| Background and experimental system

Proper transcription of genes is fundamental to the viability of all organisms and represents the primary point for controlling gene expression in a cell. In bacteria, transcription is catalyzed by an RNA polymerase (RNAP) comprised of four core subunits (two alpha, beta, and beta prime) and a dissociable subunit known as a sigma moters; they are located upstream of the transcriptional start site (designated as +1): the 35 element (consensus sequence TTGACA) and the 10 element (consensus sequence TATAAT). 2 Typically separated by 17 base pairs, these elements are recognized by two different domains of the sigma subunit. 1 Additional promoter elements make contacts with the core RNAP subunits, and the relative contributions of each promoter element to transcriptional control varies from promoter to promoter. 2 After RNAP recognizes and binds promoter factor that helps the polymerase recognize specific DNA, the DNA locally melts near the start site of transequences in the promoter DNA. The primary sigma factor in bacteria is sigma70, 1 which is used in the experi-scription to create an open region known as the transcription bubble. Using nucleotide triphosphates (NTPs), ments described here. There are two primary core the polymerase initiates RNA synthesis at +1. RNAP will continue to add nucleotides and elongate the RNA until it encounters termination signals at the end of the gene. 3 The in vitro experiment described here focuses on two steps in the process of transcription: binding of RNAP to the promoter DNA, and the transcriptional activity of the RNAP (i.e., synthesis of RNA). These two steps are probed using a unique fluorescence-based elec-circle). These are annealed to the template strand oligo, which contains a Cy5 fluorophore on its 5 0 end. Fluorescence emission from the Cy5 is quenched in the annealed construct (designated by the gray circle). When this promoter is transcribed by RNAP the quencher oligo is released due to the nick in the backbone of the nontemplate strand, thereby revealing fluorescent signal from trophoretic mobility shift assay (EMSA) that uses Escheri-the Cy5 dye on the template strand. Thus, NTPchia coli RNAP and promoter DNA, both commercially available. The EMSA, or gel shift, monitors the migration of fluorescently labeled promoter DNA through a native gel under a current in the absence and presence of RNAP. The size of the RNAP/DNA complex is larger than the DNA alone; therefore, the protein-bound DNA migrates more slowly through the gel. The gel matrix creates a "caging effect" to prevent dissociated RNAP and DNA from diffusing away from one another before rebinding occurs. The Supporting Information contains literature and video resources that discuss and demonstrate the theory and applications of EMSAs.

In addition to marking the position of the promoter DNA in the native gel, fluorescence is also used to determine whether transcription of the promoter DNA has occurred, using the approach diagrammed in Figure 1. The start site of transcription (+1) is designated with an arrow showing the direction of transcription. The positions of the 35 and 10 elements are indicated by black boxes. The nontemplate strand (top strand) consists of two oligonucleotides, resulting in a nick between bases dependent appearance of Cy5 emission (corresponding to release of the quencher oligo) is a signature for the RNA synthesis step of transcription. Note also that the first cytosines in the template strand are located at positions +36 and +37. Therefore, when transcription occurs in the absence of GTP, RNAP will stably pause at +36, thus keeping the polymerase bound to the template DNA.

By using this promoter in EMSAs with RNAP, students detect formation of the RNAP/DNA complex by monitoring the change in migration of the Cy3 signal, and detect transcriptional activity by monitoring NTPdependent appearance of the Cy5 fluorescent signal. As an inquiry-driven component of the experiment, students add the antibiotic rifampicin to reactions and determine which step in transcription it inhibits by evaluating whether rifampicin affects the Cy3 or the Cy5 signals. Their data will reveal that rifampicin does not inhibit DNA binding by RNAP, but rather the transcriptional activity of the polymerase. Depending on the curriculum, this experiment could be expanded to encompass additional concepts, as described in more detail below. +20 and +21. The upstream oligo contains a Cy3 fluorophore on its 5 0 end (green circle) and the downstream oligo contains quencher molecule on its 3 0 end (black The primary student learning goals are to: (1) Understand that promoter binding by RNAP and transcriptional activity are two experimentally distinguishable steps in transcription; (2) Understand that inhibitors of a biological process, such as transcription, typically target specific steps within that process; (3) Understand the theory and practice of using EMSAs to detect protein/DNA interactions. Achievement of the learning goals is assessed by the students' ability to provide clear, written results and data interpretations in their laboratory notebooks. This includes answering post-lab analysis questions (see Supporting Information). In doing so, students demonstrate:

(1) The ability to interpret which fluorescent signals in the EMSA represent formation of the RNAP/promoter complex and which signals represent transcriptional activity, (2) The ability to correctly interpret EMSA data to conclude how rifampicin inhibits transcription, and (3) The ability to validate their conclusions regarding rifampicin using data from the scientific literature. Instructors can emphasize additional concepts depending on the curriculum goals of individual courses. As examples, the theory behind fluorescence and quenching could be emphasized, or the central dogma and its relationship to medicine could be discussed (rifampicin is an antibi-needs, set-up instructions with buffer and gel recipes, tips to avoid common student pitfalls, resources for the theory of EMSAs, as well as pre-lab and post-lab questions for students. otic), or an analysis of structure/function could be emphasized using structures of rifampicin bound to RNAP as a guide.

| EX PERIMENTAL PR OCEDURE S

There are many ways in which this experimental system could be expanded to meet additional learning objec-2.1 | Protein (RNAP) and promoter DNA tives or to mimic a research environment more closely. For example, students could investigate how changing the sequence and/or spacing of bacterial promoter elements impacts RNAP binding and activity. 4,5 They could The E. coli RNAP holoenzyme that contains the sigma70 subunit was purchased from New England Biolabs. The oligos to construct the three-piece promoter DNA were evaluate the binding affinity of RNAP for the promoter purchased from Integrated DNA Technologies with DNA, or the potency of rifampicin inhibition by making quantitative measurements of the K D and IC 50 , respectively. 6,7 Questions about kinetic stability could be evaluated by measuring the dissociation rate constant for the RNAP/DNA complex under different salt concentrations. 8 Complementary in vivo experiments could be performed to evaluate inhibition of bacterial growth with rifampicin. 9 Finally, we have used this quencher-release technique to investigate mechanisms of transcription by human RNAP II, using an EMSA as well as singlemolecule microscopy. The research paper that describes these studies provides a nice "real-world" complement to sequences and modifications as follows. The long nontemplate strand oligo contained a Cy3 fluorophore at the 5 0 end and had the sequence 5 0 -GCTGCTCGATTAGGCTT-GACACTTTATGCTTCGGCTCGTATAATGTGTGGAT-CACTATCTTTAATCACTA-3 0 . The 35 and 10 elements are underlined, and the start site at + 1 is in bold. The short nontemplate oligo contained the Iowa Black RQ-Sp quencher at the 3 0 end and had the sequence 5 0 -CTCACACTAACCTCAGG-3 0 . The template strand oligo contained a Cy5 fluorophore at the 5 0 end and had the sequence 5 0 -CCTGAGGTTAGTGTGAGT AGTGATTAAAGATAGTGATCCACACATTATACGAG the classroom experiment, which students could discuss. 10 CCGAAGCATAAAGTGTCAAGCCTAATCGAGCAGC-3 0 . To construct the promoter DNA the three oligos were The experiment described is unique within the current education literature. Only a handful of other laboratory experiments 11,12 or problem-solving exercises 13,14 describe EMSAs. This experiment provides an experimental framework for students to explore the mechanism of transcription by distinguishing between the binding of RNAP and its transcriptional activity. This complements, but is distinct, from other experiments in the education literature that probe transcription by evaluating changes in the expression level of reporter genes such as betagalactosidase, [15][16][17] or evaluating levels of cellular RNA via RT-PCR or arrays, for example. [18][19][20][21] annealed by mixing 100 pmol of each fluorescent oligo with 1 nmol of the quencher oligo (the excess ensured there would be no unquenched template strand oligo) in a final volume of 50 μl of 1X annealing buffer (20 mM Tris pH 7.9, 2 mM MgCl 2 , 50 mM KCl). The annealing mixture was heated to 95C then slowly cooled to room temperature by removing the heating block and placing it on the benchtop. The 2 μM stock solution of promoter DNA was stored in small aliquots at 80C in dark tubes or tubes wrapped in foil to pro-tect from light. This experiment is appropriate for students in upperlevel undergraduate biochemistry or molecular biology 2.2 | Native gel laboratory courses. It is not necessary that students have been introduced to EMSAs or fluorescence, although it is helpful if they have been introduced to transcription and the central dogma in classroom coursework. From a technical perspective, micropipette skills are important for the success of this experiment. This experiment encompasses one 4-5 h laboratory session. The number of students per team/lab group is flexible. The experiment uses a purchased protein and purchased oligos, making it easily accessible from a reagent perspective. The Supporting Information contains lists of reagent and equipment

The native gel was prepared before assembling the experimental samples. Precast native gels are commercially available (BioRad, Thermofisher, for example); however, we have not tested their use in these assays. Low fluorescence glass gel plates (20 cm 22 cm) with 1.5 mm spacers were clamped or taped together. A mixture (60 ml) containing 5% acrylamide (37.5:1 acrylamide:bis ratio), 0.5X TBE (Tris-Borate-EDTA) buffer, and 5% glycerol was prepared. To polymerize, 340 μl of 10% ammonium persulfate and 80 μl of TEMED were added, the solution was mixed, then immediately poured into the gel plates and the comb was inserted. The polymerized gel was prerun at 150 V for 30-60 min prior to sample loading using 0.5X TBE as the running buffer.

| Assembly of RNAP binding/ transcription reactions

following settings: (1) to detect Cy3 emission: 532 nm excitation, 575 nm emission and (2) to detect Cy5 emission: 635 nm excitation, 665 nm emission. Any imaging system capable of detecting emission from these fluorophores can be used. The dyes and quencher can be changed as needed to meet the requirements of other imaging systems.

Reactions were assembled on ice according to Table 1.

| Hazards and safety

Components were added in the order listed (top to bottom) to give a final volume of 20 μl prior to the addition of NTPs or TE(low). Recipes for the reaction components are in the Supporting Information. The first two samples did not contain protein, and the odd numbered samples thereafter did not receive NTPs. The final four reactions contained 25 μM rifampicin (TOKU-E). Reactions contained 40 mM Tris pH 7.5, 10% glycerol, 50 mM KCl, 10 mM MgCl 2 , 1 mM DTT, 0.01% Triton X-100, 0.15 ng/ μl salmon sperm DNA (SS_DNA), and 2.5 nM of promoter DNA. The RNAP was diluted in the same buffer Personal protective equipment is recommended when working in the laboratory. Most reagents for this lab are aqueous buffers, salts, DNA, and protein, which pose minimal risk. Prior to polymerization, acrylamide is a neurotoxin, and should only be handled using gloves. The polymerized gel is not hazardous. The concentrated rifampicin stock could cause skin irritation and should be handled with gloves. When loading an actively running gel, students should not touch the running buffer with their hands. minus the DNA. After addition of RNAP, the reactions were incubated at 37C for 15 min. Next, either TE(low) or NTPs (1 mM ATP, CTP, UTP) were added as indicated

| RE SULT S AN D D ISCUSSION

and reactions were incubated an additional 15 min at 37C. Then, 18 μl of each reaction was loaded onto the native gel (no loading dye), while the gel was running at 150 V. The gel was run for 90-120 min with a garbage bag placed over the gel running apparatus to protect from light exposure. It is important the gel does not heat up while running; if necessary reduce the voltage.

The results from an EMSA to test the formation of RNAP/promoter complexes, transcriptional activity, and inhibition by rifampicin are shown in Figure 2. Reactions were assembled according to Table 1. The same native gel was scanned twice, once for Cy3 emission (left panel) and once for Cy5 emission (right panel), and the images from the two scans were aligned.

The results in lanes 1-6 enabled students to under-

| Data acquisition and analysis

stand the relationship between emission from each fluorophore, formation of RNAP/promoter complexes, and The gel was scanned for fluorescence emission (remove transcriptional activity. The Cy3 emission marked the top plate) using a Typhoon Imager 9400 position of the promoter DNA. The addition of RNAP to (GE biosciences). Two scans were performed using the

T A B L E 1 Reactions to evaluate RNAP binding to promoter DNA, the promoter DNA resulted in formation of a complex Reaction number transcriptional activity, and the effect of rifampicin 1 2 3 ddH20 10 10 8 Master mix 10 10 10 250 μM Rifampicin ---1:20 RNAP --2 4 5 6 7 8 9 10 8 8 8 6 6 6 6 10 10 10 10 10 10 10 ---2 2 2 2 2 2 2 2 2 2 2 Incubate at 37C for 15 min TE(low)

11X NTPs (-GTP)

Incubate at 37C for 15 min

Note: The components to add are listed in the first column, followed by the volume in μl to add to each reaction. The recipes for the components are in the Supporting Information. that shifted the Cy3-labeled DNA to a slower migrating band (compare lanes 1 and 2 without RNAP to the remaining lanes that all contained RNAP). The Cy5 emission marked transcriptional activity. Prior to transcription, emission from the Cy5 fluorophore was quenched (see Figure 1 and note the lack of Cy5 signal in lanes 1 and 2). The Cy5 emission remained quenched when by rifampicin, regardless of the addition of NTPs (see Cy3 signal, lanes 7-10). Therefore, rifampicin did not inhibit the ability of RNAP to bind to promoter DNA. By contrast, rifampicin strongly inhibited the appearance of the NTP-dependent Cy5 signal, showing that the antibiotic blocked the polymerase's ability to complete the RNA synthesis step of transcription (notice the lack of Cy5 signal in lanes 8 and 10, compared with lanes 4 and 6). As part of the post-lab data analysis, students can compare their conclusions regarding how rifampicin inhibits transcription to findings in the research literature. In this regard, the crystal structure of rifampicin bound to a bacterial RNAP 22 and biochemical experiments 23,24 are particularly useful. As an expansion of this experiment, the Supporting Information shows the use of salmon sperm DNA as an inhibitor of RNAP binding to promoter DNA. This could be used alongside rifampicin if testing inhibitors of both steps in transcription is preferable.

A that RNAP/promoter complexes formed. Students might observe weak signal in the Cy5 scan prior to transcription, which can arise from a small amount of DNA that does not have the quencher oligo annealed.

In the reactions that received NTPs, the nontemplate strand oligo containing the quencher was released when the polymerase transcribed the DNA, thus revealing Cy5 emission (see lanes 4 and 6). Hence displacement of the quencher oligo served as a marker for the RNA synthesis step in transcription. It is important to note that the Cy5 signal migrated at the same position as the RNAP/ promoter complexes in Cy3 scan. This shows that RNAP remained bound to the DNA after transcription, which is consistent with transcription halting at + 35 in the absence of GTP (see Figure 1). It is not necessary for RNAP to shift all the promoter DNA for the experiment to be interpreted. Even if the DNA is not fully saturated by the polymerase, students can still detect transcriptional activity (Cy5 signal) or changes in the fraction of DNA bound (Cy3 signal) when the experimental conditions are varied.

The results in lanes 7-10 enabled students to conclude how rifampicin functions as an inhibitor. The RNAP/promoter DNA complexes were largely unaffected RE FERE NCES