Swiss-Czech Proteomics Server

Proteomics Database of Streptomyces coelicolor

Basic information and experimental procedures

Here you will find basic information about cultivation conditions, sample preparation, and data treatment. Information about spot identification and mass spectrometry will be added later.
More comprehensive information can be found in - J. Vohradsky, X-M. Li, C. Thompson, Electrophoresis 1997, 15, 1418-1428.

S. coelicolor J1501 (hisA1 uraA1 strA1 pgl SCP1- SCP2-)

Growth curve
Growth curve

Typical multiphasic growth curve of S. coelicolor cultivated in liquid minimal medium. Each of the points of the growth curve correspond to three samples treeted as described below. Samples were collected in the following time intervals:12, 16, 20, 24, 26, 28, 30, 32, 36, 40, 44, 48, 54, 60, 66 and 72 hours.

Cultivation conditions

S. coelicolor liquid culture was obtained by a two-step protocol, essentially as described by Puglia et al. [1] . Seed cultures were inoculated with ca. 108 spores into a culture tube containing a stainless steel spring (for obtaining dispersed growth) and 10 ml J medium [1]. The culture was grown for 48 h at 30°C at 200 rpm in an orbital shaker. Mycelia were harvested by centrifugation, washed and resuspended in 10 ml MG medium [2] . The suspension was used to inoculate 500 ml of MG medium , supplemented with histidine (50 µg/ml) and uracil (7.5 µg/ml) in a 2 l baffled flask. At various times during growth, one ml samples were labelled for 40 min with 100 µCi of 35S-met/cys Protein Labelling Mix (Dupont). Labelled mycelia were collected, washed and then stored at -80 C.

Sample preparation and 2D gel electrophoresis

The mycelia were resuspended in a sonication buffer (50 mM TES pH 7.4, 1 mM EDTA, 50 mM NaCl, 5% glycerol, 5 mM DTT, Protease Inhibitor Cocktail (Boehringer Mannheim)) and disintegrated by sonication (3 x 0.5 min), using a Model 250 Sonifier® (Branson, USA) with output set at 25 W. After sonication, cell extract was obtained by centrifugation (15,000 g, 30 min, 4 C). Protein content was measured by Bio-Rad Protein Assay. Radioactive samples (ca.106 dpm) were resolved on high resolution 2D gels (Investigator System, Oxford Glycosystems, Oxford, UK). Before loading, the cell extract was mixed with an equal volume of the 2D sample buffer (3 g urea, 0.2 g CHAPS, 80 mg DTT, 2.3 ml H2O, 275 µl Ampholytes 3-10/2D (Oxford GlycoSystems), 25 µl Pharmalyte 8-10.5 (Pharmacia Biotech)). Isoelectric focusing (IEF) was carried out for 16,000 volt-hours. In the second dimension, proteins were separated on vertical 12.5% gels (Duracryl®, 30% acrylamide, 0.8% Bis, Oxford GlycoSystems). In general, triplicate gels were run for each sample. The gels were fixed in 40% methanol/10% acetic acid, then dried without staining. The protein spots were visualized by exposing the gel to X-ray film (Curix, AGFA). Two autoradiograms from each gel were developed after different times of exposure (about 1 week and 1 month).

Electropherogram analyses

Autoradiograms were scanned using a Molecular Dynamics Model 300 Series Computing Densitometer and Image Quant (Molecular Dynamics) image acquisition software with 74 µm/pixel resolution. Gel images were converted to eight bits/pixel and imported to pdQuest (pdi Inc., Huntington Station, NY, U.S.A.) gel analysis software. To minimize errors related to exposure and film sensitivity, two different exposures were merged using the pdQuest software. The resulting composite image was corrected for background and streaking (vertical and horizontal) and then median-smoothed. Spots were detected using "medium" sensitivity. Spot quantification is based on a 2-D Gaussian model: the quantitative measurement of the intensity of each spot is computed as the volume of the 2-D Gaussian form defined by the gel image. The faint spots volumes were corrected using a standard protocol [3] . Integrated densities of spots were computed and relative values of ppm (parts per million) were calculated by dividing integrated spot density by total dpm loaded onto a gel and multiplied by 106.


Individual samples were assigned to "gelsets" 1-16, numbered according to the time of labeling: 12 h (1), 16 h (2), 20 h (3), 24 h (4), 26 h (5), 28 h (6), 30 h (7), 32 h (8), 36 h (9), 40 h (10), 44 h (11), 48 h (12), 54 h (13), 60 h (14), 66 h (15) and 72 h (16). Each gelset contained three member gels from the sample (except gelset 7 (four repeats) and gelset 5 (two repeats). Individual member gels were thus arranged in gelsets 1-16, numbered consecutively 1-48, and assembled into an experimental matchset. A representative gel from gelset 7 (labelled at 30 h) served as the basis for construction of the reference gel. During the matching process, landmark spots were chosen both on the reference and the member gels. The automatic procedure of pdQuest was used to match remaining spots. After that, automatic matches were checked and mismatches were manually corrected. Spots of low reliability were identified by setting two limits: 75% of total ppm in a gel together with a spot quality [4] threshold of 50. Spots lying outside the chosen intervals were temporarily canceled both in reference and member gels. Unmatched spots in member gels were added to the reference gel. This was repeated for all 48 gels of the matchset. When the comprehensive reference gel was created, spots canceled in the member gels were restored and the matching process was repeated. Final matches were carefully checked again. This process ensured that all spots with reasonable quality and density above the threshold were included in a final reference gel containing 1237 spots.

Basic data pre-processing

The experiment was designed so that each spot had three values (gels) for one sample time point (gelsets; 2 gels for point No. 5 and 4 gels for point No. 7), reflecting triplicate gels. During the "clean up" procedure, each set of values for a single spot was inspected for occurrence of only one spot in a set. In this case the value was considered as artifact and the single value was assigned 0. Due to this procedure several (16) spots in the database have all values equal 0. To keep all original data in the database these spots were not excluded, but they were not used in other data analyses as e.g cluster analysis.

Spot quality value

Originaly defined by Garrels [4] is a composite criterion comprising spot image features as goodness of fit of a model function to a spot, and the degree of overlap with other spots. These parameters are combined to form a single quality criterion having a score from 0 to 100. The criterion is unbiased with respect to the protein and its physicochemical and biological functions. It gives rough estimate about the reliability of the data represented by each ppm value or a bar in the graph. The lower is the value usually the higher is the variance of spot density in the repeats.


1. Puglia, A. M., Vohradsky, J., Thompson, C. J., Mol. Microbiol., 1995, 17, 737-746.
2. Doull, J. L., Vining, L. C., FEMS Microbiol. Lett., 1989, 65, 265-268.
3. PDQUEST User's Guide for Version 5.0 , pdi, New York, 1994.
4. Garrels, J., J. Biol. Chem., 1989, 264, 5269-5282.

Other references and pdf reprints can be found on the lab web pages downloads.