Cornell iGEM

Wet Lab Overview

flavocide is fishPHARM’s biodegradeable peptide treatment for BCWD. To create our optimal treatment system, we had to first determine the most effective EcnB peptide against F. psychrophilum, the causative agent of BCWD. Multiple isoforms of the polypeptide toxin exist, each of which is produced by a different species of bacterium in nature. Zone of Inhibition trials and growth assays were then conducted to decide which species of bacterium produced the most potent EcnB isoform. A variety of organisms cause disease through their very presence in the host organism. The exploitation of toxins similar to the system developed by our team this year could pave the way for a microbiological treatment protocol for a plethora of other infectious diseases.

BioBricks

Our project relied heavily on the use of BioBricks. We successfully created 10 different BioBricks for each isoform of EcnB. We have also added our EcnB peptide downstream of stabilization proteins MBP and EDA to help stabilize EcnB production.

Our constructs have been summarized as follows:

ecnB-only series: These parts included a constitutive T7 promoter, a ribosome-binding site, the ecnB genes, a 6X Histidine tag to facilitate protein purification, and a terminator.

MBP-TEV-EcnB series: These parts included a fusion protein called Maltose Binding Protein (MBP) for additional stability. They have included a constitutive T7 promoter, a ribosome-binding site, the MBP gene, the TEV protease cutsite gene, the ecnB gene, a 6X Histidine tag to facilitate protein purification, and a terminator.

EDA-GSG series: These parts included a newly discovered fusion protein called KHG/KDPG adolase (EDA). Since EDA has never been characterized or BioBricked in iGEM’s history, we wanted to test it with a known endoglucanase called cel5a. The first BioBrick included a constitutive T7 promoter, a ribosome-binding site, the eda gene, the GSG linker sequence, BamHI/NdeI restriction sites for modularity, a 6X Histidine tag to facilitate protein purification, and a terminator. The second BioBrick included a cel5a gene at the modular site.

EcnA/EcnB: This BioBrick was created because ecnA is known to be the antidote for ecnB. When they are synthesized together, they are expected to offset the properties of each other. This part included a constitutive T7 promoter, a ribosome-binding site, the ecnA gene, the TEV protease cutsite gene, the ecnB gene, BamHI/NdeI restriction sites for modularity, a 6X Histidine tag to facilitate protein purification, and a terminator.

Chassis

BL21 is a cell strain commonly used with the T7 bacteriophage promoter system. In its chromosomal DNA is the T7 RNA polymerase gene, which can be regulated by arabinose induction and glucose inhibition of the araBAD promoter. This allows for efficient and high-level protein expression. Furthermore, the T7 Lysozyme gene in the pLysS plasmid is able to reduce basal expression by suppressing T7 RNA polymerase activity in uninduced cells [1].

EcnB Isoform & Strain List

Each of our BioBrick isoforms of the EcnB polypeptide toxin is naturally produced by a different species of bacterium as shown in the chart below.

BioBrick	Entericidin Bacterial Origin
BBa_K1595000	Ralstonia pickettii: Raslstonia pickettii is a Gram-negative, rod-shaped bacteria. This oligotroph is typically found in damp environments ranging from rivers to soils to biofilm on waterpipes. R. pickettii has the ability to sequester toxic metal ions and degrade harmful hydrocarbon molecules, and has consequently been of great interest in the bioremediation research community. There is no documentation regarding R. pickettii affecting healthy individuals, but if it contacted an individual with an already compromised immune system, it can infect the respiratory tract and bloodstream [2].
BBa_K1595001	Methyloversatilis universalis: M. universalis is a Gram-negative, facultative methylotroph. It has the ability to grow on single carbon compounds like methane and dimethylsulfide, and consequently have been of interest to greenhouse gas and environmental research. This particular strain was isolated from lake sediments [3].
BBa_K1595002	Xanthomonas arboricola: Xanthomonas arboricola is a gammaproteobacteria.These bacteria are responsible for causing the bacterial spot of stone fruit, a prevalent diseases amongst peach, plum, nectarine, apricot, and almond orchards. They are found worldwide, stretching across four continents. All strains of this bacteria are pathogenic and test positive for starch hydrolysis and quinate metabolism [4].
BBa_K1595003	Sphingobium yanoikuyae: Sphingobium yanoikuyae is a soil bacterium. It can also be isolated from clinical specimen and has unique abilities including degrading potent environmental pollutants, such as biphenyl, naphthalene, and phenanthrene. They have therefore been of interest for their bioremediation potential [5].
BBa_K1595004	Bordetella avium: Bordetella avium is a Gram-negative, motile, non-spore forming coccobacillus. It is found in patients with cystic fibrosis and have been known to causes bordetellosis (upper respiratory disease) in poultry. It preferentially binds to ciliated tracheal epithelial cells. The entericidin locus in B. avium is activated under high osmolarity and may function together with hydrolase to affect water concentration. Entericidin A and B are cell-envelope lipoproteins that also allow plasmids to be passed on to daughter cells [6].
BBa_K1595005	Azospirillum brasilense: Azospirillum brasilense is a soil bacteria that promotes plant growth via nitrogen fixation. Its motile flagellum is critical in attaching to plant roots in rhizosphere. Genome data suggests that this genus transitioned from water to land around the time of the emergence of vascular plants. Entericidin’s role in osmolarity may play a role in this transition [7].
BBa_K1595006	Escherichia coli: Escherichia coli are generally found in “animal feces, lower intestines of mammals, and even on the edge of hot springs”. This Gram-negative, rod-like bacterium produces the EcnB protein by RpoS (a sigma factor) regulation. The protein acts as a toxin to induce “programmed cell death of bacterial populations in stationary phase” [8].
BBa_K1595007	Enterobacter aerogenes: Enterobacter aerogenes is generally found in “soil, water, dairy products, and in the intestines of animals”. This highly motile, rod-shaped, Gram-negative bacterium is commonly found in respiratory, gastrointestinal, and urinary tract infections as well. It tends to be an opportunistic bacterium that infects a host whose immune system is already suppressed [9].
BBa_K1595008	Mannheimia haemolytica D174: Mannheimia haemolytica has been implicated in bovine respiratory disease. It is the primary cause of epizootic pneumonia in cattle, also commonly known as Shipping Fever. This bacterium can be found in the nasopharynx of hosts and enters the lungs when host defenses are weakened by stress or infection. M. haemonlytica can produce a cytotoxin known as leukotoxin which targets leukocytes, potentially contributing to its pathogenicity [10].
BBa_K1595009	Cedecea neteri: Cedecea neteri is a rare Gram-negative, motile, nonspore-forming bacillus. It has been associated with bacteremia and isolated from human clinical specimens [11].
BBa_K1595010	Klebsiella oxytoca: Klebsiella oxytoca is a Gram-negative, rod-shaped bacteria. It is commonly cultured from the healthy skin, mucous membranes, and intestines of healthy humans, but is also a pathogen. K. oxytoca has been found in patients with antibiotic-associated hemorrhagic colitis, urinary tract infections, and celiac disease [12].
BBa_K1595011	Thioclava : Thiocava is an aerobic and sulfur-oxidizing bacteria found in sulfidic hydrothermal regions. Thioclava are autotrophic and grow with thiosulfate as an energy source [13].
BBa_K1595012	Sinorhizobium meliloti: Sinorhizobium meliloti is a gram negative, nitrogen fixing bacterium. It exists symbiotically with legumes and works in the denitrification process. Enzymatic reactions and cellular processes change significantly when bacterium occupies root nodules of hosts. Proteins that play a significant role in the occupation process include osmoregulation proteins and potentially entericidin [14].
BBa_K1595013	Acinetobacter baumannii: Acinetobacter baumannii is a pleomorphic Gram-negative organism that is associated with nosocomial infections. It is often found in aquatic environments and is known for its antibiotic resistance [15].
BBa_K1595014	Rhodobacter capsulatus: Rhodobacter capsulatus is a purple, photosynthetic bacterium with a high capacity for “aerobic chemoautotrophic growth,” helping it to grow in a variety of conditions. It’s use of oxygen as a terminal electron acceptor allows for this ability. It is mainly found in freshwater and marine environments [16].
BBa_K1595015	Psychrobacter: Psychrobacter bacteria tends to live in very cold environments, including “Antarctic ice, soil, sediments” and the deep ocean [1]. This bacterium has been found in fish and meat products and, less commonly, in human tissue. It is a Gram-negative, non-motile bacterium [17].
BBa_K1595016	Agrobacterium: Agrobacterium is a Gram-negative, non-sporeforming, rod-shaped bacterium. It is known to cause gall disease and has been studied in its mechanism to cause tumors. Agrobacterium are usually found on root surfaces and infects wound sites in root tissues [18].
BBa_K1595017	Thalassospira: Thalassospira are Gram-negative, motile, spiral-shaped, chemoheterotrophic bacteria from the family Rhodospirllaceae and the class Alphaproteobacteria and are found in various marine environments. These environments include a variety of ecological niches, including surface and deep seawater, deep sediment, and halobios in the Pacific, Atlantic, Indian, and Arctic Oceans. They have shown potential for eliminating marine oil pollution, especially in polycyclic aromatic hydrocarbons degradation. [19]
BBa_K1595018	Erwinia: E. amylovora is a Gram-negative, rod shaped soil bacterium. It is known to be harmful to plants and is pathogenic for orchards including pears and apples [20].
BBa_K1595019	Lautropia mirabilis: First isolated in 1994, Lautropia mirabilis is a Gram-negative, motile, spherical bacterium. It has been observed in human oral and pulmonary sites, specifically in oral cavities of children with HIV.[21]

The multifarious sources of the Entericidins tested in our laboratory were extremely taxonomically diverse. Above is an evolutionary tree of the various Entericidin B isoforms tested. They originated form different bacterial species, ranging from Agrobacterium tumefaciens, a tumor-causing bacterium commonly used in genetic engineering to Acinetobacter baumanii, a pathogen of humans. The purpose of this diversity was to sample the widest range of Entericidin isoforms, to ensure that we found the most potent.

Flavobacterium Growth

Flavobacterium psychrophilum is a non-model organism that no iGEM team has worked with before. We thus needed to modify our current bacterial culture techniques in order to successfully work with this novel bacterium. We collaborated with Dr. Rod Getchell of the Aquatic Animal Health Lab at Cornell’s College of Veterinary Medicine in order to fully realize our goals in characterizing F. psychrophilum. Dr. Getchell provided us with two strains of F. psychrophilum isolates (strain 025 and strain 431), recovered from the kidneys of systemically infected Chinook and Coho salmon in the Great Lakes. Below is our characterized growth curve of F. psychrophilum strain 025 and 431 in liquid cytophaga broth without the presence of EcnB peptide.

Protein Stabilization

Because EcnB peptide has a relatively small size of approximately 5.3 kDa (or ~48 amino acids), it can be easily degraded within E. coli after inducing overexpression. To avoid this, we introduce the usage of fusion proteins for enhanced stability and yield.

The first fusion protein in question is with the maltose-binding protein (MBP) used in E. coli in the catabolism of maltodextrins. Having a MBP-fusion protein generally increases the solubility of proteins expressed in E. coli, though for us the primary usage is to artificially increase the size of the EcnB protein to avoid degradation. Once expressed and isolated from the system, it then becomes rather simple to recover the EcnB protein through the usage of a TEV protease cut site situated between MBP and ecnB. To this end, we have developed a BioBrick containing a constitutive T7 promoter, a ribosomal binding site, the MBP gene, the TEV protease cut site, the ecnB gene of interest, a 6xHis tag, and a terminator.

The second fusion protein includes the addition of the fusion expression partner KDPG aldolase (EDA), a novel solubility enhancer protein that has yet to be BioBricked in iGEM. Similar to MBP previously, EDA increases solubility of the chimeric proteins as well as limiting aggregation of the fusion partner.

Results

Expression Results

After cloning our genes into the BioBrick format, our first priority was to ensure our EcnB peptides expressed. We performed SDS-PAGE for BL21 as negative control, BBa_K1460001 as positive control, BBa_K1595001, BBa_K1595005, BBa_K1595005, BBa_K1595010, BBa_K1595013, BBa_K1595015, BBa_K1595008, BBa_K1595016, BBa_K1595018, and BBa_K1595024. Besides the sample from BBa_K1595015, there were no clear bands at 5kDa, where we would expect our protein. In addition, there was no apparent band at 47.5kDa where our EcnB-MBP construct, BBa_K1595024, was expected to be. Because we cannot confirm expression from our SDS-PAGE gels, we ran western blots. From our first western blot, we saw the expression of BBa_K1595006 and BBa_K1595016. From the second set, using newly expressed BBa_K1595006 as positive control, we saw the expression of BBa_K1595018 and BBa_K1595024.

Figure 1.1. SDS-PAGE for BL21, BBa_K1460001, BBa_K1595001, and BBa_K1595005. Protein expression is compared between pre-induction and post-induction samples. No noticeable difference was observed.

Figure 1.2. SDS-PAGE for BBa_K1595006, BBa_K1595010, BBa_K1595013, and BBa_K1595015. Similarly, there is no significant difference observed between pre-induced and post-induced samples. A dark band under post-induced BBa_K1595015 (10) suggests possible expression.

Figure 1.3. SDS-PAGE for BBa_K1595008, BBa_K1595016, BBa_K1595018, and BBa_K1595024. No significant difference was observed between pre-induction sample and post-induction samples.

Figure 1.4. Western Blot for BBa_K1595006, BBa_K1595008, and BBa_K1595016. Proteins from cell lysates were purified for SDS-PAGE, with gel samples taken at different stages of the purification. Western blot results show that BBa_K1595006 (1) and BBa_K1595016 (3) expressed successfully, but not BBa_K1595008 (2).

Figure 1.5. Western Blot for BBa_K1595006, BBa_K1595024, BBa_K1595015, and BBa_K1595018. Proteins from cell lysates were purified for SDS-PAGE, with gel samples taken at different stages of the purification, with BBa_K1595006 (1) acting as positive control. The uninduced BBa_K1595006 displayed leaky expression. Meanwhile, BBa_K1595024 (2) and BBa_K1595018 (4) showed considerable expression at around 50 kDa and 5 kDa, respectively. BBa_K1595015 (2) showed minimal to no expression.

Lysate Assay Initial results from the lysate assay displayed very promising results for the MBP+EcnB fusion protein construct, BBa_K1595024. Unfortunately, the EcnB constructs BBa_K1595006, BBa_K1595008, BBa_K1595018, and BBa_K1595015 did not display a significant change in OD600 when compared to the negative controls containing empty BL21 lysates and plain phosphate buffered saline (PBS) (Figures 2.2 - 2.5). But for BBa_K1595024, results display a greater than 50% drop in OD600 over a span of 12 hours when compared to the negative controls (Figure 2.1). Interestingly, all the constructs display a potential “acclimation period”, which appear to encompass the first four hours after inoculating the liquid cultures with the construct lysates. Furthermore, because of the highly promising results from the fusion protein construct BBa_K1595024, we decided to further study this effect by continuing with zone of inhibition testing.

Figure 2.1: Lysate Assay of BBa_K1595024
Normalized graph comparing measured OD600 over time for BBa_K1595024 as compared to several negative control benchmarks

Figure 2.2: Lysate Assay of BBa_K1595006
Normalized graph comparing measured OD600 over time for BBa_K1595006 as compared to several negative control benchmarks

Figure 2.3: Lysate Assay of BBa_K1595008
Normalized graph comparing measured OD600 over time for BBa_K1595008 as compared to several negative control benchmarks

Figure 2.4: Lysate Assay of BBa_K1595018
Normalized graph comparing measured OD600 over time for BBa_K1595018 as compared to several negative control benchmarks

Figure 2.5: Lysate Assay of BBa_K1595015
Normalized graph comparing measured OD600 over time for BBa_K1595015 as compared to several negative control benchmarks

Zone of Inhibition Data collected from the zone of inhibition assays display promising results for the effectiveness of the designed EcnB constructs (Figure 3). The lysates of EcnB constructs BBa_K1595006, BBa_K1595008, and BBa_K1595016 all showed very consistent fields of inhibition, with average inhibition diameters of 46.5mm, 33.71mm, and 30.2mm respectively. Comparatively, the negative controls containing BBa_K1460001, empty BL21, and ddH20 all displayed no zones of inhibition at all in the presence of F. psychrophilum, which was an expected result. Thus, we can conclude that these three tested EcnB constructs all are effective in combating F. psychrophilum growth due to their expression of the EcnB protein. Furthermore, the MBP+EcnB fusion protein construct, BBa_K1595024, performed exceptionally well against F. psychrophilum growth, generating an average zone of inhibition very similar in size to the average zone of inhibition from the known antibiotic, oxytetracycline. The BBa_K1595024 zone of inhibition had an average diameter of 50.7mm, as compared to 55.2mm from oxytetracycline’s zone of inhibition. A two-sample t-test shows no statistical difference between the two diameters (p > 0.05). From these results, it is rather clear that the fusion protein MBP+EcnB functions notably well in combating F. psychrophilum, with EcnB retaining its effectiveness despite being part of a fusion protein combination. Future work will include extensive zone of inhibition testing to other EcnB constructs, with a continual focus on determining the most effective isoform of EcnB.

Figure 3: Histogram of Zone of Inhibition Results.
Zone of inhibition measurements for two negative controls, empty BL21 and BBa_K1460001, the positive control, oxytetracycline, and the EcnB constructs, BBa_K1595006, BBa_K1595008, BBa_K1595016, BBa_K1595024.

Future Work

The Cornell iGEM wet lab subteam is committed to discovering the next generation of medicine for fish. Our goal is to provide our customers the most effective and cost-efficient form of flavocide. To accomplish this, we seek to perform a series of medical trials to further test our product's efficacy and we aim to streamline the manufacturing process for large-scaled production.

From the Zone of Inhibition (ZOI) assays, we have preliminary data for comparing the relative strengths of our EcnB isoforms. To ensure our medicine also provides a long lasting form of protection, we are in the process of testing the functional stability of each isoform. We aim to compare their extracellular degradation rates and hope to lengthen functional lifetimes by successfully linking them downstream of stable fusion proteins like EDA and MBP.

In addition, we understand that fish are dynamic systems. To account for these additional levels of complexity, we seek to develop a proposal for safely and ethically testing the efficacy of flavocide in live salmonids. First, we hope to test for the effective EcnB concentration range against Flavobacterium through ZOI assays. With these results, we could use our drug delivery model to predict the required effective EcnB concentration range in the salmonid bloodstream. This provides a reasonable magnitude for drugs to test in later experimental trials. With this information, we can develop an informed proposal in accordance with the IACUC regulation and code of conduct. Although we don't expect bioaccumulation and toxicity in the fish system, we would like to confirm and verify that our proposed isoform falls within this accepted assumption.

Once our chosen isoform shows progress against bacterial coldwater disease, we can optimize the manufacturing process for large scaled production. This would include characterizing bacterial growth and expression for crucial parameters to minimize time and material expenses while maximizing flavocide yields.

By continuing to advance wet lab on all fronts, we hope to provide fisheries with an even more comprehensive solution against BCWD in the near future.

References

[1] Garcia, L., & Molineux, I. (1995). Rate of translocation of bacteriophage T7 DNA across the membranes of Escherichia coli. Journal of Bacteriology, 177(14), 4066-4076.

[2] Stelzmueller, I., Biebl, M., Wiesmayr, S., Eller, M., Hoeller, E., Fille, M., Weiss, G., Lass-Floerl, C. and Bonatti, H. (2006), Ralstonia pickettii—innocent bystander or a potential threat?. Clinical Microbiology and Infection, 12: 99–101.

[3] Kittichotirat, W., Good, N., Hall, R., Bringel, F., Lajus, A., Medigue, C., . . . Kalyuzhnaya, M. (2011). Genome Sequence of Methyloversatilis universalis FAM5T, a Methylotrophic Representative of the Order Rhodocyclales. Journal of Bacteriology, 193(17), 4541-4542. doi:10.1128/JB.05331-11

[4] Boudon, S., Manceau, C., & Nottéghem, J. (2005). Structure and Origin of Xanthomonas arboricola pv. pruni Populations Causing Bacterial Spot of Stone Fruit Trees in Western Europe. Phytopathology, 95(9), 1081-1088.

[5] Gai, Z., Wang, X., Tang, H., Tai, C., Tao, F., Wu, G., & Xu, P. (2011). Genome Sequence of Sphingobium yanoikuyae XLDN2-5, an Efficient Carbazole-Degrading Strain. Journal of Bacteriology, 193(22), 6404-6405. doi:10.1128/JB.06050-11

[6] Kersters, K., Hinz, K., Hertle, A., Segers, P., Lievens, A., Siegmann, O., & Ley, J. (1984). Bordetella avium sp. nov., Isolated from the Respiratory Tracts of Turkeys and Other Birds. International Journal of Systematic Bacteriology, 34(1), 56-70. doi:10.1099/00207713-34-1-56

[7] Holguin, G., Patten, C., & Glick, B. (1999). Genetics and molecular biology of Azospirillum. Biology and Fertility of Soils, 29(1), 10-23. doi:10.1007/s003740050519

[8] Stehr-Green, J. K., Centers for Disease Control and Prevention, & National Institutes of Health. (2000). Foodborne disease outbreak investigation: epidemiologic case studies. In Foodborne disease outbreak investigation: epidemiologic case studies. Department of Health & Human Services.

[9] Enterobacter aerogenes. (2011, April 22). Retrieved August 1, 2015, from https://microbewiki.kenyon.edu/index.php/Enterobacter_aerogenes

[10] Rice, J., Carrasco-Medina, L., Hodgins, D., & Shewen, P. (2007). Mannheimia haemolytica– and Pasteurella multocida–Induced Bovine Pneumonia. Food Animal Practice, 8(2), 117-28. doi:10.1017/S1466252307001375 [19] Mannheimia haemolytica. (2012, July 18). Retrieved August, 2015, from https://en.wikivet.net/Mannheimia_haemolytica

[11] Farmer, J.J., Sheth, N., Hudzinski, J., Rose, Harold. Asbury, M. (1982). Bacteremia due to Leptotrichia trevisanii sp. nov. European Journal of Clinical Microbiology & Infectious Diseases, 16(4), 775-778. Retrieved August. 2015, from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC272471/

[12] Darby, A., Lertpiriyapong, K., Sarkar, U., Seneviratne, U., Park, D., Gamazon, E., . . . Fox, J. (2014). Cytotoxic and Pathogenic Properties of Klebsiella oxytoca Isolated from Laboratory Animals. PLoS ONE. doi:10.1371/journal.pone.0100542

[13] Sorokin, D. (2005). Thioclava pacifica gen. nov., sp. nov., a novel facultatively autotrophic, marine, sulfur-oxidizing bacterium from a near-shore sulfidic hydrothermal area. International Journal Of Systematic And Evolutionary Microbiology, 1069-1075. Retrieved August 1, 2015, from http://www.ncbi.nlm.nih.gov/pubmed/15879235

[14] Escherichia coli. (2014, November 13). Retrieved September 15, 2015, from https://microbewiki.kenyon.edu/index.php/Escherichia_coli

[15] Howard, A., O’Donoghue, M., Feeney, A., & Sleator, R. (2012, May 1). Acinetobacter baumannii: An emerging opportunistic pathogen. Retrieved August 1, 2015. [ ] Rice, L. (2008). Federal Funding for the Study of Antimicrobial Resistance in Nosocomial Pathogens: No ESKAPE. The Journal of Infectious Diseases J INFECT DIS, 197(8), 1079-1081.

[16] Escherichia coli. (2015). Retrieved August 1, 2015, from https://microbewiki.kenyon.edu/index.php/Escherichia_coli

[17] Psychrobacter. (2015). Retrieved September 15, 2015, from https://microbewiki.kenyon.edu/index.php/Psychrobacter

[18] Van Haute, G. (2003, August 1). Agrobacterium tumefaciens. Retrieved August 1, 2015, from http://users.skynet.be/albert.de.koning/agrobacterium.pdf.

[19] Lai, Q., Liu, Y., Yuan, J., Du, J., Wang, L., Sun, F., & Shao, Z. (2014). Multilocus Sequence Analysis for Assessment of Phylogenetic Diversity and Biogeography in Thalassospira Bacteria from Diverse Marine Environments. Third Institute of Oceanography State Oceanic Administration, 9(9), 1-11. doi:e106353

[20] Johnson, K. (2015). Fire blight of apple and pear. Retrieved August 1, 2015, from http://www.apsnet.org/edcenter/intropp/lessons/prokaryotes/Pages/FireBlight.aspx

[21] Rossmann, S., Wilson, P., Hicks, J., Carter, B., Cron, S., Simon, C., . . . Kline, M. (1998, June 1). Isolation of Lautropia mirabilis from Oral Cavities of Human Immunodeficiency Virus-Infected Children. Retrieved September 15, 2015.

B07 Weill Hall, Ithaca, NY

Team:Cornell/wetlab