br Results br Discussion The homing engraftment of

Results
Discussion The homing/engraftment of MSCs is the process by which cells migrate, engraft, and exert local functional effects. Homing involves a cascade of events initiated by shear-resistant adhesive interactions between flowing cells and the target endothelium or extracellular matrix. This process is mediated by adhesion receptors expressed on circulating cells that engage relevant coreceptors, resulting in cell tethering, rolling, and firm adherence (Sackstein, 2005). Similarly to induction of IDO in MSCs, we demonstrate that homing/engraftment of MSCs is an “active” process that requires cells that are metabolically (i.e., viable) and functionally fit, as dead MSCs or those with a disrupted pyruvate dehydrogenase kinase cytoskeleton do not engraft. We demonstrate that immediately after thawing, cryo MSCs have a reduced capacity to bind to both extracellular matrix molecules (i.e., fibronectin) and endothelial cells, and these binding defects can be reproduced in vivo. Consistent with our data, Castelo-Branco et al. (2012) showed that cryo MSCs are incapable of homing to the inflamed colon in a murine model of colitis, and Hattori et al. (2001) showed that HSCs also have a migratory defect post-thaw. Thus, we propose that the process of cryopreservation and thawing exerts a generic negative effect on cells with regard to their migratory/homing properties. Cryoinjury is a significant problem in many primary cell types (Durkut et al., 2013; Fortunato et al., 2013) and is likely associated with both generic and cell-specific effects. Within the mesenchymal lineage, both hMSCs (François et al., 2012) and human primary fibroblasts (Liu et al., 2000) undergo a biochemical “heat-shock stress response” when thawed from cryopreservation, which not only serves to protect the cells from death (Beere, 2005; Yenari et al., 2005) but can also result in protein synthesis inhibition (Cuesta et al., 2000; Doerwald et al., 2003). Conversely, although both HSCs and MSCs have a reduced binding potential post-thaw, mechanistically the defect appears to be cell-type dependent. For HSCs, the primary defect appears to be due to shedding of L-selectin from the membrane (De Boer et al., 1998; Koenigsmann et al., 1998) and is due primarily to DMSO exposure rather than the freezing process. Selectins comprise a family of three members (E-, P-, and L-selectin), are differentially expressed by leukocytes and endothelial cells, and are involved in the early steps of leukocyte extravasation. There is no evidence from the literature that MSCs possess selectins (Rüster et al., 2006); thus, shedding of selectin is unlikely to explain our result. On the other hand, MSCs do express numerous cell adhesion molecules, including VLA4 (CD49d and CD29 integrins) (De Ugarte et al., 2003), that can interact with VCAM-1 on endothelial cells. As such, we investigated whether receptor shedding of nonselectin adhesion receptors might explain the defect we observed. Our data suggested that receptor shedding is unlikely to be the putative cause for reduced MSC binding immediately post-thaw. Furthermore, we ruled out viability, calcium influx, and metabolism as possible explanations. Thus, the homing defect in MSCs that occurs immediately after thawing is not only different from that seen in HSCs but is also independent of basal metabolism. Although the mechanical properties of adhesive attachments are usually attributed to the ligand-receptor interaction, the strength and survival of cell attachment depend heavily on molecular connections below the plasma membrane surface (Evans and Calderwood, 2007). In particular, the interactions of the cytoplasmic integrin b tails with intracellular cytoskeletal and signaling proteins figure prominently in the regulation of integrin activation (Evans and Calderwood, 2007). Furthermore, efficient actin polymerization organization and remodeling are critically important for stabilizing integrin ligand interactions, which allows cells to resist detachment under conditions of flow (Rullo et al., 2012). Recently, Xu et al. (2012) evaluated the effect of cooling rates on the recovery of MSCs from cryopreservation and demonstrated that immediately after thawing, the F-actin component of the cytoskeletons was disrupted. They further demonstrated that the critical process that led to the changes in F-actin morphology and distribution was not the addition/removal of the cryopreservative (i.e., DMSO), but rather the freezing process itself, and that slower cooling rates of 5°C/min and 10°C/min resulted in greater disruption (Xu et al., 2012). In another study, Ragoonanan et al. (2010) demonstrated that post-thaw cryo MSCs show abnormal actin morphology as well as changes in intracellular pH and mitochondrial aggregation. This prompted us to investigate the F-actin polymerization capabilities of MSCs in our system. In agreement with the above reports, we found that (1) immediately after thawing, cryo MSCs had a significantly reduced capacity to polymerize F-actin in vitro, and (2) when the F-actin cytoskeleton of cultured live MSCs was chemically disrupted, they lost their capacity to engraft in vivo. Thus, the ability of MSCs to rapidly remodel their cytoskeleton is a significant contributor to the homing/engraftment defect we observed. Other possible contributors to the MSC binding defect we observed include alterations in integrin confirmation or clustering that could impact binding affinity or avidity, respectively, as well as alterations in the formation of macromolecular adhesion complexes that could impact signal transduction. However, these were not evaluated in this study. In future studies, understanding how cryopreservation and thawing affects the intra- and extracellular binding machinery of MSCs, and developing strategies to mitigate these binding defects may help to improve the therapeutic utility of MSCs and minimize potential safety issues.