Systems biology provides an integrated view on bacterial adaptation under changing environmental conditions, including its metabolism, its transcriptome, and proteome [1
]. Furthermore, protein complexes have already been the topic of several studies; for instance, in E. coli
(EcoCyc has a useful dataset on protein complexes [2
]), and there are always new examples on protein complexes analyzed in E. coli
] and in other prokaryotes (reviewed in [4
]). However, not much is known about protein complexes and their specific components in Staphylococcus aureus
Omics studies, in particular proteomics, are essential in understanding and revealing the life style of S. aureus
]. S. aureus
is a Gram-positive model organism and a challenging pathogen in clinical infections. It is not easy to establish a general overview on the S. aureus
proteome and protein complexes: identification of conserved and strain-specific proteins requires all-against-all sequence comparisons; structure predictions require detailed calculations even for a single protein complex. Nevertheless, in order to have a good strain overview and look at representative proteins and protein complexes we first performed a refined strain comparison combining two well-established phylogenetic markers, i.e.
, ribosomal RNA and MLST markers (including arc
). For the best phylogenetic resolution we then considered highly-conserved regions shared between S. aureus
strains. Based on this high-resolution analysis and considering the 64 S. aureus
genomes completely known we can show that there are three sub-clades (A-C) encompassing all S. aureus
strains and give a first view on the complete repertoire of proteins and complexes conserved among all these strains. In order to avoid both too complex calculations, and the annotation of all strains individually and completely for each protein, we next compare key representatives of each clade amongst each other: model strains COL, USA300, Newman, and HG001 (clade A), model strain N315 and Mu50 (clade B), and ED133 and MRSA252 (clade C). We establish strain-specific proteins that distinguish the different strains from each other and look at pathogenicity islands with a high number of strain-specific proteins. Next, we analyze important protein repertoires involved in virulence, cell wall component/glycosylation and look at individual strain-specific protein complexes in the three clades. For strain-specific protein complexes we give several detailed structure predictions. Furthermore, the sequence comparisons are complemented by predictions from bioinformatics using three different gene context methods, evidence from databases, co-expression, and text mining. We also indicate which of these interactions are of particular interest for further experimental investigation.
We find that there is surprisingly high diversity, complexity and adaptation potential of proteome and protein complexes amongst S. aureus strains. This highlights the need for detailed systems biological investigations and high-throughput experiments to better understand the suggested interactions and complexes as well as their intricate regulation. Several of these improve S. aureus adaptation and its challenging capacity for infection. As a first overview, our study shows which proteins and complexes are conserved among all three S. aureus clades and models strain-specific proteins and protein complexes from key representatives of each clade.
is an important model organism and pathogen. Its proteome is well studied [27
], however, its dynamics and regulation still present challenges and we often lack information on detailed insight into the protein complexes formed. New advances in bioinformatics and systems biology allow us to investigate proteome changes in different dimensions. Starting from comparatively solid ground, we focus on sequence evidence (genome sequences) and model strains and assemble best predictions, biochemical rules, and experimental evidence to show the conserved and strain-specific protein complexes known for eight S. aureus
strains representing the three clades of S. aureus
strains. We started from three extensive studies in systems biology to delineate a set of conserved core complexes in S. aureus
]. As more data come in from more and more detailed proteomics studies, this list will be both extended and refined.
illustrates several strain-specific proteins but shows that the overall protein functions and protein complexes for these interesting functions are shared between strains, and even between clades. The backbone of conserved functions (Supplementary File S2
) shows that S. aureus
strains share particularly well their central metabolism, a recurrent theme in bacteria [4
]. Enzymes are often well known in their basic structure and, hence, allowed us to model several interesting proteins from virulence-involved protein complexes for S. aureus
COL. The structure annotation approach used [10
] gives the detailed structure as links and pointers to conserved structural domains from which the protein is formed (see materials and methods; detailed results in Supplementary File S3
). We annotated every little segment of known structure in these 13 S. aureus
proteins. However, calculating homology models from such data, the Figure 5
, Figure 6
, Figure 7
, Figure 8
, Figure 9
, Figure 10
and Figure 11
give more detailed protein structure modelling results on strain-specific proteins from all three S. aureus
clades as pictures and views on the three dimensional structure of selected S. aureus
proteins from different protein complexes.
Furthermore, looking at the strictly strain-specific protein lists calculated from extensive sequence comparisons instead (Supplementary Files S5 and S6
) shows that the individual strains have proteins involved in membrane functions, mobile genetic elements, and virulence factors, but also a considerable portion of hypothetical proteins still requiring more experimental investigations to understand their specific function.
Analyzing S. aureus
complexes is challenging as there is also variation in S. aureus
protein complexes over time and there are different modes to regulate this (Table 2
): phosphorylation, glycosylation, and other protein modifications, regulatory interactions include RNA but also accessory proteins, shuttling complexes, metabolites, and the energy state of the cell.
Analysis, calculations of structure, and description of individual complexes are time-consuming and challenging, and it is even more important to complement these observations and predictions with follow-up experiments on the dynamics of protein complexes; a technically challenging undertaking.
Protein complexes change with time and play a crucial role in the adaptation of bacteria which should not be underestimated. Typical situations where this becomes important include adaptation of protein complexes in the diauxic shift [1
], and the use of key protein complexes as potential drug targets (Figure 10
and Figure 11
). Furthermore, several complexes (e.g., antiporters, ABC transporters, Figure 6
and Figure 11
) are heavily involved in adaptation against xenobiotics [13
]. Protein modification triggers assembly and modification of protein complexes, for instance, by protein phosphorylation (Table 2
), by system adaptation (e.g., aerobic, anaerobic), metabolism, or in ribonucleoproteins (Supplementary File S4
), and maybe also by bridging metabolites. Finally, virulence factors are generally expressed condition-specific, for instance enterotoxins (Figure 6
, N315 enterotoxin), cell wall synthesis (Figure 5
and Figure 7
for COL), secretion systems (Figure 8
, SecY of MRSA252), and methicillin resistance (Figure 9
, repressor MecI comparing N315 and Mu50).
One regulatory mechanism involved in the flexibility of protein complexes is post-translational modifications. Modifications, such as protein phosphorylation, glycosylation, and acetylation represent an efficient means to regulate the activity of the individual subunits and, thus, the entire ensemble of proteins as such. As modifying moieties could be rapidly removed or added, protein functionality and/or structure quickly becomes adapted to environmental changes, such as the transition from aerobic to anaerobic conditions. From our data it becomes clear that protein modifications might indeed play a fundamental role in the regulation of protein complexes and their assembly. For the Ser/Thr kinases, for example, we observe that, in contrast to all other strains, the S. aureus
strain COL expresses a shortened version of the kinase, presumably affecting its modifying activity/specificity. In fact, it has been reported by the Ohlsen group that methicillin resistance is affected if pknB
(synonym for stk
, Ser/Thr kinase) is deleted [29
The phylogenetic analysis points out that the three USA300 strains are assigned to the same clade, but are quite distinct from each other, in particular the first characterized USA300 isolate FPR3757 [30
]. The second is TCH1516 [31
], which is closer to a recently reported ISMMS1 [32
]. This supports the importance of strain variation for these dangerous and highly resistant S. aureus
Studying protein complexes in S. aureus
and their changes is a direct route to identify important switches involved in systems biological adaptation. With these data, more detailed investigations of these S. aureus
COL protein complexes and protein structures are possible; for instance, detailed investigations on the whole complex, its assembly and disassembly (currently done by us for pyruvate dehydrogenase complex in S. aureus
COL considering all subunits), or direct targeting of such protein complexes by different drugs, which then requires a systems biological analysis of these drug effects (e.g., [12
Regarding clinical relevant isolates such as the USA300 variants examined, our results point out promising targets for direct pharmacological intervention. For instance, to prevent protein complex formation, there is now a range of novel peptide-based or chemically-improved inhibitors available (e.g., [12
]). This could be used as new agents against MRSA. In summary, these are further arguments why the study of S. aureus
protein complexes is both interesting and challenging, and why a general overview on the protein complex repertoire available for S. aureus
strains is important though it can only focus on selected, but representative examples.