The objective of this research was to determine whether chitosan microparticles could be developed as potential satiety enhancing ingredients by investigating their swelling behaviour in the gastro-intestinal tract using magnetic resonance imaging (MRI). Previous work undertaken using a covalently cross-linked composite of chitosan and bovine serum albumin (BSA) showed this system swells in gastric conditions in vitro (Butler, Clark, & Adams, 2006). This paper describes the preparation and characterisation of chitosan/BSA microparticles cross-linked using 25 mM and 100 mM glutaraldehyde (GDA). 1.5% chitosan/15% BSA microparticles cross-linked with 25 mM GDA swell by 100% in simulated gastric conditions. We found an experimental correlation between the transverse NMR relaxation time, T(2), and swelling of the microparticles in vitro. Our study of four volunteers showed swelling of these microparticles in the human stomach, observed from increases in T(2) of the microparticles using non-invasive MRI. As such, our initial results, with 4 volunteers, suggest that this system could be used as a potential satiety enhancing agent in vivo as we would expect it to provide gastric distension due to swelling at low pH. Further studies with more volunteers would be necessary to confirm our initial findings. (C) 2011 Elsevier Ltd. All rights reserved.

This paper describes an in vivo study of the use of encapsulation to alter lipid bio-accessibility in the gastro-intestinal (GI) tract. The hypothesis that encapsulation would delay the accessibility of lipid in the GI tract was tested using subjects (N = 11) who consumed either 30 g of 13C-labelled sunflower oil encapsulated within 70 g alginate gel particles, or unencapsulated ‘free’ sunflower oil and alginate-only particles (control). Lipid accessibility was determined by the appearance of the 13C label in breath carbon-dioxide (CO2), and the particles behaviour in the GI tract was investigated using non-invasive magnetic resonance imaging (MRI). Despite apparently faster gastric emptying, encapsulation of lipid resulted in a mean 47 min delay in the time to peak of the 13C label reaching the breath and a reduction in the initial slope of the 13C breath curve (both p < 0.05). Furthermore the gall bladder ejection fraction measured from MRI data was significantly smaller for the encapsulated lipid meal. We suggest this was due to delayed accessibility of the lipid for its digestion to the free fatty acids which are required to stimulate duodenal receptors and the release of cholecystokinin (CCK). This would account for both the smaller gall bladder response and faster gastric emptying observed with the encapsulated lipid meal. This study showed that encapsulation is an effective strategy to control the GI fate of lipids. It also showed that multimodal studies including MRI are powerful tools for investigating how the body handles food components and may aid in the design of food products with controlled functionality.

The relationship between NMR relaxometry and rheology of acid alginate gels has been investigated in vitro and compared to that of ionic alginate gels. The onset and early stages of the gelation process were characterized by the time dependent changes in the storage modulus G′ and transverse relaxation rate R2 of the water, demonstrating how this NMR parameter can be used as a non-invasive/non-destructive indicator of the formation of these gels. Compression tests on the final gels showed that the acid gels were much weaker. No direct correlation was observed between the final strength of these types of gels and the measured NMR R2 values. The results suggest that in MRI studies in vivo the R2 relaxation rate could provide valuable information on the rate of gelation (but not on the actual final gel strength) of alginate-containing foods in the human gastric lumen.

This paper summarises work undertaken to visualise and characterise alginate gel beads, in vivo, in the human gastro-intestinal tract using magnetic resonance imaging (MRI). Two types of alginate beads were investigated: "strong" (strongly gelled) and "weak" (weakly gelled) beads, each prepared by long and short exposures to a calcium chloride gelling bath, respectively. Ten volunteers attended the study centre, on two different occasions, and consumed a different bead type on each visit. Before consuming the bead meal, the volunteers consumed 50 ml of Calogen® (a fatty preload) which converted gastric motility to a fed pattern and delayed the emptying of the non-nutrient model bead meal from the stomach. Gastric emptying of the bead meal was dominated by the fatty preload, with the median half-emptying time 48 min for both bead types. Gastric sieving of the strong beads occurred with the percentage of the meal remaining at 60 min significantly higher for the strong bead meal compared to the weak bead meal. Both types of gelled beads were clearly visualised by MRI in both the stomach and the small intestine, with beads being more visible in the ileum compared to the jejunum and duodenum. The water proton transverse relaxation time, T2, of the beads was measured in the stomach. After an initial increase, T2 decreased slightly over time from the initial in vivo measurement. In the first paper of this series, we correlate in vitro T2 shortening with formation of a more dense gel network. These in vivo investigations suggest that changes in the gel beads, within the GI tract, could be monitored using T2.

This paper summarises the work undertaken to characterise the physico-chemical properties of alginate gel beads in simulated gastro-intestinal (GI) conditions. Two types of alginate beads were investigated: "strong" (strongly gelled) and "weak" (weakly gelled) beads prepared by long and short exposure to a calcium chloride gelling bath, respectively. The beads were found to shrink in gastric conditions and swell in intestinal conditions due to changes in electrostatic forces in the gel matrix at the different pH and ionic strength conditions. We found a good correlation between the NMR transverse relaxation time (T2) of the water protons within the gel and alginate concentration which was dependant on environmental conditions. T2 shortening was observed in gastric conditions reflecting the formation of a more dense gel network on shrinking. In intestinal conditions, T2 increased reflecting the formation of a more open, porous gel network on swelling. This was corroborated by electron microscopy which clearly depicted the changes in gel density in simulated GI conditions. The mechanical properties of the beads similarly reflected the changes in the gel microstructure with the beads becoming stronger in gastric conditions and weaker in intestinal conditions, respectively. The beads were shown to eventually disintegrate towards the end of the intestinal phase which may make these alginate gel beads an attractive option as controlled delivery devices in the gastro-intestinal tract. The in vivo behaviour of the beads within the GI tract is investigated using non-invasive magnetic resonance imaging in the second paper of this series.

The inhomogeneity of alginate gel beads prepared by an external diffusion method has been characterised using spatially resolved nuclear magnetic resonance or “magnetic resonance imaging” (MRI) and transmission electron microscopy (TEM). The beads exhibited various degrees of inhomogeneity although reducing the length of exposure to the gelling bath and the presence of non-gelling ions decreased gel inhomogeneity. In order to gain information regarding the gastro-intestinal functionality of these beads for in vivo applications, they were exposed to simulated gastro-intestinal conditions. The increased polymer concentration at the edge of the beads was shown to persist throughout our gastro-intestinal model despite the centre of the bead becoming progressively more porous in nature. The porosity of the alginate gels has been quantified by image analysis of transmission electron micrographs and shown to depend on both location within the bead and gastro-intestinal conditions. We suggest that such changes in porosity of these alginate beads during simulated gastro-intestinal conditions may make these an attractive option for controlled delivery applications in vivo.

The objective of the work was to determine whether chitosan-tripolyphosphate (TPP) systems can be used to develop safe gel particles for in vivo applications. In particular, we are interested in the use of chitosan systems capable of swelling at low pH in vitro as potentially swellable, satiety-enhancing ingredients. The formation of homogeneous chitosan-TPP gel beads was improved by reducing the pH of the TPP gelling bath from 8.5 to 4.0 thus increasing the cationic nature of chitosan and the crosslink density. However, the mechanical strength of this system was reduced compared to the basic system. This has been related to tightening of the gel network by increased shrinking of this system. Although release studies have shown that decreasing the pore size of the gel, by increasing the crosslink density through increasing the TPP concentration, is a good way to trap our model release polymer (dextran), in simulated gastrointestinal conditions, these ionic gel beads showed marked shrinking. We propose that this is due to screening of the charge on the chitosan molecule in the presence of salt and/or an excess of TPP which encourages ionic interactions. We conclude that these ionic gels are an unfeasible alternative satiety-enhancing ingredient to covalently crosslinked systems. (C) 2008 Wiley Periodicals, Inc.

Background Unexplained nausea and vomiting is often associated with delayed gastric emptying in patients with functional dyspepsia. We hypothesized that the experience of an unpleasant, nauseating taste could lead to a delay in gastric emptying. Methods Sixteen healthy women consumed a bland liquid test meal on three separate study days. On two of the study days subjects sham fed either a bitter tasting, modified Slim-Fast bar or one with a pleasant strawberry flavour. The time for 50% gastric emptying (GE50) was non-invasively assessed by electrical impedance tomography and antral motility by electrogastrography (EGG). Results Gastric emptying was significantly delayed by sham feeding the bitter compared with the pleasant bar, GE50 24.7±3.9 versus 17.2±1.8 min, P<0.05. EGG power rose significantly during both the pleasant (basal 1.46±0.07 to 2.33±0.14 log10 μV2/min, P=0.000) and the bitter sham feed (basal 1.64±0.09 to 2.35±0.11 log10 μV2/min, P=0.000). Conclusion An unpleasant bitter taste delays gastric emptying but does not significantly impair antral motility.

Previous studies indicated that physical characteristics of food influence satiety, but the relative importance of the oral, gastric, and intestinal behaviors of the food is unclear. The aim of this study was to investigate the satiating effects of 2 types of alginates, which gel weakly or strongly on exposure to acid, compared with guar gum whose viscosity is unaffected by acid. Subjects (n = 12; 3 men, 9 women) ingested a 325-mL sweetened, milk-based meal replacer beverage on 4 separate occasions, either alone as a control or including 1% by weight alginate or guar gum. Intragastric gelling, gastric emptying, and meal dilution were assessed by serial MRI while satiety was recorded for 4 h. MR images showed that all of the meals became heterogeneous in the stomach except for guar, which remained homogeneous. The alginate meals formed lumps in the stomach, with the strong-gelling alginate producing the largest volume. Although gastric emptying was similar for all 4 meals, the sense of fullness at the same gastric volume was significantly greater for all 3 viscous meals than for the control. Compared with the control meal, the strong-gelling alginate (P = 0.031) and guar (P = 0.041) meals increased fullness at 115 min, and the strong-gelling alginate decreased hunger by the 115-min (P = 0.041) and 240-min (P = 0.041) time points. Agents that gel on contact with acid may be useful additions to weight-reducing diets. We hypothesize that this effect is due to distension in the gastric antrum and/or altered transport of nutrients to the small intestine in the lumps.

C. L. Hoad, E. F. Cox, P. Rayment, R. C. Spiller, P. J. Wright, M. Butler, L. Marciani, P. A. Gowland Sir Peter Mansfield Magnetic Resonance Centre, School of Physics and Astronomy, University of Nottingham, Nottingham, United Kingdom, Corporate Research, Unilever R & D, Colworth, Beds., United Kingdom, Wolfson Digestive Diseases Centre, University Hospital, Nottingham, United Kingdom Introduction The study of small bowel diseases using MRI is well established, but the study of normal small bowel physiology using serial MRI has not yet been realised. Small bowel disease is visualised by distending the small bowel lumen with water by using preparations that prevent absorption. This approach is not acceptable when studying normal physiology as water is naturally secreted and absorbed in the small bowel. MRI is potentially a powerful technique for studying gastrointestinal physiology, as it is relatively non invasive, and because it can measure many different, related functions within a single experiment. The aim of this study was to determine the effect of bead strength on the gastrointestinal response to a model meal containing solid beads, using MRI to visualise and quantify the beads in the stomach and small bowel. Materials and Methods Volunteer Selection: Seven healthy volunteers, with no history of gastro-intestinal disease, formed the study group. The study was approved by the Local Ethics Committee and all volunteers gave written informed consent. Meal Descriptions: Two different bead types were used as model ‘solids’ in the study; solid centre alginate beads (strongly gelled) and liquid centre alginate beads (weakly gelled). These were 2-4 millimetres in diameter and made by dropping 200 ml of 1.5% w/w solution of Manugel DMB (ISP, Koln, Germany) into 0.37% CaCl soln for different lengths of time. The meals were randomised in a double blind fashion. In addition to the beads, the volunteers consumed 500 ml of distilled water to help the volunteer swallow the beads without chewing and to provide contrast to the beads in vivo. Study Protocol: Volunteers were asked to attend at 7:45am having fasted overnight and having abstained from alcohol for 24 hours, and caffeine and strenuous exercise for 18 hours. Volunteers were scanned before consumption of the test meals to provide a baseline set of measurements for the study day. A fat pre-load meal of 50 ml Calogen (SHS International Ltd, Liverpool, UK) was given to the volunteers 15 mins before the main bead meal to turn the GI tract into a fed state. Volunteers consumed the bead meal over 15 mins. Images were acquired on a 3.0 T Philips Intera Achieva MRI scanner. Coronal RARE (TE=400ms) images of the small bowel and transverse HASTE (TE=59) images of the stomach were acquired during 2 breathholds for each image type. These acquisitions were repeated at approximately 30 minute intervals over 4 hours. Five minutes before the 4 hour scan 200 ml of water was given to the volunteer to aid visualization of any beads that were left in the stomach. A satiety questionnaire was completed by the volunteer before each imaging period. Analysis: Gastric half emptying times were measured from gastric volume measurements made from the HASTE images. Bead visualisation was scored by one operator using the RARE images in the whole image and four quadrants of the intestine (0 – no beads visible, 1 – few beads visible, 2 – many beads visible). (See figure 1) The quadrants were defined in a coronal image, with the centre being at the inter-vertebral disc between L2 and L3 of the spine. These scores were integrated over the whole time period to give a visualisation score (max 16). The time for the meal to initially reach the cecum from the mouth, was also measured from the coronal images. The areas under the satiety curves (AUC) were calculated. Results The median gastric half emptying times for the meals containing weak and strong gelling beads were 39 mins and 45 mins respectively (inter-quartile range: weak 34-50 mins, strong 39-65 mins). This was despite the weaker beads having a larger initial volume as they retained more water in the gelling process. The median time to cecum for both sets of beads was 120 mins. No beads were observed in the stomach after the water refill. The AUC for the fullness scores to 240 min was statistically higher for the weak compared to strong beads (p=0.043, N=7 Wilcoxon Signed Ranks Test). The median visualisation scores for the whole intestine were 15 and 14 for the weak and strong beads respectively with no statistical differences between them, however the integrated visualisation scores in the different quadrants were statistically different for both the weak and strong beads (p = 0.002 (weak), p=0.003 (strong), N=7 Friedman Test) with quadrants 3 and 4 having lower scores than 1 and 2. Conclusions This experiment showed that it is possible to visualise small beads in the stomach and small bowel to track the progress of a meal through the GI tract. Gastric emptying was longer for the stronger beads despite the smaller volume for this meal. These results are in good agreement with previous studies. The AUC fullness data suggest that gastric volume dominated the fullness rating for this model meal. Beads were visualised in all areas of the small bowel, however they were seen more in the lower quadrants of the small bowel; which is likely to contain mainly the ileum and distal jejunum. Both bead strengths were observed to reach the cecum intact and no differences in transit time from mouth to the cecum were observed between the bead types, suggesting that, in contrast to the stomach, the intra-intestinal forces could not differentiate the bead types. The errors for this measurement were large due to the measurement interval of 30 mins and difficulty in determining if beads had arrived at the cecum if no fluid was also present. References 1. Debatin JF, Eur. Radiol. 9; 1523-1534 (1999). 2. Wright PJ, et al. Proc. 11 Annual Meeting of British Chapter of ISMRM p41 (2005). 3. Hill AJ, et al. Int. J. Obes. 19; 361-375 (1995) 4. Marciani L, et al. Am J. Physiol. Gastrointest. Liver Physiol. 280; G844-G849 (2001) Acknowledgements This work was funded by Unilever.