The Promise of Evolutionary Systems Biology: Lessons from Bacterial Chemotaxis Orkun S.

Soyer (29 June 2010) Science Signaling 3 (128), pe23. [DOI: 10.1126/scisignal.3128pe23]

The following resources related to this article are available online at http://stke.sciencemag.org. This information is current as of 7 July 2010.

Article Tools Related Content

Visit the online version of this article to access the personalization and article tools: http://stke.sciencemag.org/cgi/content/full/sigtrans;3/128/pe23 The editors suggest related resources on Science's sites: http://stke.sciencemag.org/cgi/content/abstract/sigtrans;3/128/eg5 http://stke.sciencemag.org/cgi/content/abstract/sigtrans;3/128/ra50 This article cites 29 articles, 10 of which can be accessed for free: http://stke.sciencemag.org/cgi/content/full/sigtrans;3/128/pe23#otherarticles Look up definitions for abbreviations and terms found in this article: http://stke.sciencemag.org/glossary/ Obtain information about reproducing this article: http://www.sciencemag.org/about/permissions.dtl
Downloaded from stke.sciencemag.org on July 7, 2010

References Glossary Permissions

Science Signaling (ISSN 1937-9145) is published weekly, except the last week in December, by the American Association for the Advancement of Science, 1200 New York Avenue, NW, Washington, DC 20005. Copyright 2008 by the American Association for the Advancement of Science; all rights reserved.

PERSPECTIVE
EVOLUTION

The Promise of Evolutionary Systems Biology: Lessons from Bacterial Chemotaxis

Orkun S. Soyer
Published 29 June 2010; Volume 3 Issue 128 pe23

Bacterial chemotaxis and the signaling networks underlying it provide us with a model system for studying the molecular basis of behavior and information processing. Although chemotaxis is well characterized at both the phenotype and genotype levels in the model organism Escherichia coli, it is not yet possible to predict chemotaxis behavior in diverse bacteria on the basis of their environment or genome sequence. Moreover, we still cannot propose a plausible evolutionary trajectory from minimal systems to present-day chemotaxis networks. The analysis of all sequenced bacterial genomes provides a prediction of their chemotaxis networks and reveals substantial structural diversity. Additionally, it uncovers a set of previously unknown proteins that could be the missing link between complex present-day chemotaxis networks and simpler, ancestral systems composed of a few proteins. Further evaluation of these findings with experimental and modeling studies will allow us to distill evolutionary design principles in chemotaxis signaling networks.

Systems biology increasingly delivers a detailed understanding of structure and dynamics in specific biological networks in model organisms. Such accumulation of knowledge raises an important question in the field: Can we employ understanding from specific cases to decipher design principles applicable to all biological systems? Providing an aff irmative answer to this question is one of the key prospects of systems biology, and at the same time one of its biggest challenges. Evolutionary approaches could be crucial in surmounting it. Because biological systems are the product of evolution, their common features will be the result of evolutionary processes. In other words, the principles we seek are not those of an engineer but those of evolution. Deciphering these principles is possible only through an understanding of how evolutionary processes, both adaptive and neutral, can shape biological networks. With this understanding, we should be able to track the origins of complex biological systems and predict their dynamics and structure on the basis of the life-style, environment, or genome sequence of their bearer.
Systems Biology Program, School of Engineering, Computing and Mathematics, University of Exeter, Exeter EX4 4QF, UK. E-mail, o.s.soyer@exeter.ac.uk

Wuichet and Zhulin (1) provide a glimpse of this promise of evolutionary systems biology. By analyzing the genomes of all sequenced bacteria to identify homologs of genes associated with chemotaxis, the authors characterize the chemotaxis signaling networks in these organisms. Although this phylogenetic approach is commonly used in comparative genomics, their study stands out because it focuses on a specific and functionally well-defined system: the chemotaxis network. This network is well characterized in several model organisms including Escherichia coli (2, 3), but we still cannot claim to have achieved a full understanding of bacterial chemotaxis, because response dynamics and network structure show clear differences between these model organisms and others (4, 5). The analysis by Wuichet and Zhulin provides a broad view by predicting the wiring diagram of the chemotaxis networks as found in most present-day bacteria. The results show that although the network structure found in E. coli is evident as a main theme, there are substantial deviations from it in other species. In total, the authors identify 18 different classes of chemotaxis systems, which differ in protein constituents and their interactions and in the presence or absence of multiple pathways controlling motility. The latter finding is in line with another study (6) and highlights the potential role of signal inte-

gration for proper chemotaxis in many motile bacteria. The chemotaxis network as seen in E. coli comprises seven proteins that interact in a specific way and produce dynamical features such as high sensitivity (7, 8) and precise adaptation (9, 10). These features seem essential for proper chemotaxis behavior, making it difficult to imagine how this relatively complex signaling network and its dynamical features could have evolved and through which intermediary steps. In addition, all chemotaxis systems studied to date contain dedicated receptors with methyl-accepting domains and a specific histidine kinase (CheA), which has a domain structure different from that of the class I kinases (11). This suggests that the chemotaxis system has evolved through spontaneous innovation of these proteinsa rather unlikely prospect given our understanding of evolution. In their study, Wuichet and Zhulin shed light on this puzzling picture and the evolutionary origin of chemotaxis systems. By extending their comparative genomic analysis to gene neighborhoods and by using purpose-built databases for detecting homology, they find two sets of previously unknown proteins. One set, termed MAC (methylaccepting coiled-coil) proteins, contains methylation sites as well as domains with methyltransferase and methylesterase functionality. These functions are performed by specif ic proteins, CheR and CheB, in the E. coli network. The second set of previously unknown proteins comprises histidine kinases that combine domain architectures from both class I kinases and CheA. In particular, these kinases have N-terminal sensory domains, as in class I kinases, but their phosphorylation domain is more similar to that found in CheA. Taken together, these discoveries allow us to link seemingly complex networks in present-day bacteria to simple and potentially minimal ancestral systems comprising few signaling proteins. The analysis of Wuichet and Zhulin opens up a new and exciting research avenue. The diverse network structures should be analyzed in terms of response dynamics and chemotaxis behavior, because even small changes in the structure of signaling networks can result in substantial changes in response dynamics (1214), which would then alter cellular behavior. In the context of chemotaxis, we can expect that changes in response dynamics would lead to different chemotaxis strategies, potentially beneficial under a particular envi29 June 2010 Vol 3 Issue 128 pe23 1

Downloaded from stke.sciencemag.org on July 7, 2010

www.SCIENCESIGNALING.org

PERSPECTIVE
ronmental condition or mecha(+) Attractant (-) (+) Attractant (-) nism of motility (Fig. 1). Linking structural changes at the protein and network levels to response dynamics and physiology requires a combined effort of modeling and experimentation. Modeling studies based on the Time Time Fluctuating work of Wuichet and Zhulin attractant will lead to the generation of sources hypotheses regarding response dynamics in chemotaxis and will guide experimental work in species beyond the typical modSingle-flagellated el organisms. The value of such bacteria guidance is clear, considering Stationary that the detailed knowledge of attractant chemotaxis networks in these sources organisms took many decades (+) Attractant (-) to accumulate. To date, research in evoluMotility machinery tionary systems biology employs two main approaches. On Receptor complex the one hand are comparative CheY genomics approaches such as CheA that used by Wuichet and Time CheB Zhulin. These provide insights into differences among organCheR isms at the genetic level, which CheZ we can then try to interpret at the network level as changes in connectivity. When applied in Fig. 1. A cartoon representation of sample chemotaxis networks and their dynamics in bacteria living under different conditions and possessing different motility machinery. All motile bacteria possess a signalthe context of gene regulatory ing network to regulate their motility behavior and to achieve chemotaxis. This network is composed of six networks (15, 16), comparative regulatory proteins in the model organism Escherichia coli : a receptor complex for sensing chemoattracgenomics played an important tants, two proteins (CheR and CheB) that regulate receptor activity, a histidine kinase (CheA), a response role in achieving a global under- regulator (CheY), and its phosphatase (CheZ) that acts on the response regulator. Genomic analyses standing of how the evolution of have revealed that chemotaxis networks show substantial diversity, in terms of both structure and protein these systems links to changes content, across the bacterial kingdom. This diversity is a result of evolution; depending on their life-style, in development (17). On the environment, and motility machinery, bacteria have evolved different signaling networks to achieve reother hand, a more bottom-up sponse dynamics and chemotaxis behavior suitable for those conditions (note that the tree shown is repapproach aims to directly ex- resentative and does not reflect phylogeny of any particular species). The differences in response dynamplore the evolution of biological ics are depicted in the cartoon by the response of the network to addition and subtraction (indicated by systems in silico. By allowing down and up arrows, respectively) of a chemoattractant to the media. models derived from first prinReferences and Notes ciples to evolve in the computer (1824) [for example, using an experimentally veri1. K. Wuichet, I. B. Zhulin, Origins and diversification or by exploring the space of possible mod- fied network as a starting point for an evoof a complex signal transduction system in els (12, 25, 26), these studies allow us to lutionary simulation (27)] will be an excitprokaryotes. Sci. Signal. 3, ra50 (2010). 2. H. C. Berg, D. A. Brown, Chemotaxis in Esidentify evolutionary processes and envi- ing avenue for future research. cherichia coli analyzed by three-dimensional ronmental conditions that can lead to speEvolution drives change in all biologitracking. Antibiot. Chemother. 19, 5578 (1974). cific functional and structural features in cal systems. Those changes that are neutral 3. D. Bray, M. D. Levin, K. Lipkow, The chemotactic behavior of computer-based surrogate bacteria. biological networks. The two approaches or adaptive under the conditions experiCurr. Biol. 17, 1219 (2007). are complementary; the former can help to enced by the organism might persist in the 4. R. Schmitt, Sinorhizobial chemotaxis: A departure identify the common genetic innovations population, leading to diversity (4, 28) and from the enterobacterial paradigm. Microbiology selected for by evolution, whereas the latter complexity (29) at the network level that is 148, 627631 (2002). 5. S. L. Porter, G. H. Wadhams, J. P. Armitage, can help to discern the links between plau- not present in similar human-made sysRhodobacter sphaeroides: Complexity in chemosible selective pressures at the organism tems. If there are any design principles tactic signalling. Trends Microbiol. 16, 251260 level and the resulting effects on response that provide a unifying pattern underneath (2008). 6. R. Hamer, P. Y. Chen, J. P. Armitage, G. Reinert, dynamics and structure at the network lev- such complexity and diversity, deciphering C. M. Deane, Deciphering chemotaxis pathways el. Combining these two approaches so that them requires a firm understanding of the using cross species comparisons. BMC Syst. evolutionary simulations incorporate data effects of evolutionary processes at the Biol. 4, 3 (2010). 7. P. Cluzel, M. Surette, S. Leibler, An ultrasensitive from genomic and experimental studies network level.
Tumbling Rate Tumbling Rate Tumbling Rate

Downloaded from stke.sciencemag.org on July 7, 2010

www.SCIENCESIGNALING.org

29 June 2010

Vol 3 Issue 128 pe23

PERSPECTIVE
bacterial motor revealed by monitoring signaling proteins in single cells. Science 287, 16521655 (2000). V. Sourjik, H. C. Berg, Receptor sensitivity in bacterial chemotaxis. Proc. Natl. Acad. Sci. U.S.A. 99, 123127 (2002). N. Barkai, S. Leibler, Robustness in simple biochemical networks. Nature 387, 913917 (1997). R. G. Endres, N. S. Wingreen, Precise adaptation in bacterial chemotaxis through assistance neighborhoods. Proc. Natl. Acad. Sci. U.S.A. 103, 1304013044 (2006). A. M. Bilwes, L. A. Alex, B. R. Crane, M. I. Simon, Structure of CheA, a signal-transducing histidine kinase. Cell 96, 131141 (1999). O. S. Soyer, M. Salath, S. Bonhoeffer, Signal transduction networks: Topology, response and biochemical processes. J. Theor. Biol. 238, 416425 (2006). S. G. Peisajovich, J. E. Garbarino, P. Wei, W. A. Lim, Rapid diversification of cell signaling phenotypes by modular domain recombination. Science 328, 368372 (2010). C. A. Voigt, D. M. Wolf, A. P. Arkin, The Bacillus subtilis sin operon: An evolvable network motif. Genetics 169, 11871202 (2005). J. C. Perez, E. A. Groisman, Evolution of transcriptional regulatory circuits in bacteria. Cell 138, 233244 (2009). M. Madan Babu, S. A. Teichmann, L. Aravind, Evolutionary dynamics of prokaryotic transcriptional regulatory networks. J. Mol. Biol. 358, 614633 (2006). S. B. Carroll, Evo-devo and an expanding evolutionary synthesis: A genetic theory of morphological evolution. Cell 134, 2536 (2008). P. Franois, V. Hakim, Design of genetic networks with specified functions by evolution in silico. Proc. Natl. Acad. Sci. U.S.A. 101, 580585 (2004). M. J. A. van Hoek, P. Hogeweg, In silico evolved lac operons exhibit bistability for artificial inducers, but not for lactose. Biophys. J. 91, 28332843 (2006). T. Pfeiffer, O. S. Soyer, S. Bonhoeffer, The evolution of connectivity in metabolic networks. PLoS Biol. 3, e228 (2005). R. A. Goldstein, O. S. Soyer, S. Miyano, Evolution of taxis responses in virtual bacteria: Non-adaptive dynamics. PLoS Comput. Biol. 4, e1000084 (2008). M. Salath, O. S. Soyer, Parasites lead to evolution of robustness against gene loss in host signaling networks. Mol. Syst. Biol. 4, 202 (2008). C. Troein, J. C. Locke, M. S. Turner, A. J. Millar, Weather and seasons together demand complex biological clocks. Curr. Biol. 19, 19611964 (2009). I. Salazar-Ciudad, J. Jernvall, A computational model of teeth and the developmental origins of morphological variation. Nature 464, 583586 (2010). M. Kollmann, L. Lvdok, K. Bartholom, J. Timmer, V. Sourjik, Design principles of a bacterial signalling network. Nature 438, 504507 (2005). A. Wagner, Circuit topology and the evolution of robustness in two-gene circadian oscillators. Proc. Natl. Acad. Sci. U.S.A. 102, 1177511780 (2005). J. F. Matias Rodrigues, A. Wagner, N. D. Price, Evolutionary plasticity and innovations in complex metabolic reaction networks. PLoS Comput. Biol. 5, e1000613 (2009). A. Ogawa, R. J. Sommer, Developmental biology. Strategies to get arrested. Science 326, 944945 (2009). O. S. Soyer, S. Bonhoeffer, Evolution of complexity in signaling pathways. Proc. Natl. Acad. Sci. U.S.A. 103, 1633716342 (2006). Acknowledgments: I acknowledge support of the University of Exeter, Science Strategy, and thank R. Goldstein and S. Porter for their useful comments. Competing interests: I declare no competing interests. 10.1126/scisignal/3128pe23

25. 26. 27.

8. 9. 10.

17. 18.

19.

11. 12.

28. 29. 30.

20. 21.

13.

22. 23.

14. 15. 16.

Downloaded from stke.sciencemag.org on July 7, 2010

24.

Citation: O. S. Soyer, The promise of evolutionary systems biology: Lessons from bacterial chemotaxis. Sci. Signal. 3, pe23 (2010).

www.SCIENCESIGNALING.org

29 June 2010

Vol 3 Issue 128 pe23