Scientia - Fall 2013

Scientia

Autumn 2013 Issue III

T

he cover of this third issue of Scientia features a curvaceous detail of Henry Moore’s sculpture, “Nuclear Energy,” captured by Executive Board Member, photographer Carrie Chui. The sculpture, occupying the very historical location of what is now the corner of Ellis Avenue and 57th Street, memorializes the success of what was the first self-sustaining nuclear chain reaction. Contrary to what you might have imagined as a cataclysmic event, the initial reaction was described to be quite the opposite—it was “too weak to power even a single light bulb.” And though the reaction may have been arguably anticlimactic, its success has unarguably changed science forever.

Since its momentous exhibition, the first self-sustaining nuclear chain reaction has set off a chain of advances in a new era of science and technology, a legacy that has prevailed for over 70 years. Immediately following the success of the chain reaction was not only the development of the atomic bomb, its wellknown catastrophic derivative, but also various outgrowth initiatives reaching areas of climate change, physical science, chemical, and medical research. After World War II, scientists like Enrico Fermi who were a part of the war effort continued work in the scientific arena, making remarkable progress in chemistry and particle physics. Likewise, medical research reaped its benefits in the post-war period—such was apparent at the Argonne Hospital, which introduced the use of radiation in the treatment of cancer.

As you flip through the pages of our third edition of Scientia, you will find a discussion of various current and thoughtprovoking scientific inquiries, ranging from the mechanics of visual cognition to paternal influences on offspring. Indeed, though the pivotal reveal of the chain reaction is in the past, Nuclear Energy reminds us that science, as we hope you will notice in Scientia, is always advancing and inspiring.

Scientia Scientia Inquiries: Interview with Dr. Matthew Tirrell Microtubules: The Miniscule Workhorses DiscerningThisfromThat:VisualCognitionwithDr.DavidFreedman Dr. Peggy Mason Finds Empathy in Rats

Scientia Abstracts:

4

13

Elucidating the Mechanism for CED-4 and CED-3 Binding via Biochemical Analyses Paternal Roles in the Educational Outcomes of Sons and Daughters Diet and Health Capital: An American Case Study

Sacrificial Polymers and Their Use in Patternable Air-gap Fabrication David Goldfeld

Extension of a Phage-V. cholera Interaction Model to Include Vaccination Strategy and Seasonality in Growth Rate Kipp Johnson

Produced by The Triple Helix at the University of Chicago Layout and Design by Charles Pena, Co-Director of Production Cover Photograph by Carrie Chui, Co-Director of Production Cover Letter written by Daniel Frankel, Director of Marketing Scientia Team: Patrick Delaney, Khatcher Margossian, Luizetta Navrazhnykh, & Michael Begun

15 18

Autumn 2013

About Scientia Dear Reader, The Triple Helix, Inc. (TTH) at the University of Chicago is proud to present the third issue of Scientia, our most recent venture as a chapter. Scientia, with its team of committed editors, strives to showcase the highest quality original undergraduate research. Additionally we strive to provide a resource for students interested in pursing research in the future.

We now feature three different types of writing pieces: investigative “Scientia Inquiries,� short research abstracts, and full-length research articles. With these additions we have expanded the possibility of authorship to all students, regardless of previous research experience.

We are excited to present this issue to you, both as an example of excellent work and as a resource for our research community. Scientia is currently seeking writers and editors to contribute to future issues. We encourage all those with interest to get involved.

Sincerely, Patrick M. Delaney Editor in Chief, Scientia

2

Scientia

About The Triple Helix

at The University of Chicago

The Triple Helix, Inc. (TTH) is the world’s largest completely student-run organization dedicated to evaluating the true impact of historical and modern advances in science. Of TTH’s more than 25 chapters worldwide, the University of Chicago chapter is one of the largest and most active. We, TTH at The University of Chicago, are extremely proud of our chapter’s 2013 accomplishments.

As we enter our seventh year as a full-fledged chapter, we continue to work closely with illustrious faculty members from all departments of our University.

We have had the pleasure of interacting with Nobel laureates, winners of the Lasker-DeBakey Clinical Medical Research Award, as well as more locally-loved recipients of our University’s Quantrell Award for Excellence in Undergraduate Teaching. Enclosed in this issue of Scientia, you can find an interview with one of our closest colleagues, the founding Pritzker Director of Chicago’s Institute for Molecular Engineering, Matthew Tirrell. Without the help of him and his department, our chapter would not be what it is today.

This year has been marked by tremendous growth, as our chapter has expanded its efforts to increase our members and audiences. TTH prides itself on hosting students who have declared more than 30 of the University’s different majors and minors. Participation in The Triple Helix at UChicago is an educational and enjoyable undertaking that has proven to be a valuable experience for students interested in undergraduate research. Scientia, our journal of original University of Chicago undergraduate research, specifically highlights student work that often goes unrecognized.

Thank you for your continued support and please enjoy the third issue of Scientia.

Daniel Frankel Director of Marketing

3

Autumn 2013

Scientia Inquiries Interview with Dr. Matthew Tirrell, Pritzker Director of the Institute for Molecular Engineering Jake Russell In 2011, Dr. Tirrell came to the University of Chicago to lead the new Institute for Molecular Engineering.

A pioneering researcher in biomolecular engineering and nanotechnology, his work focuses on the surface properties of polymers, such as adhesion, friction, and biocompatibility. Dr. Tirrell’s research has generated valuable insights into polymer properties and the creation of new materials based on self-assembling synthetic and organic compounds. As the founding Pritzker Director of the Institute of Molecular Engineering – UChicago’s first real foray into engineering – Dr. Tirrell plans to tackle problems in areas ranging from health care to the environment by integrating expertise from various disciplines.

Dr. Tirrell first became interested in material science as an undergrad at Northwestern University, where he majored in chemical engineering and participated in a co-operative program between Northwestern and a polymer production company in his home state of New Jersey. This combination gave him experience working in an industrial organic materials lab and shaped his research interests. After completing his PhD in polymer science in 1977 at the University of Massachusetts, he has served as Chair of Chemical Engineering at the University of Minnesota, Dean of the College of Engineering at UC Santa Barbara, and Chair of the Bioengineering Department at UC Berkeley. The Triple Helix recently had the opportunity to speak with Dr. Tirrell about his career, current research in biomolecular engineering, and visions for the future of the Institute.

J: Thanks for your time today. To start out with, what was your thesis based on? I have here mechanical stresses on enzymes.

M: Well the idea is that enzymes are big molecules, and when you apply shearing forces or fluid mechanical

4

forces to big molecules, they deform in shape. […] I worked mainly on one enzyme, because we knew that we needed a massive molecule. So the enzyme that I worked on was urease, which converts urea to ammonia and carbon dioxide. Since it’s so big, it was more susceptible to these kinds of things.

J: So then you extrapolated the results from that to other enzymes? As in, if it deforms this enzyme then it will deform others..

M: Well, to tell you the truth, the main result of my thesis was that you didn’t have to worry about this too much. Yet people were concerned about it, because biotechnology was just taking off then in the mid-to-late70s, and people were wondering, if you put these things in reactors and stirred them, were you going to inactivate all the enzymes, and things like that. We were able to show that it really took a lot to mess these things up. And another thing that was a bigger effect was that in order to get very high shear rates, you generally have to go to situations where the ratio of the surface area containing the solution to the volume of the solution is pretty high.

In other words the things are going to be exposed to a lot of surface. It turns out that enzymes and other polymer molecules stick to surfaces. That turned out to be a bigger factor than the shearing itself – the interaction of the enzymes with the surface. You’d get a lot of deactivation of the enzymes due to the surface interactions.

J: I see. What about after that, at Minnesota?

M: Well I still started out working mainly on synthetic polymers even though I had done my thesis on enzymes.

Partly that was because of the ideas that I had and, you know, even back in the 70s when you became an assistant professor, the first thing you had to do was start applying for research grants. Its sort of another method of peer review in a way, you know, you put your ideas out there,

Scientia and see if your ideas are competitive enough for agencies to fund them. […] And so I actually got really interested in the flow of polymers in narrow channels and porous media. This was at a time when the price of oil was really high and all of the oil companies were investigating things to enhance the recovery of oil. And oil isn’t really sitting below the ground in big lakes that they can just pump out, it’s usually permeating porous rock. So since oil was so valuable and you can get such a high price for it, oil companies were investing in surfactants and polymers and pushing them into the soil to push out more oil.

J: Right, I read that they used to use water but they discovered that water just went through the oil.

M: Yeah, I mean if you try to push a really viscous liquid with a not-so-viscous liquid, the less viscous liquid just goes by and leaves the other stuff behind, so you needed to adjust the interfacial tension and viscosity of the water.

J: What was your research at Santa Barbara focused on?

M: Well by that time we had moved well into things related to interactions of synthetic materials with biological systems. So we called my lab at Santa Barbara the biomolecular surfaces lab: how do you manipulate the interactions between synthetic materials and biological systems? There were a lot of interesting angles to that.

J: I can imagine. So then you were head of bioengineering at UC Berkeley for two years. That seems like quite a big leap to come to somewhere like Chicago that had no previous engineering program.

M: But that’s just the point, that’s why I came.

J: Why don’t you tell me about that, how did that come about?

M: Well it came about by a headhunter calling me, to fill the position of founding director of the Institute for Molecular Engineering. I had never heard of this, but you know, I had respect for [the University]. I’d been here before, I’d given talks on the campus, I had friends here, so I knew this was a great place. And it turned out that my department head at Minnesota had gone to UChicago for graduate school. He was a chemist – even though he was in the chemical engineering department, he was a chemist. And he used to make fun of me for going to Northwestern. At any rate, you know, I had high regard for this place, and then when I saw what was really being contemplated, starting a whole new engineering program from scratch, that was very interesting. So this is really the chance of a career, almost nobody gets to do what I’m doing.

J: Were you intimidated at all by the size of the job?

M: I still am. Yes, I was intimidated, maybe more so now that I’m actually in the middle of it than I was when I accepted it. You know, I realize how important this is for the University, how much the university is investing in it.

I’m absolutely committed to making this a huge success.

J: Right, it’s not just the fact that we’ve never had molecular engineering before, it’s that we’ve never had any engineering at all.

M: Well, this is important, to get this right. But you know, I’ve had experience with building things before, even at Berkeley. The department was started when I came there, we hired new faculty during the years that I was there. So, you know, I like to hire people. I like to convince them that what I’ve decided to do is good.

J: How many have you hired here in the last couple of years?

M: So far there’s just four of us including me, so I’ve hired three. […] We will probably hire three or four more people this year.

J: Wow, it’s expanding fast. Let’s talk about the Institute itself and what it does. I know a lot of my friends that I talk to, science and non-science students alike, most of them know it exists, but they’re not really sure what the Institute for Molecular Engineering is designed to do.

M: The name isn’t really self explanatory.

J: All they know is that it says Engineering, and UChicago doesn’t have engineering!

M: [Laughs] Well it does now. You know, probably a more accurate name in some ways, or at least a name that would evoke the right things for people, would be Department of Molecular Engineering. That’s the sort of size we’re going to have. We have the authority to hire 25 faculty members right now, so that’s about the size of the chemistry department. On the other hand, in the university organizational chart, I have the status of a dean. […] So this is one of my standard lines: we have the size of a department, the status of a school or a division, but the style of a research institute. And by that I mean, we are not recreating other schools of engineering here. Other schools of engineering, like Berkeley, or Northwestern, or MIT, divide engineering into disciplines. Our view is exactly the opposite. We are combining disciplines into molecular engineering. So we’re about integrating and synthesizing things that come from different disciplinary traditions. So that’s the sense in which we are like a research institute. We want to bring disciplines together to tackle big problems. We’re thinking about health care, we’re thinking about water, we’re thinking about energy storage, we’re thinking about new materials, stuff like that. We’re not just going to recreate the standard model.

5

Autumn 2013 J: So I guess that still does kind of fit in with the University’s principles of combining all the disciplines, with the Core and stuff. Even though it says engineering, it’s really not like engineering at other schools.

M: It won’t be organized like it. The problems we’re tackling are being tackled at other good engineering schools, but I think we’re going to have a unique and powerful way of tackling them. We’re like a department in that we’re going to have a graduate degree program and an undergraduate degree program. Our undergraduate students in the college will be able to concentrate in Molecular Engineering.

J: I hadn’t heard about that. When is it going to take effect?

M: Well we haven’t gotten our undergraduate degree approved yet, but we will work on it next year. We’re working on getting our graduate program approved this year first.

J: So then we’ll have graduate students coming in specifically for this.

M: And they’ll get PhD’s in molecular engineering.

And then the year after, we’ll set things up so people can get bachelor’s degrees or probably minors too. You could get your bachelor’s degree in chemistry with a minor in molecular engineering.

J: And will there be new classes? Taught by the faculty that you’re hiring?

M: Absolutely. We’ll create a whole new curriculum, at the undergraduate and graduate levels. Classes too: molecular electronics, for example. Molecular therapeutics.

J: So I read that the Institute will be working very closely with Argonne. Could you tell me a bit about that and how that partnership is going to work?

M: Well you probably know that the University of Chicago manages Argonne National Laboratory for the Department of Energy. All the Department of Energy labs and other agency national labs are usually not managed directly by the government, but are managed for the government by a university or a private company. So we manage Argonne, and Fermilab, for that matter. […]

Another part of President Zimmer’s thinking in starting engineering was that this would be an effective way of attracting new personnel to Argonne as well, and making our connection with Argonne better. So everyone that’s been hired so far has a 25% appointment to Argonne. It means that Argonne pays 25% of our salary.

J: So would a lot of research then be happening at Argonne, as opposed to here?

M: Yeah, a significant fraction of it. I mean, Argonne

6

is providing space, office space, access to facilities, and probably access to DOE funding. So a national lab has some capabilities that a university never has. They have a big synchrotron there for doing really powerful x-ray scattering experiments, they have a nanofabrication center that has equipment we don’t have here. Things like that. So there are some really special capabilities that Argonne has that will be part of the IME constellation.

J: Why don’t we talk about your current research now?

I have a few titles from your recent papers: “Targeting atherosclerosis by using modular, multifunctional micelles”, “Viscosity and interfacial properties in a mussel-inspired adhesive coacervate”, and “Chain length dependence of antimicrobial peptide-fatty acid conjugate activity”. Those all sound very complex.

M: Well, part of it is on these polyelectrolyte complexes that I mentioned. The other part is in designing and making nanoparticles that interact with the body, and we have two themes within that. One concerns particles that could be injected into the bloodstream, with surface proteins… [looks through brochure and points at attached picture] … This is a schematic or artist conception of one of our particles.

J: Wow, that looks deadly.

M: [Laughs] Well it’s not supposed to be deadly.

Out here are fragments of protein that could be all the same or a mixture, with the capacity to recognize some characteristic of diseased tissue. In particular we’ve worked on ones where these peptides recognize blood clots.

Blood clots form in tumors, because the blood vessels in tumors are almost always leaky. Also atherosclerosis, which is the build-up of lipids in blood vessel walls. And we have particles that can detect those things, accumulate there, and then emit fluorescence or some other kind of signal to tell us there’s a problem here. So we’re actually trying to develop problem-seeking nanoparticles that look like this.

J: And then what would you be able to do after that?

M: Well, you could intervene in some way, depending upon what you’re talking about. Either deliver a drug locally, or put a stent in or something that would guard against the atherosclerosis. So the other thing we’re doing is making particles like that that stimulate the immune system. That might raise antibodies to these things. So we’re making a kind of synthetic vaccine.

J: So how far out are we from seeing actual benefits from this?

M: I don’t know, three to five years. I mean we’re already using these things in animals, so if something looks very good, we could start to move it towards human

Scientia

Graphic credit: Peter Allen

trials. Another one of the things we’re doing is studying polyelectrolyte complexes. So if you take a highly negatively charged polymer even in dilute solution, and a highly positively charged polymer in dilute solution, and you pour those two solutions together, you very frequently get a new phase formed. The polymers of opposite charge come together and either form a precipitate or, more interestingly, they form a separate liquid phase. The water phases combine and remain miscible, but then you get a dense, polymer rich phase.

It turns out that some of our interest in this was stimulated because a lot of marine organisms use these polymer complex phases to glue things together underwater, to attach to surfaces.

J: Like mussels.

M: Like mussels, yes. Like various kinds of worms that need to build habitats. They secrete oppositely charged proteins and use this kind of thing as glue because even these water soluble proteins phase separate in water, and they make very effective biological glue, which is hard to do. So we’ve been studying and trying to figure out how to exploit that type of stuff for man-made purposes.

J: And has your work there yielded any useful results?

M: Yeah, we’ve been able to create new synthetic materials based on what we’ve learned from mussels. You know, these coacervate mixtures are found in things like e-readers, like a Kindle or something. They create an image by applying a field that either brings a black or a white particle to a screen. And in order to get high resolution, you have to keep the particles in a very fine state of dispersion. It turns out that these polyelectrolyte complexes are often used to encapsulate these particles.

So coacervates are used in things like e-readers all the time.

J: Wow, so there’s a huge range of applications.

M: Yeah, so we’re trying to make better materials for certain kinds of applications. I think there are lots of possibilities.

Microtubules: The Miniscule Workhorses Austin Yu Microtubules are miniscule, hollow cylindrical polymers at work in all eukaryotic cells. They simultaneously serve as both structural units and dynamic regulators of cellular organization. Among other functions, microtubules transport organelles within

the cell and segregate chromosomes during mitosis.

These crucial polymers are on the order of micrometers (one-millionth of a meter) long, and mere nanometers (one-billionth of a meter) wide. As is often the case with microscopic things, we humans easily overlook

7

Autumn 2013 microtubules in our macroscopic lives. Not so for Dr. Mohan Gupta, a pioneer in the research of mechanisms underlying tubulin and microtubule function.

Dr. Gupta’s interest in microtubules began during his undergraduate and graduate studies in biochemistry at the University of Kansas. There, he became fascinated with the ability of microtubules to do physical work, and wanted to better understand how cells make microtubules do what they do. Since arriving at the University of Chicago four years ago, Dr. Gupta has been investigating microtubules and their protein subunits, tubulins, extracted from Saccharomyces cerevisiae yeast cells. One advantage of using yeast cells, he says with a grin, is that “[we] don’t have to extract the tubulin from many pounds of mashed cow and pig brains.” Moreover, says Dr. Gupta, yeast tubulin is much easier to mutate than mammalian tubulin, so it is a better model for studying mutant variants. Presently, Dr. Gupta has three main research foci: first, he is concerned with kinesin-8, a class of motor proteins that reorganize microtubules.

Second, he is researching the role of tubulin mutations in congenital disorders, and the functions of the eight types of human tubulin. Third, he is delving into the workings of antimitotic cancer drugs.

Microtubules are dynamically unstable polymers; that is, individual microtubules are constantly growing or shrinking. This dynamic instability allows microtubule networks to rapidly reorganize themselves, a process facilitated by kinesin proteins. Dr. Gupta and his colleagues discovered how kinesin-8 motor proteins align the mitotic spindle in cell division, allowing for proper chromosome segregation.

They also found that kinesin-8 proteins regulate microtubule growth and shrinkage by either binding or unlinking tubulins at the ends of the microtubule [1]. As a result, Dr. Gupta believes that drugs which inhibit kinesin-8 have excellent potential as cancer therapeutics. By inactivating kinesin-8, they deregulate the dynamic instability of microtubules, preventing the rearrangement necessary for cell

8

division and halting the division of tumor cells in culture.

It is known that humans with specific mutations in β-tubulins suffer from congenital disorders that impact the nervous system. Dr. Gupta’s second area of research investigates the impact of different tubulin mutations in humans. He and his colleagues found that mutations in TUBB2B, one variant of β-tubulin, disrupt axon formation and cause severe issues in eye movement [2].

A longstanding mystery is why humans even have eight different types of β-tubulin; the function of individual variants is currently unknown. Dr. Gupta is studying how disease-causing mutations inhibit specific cellular functions of tubulin. By doing so, he hopes to understand how mutations in each of these β-tubulins cause disease and shed light on the distinct roles that each type of tubulin plays in the microtubule and in the cell.

Dr. Gupta’s third area of research involves microtubules and the spindle assembly checkpoint (SAC). The SAC is a group of proteins that gather at the mitotic site and disperse as microtubules are attached to the kinetochores. This checkpoint prevents chromatid separation until enough microtubules are anchored to the kinetochores, ensuring that chromosomes are properly segregated. If the SAC detects issues in microtubule and kinetochore assembly, it will delay mitosis and can eventually cause programmed cell death, or apoptosis. An analogy for the function of the SAC, relates Dr. Gupta, “is holding together paired, multicolored socks with a plastic clip so that a blind person can sort them.” An important class of

Reproduced from [3]

Scientia anticancer drugs activates the SAC and induce apoptosis, whereas related drugs simply delay cell division. Dr. Gupta wants to better understand how some drugs that stabilize microtubules in the mitotic spindle also induce cell death. He hopes to bolster the development of novel anticancer drugs that can activate the SAC to block cell division. For instance, drugs which target the checkpoint directly, rather than microtubule dynamics in general, are predicted to increase specificity and have far less adverse References 1. Su X, Qiu W, Gupta ML, Pereira-Leal JB, Reck-Peterson SL, Pellman D. Mechanisms underlying the dual-mode regulation of microtubule dynamics by kip3/kinesin-8.

Molecular Cell 2011 Sep;43(5):751-63.

side effects in humans.

Dr. Gupta believes that a better understanding of molecular dynamics at a microscopic level will allow us to tackle preeminent human diseases at a macroscopic level.

For instance, determining how microtubule dynamics are regulated is one key to the development of novel cancer therapeutics. By better understanding the mechanistic aspects of microtubules, we can make great strides toward improving human health.

2. Cederquist GY, Luchniak A, Tischfield MA, Peeva M, Song Y, Menezes MP, Chan W, Andrews C, Chew S, Jamieson RV, Gomes L, Flaherty M, Grant PE, Gupta ML, Engle EC. An inherited TUBB2B mutation alters a kinesin binding site and causes polymicrogyria, CFEOM, and axon dysinnervation. Human Mol. Gen. 2012 Dec;21(26): 5484-99.

3. http://www.flickr.com/photos/29225114@N08/8658765020/

Discerning This from That: Visual Cognition with Dr. David Freedman Gloria Wang When you look around a room, what do you notice?

Do you see the stacks of the books that you have yet to read for your upcoming midterms, or the bundles of clothes waiting to be washed? You see them (and let them be), but how does your brain learn whether an object is a book or a piece of clothing, or determine that two objects are both books or both clothes? Dr. Freedman, a neurobiologist here at the University of Chicago, seeks to find out how we learn about what we see.

Dr. Freedman began studying electrical engineering at the University of Rochester, but decided that he wanted to focus on basic research. His interest in vision neuroscience sparked during his second year, when he took a psychology class on sensation and perception by Dr. David Williams, a vision expert who studies the retina.

This class exposed him to vision neuroscience, which seeks to understand the mechanisms involved in the brain during visual perception, especially the processing and interpreting of incoming stimuli. Inspired by the subject, Dr. Freedman went on to MIT for graduate school to study late visual perception and processing in the brain (important in functions such as attention) before starting his lab at the University of Chicago.

Currently, Dr. Freedman’s lab studies the higher levels of the visual processing hierarchy, focusing on more

cognitive aspects of vision, such as how the brain learns and recognizes various visual stimuli, with a specific emphasis on categorization. Recognition of visual stimuli gives what we see meaning in our minds, which affects our behaviors and decisions. They developed a MATLAB toolbox called MonkeyLogic to create programs that act as games and tasks [1]. Such tasks include a directioncategorization task, where arrows pointing in various directions are assigned to two categories; a shapepair association task, where various pairs of shapes are grouped together [2]; and a memory matching task, which involves determining whether two quickly flashed pictures are identical [3]. These tasks are used to determine how the brain assigns meaning to the objects in order to think and make decisions about them. Rhesus monkeys, a standard animal in neuroscience labs, perform these various tasks in Dr. Freedman’s lab. In some studies, the monkeys train for each task and then are evaluated based on performance; in other studies, the monkeys are observed while learning the task. The first method evaluates how the brain thinks about the visual stimuli it has already experienced, while the second evaluates how the brain attaches meaning to visual stimuli.

The data from these experiments are manyfold, involving behavioral, neural, and computational

9

Autumn 2013

Reproduced from [5]

analysis components. During these experiments, the lab members observe the behavior of the monkeys during these tasks by tracking their eye movements, button presses, and joystick movements. In addition, they measure the neural activity in various locations of the brain to see what parts of the brain are active and changing during these tasks, and the activity is then analyzed through various computational analyses. From the data, they can determine the relationship between the neural activity when learning or thinking to the behavior the monkey displays as it answers the question.

In this way, Freedman and his team can determine what parts of the brain are involved in higher-level processes like memory and categorization, and the effect of these neural activities on decision-making.

In contrast to basic sensory vision, which results from merely seeing an image, the field of cognitive vision deals with the higher-order process of interpreting and understanding what one sees. Some basic visual processing may involve identifying or paying attention to an object, while more complex forms involve recognizing, categorizing and comparing images that often are represented as complex patterns in the brain.

According to Dr. Freedman, the information we obtain from basic sensory vision is well understood. These early steps of visual processing involve the occipital lobe, specifically the primary visual cortex, the first cortex to receive such information. However, he says, “We know less and less about what’s happening in higher-level

10

visual processing areas...As you move higher up in this visual processing hierarchy, areas that are at higher levels are not only involved in visual stimulus processing, they are also involved in cognitive functions related to vision” [3]. Such cognitive functions include attention, learning, and the formation of visual memories.

Dr. Freedman’s lab has made many discoveries in the field. In one study, Freedman and colleagues used a delayed-matching-to-category task, which involved categorizing various computer-generated cat-dog hybrid morphs as cats or dogs, which suggested flexibility of the prefrontal cortex in learning and re-learning [4]. In a recent study involving a direction-categorization task, they found that lateral intraparietal (LIP) neurons, which are located in the parietal cortex, show more prominent signals during categorization than prefrontal cortex (PFC) neurons, although both appear to be involved in cognitive visual processes [5]. Unlike these areas, the middle temporal and inferior temporal cortices are involved in basic visual sensory perception, showing that different parts of the brain become involved as the visual processes become more complex [3].

Figuring out the complete mechanism and neurons involved will prove to be a very important discovery.

When an individual sees, he or she experiences a “visual environment,” and the brain has to make sense of the various objects that he or she sees. This, in essence, becomes a “visual experience,” which is important in developing our memory and our understanding of these

Scientia various objects by making associations, categories, and meanings with our previous experiences and memories.

A fuller comprehension of our surroundings leads us to make certain behaviors as a response. By figuring out which areas of the brain are involved and how they are involved, not only will these higher cognitive functions be intricately understood, but these breakthroughs can also expand beyond the field. According to Dr. Freedman, this includes “[treating] diseases that affect recognition, decision making, or memory, and that includes many diseases such as Alzheimer’s, schizophrenia, autism, and

Attention Deficit Disorder” [3].

While the broader consequences of these discoveries are large, the driving force in Dr. Freedman’s lab is the search for the answer to a basic question, the process behind, as he describes, “how the brain recognizes the meaning of information and how we make decisions based on that incoming visual information” [3]. Thanks to the brain’s remarkable ability, we are spared from accidentally washing our books and reading our clothes, and it is worth discovering how the brain does it.

References

3. Freedman, D, interview about research at University of Chicago, 2013, April 3. (G. Wang, Interviewer).

4. Freedman, D.J, Riesenhuber, M., Poggio, T., Miller, E.K. Categorical representation of visual stimuli in the primate prefrontal cortex. Science.

2001;291:312-316.

5. Fitzgerald, J.K., Swaminathan, S.K., Freedman, D.J. Visual categorization and the parietal cortex. Frontiers in Neuroanatomy. 2012;6:1-6

1. Neurophysiology of Visual Learning, Memory and Recognition [Internet]. The Freedman Laboratory, University of Chicago; 2013 [updated 2013 Jan 26; cited 2013 Apr 30]. Available from: http://www.freedmanlab.org 2. Fitzgerald, J.K., Freedman, D.J., Fannini, A., Bennur, S., Gold, J.I., Assad, J.A. Biased associative representations in parietal cortex. Neuron. 2013;77(1):180-191.

Dr. Peggy Mason Finds Empathy in Rats Michika Maeda Empathy, the capacity to recognize, identify with, and act in response to the feelings of others, plays a vital role in our social interactions. The subject of empathy also sparks lively discussions in philosophy and psychology: researchers often debate whether nonhuman animals are capable of empathy. Dr. Peggy Mason, Professor of Neurobiology, thinks yes. “I am of the belief that there is no chasm between us and other animals, other primates, other orders,” she says. “There are differences, but just in degree, not in quality” [1]. Pro-social behavior, consisting of “actions that are intended to benefit another,” is commonly motivated by empathy in humans [3]. Comparatively, recent research by Mason has found strong evidence of pro-social helping behavior in rats, suggesting that variations of empathy are shared across animal orders.

Dr. Mason didn’t always study empathy. After researching pain modulation for more than 25 years, she shifted her focus to the biological basis of empathic behavior [2]. When Science published neuroscientist Jeffrey Mogil’s research on emotional contagion in rats, the journal asked Mason to comment. Mogil’s research suggested that when mice observe other mice in pain, they feel more pain themselves. In one experiment,

when a mouse received a painful injection, seeing another mouse in pain begat more pain in the mouse that received the noxious stimulus. Defined as the tendency to sympathetically ‘catch’ the feelings of others, emotional contagion “is a really exciting thing, a very low level of empathy,” notes Mason [1].

After reading Mason’s blog post on empathy for Science, then-PhD candidate Inbal Ben-Ami Bartal contacted Mason, expressing interest in researching empathic behavior using rats. With Dr. Jean Decety, Professor of Psychology and Psychiatry, the three scientists developed the paradigm used in the recent study. Importantly, the study uses a form of psychological distress, not physical pain. “It’s shocking that this works, because while the animal is trapped, the rat is not even constrained; it’s a moderate stressor, not extreme pain” [1]. Despite the mild extent of the stressor and the fact there is absolutely nothing done directly to the free rat, rats still reacted to help their fellow cagemates.

To investigate the effect of social distress on pro-social behavior in rats, two rats were placed together in a cage: one rat was trapped in a Plexiglas restrainer that could only be opened from the outside by the other rat. The restrainer system was designed to be neither too difficult

11

Autumn 2013 nor too easy to open, though the process still challenged the rats. The unrestrained rat worked diligently to free its cagemate, a behavior observed consistently in trials over 12 days. The only reward associated with the action was that associated with freeing the other rat.

The team also tested the two rats in separate cages.

In this situation, the unrestrained rat would not be able to play with its cagemate after freeing the other rat; Mason wanted to investigate whether rats were helping each other out of concern, or rather desire for social interaction. In spite of this condition, the results did not

Reproduced from [5]

change, suggesting that the rats did not rescue their cagemates for self-benefit, but rather out of empathic concern. The experiment provided strong evidence that the rats were not freeing their cagemates only for social company.

Mason and her team also tested rats in a “cagemate versus chocolate paradigm” [3]. In the experimental group, rats were placed in a cage with two restrainers, one containing a cagemate and the other filled with chocolate, a treat for the rats. In contrast, the control placed rats in a cage with an empty restrainer and a chocolate-filled restrainer. Even when the rats were faced with an alternative prospect of chocolate chips, the unrestrained rats chose to free their cagemates and References 1. Mason, Peggy. Interview. 7 March 2013.

2. Mason lab website. < http://masonlab.uchicago.edu>.

3. Mason, Peggy et al. Science. “Empathy and Pro-Social Behavior in Rats.” 334

12

also access the chocolate, while the rats in the control group went straight for the chocolate chips and opened the empty restrainer less frequently. This experiment demonstrated that the rats valued helping their cagemates on par with obtaining chocolate chips.

Yet, how far is too far? While a healthy, happy rat may help a fellow rat in need, a starving, grumpy rat almost certainly would not. Further research could investigate the parameters of helping, exploring the necessary conditions to motivate a rat to help another in need. Describing the differences between humans and other animals, Mason notes that humans have become more Machiavellian than animals in distinguishing who to help. “Essentially, we’ve decided ‘I’m going to help that individual, but not that individual’; and that’s not a bad thing,” Mason remarks. “It’s impossible to help every single person we see in distress; it comes at a cost, not just a financial cost but a cost to the self. Putting a boundary on whose distress you pay attention to and whose you don’t, is a very tough thing” [1].

Learning more about such parameters could lead to a better understanding of the psychological phenomenon of empathy, including the biological roots of its parameters.

Mason’s study was the first to demonstrate that rodents act to relieve one another’s distress. While similar behavior has been observed in chimpanzees and monkeys, rats can be directly utilized in laboratory settings for further study on the neurological basis of empathic behavior [4]. Such research could include clinical trials in the future. “Continuing this work, we would like to extend the research to important clinical applications, including creating a model for autism,” says Mason. “Very little has been done to try and verify if current models of autism actually impair social behavior [in spite of the serious effect of autism on social interactions].” If reduced sociality is confirmed using an animal model, it would open many doors to exploring therapeutic options.

Mason and her team continue to research empathic helping behavior in rats, and plan on publishing another paper later this year. Topics include how close the rats must be in order to incite helping, as well as the aforementioned parameters to helping. Understanding how and where the biological urge to help operates in rats will let us better understand our own behavior across the spectrum.

(2011): 1427-1430. Print.

4. Ferber, Dan. “Rats Feel Each Other’s Pain.” ScienceNow. 8 Dec 2011. Web. 5. http://en.m.wikipedia.org/wiki/File:Wistar_rat.jpg

Scientia

Scientia Abtracts Elucidating the Mechanism for CED-4 and CED-3 Binding via Biochemical Analyses Raymond Dong In the Caenorhabditis elegans roundworm, four specific genes – CED-4 (CED, for Cell Death Abnormal), CED-3, CED-9, and EGL-1 (EGL, for Egg-Laying Defective) – control the pathway of apoptosis. These four genes have direct homologues in humans, and understanding these genes offers insights into a similar signaling cascade event in humans. The CED-4 apoptosome is involved in the activation of the CED-3 zymogen, which leads to programmed cell death. The focus of my research is to better understand how the CED-4 apoptosome activates CED-3. Previously, it was reported that CED-4 forms an octameric oligomer to bind to CED-3, although only a hypothetical model was presented for the specific interaction sites. The complete structure of CED-3/CED-4 complex was unknown and the mechanism describing the interaction between the two proteins was unclear. To help elucidate the structure, I report the purification of truncated forms of CED-3, CED-4, CED-4/CED-9 complex, and EGL-1.Furthermore, using gel filtration and ITC (Iso-Thermal Titration Calorimetry), I confirm the previous model’s hypothesis that there is no cross-interaction between CED-3 CARD and CED-4 HD1/WHD regions via biochemical binding assays. My work offers insight to the specific mechanism for the activation of CED-3 and confirms the previously hypothesized model by Qi et al. For future work, a high-resolution crystal structure of the CED-3/CED-4 complex must be obtained for confirmation of my research.

13

Autumn 2013

Paternal Roles in the Educational Outcomes of Sons and Daughters Victor Ma

Previous studies have indicated a growing gender inequality in higher education, in that females are outperforming males in college enrollment, GPA, and attainment of bachelor’s degrees. Studies have examined the role of parents in educational attainment; however, the absence of the father in the household is much more common than the absence of the mother. No research has previously examined the contribution of parental roles to the gender gap and how the absence of a father affects a child’s educational success. Using data from the National Longitudinal Survey of Freshmen, this study examines the paternal role in educational outcomes through bivariate relationships and regressions, demonstrating the significant effect of a father figure on male graduation rates. Differentials between paternal impact on the graduation rates of sons versus daughters may contribute to the gender gap. Methods through which the variable “father’s presence in the household” works include the cultivation of human and social capital and the development of behavioral skills and aspirations. Possible policy implications, including responsible fatherhood programs, are also discussed.

Diet and Health Capital: An American Case Study Matthew Klein

Last year, Kenneth Arrow and a group of economists (Arrow et. al. 2012) defined sustainability as a non-negative change in per capita welfare levels over time. They introduced several innovations in the measurement of welfare, including a way to determine the value of a nation’s stock of health capital. However, an in depth analysis of health capital was outside of the scope of that paper. I aim to contribute to the methods by which a nation’s capacity to be sustainable is measured by examining health capital. In particular, I seek to develop a series of equations that link the diet of a nation’s average citizen to that nation’s per capita welfare over time. I spend the majority of this paper developing this model before applying it to the American diet from 2005 to 2009. Over this four year period, the average American diet became healthier, but remained unhealthy overall such that the impact of diet on America’s capacity to be sustainable was roughly $180 billion. The findings of this research are that even a small change in diet can have a large impact in whether or not a nation is sustainable over time. The meaning of this result frames a discussion of improvements to the model and to the institutions that surround the American diet. However, a larger question remains unanswered: how important a consideration is diet as compared to the other component pieces of health capital such as access to health care, and exposure to violence and disease? I conclude by hypothesizing that diet may play a smaller role than other factors in health capital but that its importance is undiminished since diet is omnipresent and other factors are not.

References Arrow, K. J., Dasgupta, P., Goulder, L. H., Mumford, K. J., & Oleson, K. (2012) Sustainability and the Measurement of Wealth. Environment and Development Economics, 17, 317-353.

14

Scientia

Research Articles Sacrificial Polymers and Their Use in Patternable Air-gap Fabrication David Goldfeld Patternable air-gaps are added to electrical and mechanical structures in semiconductor and MEMS devices as a means to decrease the dielectric constant, add mechanical compliance, and facilitate microfluidics. In this study, sacrificial polymers and standard photolithographic techniques were used as a way to create air gaps. A photodefinable sacrificial polymer was created by adding photoacid generator (PAG) to the sacrificial polymer mixture. Air gaps were made by patterning the sacrificial polymer into the desired structure, covering it with an overcoat polymer, and decomposing the remaining sacrificial material. There are three problems associated with this fabrication technique: (i) the patterning resolution was coarse, (ii) residue was produced during the decomposition of the polymer, and (iii) wide structures tended to collapse. These problems were investigated through process optimization, quartz crystal microbalance (QCM), and overcoat modification, respectively. In addition, a new sacrificial polymer, PDM-1088, was investigated. PDM-1088 increased the resolution of the pattern, QCM measurements indicated the amount of remaining residue, and hardening the overcoat allowed the fabrication of structures several hundred micrometers wide. Micrometer scale air gaps were successfully fabricated through common photolithographic techniques, permitting the integration of these structures into semiconductor processing.

Introduction Air-gaps have yet to be created using standard electronics processing techniques. They are desirable for use in semiconductor and MEMS (microelectrochemical systems) devices as a means to decrease dielectric constant, add mechanical compliance, and facilitate microfluidics. In this study, photopatternable air-gaps were successfully fabricated by improving each of the three major problems associated with sacrificial polymers: (i) difficulty patterning the sacrificial polymer, (ii) residue after decomposition, and (iii) wide structure collapse.

Experimental Procedure Sacrificial polymers were used to create patternable air gaps. First, a new polymer, PDM-1088, was mixed with a photoacid generator (PAG) so that it could be patterned.

The PAG releases a proton upon heating or by exposure

to 248 nm UV light. The acid significantly decreases the decomposition temperature of the sacrificial polymer. As shown in Figure 1, PDM-1088 mixed with PAG has a lower decomposition temperature (170°C) than the neat polymer (260°C) and the exposed polymer/PAG mix has an even lower decomposition temperature (85°C). Through standard photolithography, as explained in Figure 2, PDM-1088 was spin coated onto a Cu-sputtered Si wafer and a pattern was exposed. Upon heating to 120°C, the exposed portions decomposed. An overcoat material was then spin-coated onto the sample. Two different overcoat polymers were used in this study; polyimide and Avatrel 8000 (polynorbornene). The overcoat was then exposed to 365 nm UV light to activate the polymer cross-linkers and was cured at 150°C for 1 hour. Finally, the sample was exposed a second time at 248 nm, activating all remaining PAG, followed by a 6 hour cure at 150°C to decompose

15

Autumn 2013

Figure 1: Thermogravimetric Analysis graph of PDM-1088 (a) without additives (b) loaded with PAG (c) loaded with PAG and exposed to 248 nm UV light

the remaining PDM-1088.

Results There are three major issues associated with this processing technique: (i) patterns are coarse due to polymer reflow and proton diffusion, (ii) the polymer leaves behind a notable residue after decomposition, and (iii) wide structures tend to collapse in the center due to surface tension. These problems were solved systematically by modifying the processing approach in three ways.

PDM-1088 has a glass transition temperature of 90°C, which is higher than previous sacrificial polymers. Decreasing the processing temperature increased control of the polymer reflow resulting in the fabrication of clean patterns.

Quartz crystal microbalance (QCM) was used to measure the mass left behind after polymer decomposition. This quantification of the residue allows modification of the processing techniques, polymer choice, and PAG loading to achieve

16

minimal residue. Polymer samples were spin-coated onto quartz crystals. The crystal was allowed to equilibrate to its resonant frequency in the QCM, and then the residue was washed off without removing the sample from the instrument. Then, the solvents were evaporated at room

Figure 2: Schematic of air gap fabrication method

Scientia the resulting polymer was able to withstand surface tension pull, even at a thickness of 7 µm. Adding TMPTGE worked significantly better than several other attempted methods, including the addition of surfactant (Triton X-100) and the addition of a second, glass-like polymer (Epoxycyclohexyl Polyhedral Oligomeric Silsesquioxane, or POSS).

Figure 3: QCM results of mass residue vs. polymer thickness (PDM-1088 with PAG)

temperature and the crystal equilibrated again to its resonant frequency. The change in frequency allowed us to use the Sauerbrey equation to calculate the mass of the residue left on the crystal. Figure 3 shows the linear relationship between polymer thickness and residue, meaning the residue is dependent on the composition of the formulation.

Wide air-gap structures were successfully fabricated by modifying the overcoat material. Avatrel 8000 was not strong enough to hold up air-gaps wider than 100 µm as the surface tension pulled the overcoat onto the substrate.

To strengthen the Avatrel, extra trimethylolpropane triglycidyl ether (TMPTGE) was added to the polymer.

TMPTGE is a trifunctional polymer crosslinker used with Avatrel 8000, and the addition significantly increased the hardness as seen in Table 1. By increasing the hardness,

Conclusion By modifying processing methods and incorporating the use of PDM-1088, we successfully fabricated air gaps several hundred micrometers wide and only 5 µm tall. For the first time, such structures were achieved with an overcoat thickness less than 10 µm.

Future Work The fabrication method we developed can now be used in direct applications including electronic interconnects, MEMS, and microfluidics. More work needs to be completed to increase resolution, reduce residue, and create wide structures for certain applications.

Acknowledgments Thank you to the Kohl Group, support staff at Georgia Institute of Technology, and Monica Hochstein, an RET that contributed to this work. This research was part of the National Nanotechnology Infrastructure Network REU and is supported by the National Science Foundation. Thank you to Promerus®, LLC for their polymer contributions.

Table 1: Hardness and elastic modulus values of modified Avatrel 8000 overcoats

17

Autumn 2013

Extension of a Phage-V. Cholera Interaction Model to Include Vaccination Strategy and Seasonality in Growth Rate Kipp Johnson Introduction Cholera is an intestinal disease caused by the pathogenic, gram-negative bacterium Vibrio cholera [1]. Upon human ingestion of food or water contaminated with the V. cholera bacterium, the cell produces an enterotoxin that causes severe, painless diarrhea which the bacteria uses as a medium to reproduce and also as a vector to propagate [2]. Worldwide, there are approximately 3-5 million cases of cholera infection and 100-300 thousand cholera related deaths each year. The mortality rate for acute cholera infection, once as high as 20-40%, currently stands at around 1-5% for those adequately treated with modern rehydration therapy [2]. Because V. cholera is primarily maintained in contaminated water reservoirs, the vast majority of human infection and death from cholera is in the developing world, where such reservoirs are more frequently found [3].

Recent studies of seasonal cholera epidemics in Bangladesh, an area devastated by endemic cholera, have found the presence of a bacteriophage that preys upon the V. cholera bacterium, Phage CTXΦ [4]. Several recent papers have focused upon the presence of phages and the dynamics of bacteria-phage interaction in the wild [1,4]. Here, we will first recreate the dynamic system of one of these recent papers, and then expand upon the model to consider other factors possibly important for control of cholera infection in the developing world.

A mathematical model incorporating the presence of phage interaction with bacterium growth is intriguing due to the role of bacteriophages in mediating cholera infection and because cholera epidemic occurrence in human populations is not well understood. Both phage and cholera concentrations are known to oscillate, but due to experimental difficulties, characterizing the relationship between the two is difficult [4]. Increases in phage concentration have been associated with decreases in cholera, but the etiological agent has not been conclusively proven. Potentially, introducing phages into often-contaminated V. cholera reservoirs

18

could serve as a type of ecological bioengineering to control cholera presence, if the phages infect and destroy V. cholera and ultimately reduce human infection and death [1]. On the other hand, it may be that phage count is only a byproduct of the human-V. cholera cycle, and bears little important to the dynamics of the situation. Finally, as we extend the model originally presented by adding bacterial growth rate seasonality and vaccination, we may be able to demonstrate a scenario whereby vaccinating humans interacts with reservoir phage concentrations, presenting important public health considerations.

Original Model Structure We begin by presenting the model for phage-cholera interaction developed by Das and Mukherjee [1], where S is susceptible population, I is infected population, V is V. cholera concentration, and P is bacteriophage concentration. The differential equations for our model are written as:

where a is a constant immigration rate, α is the rate of exposure to contaminated water, r is the recovery rate of infected individuals (moving from I to S), ε is the disease-induced mortality rate, k is the median infectious concentration of V. cholera in water, δ is the shedding rate of V. cholera from infected individuals into the water

Scientia

Figure 2: Seasonality in Cholera (red = forcing, black = no forcing)

19

Autumn 2013

Figure 3: Seasonality in Cholera (red = forcing, black = no forcing)

20

Scientia supply, γ is the rate at which phage kill infected V. cholera cells, g is the growth rate of V. cholera, and β is the number of phages produced by one infected V. cholera cell. Finally, μ1, μ2, and μ3 are the respective death rates of humans, V. cholera, and phage.

The system has three equilibria: the human-free equilibrium, the phage-free equilibrium, and the phagecholera equilibrium, where phage and cholera both persist in the system. We will look at the phage-cholera equilibrium, as we are interested in behavior in our system that approximates reality, where cholera, bacteria, and phage are endemic in the environment.

Extended Model Structure Seasonality Cholera dynamics vary strongly due to effects from seasonality [6]. Interestingly, these seasonality effects are significantly different throughout regions of the world; specifically, latitude and climate seem to be strongly predictive both of the incidence of cholera as well as of the magnitude and timing in the pattern of its seasonality [6,7]. For example, areas of South America and Africa have intermittent cholera outbreaks, but the disease is endemic in South Asian countries such as India and Bangladesh [6]. The effects of latitude and climate are likely due to the changes in precipitation and differences in extreme weather events (such as monsoons and flooding in Bangladesh, where the phage CTXΦ was originally observed) [4]. The changes in water level brought on by these differences probably serve several functions, as they create new reservoirs of V. cholera and strain the sewer systems in less-developed areas, providing a vector for contamination of drinking water by the bacterium [2]. Ongoing climate change may worsen these effects [8].

Although complex models for cholera seasonality have been developed that include factors such as upper-troposphere humidity, cloud cover, observed solar radiation, climate change, the El-Nino Southern Oscillation, and various human-caused factors such as anthropometric global warming, we will ignore the causes of cholera seasonality in our attempt to model the dynamics [6]. Instead, we first simply recognize that cholera seasonality oscillates with a period approximately equal to one year.

We shall model this change in seasonality by changing g, or the rate of growth of the cholera bacterium in water reservoirs. We do not vary the transmission α, because cholera is not spread person-to-person in any significant

fashion. Rather, humans take in V. cholera from their environment. Thus, with these considerations, varying of g is justified from our assumption that the effects of seasonality change the environment of the water reservoirs, which affects bacterium growth. Unfortunately, deciding what magnitude to force seasonality from the baseline in our function is much more difficult. The literature seems to suggest that cholera concentration can either increase or decrease drastically or not much at all; in order to account for these effects, and without the existence of robust data sets to draw parameter estimates from, we settle for exploring several different plausible numerical values for the g parameter. We thus express cholera growth as a function of time as:

where g0 is the baseline growth rate, g1 is the amplitude of seasonality, and the fraction 2/365 is chosen to give the function a period of one year. This equation requires several assumptions: first, that the growth rate of V. cholera varies sinusoidally depending upon month, which is likely an imperfect approximation. Second, because we do not have an accurate idea for the amplitude of seasonality, we numerically simulate the effects of several g1 values.

Vaccination Vaccines for cholera have existed since the late 1800s [5].

The very first cholera vaccination technique was simply the intravenous injection of the cholera bacterium, which is effective in producing an immune response but can also be dangerous [5]. In the 1980s, effective oral vaccines based upon inactivated whole-cell V. cholera combined with characteristic V. cholera toxin protein were introduced to the market [9]. The vaccines never became very popular, because their effectiveness is imperfect: Generally, they confer approximately 50-80% protection against contracting cholera in the two years after vaccination [9]. Our original model as presented by Das and Mukherjee did not include any class for vaccinated individuals; we propose to extend their model by including such a class. An intriguing question that we hope to answer is whether by vaccinating many individuals, we could perhaps decrease V. cholera presence enough that reservoir phage CTXΦ concentration would also be altered. If phage concentration is affected negatively by vaccination, it may ultimately prove unwise (from a public health perspective) to vaccinate many individuals in the

21

Autumn 2013

Figure 4: Seasonality in Cholera (red = forcing, black = no forcing)

22

Scientia first place.

The modified flow chart for our model is shown below.

First, we assume that 50 percent of those vaccinated will lose their immunity over the course of one year. To make the model more tractable, we assume that rate of vaccine loss is constant (i.e. that an individual is equally likely to lose vaccinated status one week after being introduced into the vaccinated class and at 50 weeks after vaccination. This scenario is likely untrue, but makes the model more easily computable. We also assume that the natural vaccinated class death rate is identical to the susceptible and infected natural death rate (this result may also not be true as wealthier people will have different health outcomes due to the capacity of affording vaccines and other reasons). Acknowledging these assumptions, we do not find that the model is overly compromised. Our vaccination loss rate parallels the CDCâ&#x20AC;&#x2122;s estimates, and the constant vaccine rate loss becomes insignificant as we run our systems to equilibrium. Additionally, the cholera vaccination is relatively inexpensive when carried out by organizations such as the World Health Organization, meaning that vaccines are often administered for free.

In order to examine the effectiveness of the vaccine, it is necessary to look at several possible vaccination rates. We will do this by comparing the vaccination rate to the rate with which individuals progress form the susceptible to the infected class. More specifically, we will use multiples of 0, 2/5, 4/5, 6/5, 8/5, and 2 times this rate in order to examine what protection vaccinations confer upon the population.

Our dS/dt equation thus becomes:

And we add a new equation for our vaccinated class (ν), which is given by:

Results We begin by setting the growth amplitude g1 = 0.9, 0.75, 0.01, and 0.001 and numerically integrating the models in R until an approximate equilibrium is exhibited for all results. We first see in Fig. 5 of the graphical appendix that forcing with amplitude g1 = 0.9, we arrive at a nearly chaotic fluctuation about the unforced equilibrium for susceptible humans, infected humans, V. cholerae count, and phage count. Surprisingly, the massive fluctuations in the susceptible population, vibrio, and phage are almost entirely fluctuations from the equilibrium to a higher point above the equilibrium.

On the contrary, for the infected population, the fluctuations are relatively symmetrical above and below the equilibrium axis. The order of magnitude of these fluctuations also varies enormously: human-susceptible and vaccinated populations vary by about half an order of magnitude from high to low population counts. V. cholerae and phage counts, however, display even wilder variations. We observe that vibrio content explodes from near zero to over 107 in relatively short time frames, while the phage concentration exhibits even wilder behavior, fluctuating from near zero to over 109. The high vibrio count appears to be correlated with the high phage count; high susceptible individual count appears to be correlated with low infected individual count, and vice versa.

As shown in Fig. 6, setting g1 = 0.75 gives us very different system dynamics. We now see considerably more consistent oscillating dynamics, where peak susceptible individual content is almost exactly outof-phase with the infected individual population. The rapid increases in V. cholerae count immediately before increases in the infected individual count suggests that environmental cholera growth, rather than cholera donated to reservoirs via human shedding, is the driving force for cholera dynamics with this forcing amplitude.

For the sake of space, we omit detailed descriptions of Figures 7, and 8 (which show our systemâ&#x20AC;&#x2122;s dynamics with g1 = 0.01 and 0.001, respectively), except to note that they

23

Autumn 2013

Figure 5.1: Seasonality in V. cholera (red = forcing, black = no forcing)

show what we will phrase as â&#x20AC;&#x153;oscillating oscillationsâ&#x20AC;?, or high-frequency oscillations as the system proceeds along a longer wavelength sinusoidal trajectory, and Fig. 8, which shows 3-dimensional phase portraits for the four various examined forcing amplitudes.

Figures 10-14 show results of various vaccination rates in our system, shown for g1 values of 0.0 (no seasonal forcing), 0.9, 0.75, and 0.1. Figure 15 demonstrates total survivorship observed in our system. We see that, absent seasonal forcing, a higher vaccination rate leads to more

24

individuals being maintained in the susceptible group, and also consistently fewer individuals in the infected class. However, the difference between the classes is only about 5%, from highest to lowest (176-170 in susceptibles, and 315-330 in the infected class). The real difference is the amount of humans in the vaccinated population: obviously, with a vaccination rate of 0, we get zero individuals in the vaccinated group. However, with increasing vaccination rate, we get a sustained, proportionately larger vaccinated class. This disparity

Scientia

Figure 5.2: 3D phase portraits of V. cholera count, Phage count, and Susceptible individual count. Growth rate amplitudes (clockwise from quadrant IV) are 0.9, 0.87, 0.01, and 0.001.

is due to much fewer individuals dying from cholera as more are vaccinated. We also see that vaccination rates of greater than 4/5Îą give us large fluctuations in both vibrio and phage count, while vaccination rates below that give us much more consistent numbers.

As we introduce seasonal forcing of a very large degree (figures 11 and 12) we find that for g1 = 0.9, the

magnitude of both susceptible and infected persons, as well as of V. cholerae and phage count is dependent upon the vaccination rate. More specifically, increasing the vaccination rate decreases the magnitude of high-low swings in susceptible humans, but drastically increases the amplitude of infected population swings. Even more interestingly, seasonality here appears to interact with

25

Autumn 2013 the cycles in both V. cholerae and phage populations, as we observe a consistently increased reservoir bacterium count as the population increases as well as an increased phage count. Fig.12 shows the pattern of these oscillations on a smaller time scale, allowing us to see that massive yearly oscillations coexist with seemingly unpredictable sub-year oscillations. The magnitude of these oscillations, especially in and phage count, is enormous, from minima

near zero to maxima on the scale of 107 and 109, respectively. Because of computational constraints, we were unable to run the higher vaccination rates (1.0 and .80) fully to their equilibria, but note that these vaccination rates may be achievable in practice. Regardless, in this system we observe constant, relatively large and undamped oscillations in vaccination rate, which is not the case with growth rate amplitudes of 0.75, 0.01, or 0.

Figure 6: Vaccinated population dynamics absent seasonality in growth rate. Vaccination rates: Purple = 0.0α, yellow = 2/5α, blue = 4/5α, green = 6/5α, red = 8/5α, black = 2α.

26

Scientia

Figure 7: Vaccinated population dynamics with g = 0.9. Vaccination rates: Purple = 0.0α, yellow = 2/5α, blue = 4/5α, green = 6/5α, red = 8/5α, black = 2α.

Figure 8: Vaccinated population dynamics with g = 0.75. Vaccination rates: Purple = 0.0α, yellow = 2/5α, blue = 4/5α, green = 6/5α, red = 8/5α, black = 2α.

27

Autumn 2013

Figure 9: Vaccinated population dynamics with g = 001. Vaccination rates: Purple = 0.0α, yellow = 2/5α, blue = 4/5α, green = 6/5α, red = 8/5α, black = 2α.

With growth amplitude set to 0.75, we observe continuing oscillations in susceptible and infected population counts but notice that wave amplitude decreases as vaccination rate decreases in susceptible populations but increases as vaccination rate increases in infected populations. Furthermore, both V. cholerae count and phage count appear to be uncoupled from human-infected and susceptible counts, as the numerical values for these classes are nearly identical by the time the system has reached equilibrium. We note

28

that the large, yearly oscillations are the dominant feature of this system; although small sub-year scale oscillations are still observed, they are smaller in magnitude than those in our previous simulation. The equilibrium value of vaccinated individuals is linear and directly proportional to the susceptible class vaccination rate.

Setting growth amplitude near zero (0.01) gives us results that are generally very similar to that of vaccination absent forced seasonality, except in one instance. We find that, although the susceptible, infected, and V. cholerae

Scientia classes are all but identical, we now see consistent sinusoidal oscillations in phage density, with values ranging in the approximate interval 6.3-6.6×108. This behavior is altogether different than the non-seasonal vaccinated system behavior, where we observed phage oscillations that were identical in appearance to vibrio oscillations.

The central goal of our simulations was first to discover the interactions among humans, phage, and cholera, given the presence of seasonality in cholera growth rate; and second, to observe the effects of vaccination upon human cholera infection and mortality. As previously noted, cholera infection is endemic in areas like the Indian subcontinent, but occurs in sporadic (yet often major) epidemics in places like West Africa. Simply by seasonally forcing cholera growth rate in the original model, we were able to achieve similar results. With growth amplitude of 0.9, we found that our simple sine-wave model for growth forcing led to seemingly chaotic behavior in all four compartments. Similarly, by forcing the same function at lower magnitudes, we found that we could produce complex yet lower-magnitude oscillatory characteristics in human, infected human, and vibrio counts, and regular sinusoidal oscillation in phage count. As all other parameters were kept constant in our simulations, we posit that variation in seasonal growth magnitude derived from different climactic factors may alone be sufficient to provide for this disparity in cholera dynamics seen across the world. This is an interesting result, because it shows that much of the variation in cholera dynamics between countries may possibly be explained simply by a differing maximum amplitude of growth, even absent differences in the effectiveness or comprehensiveness of public health effects to control cholera infection. However, this is not to say that public health efforts are not of use: we notice that no magnitude of seasonal forcing allowed for the infected population to go to zero in any of our seasonal forcing, suggesting that yearly climate variation alone is not capable of eliminating the disease, and that substantial efforts must be made in other areas in order to eliminate human cholera infections. The caveat, of course, is that we must first prove that seasonal growth amplitude is, in fact, significantly different between regions of the world. As we previously stated, an accurate quantitative determination of V. cholerae presence in water reservoirs has actually proven to be a difficult task, so it may be some time before we can determine the validity and real-world implications of this claim. We are also assuming that the climatic difference in cholera causes different deviations from the same baseline cholera growth rate, and does not

actually shift the baseline growth rate itself, which may be an incorrect assumption.

Starting from the phage-cholera coexistence equilibrium found by the original authors, we find that phage and V. cholerae populations interact in a complex fashion. Although we were varying vibrio growth rate sinusoidally, we find that only phage count shows characteristically sinusoidal oscillations at g1 < 0.9. This suggests that the bacterial shedding rate from infected individuals through contamination of water supplies from human sewage plays a non-trivial role in cholera dynamics. It also appears that the susceptible, infected, and vibrio compartment populations are relatively closely synchronized, while phage count is not. This is an interesting result, because it seems to demonstrate that phage presence does not directly interact with or reduce the infected population count. From a population health perspective, this means that adding additional phage CTXΦ to reservoirs known to possess cholera may not be an efficient strategy to reduce the human disease burden.

However, our model did not account for the fact that both V. cholerae and CTXΦ are not uniformly distributed in the water reservoirs that harbor them, as they are undoubtedly more concentrated in areas of high human sewage pollution. Including targeted phage application in areas of cholera pollution is far beyond the scope of this model, but may prove to be an effective strategy if the cholera concentration is high enough that the effect of phage predation can overcome phage degradation rate. Additionally, breeding more virulent or deadly phages that may not otherwise be selected for in nature could also provide an avenue for improved phage control of cholera.

We are uncertain how to interpret the patterns in all four compartments of the model of what we phrased earlier as “oscillating oscillations.” As these oscillations are variable in magnitude but much higher in frequency than the yearly seasonal oscillation, it is difficult to understand their consequences from a public health standpoint except to note that they may potentially make data collection far more difficult. For example, these highfrequency oscillations appear to occur nearly weekly at g = 0.01, and are almost of the same magnitude as the larger yearly oscillation. A foreseeable situation might occur where some field worker attempting to quantify V. cholerae content of a reservoir on a weekly basis would end up with a vastly different regression model than someone who experimentally characterized bacteria content on only the first week of every month, or where these high-frequency oscillations could mask longer-scale

29

Autumn 2013 oscillations.

Introducing a vaccinated class added very interesting dynamics to our system. Changes in susceptible population were relatively unimportant, because much of these susceptible individuals are simply redirected into our vaccinated class. Ignoring this class for now, we see very conflicting results depending upon the magnitude of g1. When this amplitude was set at 0.9 or 0.75, we see that magnitude of the swings in infected populations actually increased, compared to the magnitude absent vaccination. This is a somewhat paradoxical result, because we would expect that reducing the amount of susceptible individuals through our vaccination regimen would give us more consistent infected class dynamics.

Interestingly, we observe diverging mean infected class populations according to vaccination rates at growth

amplitudes 0.9 and 0.75. In the former case, there was not a significant difference, but in the case of the latter, increasing vaccination rate proportionally decreased mean infected population.

However, when considering vaccination, we think it more prudent to consider total survivorship. We calculated total survivorship as the amount of individuals in the susceptible, infected, and vaccinated class at the end of our simulation, and present our results in Fig. 15 of the graphical appendix. As you can see, increasing vaccination proportionally increases total survivorship at the end of the simulation, meaning that many fewer individuals succumb to death from cholera (with all human class natural-causes death rates equal to each other). This is obviously a good outcome, and can inform our public health decision. Although the magnitude of cholera

Figure 10: Total survivorship, by vaccination rate and growth-rate amplitude forcing. The total all-class survivorship is higher for g = 0.9 likely only because we did not have the computing resources to run this system to equilibrium.

30

Scientia infection will increase as vaccination rate increases, requiring more resources to control, the additional resources are worthwhile because total equilibrium survivorship is drastically increased. Furthermore, these increases are essentially independent of the amplitude of cholera growth rate.

It is interesting to note that even with 100% vaccination, cholera infection remains in an either epidemic or endemic fashion in the population, and phage and bacterium counts are not strongly altered.

This seems to imply that because the bacteria can reproduce in water reservoirs outside of the human body, it would be impossible to cause the extinction of V. cholerae simply by vaccinating any amount of individuals. This is a sobering result, because there is thus no amount of vaccination that will completely eliminate human cholera infection. Given the absence of other interventions, it may be impossible to eliminate human cholera burden.

To conclude, we find that likely the first methods for cholera relief should be the same that societies have practiced for years. The best way in our model to reduce cholera burden on human populations would be to first reduce the death rate ε for those who contract an infectious quantity of the bacterium. We already have an effective technique to do just that: Cholera oral rehydration therapy, which consists of providing large volumes of a saline solution, can be given by non-medical personnel, and is both very effective and inexpensive [10]. Second, working toward reducing the transmission rate α to near zero, as has been done in most developed

nations by sanitizing drinking water before consumption, would preclude the entrance of susceptible individuals into the infected class. Finally, vaccination does appear to be an effective intervention in terms of reducing the total human mortality from cholera and may be useful in circumstances of acute cholera outbreak, but cannot by itself cause environmental extinction of the Vibrio cholerae bacterium.

References 1. Prasenjit Das and Debasis Mukherjee, “Qualitative Analysis of a Cholera Bacteriophage Model,” ISRN Biomathematics, vol. 2012, Article ID 621939, 13 pages, 2012. doi:10.5402/2012/621939 2. Diarrhoeal Diseases. The World Health Organization. http://www.who.int/vaccine\_research/diseases/diarrhoeal/en/index3.html. Updated Feb. 2009.

3. Sack DA, Sack RB, Nair GB, Siddique AK (January 2004). “Cholera”. Lancet 363 (9404): 223–33. doi:10.1016/S0140-6736(03)15328-7 4. Mark A. Jensen, Shah M. Faruque, John J. Mekalanos, and Bruce R. Levin.

Modeling the role of bacteriophage in the control of cholera outbreaks. PNAS 2006 103 (12) 4652-4657; published ahead of print March 14, 2006, doi:10.1073/ pnas.0600166103 5. Sinclair D, Abba K, Zaman K, Qadri F, Graves PM (2011). “Oral vaccines for preventing cholera”. Cochrane Database Syst Rev (3): CD008603. doi:10.1002/14651858.

CD008603.pub2. PMID 21412922.

6. Emch M, Feldacker C, Islam MS, Ali M. Seasonality of cholera from 1974 to 2005: a review of global patterns. Int J Health Geogr. 2008 Jun 20;7:31.

7. Review of reported cholera outbreaks worldwide, 1995-2005. Am J Trop Med Hyg. 2006 Nov; 75(5):973-7.

8. Rodó et al. ENSO and cholera: A nonstationary link related to climate change?

Published online before print September 12, 2002, doi: 10.1073/pnas.182203999 PNAS October 1, 2002 vol. 99 no. 20 12901-12906.

9. Cholera vaccines. A brief summary of the March 2010 position paper. The World Heatlh Organization. http://www.who.int/immunization/

Cholera_PP_Accomp_letter__Mar_10_2010.pd 10. Alam NH, Hamadani JD, Dewan N, Fuchs GJ. Efficacy and safety of a modified oral rehydration solution (ReSoMaL) in the treatment of severely malnourished children with watery diarrhea. J Pediatr. 2003 Nov;143(5):614-9.

Acknowledgements I would like to think Dr. Greg Dwyer for helping me to choose a subject and construct the code for the model and also for providing feedback and assistance as I was writing the paper.

31

Acknowledgements

:

The Triple Helix at the University of Chicago would like to thank the following individuals for their generous and continued support: Dr. Matthew Tirrell

Founding Pritzker Director of the Institute for Molecular Engineering

Dr. Jose Quintans

William Rainey Harper Professor and Master of the Biological Sciences Collegiate Division,

Advisor to the Triple Helix

Eleanor Daugherty

Assistant Vice President for Student Life and Associate Dean of the College

Arthur Lundberg

Assistant Director of the Student Activities Center

Ravi Randhava

2012-2013 Student Activities Advisor

Brandon Kurzweg

Student Activities Advisor

We also thank the following departments and groups: The Institute for Molecular Engineering The Biological Sciences Department The Physical Sciences Department The Social Sciences Department University of Chicago Annual Allocations Student Government Finance Committee (SGFC)

Finally, we would like to acknowledge all our Faculty Review Board members for their time and effort.

Research Submission Undergraduates who have completed substantial work on a topic are highly encouraged to submit their manuscripts. We welcome both full-length research articles and abstracts. Please email submissions to uchicago.print@thetriplehelix.org. Please include a short description of the motivation behind the work, relevance of the results, and where and when you completed your research.

If you would like to learn more about Scientia and The Triple Helix, visit http://thetriplehelix.uchicago.edu or contact us at uchicago@thetriplehelix.org.

Meet the Staff Print Division

Production

Scientia

Co-Directors Charles PeĂąa Carrie Chui

Editor in Chief Patrick Delaney Managing Editors Khatcher Margossian Luizetta Navrazhnykh Michael Begun Writers Scientia Inquiries Jake Russell Austin Yu Gloria Wang Michika Maeda Scientia Abstracts Raymond Dong Victor Ma Research Matthew Klein David Goldfeld Kipp Johnson Associate Editors Jordan Theyel Nora Helfand Jake Russell Annie Marsden Megan Garvey Luke Versten Lydia Pazienza Hannah Friedman Faculty Review Board Michael Glotzer, PhD Paul Kohl, PhD Erdal Uzunlar, PhD Candidate Greg Dwyer, PhD Scientia Inquiry Faculty Matthew Tirrell, PhD Peggy Mason, PhD Mohan Gupta, PhD David Freedman, PhD

SISR Executive

Editor in Chief Austen Smith Mangaging Editors Abhi Gupta Frank Qian Jacob Ryall

Marketing

Director Daniel Frankel

Events Director Evan Jin Coordinators Catherine Castro Cecilia Jiang Stephanie Bi

E-Publishing Director Deborah Olaleye Managing Editors Austin Yu Alice Lee

Executive Co-Presidents Jawad Arshad Melissa Cheng

TTH International at UChicago

Executive Directors of Melissa Cheng Internal Affairs Khatcher Margossian Senior Financial Officer Austin Yu Senior Production Editor Charles PeĂąa