|Program and Abstracts
Back to 2015 Annual Meeting Program
Rational Identification of Novel Surface Markers to Enhance Cell Therapies
Michael Januszyk, MD, Robbert C. Rennert, MD, Michael Sorkin, MD, Melanie Rodrigues, PhD, Dominik Duscher, MD, Michael T. Chung, MD, Kevin Paik, BS, Revanth Kosaraju, BS, Michael Findlay, MD PhD, Jason P. Glotzbach, MD PhD, Zeshaan N. Maan, MBBS, MRCS, Atul J. Butte, MD, PhD, Geoffrey C. Gurtner, MD, PhD.
Stanford University, Stanford, CA, USA.
Cell-based therapies are an emerging approach for regenerative medicine and wound healing applications. Existing therapies have been developed empirically, making it difficult to improve technologies in the face of negative results. Recent advances in high-throughput, microfluidic technology have enabled single-cell resolution analytics that can provide new information to guide this process. In prior work, we have combined single cell transcriptional analysis with novel mathematical modeling to characterize heterogeneity in putatively homogeneous stem cell populations. Here we aimed to build on this work in order to develop a rational, systematic framework to improve cell-based therapies.
Utilizing human adipose derived stem cells (hASCs) as a test progenitor cell pool, we first obtained a comprehensive profile of hASC surface marker (SM) expression through single-cell transcriptional analysis of all known SMs with commercially available antibodies. This allowed us to cast the widest possible net in our search for novel subpopulation defining markers without relying on a priori assumptions of gene expression. Using this approach, we identified over 200 markers that were expressed within hASCs. Focusing on the 96 SMs with highest, non-uniform expression (which are most likely to distinguish biologically important cell subsets), we applied targeted transcriptional analysis and partitional clustering to identify subpopulations with pro-regenerative expression profiles. Linear discriminant analysis was then applied to determine the optimal subset of surface antigens necessary to prospectively isolate these cells for subsequent in vitro and in vivo functional evaluation. Cell survival, stemness, and engraftment capacity were assessed for each progenitor subset. In vivo vasculogenic capacity was evaluated using a hind-limb ischemia model. Primary cells from each group were seeded into a splinted excisional model of murine wound healing using a hydrogel scaffold and healing rates assessed.
We identified a distinct subpopulation of hASCs that was consistently present across multiple partitional clustering permutations. This subpopulation could be defined with high sensitivity and specificity by two surface marker genes (DPP4 and CD55) and was present across multiple human patients. This population was recapitulated on a protein level and could be prospectively isolated based on co-expression of DPP4 and CD55 using fluorescence-activated cell sorting (FACS). We found that these cells expressed increased levels of general stem cell markers (such as CD34 and CD73), as well as genes associated with cancer stem cells (CD99 and ITGB3) and embryonic stem cells (GGT1), suggesting that this subpopulation may have increased regenerative potential. Additionally, these cells were depleted in the setting of both diabetes and aging - pathologic states associated with impaired tissue regeneration. As predicted, subpopulation enrichment led to significant enhancement in cell survival, stemness, and engraftment capacity. Furthermore, application of FACS-enriched healthy ASCs significantly improved wound closure rates and dermal regeneration compared to application of unsorted or depleted cells, effectively normalizing diabetic healing.
This work establishes the efficacy of single cell analysis for not only the rational enhancement of cell-based therapeutics, but also for the study of progenitor cell biology in pathologic states.
Back to 2015 Annual Meeting Program