Past & Current Research Projects

Increased cellular metabolism and oxygen concentrations are characterized by a deleterious increase in reactive oxygen species (ROS) termed oxidative stress, which can damage cellular macromolecules. Paradoxically, low oxygen concentration, known as hypoxia, also produces a burst of intracellular ROS. For this project, we aimed to characterize the similarities & differences between the genomic response to oxidative stress-linked ROS and hypoxia-linked ROS. Specifically, we used CRISPR to mutate a key ROS-responsive transcr-iption factor, NRF2, and have identified genes that respond to oxidative stress/hypoxia in an NRF2-dependent or independent manner. Surprisingly, our findings suggest that the proportion of genes altered in hypoxia that require the presence of NRF2 is similar to the proportion requiring NRF2 in oxidative stress; however, the NRF2-dependent genes altered in these two cellular conditions are largely context-specific, being active in altered significantly in only one condition or the other.

manuscript in preparation

Previous research has investigated links between physiological/ psychosocial stress and a transcriptional network controlled by the transcription factor glucocorticoid receptor, or GR.  However, most of this research has looked specifically at gene expression changes within stress- and reward-linked brain circuits, including: the hypothalamic-pituitary-adrenal axis, hippocampus, para-ventricular nucleus, prefontal cortex, amygdala and striatum.  The cerebellum is a brain region that was previously thought to be involved mainly in motor learning and 'muscle memory', but is now increasingly recognized to be important in social behaviors and cognition, which are altered in stress.  In this project, we have collaborated with neurophysiologists to understand the genomic consequences of psychosocial stress and GR activation on the cerebellum.  We have correlated epigene-tic marks of gene activity (H3K27ac) with gene expression data to identify key molecular changes induced by GR activation/stress in the cerebellum.  We have also compared these data with publicly available single cell data to help distinguish the primary cell populations that may be impacted by these changes.

manuscript in preparation

Amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) are debilitating, progressive neurodegener-ative diseases with almost no current treatments. Importantly, though ALS and FTD are characterized by degeneration of distinct regions of the brain/CNS, recent insights have shown a shared etiology, genetics, and even clinical co-occurrence of these diseases. The Fernandez-Funez lab uses the fruit fly as a model organism for studying neurodegenerative disease such as these and prion diseases. For this project, they conducted a genome-wide RNAi screen of fly genes (orthologous to human genes) to identify genetic modulators of two ALS/FTD disease models: overexpression of a C9orf72 hexapeptide (PR36) or TDP-43M337V in fly retinal neurons. We found a large number of potential modifiers of ALS/FTD (>300 in total), which include both known and novel genes. These genes also converge on both known and novel biological mechanisms associated with these diseases, including mitochondrial dysfunction, proteostasis, cellular signal-ing, and more. I specifically have helped in taking this well-curated set of PR36/TDP-43 modifiers (both enhancers and suppressors) and have integrated: 1) protein and genetic interaction data using the STRING database, 2) RNA-seq datasets to determine clinical relevance, 3) functional annotation to determine the molecular and biological processes these modifiers converge on, and 4) network analysis using Cytoscape and R to identify subnetworks and functional modules within the data.

manuscript in preparation

NRF2 is a transcription factor (TF) that regulates expression of genes that protect against the deleterious effects of oxidative stress. Its activity is induced by oxidative stress, but this is not the only context in which NRF2 can be activated. Mutations in NRF2, or the pathway that activates it, can also lead to increased NRF2 activity (termed hyperactivation) and promote oncogenesis. Using an integrative genomics approach, I identified the downstream targets of hyperactive NRF2 in cancer and the ultimate consequences of their dysregulation. This approach revealed a core set of 32 direct targets that are consistently upregulated in NRF2 hyperactivated tumors, which include canonical redox-related NRF2 targets (NQO1, GCLC, etc.), as well as several target genes that have not been previously linked to NRF2 activation. The key distinguishing feature of these NRF2 “cancer target” genes is that they are regulated by “strong” NRF2 DNA binding sequences found in regions with ubiquitously permissive chromatin signatures. 

Relevant publications:

1. https://doi.org/10.1016/j.redox.2018.07.026

After discovering that the NRF2 targets upregulated across diverse cancer types are largely determined by their permissive  chromatin environment (see NRF2 target genes and activity across cancers), I next endeavored to find if a similar mechanism governed the response to acute oxidative stress. Thus, I explored the transcriptional response to acute oxidative stress using paired ATAC-seq and RNA-seq time course experiments. Utilizing a variety of analytical techniques, including:  principal component analysis, differ-ential expression/accessibility, and TF footprinting analysis, I found: 1) the transcriptional response to acute oxidative stress also takes place within a largely predetermined accessibility land-scape, and 2) a wide variety of additional transcription factors respond to oxidative stress, including AP-1, p53, SP/KLF and NF-Y factors.

Relevant publications:

1. https://doi.org/10.1016/j.freeradbiomed.2021.05.016

The fruit fly salivary gland (SG) is a useful model of both secretory organ function and of tubulogenesis (the development of tubular organs, which frequently are secretory). For this project, the Andrew lab studied how a set of transcription factors (including CrebA, Ribbon, and others) help to coordinate the devel-opment and secretory capacity of the SG. We found that CrebA and Ribbon bind to the promoters of ribosomal protein genes in the SG to boost translational capacity, but this type of direct promoter-binding is largely absent in another tubular organ (the trachea). I specifically helped in: 1) identifying binding targets of CrebA, Ribbon, etc using well-established analysis pipelines, such as the ENCODE ChIP-seq pipeline and MACS2, 2) integrating multiple data sources, such as ChIP-seq and microarray experiments, to identify high-confidence SG-specific target genes, and 3) functional annotation of the identified genes using sources such as gene ontology, Reactome, KEGG, DAVID, etc.

Relevant publications:

1. https://doi.org/10.1111/tra.12753

2. https://doi.org/10.1083/jcb.202110073

Previous studies have established that Nrf2 expression is limited in certain types of neurons, however no study has shown that Nrf2 is globally restricted in most (indeed nearly all) neurons of the brain and CNS. In this project and associated study, we utilized a variety of single cell RNA-seq datasets to show that Nrf2 expression is largely only found in non-neuronal cell populations of the CNS, and that this trend is conserved in mouse and humans. Furthermore, we show that these expression patterns aren't merely coincidence and that they are likely driven by upstream 'regulatory logic'- i.e. a restrictive chromatin environment at the Nrf2 promoter (and a nearby enhancer) in neurons, which is more variable/permissive in non-neuronal populations. Taken together, these findings imply there is a functional, and evolutionarily conserved basis for limited Nrf2 expression in neurons that should be further explored.

Relevant publications and code:

1. https://doi.org/10.1016/j.redox.2023.102830

2. https://github.com/levin252/NRF2_brain_single-cell-analysis 

As the environmentally destabilizing impacts of climate change continue to worsen, there has been a marked increase in the frequency and scale of wildfires globally. Despite  this, the impacts of wildfire smoke inhalation on human health remain understudied and under-appreciated. Here, in collaboration with a state-of-the-art wildfire smoke exposure facility in Montana, we studied the long-term and acute physiological and genomic effects of wildfire/wood smoke inhalation (specifically the PM2.5 comp-onent of smoke) on mouse lung immune cells called alveolar macrophages (AM). We found that wood smoke exposure resulted in a variety of changes to physiological processes (such as cytokine release and efferocytosis) and gene expression changes in AMs in male mice. There were less pronounced effects (both physiological and genomic) in female mouse AMs however, indicating a novel sex-specific response to wood smoke exposure.

Relevant publication:

1. https://doi.org/10.1080/08958378.2024.2354398