Stacy L. Hrizo and Nancy Kaufmann From the Department of Biological Sciences, University of Pittsburgh, Pittsburgh, Pennsylvania 15260

Transcription

1 Q 2009 by The International Union of Biochemistry and Molecular Biology BIOCHEMISTRY AND MOLECULAR BIOLOGY EDUCATION Vol. 37, No. 3, pp , 2009 Laboratory Exercises Illuminating Cell Signaling: Using Vibrio harveyi in an Introductory Biology Laboratory Received for publication, October 17, 2008, and in revised form, December 9, 2008 Stacy L. Hrizo and Nancy Kaufmann From the Department of Biological Sciences, University of Pittsburgh, Pittsburgh, Pennsylvania Cell signaling is an essential cellular process that is performed by all living organisms. Bacteria communicate with each other using a chemical language in a signaling pathway that allows bacteria to evaluate the size of their population, determine when they have reached a critical mass (quorum sensing), and then change their behavior in unison to carry out processes that require many cells acting together to be effective. Here, we describe a laboratory exercise in which the students observe the induction of bioluminescence or light production as an output of the quorum sensing pathway in Vibrio harveyi. Using both wildtype and mutant bacterial strains they explore the induction of community behavior via cell-cell communication by determining whether there is a correlation between the density of the bacterial population and the production of light by the bacterial culture. Using data from a cross-feeding assay the students make predictions about the identity of their strains and directly test these predictions using conditioned media from various liquid cultures. This two part exercise is designed for an introductory biology course to begin familiarizing students with collecting data, making predictions based upon the data and directly testing their hypotheses using a model organism with a cell signaling pathway that has a simple visual output: light production. Keywords: Undergraduate laboratory, cell signaling, bacteria, quorum sensing. Cell signaling is an abstract and varied concept that challenges introductory students in many ways. Often this results in students memorizing isolated pathways without understanding their context and value in biological systems. Few undergraduate laboratories have been developed to enhance conceptualization of cell signaling, perhaps because we traditionally think of signaling only in complex multicellular systems. However, using mammalian cells in an introductory laboratory class can be difficult because the cell culture systems are expensive and require specialized equipment. We report that bacterial cell-cell communication can be used in the undergraduate laboratory as an inexpensive and effective model for allowing students to directly test a cell signaling pathway. Bacterial cell-cell communication or quorum sensing is a fascinating process through which bacterial communities implement and coordinate behavior such as biofilm formation, virulence factor expression, and in the case of Vibrio harvyei, light production (for recent reviews see [1, 2]). Vibrio harveyi are safe, non-pathogenic marine bacteria that use a signaling pathway to regulate the expression of bioluminescence genes in response to the density of the To whom correspondence should be addressed. 536B Crawford Hall 4249 Fifth Avenue, Pittsburgh, PA nkaufman@pitt.edu. Present address of Stacy L. Hrizo, Department of Pharmacology and Chemical Biology, University of Pittsburgh Medical School, Pittsburgh, PA DOI /bmb culture (Fig. 1) [3, 4]. Of interest to students and researchers alike, this species of marine bacteria is closely related to Vibrio cholerae, which causes cholera in humans [5]. The World Health Organization reported 236, 896 cases of cholera and 6,311 deaths from the disease last year alone which is available at: ( factsheets/fs107/en/index.html). Therefore, there is significant interest and research into understanding these organisms. Importantly, the quorum sensing pathway that regulates bioluminescence in V. harveyi is very similar to the pathway that regulates virulence factor expression in V. cholerae [6, 7] making V. harveyi an important model system for human health research. It has been suggested that therapies may be developed to disrupt quorum sensing pathways in bacteria thus reducing the production of virulence factors and enabling treatment of the human disease [8]. This relevance to current medical research further enhances student interest and can be used to introduce students to careers in research. Furthermore, the V. cholerae model makes an excellent link to studies of the cell membrane, as binding of the toxin to the cell membrane causes severe osmotic fluid loss in infected humans. V. harveyi glow when their quorum sensing pathways are activated as signaling results in the production of luciferase, an enzyme that generates light in the visible range, readily observed with the naked eye in a dark room [9]. This makes them an ideal model system for the undergraduate laboratory. The research laboratory of Dr. Bonnie Bassler at Princeton University ( This paper is available on line at

2 165 labs/bassler/) has pioneered genetic and biochemical assays to dissect the quorum sensing pathway of V. harveyi. We have adapted these assays to allow students to identify V. harveyi mutants defective in two different steps of the signaling pathway so that students learn the logic of an ordered pathway. Although the wildtype strain glows on its own, the two mutants do not. One strain has a mutation eliminating its ability to produce either of the two secreted signaling molecules (AI-1, AI-2), so these bacteria cannot signal to each other and thus do not glow. The other strain has a mutation that eliminates its ability to produce the luciferase enzyme, essential for the production of light from its substrate luciferin. This strain can send and receive signals, but fails in the last step to produce light. Through the assays described below, students identify the two mutant strains and learn that the signaling output, light, is only produced when bacterial community numbers are sufficiently high and when the signaling pathway is intact. The following two section laboratory exercise uses three V. harveyi strains, one wildtype and two mutants, to study the induction of bacterial cell signaling. In the first experiment students look for a correlation between culture density and light output to determine whether bacteria indeed use cell signaling to count their neighbors. In the second experiment, they collect data from a plate crossfeeding assay using the signaling pathway output, light, to identify three unknown V. harveyi cultures 1. Based on the ability of the strains to produce light the students formulate hypotheses about the identity of their unknown strains. Using a liquid cross-feeding assay, the students then confirm their hypotheses. This assay also provides further evidence that a secreted molecule is able to stimulate bioluminescence in the bacteria [10]. FIG. 1.AI- signaling pathway in Vibrio harveyi. The signaling molecules, also known as autoinducers, are called AI-1 and AI- 2 and are recognized by their receptors, LuxN and LuxPQ, respectively. These receptors transduce the signal into the cell via a phosphorylation signaling cascade that ultimately results in the phorphorylation of the transcription factor LuxO. Phosphorylated LuxO is active to mediate the transcription of quorum sensing target genes, such as luciferase. Thus, the bacteria synthesize the light-producing enzyme when the pathway is activated. (Figure modified from ref. [15]). LABORATORY OVERVIEW AND PRE-LABORATORY SESSION The experiments are divided into two (3 hour) laboratory sessions including a half hour pre-lab and post-lab. Certain methods, such as sterile technique, streaking for single colonies and serial dilutions, are routinely found in undergraduate laboratories and are not described in detail. A complete list of reagents and recipes for media preparation are detailed below. The V. harveyi strains used in these experiments can be obtained from ATCC (wildtype: BB120 [11]; signal mutant: MM32 [12]; luciferase mutant: BB151 [4] [Table I]). Optimal growth of the cultures is at 30 8C in ATCC autoinducer bioassay media in a shaker set at 350rpm (Note that vigorous aeration is essential for culture growth and light induction). We have found that other common bacterial growth media, such as Luria-Bertani Broth (LB), while supporting bacterial growth, quench the bioluminescence of the organism. Before the start of the laboratory exercise, we discuss several major points with the students to emphasize the importance of cell signaling. Topics for discussion include: 1) Why is it important for cells in a multicellular organism to communicate with each other? 2) Why would it be useful to have mechanisms for both direct communication between neighboring cells and for long-distance communication with cells far away? 3) Why do single-celled organisms such as bacteria need to communicate with each other? What might be the advantage to bacteria to live in a community? Having introduced the idea that bacteria also possess signaling pathways, we transition to the concept of using bacteria as model organisms to understand cell to cell signaling. 4) What components of a signaling pathway would bacteria need to have? (signal, receptor, messengers, output effects) At this point we display Fig. 1, which outlines the quorum sensing pathway in V. harveyi. After going through the pathway briefly, we talk about the similarities between V. harveyi and V. cholerae, in particular, that they both use quorum sensing. We then discuss the importance of understanding bacterial signaling in relation to human disease prevention and treatment. In the first laboratory session we address whether V. harveyi can sense bacterial numbers by determining whether the concentration of bacteria in a culture affects their production of light. We determine the viable cell 1 This plate cross-feeding assay works because the signal is secreted and can diffuse through the agar plate, a great opportunity to review the process of diffusion.

3 166 BAMBED, Vol. 37, No. 3, pp , 2009 TABLE I Strains and predicted plate cross-feeding results (expected student responses are answered with yes or no) Wild-type strain Signal mutant strain (No AI-1, No AI-2) Light enzyme mutant strain (No luciferase) AI-2 signal production þ - þ AI-1 signal production þ - þ Light enzyme production (luciferase) þ þ - Prediction : Bioluminescence plated alone Yes No No Prediction: Bioluminescence plated next to another strain Yes Yes No if signal is secreted and diffusible Prediction: Bioluminescence plated next to another strain (but not touching) if signaling is contact-mediated Yes No No count of each bacterial culture and identify the concentration of bacteria necessary to initiate en masse light production. We also begin a plate cross-feeding experiment to determine if the signaling molecule is a soluble secreted molecule or whether the bacteria require direct contact to initiate the signaling pathway. In the second laboratory session, we collect and analyze our serial dilution and plate cross-feeding results. We further confirm the plate cross-feeding results in a liquid culture system and finally identify which bacteria culture corresponds to which signaling pathway mutant. LABORATORY SESSION 1 Required Materials 01) Twelve plates of ATCC Autoinducer Bioassay Media (see below) per group 02) Fifteen sterile toothpicks or plastic loops for each group 03) Six sterile cell spreaders per group 04) A container with lysol for the disposal of toothpicks and cell spreaders 05) Seven sterile 1.5 ml eppendorf tubes per group 06) Three Vibrio harveyi cultures of 25 ml grown to three different cell densities (OD 600 of 0.025, 0.15, and 0.7) in autoinducer bioassay media per group 07) Twenty milliliter sterile liquid autoinducer bioassay media per group 08) p1000 and p200 micropipettes for each group 09) Permanent markers for each group 10) Master plate with strain A (BB120-wildtype) 11) Master plate with strain B (BB151-luciferase mutant) 12) Master plate with strain C (MM32-signal mutant) 13) One room without windows or that can be sufficiently darkened to observe light production ATCC medium: 2034 Autoinducer bioassay medium (AB) NaCl g MgSO g Casamino acids, vitamin-free g Distilled water ml Adjust to ph 7.0 with KOH. Autoclave at 121 8C for 15 minutes. Cool to room temperature and aseptically add filter sterilized: 1 M Potassium phosphate, ph ml 0.1 M L-Arginine ml 50% Glycerol ml Note: To make agar plates add 15 g bactoagar to solution before autoclaving. The potassium phosphate will precipitate a little, but it goes back into solution after the plates have cooled. Serial Dilutions: Can Bacteria Sense Culture Density? Activation of the quorum sensing pathway in V. harveyi is dependent upon cell culture density because only at a threshold level will the bacteria produce light en masse [13, 14].To test this model, students visually examine three flasks of cultures with increasing cell density in a well-lit room. They denote which flask contains the most to least dense culture by visual inspection alone. One culture is below the threshold density for bioluminescence as measured by a spectrophotometer set at 600 nm wavelength (OD 600 ¼ 0.025) and will not produce much light. The other two cultures are at a higher OD 600 and do glow. (Note: Culture densities may be optimized for individual laboratory conditions.) After the students qualitatively record their cell density observations, they then go into a dark room to observe the bioluminescence in each flask. They examine the light production of each culture and write down their observations to determine if the cell density corresponds with the bioluminescence of the culture 2. The students next set up serial dilutions to confirm that their visual observations of the culture density match the actual viable cell count in the cultures. We have found that at the above culture densities, the best bacterial counts ( colony forming units) are obtained by diluting the culture and fold and spreading 100 ll of the dilutions onto agar plates. These plates are incubated overnight (approximately 18 hours) at 30 8C and stored at 4 8C until data collection. Note that all of these assays, including serial dilutions, use ATCC autoinducer bioassay media as using water or a buffer with a low salt concentration to dilute the cultures will kill these marine bacteria. Plate Cross-feeding Experiment: Is the signal molecule secreted? What happens to signaling if a pathway component is missing? A key concept in cell signaling is that soluble chemicals can diffuse through the environment over long distances and initiate cell signaling in a target cell expressing 2 If the instructor wishes the students can also use the more quantitative method of analyzing the culture density by diluting the culture 1:10 in fresh media and reading the optical density with a spectrophotometer set at 600nm wavelength.

4 167 re-stimulate bacterial growth. If the plates are left at 30 8C for too long (several days) the cultures become saturated on the plate and this slows bacterial growth which will quench the quorum sensing pathway and turn off luciferase production. During the exercise, it is important to remind students about their sterile technique as the most common mistakes made by the students is that they forget to close the lids of their tubes and plates and allowing their sterile loops to touch a non-sterile surface, thus contaminating their cultures. Other bacteria will grow but likely not glow! FIG. 2.Plate Cross-feeding Data. Bioluminescence in a plate cross-feeding experiment. The wild-type strain at the top of the plate will bioluminesce alone. Notice that the signal mutant strain produces significantly more light at the leading edge of the streak pattern where the received signal concentration would be highest. The luciferase mutant strain remains dark as it is unable to make luciferase. the receptor. Therefore, in order to experimentally prove that direct cell-cell contact is not necessary for some types of cell signaling we will use the plate crossfeeding assay and confirm our results with a liquid culture cross-feeding experiment in the second laboratory session. Students also need to learn that each component of a linear signaling pathway is essential. For these assays the students are given three strains of V. harveyi (A, B and C) that contain different combinations of the components of the signaling pathway thus altering their ability to produce luciferase in response to quorum sensing (Table I). Students are asked to assume that the signaling molecule is secreted and diffusible. Their results will either support or refute this model. Students streak each strain by itself onto an autoinducer bioassay media plate and as sectored patches onto the same plate (see Fig. 2) to determine which strain can glow when grown alone and which requires the presence of a second V. harveyi strain to glow. This will indicate that the strain requires a second strain in its vicinity to provide the missing signaling molecule to initiate the quorum sensing pathway. The students are given Table I which describes the signaling components of each strain and are asked to make predictions about results from the cross-feeding assay which they enter into Table I. On the data collection day they match their observations to their predicted results in Table I and identify the known strains (A, B, and C). The cross-feeding assay requires that the students practice aseptic inoculation techniques to maintain a pure bacterial culture. The students are given sterile plastic loops to transfer and streak bacteria from the master plates to their labeled experimental plates. These plates are incubated at 30 8C overnight and collected by the instructor the next morning and then stored at 4 8C (for up to a week). The plates are removed from 4 8C storage the day before student data analysis and put at 30 8C to LABORATORY SESSION 2 In the second session, students examine their serial dilution and cross-feeding plates to collect data. They can thus determine the identity of each strain based on the bioluminescence of each strain when plated alone or in combination. The students next verify these results by using conditioned media from V. harveyi cultures that contain the secreted signaling molecule, but no bacteria. The students set up the liquid culture experiments before data collection from the cross-feeding and serial dilution assays, as quorum sensing induction takes approximately 2 hours. Required Materials 1) Thirty milliliter of sterile autoinducer media per group 2) Twenty milliliter of filtered media from a wild-type culture at an OD 600 of >0.5 3) Note: filtered media can be prepped up to a week ahead of time and stored at 4 8C (see below) 4) Five milliliter cultures of A, B, and C strains for each group (Note: these can be saturated overnight cultures) 5) Eppendorf tubes of ml per group 6) Glass culture tubes (6) with caps 7) p1000 pipette 8) One room without windows or that can be sufficiently darkened to observe light production Liquid Cross-Feeding Experiment: Can we verify the plate cross-feeding results with liquid media containing secreted signal molecules? For the liquid cross-feeding experiment, students are given two different vials of sterile ATCC media. One vial contains liquid media which has never had prior bacterial growth. The second vial contains media conditioned by a wildtype V. harveyi strain. To prepare this media, wildtype V. harveyi cells are grown overnight to an OD 600 of >0.5 and the cells are removed from the suspension by centrifugation ( g for 10 minutes). The resultant supernatant which contains the signaling molecule is filter sterilized with a 0.45 lm cellulose acetate filter (Corning) to remove any remaining bacteria. Next, the students are given 3 ml overnight cultures of the three strains. The students transfer 1 ml of each overnight culture to two

5 168 BAMBED, Vol. 37, No. 3, pp , ml eppendorf tubes and pellet the cells by centrifugation ( g for 5 min). The students resuspend the resultant pellets in 1 ml fresh media. They add the newly resuspended cultures to glass culture tubes containing 4 ml of fresh liquid media or 4 ml of the signal containing conditioned media. They incubate their tubes in a 30 8C shaker for 2 hours and then take them into a dark room to examine the bioluminescence of each culture. (Data collection from the serial dilutions and plate crossfeeding assays can be scheduled during this incubation time). By observing the light production of cells treated with conditioned media, the students can determine whether their identification of strains from the plate cross-feeding assay was correct. Data Collection: Plate Cross-feeding Analysis The students are given their plates from the 30 8C incubator and visually examine the growth of their bacterial streaks in a well lit room to ensure the growth of strains A, B, and C. They then enter a darkened room where they observe the bioluminescence of the bacteria (Fig. 2). The students record their cross-feeding results denoting whether the bacteria produced light when plated alone or required the presence of another strain to bioluminesce. The wildtype strain has all of the components of the signaling pathway and will glow when plated alone (and next to another strain.) The signal and luciferase mutants will remain dark when plated alone. However, the dual platings will distinguish between these strains. The signal mutant will be stimulated to produce luciferase in response to signals that are secreted by the other bacteria growing on the plate. The luciferase mutant is unable to generate the light-producing enzyme and thus will remain dark under all plating conditions. This analysis enables the students to identify the quorum sensing mutants based on their phenotypes described in Table I. They will then confirm these results with the data from their liquid cross-feeding experiments. Data Collection: Serial Dilution Analysis During the 2 hour incubation for the liquid cross-feeding assay, the students also collect and analyze their serial dilution data. Each group counts the number of colonies on their plates at the dilution that generated colonies on a plate for easy and accurate counting. They then calculate the colony forming units in the initial culture using their data and the serial dilution information (For example: Number of colonies ¼ 50, Dilution factor ¼ 1: or 1:1,000,000; ,000,000 ¼ 50,000,000 or CFUs/mL), giving the students an opportunity to practice laboratory math. The students compare their serial dilution results with the results from their visual inspection of the cultures from the first laboratory session to determine the number of viable cells present in the cultures that matches their qualitative analysis of culture density. The students decide whether their data support or fail to support a threshold dependent quorum sensing model for V. harveyi. Finally, students propose additional experiments that could strengthen the argument that the bacteria indeed count their community members and only produce light at a threshold cell density. These data and ideas are discussed in the post-lab. Post-Lab Discussion In the post-lab, each group reports their findings, filling in a chart for their results with unknowns A, B, and C similar to Table I. This group data is recorded on a table on the chalkboard and we discuss the following points: 1) Is communication in V. harveyi contact-mediated or does it use a secreted, diffusible signal? What results would we have seen if signaling were contact-mediated? 2) Which unknown is the wildtype, signal mutant, and luciferase mutant strain? 3) Do our data support a model for quorum sensing in V. harveyi? 4) What further experiments should be done to strengthen or confirm this model? CONCLUSIONS This two-session research module has the advantage that it does not require expensive laboratory equipment for effective completion. Many of the steps in data analysis require only the visual inspection of the student in a lit room and a darkened room, thus making it accessible to many educational institutions. Also, the modules were designed for an introductory biology course to train the students to create hypotheses based upon their observations and to test those hypotheses in a hands-on manner. A common pitfall for first-year science students is a lack of time for analysis and discussion of their data; as a result the students do not fully understand the meaning of their experimental results. To address this we have included a thirty minute pre-laboratory discussion during session 1, and a data analysis and discussion during the 2-hour incubation period in session 2. This gives the students ample time to collect their data, critically analyze their results, and ask questions. We do not require a formal laboratory report for this exercise, rather we have created a study guide with questions to drive home important skills (i.e. calculating colony forming units of the original culture), major points (i.e. Do your data support that it is an extracellular signal and if so what type of molecule could the signal be?) and an open-ended question that requires the students to research their answer (i.e. Why do you think it is important that bacteria talk to each other?). Overall, this set of experiments familiarizes introductory level students with a broad range of laboratory skills including sterile inoculation techniques with both solid and liquid media, serial dilutions, and the use of basic laboratory equipment such as pipettes, centrifuges and incubators. At the end of the unit students have observed stimulation of a signaling pathway from an extracellular signal and correlated signaling output with the requirement of all components of the signaling pathway.

6 169 Acknowledgment We thank Dr. Bonnie Bassler at Princeton University for providing us with the Vibrio harveyi strains used in this article. We also thank Ms. Carole LaFave and Dr. Melanie Popa for assistance with protocol development and Ms. Shruthi Vembar for her critical review of the article. REFERENCES [1] B. L. Bassler, R. Losick (2006) Bacterially speaking, Cell 125, [2] T. Defoirdt, N. Boon, P. Sorgeloos, W. Verstraete, P. Bossier (2008) Quorum sensing and quorum quenching in Vibrio harveyi: Lessons learned from in vivo work, ISME. J. 2, [3] B. L. Bassler, M. Wright, R.E. Showalter, M.R. Silverman (1993) Intercellular signalling in Vibrio harveyi: Sequence and function of genes regulating expression of luminescence, Mol. Microbiol. 9, [4] B. L. Bassler, M. Wright, M. R. Silverman (1994) Multiple signalling systems controlling expression of luminescence in Vibrio harveyi: Sequence and function of genes encoding a second sensory pathway, Mol. Microbiol. 13: p [5] S. M. Faruque, M. J. Albert, J. J. Mekalanos (1998) Epidemiology, genetics, and ecology of toxigenic Vibrio cholerae, Microbiol. Mol. Biol. Rev. 62, [6] M. B. Miller, K. Skorupski, D.H. Lenz, R.K. Taylor, B.L. Bassler (2002) Parallel quorum sensing systems converge to regulate virulence in Vibrio cholerae, Cell 110, [7] J. Zhu, M B. Miller, R.E. Vance, M. Dziejman, B.L. Bassler, J. J. Mekalanos, (2002) Quorum-sensing regulators control virulence gene expression in Vibrio cholerae, Proc. Natl. Acad. Sci. U. S. A. 99, [8] M. Hentzer, M. Givskov (2003) Pharmacological inhibition of quorum sensing for the treatment of chronic bacterial infections, J. Clin. Invest. 112, [9] K. H. Nealson, J. W. Hastings (1979) Bacterial bioluminescence: Its control and ecological significance, Microbiol. Rev. 43, [10] M. G. Surette, B. L. Bassler (1998) Quorum sensing in Escherichia coli and Salmonella typhimurium, Proc. Natl. Acad. Sci. U. S. A. 95, [11] B. L. Bassler, E. P. Greenberg, A. M. Stevens (1997) Cross-species induction of luminescence in the quorum-sensing bacterium Vibrio harveyi, J. Bacteriol. 179, [12] S. T. Miller, K.B. Xavier, S.R. Campagna, M.E. Taga, M.F. Semmelhack, B.L. Bassler, F.M. Hughson (2004) Salmonella typhimurium recognizes a chemically distinct form of the bacterial quorum-sensing signal AI-2, Mol. Cell. 15, [13] K. H. Nealson, T. Platt, J. W. Hastings (1970) Cellular control of the synthesis and activity of the bacterial luminescent system, J. Bacteriol. 104, [14] B. N. Lilley, B. L. Bassler (2000) Regulation of quorum sensing in Vibrio harveyi by LuxO and sigma-54. Mol. Microbiol. 36, [15] B. L. Bassler (1999) How bacteria talk to each other: Regulation of gene expression by quorum sensing, Curr. Opin. Microbiol. 2,