349920 Investigation of Human Embryonic Stem Cell Differentiation Induced By Chitosan Nanoparticles

Monday, November 4, 2013
Grand Ballroom B (Hilton)
Kimaya Padgaonkar, Department of Chemical Engineering, University of Pittsburgh, Pittsburgh, PA, Thomas Richardson, Chemical Engineering, University of Pittsburgh, Pittsburgh, PA, Joseph E. Candiello, Bioengineering, University of Pittsburgh, Pittsburgh, PA, Shilpa Sant, Department of Pharmaceutical Sciences, University of Pittsburgh School of Pharmacy, Pittsburgh, PA and Ipsita Banerjee, Chemical and Petroleum Engineering, University of Pittsburgh, Pittsburgh, PA

Investigation of Human Embryonic Stem Cell Differentiation Induced by Chitosan Nanoparticles

Kimaya Padgaonkar1, Thomas Richardson1, Joseph Candiello2, Shilpa Sant2,3,4, Ipsita Banerjee1,2,4
Department of Chemical Engineering1 & Department of Biongineering2, Swanson School of Engineering; Department of Pharmaceutical Sciences3; McGowan Institute for Regenerative Medicine4, University of Pittsburgh

Type 1 diabetes results in the destruction of insulin producing cells, leaving the individual insulin dependent.  These people require alternative sources of insulin, such as manual insulin injection or transplantation of donor organs. However, insulin injections are inconvenient while donor organs and islets for transplantation are scarce, therefore, an alternative option—the transplantation of insulin producing cells from a regenerative source—could be highly influential. With their unlimited self-renewal capacity, hESCs can potentially alleviate the problem of current insulin sources, after undergoing successful differentiation to the proper functional cell type. The first step in the differentiation of hESCs is inducing germ layer determination to mesendoderm before differentiating into further germ layers including definitive endoderm. Directed differentiation by addition of soluble chemical factors is one method that allows for induction of hESC to the mature functional cell type. However, these soluble growth factors, such as Activin A typically used to induce definitive endoderm, are expensive. We investigated the possibility of using less expensive, insoluble materials such as chitosan to induce differentiation. We investigated the possibility of differentiating hESCs in presence of chitosan nanoparticles (CNPs) using alternate culture configurations.

Monolayer 2-D cultures on tissue culture plastic (TCP), a common method for culturing cells, was utilized as well as two different 3-D methods---calcium alginate encapsulation and cellular aggregation. The 2D cultures, with CNPs added in the culture media, elicited no significant effect in the hESC population. The viability and proliferation of the hESCs at the CNP concentration used was unchanged compared to cultures without CNPs, and the hESCs on TCP did not undergo differentiation compared to culture without CNPs.  For 3D culture configuration, first we co-encapsulated hESCs with CNPs in a calcium alginate microcapsule. The co-encapsulated cells were allowed to differentiate with and without the growth factor, Activin A. hESCs encapsulated in calcium alginate without CNPs were used as differentiation spontaneously differentiating control. Very interestingly, the CNP co-encapsulated hESCs showed significant difference in differentiation, compared to the spontaneous control.  In 3-D encapsulation, gene expression analysis showed that ectoderm and mesoderm specific differentiation were not induced. Quite surprisingly however, mesendoderm and endoderm specific differentiation were induced on a level similar to chemically induced differentiation by Activin A. Definitive Endoderm differentiation was further confirmed by analysis of protein expression for proteins indicative of the definitive endoderm germ layer. Since chitosan was observed to induce endoderm differentiation even without any growth factor, we began an investigation of the effect of chitosan on hESC aggregates alone.

To isolate the cause of differentiation, calcium alginate was eliminated in our subsequent experiments, and hESCs were directly aggregated with CNP using various formation methods. Confluent H1 hESC colonies were made into a single cell suspension with Accutase and treated with ROCK inhibitor. Three methods were utilized to form aggregates. The hanging drop method had a cell seeding density of 10x105 cells/mL and 30 µL of media for each hanging drop. Aggregates were collected after 4 days. CNP to hESCs mixed ratios were 1:1, 1;10, and 1:40. The microwell method had a cell seeding density of 10x106 cells/mL in  450 nm microwells, and were seeded using two sequential seedign steps. Aggregates were collected after 4 days. Three techniques were used: CNPs mixed with hESCs (1:10) and then seeded in two steps; CNPs seeded between two layers of hESCs (1:40); CNPs mixed with hESCs (1:40) and then seeded in two steps. Finally, the stirred suspension method was used to form hESC-CNP aggregates by suspension of 1x106cells/ml in media containing CNPs at a ratio of 1:10 and 1:40 (CNPs to cells) in a non-adherent tissue tissue culture dish.  The dish was placed on a plate shaker at 40 rmp in the incubator for four days to induce aggregate formation.

CNPs were labled with FITC to monitor the incorporation within the aggregate by microscopy.  For the hanging drop method, 1:40 CNPs to cell ratio had the best EB formation and CNP incorporation of the three concentration ratios tested. For the microwell and stirred suspension method, 1:40 CNPs to cells ratio had the best EB formation and CNP incorporation of the two concentration ratios tested. The mixed approach was more straightforward than the layered approach. The best technique creating aggregates in microwells had the greater ratio between CNPs and cells (1:40) and incorporated CNPs with cells prior to seeding. It created uniform aggregates and was more straightforward compared to the layering method. The microwell method showed better incorporation of CNPs within aggregates compared to the hanging drop method which created smaller, less uniform aggregates even within the same CNP ratio trials. However, the stirred suspension method showed the most uniform distrubtion of CNPs throughout the aggregates. 

Aggregates formed by the microwell and stirred suspension method were further analyzed via gene expression and the results were not as expected as compared to the alginate encapsulation gene expression data. We observed little to no upregulation in the definitive endoderm stage or any germ lineage even though the encapsulation data had showed strong upregulation.  Recent data suggests that differentiation of the hESC encapsulated with CNPs could be the results of CNP endocytosis of CNPs.  Aggregate formation methods may be inducing aggregation of the CNPs themselves into larger constructs, resulting in a differing biophysical interaction between the chitosan particles and hESCs. The microwell method is still under investigation so trials will be repeated and further analyzed. Based on those results, exploration of microwell (sizes) and CNP incorporation (ratios) can be continued



Kimaya Padgaonkar: kip15@pitt.edu

Thomas Richardson: tommyr87@gmail.com

Joseph Candiello: joe.e.candiello@gmail.com

Shilpa Sant:shs149@pitt.edu

Ipsita Banerjee: banerjee.ipsita@gmail.com

Extended Abstract: File Not Uploaded