RESEARCH
FALL 2014 | PENNSCIENCE JOURNAL 1
Contents Features Disorder: A Review of Our 06 Bipolar Understanding of the Disease The Genetics of Depression 08 Correlation between 10 The Alcoholism and Mood Disorders Neurostimulation 12
14 Anxiety: The Feeling of Being Chased
Interview Dr. Jeffery Greeson
16
Professor of Clinical Psychology Perelman School of Medicine
Research and Cyclic Axial Mechanical Properties of Porcine 18 Single Common Carotid Artery Development of a single-plasmid system for screening site-specific DNA methylases
24
Characterization of the Neuroblastoma Driver 28 Epigenetic Genes ARID1A and ARID1B
Cover image: edited CC Image courtesy of 55Laney69 on Flickr under an Attribution 2.0 Generic CC License.
2
PENNSCIENCE JOURNAL | FALL 2014
RESEARCH
Editorial Staff EDITORS-IN-CHIEF Vivek Nimgaonkar Donald Zhang
WRITING MANAGERS Rami Ezzibdeh Natalie Neale
EDITING MANAGERS Ishmam Ahmed Kartik Bhamidipati Karanbir Pahil
DESIGN MANAGERS Courtney Connolly Carolyn Lye
Angela Chang Zoe Daniels Joseph Gao Jenna Harowitz Michal Hirschorn Ila Kumar Ajay Patel Grace Ragi Samip Sheth Kejia Wang Edward Zhao
RESEARCH
BUSINESS MANAGER Claudia Cheung
EDITING
Sharon Kim Yishan Liu Josh Tycko
WRITING
Coby Basal Ritwik Bhatia Sebastian de Armas Richard Diurba Lucy Li Antoinette Radcliffe
LAYOUT
Emily Chen Grant Shao
BUSINESS
Anand Desai Tina Huang Samip Sheth
FACULTY ADVISORS
Dr. M. Krimo Bokreta Dr. Jorge Santiago-Aviles
About PennScience PennScience is a peer-reviewed journal of undergraduate research published by the Science and Technology Wing at the University of Pennsylvania. PennScience is an undergraduate journal that is advised by a board of faculty members. PennScience presents relevant science features, interviews, and research articles from many disciplines including biological sciences, chemistry, physics, mathematics, geological sciences, and computer sciences. PennScience is a SAC funded organization. For additional information about the journal including submission guidelines, visit www.pennscience.org or email us at pennscience@gmail.com. FALL 2014 | PENNSCIENCE JOURNAL 3
Letter from the Editors Dear Readers, It is our pleasure to bring to you the first print issue of the 13th volume of PennScience. For this issue, we have written a series of feature articles around the topic of mood disorders. Psychiatric disorders and the biological basis of our mind’s function have proven to be major challenges in biomedical research despite the efforts of countless researchers in the field. We hope that we are able to share with you some of the complexity inherent in this area of scientific research while also illustrating some of the strides that have been made in unraveling and decomposing those complexities. This year we were fortunate to have an amazing staff, and we are excited for you to be able to see some of the products of their work in exploring mood disorders over the semester. Our first feature article of the issue is a review of the literature on Bipolar disorder. In the article, Antoinette Radcliffe sheds some light on the current standards for diagnosis and definition of the disease, and in doing so, she gives us a chance to consider the psychiatric affliction from a purely medical and scientific approach that avoids the stigma sometimes attached to public descriptions of the disorder. Lucy Li follows this with an examination of the genetics behind depression, noting the interactions between environmental and hereditary factors in the origination of depression. In the third article, Coby Basal also delves into the literature to investigate and understand the relationship between alcohol and mood disorders. Our fourth feature article offers a consideration of neurostimulation as Ritwik Bhatia weighs the benefits of the approach. In our final feature article, Richard Diurba analyzes anxiety, writing about some of the biological underpinnings of stress and the clinical frameworks for anxiety. We are extremely lucky to also have an interview with Dr. Jeffrey Greesen, whose research and work at the Hospital of the University of Pennsylvania is rooted in clinical psychology. In addition to the articles in the print issue, we will also be releasing an article by Sebastian de Armas, reviewing the various treatment options existing for mood disorders, on our website as part of our expanded web content associated with this fall issue. We are also honored to share three research papers in this print issue. Yishan Liu presents a study conducted upon porcine arteries to assess the effects of trauma. Our second and third research articles in the issue deal with DNA methylation. Josh Tycko discusses an approach undertaken to develop a screen for site-specific DNA methylases using a single plasmid system. Sharon Kim offers a study of the epigenetics associated with ARID1A and ARID1B neuroblastoma cell lines. This semester, PennScience has also been considering other ways to foster scientific discussion on campus. We held PennScience’s first-ever coffee chat with a professor. In holding coffee chats, we hope to give students a place to talk to science professors outside of the classroom about their research and their careers in science. We were lucky to be joined by Professor Philip Rea of the biology department this semester, and we are excited to host more coffee chats in the spring, when we are also hoping to launch additional web content and a journal club series. Throughout this semester, we have enjoyed the opportunity to work with everyone connected to the journal, and we would like to thank the many groups and individuals who make PennScience possible. To all of the staff and managers for the journal, we extend our thanks for their continued drive and energy. We would also like to thank the Student Activities Council and the Science and Technology Wing for continued support of the journal. We are also extremely grateful to our fantastic faculty advisors, Dr. Krimo Bokreta and Dr. Jorge Santiago-Aviles for their consistent guidance and backing. We finally would like to thank the Penn faculty and student community for engaging with the journal and for joining us in furthering the scientific discourse at Penn. Sincerely, Vivek Nimgaonkar and Donald Zhang Co-Editors-in-Chief
4
PENNSCIENCE JOURNAL | FALL 2014
Call for Submissions Looking for a chance to publish your research? PennScience is accepting submissions for our upcoming Spring 2015 issue! Submit your independent study projects, senior Design projects, reviews, and other original research articles to share your work with fellow undergraduates at Penn and beyond. Email submissions and any questions to pennscience@gmail.com.
Research in any scientific field will be considered, including but not limited to:Â
Biochemistry | Biological Sciences Biotechnology | Chemistry Computer Science | Engineering Geology | Mathematics | Medicine Physics | Psychology
FALL 2014 | PENNSCIENCE JOURNAL 5
FEATURES
BIPOLAR DISORDER:
A Review of Our Understanding of the Disease by Antoinette Radcliffe
Introduction: Bipolar Disorder (BD) is the sixth leading cause of disability in the world, affecting about 5.7 million American adults (1). One in five patients with BD commits suicide, reducing their lifespan by 9.2 years (2). BD, or manic-depressive illness, is a mood disorder characterized by its extreme mood swings and affects more than 1 percent of the population (3). The illness usually begins in adolescence and affects men and women equally (1). Although the cause remains unknown, having a relative with Bipolar Disorder increases one’s risk for the illness (4). The term “bipolar” is currently used as a colloquialism for moody; yet, BD is a severe mood disorder and it is necessary to understand its diagnosis, treatment, and genetic basis in order for one to spread awareness for the disease. Diagnosis: For a diagnosis to be made, the patient must be exhibiting major deviations from his or her normal behavior. In a manic state, one might experience distraction, recklessness, poor judgment and temper, little need for sleep, and irritability. In a depressive state, one could notice sadness, fatigue, feelings of worthlessness, changes in appetite, low self-esteem, and thoughts of death or suicide (5). These states can occur together (mixed state), right after each other (rapid cycling), or a long time after each other, as the pattern is different for each individual with the disorder. There appear to be four main types of BD. Bipolar I Disorder is defined by either a manic or mixed episode lasting at least seven days or requiring hospitalization due to severe mania. It is often accompanied by a depressive episode lasting at least two weeks. On the other hand, Bipolar II Disorder consists of depressive and hypomanic episodes with no extensive manic or mixed states. Bipolar Disorder Not Otherwise Specified (BP-NOS) is diagnosed when the symptoms exist but fail to meet the criteria for Bipolar I or II. Cyclothymic Disorder, or Cyclothymia, is a mild form of BD in which hypomanic and mild depressive episodes occur for at least two years, but do not fulfill the diagnostic requirements for any other type of bipolar disorder. A less typical, severe form of the illness called Rapid-Cycling Bipolar Disorder is characterized by four or more episodes within a year (6). BD also has many comorbidities, 6
PENNSCIENCE JOURNAL | FALL 2014
FEATURES including substance abuse, anxiety disorders (especially posttraumatic stress disorder and social phobia), attention deficit hyperactive disorder, thyroid disease, migraine headaches, heart disease, diabetes, and obesity (7, 8). The psychiatric and medical communities have created the impression that BD has risen in its incidence and prevalence. However, because BD relies solely upon clinical diagnosis, lacking pathophysiologic indicators, it is difficult to prove this potential over-diagnosis. Some have also critiqued the categorical distinction between schizophrenia spectrum disorders due to growing evidence from longitudinal and genetic studies (9). Genetics: More than two thirds of those affected by BD have one close relative with the disorder or with unipolar major depression (1). In the largest linkage sample for BD, 1,152 individuals from 250 families with BD and related affective illnesses were analyzed using genome linkage. Researchers discovered significant evidence of linkage at chromosome 17q and chromosome 6q, and suggestive evidence of linkage was observed on chromosomes 2p, 3q, and 8q (6). In 2007, researches assembled and validated a database of phenotypic variables from families with BD. The database contains 197 variables on 5,721 subjects in 1,177 families. Within these 5,721 subjects, the clinical presentation of BD varied greatly. They discovered that many phenotypic variables are familial and some quantitative variables are heritable, arising from shared genotypes. This database is available to the entire research community to explore the connections between phenomenology and genetics of BD (10). Treatment: Because there is no cure for BD, it requires continuous, lifelong treatment (11). Psychotherapy aims to provide education, guidance, and support to patients and their families. The most commonly used types of psychotherapy for BD are cognitive behavioral therapy, family-focused therapy, and interpersonal and social rhythm therapy. Cognitive behavioral therapy teaches patients how to change negative or harmful behaviors and thought patterns. Family-focused therapy aims to enhance family coping strategies by improving familial communication and problem solving. Interpersonal and social rhythm therapy improves the relationships of those with BD and manages their daily routines (12). Medications are used in conjunction with psychotherapy. Most psychiatrists typically first treat BD with mood stabilizers, such as lithium, anticonvulsants, and valproic acid, which decrease mood swings. Anti-depressants are used to combat the symptoms of depressive episodes. Common antidepressants used for BD are fluoxetine (Prozac®), paroxetine (Paxil®), sertraline (Zoloft®), and bupropion (Wellbutrin®). However, anti-depressants can increase the risk of mania or hypomania. To reduce this risk, anti-depressants are often paired with a mood stabilizer or atypical anti-psychotics. To treat mania, olanzapine (Zyprexa®) is often used. Aripiprazole (Abilify®) treats manic or mixed states. These anti-psychotics can be taken as a pill or an injection. Injections are often used only in severe cases, whereas pills are often used for maintenance treatment (13).
A controversial alternative treatment is electroconvulsive therapy (ECT), formerly known as shock therapy. ECT provides relief for those with severe BD who have yet to benefit from other treatments. Today, the patient takes a muscle relaxant and is put under anesthesia before ECT is administered. The treatment usually lasts from thirty to ninety seconds and patients recover within five to fifteen minutes. ECT is also used for those experiencing bipolar symptoms along with other medical conditions like pregnancy, in which medication use is risky (14). Conclusion: Overall, much progress has been made in standardizing the diagnosis and enhancing the treatment of BD. However, as with many mood disorders, the stigma of mental illness has kept both the public and medical community focused solely on the symptoms. The National Institute of Mental Health’s creation of the Research Domain Criteria (RDoC) aims to classify psychopathology by neurobiological measures and observable behavior. The goal is to outline basic dimensions of functioning, analyzed from behavior to neural circuits to genes. When completed, the RDoC will improve the understanding of psychopathy and its treatment (15). However, educating the public will likely be the most effective approach to reduce the stigma associated with BD and its insulting colloquial use. Resources:
1. R. C. Kessler, et al., Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry.62, 593–602 (2005). 2. K.R. Jamison, Suicide and Bipolar Disorder. J Clin Psychiatry.61, Suppl 9:47-51 (2000). 3. N. Craddock and P. Sklar, Genetics of bipolar disorder: successful start to a long journey. Trends in Genetics.25, 99-105 (2009). 4. P. Sklar, et al., Large-scale genome-wide association analysis of bipolar disorder identifies a new susceptibility locus near ODZ4. Nat Genet.43, 977–983 (2011). 5. Diagnostic and Statistical Manual of Mental Disorders (American Psychiatric Association, Fifth Edition. Washington, DC, 2013). 6. D. M. Dick et al., Genomewide Linkage Analyses of Bipolar Disorder: A New Sample of 250 Pedigrees from the National Institute of Mental Health Genetics Initiative. Am J Hum Genet.73, 107-114 (2003). 7. K. R. Krishnan, Psychiatric and medical comorbidities of bipolar disorder. Psychosom Med.67, 1–8 (2005). 8. D. J. Kupfer, The increasing medical burden in bipolar disorder. JAMA.293, 2528–2530 (2005). 9. I. Iordache, N.C. Low, The overdiagnosis of bipolar disorder. J Psychiatry Neurosci.35, E3-E4 (2010). 10. J.B. Potash, et al., The Bipolar Disorder Phenome Database: A Resource for Genetic Studies. Am. J. Psychiatry.164, 1229–1237 (2007). 11. D. J. Miklowitz, A review of evidence-based psychosocial interventions for bipolar disorder. J Consult Clin Psychol.67, 28–33 (2006). 12. D. J. Miklowitz, et al., Psychosocial treatments for bipolar depression: a 1-year randomized trial from the Systematic Treatment Enhancement Program (STEP). Arch Gen Psychiatry.64, 419–426 (2007). 13. M. E. Thase, G. S. Sachs, Bipolar depression: pharmacotherapy
FALL 2014 | PENNSCIENCE JOURNAL 7
FEATURES
The Genetics of
DEPRESSION By Lucy Li
W
hen we think about depression, we often consider symptoms and environmental stressors. Yet popular discourse often fails to touch upon the genetic basis of the disorder. In scientific discourse, there is much discussion about expression of genes and their effects on phenotype, especially in relation to disease. The rise of genetic screening in the past decade has made research on diseases with a genetic basis more viable. However, the understanding of the genetic basis of depression is still in its early stages. Although it is widely believed that Major Depressive Disorder (MDD), also known as clinical depression, can have a genetic basis, the details have been widely contested. Because depression is defined by how a person feels, diagnosis is traditionally more difficult than simple biological testing. Genetic analysis of MDD will lead to better understanding of the disorder and can potentially open a new path for diagnosis and treatment. For quite some time, it has been noted that depressive disorders tend to run in families. Family-focused studies have shown that the relatives of people with MDD have almost three times the risk of being diagnosed when compared to those without a family history of MDD (1). In comparison, the familial risk associated with schizophrenia has a proportion
8
PENNSCIENCE JOURNAL | FALL 2014
FEATURES that is ten-fold greater (2). Studies of genetically identical twins have shown that genes can account for one-third to two-thirds of the risk of developing MDD, with the rest of this risk resulting from non-genetic environmental factors (2). Unlike diseases that follow a simple Mendelian pattern of inheritance, MDD potentially involves complicated interactions between multiple genes that are presently not fully understood. Most genomics studies conducted for MDD have been either genome-wide association studies (GWAS) or structural variation (SV) studies (3). GWAS are typically done to observe genetic variation from individual to individual in a population to find any variation of DNA associated with a particular trait, while SV studies look at the variation in structure of chromosomes (4,5). These studies have shown that certain regions, such as parts of chromosomes 1, 11 and 12, may contain susceptibility loci (2). They have also allowed scientists to look at differences between genes of normal and diseased phenotypes. There are many potential genes involved in the mechanism of MDD. Currently, six main genes have been linked to MDD: MTFHR (related to folate metabolism), SLC6A3 (dopamine transporter), DRD4 (dopamine receptor type 4), GNB3 (G-protein beta subunit 3), SLC6A4 (serotonin transporter), and 5-HTTLPR (serotonin transporter) (2). This relationship is easily understood since neurotransmitters potentially play a role in the development of MDD. In particular, the serotonin transporter gene, 5-HTTLPR, is important. A 44-base pair insertion/deletion in 5-HTTLPR, which occurs in the promoter region (where initiation of transcription begins on DNA), is a polymorphism that may cause symptoms of MDD by affecting how serotonin is transported from the synaptic cleft to the pre-synaptic neuron (6,7). However, conclusions have been difficult to draw due to conflicting data and a lack of robust studies (2,6,8). Many factors contribute to the weakness of these studies, including small sample sizes, population stratification, rare genomic variation, and lack of control for environmental factors (8). Beyond pure genetics, it is believed that interactions between genetic predispositions and the environment cause the development of mental illnesses like MDD. A study done by Capsi et al., found that people with shorter versions of 5-HTTLPR can be up to two times more likely to be diagnosed with depression after stressful events such as loss of a job or death of a loved one (6). Similar numbers were reported for the effect of childhood malnutrition on MDD. Childhood abuse was another environmental factor found to significantly increase the probability of depression later in life for those with the same 5-HTTLPR polymorphism (6). In the past, depression has been linked to anxiety and depressive-traits, which have in turn been associated with allelic variation in the 5-HTTLPR function. However, the interactions between genetics and the environment are complex, and MDD cannot be predicted by genotype alone. Another complexity in accurately identifying the genetic basis of depression is the inherent difficulty in the diagnosis of depression. Psychological disorders
have traditionally been diagnosed and categorized through the Diagnostic and Statistical Manual of Mental Disorders on the basis of the symptoms experienced and the progression of the disease, even though these boundaries have been difficult to define (9). More recent findings through GWAS studies provide evidence that psychological disorders may be more interlinked than previously believed. Some studies suggest that some SNPs (Single Nucleotide Polymorphisms) show cross-disorder effects such as bipolar disorder, MDD, and schizophrenia (9). Thus, the same variation in genes can potentially lead to increased risk of multiple disorders, since an alternation of one biological system can have many effects on downstream molecular processes. The interconnected nature of psychological disorders has traditionally been overlooked but has the ability to greatly advance our understanding. The potential for treating genetically based MDD is still largely unexplored. Depression is one of the most common psychiatric disorders, yet treatments are still unsatisfactory. Another potential area of study involves analyzing the genetics behind current antidepressants, which could lead to a more personalized treatment of MDD (10). In the future, large-scale, genome-wide studies with greater sample sizes are needed to further understand the relationship between genetics and MDD (8,2). Increased knowledge of the genetic factors behind MDD could contribute to an improvement in diagnosis and prevention of depression as well as the discovery of molecular targets for a new generation of psychotropic drugs. Bibliography 1. M.C. Neale, L. R. Cardon, Methodology for Genetic Studies of Twins and Families (Springer, New York, 1992). 2. J.M. Hettema, Genetics of Depression. FOCUS.8, 316322 (2010). 3. P.F. Sullivan, M.J. Daly, M. O’Donovan, Genetic Architectures of Psychiatric Disorders: The Emerging Picture and Its Implications. Nature Reviews Genetics.13, 537-551 (2012). 4. W.S. Bush, J.H. Moore, Genome-Wide Association Studies. PLoS computational biology.8, e1002822 (2012). 5. L. Feuk, A.R. Carson, S.W Scherer, Structural Variation in the Human Genome. Nature Reviews Genetics.7, 8597 (2006). 6. D.F. Levinson, The Genetics of Depression: A Review. Biological psychiatry.60, 84-92 (2006). 7. B. Schneider, D. Prvulovic, Novel Biomarkers in Major Depression. Current opinion in psychiatry 26, 47-53 (2013). 8. E.C. Verbeek, et al., The Genetics of MDD–A Review of Challenges and Opportunities. Journal of Depression and Anxiety.3, doi:10.4172/2167-1044.1000150 (2014). 9. Cross-Disorder Group of the Psychiatric Genomics Consortium, Identification of Risk Loci with Shared Effects on Five Major Psychiatric Disorders: A GenomeWide Analysis. Lancet.381, 1371 (2013). 10. A. Serretti, C. Fabbri, Shared Genetics among Major Psychiatric Disorders. The Lancet.381, 1339–134 (2013).
FALL 2014 | PENNSCIENCE JOURNAL 9
FEATURES
The Correlation between
Alcoholism and Mood Disorders
M
ood disorders are identified by persistent abnormal mood (4). They include psychiatric disorders such as major depressive disorder, bipolar disorder, and substance-induced mood disorder. Among people with mood disorders, alcoholism is common (1,3,6). It is believed that this may be the result of consuming alcohol as a means of selfmedication to reduce emotional stress (1,2,3). However, alcohol consumption is also common among people without mood disorders. Moreover, research shows that alcohol consumption among people without mood disorders can play a role in the development of mood disorders (3). However, studies do not clearly show in which direction the arrow of causality points. Therefore, this article will attempt to analyze the evidence in order to better understand whether mood disorders lead to alcoholism or alcoholism to mood disorders. Many studies have found that people with mood disorders are more likely to abuse alcohol than those without mood disorders. Those with mood disorders often use alcohol as “treatment” for their mood disorders; one study
10 PENNSCIENCE JOURNAL | FALL 2014
By Coby Basal
suggests that 25 percent of individuals with mood disorders use alcohol or drugs to treat their own symptoms (1). This rate of abuse among people with mood disorders is two to three times higher than the rate of alcohol abuse among individuals without mood disorders. Those with bipolar disorder are even more likely to abuse alcohol than those with other mood disorders (1). In addition, a second study indicates that 37 percent of individuals with mood disorders abuse alcohol (2). This study also found that people with bipolar disorder are disproportionately more likely to use alcohol or drugs as self-medication. A third study found a particularly powerful effect in women as depression preceded alcoholism in 66 percent of cases (6). Together, these studies provide compelling evidence that those with mood disorders are more likely to abuse alcohol. Conversely, other studies have found that people who consume alcohol — who otherwise would not have mood disorders — are more likely to develop such disorders. For example, in a study in which heavy doses of alcohol were administered over several weeks, participants reported severe depressive symptoms (3). These same participants, after stopping their consumption of high doses of alcohol, demonstrated improved mood stability within a few days to a few weeks (3). In another study involving men and women, subjects reported positive effects such as vigor
FEATURES The Prevalence of Alcohol, Other Drug, and Mental Disorders in the US Total Community
22.5%: non-substance abuse mental disorder 58.9%: no prevalence
13.5%: alcohol-dependence disorder 6.1%: other drug-dependence disorder 25%: individuals with mood disorders use alcohol or drugs to treat their own symptoms
and elation upon moderate alcohol consumption, but then also reported negative effects such as depression, anxiety, and fatigue shortly after moderate alcohol consumption (5). These findings suggest that continuous alcohol consumption leads to fluctuations between positive and negative moods. It is easy to imagine how this could potentially be linked to the onset of mood disorders. There is some evidence that there may be gender differences in the onset of mood disorders and alcoholism. The study alluded to above, which found evidence of depression preceding alcoholism in women, found that in men, alcoholism preceded depression in 78 percent of cases (6). Recent research thus suggests that the relationship between alcoholism and mood disorders is very complex. It appears that one could claim that mood disorders can lead to alcoholism, but one would not be wrong to assert that alcoholism can precede mood disorders. What is clear is that effective treatment is required to treat people who suffer from alcoholism — whether their alcoholism was the result or part of the cause of their mood disorder. One common treatment involves the use of detoxification drugs such as benzodiazepines to alleviate withdrawal symptoms (7). Such treatment is considered to be a safe and effective method to treat alcoholism and its withdrawal symptoms. What makes benzodiazepines ideal is that they can be used to treat anxiety disorders as well (8). Thus, this class of drugs can potentially effectively address both alcoholism and mood disorders. However, researchers should still search for other, more effective treatments. For example, since patients with mood disorders often say that alcohol improves their moods, alleviates tension, maintains euphoria, and provides energy, researchers should search for safe treatments that can accomplish all of these
goals (1). Such treatments could replace alcohol for individuals, and as a result, lessen drug and alcohol abuse among victims of mood disorders (3). References: 1. J.M. Bolton, J. Robinson, J. Sareen, Self-medication of mood disorders with alcohol and drugs in the National Epidemiologic Survey on Alcohol and Related Conditions. Journal of Affective Disorders.115, 367-375 (2009). 2. D.A. Regier, et al., Comorbidity of Mental Disorders With Alcohol and Other Drug Abuse: Results From the Epidemiologic Catchment Area (ECA) Study. JAMA..264, 2511-2518 (1990). 3. E.B. Raimo, M.A. Schuckit, Alcohol dependence and mood disorders,.Addictive Behaviors.23, 933-946 (1998). 4. I. Macmillan, A. Young, I.N. Ferrier, Mood (affective) Disorders. Medicine.32, 14-16 (2004). 5. P.B. Sutker, B.Tabakoff, K.C. Goist, C.L. Randall, Acute Alcohol Intoxication, Mood States and Alcohol Metabolism in Women and Men. Pharmacology Biochemistry and Behavior.18, 349-354 (1983). 6. J.E. Helzer, T.R. Pryzbeck, The Co-Occurrence of Alcoholism with other Psychiatric Disorders in the General Population and Its Impact on Treatment. Journal of Studies on Alcohol and Drug .49, 219-224 (1988). 7. M. Hayashida, et al., Comparative Effectiveness and Costs of Inpatient and Outpatient Detoxification of Patients with Mild-toModerate Alcohol Withdrawal Syndrome. New England Journal of Medicine. 320, 358-365 (1989). 8.Schatzberg, J. O. Cole, Benzodiazepines in Depressive Disorders. Archives of General Psychiatry.35, 1359-365 (1978).
FALL 2014 | PENNSCIENCE JOURNAL 11
FEATURES
Neurostimulation I
n 2014, about 9 percent of Americans from various backgrounds will suffer from some form of depression. According to a report published in 2011, the rate of depression in France was an astounding 21 percent, with other European and South American countries logging in rates in the double digits (1). Prevention and treatment are certainly needed for what has become an epidemic in a global context. Although we often think of someone suffering from depression as simply being “sad” or “feeling down”, depression does indeed have a very strong biological basis. For example, through extensive research over the last two decades, we have come to understand how neurostimulators — specifically fluctuations in their concentrations — result in the symptoms we most commonly identify with depression. While there are a multitude of treatment options, including cognitive behavior therapy, electroconvulsive therapy, and medications aimed at creating a balanced chemical environment in the brain, neurostimulation remains a steadily growing though highly controversial method. Neurostimulation involves modulating the central or peripheral nervous system through the use of electrical or magnetic impulses. As much as the brain functions in a chemical manner, it also functions in a highly electrical manner through synapses and voltage-mediated action potentials. It is through manipulation of these electrical maneuvers that neurostimulation is most effective. The use of neurostimulation leads to significant enhancements in brain metabolism, and PET scans have shown that structures in the brain also change following the initial use of neurostimulation. It is interesting to note that these structures are known regions of high concentrations of dopamine, a neurotransmitter most commonly associated with emotional responses and pleasure, and thus stimulating these structures reduce depressive symptoms. Currently, neurostimulation is most commonly used in
12 PENNSCIENCE JOURNAL | FALL 2014
pain management during surgical processes. It is also an established option for the treatment of conditions such as depression, Parkinson’s disease, and epilepsy. Although most patients turn to neurostimulation after the failed use of medications, neurostimulation techniques provide more relief in cases of chronic and recurrent depression. There are two major neurostimulation techniques, each with its own benefits, uses, and histories. Vagus Nerve Stimulation, or VNS, involves implanting a battery-powered device into the chest, with electrical leads tunneling from under the skin to a nerve in the neck. The stimulation lasts the entire day, in increments of 30 seconds per 5 minutes of elapsed time. VNS has shown to take about 5 months to have an effect, and has led to gradual improvement over the course of five years. VNS, however, has been shrouded in controversy despite receiving FDA approval over 9 years ago. In one study, 53.1 percent of subjects fulfilled the criteria of response of lower depressive symptoms, and 38.9 percent of subjects even fulfilled the remission criteria (2). However, Medicare, as well as other insurance carriers, believes that the benefits of VNS do not justify its high cost. As time passes, and more data accumulates, researchers hope to fortify their claims and increase the use of VNS in patients who are dealing with chronic mood disorders, such as bipolar depression. Transcranial Magnetic Stimulation, or TNS, on the other hand, gained FDA approval in 2009, and requires a rapidly moving magnetic field to cause induction of a current in the prefrontal cortex of the brain. Activity rates of the prefrontal cortex greatly decline when a person is depressed. The Neuronetics device, approved by the FDA in 2009, has achieved a
FEATURES
by Ritwik Bhatia
58 percent response rate in studies, while the newer Brainways device, approved in 2013, has not been researched extensively in any studies as of yet. While insurance companies are still actively against VNS, insurers are increasingly supportive of the use of TMS
treatments. Increased coverage stems from the fact that TMS is less costly, has fewer side effects and can be used to treat a larger population of patients who suffer from milder forms of depression. What does the future hold for neurostimulation? Product design and development firm Cambridge Consultants recently released a report, detailing its new generation of devices that promise minimal side effects in the treatment of various mood disorders. However, with funding and investment backed by groups such as Cambridge Consultants, the goal is to expand neurostimulation to treat conditions such as traumatic brain injury and migraines, and to improve quality of life. The journey to making neurostimulation accessible and acceptable to the public has certainly been an uneven road. Once heralded as a unique way to treat serious conditions, then viewed as a highly dangerous technique, and now seen as a viable, yet expensive, method to treating the epidemic of mood disorders globally, neurostimulation remains at the forefront of medical controversy and innovation. References 1. E. Bromet, et al., Cross-national epidemiology of DSM-IV major depressive episode. BMC medicine.9, 90(2011). 2. M. Bajbouj, et al., Two-year outcome of vagus nerve stimulation in treatmentresistant depression. Journal of clinical psychopharmacology.30, 273-281 (2010).
FALL 2014 | PENNSCIENCE JOURNAL 13
FEATURES
Anxiety: The Feeling of Being Chased By Richard Diurba
O
ver ten thousand years ago, lions would chase humans across the plains. These humans would immediately respond with shock and run, literally, for their lives. However, now the lion is thousands of miles away and the closest threat may be a midterm exam, and while midterms can leave students feeling burned out and exhausted, they do not pose the risk to life of a full-grown lion. So how should we view anxiety? Is this seemingly uncalibrated artifact of evolution a danger to our health? Anxiety is characterized by a combination of two factors — the general concern for the well being of oneself and the concern for how one arrived to a general state of anxiousness. The anxiety aggregates from multiple stressors over time. For example,
14 PENNSCIENCE JOURNAL | FALL 2014
FEATURES facing many midterms or coping with a serious family emergency creates the stress that causes anxiety to increase. If one is exposed to stress, then one becomes anxious about the outcome of the stressor. When stress builds, a negative biological phenomenon occurs. The amygdala, which is the stress and fear center of the brain, undergoes higher stimulation than that of daily levels; physiologically, the body acts to compensate with anxiety (1). This natural response increases tension in the brain and body. The inevitability of activation in the amygdala is high; on average, the amygdala’s “anxiousness” will be activated 28.8% of a lifetime (1). There exist several forms of anxiety with different severities. The first is generalized anxiety disorder (GAD). According to H.S. Akiskal of the University of California, San Diego, “Generalized anxiety disorder is defined as an uncontrollable disposition to worry about one’s welfare . . . Associated manifestations include arousal, vigilance, tension, irritability, unrestful sleep and gastrointestinal distress” (2). If an individual sits endlessly contemplating his or her actions or purpose, he or she may be exhibiting GAD symptoms. These symptoms have a neurological basis. While worrying about a particular issue, the brain places a higher amount of energy in the cerebral cortex, the thinking area of the brain. As a consequence, the cerebellum, which controls key bodily processes, is ignored during the energy transfer. This means overthinking due to stress and anxiety literally saps the body’s ability to successfully regulate itself, leading to the symptoms described above (1). An even more severe form of anxiety beyond GAD is seen at the clinical level. Clinical anxiety is diagnosed through various tests for anxiety. A combined study from the University of Medicine and Dentistry of New Jersey, the University of Maryland and the University of Pennsylvania attempted to create the ideal test for anxiety. Their test for clinical anxiety includes a total of nine self-report tests. The patient is then screened for around twenty minutes for a suitable treatment program (3). One issue with the second wave of tests is that the vast majority of the tests also capture the effects of depression; the overlap in results is in part due to shared symptoms and causes. As the team of Beck, Brown, Epstein, and Steer stated in their report, “a number of studies have reported high correlations between the widely used rating scales of anxiety and depression” (3). Akiskal further posits that, “if one is anxious for life, one is likely to have a superimposed depressive disorder” (1). It is suggested that those experiencing signs of manic anxiety should see a clinical psychologist. Furthermore, Psychology Today has created a neat survey that allows the user to self-report stress levels and offers suggestions based on the results (http://psychologytoday.tests.psychtests.com/take_test. php?idRegTest=1597) (4). The questions address feelings and fatigue levels. A test taker that scores around 80 to 90 out of 100 is considered highly anxious. According to the test, high scoring individuals typically have the feeling that “life may have no meaning at all.” In contrast, a low
score indicates the ability to handle stress. One way to lower anxiety is to try to discontinue rumination on life events. Susan Nolen-Hoeskema of the University of Michigan stated, “Several longitudinal studies have shown that people who engage in more ruminative responses when they are sad, blue, or depressed have higher levels of depressive symptoms over time” (5). The underlying biology also supports this form of selftreatment. When there is greater metabolic activity in the cerebral region, stress levels go up. By reducing the amount of activity in the cerebral cortex, stress will decrease. Furthermore, over time, the hippocampus forgets the conditioned response from the stressor (1). Continually conditioning the brain not to react to stress will slowly yet significantly reduce general anxiety by mitigating stressors. A more clinical way to treat anxiety includes therapy or prescribed medication. Therapy, the more popular of the two, consists of three possible strategies: cognitive behavioral training (CBT), applied relaxation (AR), and nondirective therapy (ND). CBT conditions the brain to no longer be fearful and to learn strategies to reduce stress. AR focuses on strategies to reduce stress when a stressful situation occurs, while ND is a more conventional type of psychotherapy (6). Drugs often prescribed for anxiety include Librium (chlordiazepoxide), Nembutal (sodium pentorbarbital), meprobamate, and diazepam. Anxiety remains a constant factor in life. In a world of ever-increasing societal pressures, excess anxiety will always remain an issue. It has evolved as an adaptation for human survival. However, with the right tools and mindset, people can mitigate anxiety and escape the “lions” of our day. Works Cited 1. L.M. Shin, I. Liberzon, The Neurocircuitry of Fear, Stress, and Anxiety Disorders. Neuropsychopharmacology.35, 169-91 (2009). 2. H.S. Akiskal, Toward a Definition of Generalized Anxiety Disorder as an Anxious Temperament Type. Acta Psychiatrica Scandinavica.98, 66-73 (1998). 3. A.T. Beck, N. Epstein, G. Brown, R.A. Steer, An Inventory for Measuring Clinical Anxiety: Psychometric Properties. Journal of Consulting and Clinical Psychology.56, 893-97 (1988). 4. J.A. Taylor, A Personality Scale of Manifest Anxiety. The Journal of Abnormal and Social Psychology.48, 28590 (1953). 5. S. Nolen-Hoeksema, The Role of Rumination in Depressive Disorders and Mixed Anxiety/depressive Symptoms. Journal of Abnormal Psychology. 109, 504-11 (2000). 6. T.D. Borkovec, E. Costello, Efficacy of Applied Relaxation and Cognitive-behavioral Therapy in the Treatment of Generalized Anxiety Disorder. Journal of Consulting and Clinical Psychology.61, 611-19 (1993). 7. M.E. Olds, J. Olds, Effects of Anxiety-relieving Drugs on Unit Discharges in Hippocampus, Reticular Midbrain, and Pre-optic Area in the Freely Moving Rat. International Journal of Neuropharmacology.8, 87-IN3 (1969).
FALL 2014 | PENNSCIENCE JOURNAL 15
INTERVIEW
An Interview with
Dr. Jeffery Greeson CONDUCTED BY RAMI EZZIBDEH How did you get interested in the field of behavioral medicine? I have long been interested in the effects of stress on health. Stress, particularly chronic stress, can contribute to many common health problems, including heart disease, hypertension, diabetes, asthma, arthritis, obesity, depression, and chronic pain, among others. Although there are medications for each of these conditions, healthy lifestyle change is often prescribed, yet relatively few people are able to maintain lifestyle changes long-term. As a prospective graduate student, I wondered if reducing stress could not only help people engage in healthy lifestyle changes, but also potentially mitigate stress-related biochemical and physiological pathways implicated in many chronic diseases. Behavioral medicine is a field of research and clinical practice that addresses this area, so I knew it was a great fit for me. You actively research mindfulness meditation from a clinical perspective. Can you please explain what it is and the mechanisms that link mindfulness to reduced stress levels.
Dr. Jeffery Greeson is an Assistant Professor of Clinical Psychology in the Psychiatry Department of the Perelman School of Medicine at Penn. His research interests include psychoneuroimmunology, mindfulness meditation, and advanced statistical modeling. Dr. Greeson’s clinical work focuses on combining cognitive behavioral therapy with mindfulness to help clients in treating and preventing stress related health problems.
16 PENNSCIENCE JOURNAL | FALL 2014
Mindfulness meditation is an increasingly popular way to develop the core skills of present-focused attention, nonjudgmental awareness, and compassion, with implications for both mental and physical health. The most well researched mindfulness meditation training program is an 8-week course called Mindfulness-Based Stress Reduction (MBSR), developed by Dr. Jon Kabat-Zinn at the University of Massachusetts Medical School. One needs to undertake professional training to become an MBSR instructor, because courses can often include 20-40 people with a variety of stress-related conditions, and a teacher must be able to teach meditation from their own personal experience, coupled with supervision in how to deliver the curriculum with both ‘fidelity’ and ‘fluidity’, as Dr. Kabat-Zinn likes to say at the MBSR teacher trainings. Mindfulness meditation practices typically include: (1) mindful breathing (sensory awareness of breath sensations, intentionally returning one’s focus to the breath whenever one notices the mind has wandered), (2) body scan (sensory awareness that is systematically moved through each area of the body, while imagining breathing into and out of each area as one shifts attention from place to place), (3) gentle yoga (in which the body is the vehicle for mindful attention, awareness, and compassion, as one explores various stretches, postures, and the thoughts, feelings, and habitual judgments and reactions associated with recognizing one’s limits and capacities), (4) mindful eating (sensory awareness of a single
INTERVIEW bite of food, such as a raisin, a piece of fruit, or a chocolate, while also being mindfully aware of thoughts, feelings, and impulses such as craving or aversion that often arise in the presence of food), (5) mindful walking (sensory awareness of the body in motion as one deliberately walks more slowly, then more quickly, than one’s usual pace, again noticing thoughts, judgments, impatience, and sensations throughout the body as one moves), and (6) loving-kindness, also known as metta or compassion meditation (sending wishes for health, happiness, safety, and peace to oneself, loved ones, mentors or benefactors, strangers, communities, and even all beings). Because stress is typically rooted in thinking about the future (worry) or the past (rumination), mindfulness can reduce stress, in part, by helping us reinhabit the present, which is the only moment we ever have. Mindfulness can also help reduce stress by helping us not identify with thoughts, worries, or negative beliefs; we learn to see thoughts as “just thoughts” or “mental events”, that aren’t necessarily true. This can relieve a tremendous amount of stress and psychological burden for many people. Mindfulness meditation practice can also help people better recognize and regulate emotions and emotional reactions. Several neuroimaging studies, for example, have shown that mindfulness practice can produce neuroplastic changes in the brain, including increased grey matter density, functional connectivity, and enhanced prefrontal control of limbic system activity, which is implicated in fear, anxiety, and a variety of stress-related physiological responses and symptoms. In fact, some studies have now shown that mindfulness meditation training can reduce a variety of stress-related biomarkers, including blood pressure, inflammatory gene expression, and catecholamine (adrenaline, norepinephrine) responses to stress. Can you provide us with a brief overview on the field of psychoneuroimmunology, specifically in the context of stress responses? Psychoneuroimmunology -- or PNI -- is an interdisciplinary field that studies interactions between the mind, the brain, the immune system, behavior, and health. It is a field that spans studies of cells (in vitro), animals, and humans. Because stress can affect nearly all systems of the body, with important implications for health and health care, stress is a major topic of investigation in the field of PNI. The stress response not only includes the release of adrenaline (from the adrenal gland), cortisol (from the adrenal cortex), and a characteristic “fight-or-flight” reaction in which breathing is faster and shallower, heart rate increases, and muscles tense up, but also an inflammatory response -- in case you get bitten or scratched as you run away from the Saber Tooth Tiger! In the short term, acute stress responses are critical to survival. However, chronic activation of the stress response -- due to constant worrying, or feeling stressed about work, school, relationships or money, for example -- can lead to wear & tear on the body’s systems,
eventually predisposing to stress-related diseases, like heart disease, diabetes, and many others. Researchers in the field of PNI are actively investigating what types of stressors elicit what types of responses in the brain and body, with what types of implications for health and disease risk. Clinical researchers and clinical practitioners can translate that basic knowledge into new therapies -whether pharmacological or behavioral -- to specifically target stress-related pathways implicated in symptoms and illness. What do you think are some of the biggest research and clinical challenges facing psychiatrists who work in the field of mood disorders? In my view, one of the biggest challenges that cuts across research and clinical practice is how to best personalize treatment and prevention. Personalized medicine typically infers treatment based on one’s genome. However, in the relatively near future, it may be conceivable to personalize treatment for mood disorders based on a person’s specific pattern of brain function, and corresponding psychological, physiological, and behavioral traits or qualities. We know that many different types of treatments change the brain -- from medication to meditation, exercise to psychotherapy. Now, we just need to know what type of treatment, or combination or sequence of treatments, works best for whom, under what conditions. That is a much taller and far more complex question. Another formidable challenge is that a substantial minority of patients do not seem to respond to first-line treatments, again, be they pharmacotherapy or behavioral therapies like psychotherapy or mindfulness meditation. Like personalized medicine, this is another area of vigorous scientific investigation at present. Is there an example of a recent development your field that you believe will have great clinical translatability and alter the way psychiatric treatments are conducted? I know some researchers are studying how taking a functional brain scan BEFORE starting a medication or psychotherapy can help determine what type of treatment is likely to work best, depending on one’s pattern of brain (dys)function. Should this approach eventually make its way into routine clinical practice, it would revolutionize mental health treatment by personalizing and targeting treatments, and possibly preventive interventions, to the specific functional roots that underlie disorder. As a clinical psychologist and a behavioral medicine researcher, it will be fascinating to see whether behavioral therapies like mindfulness meditation and psychotherapy are capable of targeting the same pathways and underlying functional brain mechanisms that drug treatments and neuromodulation therapies, like deep brain stimulation (DBS) and transcranial magenetic stimulation (TMS), are now being developed to address.
FALL 2014 | PENNSCIENCE JOURNAL 17
RESEARCH Single and Cyclic Axial Mechanical Properties of Porcine Common Carotid Artery Yishan Liua, Stephanie Pasquesia, Susan Marguliesa Department of Bioengineering, University of Pennsylvania, Philadelphia, PA 19104, USA
a
When infants suffer vigorous shaking injuries, the head rotates in an anterior-posterior fashion about the neck. The neck must bear large repeated loads as a result of the rapid rotational accelerations of the head. The goal of this study was to investigate the cyclic material properties of the porcine common carotid artery to reveal how neck vessels behave under repeated loads and this differs from its behavior under single loads. Arteries from eight different toddler pigs were pulled to failure at approximately 60 s-1 and cyclically to 0.5 strain at a frequency of 3Hz. The single pull low and high strain moduli were found to be 4.156 ± 3.374 and 5.812 ± 4.492 MPa, respectively. For the cyclic pulls, a mean low strain modulus of 0.686 ± 0.402 MPa and high strain modulus of 1.738±0.990 MPa were obtained for the second cycle. Results from the cyclic tests indicate that both the low and high strain moduli decay in an exponential fashion over time; therefore, the vessels do indeed experience fatigue under oscillating loads. Furthermore, it was found that high strain modulus decayed at a faster rate than the low strain modulus, indicating that the effects of fatigue may be more pronounced at higher displacements. Results from this study are useful in understanding how neck vessels respond during repeated extensions of the neck in shaking injuries, information that is lacking in current research. This preliminary understanding may ultimately be useful in the long term for both the diagnosis and prevention of head and neck injuries.
1. Introduction Shaken baby syndrome (SBS) is the forceful shaking of infants that may often lead to severe brain injuries. It is the most common cause of death and disability of children in abusive households (3). These shaking type injuries may lead to the tearing and rupture of the cerebral bridging veins due to the relative motion between the brain and skull which may result in subdural hematoma or cerebral edema. In non-fatal cases, injuries caused by SBS may also lead to longterm consequences such as motor impairment, blindness, and paralysis. During a shaking event, the weak muscles and vessels in the neck must bear the inertial forces from the motion of the head as it rapidly rotates back and forth (Fig. 1). During typical rotational accelerations of the head, the neck must bear cyclic subcatastrophic loads, sometimes as high as 35,910N as it extends and shortens (1). This type of cyclic damage to the neck and head is not confined to the study of shaken baby syndrome; it is also prevalent in athletes. For example, forces are transmitted to the neck when a soccer player heads a ball. The forces seen by the neck may occur enough times repeatedly that it may result in fatigue damage of vessels. In addition, football players endure multiple tackles during games and practices in which the head and neck rotate violently, leading to concussions and other head injuries. One recent study used accelerometers on college football players to measure the frequency and location of the repeated head impacts. The study ultimately found that on average, each player endured over 1,000 head impacts per season and that the greatest number of blows occurred at the front of the helmet (4). To date, there have been none or few previous studies on the cyclic axial loading of vessels as most research has been limited to studying vessels under either monotonic or circumferential-direction loading (2,7). The objective of this study is to therefore characterize the behavior of 18 PENNSCIENCE JOURNAL | FALL 2014
Fig.1. Snapshots of deformations to the neck under shaking injuries [1]
the porcine common carotid artery under both single and cyclic loading in the axial direction, an understanding that is lacking in current research. Properties of interest include ultimate stress, strain at ultimate stress, and low and high strain moduli for single axial loading, and low and high strain moduli for cyclic loading. Stress is the force applied divided by the cross sectional area perpendicular to that force, or σ=F/A. Strain characterizes the deformation of a material and is calculated as the change in length over the initial length, or ε = dl / lo. Ultimate stress is the maximum stress that can be withstood by a material before failure and is taken to be the peak of the stress-strain curve for each tissue specimen; strain at ultimate stress is the corresponding strain to this maximum stress. Finally, the modulus is the ratio of the stress to strain, or E=σ/ε. The low and high strain moduli are the modulus values at low deformations (.3-.4 strain) and at high deformations (.4-.5 strain), respectively. These key properties were measured and compared under both single and cyclic pull conditions. This work is central to determining the effects of infant shaking and sports injuries which cause repeated subcatastrophic
RESEARCH damage to the neck and head. The results from this study enhance current understanding of the fatigue properties of blood vessels and reveal if and how their material properties change after cyclic damage. The project is also an important first step in understanding blood vessel behavior before studying the smaller and more delicate cerebral bridging veins.
2. Methods
2.1. Vessel Preparation Left and right common carotid arteries were dissected from eight 4-week-old female farm pigs; each artery was cut into 3 sections and all specimens were frozen in a 90% DMEM-10% DMSO cryopreservation solution until testing. Thirty minutes before testing, vessels were thawed in serial dilutions of decreasing DMSO concentrations (10%, 7.5%, 5%, 2.5%, and 0%). The dimensions of each sample were then measured. Each sample was placed flat between two glass slides and three measurements of the sample width were taken with calipers. The circumference was then calculated as C=2*Wavg, where C is the circumference of the vessel and Wavg is the average of the width measurements. The thickness of vessel wall was calculated by measuring the thickness of the specimen in between the slides, subtracting the thicknesses of each slide, and dividing by two, or tvessel wall=(1/2)*(tspecimen+slides-tslide 1-tslide 2). Table 1 displays the average dimensions of each vessel used for the single and cyclic tests. The vessels were affixed to 14-gauge blunt-tipped needles with
circumferential grooves etched around the tip and secured by tying size 6-0 silk sutures around the vessel circumference and dabbing cyanoacrylate on the distal ends. The vessels and needles then were attached to the testing device with custom blocks with internal channels drilled through them to allow the samples to be pressurized. This grip system was similar in design to that used by Monson et al (6). 2.2 Experimental Apparatus Single-Pull Device: Single pull tests were conducted on a custom designed drop test apparatus (Fig. 2) that consisted of a metal plate which slid down three shafts and hit a lever arm when released; the opposite end of the lever arm was attached to one of the vessel blocks so the top end of the vessel was pulled when the lever arm was struck. The top vessel block was attached via a universal joint to the lever arm end to ensure linear vessel displacement. The strain rate of the test was controlled by adjusting the height from which the metal plate fell. The other, stationary end of the vessel and grip was mounted on a 5lb capacity load cell (Model 31, HoneywellSensotec, Columbus, OH) and an adjustable lab jack. A laser displacement sensor (LC 1607-50, Micro Epsilon, Ortenburg, Germany) was used to measure the displacement of the lever arm; this motion was calibrated to account for the linear motion of the grip when calculating strain. Vessels were manually pressurized with saline from a syringe connecting to the bottom vessel grip. The vessels were pressurized to about 100 mmHg for each test as measured by the signals on pressure transducers (26PCBFM6G, Honeywell, Columbus, OH) attached to both the top and bottom blocks.
Table 1 Dimensions of the carotid artery samples used for single and cyclic tests for each pig
Pig #
505 451 454 479 503 399 452 477 Â
Artery for Single Pull Average Average Average Circumference Vessel Wall Cross of Vessel (mm) Thickness Sectional (mm) Area (mm2) 10.147 0.540 4.560 6.727 0.298 1.727 7.690 0.365 2.386 7.287 0.310 1.955 7.093 0.441 2.518 11.700 0.659 6.344 7.507 0.340 2.191 8.340 0.581 3.786
Cyclic Pull Device: The cyclic device used an Alphastep closed loop stepper motor (Vexta #AS46AAP, Oriental motor, Braintree, MA) to turn four sets of gears, which then allowed a plate attached to rotating threaded shafts to move up and
Artery for Cyclic Pull Average Average Average Circumference Vessel Wall Cross of Vessel (mm) Thickness Sectional (mm) Area (mm2) 8.153 0.300 2.166 7.793 0.320 2.175 8.267 0.327 2.368 9.440 0.495 3.906 8.187 0.314 2.259 8.187 0.319 2.290 8.907 0.462 3.445 7.673 0.317 2.117 down linearly; the bottom grip of the vessel was attached to this moving plate. The motion of the motor was computercontrolled via a driver (Vexta #ASD13A-AP, Oriental motor, Braintree, MA). Elements from the single pull device were transferred to the cyclic device for the cyclic pulls including
FALL 2014 | PENNSCIENCE JOURNAL 19
RESEARCH the blocks, blunt tipped needles, pressure sensors, salinefilled syringe, lab jack, and laser displacement sensor. The cyclic device was calibrated by measuring the linear distance moved by the plate for each step of the motor. A picture of the assembled cyclic device is shown in Figure 3. Laser Displaceme nt Sensor
Lever Arm
Pressure Transducers
Syringe with Saline
2.3 Test setup: Single and cyclic pulls were conducted on right common carotid arteries from eight different four-week old pigs. Once a carotid artery was properly mounted into either device, the height of the lab jack was adjusted until the force reading on the load cell just began to register a load. The length of the vessel was then measured with calipers and this was taken as the zero load, or gauge, length. The vessel was pressurized until the top and bottom pressure sensors read about 100 mmHg and then the test was initiated.
Lab Jack Laser Sensor and Stand Custom Grips
Cell Stretcher
(a)
Load Cell
Pressure Sensors
Syringe with Saline Vessel
Custom Grips
Fig.2. Single Pull Device
20 PENNSCIENCE JOURNAL | FALL 2014
(b) Fig.3. (a) Cyclic Pull Device and (b) Close up of device components
For the single tests, vessels were pulled axially at a strain rate of approximately 60s-1; this value represents the median strain rate experienced by vessels in the brain and was measured from previous finite element analyses conducted in the lab. For the cyclic tests, vessels were repeatedly pulled to 0.5 strain for 30 seconds at 3 Hz, a frequency commonly seen during shaking injuries (8). Measurements of the force, displacement, and pressure inside the vessels were taken during each test.
3. Results: In total, there were 8 successful single tests and 7 successful cyclic tests. 3.1 Single Pull Results: The mean ultimate stress and strain at ultimate stress were 4.221 ± 0.921 MPa and 0.983 ± 0.229 MPa, respectively (Table 2). The low strain (0.3 to 0.4 strain) and high strain (0.4 to 0.5 strain) moduli were 4.156 ± 3.374 and 5.812 ± 4.492 MPa, respectively (Table 3). 3.2 Cyclic Pull Results: For the cyclic pulls at 3Hz, the decay of low strain and high strain moduli was graphed and a mono-exponential curve was fit to this decay. Modulus values were taken from the second cycle to the second to last cycle. For the second cycle, the mean low strain modulus was 0.686 ± 0.402 MPa and the mean high strain modulus was 1.738±0.990 MPa (Table 3). While data analysis is still ongoing, Figures 4 and 5 below show representative low and high strain modulus decay curves for a carotid artery sample, respectively. As shown in Table 4, the mean low strain strain rates at which the vessels were pulled were 44.933 ±6.272 and 3.745 ± 1.101 s-1 for single and cyclic tests, respectively. The mean high strain strain rates for single and cyclic pulls were 46.538 ± 5.020 and 1.728± 0.343 s-1, respectively.
RESEARCH Table 2 Ultimate Stress and Strain for Porcine Carotid Arteries Under Single Pull Single Pull Data Ultimate Strain at Stress Ultimate (MPa) Stress
Pig #
Artery505 0.55
exponential fit lowermodulus vs. cyclenumber
0.5
2.440
1.392
451
5.242
1.078
0.45
454
3.773
0.928
0.4
479
5.094
0.685
503
3.793
1.062
399
4.108
1.128
452
4.958
0.767
477
4.357
0.819
low modulus (Mpa)
505
f(x) = a*exp(-b*x)+c a= 0.3519 b = 0.06803 c= 0.1552 R-square: 0.9071
0.35 0.3 0.25 0.2 0.15
Average
4.221
0.983
Median Standard Deviation Coefficient of Variance
4.232
0.995
0.921
0.229
0.218
0.234
0.1 0
10
20
30
40 50 cycle number
60
70
80
90
Fig.4. Low Strain Modulus vs. Cycle for Carotid Artery Sample from Pig 505
Table 3 Single and Cyclic Low and High Strain Modulus of Porcine Carotid Arteries
Pig #
Cyclic Pull Low High Strain Strain Modulus Modulus (MPa) (MPa)
505 451 454 479 503 399 452 477
0.711 1.247 4.476 9.732 1.901 1.407 7.438 6.335
1.192 1.654 5.095 13.200 6.385 1.348 10.964 6.661
0.556 1.553 0.608 0.287 0.577 0.717 0.303 0.885
1.985
Average Median Standard Deviation Coefficient of Variance
4.156 3.189
5.812 5.740
0.686 0.592
1.738 1.740
3.374
4.492
0.402
0.990
0.812
0.773
0.586
0.569
1.740 0.632 2.664 1.556 0.437 3.155
Artery505 2
Exp Fit highermodulus vs. cyclenumber
1.9 1.8 high modulus (Mpa)
Single Pull Low High Strain Strain Modulus Modulus (MPa) (MPa)
f(x) = a*exp(-b*x)+c a= 0.6546 b = 0.08169 c= 1.246 R-square: 0.8619
1.7 1.6 1.5 1.4 1.3 1.2 1.1 0
10
20
30
40 50 cycle number
60
70
80
90
Fig.5. High Strain Modulus vs. Cycle for Carotid Artery Sample from Pig 505
FALL 2014 | PENNSCIENCE JOURNAL 21
RESEARCH Table 4 Single and Cyclic Low and High Strain Strain Rates of Porcine Carotid Arteries  that for Pig #451, there is no data under the cyclic tests because they were not performed successfully. Note
Single Pull Strain Rate (s-1)
Pig #
Low Strain 505 451 454 479 503 399 452 477
Average Median Standard Deviation Coefficient of Variance
High Strain
41.202 54.254 36.115 46.905 42.631 43.180 53.522 41.657 44.933 42.906 6.272
46.183 51.383 37.281 45.817 44.589 48.589 53.925 44.540 46.538 46.000 5.020
0.140
0.108
4. Discussion: This study aimed to characterize the axial single and cyclic material properties of porcine common carotid arteries and investigate the effects of repeated loading on blood vessels of the neck and head. For single pull, these properties include ultimate stress, strain at ultimate stress, and low and high strain moduli. For cyclic pull, the properties include low and high strain moduli. Both average low strain and high strain moduli in the cyclic studies (0.686 and 1.738 MPa, respectively) were found to be significantly lower than in single pull (4.156 and 5.812 MPa, respectively). While reasons for this are currently unknown, one possible explanation is that the vessels were stretched at a higher strain rate in the single pull tests than in the cyclic pulls and strain rate has been shown in some circumstances to have an effect on measured properties (10). Both low strain and high strain cyclic moduli were also found to decay at a monoexponential fashion. Therefore, the porcine common carotid arteries were indeed found to soften, or undergo fatigue, under repeated stretching. However, further experiments need to be conducted to see if this trend extends to other vessels. The data thus far is consistent with previous research on general fatigue behaviors of polymers, which has found that the peak tensile and compressive stresses of polymers typically decrease after a number of stretch cycles at a set displacement. Current research also suggests that under oscillating loads, polymers fail at lower stress levels than their yield stress under monotonic loads (9). However, it is necessary to stretch the carotid arteries cyclically for longer periods
PENNSCIENCE JOURNAL | FALL 2014
Cyclic Pull Strain Rate (s-1) Low Strain High Strain nd 2 cycle 2nd Cycle -
4.535 4.532 4.348 3.841 3.209 4.626 3.520 3.745 4.094 1.101 0.294
-
1.956 1.901 1.839 1.443 1.330 2.236 1.391 1.728 1.839 0.343 0.199
of time to determine their fatigue life before this can be determined with more certainty. Furthermore, the two mono-exponential equation constants, 0.1552 and 1.246 MPa, from the exponential decay curves in Figures 3 and 4, may represent the steady state low and high strain moduli, respectively. These steady state moduli represent the long-term moduli of the vessel after initial cyclic softening and before catastrophic failure, when tensile stress amplitude reaches zero. Another interesting observation from the data is that the high strain modulus generally decayed at a faster rate than the low strain modulus. For example, in Figures 4 and 5, the rate of exponential decay for the high strain modulus was 0.08169 compared to 0.06803 for the low strain modulus. This could imply that the effects of fatigue are greater as the vessels are stretched to higher displacements. The biggest challenge in conducting these tests was measuring the dimensions of the vessels under zero load conditions. It was difficult to measure dimensions consistently due to differences in the degree to which the vessels were compressed between the glass slides. The method could be improved by using a micrometer with a force transducer to apply the same force to the vessels before the measurements are taken, similar to the technique described by Lee and Haut (5). Another difficulty was determining the zero force length; sometimes, the vessels still looked slack even when the force transducer registered a load, and in these cases, we loaded the vessel to the length where it no longer appeared slack. Therefore, a more systematic method is needed to determine the length of vessels at zero load.
RESEARCH While this project did elucidate differences between single and cyclic properties of the carotid artery, more work is needed to fully understand the cyclic behavior of vessels. In particular, vessels in this project were only displaced to 0.5 strain for the cyclic tests and it would be interesting to see how their behavior changes under higher strains like 0.7 or 0.8. Additionally, vessels could be stretched for longer periods of time to determine the fatigue life at multiple strains and frequencies. It may also be useful to perform histological tests on the vessels before and after they are pulled to analyze how the microscopic structure of the tissues change after extended loads. 5. Conclusion: This experiment seeks to study the behavior of porcine common carotid artery under axial single and cyclic loading. It was found that both average low and high strain moduli for the cyclic pulls were lower than for single pulls. Furthermore, cyclic low and high strain moduli were found to decay in a mono-exponential fashion, suggesting that the arteries do undergo fatigue under repeated loading. The decay rate was also higher at high strains than lows strains, so the fatigue effects seem to increase with strain. In the long term, an understanding of blood vessel behavior may be useful in the diagnosis, treatment, and prevention of head and neck injuries. Using knowledge of the material properties of relevant tissues, it may become possible in the future to understand why and how injuries occur and the responses of tissues to certain injury scenarios. If an infant is sent to the emergency room due to head trauma, for example, this and future work may aid in differentiating between accidental and abusive trauma based on observation of the injury. Furthermore, research in injury biomechanics can be applied to the design of helmets, car seats, and other protective equipment for those who are susceptible to these types of injuries. Conflict of Interest Statement The authors have no conflict of interest. Acknowledgements This project was funded by Professor Margulies’ NIH grant, an American Heart Association Pre-Doctoral Fellowship, and the Rachleff Scholars program at the University of Pennsylvania. The authors would also like
to thank Sarah Sullivan and David Gabrieli for their guidance in the finite element analysis and Jill Ralston for her assistance in acquiring the tissue samples. References 1. F. Bandak, Shaken baby syndrome: A biomechanical analysis of injury mechanisms. Forensic Science International.151, 71-79 (2005). 2. D. Bell, R. Kunjir, K. Monson, Biaxial and failure properties of passive rat middle cerebral arteries. Journal of Biomechanics.46, 91-96 (2013). 3. S. Carbaugh, Understanding Shaken Baby Syndrome. Advances in Neonatal Care.4, 105-116 (2004). 4. J. Crisco, et al., Frequency and Location of Head Impact Exposures in Individual Collegiate Football Players. Journal of Athletic Training.45, 549-559 (2010). 5. M. Lee, R. Haut, Insensitivity of tensile failure properties of human bridging veins to strain rate: implications in biomechanics of subdural hematoma. Journal of Biomechanics.22, 537-542 (1989). 6. K. Monson, N. Barbaro, G. Manley, Biaxial Response of Passive Human Cerebral Arteries. Annals of Biomedical Engineering.36, 2028-2041 (2008). 7. K. Monson, W. Goldsmith, N. Barbaro, G. Manley, Axial mechanical properties of fresh human cerebral blood vessels. Journal of Biomechanical Engineering.125, 288-294 (2003). 8. M. Prange, B. Coats, A. Duhaime, S. Margulies, Anthropomorphic stimulations of falls, shakes, and inflicted impacts in infants. J Neurosurgery.99, 143-500 (2003). 9. J. Sauer, G. Richardson, Fatigue of Polymers. International. Journal of Fracture.16, 499-532 (1980). 10. B. Stemper, N. Yoganandan, F. Pintar, Mechanics of arterial subfailure with increasing loading rate. Journal of Biomechanics.40, 1806-1812 (2007).
FALL 2014 | PENNSCIENCE JOURNAL 23
RESEARCH Development of a single-plasmid system for screening site-specific DNA methylases Josh Tycko, Daniel Cabrera, Danielle Fields, Brad Kaptur, Mahamad Charawi, Spencer Glantz, Michael S. Magaraci, Avin Veerakumar, Jordan S. Miller, Brian Y. Chow * Department of Bioengineering, University of Pennsylvania, Philadelphia, PA 19104 * Correspondence: bchow@seas.upenn.edu Targeted methylases are engineered fusion proteins that catalyze sequence-specific methylation – a useful tool to research DNA methylation and edit epigenomes. To accelerate the development of targeted methylases, we created and validated a biobrick assay, and enzymatically active fusion proteins for use as a positive control. Our modular single-plasmid system allows methylase fusions to be easily cloned, expressed, and tested via inexpensive digestion and gel electrophoresis. Our biobrick enables quickly measuring the existence and extent of targeted methylation. It additionally includes validated primer-binding sites for methylation-sensitive bisulfite sequencing, and our E. coli chassis effectively eliminated noise associated with methylation studies. The rapid and affordable workflow enables the quick iterations of the design-build-test cycle necessary for protein engineering.
Introduction In the past, DNA methylases, the enzymes responsible for DNA methylation, have been fused to zinc fingers to confer sitespecificity for manipulating methylation patterns in bacteria, mammalian cells, and ex vivo (3-5). However, these zinc finger fusion proteins are not fully optimized in terms of methylation efficacy and site-specificity, and many researchers have written that nonspecific methylation should be studied further (3, 5). Given the higher specificity and lower cost of TALE and dCas9 as alternatives to zinc fingers for genome editing, we expect researchers will soon begin the lengthy and expensive development process for optimized targeted methylases using these newer DNA binding domains (2). Synthetic biology is characterized by the application of engineering principles such as abstraction, modularity and standardization to biological problems. Here, we propose a modular single-plasmid design, with a noiseless restriction-based assay, as an affordable and accelerated method for screening libraries of targeted methylases. Additionally, we demonstrate the results of using our assay for preliminary analysis of a novel fusion protein based on the TALE binding domain. Our assay is available as a biobrick (BBa_K1128001), and it follows a similar workflow as previously developed multi-plasmid systems for developing site-specific DNA methylases (9).
Materials and Methods Plasmid Assembly. The following experiments used our Biobrick backbone (BBa_K1128001) and inserts available from the Registry of Biological Parts and Addgene. The backbone is a modified pET-26b(+) plasmid, an IPTG-inducible T7 expression vector including the kanamycin resistance gene and a low-copy pBR322 origin of replication. It was purchased from Novagen and modified to include the Biobrick prefix and suffix for Assembly Standard 10 cloning, an sgRNA expression cassette for use with dCas9 (including a constitutive promoter and T7 terminator) downstream of the original expression cassette, and a “target site” with the recognition sequences for the TALE1 and Zif268 24 PENNSCIENCE JOURNAL | FALL 2014
DNA binding domains. The gene for the prokaryotic DNA CpG methylase M.SssI with a 5’ linker was inserted downstream of the T7 promoter. Depending on the experiment, either Zif268 or TALE1 was cloned into the vector upstream of the methylase by restriction digest and ligation. To express M.SssI alone, the linker sequence was removed and a start codon was added by PCR. Zinc Finger 268, was purchased from Addgene (plasmid 12612), to which it had been donated by the Scot Wolfe lab (1). TALE 1 was purchased from Addgene (plasmid 27969),to which it had been donated by the Feng Zhang lab (2). Bacterial Strains: Escherichia coli T7 Express [fhuA2 lacZ::T7 gene1 [lon] ompT gal sulA11 R(mcr-73::miniTn10--TetS)2 [dcm] R(zgb-210::Tn10--TetS) endA1 Δ(mcrC-mrr)114::IS10] was obtained from New England Biolabs and used throughout the studies because of the T7 polymerase expression system and methylation sensitive restriction enzyme knockouts. NEB5alpha E. coli from New England Biolabs were used as expression negative controls because they lack an inducible T7 expression system. In vitro methylation: Purified M.SssI from NEB was incubated with plasmid DNA and S-adenosylmethionine (SAM) according to the manufacturer’s protocol. In vivo methylation: The plasmid was transformed into T7 Express competent cells, induced with varying concentrations of IPTG (0.1mM to 2mM) for varying amounts of time (2 to 24 hours). Methylation sensitive digestion protocol: 600 ng of purified DNA were digested with 1 ul of AvaI and 1ul of XbaI, 2 ul of NEBuffer 2.1, and nuclease free water to total 15 ul per reaction. The solution was incubated for one hour at 37oC. Gel electrophoresis was performed with 1.0% agarose gel in TAE buffer. Additional methods and protocols: Other techniques we used are available for download at: http://2013.igem.org/Team:Penn/ Protocols
RESEARCH Figure Legends
Â
Figure 1. Design of a single-plasmid system for the construction, expression, and measurement of site-specific methylases. A) The singleplasmid system (BBa_K1128001) includes the necessary genetic parts to test site-specific methylases. The methylase, M.SssI with an upstream linker sequence (BBa_K1128002), can be cloned in by standard biobrick assembly with EcoRI and PstI. B) A model showing how methylation sensitive restriction digest followed by gel electrophoresis distinguishes between methylation states.
FALL 2014 | PENNSCIENCE JOURNAL 25
RESEARCH Results
3). We then constructed the first TALE-methylase fusion protein, to our knowledge. After a 4-hour induction, the two most visible bands corresponded to no methylation and non-specific methylation. Less visible were the highest band corresponding to full methylation and the lowest band corresponding to site-specific methylation (Figure 4). This experiment was repeated for 0, 2, 6, and 24 hour inductions with 0, .1, 1, and 2 µM IPTG. Greater induction time and IPTG concentration corresponded to greater intensity for the full methylation band (not shown).
The completed plasmid (BBa_K1128001) includes all the genetic parts necessary to both express a sitespecific methylase and report its activity in terms of site-specificity and methylation efficacy (Figure 1A). By methylating the plasmid in vitro with purified M.SssI as well as by methylating the plasmid in vivo by expressing M.SssI with the T7 promoter, we verified the digestion band pattern appeared as expected (Figure 2A). To verify the plasmid could report analog levels of methylation, it was methylated in vitro for varying degrees of time, and a linear relationship between time methylated and the band’s normalized intensity level after quantitative image analysis was observed (Figure 2B).
Figure 3. Zinc finger-M.SssI does not show site-specific methylation. The plasmid was modified to make a negative control with the zinc finger’s binding site removed. Lanes 3 and 4 are controls to show the zinc finger-M.SssI plasmid digests as expected when linearized and when expression is not induced. Lanes 5 and 6 repeat those controls for the plasmid with the binding site removed. Lanes 1 and 2 show the two plasmids after 10 hours of induced expression of zinc finger-M.SssI and digestion with XbaI and AvaI.
Figure 2. Detection of methylation in vitro and in vivo with the single-plasmid system. A) M.SssI was cloned and expressed in T7 Express cells in vivo, and M.SssI was incubated with the plasmid in vitro. As a negative control, the same plasmid was transformed into NEB 5 alpha competent E. coli.These groups were digested with AvaI and XbaI. As a positive control for the fully methylated band, the plasmid was linearized by digestion with only XbaI, mimicking complete protection against AvaI digestion. B) The plasmid was methylated in vitro for increasing amounts of time with M.SssI. The bands shown and quantified with imageJ are the bands corresponding to full methylation of the plasmid.
Then, the zinc finger-M.SssI fusion was expressed from two plasmids, which were the same except for the presence or lack of the zinc finger’s nine base pair binding site upstream of the “target” AvaI cut site. After a 10-hour induction, only the fully methylated band was visible, regardless of the presence or lack of the zinc finger’s binding site (Figure 26 PENNSCIENCE JOURNAL | FALL 2014
Figure 4. TALE-methylase shows non-specific methylation and lower enzymatic activity. The TALE-methylase was induced and expressed for 4 hours or not induced, then digested with XbaI and AvaI.
Discussion The modular plasmid proved to be an effective system to rapidly generate data on targeted methylases. Including the different DNA binding domains, presence or absence of the target binding site, and varied induction
RESEARCH parameters, we were able to produce quantifiable data for over 100 conditions in 3 weeks, including time for cloning. Site-specific methylation is reported by this assay only if greater than 10% of the plasmids are methylated at only the target AvaI site, based on the limits of visibility with our gel electrophoresis set up and loading 600 ng of DNA. In the future, this system could be used to accelerate and lower costs of library screening as researchers optimize promising TALEmethylases or Cas-methylases. Optimization could include varying linker lengths, performing directed evolution, or creating a split-reconstitution system with two DNA binding domains fused to subunits of the methylase. This latter approach could overcome the offtarget methylation inherent to any bifunctional system with an active methylase effector (7). In many cases, the ultimate application of targeted methylation will be in mammalian systems, but the E. coli chassis is effective for library screening because the lack of endogenous CpG methylase makes the assay noiseless. Recently, researchers produced other tools for engineering other aspects of the epigenome, including: TALE-histone modifiers and TALE-DNA demethylases (8, 10, 11). Our new biobrick could be the enabling tool that allows researcher to complete a new toolkit for investigating the role of epigenetics in development and disease by direct manipulation of the primary epigenetic patterns. Acknowledgments: The authors thank Marisa Bartolomei, Rebecca Simmons, and Christopher Krapp for helpful discussion; Adam Peritz and the Penn Genome Frontiers Institute (PGFI) for wet lab training; Micah Kaats for graphic design of Figure 1; the Department of Bioengineering, Sevile Mannickarottu and Henry Ma for generous use of the Stephenson Family Foundation Undergraduate Teaching Laboratory; and Erik Toorens and the Penn Medicine DNA sequencing facility; This research was funded by the University of Pennsylvania’s: School of Engineering, the Office of the Vice Provost for Research, the Mack Institute for Innovation Management, the Jerome Fisher Program in Management and Technology, and undergraduate research grants from the Ben Franklin Scholars, College Alumni Society, and Vagelos Fund of the Center for Undergraduate Research and Fellowships (CURF). We are grateful for sponsorships from Addgene, Biomatters, Integrated DNA Technologies, and New England Biolabs. Research materials were provided by David Gdula & New England Biolabs, the Zhang Lab (MIT), and the Wolfe Lab (Univ. of Massachusetts Medical School).
Author Contributions JT and DC conceived the project idea and designed the BioBricks in consultation with STG, JSM, and BC. JT, DC, BK, MC, and DF designed and performed all experiments. All authors contributed to data analysis and writing the manuscript. STG, MSM, AV, and BYC supervised all aspects of the projects. Notes The authors declare no competing financial interest. References 1. X. Meng, M.H. Brodsky, S. A. Wolfe, A bacterial one-hybrid system for determining the DNA-binding specificity of transcription factors. Nat. Biotechnoly.23, 988-994 (2005). 2. F. Zhang, L. Cong, S. Lodato, G.M. Church, P. Arlotta, Efficient construction of sequence-specific TAL effectors for modulating mammalian transcription. Nat. Biotechnol.29, 149-153 (2011). 3. G. L. Xu, T.H. Bestor, Cytosine methylation targeted to pre-determined sequences. Nat. Genet.17, 376-378 (1997). 4. A.E. Smith, P.J. Hurd, A.J. Bannister, T. Kouzarides, K.G. Ford, Heritable Gene Repression through the Action of a Directed DNA Methyltransferase at a Chromosomal Locus. J. Biol. Chem.283, 9878-9885 (2008). 5. F. Li, et al., Chimeric DNA Methyltransferases Target DNA Methylation to Specific DNA Sequences and Repress Expression of Target Genes. Nucleic Acids Res.35, 100-112 (2007). 6. A.G. Rivenbark, et al., Epigenetic reprogramming of cancer cells via targeted DNA methylation. Epigenetics.7, 350-360 (2012). 7. W. Nomura, C. F. Barbas, In Vivo Site-Specific DNA Methylation with a Designed Sequence-Enabled DNA Methylase. J. Am. Chem. Soc.129, 8676-8677 (2007). 8. M.L. de Groote, P.J. Verschure, M.G. Rots, Epigenetic Editing: Targeted Rewriting of Epigenetic Marks to Modulate Expression of Selected Target Genes. Nucleic. Acids. Res.40, 10596-10613 (2012). 9. A.E. Smith, K.G. Ford, Specific targeting of cytosine methylation to DNA sequences in vivo. Nucleic Acids Res.35, 740-754 (2007). 10. P.D. Hsu, et al., DNA targeting specificity of RNAguided Cas9 nucleases. Nature biotechnology.31, 827832 (2013). 11. Y.W. Hwang, et al., Efficient genome editing in zebrafish using a CRISPR-Cas system. Nature biotechnology.31, 227-29 (2013).
Author Information Corresponding Authors Tel: (215) 898-5159; E-mail: bchow@seas.upenn.edu
FALL 2014 | PENNSCIENCE JOURNAL 27
RESEARCH Epigenetic Characterization of the Neuroblastoma Driver Genes ARID1A and ARID1B Sharon Kim University of Pennsylvania Analyzing oncogenic drivers and tumor suppressing functionality for any cancer provides insight on improved therapies. A recently identified novel oncogenic pathway in highrisk neuroblastoma (NB) has been identified in wholegenome sequencing: mutually exclusive mutations of ARID1A and ARID1B protein subunits in the Swi/Snf chromatin remodelling BAF complex (BRG1Associated Factors) (1). The BAF protein complex controls the accessibility of DNA sequences to transcription factors through combinatorial assembly (2). Exchanges of subunits within the BAF complex accompany transitions from pluripotent stem cell to neural stem cell and to postmitotic neuron. Our goal was to provide genomewide epigenetic characterization of NB cell lines that had ARID1A or ARID1B alterations or were wild type to define the chromatin and methylation states arising downstream of BAF complex mutations. With genomewide histone modifications and methylation maps, we can then attempt to define how BAF aberrations alter chromatin states and transcriptional programs in NB. We also aimed to analyze the effect of EZH2 inhibitors in ARID1A and ARID1B mutated NB cell lines. EZH2 is a subunit of polycomb repressor complex 2 (PRC2), a multiprotein histone methyltransferase (HMT complex). Overexpression of EZH2 has been shown to lead to tumorigenesis, suggesting that EZH2 enzymatic activity is a required driver of mutantbearing cells (3). Thus, in our experiment, we hypothesized that EZH2 inhibitors would potentially downregulate mutant BAF complex activity in neuroblastoma cell lines, by sustaining BAF complex tumor suppressor functionality. Interestingly, results from cytotoxicity tests show that EZH2 inhibitors have no significant potency against mutant BAF neuroblastoma cell lines. We conclude that EZH2 inhibitors may potentially alter the epigenetic characteristics of mutant NB cell lines, making them more vulnerable to cancer treatment, such as chemotherapy.
Introduction A crucial biological function in mammalian cells is DNA packaging, which is controlled by at least three process: (1) DNA methylation, (2) histone modification, and (3) ATP-dependent chromatin remodeling (2). The BAF protein complex, part of the SWI/SNF chromatin remodeling family, controls the accessibility of DNA sequences to transcription factors through combinatorial assembly: exchanges of subunits within BAF complex accompany transitions from pluripotent stem cell to neural stem cell and to post-mitotic neuron [Figure 1] (4). In essence, the BAF complex is responsible for various biological functions in neural development. For example, esBAF complexes are crucial to the core pluripotency transcriptional circuit and interact with embryonic stem cell specific transcription factors [Figure 2]. Likewise, npBAF (neural progenitor) promotes selfrenewal of neural stem cells, and nBAF is important to neuron specific functions, such as proper dendritic growth [Figure 2]. Approximately 19.6% of all human tumors contain mutated SWI/SNF subunits, suggesting that SWI/SNF subunits are tumor suppressors rather than oncogenes (5). However, mutated SWI/ SNF subunits may produce gainoffunction properties, resulting in oncogenic activity. For instance, ARID1A and ARID1B are mutually exclusive subunits that are mutated with different frequencies in different cancer types (1). Mutations in ARID1A and ARID1B lead to exceptionally poor outcome neuroblastoma, a disease of attenuated neural differentiation (Figure 2) (Figure 3) (1). We thus hypothesize that ARID1 alterations disrupt progression from the neural progenitor BAF (npBAF) to the post-mitotic neuron (nBAF) complex, preserving a stem cell-like state. In this project, we aimed to characterize BAF complex cell 28 PENNSCIENCE JOURNAL | FALL 2014
lines (WT, ARID1A mutant, and ARID1B mutant) using nuclear protein extracts. Utilizing this data, our subsequent aim was to further define the chromatin and methylation states arising downstream of these BAF complex mutations. We also aimed to analyze the effect of EZH2 inhibitors on mutated ARID1A and ARID1B neuroblastoma BAF whole cell lysates. Posttranslational modifications of histone proteins play an important biological role in cells. One important type of transcriptioncontrolling modification is the PRC2 complex, a multiprotein histone methyltransferase (HMT) complex . EZH2 is a catalytic subunit of polycomb repressive complex 2 (PCR2) and is involved in repressing gene expression (6,7). However, overexpression of EZH2 has shown to lead to tumorigenesis, suggesting that EZH2 enzymatic activity is a required driver of mutantbearing cells (3). It has been shown that tumorigenesis can be completely suppressed by tissue-specific deletion of EZH2, suggesting that there might be an inverse relationship between EZH2 activity in PRC2 and SWI/SNF. In other words, overexpression of EZH2 may interfere with the tumor suppressing effects of SWI/SNF complexes. Thus, in our experiment, we hypothesized that EZH2 inhibitors would kill off mutant BAF complex neuroblastoma cells, by sustaining BAF complex tumor suppressor functionality. We used GSK126, which has been shown to be an inhibitor of EZH2 methyltransferase activity in EZH2 mutant DLBCL cell lines and xenografts in mice (6). Ultimately, we attempted to see whether GSK126 would have any kind of potency against ARID1 mutant neuroblastoma BAF whole cell lysates, which may direct us toward novel and improved drug treatments.
RESEARCH
Figure 1. (Hogarty, 2013)1: BAF complexes relevant to neural differentiation. BAF complexes are chromatin remodellers through ATPdependent Brg1 activities. They have combinatorial assemblies that provide for discriminant DNA and histone binding via subunit motifs. ARD1A and ARID1B are mutually exclusive subunits providing DNA binding, histone modifying, and subunit-scaffold functions to the complex. Multiple subunits are exchanged during progression from a stem cell to a neural progenitor and finally to a mature neuron. Compelling data that retention of npBAF is associated with an aggressive phenotype comes from multiple NB transcriptome datasets (2). This points to the oncogenic role for deregulated BAF signaling in NB.
Figure 2. (Hogarty, 2013)1: Overall survival (OS) according to ARID1 gene mutation status. Median OS is 386d from diagnosis for individuals with ARID1 mutant tumors, compared to 1,689d for those with ARID1 wild type neuroblastomas.
FALL 2014 | PENNSCIENCE JOURNAL 29
RESEARCH Materials and Methods: [1] Cell lines Mutation Prevalence ARID1A and ARID1B sequence and copy number were sought in NB cell lines using multiple platforms including WES, WGS, targeted sequencing, and genomic SNP-chip. All mutations were verified using an orthogonal technique. 20 NBderived cell lines were studied overall [Figure 3]. Whole cell lysates specific to the BAF Swi/Snf complex were extracted from these cell lines. Figure 3. (Kim, 2014): Table of neuroblastoma cell lines with different ARID1A and ARID1B statuses.
30 mcg of protein samples were loaded into NuPage 48% BisTris polyacrylamide nupage gels and run under 200 voltage for 1 hour. Separated proteins within the precast gel were then transferred onto PVDF membranes using the iBlot Dry Blotting System. Membranes were incubated overnight in variously concentrated primary antibodies in 5.0% blocking buffer. After several wash steps using the TBST buffer, the membrane was then incubated in 1:3000 concentrated secondary antibodies made in goat, mouse, or rabbit for at least one hour in room temperature. Signal was then detected through luminescent light, activated through ECL chemiluminescent substrate reagents, and captured on XRay film paper through the film developer machine. Immunoblotting was done to characterize the relative abundance of specific proteins in the cell lines. Immunoblotting test runs were done several times to determine optimal conditions that would produce signal, such as primary antibody concentration, film exposure times, or type of chemiluminescence reagent. The list of antigens we attempted to detect in our nuclear whole cell lysate samples is shown in Figure 4. [4] Cytotoxicity Test The effect of EZH2 inhibitor on various neuroblastoma cell lines was also analyzed with cell cytotoxicity tests. Two experimental trials were set up that ultimately led to the similar results. For the first experiment, cell lines NB1643, CHLA15, and NAS were plated at 20,000 cells per well, while NLF was plated at 12,500 cells per well due to lack of cell quantity. GSK126 is a selective, Sadenosylmethioninecompetitive small molecule that was used to inhibit EZH2 methyltransferase activity. GSK126 product of Active Biochemicals (Cat # A 1275 Cat No. 1346574579) was utilized. The CellTiterGlo Luminescent Cell Viability Assay was used to quantify ATP levels, an indicator of metabolically active cells. This assay results in cell lysis and generation of a luminescent signal proportional to the amount of ATP present. The amount of ATP is directly proportional to the number of cells present in culture. In order to analyze the effect of EZH2 inhibition, different concentrations of GSK126 were utilized: 5 uM, 1 uM, 500M, 100nM, and 20 nM. Promega plate reader was then used to quantify ATP levels in the tested neuroblastoma cell lines: NB1643, NAS, NLF, and CHLA15.
Results: Figure 4 shows the list of antibodies that provided optimal signal. Company name, clone type, concentrations, monoclonality, polyclonality, and exposure times were all factors taken into consideration when attempting to find optimal conditions for antibody usage. While several antibodies proved to be effective, our experiments are still [2] Whole Cell Lysate Extraction – Whole cell lysate proteins inconclusive on BCL7A and BCL11A antibodies. Different were extracted from these cell lines with specific ARID1A antibodies and conditions needed to be tested first before proceeding in more experimental depth and characterizing and ARID1B statuses (8). mutant BAF complex members through immunoblotting. [3] Immunoblotting Protein concentrations of these nuclear Immunoblot analyses are ongoing to assess for altered BAF lysates were then calculated through an optical density test. complex membership in ARID1 mutant versus wild type 30 PENNSCIENCE JOURNAL | FALL 2014
RESEARCH Figure 4. (Kim, 2014): Antibodies specific to BAF whole cell lysates extracted from neuroblastoma cell lines. Antibodies were tested to determine conditions for optimal signal. Antibodies were also used to characterize differences between neuroblastoma cell lines with different ARID1A and ARID1B statuses.
tumors. EZH2 inhibitors do not show activity against ARID1 mutant or wild type neuroblastoma cell lines, in contrast to other tumors with BAF complex mutations [Figure 5]. With increasing levels of GSK-126 inhibition, the percentage of cell death with increasing concentrations of [EZH2] was statistically insignificant in cell lines NAS, NLF, NB1643, and CHLA15. GSK126 is a known EZH2 inhibitor drug and increasing concentrations were used to treat cells of different neuroblastoma cell lines. As shown in figure 5, even with high concentrations of 5000 nM [GSK126], there does not seem to be significant reduction of relative cell survival. Our results show that even with high concentrations of the drug, no less than 85% of cells relatively survived in our cytotoxicity test assays. Ultimately, Cell TiterGlo Assay was done to quantify cell ATP levels after exposure to inhibition and results show that there is no significant potency of the drug against treated cells.
Figure 5. (Kim, 2014): Graphs derived from cytotoxicity tests, examining the effects of GSK126 (EZH2 inhibitor) on various neuroblastoma cell lines (NAS, NLF, NB1643, CHLA15).
FALL 2014 | PENNSCIENCE JOURNAL 31
RESEARCH
Discussion Immunoblot analyses are ongoing to assess for altered BAF complex membership in ARID1 mutant versus wild type neuroblastoma tumors. We are continuing antibody tests to search for conditions that would provide optimal signal. Future work from this project includes co immunoprecipitation in order to analyze how strongly proteins within the BAF complex are bound together. Coimmunoprecipitation data can then be compared to immunoblot data to confirm the presence of proteins of specific BAF complex type during time-specific phases of neural development. Furthermore, future work should entail CHIPS equencing to define the chromatin and methylation states arising downstream of BAF complex mutations [Figure 6]. The tumor DNA can be cross-linked to a protein of interest and sheared. Target-protein bound DNA can then be captured by immunoprecipitation and then released for highthroughput sequencing. Reads from each end of DNA fragment can then be genomically identified. The output would provide regions of target protein binding, allowing for comparisons of genes with select binding proteins or histone marks of interest. Marks among neuroblastoma cell lines of different ARID1 mutation status as well as their gene expression can be compared. In essence, the goal of CHIP-Sequencing is to analyze whether BAF complexes abnormally localize on the genome and whether they are dysfunctional at the sites that they bind. We hypothesize that mutated BAF complexes presumably bind to DNA differently from the way that wild type BAF complexes bind. However, if it is the case that mutated BAF complexes bind to DNA similarly to 32 PENNSCIENCE JOURNAL | FALL 2014
wild type BAF complexes, we predict abnormal genetic activity downstream of mutated BAF complexes. Further information on how mutated BAF complexes behave on the genome can point to new scientific proposals. Interestingly, our data shows that EZH2 inhibitors do not have efficacy against ARID1 mutant or wild type neuroblastomas, in contrast to other tumors with BAF complex mutations, as demonstrated in scientific literature. Thus, from our experiment, we conclude that EZH2 inhibitor has no efficacy against ARID1 mutant or wild type neuroblastoma cell lines. However, it is possible that cells treated with EZH2 inhibitors have altered epigenetic functionality. For example, chemotherapy has strong efficacy against cells with wild type ARID1 genes, while it has significantly decreased efficacy against mutant neuroblastoma tumor cells. It would be interesting to analyze whether EZH2 inhibitors alter the epigenetic behavior of mutant ARID1 genes, such that the treated mutated cells are more susceptible to death when exposed to chemotherapy [Figure 7]. Thus, future experimentation should aim to analyze the behavior of mutant ARID1 neuroblastoma cell lines when treated with EZH2 inhibitors. This may ultimately be promising in deriving more effective treatment plans for neuroblastoma cancer patients.
Conclusion High r isk neuroblastoma remains a lethal cancer currently treated with intensive and highly morbid cytotoxic therapy. Thus, new therapies are needed. Neuroblastoma can be considered a cancer of aborted neural differentiation, and emerging data suggest that altered epigenetic programs mediate this. Furthermore, improved outcomes using retinoids
RESEARCH new therapeutics targeting these disrupted epigenetic programs.
Figure 6. Workflow for CHIPSeq. (Hogarty, 2013): Tumor DNA is crosslinked to protein of interest and sheared. Targetprotein bound DNA is captured by IP and then released for highthroughput sequencing. Reads from each end of DNA fragment map to the genome are identified. The output regions enriched for binding of the target protein allowing for comparisons of genes with select binding proteins or histone marks of interest. We will further compare such marks among NBs of different ARID1 mutation status, and compare marks with expression of the proximate gene. [Note: this is not experimental data but an example of potential output].
Acknowledgements I would like to sincerely thank Dr. Michael Hogarty, Xueyuan Liu, and Annette Vu for supporting, teaching, and encouraging me throughout this research project. I would also like to thank CURF for funding this project in part through the Pincus Magaziner Undergraduate Research Grant.
References
Figure 7. Workf low for EZH2 Inhibitor Efficacy. (Kim, 2014): Future experimentation should aim to analyze efficacy of chemotherapy on mutant ARID1 neuroblastoma cell lines pre treated with EZH2 inhibitor. [Note: this is not experimental data but an example of potential output].
1. M. Sausen, et al., Integrated genomic analyses identify ARID1A and ARID1B alterations in the childhood cancer neuroblastoma. Nature Genetics.45, 121 9 (2013). 2. N. Singhal, et al., ChromatinR emodeling Components of the BAF Complex Facilitate Reprogramming. Cell.141, 9439 55 (2010). 3. S.K. Knutson, N.M. Warholic, V.M. Richon, Durable tumor regression in genetically altered malignant rhabdoid tumors by inhibition of methyltransferase EZH2. Proceedings of the National Academy of Sciences.110, 79227 927 (2013). 4. J. Wu, et al., Proteomic and bioinformatic analysis of mammalian SWI/SNF complexes identifies extensive roles in human malignancy. Cell.4, 941 08 (2007). 5. C. Kadoch, et al., Proteomic and bioinformatic analysis of mammalian SWI/SNF complexes identifies extensive roles in human malignancy. Nature Genetics.45, 11 1 (2013). 6. H. Li, R. Zhang, Role of EZH2 in epithelial ovarian cancer: from biological insights to therapeutic target. Frontiers in Oncology.3, 2-7 (2013). 7. M.T. McCabe, et al., (2012). EZH2 inhibition as a therapeutic strategy for lymphoma with EZH2activating mutations. Nature.492, 1081 12 (2012). 8. Abcam. (2014). Sample preparation (WB guide). Retrieved from Abcam website: http://www.abcam.com/index.html?pageconfig=resourc e&rid=11379 9. L. Ho, G.R. Crabtree, (2010). Chromatin Remodelling During Development. Nature.463, 4744 83 (2010). 10. J.I. Wu, et al., Regulation of Dendritic Development by NeuronSpecific Chromatin Remodeling Complexes. Neuron.56, 941 08 (2007). 11. A.S. Yoo, G.R. Crabtree, (2009). ATPD ependent Chromatin Remodeling in Neural Development. Current Opinion in Neurobiology.19, 120 126 (2009).
show that differentiation therapy can have efficacy in neuroblastoma. This research project and future related work provide a unique opportunity to learn more about the aborted differentiation program in neuroblastoma, to generate new hypotheses regarding neuroblastoma etiology, and to provide model systems to develop
FALL 2014 | PENNSCIENCE JOURNAL 33
RESEARCH
34 PENNSCIENCE JOURNAL | FALL 2014
Call for Submissions Looking for a chance to publish your research? PennScience is accepting submissions for our upcoming Spring 2015 issue! Submit your independent study projects, senior Design projects, reviews, and other original research articles to share your work with fellow undergraduates at Penn and beyond. Email submissions and any questions to pennscience@gmail.com.
Research in any scientific field will be considered, including but not limited to:Â
Biochemistry | Biological Sciences Biotechnology | Chemistry Computer Science | Engineering Geology | Mathematics | Medicine Physics | Psychology
FALL 2014 | PENNSCIENCE JOURNAL 35
www.pennscience.org PennScience is sponsored by the Science and Technology Wing at the University of Pennsylvania.