InteractiveFly: GeneBrief

smog: Biological Overview | References

BIOLOGICAL OVERVIEW

Tissue morphogenesis arises from controlled cell deformations in response to cellular contractility. During Drosophila gastrulation, apical activation of the actomyosin networks drives apical constriction in the invaginating mesoderm and cell-cell intercalation in the extending ectoderm. Myosin II (MyoII; Zipper) is activated by cell-surface G protein-coupled receptors (GPCRs), such as Smog and Mist, that activate G proteins, the small GTPase Rho1, and the kinase Rok. Quantitative control over GPCR and Rho1 activation underlies differences in deformation of mesoderm and ectoderm cells. The GPCR Smog activity is concentrated on two different apical plasma membrane compartments, i.e., the surface and plasma membrane invaginations. Using fluorescence correlation spectroscopy, the surface of the plasma membrane was probed, and it was shown that Smog homo-clusters in response to its activating ligand Fog. Endocytosis of Smog is regulated by the kinase Gprk2 and beta-arrestin-2 that clears active Smog from the plasma membrane. When Fog concentration is high or endocytosis is low, Smog rearranges in homo-clusters and accumulates in plasma membrane invaginations that are hubs for Rho1 activation. Lastly, this study found higher Smog homo-cluster concentration and numerous apical plasma membrane invaginations in the mesoderm compared to the ectoderm, indicative of reduced endocytosis. Dynamic partitioning of active Smog at the surface of the plasma membrane or plasma membrane invaginations has a direct impact on Rho1 signaling. Plasma membrane invaginations accumulate high Rho1-guanosine triphosphate (GTP) suggesting they form signaling centers. Thus, Fog concentration and Smog endocytosis form coupled regulatory processes that regulate differential Rho1 and MyoII activation in the Drosophila embryo (Jha, 2018).

Tissue morphogenesis requires control over changes in cell shape and cell-cell contacts, which depend on the spatiotemporal regulation of actomyosin contractility. In Drosophila embryos, mesoderm invagination is driven by apical constriction, a geometric cell shape change facilitated by medial-apical Myosin II activation. In the ectoderm, tissue extension arises from cell-cell intercalation, whereby cells undergo neighbor exchange through the polarized remodeling of cell junctions. Junction remodeling is driven by medial-apical MyoII contractile pulses and MyoII planar polarized accumulation (Jha, 2018).

Actomyosin contractility is regulated by conserved signaling pathways. MyoII regulatory light chain is activated by Rho-kinase (Rok) downstream of the small GTPase Rho1, which in turn is regulated by GTPase-activating proteins (GAPs) and guanine nucleotide exchange factors (GEFs). This conserved pathway was shown to be under the direct control of signaling at the cell surface, such as Celsr in vertebrate neural tube formation and G protein-coupled receptors (GPCRs) in early Drosophila embryos. The GPCRs Mist (Manning, 2013) and Smog (Kerridge, 2016) transduce signals from the secreted ligand Fog in the Drosophila presumptive mesoderm (Mist and Smog) and ectoderm (Smog). Medial-apical MyoII activation progresses downstream of hetero-trimeric G proteins Gα12/13, Gβ13F, and Gγ1 in both mesoderm and ectoderm. In the mesoderm, high medial-apical MyoII activation is under a stable regime that ensures persistent apical constriction, while in the ectoderm, intermediate medial-apical MyoII activation is under a pulsatile regime that enables cell-cell intercalation. Therefore, to understand how quantitative activation of MyoII is generated and its temporal dynamics encoded, it is necessary to decipher the regulation of GPCR signaling (Jha, 2018).

Differential MyoII activation in the mesoderm and ectoderm is partly imparted by the ligand Fog, co-expression of Mist and Smog in the mesoderm, as well as by the mesoderm-specific transmembrane protein T48, which enhances apical recruitment of RhoGEF2 and, thereby, is proposed to potentiate Rho1 and MyoII activation. High Fog expression in mesoderm activates high MyoII, while in the ectoderm low Fog expression leads to low activation of MyoII. However, in general, ligand availability is one of several mechanisms impacting GPCR activation and signaling. Various cell culture studies have focused on the other modalities that regulate GPCR signaling. The major regulators of GPCR signaling are G protein-coupled receptor kinases (GRKs) that phosphorylate GPCRs and trigger signal termination, by allowing β-arrestin binding and recruitment of other adaptor proteins. In turn, β-arrestins direct activated receptors to clathrin-coated pits and remove them from the plasma membrane by endocytosis. While removal of activated GPCRs from the plasma membrane via endocytosis terminates GPCR signaling, it also reduces the number of receptors present on the surface for ligand stimulation. This effectively sets a quantitative control over GPCR signaling via endocytosis. Drosophila has only one non-visual GRK (Gprk2) and one non-visual β-arrestin-2 (kurtz). Gprk2 mutant mothers show aberrant contractility in the mesoderm lateral cells, and it was suggested that Gprk2 attenuates Fog-dependent MyoII activation in these cells. Eggs lacking Kurtz display cuticle phenotypes and suggest gastrulation defects. These data indicate that Kurtz plays a role with Gprk2 to terminate Fog signaling and could control Rho1 and MyoII via GPCR endocytosis. Its function in the mesoderm and ectoderm has not been addressed (Jha, 2018).

Conventionally, GPCR signaling from the plasma membrane is thought to occur via ligand binding and subsequent signal transduction via G proteins that relay the information to the interior of the cell. Apart from GPCR endocytosis, the localization of GPCR within the cell membrane will influence GPCR signaling. Lateral movement of GPCRs within the plasma membrane is often restricted to specific nano-domains, suggesting that selective compartmentalization is necessary for efficient signaling as it can increase GPCR localization and clustering. GPCR clustering in the form of homo- and hetero-oligomers has been reported to control both signal amplification as well as receptor recycling. Whether the main role of GPCR clustering is for chaperoning active receptors for transport or to control GPCR signaling specificity remains unclear, especially during development. To understand GPCR signaling during tissue morphogenesis, it is important to elucidate both the clustering of GPCRs at the plasma membrane and the role of endocytosis (Jha, 2018).

This study investigated the quantitative regulation of the GPCR Smog signaling by endocytosis in both the ectoderm and the mesoderm. Fog was shown to promote homo-clusters of Smog, while endocytosis rapidly removes Smog homo-clusters from the surface of the plasma membrane in the ectoderm. Dynamic partitioning of active Smog homo-clusters in two plasma membrane compartments, the surface or the plasma invaginations, was shown to directly impact Rho1 and MyoII activation. In the mesoderm, numerous apical plasma membrane invaginations and high Smog homo-clusters correlate with high Rho1 and MyoII activation compared to the ectoderm (Jha, 2018).

Epithelial cells exhibit different types of cell deformations owing to quantitative control over cell contractility that arises from contraction of the actomyosin cytoskeleton. GPCR signaling relays information conveyed by tissue-specific factors in the mesoderm and ectoderm to control this quantitative regulation during tissue morphogenesis. Rho1-dependent activation of MyoII during both apical constriction in the mesoderm and cell-cell intercalation in the ectoderm is controlled by GPCR signaling. Activation of the GPCR Smog underlies Rho1 activation in both mesoderm and ectoderm. It is believed that differential regulation of the GPCR Smog and other GPCRs underlies these tissue-specific differences in MyoII activation. This partly relies on the fact that Fog, the activating ligand, is present at higher levels in the mesoderm than in the ectoderm. This work sheds new light on this process by probing the plasma membrane organization and distribution of Smog in conditions that affect both endocytosis and production of the ligand Fog (Jha, 2018).

Probing the ectodermal cells with FCS, Smog homo-clusters on the surface of apical plasma membrane is reported and this process depends on Fog. When Fog is absent, such as in a fog-dsRNA, the brightness per Smog::GFP unit is lower, suggesting that Fog induces the formation of Smog homo-clusters. Dynamic exchange of homo-clustered Smog occurs sbetween the surface and plasma membrane invaginations. This dynamic distribution of Smog between the two plasma membrane compartments is strongly dependent upon both the rate of Smog endocytosis and Fog concentration. Increasing Fog concentration or reducing Smog endocytosis enhances the presence of Smog homo-clusters in apical plasma membrane invaginations, which results in an apparent decrease in Smog homo-clusters at the cell surface. When Fog concentration is high under conditions where Smog endocytosis is reduced, for example, when β-arrestin-2 is knocked down, Smog homo-clusters accumulate at the surface as well as in the plasma membrane invaginations. Thus, Fog concentration and Smog endocytosis form coupled regulatory processes that control the Smog cluster formation and influence the distribution of active Smog in different plasma membrane compartments. Importantly, this controls the quantitative activation of Rho1 and MyoII. Under low-endocytosis regimes (Gprk2 or β-arrestin-2 knockdowns in the ectoderm), high levels of active Rho1 accumulate in the apical plasma membrane invaginations. It is proposed that the apical plasma membrane invaginations are signaling hubs, where signaling components could concentrate, give rise to high G protein signaling (e.g., Gα12/13), and sustain high MyoII activation. The size and stability of these signaling invaginations is tuned by endocytosis, and they may provide a means to control the strength and persistence of signaling. Pulsatile active Rho1 in the ectodermal cells requires intermediate Rho1 activation. In the ectoderm, low Fog expression and rapid Smog endocytosis by Gprk2 and β-arrestin-2 lead to intermediate activation of Rho1. In turn, intermediate Rho1 activation at the apical plasma membrane creates the conditions required for self-organized actomyosin dynamics associated with pulsation (Jha, 2018).

This study also points to the possibility of tissue level regulation of endocytosis and plasma membrane compartmentalization of GPCRs. Large apical plasma membrane invaginations are observed in the mesoderm compared to the ectoderm. In the mesoderm, Smog accumulates in larger, more numerous, apical plasma membrane invaginations, and it displays larger Smog homo-clusters compared to in the ectoderm. In the mesoderm, Rho1 and MyoII activation is higher. Another GPCR, Mist produced in the mesoderm, works synergistically with Smog to boost Rho1 and MyoII activation (Manning, 2013). This is also due to the expression of another GPCR Mist in the mesoderm and to Fog being present at higher levels in the mesoderm. Ectodermal cells have similar properties of high Smog homo-clusters when Fog is overexpressed and GPCR endocytosis is slowed down. An intriguing possibility is that Smog and potentially Mist endocytosis is downregulated in the mesoderm compared to the ectoderm. Interestingly, the E3 ubiquitin ligase Neuralized (Neur), which is uniformly expressed in the embryo, is inhibited in the ectoderm by the small proteins of the Bearded (Brd) family. Brd genes are repressed by the mesoderm transcription factor Snail, so that Neur is only active in the mesoderm. In a Brd mutant, where Neur becomes active in the ectoderm, MyoII activation is increased and Neur degradation or repression in the mesoderm following Brd overexpression both reduce MyoII activation. Previous studies have shown that the E3 ubiquitin ligase targets β-arrestin-2 for ubiquitination and degradation, and, thereby, it affects endocytosis and signaling by GPCRs. It is possible that GPCR endocytosis could be reduced in the mesoderm due to increased Neur activity in this tissue. This may depend on the downregulation of several target proteins, such as β-arrestin-2 (Jha, 2018).

Selective compartmentalization of GPCR on the plasma membrane as in the case of large apical plasma membrane invaginations can increase the concentration and the probability of GPCR clustering and oligomerization. The current data suggest that the dynamic modulation of GPCR signaling can be achieved by a change in their cluster/oligomer formation. Receptor oligomerization may enlarge the signaling capacities by the recruitment of more downstream signaling components during GPCR signaling. G proteins are reported to be expressed at low concentration, and selective compartmentalization of GPCRs on the plasma membrane further increase the probability of GPCR clustering and oligomerization for efficient signaling. Investigation of G protein activation by different GPCRs in vivo will be needed to test if a similar mechanism is in place during epithelial morphogenesis (Jha, 2018).

Modular activation of Rho1 by GPCR signalling imparts polarized myosin II activation during morphogenesis

Polarized cell shape changes during tissue morphogenesis arise by controlling the subcellular distribution of myosin II. For instance, during Drosophila gastrulation, apical constriction and cell intercalation are mediated by medial-apical myosin II pulses that power deformations, and polarized accumulation of myosin II that stabilizes these deformations. It remains unclear how tissue-specific factors control different patterns of myosin II activation and the ratchet-like myosin II dynamics. This study reports the function of a common pathway comprising the heterotrimeric G proteins Gα12/13 (Concertina), Gβ13F and Gγ1 in activating and polarizing myosin II during Drosophila gastrulation. Gα12/13 and the Gβ13F/γ1 complex constitute distinct signalling modules, which regulate myosin II dynamics medial-apically and/or junctionally in a tissue-dependent manner. A ubiquitously expressed GPCR called Smog (Poor gastrulation, Pog & CG31660) was identified as being required for cell intercalation and apical constriction. Smog functions with other GPCRs to quantitatively control G proteins, resulting in stepwise activation of myosin II and irreversible cell shape changes. It is proposed that GPCR and G proteins constitute a general pathway for controlling actomyosin contractility in epithelia and that the activity of this pathway is polarized by tissue-specific regulators (Kerridge, 2016).

During tissue morphogenesis, cells rearrange their contacts to invaginate, intercalate, delaminate or divide. During Drosophila gastrulation, invagination of the presumptive mesoderm in the ventral region of the embryo and of the posterior midgut requires apical cell constriction, a geometric deformation that occurs in different organisms. Elongation of the ventral-lateral ectoderm requires cell intercalation, a general topological deformation associated with junction remodelling. In the ectoderm, the so-called 'vertical junctions', oriented along the dorsal-ventral axis, shrink, followed by extension of new 'horizontal' junctions along the anterior-posterior axis. Despite differences in the cell deformations associated with intercalation and apical constriction, recent studies revealed that both processes require myosin II (MyoII) contractility. Cell shape changes rely on the pulsatile activity of MyoII in the apical-medial cortex, whereby MyoII undergoes cycles of assembly and disassembly allowing stepwise deformation1. Moreover, each step of deformation is stabilized and thereby retained, contributing to the irreversibility of tissue morphogenesis. In the mesoderm, each phase of apical area constriction mediated by MyoII pulses is followed by a phase of shape stabilization involving persistence of medial MyoII. In the ectoderm, medial-apical MyoII pulses flow anisotropically towards vertical junctions resulting in steps of shrinkage that are stabilized by a planar-polarized pool of junctional MyoII. This ratchet-like behaviour of MyoII is regulated by the Rho1-Rok pathway and requires quantitative control over MyoII activation. Low Rho1/Rok activity fails to form actomyosin networks, intermediate activation establishes MyoII pulsatility and high activation confers stability. The signalling mechanisms that cause stepwise activation of MyoII by Rho1 remain unknown. It is also unclear whether different pathways for Rho1 activation operate in the mesoderm and in the ectoderm as indeed Rho1 can be activated by numerous signalling mechanisms or whether a common pathway might exist (Kerridge, 2016).

Tissue-specific factors can result in polarized shape changes by signalling through cell surface receptors. For instance, in Drosophila ectoderm, pair rule genes encoding transcription factors control planar-polarized enrichment of MyoII through the combinatorial expression of the surface proteins Toll2, Toll6 and Toll8 in stripes. Likewise, in the mesoderm, Twist and Snail induce expression of Fog, a secreted ligand, and a G-protein-coupled receptor (GPCR) Mist (methuselah-like 1), which is reported to transduce Fog. The downstream G protein Gα_12/13 (known as Concertina (Cta) in Drosophila) is required for RhoGEF2 and thereby MyoII apical recruitment. As RhoGEF2 is a known GEF for Rho1, the requirement of Gα_12/13 for RhoGEF2 apical recruitment suggests that GPCRs and G-protein signalling mediate MyoII activation through the Rho1 pathway. These considerations prompted asking whether G-protein signalling directly controls the different regimes of MyoII dynamics (pulsatility and/or stability) in the mesoderm and planar polarized activation of Rho1 and MyoII in the ectoderm (Kerridge, 2016).

This study reports the function of the heterotrimeric G proteins Gα_12/13, Gβ13F and Gγ1 in activating and regulating MyoII dynamics both in the mesoderm and in the ectoderm. Receptor activation, through the GEF activity of the GPCR, converts Gα from an inactive GDP-bound state, in a complex with Gβγ, to an active GTP-bound state. This results in dissociation of Gβγ, enabling binding of both Gα-GTP and Gβγ to their respective effectors for signalling. This study found that Gα_12/13 and the Gβ13F/Gγ1 complex constitute distinct signalling modules, which regulate MyoII dynamics medial-apically and/or junctionally in a tissue-dependent manner. A ubiquitously expressed GPCR called Smog, was found to be required for cell shape changes associated with both mesoderm invagination and ectoderm elongation. During these morphogenetic events, Smog functions with other GPCRs, Mist in the mesoderm and an as yet unknown GPCR in the ectoderm, to activate the Rho1-Rok pathway. This results in stepwise activation of Rho1 and MyoII, ensuring irreversible cell shape changes (Kerridge, 2016).

First, this study reports that Gα_12/13 and Gβ13F/Gγ1 function as distinct signalling modules that control Rho1 and MyoII in different domains. Gα_12/13 activates medial-apical MyoII through its effector RhoGEF2 both in the ectoderm and the mesoderm. In mammals, p115-RhoGEF interacts directly with Gα12 suggesting that this may be a conserved signalling module. In contrast, Gβ13F/Gγ1 activates MyoII both at cell junctions and in the medial-apical domain. This modularity may provide distinct regulatory mechanisms for the activation of MyoII in different subcellular compartments owing to the existence of different molecular effectors of Gα-GTP and Gβγ. Second, stepwise activation of Rho1 by multiple GPCRs and their ligands determines the emergence of a pulsatile regime medial-apically, or stable activation. In the mesoderm, Smog and Mist GPCRs, together with high expression of their ligand Fog, ensure stabilization and rapid (<5 min) accumulation of MyoII ensuring apical constriction. In the ectoderm, low Fog expression and thus lower activation of Gα_12/13 and RhoGEF2 is responsible for intermediate medial-apical activation of MyoII and pulsatility. Indeed, Fog, constitutively active Gα_12/13QL and RhoGEF2 overexpression all lead to stable accumulation of MyoII instead of pulsation, similar to constitutively active RhoV14 (Kerridge, 2016).

Interestingly, the same receptor Smog controls MyoII activation in different subcellular domains during intercalation and apical constriction begging the question of how activation of Gα_12/13 and Gβγ is differentially achieved in the ectoderm and the mesoderm. The polarization of Smog activation is to some extent imparted by the ligand. Fog/Smog regulates medial-apical accumulation of MyoII in the two tissues: Fog induces medial Rho1 and Rok activation in the mesoderm and ectoderm and, when ectopically expressed in the ectoderm, it can increase Rho1 and Rok in the medial cortex. This argues that another mechanism results in junction-specific activation of Smog, Gβ13F/Gγ1, Rho1 and Rok in the ectoderm (Kerridge, 2016).

It is possible that an unknown ectoderm-specific ligand activates Smog specifically at junctions. Junctional localization of the Rho1 pathway by Smog may also be imparted by subcellular processing of Smog signalling, such as localization/activation of downstream effectors of Gα_12/13 and Gβγ. The recently identified Toll receptors required for MyoII planar-polarized activation may bias Smog signalling although the molecular mechanisms remain unclear. This could be through localization of RhoGEFs. In the mesoderm, the transmembrane protein T48 localizes RhoGEF2 apically through binding to its PDZ domain, and is required for apical MyoII activation in parallel with Smog, Gα_12/13 and Gβγ. Similarly, other GEFs may be required for junctional Rho1 activation by Smog (Kerridge, 2016).

What might be the advantage of having multiple GPCRs? Gastrulation sets the foundation for all other future processes in development and hence requires robustness. GPCRs with similar functions yet subtle differences such as ligand specificity may offer advantages compared with single ligand-receptor pairs. For instance, high cortical tension associated with mesoderm invagination may require multiple GPCRs activating parallel pathways to attain efficiency of the process. Moreover, multiple GPCRs may concede tissue-specific regulation of the common G-protein subcellular pathways. Finally, multiple GPCRs can allow stepwise activation of MyoII. Although activation by one GPCR is sufficient to induce pulsatility, more GPCRs are required to shift the actomyosin networks to more stable regimes (Kerridge, 2016).

The discovery that Smog and heterotrimeric G protein activate Rho1 and MyoII in two different morphogenetic processes provides a potentially general molecular framework for tissue mechanics. It is proposed that different developmental inputs tune a common GPCR/G-protein signalling pathway to direct specific patterns and levels of Rho1 activation. Quantitative control specifies the regime of MyoII activation through Rho1, namely pulsatility or stability of MyoII. Modular control defines the subcellular domains where MyoII accumulates (medial-apical or junctions) depending on molecular effectors. How developmental signals tune GPCR signalling will be important to decipher (Kerridge, 2016).

Search PubMed for articles about Drosophila Smog

Jha, A., van Zanten, T. S., Philippe, J. M., Mayor, S. and Lecuit, T. (2018). Quantitative control of GPCR organization and signaling by endocytosis in epithelial morphogenesis. Curr Biol 28(10): 1570-1584 PubMed ID: 29731302

Kerridge, S., Munjal, A., Philippe, J. M., Jha, A., de Las Bayonas, A. G., Saurin, A. J. and Lecuit, T. (2016). Modular activation of Rho1 by GPCR signalling imparts polarized myosin II activation during morphogenesis. Nat Cell Biol 18(3): 261-70. PubMed ID: 26780298

Manning, A. J., Peters, K. A., Peifer, M. and Rogers, S. L. (2013). Regulation of epithelial morphogenesis by the G protein-coupled receptor mist and its ligand fog. Sci Signal 6: ra98. PubMed ID: 24222713

date revised: 5 August 2021

Home page: The Interactive Fly © 2011 Thomas Brody, Ph.D.