Mini-array Transcriptional Analysis of the Listeria monocytogenes Lecithinase Operon as a Class Project

Transcription

1 2006 by The International Union of Biochemistry and Molecular Biology BIOCHEMISTRY AND MOLECULAR BIOLOGY EDUCATION Printed in U.S.A. Vol. 34, No. 3, pp , 2006 Laboratory Exercises Mini-array Transcriptional Analysis of the Listeria monocytogenes Lecithinase Operon as a Class Project A STUDENT INVESTIGATIVE MOLECULAR BIOLOGY LABORATORY EXPERIENCE* Received for publication, October 13, 2005, and in revised form, January 4, 2006 Douglas Christensen and Marko Jövic From the Department of Life Sciences, Wayne State College, Wayne, Nebraska This report describes a molecular biotechnology-based laboratory curriculum developed to accompany an undergraduate genetics course. During the course of a semester, students researched the pathogen, developed a research question, designed experiments, and performed transcriptional analysis of a set of genes that confer virulence to the food-borne pathogen, Listeria monocytogenes. Gene fragments were amplified via PCR and utilized in mini-arrays, a dot-blot-based format suitable for the simultaneous transcriptional analysis of multiple genes. The project provides exposure to a wide range of molecular techniques and can be easily modified for variations in class size. Data are generated at various steps of the process, allowing for student interpretation, troubleshooting, and assessment opportunities. Keywords: Dot-blots, hybridizations, arrays, molecular genetics laboratory, transcriptional regulation. Normal or pathological functions of tissues are largely determined by variations of gene expression. DNA microarray technology allows simultaneous determination of expression levels in tens of thousands of genes. This approach has dramatically increased the data generated in the field of functional genomics. Although this technology enables researchers to assess the relative expression of thousands of genes, it is also highly dependent on expensive microarray equipment such as high throughput DNA array robots, multicolor fluorescence scanners, and skilled staff capable of performing array-based experiments [1], all of which are not conducive to hands-on learning at the undergraduate level. Many of the concepts of microarray technology and its applications in an actual research setting can be carried out by utilizing much smaller data sets, limited to several genes rather than thousands. In this case, we utilized a dot-blot-based array called mini-array technology to introduce students to laboratory techniques in microarray technology. This report describes the laboratory experience of sophomore and junior level students in molecular genetics. These students have taken general biology and chemistry but have no prior experience with molecular genetics. Students learn procedures utilized in the development of a mini-array that introduces them to molecular techniques * This project was supported by National Institutes of Health Grants 1 P20 RR16469 and 2 P20 RR from the Biomedical Research Infrastructure Network (BRIN) and Institutional Development Award (IDeA) Network of Biomedical Research Excellence (IN-BRE) Programs, respectively, of the National Center for Research Resources. To whom correspondence should be addressed: Dept. of Life Sciences, 223 Carhart Science, Wayne State College, 1111 N. Main, Wayne, NE dochris1@wsc.edu. This paper is available on line at and interpretation of data. As the semester-long project progresses, students are exposed to literature review techniques, media preparation, DNA isolation procedures, National Center for Biotechnology Information (NCBI) 1 - based sequence analysis, primer design, PCR techniques, gel electrophoresis, gel excision/purification techniques, RNA isolation procedures, probe labeling techniques, hybridization procedures, and most critically, data analysis/ interpretation and troubleshooting. SESSION ONE In the first session (2 3 h), the instructor suggests an organism/pathway of interest for subsequent study. One of the benefits of this experience is that variations in class size can be easily accommodated by use of larger gene sets or multiple gene systems from various organisms. In this case, we developed a gene expression analysis project with an objective being the lecithinase operon in strains of Listeria monocytogenes that exhibit either high pathogenicity (L. monocytogenes 4b) or low pathogenicity (L. monocytogenes 1/2a). This operon is composed of six genes (Fig. 1), allowing for our laboratory ( 14 students per laboratory) to be assigned a pet gene (gene fragment) for every two or three students in the laboratory. The professor directed a discussion centering around the idea that research could provide insight as to why some strains of L. monocytogenes are pathogenic, whereas others are not. This information provided the students with a basis for gathering journal articles relevant to the subject at hand. Each student was given 1 week to obtain two peer-re- 1 The abbreviation used is: NCBI, National Center for Biotechnology Information.

2 222 BAMBED, Vol. 34, No. 3, pp , 2006 and often manifested as uncomplicated, mild gastrointestinal symptoms. Listeriosis in immunocompromised individuals can cause a wide range of diseases due to its ability for efficient cell-to-cell movement across normal barriers to infection such as the bloodbrain barrier [4]. FIG. 1. Lecithinase operon. Clustered virulence genes in a pathogenesis island of L. monocytogenes. The PrfA protein (depicted with an oval) is thought to promote ( ) the transcription of other genes involved in virulence including: acta (actin polymerization), hly (listeriolysin O), and plca (phosphatidylinositol-specific phospholipase A). This figure was recreated from Salyers and Whitt [6]). viewed journal articles relevant to the assigned topic. These can be obtained from electronic or print sources depending on your local resources. At the end of the first week, copies of each paper were collected along with a brief (1 2 pg) summary of each article for assessment purposes. This information serves as a tool for studentorganized experimental design. Technical skills introduced during the first laboratory session included training in media preparation and autoclave sterilization. SESSION TWO Session 2 ( 1 h) begins with a classroom question-andanswer session, led by the instructor. By asking specific questions geared toward the lecithinase operon, L. monocytogenes 4b and 1/2a, the students are able to utilize the information gathered the previous week to address basic questions regarding the function of specific genes associated with virulence. The instructor assembles a summary in outline form. The outline is used to generate experiments that are fundamentally relevant to the core of the project. From the student-guided outline, the professor writes a background paper used to frame the research questions and determine the experimental objectives for the semester. These will be set up in Session Three. The following passages reflect the results of this class discussion. The genus Listeria contains six species, namely L. monocytogenes, Listeria ivanovii, Listeria seeligeri, Listeria grayi, Listeria innocua, and Listeria welshimeri. This genus belongs to the low G C content Grampositive bacteria and is closely related to the Bacillus and Streptococcus genera [2, 3]. Only two species in the genus are pathogenic; L. monocytogenes is associated with human and animal disease, and L. ivanovii is associated only with animal sickness. L. monocytogenes is a human pathogen commonly found in the environment and sometimes carried asymptomatically in the human gut. Listeriosis was first described in 1926 as a disease of rabbits and has generally been thought of as a veterinary disease and an occasional infection in humans. However, in most reported cases of human illness, there is no history of direct contact with infected animals, and most human listerial infections have occurred in urban populations [3]. In healthy individuals, successful infection is rare L. monocytogenes mediates its own internalization into the host cell by producing internalin A (inla) and internalin B (inlb), two homologous invasion proteins. Host cell vacuoles are then lysed by the pore-forming toxin listeriolysin O (hlya) and phospholipase C, enabling bacteria to escape [3]. Once in the cytoplasm, bacteria can multiply and rapidly move around the cell by polymerization of membrane-based actin protein (acta) [5]. Upon its interaction with the host cell membrane, the bacterium forces its way into a neighboring cell, where it lyses the adjacent cell s membrane, completing the cycle. Although dependent on expression of a number of virulence genes, Listeria infection is believed to be regulated by a single factor, a transcriptional activator protein known as PrfA. It has been shown that the prfa gene codes for the key component of L. monocytogenes pathogenicity, responsible for gene expression in lecithinase operon [5]. This set of genes contains the major Listeria virulence factors, consecutively aligned, with the prfa gene being at the very beginning of the operon sequence (Fig. 1) [6]. Regulation of these genes by PrfA is supported in recent research, although there are significant data indicating that inla and inlb are partially thermoregulated. Generally, the method of regulation consists of PrfA binding to what have been named PrfA boxes by many scientists. These boxes consist of 14-bp sequences located around 41 bp upstream. There has been evidence of the lecithinase operon genes being differentially regulated, with hlya and plca being better transcribed than mpl, a zinc-dependent metalloprotease and acta. This discrepancy is a result of the alignment of their PrfA boxes, which are being perfectly symmetrical with plca and hly sequences extending in opposite directions [3, 5]. One experiment actually demonstrated that more symmetrical boxes produced genes with higher expression [7]. It has also been suggested that PrfA is capable of repression of the same genes it activates through cooperative binding of several PrfA molecules [5, 7]. This is thought to be the reason that gene expression decreases late in the stationary phase of L. monocytogenes and also suggests a mechanism by which PrfA could downregulate its own synthesis [7]. Listeria species are well adapted for their saprophytic lifestyle in decaying vegetation. Under such innate circumstances, the mammalian virulence genes are repressed. The plant-derived disaccharide cellibiose was thought to be the basis of the signal that the bacterium is in its natural soil environment. However, this control is now thought to be a more general mechanism of carbon repression rather than a spe-

3 223 cific signal, with sugars using two independent mechanisms of virulence gene repression in Listeria, one responding to any carbon source and another specific for -glucosides [3]. However, for the purpose of our research, we focused exclusively on prfa gene-associated regulation in the lecithinase operon, with an emphasis on assessing the capabilities of new gene expression analysis technique, namely mini-array technology. These background papers are disseminated to each student with the assigned task of preparing a short summary of experimental conditions they think may be relevant to the study of genes associated with virulence. SESSION THREE This session is critical and can be handled in dramatically different fashions depending on the personality of the class. Ideally, the class is provided a single task. In this case, they were to design a study that will lead to analysis of the transcriptional regulation of genes that may be involved in the virulence of L. monocytogenes 4b and lack of virulence in 1/2a. More guidance may be required in many cases. Ideally, students are allowed to discuss options in an instructor-free setting and generate a basic set of genes and experimental objectives, design, and controls that will determine the source of L. monocytogenes virulence. After an hour of discussion, the instructor can critique the logic and objectives identified in the student-directed discussion. The resulting experimental design needs to be reasonable and must be highly adaptable to variable class sizes. This is most easily accomplished by altering either the gene numbers and/or the experimental conditions, ensuring that all students will get a hands-on learning experience. Final decisions will prepare students for subsequent wet laboratories as described below. MATERIALS AND METHODS SESSION FOUR A 3-h wet laboratory begins with a DNA isolation procedure carried out as described previously [8]. In preparation, the instructor has inoculated the bacteria of interest into the student-prepared media (from session 1) the previous day. Because of its pathogenicity (L. monocytogenes 4b and 1/2a), the DNA isolation from this organism was carried out prior to laboratory by the professor. However, students were directed to isolate DNA from Salmonella control cultures under the supervision of the professor during this particular laboratory procedure. Isolated samples were stored at 20 C prior to analysis in session 5. SESSION FIVE In this session (an 3-h laboratory), three tasks are to be accomplished, which included: the determination of DNA concentrations (isolated in session 4), basic NCBI gene analysis instructions, and primer design instructions. DNA concentrations were determined via 260 nm spectrophotometer analysis (Ultra spec 2000, Amersham Biosciences, Cambridge England) utilizing the following calculation: DNA concentration ( g/ml) (A 260 ) (dilution factor) (50 g of DNA per ml/1 A 260 unit). The remainder of the laboratory is utilized to instruct students how to use the NCBI data base website and design primers. In this case, students are taught simply to conduct specific NCBI gene searches and identify potential genes or gene regions relevant to the transcriptional regulatory analysis of the lecithinase operon. Primer design instructions are purposely vague. Students are asked to design primers that are nucleotides in length. Instructions are limited to describing 5-3 orientation and its implications as well as the importance of representing each of the complementary strands of DNA. Finally, the students are encouraged to think about possible primer flaws, but a great deal of leeway was allowed here. Primer designs were returned within 1 week along with the NCBI-obtained sequence of the gene fragments of interest. Annealing sites were highlighted on the accompanying sequence. Understandably, student-designed primers were often flawed with the major problems being designs that would result in priming of one strand only, highly unbalanced GC ratios (melting temps) within a primer set, and probable secondary structure formations. By holding a short classroom discussion after the primer designs are submitted, many potential errors can be addressed and become a valuable learning experience for the class. After some modification, several primer sets were ordered (Sigma Genosys). We have used the following primers successfully (listed below in a 5-3 orientation) for this particular exercise: hly, GGAGATGCAGTGAC and GCTTGCAACTGCTC; plcb, GGT- ATGTGCTTGACC and GGTAATCAGTCACC; prfa, CGCTCAAG- CAGAAG and GCTGAGCTATGTG; ccpa, GAGTGAATAGAGATG and CGTCCGTTCATAGTC; acta, CGAAACTGCACGTG and GG- GAAGTCCGAAGC; flic, GCGCTGTCGAGTTCTATCGAGC and CAACGGTGACTTTATCGCCATTCC; InvA, GCTCGTTTACGACC and CGACGGACATCGAC. SESSION SIX Standard PCR was carried out using the primers developed in Session Five ( 1-h setup). Utilizing Sigma Taq polymerase and reagents, standard (30- l) PCR reactions included a denaturing step (94 C, 4 min), 30 cycles of denaturing (94 C, 1 min), annealing (55 C, 30 s), and extension (72 C, 1 min). Amplification requires 3 h; therefore, analysis is conducted in Session Seven. SESSION SEVEN Validation of gene fragment presence and size was confirmed using a standard 1-kbp marker (Invitrogen) in 0.8% Tris Acetate EDTA or ethylenediaminetetraacetic acid agarose gel electrophoresis (Sigma). Gene fragments were excised and purified with Qiaex R II gel extraction kit (Qiagen, Valencia, CA). Students were also allowed to verify the concentration of harvested PCR product through UV spectrophotometry. Typical yields are ng/ l. All samples were diluted to a standard concentration of 5 ng/ l using nuclease-free water (Ambion). SESSION EIGHT After providing students with background information on the capabilities of microarray technology, we set up a modified experimental design, as represented in Fig. 2. This dot-blot-based technique is introduced by a discussion comparing Northern blot techniques with microarray analysis. A volume (1 l) of each heat-denatured PCR-generated gene product (purified from sessions 6 and 7) were spotted (resulting in a final of 5 ng) onto a positively charged nylon membrane (Roche Applied Science) and fixed through UV cross-linking (Spectrolinker, Spectronics Corp.). Procedurally, probe generation required the isolation of total RNA from L. monocytogenes 4b and 1/2a grown under the experimental conditions outlined in Figs. 3 and 4. Cells were grown in tryptic soy broth with yeast extract at 37 C for 6 h. RNA was harvested utilizing the FastRNA Blue kit R (Bio 101 Systems, Carlsbad, CA). Since this particular project employed a pathogen, this work was carried out by a supervised senior research student only. Upon RNA harvest, however, the class resumed hands-on duties of all procedures. Chemiluminescent probe generation and detection was performed, utilizing the Gene Images TM random prime labeling and detection system (Amersham Biosciences). Kit

4 224 BAMBED, Vol. 34, No. 3, pp , 2006 FIG. 2. Mini-array procedure. Gene fragments are PCR-amplified and purified through agarose gel purification (step A). Bands are cleaned, heat-denatured, and transferred directly in 1- l amounts to nylon membrane (step B), which will act as a hybridization platform in step E. Total RNA is isolated from a bacterial organism under various experimental conditions (step C). The total RNA is DNAse-treated (not shown) and fluorescein II-dUTP labeled (step D) (Amersham Biosciences). The labeled product is hybridized to the nylon mini-array platform, and detection is carried out. FIG. 3.Mini-array analysis of acta, ccpa, hly, plca, and prfa (no glucose). In the absence of glucose, both L. monocytogenes 4b (mono 4b) and L. monocytogenes 1/2a (mono 1/2a) appear to have nearly identical transcriptional regulation of acta, ccpa, hly, and prfa genes. Gene plca (circled), however, was not transcribed in the low virulent 1/2a strain, whereas the highly virulent 4b strain continued to transcribe this gene product. Negative controls (cont.) (flic, InvA, and lambda phage) expectedly indicated no background binding. modifications were minor and included a 1-h DNase treatment (Invitrogen) followed by AutoSeq TM G-50 column purification prior to handling the RNA as if it were DNA according to kit instructions. The final product (cdna) was to be utilized in hybridization and washing steps. To accommodate timing issues, small student groups were assigned tasks throughout the 2-day washing and hybridization periods. The resulting images were captured on a UVP Biochemi gel documentation system, and printed images were distributed to the class for individual student interpretation. SESSIONS NINE AND TEN Images obtained during session 8 (Figs. 3 and 4) were handed individually to each member of the class. Each student was asked to interpret data based on the original research objectives and write a short summary addressing the following questions. 1) Can probes generated from total RNA be used successfully at a single temperature for transcriptional analysis? 2) Do the results associated with the negative controls affect your confidence in the data (how so)? 3) How can we be sure that all of the genes associated with the lecithinase operon (on this hybridization platform) can be successfully detected on this platform? 4) What do these results suggest in terms of variations (if any) between the FIG. 4.Mini-array analysis of acta, ccpa, hly, plca, and prfa (bile or glucose). In the presence of glucose (10 mm), both L. monocytogenes 4b (mono 4b) and L. monocytogenes 1/2a (mono 1/2a) appear to have nearly identical transcriptional regulation of acta, ccpa, hly, plca, and prfa genes. In the presence of bile (0.1 mg/ml), however, plca (circled) appears to be inconsistently regulated, comparing high virulent strain 4b and low virulent strain 1/2a. Negative controls (flic and InvA) were expectedly negative. transcriptional regulation of L. monocytogenes 4b versus 1/2a? Of course, the number and intensity of questions in any session can vary greatly. Appropriate interpretations of experimental data vary from student to student. However, this provides an ideal assessment of each student s comprehension of the project as a whole. After assessment, the instructor can lead a group discussion (session 10, 1 h) concerning the results, allowing students to see and explain alternative viewpoints. RESULTS AND DISCUSSION Mini-array design is a modification of a dot-blot in that denatured gene fragments are directly transferred onto charged membrane, in this case without a blotting apparatus. Probes are generated utilizing random priming of prokaryotic total RNA isolates as described above. Because probe generation utilizes total RNA as template, it is similar to prokaryotic microarray probe technology. However, unlike microarray technology, mini-arrays are not capable of accurate depiction of expression ratios between experimental and control groups. The limitation is due to the inability to differentiate between control and experimental probes with any level of confidence on a single hybridization platform. The importance of this limitation should be discussed with the class. Despite its limitations, mini-arrays provide an avenue to discuss the benefits and shortcomings of dot-blotting, microarray technology, and Northern blotting. In this laboratory experience, transcriptional analyses of multiple genes associated with the lecithinase operon were analyzed through several mini-arrays. RNA was isolated from L. monocytogenes 1/2a and 4b RNA grown in the presence (10 mm) and absence of glucose or with the addition of 0.1 mg/ml bile (Oxgall, BD Diagnostic Systems). These specific conditions were chosen as a result of class discussions based on literature review information obtained by the students. The reasoning of each condition, by student consensus, was that L. monocytogenes that has infected human cells may find itself in competition for glucose within human host cells and, if ingested on a food source, would contact bile in the digestive tract. Although these conditions may not provide the ideal conditions in the instructor s mind, it is important to respect the class consensus and proceed so that students develop a degree

5 225 of ownership for the research. This session also allows for significant troubleshooting discussions to take place. If RNA harvests are low or contain significant contamination with DNA and/or proteins, students can be introduced to experimental troubleshooting. Low harvests, for example, were readily eliminated by subjecting the cells to additional physical disruption (2 60 s) in a bead-beater (Mini- Beadbeater TM, BioSpec Products, Inc.). In all cases, RNA was harvested during log phase growth and resulted in successful annealing to all target genes associated with Listeria. Additionally, no background binding was observed with negative controls (Figs. 3 and 4). Hybridization techniques also allowed for additional troubleshooting to be performed by the students. High background levels on some mini-arrays were solved via increased washing times and temperatures (data not shown). In addition to the success of this multiple probe mini-array, insight was gathered in these same figures as to the regulation of the lecithinase operon in L. monocytogenes in response to variable glucose and bile levels in the environment. The class, with instructor guidance, is asked to draw final conclusions. In this case, the results indicate that multiple probes can be used successfully in a simultaneous reaction when using prokaryotic targets. Standard hybridization and washing conditions were successful and could be modified slightly to accommodate individual samples (Figs. 3 and 4). Results indicate that this dot-blotbased mini-array has the potential to become an affordable and accessible technique of prokaryotic transcriptional analysis for functional genomics studies. It may also provide an alternative approach to confirming microarray data, which is commonly carried out through reverse transcription-pcr and is often less inaccurate than the reverse transcriptase step [9] and less time-consuming than Northern blot-based analysis. Larger gene expression data sets are necessary to further verify findings on a wider range of both eucaryotic and prokaryotic organisms. Mini-array technology is a highly efficient and affordable method. In comparison with a standard Northern blot procedure, the mini-array eliminates the overnight blotting step through direct transfer of the gene fragments onto membrane. However, dot-blotbased mini-array performance dramatically increases gene analysis data with the use of multiple probes, making it a powerful, affordable technique and a powerful teaching tool. In terms of gene regulation in both low and highly pathogenic strains of L. monocytogenes, it has been hypothesized that the PrfA protein is the key regulatory factor controlling expression of all of the virulence genes in the lecithinase operon [7]. Representative results from this experimentation example (Figs. 3 and 4) demonstrate that transcription of plca differs between high and low pathogenic strains of L. monocytogenes. Highly pathogenic strain 4b continues to transcribe plca under low glucose conditions, whereas strain 1/2a does not (Fig. 3). This result would indicate that a possible scenario for variation in virulence between these two strains is the inability of L. monocytogenes to produce phospholipase C during conditions of limited glucose. This would reduce the organism s ability to escape the vacuole, an act normally aided by the plca protein [10]. Further analysis of the lecithinase operon revealed other variations between L. monocytogenes 1/2a and 4b. Fig. 4 demonstrates that when glucose is available, both strains are capable of transcribing all genes associated with the operon. However, it is interesting to note that the presence of bile down-regulates the transcription of only one gene, plca, but only in the highly pathogenic strain 4b. These data indicate that at the time L. monocytogenes is exposed to bile, its virulence may be significantly reduced. Furthermore, since bile production may be affected by general health, these results correlate with the fact that healthy individuals seldom suffer from listeriosis. This research would not be the first to indicate independent mechanisms of lecithinase gene operon regulation. Resent research suggests significant up-regulation of virulence genes, specifically hly (20-fold) and acta (226- fold), upon entry of L. monocytogenes into host cell cytosol [10]. This model corresponds to other studies suggesting independent gene regulation under conditions corresponding to those of the host cell, such as iron concentration of extracellular compartments and cell ph [11, 12]. Furthermore, thermal regulation has also proven to have a substantial effect, especially under conditions above 26 C, when hly and potentially some other virulence genes are highly expressed regardless of other factors. This report presents a novel teaching application using mini-array technology to explore relationships in transcriptional regulation of virulence genes associated with high and low pathogenic strains of bacteria. Furthermore, it demonstrates effective transcriptional analysis using a low cost, relatively high throughput system to teach undergraduates cutting edge research and molecular biotechnology. Finally, investigative activities such as this have played a large role in dramatically increasing our students awareness of Ph.D. research programs. In the last several years, Wayne State College has successfully placed a number of students in Ph.D. research laboratories around the country. Activities such as the series described above have been instrumental in shaping the future goals of our students. They have provided students with the confidence they need to hit the ground running in premiere graduate research and professional programs throughout the country. Acknowledgments We thank the University of Nebraska-Lincoln, Food Science Culture Collection for providing specialized bacterial cultures used in this investigation. Also, we thank Drs. Barb Engebretsen and Shawn Pearcy for the contributions to this study and hard work on the grant. REFERENCES [1] J. A. Macoska [2002] The progressing clinical utility of DNA microarrays, CA-Canc. J. Clin. 52, [2] L. W. Greiffenberg, Goebel, K. S. Kim, I. Weiglein, A. Bubert, F. Engelbrecht, M, Stins, M. Kuhn [1998] Interaction of Listeria monocytogenes with human brain microvascular endothelial cells: InlB-dependent invasion, long-term intracellular growth, and spread from macrophages to endothelial cells, Infect. Immun. 66, [3] J. Vázquez-Boland, M. Kuhn, P. Berche, T. Chakraborty, G. Domínguez-Bernal, W. Goebel, B. González-Zorn, J. Wehland, J. Kreft [2001] Listeria pathogenesis and molecular virulence determinants,

6 226 BAMBED, Vol. 34, No. 3, pp , 2006 Clin. Microbiol. Rev. 14, [4] D. Portnoy, V. Auerbuch, I. Glomski. [2002]. The cell biology of Listeria monocytogenes infection: The intersection of bacterial pathogenesis and cell-mediated immunity, J. Cell Biol. 158, [5] J. Williams, C. Thayyullathil, N. Freitag [2000] Sequence variations within PrfA DNA binding sites and effects on Listeria monocytogenes virulence gene expression, J. Bacteriol. 182, [6] A. Salyers, D. Whitt [1994] Bacterial Pathogenesis: A Molecular Approach, pp , American Society of Microbiology Press, Washington, D. C. [7] A. Renzoni, A. Klarsfeld, S. Dramsi, P. Cossart [1997] Evidence that PrfA, the pleiotropic activator of virulence genes in Listeria monocytogenes, can be present but inactive. Infect. Immun. 65, [8] M. O. Byon, J. K. Kaper, L. O. Ingram [1986] Construction of a new vector for expression of foreign genes in Zymomonas mobilis, J. Ind. Microbiol. 1, [9] D. M. Mutch, A. Berger, R. Mansourian, A. Rytz, M. A. Roberts [2002] The limit fold change model: A practical approach for selecting differential expressed genes from microarray data, BMC Bioinformatics 3, 17. [10] M. Herler, A. Bubert, M. Goetz, Y. Vega, J. A. Vazquez-Boland, W. Goebel [2001] Positive Selection of mutations leading to loss or reduction of transcriptional activity of PrfA, the central regulator of Listeria monocytogenes virulence. J. Bacteriol. 183, [11] J. Behari, P. Youngman [1998] Regulation of hly expression in Listeria monocytogenes by carbon sources and ph occurs through separate mechanisms mediated by PrfA. Infect. Immun. 66, [12] P. Cossart, M. Lecuit [1998] Interactions of Listeria monocytogenes with mammalian cells during entry and actin-based movement: Bacterial factors, cellular ligands and signaling. EMBO J. 17,