Osteoarthritis (OA) is the most prevalent chronic disease of the joints and leads to degeneration of articular cartilage surfaces. While physical therapy and weight loss have demonstrated improved functionality in patients with OA, current drugs are limited to providing symptomatic relief. Thus, there is a notable need for the development of a disease modifying therapeutic for OA. Mesenchymal stromal cells (MSC) offer a promising treatment strategy for OA due to the regenerative and immunomodulatory capacity these cells possess. MSC therapeutics for cartilage regeneration have been widely studied, in both the pre-clinical and clinical environment. While pre-clinical studies have shown improved cartilage repair with MSC treatment, effective translation into the clinic has been limited by numerous factors ranging from high variability and heterogeneity of MSCs to poor understanding of critical quality and potency attributes. This knowledge gap motivates the overall objective of this thesis to identify cellular attributes that relate to the therapeutic efficacy of human MSC (hMSC) for treatment of OA. The overarching hypothesis is that a profile of hMSC secreted cytokines, ribonucleic acid (RNA) transcripts, and intracellular signaling phospho-proteins can be identified that relate to the therapeutic efficacy of hMSCs for treatment of OA.

To test this hypothesis, bone marrow derived hMSCs were encapsulated in sodium alginate to independently assess the paracrine signaling properties of these cells, by preventing their direct engraftment into the native tissue, in both developing OA [3-week medial meniscal transection (MMT) study] and established OA (6-week MMT study), in Specific Aim 1. The paracrine mechanisms of these hMSCs showed a chondroprotective role for articular cartilage in both stages of OA, attenuating increases in articular cartilage swelling and surface roughness in developing OA and reducing increases in articular cartilage surface roughness and exposed bone area in established OA. Though the encapsulated hMSCs provided a disease modifying protective effect, the treatment did not regenerate or restore the cartilage back to levels comparable to sham controls in either model (3-week and 6-week MMT studies) of disease assessed. Though protective effects were observed on the cartilage, encapsulated hMSCs yielded increased osteophyte volumes in both developing OA and established OA, which have been identified as an unwanted phenotype for restoring joint function. The effects of biomaterial encapsulation on hMSC cytokine secretion was also assessed in an OA simulated microenvironment [IL (interleukin)-1β conditioning]. The immunomodulatory potential of biomaterial encapsulation on hMSC function demonstrated a targeted paracrine response, while nonencapsulated hMSCs showed an indiscriminate upregulation of all cytokines quantified. Overall, these studies demonstrated that biomaterial encapsulation of hMSCs mediated the paracrine response to a simulated OA microenvironment and enhanced the in vivo therapeutic efficacy of the hMSCs in both delaying the onset of OA and in preventing further disease progression in established OA.

The findings in Specific Aim 1 raised important questions regarding how these hMSC treatments would translate into the clinic, as variability in hMSC source (donor heterogeneity) remains a pressing challenge in effectively translating these MSC therapeutics into viable clinical treatments. Thus, in Specific Aim 2 donor heterogeneity was assessed to identify secreted cytokines, RNA transcripts, and intracellular signaling phospho-proteins from the MAPK and Akt pathway (cellular attributes) of hMSCs that relate to the therapeutic outcomes of hMSC therapeutics in OA, as determined in the MMT model of OA. Multivariate analyses using partial least squares regression (PLSR), principal component analysis (PCA), and correlation analysis were used to relate cellular attributes and therapeutic outcomes. These studies demonstrated that hMSC secretion of granulocyte macrophage colony stimulating factor (GM-CSF), chemokine ligand 1 (GRO), IL-4, platelet derived growth factor (PDGF)-AA, and transforming growth factor (TGF) β3 relate to the cytokine secretory profile of more therapeutic hMSCs. The findings of this cytokine analysis were further supported with the outcomes of RNA-Seq which demonstrated enrichment of the majority of the therapeutic cytokine gene expression pathways identified. Furthermore, RNA-Seq implicated key differences in signaling gene expression pathways, motivating the study of intracellular signaling. Thus, signaling was quantified for both the MAPK and Akt pathways was studied in Specific Aim 3. Overall, these studies demonstrated that a profile of hMSC cellular attributes could be identified in an in vitro environment that relate to the therapeutic efficacy of these hMSCs in vivo.

Building off the exciting findings in Specific Aim 2, the intracellular signaling pathways mediating the secretion of therapeutic related cytokines were explored in Specific Aim 3. More specifically, phospho-proteins in the MAPK and Akt intracellular signaling pathways were quantified due to their key role in modulating the MSC immune response and due to the regulatory role, these pathways have upstream of the immunomodulatory cytokines analyzed in Specific Aim 2. These studies demonstrated that more therapeutic hMSCs demonstrated increased p-MSK1, p-p53, p-Atf-2, p-p38, p-HSP-27, and p-JNK signaling relative to less therapeutic hMSCs. Furthermore, less therapeutic hMSCs demonstrated increased p-Akt signaling levels, only, relative to more therapeutic hMSCs. To confirm that these signaling phospho-proteins identified were mediating the paracrine signaling of hMSCs,signaling phospho-proteins were then perturbed with pharmacological interventions to confirm their mediating role. In the MAPK pathway the p-JNK phosphoprotein in more therapeutic hMSCs was targeted with the SP600125 p-JNK inhibitor and showed these treated hMSCs demonstrated reduced secretion of the therapeutic related cytokines GM-CSF, GRO, IL-4, and PDGF-AA, which related to therapeutic outcomes in a pre-clinical model of OA. In the Akt pathway the p-Akt phospho-protein in more therapeutic hMSCs was targeted with the SC79 p-Akt activator and showed that these treated hMSCs showed increased secretion of nonspecific cytokines. Based on these findings a combination therapy of a Metformin p-JNK activator and MK-2206 p-Akt inhibitor were used to treat less therapeutic hMSCs and demonstrated that treating these hMSCs with this intervention strategy yielded a more therapeutic phenotype. More specifically, less therapeutic hMSCs treated with this combination therapy yielded a more targeted paracrine profile with increased secretion of GM-CSF, GRO, IL-4, and PDGF--AA;​ thus, yielding a cytokine profile similar to that of more therapeutic hMSCs.

Overall, this thesis increases the fundamental knowledge of hMSCs as therapeutics for use in OA and provide therapeutic targets that are translatable into the clinic. More specifically, MSC cellular attributes that relate to the efficacious outcomes of these cellular therapeutics were identified and intervention strategies were developed to increase the potency of MSCs exhibiting less therapeutic potential.