Direct Proteomic Mapping of Streptomyces Luteogriseus Strain 103

Oct 28, 2005 - Effect of potato suberin on Streptomyces scabies proteome. ANNIE LAUZIER , ANNE-MARIE SIMAO-BEAUNOIR , SYLVIE BOURASSA , GUY ...
0 downloads 0 Views 332KB Size
Download PDF
Direct Proteomic Mapping of Streptomyces Luteogriseus Strain 103 and cnn1 and Insights into Antibiotic Biosynthesis Yu-xia Wang and Ying-Jin Yuan* Department of Pharmaceutical Engineering, School of Chemical Engineering and Technology, Tianjin University, P.O. Box 6888, Tianjin 300072, Peoples Republic of China Received June 9, 2005

Global proteome of the antibiotic-production strain 103 and the nonantibiotic-production mutant cnn1 were directly analyzed using 2D LC-MS/MS. A total of 726 and 809 proteins have been identified, respectively. Physical and chemical characterization, subcellular location and functional classification of the global protein were carried out. By searching the key enzymes of several probable antibiotic biosynthesis pathways in the identified proteins of strain 103, only polyketide synthase was found, which suggested that Maituolaimysin be synthesized through polyketide pathway. The same searching result was obtained in strain cnn1, which confirmed the conclusion drew from strain 103. Other proteins associated with the polyketide pathway of the two strains were searched according to the protein classification scheme of Streptomyces coelicolor (available at http://www.sanger.ac.uk/Projects/ S_coelicolor/) and most of them were found. The activity inhibition of beta-ketoacyl ACP synthase, a key enzyme in the polyketide pathway, directly resulted in the decrease of Maituolaimysin production, which proved the conclusion obtained in the proteomic research. Keywords: proteomics • 2D LC-MS/MS • Streptomyces luteogriseus • polyketide pathway • microlide

Introduction Streptomycetes are the most important gram-positive mycelial soil bacteria, which can produce a range of diverse secondary metabolites with important applications in human or veterinary medicine and agriculture as antitumor, antibacterial and antifungal agents.1 One new strain of Streptomyces luteogriseus, which produces a promising antibiotic named Maituolaimysin with promising activity of anti-HIV PR (IC50 at 39.8 µg/mL) and anti-Coxsackie-virus B6 (IC50 at 231.1 µg/mL),2 was isolated from soil in our laboratory. One set of nonantibiotic-production mutant cnn1 with heredity stability was obtained from strain 103 by dealing with LiCl solution and UV radiation. Although Maituolaimysin produced by Streptomyces luteogriseus has the activities mentioned above, the yield from the original strain is not yet very high. It is very difficult to sharply improve the antibiotic production only by the traditional fermentation optimization and mutagenesis treatment. So it is urgent for us to make clear the Maituolaimysin biosynthesis pathway to further improve its production. There are about five major antibiotic biosynthesis pathways including shikimate, polyketide, peptide synthesizing pattern, glyco-diriving and mevalonate pathway.3-7 For each pathway, there are several special enzymes catalyzing the reactions of it. Proteomic technologies make it possible to characterize and identify many more proteins expressed by a strain, cell, or tissue under defined conditions than was capable just a few years * To whom correspondence should be addressed. Tel: 86-22-27401149. 27403888. Fax: 86-22-27403888. E-mail: [email protected] or yjyuan@ tju.edu.cn. 10.1021/pr050169h CCC: $30.25

 2005 American Chemical Society

ago, which can provide the protein expression map conveniently to find out proteins we interested. The objective of the proteome characterization of Streptomyces luteogriseus was to obtain comprehensive protein expression profiles so as to find out the enzymes catalyzing the Maituolaimysin biosynthesis. The profiling of mutant cnn1 is to confirm the conclusions drew from the strain 103. By searching the key enzymes of the possible pathways in the total protein profiles of Streptomyces luteogriseus, Maituolaimysin biosynthesis related enzymes and pathway would be discovered. By comparing the searching results of strain 103 and cnn1, the pathway would be confirmed. That would be great help to improve the production of Maituolaimysin. To date, microorganism proteomics research mainly focus on the “model” microbes or the microorganisms whose genome is completely sequenced.8-10 After the accomplishment of genome sequencing of Streptomyces coelicolor A3(2), the study of overall gene expression at the mRNA and protein level was performed immediately.11 However, Streytomyces proteome researchers mainly use the traditional method by using two-dimensional gel electrophoresis (2DE) coupled with matrixassisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry.12-15 Although this method is widely used in protein separation and identification, it has significant limitations. First, it is impossible to analyze the entire proteome. Proteins with extreme pI and molecular weight, hydrophobic proteins and low abundance proteins are rarely seen in 2DE based study. Second, 2DE is time and labor consuming and it is not easy to automate. Third, its reproducibility is poor.16 Journal of Proteome Research 2005, 4, 1999-2006

1999

Published on Web 10/28/2005

research articles A nongel based “shotgun” protein identification technology has been developed and named Multidimensional Protein Identification Technology (MudPIT).17,18 In this method, the protein mixture extracted from the cells or tissues is digested first and loaded into the columns to be separated, then the separated peptides flow directly into the mass spectrum. MudPIT is a direct and rapid protein identification method and can overcome the 2DE shortcoming mentioned above. It has undergone a rapid development and been extensively used since it appeared.19,20 In this study, we analyzed the whole cell protein extraction of the strain 103 and cnn1 using MudPIT and identified a total of 726 and 809 proteins, respectively. For the first time, proteins and biosynthesis pathway associated with Maituolaimysin production were discussed.

Experimental Section Materials. Ammonium bicarbonate, sequence grade modification trypsin, Guanidine, iodacetylamine, urea, Thiourea, and ammonium chloride were from Sigma (St. Louis, MO). HPLC grade formic acid was from Acros (Loughborough, UK). HPLC grade acetonitrile (ACN) was from Merck (Darmstadt, Germany), 1,4-Dithiothreitol (DTT) was from Roche (Mannheim, Germany). 3-[(3-cholamidopropyl)dimethylammonio]-1- propanesulfonate (CHAPS) and PMSF were from Amresco (Solon, OH). Centricon filters (10 kDa) were from Millipore (Bedford, MA). Water was purified using a Milli-Q system (Millipore, Bedford, MA). Bio-basic strong cation exchange HPLC column (0.32 × 100 mm, 5 µm) and the reverse phase HPLC column (0.18 × 100 mm, 5 µm) were from ThermoHypersil (Hemel Hempstead, UK). All other chemicals used were of analytical grade and obtained commercially. Growth of S. luteogriseus. Streptomyces luteogriseus spores were grown in 250 mL conical flasks containing 50 mL seed medium consisting of soluble starch (40.0 g/L), glucose (5.0 g/L), peptone (4.0 g/L), K2HPO4 (0.5 g/L), MgSO4‚7SO4 (0.5 g/L), NaCl (0.5 g/L), and corn steep (2 mL/L) on a rotary shaker at 28 °C, 220 rpm for 48 h. Then, 5 mL seed cultures were moved into 250 mL conical flasks containing 50 mL fermentation medium on a rotary shaker at 28 °C, 220 rpm for 72 h. The fermentation medium is composed of cornstarch (40.0 g/L), glucose (5.0 g/L), soybean meal (16.0 g/L), K2HPO4 (1.0 g/L), MgSO4‚7SO4 (0.5 g/L), NaCl (0.5 g/L), amylase (0.05 g/L) and corn steep (2 mL/L) at pH 6.5. The mycelia were collected by centrifugation at 5000 × g for 10 min at 4 °C. Then the mycelia were washed three times by suspension in previously chilled phosphate buffered saline (PBS, 100 mM, pH 7.4) and collected in the same centrifuga- tion condition mentioned above. Analysis of Maituolaimysin by HPLC. The solution of the culture (10 mL) was extracted three times with 3 × 5 mL ethyl acetate after being adjusted to pH 7.0. The ethyl acetate extracts were collected and evaporated at reduced pressure. The residue was dissolved in 1 mL methanol and loaded onto the C18 column (Kromasil, 250 × 4.6 mm, 5 µm), then eluted with MeOH/H2O/IPA (72:25.5:2.5, v/v/v) at 1 mL/min. The detection wavelength was 330 nm. Determination of Cell Mass. A fifty milliliter portion of sterilized water was added to a 20 mL fermentation culture. The mixture was filtrated, and washed three times with sterilized water. The sediment was placed into dry oven at 105 °C until it weight stabilized. Then it was cooled in the desiccator and weighed. 2000

Journal of Proteome Research • Vol. 4, No. 6, 2005

Wang and Yuan

Extraction of Whole Cell Proteins. The harvested mycelia were placed into a pre-chilled pestle and liquid nitrogen was scooped on top of the cells to flash freeze them. The cells were ground into a powder with a mortar and pestle. Cell powder (0.2 g) was suspended in 0.5 mL lysis buffer consisting of 7 M urea, 2 M thiourea, 4% (m/v) CHAPS and 40 mM Tris, 1 mM PMSF, and 65 mM DTT in a sterilized Eppendorf tube at 4 °C for 1 h. The suspension was centrifuged at 4 °C and 18000 × g for 30 min. The supernatant was precipitated overnight by adding 6× volumes of ice-cold acetone at -20 °C. After being centrifuged and lyophilized to dryness the pellets were stored at -25 °C until use. Typsin Digestion. The pellets were dissolved in 0.5 mL reducing solution (6 M Guanidine hydrochloride, 100 mM ammonium bicarbonate, pH8.5). Protein concentration was determined by the Bradford assay. Two-hundred micrograms of protein was adjusted to 1 µg/µL. The proteins were reduced with 5 mM DTT at 37 °C for 1 h, and carboxyamido-methylated using 10 mM idoacetamide in the dark at room temperature for 30 min. The proteins were exchanged into 50 mM ammonium bicarbonate buffer pH 8.5, and incubated with sequencing grade trypsin (50:1) at 37 °C overnight. Digestion was terminated with formic acid. 2D-LC-MS/MS analysis. The protein digests were analyzed with 2D-LC-ESI-MS/MS (LCQ DecaXP MAX Thermo Finnigan, Palo Alto, CA). The whole process was controlled by the Xcalibur data system (Thermo Finnigan, Palo Alto, CA). The peptide mixture was centrifuged at 6000 × g and 4 °C for 1 min and the supernatant was transferred into 100 µL vial insert. 100 µg of sample was injected into the column followed by a 20 min-washing without salt. After that, the peptides were separated by a 12-step-elution from the strong cation exchanger followed by a gradient elution from the reversed-phase chromatography. The salt steps used were 25, 50, 75, 100, 125, 150, 175, 200, 250, 300, 400, and 600 mM ammonium chloride, respectively. The reversed-phase elution gradient procedure was 1 min 100% buffer A (5% (vol/vol) acetonitrile 0.1% (vol/ vol) formic acid in water); a 70-min gradient to 65%buffer B (0.1% (vol/vol) formic acid in acetonitrile); a 20-min gradient to 80% buffer B; 5 min 80% buffer B; 1 min gradient to 100% buffer A; and 12 min reequilibration at 100% buffer A. Peptides eluted from the capillary column were electrosprayed directly into a mass spectrometer. A continuous scan event consisting of one full MS scan (400-2000 m/z) followed by three data-dependent MS/MS was carried out. The first, second and third most intense ions from the MS scan were selected individually for collision-induced dissociation (CID) at 35% normalized collision energy. The temperature and voltage for the capillary of the ion source were maintained at 150 °C, 3.0 KV, respectively. The zoom scan function was set off and the dynamic exclusion was applied. Protein Identification. Protein identification was performed with Bioworks version 3.1 (Thermo Finnigan) and SEQUEST algorithm.21 The raw MS/MS data were searched against the nonredundant protein database (updated 19 November, 2004) that were download as FASTA formatted sequences from the National Center for Biotechnology Information (http:// www.ncbi.nih.gov). DTA files were generated from the MS/MS threshold of 100 000 peptide mass tolerance 2.5 Da, fragment ion tolerance 0.5 Da, and minimum ion count of 35%. A tryptic enzyme restriction with maximum of two internal missed cleavage sites was used. A molecular mass of 57 Da was added to the static search of all cysteins to account for carboxami-

Proteomic Mapping of S. Luteogriseus Strain 103

research articles

Figure 1. Process curve of Streptomyces luteogriseus. The bacteria were grown in 250 mL flasks at 28 °C, 220 rpm by flaskshaking batch fermentation. The yield of Maituolaimysin the two strains against fermentation time was plotted (right y-axis); the biomass and residual sugar of the two strains against the fermentation time were also plotted (left y-axis). 9 yield of Maituolaimysin; b biomass; 2 residual sugar.

domethylation. All of the peptide matches were filtered based on their false-positive rate (less than 1%).22 The lowest Xcorr values of the peptide were set to 1.9(+1 charge), 2.2(+2 charge), 3.75(+3 charge), and DeltCn must larger than 0.08. The MS/ MS spectrum must be of good quality with fragment ions clearly above the baseline noise and there must be a continuous yand b- ion serious. Bioinformatics Annotation and Protein Classification. The theoretical pI and molecular weight based on the primary amino acid sequence of the identified proteins were calculated using the Expert proteins Analysis System (Expasy) web server (http://us.expasy.org/tools/pi_tool.html).23 The actual molecular weights and pIs may differ slightly from the calculated values because post-translational modifications are not considered in the program. Trans-membrane spanning alpha helices of the membrane proteins were determined through the web-based prediction program TMHMM v2.0 (provided by the Center for Biological Sequence Analysis of the Technical University in Denmark, http://www.cbs.dtu.dk/ services/TMHMM-2.0).24 The protein subcellular location annotation was elucidated using the program PSORT (http:// psort.nibb.ac.jp/form2.html).25,26 The protein function family was categorized according to the protein classification scheme of Streptomyces coelicolor (available at http:// www.sanger.ac.uk/Projects/S_coelicolor/, last modified on Oct 21st, 2003). Proteins that were unclassified according to the above scheme were categorized using the AmiGO tool of Gene Ontology (GO) (http://www.geneontology.org/).27

Results and Discussion Fermentation of the Two Strains and Production of Maituolaimysin. In this section, we focused on biomass, residual sugar, and the yield of Maituolaimysin of both the S. luteogriseus strain 103 and cnn1. These were analyzed during flask-shaking batch fermentation in 250 mL flasks at 28 °C, 220 rpm (shown in Figure 1). The antibiotic production of strain 103 was very low in the prophase of the fermentation, and it

Figure 2. Calculated pI and Mw of the total proteins of the two strains. pI was plotted against Mw on a logarithmic scale. The box indicates the typical 2DE range. (a) strain 103; (b) strain cnn1.

increased sharply after 48 h and reached a maximum yield of 18.6 mg/L. Biomass and residue sugar in the fermentation of the two strains were determined at the same time. During the first 24 h, strains grew slowly and almost no Maituolaimysin was secreted and a small amount of sugar was utilized. Then the biomass increased quickly and reached its peak at about 72 h. During this period, lots of sugar was used and Maituolaimysin began to accumulate. After 72 h, biomass and sugar utilization changes became smooth. However, Maituolaimysin accumulated continually in this late log phase and maximum antibiotic production (18.6 mg/L) was achieved at 96 h. Global Identification of Proteins by MudPIT. The whole cell extracts of the two parallel strains 103 and cnn1 were analyzed using 2D-LC-MS/MS. For each sample, a total of 13 individual chromatograms were obtained. Each chromatogram consisted of a salt elution of a peptide fraction from the SCX dimension followed by an organic gradient to resolve the peptides on the reverse phase C18 column. The peak capacity of the twodimensional on-line system was greatly increased compared with the one-dimensional chromatographic method. The MS/ MS spectra were analyzed with SEQUEST software. To compare the protein identification bias of the method with that of the 2Dgel, we calculated the molecular weights (Mw) and isoelectric points (pI) of the identified proteins based on their primary amino acid sequence. Figure 2a and b showed the calculated Mw’s and pI’s of all identified proteins of the two strains and the rectangle frame included the 2D-gel range. A total of 157 (21.73%) and 196 (24.3%) points fell outside the Journal of Proteome Research • Vol. 4, No. 6, 2005 2001

research articles

Wang and Yuan

Figure 4. Cellular location of the identified proteins of strain 103 and cnn1. Table 1. TMDs of the Identified Membrane Proteins Strain 103

Figure 3. Physicochemical properties of the identified proteins of the two strains. (a) pI distribution of identified proteins; (b) Molecular weight distribution of identified proteins.

frame respectively, which represented the proteins beyond the resolution of 2D-gel electrophoresis. Figure 3, parts a and b, showed the pI and Mw distribution of the identified proteins of the two strains. As is shown in Figure 3a, pI of the most identified proteins ranged from 4.5 to 7.0 and 8.0 to 10.5. Only 40 and 35 proteins with pI between 7.0 and 8.0 were identified, respectively. Molecular mass of the most identified proteins ranged from 10 KDa to 110 Kda (Figure 3b). Proteins with extreme pI and Mw were also detected, which were very difficult to be solved using the 2D gel methods. For strain 103, 25 proteins with pI > 10 and 14 proteins with pI < 4.5 were identified, the highest pI was 11.63 and the lowest pI was 3.97. A total of 22 proteins with Mw < 10 KDa and 101 proteins with Mw > 110 KDa were seen in the results. For strain cnn1, 54 proteins with pI > 10, 13 proteins with pI < 4.5, extremely basic protein (pI ) 11.73), and acid protein (pI ) 3.84) were found. A total of 20 proteins with Mw < 10 KDa and 104 proteins with Mw > 110 KDa were identified. Subcellular Location and Membrane Proteins. In this study, we used the PSORT algorithm to predict the subcellular location of the total identified proteins. Prediction results of strain 103 and cnn1 were shown in Figure 4. Of the total predicted proteins, 77.76% (563) and 72.2% (584) were cytoplasmic proteins, respectively. The percent of membrane proteins of strain 103 (12.56%, 91) was less than that of cnn1 (20.7%, 168) and the percent of its outside proteins (5.21%, 38) was about two times that of cnn1 (2.9%, 23). For the predicted membrane proteins, the TMHMM v2.0 program was used to predict the number of TMDs in order to evaluate their hydropathic character. The TMDs of the membrane proteins of the two strains were listed in Table 1. We can see from the table that the number of the TMDs ranged from 1 to 15. A total of 33 (36.26%) and 92 (54.76%) membrane proteins contained only one TMD, 37 (40.6%) and 45 (26.8%) contained more than five TMDs, respectively. 2002

Journal of Proteome Research • Vol. 4, No. 6, 2005

Strain cnn1

TMDs

no. of proteins

% of proteins

no. of proteins

% of proteins

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 total

33 10 9 2 5 7 7 3 3 5 3 2 1 0 1 91

36.26 10.99 9.89 2.20 5.49 7.69 7.69 3.30 3.30 5.49 3.30 2.20 1.10 0 1.10 100

92 19 6 6 8 10 3 4 5 4 5 4 0 1 1 168

54.76 11.31 3.57 3.57 4.76 5.95 1.78 2.38 2.98 2.38 2.98 2.38 0 0.60 0.60 100

Functional Classification. The identified proteins of the S. Luteogriseus and the mutant were classified according to the classification scheme of the S. coelicolor, which was available at http://www.sanger.ac.uk/Projects/S_ coelicolor. As is shown in Figure 5 parts a and b, a high percent (46% and 42%) of the total identified proteins consisted of unknown function proteins, including entries described as genes of no characterization and hypothetical or unnamed proteins. This is probably because of the lack of the special database. Of the characterized part, proteins involved in cell processes took the largest part of 21% and 19%, followed by proteins associated with the small molecular metabolism of 14% and 13% of the two strains. Proteins participating in regulation occupied only 4% and 3% of the total part due to their low abundance. The small molecular metabolism proteins were further sorted depending on the metabolism passes they were involved in (Figure 5, parts c and d). The unclassified part included proteins that participated in one or more metabolic passes. Of the proteins associated with the small molecular metabolism, 32% and 24% were energy metabolism proteins, which took the largest part of the classified proteins of both strains. The order of the following classes was a little different between strain 103 and cnn1. For strain 103, amino acid biosynthesis proteins took the second largest part (17%), followed by the proteins of degradation of small molecules (11%). The proteins of biosynthesis of cofactors and carriers, central intermediary metabolism, and

Proteomic Mapping of S. Luteogriseus Strain 103

research articles

Figure 6. MS/MS spectrum of double charged peptide NLLFEQPLILEK of polyketide synthase. All major peaks have been labled as ether b- or y-ions.

Figure 5. Functional description of the total identified proteins. (a,b) Classes of the total identified proteins, (a) strain 103, (b) strain cnn1; (c,d) Subclasses of the metabolic proteins, (c) strain 103, (d) strain cnn1.

secondary metabolism occupied the same part: 6%. The number of nucleotide and fatty acid biosynthesis proteins was smallest. For strain cnn1, 13% were involved in biosynthesis of carriers and 12% were secondary metabolism proteins. Amino acid, fatty acid biosynthesis, and degradation of small molecules took the lower part of less than 10%. Nucleotide biosynthesis proteins (3%) were the lowest part in the small molecular metabolism proteins. The proportion of each class was consistent with the expression abundance according to the function classification. Enzymes Involved in Maituolaimycin Synthesis Pathway. To investigate the pathway of the Maituolaimysin biosynthesis we searched the special proteins of the possible pathways in the total proteins of the two strains, respectively. According to the major structure of Maituolaimysin2 and the five reported antibiotic biosynthesis pathways, the peptide synthesizing pattern and the glyco-diriving pathway were impossible and out of consideration. The shikimate, polyketide, and mevalonate pathway were all possible pathways only by the judgment of the antibiotic structure. The key enzymes of the three

pathways were aminobenzoic acid synthase,3 polyketide synthase,4 and phosphomevalonate kinase.7 The three proteins were searched in the total obtained proteins of the both strains, respectively. The result is that only the polyketide synthase existed in both of the strains (Figure 6). The other two proteins were found neither in strain 103 nor in strain cnn1. To further prove the result, we searched all the other enzymes associated with the polyketide biosynthesis pathway according to the protein classification scheme of Streptomyces coelicolor and most of them were found. This confirmed that Maituolaimysin was synthesized through the polyketide pathway. Proteins probably associated with polyketide pathway were listed in Table 2. Type I polyketide synthase, which was reported as the key enzyme of the microlide biosynthesis,28 were found in both strain 103 and cnn1. There are two possible explanations for strain cnn1 having polyketide synthase expression but no antibiotic production. One is that polyketide synthase has been expressed but has not been activated, that is, the expressed polyketide synthase has no catalytic activity. The other is that polyketide synthase has been expressed and has catalytic activity. The microlide has been synthesized but the subsequent reactions are interrupted and the end product is not the expected antibiotic Maituolaimysin. For the proteins only found in strain 103 in Table 2, one probable reason was the lower copy number of those proteins or lost in strain cnn1, the other reason was that the MS analysis did not detect them sufficiently. β-Ketoacyl synthase and 3-oxoacyl-[acyl-carrier protein] reductase were two important proteins to the polyketide pathway.29 It was reported that 3-oxoacyl-[acyl-carrier protein] reductase catalyzes the first reduction step in each cycle of the fatty acid elongation which was often performed in polyketide biosynthesis.30,31 The lower or no expression of them would result in the block of the pathway and no antibiotic produced. This maybe one of the reason that strain cnn1 is the noantibiotic-production strain. Further study should be carried out to prove that. On the whole, the overlap of the proteins identified was only 64% and there were still about 250 proteins were unique to each strain. Some of them were the true changes of the protein expression of the strain. Others were owing to the shortcomings of the MudPIT.32 What is more, the comparison of the total proteins of the two strains could not tell us the true differences Journal of Proteome Research • Vol. 4, No. 6, 2005 2003

research articles

Wang and Yuan

Table 2. Proteins Probably Associated with the Maituolaimysin Biosynthesis Strain 103

Protein description/peptide sequence polyketide synthase R.NLLFEQPLILEK.N Oxidoreductase, aldo/keto reductase family K.DHIDEILGALGWNLSDEDYERISK.I 8-amino-7-oxononanoate synthase K.LIAVDAVYSMDGDQAPLPALLALAER.Y probable oxidoreductase R.LMFAGTPQAIADLR.A putative methyltransferase K.EVYRILKPGGLFAFIDLHK.P putative decarboxylase R.YRLPVTVVILNNGGVYR.G transcriptional regulatory protein K.MSGLRSEITANALDGINPADAHR.L putative hydroxylase K.DAVSSMIFSVTAKKLISYDSLASDGK.M beta-ketoacyl synthase -.MTKYTDHLGRPIVAITGAGVVSSLGQGK.D hydroxylacyl-CoA dehydrogenase R.WVPVHAQVELQARTVVAAAECALTP.3-oxoacyl-[acyl-carrier protein] reductase R.LQGRVALVTGGAGGIGLAVCR.R putative acyl-CoA oxidase K.ALSRAVCIAVRYSAIR.K probable acyltransferase R.IYFANHTSHLDAVVLWSALPHEIR.L

no. of peptides

2 2 1 3 3 2 1 1

Strain cnn1

Protein description/peptide sequence polyketide synthase R.NLLFEQPLILEK.N Oxidoreductase, aldo/keto reductase family R.EFGRTGVKTSLLGFGAMR.L 8-amino-7-oxononanoate synthase R.ARVVVTPHRDVDAVDAALR.S probable oxidoreductase R.NRWKLGGVTQELSDR.M putative methyltransferase K.RVVDLLR.P putative decarboxylase R.VDLVHTDIGEDAHR.V transcriptional regulatory protein K.MLGVNRRSVAAILAR.L putative hydroxylase K.KDPSPTPLLGLPVELEPR.C

no. of peptides

2 3 1 3 3 1 2 2

1 2 1 3 1

in the protein abundance although we could derive some relative quantitative information using the spectral counting method.33 But all of the above-mentioned problems were not really concerned. We cared for only the proteins associated with certain antibiotic biosynthesis pathway. The result told us that the two strains both had the proteins involved in one pathwaypolyketide pathway. To testify the facticity of the result obtained above, the key enzyme inhibitor experiment was carried out. The activity of β-ketoacyl ACP synthase, a key enzyme in the polyketide pathway, were inhibited by adding the special inhibitorss cerulenin or iodoacetamidesinto the solution of the culture of strain 103. Biomass and yield of Maituolaimysin were detected (shown in Figure 7). Under the existence of cerulenin or iodoacetamide, the biomass changed a little (Figure 7a) while the yield of Maituolaimysin was decreased sharply (Figure 7b). Especially in the 1 mmol/L iodoacetamide existed fermentation system, there was almost no Maituolaimysin produced. The decrease of Maituolaimysin production under the existence of the key enzyme inhibitor of polyketide pathway proved that Maituolaimysin was synthesized through polyketide pathway, which was obtained in the proteomic research.

help investigate the quantitative difference of the proteins involved in Maituolaimycin biosynthesis of the two strains. This will help to better understand the antibiotic biosynthesis pathway.

Conclusion The proteome of strain 103 and cnn1 have been globally analyzed using nongel-based 2D-LC-MS/MS technology. As far as we know, it’s the first time that the proteome of Streptomyces has been analyzed with shotgun methods. A total of 726 and 809 proteins were identified, respectively. Proteins associated with the polyketide pathway were found in both strain 103 and cnn1, which suggested that the Maituolaimysin be synthesized through polyketide pathway. The inhibition of β-ketoacyl ACP synthase directly resulted in the decrease of yield of Maituolaimycin, which proved the results of the proteomic research. Unfortunately, quantitative difference could not be obtained in this study. ICAT34 or H218O35 label will 2004

Journal of Proteome Research • Vol. 4, No. 6, 2005

Figure 7. Effects of cerulenin and iodoacetamide on Maituolaimysin production. 9: control; 0: 0.1 mmol/L; b: 0.5 mmol/L Iodoacetamide; O: 1 mmol/L Iodoacetamide.

Proteomic Mapping of S. Luteogriseus Strain 103

It is a far and difficult way to discover an antibiotic biosynthesis pathway. There are many methods and technologies to investigate it, such as radioactive and stable isotope label method,36,37 block mutant method,38 key enzyme inhibitor method,39 DNA Shuffling,40 and so on. With the development of the proteomics, it is possible to use the idea of proteomics to study the biosynthesis pathway. This research provided a feasible proteomic way to investigate the antibiotic biosynthesis pathway and proved the results using the key enzyme inhibitor method. Meanwhile, works on the gene and metabolic level have been carried out in our lab to further understand the antibiotic biosynthesis pathway and to improve the production of the antibiotic. Abbreviations. 2D-LC, two-dimensional liquid chromatography; SCX, strong cation exchange; RP, reversed phase; MS, mass spectrometry; MS/MS, tandem mass spectrometry; ESI: electrospray ionization; CID: collision-induced dissociation; 2DE, two-dimensional electrophoresis

Acknowledgment. We thank Yuheng Ben for technical assistance and Fengming Yu, Liangzhi Li for helpful discussion. We are grateful for the financial support from the National Natural Science Foundation of China (Grant No. 20425620). References (1) Champness, W. C.; Chater, K. F. “Regulation and integration of antibiotic production and morphological differentiation in Streptomyces.” In Regulation of Bacterial Differentiation; Piggot, P. J., Moran, C. P., Youngman, P., Eds; American Society for Microbiology, Washington DC, 1994, 61-93. (2) Yuan, Y. J.; Wang, Z. P.; Wang, L.; Li, X.; Fan, J. Y. A novel macrolide antibiotic production by microbial fermentation. Chinese Patent. 2003, 148863. (3) Chang, Z.; Sun, Y.; He, J.; Vining, L. C. p-Aminobenzoic acid and chloramphenicol biosynthesis in Streptomyces venezuelae: gene sets for a key enzyme, 4-amino-4-deoxy chorismate synthase. Microbiology 2001, 147, 2113-2126. (4) Yu, J. H.; Leonard, T. J. Sterigmatocystin biosynthesis in Aspergillus nidulans requires a novel type I polyketide synthase. J. Bacteriol. 1995, 177, 4792-4800. (5) Stachelhans, T.; Schneider, A.; Marahiel, M. A. A. Engineered biosynthesis of peptide antibiotics. Biochem. Pharm. 1996, 52, 177-186. (6) Gould, S. J. Biosynthesis of the Kanamycin. Chem. Rev. 1997, 97, 2499-2510. (7) Wilding, E. I.; Brown, J. R.; Bryant, A. P.; Chalker, A. F.; Holmes, D. J.; Ingraham, K. A.; Iordanescu, S.; So, C. Y.; Rosenberg, M.; Gwynn, M. N. Identification, Evolution, and Essentiality of the Mevalonate Pathway for Isopentenyl Diphosphate Biosynthesis in Gram-Positive Cocci. J. Bacteriol. 2000, 182, 4319-4327. (8) Washburn, M. P.; Wolters, D.; Yates, J. R., III. Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nat. Biotechnol. 2001, 19, 242-247. (9) Zhu, W. H.; Reich, C. I.; Olsen, G. J.; Giometti, C. S.; Yates, J. R., III. Shotgun Proteomics of Methanococcus jannaschii and Insights into Methanogenesis. J. Proteome Res. 2004, 3, 538-548. (10) Wang, J. Q.; Xue, Y. F.; Feng, X. L.; Li, X. L.; Wang, H.; Li, W.; Zhao, C. F.; Cheng, X. J.; Ma, Y. H.; Zhou, P. J.; Yin, J. N.; Bhatnagar, A.; Wang, R.; Liu, S. Q. An analysis of the proteomic profile for Thermoanaerobacter tengcongensis under optimal culture conditions. Proteomics 2004, 4, 136-150. (11) Orsaria, L.; Paoletti, L.; Gramajo, H. C. Characterization of stationary- phase proteins in Stryptomyces coelicolor A3(2). FEMS Microbiol. Lett. 1998, 162, 275-281. (12) Li, X. M.; Novotna, J.; Weiser, V. J. Major proteins related to chlortetracyline biosynthesis in a Stryptomyce aureofaciens production strain studied by quantitative proteomics. Appl. Microbiol. Biotechnol. 2001, 57, 717-724. (13) Hesketh, A. R.; Chandra, G.; Shaw, A. D.; Rowland, J. J.; Kell, D. B.; Bibb, M. J.; Charter, K. F. Primary and secondary metabolism, and post-translational protein modifications as potrayed by proteomic analysis of Stryptomyces coelicolor. Mol. Microbiol. 2002, 46, 917-932.

research articles (14) Hesketh, A. R.; Chater, K. F. Evidence from proteomics that some of the enzymes of actinorhodin biosynthesis have more than one form and may occupy distinctive cellular locations. J. Ind. Microbiol. Biotechnol. 2003, 30, 523-529. (15) Novotna, J.; Vohradsky, J.; Berndt, P.; Gramajo, H.; Langen, H.; Li, X. M.; Minas, W.; Orsaria, L.; Roeder, D.; Thompson, C. J. Proteomic studies of diauxic lag in the differentiating prokaryote Stryptomyces coelicolor reveal a regulatory network of stressinduced proteins and central metabolic enzymes. Mol. Microbiol. 2003, 48, 1289-1303. (16) Sarka, B. G. Proteome analysis by two-dimensional gel electrophorasis and mass spectrometry: strengths and limitations. Trends Anal. Chem. 2003, 22, 273-281. (17) Wolters, D. A.; Washburn, M. P.; Yates, J. R., III. An automated multidimensional protein identification technology for shotgun proteomics. Anal. Chem. 2001, 73, 5683-5690. (18) Wu, C. C.; MacCoss, M. J.; Howell, K. E.; Yates, J. R., III. A method for the comprehensive proteomic analysis of membrane proteins. Nat. Biotechnol. 2003, 21, 532-538. (19) Jiang, X. S.; Zhou, H.; Zhang, L.; Sheng, Q. H.; Li, S. J.; Li, L.; Hao, P.; Li, Y. X.; Xia, Q. C.; Wu, J. R.; Zeng, R. A High-throughput Approach for Subcellular Proteome. Mol. Cell. Proteomics 2004, 1, 441-445. (20) Blonder, J.; Rodriguez-Galan, M. C.; Chan, K. C.; Lucas, D. A.; Yu, L. R.; Conrads, T. P.; Issaq, H. J.; Young, H. A.; Veenstra, T. D. Analysis of Murine Natural Killer Cell Microsomal Proteins Using Two-Dimensional Liquid Chromatography Coupled to Tandem Electrospray Ionization Mass Spectrometry. J. Proteome Res. 2004, 3, 862-870. (21) Eng, J. K.; McCormack, A. L.; Yates, J. R., III. An approach to correlate tandem mass-spectral data of peptides with amino acid sequences in a protein database. J. Am. Soc. Mass Spectrum. 1994, 5, 976-989. (22) Peng, J.; Elias, J. E.; Thoreen, C. C.; Licklider, L. J.; Gygi, S. P. Evaluation of Multidimensional Chromatography Coupled with Tandem Mass Spectrometry (LC/LC-MS/MS) for Large-Scale Protein Analysis: The Yeast Proteome. J. Proteome Res. 2003, 2, 43-50. (23) Wilkins, M. R.; Gasteiger, E.; Bairoch, A.; Sanchez, J. C.; Williams, K. L.; Appel, R. D.; Hochstrasser, D. F. “Protein Identification and Analysis Tools in the ExPASy Server.” In 2-D Proteome Analysis Protocols; Link, A. J., Ed.; Humana Press: New Jersey, 1998. (24) Krogh, A.; Larsson, B.; Heijne, G. V.; Sonnhammer, E. L. L. Predicting transmembrane protein topology with a hidden Markov model: Application to complete genomes. J. Mol. Biol. 2001, 305, 567-580. (25) Nakai, K.; Horton, P. PSORT: A program for detecting sorting signals in proteins and predicting their subcellular localization. Trends Biochem. Sci. 1999, 24, 34-35. (26) Emanuelsson, O.; Heijne, G. V. Prediction of organellar targeting signals. Biochim. Biophys. Acta 2001, 1541, 114-119. (27) Ashburner, M.; Ball, C. A.; Blake, J. A.; Botstein, D.; Butler, H.; Cherry, J. M.; Davis, A. P.; Dolinski, K.; Dwight, S. S.; Eppig, J. T.; Harris, M. A.; Hill, D. P.; Issel-Tarver, L.; Kasarskis, A.; Lewis, S.; Matese, J. C.; Richardson, J. E.; Ringwald, M.; Rubin, G. M.; Sherlock, G. Gene Ontology: Tool for the unification of biology. The Gene Ontology Consortium. Nat. Genet. 2000, 25, 25-29. (28) Anzai, Y. J.; Saito, N.; Tanaka, M.; Kinoshita, K. J.; Koyama, Y.; Kato, F. Organization of the biosynthetic gene cluster for the polyketide macrolide mycinamicin in Micromonospora griseorubida. FEMS Microbiol. Lett. 2003, 218, 135-141. (29) Lombo, F.; Brana, A. F.; Salas, J. A.; Mendez, C. Genetic organization of the biosynthetic gene cluster for the antitumor angucycline oviedomycin in Streptomyces antibioticus ATCC 11891. Chembiochem. 2004, 5, 1181-1187. (30) LopezLara, I. M.; Geiger, O. The nodulation protein NodG shows the enzymatic activity of an 3-oxoacyl-acyl carrier protein reductase. Mol. Plant Microbe Interact. 2001, 14, 349-357. (31) Fisher, M.; Kroon, J. T.; Martindale, W.; Stuitje, A. R.; Slabas, A. R.; Rafferty, J. B. The X-ray structure of Brassica napus beta-keto acyl carrier protein reductase and its implications for substrate binding and catalysis. Structure Fold Des. 2000, 8, 339-347. (32) Durr, E.; Yu, J. Y.; Krasinska, K. M.; Carver, L. A.; Yates, J. R., III.; Testa, J. E.; Oh, P.; Schnitzer, J. E. Direct proteomic mapping of the lung microvascular endothelial cell surface in vivo and in cell culture. Nat. Biotechnol. 2004, 22, 985-992. (33) Liu, H.; Sadygov, R. G.; Yates, J. R., III. A Model for Random Sampling and Estimation ofRelative Protein Abundance in Shotgun Proteomics. Anal. Chem. 2004, 76, 4193-4201.

Journal of Proteome Research • Vol. 4, No. 6, 2005 2005

research articles (34) Yan, W.; Lee, H.; Deutsch, E. W.; Lazaro, C. A.; Tang, W.; Chen, E.; Fausto, N.; Katze, M. G.; Aebersold, R. A dataset of human liver proteins identified by protein profiling via isotope-coded affinity tag (ICAT) and tandem mass spectrometry. Mol. Cell Proteomics 2004, 3, 1039-1041. (35) Zang, L.; Palmer, T. D.; Hancock, W. S.; Sgroi, D. C.; Karger, B. L. Proteomic analysis of ductal carcinoma of the breast using laser capture microdissection, LC-MS, and 16O/18O isotopic labeling. J. Proteome Res. 2004, 3, 604-612. (36) Steffensky, M.; Li, S. M.; Vogler, B. Novobiocin biosynthesis in Streptomyces spheroids. FEMS Microb Lett. 1998, 161, 69-74. (37) Marfori, E. C.; Bamba, T.; Kajiyama, S. Biosynthetic studies of the tetramicacid antibiotic trichosetin. Tetrahedron 2002, 58, 6655-6658.

2006

Journal of Proteome Research • Vol. 4, No. 6, 2005

Wang and Yuan (38) Gerth, K.; Steinmetz, H.; Hofle, G. Studies on the biosynthesis of epothilones: the PKS and epothilone C/D monooxygenase. J. Antibiotics 2001, 54, 144-148. (39) Nonaka, K.; Kumasaka, C.; Okamoto, Y. Bioconversion of milbemycin-related compounds: biosynthetic pathway of milbemycins. J. Antibiotics 1999, 52, 109-116. (40) Brautaset, T.; Sekurova, O. N.; Sletta, H. Biosynthesis of the polyene antifungal antibiotic nystatin in Streptomyces noursei ATCC 11455: analysis of the gene cluster and deduction of the biosynthetic pathway. Chem. Bio. 2000, 7, 395-403.

PR050169H