RESEARCH LETTER
Quantitative proteomics of Chlorobaculum tepidum: insights into the sulfur metabolism of a phototrophic green sulfur bacterium Lasse G. Falkenby1, Monika Szymanska1, Carina Holkenbrink2, Kirsten S. Habicht3, Jens S. Andersen1, Mette Miller1 & Niels-Ulrik Frigaard2 1
Department of Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark; 2Department of Biology, University of Copenhagen, Helsingør, Denmark; and 3Institute of Biology, University of Southern Denmark, Odense, Denmark
Correspondence: Niels-Ulrik Frigaard, Section for Marine Biology, Department of Biology, University of Copenhagen, Strandpromenaden 5, DK-3000 Helsingør, Denmark. Tel.: +45 35 32 19 57; fax: +45 35 32 19 51; e-mail:
[email protected] Received 21 May 2011; revised 19 July 2011; accepted 25 July 2011. Final version published online 25 August 2011. DOI: 10.1111/j.1574-6968.2011.02370.x
MICROBIOLOGY LETTERS
Editor: Christiane Dahl Keywords quantitative proteomics; photosynthetic bacteria; sulfur metabolism; dissimilatory sulfite reductase; SOX enzyme system; sulfite oxidation.
Abstract Chlorobaculum (Cba.) tepidum is a green sulfur bacterium that oxidizes sulfide, elemental sulfur, and thiosulfate for photosynthetic growth. To gain insight into the sulfur metabolism, the proteome of Cba. tepidum cells sampled under different growth conditions has been quantified using a rapid gel-free, filteraided sample preparation (FASP) protocol with an in-solution isotopic labeling strategy. Among the 2245 proteins predicted from the Cba. tepidum genome, approximately 970 proteins were detected in unlabeled samples, whereas approximately 630–640 proteins were detected in labeled samples comparing two different growth conditions. Wild-type cells growing on thiosulfate had an increased abundance of periplasmic cytochrome c-555 and proteins of the periplasmic thiosulfate-oxidizing SOX enzyme system when compared with cells growing on sulfide. A dsrM mutant of Cba. tepidum, which lacks the dissimilatory sulfite reductase DsrM protein and therefore is unable to oxidize sulfur globules to sulfite, was also investigated. When compared with wild type, the dsrM cells exhibited an increased abundance of DSR enzymes involved in the initial steps of sulfur globule oxidation (DsrABCL) and a decreased abundance of enzymes putatively involved in sulfite oxidation (Sat-AprAB-QmoABC). The results show that Cba. tepidum regulates the cellular levels of enzymes involved in sulfur metabolism and other electron-transferring processes in response to the availability of reduced sulfur compounds.
Introduction Quantitative mass spectrometry-based proteomics has become widely used for examining differences in global expression level of proteins in various cellular states (Bantscheff et al., 2007; Elliott et al., 2009; Walther & Mann, 2010). In this method, proteins from samples obtained from different experimental conditions can be distinguished by incorporation of unique, stable isotopes with disparate masses in one of the samples. In this way, various samples can be combined and analyzed in a single LC-MS/MS analysis allowing estimation of the relative intensities of the peptides of interest from the labeled and FEMS Microbiol Lett 323 (2011) 142–150
unlabelled samples. Metabolic (Ong et al., 2002) and chemical (Boersema et al., 2009) labeling are two common procedures used for introducing heavy isotopes into cellular proteins. A pre-requisite for metabolic labeling of proteins is that the cells efficiently take up a labeled substrate in culture and incorporate it into proteins. However, this approach does not always result in a sufficient degree of labeling. Alternatively, as used in the present work, isotopic labeling can be performed by chemical labeling of peptides resulting from post-digestion of the cellular protein fractions. The green sulfur bacterium (GSB) Chlorobaculum (Cba.) tepidum is a strictly anaerobic, photosynthetic ª 2011 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved
143
Quantitative proteomics of Chlorobaculum tepidum
bacterium that lives in anaerobic aquatic environments, where reduced sulfur compounds, predominantly sulfide and light occur at the same time (Wahlund et al., 1991; Overmann, 2008). Chlorobaculum tepidum oxidizes sulfide, elemental sulfur, and thiosulfate for use as electron donor in its photosynthesis. The 2.15-Mbp genome of Cba. tepidum has been sequenced and revealed about 2245 protein-encoding genes (Eisen et al., 2002). Currently, 15 genome sequences of GSB have been determined (Gregersen et al., 2011). This information has allowed a detailed analysis of the sulfur metabolism of GSB, but many processes are still poorly described (Frigaard & Bryant, 2004, 2008; Frigaard & Dahl, 2009; Sakurai et al., 2010). Table 1 lists 57 enzymes putatively involved in the oxidative sulfur metabolism of Cba. tepidum, some of which have been functionally investigated. Figure 1 shows a simplified scheme of the pathways and enzymes of the oxidative sulfur metabolism of Cba. tepidum. Sulfide is oxidized by sulfide:quinone oxidoreductases (SQR; Chan et al., 2009); additional unknown enzyme activity contributes to sulfide oxidation (Holkenbrink et al., 2011). Thiosulfate is oxidized exclusively by the sulfur oxidation (SOX) enzyme system in the periplasm (Ogawa et al., 2008, 2010; Azai et al., 2009). Both of these processes give rise to a putative oligosulfide pool, which presumably is in equilibrium with an extracellular pool of sulfur globules that sometimes is referred to as ‘elemental sulfur’ (‘S0’). Oxidation of the oligosulfide pool is dependent on the dissimilatory sulfite reductase (DSR) enzyme system (Holkenbrink et al., 2011), which consists of at least 15 subunits (Table 1). The DSR system presumably forms intracellular sulfite that is oxidized by an enzyme system consisting of Sat, Apr, and Qmo proteins (Rodriguez et al., 2011). The electron acceptors, cytochrome c, and menaquinone (Fig. 1) are ultimately oxidized by the photosynthetic reaction center. In cultures of Cba. tepidum that contain both sulfide and thiosulfate, sulfide is oxidized preferentially while sulfur globules are formed (Chan et al., 2008; Azai et al., 2009; Holkenbrink et al., 2011). Following sulfide depletion, thiosulfate and sulfur globules are oxidized to sulfate. The molecular mechanism of this phenomenon is poorly understood. Sulfide possibly inhibits thiosulfate oxidation either by substrate competition between sulfide and thiosulfate (the SOX system oxidizes sulfide in vitro; Ogawa et al., 2010) or by saturation of the electron acceptor pool. Regulation of sulfur metabolism genes in GSB is poorly described, but it is known that SoxA is induced by thiosulfate in Chlorobaculum thiosulfatiphilum (Verte´ et al. 2002). In the purple sulfur bacterium, Allochromatium vinosum, sox and dsr genes are expressed at a low constitutive level in the absence of reduced sulfur FEMS Microbiol Lett 323 (2011) 142–150
substrates and are induced by thiosulfate and sulfide, respectively (Grimm et al., 2010, 2011).
Materials and methods Cell cultures and chemical measurements
Chlorobaculum tepidum TLS was grown under incandescent illumination in CL medium (Frigaard et al., 2004). For experiments comparing early and late exponential growth phase, wild-type cells were grown at 45 °C in 1-L flasks under a light intensity of 200 lmol photons m 2 s 1. For experiments comparing wild type and the dsrM mutant (Holkenbrink et al., 2011), cells were grown at 42 °C in 15-mL tubes under a light intensity of 50 lmol photons m 2 s 1 and harvested in the late exponential growth phase. Cells were harvested by centrifugation and stored at 20 °C prior to analysis. Bacteriochlorophyll c was determined by extracting the cell pellet with acetone : methanol (7 : 2 by vol) (Frigaard et al., 1997). Sulfide was measured using the colorimetric methylene blue method (Cline 1969). Sulfate and thiosulfate were measured by ion chromatography (Dionex, Hvidovre, Denmark) using a carbonate buffer as eluent. Samples for analysis of elemental sulfur were dissolved in methanol and analyzed as S8 by HPLC using a Sykam pump (S1100), UV–VIS detector (Sykam S3200), Zorbax ODS-column (125 9 4 mm, 5 lm; Knauer, Berlin, Germany) and methanol as the eluent at a flow rate of 1 mL min 1. Elemental sulfur was detected at 265 nm. Protein preparation for proteome analysis
Cell pellets were thawed and extracted in acetone : methanol (7 : 2 by vol) to remove pigments. The colorless cell pellets were solubilized in an SDS-containing buffer (Laemmli, 1970) supplemented with a complete protease inhibitor cocktail (Roche, Hvidovre, Denmark) at 95 °C for 3–5 min and cleared by centrifugation. Prior to digestion, proteins were reduced with dithiothreitol (1 mM) and alkylated with iodoacetamide (5 mM). The filter-aided sample preparation (FASP) protocol was used essentially as described (Wisniewski et al., 2009). Protein extract (20 lL) was mixed with solution UA (200 lL; 8 M urea in H2O, pH 8.5). This solution was loaded onto a 10-kDa cut-off filter spin filter and centrifuged (14 000 g, 40 min). The retentate was washed three times with solution UA and the flow-through discarded. Then a solution of iodoacetamide (100 lL; 0.05 M in-solution UA) was added to the filter and incubated for 5 min. The filters were then centrifuged (14 000 g, 30 min) and washed twice with a urea solution (100 lL; 8 M in H2O, pH 8.0). After each wash, the filter units were centrifuged (14 000 g; 40 min). ª 2011 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved
144
L.G. Falkenby et al.
Table 1. Gene products in Chlorobaculum tepidum known or suggested to be involved in oxidative sulfur metabolism* Detected in unlabeled samples†
Detected in labeled samples†
References
Membrane-bound cytochrome c-555, protein CycB Soluble cytochrome c-555, protein CycA (periplasmic) Sulfide:quinone oxidoreductase SqrD
Yes Yes
Yes Yes
Azai et al. (2009)
Yes
Yes
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
No No No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Yes
Yes
Rodriguez et al. (2011)
Yes
Yes
Rodriguez et al. (2011)
CT0876
Polysulfide-reductase-like enzyme, subunit PsrC Polysulfide-reductase-like enzyme, subunit PsrB Polysulfide-reductase-like enzyme, subunit PsrA Dissimilatory sulfite reductase protein DsrC, copy 1‡ Dissimilatory sulfite reductase protein DsrA, copy 1‡ Dissimilatory sulfite reductase protein DsrB, copy 1‡ Dissimilatory sulfite reductase protein DsrL, copy 1‡ Dissimilatory sulfite reductase protein DsrE Dissimilatory sulfite reductase protein DsrF Dissimilatory sulfite reductase protein DsrH Sulfate adenylyltransferase, protein Sat Adenylylsulfate reductase, subunit AprB Adenylylsulfate reductase, subunit AprA Quinone-modifying oxidoreductase, subunit QmoA Quinone-modifying oxidoreductase, subunit QmoB Quinone-modifying oxidoreductase, subunit QmoC Sulfide:quinone oxidoreductase SqrE§
Yes
Yes
CT1015
Sulfur-oxidizing protein SoxJ
Yes
Yes
CT1016 CT1017 CT1018 CT1019 CT1020 CT1021
Sulfur-oxidizing Sulfur-oxidizing Sulfur-oxidizing Sulfur-oxidizing Sulfur-oxidizing Sulfur-oxidizing
Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes
Chan et al. (2009), Gregersen et al. (2011) Ogawa et al. (2010), Gregersen et al. (2011) Ogawa et al. (2008, 2010) Ogawa et al. (2008, 2010) Ogawa et al. (2008, 2010) Ogawa et al. (2008, 2010) Ogawa et al. (2008, 2010) Azai et al. (2009), Ogawa et al. (2008, 2010)
CT1023 CT1025 CT1087
Sulfur-oxidizing protein SoxW Sulfide dehydrogenase (SoxF/SqrC homolog) Sulfide:quinone oxidoreductase SqrF
Yes Yes No
Yes Yes No
CT1245
Heterodisulfide-reductase-like enzyme, subunit D Heterodisulfide-reductase-like enzyme, subunit A Heterodisulfide-reductase-like enzyme component Heterodisulfide-reductase-like enzyme component Heterodisulfide-reductase-like enzyme, subunit B Heterodisulfide-reductase-like enzyme, subunit G RuBisCO-like protein (RLP) Sulfhydrogenase-like enzyme, subunit B Sulfhydrogenase-like enzyme, subunit G Sulfhydrogenase-like enzyme, subunit D Sulfhydrogenase-like enzyme, subunit A
Yes
Yes
Yes
Yes
Yes Yes
Yes No
Yes
Yes
Yes
Yes
Yes No No No Yes
Yes No No No Yes
Locus
Name
CT0073 CT0075 CT0117 CT0494 CT0495 CT0496 CT0851 CT0852 CT0853 CT0854 CT0855 CT0856 CT0857 CT0862 CT0864 CT0865 CT0866 CT0867 CT0868
CT1246 CT1247 CT1248 CT1249 CT1250 CT1772 CT1891 CT1892 CT1893 CT1894
protein protein protein protein protein protein
SoxX SoxY SoxZ SoxA SoxK SoxB
ª 2011 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved
Chan et al. (2009), Holkenbrink et al. (2011)
Holkenbrink Holkenbrink Holkenbrink Holkenbrink
et et et et
al. al. al. al.
(2011) (2011) (2011) (2011)
Gregersen et al. (2011) Chan et al. (2009), Holkenbrink et al. (2011)
Hanson and Tabita (2001)
FEMS Microbiol Lett 323 (2011) 142–150
145
Quantitative proteomics of Chlorobaculum tepidum
Table 1. Continued
Locus
Name
Detected in unlabeled samples†
Detected in labeled samples†
CT2080 CT2081 CT2238 CT2239 CT2240 CT2241 CT2242 CT2243 CT2244 CT2245 CT2246 CT2247 CT2248 CT2249 CT2250 CT2251
Flavocytochrome c, subunit FccA Flavocytochrome c, subunit FccB Dissimilatory sulfite reductase protein DsrW Dissimilatory sulfite reductase protein DsrV Dissimilatory sulfite reductase protein DsrP Dissimilatory sulfite reductase protein DsrO Dissimilatory sulfite reductase protein DsrJ Dissimilatory sulfite reductase protein DsrK Dissimilatory sulfite reductase protein DsrM Dissimilatory sulfite reductase protein DsrT Dissimilatory sulfite reductase protein DsrU Dissimilatory sulfite reductase protein DsrL, copy 2‡ Dissimilatory sulfite reductase protein DsrB, copy 2‡,¶ Dissimilatory sulfite reductase protein DsrA, copy 2‡ Dissimilatory sulfite reductase protein DsrC, copy 2‡ Dissimilatory sulfite reductase protein DsrN
No No Yes No No Yes Yes Yes Yes No Yes Yes Yes Yes Yes No
No No No No No Yes No Yes No No Yes Yes Yes Yes Yes No
References
Holkenbrink et al. (2011) Holkenbrink et al. (2011)
Holkenbrink Holkenbrink Holkenbrink Holkenbrink Holkenbrink Holkenbrink Holkenbrink
et et et et et et et
al. al. al. al. al. al. al.
(2011) (2011) (2011) (2011) (2011) (2011) (2011)
*Based on Eisen et al. (2002), Frigaard & Bryant (2008), and references in the table. This work. ‡ The DsrA, DsrB, and DsrC proteins are each encoded by two nearly identical gene copies resulting in identical proteins, whereas two dsrL genes encode DsrL proteins with 99.2% amino acid sequence identity. § SQR activity has not been demonstrated for this protein (Chan et al., 2009). ¶ A frameshift mutation in gene CT2248 was reported by Eisen et al. (2002). However, our Cba. tepidum strain has a restored reading frame (Holkenbrink et al., 2011). †
H2S
“S0”
2–
MK
SQR
SO4 Cyt c
SOX
2–
Sn
Periplasm
2–
S2O3
DsrTMKJOP
Cytoplasm
DsrABLU DsrEFHC
MK 2–
SO3
QmoABC AprAB Sat
2–
SO4
Fig. 1. Simplified overview of the predicted pathways and enzymes of the oxidative sulfur metabolism in Chlorobaculum tepidum (Frigaard & Bryant, 2008; Holkenbrink et al., 2011). Cyt c, cytochrome c; MK, menaquinone.
Chemical labeling of peptides
Dimethyl labeling was performed essentially as described by Boersema et al. (2009). Briefly, the isolated proteins on the filter device were subjected to a Lys-C digestion. The resulting peptides were reconstituted in 100 mM TEAB buffer (Sigma, St. Louis, MO). Samples for ‘light’ labeling were mixed with formaldehyde (4% in H2O; Sigma). Samples for ‘heavy’ labeling were mixed with formaldehyde-D2 (4% in H2O; Sigma). Both samples were FEMS Microbiol Lett 323 (2011) 142–150
then mixed with freshly prepared sodium cyanoborohydride (0.6 M). After incubation for 1 h at room temperature, the reaction was quenched with ammonia solution (1% v/v) and TFA. The acidified samples were desalted on StageTips made from C18 disks excised from Empore High Performance Extraction Disks (3M, St. Paul, MN) in a pipette tip (Rappsilber et al. 2007). Peptide purification and mass spectrometry
Peptide mixtures were separated by nanoLC using an Agilent 1200 nanoflow system connected to either an LTQ Orbitrap XL or LTQ FT Ultra mass spectrometer (both from Thermo Electron, Bremen, Germany) equipped with a nanoelectrospray ion source (Proxeon Biosystem, Odense, Denmark). Chromatographic separation of the peptides took place in an in-house packed 20 cm fused silica emitter (75-lm i.d.) with reverse-phase ReproSil-Pur C18-AQ (3 lm) resin (Maisch GmbH, Ammerbuch-Entringen, Germany). Peptides were injected onto the column (flow rate 500 nL min 1) and eluted with a flow of 250 nL min 1 from 5% to 40% acetonitril in 0.5% acetic acid over 2 h. A ‘top 6’ acquisition method was set up on the mass spectrometer, utilizing the high mass accuracy of the Orbitrap for intact peptides and the speed and sensitivity ª 2011 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved
146
of the LTQ (iontrap) for fragment spectra. The initial scan event was the intact peptide mass spectrum in the Orbitrap with range m/z 300–1800 and resolution R = 60 000 at m/z 400. Six CID fragmentation spectra in the iontrap (AGC target 5000, maximum injection time 150 ms) of the six most intense ions from the Orbitrap scan were recorded. Dynamic exclusion (2.5 min) and charge state screening requiring charge 2+ or more were enabled. The obtained tandem MS spectra were matched against theoretical spectra from a protein sequence database derived from the Cba. tepidum genome (GenBank acc. no. NC_002932) using Mascot (Matrix Science Ltd; www. matrixscience.com). Results were validated using the MSQUANT software package (Mortensen et al., 2010) requiring minimum two unique peptides per protein and minimum six amino acids per unique peptide. In silico analyses showed that a maximum 2159 of the 2245 proteins (96%) encoded by the Cba. tepidum genome are theoretically detectable using this approach. Nearly all theoretically undetectable proteins were small hypothetical proteins (2 or
