Growth and differentiation on glass and zirconia beads
In this work, we aimed to determine whether the hydrophobicity and roughness of a surface on which streptomycetes may attach in their natural environments influences their growth and differentiation. We have chosen S. granaticolor as a model organism because, in addition to the fact that it produces the easily quantified antibiotic granaticin, it is one of the few streptomycetes that grows planktonically without forming clumps. This was important for the analysis of the adhesion process of individual hyphae on the surface of the beads.
The microbeads we used differed in hydrophobicity, which was quite apparent during our manipulation of them. Streptomyces granaticolor mycelia inoculated onto the surface of wetted microbeads formed a continuous lawn of completely differentiated culture on glass, while on zirconia/silica beads distinct but also completely differentiated colonies were formed (Fig. 1). We made the same observation in the cases of S. coelicolor A3(2) and S. ambofaciens (Supporting Information, Fig. S1). The difference in hydrophobicity between the two types of beads also caused different levels of surface hydration. Thus, mycelia on glass, which is more hydrophilic, spread more readily on the surface of the bead. In contrast, on zirconia/silica, where there is less hydration, the mycelia soon cease to grow and start to differentiate, leading to the formation of smaller distinct colonies (Figs 1 and S1).
Figure 1. Cultivation of Streptomyces granaticolor on glass (a) and zirconia/silica beads (b) beads for 96 h at 30 °C. The beads were surrounded by medium 8 supporting the differentiation of streptomycetes. In the inset, the detailed shapes of the colonies are shown.
Download figure to PowerPoint
There is no simple correlation between the hydrophobicity of the support surface to which the bacteria adhere and the strength of adhesion. It was shown previously that microbial adhesion is closely associated with the overall surface hydrophobicity of the microorganisms (Zita & Hermansson, 1997). To describe these interactions, Liu et al. (2004) developed a mathematical model that demonstrates that a high cellular hydrophobicity strongly facilitates microbial adhesion on both hydrophobic and hydrophilic support surfaces.
Surface roughness, which is also one of the differences between the types of beads that we used (Fig. 3), has an influence on biofilm formation and stability but appears to be a minor factor during initial adhesion (Bos et al., 1999). In the case of Streptococcus thermophilus, it was shown previously that biofilm structures do not adhere preferentially to scratches or grooves on their support surface (Van Hoogmoed et al., 1997).
Granaticin production on glass and zirconia/silica beads
In our previous work (Nguyen et al., 2005), we showed that S. coelicolor A3(2) produces a larger amount of the antibiotic actinorhodin on glass beads than in liquid culture (Fig. S1). In this study, we quantified and compared the production of granaticin on glass and zirconia/silica beads to determine whether the quality of the bead surfaces influences the production of this pigmented antibiotic.
We compared biomass and granaticin production after 72 h of stationary cultivation, using the antibiotic production-supporting medium, MG. Figure 2 shows more intense blue pigmentation on the plate containing glass beads (a) than was observed on the plate containing zirconia/silica beads (b). It is important to note that the white colour of the zirconia/silica beads, which are not transparent like the glass beads, contributes to this observed colour difference.
Figure 2. Cultivation of Streptomyces granaticolor in minimal MG medium, which supports antibiotic production on glass (a) and zirconia/silica (b) beads for 72 h at 30 °C. In the inset, the produced pigmented antibiotic granaticin on the backgrounds of the glass and zirconia/silica beads is shown.
Download figure to PowerPoint
However, the measurement of cell biomass, as assessed by protein content, for the whole plate showed that there was more biomass produced on glass beads as well as more granaticin per milligram of protein (Table 1). Comparison with liquid culture showed that there was more biomass and antibiotic production on glass microbeads than in liquid culture (Table 1).
Table 1. Production of biomass and granaticin by Streptomyces granaticolor in minimal MG medium, which supports antibiotic production, on glass and zirconia/silica beads and in a liquid medium incubated for 72 h at 30 °C. Production of granaticin was measured spectrophotometrically and expressed relative to the milligrams of protein in each sample
|Cultivation system||Protein mg mL−1||Granaticin μg mg−1 protein|
Our results show that the qualities of the support surface used during attached growth, along with the cultivation conditions and media composition can influence antibiotic biosynthesis.
Adhesion to glass and zirconia/silica beads
The process of bacterial attachment and the subsequent development of a biofilm can be divided into two stages: the primary or docking stage and the secondary or locking phase. The second stage of adhesion employs molecularly mediated binding between specific adhesins and the surface (Dunne, 2002). In the mature biofilm, cells are encased in an extracellular matrix composed of proteins, exopolysaccharides and extracellular DNA (Flemming & Wingender, 2010; McDougald et al., 2012).
Because we have shown that differences in the physicochemical properties of microbeads can influence both the growth and secondary metabolism of streptomycetes, we focused our attention on protein expression during adhesion and recognition of these surfaces by streptomycetes.
The development of streptomycetes is accompanied by a change in cell surface properties: vegetative hyphae growing in moist substrates have hydrophilic cell surfaces, whereas the aerial hyphae and spores are hydrophobic (Elliot et al., 2003; Del Sol et al., 2007). Del Sol et al. (2007) studied the cell surface of S. coelicolor using atomic force microscopy and showed that young, branching vegetative hyphae have a relatively smooth surface and are attached to an inert silica surface by means of a secreted extracellular matrix.
Our SEM observations showed that young mycelia (6 h) adhere more readily to the surface of glass beads, and we also observed more massive growth on glass than on zirconia/silica (Fig. 3). This is in agreement with our observations using a light stereo microscope (Fig. 1) and our biomass measurements (Table 1).
Figure 3. SEM images of Streptomyces granaticolor growing on glass (a, b) and zirconia/silica (c, d) surfaces for 6 h. Scale bars are indicated in each panel.
Download figure to PowerPoint
Proteome comparison of streptomycetes grown on glass vs. zirconia/silica beads
There are a number of proteomic studies of streptomycetes grown in liquid culture (Hesketh et al., 2002, 2007; Manteca et al., 2011; Zhou et al., 2011); however, there are only a very limited number of studies of streptomycetes proteomes obtained during their sessile growth and differentiation phases (Manteca et al., 2006, 2010).
Because we used the beads to disintegrate the mycelia growing on them, there was a good chance of also recovering the soluble proteins secreted by the cells from the surface of the beads.
Comparison of proteomes from 6 h S. granaticolor mycelia growing on glass and zirconia/silica beads did not reveal many differences (a total of 4 spots with different densities out of 300–400 protein spots detected in each gel). We identified one unique protein present in the glass bead proteome, spot no. 3303, and three proteins differing in expression (different by a factor of 2 and a statistical confidence level of 95%). Spot no. 7506 was up-regulated in the glass bead proteome, and spots no. 3504 and 6604 were up-regulated in the zirconia/silica bead proteome (Fig. 4).
Figure 4. Representative reference 2-D gel of the zirconia/silica proteome sample of Streptomyces granaticolor (300 μg of proteins were loaded). Proteins were stained with fluorescent dye SYPRO® Ruby and scanned at 100-μm resolution using a Molecular Imager FX (Bio-Rad) and 300–400 protein spots were detected in each gel. The numbered spots representing differences between glass and zirconia/silica beads were cut out, and the proteins were identified by MALDI-TOF (the results are shown in Table 2). Spot no. 3303 was unique and spot no. 7506 was up-regulated in the glass bead proteome (solid circles). Spots no. 3504 and 6604 were up-regulated in the zirconia/silica bead proteome (dashed circles). The experimentally obtained MW (kDa) scale is indicated. The inserted graphs show the quantitative differences between the proteins of interest in the glass (solid columns) and zirconia/silica (striped columns) proteomes, as determined using the PDQuest™ software. The expansion shows the detail of the area where unique protein no. 3303 is located.
Download figure to PowerPoint
The identification of these proteins by mass spectrometry revealed that they do not have very interesting functions at first glance (Table 2); however, a search of the literature revealed that all of these enzymes are multifunctional and that orthologs were described to be involved in adhesion or biofilm formation in different microorganisms.
Table 2. MALDI-TOF peptide mass mapping identification of different proteins selected from glass and zirconia/silica beads proteomes as analysed by 2-D gel electrophoresis. The positions of selected spots are marked on the gel in Fig. 4
|Spot no.||Protein||NCBI DTB no.||Description||References|
|3303||Glyceraldehyde-3-phosphate dehydrogenase, type I||302542762||A tetrameric NAD-binding enzyme common to both the glycolytic and gluconeogenic pathways|| http://www.ncbi.nlm.nih.gov/protein/302542762 |
|3504||Cystathionine beta-synthase||12082815||A multifunctional enzyme: catalyses beta-replacement reaction between l-serine, l-cysteine, cysteine thioethers or some other beta- substituted alpha-l-amino acids and a variety of mercaptans|| http://www.ncbi.nlm.nih.gov/protein/12082815 |
|6604||Amidophosphoribosyl-transferase||408679559||An enzyme in the de novo pathway of purine ribonucleotide synthesis and is regulated by feedback inhibition by AMP and GMP|| http://www.ncbi.nlm.nih.gov/protein/408679559 |
|7506||Dihydrolipoamide dehydrogenase||408679449||A component of the multienzyme 2-oxo acid dehydrogenase complexes; catalyses oxidation of its dihydrolipoyl groups|| http://www.ncbi.nlm.nih.gov/protein/408679449 |
The most interesting of the proteins identified is dihydrolipoamide dehydrogenase (DLDH, spot no. 7506), the enzyme classically involved in the reoxidation of dihydrolipoamide during the conversion of 2-oxo acids, such as pyruvate, in several multienzyme complexes in central metabolism (Perham et al., 1987). However, it was shown previously that DLDH is not involved in metabolising 2-oxo acids in the Pneumococcus, indicating that DLDH may serve other functions (Smith et al., 2002), which is supported by evidence that organisms that lack 2-oxo acid dehydrogenases still express a DLDH protein (Danson, 1988).
Siegmann et al. (2009) identified DLDH in Rhodococcus ruber as an exocellular protein involved in binding titanium dioxide particles by nonelectrostatic forces.
Rhodococcus is a genus closely related to Mycobacteria, belonging to the Corynebacteria, which together with Streptomyces belong to Actinomycetales.
These findings and our results support the hypothesis that DLDH serves as an adhesion or binding protein in certain bacteria.
The unique protein we found on glass beads was identified as glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which is known as a classical glycolysis enzyme, but is also known to be a cell surface-exposed protein that can act as an adhesin and is involved in bacterial pathogenicity (Bergmann et al., 2004). Minor density of GAPDH spot which is just next to a major spot of the same molecular weight (see Fig. 4) might indicate posttranslation modification of this essential enzyme. Phosphorylation of GAPDH by protein kinase Cι/λ was found in rabbit muscle (Tisdale, 2002) or in wheat endosperm by SnRK1 kinase (Piattoni et al., 2011), in both cases at serine residues. It was demonstrated previously that GAPDH possesses plasmin- and plasminogen-binding activities in Streptococcus pneumoniae (Bergmann et al., 2004), and in a proteomic study of Streptococcus pyogenes surface-associated proteins, GAPDH was identified among other 15 proteins that were shown to have dual localisation in bacterial cells, both in the cytoplasm and on the bacterial surface (Severin et al., 2007).
Two up-regulated proteins in the proteome from S. granaticolor grown on zirconia/silica beads were amidophosphoribosyltransferase, which is also known as glutamine phosphoribosylpyrophosphate amidotransferase and is involved in purine metabolism, and cystathionine beta-synthase, which is a cytoplasmic enzyme that functions in cysteine biosynthesis.
Amidophosphoribosyltransferase was found in a biofilm formed by Staphylococcus aureus on human keratinocytes. Staphylococcus aureus is prevalent in cutaneous infections, such as chronic wounds, and is an important human pathogen (Secor et al., 2011).
Cystathionine beta-synthase was found to be highly up-regulated in Neisseria gonorrhoeae, the causative agent of gonorrhoea, in the proteome from a biofilm produced in a continuous-flow apparatus in comparison with the proteome of cells grown in planktonic culture (Phillips et al., 2012).