Regenerative Medicine

EBioMedicine 2015; 24;4:62-73.

Diagram of stem cell population

Mesenchymal stem cells (MSCs) have gained appeal as an experimental therapy for a diverse array of human maladies due to their broad-based efficacy in pre-clinical models and highly favorable safety profile in human patients. Although MSC-based therapies have been evaluated in >900 human clinical trials, in many cases patient outcomes have been suboptimal. To improve efficacy, we have developed a Clinical Indications Prediction (CLIP) scale, which simultaneously classifies human MSC donor populations based on their intrinsic biological activity and predicts how large-scale manufacturing alters the composition and function of these populations. The basis for the CLIP scale is rooted in the transcription factor TWIST1, which we showed coordinately regulates stem/progenitor and paracrine functions of MSCs. Specifically, intrinsic levels of TWIST1, which vary significantly between human MSC donor populations, predicts differences in growth rates, colony forming unit-fibroblast activity, tri-lineage differentiation potential, and pro-angiogenic, anti-inflammatory and immuno-modulatory activity as determined in relevant cell-based assays and in vivo models. Consequently, the CLIP scale is transformative in that it provides a means to pre-select human MSC donor populations and manufacturing schemes to develop therapies tailored to a specific disease indication. Currently, the lab is evaluating the ability of the CLIP scale to reconcile the biological activity of human MSC isolates with patient outcomes in completed human clinical trials via retrospective analysis. It is also being used to evaluate how existing cGMP manufacturing protocols alter the composition and biological activity of human MSCs, and to develop novel protocols to generate MSCs of defined potency for different disease indications. The lab is also pursuing studies to further interrogate the complex biology of TWIST proteins in MSCs in order to expand the scope and enhance the predictive value of the CLIP scale.

Skeletal Biology

Diagram of skeletal biology

Skeletal involution and marrow adiposity are significant clinical pathologies with multiple causal stressors in adults including chronological aging, obesity, type 2 diabetes, mechanical unloading (disuse), menopause, and starvation. These conditions result in decreased bone mineral density (BMD) and alterations in bone architecture that lead to increased fracture risk and delayed fracture healing. The later conditions are debilitating to patients, result in increased mortality rates, and are costly to treat and manage. Bone marrow has long been recognized to contain adipose tissue, and emerging studies indicate that obesity, disuse, osteoporosis and aging also result in observable increases in marrow adiposity, which is believed to further exacerbate bone loss. Bone marrow is the only tissue where bone and fat coexist in the same micro-environment, and it is now well-established that these tissues arise from a common skeletal stem cell (SSC), and recent studies indicate that SSC dysfunction may contribute to the pathophysiology of several skeletal-related diseases. Our lab is currently using mouse models of diet-induced obesity, mechanical unloading (disuse), and chronological aging to interrogate how alterations in SSC frequency and function contribute to skeletal pathophysiology. In these studies, we employ biophysical (microCT, FACS), molecular (RNA-Seq) and genetic approaches to identify key regulatory pathways and candidate genes that may represent useful drug targets for intervention.

Stem Cell Biology

Cell Death Differ 2018; 25:677-690.

Diagram: Stem cell biology

Our lab developed a reliable method to isolate primary MSCs from mouse bone marrow and has exploited this technology to interrogate pathways critical for MSC self-maintenance. These studies have uncovered important roles for p53, JNK1, TWIST2, and IP6K1, the function of which are being evaluated in more detail using genetics-based approaches. For example, the lab recently reported that mouse MSCs are highly sensitive to oxygen-induced stress via a p53-dependent mechanism, and careful examination of MSCs from p53 null mice uncovered an indispensable role for this protein in normal MSC maintenance. These findings have important implications for use of immortalized cell lines as surrogates to study MSC biology and reveal new mechanisms by which p53 regulates multi-potency and the cellular redox balance. Similarly, the lab also reported that loss-of-function of IP6K1 enhances MSC fitness in vitro, and skews cells toward an osteogenic fate at the expense of adipogenesis. We have also shown that TWIST2 functions to promote growth and inhibit multi-lineage differentiation of MSCs. Consequently, the lab continues to employ mouse and human MSCs to conduct basic studies in stem cell biology. We are also developing cell-based screening applications for MSCs to be used in drug development.

Cancer Therapeutics

ACS Chem Biol 2015; 10:2267-2276.

Diagram: Cancer therapeutics

Breast cancer is among the most common cancers in the U.S. and while early diagnosis is critical to successful treatment outcomes, survival rates for patients with advanced disease remain low. Our laboratory identified a microRNA cluster located within the highly imprinted DLKI-DIO3 genomic locus whose expression was found to be dysregulated in invasive ductal carcinoma. One microRNA in the cluster, miR-544 was found to be highly induced in breast adenocarcinoma cells in response to hypoxic stress and subsequent studies revealed miR-544 suppressed growth regulatory and energy utilization pathways thereby driving tumor adaptation to such stress. The lab has since pursued a two-pronged approach to develop RNA-targeting therapeutics capable of modulating the activity of miR-544 in cancer. Specifically, a rationale-design based approach pioneered by the Disney lab was used to query over 900 small molecule–RNA motif binding partners to identify compounds that are predicted to bind UU internal loops within the miR-544 hairpin precursor. Five potential heterocyclic lead small molecules were identified using this approach, and their bioactivity and in cellulo selectivity for miR-544 were evaluated using a novel fluorescent-based miRNA-drug interaction screening platform developed by in our lab that is capable of high-throughput application. We have shown that several promising lead compounds inhibit miR-544 maturation with high selectivity and specificity and sensitize breast cancer cell lines to killing by 5-fluorouracil and hypoxia at a 25-fold lower concentration than a 2’-O-methyl RNA antagomir designed to target miR-544. These small molecules also block growth of tumor xenografts in mice by direct modulation of the target microRNA, thereby providing proof-of-concept that RNA- directed chemical probes can serve as novel
lead therapeutics. Recent studies in the lab have shown that other lead compounds may enhance the potency and expand the range of immuno-therapies by augmenting antibody dependent cellular cytotoxicity (ADCC) killing of breast cancer cells. Studies are ongoing to develop optimized 2nd generation compounds with improved pharmacokinetics for clinical development.