Factors influencing alginate gel biocompatibility.

Alginate remains the most popular polymer used for cell encapsulation, yet its biocompatibility is inconsistent. Two commercially available alginates were compared, one with 71% guluronate (HiG), and the other with 44% (IntG). Both alginates were purified, and their purities were verified. After 2 days in the peritoneal cavity of C57BL/6J mice, barium (Ba)-gel and calcium (Ca)-gel beads of IntG alginate were clean, while host cells were adhered to beads of HiG alginate. IntG gel beads, however, showed fragmentation in vivo while HiG gel beads stayed firm. The physicochemical properties of the sodium alginates and their gels were thoroughly characterized. The intrinsic viscosity of IntG alginate was 2.5-fold higher than that of HiG alginate, suggesting a greater molecular mass. X-ray photoelectron spectroscopy indicated that both alginates were similar in elemental composition, including low levels of counterions in all gels. The wettabilities of the alginates and gels were also identical, as measured by contact angles of water on dry films. Ba-gel beads of HiG alginate resisted swelling and degradation when immersed in water, much more than the other gel beads. These results suggest that the main factors contributing to the biocompatibility of gels of purified alginate are the mannuronate/guluronate content and/or intrinsic viscosity.

Role of protein contaminants in the immunogenicity of alginates.

Alginate is widely used for cell microencapsulation and transplantation. There is a lack of standardization of alginate purity and composition. In a previous study, we compared different alginate purification methods and concluded that polyphenol and endotoxin contaminants were eliminated efficiently but residual protein contaminants persisted with all of the methods under evaluation. The objective of this study was to test the hypothesis that residual proteins play a role in the immunogenicity of certain alginate preparations. Using preparative size exclusion chromatography (SEC) and a large scale purification protocol that was derived from the findings obtained with SEC, we substantially decreased the protein content of alginate preparations. When implanted into mouse peritoneum, barium alginate beads made of alginates that were purified using SEC or the derived large scale protocol induced significantly less pericapsular cell adhesion than those made with control alginates. In conclusions, these results suggest that removing residual protein contamination may decrease the immunogenicity of certain alginate preparations. The measurement of proteins could be used as a screening method for evaluating alginate preparations.

Direct effect of alginate purification on the survival of islets immobilized in alginate-based microcapsules.

Alginate purification has been shown to decrease the host immune response to implanted alginate-based microcapsules, but the direct effect of contaminants on islet cell survival remains unknown. Wistar rat islets were immobilized in calcium alginate beads made with crude vs. purified alginate and then incubated in CMRL culture medium. Islet survival was evaluated at 1, 4, 7, 14 and 27 days post-encapsulation. Islet viability was investigated using a dual staining assay (propidium iodide and orange acridine). The islet cell necrosis and the proportion of apoptotic cells were quantified under optical microscopy and with a TUNEL assay, respectively. Islets immobilized in purified alginate were more viable, and had fewer necrotic centers, a smaller area of central necrosis and a lower number of apoptotic cells. At day 14 and 27 post-encapsulation, respectively, 48% and 23% of islets were viable with purified alginate vs. 18% and 8% with crude alginate (p<0.05). At day 14, the surface area of central necrosis and the number of necrotic islets were more important with the impure alginate (65% vs. 45% and 73% vs. 53%, respectively; p<0.05). We conclude that alginate purification improves the survival of islets that are immobilized in alginate-based microcapsules. These findings indicate that caution should be taken in the interpretation of in vivo experiments, as the results could be explained by either a direct effect on islet survival or a modification of the host reaction, or both. Moreover, it suggests that the effect on islet viability should be assessed during the development of biomaterials for cell encapsulation.

Absence of CD47 in vivo influences thymic dendritic cell subset proportions but not negative selection of thymocytes.

CD47 is a ubiquitously expressed molecule which has been attributed a role in many cellular processes. Its role in preventing cellular phagocytosis has defined CD47 as an obligatory self-molecule providing a 'don't-eat-me-signal'. Additionally, CD47-CD172a interactions are important for cellular trafficking. Yet, the contribution of CD47 to T cell stimulation remains controversial, acting sometimes as a co-stimulator and sometimes as an inhibitor of TCR signalling or peripheral T cell responses. Most of the experiments leading to this controversy have been carried in in vitro systems. Moreover, the role of CD47 on thymocyte differentiation, which precisely relies on TCR signal strength, has not been evaluated. Here, we examine the in vivo role of CD47 in T cell differentiation using CD47-deficient mice. We find that, in the absence of CD47, thymocyte positive and negative selection processes are not altered. Indeed, our data demonstrate that the absence of CD47 does not influence the strength of TCR signalling in thymocytes. Furthermore, in agreement with a role for CD47-CD172a interactions in CD172a(+) dendritic cell migration, we report a reduced proportion of thymic dendritic cells expressing CD172a in CD47-deficient mice. As the total proportion of dendritic cells is maintained, this creates an imbalance in the proportion of CD172a(+) and CD172a(low) dendritic cells in the thymus. Together, these data indicate that the altered proportion of thymic dendritic cell subsets does not have a primordial influence on thymic selection processes.

The effect of covalent cross-links between the membrane components of microcapsules on the dissemination of encapsulated malignant cells.

Stem cells and immortalized cells have considerable therapeutic potential but present risks of malignant transformation. Cell microencapsulation allows transplantation without immunosuppression. We have developed a method for microencapsulating living cells within covalently cross-linked membranes that are chemically and mechanically extremely resistant. We provide herein direct evidence that these microcapsules can prevent malignant cell dissemination. When 20,000 or more nonencapsulated EL-4 thymoma cells were implanted intraperitoneally in mice, all recipients died with widespread metastasis within 26.3+/-1.0 days. All recipients of 250,000 EL-4 cells microencapsulated in covalently cross-linked membranes were living and disease-free, 150 days post-implantation. Encapsulation in standard microcapsules only slightly delayed the recipient death. Pancreatic islets transplanted using either type of microcapsule presented similar survival. We conclude that microencapsulation in covalently cross-linked membranes prevents malignant cell dissemination.