The angiopoietin-Tie2 signaling pathway is central to regulating vascular stability, remodeling, and permeability. Angiopoietin-1 (Ang1) promotes pAKT activation and vascular stability and regeneration, whereas Ang2 antagonizes these effects, leading to leaky vasculature. Although Tie2’s association with α5β1 integrin has been implicated in mediating these divergent outcomes, the requirement of direct F-domain ligand binding for integrin recruitment remains unclear. Here, we report the development and mechanistic evaluation of a de novo designed Tie2 mini binder (Tmb) that selectively targets the Tie2 receptor without engaging α5β1 integrin. Using an AI-based protein design pipeline, we designed Tmb with high affinity (KD ≈ 0.65 nM) for Tie2, as confirmed by CryoEM analysis, which demonstrated that Tmb accurately recapitulates its designed structure. When conjugated to multivalent scaffolds, Tmb effectively clusters Tie2 receptors, recapitulating the signaling profile of native Ang1. Notably, high valency Tmb constructs (e.g., H8T) robustly activated pAKT and induced nuclear FOXO1 exclusion, mirroring the pro-survival and vascular stabilizing effects of Ang1, despite lacking the capacity to bind α5β1 integrin directly. Detailed cellular assays revealed that Tie2 clustering leads to the formation of two distinct complexes: a Tie2–α5β1 integrin complex that facilitates focal adhesion assembly and cell migration via pCAS recruitment, and a Tie2–tight junction complex (comprising ZO1, claudin-5, and occludin) that underpins vascular barrier integrity. Importantly, competitive binding studies demonstrated that integrin recruitment to the Tie2 complex does not require direct F-domain engagement. In human iPSC-derived diabetic blood vessel organoids, treatment with Tmb-based Tie2 agonists ameliorated diabetic vasculopathy phenotypes by reducing pathogenic collagen IV deposition, restoring tight junction organization, and lowering nuclear FOXO1 levels. These findings provide novel insights into the mechanistic interplay between Tie2, integrin, and junctional proteins, and underscore the therapeutic potential of synthetic Tie2 agonists in vascular repair and diabetic vasculopathy.

Downregulation of the mTOR complex has been shown to increase lifespan and delay development of multiple organisms, including Drosophila melanogaster. Rapamycin, an inhibitor of this complex, is undergoing FDA-approved clinical trials as a promising anti-aging drug. However the impact of genetic variation on rapamycin's response is unknown. Our study of 140+ genetically diverse Drosophila strains revealed significant variation in pupation time after rapamycin exposure, however, the underlying mechanisms of this variation remain poorly understood. Surprisingly, this sensitivity does not correlate with genetic variation in or around the mTOR gene. We therefore hypothesize that differences in phosphorylation of downstream mTOR targets may explain this variation. Currently, we are using multiple approaches to investigate how activation of downstream targets differs between highly resistant and sensitive strains. We aim to characterize the phosphoproteome of first instar Drosophila larvae from highly sensitive and resistant strains. First instar larvae were treated with rapamycin for 12 hours, followed by mass spectrometry analysis to identify phosphorylation changes in mTOR pathway targets. To validate that 12 hours of treatment induces a rapamycin response, we monitored the growth of a parallel group of larvae until 72 hours and measured their size. Sensitive DGRP strains, 348 and 517, showed a twofold reduction in length when treated with 20uM rapamycin compared to control (p-value <0.0001), while the resistant strain, 441, showed no significant decrease. Comparing the phosphoproteome of multiple resistant and sensitive lines will uncover molecular factors associated with resistance or sensitivity. Additionally, whole-larvae RNA-seq will assess the expression profile of these factors, revealing whether gene expression of tor pathway-related genes contributes to sensitivity. Understanding the mechanisms behind rapamycin resistance or sensitivity is critical for its clinical application. This project highlights the value of accounting for genetic variation in drug development, guiding future approaches for developing new drugs.

Upon nerve injury and neurodegeneration, neuron regeneration is crucial to maintain proper function. However, this natural process happens infrequently and slowly. Neuron regeneration is known to be mediated by the activity of nerve growth factor (NGF) in neurons, which binds to two receptors: tropomyosin receptor kinase A (TrkA) and p75 neurotrophin receptor (p75NTR). Previous studies have shown that engaging the receptor p75NTR activates a signaling pathway that also triggers a pain response, thus it would be ideal to have a ligand that only activates TrkA for neuron regeneration without initiating the pain response. In collaboration with the Institute for Protein Design (IPD), this study investigated an AI-designed TrkA agonist that specifically binds to and activates only the TrkA receptor. We used fibroblasts transdifferentiated into neurons as a model to study the efficiency of this TrkA agonist. Western blotting was used to study the phosphorylation of the proteins downstream of TrkA in the signaling pathway, such as pPLCγ, pAkt, and pErk, and the activity of transient receptor potential vanilloid 1 (TRPV1), a calcium channel that indicates the sensitivity of a neuron. Immunofluorescence staining was used to examine the expression of calcitonin gene-related peptide (CGRP), a neuropeptide involved in pain perception. We found that the designed TrkA agonist generates a similar level of activation of downstream proteins as NGF while successfully preventing the expression of pain response markers. Directly injecting NGF as a treatment for neurodegenerative diseases is generally not considered viable as it often induces significant pain, therefore this TrkA agonist has the potential for therapeutic use.

Directly converting fibroblasts (that make up scar tissue) into skeletal or heart muscle without a pluripotent intermediate (direct skeletal muscle or cardiac reprogramming) is one of the most promising methods for regenerating lost muscle tissue, but its low efficiency in human cells remains a significant obstacle toward clinical application. In collaboration with the Institute of Protein Design, UW, we have designed several synthetic minibinders against receptor kinases which are highly specific to their cognate receptor. Utilizing these minibinders we have created a new class of designed protein, called heterofusions, that fuse two unrelated minibinders together to force the two cognate receptor kinases together in an unnatural pairing, which could elicit novel signaling responses not achievable using natural ligands. However, which heterofusions elicit novel signaling is unknown. We aim to use direct skeletal muscle and cardiac reprogramming systems, which would benefit from this novel signaling, to screen which heterofusions elicit novel signaling to increase efficiency. To do this I developed an inducible direct cardiac reprogramming system and we also used a previously established inducible direct skeletal muscle reprogramming system to be backgrounds for screening heterofusions, with efficiency determined by imaging cardiac and skeletal muscle development makers. We found a few heterofusions, including that which brings together TrkA and BMPRII (TAB2), increased the efficiency of skeletal muscle reprogramming. I found in signaling experiments using Chinese hamster ovary cells modified to express human TrkA and BMPRII that TAB2 upregulates pERK and pCREB. Interestingly, pCREB is not part of native TrkA or BMPRII signaling, meaning novel signaling is occuring. Additionally, I have shown pCREB inhibition with a small molecule impairs direct skeletal reprogramming and TAB2’s ability to increase efficiency, showing pCREB is TAB2’s mechanism of increasing efficiency. These results show heterofusions novel signaling abilities and its applications in revolutionizing regenerative therapies.

Natural growth factors like fibroblast growth factor (FGF) are essential for maintaining pluripotency in induced pluripotent stem cells (iPSCs). However, current limitations of native growth factors include signal instability, off-target pathway activation, and dependence of xenogenic components for production. To address these issues, we developed a synthetic protein, C6-79C, which consists of six scaffolded subunits of a de novo designed FGFR1/2c binder, mb7. While mb7 functions as an FGF pathway inhibitor, the hexameric C6-79C acts as a receptor tyrosine kinase (RTK) agonist, providing more isoform-specific and prolonged signaling compared to native FGF. We formulated SynGrow, replacing FGF with C6-79C in minimal E8 media, and compared its performance against commercial media. Our study focused on three objectives: (1) comparing the expression of pluripotency markers (Oct4, NANOG, SOX2, and TRA1-60) in cells grown in SynGrow versus commercial media, and (2) evaluating morphology and viability under different media change regimens (daily, every other day, or no change). iPSCs grown in SynGrow exhibited superior morphology compared to those in mTeSR (commercial media). Pluripotency markers (Oct4, NANOG, and SOX2) were expressed at similar levels in both media, with SynGrow also showing higher expression of TRA1-60 across passages, confirmed by flow cytometry. Future evaluations will assess germ layer marker expression following directed differentiation. Our findings demonstrate that synthetic protein-based media formulations, like SynGrow, can effectively replace native growth factor-based media. This approach offers stable, prolonged, and xeno-free alternatives for stem cell culture, with broad implications for improving reproducibility and safety in regenerative medicine and cell-based therapies. 

Salivary glands are organs in the mouth which produce and secrete saliva, a multifunctional fluid crucial for processes including oral cavity lubrication, digestion, and antimicrobial functions. Diabetes mellitus has been associated with salivary gland dysfunction and harmful oral consequences including severe tooth decay and disrupted wound healing, yet it is not currently known what cell populations are affected in salivary glands and how this disease affects cell organization, function, and metabolic response. One model for diseases in human tissues are organoids, three-dimensional multicellular systems derived from stem cells which self-organize to mimic the structure and function of tissues in vivo when given the right cues. Dr. Devon Ehnes in the Ruohola-Baker Lab recently created a protocol to develop salivary gland organoids from induced pluripotent stem cells (iPSCs), and through additional culture in a high-glucose media along with inflammatory cytokines, this organoid has been used to study how diabetes affects salivary glands. Preliminary analysis has suggested acinar and ductal cell dysfunction and mitochondrial stress as causes of salivary gland dysfunction, but further work is necessary to understand how this diabetic environment leads to changes in cell function and mitochondrial activity. Here, I use a human iPSC-derived organoid model to assess how diabetic conditions affect the expression and localization of the acinar marker AMY1A, the ductal marker KRT19, the cell stress marker FOXO1, and the mitochondrial marker ATPB to determine the mechanisms for salivary gland dysfunction in diabetes.

Human tooth development is a complex and tightly regulated process that involves multiple signaling pathways and specialized proteins coordinating enamel formation. Enamel, the hardest tissue in the human body, is secreted by ameloblasts, which follow a distinct developmental process. Disruptions in these processes can lead to enamel-related disorders, such as amelogenesis imperfecta, a genetic condition characterized by defective enamel formation. A key factor in this disorder is WDR72, a gene that encodes the tryptophan-aspartate repeat domain 72 (WDR72) protein, which is critical for intracellular trafficking during enamel maturation. Although WDR72 has been studied in animal models, its precise localization and function in human fetal tooth buds remain incompletely understood. To address this question, I cryosectioned human fetal tooth samples at 19 and 22 gestational weeks and performed immunochemistry staining to visualize WDR72 alongside key enamel proteins. I performed cryosectioning to prepare thin tissue sections of each tooth bud sample, followed by immunohistochemical staining with antibodies specific to WDR72. I then imaged selected sections under a fluorescence microscope. Preliminary results suggest distinct WDR72 distribution in regions corresponding to secretory ameloblasts. These findings offer insights into the localization of WDR72 during tooth formation and lay the groundwork for future studies on the mechanisms of tooth regeneration. 

Under acute genotoxic stress, such as chemoradiation, stem cells can undergo cell cycle arrest at the G1/S phase to avoid apoptosis. This protective state, called quiescence, is reversible once stress-free conditions allow re-entry into the cell cycle to regenerate daughter cells. We have previously demonstrated a common mechanism by which two types of stem cells—Drosophila germline stem cells (GSCs) and human-induced pluripotent stem cells (hiPSCs)—enter quiescence. Recently, we found Cyclin E (CycE) associated with the outer mitochondrial membrane (OMM) in both GSCs and hiPSCs. We are interested in studying the interaction between CycE mitochondrial localization domains and mitochondrial proteins responsible for CycE localization.To map the CycE mitochondrial localization domain, I have generated four CycE truncations tagged with GFP: ΔN-terminus, ΔCyclin Box_N terminus, ΔCyclin Box_C terminus, and ΔC-terminus. I have tested these constructs in various cell lines, including Rcc4, HCT116, MCF10A, HEK, and HeLa, and found that HCT116 exhibits mitochondrial localization of CycE. I will compare the localization of wild-type CycE-GFP versus mutant CycE using immunofluorescent staining of CycE and mitochondria in HCT116, as this cell line is well-suited for transfection studies. We have shown that mitochondrial CycE is degraded in quiescent stem cells through PINK1/PARKIN-mediated mitophagy. We propose that CycE degradation is necessary for quiescence entry. In Drosophila GSCs, we observe that upon irradiation, cells overexpressing non-degradable CycE continue cell division, whereas control cells undergo quiescence. Understanding the mechanism by which Cyclin E localizes to the OMM will enhance our knowledge of how it prevents quiescence entry, thereby contributing to the development of anti-cancer treatments.