- Email: [email protected]
The Secret Life of Smoothened
Previews 823
The Secret Life of Smoothened
The membrane protein Smoothened is the signal-generating component of the Hedgehog pathway. How Smoothened transmits the signal to downstream effectors in the cytoplasm has been a long-standing mystery. Now, with recent reports demonstrating a direct interaction between Smoothened and Costal-2, we appear to be significantly closer to solving this problem. During the development of multicellular organisms, individual cells adopt specific fates that are appropriate to their spatial and temporal position. This, naturally, requires that cells have knowledge of their position, and are able to relay this information to neighboring cells. Among the tools used by cells to communicate positional information are morphogens. These are secreted signaling molecules that move through a field of cells and induce responses up to several dozen cell diameters away. The defining characteristic of a morphogen is its ability to elicit different responses in different cells in a concentration-dependent manner. Because the concentration of a morphogen decreases as it moves away from its source, several different positional territories can be defined by a single morphogen gradient. Many of the proteins and mechanisms used by cells to transmit morphogen signals from the cell surface to the nucleus have been identified, and a great deal has been learned about the mechanisms by which this information flow is regulated. From this research, it appears the morphogens use the same receptors and the same signal transduction machinery to activate both low- and high-threshold responses. Thus, an intriguing question remains: how do the components of a signal transduction pathway translate the concentration of a morphogen at the cell surface into a differential response in the nucleus? For one important morphogen in Drosophila, Hedgehog (Hh), part of the answer to this question lies in the dual activity of its nuclear effector, the zinc finger transcription factor Cubitus interruptus (Ci; for a review, see Ingham and McMahon, 2001). In the absence of the Hh signal, Ci is sequestered in the cytoplasm by a protein complex containing the kinesin-like protein Costal-2 (Cos2), the serine/threonine kinase Fused (Fu), and the novel protein Suppressor of fused [Su(fu)]. This complex targets the full-length Ci protein (Ci-155) for proteolytic cleavage into a smaller repressor form (Ci-75). Ci-75 translocates into the nucleus and represses the expression of Hh target genes. When Hh binds to its receptor Patched (Ptc), it relieves the repression that Ptc normally exerts on the seven-pass transmembrane protein Smoothened (Smo). Freed of Ptc inhibition, Smo promotes the stabilization of full-length Ci, resulting in the subsequent loss of repressor Ci-75. This derepression is sufficient for the expression of low-threshold Hh target genes, such as decapentaplegic (dpp) in the wing imaginal disc. The expression of high-threshold target genes (e.g., engrailed [en] and ptc), however, requires the
translocation of Ci-155 into the nucleus and its conversion into a transcriptional activator. This activation is accomplished, at least in part, by relieving the inhibitory effects of Su(fu) on Ci-155. Importantly, these highthreshold responses also require Smo, suggesting that Smo plays a pivotal role in translating the Hh gradient by regulating the two different activities of Ci proteins. But how Smo activity regulates these downstream effects, or what actually is Smo activity for that matter, has been an enduring mystery. Several recent reports are beginning to reveal the secrets of Smo activity. In the November issue of Molecular Cell (Lum et al., 2003), Philip Beachy and colleagues report on the functional interaction between the C-terminal tail of Smo and Cos2. These results echo the work recently published by several other groups (Jia et al., 2003; Ruel et al. 2003; Ogden et al. 2003). Together these results, along with those of Mathew Scott and colleagues (Zhu et al., 2003), demonstrate that in the absence of Hh signal, Smo resides in cytoplasmic vesicles where it complexes with the Cos2, Fu, Su(fu), and Ci-155 proteins. Upon Hh stimulation, this complex translocates to the plasma membrane and the Smo, Cos2, Fu, and Su(fu) components become phosphorylated. They also show that both the binding of Smo to Cos2 and the movement of Smo to the cell surface are necessary for the transduction of the Hh signal. These findings are both satisfying and puzzling. They are satisfying because the evidence of a direct association between Smo and Cos2 closes a long-standing gap in this signal transduction pathway. We now have a direct connection between the essential signal generator Smo and the effectors that regulate Ci. In much the same way, Wong et al. (2003), also in the November issue of Molecular Cell, have closed a gap in the Wnt pathway by demonstrating physical interaction between the receptor Fz and its downstream effector Dsh. While demonstrating a direct interaction of Smo with Cos2 is an important step forward, these papers do not give us a simple explanation as to how the association between these proteins transmits the Hh signal. A number of observations in these papers do, however, give us some clues. In the absence of the Hh signal, Smo is present mostly in cytoplasmic vesicles, where it associates with the Cos2 complex. Stimulation of cells with Hh leads to the translocation of the Smo/Cos2 complex to the plasma membrane, and to an increase in the phosphorylation of most of the components. The association of Cos2 with Smo is required for the phosphorylation of Smo and its accumulation at the plasma membrane. Thus, Cos2 has a positive role in Hh signaling in addition to its more familiar negative role in the targeting Ci for cleavage. If these roles of Cos2 are mutually exclusive they could form part of a feed-forward loop, where Cos2 promotes the activation of Smo and this activated Smo diverts Cos2 away from its Ci-processing role and into the Smo activation role. This promotes the activation of more Smo, which then diverts more Cos2 and so on. The decrease in Cos2 involved in Ci cleavage would lead to the stabilization of Ci-155. The interaction of Smo with Cos2 could divert Cos2 and stabilize Ci several ways. The binding of Cos2 to Smo could promote the dissociation of Ci from Cos2, or the movement of Smo to the plasma membrane could
Developmental Cell 824
pull the Ci/Cos2 complex away from other components of the proteolytic processing machinery. The finding by Ruel et al. (2003) that Ci was largely absent from Smo and Cos2 complexes in Hh-stimulated cells is consistent with the former, whereas the opposite result (Ci accumulating in Smo/Cos2 complexes in response to Hh stimulation) presented by Lum et al. (2003) is consistent with the latter. These differences, and others, between the papers seems to indicate that the complexes formed by these proteins are quite dynamic, and that a more detailed understanding of who is binding to whom, and when and where, is necessary to understand how the association of Smo with Cos2 promotes the stabilization of Ci. Does the recruitment of the Cos2 complex to the plasma membrane by Smo constitute the role of Smo in transmitting the Hh signal? Apparently, only partially. Jiang and coworkers (Jia et al., 2003) found that they could induce stabilization of Ci and the expression of the low-threshold target gene dpp by expressing the cytoplasmic tail of Smo at the plasma membrane. This generated the same level of activity as can be obtained by preventing Ci cleavage. But stabilization of Ci is only part of the story. It is only sufficient for the induction of low-threshold responses, and the expression of highthreshold target genes requires conversion of Ci-155 into a transcriptional activator, and its translocation into the nucleus. Interestingly, Jiang’s group found that expression of the Smo cytoplasmic tail, even at very high levels, could not induce the expression of the highthreshold genes en and ptc. However, when the same construct was expressed in wing discs lacking Su(fu), ectopic expression of en and ptc was induced. While the binding of Cos2 to the C-terminal tail of Smo and the translocation of Smo to the plasma membrane is necessary for the transduction of the Hh signal, it appears that this only generates partial activity, sufficient for low-threshold responses. The expression of highthreshold target genes must require additional activities of Smo. These may be reflected by the alterations in the
Auxin Transport: The Fountain of Life in Plants?
The signaling molecule auxin is intimately associated with morphogenic processes in plants but how it influences pattern formation has been unclear. Recent studies by Friml et al. (2003), Benkova´ et al. (2003), and Reinhardt et al. (2003) have elegantly illustrated that auxin transport plays a key role during embryonic, root, and shoot organogenic processes, respectively.
Genetic and pharmacological lines of evidence suggest that auxin is intimately associated with pattern formation in plants (Friml, 2003; Bhalerao and Bennett, 2003).
phosphorylation states of Cos2, Fu, and Smo as well as Su(fu). Maximal activity can be elicited by removal of Su(fu), and clearly membrane recruitment of the Cos2 complex by Smo is not sufficient to alleviate Su(fu) suppression of Ci activity. What Su(fu) does to limit Ci activity, and how Smo regulates this function of Su(fu) are questions that remain for future work. Another important question that remains to be fully answered is how Hh binding its receptor leads to the initial activation of Smo. The changes in protein complex formation, localization, and phosphorylation appear to shed light on the effector pathway downstream of activation, but say less about the upstream part of the activation process. While these recent papers are a great step forward in our understanding of the role of Smo in Hh signaling, they also point out that Smo still has some secrets that remain to be revealed. Russell T. Collins and Stephen M. Cohen European Molecular Biology Laboratory Meyerhofstrasse 1 69117 Heidelberg Germany Selected Reading Ingham, P.W., and McMahon, A.P. (2001). Genes Dev. 15, 3059– 3087. Jia, J., Tong, C., and Jiang, J. (2003). Genes Dev. 17, 2709–2719. Lum, L., Zhang, C., Oh, S., Mann, R.K., von Kessler, D.P., Tiapale, J., Weis-Garcia, F., Gong, R., Wang, B., and Beachy, P.A. (2003). Mol. Cell 12, 1261–1274. Ogden, S.K., Ascano, M., Jr., Stegman, M.A., Suber, L.M., Hooper, J.E., and Robbins, D.J. (2003). Curr. Biol. 13, 1998–2003. Ruel, L., Rodriguez, R., Gallet, A., Lavenant-Staccini, L., and Therond, P.P. (2003). Nat. Cell Biol. 5, 907–913. Wong, H.-C., Bourdelas, A., Krauss, A., Lee, H.-J., Shao, Y., Wu, D., Mlodzik, M., Shi, D.-L., and Zheng, J. (2003). Mol. Cell 12, 1251–1260. Zhu, A.J., Zheng, L., Suyama, K., and Scott, M.P. (2003). Genes Dev. 17, 1240–1252.
Asymmetric distributions of auxin (termed auxin maxima) have been described to overlay developmental gradients in roots (Casimiro et al., 2001; Sabatini et al., 1999), leading to suggestions that auxin may act like a plant equivalent of animal morphogen and confer patterning information in a concentration threshold-dependent manner (Friml, 2003; Bhalerao and Bennett, 2003). Nevertheless, it was unclear whether auxin maxima are common to all plant organogenic events and even more importantly, how auxin influences pattern formation. Recent studies (Friml et al., 2003; Benkova´ et al., 2003; Reinhardt et al., 2003) have provided new answers to how auxin plays a key patterning function during embryonic, root, and shoot organogenic processes, respectively. The initial formation of auxin maxima appears to be common to all plant organogenic events and precedes cell specification. Friml et al. (2003) and Benkova´ et al. (2003) used an auxin-responsive reporter, DR5rev::GFP,
The Secret Life of Smoothened
The membrane protein Smoothened is the signal-generating component of the Hedgehog pathway. How Smoothened transmits the signal to downstream effectors in the cytoplasm has been a long-standing mystery. Now, with recent reports demonstrating a direct interaction between Smoothened and Costal-2, we appear to be significantly closer to solving this problem. During the development of multicellular organisms, individual cells adopt specific fates that are appropriate to their spatial and temporal position. This, naturally, requires that cells have knowledge of their position, and are able to relay this information to neighboring cells. Among the tools used by cells to communicate positional information are morphogens. These are secreted signaling molecules that move through a field of cells and induce responses up to several dozen cell diameters away. The defining characteristic of a morphogen is its ability to elicit different responses in different cells in a concentration-dependent manner. Because the concentration of a morphogen decreases as it moves away from its source, several different positional territories can be defined by a single morphogen gradient. Many of the proteins and mechanisms used by cells to transmit morphogen signals from the cell surface to the nucleus have been identified, and a great deal has been learned about the mechanisms by which this information flow is regulated. From this research, it appears the morphogens use the same receptors and the same signal transduction machinery to activate both low- and high-threshold responses. Thus, an intriguing question remains: how do the components of a signal transduction pathway translate the concentration of a morphogen at the cell surface into a differential response in the nucleus? For one important morphogen in Drosophila, Hedgehog (Hh), part of the answer to this question lies in the dual activity of its nuclear effector, the zinc finger transcription factor Cubitus interruptus (Ci; for a review, see Ingham and McMahon, 2001). In the absence of the Hh signal, Ci is sequestered in the cytoplasm by a protein complex containing the kinesin-like protein Costal-2 (Cos2), the serine/threonine kinase Fused (Fu), and the novel protein Suppressor of fused [Su(fu)]. This complex targets the full-length Ci protein (Ci-155) for proteolytic cleavage into a smaller repressor form (Ci-75). Ci-75 translocates into the nucleus and represses the expression of Hh target genes. When Hh binds to its receptor Patched (Ptc), it relieves the repression that Ptc normally exerts on the seven-pass transmembrane protein Smoothened (Smo). Freed of Ptc inhibition, Smo promotes the stabilization of full-length Ci, resulting in the subsequent loss of repressor Ci-75. This derepression is sufficient for the expression of low-threshold Hh target genes, such as decapentaplegic (dpp) in the wing imaginal disc. The expression of high-threshold target genes (e.g., engrailed [en] and ptc), however, requires the
translocation of Ci-155 into the nucleus and its conversion into a transcriptional activator. This activation is accomplished, at least in part, by relieving the inhibitory effects of Su(fu) on Ci-155. Importantly, these highthreshold responses also require Smo, suggesting that Smo plays a pivotal role in translating the Hh gradient by regulating the two different activities of Ci proteins. But how Smo activity regulates these downstream effects, or what actually is Smo activity for that matter, has been an enduring mystery. Several recent reports are beginning to reveal the secrets of Smo activity. In the November issue of Molecular Cell (Lum et al., 2003), Philip Beachy and colleagues report on the functional interaction between the C-terminal tail of Smo and Cos2. These results echo the work recently published by several other groups (Jia et al., 2003; Ruel et al. 2003; Ogden et al. 2003). Together these results, along with those of Mathew Scott and colleagues (Zhu et al., 2003), demonstrate that in the absence of Hh signal, Smo resides in cytoplasmic vesicles where it complexes with the Cos2, Fu, Su(fu), and Ci-155 proteins. Upon Hh stimulation, this complex translocates to the plasma membrane and the Smo, Cos2, Fu, and Su(fu) components become phosphorylated. They also show that both the binding of Smo to Cos2 and the movement of Smo to the cell surface are necessary for the transduction of the Hh signal. These findings are both satisfying and puzzling. They are satisfying because the evidence of a direct association between Smo and Cos2 closes a long-standing gap in this signal transduction pathway. We now have a direct connection between the essential signal generator Smo and the effectors that regulate Ci. In much the same way, Wong et al. (2003), also in the November issue of Molecular Cell, have closed a gap in the Wnt pathway by demonstrating physical interaction between the receptor Fz and its downstream effector Dsh. While demonstrating a direct interaction of Smo with Cos2 is an important step forward, these papers do not give us a simple explanation as to how the association between these proteins transmits the Hh signal. A number of observations in these papers do, however, give us some clues. In the absence of the Hh signal, Smo is present mostly in cytoplasmic vesicles, where it associates with the Cos2 complex. Stimulation of cells with Hh leads to the translocation of the Smo/Cos2 complex to the plasma membrane, and to an increase in the phosphorylation of most of the components. The association of Cos2 with Smo is required for the phosphorylation of Smo and its accumulation at the plasma membrane. Thus, Cos2 has a positive role in Hh signaling in addition to its more familiar negative role in the targeting Ci for cleavage. If these roles of Cos2 are mutually exclusive they could form part of a feed-forward loop, where Cos2 promotes the activation of Smo and this activated Smo diverts Cos2 away from its Ci-processing role and into the Smo activation role. This promotes the activation of more Smo, which then diverts more Cos2 and so on. The decrease in Cos2 involved in Ci cleavage would lead to the stabilization of Ci-155. The interaction of Smo with Cos2 could divert Cos2 and stabilize Ci several ways. The binding of Cos2 to Smo could promote the dissociation of Ci from Cos2, or the movement of Smo to the plasma membrane could
Developmental Cell 824
pull the Ci/Cos2 complex away from other components of the proteolytic processing machinery. The finding by Ruel et al. (2003) that Ci was largely absent from Smo and Cos2 complexes in Hh-stimulated cells is consistent with the former, whereas the opposite result (Ci accumulating in Smo/Cos2 complexes in response to Hh stimulation) presented by Lum et al. (2003) is consistent with the latter. These differences, and others, between the papers seems to indicate that the complexes formed by these proteins are quite dynamic, and that a more detailed understanding of who is binding to whom, and when and where, is necessary to understand how the association of Smo with Cos2 promotes the stabilization of Ci. Does the recruitment of the Cos2 complex to the plasma membrane by Smo constitute the role of Smo in transmitting the Hh signal? Apparently, only partially. Jiang and coworkers (Jia et al., 2003) found that they could induce stabilization of Ci and the expression of the low-threshold target gene dpp by expressing the cytoplasmic tail of Smo at the plasma membrane. This generated the same level of activity as can be obtained by preventing Ci cleavage. But stabilization of Ci is only part of the story. It is only sufficient for the induction of low-threshold responses, and the expression of highthreshold target genes requires conversion of Ci-155 into a transcriptional activator, and its translocation into the nucleus. Interestingly, Jiang’s group found that expression of the Smo cytoplasmic tail, even at very high levels, could not induce the expression of the highthreshold genes en and ptc. However, when the same construct was expressed in wing discs lacking Su(fu), ectopic expression of en and ptc was induced. While the binding of Cos2 to the C-terminal tail of Smo and the translocation of Smo to the plasma membrane is necessary for the transduction of the Hh signal, it appears that this only generates partial activity, sufficient for low-threshold responses. The expression of highthreshold target genes must require additional activities of Smo. These may be reflected by the alterations in the
Auxin Transport: The Fountain of Life in Plants?
The signaling molecule auxin is intimately associated with morphogenic processes in plants but how it influences pattern formation has been unclear. Recent studies by Friml et al. (2003), Benkova´ et al. (2003), and Reinhardt et al. (2003) have elegantly illustrated that auxin transport plays a key role during embryonic, root, and shoot organogenic processes, respectively.
Genetic and pharmacological lines of evidence suggest that auxin is intimately associated with pattern formation in plants (Friml, 2003; Bhalerao and Bennett, 2003).
phosphorylation states of Cos2, Fu, and Smo as well as Su(fu). Maximal activity can be elicited by removal of Su(fu), and clearly membrane recruitment of the Cos2 complex by Smo is not sufficient to alleviate Su(fu) suppression of Ci activity. What Su(fu) does to limit Ci activity, and how Smo regulates this function of Su(fu) are questions that remain for future work. Another important question that remains to be fully answered is how Hh binding its receptor leads to the initial activation of Smo. The changes in protein complex formation, localization, and phosphorylation appear to shed light on the effector pathway downstream of activation, but say less about the upstream part of the activation process. While these recent papers are a great step forward in our understanding of the role of Smo in Hh signaling, they also point out that Smo still has some secrets that remain to be revealed. Russell T. Collins and Stephen M. Cohen European Molecular Biology Laboratory Meyerhofstrasse 1 69117 Heidelberg Germany Selected Reading Ingham, P.W., and McMahon, A.P. (2001). Genes Dev. 15, 3059– 3087. Jia, J., Tong, C., and Jiang, J. (2003). Genes Dev. 17, 2709–2719. Lum, L., Zhang, C., Oh, S., Mann, R.K., von Kessler, D.P., Tiapale, J., Weis-Garcia, F., Gong, R., Wang, B., and Beachy, P.A. (2003). Mol. Cell 12, 1261–1274. Ogden, S.K., Ascano, M., Jr., Stegman, M.A., Suber, L.M., Hooper, J.E., and Robbins, D.J. (2003). Curr. Biol. 13, 1998–2003. Ruel, L., Rodriguez, R., Gallet, A., Lavenant-Staccini, L., and Therond, P.P. (2003). Nat. Cell Biol. 5, 907–913. Wong, H.-C., Bourdelas, A., Krauss, A., Lee, H.-J., Shao, Y., Wu, D., Mlodzik, M., Shi, D.-L., and Zheng, J. (2003). Mol. Cell 12, 1251–1260. Zhu, A.J., Zheng, L., Suyama, K., and Scott, M.P. (2003). Genes Dev. 17, 1240–1252.
Asymmetric distributions of auxin (termed auxin maxima) have been described to overlay developmental gradients in roots (Casimiro et al., 2001; Sabatini et al., 1999), leading to suggestions that auxin may act like a plant equivalent of animal morphogen and confer patterning information in a concentration threshold-dependent manner (Friml, 2003; Bhalerao and Bennett, 2003). Nevertheless, it was unclear whether auxin maxima are common to all plant organogenic events and even more importantly, how auxin influences pattern formation. Recent studies (Friml et al., 2003; Benkova´ et al., 2003; Reinhardt et al., 2003) have provided new answers to how auxin plays a key patterning function during embryonic, root, and shoot organogenic processes, respectively. The initial formation of auxin maxima appears to be common to all plant organogenic events and precedes cell specification. Friml et al. (2003) and Benkova´ et al. (2003) used an auxin-responsive reporter, DR5rev::GFP,