Coupling gene expression and multicellular morphogenesis during fruiting body formation in Myxococcus xanthus

Transcription

1 Downloaded from orbit.dtu.dk on: Jul 04, 2018 Coupling gene expression and multicellular morphogenesis during fruiting body formation in Myxococcus xanthus Søgaard-Andersen, Lotte; Overgaard, Martin; Lobedanz, Sune; Ellehauge, Eva; Jelsbak, Lars; Rasmussen, Anders Aa. Published in: Molecular Microbiology Publication date: 2003 Document Version Publisher's PDF, also known as Version of record Link back to DTU Orbit Citation (APA): Søgaard-Andersen, L., Overgaard, M., Lobedanz, S., Ellehauge, E., Jelsbak, L., & Rasmussen, A. A. (2003). Coupling gene expression and multicellular morphogenesis during fruiting body formation in Myxococcus xanthus. Molecular Microbiology, 48(1). General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. Users may download and print one copy of any publication from the public portal for the purpose of private study or research. You may not further distribute the material or use it for any profit-making activity or commercial gain You may freely distribute the URL identifying the publication in the public portal If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

2 Blackwell Science, LtdOxford, UKMMIMolecular Microbiology XBlackwell Publishing Ltd, Review ArticleFruiting body morphogenesis in M. xanthusl. Søgaard-Andersen et al. Molecular Microbiology (2003) 48(1), 1 8 MicroReview Coupling gene expression and multicellular morphogenesis during fruiting body formation in Myxococcus xanthus Lotte Søgaard-Andersen,* Martin Overgaard, Sune Lobedanz, Eva Ellehauge, Lars Jelsbak and Anders Aa. Rasmussen Department of Biochemistry and Molecular Biology, University of Southern Denmark, Campusvej 55, 5230 Odense M, Denmark. Summary Accepted 26 November, *For correspondence. sogaard@bmb.sdu.dk; Tel. (+45) ; Fax (+45) A recurring theme in morphogenesis is the coupling of the expression of genes that drive morphogenesis and the morphogenetic process per se. This coupling ensures that gene expression and morphogenesis are carried out in synchrony. Morphogenesis of the spore-filled fruiting bodies in Myxococcus xanthus illustrates this coupling in the construction of a multicellular structure. Fruiting body formation involves two stages: aggregation of cells into mounds and the position-specific sporulation of cells that have accumulated inside mounds. Developmental gene expression propels these two processes. In addition, gene expression in individual cells is adjusted according to their spatial position. Progress in the understanding of the cell surface-associated C-signal is beginning to reveal the framework of an intercellular signalling system that allows the coupling of gene expression and multicellular morphogenesis. Accumulation of the C-signal is tightly regulated and involves transcriptional activation of the csga gene and proteolysis of the full-length CsgA protein to produce the shorter cell surface-associated 17 kda C-signal protein. The C-signal induces aggregation, sporulation and developmental gene expression at specific thresholds. The ordered increase in C-signalling levels, in combination with the specific thresholds, allows the C-signal to induce these three processes in the correct temporal order. The contactdependent C-signal transmission mechanism, in turn, guarantees that C-signalling levels reflect the spatial position of individual cells relative to other cells and, thus, allows the cells to decode their spatial position during morphogenesis. By this mechanism, individual cells can tailor their gene expression profile to one that matches their spatial position. In this scheme, the molecular device that keeps gene expression in individual cells in register with morphogenesis is the C-signalling system, and the morphological structure, which is assessed, is the spatial position of individual cells relative to that of other cells. Introduction Morphogenesis in biological systems involves the conversion of genetic information to shape. This conversion often involves temporally ordered changes in the expression of genes, the products of which direct morphogenesis. The reverse information flow, in which morphology controls gene expression, plays an equally important role in the conversion of genetic information to shape (Rudner and Losick, 2001). According to this scheme, morphogenesis involves a two-way interaction in which gene expression drives morphology, which in turn regulates the expression of new sets of genes. The significance of this two-way interaction is twofold: it keeps genetic programmes and morphogenesis in register. Moreover, it greatly simplifies the logistics of the construction processes by ensuring that the building blocks are synthesized at the right time and at the right location (Shapiro and Losick, 1997). Examination of the temporally ordered expression of genes in the flagellar hierarchy in Salmonella typhimurium and Escherichia coli (Chilcott and Hughes, 2000) and of the temporally ordered and spatially controlled patterns of gene expression in the mother cell and forespore during sporulation in Bacillus subtilis (Rudner and Losick, 2001) have been instrumental in elucidating how the two-way interaction between gene expression and 2003 Blackwell Publishing Ltd

3 2 L. Søgaard-Andersen et al. morphology might operate at the molecular level. In both systems, the completion of specific morphological structures signals the activation of transcriptional regulators, which subsequently induce downstream genes. In the flagellar hierarchy, early gene expression results in the formation of the hook basal body structure, an intermediate structure in the construction of a flagellum. Once completed, this structure signals the activation of late genes in the hierarchy. During sporulation in B. subtilis, early sporulation genes drive the formation of the polar septum, which separates the developing cell into a mother cell and a forespore. Formation of this structure signals the activation of early forespore-specific genes. At a later stage, completion of the engulfment process, which pinches off the forespore within the mother cell, signals the activation of late forespore-specific genes. These two systems illustrate that the morphological checkpoints that govern gene expression might be subcellular in nature, as is the case in the flagellar hierarchy, or they might involve the formation of specialized cell types, as is the case during sporulation in B. subtilis. Morphogenesis of the spore-filled, multicellular fruiting bodies in Myxococcus xanthus involves an additional layer of complexity, as it requires the co-ordinated efforts of thousands of cells. Examination of fruiting body formation suggests that bacteria are capable of decoding multicellular morphological checkpoints and of transforming this information into appropriate changes in gene expression. The goal of this review is to summarize the current view of how the exquisite use of a cell surface-associated morphogen, the C-signal, allows M. xanthus cells to decode multicellular morphological checkpoints and, thereby, synchronize developmental gene expression and morphogenesis. Fruiting body morphogenesis in M. xanthus The life cycle of M. xanthus consists of separate cycles of growth and development (Dworkin, 1996). In the presence of nutrients, the motile, rod-shaped vegetative cells grow and divide. Myxobacteria move by gliding (Spormann, 1999; Wolgemuth et al., 2002) and, if cells are present on a solid surface, they form co-operatively feeding swarms. The developmental cycle is initiated when cells are starved at a high cell density on a solid surface. If these three conditions are met, growth ceases, and the developmental programme that culminates in the morphogenesis of the spore-filled fruiting bodies is initiated. The first morphological manifestations of fruiting body formation are evident after 6 h with changes in cell behaviour (Jelsbak and Søgaard-Andersen, 2002) as the cells begin to aggregate and form small aggregation centres (Dworkin, 1996) (Fig. 1). As more cells enter the aggregation centres, they increase in size and eventually become symmetric mounds. By 24 h, the aggregation process is complete, and the mounds hold 10 5 densely packed cells. Inside the mounds, the rod-shaped, motile cells differentiate to non-motile spores resulting in mature fruiting bodies. Although all cells have the capacity to become a spore (Dworkin and Gibson, 1964), only cells that have accumulated inside the mounds sporulate. Cells left outside the fruiting bodies, called peripheral rods, remain rod shaped (O Connor and Zusman, 1991) (Fig. 1). Two lines of evidence suggest that the temporal and spatial co-ordination of aggregation and sporulation reflects regulatory interactions. First, even though most developmental mutants display aggregation as well as Fig. 1. Morphogenesis of multicellular fruiting bodies in M. xanthus. On the developmental timeline, the black triangles indicate the expression time of developmental genes expressed in all cells before the initiation of C-signalling; the hatched triangles indicate the expression time of C- signal-dependent genes, all of which are only expressed in sporulating cells. The time of action of the A to E signals is indicated by the arrows below the timeline (Shimkets, 1999). Cellular arrangements during the different stages of fruiting body formation are shown above the timeline. The greyscale colouring of cells according to the intensity profile shown on the right indicates the level of C-signalling in individual cells during the different morphological stages. At 72 h, cells inside the fruiting bodies have matured to spores, and the cells remaining outside the fruiting bodies are the peripheral rods.

4 Fruiting body morphogenesis in M. xanthus 3 sporulation defects, several mutants have been isolated in which sporulation occurs outside mounds (Morrison and Zusman, 1979; Cho and Zusman, 1999; Kruse et al., 2001). Secondly, in non-motile mutants, which are unable to aggregate and sporulate (Kroos et al., 1988), the addition of exogenous C-signal (see below) restores sporulation without restoring aggregation (Kim and Kaiser, 1990a). Therefore, during starvation, wild-type cells are capable of sporulating outside mounds; however, they are prevented from doing so by regulatory mechanisms. As mound formation is a relatively slow process, M. xanthus cells are faced with the problem of postponing sporulation until aggregation has resulted in mound formation. Fruiting body morphogenesis involves temporally coordinated changes in gene expression, in which genes are turned on at specific time points during development (Dworkin, 1996) (Fig. 1). Moreover, developmental gene expression is spatially controlled (Fig. 1). Genes turned on after 6 h are preferentially expressed in sporulating cells, whereas genes activated before 6 h are expressed in all cells including peripheral rods (Julien et al., 2000). Interestingly, the genes that are specifically expressed in sporulating cells depend on the C-signal (see below), whereas genes expressed in all cells are expressed independently of the C-signal (Julien et al., 2000). To co-ordinate the efforts of thousands of cells during fruiting body formation, cells communicate with each other using at least five intercellular signals (A to E) (Shimkets, 1999). Analyses of developmental gene expression in signalling mutants suggest that these signalling systems are arranged in a time-based hierarchy and that they lie on the same developmental pathway. The B-, A-, D- and E-signals are essential for progression through the first 5 h of development. The functions and biochemical nature of the B-, D- and E-signals remain to be clarified. The A-signal consists of a subset of amino acids and is part of a cell density sensing system that measures the density of starving cells. Development proceeds once a threshold concentration of A-signal is reached (Plamann and Kaplan, 1999). This mechanism guarantees that cells only progress through early development if they are starved at a sufficiently high density. The C-signal, the last of the known signals to act, becomes important for development after 6 h, coinciding with the first overt signs of multicellular morphogenesis (Kroos and Kaiser, 1987; Jelsbak and Søgaard-Andersen, 2002). The C-signal is used repeatedly during development, first to induce aggregation and later sporulation. Tied to the induction of these processes, the C-signal induces the expression of most genes turned on after 6 h (Kroos and Kaiser, 1987; Kim and Kaiser, 1991; Li et al., 1992; Kruse et al., 2001). The C-signal is exposed on the cell surface (Kim and Kaiser, 1990b; Shimkets and Rafiee, 1990), and C-signal transmission occurs by a contact-dependent mechanism that involves end-to-end contacts between cells (Kim and Kaiser, 1990b,c; Sager and Kaiser, 1994). Thus, these five signalling systems may be viewed as having distinct functions. The early signals that act before the initiation of morphogenesis appear to evaluate cell density and possibly also starvation conditions, e.g. severity and duration, thus ensuring that fruiting body formation is only initiated if a sufficient number of cells are starved and while there is still sufficient capacity for protein synthesis. In contrast, C-signal is the morphogen that induces and co-ordinates aggregation, sporulation and gene expression after 6 h. Morphogenesis and developmental gene expression are coupled by regulatory mechanisms Fruiting body morphogenesis and developmental gene expression could be directed by dedicated, parallel pathways, each following their own agenda and timing mechanism, or by linked regulatory pathways. Several lines of evidence support the latter model. First, most developmental mutants display defects in morphogenesis as well as in developmental gene expression (Dworkin, 1996). Secondly, in several cases, developmentally regulated genes have been shown not only to direct the expression of downstream developmental genes but also to be important for morphogenesis (e.g. Kroos et al., 1990). Morphogenesis also directs developmental gene expression. Consistent with the C-signal-dependent initiation of morphogenesis, this coupling is only evident in the expression of C-signal-dependent genes, whereas genes expressed independently of the C-signal are also expressed independently of morphogenesis. One line of evidence for morphogenesis-dependent developmental gene expression comes from the analyses of non-motile mutants. These mutants neither aggregate nor sporulate (Kroos et al., 1988). Moreover, in these mutants, expression of C-signal-dependent genes is reduced or abolished, whereas expression of C-signal-independent genes is unaffected (Kroos et al., 1988). Interestingly, the addition of exogenous C-signal restores developmental gene expression and sporulation in these mutants without restoring the aggregation defect (Kim and Kaiser, 1990a). In these mutants, gene expression and sporulation are also restored by artificially aligning the cells in a pattern that increases the frequency of end-to-end contacts (Kim and Kaiser, 1990c). These observations indicate the existence of one or more morphological checkpoints, in this case involving a special spatial arrangement of cells, important for the expression of C-signal-dependent genes. A second line of evidence for the morphogenesisdependent expression of C-signal-dependent genes comes from the observation that peripheral rods and sporulating cells display different gene expression profiles

5 4 L. Søgaard-Andersen et al. in which C-signal-dependent genes are only expressed in sporulating cells (Julien et al., 2000). How then does the C-signal co-ordinate morphogenesis and gene expression? The C-signal The molecular nature of the C-signal has been controversial. Mutants unable to synthesize C-signal carry mutations in the csga gene (Shimkets et al., 1983). csga encodes two proteins, one with a size of 25 kda corresponding to the full-length CsgA protein (referred to as p25), and one with a size of 17 kda (referred to as p17) (Kruse et al., 2001). Recent evidence suggests that p17 is the C-signal and is formed by developmentally regulated proteolysis of p25 (S. Lobedanz and L. Søgaard-Andersen, unpublished observations). Originally, the C-signal was purified as a csga-encoded, 17 kda protein from starving cells on the basis of its ability to rescue development of csga cells (Kim and Kaiser, 1990b). The identity of the C-signal was questioned by several observations, which suggested that p25 acts as an enzyme to produce the C-signal. First, exogenous p25 rescues the development of csga cells (Lee et al., 1995). p25 shares homology with NAD(P)- dependent short-chain alcohol dehydrogenases (SCAD) and binds NAD + in vitro (Lee et al., 1995). Secondly, p25 derivatives carrying substitutions in either the cofactor binding pocket or the catalytic site are unable to rescue the development of csga cells (Lee et al., 1995). Finally, overproduction of the SocA protein, which also shares homology with SCAD, in csga mutants restores development in these cells (Lee and Shimkets, 1994). Recent results have shown that the p17 corresponds to the C-terminal 17 kda of p25 and lacks the cofactor binding pocket (S. Lobedanz and L. Søgaard-Andersen, unpublished observations). Consistently, p17 is unable to bind NAD + in vitro (S. Lobedanz and L. Søgaard- Andersen, unpublished observations). Nevertheless, exogenous p17 restores the development of csga cells (S. Lobedanz and L. Søgaard-Andersen, unpublished observations). These findings suggest that p17 is the C- signal rather than acting as an enzyme to produce the C-signal. Moreover, in vitro analyses suggest that p17 is produced by proteolysis of p25 and that the activity of the protease is developmentally regulated (S. Lobedanz and L. Søgaard-Andersen, unpublished observations). These findings raise questions about the function of p25. It remains a possibility that p25 possesses SCAD activity. The observation that exogenous p25 restores the development of csga cells could be explained in either of two ways: p25 could be processed to p17, or p25 may imitate p17. By the same token, the ability of SocA to substitute for the C-signal could be explained by the ability of SocA to imitate p17. Likewise, in the case of the non-functional p25 proteins that carry substitutions in either the cofactor binding site or the catalytic Fig. 2. Model of the C-signal transduction pathway. The schematic illustrates the C-signal transmission between two neighbouring cells. For simplicity, the components in the pathway are only shown in the cell on the right. These components are also present in the cell on the left. In this cell, only csga is shown to illustrate the signal amplification loop labelled 2. The second signal amplification loop is labelled 1. HPK indicates the hypothetical FruA histidine protein kinase. W7536 symbolizes the sporulation gene monitored by the Tn5lacW7536 insertion. See the text for a detailed explanation of the pathway.

6 Fruiting body morphogenesis in M. xanthus 5 site, the substitutions might interfere with p25 processing or with the ability of p25 to imitate p17. The C-signal transduction pathway The identification of components in the C-signal transduction pathway has provided a framework for understanding how the reception of the C-signal is transformed into different responses (Fig. 2). The receptor of the C-signal has yet to be identified. The DNA-binding response regulator protein FruA is the first recognized component in the pathway (Ellehauge et al., 1998). FruA mutants are unable to aggregate and sporulate and are deficient in C- signal-dependent gene expression (Ogawa et al., 1996; Ellehauge et al., 1998; Horiuchi et al., 2002). FruA is subject to two regulatory mechanisms. In one mechanism, the early acting A- and E-signals induce transcription of frua (Ogawa et al., 1996; Ellehauge et al., 1998). In the second mechanism, C-signal activates the FruA protein, conceivably by phosphorylation, so that it may interact with downstream targets (Ellehauge et al., 1998). The pathway branches downstream from FruA. One branch leads to aggregation. In this branch, C-signal acts as a FruA-dependent input signal to the cytoplasmic Frz signal transduction system and induces methylation of the FrzCD protein (Søgaard-Andersen and Kaiser, 1996), a methyl-accepting chemotaxis protein (McBride et al., 1989). The second branch leads to C-signal-dependent gene expression and sporulation (Søgaard-Andersen et al., 1996). C-signal and FruA jointly regulate the expression of at least 53 genes during development (Horiuchi et al., 2002). Two of these genes constitute the devrs operon (Ellehauge et al., 1998), which encodes the DevRS proteins and is essential for sporulation (Kroos et al., 1990). In turn, DevRS are required for expression of the sporulation gene monitored by the Tn5lacW7536 insertion (Licking et al., 2000). A third branch in the pathway is located upstream from FruA and leads to increased transcription of csga during development (Kim and Kaiser, 1991; Crawford and Shimkets, 2000; Gronewold and Kaiser, 2001; Kruse et al., 2001). The four proteins encoded by the act operon have important functions in csga expression. ActA and ActB, which is a DNA-binding response regulator, are required for full transcription of csga during development, whereas ActC and ActD are important for the correct timing of csga transcription (Gronewold and Kaiser, 2001). It has been suggested that the Act proteins are specifically involved in the C-signal-dependent increase in csga transcription (Gronewold and Kaiser, 2001). After csga transcription, p25 accumulates, and then p25 is processed to produce the C-signal, p17. p17 is exposed on the cell surface and may interact with a C-signal receptor on a neighbouring cell. Temporal coupling of morphogenesis and gene expression How, then, does the C-signal induce three qualitatively different responses that are separated in time and space, and how is the C-signal transduction pathway configured to ensure the correct temporal order of aggregation, sporulation and gene expression? Overexpression of the csga gene early during development results in premature aggregation, premature sporulation and premature C- signal-dependent gene expression (Kruse et al., 2001), suggesting that an ordered increase in C-signalling levels during development holds the key to these questions. The C-signal transduction pathway contains two signal amplification loops. During the aggregation process, cells are organized in streams, in which cells are arranged end-to-end and with their long axes roughly parallel to each other (O Connor and Zusman, 1989). Inside the mounds, cells move in circular tracks also making extensive end-to-end contacts (Sager and Kaiser, 1993). As the aggregation process results in an ordered increase in cell density until cells have accumulated at the highest cell density inside the mounds, the conditions for C- signal transmission are gradually optimized during the aggregation process. Therefore, the prediction is that, during the aggregation process, the level of C-signalling that a cell is exposed to increases (loop 1 in Figs 2 and 3). In the second amplification loop (loop 2 in Figs 2 and 3), C-signalling results in increased csga transcription. This, in turn, results in p25 accumulation, which is subsequently processed to p17 and may then engage in C- signal transmission. The outcome of these two signal amplification loops is that cells engaged in C-signalling are exposed to an ordered increase in the level of C- signalling during development. Different thresholds of C-signal are required for the induction of the different C-signal-dependent responses. Intermediate levels of C-signalling are required for early C-signal-dependent gene expression and aggregation; and high levels are required for late C-signal-dependent gene expression and sporulation (Kim and Kaiser, 1991; Li et al., 1992; Kruse et al., 2001). The C-signal thresholds, in combination with the ordered increase in the level of C-signalling, allow the C-signal to act as a timer that induces first aggregation and early C-signal-dependent gene expression and then sporulation and late C- signal-dependent gene expression (Fig. 3). The timing mechanism would work as follows. After 6 h, FruA and the Act proteins have accumulated, and p17 is beginning to accumulate. This results in the initiation of C-signalling. Once the threshold level for aggregation and early C- signal-dependent gene expression is reached, these two processes are turned on. As cells aggregate, they are exposed to an ordered increase in C-signalling levels, and

7 6 L. Søgaard-Andersen et al. Fig. 3. Quantitative model of the C-signal transduction pathway. 1 and 2 indicate the two signal amplification loops that govern the increase in C-signalling levels. The amplification loops are set in motion by the starvation-induced expression of csga (Crawford and Shimkets, 2000). The outputs from the C-signal transduction pathway are shown in the lower part of the figure and depend on signalling levels. See the text for a detailed explanation of the pathway. C-signal-dependent genes are progressively induced. Once cells are densely packed inside the mounds, they are exposed to the highest level of C-signalling, and the threshold level that results in the induction of late gene expression is exceeded. This results in devrs expression and, subsequently, expression of the gene containing the Tn5lacW7536 insertion, and sporulation ensues. Thus, the ability of the C-signalling system to induce different responses can be explained by different quantitative requirements of the different responses in combination with the ordered increase in C-signalling levels during development. At the molecular level, it is tempting to speculate that the ordered increase in C-signalling levels is transformed to an ordered increase in the activated forms of the two transcription factors FruA and ActB. Activated ActB would stimulate csga expression and, ultimately, p17 accumulation. The function of activated FruA is twofold: it interacts directly or indirectly with the Frz system to modulate the activity of this signal transduction system in such a way that aggregation is induced (for a detailed description of how the C-signal modulates cell behaviour and induces aggregation, see Jelsbak and Søgaard-Andersen, 2002). In addition, the ordered increase in activated FruA would result in the temporally ordered induction of C- signal-dependent genes. Spatial coupling of morphogenesis and gene expression A hallmark in fruiting body formation is the position-specific sporulation of cells that have accumulated inside the fruiting bodies. How might the C-signal ensure the spatial co-ordination of aggregation, sporulation and gene expression, and how is the expression of C-signaldependent genes coupled to morphogenesis? The spatial co-ordination of aggregation, sporulation and gene expression seems to be a direct consequence of the contact-dependent C-signal transmission mechanism involving specific end-to-end contacts between cells. This transmission mechanism ensures that the level of C- signalling to which a cell is exposed reflects the spatial position of that cell relative to other cells. Therefore, the gene expression profile of individual cells closely matches the position of individual cells during morphogenesis. Only cells that are closely packed inside the mounds reach the high level of C-signalling that induces late gene expression and sporulation. Consequently, only cells that have accumulated inside the mounds express late genes and undergo sporulation. This model is in agreement with that proposed by Kaiser and coworkers to explain the different gene expression profiles in sporulating cells and peripheral rods (Julien et al., 2000). These authors observed that the cellular organization of peripheral rods favours side-to-side contacts as opposed to end-to-end contacts (Fig. 1). Therefore, peripheral rods experience a lower frequency of end-to-end contacts compared with the frequent end-toend contacts in aggregating cells and in cells inside mounds. Consequently, the peripheral rods experience a level of C-signalling that allows neither C-signal-dependent gene expression nor aggregation and sporulation. Morphological checkpoints in fruiting body morphogenesis The C-signal has dual functions in fruiting body formation:

8 Fruiting body morphogenesis in M. xanthus 7 it induces aggregation, sporulation and gene expression in the correct temporal order. Moreover, the mechanism of the C-signal allows cells to assess and verify their spatial position relative to other cells and, subsequently, to tailor their gene expression profile accordingly. Therefore, in terms of morphological checkpoints that couple gene expression to morphogenesis during fruiting body formation, the structure that is evaluated is the position of individual cells in a field of other cells, and the molecular device that computes the position of individual cells is the C-signalling system. Conclusions Bacteria have the ability to decipher morphological checkpoints of different complexities and to transform this information to changes in gene expression. In the case of the subcellular morphological checkpoint in the flagellar hierarchy, completion of this structure results directly in the activation of s 28 by allowing the export of the FlgM antisigma factor. In the case of the morphological checkpoints in B. subtilis sporulation, formation of the specialized cell types signals the activation of intracellular signal transduction pathways, which activate downstream transcription factors. In the case of the multicellular morphological checkpoint in M. xanthus, cell position is deciphered by means of an intercellular signalling system that activates an intracellular signal transduction pathway which, in turn, activates downstream transcription factors. Thus, the characteristics of the morphological checkpoints are tailored to the characteristics of each individual system. In M. xanthus, the identification of components important for development has so far depended on the painstaking identification of relevant genes and proteins on a step-by-step basis. The recent advent of the M. xanthus genome sequence (B. Goldman, personal communication), proteome analysis (Horiuchi et al., 2002) and plans for constructing an M. xanthus DNA microarray (H. Kaplan, personal communication) promise to pave the way for a large-scale, systematic identification of additional important players in the regulatory pathways that drive and co-ordinate the temporally and spatially regulated changes in gene expression, which ultimately result in fruiting body morphogenesis. Acknowledgements We thank Stephen Douthwaite and Birgitte Kallipolitis for valuable criticism of the manuscript. Work in the authors laboratory was supported by the Female Researchers in Joint Action Research Program, by the Danish Natural Science Research Council and by the Carlsberg Foundation. References Chilcott, G.S., and Hughes, K.T. (2000) Coupling of flagellar gene expression to flagellar assembly in Salmonella enterica serovar Typhimurium and Escherichia coli. Microbiol Mol Biol Rev 64: Cho, K., and Zusman, D.R. (1999) Sporulation timing in Myxococcus xanthus is controlled by the espab locus. Mol Microbiol 34: Crawford, E.W., and Shimkets, L.J. (2000) The Myxococcus xanthus soce and csga genes are regulated by the stringent response. Mol Microbiol 37: Dworkin, M. (1996) Recent advances in the social and developmental biology of the Myxobacteria. Microbiol Rev 60: Dworkin, M., and Gibson, S.M. (1964) A system for studying microbial morphogenesis: rapid formation of microcysts in Myxococcus xanthus. Science 146: Ellehauge, E., Nørregaard-Madsen, M., and Søgaard-Andersen, M. (1998) The FruA signal transduction protein provides a checkpoint for the temporal co-ordination of intercellular signals in Myxococcus xanthus development. Mol Microbiol 30: Gronewold, T.M.A., and Kaiser, D. (2001) The act operon controls the level and time of C-signal production for Myxococcus xanthus development. Mol Microbiol 40: Horiuchi, T., Taoka, M., Isobe, T., Komano, T., and Inouye, S. (2002) Role of frua and csga genes in gene expression during development of Myxococcus xanthus. Analysis by two-dimensional gel electrophoresis. J Biol Chem 277: Jelsbak, L., and Søgaard-Andersen, L. (2002) Pattern formation by a cell surface-associated morphogen in Myxococcus xanthus. Proc Natl Acad Sci USA 99: Julien, B., Kaiser, A.D., and Garza, A. (2000) Spatial control of cell differentiation in Myxococcus xanthus. Proc Natl Acad Sci USA 97: Kim, S.K., and Kaiser, D. (1990a) Cell motility is required for the transmission of C-factor, an intercellular signal that coordinates fruiting body morphogenesis of Myxococcus xanthus. Genes Dev 4: Kim, S.K., and Kaiser, D. (1990b) C-factor: a cell-cell signaling protein required for fruiting body morphogenesis of M. xanthus. Cell 61: Kim, S.K., and Kaiser, D. (1990c) Cell alignment required in differentiation of Myxococcus xanthus. Science 249: Kim, S.K., and Kaiser, D. (1991) C-factor has distinct aggregation and sporulation thresholds during Myxococcus development. J Bacteriol 173: Kroos, L., and Kaiser, D. (1987) Expression of many developmentally regulated genes in Myxococcus depends on a sequence of cell interactions. Genes Dev 1: Kroos, L., Hartzell, P., Stephens, K., and Kaiser, D. (1988) A link between cell movement and gene expression argues that motility is required for cell cell signaling during fruiting body development. Genes Dev 2: Kroos, L., Kuspa, A., and Kaiser, D. (1990) Defects in fruiting body development caused by Tn5 lac insertions in Myxococcus xanthus. J Bacteriol 172:

9 8 L. Søgaard-Andersen et al. Kruse, T., Lobedanz, L., Berthelsen, N.M.S., and Søgaard- Andersen, L. (2001) C-signal: a cell surface-associated morphogen that induces and coordinates multicellular fruiting body morphogenesis and sporulation in M. xanthus. Mol Microbiol 40: Lee, B.-U., Lee, K., Mendez, J., and Shimkets, L.J. (1995) A tactile sensory system of Myxococcus xanthus involves an extracellular NAD(P)+-containing protein. Genes Dev 9: Lee, K., and Shimkets, L.J. (1994) Cloning and characterization of the soca locus which restores development to Myxococcus xanthus C-signaling mutants. J Bacteriol 176: Li, S., Lee, B.-U., and Shimkets, L.J. (1992) csga expression entrains Myxococcus xanthus development. Genes Dev 6: Licking, E., Gorski, L., and Kaiser, D. (2000) A common step for changing cell shape in fruiting body and starvationindependent sporulation in Myxococcus xanthus. J Bacteriol 182: McBride, M.J., Weinberg, R.A., and Zusman, D.R. (1989) Frizzy aggregation genes of the gliding bacterium Myxococcus xanthus show sequence similarities to the chemotaxis genes of enteric bacteria. Proc Natl Acad Sci USA 86: Morrison, C.E., and Zusman, D.R. (1979) Myxococcus xanthus mutants with temperature-sensitive, stage-specific defects: evidence for independent pathways in development. J Bacteriol 140: O Connor, K.A., and Zusman, D.R. (1989) Patterns of cellular interactions during fruiting-body formation in Myxococcus xanthus. J Bacteriol 171: O Connor, K.A., and Zusman, D.R. (1991) Development in Myxococcus xanthus involves differentiation into two cell types, peripheral rods and spores. J Bacteriol 173: Ogawa, M., Fujitani, S., Mao, X., Inouye, S., and Komano, T. (1996) FruA, a putative transcription factor essential for the development of Myxococcus xanthus. Mol Microbiol 22: Plamann, L., and Kaplan, H.B. (1999) Cell-density sensing during early development in Myxococcus xanthus. In Cell Cell Signaling in Bacteria. Dunny, G.M., and Winans, S.C. (eds). Washington, DC: American Society for Microbiology Press, pp Rudner, D.R., and Losick, R. (2001) Morphological coupling in development: lessons from prokaryotes. Dev Cell 1: Sager, B., and Kaiser, D. (1993) Two cell-density domains within the Myxococcus xanthus fruiting body. Proc Natl Acad Sci USA 90: Sager, B., and Kaiser, D. (1994) Intercellular C-signaling and the traveling waves of Myxococcus. Genes Dev 8: Shapiro, L., and Losick, R. (1997) Protein localization and cell fate in bacteria. Science 276: Shimkets, L.J. (1999) Intercellular signaling during fruitingbody development of Myxococcus xanthus. Annu Rev Microbiol 53: Shimkets, L.J., and Rafiee, H. (1990) CsgA, an extracellular protein essential for Myxococcus xanthus development. J Bacteriol 172: Shimkets, L.J., Gill, R.E., and Kaiser, D. (1983) Developmental cell interactions in Myxococcus xanthus and the spoc locus. Proc Natl Acad Sci USA 80: Søgaard-Andersen, L., and Kaiser, D. (1996) C factor, a cell-surface-associated intercellular signaling protein, stimulates the Frz signal transduction system in Myxococcus xanthus. Proc Natl Acad Sci USA 93: Søgaard-Andersen, L., Slack, F., Kimsey, H., and Kaiser, D. (1996) Intercellular C-signaling in Myxococcus xanthus involves a branched signal transduction pathway. Genes Dev 10: Spormann, A.M. (1999) Gliding motility in bacteria: insights from studies of Myxococcus xanthus. Microbial Mol Biol Rev 63: Wolgemuth, C., Hoiczyk, E., Kaiser, D., and Oster, G. (2002) How myxobacteria glide. Curr Biol 12: