Global Analysis of the Membrane Subproteome of Pseudomonas aeruginosa Using Liquid Chromatography-Tandem Mass Spectrometry Josip Blonder, † Michael B. Goshe, ‡ Wenzhong Xiao, § David G. Camp II, | Mark Wingerd, | Ronald W. Davis, § and Richard D. Smith* ,| Biological Sciences Division and Environm ental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, P.O. Box 999, Richland, Washington 99352 Received September 10, 2003 Pseudomonas aeruginosa is one of the most significant opportunistic bacterial pathogens in humans causing infections and premature death in patients with cystic fibrosis, AIDS, severe burns, organ transplants, or cancer. Liquid chromatography coupled online with tandem mass spectrometry was used for the large-scale proteomic analysis of the P. aeruginosa membrane subproteome. Concomitantly, an affinity labeling technique, using iodoacetyl-PEO biotin to tag cysteinyl-containing proteins, permitted the enrichment and detection of lower abundance membrane proteins. The application of these approaches resulted in the identification of 786 proteins. A total of 333 proteins (42%) had a minimum of one transmembrane domain (ranging from 1 to14) and 195 proteins were classified as hydrophobic based on their positive GRAVY values (ranging from 0.01 to 1.32). Key integral inner and outer membrane proteins involved in adaptation and antibiotic resistance were conclusively identified, including the detection of 53% of all predicted opr-type porins (outer integral membrane proteins) and all the components of the mexA-mexB-oprM transmembrane protein complex. This work represents one of the most comprehensive proteomic analyses of the membrane subproteome of P. aeruginosa and for prokaryotes in general. Keywords: proteome • membrane proteins • low abundance • LC-MS/MS • affinity labeling Introduction Pseudomonas aeruginosa is a gram-negative bacterium of increasing importance because it is one of the top three opportunistic pathogens in humans causing premature death in patients with cystic fibrosis, HIVinfection, organ transplants, or cancer. 1 It is the most common cause of various nosocomial infections including sepsis in burned or extensively injured patients. Its intrinsic adaptation ability complicates therapeutic strategies even more, with the emerging occurrence of P. aeruginosa strains exhibiting a high level of multidrug resis- tance to several classes of antibiotics. 2 Infections caused by this pathogen have an adverse impact on the mortality and economics of the above-mentioned diseases and clinical conditions. 3-8 The complexity of pathophysiological processes involved in P. aeruginosa pathogenicity is emphasized by the diversity of the diseases associated with this organism and is indicative of the multiplicity of virulence factors that it is able to produce. 9-14 Consequently, there is significant interest in the proteins responsible for this virulence, particularly those of the cellular membrane. 15 The phenotypic resistance of gram-negative bacteria is a direct consequence of the complex structure of their cell envelope, which acts as a barrier and prevents drug molecules from reaching their target sites or increases their active efflux, primarily mediated by various classes of integral membrane proteins. 16,17 It has been indicated that antibiotic resistance is due to integral outer membrane protein channels (porins) and membrane protein complexes known as multidrug efflux pumps that transport antibiotic and biocide molecules out of the cell. 18 Currently, about seven hundred genes are implicated in the infection process with a significant number of them expected to code for membrane and secretory proteins. 19 Large-scale analysis of hydrophobic integral membrane proteins from complex protein mixtures is an important and challenging aspect of mass spectrometry-based membrane proteomics. Although, large-scale proteomic studies based on in-gel digestion of proteins separated using two-dimensional polyacrylamide gel electrophoresis (2-D PAGE) have recently shown significant improvements in protein coverage, very hydrophobic integral membrane proteins are generally not amenable using this approach primarily due to the issues *To whom correspondence should be addressed. E-mail: rds@pnl.gov. † SAIC-Frederick Inc., Laboratory for Proteomics and Analytical Technolo- gies, Mass Spectrometry Center, National Cancer Institute at Frederick, P.O. Box B, Frederick, MD 21702. Phone: (301) 846-7211. ‡ Department of Molecular and Structural Biochemistry, North Carolina State University, 128 Polk Hall, Campus Box 7622, Raleigh, NC 27695. § Stanford University School of Medicine, Stanford Genome Technology Center, Stanford, CA 94305. | Environmental Molecular Sciences Laboratory, P.O. Box 999, MSIN: K8- 98, Richland, WA, 99352. 434 Journal of Proteome Research 2004, 3, 434-444 10.1021/pr034074w CCC: $27.50  2004 American Chemical Society Published on Web 02/07/2004