CMB Training Program

CMBTG Annual Retreat 2019

Oral Presentations

SESSION I, 9:00 – 10:00 AM

Julia Berkson . . . Prlic Lab

MAIT cell and MR1-expressing cell dynamics in healthy human colon

Julia D. Berkson1, Anneta Naidoo, Valentin Voillet, Shamiska Rohith, Erica Andersen-Nissen, Martin Prlic
1Vaccine and Infection Disease Division, Fred Hutchinson Cancer Research Center

T-cells are vital for fighting infections by coordinating and participating in the elimination of infected host cells. Recognition of foreign antigens is accomplished through peptide fragments presented by major histocompatibility complex (MHC) molecules to the T-cell receptor (TCR), which is unique for each T cell. In contrast, the totality of unconventional human mucosal-associated invariant T (MAIT) cells solely recognize microbial metabolites presented by MHC-class I-related protein (MR1). Importantly, commensal species of bacteria and yeast of a healthy microbiota generate these metabolites with agonist properties for MAIT cells. It is unclear whether MAIT cells receive a TCR signal and how their phenotype compares to the circulating population. Open questions also include whether MR1 is expressed in healthy mucosal tissues and by what cell types.

To better understand how MAIT cells behave in healthy, microbe-rich mucosal tissues, we obtained colon tissue biopsies and matched blood from healthy donors. Using flow cytometry, we define MR1 expression on colonic immune cells and surprisingly find extensive expression among various immune subsets. We also show MAIT cells in the colon have increased expression of activation markers compared to matched blood though do not display signs of an effector program. We further define the phenotype of colonic MAIT cells by RNA-sequencing and find they have a unique signature compared to blood MAIT cells as well as conventional T cell in the colon. Our data suggests that MAIT cells may be continually receiving signals from the microbiota in a healthy colon and highlights the importance of MAIT cell function at the interface of immune cell and microbial interactions during homeostasis.

Sarah Crist . . . Ghajar Lab

How do breast cancer cells overcome the ultimate suppressive microenvironment: skeletal muscle?

Sarah Crist1,2, Ilsa Coleman1, Megan Kufeld1, Patrick Paddison1, Peter Nelson1, and Cyrus Ghajar1
1Public Health Sciences Division/Translational Research Program, Fred Hutchinson Cancer Research Center, Seattle, WA,; 2Program in Molecular and Cellular Biology, University of Washington, Seattle, WA.

Skeletal muscle is a striking example of a tissue that resists metastasis: only 0.2% of malignant cancer cases involve skeletal muscle metastases. Upon careful autopsy, however, microscopic lesions in skeletal muscle are detected almost 100-times more frequently. Taken together, these data suggest that disseminated tumor cells (DTCs) successfully reach skeletal muscle and survive within this tissue, but rarely colonize it. Skeletal muscle has been recognized since the 1950s as an inhospitable tissue for tumorigenic outgrowth, yet its anti-metastatic nature remains a biological mystery. We posit that uncovering the mechanisms utilized by tumor cells to overcome growth suppression in skeletal muscle will subsequently implicate the features of the skeletal muscle microenvironment necessary to suppress DTC outgrowth. Here, we employed mouse models and organotypic cultures to identify a niche for DTCs within skeletal muscle, and factors within this tissue microenvironment that inhibit outgrowth of breast cancer cells. The generation and transcriptional analyses of skeletal muscle-metastatic mammary cancer sublines enabled our identification of genes that confer skeletal muscle-metastatic capacity and implicate molecular pathways involved for skeletal muscle-driven suppression. Candidate genes were functionally screened for their necessity in skeletal muscle colonization using the CRISPR/Cas9 system in both the organotypic co-culture and in vivo setting. Interestingly, half of the candidates that significantly promote tumor cell outgrowth in skeletal muscle are required for collagen trimerization and extracellular matrix organization. Thus, the data suggest that the remodeling of skeletal muscle into a collagen-rich environment may be critical for tumor cell colonization within this tissue. Our hope is that the thorough characterization of these identified mechanisms could inspire therapies that diminish DTC colonization in metastasis-prone sites in an enduring manner.

Janani Gopalan . . . Scott Lab

The AKAP220-PP1-HDAC6 signaling axis regulates actin-microtubule crosstalk

Janani Gopalan1, Benjamin Freedman2, Ankita Roya2, Katherine Forbush1 & John Scott1
1Department of Pharmacology, University of Washington 2Kidney research institute, University of Washington

In the kidney, primary cilia jutting into the lumen of cells lining the inner medullary-collecting duct bend with urine flow to initiate intracellular signaling events. My studies as a rotation student in the Scott lab led to an interesting discovery, that loss of the A-Kinase Anchoring protein AKAP220, correlates with a 6 ± 0.3fold increase in primary cilia number and changes in cilia morphology. Surprisingly, this effect was recapitulated in gene-edited cells that express a mutated form of the anchoring protein that is unable to target Protein Phosphatase 1 (PP1). Exploring this further, I found that histone deacetylase 6 (HDAC6), the enzyme that depolymerizes cilia, is regulated by PP1 (by dephosphorylation). Thus, I envisage a molecular mechanism whereby HDAC6 is part of the AKAP220 signaling complex and perturbing PP1-targeting by AKAP220 alters HDAC6 activity.

HDAC6 inhibitors have been used in cancer therapy, to promote epithelial traits in cells. A western blot of total HDAC6 showed that the AKAP220-KO and AKAP220-ΔPP1 cells had lower levels of HDAC6 compared to the wildtype. Western blots for epithelial markers such as E-cadherin (an adherens junction marker) and ß4 integrin (a cell-matrix junction marker) showed a drastic increase of both proteins in the mutant cells compared to wildtype.

Actin polymerization is an important precursor for cilia biogenesis as well as cell-cell junction formation. Further, HDAC6 as well as PP1 have both been shown to regulate actin polymerization. Preliminary imaging analysis showed that there is a loss of actin in stress fibers and lamellipodia of the mutant cells. Further, mutants treated with an actin stabilizing drug (to mimic the wildtype), show reduced percentage of ciliated cells as well as the levels of epithelial markers in the mutant cells. The converse experiment where wildtype cells were treated with actin depolymerizing drugs (to mimic the mutants) showed an increase in ciliated cells and levels of epithelial markers. This solidifies the idea that altered HDAC6 activity in the mutant cells causes poor actin polymerization, leading to upregulation of primary cilia and epithelial proteins in the mutant cells.
Collectively, this project will address the idea that AKAP220-PP1-HDAC6 complex regulates both actin and microtubule networks.

Ashley Hall . . . Queitsch Lab

The role of ribosomal DNA copy number in aging

Ashley N. Hall1,2, Elizabeth A. Morton1, and Christine Queitsch1.
1Department of Genome Sciences, University of Washington, Seattle WA 2Molecular and Cellular Biology PhD Program, University of Washington, Seattle WA

Ribosome biogenesis is known to affect lifespan. The ribosome complement of the cell requires copious amounts of ribosomal RNA, which is encoded in large repetitive ribosomal DNA (rDNA) arrays in all eukaryotes. In C. elegans, N2 has 100 copies of the rDNA repeat unit, while wild isolates have ~70-400 copies. Phenotypically, severe reductions in rDNA copy number that limit ribosome biogenesis can be fatal. However, it is poorly understood how rDNA copy number variation within the range permissible for life affects phenotype. We seek to determine if differences in rDNA copy number within the range typically found in C. elegans wild isolates affect aging.

To explore the potential role of rDNA copy number in aging, we have generated a set of 118 recombinant inbred lines (RILs) with high (420 copies, strain MY1) and low (130 copies, strain SEA51) rDNA copy number. We selected individuals homozygous for either 130 or 420 rDNA copies and use an automated worm imaging system (the WormBot) to determine lifespan in high-throughput. Our data will allow us to test if rDNA copy number affects lifespan, and if rDNA copy number interacts with other genetic loci. Our data indicate that the RILs show a wide range of lifespans. We are currently exploring what genetic variation contributes to this variation in lifespan.

To supplement this RIL resource, we backcrossed the 130-copy rDNA array into the background of the parental 420-copy donor, and the 420-copy array into the parental 130-copy donor. These strains allow us to isolate rDNA copy number from other genetic variation. We will use this resource to test the effect of rDNA copy number variation on healthspan traits less amenable to high-throughput testing.

SESSION II, 10:15 – 11:15 AM

Alex Pollock . . . Woodward Lab

Production and Application of a FRET-Based Protein Biosensor to Detect Intracellular cGAMP Accumulation at the Single Cell Level.

Alex Pollock1,*, Shivam Zaver2,*, Joshua Woodward1
1Microbiology Graduate Program, 2Molecular and cell Biology Graduate Program, University of Washington

Innate immune signaling is in a renaissance largely due to the recognized importance of cancer immunotherapy. cGAMP is an innate signaling molecule made by cGAS upon recognition of viral or other mislocalized cytoplasmic DNA which initiates an IFN- inflammatory response. Additionally, cGAMP has been found to be anti-tumorigenic both itself and in synergy with immunoblockade therapy. Despite recognition of the importance of this signaling molecule there is minimal understanding of the kinetics and population level stochasticity of cGAMP production largely due to the lack of a sensitive single cell tool to quantitate intracellular levels of cGAMP.

We have produced nSTING – an intramolecular protein-based FRET biosensor which fills this gap and will allow for new avenues of research. nSTING is based on the cGAMP binding protein STING, which was truncated, given optimized linkers, and fused to the FRET pair mKO2-Teal. Our in vitro studies show that nSTING robustly, reversibly, and specifically responds to cGAMP. We have successfully translated this tool into mammalian cell culture with similar responsiveness and kinetics. We are currently exploring the limits of this sensor to determine the ability to use nSTING as a tool to study cGAMP localization, cellular stochasticity, as a drug screening platform, and as a CRISPR screening platform to identify new regulators of cGAMP production.

Brittany Ruhland . . . Reniere Lab

Whole-cell proteomics in Listeria monocytogenes reveals novel binding partners for chaperone protein YjbH

Brittany R. Ruhland1 and Michelle L. Reniere1
1University of Washington Department of Microbiology

Protein degradation is an essential regulatory mechanism employed by bacteria to facilitate rapid responses to environmental changes, such as oxidative stress. The Gram-positive, facultative intracellular pathogen Listeria monocytogenes (Lm) is an excellent model to investigate the oxidative stress response, as this bacterium faces abrupt changes in oxidative stressors during transitions between host environmental niches. Recently, the uncharacterized protein YjbH was shown to disrupt the post-transcriptional regulation of a Lm virulence factor. In other Firmicutes, YjbH homologs mediate the degradation of the redox-responsive transcriptional regulator Spx. This work demonstrates for the first time the interaction of Lm YjbH and Spx proteins, and shows eight additional proteins with which YjbH interacts. These nine proteins are all more abundant in a ∆yjbH background compared to wildtype, as determined by whole-cell proteomics. Given the described role of YjbH as an adaptor protein that increases degradation of Spx in the closely related Bacillus subtilis, we propose that YjbH in Lm acts as a general chaperone protein and that through this role YjbH is indirectly involved in proper redox and virulence regulation mechanisms. YjbH has additionally been described as a thioredoxin family protein due to its conserved Cysteine-X-X-Cysteine motif, although thioredoxin activity has never been shown to be critical to YjbH function. Using the bacterial-two-hybrid system to measure protein-protein interactions, in parallel with in vitro oxidative stress and virulence assays, we suggest here that the cysteines of YjbH are critical for its proper structure, but are not participating in the well-characterized thioredoxin activity described for C-X-X-C motifs. This work highlights the importance of further study on the general protease adaptor YjbH to fully understand redox and virulence regulation in Lm.

Frank Soveg . . . Savan Lab

Differential subcellular targeting determines the antiviral specificity of OAS1

Frank W. Soveg1, Johannes Schwerk1, Katharina Esser-Nobis1, Alison Kell1, Adriana Forero1, Jonathan Clingan1, Daniel B. Stetson1, Saumendra Sarkar2, Michael Gale Jr.1 and Ram Savan1
1Department of Immunology, University of Washington, Seattle, WA. 2Department of Microbiology and Molecular Genetics, University of Pittsburgh, Pittsburgh, PA.

Oligo adenylate synthetase (OAS) proteins comprise an ancient antiviral system important for viral restriction. Upon sensing viral RNA, the catalytically active OAS proteins generate the second messenger 2-5A, which activates the latent ribonuclease RNaseL. Activated RNaseL potently restricts viral replication by degrading cellular RNAs to block host translation and induce apoptosis. Human OAS1 is spliced into several different isoforms, but it is unclear if these isoforms have overlapping or unique functions during a viral infection. A SNP in the splice acceptor site of exon 6 in the OAS1 gene (A/G, rs10774671) is associated with West Nile Virus resistance. The susceptibility allele (A) controls p42 expression and the resistance allele (G) controls p46 expression, which suggests OAS1 isoforms have different functions in antiviral immunity. Interestingly, p46 contains a C-terminal membrane-targeting CaaX motif. Proteins containing CaaX motifs at their C-termini are prenylated and targeted to membranes. Our study shows the p46 isoform localized to the Golgi apparatus, while p42 and other OAS1 isoforms are present in the cytosol. We found the differential localization of OAS1 isoforms was dependent on the CaaX motif. Strikingly, we observed p46 is reassorted to viral replication organelles (VROs) during West Nile and Zika virus infection. We hypothesize the membrane-targeting CaaX motif on p46 allows the protein to infiltrate viral replication organelles, giving it enhanced access to viral RNA. We show p46, compared to cytosolic OAS1 isoforms, possesses significantly stronger antiviral activity against RNA viruses that replicate on cellular membranes. Collectively, these data show isoform-specific prenylation localizes OAS1 isoforms to unique subcellular compartments and defines their antiviral functions. In summary, this work highlights how subcellular targeting of antiviral proteins is an important feature of the host-pathogen interface. 

Dylan Udy . . . Bradley Lab

Novel cellular quality control mechanisms to prevent accumulation of truncated proteins

Dylan B. Udy1,2,3, Qing Feng1,2,3, and Robert K. Bradley1,2
1Computational Biology Program, Public Health Sciences Division, Fred Hutchinson Cancer Research Center 2Basic Sciences Division, Fred Hutchinson Cancer Research Center 3Molecular and Cellular Biology Graduate Program, University of Washington, Seattle, WA 98109, USA

Cells have evolved many elaborate strategies to control gene expression and homeostasis. One such strategy is nonsense-mediated mRNA decay (NMD), which is a conserved eukaryotic cellular surveillance system that identifies and degrades mRNA transcripts containing premature termination codons (PTCs). This quality control pathway acts to prevent the accumulation of mRNAs that would otherwise be translated to produce non-functional or deleterious truncated proteins. In addition to its canonical role in mRNA quality control, NMD has been found to impact gene expression through the degradation of mRNAs that have been alternatively spliced to include a PTC and mRNAs that appear to encode functional, full-length proteins.

Although the presumed purpose of NMD is to prevent aberrant truncated protein accumulation, previous studies have limited their measurements to mRNA levels and degradation rates rather than direct measurements of the proteins translated from those transcripts. Whether cells have additional mechanisms to degrade or prevent production of these proteins, which we term “NMD peptides,” is unknown. We hypothesize that cells utilize specific molecular mechanisms to target and degrade NMD peptides, NMD transcripts are translationally repressed to prevent NMD peptide production, and the truncated protein sequence confers inherent instability to the NMD peptide. To test these possibilities, we designed reporters to measure NMD transcript and peptide levels when cis elements and trans factors are modulated and used computational analyses to examine features shared among endogenous NMD peptides that could facilitate their degradation. These results provide evidence for novel protein quality control mechanisms intimately related to a well-characterized mRNA quality control pathway.


11:30 – 12:30 PM

Dr. Kellie Bickel

Senior Director of Intellectual Property
TwinStrand Biosciences

Kellie Bickel, Ph.D. is the Senior Director of Intellectual Property at TwinStrand Biosciences, a young, dynamic company that is enthusiastically developing technology for a new generation of high-accuracy clinical diagnostics and innovative tools for scientific discovery. At TwinStrand Biosciences, she is responsible for developing and implementing an Intellectual Property (IP) strategy to protect TwinStrand’s core Duplex Sequencing™ technology, as well as increase and drive value from the company’s IP assets. She works with and advises the company’s executive team and Board of Directors on all IP matters, including: patent and trademark procurement and enforcement; IP portfolio growth and management; and IP business development. Kellie enjoys working closely with both research and business teams to identify and develop key patent positions around life science technology.

Kellie has previously worked in nationally-recognized legal teams at both Perkins Coie, LLP and at the University of Washington Center for Commercialization where she partnered with a variety of clients from start-ups to large, publicly-traded companies to tailor IP pursuits to business goals. Kellie holds a Ph.D. in Molecular & Cellular Biology from the University of Washington where she was the recipient of a National Science Foundation pre-doctoral fellowship and a Cellular and Molecular Biology, NIH pre-doctoral grant.

Dr. Tracie Delgado

Associate Professor of Biology
Seattle Pacific University

Dr. Tracie Delgado graduated with her PhD in Microbiology in 2011.  For the last 8 years, she has been a professor at Northwest University in Kirkland WA.  Starting this fall, Dr. Tracie Delgado will be an Associate Professor of Biology at Seattle Pacific University.  Her research focuses on understanding how viruses cause cancer.  She has mentored 11 undergraduates in her lab and many of them are in medical and graduate school.  Her teaching focus is in the field of Molecular and Cellular Biology.  She is a member of the Society for the Advancement of Chicanos and Native American’s in the Sciences (SANCAS) and is an advocate for increasing the number of underrepresented minorities in the sciences. 

Dr. Yesenia Green

Business Development Specialist
EMulate Therapeutics

Yesenia Green joined EMulate Therapeutics in January 2017, as a Business Development Specialist. Dr. Green leverages her strong scientific background to support the company’s partnering and preclinical strategies, partnership negotiations, and strategic alliances. Prior to joining EMulate Therapeutics, Dr. Green was Associate Director of Capital Formation and Partnerships for Life Science Washington.

Dr. Green received her PhD in Molecular and Cellular Biology from University of Washington. During her time at University of Washington, Dr. Green also worked as a Technology Licensing Fellow for CoMotion.

Poster Session

SESSION A, 12:30 – 1:15 PM

Joanna Maltbaek . . . Stetson Lab

Human DNA-PK activates a STING-independent DNA sensing pathway

Katelyn Burleigh1, Joanna H. Maltbaek1, Stephanie Cambier1, Richard Green1, Michael Gale, Jr.1,4, Richard C. James2,3, and Daniel B. Stetson1,4*

1Department of Immunology, University of Washington School of Medicine
2Center for Immunity and Immunotherapies, Seattle Children’s Research Institute
3Department of Pediatrics, University of Washington School of Medicine
4Center for Innate Immunity and Immune Disease, University of Washington School of Medicine

Detection of intracellular DNA by the cGAS-STING pathway activates a type I interferon-mediated innate immune response that protects from virus infection and can be harnessed to promote anti-tumor immunity. Whether there are additional DNA sensing pathways, and how such pathways might function, remains controversial. We show here that humans – but not mice – have a second, potent, STING-independent DNA sensing pathway that is blocked by the E1A viral oncogene of human adenovirus 5. We identify human DNA-PK as the sensor of this pathway and demonstrate that DNA-PK kinase activity drives a robust and broad antiviral response. We discover that the heat shock protein HSPA8/HSC70 is a unique target of DNA-PK. Finally, we demonstrate that detection of foreign DNA and DNA damage trigger distinct modalities of DNA-PK activity. These findings reveal the existence, sensor, unique target, and viral antagonists of a STING-independent DNA sensing pathway (SIDSP) in human cells.

Vanessa Montoya . . . Emerman Lab

A role for the SETD1B histone trimethyl transferase in HIV Replication

Vanessa Montoya and Michael Emerman

At different points in the HIV life cycle, host cellular proteins are hijacked by the virus to establish and enhance infection. To identify genes that encode these dependency factors, the Emerman Lab has developed a virus-packageable HIV-CRISPR screen that targets the entire human genome. We screened for HIV dependency factors with two viral strains, one isolated late in disease progression, that uses CXCR4 as a co-receptor, and another isolated early in infection that uses CCR5 as a co-receptor. In addition to the receptor CD4, and the respective co-receptors that came through the screen as expected dependency factors, the SETD1B, scored as one of the highest hits for both strains (ranked 4th and 5th). SETD1B is a component of a histone methyltransferase complex known to produce trimethylated histone H3 at Lys4, a marker for transcriptional activation. However, its role in HIV has not been described before. To validate this observation, I knocked out SETD1B in a T cell line and consistently observed an average 5.7-fold decrease of HIV infection. During the viral life cycle, the HIV genome integrates into the cellular chromosome at transcriptionally active sites of chromatin. Therefore, I hypothesize that during HIV infection, SETD1B plays a role in regulation of HIV transcriptional activation, or alternatively that it actively methylates one of the viral proteins in a manner that favors virus replication. Experiments are underway to distinguish these two hypotheses and describe the role of SETD1B in virus replication.

Jeet Patel . . . Wills Lab

Hif1a regulates expression of posterior patterning genes during vertebrate development and regeneration

Jeet H. Patel1,2, Evan E. Takayoshi2, Andrea E. Wills2
1Molecular and Cellular Biology Program, University of Washington 2Department of Biochemistry, University of Washington

Proper patterning of tissues during development requires precise spatial and temporal coordination of numerous signaling pathways. These cues that must be reestablished in regenerative organisms to correctly replace lost structures. While many embryonic signaling pathways, such as Notch, Wnt, and FGF, are necessary for regeneration, it is not well understood how these pathways are activated in response to injury. Here, we begin to bridge the gap between stress signaling and the regenerative response though a comparative analysis of the role of the transcription factor Hif1a. Hif1a is a known regulator of metabolic and stress responses. Here, we sought to identify shared and differential regulatory roles of this transcription factor in the complementary processes of tail formation and regeneration in the frog, Xenopus tropicalis. Perturbation of Hif1a stability during development using genetic, biomedical, and environmental tools leads to pronounced defects in anteroposterior patterning. Increased Hif1a leads to embryonic posteriorization, while loss of Hif1a leads to anteriorization. Hif1a loss of function leads to reduced expression of canonical Wnt target genes, including Wnt-dependent organizer genes at the gastrula stage, as well as posterior patterning markers during neurulation. During tail regeneration, we show that Hif1a is necessary for proper outgrowth of the regenerate. Analysis of markers identified in development shows that Hif1a is necessary for activation of similar targets in both posterior development and regeneration. This suggests a novel model where Hif1a regulates gene expression programs downstream of Wnt signaling which lead to proper formation and patterning of the tail.

Samantha Schuster . . . Hsieh Lab

The Role of Somatic 3’UTR Mutations in Prostate Cancer

Prostate cancer is the most commonly diagnosed cancer in men, killing almost 30,000 men yearly. Almost all of these men die from metastatic, castration-resistant prostate cancer (mCRPC), which is currently incurable and incompletely understood. ¬In order to better treat these patients, new models of prostate cancer progression must be developed. Importantly, though coding sequences mutations have been widely studied in the field of cancer genomics, significantly less is known about the effect of somatic mutations in regulatory regions on cancer progression. One regulatory region of particular interest is the 3’ untranslated region (UTR), which controls many post-transcriptional processes including mRNA stability and translation efficiency, as these have been previously associated with prostate cancer. We propose that elucidating the effects of somatic mutations on translational regulatory regions of the genome, namely the 3’UTR, may uncover novel paradigms of prostate cancer progression.

In order to probe the function of the 3’UTR in prostate cancer, we obtained and sequenced genome-wide UTRs in matched cancer and normal tissue samples for 230 prostate cancer patients. We performed somatic mutation calling on these samples, discovering 13,104 different 3’UTR mutations in 6,526 genes across the patient cohort. Many of these mutations are found in multiple patients and in known cancer-related genes, indicating that they might contribute to cancer progression in these patients. Some 3’UTR mutations also cluster in potential 3’UTR hotspots indicative of functional motifs, such as RNA-binding protein or miRNA binding sites. Many mutations are also associated with functional changes of either transcript abundance or translational efficiency. I will take both a candidate gene approach and a screen-based approach to functionally test and validate these mutations. First, I will test a set of high-confidence, recurrent mutations for whether they induce post-transcriptional changes and characterize top hits by generating endogenous mutants in prostate cancer cell lines. I will also perform a massively parallel functional screen of all the recurrent 3’UTR mutations we discovered to determine how they change mRNA stability and translation efficiency on a genome-wide scale. We hypothesize that using these approaches, we will be able to determine how 3’UTR somatic mutations contribute to prostate cancer progression, both at molecular and genome-wide levels.

Ami Yamamoto . . . Cheung Lab

Investigating the metastatic potential of tumor cell clusters

Ami Yamamoto1,2, Emma Wrenn1,2, Breanna Moore1, Kevin Cheung1
1Translational Research Program, Public Health Sciences and Human Biology Divisions, Fred Hutchinson Cancer Research Center 2Molecular and Cellular Biology Graduate Program; University of Washington

Metastasis has conventionally been thought of as a single-cell process. However, more recent studies in our lab and others have established that tumor cell clusters give rise to the majority of metastases in different cancer mouse models. Importantly, the presence of circulating tumor cell clusters in patient blood samples have been shown to correlate with poorer prognosis in multiple cancers, yet the mechanisms that underlie clusters’ metastatic advantage remain largely unknown. Recently, I have used tail vein injection assays of highly metastatic breast cancer cells to show that tumor cell clusters exhibit a 536-fold increase in metastatic potential compared to single cells. Turning to gene expression to dissect molecular difference between tumor cell clusters and single cells, I have analyzed RNA-sequencing data generated by our lab comparing gene expression between single and clusters. From this analysis, uPAR, a multidomain glycoprotein that is part of the plasminogen activation system, is the top hit for genes upregulated in clusters compared to single cells in mouse and human breast tumor models. I have generated shRNA knockdowns (KD) of uPAR or uPAR ligand, uPA, in mouse primary breast tumor organoids. Relative growth of uPAR-KD or uPA-KD organoids was significantly lower compared to controls. Taken together, my studies suggest that uPAR and uPA may contribute to the metastatic advantage of clusters.  

SESSION B, 1:15 – 2:00 PM

Marcella Cline . . . Zweifel Lab

Genetic disruption of Dcc and Ntn1 expression in VTA dopamine neurons produces significant behavioral changes

Cline, M. M1,3, Hunker, A.2, Juarez, B.2, Zweifel, L1,2.
1 University of Washington, Seattle; Department of Psychiatry and Behavioral Sciences , 2Department of Pharmacology , 3 Molecular and Cellular Biology

Netrin1 signaling through the netrin receptor deleted in colorectal cancer (DCC) promotes growth cone attraction, axon elongation and branching, and synaptic development during central nervous system development. Emerging evidence suggests DCC and secreted netrins may play an additional role in maintaining excitatory synaptic connections in the adult brain. As neurodevelopmental regulators, netrin 1 and DCC are highly expressed in many brain regions throughout neurodevelopment and their expression decreases in adulthood. Interestingly, however, the adult ventral tegmental area (VTA) maintains a high expression of both netrin1 and DCC. Netrin 1 and DCC co-label with TH expression suggesting expression in dopamine neurons of the VTA, which are critical modulators of a variety of behaviors. To explore the function of netrin1 and DCC in adult VTA dopamine neurons, we used a combination of viral-mediated, cre-inducible CRISPR/Cas9 technologies and transgenic mice to selectively disrupt Ntn1 and Dcc expression exclusively in midbrain dopamine neurons of adult mice. Eight-week old DAT-IRESCre/+ mice were injected bilaterally into the VTA with AAV-FLEX-saCas9-HA-sgNtn1 or -sgDcc. Control DAT-IRESCre/+ mice received bilateral injections of AAV-FLEX-YFP. Four weeks following viral injections, mice were screened for alterations in dopamine-mediated behaviors. Open field testing revealed Ntn1 and Dcc conditional knockout (cKO) mice spent significantly less time in the arena center compared to YFP controls, and significantly more time on the arena edge, suggesting the promotion of an anxiety-like phenotype. Locomotor behaviors measured across three nights and two days showed significantly reduced day distance traveled in Ntn1 and Dcc cKOs compared to controls on day one, suggesting a hypolocomotive phenotype with genetic disruption of Ntn1 and Dcc. Curiously, Dcc cKO, but not Ntn1 cKO, mice displayed significant impairment in novel object recognition and social recognition. These data suggest that both Ntn1 and Dcc play an important role in regulating specific aspects of dopamine-regulated behaviors through their actions in dopamine neurons of the adult VTA. Future experiments will be aimed at establishing a role for these proteins in maintaining excitatory synaptic connectivity in dopaminergic midbrain neurons in the adult brain.
This investigation was supported in part by Public Health Service, National Research Service Award, T32GM007270, from the National Institute of General Medical Sciences.

Jesse Hansen . . . Kollman Lab

Molecular and cellular basis for assembly of the metabolic filament CTP Synthase

Jesse M. Hansen1,2, Eric M. Lynch2, Avital Horowitz2, Joel Quispe2, Dan Farrell2, Justin Kollman2
1Biological Physics, Structure, and Design (BPSD) Graduate Program, University of Washington 2Department of Biochemistry, University of Washington

Compartmentalization of biochemical processes is a distinguishing feature of the cell. Enzymes have typically been depicted as largely disordered collection of freely diffusible molecules, but this view has been challenged in recent years. While many metabolic processes are well-known to be confined to membrane-bound organelles that sequester enzymatic reactions, more recent work has highlighted the importance of directed self-assembly of enzymes in organizing metabolism. At least a dozen core metabolic enzymes form filaments in vivo, but for most the molecular mechanisms of assembly remain unclear. CTP synthase (CTPS) catalyzes the final and rate-limiting step of CTP de novo biosynthesis, and is inhibited by its product CTP. The enzyme assembles into filaments in every organism studied so far, suggesting an important role in the cell. Fluorescence-tagging studies have shown that enzyme polymerization is generally a stress response, which suggest that filaments are important for maintaining homeostasis.  We use yeast as a complete system to study structure, biochemistry, and in vivo assembly of the prototypical metabolic filament CTPS.  Here we show that yeast CTPS forms filaments in both an active and inactive states, and polymerization is induced by lowering pH. To better understand why CTPS assembles into filaments, we have solved four high resolution cryo-EM structures corresponding to the active/inactive states of both yeast CTPS isoforms. Unexpectedly, the yeast filament geometry and assembly contacts show drastic differences in comparison to the human protein.  Using these structures as guidance, we have generated non-polymerizing and hyper-polymerizing mutants of yeast CTPS.  Preliminary characterization of these mutants has been done, and their GFP-tagged versions have been introduced into yeast to probe in vivo consequences of CTPS assembly.  This work is beginning to unravel the complexities of CTPS oligimerization, and will provide a framework for future studies to better understanding the importance of other metabolic filaments.

Kristin Holmes . . . Berger Lab

The catalytic Ras G-domain forms dimers

Kristin Holmes, Alice Berger

Ras family genes (K-, N- and H-RAS) encode small GTPases and are mutated in ~30% of human tumors. KRAS mutations are found in approximately 30% of lung adenocarcinomas and over 90% of pancreatic cancers. Currently no small molecule inhibitors have been established as effective therapeutics in patients with KRAS-driving mutations. As a result, patients with KRAS-mutated lung cancer are primarily treated with chemotherapy, or with PD-1/PD-L1 immunotherapy. Recently it has been shown that KRAS forms homodimers, and KRAS homodimers are required for mutant KRASdriven tumorigenesis. Our preliminary mass spectrometry data identified NRAS as a KRAS interacting protein, raising the possibility that Ras proteins can also heterodimerize. Here we propose the hypothesis that KRAS and NRAS heterodimerize, and this KRAS/NRAS heterodimer modulates lung adenocarcinoma progression. I have employed the biochemical assay of size exclusion chromatography to investigate the ability of KRAS and NRAS to physically heterodimerization. By using size exclusion chromatography, I have been able to resolve putative Ras dimers from monomeric fractions, indicating the innate ability of the catalytic Ras G-domain to form dimers. The results of our studies will make a significant impact on the field of Ras biology and may uncover opportunities to abrogate Ras heterodimers for therapeutic benefit.

Jessica Huang . . . Gerner Lab

Monocytes promote the generation of effector T cells through localized IL-12 production in draining lymph nodes

Jessica Huang1, Joseph Leal1, Karan Kohli1, Michael Gerner1
1University of Washington, Department of Immunology

Cells of the innate immune system are integrally involved in the generation of adaptive immunity. Particularly, conventional dendritic cells (cDCs) are known to mediate T cell activation and differentiation in lymph nodes (LNs) during inflammation. While other innate cell subsets can contribute, their roles remain less well-defined. Here, we utilize immunization models and Type-1 inflammatory adjuvants to study the responses of different innate cell populations in the draining LNs. We found that within hours of immunization with distinct TLR agonists, monocytes rapidly migrated to the draining LNs in large numbers and differentiated into monocyte-derived DCs (MoDCs). These MoDCs further infiltrated the deep T cell zone where they physically interacted with T cells undergoing activation by cDCs. While MoDCs did not capture large quantities of antigen, they did constitute a major source of IL-12 production in the T cell zone, suggesting their likely role in T cell differentiation. Indeed, early-effector T-betHI T cells were preferentially enriched in regions heavily infiltrated by MoDCs, suggesting localized T cell polarization. Blockade of monocyte trafficking into LNs with a CCR2 blocking antibody resulted in decreased IL-12 levels in the T cell zone, and in a significant reduction in effector CD4+ and CD8+ T cells 4 days post-immunization. Together, these data suggest that during the generation of adaptive immune responses, monocytes in draining LNs create a localized spatial niche and cooperate with cDCs to promote the production of optimally differentiated effector T cells.

Sigal Kofman . . . Oberst Lab

Elucidating the role of necroptotic signaling in Zika virus induced pathology

Sigal B. Kofman1, Brian P. Daniels1,2, and Andrew Oberst1
1Department of Immunology, University of Washington 2Department of Cell Biology and Neuroscience, Rutgers University

In the recent epidemic, Zika virus (ZIKV) infection during pregnancy has been linked with congenital abnormalities, including microcephaly, intrauterine growth restriction (IUGR) and fetal demise. However, the mechanism underlying these birth defects remains unknown. Type I interferons (IFN) are central to a potent anti-viral immune response, and while this response is key to preventing dissemination of virus, it can also exacerbate immune pathologies. In a study investigating the role of IFN signaling in ZIKV-induced birth defects, Akiko Iwasaki’s group demonstrated that fetal IFN signaling led to resorption and IUGR.  Specifically, ZIKV infected fetuses with intact IFNAR signaling were resorbed between E10.5 and E12.5. Concurrently, our group has found that ZIKV infection can activate the nucleotide sensor ZBP1 and the RIP kinases, components of the necroptotic cell death pathway. This is notable because ZBP1 is potently IFN-inducible. Furthermore, in genetic models of unrestrained RIPK signaling, lethality is observed at the same developmental stage and with the same tissue involvement as that triggered by ZIKV-induced, IFN-dependent IUGR.  Together, these observations lead us to the hypothesis that engagement of necroptosis contributes to ZIKV-induced IFN-mediated embryonic pathologies.

Keynote Speaker: Dr. Matt Welch

Mobilization of the cytoskeleton by microbial pathogens

Professor and Head
Division of Cell & Developmental Biology
Department of Molecular and Cellular Biology
University of California, Berkeley

Dr. Matt Welch received his PhD from UC Berekely in 1993. He started his lab at UC Berekeley in 1998. His research focuses on how the actin cytoskeleton in cells drives changes in shape to cells and organelles. Dr. Welch has mentored trainees of all levels- undergraduates, graduate students and postdocs, of which all are still pursuing scientific careers. He is a member of the American Society for Rickettsiology and an elected fellow to the American Academy of Microbiology. Dr. Welch also serves on the editorial boards for Cell and The Journal of Cell Biology.