Environmental Microbiology (2006) 8 (1), 165���177 doi:10.1111/j.1462-2920.2005.00890.x �� 2005 Society for Applied Microbiology and Blackwell Publishing Ltd Blackwell Science, LtdOxford, UKEMIEnvironmental Microbiology 1462-2912Society for Applied Microbiology and Blackwell Publishing Ltd, 2005 8 1165177 Original Article Pseudomonas putida transcriptional apparatusL. Yuste et al. Received 28 March, 2005 accepted 23 June, 2005. *For correspon- dence. E-mail frojo@cnb.uam.es Tel. ( + 34) 91 585 45 39 Fax ( + 34) 91 585 45 06. Growth phase-dependent expression of the Pseudomonas putida KT2440 transcriptional machinery analysed with a genome-wide DNA microarray Luis Yuste, 1 Ana B. Herv��s, 2 In��s Canosa, 2 Raquel Tobes, 3 Jos�� Ignacio Jim��nez, 4 Juan Nogales, 4 Manuel M. P��rez-P��rez, 5 Eduardo Santero, 2 Eduardo D��az, 4 Juan-Luis Ramos, 3 V��ctor de Lorenzo 1 and Fernando Rojo 1 * 1 Departamento de Biotecnolog��a Microbiana, Centro Nacional de Biotecnolog��a, CSIC, Campus de la Universidad Aut��noma de Madrid, Cantoblanco, 28049 ��� Madrid, Spain. 2 Centro Andaluz de Biolog��a del Desarrollo, Facultad de Ciencias Experimentales, Universidad Pablo de Olavide. Ctra. Utrera, Km. 1. 41013 ��� Sevilla, Spain. 3 Estaci��n Experimental del Zaid��n, CSIC, Profesor Albareda 1, 18008 ��� Granada, Spain. 4 Centro de Investigaciones Biol��gicas, CSIC, Ramiro de Maeztu 9, 28040 ��� Madrid, Spain. 5 Servicio de Gen��mica, Centro Nacional de Biotecnolog��a, CSIC, Campus de la Universidad Aut��noma de Madrid, Cantoblanco, 28049 ��� Madrid, Spain. Summary Bacterial transcriptional networks are built on a hier- archy of regulators, on top of which lie the compo- nents of the RNA polymerase (in particular the sigma factors) and the global control elements, which play a pivotal role. We have designed a genome-wide oli- gonucleotide-based DNA microarray for Pseudomo- nas putida KT2440. In combination with real-time reverse transcription polymerase chain reaction (RT- PCR), we have used it to analyse the expression pat- tern of the genes encoding the RNA polymerase sub- units (the core enzyme and the 24 sigma factors), and various proteins involved in global regulation (Crc, Lrp, Fur, Anr, Fis, CsrA, IHF, HupA, HupB, HupN, BipA and several MvaT-like proteins), during the shift from exponential growth in rich medium into starvation and stress brought about by the entry into stationary phase. Expression of the genes encoding the RNA polymerase core and the vegetative sigma factor decreased in stationary phase, while that of s S increased. Data obtained for s N , s H , FliA and for the 19 extracytoplasmic function (ECF)-like sigma factors suggested that their mRNA levels change little upon entry into stationary phase. Expression of Crc, BipA, Fis, HupB, HupN and the MvaT-like protein PP3693 decreased in stationary phase, while that of HupA and the MvaT-like protein PP3765 increased signifi- cantly. Expression of IHF was indicative of post- transcriptional control. These results provide the first global study of the expression of the transcriptional machinery through the exponential stationary-phase shift in P. putida . Introduction Pseudomonas putida is a ubiquitous Gram-negative bac- terium, metabolically very versatile and adapted to thrive in very diverse habitats. It can be found in soils, aquatic systems or associated to plants (Timmis, 2002). Pseudomonas putida KT2440 is non-pathogenic and has been certified as a safety strain by the Recombinant DNA Advisory Committee. It is widely used as experimental model, as host for gene cloning and expression of heter- ologous genes, and for biotechnological applications such as bioremediation or biotransformations (Wackett, 2003 Jim��nez et al ., 2004 Pieper et al ., 2004). This strain has been extensively characterized at the physiological, bio- chemical and genetic levels. KT2440 is a plasmid-free derivative of P. putida mt-2, which contains the TOL plas- mid pWW0 (Franklin et al ., 1981 Nakazawa, 2002 Regenhardt et al ., 2002). This plasmid harbours a cata- bolic pathway for toluene and xylenes that has been extensively studied from the biochemical and molecular points of view (Ramos et al ., 1997). The sequence of P. putida KT2440 genome has been determined and anno- tated (Nelson et al ., 2002). The genome (6.2 Mb) contains 105 distinguishable genomic islands that provide increased metabolic proficiency as well as defence against several kinds of biotic and abiotic stresses (Weinel et al ., 2002). Its metabolic versatility, as well as the need to adapt to changing environmental conditions, suggests that P. putida should be endowed with sophisticated

166 L. Yuste et al. �� 2005 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology , 8 , 165���177 mechanisms to regulate gene expression. Several obser- vations support this idea. The number of transcriptional regulators is very large in relation to the genome size. Twenty-four sigma factors have been identified or predicted, 19 of which correspond to the subfamily of extracytoplasmic function (ECF) sigma factors (Mart��nez- Bueno et al ., 2002). Known ECF sigma factors coordinate transcription in response to extracytoplasmic stimuli (Raivio and Silhavy, 2001). Genes corresponding to sev- eral global regulators are also present. Genomic DNA microarrays are now available for several bacterial species (Khodursky et al ., 2000 Denef et al ., 2004 Goodman and Lory, 2004 Pappas et al ., 2004 Zhou et al ., 2004 Kang et al ., 2005), and are helping to better understand bacterial physiology and lifestyle from a genome-wide perspective. A DNA array for P. putida has been described based on ~ 2 kb DNA fragments (Stjepan- dic et al ., 2002). However, each spot on this array does not correspond to a defined gene, but to a pair of genome coordinates, which limits its usefulness for gene expres- sion analyses. We have constructed a P. putida genome- wide oligonucleotide-based DNA microarray that contains spots corresponding to all the P. putida KT2440 open reading frames (ORFs) annotated at the NCBI Microbial Genome database, and those of the pWW0 TOL plasmid. In combination with real-time reverse transcription poly- merase chain reaction (RT-PCR) analyses, this microarray was used to visualize the expression profiles of the prin- cipal components of the basic transcriptional machinery when cells leave exponential growth in a complete medium and enter stationary phase. This transition is environmentally relevant, as cells pass from an unre- stricted growth to a situation of nutrient limitation and diverse stresses. In particular, we focused on the genes encoding the different subunits of RNA polymerase (RNAP), including the 24 sigma factors described in this strain, and on genes encoding several global regulators known or presumed to be important for the expression of metabolic pathways. The results provide a global picture of the changes that occur in the expression of the main components of the transcriptional regulatory network when this bacterial strain enters the stationary phase of growth. Results Expression of genes coding for components of the RNAP upon entry into stationary phase The P. putida oligonucleotide-based genomic microarray developed is described in detail in Experimental proce- dures . We used the microarray for analysing the entry into stationary phase of cells growing in batch cultures con- taining a complete LB medium. These growth conditions were chosen because they are the most frequently used in previous reports on the expression of the P. putida transcriptional machinery. All microarray analyses were performed with RNA samples obtained from three inde- pendent cultures grown under identical conditions. The cDNA obtained from each RNA preparation was hybrid- ized to a minimum of two microarray slides. Therefore, each data set corresponds to the average of the data obtained in a minimum of six hybridizations. In addition, the expression profile of many of the genes was analysed as well by real-time RT-PCR on RNA samples obtained from at least two additional different cultures. Total RNAs were prepared from culture samples taken at mid-exponential phase (A 600 of 0.5), late exponential phase (A 600 of 1.2), early stationary phase (A 600 of 2.2) and late stationary phase (A 600 of 4). The transcriptome profiles of cells collected at turbidity values of 1.2, 2.2 and 4 were compared with that of cells collected at mid- exponential phase (A 600 of 0.5). The microarray data were normalized and statistically analysed with the software package ���LIMMA��� (Smyth, 2004), as detailed in Experi- mental procedures . In brief, LIMMA deduces the differen- tial expression values of the genes in the microarray using linear models and moderated t -statistics using the empir- ical Bayes approach. The probability values obtained ( P - values) were adjusted for multiple testing to control the false discovery rate (Benjamini and Hochberg, 1995). In this work we have concentrated in a detailed analysis of the changes detected for the genes corresponding to the RNAP and its sigma factors, as well as for several known or predicted global regulators or chromatin-associated proteins that affect gene expression. The results obtained are summarized in Table 1. The mRNA levels corresponding to genes of the RNAP core ( a , b , b�� and w ) and to all the sigma factors were very similar in cells collected at the mid-exponential (A 600 of 0.5) and at the late-exponential (A 600 of 1.2) phases of growth, showing small fold changes and, in many cases, relatively high P -values. However, the mRNA levels for many of the RNAP genes clearly changed upon entry into stationary phase (A 600 of 2.2) and in late stationary phase (A 600 of 4), with P -values that were very low in most cases (Table 1 and Fig. 1). These low P -values are indicative of a high probability of these genes being differentially expressed in the two conditions considered. In the case of the genes corresponding to the RNAP core ( rpoA, rpoB, rpoC and rpoZ, encoding the a -, b -, b�� - and w - subunits respectively), the microarray showed a modest decrease (about 1.5-fold) in their mRNA levels upon entry into early stationary phase. The levels of rpoA and rpoB diminished close to fourfold in late stationary phase, although for rpoC and rpoZ the decrease was lower. Expression of rpoA and rpoB was further analysed by quantitative real-time RT-PCR. The results confirmed that