A Concrete Abstract

Hey readers! This week’s blog post is a draft of my abstract for my summer project. With our poster presentation in less than two weeks, my projects are beginning to wrap up, so I wrote this draft to encapsulate the work I’ve done so far in a concise paragraph:

There are current inadequacies in the understanding and treatment of brain metastases. This study explored the role of ABL kinase, a non-receptor tyrosine kinase, in lung to brain metastasis with the goal of identifying novel targets for anti-metastatic therapy. We explored this concept using two approaches: 1) assessing the role of ABL in an EML4-ALK fusion cell line (unexplored to date) and 2) screening genes linked to ABL and examining their role in promoting metastasis. These questions were explored in vitro, either in the EML4-ALK fusion cell line or in a brain metastatic-derived cell line (respectively for each approach). Many properties of the EML4-ALK fusion line were found to be impacted by ABL, identifying ABL as an important, and targetable, regulator of this cancer subtype. In comparison to the parental cell line, multiple genes were shown to be upregulated in the brain metastatic-derived cell line. Furthermore, many of these targets were significantly attenuated in an ABL knockdown line. These results implicate ABL as an influential part of the EML4-ALK fusion cancer type, as well as establish novel mechanisms of ABL-regulated metastasis.

Next week I’ll be reflecting on one of the faculty talks we have had so far this summer. See ya then!


Amitifadine, a Triple Re-uptake Inhibitor, Reduces Self-administration of the Opiate Remifentanil in Rats: Abstract

Drug addiction is do to neural systems that may be altered by different drugs. Similar drugs that can affect nicotine self admin can also affect opioid self-administration. One such drug is amitifadine. We have found a correlation between reduced remifentanil use both acute and chronic when being treated with amitifadine doses of 5, 10 and 20 mg/kg. Repeated treatment with 10 mg/kg of amitifadine reduced remifentanil self-administration chronically, even after cessation of treatment. There was no significant affect on feed motivated responding when being treated with the 10 mg/kg amitifadine dose. Amitifadine extended and maintained anti-nociceptive effects while not attenuating remifentanil-induced analgesia. These studies with amitifadine show promise as the drug can be used to reduce opioid self-administration in patients while not negating the pain killing properties that make opioids so desirable. Further studies are needed to determine the efficiency of amitifadine as a drug that combats opioid addiction.



(Levin, et al. 2019)


Curiosity, the desire for new information, has been shown to encourage exploration and benefit learning. While the relationship between curiosity and episodic memory seems intuitive, the present study examined this relationship with active engagement as a mediating variable. In this study, participants watched videos of continuous line drawings wherein an object was formed as the video progressed. Participants were encouraged to guess frequently during the video regarding their prediction of the final image. After each video concluded, participants reported a final guess and provided subjective ratings of how they experienced the video, including how curious, frustrated, and surprised they felt. A memory test was conducted 24 hours following the first task, in which participants were asked to identify partial images from the videos of the previous task versus novel partial images taken from other unseen videos. It is expected that videos that elicit higher ratings of curiosity will have a higher rate of recall. Videos with more guesses, indicating more active engagement, are also expected to have higher recall. Future experimental conditions will examine how the regulation of agency will further impact curiosity and memory. Delineating the interaction between curiosity and memory may ultimately improve learning in educational settings.


Place cells are a type of pyramidal neurons in the hippocampus that activate in a pattern that encodes for space as an animal moves through its environment. Our current project examines how the brain reacts to unexpected visual manipulations by looking at place cell activity. We hypothesize that the greater the visual spatial manipulation, the greater the place cell activity will deviate from normal activity. A genetically encoded indicator recorded neural activity in the mouse’s brain by binding to calcium ions and fluorescing. An algorithm called constrained non-negative matrix factorization identified areas of fluorescence that may represent neurons, but also generated many false positives. Eliminating these false positives is necessary for a more accurate analysis of place cell activity. The percentage of false positives will also allow us to reexamine the effectiveness of the algorithm and identify areas of improvement to generate more accurate neuron selections. By analyzing only the neurons that represent place cells, we hope to better understand how the brain uses vision and how vision is being used to correct for errors. In particular, these findings may have significant impact in disease models, such as Alzheimer’s research, in which the patients have difficulty with episodic memory.


Brachyury is a transcription factor that has been determined to be an essential component in the course of sea urchin embryonic development. We know that it is expressed at both the blastopore and stomodael opening of the embryo, but inactivates in cells that leave those respective areas. Additionally, in its absence gastrulation does not occur. What makes this protein so integral to these morphogenetic movements, however, remains a mystery. We hope to find an answer to this question by examining how brachyury plays a role in gene transcription and protein synthesis. In this study, we will be investigating a number of genes known to be associated with the formative stage of gastrulation and analyzing the change in their expression with brachyury both present and knocked down. Through in-situ hybridization, we will observe the location and expression patterns of each gene in regards to the presence of brachyury, and use that information to draw associations. We hope to eventually generate a detailed and extensive gene regulatory network on the brachyury gene in order to establish a deeper understanding of the processes that prompt its behavior. 


The neurobiological mechanisms of the visual system involves the primary visual cortex (V1) transmitting retinal input to multiple higher visual areas (HVAs) such as the posteromedial (PM) area. Each V1 or HVA neuron has a unique receptive field, or region of sensory space that affects the neuron’s firing rate in response to incoming stimuli. When the stimulus enlarges to a certain size, the average firing rate of individual neurons will drastically decrease; however, it is unclear why this pattern, or surround suppression, is not as radical in PM neurons. We hypothesize that V1 axonal inputs converge, or overlap, more to PM than those to alternative HVAs. We inject fluorescent tags, tdTomato, in-vivo in mouse V1 and wait for the tags to express in V1 to HVAs axons. The brains are then sliced, and tdTomato is imaged in-vitro under a microscope. MatLab is then used to analyze the width of the fluorescent tdTomato area, or the width of axon spread for each HVA. We expect PM’s axon spread width will be larger than the other HVAs, suggesting higher convergence of V1 to PM input. These findings will ultimately benefit our neurobiological understanding of the visual circuits that lead to perception.


Generating a library of mutant enzymes to degrade plastic

As plastic pollution accumulates in the ocean, it becomes a growing threat to marine and human health. Ideonella sakaiensis is a recently discovered bacterium that digests polyethylene terephthalate (PET), the plastic primarily found in single-use water bottles. This bacterium secretes two enzymes, PETase and MHETase, to digest PET and use it as a source of carbon and energy. The process by which PETase and MHETase digest PET is currently too inefficient to be useful for pollution clean-up. The goal of my project was to generate a library of mutant PETase enzymes in order to create a variant of PETase capable of digesting plastic at a rate rapid enough for practical use. To generate this library, we used error-prone PCR, in which Taq and Mutazyme polymerases introduce random mutations while replicating the PETase gene. Individual sequences from the PCR product were obtained through molecular cloning using a 2.1-TOPO vector system and competent DH5alpha E. coli, followed by Sanger sequencing. Sequencing confirmed that error-prone PCR successfully introduced mutations into the PETase gene. Furthermore, successive rounds of error-prone PCR caused an accumulation of mutations within the gene. These mutant PETases will be tested for efficiency in PET digestion in a future study.


The activation of type I interferons has long been reported to regulate pain, but the precise mechanism has remained unclear. We tested the hypothesis that type I interferons are critical pain regulators and function via a neuronal mechanism. Sensory, motor, and behavioral testing in addition to immunohistochemistry were performed on type I interferon receptor knockout (Ifnar1 KO) and wild type (WT) mice. Heightened sensitivity in KO mice suggests an importance of type I interferons in regulating pain under basal conditions. The lack of any difference in motor performance and quantity of nerve fibers in tissue of WT and Ifnar1 KO, along with RNA scope, support a neuronal-specific mechanism of interferons in regulating pain. These findings suggest that type I interferons are critical in regulating both pain and neuroinflammation by a neuronal mechanism, even under normal, healthy conditions. These findings redefine our understanding of the role of interferons in pain and could translate to improved treatment of a heterogeneous array of chronic pain.

Abstract: Multiplexing the D4 Assay for HER2 and GADPH Detection

Breast cancer is the leading cause of cancer mortality among women. The majority of cases and resulting deaths occur in low-resource settings. Effective breast cancer treatment requires a detailed assessment of the tumor. Unfortunately, the lack of clinical infrastructure in low-resource settings prohibits the use of traditional breast cancer pathological methods. The D4 assay is a miniaturized, self-contained assay that can measure protein biomarkers from complex biological milieu with high sensitivity and specificity without the need for equipment (other than a smartphone) and can be performed with minimal user training. Previously, a D4 assay for HER2 detection from fine needle aspirate (FNA) samples was developed. However, due to variation in FNA sampling, it is important to normalize HER2 concentration in clinical settings. GADPH, a housekeeping protein, provides a normalization standard. Thus, I constructed a multiplexed assay that simultaneously detects HER2 and GADPH. The assay was tested for cross-reactivity and optimal antibody concentrations. We found that there is minimal cross reactivity between HER2 and GADPH. Furthermore, multiplexing does not compromise the analytical performance of the test. The ability to multiplex HER2 and GADPH will make the D4 assay a more accurate diagnostic tool to enable effective breast cancer treatment.


Here is my drafted abstract for my poster, I can’t wait to read about everyone’s research and results!

Male mice produce ultrasonic vocalizations (USVs) during mating, but it is not known how inhaling the helium affects the USVs besides changing the frequency or whether changes in USVs are accompanied by changes in mating behaviors. It is also unknown whether the animal gradually corrects the vocal and behavioral changes after continuously experiencing helium, such as trying to lower the pitch of their voice. Through placing mice in an air-filled or heliox-filled chamber and recording their vocalizations and behavior, this study tests the hypotheses that 1) behavior changes in helium, 2) USVs get shorter and quieter, and 3) the mice show evidence of learning by correcting changes in their USVs. We predict that while inhaling helium, males will display lower proportions of sexually aggressive mating behaviors. Additionally, we predict that the USVs produced during helium sessions will have higher frequencies, lower amplitudes, and have fewer syllables in one bout of vocalizing. Finally, with our current data we do not expect to find evidence of the mice correcting the vocal and behavioral changes. Future studies will continue to research this third hypothesis, which if supported, would have broader implications for the neurobiology behind differentiation of self-produced sounds from external environmental sounds.


Parkinson’s Disease is a common neurodegenerative disease characterized by the death of dopaminergic neurons, which can be caused by oxidative stress in the mitochondria and subsequent DNA damage. It is known that the mutated variant of the LRRK2 gene causes Parkinson’s Disease; however, the role of the unmutated gene, outside of its kinase activity, is relatively unknown. We seek to determine the role of the unmutated LRRK2 gene under oxidative stress and DNA damage by simulating these conditions with toxins in both wild type and CRISPR edited LRRK2-KO cell lines. We will be using western blots to measure the differences in phosphorylated proteins known to respond to DNA damage and different levels of cyclins and related proteins that may contribute to the cessation of the cell cycle. We expect to find that the resistance to DNA damage and the percent of cell death is different in the two cell lines. We also expect to find that the LRRK2-KO cell line, under stressful conditions, creates a checkpoint in the cell cycle that stops growth and division. Understanding LRRK2’s role and its pathways can help us better understand how the increased kinase activity of the mutated gene causes Parkinson’s Disease.

Abstract: How does DIP-alpha contribute to the development of the olfactory circuit in Drosophila?

The assembly of neural circuits is a highly complex process that has not yet been fully understood. The olfactory system of Drosophila offers an excellent model system to better understand this organization, as 50 classes of olfactory receptor neurons (ORNs) synapse with their target projection neurons within 50 class-specific and uniquely positioned glomeruli in the brain’s antennal lobes. By analyzing the antennal transcriptome data from four stages of the antennal lobe development, we have identified a gene family of interest containing DIPs and DPRs, which are known to have homophilic and heterophilic interactions that regulate the assembly of several neural circuits. Through an RNAi mediated screening of this Ig Superfamily, we hypothesized DIP-alpha is important to the organization of ORNs into specific glomerular formations via axon terminal guiding.  We used several genetic methods to test this hypothesis, including the downregulation and knockdown of DIP-alpha in all ORNs and explored glomerular formation in various DIP-alpha mutant flies. In addition, we used genetic labelling techniques and antibody staining to identify the expression pattern of DIP-alpha. Although we were unable to identify a clear expression pattern, our results from the glomeruli structures in flies where the function of DIP-alpha was genetically perturbed suggests that DIP-alpha is important to glomerular formation. Further study of its roles in class sorting will give us more insight into the molecular basis of this complex circuit assembly.


CTNND2 is an astrocyte-enriched Autism risk gene that encodes for the protein delta-catenin. Preliminary experiments revealed that knockdown of delta-catenin in astrocytes impairs astrocyte survival and decreases astrocyte processes both in vitro and in vivo. Astrocyte morphology coincides with synapse formation during cortical development. Delta-catenin was assumed to be neuron-specific and has been implicated in modulating cadherin-based homophilic interactions between pre and post-synapses. We hypothesis that delta-catenin stabilizes astrocyte-neuron adhesion signalling via cadherin molecules in order to control astrocyte and synapse development. We have verified the presence of endogenous Ctnnd2 mRNA in both cortical and hippocampal astrocytes through fluorescence in situ hybridisation and detail our attempts to visualize N-cadherin through immunohistochemistry.


Although dystonia is the third most common movement disorder and causes muscles to contract uncontrollably, the exact mechanisms for dystonia are poorly understood. Past research on multiple forms of dystonia have implicated phospho-eIF2a pathway activity in the brain as a central source of dysfunction. This pathway is involved in responding to cellular stressors and mediating synaptic plasticity responses in the brain. The goal of this study is to identify the brain regions and developmental periods in which the pathway’s activation is disrupted in a DYT1 mouse model of childhood-onset dystonia. Western Blot analysis was used to determine the expression levels of phospho-eIF2a and total eIF2a of DYT1 and control mice at various time points in development including p0, p5, p14, p21 and p30 and in various brain regions including the striatum, cortex, cerebellum, and midbrain to determine where pathway dysregulation was most predominant. This knowledge will advance our understanding of the cellular mechanisms of dystonia and provide proof-of-principle experiments to determine whether targeting this pathway is beneficial.


In Drosophila Melanogaster (fruit flies), wingless(wg) a Wnt growth factor responsible for cell to cell communication operates in two parts, a signal transduction and movement (of this signal) to other neighboring cells. What remains unclear to researchers who study wingless is how wingless protein is moved and the molecules that help move it. The Bejsovec Lab’s hypothesis is that molecules that aid the movement of mutant Wg protein will be able to influence the wild type Wg protein. In order to test this hypothesis, cuticle preparations of mutant for wingless and these suppressor genes were scored for improved cuticle patterning in double mutant embryos. In order to confirm that the changes in cuticle pattern were due to increased movement of the wgNE2 protein, antibody staining was performed at early developmental stages. The findings of this project may uncover that these molecules are able to aid not only mutant wingless but wild type as well.


This week, I’ve prepared an early draft of the abstract for my poster:

The anterior cingulate cortex (ACC) is a cortical brain region implicated in a variety of functions, including emotional self-regulation and the reward pathway. This project utilizes optogenetic stimulation techniques to both excite and inhibit the ACC in the mouse brain. Two opsins, channelrhodopsin and Guillardia theta anion channelrhodopsin 2 (GtACR2), were used to excite or inhibit the ACC respectively. Food deprived mice were trained to press a lever under different fixed ratio schedules. The mice were then given the same lever pressing task during intermittent periods of high frequency stimulation. Preliminary analysis of the lever press rate indicates a clear reduction in the amount of presses during stimulation periods when compared to nonstimulation periods in mice with the excitatory channelrhodopsin expressed, with no similar reduction in control mice. Given that mice are pressing less when the ACC is excited, there is reason to believe that, absent of motor impairment, the ACC is implicated in the effort-reward decision making process in the rodent brain.

Abstract thinking

Inhibition of HipA to Reduce Multidrug Tolerance in E. coli

The HipBA operon is a bacterial toxin-antitoxin module that plays a crucial role in multidrug tolerance in E. coli.  HipA, the toxin, is a kinase that functions by phosphorylating translation factors, inhibiting translation in the cell and inducing a state of dormancy in which cellular processes that are targeted by antibiotics are shut down, allowing the cell to evade antibiotic poisoning.  The goal of my research is to identify molecules that bind to HipA and inhibit its kinase activity.  HipA autophosphorylates, so its activity can be measured by its phosphorylation.  To measure phosphorylation, wild-type HipA is completely dephosphorylated using a phosphatase enzyme, and its phosphorylation after the addition of ATP can be visualized through a ProQ Diamond Phosphoprotein Gel Stain.  This assay can be repeated with wild-type HipA in the presence of target molecules that are believed to inhibit its kinase activity.  A lack of autophosphorylation in the presence of a molecule indicates that this molecule inhibits HipA and could be a potential molecule of study for the development of an antibiotic.

There are Good Days, and There are Bad Days for Science

Hello everyone! For this week’s blog entry I’ll bring you through a day in my life working in the Bilbo/ Eroglu lab. My day begins with an 8am wake-up and a half hour of frantic pre-work preparations. Somehow during this time I manage to shower, dress, make breakfast, pack lunch and run out the door to catch the 8:30am Swift Express bus to West Campus. This bus is crucial. Missing the Swift Express means being late for the 9am start of the work day. 

Once in lab, I check in quickly with my mentor, a final year PhD student, Carina, about my daily schedule. Because most of my time is spent on software/ equipment owned through Duke’s Core Research facility (expensive and/ or large equipment which the University only owns a few of and are thus shared between all of the labs on campus) I am able to dictate my own daily schedule based upon availability of the Core machines. It is through this University program that I am able to image slices of brain tissue on a Zeiss 880 Confocal (which is super cool!), deconvolude these images (reduce background noise and make more clear) on a software called Huygens, and create 3D reconstructions of the microglia I image in software called Imaris. 

By 10am I am set up to use one of these softwares – Zen (used on the confocal microscope), Huygens, and Imaris which takes up most of my days, depending on what stage in analysis I am in. While I am grateful that the University has licenses to these programs which allow me to visualize microglia cells and synapses within the brain, it is incredible how much of my day is spend troubleshooting such expensive software. 

By noon, Imaris has probably crashed at least twice (see attached photo), and Huygens failed to process my images. While I troubleshoot or re-do lost work, I eat lunch, usually snacking throughout the day as opposed to taking a long break. I take “lunch” time to organize data which I’ve acquired, schedule more time on Duke Core Research machines, or chat with lab members. 

3:30pm: Coffee time!

You can usually find me in lab until around 6pm, after which I’m either too hungry to go on without proper dinner, or too annoyed at how long a particular program is taking to render my data. I trek home, catching the Swift Express back.

In the evening I gym, cook dinner, and watch some Netflix until I crash for the night, ready to do it all over again tomorrow!

Sometimes if I wish hard enough, it responds

Me with the scope!