Shotgun proteomics of Orientia tsutsugamushi

Abstract

Orientia tsutsugamushi is one of the obligate intracellular bacteria, and causes Tsutsugamushi disease (scrub typhus). The whole genome sequence of Boryong strain revealed some characteristics of this bacterium, such as a type IV secretion system, many histidine kinases, SpoT, Tra, and ankyrin repeatand tetratricopeptide repeat (TPR)-containing proteins [1]. Additionally, the structure of the genome of O. tsutsugamushi was completely different from those of other rickettsia that have been previously reported. We performed a combination assay of SDS-PAGE and LC-MSMS (GeLC-MSMS) to identify expressed proteins and study the protein profile of this bacterium. The Kuroki strain of O. tsutsugamsuhi, which is very close to the Boryong strain, was propagated in L929 cells and purified by the Renografin density method. About 50 lg of the purified rickettsia was subjected to SDS-PAGE, and the gel was cut into about 40 pieces. Then, the pieces of gel were subjected to in-gel digestion with trypsin or endoproteinase Lys-C. Trypsin digests proteins at the arginine and lysine sites, whereas Lys-C digests only at the lysine site. Finally, the digested peptides were subjected to LC-MSMS analysis (Magic2002 Michrom Biosource Inc., Auburn, CA, USA and LCQ-DECAXP Thermo Fisher Scientific, San Jose, CA, USA). Two different procedures were used, involving different sizes of column (InertsilODS3; GL Science Inc., Shinjuku, Tokyo, Japan) and acetonitrile concentration gradient times: a 0.1 mm · 5 cm column and a 60-min gradient time; and a 0.1 mm · 15 cm column and a 120-min gradient time. For data analysis, proteins were identified using the TurboSequest algorithm in the Bioworks 3.1 package software (Thermo fisher Scientific) and the O. tsutsugamushi database (NCBI, National Center for Biotechnology Information, Bethesda, MD, USA). The criteria for positive identification of peptides were cross-correlation number (Xcorr) >1.0 for singly charged ions, Xcorr >1.5 for doubly charged ions, and Xcorr >2.0 for triply charged ions. Only the best-matched peptides were considered. The strategy using both trypsin and Lys-C for in-gel digestion and two sets of columns and gradient times for LC-MSMS had the great advantage of identifying proteins, especially proteins of lower molecular mass (<50 kDa). One hundred and sixty-five proteins were identified by both trypsin and Lys-C digestion, 68 proteins only by Lys-C, and 351 only by trypsin digestion. Finally, the strategy allowed us to identify 584 of 1152 proteins that are annotated on the genome of O. tsutsugamsuhi (49.4%). The identified proteins included 67.9% (127 ⁄ 187) of essential proteins of the potential minimum bacterium as previously proposed by Gil et al. [2] (Table 1). The identified proteins were classified into COG functional categories [3,4] and compared with those of other rickettsia and bacteria [5]. Comparatively, in O. tsutsugamushi, the category ‘translation’ was higher than the other categories, whereas categories of the entire metabolism were not as high as in other rickettsia and obligate intracellular