This item is under embargo and not available online per the author's request. For access information, please visit


Date of Award

Spring 5-15-2021

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Molecular Cell Biology)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



End-stage organ failures remain a clinical challenge with an unmet need for medical therapies, with transplantation often being the only curative option. Despite advances in transplantation outcomes, organ shortage continues to limit the availability of cures to patients in need. The direct lineage reprogramming of one cell type to another is a promising avenue for therapy with the following advantages: (1) patient-specific cell sources, (2) direct conversion without reverting to pluripotency and the associated risk of teratoma formation, and (3) utilization of the cell type responsible for fibrotic scar formation for the engineering towards the desired cell fate. Nonetheless, many questions remain in the field, with open issues related to reprogramming trajectory, efficiency, and specificity. These issues contribute to the limited utility of directly reprogrammed cells, which oftentimes do not engraft target organs successfully or do so with only partial functionality.

Of note, many reprogramming strategies are extremely inefficient and produce heterogeneous cells, with much of the molecular mechanism remaining unknown. A prototypical engineering approach is reprogramming fibroblasts to induced endoderm progenitors (iEPs). This lineage conversion is achieved by overexpressing transcription factor Hnf4α, and pioneer transcription factor Foxa1, 2, or 3. In addition to binding to their gene targets, pioneer transcription factors can bind compacted chromatin, increase target site accessibility, and recruit cooperative transcription factors. Pioneer transcription factor binding often precedes transcriptional activation during development and is thought to be important for establishing competence for developmental programs.

Interestingly, when iEPs are transplanted into the mouse, they can engraft both liver and colon, suggesting that they consist of cells with hepatic and intestinal potentials. This is unsurprising, as both Foxa1/2/3 and Hnf4α are known to be important for endoderm development. However, not all cells become reprogrammed despite abundant overexpression of reprogramming factors. This raises important questions regarding the mechanism of reprogramming: What gene targets are bound by reprogramming factors? Are different outcomes due to differences in binding? Is pioneer transcription factor binding important for reprogramming success? My hypothesis for the observed inefficiency and heterogeneity of direct lineage reprogramming is that pioneer transcription factors bind developmental gene targets inefficiently and variably, influenced by variable chromatin contexts. To test my hypothesis, I utilized several recent technologies, including: single-cell RNA-sequencing, to measure transcriptional changes at a resolution needed to reveal population heterogeneity and reprogramming dynamics; our novel cell tracking approach ‘CellTagging’, to study the dynamics of clonal expansion during reprogramming and to track cell identity changes in a competitive transplant setting; and the single cell Calling Cards assay, to record transcription factor binding in different reprogramming trajectories and reveal important gene regulatory events that influence reprogramming outcomes during the early phases of reprogramming.

We found that direct reprogramming is characterized by distinct paths: one leading to successful reprogramming, and a ‘dead-end’ trajectory. Trajectory bifurcation is deterministic early on, by day 13. This also led to the discovery of Mettl7a1, a putative methyltransferase identified to be associated with the successful trajectory, which increases reprogramming efficiency when added to the reprogramming cocktail. We later used CellTag Indexing to track cells from the reprogramming and dead-end trajectories transplanted into the colon, and discovered that successfully reprogrammed iEPs engraft via an intestinal stem cell state. Finally, active enhancer recording in early iEP reprogramming showed that the reprogramming trajectory is associated with rapid activation of target tissue-specific enhancers in regions that are repressed in the starting cell type, suggesting that overcoming the chromatin barrier might an important event for reprogramming outcome.

In summary, I hope to have revealed some insight into the observed inefficiency and heterogeneity in direct lineage reprogramming. This is a nascent but promising field where much of the molecular mechanism is still poorly understood, with many more remaining questions to be answered before advances can be brought from bench to bedside. Further understanding the mechanism of direct reprogramming, by studying the early actions of pioneer transcription factors, may reveal additional roadblocks that are limiting reprogramming efficiency, and may one day lead to novel strategies for improved cell fate engineering and application in regenerative medicine.


English (en)

Chair and Committee

Samantha Morris

Committee Members

Benjamin Humphreys, Robi Mitra, Laura Schuettpelz, Andrew Yoo,

Available for download on Sunday, May 15, 2022