Volume IV Issue I FALL 2018 by JUST Journal

JUST VOL IV // ISSUE I // FALL 2018

LETTER FROM THE EDITOR-IN-CHIEF Dear Reader, I am elated to present to you with Volume IV, Issue I of the Journal of Undergraduate Science and Technology (JUST). JUST is truly a campus wide effort and a celebration of not only the extraordinary research that takes place on this campus, but the work conducted by undergraduates specifically. I would like to extend my sincerest thanks to the undergraduate researchers who submitted their work along with the faculty and staff who supported them. I would also like to express my gratitude for the JUST staff that have chosen to make JUST a part of their undergraduate experience and worked diligently to bring you this publication. Additionally, without the generous support of the Wisconsin Institute for Discovery, the Holtz Center for Science and Technology Studies, the College of Agriculture and Life Sciences, and Associated Students of Madison, the publication of this journal would not have been possible. JUST’s mission has always been to support undergraduate researchers and make science accessible to broader audiences. At UW-Madison, we have been uniquely able to provide the opportunity for undergraduates to publish their work in a peer-reviewed journal and give students a glimpse into the publication process of an academic journal. On the other hand, our staff gain exceptional skills and experience the publication process from the perspective of a producer in a scholarly journal. We believe that these experiences are an invaluable supplement to a traditional undergraduate education, especially for those students who wish to continue research. As for the second part of our mission, I believe that scientific literacy is more important than ever in today’s advancing society. STEM topics have immersed themselves in all aspects of daily life, and all of our lives can only be enriched by a solid understanding of scientific thought. Effective communication of research and science is key to this. We are honored to be a small part in a much larger effort to make research and scientific achievement more accessible to non-expert communities beyond academia. In many ways, the space we occupy on campus mirrors the tenets of the Wisconsin Idea: that the influence of the university should better people’s lives outside of the classroom and across the state. We believe that by helping to train the next generation of researchers and assisting in the dissemination of scientific knowledge, JUST is helping to realize and advance the Wisconsin Idea. In this issue of JUST, you will find a wide range of scientific disciplines represented both by our peer reviewed reports and our shorter editorials as well as the visual pleasure of scientific imagery. Please join us in making it a tradition to recognize the incredible research and thoughtful written pieces presented by UW-Madison undergraduates, and in our larger pursuit to support science literacy. Sincerely,

Helen Heo JUST Editor-in-Chief

JUST VOL IV // ISSUE I // FALL 2018

SPONSORS & PARTNERS EDITOR-IN-CHIEF Helen Heo MANAGING EDITORS Teja Karimikonda Aliyah Keval

We would like to sincerely thank CALS, the Associated Students of Madison, the Wisconsin Institute for Discovery, and the Holtz Center for Science and Technology Studies for financially supporting the production of JUSTâ&#x20AC;&#x2122;s issue. Thank you!

DIRECTOR OF FINANCE Aditya Singh DIRECTOR OF MARKETING Tammy Zhong DIRECTOR OF DESIGN Elizabeth Owens WEBMASTER Cayman McKee MARKETING ASSISTANT Annika Peterson EDITORS OF CONTENT Bailey Spiegelberg, Senior Kort Driessen, Senior Eric Leisten, Senior Haley Dagenais, Associate Stephen Halada, Associate Sydney Ring, Associate Reilly Mooney, Associate Madison Knobloch, Associate & Copy COPY EDITOR Mary Magnuson STAFF WRITERS Annika Peterson, Head Staff Writer Hae Rin Lee Aislen Kelly Ryan Brown Aadhshire Kasat

Special Thank-You to: Mills Botham, Wisconsin Union President Fernanda Martinez, Publications Committee Director Jen Farley, Publications Committee Advisor Sadeq Hashemi, Creative Associate Director Through the publishing of our seven student-run journals and magazines, the Publications Committee of the Wisconsin Union Directorate provides a creative outlet for UW-Madison students interested in creating poetry and prose, reporting on music and fashion, or delving into research in science and public policy. We celebrate creativity on campus by providing hands-on experience in publishing, editing, writing, and artmaking.

JUST VOL IV // ISSUE 1 // FALL 2018

TABLE OF CONTENTS EDITORIALS Algal Blooms: A Growing Concern......................................................................3 Annika Peterson

Bio-renewable Alternative to Polyethylene Terephthalate (PET).......................5 Aislen Kelly

Implications for Regenerative Medicine Under the 21st Century Cures Act......7 Ryan Brown

Moving towards Infinity......................................................................................10 Aadhishre Kasat

PIXELS Alder Levin..........................................................................................................12 Anna Schmidt.....................................................................................................12 Margaret Jameson...............................................................................................13

REPORTS Impacts of Warmer Spring Temperatures on Flowering Times of Individual Native Wisconsin Prarie Plants .........................................................................16 Alder Levin and Olympia Mathiaparanam

Soil Moisture Effects on Portable X-ray Fluorescence Spectroscopy and Visible-Near Infrared Spectroscopy Measurements................................................21 Sean Fischer, Alfred E. Hartemink, Yakun Zhang, and Jenifer L. Yost

Cost-efficiency of Treating Vivax Malaria in Pregnancy in Thai Refugee Camps using Glucose-6-Phosphate-Dehydrogenase testing and selective Primaquine treatment.............................................................................................28 Diane Xue

JUST VOL IV // ISSUE I // FALL 2018

ENVIRONMENTAL SCIENCE

Harmful Algal Bloom in Western Basin of Lake Erie. September 20, 2017. Credit: Aerial Associates Photography Inc. by Aackary Haslick.

ALGAL BLOOMS: A GROWING CONCERN By Annika Peterson

Locals and those traveling to the city can swim and boat on the lakes, or bike and walk along their shores. However, many Madison residents notice that occasionally, the lakes change a disconcerting green color, and the shore has an unpleasant smell. are cyanobacteriaâ&#x20AC;&#x201D;a photosynthetic bacteria that are often called blue-green algae, though they are not technically classified as algae and can be colors other than blue-green [1]. Cyanobacteria have characteristics that allow them to thrive in the lake environment and produce the bloom. Unlike many eukaryotes, Cyanobacteria are able to fix their own nitrogen, giving them a competitive advantage. They have vesicles that can be filled with gas, so they float to be close to their food source, sunlight [2]. Finally, they produce compounds that are toxic to avoid being eaten, though these compounds can also be harmful to humans JUST VOL IV // ISSUE 1 // FALL 2018

EDITORIAL

Madison is known as the City of Four Lakes. The lakes Mendota, Monona, Waubesa, and Kegonsa surround Madison and are connected by the Yahara River. Because of these plentiful waterways, outdoor recreation is part of Madisonâ&#x20AC;&#x2122;s culture. Locals and those traveling to the city can swim and boat on the lakes, or bike and walk along their shores. However, many Madison residents notice that occasionally, the lakes change a disconcerting green color, and the shore has an unpleasant smell. The green color and smell are part of an event that is called an algal bloom, or cyanobacterial bloom. The organisms that cause the blooms

limnologists at UW-Madison and around the country. and animals. The combination of high nutrient levels and Climate change may also a play a role in cyanobaccyanobacteria’s arsenal of evolved skills make the conditions terial blooms. Higher temperatures make it easier for cyain Madison’s lakes right for algal blooms. nobacteria to grow faster than other microorganisms, while Algal blooms can be an inconvenience and an warmer temperatures reduce mixing in the lake, making it eye-sore to those who wish to enjoy lakes, but they also afeasier for cyanobacteria to flourish at the surface. Extreme fect water resources and the lake ecosystem. Cyanobacteria weather, for example, more frequent extremely heavy rains produce toxins (cyanotoxins) which can be toxic to human wash more phosphorus or animal livers, nervous into waterways, and cyasystems, and skin, although nobacteria can thrive. As scientists do not fully unclimate change causes temderstand their impacts. This peratures to be higher and can present problems when rainfall events to be heavier lakes are used as sources in Wisconsin, the incidence for drinking water [3]. Algal of algal blooms will rise. blooms dramatically impact other species. When the alCyanobacterial blooms are gae proliferate, there is an a concerning phenomenon excess of organic matter in that can impact the lakes of the water and as it decomMadison and are progressively becoming a more serious isposes, it ties up all the oxygen in the water, killing fish and sue. Green, smelly water is not pleasant for swimmers and other aquatic animals in a process is called eutrophication boaters, and can also impact drinking water quality. Eutro[1,2]. phication throws the lake ecosystem by killing fish an aquat Many environmental factors create the right conic organisms following a bloom. Although efforts have been ditions for cyanobacterial blooms, including rain fall events made through regulation related to phosphorus use in ferand temperature, but an important driver is an excess of tilizers, algal blooms are still occurring. Research has helped phosphorus in the environment. The element phosphoscientists understand how and why blooms occur, but more rus is a crucial part of aquatic ecosystems which are often research is needed to prevent these events in a changing cli“phosphorus limited”, meaning that the organisms at the mate. The increasing occurrence of algal blooms in lakes bottom of the food web, including cyanobacteria, need it to such as Lake Mendota can allow the public to have greater grow. Excess phosphorus causes their population to grow understanding how the dydramatically [2]. The most namics of the lake ecosyssignificant source of phostem can change, and how phorus in the Yahara River our actions impact local watershed is runoff from waters. agricultural areas that surround the city of Madison. References: Although phosphorus-rich 1. Adam Hinterthuer. What manure from dairy farms Causes the Algae Blooms or fertilizer used to grow in Madison’s Lakes? UW corn and soybeans are imCenter for Limnology. June portant to agriculture, both 11 2018. Available from: can have negative impacts http://blog.limnology.wisc. on waterways. edu/what-causes-the-algae Since cyanobacblooms-in-madisons-lakes/. terial blooms raise many North Shore Yacht Club- Tilapia. November 26, 2010. Cited September 12 2018. concerns, the people living Credit: Matthew Dillon 2. Huisman J, Codd G, Paerl in surrounding areas and usH, Ibelings B, Verspagen J, Visser P. Cyanobacterial ing the lakes have tried to reduce the issue through legislaBlooms. Nature Reviews. 2018; 16: 471-483 tion and education. In the Yahara Watershed, which feeds 3. Miller T, Beversdorf L, Weirich C, Bartlett, S. CyanobacteLake Mendota, legislation from the Wisconsin DNR has rial Toxins of the Laurentian Great Lakes, Their Toxicohelped farmers and landowners manage their use of fertilizlogical Effects, and Numerical Limits in Drinking Water. ers containing phosphorus. Algal blooms were significantly Marine Drugs. 2017; 15 reduced in the 1990’s because of these actions. However, 4. Lathrop R, Carpenter S, Stow C, Soranno P, Panuska J. algal blooms are still an issue in may bodies of water and Phosphorus loading reductions needed to control bluetheir frequency is increasing in some areas [2]. Periods of green algal blooms in Lake Mendota. Canadian Journal of heavy rain are still the biggest predictor in whether a bloom Fisheries and Aquatic Sciences. 1998; 55: 1169-1178● will happen because nutrients are away as rain water runs off fields [4]. Understanding the species of cyanobacteria that cause the bloom is a continuing area of research for

EDITORIAL

Algal blooms can be an inconvenience and an eye-sore to those who wish to enjoy lakes, but they also affect water resources and the lake ecosystem

JUST VOL IV // ISSUE I // FALL 2018

SUSTAINABILITY

Plastic Waste in Body of Water. Image: Pixabay. Creative Commons.

BIO-RENEWABLE ALTERNATIVE TO POLYETHYLENE TEREPHTHALATE (PET) By Aislen Kelly

JUST VOL IV // ISSUE 1 // FALL 2018

EDITORIAL

The pliability and low-cost of synthetic thermoplastics have rendered them nearly irreplaceable in the modern world. The discovery of polymeric materials was an incredible feat for humankind. But in spite of their many favorable properties, one thing has become abundantly and alarmingly clear: they are incapable of degradation.

The rate at which we discard these materials, coupled with the colossal amount annually produced anew, make an unsettling pair. Only 30% of PET plastic is successfully recycled each year in the US, despite being arguably the most recyclable plastic in public circulation [1]. Upon collection, the “garbage” merely requires a system of sorting, washing, grinding and melting processes to retain its initial, seemingly-untouched form. PET is primarily used in single-use beverage containers due to its inert properties and material strength, but can also be found in fabrics, construction materials, or various packaging applications [1]. It’s a very versatile polymer with many favorable qualities, yet its environmental damage can no longer be overlooked as pollution rates rise and feedstocks of raw, non-renewable materials diminish. Jim Dumesic, a Chemical and Biological Engineering professor at UW-Madison, has developed a system in which biomass, rather than crude oil and natural gas, feeds the synthesis reaction of a material practically indistinguishable from PET, called polyethylene furanoate (PEF). PEF is created by the synthesis of ethylene glycol and furandicarboxylic acid (FDCA). Forming high concentrations of this latter component normally requires the addition of a homogenous base to improve its solubility, resulting in the formation of unwanted salts in the effluent of the reactor that require further separation and resources. However, the Dumesic Research Group discovered that the presence of a plant-derived solvent denoted GVL (γ-valerolactone) enabled a high solubility of FDCA, alleviating the need for a homogenous base [2]. Conducting the dehydration reaction of fructose, a natural sugar, in this organic mixture along with some of the desired product itself (FDCA) resulted in a high yield of FDCA’s precursor, HMF. They applied a Pt/C catalyst to the system in place of a corrosive mineral acid, as often used in fructose dehydration reactions, after finding they could recover and recycle it. HMF, when subsequently oxidized in these conditions, formed a high concentration of FDCA without the formation of waste products as aforementioned. This system introduces PEF as a viable replacement to PET as it is similar in cost and richer in properties, without the environmental baggage. This bio-renewable plastic has a lower melting temperature and a higher glass transition temperature than that of PET. This means that PEF can be morphed at less extreme conditions in its unit operations and has the ability to withstand higher temperatures without losing its structural integrity. In other words, manufacturing sites would reduce their energy consumption and product applications could expand, yielding more profit. Further, regulatory ex-

periments have proven that PET is much more susceptible to oxygen and carbon dioxide permeation than PEF, making PEF a much more attractive candidate in applications concerning food packaging. Though both require enzymatic treatment to induce degradation, replacing PET with PEF would greatly reduce associated greenhouse gas emissions as well as our collective dependence on nonrenewable resources. UC-Berkeley is currently conducting research on the manipulation of PETase (a PET-digesting enzyme) to lower its cost and alter its substrate specificity to encompass and degrade a variety of synthetic plastics, which would revolutionize the modern world [3]. This development would reduce current plastic waste and its effects while entirely halting its accumulation. Further, the Huber Research Group under CBE Professor George Huber at UW-Madison is working to optimize plant biomass conversion pathways with the goal to develop cleaner fuels and chemicals via more economical processes [4]. Shifting from a petroleum-dependent world is no longer a distant dream. The world is at an immensely pivotal moment in time. Anthropogenic environmental degradation has rendered many ecosystems uninhabitable over the years, soon to be ours. Unfortunately, polymeric materials add to the destruction. Who says they have to? Deriving the bulk of our packaging pollution from biomass is a sizeable way to reduce environmental harm. Constructing enzymes to reduce current and future waste accumulation will be another. Truly significant sustainable development is on the rise and climbing, yet much is still to be done.

EDITORIAL

PET is primarily used in single-use beverage containers due to its inert properties and material strength, but can also be found in fabrics, construction materials, or various packaging applications.

JUST VOL IV // ISSUE I // FALL 2018

References: 1. Plastics. How Plastics Are Made. Available: https://plastics.americanchemistry.com/How-Plastics-Are-Made/ 2. Made HI. YouTube. YouTube; 2013. Available: https:// www.youtube.com/watch?v=ed7XJeXl3b4 3. Motagamwala AH, Won W, Sener C, Alonso DM, Maravelias CT, Dumesic JA. 2018. Science Advances, 4(1): eaap9722. Available: https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC5775026/ 4. Welcome to the Dumesic Research Group. Index. Available: http://jamesadumesic.che.wisc.edu/ 5. Austin HP, Allen MD, Donohoe BS, Rorrer NA, Kearns FL, Silveira RL, et al. 2018 Characterization and engineering of a plastic-degrading aromatic polyesterase. PNAS, 115 (19) E4350-E4357. Available: http://www. pnas.org/content/early/2018/04/16/1718804115 6. Huber Group Research. Huber Group Research. Available: http://biofuels.che.wisc.edu/ ●

BIOETHICS

A mouse neural stem cell (blue and green) sits in a lab dish, atop a special gel containing a mat of synthetic nanofibers (purple).The cell is growing and sending out spindly appendages, called axons (green), in an attempt to re-establish connections with other nearby nerve cells. 2016. Mark McClendon, Zaida Alvarez Pinto, Samuel I. Stupp, Northwestern University, Evanston, IL

IMPLICATIONS FOR REGENERATIVE MEDICINE UNDER THE 21ST CENTURY CURES ACT The passing of the 21st Century Cures Act was the culmination of historical, political, and social movements. These movements illustrate the desire for the reevaluation of what counts as evidence and what determines the level of safety of a therapy. JUST VOL IV // ISSUE 1 // FALL 2018

EDITORIAL

By Ryan Brown

This is especially important in largely untested, emerging fields of treatment such as regenerative medicine. Thus, it is important to investigate the possible ramifications of this legislation, and how it may affect patients. To receive the designation in the 21st Century Cures Act, a regenerative medicine therapy must first be “intended to treat a serious or life threatening disease or condition” as well as have “preclinical evidence that the drug has the potential to address unmet medical needs” [1]. This allows for several important considerations: Firstly, these therapies receive additional and quicker guidance and collaboration with the FDA. Secondly, regenerative medicine therapies can receive the accelerated approval designation. This designation would grant the regenerative medicine field access the same deregulatory standards seen in the rest of medicine, primarily allowing the ability to evaluate diseases in a quicker fashion. This is achieved by choosing surrogate endpoints and biomarkers, which are simply ways of evaluating points that strongly correlate with a desired effect rather than the effect itself. For example, a change in biochemical response that is typically correlated with survival or life outcomes allows for quicker studies, but also less accurate. The 21st Century Cures Act thus changes the bar of evidence needed to pass clinical trials. These alterations make passing trials easier, which can lead to a number of important ethical questions. The accelerated approval process’ use of different endpoints has been shown as considerably less accurate than that of hard clinical endpoints [2]. The FDA states in their guidance that endpoints are primarily used when “the disease course is long, and an extended period of time would be required to measure the intended clinical benefit of a drug” [3]. Cell therapies definitely meet this criteria; however, finding a valid clinical surrogate for new therapies is difficult to determine. This is because clinical surrogates are only useful when the pathophysiology of the disease and mechanism of intervention are thoroughly understood [4]. Due to the many uncertainties that are within regenerative medicine, it is unclear how the pathway of treatments work. Surrogate endpoints are especially problematic in regenerative medicine because there are often visible effects of regrowth while not effecting overall outcomes on survival. Furthermore, biomarkers and surrogate endpoints can often show indications that cell function has changed, but this rarely signifies an intended change of function at an organ or tissue level. Such short term endpoints also fail to take into account possible negative side effects that only proliferate over long periods of time. These concerns put patients at risk to receive non-understood treatments from such an accelerated

process. This possibility is troubling because of the 36 drugs approved to market between the start of 2008 to the end of 2012 that used surrogate endpoints, only half of the therapies showed any benefit [5]. This means that treatments may get to patients faster, but also result in therapies that provide no benefit for the patient. These endpoints are further coupled with reduced need for phase three trials. This is troubling because it has been shown that these trials are pivotal in determining effectiveness between phase two and three trials. Often times the results of the phase two and three trials can vary widely in safety and effectiveness, often hurting patients [6]. Here, both the type and amount of evidence are reduced considerably. This sets an interesting precedent as typically innovative, untested fields that are developing move much slower through the FDA, rather than faster. Essentially, this means that the patients who take these therapies would have to take on a larger amount of risk. In an attempt to account for this, the FDA states that this pathway is for any regenerative medicine therapy that is considered a serious or life threatening disease, and shows preliminary evidence for its benefits [3]. This indicates that there is consideration by regulators in determining which patients can bear the extra risk, but the bill fails to layout a strong definition of what “serious” means. A “serious” condition is subjective and refers to any disease that is “long lasting and has a serious impact on day to day functioning” [7]. This is not a particularly difficult standard to meet as poor vision, back pain, or any age-related problems would fit this definition. While these problems are definitely serious and prohibitive for the individual with them, they are not likely to cause death, therefore making the potential risks associated with accelerated regenerative medicine therapies questionable. While it is hopeful that cures reach patients faster, they may not warrant intervention of technology that is not fully understood. The risk increases when looking at how new the field is, and the unique risk factors associated with cellular therapies. Any field in its early stages has been known to illustrate higher rates of failure. For example, when surgical techniques were still being discovered they had extremely high chances of failure, but today they stand as one of the focal points of medicine. This same idea is what makes regenerative medicine so promising and at the same time, concerning. The same properties that make stem cells so valuable, through proliferation or differentiation, are what makes them equally deadly [8]. These cells are not metabolized and disposed of by the body; they are instead integrated into the individual, making the effects long lasting [9]. This means that benefits can be kept throughout one’s lifetime without follow-up treatments, but the negative effects can likewise reside and fester.

The 21st Century Cures Act thus changes the bar of evidence needed to pass clinical trials.

EDITORIAL

Treatments may get to patients faster, but also result in therapies that provide no benefit for the patient.

JUST VOL IV // ISSUE I // FALL 2018

Due to this concern, the FDA has historically been very slow and cautious with the introduction of this technology. Only 15 regenerative medicine technologies have been approved by the FDA, 7 of which are related to hematopoietic stem cells (HSC). This preferential approval is likely because HSCs have been used for transplantation since the 1960s, and are the most understood stem cell therapy [10]. Beyond HSC’s, most conditions have success rates that remain frustratingly low while morbidity and mortality rates are unacceptably high [11]. Also a large number have only just began phase I trials, demonstrating a field that is just moving from preclinical research into experimental trials [12]. Most new technologies face these obstacles, making faster and quicker approval times of possible concern. While the FDA has the capacity and resources to make smart choices, poor wording and unclear direction still leave patient safety at risk. This is because regenerative medicine has many specific problems due to both the complexity and recent emergence of the field. The 21st Century Cures Act has created pathways for these therapies to more quickly get through the FDA review process in order to address the needs and desires of society. The high hopes and potential of regenerative medicine has driven physicians, patients, and industry to all dream of the day when science can develop the cures they envision. The type of evidence that is needed to evaluate these goals is up to society. While the basis of evidence is always changing, the end goal will always be the same: ensuring that harm will be minimized. References: 1. 21st Century Cures Act. 2016. Public Law 114–255. 2. Svensson, S., Menkes, D. B. & Lexchin, J. 2013. Surrogate outcomes in clinical trials: a cautionary tale. JAMA

Intern Med, 173, 611-612. 3. US FDA. Regenerative medicine advanced therapy designation 2017. www.fda.gov/BiologicsBloodVaccines/ CellularGeneTherapyProducts/ucm537670.htm. 4. Aronson, J. K. 2005a. Biomarkers and surrogate endpoints. Br J Clin Pharmacol. England. 5. Kim, C. & Prasad, V. 2015. Cancer drugs approved on the basis of a surrogate end point and subsequent overall survival: An analysis of 5 years of US food and drug administration approvals. JAMA Intern Med, 175, 19921994 6. US FDA. 22 Cases Where Phase II and Phase III trials Had Divergent Results 2017. https://www.fda.gov/ downloads/AboutFDA/ReportsManualsForms/Reports/UCM535780.pdf 7. US FDA. FDA’s Origin and Function 2017. https:// www.fda.gov/AboutFDA/WhatWeDo/History/Origin/default.htm 8. Sipp, D. & Turner, L. 2012. Stem cells. U.S. regulation of stem cells as medical products. Science, 338, 1296-1297. 9. Dlouhy, B. J., Awe, O., Rao, R. C., Kirby, P. A. & Hitchon, P. W. 2014. Autograft-derived spinal cord mass following olfactory mucosal cell transplantation in a spinal cord injury patient: Case report. J Neurosurg Spine, 21, 618-622. 10. Okayanos. 2015. History of Cell Therapy. Freeport, Grand Bahama. https://okyanos.com/history-of-cell-therapy/ 11. Daley, G. Q. 2012. The promise and perils of stem cell therapeutics. Cell Stem Cell, 10, 740-749. 12. Li, M. D., Atkins, H., & Bubela, T. (2014). The global landscape of stem cell clinical trials. Regen Med, 9(1), 2739. ●

EDITORIAL JUST VOL IV // ISSUE 1 // FALL 2018

CHEMISTRY

Nobel Prize, Image: Adam Baker, Creative Commons

MOVING TOWARDS INFINITY

EDITORIAL

By Aadhishre Kasat

Arnold’s leading question was to resolve which of these variants work best in organic solvents. The principles of evolution state that in a given environment, only the fittest survive. Arnold applied the same principle to enzymes and called this stage selection. Some say that we use science and research to constantly feed our growing curiosity; that when a discovery is made, we only get a glimpse of the entire truth. The idea of research necessitates accepting that our knowledge is not definite—accepting that we have to learn, un-learn, and re-learn because what we know for sure today may change tomorrow. Our knowledge is imperfect, but it evolves just like any natural process.

JUST VOL IV // ISSUE I // FALL 2018

However, due to Frances H. Arnold’s research in protein engineering, sustained by her ever-growing curiosity, she has refined some of that knowledge. The 2018 Chemistry Nobel Prize was awarded to Arnold for the “Directed Evolution of Enzymes”, as well as George P. Smith and Sir Gregory P. Winter for the “Phage Display of Peptides and Antibodies”. This year, the Laureates incorporated the ideas of genetic evolu-

JUST VOL IV // ISSUE 1 // FALL 2018

EDITORIAL

the catalyst [6]. tion, mutation and selection to develop new proteins. However, procurement of such a catalyst isn’t Genetic selection is the process by which certain traits always easy. But due to Arnold’s mechanism for the evobecome more prevalent in a species than other traits [1]. lution of directed enzymes, we can develop catalyst for Arnold’s protein experimentation began in reactions we did not previously have by adapting exist1993 when she developed the first directed evolution ing catalysts to work in reactions they did not previously of enzymes (proteins that catalyze chemical reactions) work in. Consequently, the efficiency of several industrias well as refined the methods now routinely used to al processes will increase, will consume less energy and develop new catalysts (substances that alter the rate of resources, and ultimately save money. chemical reactions) [2,3]. Catalysts and enzymes work We live in such a sophisticated world, most of only in specific chemical reactions and reaction condiwhich we still aren’t able to comprehend. But we must tions, and on specific substrates (substances to which not give up; we must continue to wonder why. We should catalysts bind onto) [4]. Arnold began her work with constantly ask ourselves, “what do we know, how do we an enzyme called subtilisin, which is known to catalyze know this, and how can we be sure?” Arnold opened a chemical reactions in aqueous solutions. Arnold’s chaldoor in protein engineering that led her to potentially lenge was to manipulate the enzyme in a manner so that unlimited directed enzyme evolution. In the end, it is it works effectively in organic solvents. She created rancuriosity that opens doors for us, leading us to places we dom changes in the sequence of base pairs that transnever even knew existed. Our knowledge may be imperlated to the formation of subtilisin. She then introduced fect, but that is why we conthese mutated genes into bactetinue to do research. Research ria that produced thousands of may just be the bridge between different variants of subtilisin. us and infinity. Arnold’s leading question was to resolve which of References: these variants work best in or1. The Nobel Prize in ganic solvents. The principles Chemistry 2018. NobelPrizeoof evolution state that in a givrg. Nobel Media AB 2018; en environment, only the fittest 2018. Available: http://www. survive. Arnold applied the nobelprize.org/prizes/chemissame principle to enzymes and try/2018/press-release/. called this stage selection. 2. Newman T. En We know that subtilisin zymes: Function, definition, breaks down milk. So, Arnold and examples. Medical News investigated which variant of Today. MediLexicon Intersubtilisin works most effectively national; 2018. Available: to break down milk in an orhttps://www.medicalnewstoganic solvent. She subsequentday.com/articles/319704.php ly introduced a new round of 3. Britannica TEof E. random mutations to subtilisin, Catalyst. Encyclopædia Britanwhich yielded a variant with imnica. Encyclopædia Britannica, proved function in the same orFrances Arnold at CalTech 2008, Image: Wikipedia inc.; 2017. Available: https:// ganic solvent. In the third genCommons www.britannica.com/science/ eration of subtilisin, she found catalyst a variant that worked 256 times better in the organic 4. Substrate. Encyclopædia Britannica. Encyclopædia solvent than the original enzyme. This particular variant Britannica, inc.; Available: https://www.britannica. had a combination of ten varied mutations. This invescom/science/substrate-enzymatic-reactions tigation demonstrates the power of allowing chance and 5. THE ROYAL SWEDISH ACADEMY OF SCIENCdirected selection, instead of solely human rationality, ES. NobelPrizeorg. 2018. Available: https://www. to govern the development of new enzymes [5]. nobelprize.org/uploads/2018/10/advanced-chem Research in this field focuses on optimization istryprize-2018.pdf. Accessed 3Nov2018. and efficiency. It infuses high specificity, limited side 6. The Nobel Prize in Chemistry. NobelPrizeorg. Noreactions, and improves the tolerance of proteins in a bel Media AB 2018; Available: https://www.nobelrange of reaction conditions. Arnold’s work shows that prize.org/prizes/chemistry/ ● enzymes can catalyze new reactions as well as catalyze reactions in conditions very different from their optimal conditions. This discovery has the potential to affect highly marketed chemical reactions, including those involved in the production of sulfuric acid and the production of fertilizers requiring vanadium pentoxide as

PIXELS

where science and art collide

Alder Levin: Pictured is a honey bee on showy goldenrod at the Biocore Prairie. This species is responsible for much of the pollination that maintains both our wild flora and our domesticated crops. Victims of pesticide pollution, climate change, and habitat destruction, it's important to protect them from these human derived threats.

Anna Schmidt: This is a photo of the surface of Lake Mendota during a large blue-green algae (cyanobacteria) bloom in June 2018.The photo was taken on a sampling trip for a project that focuses on long-term trends in microbial community composition of Lake Mendota. A sample of the surface water was collected for analysis of cyanotoxins, toxic substances produced by cyanobacteria.

JUST VOL IV // ISSUE I // FALL 2018

Margaret Jameson: iPSC cardiomyocytes stained for Cav1.2 (cyan), RYR2 (green), Actin (red), and DAPI (blue).

Alder Levin: This grasshopper was found in the short grass at the Biocore Prairie. It is just one of the many insect species that can be found in this ecosystem when taking a closer look.

JUST VOL IV // ISSUE 1 // FALL 2018

REPORT

REPORTS

Impacts of Warmer Spring Temperatures on Flowering Times of Individual Native Wisconsin Prarie Plants .........................................................16

The flowering times of plants are influenced differently by climatic factors and genetic regulatory systems. Consequently, some species may exhibit phenotypic plasticity while others reach physiological thresholds that limit responses to changing climate. This could cause ecological mismatches, influencing ecosystem processes including interactions with pollinators and herbivores, water uptake, nutrition cycling, and carbon sequestration.To study relationships between climate and biological events, Aldo Leopold began documenting flowering dates of plants at the University of Wisconsin–Madison nearly one hundred years ago. Since then, there have been increased efforts to record flowering dates of Wisconsin wildflowers, and new data shows many prairie species flowering earlier in response to warmer temperatures. Similar to Aldo Leopold and Nina Leopold Bradley’s work, we examined the relationship between spring temperature and flowering of individual prairie plants for three years. Tracking individual plants, as opposed to monitoring broad prairie populations, limits confounding variables related to genetic diversity and microclimate, thus our phenological observations are more closely associated with yearly environmental fluctuations. Data were analyzed from 2016 to 2018, when spring temperatures varied from 7.21- 8.93º C. Analysis of this data reveals no significant relationship between spring temperature and flowering time for individual prairie plants during those years. However, comparing two years of our data shows large differences between flowering dates for the same individual of some species and very similar flowering dates for the same individuals of other species. This indicates that the relationship between climate and flowering time is species-dependent. Future data collection will continue to shed light on this relationship.

Soil Moisture Effects on Portable X-ray Fluorescence Spectroscopy and Visible-Near Infrared Spectroscopy Measurements................................21 Portable X-ray fluorescence (pXRF) and visible-near infrared spectroscopy (vis-NIR) are used to measure or predict soil properties such as soil organic carbon, nutrients, soil texture, and environmental pollutants such as heavy metals. These characteristics have great importance in soils under agriculture as well as environmental remediation. When taken in the field, vis-NIR and pXRF measurements are affected by the moisture conditions of the soil.This study seeks to quantify the effects of soil moisture on pXRF and vis-NIR measurements and propose correction equations of moisture effects.Topsoil and subsoil samples from an Entisol (~sand) and a Mollisol (~silty loam) were wetted in the laboratory to six different moisture contents, and subsequently scanned by the pXRF and vis-NIR.The elemental intensity in spectra from the pXRF was significantly reduced by higher moisture contents, yielding lower elemental concentrations. This was linearly related to gravimetric water content. The effect was stronger in the silty loam soils than in the sandy soils. The reflectance of vis-NIR spectra decreased with increasing soil moisture content, and the reduction in reflectance of two spectral ranges (1410-1450 and 1910-1930 nm) was modeled by non-linear logarithmic regression. It is concluded that soil moisture reduces the measurements of pXRF and reflectance spectra of vis-NIR. The elemental concentrations measured by pXRF could be corrected by linear model with known gravimetric water content, whereas more sophisticated models are needed for vis-NIR spectra. The results of this study may improve the accuracy of pXRF and vis-NIR spectroscopy soil analysis used in the field, and could influence the way soil data is collected.

JUST VOL IV // ISSUE I // FALL 2018

Cost-efficiency of Treating Vivax Malaria in Pregnancy in Thai Refugee Camps using Glucose-6-Phosphate-Dehydrogenase testing and selective Primaquine treatment..............................27

REPORT

Pregnant women have an increased risk of contracting malaria, especially in densely populated areas. The inability to prescribe Primaquine, the only known treatment of vivax malaria, to pregnant women is a major complication for malaria eradication. Primaquine is contraindicated for pregnant women due to the unknown risk of Glucose-6-Phosphate Dehydrogenase (G6PD) deficiency of the fetus. G6PD deficient individuals are at risk of severe hemolytic anemia if exposed to Primaquine. Genetic screening for G6PD could be done on pregnant women to determine the G6PD status of a male fetus due to the X-Linked nature of the gene. However, cost remains a hurdle for implementing new genetic screening protocols. In this study, I evaluated the cost-effectiveness of implementing point-of-care testing for G6PD in refugee camps on the Thai-Myanmar border using recent data from the Shoklo Malaria Research Unit on the effects of malaria in pregnancy on newborn mortality. The cost-effectiveness ratios based on infant mortality rates and value of statistical life measurements are in comparison to the current strategy of no screening. My results suggest that the screening strategy is cost-effective relative to no screening. To the best of my knowledge, this is the first cost-effectiveness analysis for evaluating genetic testing of pregnant women for G6PD as a means of malaria eradication in Thai-Myanmar border refugee camps.

JUST VOL IV // ISSUE I // FALL 2018

Impacts of Warmer Spring Temperatures on Flowering Times of Individual Native Wisconsin Prarie Plants Alder Levin and Olympia Mathiaparanam ABSTRACT The flowering times of plants are influenced differently by climatic factors and genetic regulatory systems. Consequently, some species may exhibit phenotypic plasticity while others reach physiological thresholds that limit responses to changing climate. This could cause ecological mismatches, influencing ecosystem processes including interactions with pollinators and herbivores, water uptake, nutrition cycling, and carbon sequestration. To study relationships between climate and biological events, Aldo Leopold began documenting flowering dates of plants at the University of Wisconsin–Madison nearly one hundred years ago. Since then, there have been increased efforts to record flowering dates of Wisconsin wildflowers, and new data shows many prairie species flowering earlier in response to warmer temperatures. Similar Similar to the work of Aldo Leopold and other notable phenologists, we examined the relationship between spring temperature and flowering of individual prairie plants for three years. Tracking individual plants, as opposed to monitoring broad prairie populations, limits confounding variables related to genetic diversity and microclimate, thus our phenological observations are more closely associated with yearly environmental fluctuations. Data were analyzed from 2016 to 2018, when spring temperatures varied from 7.21- 8.93º C. Analysis of this data reveals no significant relationship between spring temperature and flowering time for individual prairie plants during those years. However, comparing two years of our data shows large differences between flowering dates for the same individual of some species and very similar flowering dates for the same individuals of other species. This indicates that the relationship between climate and flowering time is species-dependent. Future data collection will continue to shed light on this relationship.

REPORT

INTRODUCTION

Recording the timing of seasonal phenomena has been an important human endeavor for thousands of years. Detailed records of Japanese cherry blossoms dating back to 900 B.C.E., grape harvesting times in Switzerland during the 1500s, and the spring day that the ice broke on Henry David Thoreau’s Walden Pond in 1857 are all examples of phenological data that offer scientists a glimpse into seasons past [1]. In recent years, ecologists have turned to the historical records of notable phenologists like Robert Marsham, Gilbert White, Henry David Thoreau, and Aldo Leopold in an effort to understand the effects of a changing climate on current biological events [2]. While datasets can include a wide range of events like bird migration and changing leaf colors, one of the most biologically important phenological measures is the first flowering dates of native plants. Flowering time is influenced by many climatic factors including sunlight, rainfall and snowfall, but perhaps its strongest relationship is with temperature [3]. The sensitivity of flowering times to changes in seasonal temperature has made this measurement especially valuable, and recent research demonstrates that plant flowering in the United States has occurred at earlier dates in response to warmer temperatures [4]. These observations raise questions regarding how changes in climate could influence the synchronization of flowering plants with herbivores, pollinators, water uptake, 18

JUST VOL IV // ISSUE I // FALL 2018

carbon sequestration or the removal of carbon dioxide from the atmosphere, and nutrient cycling as well as the potential consequences of ecological mismatches [5; 6]. Aldo Leopold’s data on flowering times of native Wisconsin plants is of particular importance to this study. Leopold, along with his students and members of his family, kept records of seasonal events in Wisconsin, primarily in Madison, Wisconsin and at his shack in Sauk County, Wisconsin spanning from 1935 to 1945 [7]. Data collection was continued in these counties between 1977 and 2012 by his daughter, Nina Leopold Bradley, Dr. Stanley A. Temple of the University of Wisconsin–Madison, and staff from the Aldo Leopold Foundation [3]. These data demonstrate some of the long-term effects of a changing climate on biological events, including flowering time. The historical data indicate a significant relationship between spring temperature and flowering time for spiderwort (Tradescantia ohiensis), prairie phlox (Phlox pilosa), and shooting star (Primula fassettii) ( Figure 1a) [8]. Temperature is important because warmer temperatures increase plant growth rates, which can accelerate plant maturation [9]. However, large increases in temperature can begin to negatively impact plant growth; proteins necessary for plant function have been shown to denature when exposed to warm temperatures, consequently hindering growth and reproductive maturation [10]. Additionally, fluctuating temperatures in winter and early spring may prompt early germination of seeds. When late frost events occur following

these fluctuations in temperature, ice can damage cell walls as well as the overall plant structure [11], which influences the number of plants that will mature to reproductive age. In addition to temperature, other factors also play an important role in determining flowering time. For example, increased rainfall can increase the rate of plant development, which may result in earlier flowering. However, too much rain decreases oxygen concentration in the soil and damages roots, which may reduce health, ability to flower, and reproductive success of the plant [12]. The Leopold and Bradley data suggests a relationship between spring precipitation and flowering time such that earlier flowering dates were found in years with more precipitation for spiderwort and shooting star (Figure 1b). In contrast, later flowering dates were found in years with more

phenological data based on when they noticed a first flower in a particular natural area. In contrast, our study captures data on first flower dates of individual plants, rather than on prairie populations, from many prairie species. By tracking individual plants, we reduced the effects of confounding variables related to genetic diversity, microclimate, and observer bias, allowing us to better examine phenological observations in relation to yearly spring temperature and other environmental changes (e.g. precipitation, snowfall, winter temperature). Following three years of data collection, we found no significant relationship between spring temperature and flowering time for individual prairie plants. However, our data indicate that the relationship between environmental fluctuations and flowering time is species-dependent.

METHODS Species were selected based on their presence at the Biocore Prairie and the ease with which an exact flowering date could be determined in a repeatable way. Individual plants of each species were located and identified in the Biocore Prairie. To account for variation within the site, replicates of the same species were located in different areas of the prairie whenever possible. Individuals were marked with wire flags for easy locating in the short term, as well as numbered metal stakes to serve as permanent markers. Plant coordinates were recorded in accordance with the Biocore Prairie grid system (Figure 2). Plants were then observed in the days leading up to their flowering, at which time the flowering date was recorded and an accompanying photograph was taken. The images serve as data points to illustrate the life stage and flower maturity for each plant on the day

precipitation for prairie phlox. This is one example of how climatic factors can have dissimilar effects on flowering time for different plant species. Previous studies have focused on location-wide phenological observations in which researchers recorded

Figure 2: Map of the Biocore Prairie with grid overlay. Each of the six lupine (Lupinus perennis) are indicated by a red pin.Their coordinates correspond to their location within the grid: L20, 7.0m E, 7.1m N; L20, 13.5m E, 16.6m N; M20, 19.3m E, 4.9m N; Q17, 1.0m E, 0.0m N; Q17, 3.4m E, 0.0m N; and R18, 2.7m E, 16.9m N.

JUST VOL IV // ISSUE 1 // FALL 2018

REPORT

Figure 1: First flowering dates of spiderwort (T. ohiensis, max=162, min=127, mean=148.1), prairie phlox (P. pilosa, max=159, min=118, mean=137.3), and shooting star (P. fassettii, max=147, min=122, mean=134.0). Data were collected in 1935-1945 by Aldo Leopold, his students, and members of his family and in 1977-2012 by Nina Leopold Bradley, Stanley A. Temple, and members of the Aldo Leopold Foundation. A) First flowering date compared to average spring temperature (spiderwort: R2 = 0.4023, F(1, 45) = 30.29, p < 0.001; prairie phlox: R2 = 0.343, F(1, 45) = 23.50, p < 0.001; shooting star: R2 = 0.7029, F(1, 45) = 106.48, p < 0.001). Average spring temperature was calculated using temperatures in March, April, and May. B) First flowering date compared to spring precipitation (spiderwort R2 = 0.0264, F(1, 45) = 1.22, p = 0.275; prairie phlox: R2 = 0.0012, F(1, 45) = 0.054, p = 0.817; shooting star: R2 = 0.0851, F(1, 45) = 4.19, p = 0.047). Spring precipitation includes both rainfall and snowfall for March, April, and May (13). First flowering dates are given as days of the year, where January 1st is day 1.

in which flowering was determined. The photos were saved to ensure data collection consistency in subsequent years of this study. After the first year of data collection, researchers returned to the same individual plants to observe their flow-

REPORT

Figure 3: First flowers of plants whose flowering dates were recorded in 2016, 2017, and 2018. Each column includes the images for an individual plant in each year of the study: purple prairie clover (Dalea purpurea), meadow rue (Thalictrum dasycarpum), prairie phlox (Phlox pilosa), and clasping milkweed (Asclepias amplexicaulis).

Figure 4: Graph of preliminary data. Each line represents an individual plant’s first flowering dates in 2017 and 2018. Lines of the same color are individuals of the same species. In 2017, average spring temperature was 8.21°C, spring rainfall was 27.84 cm, spring snowfall was 21.11 cm, and total spring precipitation was 48.95 cm. In 2018, average spring temperature was 7.21°C, spring rainfall was 32.16 cm, spring snowfall was 43.05 cm, and total spring precipitation was 75.21 cm. Spring temperatures and precipitation were calculated using values from March, April, and May.

ering dates in subsequent years. On the day of first flower, the date was recorded and a photograph of the flower was taken (Figure 3). Flowering dates were analyzed using a regression curve and were compared to average spring temperature and spring precipitation. Climate data was gathered from the Wisconsin State Climatology Office. In future years of this study, researchers will continue to monitor the flowering dates of the same individuals. We hope that they will also expand the inventory of individual plants so that each of the 65 selected species has 3 to 6 replicates; although, this number will depend on the number of plants of that species at the Biocore Prairie and on the life cycle of the species.

prairie clover (Dalea candida), rosinweed (Silphium integrifolium), prairie blazing star (Liatris pycnostachya), and flowering spurge (Euphorbia corollata) (Figure 4b). First flowering dates were also recorded for a small number of individuals in 2016, allowing for a threeyear comparison of 2016, 2017, and 2018. These data can be compared to many climatic factors, including average spring temperature and spring precipitation (Figure 5).

RESULTS

DISCUSSION

Although phenological data must be collected over the course of many years to reveal strong trends between climatic factors and seasonal events, data from the first years of this study illustrate differences in flowering time changes for different species. Figure 4 displays the flowering times of seven prairie species in 2017 and 2018. The flowering dates of some species deviated minimally between the two years: meadow rue (Thalictrum dasycarpum), butterfly milkweed (Asclepias tuberosa), and yellow coneflower (Ratibida pinnata) (Figure 4a). The flowering dates of other species changed by as much as 43 days between the two years: white

Nina Leopold Bradley’s work comparing her observations (1977 to 2012) and her father’s (1935 to 1945) indicated that many southern Wisconsin plant species flowered at earlier dates in response to the warming climate over that 61-year period [8]. Building on this work, we investigated the relationship between climate and flowering times of individual prairie plants at the Biocore Prairie over three years. Our data shows that species reacted differently to the variable seasonal conditions: some plants flowered earlier in warmer years, as expected, while others showed no change. Compared to spring 2017, spring 2018 was colder by 1.0º

JUST VOL IV // ISSUE I // FALL 2018

Figure 5: Graph of preliminary data. Each line represents an individual plant’s first flowering dates in 2016, 2017, and 2018. A) First flowering date compared to average spring temperature. Average spring temperature was calculated using temperatures in March, April, and May (8.93°C in 2016, 8.21°C in 2017, and 7.21°C in 2018). B) First flowering date compared to spring precipitation. Spring precipitation included both rainfall and snowfall for March, April, and May (41.96 cm in 2016, 48.95 cm in 2017, and 75.21 cm in 2018).

JUST VOL IV // ISSUE 1 // FALL 2018

REPORT

C and had more precipitation (4.31 cm of rain, 21.95 cm of snow). We expected some plants to flower later in 2018 compared to their flowering times in 2017. Of our selected plants, this trend was only observed for white prairie clover which flowered 35 days later in 2018 than it did in 2017. Interestingly, some prairie plants (prairie blazing star, flowering spurge, and rosinweed) flowered as much as 43 days earlier in 2018 compared to their flowering times in 2017. In contrast, other species (yellow coneflower, meadow rue, butterfly milkweed) had little deviation in flowering times between the 2017 and 2018 summers. These preliminary findings suggest that native plant species may have differing genetic ability, ecological mechanisms, or morphological structures (e.g. depth of root systems, thickness of leaves, and distance from ground) that influence the degree to which these plants demonstrate phenotypic plasticity with climatic changes. Some plants have developmental mechanisms which are not primarily reliant on temperature and rainfall. As a result, they are limited in their ability to alter their flowering time during fluctuations in spring temperature and precipitation. Other factors that can influence flowering date include amount of sunlight per day (photoperiods), site conditions (prescribed fires, mowing, grazing), and genetic regulatory systems. Whereas temperature and rainfall appear

stochastic short term and increasing incrementally longterm [14], genetic systems and photoperiod remain stable. Thus, the differences in plant flowering trends in relation to climate could be due to these differing factors cueing growth, leading some plants to exhibit phenotypic plasticity and others to reach a physiological threshold that limits the ability to respond to a changing climate. Species such as yellow coneflower, meadow rue, and butterfly milkweed, which demonstrated consistency in flowering, may lack the physiological ability to vary flowering times, or perhaps they have developmental mechanisms that are predominantly regulated by photoperiod. Aside from mechanistic differences, it is possible that other climatic factors can alter flowering times. For example, some flowering plants may flower earlier with more precipitation. The amount of snow and rain during the 2018 spring season was larger than that of spring 2017. It is possible that the precipitation during the 2018 spring season helped plants such as prairie blazing star, flowering spurge, and rosinweed start their growth cycles sooner, which may have led to their earlier flowering dates. It is possible that climatic factors throughout the summer (precipitation, temperature, humidity) also influenced plants to grow and flower earlier than previous years. As demonstrated, flowering time is not necessarily tied to a single mechanism or influencer, making characterizing short-term trends of wildflower blooms difficult. Additionally, given the small fluctuations of less than 2ºC in spring temperatures between 2016 and 2018, it is questionable whether we would see significant biological differences in our data. Regardless, we observed differences in flowering times spanning as far as 43 days apart for individual plants even over just two years. Additional literature further supports that climate changes can influence phenological cycles in flora and fauna in different ways [15; 16]. Our data suggest that shifts in flowering time may arise in conjunction with numerous environmental life cycle regulators. These phenological shifts pose ecological problems as cues such as temperature, rainfall, and photoperiod align differently compared to previous years. Shifts in the timing of climatic factors that have historically been linked (e.g. temperature, day length, and rainfall) may result in ecological mismatches [17]. Organisms that used to be present at the same period over the season may become less synchronized, decoupling important interactions between symbionts such as flowers and pollinators and herbivores and host plants. Other effects could include water uptake, nutrition cycling, and carbon sequestration. A major consequence of the decoupling of ecological interactions and processes includes lowered seed and fruit production in plants, more competition between organisms for resources, and ultimately reduced population sizes and fitness with greater risk for extinction of species. Previous studies have demonstrated this loss of plant abundance specifically in species that display less phenotypic plasticity in response to changes in climate over time [18]. Thus, a major potential consequence of a desynchronized ecosystem is a loss of biodiversity. To better understand trends in flowering time, we

REPORT

encourage the continuation of this project over many years. Some of the limitations of this study stem from the limited data pool at the current time. Increasing the number of replicates for each species will greatly increase our confidence and ability to determine trends. We aim to increase the replicates of plant species during each year of data collection to improve upon our confidence in our results. While we aimed to reduce confounding variables that could influence flowering time such as varied management techniques (burning, mowing, trampling), the potential for these elements to confound our study will be mediated with the continuation of data collection over many years. Additionally, continuation of our study may allow us to observe biological changes in response to larger fluctuations in climate. Finally, our results were evaluated in consideration of spring temperature and precipitation. We hope to analyze weather data in other ways

(e.g. humidity, summer temperatures, and precipitation) to further investigate relationships between climatic factors and plant flowering.

REFERENCES

Extremes. 2015;10: 4–10. doi:10.1016/j.wace.2015.08.001 11. Pearce R. Plant Freezing and Damage. Annals of Botany. 2001;87: 417–424. doi:10.1006/anbo.2000.1352 12. Drew M, Armstrong W. Root Growth and Metabolism Under Oxygen Deficiency. Plant Roots. 2002;: 729–761. doi:10.1201/9780203909423.ch42 13. Hopkins EJ. Weather and climate data files: temperature and precipitation for south central Wisconsin. Wisconsin State Climatology Office. Wisconsin State Climatology Office; Available: http://www.aos.wisc.edu/~sco/ clim-history/data-portal.html 14. Kucharik CJ, Serbin SP, Vavrus S, Hopkins EJ, Motew MM. Patterns of Climate Change Across Wisconsin From 1950 to 2006. Physical Geography. 2010;31: 1–28. doi:10.2747/0272-3646.31.1.1 15. Sparks TH, Carey PD. The Responses of Species to Climate Over Two Centuries: An Analysis of the Marsham Phenological Record, 1736-1947. The Journal of Ecology. 1995;83: 321. doi:10.2307/2261570 16. Hunter AF, Lechowicz MJ. Predicting the Timing of Budburst in Temperate Trees. The Journal of Applied Ecology. 1992;29: 597. doi:10.2307/2404467 17. Primack RB, Miller-Rushing AJ. Tracking climate change with the help of Henry David Thoreau: Notebooks of the 19th-century naturalist help show changes to the flowers and fauna of Concord, Massachusetts. Elsevier. 2013; Available: https://www.elsevier.com/connect/ tracking-climate-change-with-the-help-of-henry-davidthoreau 18. Willis CG, Ruhfel B, Primack RB, Miller-Rushing AJ, Davis CC. Phylogenetic patterns of species loss in Thoreau’s woods are driven by climate change. Proceedings of the National Academy of Sciences. 2008;105: 17029–17033. doi:10.1073/pnas.0806446105 ●

1. Havens K, Henderson S. Citizen Science Takes Root. American Scientist. 2013;101: 378. doi:10.1511/2013.104.378 2. Tooke F, Battey NH. Temperate flowering phenology. Journal of Experimental Botany. 2010;61: 2853–2862. doi:10.1093/jxb/erq165 3. Ellwood ER, Temple SA, Primack RB, Bradley NL, Davis CC. Record-Breaking Early Flowering in the Eastern United States. PLoS ONE. 2013;8. doi:10.1371/journal. pone.0053788 4. Fitchett JM, Grab SW, Thompson DI. Plant phenology and climate change. Progress in Physical Geography. 2015;39: 460–482. doi:10.1177/0309133315578940 5. Craine JM, Wolkovich EM, Towne EG, Kembel SW. Flowering phenology as a functional trait in a tallgrass prairie. New Phytologist. 2011;193: 673–682. doi:10.1111/ j.1469-8137.2011.03953.x 6. Biederman L, Mortensen B, Fay P, Hagenah N, Knops J, La Pierre K, et al. Nutrient addition shifts plant community composition towards earlier flowering species in some prairie ecoregions in the U.S. Central Plains. PloS ONE. 2017, May 26; 12(5): e0178440. 7. Leopold A, Jones SE. A Phenological Record for Sauk and Dane Counties, Wisconsin, 1935-1945. Ecological Monographs. 1947;17: 81–122. doi:10.2307/1948614 8. Bradley NL, Leopold AC, Ross J, Huffaker W. Phenological changes reflect climate change in Wisconsin. Proceedings of the National Academy of Sciences. 1999;96: 9701–9704. doi:10.1073/pnas.96.17.9701 9. Knox County Master Gardeners. Garden Tips from Knox County Master Gardeners. Galesburg, IL: University of Illinois, U.S. Department of Agriculture ; 2014. 10. Hatfield JL, Prueger JH. Temperature extremes: Effect on plant growth and development. Weather and Climate

JUST VOL IV // ISSUE I // FALL 2018

ACKNOWLEDGEMENTS Funding was received from the UW Lakeshore Nature Preserve through the Student Engagement Grant program. We would like to thank Allie Hung for her work in establishing the study. Special gratitude is owed to Dr. Stanley A. Temple for giving us access to the historical data, which was collected between 1977 and 2012 and for providing exceptional insight into the field of phenology. Additionally, we would like to thank Seth McGee for his endless advice and unwavering support.

Soil Moisture Effects on Portable X-Ray Fluorescence Spectroscopy and Visible-Near Infrared Spectroscopy Measurements Sean Fischer, Alfred E. Hartemink, Yakun Zhang, and Jenifer L. Yost

University of Wisconsin-Madison, Department of Soil Science, F.D. Hole SoilsLab, 1525 Observatory Drive, Madison, WI 53706 USA. Corresponding author: Sean Fischer, sfischer5@wisc.edu.

ABSTRACT Portable X-ray fluorescence (pXRF) and visible-near infrared spectroscopy (vis-NIR) are used to measure or predict soil properties such as soil organic carbon, nutrients, soil texture, and environmental pollutants such as heavy metals. These characteristics have great importance in soils under agriculture as well as environmental remediation. When taken in the field, vis-NIR and pXRF measurements are affected by the moisture conditions of the soil. This study seeks to quantify the effects of soil moisture on pXRF and vis-NIR measurements and propose correction equations of moisture effects. Topsoil and subsoil samples from an Entisol (~sand) and a Mollisol (~silty loam) were wetted in the laboratory to six different moisture contents, and subsequently scanned by the pXRF and vis-NIR. The elemental intensity in spectra from the pXRF was significantly reduced by higher moisture contents, yielding lower elemental concentrations. This was linearly related to gravimetric water content. The effect was stronger in the silty loam soils than in the sandy soils. The reflectance of vis-NIR spectra decreased with increasing soil moisture content, and the reduction in reflectance of two spectral ranges (1410-1450 and 1910-1930 nm) was modeled by non-linear logarithmic regression. It is concluded that soil moisture reduces the measurements of pXRF and reflectance spectra of vis-NIR. The elemental concentrations measured by pXRF could be corrected by linear model with known gravimetric water content, whereas more sophisticated models are needed for vis-NIR spectra. The results of this study may improve the accuracy of pXRF and vis-NIR spectroscopy soil analysis used in the field, and could influence the way soil data is collected.

INTRODUCTION

JUST VOL IV // ISSUE 1 // FALL 2018

REPORT

Soil measurements are often costly and time-intensive [1], and there is a need for quicker and more accurate techniques to measure soil properties in the field [2]. Soil information is needed for precision agriculture, modeling, and other soil management that requires large amounts of accurate soil data. These management activities are important to a variety of industries, including agriculture. However, a great volume of spatial data is needed for this management. Proximal soil sensors collect data rapidly and are a cost-effective alternative to traditional, outsourced laboratory analysis [3]. Proximal soil sensing is defined as the use of fieldbased sensing equipment to, directly and indirectly, obtain data about properties of the soil [4]. Proximal soil sensing allows for faster and greater collection of data points, enabling large amounts of spatial data across broad areas to be collected to fill industry needs. Although individual sensor measurements may not be as accurate as traditional analysis, more data can be collected and thus more information on the soil resource is provided. This study attempts to improve the accuracy of field measurements of pXRF and visNIR proximal sensors by studying the possible correction of the effects of moisture on field measurements. Portable X-ray fluorescence (pXRF) spectrometer and visible near-infrared (vis-NIR) spectrometer are two examples of proximal sensors. The pXRF gives estimates of

total elemental concentrations in the soil (i.e., Mg, Si, Al and Ca). This is accomplished through the ionization of electrons of elements in the soil and the subsequent interpretation of properties of photo-electrons emitted following ionization [5]. Knowledge of elemental concentrations can be used in the diagnosis of contaminated soils or nutrient deficiencies. pXRF has also been used to characterize soil horizons, which are definable layers in the soil associated with soil properties [6, 7, 8]. The vis-NIR measures the diffuse reflectance of the soil in the visible and near-infrared portions (350-2500 nm) of the electromagnetic spectrum [9]. Spectral graphs are produced from measurements which can be analyzed to indirectly acquire information about soil properties, including soil organic carbon (SOC) concentration, texture, or moisture content [9, 10]. Both pXRF and vis-NIR spectrometers are especially useful because of their portability and ability to be used in the field. Measurements collected from the vis-NIR and pXRF are affected by soil moisture. In vis-NIR analyses soil moisture absorbs visible and near-infrared light, causing a decrease of reflectance in the vis-NIR spectra. In X-ray fluorescence soil moisture causes a linear decrease in the intensity (amplitude) of photoelectrons emitted by the soil [12]. This results in lower measured elemental concentrations [13], and the degree of this effect varies between elements, soil texture, and soil types [5]. The effect of soil moisture

conditions on these measurements remains to be quantified for different soils. Effects of moisture in different soils must be measured because of the variability of soils, which could affect sensor results differently. In this paper, the effects of soil moisture on elemental concentrations and reflectance spectra using pXRF and vis-NIR spectrometers was investigated in two soils with contrasting textures. The objectives of this research were to: (i) determine the variation in elemental concentrations and reflectance spectra with varying soil water contents, and (ii) develop a function to correct for soil moisture effects on pXRF and vis-NIR measurements.

orescence [5]. The process involves X-rays being generated within an X-ray tube inside the pXRF, and then directed at the soil surface. Energy is released and is measured by the pXRF as fluorescence. By measuring the frequency of the energy emitted, the pXRF can determine the types of elements struck by the X-rays. By measuring the intensity (or amplitude) of different waves, the pXRF can estimate the amount of those elements present in the sample in parts per million [5].

REPORT

METHODS The Soils Two soils from Wisconsin were used in this experiment: an Entisol (Soil series name: Plainfield sand; Classification: mixed, mesic Typic Udipsamments) and a Mollisol (Soil series name: Troxel silt loam; Classification: fine-silty, mixed, superactive, mesic Pachic Argiudolls). Entisols are recently formed soils, usually with a relatively thin topsoil layer. Mollisols are soils that have a thick topsoil and properties useful in agriculture. The Entisol (sand) was collected from Wallendal Farms in the Wisconsin Central Sand Plains (WGS84 43.91째N, 89.68째W) at an elevation of approximately 311 meters. The soil was under agriculture at the time of sampling. The soil is excessively drained and sandy throughout and was formed in a sandy drift on glacial outwash plains. Mean annual precipitation is about 833 mm. Typical horizons are Ap (0-24 cm), Bw1 (24-55 cm), Bw2 (55-110 cm), C (110+ cm). The Mollisol (silty loam) was located at University of Wisconsin West Madison Agricultural Research Station in south-central Wisconsin (WGS84 43.07째N, 89.54째W). The soil was formed in windblown deposits over a glacial outwash plain which was underlain by dolostone bedrock (approximately 3 m). The soil was moderately well-drained to well-drained and was under grassland at the time of sampling. Six horizons were identified in the field: Ap (0-20 cm), A2 (20-55 cm), Ab (55-62 cm), E (62-70 cm), Bt1 (70-85 cm), and Bt2 (85-102 cm). For each soil, topsoil (0-10 cm for Entisol, 0-9 cm for Mollisol) and subsoil (40-50 cm for Entisol, 99-102 cm for Mollisol) were selected. In the Entisol, the texture was uniform throughout the profile, with approximately 91% sand, 7% silt, and 4% clay content. SOC and pH decreased with depth (2.4% to 0.2% SOC; pH 6.4 to 5.8). In the Mollisol, the topsoil had approximately 13% sand, 67% silt, and 20% clay content and pH and SOC decreased with depth (2.8% to 0.3%; pH 7.1 to 5.5). The subsoil in the Mollisol had approximately 8% sand, 61% silt, and 31% clay content. pXRF Measurements Portable X-ray fluorescence spectroscopy nondestructively measures the presence and abundance of multiple elements in a sample simultaneously by using X-ray flu24

JUST VOL IV // ISSUE I // FALL 2018

Figure 1: Mechanics of portable x-ray fluorescence (pXRF) spectrometry. An x-ray photon (a) emitted by a pXRF machine ejects an inner-shell electron (b), causing a cascade effect for the electrons in the outer shells to fall inward (c). Photograph of the portable X-ray fluorescence spectrometer used (d).

The elemental (Zn, Ti, Ni, Fe, Zr, Sr, Mn, and Cu) concentrations were determined by using a Delta Premium portable X-ray fluorescence (pXRF) spectrometer (Olympus Scientific Solutions Americas Inc., Waltham, MA). The Geochem mode was used in this study and operated for a duration of 60 s in a two-beam configuration at 40 and 10 kV [17]. The spectrometer was calibrated using a 316-stainless steel calibration check coupon prior to taking measurements. The elemental concentrations were measured using an internal factory-installed calibration procedure (the Compton normalization method) [17]. vis-NIR Spectroscopy Visible-near infrared diffuse reflectance spectroscopy provides information about the soil by measuring the light it reflects in the visible and near-infrared parts (3502500 nm) of the electromagnetic spectrum. Radiation with frequencies corresponding to these parts of the spectrum is directed at a sample. The spectrometer reads the diffuse reflected light that was not absorbed by the sample and computes a percent reflectance value for each wavelength in the range 350-2500 nm. These values can be plotted as a spectral curve. This curve can be analyzed to gain information about important soil properties such as soil minerology, carbon content, soil texture, and soil moisture characteristics [9].

Figure 2: Photograph of the visual-near infrared spectrometer used (a). Example of a reflectance spectra produced by the vis-NIR spectrometer (b)

A PSR-3500 spectrometer (Spectral Evolution, Lawrence, MA) was used to scan the samples in the laboratory. The vis-NIR operates in the range of 350-2500 nm with three detectors: 1) a 512-element silicon PDA covering the visible range and part of the near infrared (350-1000 nm) with a resolution of 3 nm; 2) a 256-element InGaAs array covering 1000-1900 nm with a resolution of 8 nm; and 3) a 256-element InGaAs array covering 1900-2500 nm with a finer spectral resolution of 6 nm. The reflectance data were resampled to 1 nm for output and this results in 2151 spectral points. The reflectance spectrum was recorded by averaging 30 readings per soil sample measurement, and three replicates were taken for each sample by repositioning the probe between each scan. The vis-NIR was calibrated by a white plate made of polytetrafluoroethylene and was recalibrated every 10 samples. Continuum removal technique was applied to the reflectance spectra using the “prospectr” package [14] in R version 3.4.3 [15] to remove the general trend of the spectra and sharpen major absorption features. Experimental Design The four soil samples (Entisol topsoil 0-10 cm depth, Entisol subsoil 10-50 cm depth, Mollisol topsoil 0-9 cm, and Mollisol subsoil 99-102 cm) were air-dried, ground, and passed through a 2-mm mesh sieve. Each sample was split into six subsamples with three replicates to achieve six gradients of water contents. Each replicate contained 25 g of soil, which ensured a minimum soil depth of 5 mm for the scan. The Entisol samples (sand) received 0, 1, 2, 3, 4, and 5 ml of deionized water, while the Mollisol samples (silt loam) received 0, 2, 4, 6, 8, and 10 ml of deionized water. All samples were sealed in plastic wrap and left for 24 hours in order for moisture to spread in the soil sample. Each replicate was weighed and scanned using the pXRF and the vis-NIR spectrometers. Samples were transferred to metal containers and oven-dried at 105°C for 24 hours. The oven-dried samples were weighed and the gravimetric water content (θ_m) was calculated as follows: θ_m= ((wet weight) - (dry weight)) / ((dry weight)) = (mass of water)/(mass of oven-dried soil)

Moisture Effects on pXRF Soil moisture reduced the pXRF measurements in both soils. All elemental concentrations were lower in the Entisol (Fig. 3) than in the Mollisol (Fig. 4) at similar moisture contents. Measured elemental concentrations decreased linearly with increasing moisture content in both the topsoil and subsoil for the Entisol for most elements (Fig. 3). In the Mollisol, the elemental concentrations decreased linearly with increased moisture content for all elements (Fig. 4). In the Entisol, Ni, Sr, Zr, and Cu displayed poor correlation in the topsoil, subsoil, or in both, while Mn, Ti, Fe, and Zn were slightly correlated (R2 > 0.2) (Fig. 3). Stronger correlations were observed in the Mollisol, and the subsoil tended to have stronger correlations than in the topsoil. In general, there was greater variation in the Entisol than in the Mollisol. The best-fit line was linear and established the relationship between gravimetric water content and elemental concentrations (Tables 1 and 2).] Moisture Effects on Vis-NIR Spectroscopy The reflectance of the soil decreased with increasing soil moisture in the topsoil and subsoil of both soils (Fig. 5). In the topsoil, soil moisture influenced the reflectance between 350-1000, 1350-1700, and 1700-2200 nm. In the subsoil, soil moisture influenced the reflectance between 1350-1700, 1700-2200, and 2300-2500 nm. Between 13002200 and 2200-2500 nm, the reflectance decreased more in the subsoils than in the topsoils with increasing soil moisture. However, in the visible wavelength (350-700 nm), the moisture effect was less in the subsoils. The reduction in reflectance from maximum reflection (equal to “1”) was plotted for the two wavelength ranges (1410-1440 nm and 1910-1930 nm) which experienced the greatest changes due to moisture (Fig. 6). A logarithmic best fit line was applied to each of these plots, indicating that as soil moisture increases, reflection is reduced less than at lower moisture levels.

DISCUSSION pXRF A decrease in elemental concentrations (decrease in X-ray intensity) was observed as a result of increased moisture content. Moisture is known to affect pXRF measurements in two ways: 1) water enhances a substance’s absorption of X-rays, resulting in decreased intensity, and 2) water particles cause X-ray scattering which increases X-ray intensity [12]. Results from this study indicate that the former effect is more pronounced than the latter, and can significantly decrease X-ray intensity, leading to a decrease in elemental concentrations. Limited correlations between soil moisture and measured elemental concentrations in the Entisol might be due to the high content of sand in the soil. Coarser-textured soils tend to cause higher variability in elemental concentrations than in finer-textured soils [5]. Sand has low water JUST VOL IV // ISSUE 1 // FALL 2018

REPORT

Data Analysis The relationship between elemental concentration and gravimetric water content was studied for each of the elements: Ni, Mn, Fe, Ti, Cu, Zn, Sr, and Zr in every soil sample (Entisol topsoil, Entisol subsoil, Mollisol topsoil, Mollisol subsoil). The coefficient of determination (R2) was calculated to quantify the fit of the trendlines created. A linear model for pXRF was used because of success correcting results in past studies [12]. The continuum removed spectra for vis-NIR were averaged over three ranges of gravimetric water contents (<6%, 10-25%, 25-40%). The nonlinear relationship between the reflectance reduction at two major troughs from 1410-1450 nm and 1910-1930 nm were studied for the mean spectra of the three replicates and the R2 was calculated for every soil sample.

RESULTS

Figure 3.Topsoil (0-10 cm depth) and subsoil (40-50 cm depth) elemental concentrations by varying gravimetric water content (GWC) for the Entisol.

holding capacity and can lose moisture quickly to evaporation. Evaporation during the experiment may have been higher in the sandy soils than in the silty loam soils, causing higher variability in the sandy soils.

REPORT

Table 1. Linear relationships between elemental concentration and gravimetric water content for the Entisol.

Table 2. Linear relationships between elemental concentration and gravimetric water content for the Mollisol.

JUST VOL IV // ISSUE I // FALL 2018

vis-NIR Spectroscopy The results found in this study agree with those presented in Stenberg et al. [9] and Minasny et al. [11] that absorbance in the 1400 nm and 1900 nm areas are affected by soil moisture, resulting in a decrease in reflectance. The reflectance decreased due to O-H bond stretching and bending in response to the light energy [9]. Water generally decreases the albedo of most surfaces due to an increase in light refraction as the photons pass through films of water which form on the surfaces. This creates a scattering effect, increasing the path of travel of the photons and further increasing the chances that the light will be absorbed [9]. Soil moisture had an effect on the results for topsoil measurements in the visible range but did not have a considerable effect in the subsoils, as can be seen in Fig. 5 between 350-600 nm. The reason for this could be linked to higher soil organic carbon (SOC) concentrations in the topsoil, where SOC responds with a decrease in reflectance with increased soil moisture [9].

Figure 4.Topsoil (0-9 cm depth) and subsoil (99-102 cm depth) elemental concentrations by varying gravimetric water content (GWC) for the Mollisol.

REPORT Figure 5. Continuum-removed soil reflectance spectra in the visible-rear infrared wavelength range (350-2500 nm) at three different gravimetric water content ranges for the Entisol (topsoil: 0-10 cm depth, subsoil: 40-50 cm depth) and Mollisol (topsoil: 0-9 cm depth, subsoil: 99-102 cm depth).

JUST VOL IV // ISSUE 1 // FALL 2018

high R2 values of the regressions, while the correction for the Entisol would not be as accurate, as can be seen in the low R2 values of the regressions. Vis-NIR results skewed by soil moisture can be corrected through complex nonlinear analysis, greatly assisting measurements in the field. The effect on absorbance varies between soils due to differing mineralogy, complicating corrections [9]. Minasny et al. [11] and Ge et al. [16] have used external parameter orthogonalization (EPO) to successfully remove the moisture effects. In this study EPO was not used, instead the reflectance of two spectral ranges (1410-1450 and 1910-1930 nm) was described using logarithmic regression with known gravimetric water content. This model was successful in describing the moisture effects in these ranges, shown in the high R2 values of regressions in Figure 6.

CONCLUSION Figure 6. Measured reduction in reflectance at two wavelength ranges (1410-1440 nm and 1910-1930 nm) as affected by gravimetric water content (GWC %) for the Entisol (topsoil: 0-10 cm depth, subsoil: 40-50 cm depth) and Mollisol (topsoil: 0-9 cm depth, subsoil: 99-102 cm depth).

REPORT

Correction Function An empirical correction function may be found for pXRF results, enabling more accurate measurements in the field. Achieving higher accuracy with pXRF in situ has broad implications for soil science. For example, spatial distributions of soil nutrients under agricultural land could be developed efficiently, allowing fertilization and other soil remediation activities to occur with precision. The pXRF inference models for improved accuracy could be developed easily because of the linear relationship between decreasing intensity and soil moisture. Stockmann et al. [12] used 10 subsamples from a Dermosol to derive a correction function for Fe using linear regression and applied it to successfully correct the measurements of 120 samples taken in situ. A similar correction can be done for the soils studied here with known gravimetric water content (Table 1 and Table 2) since the elemental concentration decreased linearly with increasing water content. This study created correction functions for Fe concentrations, as well as 7 additional elements not studied in Stockman et al. [12]. A correction function for the Mollisol would be highly accurate, as can be seen in the

REFERENCES 1. Adamchuk, V., Viscarra Rossel, R. Development of onthe-go proximal soil sensor systems. Proximal soil sensing. Springer. 2010; pp. 15-28. 2. Viscarra Rossel, R.A.V., McBratney, A.B., Minasny, B. Proximal soil sensing. Springer. 2010. 3. Sahraoui, H., Hachicha, M. Effect of soil moisture on trace elements concentrations using portable X-ray fluorescence spectrometer. Journal. 2017; 9(1): 17. 4. Viscarra Rossel, R.A., Adamchuk, V.I., Sudduth, K.A., 28

JUST VOL IV // ISSUE I // FALL 2018

Soil moisture content affects pXRF and vis-NIR measurements. The elemental intensity in spectra given by pXRF was reduced by higher moisture contents, yielding lower elemental concentrations. The reflectance of vis-NIR spectra decreased with increasing soil moisture content due to higher absorbance of light by the vibration of water molecules. The elemental concentrations were shown to have linear relationships with gravimetric water content, and the effect was more pronounced in the silt loam soils (Mollisol) than in the sandy soils (Entisol). The linear regression could be used to correct for elemental concentrations with known gravimetric water content. The vis-NIR spectra of two spectral ranges (1410-1450 and 1910-1930 nm) had a nonlinear correlation with gravimetric water content which may need more sophisticated correction methods to account for the soil moisture effects. In the future, these instruments could be further developed to perform corrections in situ following calibration from site-specific samples. High-resolution soil data could be gathered efficiently, greatly assisting management activities requiring soil data.

ACKNOWLEDGEMENTS We would like to thank the F.D. Hole Soil Lab for making this research possible. McKenzie, N.J., Lobsey, C. Proximal Soil Sensing: An Effective Approach for Soil Measurements in Space and Time. Advances in Agronomy. 2011; 113: 243-291. 5. Weindorf, D.C., Bakr, N., Zhu, Y. Advances in Portable X-ray Fluorescence (pXRF) for Environmental, Pedological, and Agronomic Applications. Advances in Agronomy. 2014; 128: 1-45. 6. Grauer-Gray, J., Hartemink, A.E., Raster sampling of soil profiles. Geoderma. 2018; 318: 99-108. 7. Weindorf, D.C., Zhu, Y., McDaniel, P., Valerio, M., Lynn, L., Michaelson, G., Clark, M., Ping, C.L. Characterizing

soils via portable X-ray fluorescence spectrometer: 2. Spodic and Albic horizons. Geoderma. 2012; 189–190: 268-277. 8. Zhu, Y., Weindorf, D.C., Zhang, W. Characterizing soils using a portable X-ray fluorescence spectrometer: 1. Soil texture. Geoderma. 2011; 167–168: 167-177. 9. Stenberg, B., Viscarra Rossel, R.A., Mouazen, A.M., Wetterlind, J. Visible and near infrared spectroscopy in soil science. Advances in Agronomy. 2010; 107: 163-215. 10. Zhang, Y., Hartemink, A.E. Sampling designs for soil organic carbon stock assessment of soil profiles. Geoderma. , 2017; 307: 220-230. 11. Minasny, B., McBratney, A.B., Bellon-Maurel, V., Roger, J.-M., Gobrecht, A., Ferrand, L., Joalland, S. Removing the effect of soil moisture from NIR diffuse reflectance spectra for the prediction of soil organic carbon. Geoderma. 2011; 167–168: 118-124. 12. Stockmann, U., Jang, H.J., Minasny, B., McBratney, A.B. The Effect of Soil Moisture and Texture on Fe Concentration Using Portable X-Ray Fluorescence Spectrome-

ters. In: A.E. Hartemink, B. Minasny (Eds.), Digital Soil Morphometrics. Springer, Dordrecht. 2016; pp. 63-71. 13. Santana, M. L. T. Rieberto, B. T., Silva S. H. G., Poggere, G. C., Guilherme, L. R. G., and Curi, N. Conditions affecting oxide quantification in unknown tropical soils via handheld X-ray fluorescence spectrometer. CSIRO Publishing. 2018. 14. Stevens, A., Ramirez-Lopez, L., Stevens, M.A., Rcpp, L. Prospectr: Miscellaneous functions for processing and sample selection of vis-NIR diffuse reflectance data. 2013; R package version 0. 1. 3, pp. 32. 15. R Core Team. R: A Language and Environment for Statistical Computing, Vienna, Austria. 2016. (At: http:// www.R-project.org/). 16. Ge, Y.F., Morgan, C.L.S., Ackerson, J.P. VisNIR spectra of dried ground soils predict properties of soils scanned moist and intact. Geoderma. 2014; 221: 61-69. 17. Olympus. 2012. DELTA Family Handheld XRF Analyzer: User Interface Guide. Waltham, MA: Olympus NDT.●

REPORT JUST VOL IV // ISSUE 1 // FALL 2018

Cost-Efficiency of Treating Vivax Malaria in Pregnancy in Thai Refugee Camps Using Glucose-6-PhosphateDehydrogenase Testing and Selective Primaquine Treatment Diane Xue ABSTRACT Pregnant women have an increased risk of contracting malaria, especially in densely populated areas. The inability to prescribe Primaquine, the only known treatment of vivax malaria, to pregnant women is a major complication for malaria eradication. Primaquine is contraindicated for pregnant women due to the unknown risk of Glucose-6-Phosphate Dehydrogenase (G6PD) deficiency of the fetus. G6PD deficient individuals are at risk of severe hemolytic anemia if exposed to Primaquine. Genetic screening for G6PD could be done on pregnant women to determine the G6PD status of a male fetus due to the X-Linked nature of the gene. However, cost remains a hurdle for implementing new genetic screening protocols. In this study, I evaluated the cost-effectiveness of implementing point-of-care testing for G6PD in refugee camps on the Thai-Myanmar border using recent data from the Shoklo Malaria Research Unit on the effects of malaria in pregnancy on newborn mortality. The cost-effectiveness ratios based on infant mortality rates and value of statistical life measurements are in comparison to the current strategy of no screening. My results suggest that the screening strategy is cost-effective relative to no screening. To the best of my knowledge, this is the first cost-effectiveness analysis for evaluating genetic testing of pregnant women for G6PD as a means of malaria eradication in Thai-Myanmar border refugee camps.

INTRODUCTION

REPORT

The World Health Organization (WHO) estimates that annually, over 300,000 mothers and 2.7 million newborns die around the time of childbirth. In Thailand, the infant mortality rate (IMR) is 11 per 1000 live births. The IMR is particularly high in refugee camps on the Thai-Myanmar border, where health care services are limited and disease spreads quickly. Malaria, the most prevalent parasitic disease in the world, increases complications during pregnancy in the refugee camps [1]. Malaria during pregnancy significantly increases risk of preterm birth, which is the strongest predictor of infant mortality and morbidity, including disability, stunting, and non-communicable diseases later in life [2, 3]. The primary focus of scientists studying malaria has been on infections with Plasmodium falciparum, however, malaria caused by Plasmodium vivax is particularly devastating for pregnant women because it can relapse multiple times even after treatment. Additionally, pregnant women are significantly more susceptible to malaria due to changes in their immune system. Each time a woman becomes infected with malaria throughout the course of the pregnancy, her health and the health of the fetus are under greater stress [4, 5]. Relapse of vivax malaria is caused by dormant stages of the malaria Plasmodium vivax life cycle called hypnozoites. Hypnozoites stay in the liver and activate months or even years after the first episode. There is only one anti30

JUST VOL IV // ISSUE I // FALL 2018

malarial proven to target hypnozoites and prevent relapse: a drug called Primaquine. Unfortunately, for individuals who are deficient in the enzyme G6PD, Primaquine can trigger hemolysisâ&#x20AC;&#x201D;the bursting of red blood cellsâ&#x20AC;&#x201D;causing hemolytic anemia and further health complications. Primaquine is never prescribed to pregnant women due to the unknown risk of G6PD deficiency in the fetus and increased severity of anemia-related health risks during pregnancy [6]. In Thailand, where G6PD deficiency ranges from ten to fifteen percent, country-wide G6PD screening of newborns is not universal, nor is screening for G6PD deficiency at the time of malaria diagnosis. I hypothesized that by implementing a new standard in which women who contract malaria in pregnancy are screened for G6PD deficiency, then prescribed Primaquine if they are G6PD normal, the benefits in reducing the risk of infant mortality would outweigh the costs of implementing routine G6PD testing. In this paper, I conducted a review on the costs associated with implementing genetic testing for G6PD deficiency and providing Primaquine treatment, weighed against the benefit of lives saved from preventing malaria relapse. The Biology of Glucose-6-Phosphate-Dehydrogenase Glucose-6-phosphate dehydrogenase (G6PD) is a housekeeping enzyme, meaning it is expressed in all cells and required for the maintenance of cell function [7]. The G6PD enzyme is a part of the Pentose Phosphate Pathway, a pathway necessary for cell metabolism. As a part of the

Pentose Phosphate Pathway, G6PD produces reduced Nicotinamide adenine dinucleotide phosphate (NADPH). NADPH protects cells from reactive oxygen species. NADPH is crucial in red blood cells, because when reactive oxygen species build-up, oxidative stress causes hemolysis (bursting of red blood cells) which can cause low blood pressure, skin and eye yellowing, and heart disease. In red blood cells, the Pentose Phosphate Pathway is the only source of NADPH. Therefore, without functional G6PD enzymes, there is no line of defense against reactive oxygen species building up to toxic levels in red blood cells [8]. G6PD deficiency itself does not cause hemolysis, but it makes the cells susceptible to external triggers such as Primaquine. The G6PD enzyme is made from the G6PD gene, which is located on the X-Chromosome. G6PD deficiency is widespread and heterogeneous across the world because there are over 200 single-nucleotide-polymorphism (SNP) mutations of the G6PD gene that can lead to G6PD deficiency [9]. Because G6PD is on the X-chromosome men can either be affected (hemizygous) with G6PD deficiency or unaffected. Because females have two X-chromosomes, females can be unaffected (homozygous), intermediately affected (heterozygous), or affected (homozygous). Due to the wide range of intermediate phenotype in females caused by mosaicism—random X-chromosome inactivation—the cost-benefit model in this paper will only include mothers who are pregnant with males. The G6PD status of a male fetus can be screened by testing the mother alone. The mother will pass one of two X-chromosomes to her offspring. If the mother is found to be intermediately or fully deficient, it would be unsafe to prescribe Primaquine. But any mother who tests as G6PD normal would be safe to treat.

Thailand’s Current Malaria Treatment Guidelines for Pregnant Women Today there is a conservative estimate of 97,000 refugees spread throughout 9 camps along the Thai-Myanmar border [11]. Many refugees seek treatment for malaria from the Shoklo Malaria Research Unit (SMRU) Clinics. SMRU has five clinics throughout the refugee camps, and provides care focused on maternal-child health and infectious disease [12]. SMRU also conducts research and collects longitudinal data tracking birth outcomes. From 1986 to 2015, SMRU tracked 50,060 pregnant women and their birth outcomes [13]. Out of the 50,060, there were 4,871 preterm births. 8,221 of the babies were born to mothers with malaria during pregnancy, and 1,335 women had relapsing malaria during pregnancy (figure 1). They found that odds of preterm birth, the world's leading cause of infant mortality and death in children under 5, increased linearly by 1.79 fold with each relapse. These relapses could have been prevented with appropriate Primaquine treatment [13]. G6PD Screening for Pregnant Women According to the Thai Ministry of Public Health, current treatment guidelines state that patients “should be screened for G6PD deficiency where they are able, but screening is not mandatory” [14]. Without knowing the G6PD status of the mother, and by proxy the unborn fetus, it is unsafe to prescribe Primaquine to pregnant women. In Thailand, pregnant women who contract malaria are currently treated with Chloroquine, an antimalarial that is not effective in preventing relapse [15, 16]. In the past, it would have been implausible for all pregnant women to be tested for G6PD deficiency upon malaria diagnosis since the genetic screen relied on sophisticated hospital and laboratory facilities. However, in recent years, new rapid diagnostic G6PD genetic screens have been developed. The most widely used test is the CareStart Qualitative Point-of-Care test which only requires 2 µL of blood and has results of either normal or deficient within 10 minutes [17]. This test does not require a sample to be sent to a lab, and therefore, can be easily used by community health workers and clinicians at the Shoklo Malaria Research Unit where refugees seek medical treatment for pregnancy and malaria. Implementing a New Treatment Workflow for Treating Malaria in Pregnancy With the development of a point-of-care test that can be used in a field setting, I propose the following workflow: For all pregnant women who contract malaria, first conduct an antenatal ultrasound to determine the sex of the baby. If the baby is a boy, screen the mother for G6PD deficiency using the CareStart Rapid Diagnostic Test. If the test comes out positive, the pregnant women will be treated with Chloroquine. The CareStart Rapid Diagnostic Test has 100% sensitivity and 97% specificity.

• •

JUST VOL IV // ISSUE 1 // FALL 2018

REPORT

G6PD Deficiency and Malaria G6PD deficiency is one of the most common enzymopathies (deleterious defect of a specific enzyme) in the world, with high prevalence in African, Mediterranean, and Southeast Asian populations. In particular, G6PD deficiency is prevalent in areas where malaria is endemic or has been endemic in the recent past. The reason for the linked distribution is because malaria, an infection of the red blood cells, causes G6PD deficiency to be more prevalent in a population because weakened blood cells confer some protection against infection by malarial parasites [10]. Conversely, in areas where malaria is not highly transmitted, G6PD deficiency does not provide any advantages, and is not evolutionarily selected for. This phenomenon, called “balancing selection” may explain why many different SNP mutations causing G6PD deficiency have evolved independently among populations in areas with malaria transmission. Unfortunately, G6PD deficiency does not prevent one from contracting malaria. G6PD deficient individuals who contract vivax malaria are faced with two unfortunate options without G6PD screening. They either risk hemolysis to completely treat malaria using Primaquine or use an antimalarial that is not as effective and risk relapse [9]. The WHO has suggested that all countries implement G6PD screening prior to

prescribing malaria treatment. However, Thailand has not adopted a universal screening policy.

â&#x20AC;˘ With an average G6PD deficiency rate of

15%, the majority of women will be able to be treated with Primaquine, and prevent relapse. Primaquine is 96% effective at preventing relapse [15]. Figure 1 shows calculations for the proposed workflow using the data from the SMRU collected from 1986 to 2015. The Cost of Implementing Mandatory G6PD Screening and Primaquine Treatment In order to estimate the costs of implementing G6PD screening for refugees who contract malaria in pregnancy on the Thai-Myanmar border, the following costs must be considered (see table below):

REPORT

Calculating Preventable Costs from G6PD Screening using Value of a Statistical Life Measurements Value of Statistical life (VSL) is often used when governments or other agencies need to measure the benefits of a new proposed policy that reduces health risks. In general, VSL is the benefit from preventing a death. VSL is often estimated at 14-20 times the GDP Per Capita for a 40 year working life and 10% discount rate [20]. For Thailand, the GDP per capita in 2016 US dollars is $5907.91. Therefore, the estimated value of statistical life ranges from $2,977,586 to $4,253,695: 14 to 20 (5907.91)(40)(.9)=$2,977,586 to $4,253,695 The cost-benefit model is comparing the newborn lives saved under G6PD screening/Primaquine treatment to the cost of implementing the new test and prescribing Primaquine. VSL will be used to put a monetary estimate for each newborn life to allow this comparison.

Globally, 12.3 percent of preterm births end in neonatal death, while .2 percent of full-term births do. This means that babies born preterm are 61.5 times more likely to face neonatal death. From our proposed model, using the data SMRU malaria in pregnancy data, 112 newborns could be carried to term when they would have faced preterm birth under the current protocol. It can be estimated that .123(112) = 13 newborn lives could be saved. The Cost-Benefit Analysis Model The cost-benefit analysis will be used to assess whether screening women who contract malaria during pregnancy for G6PD deficiency prior to deciding on treatment is worth the cost. Calculating Total Benefits according to the VSL estimate = total newborn lives saved*VSL low estimate =13 *2,977,586 USD=$38,708,618 Calculating Total Costs of screening for G6PD and prescribing Primaquine = (lab technician to handle G6PD testing(1 per clinic(5 clinics))) + ( CareStart RDT(# of women with malaria pregnant with boys)) + (((14 day Primaquine treatment+supervised therapy session))(# of women who do not test positive)) =$337,931. 72(5)+$1.75(253)+(($.84+$1.67)(209))=$1,690,625.94 Calculating the Cost-Benefit Ratio =(total benefit)/(total cost)=38,708,618/1,690,625.94=22.896

DISCUSSION The Cost-Benefit Ratio shows that when considering the value of statistical life, implementing a new G6PD Deficiency screening protocol in Thai-Myanmar refugee camps that allows some women to be treated with Pri-

Cost or Value

Source

Pregnancy test

[18]

Testing for malaria during pregnancy

[18]

Antenatal ultrasound (gender reveal)

Employing extra personnel to handle the new testing service.

Typical cost of lab tech in Thailand= 575,018 (THB) = 17785.88 USD (2016). Calculate for 1986 to 2015. = 337,931.72 USD

https://www.salaryexpert.com/ salary/job/medical-and-clinicallaboratorytechnologist/thailand

CareStart G6PD test

~1.75 per test

[19]

Supervised Primaquine therapy

1.67 per session with a community health worker

[19]

Primaquine treatment

.06 per pill x 14 days = .84 per person

[19]

[18]

** SM RU provides malaria tests, pregnancy tests, and antenatal care throughout the course of pregnancy. For all pregnant women, they encourage participating in their antenatal care program, which includes ultrasounds between weeks 8-13 and weeks 18-24.Therefore, in calculating costs, I have assumed that the costs for the first three rows would be the same regardless of malaria treatment, and would not be an additional cost to the new protocol.

JUST VOL IV // ISSUE I // FALL 2018

maquine would be worth the cost. The total costs include hiring new personnel (one new lab tech for each clinic site), implementing a rapid diagnostic test, Primaquine treatment, and the personnel to supervise the Primaquine therapy. The cost estimates are based on the data collected by the SMRU from 1986 to 2015 (figure 1). The total benefits are derived from the benefit of preventing infant mortality. Prescribing Primaquine and preventing relapse will not only benefit the mother and child’s health, but also help malaria eradication in the region. In the future, I plan on fine-tuning the value of statistical life estimate. There are more accurate ways to calculate the VSL, but for the purposes of simplicity here, I have used a more general estimate. There are many different factors that could impact VSL in Thai-Myanmar refugee camps, most notably, the lack of employment in the camps. Additionally, I would also like to research extra costs and

costs saved. Other costs that could be considered would be travel time, education about the genetic screening process, and value of future health outcomes.

ACKNOWLEDGEMENTS I’d like to thank my research mentor Dr. Jason Fletcher and the WiscPAL research group for their support and feedback throughout the course of this project. I’d also like to thank the Shoklo Malaria Research Unit for all of their work in the refugee camps on the Thai-Myanmar border, and their dedication to improving health outcomes through research and data collection, which I have used in this project. Finally, thank you to the College of Agricultural and Life Sciences for the generous CALS Research Award that supported my time devoted to research.

FIGURES AND TABLES

Women with Malaria in Pregnancy

Antenatal Ultrasound (Gender Reveal)

G6PD Screen

Breaking the 4,871 down to a 1.79:1 ratio comparing those caused by malaria during pregnancy, to those that are caused by other factors, about 3,125 of the pre-term births were due to malaria in pregnancy. Assuming 16.24% of mothers had multiple episodes and gender selection doesn't play a role, (.1624)(.5)(3125)= 253.75 of preterm births due to multiple episodes of malaria were boys.

The women pregnant with boys will be screened for G6PD deficiency using the CareStart Rapid Diagnostic Test, which has 100% sensitivity and 97% specificity. 253.75 246.14 women will be accurately diagnosed. With an average G6PD deficiency rate of 15%, 209.22 women can be treated with Primaquine upon their first episode of malaria.

209.22 pregnant women with malaria would be treated with Primaquine. Primaquine is 96% effective at preventing relapse [15]. 209.22. With the proposed workflow, 200.85 babies will be 55.86% (calculated by 1/1.79) less likely to face preterm birth. About 112.2 babies would be carried to term and face a significantly lower chance of ending in neonatal death.

Figure 1: Using the data from the Shoklo Malaria Research Unit on the effects of malaria in pregnancy, I calculated the results of each stage of the proposed workflow and found that approximately 112 male babies would be carried to term that would have previously faced neonatal death without G6PD Screening.

JUST VOL IV // ISSUE 1 // FALL 2018

REPORT

Malaria Treatment Prescription

From 1986 to 2015, SMRU tracked 50,060 pregnant women. Out of the 50,060, there were 4,871 preterm births. 8,221 of the babies were born to mothers with malaria during pregnancy. Of the 8,221, 16.24% of the mothers had multiple episodes. The odds of preterm birth increased linearly by 1.79 fold with each relapse. In other words, with each relapse, a pregnant woman was 1.79 times more likely to give birth to a pre-term baby.

REFERENCES

REPORT

1. Lives at risk: Malaria in pregnancy. (2013, April 29). Retrieved from http://www.who.int/features/2003/04b/ en/ 2. Katz J, Lee ACC, Kozuki N, Lawn JE, Cousens S, Blencowe H, et al. (2013) Mortality risk in preterm and small-for-gestational-age infants in low-income and middle-income countries: a pooled country analysis. Lancet; 382:417–25. 3. Lawn JE, Blencowe H, Oza S, You D, Lee ACC, Waiswa P, et al. (2014). Every newborn: progress, priorities, and potential beyond survival. Lancet;384:189–205. 4. Nosten F, McGready R, Simpson JA, Thwai KL, Balkan S, Cho T, et al.(1999) Effects of Plasmodium vivax malaria in pregnancy. Lancet;354:546–9. 5. Desai M. (2007). “Epidemiology and Burden of Malaria in Pregnancy”. Lancet;7(2): 93-104 6. WHO, 2015. Guidelines for the Treatment of Malaria, 3rd edition. Geneva, Switzerland: World Health Organization. Available at: http://apps.who.int/iris/ bitstream/10665/162441/1/9789241549127_eng.pdf?ua=1&ua=. 7. Cappellini M. (2008). "Glucose-6-phosphate dehydrogenase deficiency". Lancet; 371: 64-74 8. Naylor C. (1996). "Glucose 6-phosphate dehydrogenase mutations causing enzyme deficiency in a model of the tertiary structure of the human enzyme". Blood; 87: 2974 – 2982 9. McGready R, Lee SJ, Wiladphaingern J, Ashley EA, Rijken MJ, Boel M, et al. (2012) Adverse effects of falciparum and vivax malaria and the safety of antimalarial treatment in early pregnancy: a population-based study. Lancet;12:388–96. 10. Baird K. (2015). “Origins and Implications of Neglect of G6PD deficiency and primaquine toxicity in Plasmodium vivax malaria”. Pathogens and Global Health;109(3): 93-106.

JUST VOL IV // ISSUE I // FALL 2018

11. Refugees in Thailand. (2018, August 31). Retrieved from https://www.unhcr.or.th/en 12. About SMRU. (2018). Retrieved from http://www.shoklo-unit.com/about 13. Moore KA. (2017). Influence of the number and timing of malaria episodes during pregnancy on prematurity and small-for-gestational-age in an area of low transmission. BMC Med;15: 117. 14. Malaria diagnosis and case management guidelines, Thailand (2015). Nonthaburi: Department of Disease Control, Thai Ministry of Public Health; 2015. 15. Kitchakarn S, Lek D, Thol S, Hok C, Saejeng A, Huy R, Chinanonwait N,Thimasarn K , Wongsrichanalai C. (2017). Implementation of G6PD testing and primaquine for P. vivax radical cure: operational perspectives from Thailand and Cambodia. WHO South-East Asia J Public Health 2017;6(2): 60–68. 16. Malaria diagnosis and case management guidelines, Thailand, 2015. Nonthaburi: Department of Disease Control, Thai Ministry of Public Health. 17. CareStart™ G6PD RDT*. (n.d.). Retrieved from http:// www.accessbio.net/eng/products/products01_01.asp 18. Kim, E. T., Singh, K., Moran, A., Armbruster, D., & Kozuki, N. (2018). Obstetric ultrasound use in low and middle income countries: a narrative review. Reproductive Health;15:129. http://doi.org/10.1186/s12978-0180571-y 19. Devine A, Parmiter M, Chu CS, Bancone G, Nosten F, Price RN, et al. (2017) Using G6PD tests to enable the safe treatment of Plasmodium vivax infections with primaquine on the Thailand-Myanmar border: A cost-effectiveness analysis. PLoS Negl Trop Dis;11(5): e0005602. https://doi.org/10.1371/journal.pntd.0005602 20. M Cameron, J Gibson, K Helmers, S Lim, J Tressler, K Vaddanak.(2010) The value of statistical life and cost-benefit evaluations of landmine clearance in Cambodia. Environ Dev Econ;15:395-416 ●

uw-madison's only undergraduate STEM research & communication journal

The Journal of Undergraduate Science and Technology (JUST) is an interdisciplinary journal for the publication and dissemination of undergraduate research conducted at the University of Wisconsin-Madison. Encompassing all areas of research in science and technology, JUST aims to provide an open-access platform for undergraduates to share their research with the university and the Madison community at large. ___ On the Cover: Monarch Butterfly at the Biocore Prarie, photograph by Alder Levin. While the loss of North American prairies has hurt this species, significantly decreasing its population size, restoration projects like this one are providing plants such as milkweed to help rebuild their numbers.