Difference between revisions of "Team:UC Davis/Design"

POLICY & PRACTICES	PROJECT	ACHIEVEMENTS	THE TEAM	SAFETY & PROTOCOLS

Project Overview:

Human practice issues motivated our project and have informed its design every step of the way. We realized early on that if we wanted to promote responsible chemical use, we had to go about the issue responsibly ourselves. This meant making a concerted effort to understand the complexities of the issue, learning from the past and present in developing a viable solution for the future, and assessing the implications of our proposed solution.

After identifying public awareness as an effective catalyst for change, we researched civic engagement initiatives and identified that successful civic engagement measures fostered a sense of individual and collective responsibility and provided a sense of agency that's grounded in reality. We were particularly inspired by Greg Neimeyer’s Black Cloud initiative, which effectively inspired a new cohort of environmental stewards at L.A.'s Manual Arts High School. As a result, we identified high school students as a potential audience and a lesson plan as a viable medium for promoting responsible chemical use. To instill a sense of agency, we wanted to engage students in a hands on activity; however, current methods of detection are not well suited for a classroom environment. GCMS has a cost prohibitive capital cost and even ELISAs at $7.50/sample begin to add up for a class of thirty students. What we needed was a tool that could rival current methods of detection in both sensitivity and cost. This need motivated our wetlab effort in developing a reliable and cost effective triclosan biosensor that could be used in conjunction with our lesson plan to raise awareness and accountability around triclosan use.

We then took a step back to assess the possible implications of our lesson plan. We harkened back to Bruce Hammock’s warning about playing into consumer fears and Arlene Blum’s sentiment about the futile cycle of replacing one toxic chemical with another. Cognizant of these two points, we developed an “antimicrobial awareness app” to extend our solution more broadly.

In summary, we realized that in order to effectively engage students in the dialogue regarding chemical use, we needed to develop a biosensor that students could use to monitor environmental levels of triclosan (nanomolar levels). Given school budget constraints we needed to drive the cost of our system (both capital cost and per sample) down. We needed a user friendly device. And lastly, we wanted to extend the scope of our solution to more than just one chemical and more than just students.

To achieve our goals, we pursued the following three measures:
1. Lesson Plan
2. Biosensor
3. Antimicrobial Footprint App
*Click through to jump to the section!

*Continue scrolling to read about our design process or jump to the lesson plan material! Motivation Our motivation for developing a lesson plan to be used in conjunction with our biosensor was two-fold: (1) to engage students in STEM by demonstrating that biology and chemistry have real world applications and (2) to raise awareness and accountability around chemical use by enlisting the help of students to monitor environmental levels of triclosan. Development To aid in our lesson plan development, we reached out to Community Resources for Science (CRS), an organization that works with scientists to bring their work into the classroom and engage students in hands on, inquiry based learning experiences. Sasha Stackhouse and Morgan Seag at CRS provided valuable feedback on how to present our project more clearly and how to fit our lesson plan to Next Generation Science Standards. Once we got the approval that our lesson plan met their standards, we connected with CRS’ network of teachers in our area -- just a portion of their network of 1,400 teachers. While we are excited to share our lesson with schools in the area, delivering the instructions ourselves somewhat limits our scope. Ultimately we want to reach students around the nation. To better understand how we could refine our lesson plan to allow for wider distribution, we spoke with Ann Moriarty, an AP Biology/biotechnology teacher at Davis Senior High School. Ms. Moriarty has a background in research herself and propounds the value of hands on activities in engaging students. In her classroom, for example, students develop an appreciation for biotechnology through running protein assays and performing gel electrophoresis to verify that their restriction enzymes cut the plasmid in the right location. Ms. Moriarty provided valuable insight on the considerations that go into developing a lesson plan- the most salient being constraints imposed by lack of funding and needing to teach to a curriculum. In regards to budget constraints Mr. Moriarty talked about the difficulty in planning field trips due to lack of funding. In explaining the first iteration of the lesson plan, which involved taking a field trip out to a local body of water where students would collect a sample for measurement, Ms. Moriarty raised some valid concerns. For example: what happens to students who don't want to go on field trip? Who is going to cover the cost of a substitute teacher, transportation costs, insurance? In sum, field trips are a logistical nightmare. So we brainstormed a work-around. Why not have students go out and collect water samples from their local environment as a pre-activity assignment? Not only did this circumvent the difficulty of organizing a field trip, this also gave students a sense of ownership - a sense of individual responsibility that we had previously identified as an important factor in a successful civic engagement initiative. The second obstacle Ms. Moriarty mentioned are the constraints imposed by teaching to a curriculum. However we identified the following AP Biology standards that the lesson plan could fulfill, allowing our lesson plan to be more easily integrated into an AP Biology curriculum: Essential Knowledge: 4.B.1 - Interactions between molecules affect their structure and function Science Practice: 5.1 - The student can analyze data to identify patterns or relationships Learning Objective: 4.17 - The student is able to analyze data to identify how molecular interactions affect structure and function In regards to improving our lesson plan for wider acceptance, Ms. Moriarty made the following four suggestions: Providing a teacher’s guide with detailed background information about the project because teachers have to fully understand the project before they can teach it, Providing an outline of how to carry out the lesson with suggested prompts and a sense of timing, Providing a deck of powerpoint slides to show to the class, and Providing assessment questions/worksheets Through collaborating with CRS and talking with Ms. Ann Moriarty, we gained an appreciation for the depth of thought required in developing an engaging and implementable lesson plan. This appreciation was deepened through our conversation with Dr. Chris Pagliarulo, Associate Director of Instruction & Assessment at the Office of Undergraduate Education at UC Davis. Dr. Pagliarulo talked about how designing an educational activity is a process by which you identify objectives and criteria for success, come up with a prototype, test it, evaluate it based on criteria, and iteratively refine it. This sounded just like the engineering design cycle! We were engineering an educational plan. Lesson Plan Engineering As in any other engineering process, Dr. Pagliarulo highlighted the importance of defining lesson plan objectives (design goals or objectives) to more easily assess whether the lesson plan achieved its goal. We identified several design objectives for the lesson plan: The lesson plan must help an instructor teach or reinforce one or more of the Next Generation Science Standards and/or meet address some learning goals from an Advanced Placement topic in STEM. This was a necessary requirement to convince teachers to allow the activity into an already crowded class schedule. The lesson plan should shift student appreciation and perception of: the interaction between science, technology and the use of technologies. The experience should show some influence on “long term” behavior or attitude towards the responsible use of chemicals. Assessments are Critical Dr. Pagliarulo also made it clear that we absolutely needed to create assessments to evaluate the success of our lesson objectives. He gave us some insight: crafting “layered” questions that assess different levels on Bloom’s taxonomy of cognitive understanding can be useful for probing what students understand and where they are struggling: for example low Bloom’s level questions might test memorization and vocabulary while some higher level Bloom’s questions might test a student’s ability to apply the core lessons to new situations different objectives might require different types of assessments: in our case assessing whether students have different two different types of objectives. (1) we are assessing the ability to teach certain concepts associated with science standards and (2) we are assessing student attitudes. These require different assessment instruments. assessment instruments themselves need to be assessed [REF #.1 page 191]: for instance, if a question is confusing to many students it may not be testing the knowledge or skill that we’re interested in assessing but rather the student’s ability to decipher the questions. A good educational plan has a plan for assessing the quality of the assessments too. Dr. Pagliarulo mentioned that one way to measure the outcomes of a lesson plan is to develop what is called a “pre post test”. These are sets of questions that are given to students before (pre) and after (post) and activity to assess any learning gains. We decided to create two sets of assessment instruments: 1. Standard multiple choice questions that an instructor could use to assess state standards 2. A survey for assessing student attitudes towards responsible chemical use and their role in this process. The survey we developed was based on pre-existing and validated assessment instruments on student attitudes towards science [REF #.2] Note: Validating the assessment instruments is apparently a long process and would require us to run the lesson and collect data numerous times and to analyze the results. Validation of the multiple choice questions would require us to examine whether certain answer choices are “good” or “poor” distractors, answer choices designed to test common misconceptions, and common statistics like the “facility factor” and “discrimination factor” that test how easy or hard a particular question is and how well a question help to discriminate different skill levels in a class, respectively. Validation of the survey instrument would require us to assess statistics known as “validity” and “reliability” [REF #.3]. Validity basically refers to whether or not the survey is measuring the concepts we think it’s measuring while reliability assesses how consistently the survey produces valid results. Dr. Pagliarulo has offered to help with this process if we can collect enough data. Putting It All Together The insights gained from these numerous conversations informed the shape and development of our lesson plan at multiple levels, including the teaching resources that are required to use it in the classroom. These include a teacher's guide, classroom powerpoint, and assessment questions -- all of which can be downloaded below: Click here to download our lesson plan Click here to download our lesson plan assessment instruments Click here to download our instructional power-point as a PDF or view as a presentation Sources: Ref #.1 J. W. Pellegrino, M. R. Wilson, J. A. Koenig, A. S. Beatty, Developing assessments for the next generation science standards (National Academies Press, 2014). Ref #.2 Russell, J. & Hollander, S. (1975). A biology attitude scale. The American Biology Teacher, 37 (5), 270-273. via (http://www.flaguide.org/tools/attitude/biology_attitude_scale.php) Ref #.3 Alwin, DF & Krosnick JA. 1991. The reliability of survey attitude measurement the influence of question and respondent attributes Sociological Methods & Research 20, 139-81. Our goal for the wet lab portion of the project was to develop an inexpensive triclosan biosensor for use in a high school laboratory setting. To achieve our goal we first needed to find a molecule that interacted specifically with triclosan and whose interaction with triclosan could be coupled to a measurable readout. In researching the biology of triclosan, we discovered that its natural target fit the above requirements! Triclosan’s natural target, the FabI enzyme, is directly inhibited by triclosan and its activity also depends on a commonly measured co-factor, NADH. It was unclear however which of the thousands of FabI enzymes we should use for our biosensor since all Bacteria and Eukaryotes have a FabI gene. We decided to use our device’s requirements to define our enzyme selection criterion. The chosen FabI enzyme would ideally: over-express well in E.coli, show activity on its respective substrate at <2nM level of enzyme, and be inhibited by triclosan at the >2nM level of inhibitor (the higher level of TCS found in treated wastewater and toxic to algae).[18] In addition, the commercially available native substrate for FabI (crotonyl-CoA) is unstable and costly. An alternative substrate was thus necessary for our biosensor to be used in a practical setting. To achieve our goal of developing a cost effective triclosan biosensor then, we were tasked with (a) discovering an enzyme that met all of our criteria and (b) identifying an alternative substrate that would enable practical use. Our wet lab section is organized into five subsections: 1. Enzyme Selection 2. Chemical Biology Substrate Screening 3. Enzyme Engineering 4. Prototyping in Real World Waste Water 5. Future Directions *Click through to jump to the section!

        ↥

The Need: A FabI enzyme that can show nanomolar inhibition using triclosan (Why nanomolar inhibition? Work done by Chalew and Halden showed that levels of triclosan leaving waste-water treatment plants was up to 9 nM, which happens to also be the toxicity threshold level for algae. [18]) Strategy 1a: Scan the registry to find characterized FabIs with inhibition data We first scanned through the iGEM parts registry for existing BioBrick Parts that code for FabI. We found Bba_K771303 from the 2012 Shanghai Jiao Tong University iGEM team, however we were unable to find enzymatic characterization data on the part. We therefore proceeded with our literature search for FabI enzymes with inhibition data. Strategy 1b: Alternate candidates were found by mining the literature to find characterized FabIs with inhibition data Triclosan inhibits type 2 fatty acid synthesis (FASII), an essential pathway in the Bacterial and Eukaryotic domains by interacting directly with the enoyl acyl carrier protein reductase (FabI) [3]. Evidence that we could use the enzyme to detect triclosan came from binding studies and crystallographic data initially from Heath et al. They showed that triclosan binding increases the enzyme’s affinity for NAD+ and triclosan’s role as an effective inhibitor is due to the formation of a stable ternary complex between FabI, triclosan, and NAD+ [4]. The basis of our biosensor, therefore, is to use triclosan’s mechanism of action as an inhibitor of enoyl acyl carrier protein reductase (FabI) in order to detect it. To screen a representative subset of FabI’s from all available domains of life, we mined the literature and found every characterized FabI with published inhibition data. We found all living organisms except for the Archaea, who synthesize lipids based on isoprenoids, have a fabI gene [5]. Unlike the other enzymes of the FASII pathway, it’s very important to note there is considerable diversity in the structure of FabI’s from different organisms[5]. AND, not all of them have the same level of sensitivity towards triclosan[23]. This is why we needed to screen a panel of enzymes in order to find the enzyme most sensitive towards triclosan with a nanomolar inhibition constant. (See above for why we wanted to see nanomolar inhibition) Reported triclosan inhibition constants: Terms: Ki is the dissociation constant of the enzyme- inhibitor complex. IC50 is the amount of an inhibitor needed to inhibit a biological process by 50% From our literature search, we were able to find enzymatic activity and triclosan inhibition data on the E. coli FabI. We also found 2 other FabI proteins that had at least nanomolar triclosan inhibition kinetics to screen: S. aureus and P. falciparum. Even though, H. influenzae FabI does not have reported nanomolar inhibition kinetics, we still wanted to verify the literature value so we included it in our set. We also acknowledge that A. thaliana has nanomolar triclosan inhibition, but we didn’t learn about its inhibition kinetics until we were near the end of our project Of the organisms studied, algae is most sensitive to triclosan, where toxicity levels are roughly 9 nanomolar which are the same concentrations reported by Chalew and Halden of triclosan leaving Wastewater Treatment Plants (WWTPs)[18]. From our literature search, we noticed that there were no characterized FabI’s from algae, nor is it even known whether triclosan exerts its effects on algae by inhibiting its FabI enzyme [19]. Due to their extreme sensitivity, we hypothesized that triclosan exerts its effects on algae through inhibition of the FabI enzyme. Therefore we decided to test 3 FabIs from algae: P. tricornutum, T. pseudonana, and A. protothecoides. This left us with a candidate list of 7 FabI proteins. 1 previously submitted FabI gene in the registry (Bba_K771303), 3 FabI proteins taken from the literature, and 3 FabI proteins from algae. We decided to codon optimize and synthesize these genes to clone into a common E. coli expression vector pET29b+. A nucleotide alignment showing the codon optimizations to Bba_K771303 is shown below. We screened each enzyme based on three criteria: The enzymes had to over-express well in E. coli, They had to have activity at the nanomolar level, And they had to be inhibited by nanomolar levels of triclosan, the concentration leaving wastewater plants [18] 1. Overexpression Protein purity assessed through SDS-PAGE gels shown below: Protein concentrations were measured spectrophotometrically from absorbance at 280 nm (A280). A280 readings converted enzyme concentration to mg/ml. The molarity of each enzyme was determined by dividing mg/ml by the enzyme’s extinction coefficient, theoretically derived from each enzyme’s amino acid sequence from http://web.expasy.org/protparam/ Extinction Coefficients used: influenzae 17,420 falciparum 46,885 protothecoides 34,380 tricornutum 37,360 coli 15,930 B. pseudomallei and T. pseudonana did not express under our conditions, so they were eliminated from our FabI team. 2. Enzyme Activity Screening Enzyme activity was measured spectrophotometrically through the decrease in NADH absorbance over time on the native substrate analog crotonyl CoA. Activity is defined as the change in optical density (absorbance) per minute. Activity is normalized by dividing activity by the microgram of enzyme used for the assay. Each enzyme was assayed with 100 uM NADH and 100 uM crotonyl-CoA. Negative control was 100 uM NADH, 100 uM crotonyl-CoA, no enzyme. Observed enzyme activities were subtracted from negative control and plotted on a log scale. Two biological replicates of each enzyme was used. S. aureus FabI showed no activity even though the enzyme has been previously characterized [11][12]. We later found out we didn’t see activity because S. aureus FabI is NADPH dependent rather than NADH dependent! Notwithstanding, NADPH is significantly more expensive than NADH[24], so instead of accommodating for S. aureus FabI, we decided to remove it from the FabI team. 3. Triclosan Inhibition Screening We were down to the "Fab 5". Since Chalew et al showed the levels of triclosan leaving Waste Water Treatment Plants (WWTPs) was up to 9 nanomolar [18], so we wanted to measure enzyme inhibition using a nanomolar level of triclosan. Under our conditions, however, not all of the fab 5 had measurable activity with a nanomolar amount of enzyme, and in order to see inhibition using a nanomolar amount of triclosan we needed to use a nanomolar amount of enzyme. Triclosan inhibition was measured by running our standard enzyme activity assay with no triclosan and 1 nM triclosan. Negative control was 100 uM NADH, 100 uM crotonyl CoA, no enzyme, no triclosan. Observed enzyme activities were subtracted from negative control activities. Percent inhibition was calculated by: ( (uninhibited activity - inhibited activity) / uninhibited activity ) * 100 The enzyme concentrations ranged from 1.9 - 3.3 nM using two biological replicates of each enzyme. Nanomolar inhibition from P. falciparum has been previously reported [8], but we are the first to show triclosan inhibition using A. protothecoides Fabi! Interestingly, with P. tricornutum, a marine diatom, which have been shown to be sensitive towards triclosan [29] we were unable to see triclosan inhibition, which suggests there is a different biological mechanism of action for triclosan inhibition. However, nanomolar inhibition was also not observed with H. influenza, and E.coli, even though previously reported [6][7][10]. Therefore more detailed studies on conditional dependencies of inhibition are needed in order to elucidate the mechanism of action. However, the A. protothecoides inhibition data clearly indicates that triclosan effects FabI from algae and that this could be the biological mechanism of toxicity. For our biosensor, P. falciparum FabI, being the enzyme most inhibited by triclosan, appears to be the best enzyme to use for our biosensor! Additionally, we further characterized the previously submitted E. coli FabI biobrick Bba_K771303. We have added our characterization data to the experience section on the parts registry!

• Confirm inhibition data in waste water correlates to known levels of triclosan in a wide variety of waste water samples
• Continue rounds of enzyme engineering to enhance another 60-fold (~10 more 1.5 folds… or 1 60-fold)
• Forward predictions based on unknown samples where the biosensor, ELISA, and MS are used in parallel