has served as a useful microscopic tool to investigate individual molecules and their interactions, particularly in studies of large assemblies of polysaccharides.27-31 The use of AFM allows imaging of hydrated biopolymers and polymer networks even at a relatively high concentration sufficient to form hydrated films or gels.12,13 The goal of this study was to visualize aggregated and nonaggregated κ-carrageenans using AFM in order to examine the proposed gelation mechanisms3-5,12,13 of helixforming polysaccharides.

Ikeda et al.

Materials and Methods Materials and Sample Preparation. κ-Carrageenan, extracted from Eucheuma cottonii (Product no. C-1263, Lot 16H0616), was purchased from Sigma Chemicals (St Louis, MO) and used without further purification unless otherwise specified. Other chemicals were of reagent grade quality. Powdered κ-carrageenans were dispersed in distilled water containing either 0.1 mol/dm3 KCl or 0.1 mol/dm3 NaI to a concentration of ca. 100 µg/mL and heated at 85 °C for an hour to ensure complete dissolution. A portion of hot κ-carrageenan solution in 0.1 mol/dm3 KCl was immediately diluted into distilled water at room temperature to a concentration of 10 µg/mL (0.01 mol/dm3 KCl). The rest of the solutions were gradually cooled to room temperature, left at room temperature at least for an hour, and diluted to 10 µg/mL (0.01 mol/dm3 KCl or NaI). Aliquots (2 µL) of the diluted sample solutions were immediately spread onto freshly cleaved mica and then imaged by AFM. The period of time between preparation and imaging was generally less than 10 min. Tetramethylammonium (TMA) κ-carrageenan was prepared using the following procedure. Powdered κ-carrageenan was dispersed in distilled water (ca. 0.3% w/w), dissolved by stirring at 95 °C for an hour, and gradually cooled to room temperature. The solution was further cooled in ice water to reduce the risk of acid degradation of the polymer and then run through a column of cation-exchange resin beads (Dowex 50W) in the hydrogen form. The temperature of the column was maintained below 10 °C. The eluted solution was kept in an ice water bath, titrated to neutral pH by addition of TMA hydroxide, concentrated by rotary evaporation at 60 °C, and then freeze-dried. Powdered TMA-κ-carrageenan was dispersed in distilled water to a concentration of ca. 100 µg/mL, heated at 85 °C for an hour, gradually cooled, left at room temperature for an hour, and diluted to 10 µg/mL. Aliquots (2 µL) of the diluted sample solutions were immediately spread onto freshly cleaved mica, and then imaged by AFM. Atomic Force Microscopy. An AFM designed and manufactured by East Coast Scientific (Cambridge, U.K.) was used for imaging with short narrow Nanoprobe cantilevers with a quoted force constant of 0.38 N/m (Digital Instruments, Santa Barbara, CA). Direct current (dc) contact mode was employed in the present study: topographical images were obtained at constant cantilever deflection31 and these topographical images were complemented by their equivalent error signal mode images. The samples spread on freshly cleaved mica were imaged under butanol. Samples spread on mica and examined in air remain hydrated and, under normal humidity conditions, a thin layer of water will also be present on the end of the AFM tip. When the tip is brought “into contact” with the sample, these hydration layers coalesce, resulting in strong adhesive forces which can damage or displace molecules when the samples are scanned in the AFM. These adhesive forces need to be eliminated in order to allow reproducible imaging.31 In the present studies, imaging under butanol was

Microstructure of κ-Carrageenan Helices

Biomacromolecules, Vol. 2, No. 4, 2001 1333

used in order to avoid water condensation between the sample surface and the tip. Butanol is one of the most favorable liquids used to improve imaging since it is a poor solvent for polysaccharides and thus retains the polymers on the mica surface during measurements and satisfactorily eliminates the undesirable capillary-induced adhesive forces between the sample surface and the cantilever tip.31,32 In the absence of capillary-induced adhesive forces, it is possible to adjust the imaging force in order to optimize contrast in the image. Results and Discussion Direct information on polysaccharide association, and the nature of gel networks, has come largely from electron microscopy 1,4,5,17,33-35 and some AFM studies.12,13,32 In both techniques, imaging requires depositing the samples onto a flat substrate. Aqueous samples can be spread onto mica or carbon substrates. For hydrophobic substrates such as carbon, surfactants have been used to aid spreading of the sample.35 Electron microscopy involves imaging the samples under a vacuum, and several procedures have been developed to preserve the nature of the biological specimen. These include shadowing and the production of replicas, vitrification for cryogenic temperature transmission electron microscopy (cryo-TEM), or rapid freezing of layers within a mica sandwich followed by sublimation at low temperatures in a vacuum. Samples for imaging by AFM are normally spread on freshly cleaved mica. The images can be acquired directly in an aqueous or buffered medium. Such studies on polysaccharides36 have been used to observe polysaccharides in motion during desorption from mica surfaces. The motion of the polysaccharides, which is rapid compared to the image acquisition times, is inconvenient and it is preferable to inhibit such motion by imaging the hydrated samples on mica in air using tapping conditions, or under liquids such as butanol.31 Such AFM studies on individual molecules have been validated against data obtained directly under buffers, TEM studies, and cryo-AFM, itself directly comparable with cryo-TEM investigations.31 Furthermore, images of polysaccharide aggregates have been validated against AFM images of hydrated films formed on mica and, under favorable conditions, direct images of polysaccharide networks in hydrated gels.12,13,31 Since the κ-carrageenan to be examined normally formed gels even at 0.05% w/w (ca. 500 µg/mL) in 0.1 mol/dm3 KCl at room temperature, κ-carrageenan helix formation was initiated at a slightly lower concentration (ca. 0.01% w/w), where no recognizable macroscopic gel formation was observed at room temperature, to capture gel precursors or microgels (local networks or aggregates) that were produced in the sol and then to sample them by collecting them onto mica surfaces. Although the adopted sample preparation procedure is reasonably mild, it may influence the images obtained. Three-dimensional structures in a sol will be flattened into two-dimensional objects. Furthermore, polymers and polymer aggregates may orientate, flow, or extend during the spreading of the sample on the substrate. Spread-

Figure 1. (a) Topographical image of κ-carrageenan helices formed on diluting a hot solution (85 °C) containing KCl. Image size is 2 µm × 2 µm. Arrows indicate crossover points of fibers. (b) Equivalent error signal mode image.

ing will lead to a large surface area and some drying of the polysaccharide film. Despite a very low starting polymer concentration, that is definitely below the critical concentration for macroscopic gelation, spreading 10 µg/mL polysaccharide solutions onto mica leads to the formation of network structures even without added salts.12,13,25,30-32 The aim of the present study is to examine the nature of such networks and their precursors. Preliminary AFM studies of κ-carrageenan network structures revealed orientation of the strands of the network,32 suggesting that κ-carrageenan networks are aligned and extended during film formation. Figure 1 shows results on κ-carrageenan that has been diluted as a hot sol, deposited onto mica, and allowed to cool to room temperature. Cooling promotes helix formation. The dilution step should minimize side-by-side helix aggregation. These κ-carrageenan samples in the presence of KCl showed oriented networks similar to previous AFM images of κ-carrageenan networks32 (Figure 1). The network is fairly continuous and is composed of occasionally branched fibers of which ends are seldom seen. Sporadically, a few fibers lie one upon another, while the width and height of the fibers are fairly uniform. Fibers that cross over one another lead to a doubling of the height and are characterized by bright dots at the crossover points. Examples of such crossover points are indicated by arrows in the Figure 1. The heights of the fibers are typically between 1 and 2 nm suggesting that the fibers are likely to be carrageenan helices. The observed widths are almost an order of magnitude larger than the measured heights. This is due to probe broadening

1334

Biomacromolecules, Vol. 2, No. 4, 2001

Ikeda et al.

Figure 2. (a) Topographical image of local networks of κ-carrageenan helices formed in the presence of KCl at 20 °C. Image size is 2 µm × 2 µm. (b) Equivalent error signal mode image.

Figure 3. (a) Topographical image of superhelical aggregates of κ-carrageenan formed in the presence of KCl at 20 °C. Image size is 2 µm × 2 µm. (b) Equivalent error signal mode image.

effects, the extent of broadening being dependent on the shape and curvature of both the probe tip and the molecules.31 The measured heights provide a more realistic indication of molecular dimensions, although the heights are generally less than the expected molecular diameter due to some degree of compression of the molecules during imaging. The structures shown are typical of those found on the mica surface, and there was no indication of the existence of large aggregates. Figures 2 and 3 show AFM images of κ-carrageenan solutions containing KCl which were equilibrated at room temperature before being spread on the mica. Localized aggregation was observed in the images which appeared to be quite different from those of the samples seen in Figure 1, or previously reported AFM images of κ-carrageenan networks.32 Two types of structures were reproducibly observed. The first consists of localized networks composed of rather tenuous strands showing some degree of branching, and with observable ends (Figure 2). The width and heights of these fibers varies, suggesting that these fibers may involve some degree of side-by-side aggregation of helices. Measured heights ranged between 1 and 3 nm suggesting only a low level of side-by-side association of helices. Absence of preferential orientation also supports the idea that the network strands were composed of aggregated helices that were rigid enough to resist being extended and, hence, oriented when spread on mica. Another type of local aggregate observed is believed to be bundles of intensively aggregated helices (Figure 3). The bundles, for which width and height (measured values up to 6 nm) vary greatly, exhibit branches

and kinks. These rigid aggregated helices would be sufficient to form a macroscopic network at a sufficiently high polymer concentration without assuming distinct junction zones and disordered flexible chains. This supports the fibrous model of gelation proposed for carrageenan samples from electron microscopy1,3-5 and for gellan gum, based on AFM studies.12,13 Although it is not clear at this point whether the finer aggregated fibers (Figure 2) further transform into much thicker bundles (Figure 3), electron microscopic studies have shown that similar “superstrands” of aggregated κ-carrageenan helices are formed in the background of fine networks in the presence of KCl.4,5 The elasticity of polysaccharide gels has often been explained based on rubber elasticity theories, while polysaccharide gels are usually much more brittle than rubbers: the fracture strain of polysaccharide gels is often less than unity,18,37-39 while even the linear viscoelastic strain region of rubbers and chemically cross-linked polymer gels extends up to 2-3.40,41 The brittleness of polysaccharide gels may be attributed to the brittleness of aggregated helices that transmit the load on the entire system and concentrate it at the weakest “branched” regions of the network. Other attempts have been made toward understanding the dynamics of polysaccharide gel networks, based on the collective diffusion of the networks.39 Such perspectives, however, also need to be amended since the polysaccharide gel networks appear to be nothing like disordered single polymer chains that are cross-linked at pointlike junctions. The results in this study are in line with the gelation mechanism of κ-carrageenan proposed more recently, that

Microstructure of κ-Carrageenan Helices

Figure 4. (a) Topographical image of κ-carrageenan helices formed in the presence of NaI at 20 °C. Image size is 2 µm × 2 µm. Arrows indicate branching points of fibers. (b) Equivalent error signal mode image.

is, that the helices aggregate to form superhelical networks in the presence of gel-promoting salts.3-5 Iodide ions are believed to bind to κ-carrageenan double helices and are known to prevent their aggregation.11 Thus, in the presence of iodide ions, the molecular weight of κ-carrageenan is approximately doubled on the coil-helix transition;2 no thermal hysteresis between the coil-to-helix and helix-to-coil transitions is observable16,17 and no gel is formed at a moderately high concentration where a gel is formed in the presence of gel-promoting salts.16-18 Recently it has been found, however, that even nonaggregated κ-carrageenan helical systems stabilized in the presence of iodide ions can exhibit solidlike elasticity if the concentration is further increased.16-18 An interesting question is whether such a system is a dispersion of entangled individual helical rods or a network? Figure 4 shows nonaggregated κ-carrageenan helices formed in the presence of NaI. The strands are fairly uniform in width and height with visible ends, and familiar orientation is clearly recognizable, suggesting the absence of side-by-side aggregation of helices. This is also consistent with the measured heights that lie between 1 and 2 nm. There was no indication of the existence of large aggregates over the entire sample surfaces. Rather unexpectedly, however, many branches are seen in the images (examples arrowed in Figure 4). These results suggest that κ-carrageenan can form a (local) network even without side-by-side aggregation of helices, although such a network may not be so rigid as the one formed in the presence of aggregating salts. The results are consistent with the electron microscopy data on

Biomacromolecules, Vol. 2, No. 4, 2001 1335

sodium κ-carrageenan in which side-by-side association of helices should be suppressed. Therefore helix formation may be regarded as leading to the formation of a mixture of entangled linear and branched polymer assemblies. Since side-by-side aggregation of helices has been inhibited, the only mode of branch formation must, in the present case, be via the coil-helix transition.13 If only molecules with perfectly matching lengths associate into double helices, the molecular weight must be doubled and no branching can occur: an unlikely scenario for a normal polydisperse sample. Association of unequal length polymer chains during double helix formation would lead to elongated fibers and to occasional branching. For gellan gum, it has been established that the coil-helix transition in the absence of interhelical aggregation can lead to the formation of local networks.12,13,15 ι-Carrageenan, which has a disaccharide repeating unit similar to κ-carrageenan with an additional 2-sulfate on the anhydrogalactose residue, is also believed to form a network in the absence of interhelical aggregation, since ι-carrageenan can form a gel but the gel does not show thermal hysteresis between the setting and melting temperatures.1 The iodide ion is known to inhibit the normal gelation of κ-carrageenan even in the presence of equal amounts of potassium ions, one of the most effective gel-promoting cation species.11 Although an estimated molar ratio of the iodide to the potassium ion is over several hundred in the case that is shown in Figure 4, it may still be possible that even trace amounts of potassium ions pre-existing in the commercial κ-carrageenan sample may be also responsible for a limited extent of side-by-side aggregation leading to formation of the observed branches. Therefore, to eliminate this possibility and to remove any gel promoting cations that may be present in the commercial samples, κ-carrageenan samples were further purified and transformed into the TMA salt form. It is also known that TMA ions prevent the gelation of κ-carrageenan since they are too bulky to allow side-byside aggregation of κ-carrageenan helices.1,9,10 AFM images show that such nonaggregated TMA-κ-carrageenan helices exhibited fairly uniform strands with occasional branches and familiar orientation (panels a and b of Figure 5), which is consistent with the results of nonaggregating helices in the presence of NaI (Figure 4), and the data in Figure 1 where side-by-side association of helices has also been discouraged. The images in Figure 5a contain regions where individual strands cross one another. At such points (arrowed in Figure 5) the heights double, as indicated by the bright spot at the crossover point. This reinforces the conclusion that the individual strands are generally real branched structures: the heights of the backbone and branch are the same, and there is no increase in height at the branch points. There were a few areas on the mica where a continuous network covered the entire observation area (2 µm × 2 µm) (panels c and d of Figure 5). The images in Figure 5c contain numerous crossover points at which there are bright spots indicative of an increase in height. Such images could result from threedimensional microgels or gel particles which have collapsed and flattened on the mica surface. Since the examined κ-carrageenan is expected to have an average molecular

1336

Biomacromolecules, Vol. 2, No. 4, 2001

Ikeda et al.

Figure 5. (a) Topographical image of tetramethylammonium κ-carrageenan helices formed at 20 °C. Image size is 2 µm × 2 µm. Arrows indicate crossover points of fibers. (b) Error signal mode image equivalent to that in panel a. (c) Topographical image of tetramethylammonium κ-carrageenan networks formed at 20 °C. Image size is 2 µm × 2 µm. (d) Error signal mode image equivalent to that in panel c.

weight of ca. 300 000, the average contour length of an individual double helix would be at most several hundred nanometers.14 Thus, the formation of long fibers that span over a micrometer-scale area must be an outcome of endto-end association of molecules,15 supporting once again the hypothesis that multiple chains are involved in the formation of individual “unaggregated” double helices. These results confirm that the coil-helix transition produces a variety of structures including local networks, less-developed but branched fibers, and linear fibers with varied lengths. In the presence of gel-promoting cations, side-by-side aggregation of helices results in thicker fibrous aggregates and networks. Conclusion AFM has been shown to be a versatile method for investigating the gelation mechanisms of κ-carrageenan. By selection of an appropriate polymer concentration, ionic species, and concentration, gel precursors or microgels composed of κ-carrageenan helices were obtained. Our results support the previously proposed two-stage model of κ-carrageenan gelation:3-5 the first stage being the coil-helix transition leading to the formation of primary fibers and a second stage involving side-by-side aggregation of helices into thicker branched fibers that can percolate throughout the entire space. However, this latest gelation model for κ-carrageenan3-5 seems to require a slight modification, since occasional branches were found in κ-carrageenan systems in which side-by-side association of helices had been

suppressed, indicating that it is possible that the coil-helix transition itself can result in a weak and/or local network. Acknowledgment. The authors thank A. P. Gunning, A. R. Kirby, M. J. Ridout, P. A. Gunning, and A. Jay for their advice and assistance. S.I. is indebted to Osaka City University for supporting his stay at the IFR, Norwich, U.K. References and Notes (1) Piculell, L. In Food Polysaccharides and Their Applications; Stephen, A. M., Ed.; Marcel Dekker: New York, 1995; pp 205-244. (2) Viebke, C.; Borgstro¨m, J.; Piculell, L. Carbohydr. Polym. 1995, 27, 145-154. (3) Viebke, C.; Piculell, L.; Nilsson, S. Macromolecules 1994, 27, 41604166. (4) Hermansson, A.-M. Carbohydr. Polym. 1989, 10, 163-181. (5) Hermansson, A.-M.; Eriksson, E.; Jordansson, E. Carbohydr. Polym. 1991, 16, 297-320. (6) Nilsson, S.; Piculell, L. Macromolecules 1991, 24, 3804-3811. (7) Ikeda, S.; Kumagai, H. J. Agric. Food Chem. 1998, 46, 3687-3693. (8) Stephen, A. M., Ed.; Food Polysaccharides and Their Applications; Marcel Dekker: New York, 1995. (9) Morris, E. R.; Rees, D. A.; Robinson, G. R. J. Mol. Biol. 1980, 138, 349-362. (10) Robinson, G. R.; Morris, E. R.; Rees, D. A. J. Chem. Soc., Chem. Commun. 1980, 152. (11) Grasdalen, H.; Smidsro¨d, O. Macromolecules 1981, 14, 1842-1845. (12) Gunning, A. P.; Kirby, A. R.; Ridout, M. J.; Brownsey, G. J.; Morris, V. J. Macromolecules 1996, 29, 6791-6796; 1997, 30, 163-164. (13) Morris, V. J.; Kirby, A. R.; Gunning, A. P. Prog. Colloid Polym. Sci. 1999, 114, 102-108. (14) Slootmaekers, D.; De Jonghe, C.; Reynaers, H.; Varkevisser, F. A.; Bloys van Treslong, C. J. Int. J. Biol. Macromol. 1988, 10, 160168. (15) Gunning, A. P.; Morris, V. J. Int. J. Biol. Macromol. 1990, 12, 338341.

Microstructure of κ-Carrageenan Helices (16) Chronakis, I. S.; Piculell, L.; Borgstro¨m, J. Carbohydr. Polym. 1996, 31, 215-225. (17) Piculell, L.; Borgstro¨m, J.; Chronakis, I. S.; Quist, P.-O.; Viebke, C. Int. J. Biol. Macromol. 1997, 21, 1141-153. (18) Ikeda, S.; Nishinari, K. J. Agric. Food Chem. 2001, 49, 4436-4441. (19) Ross-Murphy, S. B. J. Rheol. 1995, 39, 1451-1463. (20) Morris, V. J. In Food Polysaccharides and Their Applications; Stephen, A. M., Ed.; Marcel Dekker: New York, 1995; pp 341375. (21) Morris, E. R. In Food Polysaccharides and Their Applications; Stephen, A. M., Ed.; Marcel Dekker: New York, 1995; pp 517546. (22) Hirashima, M.; Takaya, T.; Nishinari, K. Thermochim. Acta 1997, 306, 109-114. (23) Norton, I.; Foster, T. Brown, R. Gums Stab. Food Ind. 1998, 9, 2590.268. (24) Tatsumi, D.; Ishioka, S.; Matsumoto, T. Nihon Reoroji Gakkaishi 1999, 27, 243-248. (25) Morris, V. J.; Gunning, A. P.; Kirby, A. R.; Mackie, A. R.; Wilde, P. J. In HydrocolloidssPart 1; Nishinari, K., Ed.; Elsevier Science B.V.: Amsterdam, 2000; pp 99-109. (26) Morris, V. J.; Mackie A. R.; Wilde, P. J.; Kirby, A. R.; Mills, E. C. N.; Gunning, A. P. Lebensm.sWiss.sTechnol. 2001, 34, 3-10. (27) Round, A. N.; MacDougall, A. J.; Ring, S. G.; Morris, V. J. Carbohydr. Res. 1997, 303, 251-253. (28) Gunning, A. P.; Mackie, A. R.; Kirby, A. R.; Kroon, P.; Williamson, G.; Morris, V. J. Macromolecules 2000, 33, 5680-5685.

Biomacromolecules, Vol. 2, No. 4, 2001 1337 (29) Kirby, A. R.; Gunning, A. P.; Waldron, K. W.; Morris, V. J.; Ng, A. Biophys. J. 1996, 70, 1138-1143. (30) Gunning, A. P.; Cairns, P.; Kirby, A. R.; Round, A. N.; Bixler, H. J.; Morris, V. J. Carbohydr. Polym. 1998, 36, 67-72. (31) Morris, V. J.; Kirby, A. R.; Gunning, A. P. Atomic Force Microscopy for Biologists; Imperial College Press: London, 1999. (32) Kirby, A. R.; Gunning, A. P.; Morris, V. J. Biopolymers 1996, 38, 355-366. (33) Chronakis, I. S.; Borgstro¨m, J.; Picullel, P. Int. J. Biol. Macromol. 1999, 25, 317-328. (34) Borgstro¨m, J.; Picullel, P.; Viebke, C.; Talmon, Y. Int. J. Biol. Macromol. 1996, 18, 223-229. (35) Sugiyama, J.; Rochas, C.; Turquois, T.; Taravel, F.; Chanzy, H. Carbohydr. Polym. 1994, 23, 261-264. (36) Gunning, A. P.; Mackie, A. R.; Kirby, A. R.; Kroon, P.; Williamson, G.; Morris, V. J. Macromolecules 2000, 33, 5680-5685. (37) Kaletunc, G.; Normand, M. D.; Nussinovitch, A.; Peleg, M. Food Hydrocolloids 1991, 5, 237-247. (38) Truong, V. D.; Daubert, C. R. J. Texture Stud. 2000, 31, 335-353. (39) Nakamura, K.; Shinoda, E.; Tokita, M. Food Hydrocolloids 2001, 15, 247-252. (40) Foegeding, E. A.; Gonzalez, C.; Hamann, D. D.; Suzanne, C. Food Hydrocolloids 1994, 8, 125-134. (41) Walstra, P.; van Vliet, T. In Food Polymers, Gels and Colloids; Dickinson, E., Ed.; The Royal Society of Chemistry: Cambridge, 1991; pp 369-382.

BM015610L