Phylogenetic Classification of Pseudomonas putida Strains by MALDI-MS Using Ribosomal Subunit Proteins as Biomarkers Kanae Teramoto, † Hiroaki Sato,* ,† Liwei Sun, † Masaki Torimura, † Hiroaki Tao, † Hiromichi Yoshikawa, ‡ Yudai Hotta, § Akifumi Hosoda, § and Hiroto Tamura § Research Institute for Environmental Management Technology, National Institute of Advanced Industrial Science and Technology (AIST), 16-1 Onogawa, Tsukuba, Ibaraki 305-8569, Japan, Fukuoka Institute of Technology, 3-30-1 Wajirohigashi, Higashi-ku, Fukuoka 811-0295, Japan, and Department of Agriculture, Meijo University, 1-501 Shiogamaguchi, Tenpaku-ku, Nagoya, Aichi 468-8502, Japan A new method for phylogenetic classification of bacterial strains using matrix-assisted laser desorption/ionization- mass spectrometry (MALDI-MS) is proposed. This method was developed using a bioinformatics-based approach to the rapid identification of bacteria as previously proposed by Demirev and co-workers, which uses ribosomal pro- teins composed of ∼50 subunit proteins as biomarkers. Although the amino acid sequences of ribosomal proteins are highly conserved, slight sequence variations can occur at the strain level. Since ribosomal subunit proteins are a complex of housekeeping proteins that have different phylogenetic evolution rates, sequence variation detected as mass differences by MALDI-MS may be useful for the phylogenetic classification of bacteria at strain level. In our proposed method, the first step is the selection of reliable biomarkers through characterization of the ex- pressed ribosomal subunit proteins of a reference strain (usually a genome-sequenced strain) by MALDI-MS. The observed masses in the MALDI mass spectra of cell lysates of sample strains are then compared with the biomarker masses of the reference strain. The biomarkers for each sample strain were designated as present or absent at the reference masses, indicated by 1 or 0, respectively, which were summarized in a table. This table is processed by cluster analysis, generating a phylogenetic tree. In this study, the success of this approach was confirmed by classification of Pseudomonas putida strains because its classification is much more compli- cated than that of other bacterial strains. Forty-three reliable biomarkers were selected from ribosomal sub- unit proteins of a genome-sequenced strain, P. putida KT2440. The numbers and kinds of biomarkers observed for 16 strains of P. putida, including different biovars, were markedly different, reflecting the variety of the strains. The classification results by the proposed method were highly comparable to those based on the DNA gyrase subunit B gene (gyrB) sequence analysis, suggesting our proposed method would be a useful high-throughput method for phylogenetic classification of newly isolated bacteria. The classification or typing of bacterial strains is of key importance in the fields of environmental microbiology, molecular epidemiology, and the fermentation industry. Two DNA sequence- based methods, DNA fragment analyses by electrophoresis and DNA sequence analyses, are currently employed for the genotypic classification of bacteria. Pulsed-field gel electrophoresis, 1 ran- domly amplified polymorphic DNA, 2 amplified fragment length polymorphism, 3 and ribotyping 4 are the chief forms of DNA fragment analysis used for characterization of bacterial strains. Since fragment analyses rely on gel electrophoretic patterns of DNA fragments by restriction enzyme treatment, band broadening and poor reproducibility of electrophoretic mobility risk impeding the acquisition of reproducible results. Automated DNA sequence analysis is, nowadays, a powerful method for classifying bacteria. The 16S rRNA gene sequence analysis is generally used for classifying bacteria at the genus and species levels. To classify bacteria at the subspecies and strain levels, the use of housekeeping genes with a higher evolution rate, such as the DNA gyrase subunit B gene (gyrB) 5,6 and the RNA polymerase σ 70 factor gene (rpoD) 7 have also been proposed. The limitation of a single gene sequence analysis has, however, been pointed out for the genotypic classification of nonclonal population whose genes are modified by horizontal transmission and homologous recombination. 8-11 * To whom correspondence should be addressed. Fax: +81 (29) 861-8308. E-mail: sato-hiroaki@aist.go.jp. † AIST. ‡ Fukuoka Institute of Technology. § Meijo University. (1) Olson, M. V. J. Chromatogr. 1989, 470, 377-383. (2) Bingen, E.; Barc, M. C.; Brahimi, N.; Vilmer, E.; Beaufils, F. J. Clin. Microbiol. 1995, 33, 1657-1659. (3) Keim, P.; Kalif, A.; Schupp, J.; Hill, K.; Travis, S. E.; Richmond, K.; Adair, D. M.; HughJones, M.; Kuske, C. R.; Jackson, P. J. Bacteriol. 1997, 179, 818-824. (4) Faruque, S. M.; Roy, S. K.; Alim, A. R. M. A.; Siddique, A. K.; Albert, M. J. J. Clin. Microbiol. 1995, 33, 2833-2838. (5) Yamamoto, S.; Harayama, S. Appl. Environ. Microbiol. 1995, 61, 1104- 1109. (6) Yamamoto, S.; Harayama, S. Int. J. Syst. Bacteriol. 1998, 48, 813-819. (7) Lonetto, M.; Gribskov, M.; Gross, C. A. J. Bacteriol. 1992, 174, 3843- 3849. (8) Urwin, R.; Maiden, M. C. J. Trends Microbiol. 2003, 11, 479-487. Anal. Chem. 2007, 79, 8712-8719 8712 Analytical Chemistry, Vol. 79, No. 22, November 15, 2007 10.1021/ac701905r CCC: $37.00 © 2007 American Chemical Society Published on Web 10/16/2007