Response dynamics of phosphorelays suggest their potential utility in cell signalling

EvgS/EvgA, the unorthodox two-component system regulating bacterial multiple resistance

IMPORTANCE

EvgS/EvgA, one of the five unorthodox two-component systems in Escherichia coli, plays an essential role in adjusting bacterial behaviors to adapt to the changing environment. Multiple resistance regulated by EvgS/EvgA endows bacteria to survive in adverse conditions such as acidic pH, multidrug, and heat. In this minireview, we summarize the specific structures and regulation mechanisms of EvgS/EvgA and its multiple resistance. By discussing several unresolved issues and proposing our speculations, this review will be helpful and enlightening for future directions about EvgS/EvgA.

A theoretical framework to analyse the flow of particles in a dynamical system with stochastic transition rates and site capacities

We study the stochasticity in a dynamical model: ribosome flow model with different site sizes that models the unidirectional movement of particles controlled by transition rates along a lattice having different site sizes. Our work models the parameters as random variables with known distributions and investigates the steady-state flow rate under this notion by using tools from the random matrix theory. Some closed-form theoretical results are derived for the steady-state flow rate under some restrictive assumptions such as random variables being independent and identically distributed. Furthermore, for arbitrary but bounded stochastic transition rates, stochastic site capacities, or both, we establish bounds for the steady-state flow rate. Our analysis can be generalized and applied to study the flow of particles in numerous transport systems in the stochastic environment.

Systematic analysis of noise reduction properties of coupled and isolated feed-forward loops

Cells can maintain their homeostasis in a noisy environment since their signaling pathways can filter out noise somehow. Several network motifs have been proposed for biological noise filtering and, among these, feed-forward loops have received special attention. Specific feed-forward loops show noise reducing capabilities, but we notice that this feature comes together with a reduced signal transducing performance. In posttranslational signaling pathways feed-forward loops do not function in isolation, rather they are coupled with other motifs to serve a more complex function. Feed-forward loops are often coupled to other feed-forward loops, which could affect their noise-reducing capabilities. Here we systematically study all feed-forward loop motifs and all their pairwise coupled systems with activation-inactivation kinetics to identify which networks are capable of good noise reduction, while keeping their signal transducing performance. Our analysis shows that coupled feed-forward loops can provide better noise reduction and, at the same time, can increase the signal transduction of the system. The coupling of two coherent 1 or one coherent 1 and one incoherent 4 feed-forward loops can give the best performance in both of these measures.

Cellular behavior can be affected by noise in molecular interactions. Signaling pathways should process noisy input signals and support cellular decision making by properly transducing the signals, while removing noise from them. Three component networks of feed-forward loops (FFLs) have been proposed to serve as ideal noise reducers, while linear pathways were shown to be good signal transducers. These signaling units do not work in isolation, so there is a possibility that a combination of various feed-forward loops can provide good noise reduction, while maintaining good signal transduction. To test this hypothesis, we have systematically tested the noise reducing and signal transducing capabilities of all possible combinations of feed-forward loops and compared them with the performance of individual FFLs. We built mathematical models of all these systems and compared their capabilities at reducing noise in the input signal while maintaining responses to meaningful changes in the incoming signal. We found that a combination of two copies of a special type of fully positive signaling FFLs is the best noise reducer, while a combination of two incoherent (one positive, one negative signal) FFLs can provide the best signal transduction. The combination of these two FFLs could provide good signal processing where both noise reduction and signal transduction are achieved.

Maximization of information transmission influences selection of native phosphorelay architectures

Phosphorelays are signal transduction circuits that sense environmental changes and adjust cellular metabolism. Five different circuit architectures account for 99% of all phosphorelay operons annotated in over 9,000 fully sequenced genomes. Here we asked what biological design principles, if any, could explain selection among those architectures in nature. We began by studying kinetically well characterized phosphorelays (Spo0 of Bacillus subtilis and Sln1 of Saccharomyces cerevisiae). We find that natural circuit architecture maximizes information transmission in both cases. We use mathematical models to compare information transmission among the architectures for a realistic range of concentration and parameter values. Mapping experimentally determined phosphorelay protein concentrations onto that range reveals that the native architecture maximizes information transmission in sixteen out of seventeen analyzed phosphorelays. These results suggest that maximization of information transmission is important in the selection of native phosphorelay architectures, parameter values and protein concentrations.

A Radical Reimagining of Fungal Two-Component Regulatory Systems

Bacterial two-component regulatory systems (TCSs) mediate signal transduction by transferring phosphoryl groups between sensor kinase and response regulator proteins, sometimes using intermediary histidine-phosphotransferase (Hpt) domains to form multistep phosphorelays. Because (i) almost all known fungal sensor kinases exhibit a domain architecture characteristic of bacterial TCS phosphorelays, (ii) all known fungal Hpts are stand-alone proteins suited to shuttle between cytoplasm and nucleus, and (iii) the best-characterized fungal TCS is a canonical phosphorelay, it is widely assumed that most or all fungal TCSs function via phosphorelays. However, fungi generally encode more sensor kinases than Hpts or response regulators, leading to a disparity between putative phosphorelay inputs and outputs. The simplest resolution of this paradox is to hypothesize that most fungal sensor kinases do not participate in phosphorelays. Reimagining how fungal TCSs might function leads to multiple testable predictions.

The Signaling Molecule Indole Inhibits Induction of the AR2 Acid Resistance System in Escherichia coli

Induction of the AR2 acid response system of Escherichia coli occurs at a moderately low pH (pH 5.5) and leads to high levels of resistance to pH levels below 2.5 in the presence of glutamate. Induction is mediated in part by the EvgAS two component system. Here, we show that the bacterial signaling molecule indole inhibits the induction of key promoters in the AR2 system and blocks the development of glutamate-dependent acid resistance. The addition of tryptophan, the precursor for indole biosynthesis, had the same effects, and this block was relieved in a tnaA mutant, which is unable to synthesize indole. Expression of a constitutively active EvgS protein was able to relieve the inhibition caused by indole, consistent with EvgS being inhibited directly or indirectly by indole. Indole had no effect on autophosphorylation of the isolated cytoplasmic domain of EvgS. This is consistent with a model where indole directly or indirectly affects the ability of EvgS to detect its inducing signal or to transduce this information across the cytoplasmic membrane. The inhibitory activity of indole on the AR2 system is not related to its ability to act as an ionophore, and, conversely, the ionophore CCCP had no effect on acid-induced AR2 promoter activity, showing that the proton motive force is unlikely to be a signal for induction of the AR2 system.

A deterministic mathematical model for bidirectional excluded flow with Langmuir kinetics

In many important cellular processes, including mRNA translation, gene transcription, phosphotransfer, and intracellular transport, biological "particles" move along some kind of "tracks". The motion of these particles can be modeled as a one-dimensional movement along an ordered sequence of sites. The biological particles (e.g., ribosomes or RNAPs) have volume and cannot surpass one another. In some cases, there is a preferred direction of movement along the track, but in general the movement may be bidirectional, and furthermore the particles may attach or detach from various regions along the tracks. We derive a new deterministic mathematical model for such transport phenomena that may be interpreted as a dynamic mean-field approximation of an important model from mechanical statistics called the asymmetric simple exclusion process (ASEP) with Langmuir kinetics. Using tools from the theory of monotone dynamical systems and contraction theory we show that the model admits a unique steady-state, and that every solution converges to this steady-state. Furthermore, we show that the model entrains (or phase locks) to periodic excitations in any of its forward, backward, attachment, or detachment rates. We demonstrate an application of this phenomenological transport model for analyzing ribosome drop off in mRNA translation.

Structural and Functional Analysis of the Escherichia coli Acid-Sensing Histidine Kinase EvgS

The EvgS/EvgA two-component system of Escherichia coli is activated in response to low pH and alkali metals and regulates many genes, including those for the glutamate-dependent acid resistance system and a number of efflux pumps. EvgS, the sensor kinase, is one of five unconventional histidine kinases (HKs) in E. coli and has a large periplasmic domain and a cytoplasmic PAS domain in addition to phospho-acceptor, HK and dimerization, internal receiver, and phosphotransfer domains. Mutations that constitutively activate the protein at pH 7 map to the PAS domain. Here, we built a homology model of the periplasmic region of EvgS, based on the structure of the equivalent region of the BvgS homologue, to guide mutagenesis of potential key residues in this region. We show that histidine 226 is required for induction and that it is structurally colocated with a proline residue (P522) at the top of the predicted transmembrane helix that is expected to play a key role in passing information to the cytoplasmic domains. We also show that the constitutive mutations in the PAS domain can be further activated by low external pH. Expression of the cytoplasmic part of the protein alone also gives constitutive activation, which is lost if the constitutive PAS mutations are present. These findings are consistent with a model in which EvgS senses both external and internal pH and is activated by a shift from a tight inactive to a weak active dimer, and we present an analysis of the purified cytoplasmic portion of EvgS that supports this.

IMPORTANCE One of the ways bacteria sense their environment is through two-component systems, which have one membrane-bound protein to do the sensing and another inside the cell to turn genes on or off in response to what the membrane-bound protein has detected. The membrane-bound protein must thus be able to detect the stress and signal this detection event to the protein inside the cell. To understand this process, we studied a protein that helps E. coli to survive exposure to low pH, which it must do before taking up residence in the gastrointestinal tract. We describe a predicted structure for the main sensing part of the protein and identify some key residues within it that are involved in the sensing and signaling processes. We propose a mechanism for how the protein may become activated and present some evidence to support our proposal.

Optimal Down Regulation of mRNA Translation

Down regulation of mRNA translation is an important problem in various bio-medical domains ranging from developing effective medicines for tumors and for viral diseases to developing attenuated virus strains that can be used for vaccination. Here, we study the problem of down regulation of mRNA translation using a mathematical model called the ribosome flow model (RFM). In the RFM, the mRNA molecule is modeled as a chain of n sites. The flow of ribosomes between consecutive sites is regulated by n + 1 transition rates. Given a set of feasible transition rates, that models the outcome of all possible mutations, we consider the problem of maximally down regulating protein production by altering the rates within this set of feasible rates. Under certain conditions on the feasible set, we show that an optimal solution can be determined efficiently. We also rigorously analyze two special cases of the down regulation optimization problem. Our results suggest that one must focus on the position along the mRNA molecule where the transition rate has the strongest effect on the protein production rate. However, this rate is not necessarily the slowest transition rate along the mRNA molecule. We discuss some of the biological implications of these results.

Arabidopsis response regulator 22 inhibits cytokinin-regulated gene transcription in vivo

Cytokinin signaling in Arabidopsis is carried out by a two-component system (TCS) multi-step phosphorelay mechanism that involves three different protein families: histidine kinases (AHKs), phosphotransfer proteins (AHPs), and response regulators (ARRs) that are in turn, subdivided into A-, B- and C-type ARRs depending on their function and structure. Upon cytokinin perception, AHK proteins autophosphorylate; this phosphate is then transferred from the AHKs to the AHPs to finally reach the ARRs. When B-type ARRs are activated by phosphorylation, they function as transcription factors that regulate the expression of cytokinin-dependent genes such as the A-type ARRs, among many others. In cytokinin signaling, while A- and B-type ARR function is well understood, it is still unclear if C-type ARRs (ARR22 and ARR24) play a role in this mechanism. Here, we describe a novel method suitable to study TCS activity natively as an in vivo system. We also show that ARR22 inhibits gene transcription of an A-type ARR upon cytokinin treatment in vivo. Consequently, we propose that ARR22, by acting as a phosphatase on specific AHPs, disrupts the TCS phosphorelay and prevents B-type ARR phosphorylation, and thus their activation as transcription factors, explaining the observed deactivation of cytokinin-responsive genes.

Stochastic analysis of Chemical Reaction Networks using Linear Noise Approximation

Stochastic evolution of Chemical Reactions Networks (CRNs) over time is usually analyzed through solving the Chemical Master Equation (CME) or performing extensive simulations. Analysing stochasticity is often needed, particularly when some molecules occur in low numbers. Unfortunately, both approaches become infeasible if the system is complex and/or it cannot be ensured that initial populations are small. We develop a probabilistic logic for CRNs that enables stochastic analysis of the evolution of populations of molecular species. We present an approximate model checking algorithm based on the Linear Noise Approximation (LNA) of the CME, whose computational complexity is independent of the population size of each species and polynomial in the number of different species. The algorithm requires the solution of first order polynomial differential equations. We prove that our approach is valid for any CRN close enough to the thermodynamical limit. However, we show on four case studies that it can still provide good approximation even for low molecule counts. Our approach enables rigorous analysis of CRNs that are not analyzable by solving the CME, but are far from the deterministic limit. Moreover, it can be used for a fast approximate stochastic characterization of a CRN.

Phosphoryl Group Flow within the Pseudomonas aeruginosa Pil-Chp Chemosensory System: DIFFERENTIAL FUNCTION OF THE EIGHT PHOSPHOTRANSFERASE AND THREE RECEIVER DOMAINS

Bacterial chemosensory signal transduction systems that regulate motility by type IV pili (T4P) can be markedly more complex than related flagellum-based chemotaxis systems. In T4P-based systems, the CheA kinase often contains numerous potential sites of phosphorylation, but the signaling mechanisms of these systems are unknown. In Pseudomonas aeruginosa, the Pil-Chp system regulates T4P-mediated twitching motility and cAMP levels, both of which play roles in pathogenesis. The Pil-Chp histidine kinase (ChpA) has eight "Xpt" domains; six are canonical histidine-containing phosphotransfer (Hpt) domains and two have a threonine (Tpt) or serine (Spt) in place of the histidine. Additionally, there are two stand-alone receiver domains (PilG and PilH) and a ChpA C-terminal receiver domain (ChpArec). Here, we demonstrate that the ChpA Xpts are functionally divided into three categories as follows: (i) those phosphorylated with ATP (Hpt4-6); (ii) those reversibly phosphorylated by ChpArec (Hpt2-6), and (iii) those with no detectable phosphorylation (Hpt1, Spt, and Tpt). There was rapid phosphotransfer from Hpt2-6 to ChpArec and from Hpt3 to PilH, whereas transfer to PilG was slower. ChpArec also had a rapid rate of autodephosphorylation. The biochemical results together with in vivo cAMP and twitching phenotypes of key ChpA phosphorylation site point mutants supported a scheme whereby ChpArec functions both as a phosphate sink and a phosphotransfer element linking Hpt4-6 to Hpt2-3. Hpt2 and Hpt3 are likely the dominant sources of phosphoryl groups for PilG and PilH, respectively. The data are synthesized in a signaling circuit that contains fundamental features of two-component phosphorelays.

Deciphering Parameter Sensitivity in the BvgAS Signal Transduction

To understand the switching of different phenotypic phases of Bordetella pertussis, we propose an optimized mathematical framework for signal transduction through BvgAS two-component system. The response of the network output to the sensory input has been demonstrated in steady state. An analysis in terms of local sensitivity amplification characterizes the nature of the molecular switch. The sensitivity analysis of the model parameters within the framework of various correlation coefficients helps to decipher the contribution of the modular structure in signal propagation. Once classified, the model parameters are tuned to generate the behavior of some novel strains using simulated annealing, a stochastic optimization technique.

Phosphorelay of non-orthodox two component systems functions through a bi-molecular mechanism in vivo: the case of ArcB

Two-component systems play a central part in bacterial signal transduction. Phosphorelay mechanisms have been linked to more robust and ultra-sensitive signalling dynamics. The molecular machinery that facilitates such a signalling is, however, only understood in outline. In particular the functional relevance of the dimerization of a non-orthodox or hybrid histidine kinase along which the phosphorelay takes place has been a subject of debate. We use a combination of molecular and genetic approaches, coupled to mathematical and statistical modelling, to demonstrate that the different possible intra- and inter-molecular mechanisms of phosphotransfer are formally non-identifiable in Escherichia coli expressing the ArcB non-orthodox histidine kinase used in anoxic redox control. In order to resolve this issue we further analyse the mathematical model in order to identify discriminatory experiments, which are then performed to address cis- and trans-phosphorelay mechanisms. The results suggest that exclusive cis- and trans-mechanisms will not be operating, instead the functional phosphorelay is likely to build around a sequence of allosteric interactions among the domain pairs in the histidine kinase. This is the first detailed mechanistic analysis of the molecular processes involved in non-orthodox two-component signalling and our results suggest strongly that dimerization facilitates more discriminatory proof-reading of external signals, via these allosteric reactions, prior to them being further processed.

Characterization of mutations in the PAS domain of the EvgS sensor kinase selected by laboratory evolution for acid resistance in Escherichia coli

Laboratory-based evolution and whole-genome sequencing can link genotype and phenotype. We used evolution of acid resistance in exponential phase Escherichia coli to study resistance to a lethal stress. Iterative selection at pH 2.5 generated five populations that were resistant to low pH in early exponential phase. Genome sequencing revealed multiple mutations, but the only gene mutated in all strains was evgS, part of a two-component system that has already been implicated in acid resistance. All these mutations were in the cytoplasmic PAS domain of EvgS, and were shown to be solely responsible for the resistant phenotype, causing strong upregulation at neutral pH of genes normally induced by low pH. Resistance to pH 2.5 in these strains did not require the transporter GadC, or the sigma factor RpoS. We found that EvgS-dependent constitutive acid resistance to pH 2.5 was retained in the absence of the regulators GadE or YdeO, but was lost if the oxidoreductase YdeP was also absent. A deletion in the periplasmic domain of EvgS abolished the response to low pH, but not the activity of the constitutive mutants. On the basis of these results we propose a model for how EvgS may become activated by low pH.

Stochastic flux analysis of chemical reaction networks

Background

Chemical reaction networks provide an abstraction scheme for a broad range of models in biology and ecology. The two common means for simulating these networks are the deterministic and the stochastic approaches. The traditional deterministic approach, based on differential equations, enjoys a rich set of analysis techniques, including a treatment of reaction fluxes. However, the discrete stochastic simulations, which provide advantages in some cases, lack a quantitative treatment of network fluxes.

Results

We describe a method for flux analysis of chemical reaction networks, where flux is given by the flow of species between reactions in stochastic simulations of the network. Extending discrete event simulation algorithms, our method constructs several data structures, and thereby reveals a variety of statistics about resource creation and consumption during the simulation. We use these structures to quantify the causal interdependence and relative importance of the reactions at arbitrary time intervals with respect to the network fluxes. This allows us to construct reduced networks that have the same flux-behavior, and compare these networks, also with respect to their time series. We demonstrate our approach on an extended example based on a published ODE model of the same network, that is, Rho GTP-binding proteins, and on other models from biology and ecology.

Conclusions

We provide a fully stochastic treatment of flux analysis. As in deterministic analysis, our method delivers the network behavior in terms of species transformations. Moreover, our stochastic analysis can be applied, not only at steady state, but at arbitrary time intervals, and used to identify the flow of specific species between specific reactions. Our cases study of Rho GTP-binding proteins reveals the role played by the cyclic reverse fluxes in tuning the behavior of this network.

Phosphorelays provide tunable signal processing capabilities for the cell

Achieving a complete understanding of cellular signal transduction requires deciphering the relation between structural and biochemical features of a signaling system and the shape of the signal-response relationship it embeds. Using explicit analytical expressions and numerical simulations, we present here this relation for four-layered phosphorelays, which are signaling systems that are ubiquitous in prokaryotes and also found in lower eukaryotes and plants. We derive an analytical expression that relates the shape of the signal-response relationship in a relay to the kinetic rates of forward, reverse phosphorylation and hydrolysis reactions. This reveals a set of mathematical conditions which, when satisfied, dictate the shape of the signal-response relationship. We find that a specific topology also observed in nature can satisfy these conditions in such a way to allow plasticity among hyperbolic and sigmoidal signal-response relationships. Particularly, the shape of the signal-response relationship of this relay topology can be tuned by altering kinetic rates and total protein levels at different parts of the relay. These findings provide an important step towards predicting response dynamics of phosphorelays, and the nature of subsequent physiological responses that they mediate, solely from topological features and few composite measurements; measuring the ratio of reverse and forward phosphorylation rate constants could be sufficient to determine the shape of the signal-response relationship the relay exhibits. Furthermore, they highlight the potential ways in which selective pressures on signal processing could have played a role in the evolution of the observed structural and biochemical characteristic in phosphorelays.

Two-component phosphorelays constitute the key signaling pathways in all prokaryotes, lower eukaryotes, and plants, where they underline diverse physiological responses such as virulence, cell-cycle progression and sporulation. Despite such prevalence, our understanding of the dynamics and function of these systems remains incomplete. In particular, it is not clear why all phosphorelays studied to date embed a four-layer architecture and how their dynamics could relate to phenotypic variability in the resulting responses. Here, we use analytical approaches and numerical simulations to analyze all possible phosphorelay topologies of length four and embedding reverse phosphorylation. We find that only two topologies can embed both hyperbolic and sigmoidal signal-response relationships, and that one of these can underlie high noise (i.e. phenotypic variability) in population responses. All of the remaining topologies are either non-functional or can embed only a hyperbolic signal-response relationship. Using analytical solutions of relay dynamics, we find that reverse phosphorylation from the third layer, a topological featured commonly observed in nature, is a necessary condition for sigmoidal signal-response relationship.

Simplifying biochemical models with intermediate species

Mathematical models are increasingly being used to understand complex biochemical systems, to analyse experimental data and make predictions about unobserved quantities. However, we rarely know how robust our conclusions are with respect to the choice and uncertainties of the model. Using algebraic techniques, we study systematically the effects of intermediate, or transient, species in biochemical systems and provide a simple, yet rigorous mathematical classification of all models obtained from a core model by including intermediates. Main examples include enzymatic and post-translational modification systems, where intermediates often are considered insignificant and neglected in a model, or they are not included because we are unaware of their existence. All possible models obtained from the core model are classified into a finite number of classes. Each class is defined by a mathematically simple canonical model that characterizes crucial dynamical properties, such as mono- and multistationarity and stability of steady states, of all models in the class. We show that if the core model does not have conservation laws, then the introduction of intermediates does not change the steady-state concentrations of the species in the core model, after suitable matching of parameters. Importantly, our results provide guidelines to the modeller in choosing between models and in distinguishing their properties. Further, our work provides a formal way of comparing models that share a common skeleton.

Split histidine kinases enable ultrasensitivity and bistability in two-component signaling networks

Bacteria sense and respond to their environment through signaling cascades generally referred to as two-component signaling networks. These networks comprise histidine kinases and their cognate response regulators. Histidine kinases have a number of biochemical activities: ATP binding, autophosphorylation, the ability to act as a phosphodonor for their response regulators, and in many cases the ability to catalyze the hydrolytic dephosphorylation of their response regulator. Here, we explore the functional role of "split kinases" where the ATP binding and phosphotransfer activities of a conventional histidine kinase are split onto two distinct proteins that form a complex. We find that this unusual configuration can enable ultrasensitivity and bistability in the signal-response relationship of the resulting system. These dynamics are displayed under a wide parameter range but only when specific biochemical requirements are met. We experimentally show that one of these requirements, namely segregation of the phosphatase activity predominantly onto the free form of one of the proteins making up the split kinase, is met in Rhodobacter sphaeroides. These findings indicate split kinases as a bacterial alternative for enabling ultrasensitivity and bistability in signaling networks. Genomic analyses reveal that up 1.7% of all identified histidine kinases have the potential to be split and bifunctional.

Transcriptional regulation is a major controller of cell cycle transition dynamics

DNA replication, mitosis and mitotic exit are critical transitions of the cell cycle which normally occur only once per cycle. A universal control mechanism was proposed for the regulation of mitotic entry in which Cdk helps its own activation through two positive feedback loops. Recent discoveries in various organisms showed the importance of positive feedbacks in other transitions as well. Here we investigate if a universal control system with transcriptional regulation(s) and post-translational positive feedback(s) can be proposed for the regulation of all cell cycle transitions. Through computational modeling, we analyze the transition dynamics in all possible combinations of transcriptional and post-translational regulations. We find that some combinations lead to ‘sloppy’ transitions, while others give very precise control. The periodic transcriptional regulation through the activator or the inhibitor leads to radically different dynamics. Experimental evidence shows that in cell cycle transitions of organisms investigated for cell cycle dependent periodic transcription, only the inhibitor OR the activator is under cyclic control and never both of them. Based on these observations, we propose two transcriptional control modes of cell cycle regulation that either STOP or let the cycle GO in case of a transcriptional failure. We discuss the biological relevance of such differences.

Bayesian design of synthetic biological systems

Here we introduce a new design framework for synthetic biology that exploits the advantages of Bayesian model selection. We will argue that the difference between inference and design is that in the former we try to reconstruct the system that has given rise to the data that we observe, whereas in the latter, we seek to construct the system that produces the data that we would like to observe, i.e., the desired behavior. Our approach allows us to exploit methods from Bayesian statistics, including efficient exploration of models spaces and high-dimensional parameter spaces, and the ability to rank models with respect to their ability to generate certain types of data. Bayesian model selection furthermore automatically strikes a balance between complexity and (predictive or explanatory) performance of mathematical models. To deal with the complexities of molecular systems we employ an approximate Bayesian computation scheme which only requires us to simulate from different competing models to arrive at rational criteria for choosing between them. We illustrate the advantages resulting from combining the design and modeling (or in silico prototyping) stages currently seen as separate in synthetic biology by reference to deterministic and stochastic model systems exhibiting adaptive and switch-like behavior, as well as bacterial two-component signaling systems.