Rho1
The steroid hormone 20-hydroxyecdysone (ecdysone) is the key regulator of postembryonic developmental transitions in insects and controls metamorphosis by triggering the morphogenesis of adult tissues from larvae. The Rho GTPase, which mediates cell shape change and migration, is also an essential regulator of tissue morphogenesis during development. Rho activity can modulate gene expression, in part, by activating LIM kinase (LIMK) and consequently affecting actin-induced SRF transcriptional activity. A link has been established between Rho-LIMK-SRF signaling and the ecdysone-induced transcriptional response during Drosophila development. Specifically, Rho GTPase, via LIMK, regulates the expression of several ecdysone-responsive genes, including those encoding the ecdysone receptor itself, a downstream transcription factor (Br-C), and Stubble, a transmembrane protease required for proper leg formation. Stubble and Br-C mutants exhibit strong genetic interactions with several Rho pathway components in the formation of adult structures, but not with Rac or Cdc42. In cultured SL2 cells, inhibition of Rho, F-actin assembly, or SRF blocks the transcriptional response to ecdysone. Together, these findings indicate a link between Rho-LIMK signaling and steroid hormone-induced gene expression in the context of metamorphosis and thereby establish a novel role for the Rho GTPase in development (Chen, 2004).
Metamorphosis in Drosophila is stringently controlled by pulses of the steroid hormone ecdysone at discrete developmental stages. During larval-pupal transition, ecdysone triggers coordinated changes in tissue morphology that involve histolysis of larval tissues and the initiation of adult structures. Rho GTPase-mediated signaling pathways have been implicated in several aspects of morphogenesis during Drosophila embryo formation. However, a role for Rho signaling in metamorphosis has not yet been reported. Among the downstream mediators of Rho signaling are the LIM kinases, and a closely related Drosophila ortholog of mammalian LIM kinases (designated Dlimk) is specifically expressed at relatively high levels in late larval and pupal stages, suggesting a potential role in Rho-LIMK signaling during this transition. In adult flies, Dlimk is expressed at substantially higher levels in males than in females, consistent with a potential evolutionarily conserved role in spermatogenesis, a process in which mammalian LIMK2 has been implicated. Dlimk mRNA is uniformly expressed throughout eye, wing, and leg imaginal discs (Chen, 2004).
The GAL4/UAS transgene system was used to examine Dlimk function in vivo. Overexpression of Dlimk in the imaginal wing disc via several different wing-specific GAL4 drivers causes notched wings, missing wing veins, vein fusion, and blistered wings. The notched wing phenotype appears to reflect an increase in apoptosis and is rescued by the p35 viral caspase inhibitor. In addition, wings exhibit enlarged cells (indicated by low wing hair density) and alterations in the number and polarity of wing hairs. Notably, similar defects in wing hair number and polarity are also seen in rho1 and Drok (the Rho effector kinase that activates LIMK) mutants, suggesting that Dlimk functions in the same pathway. Mammalian LIMKs promote actin assembly in cultured cells, and prominent F-actin accumulation and aberrant actin organization are observed in the wing discs of transgenic flies specifically overexpressing Dlimk. Thus, Dlimk can regulate actin assembly in developing tissues (Chen, 2004).
To verify that Dlimk normally regulates morphogenesis during the larval-pupal transition, a kinase-deficient form of Dlimk (DlimkD522A) was used as a dominant-negative protein. An analogous mutation in mammalian LIMK1 gives rise to a protein that specifically interferes with LIMK function in cultured cells. Use of a T80-GAL4/UAS-DlimkD522A transgenic line to express DlimkD522A during development results in viable and fertile animals, with approximately 85% of adults exhibiting malformed wings and legs, consistent with a normal requirement for Dlimk in proper disc morphogenesis. In wild-type adult legs, the femur and tibia are elongated and slender structures; however, in DlimkD522A mutant flies, the femur is bent and twisted, and the tibia is often shorter and twisted. In addition, wings are malformed and are approximately 40% smaller than those of wild-type flies. Coexpression of DlimkD522A and wild-type Dlimk results in flies whose wings appear normal, indicating that the effects of dominant-negative Dlimk result from specific inhibition of the endogenous wild-type Dlimk as opposed to nonspecific interference with an unrelated signaling pathway (Chen, 2004).
The malformed legs in DlimkD522A flies closely resemble leg defects in flies in which Rho signaling is perturbed through genetic disruption of Rho1, DrhoGEF2 (a guanine nucleotide exchange factor for Rho1), sqh (myosin light chain), and zipper (nonmuscle myosin heavy chain). Sqh and zipper are downstream targets of Drok and regulate actomyosin contractility. Loss-of-function mutants of Rho1 or DrhoGEF2 strongly suppress the severity of wing defects associated with Dlimk expression. Reducing Rho activity by overexpressing the potent Rho inhibitor, p190 RhoGAP, also efficiently suppresses Dlimk-induced wing defects. Moreover, reducing levels of Diaphanous or Drok, two Rho targets that promote actin assembly, also substantially reduces the severity of Dlimk-induced wing defects. A loss-of-function allele of blistered, the Drosophila SRF ortholog, also suppresses the Dlimk-induced wing defects, suggesting that regulation of SRF-dependent transcription by Rho-LIMK signaling plays a role in wing morphogenesis. Significantly, in mammalian cells, LIMK and Diaphanous cooperate to regulate SRF activity (Geneste, 2002). Reducing levels of the Rho-related GTPases, Rac1, Rac2, and Cdc42, or the Rac activator, Myoblast city (Mbc), or the Rac/Cdc42 effector target, PAK, has very little effect on the Dlimk-induced wing phenotype. Thus, it appears that in the developing leg and wing, Dlimk specifically mediates a Rho-actin signaling pathway required for imaginal-disc morphogenesis (Chen, 2004).
Defects in leg morphogenesis resembling those in DlimkD522A flies are seen in mutants of several ecdysone-inducible genes, including those encoding the transcription factor, Broad-complex (BR-C or br), and Stubble (Sb), a transmembrane serine protease (Appel, 1993). Both genes are required for disc morphogenesis during larval-pupal transition. Significantly, br mutants interact genetically with mutants of Sb, zipper, and blistered during imaginal-disc morphogenesis (Beaton, 1988: Gotwals, 1991), suggesting that the observed role for a Rho-Dlimk pathway in leg morphogenesis could reflect a requirement for this pathway in the response to ecdysone (Chen, 2004).
To determine if the Rho-Dlimk pathway interacts genetically with br or Sb, a heat-shock-inducible DlimkD522A transgene that exhibits a low-penetrance malformed leg phenotype was crossed with br and Sb mutants. DlimkD522A and mutants of several Rho signaling components strongly interact with Sb63b and Sb70, two dominant-negative alleles of Stubble, to produce malformed legs at a high frequency. However, mutants of the Rho-related GTPases, Rac1, Rac2, and Cdc42, do not enhance the frequency of leg defects. Similarly, several components of the Rho-Dlimk pathway, but not Rac and Cdc42, strongly interact with br1 in an analogous genetic interaction test (Chen, 2004).
The observed interactions among Rho1, Dlimk, br, and Sb support a role for Rho signaling in ecdysone-regulated metamorphosis. However, neither Rho1 expression nor activation is ecdysone inducible. In light of studies linking Rho-LIMK signaling to effects on gene expression (Sotiropoulos, 1999), BR-C and Sb expression were examined in flies overexpressing Rho1, Dlimk, or p190 RhoGAP during early puparium stages, when disc morphogenesis is underway. Expression of BR-C and Sb mRNA normally peaks approximately 2-4 hr after puparium formation. However, in flies overexpressing Rho1 or Dlimk, expression of these genes persists well beyond the normal peak of expression seen in 'driver-only' control flies (approximately 8–10 hr after puparium formation. Moreover, expression of these genes is greatly reduced at all stages of pupation in flies expressing p190 RhoGAP. Significantly, although most of the transgenic flies that overexpress p190 RhoGAP die at a late pupal stage, the few 'escapers' that eclose exhibit malformed wings and twisted and bent leg phenotypes that are very similar to those seen in flies expressing DlimkD522A . In addition, the pupal lethality that is frequently observed with overexpression of p190 RhoGAP is efficiently rescued by coexpressing Dlimk, indicating that the late developmental defects that arise as a consequence of Rho inactivation largely reflect defects in Rho-LIMK signaling (Chen, 2004).
Expression of the ecdysone receptor (EcR) itself is similarly regulated by Rho1 and Dlimk. However, loss-of-function alleles of the EcR or Sb fail to rescue the effects of overexpressing Dlimk, suggesting that Rho-LIMK signaling controls additional aspects of metamorphosis independently of its effects on the ecdysone response. Thus, Rho-LIMK signaling may play a role in coordinating Rho-directed cell shape changes and movements with ecdysone-induced gene expression during tissue morphogenesis (Chen, 2004).
Many of the identified transcriptional targets of the ligand-activated ecdysone receptor are, themselves, transcription factors, which are not the actual effectors of tissue morphogenesis. However, the Stubble gene, which is highly sensitive to Rho-LIMK signaling, encodes a protein that participates directly in morphogenesis through its ability to promote remodeling of the extracellular matrix. The ability of Rho to direct both actin-mediated cell shape changes and the expression of a cell surface protease provides a potential mechanism for coordinately regulating these two major components of tissue morphogenesis during development (Chen, 2004).
To examine more directly a requirement for a Rho-actin-SRF pathway in the transcriptional response to ecdysone, Drosophila SL2 cells were used. In SL2 cells, as in developing discs, ecdysone induces the expression of EcR mRNA. Transfection of cells with the Rho-inhibitory C3 toxin or pretreatment with the actin polymerization inhibitor, latrunculin B, substantially reduces the ecdysone-induced increase in EcR mRNA but does not affect transcription of the ecdysone-insensitive gene rp49 or the Rho1 gene. As expected, latrunculin B completely inhibits morphogenesis of leg appendages, indicating a requirement for F-actin assembly. To examine the role of SRF in ecdysone-induced EcR expression, SL2 cells were treated with RNAi corresponding to the blistered gene. RNAi-treated cells exhibit reduced SRF expression and an absence of ecdysone-induced EcR mRNA expression. Together, these results suggest that the ability of Rho and Dlimk to promote F-actin assembly and SRF activation is responsible for their effects on ecdysone-responsive gene expression and tissue morphogenesis. In addition, the findings in SL2 cells indicate that the observed effects of Rho-SRF signaling on the ecdysone response are cell-autonomous effects. Interestingly, genetic interactions have been observed between zipper and sb and between zipper and br, suggesting that Rho-regulated actomyosin contractility, in addition to F-actin assembly, may also influence the ecdysone response. In this regard, it is interesting to note that mechanical stretching of cells reportedly promotes SRF activity. Alternatively, actomyosin contractility may play a parallel role in disc morphogenesis that is independent of any direct regulation of the ecdysone response (Chen, 2004).
No motif has been identified within the 5′ and 3′ regulatory sequences (2 kb each) of the EcR gene has been identified that matches the reported SRF binding consensus site. Hence, it remains possible that an SRF-regulated coactivator of ecdysone receptor gene expression is a primary target of Rho-Dlimk signaling. It is interesting to note that the Drosophila transcription factor, Crooked legs, regulates expression of ecdysone receptor mRNA and is encoded by an ecdysone-inducible gene that is also required for wing and leg morphogenesis. Such findings highlight the complexity of the gene expression hierarchy involved in the morphogenetic response to ecdysone and indicate a likely role for transcriptional feedback mechanisms (Chen, 2004).
It is concluded that Rho GTPase-mediated signal transduction to the actin cytoskeleton and ecdysone-induced gene expression are both critical regulatory components of tissue morphogenesis during Drosophila development. A direct relationship has been described between these two pathways in the context of metamorphosis. Specifically, these findings indicate that Rho, through its ability to activate LIMK and promote actin polymerization, regulates the expression of several ecdysone-responsive genes, including the ecdysone receptor itself. By modulating the expression of ecdysone-responsive genes, including a cell surface protease, the Rho-LIMK signaling pathway appears to play a critical role in regulating the proper morphogenesis of adult structures from the imaginal discs of larvae. This connection represents a previously unrecognized link between Rho GTPase signaling and nuclear hormone signaling that potentially plays a broader role in additional developmental contexts (Chen, 2004).
Paxillin is a prominent focal adhesion docking protein that regulates cell adhesion and migration. Although numerous paxillin-binding proteins have been identified and paxillin is required for normal embryogenesis, the precise mechanism by which paxillin functions in vivo has not yet been determined. An ortholog of mammalian paxillin in Drosophila (Dpax) has been identified and a genetic analysis of paxillin function during development was undertaken. Overexpression of Dpax disrupts leg and wing development, suggesting a role for paxillin in imaginal disc morphogenesis. These defects may reflect a function for paxillin in regulation of Rho family GTPase signaling since paxillin interacts genetically with Rac and Rho in the developing eye. Moreover, a gain-of-function suppressor screen identified a genetic interaction between Dpax and center divider cdi in wing development. Cdi belongs to the cofilin kinase family, which includes the downstream Rho target, LIM kinase (LIMK). Significantly, strong genetic interactions were detected between Dpax and Dlimk, as well as downstream effectors of Dlimk. Supporting these genetic data, biochemical studies indicate that paxillin regulates Rac and Rho activity, positively regulating Rac and negatively regulating Rho. Taken together, these data indicate the importance of paxillin modulation of Rho family GTPases during development and identify the LIMK pathway as a critical target of paxillin-mediated Rho regulation (Chen, 2004).
Paxillin is a scaffolding protein found in focal adhesions. Targeted disruption of paxillin in mice results in an early embryonic lethal phenotype with defects in multiple mesodermally derived structures. The recent completion of the Drosophila genome revealed the evolutionary conservation of many of the key molecules found in focal adhesions, including integrins, paxillin, vinculin, FAK, p130CAS, and ILK. The Drosophila paxillin is predominantly expressed in embryos, pupae, and male adults. In situ analysis of staged embryos reveals a restricted expression pattern of Dpax. In particular, Dpax is highly expressed in tissues undergoing cell shape changes or cell migration. Overexpression of Dpax in late larval stages results in a pupal lethal phenotype with few escapers bearing malformed phenotypes, suggesting that Dpax also plays an important role during later stages of development (Chen, 2004).
A loss-of-function mutant of Drosophila paxillin has not yet been reported. Therefore, the UAS/GAL4 system was employed to investigate the function of Dpax in the later stages of development. As has been reported for Drosophila FAK, overexpressing Dpax results in a blistered-wing phenotype. In mammals, paxillin is a substrate of FAK in transducing signals from integrins. FAK regulates focal adhesion disassembly and has been shown to be involved in Drosophila Wnt4-mediated cell movement during ovarian morphogenesis and is also required for border cell migration during oogenesis. The function of Dpax in oogenesis is not clear; however, Dpax is also highly expressed in the border cells (Chen, 2004).
The blistered-wing phenotype is also found in integrin mutant flies. In the prepupal stage, the wing is a single epithelial sheet, and integrins have been suggested to play a regulatory role. As development progresses this sheet folds into a dorsal and ventral side, and the integrins play an adhesive role at these later stages. Using drivers that are expressed at different stages of development, the studies suggest that paxillin could be important for both the regulatory and adhesive functions of the integrins. Such functions would be consistent with studies of mammalian systems in which paxillin functions downstream of multiple integrins and can regulate both inside out and outside in signaling. In addition, both paxillin and FAK are important for focal adhesion turnover. Thus, too much paxillin or FAK may increase the turnover of focal complexes and perturb the stable adhesion between two epithelia, thereby resulting in the blistering phenotype (Chen, 2004).
Using a gain-of-function screen for modifiers that can rescue the Dpax-induced wing blistering, Cdi/TESK was identified. Like LIMK, Cdi/TESK phosphorylates the actin-depolymerizing factor cofilin and stabilizes F-actin. Cdi/TESK is highly homologous to LIMK in the kinase domain; however, a recent study has demonstrated that Cdi/TESK functions downstream of Rac1 during spermatogenesis (Raymond, 2004). Drosophila LIMK functions downstream of Rho1 in regulating disk morphogenesis (Chen, 2004). Dlimk and components in the Rho-LIMK pathway, including ssh, tsr, and bs/DSRF, also rescue the blistering phenotype. In addition, another regulator of SRF and actin, diaphanous, also shows genetic interactions with Dpax. Diaphanous is a direct effector of Rho which cooperates with LIMK to regulate SRF activation. All of these components play important roles in regulating F-actin synthesis. Taken together, these data indicate that it is possible that an increase in actin levels can prevent the increase in focal adhesion turnover caused by the excess level of paxillin, therefore suppressing the blistering phenotype. It is possible that simply overexpressing actin might be sufficient to rescue the blistering phenotype, although the results suggest that paxillin itself does not affect F-actin synthesis or actin organization. The ability of paxillin, however, to coimmunoprecipitate with LIMK and the increased cofilin phosphorylation in Pxl–/– MEFs suggests that paxillin can modulate LIMK function. These data, combined with the genetic and biochemical evidence that paxillin can regulate Rho, suggest that paxillin could act at multiple points to regulate the Rho pathway (Chen, 2004).
Interestingly, while modulation of some components downstream of Rho is able to suppress the blistering phenotype, overexpression of other components such as ROK does not alter this phenotype. While this could reflect insufficient expression levels or more complex regulation of ROK, the data suggest that paxillin's regulation of the Rho pathway may involve either modulation of only certain downstream components or a lack of function for these components in the paxillin-induced phenotypes (Chen, 2004).
Rho GTPases play an important role in regulating actin cytoskeleton organization. Genetic and biochemical analysis reveal that paxillin activates Rac signaling but inactivates Rho signaling. Previous binding and localization studies suggest that mammalinan paxillin may regulate Rac through its indirect association with at least two Rac exchange factors. Pix/Cool is linked to paxillin via PKL/Git2, the ARF-GAP, and overexpression studies with mutants of paxillin and other members of this complex have led to the suggestion that paxillin may be important for recruiting this complex to focal contacts. A second binding partner, Crk, can also link paxillin to Rac activation via a nontraditional exchange factor, Dock180. Mislocalization of one or both complexes in Pxl–/– mouse embryo fibroblasts (MEF)s could therefore lead to defects in Rac activation and subsequent defects in lamellipodium dynamics and migration. Both Pix/Cool and Crk localization were examined in rescued and Pxl–/– MEFs and only a minor decrease was detected in Cool and Crk positive peripheral adhesions in Pxl–/– cells. In MEFs, therefore, paxillin is not required for localization of these proteins to peripheral adhesions. This may be due to functional redundancy, as the paxillin family member Hic-5 can also bind the PKL-Pix complex and Crk can bind to other focal adhesion proteins, including p130Cas. In any case, mislocalization of these complexes is unlikely to account for the differences in Rac activation. In contrast, genetic studies of Drosophila have shown that deletion of a region encompassing the Drosophila homolog of Cool was able to suppress the Dpax-induced blistering. Thus, one potential mechanism by which paxillin may control Rac activation in Drosophila is through regulation of Pix/Cool. Since Rac and Rho have been shown to antagonize each other, it remains possible that in higher eukaryotes, paxillin could indirectly regulate Rac via regulation of Rho (Chen, 2004).
It is not clear how paxillin down-regulates Rho activity. Paxillin might be important for spatial regulation of Rho activity and/or controlling the activity or localization of a Rho GAP or GEF. Two Rho GAPs have been linked to mammalian paxillin. Graf is a Rho GAP that was originally identified as a Fak-binding partner, and a homolog of this protein has been identified in Drosophila studies. Since paxillin can interact with Fak, it is possible that loss of paxillin may somehow affect Graf localization or activation. While Fak localization to focal adhesions is less efficient in Pxl–/– MEFs, the effects are minimal and thus this is unlikely to account for the enhanced Rho activity. It is worth noting that it has recently been reported that mammalian paxillin binds to the p120 RasGAP and competes with p120 RasGAP for binding to p190 RhoGAP. It has been suggested that paxillin inhibits Rho by promoting the formation of free p190 RhoGAP. The Drosophila ortholog of p190 RhoGAP does not bind to the Drosophila p120 RasGAP. In addition, only minor changes in p190 localization to the leading edge were detected in Pxl–/– MEFs. Thus, paxillin may antagonize Rho function through multiple distinct regulatory mechanisms (Chen, 2004).
Taken together, these data suggest that while paxillin has the ability to interact with multiple proteins involved in diverse signaling pathways, a major function of this scaffolding protein in vivo is to regulate Rho family GTPases. Thus, misregulation of these GTPases is likely to account for the adhesion defects observed during development in mouse and Drosophila studies (Chen, 2004).
The correct localization of myosin II to the equatorial cortex is crucial for
proper cell division. A collection of genes was examined that causes defects
in cytokinesis and revealed (with live cell imaging) two distinct phases of myosin
II localization. Three genes in the rho1 signaling pathway, pebble (a Rho
guanidine nucleotide exchange factor), rho1, and rho kinase, are
required for the initial recruitment of myosin II to the equatorial cortex. This
initial localization mechanism does not require F-actin or the two components of
the centralspindlin complex, the mitotic kinesin pavarotti/MKLP1 and
racGAP50c/CYK-4. However, F-actin, the centralspindlin complex, formin
(diaphanous), and profilin (chickadee) are required to stably
maintain myosin II at the furrow. In the absence of these latter genes, myosin
II delocalizes from the equatorial cortex and undergoes highly dynamic
appearances and disappearances around the entire cell cortex, sometimes
associated with abnormal contractions or blebbing. These findings support a model
in which a rho kinase-dependent event, possibly myosin II regulatory light chain
phosphorylation, is required for the initial recruitment to the furrow, whereas
the assembly of parallel, unbranched actin filaments, generated by
formin-mediated actin nucleation, is required for maintaining myosin II
exclusively at the equatorial cortex (Dean, 2005).
This study has discovered three steps in the myosin II
localization/activation process that involve distinct groups of genes:
(1) an initial recruitment of myosin II to the equatorial cortex that is
independent of F-actin and centralspindlin but requires rho1 signaling;
(2) a secondary stabilization of myosin II at the midzone that requires
F-actin and a second set of genes that are likely involved in building a
specific type of actin network, and (3) the activation of furrowing
once myosin II is localized that depends on centralspindlin (Dean, 2005).
Rho1, its
activating guanidine nucleotide exchange factor pebble, and rho kinase are each
required for the initial recruitment of myosin II to the equatorial cortex. Rho1
has been implicated in two pathways that are important for cytokinesis.
In the first pathway, rho1 signals to F-actin
through the formin diaphanous. However, proteins on this F-actin
pathway, including F-actin itself, are not essential for the initial myosin II
recruitment to the equatorial cortex. However, rho kinase, another
downstream target of rho1, is essential. Because rho kinase phosphorylates the
myosin II RLC, it is possible that
phosphorylation of the RLC is essential for myosin II recruitment to the furrow.
This hypothesis could not be directly tested, because the myosin II heavy
chain forms large aggregates when the RLC is depleted by RNAi (Dean, 2005).
Phosphorylation of the RLC both activates the motor domain and, in some myosins,
increases bipolar thick filament formation. Because
F-actin is not required for myosin II recruitment, activation of the motor is
unlikely to be the mechanism by which phosphorylation of the RLC would cause
recruitment of myosin II to the equatorial cortex. It is quite possible,
however, that the rho kinase-mediated myosin II phosphorylation leads to thick
filament assembly and that this assembly is important for localization of myosin
to the equatorial cortex. Indeed, in Dictyostelium, it is clear that
bipolar thick filament formation is sufficient for myosin II localization to the
midzone of a mitotic cell. The nonactin-based mechanism of recruitment of myosin II filaments remains
unknown (Dean, 2005).
In contrast to the lack
of F-actin involvement in the early recruitment of myosin II to the equatorial
cortex at anaphase, F-actin disruption by Latrunculin A results in a failure to maintain
myosin II in the equatorial region. Interestingly, the downstream rho1 effectors
diaphanous/formin and chickadee/profilin are also necessary for myosin II
maintenance at the equatorial midzone. Although the loss of these genes could
deplete F-actin, phalloidin staining has shown that F-actin is still present in all
of the RNAi-treated cells. In addition, these RNAi-treated
cells still contract, unlike when F-actin is completely disrupted with LatA.
Thus, myosin II appears to be interacting with F-actin in the cortex as it
disperses in dynamic patches throughout the cortex of these diaphanous- or
chickadee-depleted cells (Dean, 2005).
It is suggested that the role of diaphanous/formin and chickadee/profilin in
maintaining the myosin II contractile ring is through the creation of specific
F-actin structures. In particular, formin- and profilin-mediated nucleation
results in unbranched actin filaments because profilin
promotes the barbed-end growth of formin-capped actin filaments. Indeed, electron microscopy has
shown that F-actin in the cleavage furrow mainly consists of unbranched, bundled
filaments. These parallel
filaments contrast with Arp2/3-mediated nucleation, which creates a highly
branched actin filament network. Indeed, Arp2/3, although essential for
lamellipodia formation, is not required for cytokinesis in Drosophila
cells. The hypothesis here
is that once myosin II is recruited to the equatorial cortex of
the cell by a rho kinase-dependent mechanism, possibly localized activation of
RLC phosphorylation, it is retained there because of its higher affinity for
parallel, unbranched actin filaments than to branched actin networks. Consistent
with this hypothesis, myosin II is depleted from the lamellipodia in migrating
cells where Arp2/3 is localized and branched F-actin networks are formed but is enriched in the lamella where F-actin filaments are
more likely to be aligned in parallel bundles. Thus,
it is proposed that high rho1 signaling to Diaphanous at the cleavage furrow
maintains a higher concentration of parallel actin filaments in this region
compared with the rest of the cortex, and these parallel filaments serve to
selectively retain myosin II at the equator to form a stable contractile ring.
In the absence of these parallel actin filaments, myosin II can bind branched
F-actin throughout the cortex, perhaps occasionally organizing them into
parallel bundles that cause increased myosin recruitment corresponding to the
flashes of cortical myosin accumulation, but these interactions are unstable (Dean, 2005).
Live-cell imaging shows that when pavarotti or racGAP50c
are depleted, the cells do not display significant contractions despite
recruiting myosin II to the equatorial cortex. Although there is some modest
membrane contractile activity in these cells, it is clear that significant
contraction or furrowing requires both components of the centralspindlin
complex. It is surprising that only these proteins were found to be necessary
for cortical contraction at sites of myosin II localization. Data from fixed
cells, as well as earlier studies, indicated that Drosophila cells do not
undergo equatorial contractions during mitosis when Diaphanous or Chickadee is
depleted. However, live-cell imaging shows that
when either of these two genes is depleted in S2 cells, not only is myosin II
transiently localized to the equatorial cortex before dispersing, but cells do
indeed display transient equatorial contraction. It is difficult to recognize
these events in fixed cells because of their transient nature and the somewhat
irregular shapes of cells depleted of these proteins. This work highlights the
importance of live-cell imaging in the study of dynamic processes such as
cytokinesis (Dean, 2005).
In addition to the suppression of furrowing, depletion of centralspindlin also
leads to an inability to retain F-actin exclusively at the equatorial cortex
during cytokinesis. This similar phenotype of the centralspindlin complex and
the F-actin affecting proteins suggests that centralspindlin may be an upstream
regulator of F-actin filament formation. Indeed kinase-dead mutants of Pavarotti
have been shown to accumulate at the spindle poles and are associated with an
abnormal accumulation of F-actin near the centrosomes.
Centralspindlin may be acting indirectly by helping to localize an important
actin-affecting protein at the central spindle, or it may act more directly on
the cortex. Because RacGAP50c has been shown to bind Pebble in vitro, it has been
hypothesized that centralspindlin affects the F-actin cortex through rho1
signaling by the localization and/or activation of Pebble. However, RacGAP50c depletion does
not lead to a lack of myosin II recruitment as does Pebble or Rho1 depletion,
and, thus, centralspindlin must act in a rho1-independent manner. For instance, the racGAP activity of
centralspindlin may itself be important for signaling to the F-actin cortex.
Finally, centralspindlin cannot be the major actomyosin ring positioning signal
because myosin II is properly recruited in its absence (Dean, 2005).
Morphogenesis of the Drosophila embryo is associated with a dynamic reorganization of the actin cytoskeleton that is mediated by small GTPases of the Rho family. Often, Rho1 controls different aspects of cytoskeletal function in parallel, requiring a complex level of regulation. The guanine triphosphate (GTP) exchange factor DRhoGEF2 is apically localized in epithelial cells throughout embryogenesis. DRhoGEF2, which has previously been shown to regulate cell shape changes during gastrulation, recruits Rho1 to actin rings and regulates actin distribution and actomyosin contractility during nuclear divisions, pole cell formation, and cellularization of syncytial blastoderm embryos. It is proposed that DRhoGEF2 activity coordinates contractile actomyosin forces throughout morphogenesis in Drosophila by regulating the association of myosin with actin to form contractile cables. These results support the hypothesis that specific aspects of Rho1 function are regulated by specific GTP exchange factors (Padash Barmchi, 2005; full text of article).
Guanine nucleotide exchange factors regulate the activity of the small GTPase Rho1, which is thought to act as a molecular switch in a broad spectrum of morphogenetic processes that require a complex reorganization of the actin cytoskeleton. However, the manner in which different aspects of Rho1 function are regulated by RhoGEFs is not well understood. This study found that DRhoGEF2 protein is broadly distributed in epithelia during oogenesis and embryonic development and concentrated at the apical surface of cells, suggesting that it may regulate Rho1 throughout morphogenesis. The defects of DRhoGEF2 mutants are less severe than those of Rho1 mutants, suggesting that DRhoGEF2 regulates specific aspects of Rho1 function (Padash Barmchi, 2005).
DRhoGEF2 has been shown to regulate cell shape changes during gastrulation, and DRhoGEF2 is implicated in epithelial folding during imaginal disc development, a process that depends on cell shape changes that are similar to those driving invagination of the germ layers. This paper shows that DRhoGEF2 regulates cytoskeletal reorganization and function during pole cell formation and blastoderm cellularization. All of these processes require the contraction of actomyosin rings. It is proposed that DRhoGEF2 regulates Rho1 activity during cell shape changes requiring actomyosin contractility. The results support the hypothesis that individual RhoGEFs may regulate specific aspects of Rho1 function during development (Padash Barmchi, 2005).
Interestingly, DRhoGEF2 has been found to be nonessential during cytokinesis, which also involves the function of contractile actin rings. The function of Rho1 during cytokinesis is regulated by the RhoGEF pebble that initiates actin ring assembly. In pebble mutants, cytokinesis is blocked at mitotic cycle 14 and subsequent mitoses occur without cytokinesis, creating polyploid, multinucleated cells. Although large multinucleated cells are also observed in DRhoGEF2 mutants at the extended germ band stage it is not clear whether these cells are caused by a block in cytokinesis or are caused by earlier defects during cellularization. In contrast to pebble, DRhoGEF2 may not be required for the assembly of actin rings, but may play a nonessential role in the separation of daughter cells. This is reminiscent of observations during cellularization. Although the function of actin rings appears compromised throughout cellularization, the data suggest that some contractile activity remains that leads to the basal closure of blastoderm cells and is responsible for the cellularized appearance of DRhoGEF2 mutants at the onset of gastrulation (Padash Barmchi, 2005).
At the retracted germ band stage, DRhoGEF2 is enriched at the apical cortex of cells in the leading edge of the lateral epidermis, which is consistent with the view that it may regulate Rho1 during dorsal closure. Rho1 function is essential for dorsal closure, and the cuticles of zygotic Rho1 mutants show dorsal holes. In DRhoGEF2 mutants, the lateral epithelial sheets closed the embryo dorsally. This does not exclude the possibility that constriction of actin cables may contribute to dorsal closure and that DRhoGEF2 may play a role in this process. Overall, the data suggest that DRhoGEF2 function may not be essential for the generation of contractile force, but rather regulate the temporal and spatial coordination of actomyosin contractility (Padash Barmchi, 2005).
During syncytial nuclear divisions and cellularization, DRhoGEF2 is localized specifically at the invaginating furrows. In DRhoGEF2 mutants, actin is irregularly distributed and metaphase furrow formation is less uniform than in the wild type. The defects in furrow formation lead to mitotic defects and the subsequent elimination of abnormal nuclei from the cortex so that, at the onset of cellularization, ~20% of the nuclei have been lost. These phenotypes are reminiscent of the defects seen in mutants of the nonreceptor tyrosine kinase Abelson (Abl). The abnormalities in actin distribution observed in abl mutants are likely caused by the mislocalization of Dia, which leads to ectopic actin polymerization at the apical end of cells. Changes in Dia distribution were not observed in DRhoGEF2 mutants, suggesting that DRhoGEF2 may regulate actin distribution by a different mechanism . Perturbations in actin distribution are observed throughout early development in DRhoGEF2 mutants. During cellularization, significant amounts of actin fail to redistribute to the base of the furrow canal. These observations show that one of the roles of DRhoGEF2 is to regulate furrow assembly. The defects in actin distribution also affect the pole cells, which fail to reorganize their cortical actin cytoskeleton and remain embedded in the somatic nuclear layer rather than sitting on top of it. Consequently, they are obliterated during invagination of the cellularization front (Padash Barmchi, 2005).
It is speculated that DRhoGEF2 may have a function in the assembly of actin cables by regulating the association of actin with other proteins such as myosin II. The mislocalization of actin observed in DRhoGEF2 mutants may be caused by failure of actin to associate with myosin. Interestingly, although myosin II is present at the metaphase furrows, it plays no essential role in their formation, and this suggests that the function of DRhoGEF2 in furrow assembly may be independent of actomyosin contractility (Padash Barmchi, 2005).
Phenotypic analysis suggests that DRhoGEF2 regulates actomyosin contractility during cellularization. Previously, the actin-binding protein Bnk has been implicated in the regulation of contractile forces. In bnk mutants, actin hexagons detach from each other and constrict prematurely. Based on this phenotype, it is suggested that, during the slow phase, cortical actin hexagons are linked to each other through Bottleneck (Bnk), and that actomyosin constriction causes the network to contract as a whole, thereby pulling the membrane front inwards. Once the cellularization front has reached the base of the nuclei and Bnk is degraded, actin hexagons detach from each other and contract as individual rings, thereby closing the blastoderm cells basally. It is proposed that DRhoGEF2-mediated activation of Rho1 may regulate the force that keeps actin hexagons under tension. Bnk counteracts contraction during the slow phase by linking individual actin rings to each other. Degradation of Bnk during the fast phase releases individual actin rings, and the DRhoGEF2-mediated contractile force now contributes to basal closure. Therefore, DRhoGEF2 and bnk act in concert to coordinate actin ring contraction during cellularization. In DRhoGEF2-bnk double mutants, the actin network disintegrates progressively, suggesting that DRhoGEF2 and bnk may play an additional role in the assembly or stabilization of actomyosin filaments (Padash Barmchi, 2005).
It has been proposed that actin network contraction contributes to the inward movement of the furrow canal. Although the data suggest that network tension is severely reduced in DRhoGEF2 mutants, the rate of membrane invagination is unaffected. This is consistent with reports on the role of myosin II during cellularization, suggesting that network tension may not contribute to membrane invagination. In the wild type, actin rings squeeze the nuclei slightly and push them basal-wards as the actin network moves over them. This may contribute to the parallel alignment of astral microtubules surrounding the nuclei and to nuclear elongation. In DRhoGEF2 mutants, nuclei are wider than in the wild type and irregularly aligned. It is proposed that network tension may create an ordered hexagonal array of actin rings that contributes to a parallel alignment of nuclei during cellularization. The force moving the actin network inward may be created by plus end-directed tracking of actin on astral microtubules and by membrane insertion as previously suggested. These observations suggest that actomyosin contractility plays a role in the spatial coordination of cytoskeletal function during cellularization (Padash Barmchi, 2005).
Two effector pathways have been implicated in the transduction of Rho1 activation to the actin cytoskeleton. During cytokinesis, which is mechanistically related to cellularization, a linear pathway including profilin and Dia have been proposed to link Rho1 to the contractile actomyosin ring. The maternally supplied Dia plays a role in a spectrum of cytoskeletal functions during early embryogenesis that also require DRhoGEF2 function, such as metaphase furrow formation, pole cell formation, and cellularization. Dia is localized at the cellularization front and is necessary for the recruitment of cytoskeletal components such as the actin-binding protein anillin and the septin homologue Pnut. The phenotypes of dia mutants suggest that dia is necessary for the assembly of contractile actin rings at sites of membrane invagination (Padash Barmchi, 2005).
The similarities between dia and DRhoGEF2 mutants might suggest dia as a downstream effector of DRhoGEF2. However, the defects of dia mutants are morphologically different from those of DRhoGEF2 mutants. In dia mutants, metaphase furrows do not form and contractile rings at the base of polar cytoplasmic buds fail to assemble. During cellularization, actin fails to condense into individual rings, and the network disintegrates during the second phase of cellularization. In DRhoGEF2 mutants actin rings form and remain largely intact but fail to constrict. In addition, the temporal and spatial localization of Dia and Pnut to the cellularization front was unaffected in DRhoGEF2 mutants and dia was not required for the localization of DRhoGEF2. These findings do not exclude that DRhoGEF2 activity may in part be mediated by dia, however, they suggest that some dia-dependent aspects of Rho1 function are still active in DRhoGEF2 mutants and that another pathway may be involved in transduction of the DRhoGEF2 signal (Padash Barmchi, 2005).
A well-characterized pathway regulating actomyosin contractility in mammalian cells and in C. elegans links Rho1 to actin via Rho kinase, the regulatory subunit of myosin light chain phosphatase (MBS) and myosin II. Rho kinase-mediated phosphorylation inhibits the activity of MBS and induces a conformational change in myosin II allowing it to form filaments that promote sliding of antiparallel actin filaments. The data are consistent with a model in which DRhoGEF2 regulates the association of actin with myosin II, thereby stabilizing actomyosin cables. It is proposed that failure to activate the Rho kinase pathway may compromise the recruitment of actin into contractile cables. This may destabilize actin cables and lead to the mislocalization of actin and to the defects in actomyosin contractility observed in DRhoGEF2 mutants. The Drosophila homologue of Rho kinase, Drok, and myosin II have recently been identified as downstream effectors of DRhoGEF2 during the regulation of actomyosin contractility in Schneider (S2) cells. In addition, myosin II is required for basal closure of blastoderm cells and the myosin II heavy chain encoded by zipper (zip) interacts genetically with DRhoGEF2. These data support the model that DRhoGEF2 may regulate actomyosin contractility through the Rho kinase pathway. Mutants in Drok and Drosophila myosin light chain phosphatase have been identified, however, their role during early embryogenesis has not been reported. Interestingly, inhibition of Drok activity by injection of the specific Rho kinase inhibitor Y-27632 into embryos before cellularization disrupts the localization of myosin II. Similar observations have been made in Drok mutant cell clones in imaginal discs. By contrast, DRhoGEF2 mutants reveal no significant changes in the localization of myosin II during cellularization. It is possible that the differences in myosin II localization between DRhoGEF2 and Drok mutants are due to different mechanisms of action at the molecular level. In mammalian cells myosin II phosphorylation is required for the generation of contractile force but not for its localization. Further investigations will be necessary to resolve how the DRhoGEF2 signal is transduced to the cytoskeleton (Padash Barmchi, 2005).
Little is known about the events that regulate the specific subcellular localization and activation of DRhoGEF2. It has recently been shown that DRhoGEF2 particles are transported from the cytoplasm to the cell periphery by tracking microtubule plus ends in Drosophila S2 cells. DRhoGEF2 particles have been observed during syncytial development that may be involved in a similar process in the embryo. It is speculated that DRhoGEF2 may be delivered to specific membrane subdomains at the cellularization front by microtubules. The G-protein α-subunit encoding gene concertina (cta) has been shown to regulate the dissociation of DRhoGEF2 from microtubules. cta has been implicated in the activation of DRhoGEF2 during gastrulation, but is not required during cellularization. It has been suggested that the force moving the actin network inward may be generated by plus end-directed crawling of actin on astral microtubules. It is speculated that DRhoGEF2 may regulate actin ring constriction during cellularization while associated with the tip of astral microtubules by recruiting Rho1 to the site of actin rings (Padash Barmchi, 2005).
DRhoGEF2 is concentrated in actin-rich regions throughout development and the human orthologue of DRhoGEF2, PDZ-RhoGEF, has been shown to bind to actin directly. Although the domain structure of DRhoGEF2 and PDZ-RhoGEF is very similar, the actin-binding region of PDZ-RhoGEF is not conserved in DRhoGEF2. Nevertheless, the localization of DRhoGEF2 is consistent with the view that it may associate with actin, however, further experiments are needed to corroborate this theory (Padash Barmchi, 2005).
During development, small RhoGTPases control the precise cell shape changes and movements that underlie morphogenesis. Their activity must be tightly regulated in time and space, but little is known about how Rho regulators (RhoGEFs and RhoGAPs) perform this function in the embryo. Taking advantage of a new probe, which recognizes the active, GTP-bound form of Rho1 and that can be expressed in vivo using the UAS/GAL4 system, thus facilitating the visualisation of RhoGTPase, evidence is presented that Rho1 is apically activated and essential for epithelial cell invagination, a common morphogenetic movement during embryogenesis. In the posterior spiracles of the fly embryo, this asymmetric activation is achieved by at least two mechanisms: the apical enrichment of Rho1; and the opposing distribution of Rho activators and inhibitors to distinct compartments of the cell membrane. At least two Rho1 activators, RhoGEF2 and RhoGEF64C are localised apically, whereas the Rho inhibitor RhoGAP Cv-c localises at the basolateral membrane. Furthermore, the mRNA of RhoGEF64C is also apically enriched, depending on signals present within its open reading frame, suggesting that apical transport of RhoGEF mRNA followed by local translation is a mechanism to spatially restrict Rho1 activity during epithelial cell invagination (Simoes, 2006).
Using a probe that allows the visualisation of Rho1 activity in the course
of normal development, evidence is presented that this GTPase is active at the
apical side during the process of cell invagination. In the spiracles Rho1
activity is essential to control this movement, similarly to that previously
shown during Drosophila gastrulation, when mesodermal cells fail to
invaginate after inhibition of Rho1 function. An apical enrichment is observed of Myosin II, a possible target of
activated Rho1, analogous to that reported in other tissues of the fly embryo
where this type of movement occurs, such as the mesoderm and the salivary
glands. Inhibition of Rho1 activity results in a disorganised
pattern of apical Myosin II and F-Actin in the spiracle cells. It is suggested that
concentration of active Rho1 at the apical side organises the Actin
cytoskeleton and promotes high Myosin II accumulation/activity in this region,
leading to a contractile Actin-Myosin based force to produce a wedge-shaped
cell (Simoes, 2006).
These data show that spatial restriction of Rho1 activity is achieved by
distinct mechanisms. (1) Albeit ubiquitous, Rho1 protein is strongly
enriched on the apical side of the invaginating spiracle cells. (2) To
ensure that this GTPase is active exclusively on that side of the cell,
opposing Rho regulators are differentially distributed in two distinct
membrane domains: two Rho activators, RhoGEF64C and RhoGEF2, are apically
localised, whereas a Rho inhibitor, the RhoGAP Cv-c, occupies the
complementary, basolateral domain. Cell shape changes and
inward cell movements driving invagination are impaired if Rho1 becomes
activated in a spatially unrestricted manner. These observations stress the
importance of finely tuning Rho1 localisation and activation during normal
tissue morphogenesis (Simoes, 2006).
Several mechanisms might be at work to achieve the specific localisation of
the Rho regulators that direct cell invagination. In the case of RhoGEF64C, its mRNA and protein are apically localised, suggesting that apical
transport of RhoGEF mRNA followed by local translation is a mechanism to
activate Rho1 in a spatially restricted manner. Recent studies show that the
mRNA of RhoA can also be transported and locally translated in the axons and
growth cones of embryonic rat neurons, where RhoA controls growth cone
collapse in response to Semaphorin 3A. This shows that intracellular mRNA transport of Rho GTPases and of their regulators may be an important mechanism to control
spatial GTPase activation (Simoes, 2006).
Loss of function of the RhoGEFs involved in spiracle invagination leads to
variable apical defects, which are compatible with a partial loss of Rho1
function: knocking out RhoGEF64C resulted in a mild disruption of cortical
Actin without blocking invagination, while the absence of RhoGEF2 could result
in a complete failure of the invagination process. These results suggest that
several RhoGEFs are required to properly activate Rho1 during spiracle cell
movement and organ shaping (Simoes, 2006).
One interesting observation from these studies is the fact that mutants for
the RhoGAP Cv-c did not show ectopic activated Rho1 on the basolateral
membrane where this RhoGAP was localised. Thus, several mechanisms must be at
work to ensure that Rho1 activity is excluded from the basolateral domain
during cell invagination: the presence of at least one RhoGAP on the basal
membrane, the apical restriction of RhoGEFs and the existence of low levels of
Rho1 protein on the basolateral side of the cells. In addition, it was also
observed that spiracles from severe cv-c mutants showed lower levels
of apical Rho1-GTP than their wild-type counterparts, correlating with the
disruption of their apical Actin. Defects in apical Actin/Myosin II have also been
reported during invagination of the tracheal pits in cv-c mutants.
Taken together, these observations suggest that GTP hydrolysis is a necessary
step in the regulation of Rho1 function during cell invagination and the
RhoGAP Cv-c may help to maintain a steady state level of apical Rho1-GTP (Simoes, 2006).
Based on the differential distribution of Rho1 GEFs and GAPs, a
model is proposed in which Rho1 must shuttle back and forth between two membrane
compartments, being GTP-bound on the apical cell membrane and GDP-bound on the
basolateral side.
Thus, during tissue morphogenesis, epithelial cells can couple their
apical-basal polarity to the spatial control of small RhoGTPase function (Simoes, 2006).
RhoGTPases act as dynamic switches in many developmental and cellular
contexts. In order to understand how they orchestrate these dynamic processes,
their activity states needs to be visualised over time. It is anticipated that
this work and the tools described will provide a basis for studying Rho1
activity in vivo. It will be interesting to extend this analysis to other
contexts in which Rho GTPases are known to act -- such a dorsal closure,
neurulation, wound healing -- and to identify the Rho regulators involved in
each case, relating their spatial/temporal distribution with the patterns of
Rho GTPase activity (Simoes, 2006).
The construction and maintenance of normal epithelia relies on local signals that guide cells into their proper niches and remove unwanted cells. Failure to execute this process properly may result in aberrant development or diseases, including cancer and associated metastasis. This study shows that local environment influences the behavior of dCsk-deficient cells. Broad loss of dCsk leads to enlarged and mispatterned tissues due to overproliferation, a block in apoptosis, and decreased cadherin-mediated adhesion. Loss of dCsk in discrete patches leads to a different outcome: epithelial exclusion, invasive migration, and apoptotic death. These latter phenotypes required sharp differences in dCsk activity between neighbors; dE-cadherin, P120-catenin, Rho1, JNK, and MMP2 mediate this signal. Together, these data demonstrate how the cellular microenvironment plays a central role in determining the outcome of altered dCsk activity, and reveal a role for P120-catenin in a mechanism that protects epithelial integrity by removing abnormal cells (Vidal, 2006).
The mechanisms that regulate organ size and shape are not well understood, but recent studies have pointed to the importance of local interactions between neighboring cells. For example, in the process known as 'cell competition', cells with relatively higher proliferative rates actively eliminate their neighbors by programmed cell death. Conversely, apoptotic cells send proliferative signals to their neighbors to compensate for their loss. In this way, normal tissue size is achieved. The misregulation of such mechanisms may contribute to the development of cancer, since most solid tumors arise from intact epithelia and are resistant to size-control signals. Tumors are particularly dangerous when linked to metastasis, a process in which cells leave the primary tumor and invade distant tissues. These processes are best understood within the context of an intact epithelium, in which the full range of cell interactions is retained. Work in Drosophila has provided an important in situ view of the action of oncogenes within epithelia (Vidal, 2006).
Src family kinases (SFKs) are active in a broad range of cancer types, including tumors of the breast, colon, and hematopoietic systems. SFK activity typically increases as tumorigenesis progresses and is associated with metastatic behavior. The major inhibitor of SFK activity is C-terminal Src kinase (Csk) and its paralog Chk; these may act as tumor suppressors in, e.g., breast cancer, presumably through their ability to inhibit Src activity and perhaps other pathways. Drosophila Csk acts primarily or exclusively through Src pathway regulation, and the reduction of dCsk activity by itself led to increased organ size, organismal lethality, and increased cell proliferation due to a failure to exit the cell cycle. However, neither Csk loss nor Src activation has been clearly linked to early events in tumorigenesis, bringing into question the role of Csk/Src in proliferation in vivo. Instead, Src is currently thought to be a major player in the metastatic events that occur later in oncogenesis. How Csk or Src promotes the metastatic behavior of cells in situ remains largely unknown (Vidal, 2006).
This study analyzed the phenotypes of dCsk in the context of developing epithelia. The outcome of a cell's loss of dCsk is linked to its cellular microenvironment. When dCsk activity is reduced broadly in the developing eye or wing, the result is overproliferation, inhibition of apoptosis, and decreased cell adhesion. Tissue integrity is retained, but dCsk cells become inappropriately mobile and fail to maintain their appropriate contacts. The outcome of these effects is an overgrown and mispatterned adult tissue. By contrast, loss of dCsk in discrete patches results in epithelial exclusion, invasive migration through the basal extracellular matrix, and eventual apoptotic death; these events occur exclusively at the boundary between dCsk and wild-type cells. Further emphasizing the unique nature of cells at this boundary, a specific requirement was found for a signal that includes Drosophila orthologs of E-cadherin, P120-catenin, RhoA, JNK, and the metalloprotease MMP2. Hence, this study explores the mechanisms by which the cellular microenvironment can direct different behaviors of cells, both in the regulation of apoptosis and epithelial integrity. It also uncovers a mechanism for the removal of abnormal cells from a normal epithelium (Vidal, 2006).
This study shows that reducing dCskactivity results in a blockade of apoptosis and downregulation of cellular adhesion. The work is consistent with the view that Csk is a tumor suppressor that acts at multiple steps. Mutations in the locus encoding the Csk paralog Chk have been described in breast tumors, and, in this study, it has been observed that human Chk can functionally replace dCsk. Therefore the experimental advantages of developing Drosophila imaginal epithelia were used to explore specific aspects of dCsk function that are relevant to the behavior of tumor cells (Vidal, 2006).
Visualization studies suggest that a reduction in dCsk activity leads to a failure of cells to stably retain associations with their neighbors, resulting in prolonged cell movement as cells slide across each other in a manner not observed in wild-type tissue. This may reflect a failure to establish stable junctions, excess cell motility, or both. Recent work has demonstrated a critical and dynamic role for the cadherin-based apical junctions in patterning the Drosophila retina. Misexpressing dE-cadherin prevents patterning defects in GMR>dCsk-IR retinas, suggesting that dCsk cells have reduced dE-cadherin function. Links between Csk, Src, cadherins, and junctional integrity have been reported in mammalian cell culture, and an association has been observed between Drosophila Src42A and dE-cadherin during embryonic development (Takahashi, 2005). The data are consistent with this view: misexpression of a kinase-dead form of Src42A leads to a disruption in the localization of the dE-cadherin-associated protein Armadillo; also, reduced Armadillo levels observed in dCsk retinas is rescued by dE-cadherin misexpression. Together, these data suggest that altering dCsk/Src activity affects cell movements by decreasing dE-cadherin adhesion (Vidal, 2006).
The mechanism by which dCsk alters dE-cadherin function is not clear, but it is relevant to note that Src activation can shift cadherin-based cell adhesion from a 'strong' to a 'weak' adhesive state in mammalian cultured cells. Phosphorylation of cadherins and catenins may mediate 'inside-out' signaling that can alter the adhesive strength of the homophilic bond between cells (reviewed in Gumbiner, 2005). Evidence for such a mechanism has been provided for integrin-mediated focal adhesions (reviewed in Hynes, 2002), and Src activity can alter focal adhesions (Yeatman, 2004). However, normal basal membrane architecture was observed in dCsk cells, as assessed both by anti-integrin staining and by transmission electron microscopy, indicating that at least the gross structure is not affected (Vidal, 2006).
The ability of dCsk to influence cell proliferation, apoptosis, and cell adhesion is consistent with its ability to direct tissue overgrowth: reducing dCsk activity throughout a tissue (or the entire organism) leads to significantly enlarged tissues. This ability demonstrates that dCsk can participate in the mechanisms that set tissue size. A small number of other proteins have been implicated in this process, including Salvador, Hippo, and Lats/Warts, which show phenotypes that are strikingly similar to dCsk. Furthermore, dCsk can directly phosphorylate Lats/Warts (Stewart, 2003) in vitro (Vidal, 2006).
However, reduction of dCsk activity shows some important differences. Mutations in salvador, hippo, or lats/warts lead to an increase in Diap1 levels, which, in turn, blocks apoptotic cell death. By contrast, reductions in dCsk does not significantly alter Diap1 protein levels. Furthermore, although both Hippo and dCsk are required to exit the cell cycle, the cell cycle profile from hippo mutant cells is normal, while dCsk cells contain a significant shift toward G2/M (Read, 2004; Stewart, 2003). Perhaps the most striking difference is the effects of these factors on discrete mutant patches. While broad loss of dCsk activity leads to expanded tissues, surprisingly discrete patches of dCsk tissue are eliminated by neighboring cells. Unlike salvador, hippo, or lats/warts, clonal patches of dCsk cells fail to survive to adulthood. The effects of dCsk reduction are more similar to those reported for the tumor suppressor gene scribble. The scribble locus encodes a component of the septate junction that regulates cell polarity and proliferation; mutant cells display neoplastic overgrowth in a homotypic environment, but are removed by JNK-dependent apoptosis in discrete clonal patches abutting wild-type tissue (Vidal, 2006).
This work provides evidence that neighboring wild-type tissue provides a locally nonautonomous signal that leads to the removal of dCsk mutant cells. For example, FRT-derived clones of dCsk cells were out-competed by neighbors with normal levels of dCsk: this was most easily seen by the clonally related 'twin spot' of wild-type tissue that was consistently larger than the few surviving dCsk clones. In contrast, FRT-mediated dCsk clones that encompassed the entire eye survived and overproliferated. In the developing wing, cells at the periphery of sd>dCsk-IR or ptc>dCsk-IR expression domains were preferentially removed by apoptosis. This death is dependent not on absolute dCsk activity, but on the juxtaposition of cells that are starkly different in their levels of dCsk. Small differences, for example across the ptc>dCsk-IR or omb>dCsk-IR graded expression domains, did not trigger cell death (Vidal, 2006).
This translocation and death of dCsk-IR cells at the patched/wild-type boundary requires at least two steps. At boundaries with wild-type tissue, dCsk cells initially lose their apical profile, shift downward, and eventually become basally excluded from the epithelium. Such excluded cells then migrate away from the boundaries in both directions and eventually die by apoptosis. These events are strikingly reminiscent of those described for tumor cells undergoing metastasis. Altered activity of both Csk and Src has been implicated in a broad variety of tumors. Typically, however, increased Src activity is associated with later events in tumorigenesis, particularly metastasis. Although the connections between high Src activity and metastases are not understood, they likely include Src's ability to break cell-cell junctions and increase cell motility. Another hallmark of metastatic behavior is the ability to degrade basal extracellular matrix: this study also demonstrate a functional requirement for MMP2 activity during the translocation of mutant cells out of the wing epithelium (Vidal, 2006).
While evidence supports the view that the activity of Csk -- and presumably Src and perhaps other effectors -- can regulate metastatic behavior, it alone is not sufficient. First, reducing dCsk activity by itself is not sufficient to allow migrating cells to survive; the data suggest that most or all eventually die. This is consistent with previous work highlighting the importance of a 'two-hit' model to allow for stable tumor overgrowth and metastasis. A second mutation that prevents apoptotic cell death would be minimally required. Second, all cells within a discrete dCsk patch are not equivalent: cells at the boundary of the clone that border cells of strongly differing dCsk levels are exclusively prone to release from the epithelium. This work predicts that cells at the borders of some human tumors are especially prone toward metastatic behavior. Metastasis is often the most serious aspect of a tumor, and approaches that address the metastatic behavior of cells may need to take into account the properties of cells at the periphery. Understanding whether and how these cells are unique may help to more effectively target therapeutic intervention (Vidal, 2006).
In addition to enabling a detailed examination of dCsk cells and their behavior within an epithelium, this model system permitted identification of signaling components that are necessary to execute the aberrant cell mobility and cell death. The results indicate important roles for dE-cadherin, dP120ctn, Rho1, dJnk, and MMP2 (Vidal, 2006).
JNK-dependent apoptosis is required for a broad palette of related mechanisms such as cell competition in developing tissues and the removal of scribble mutant cells. JNK signaling is also associated with the movement of cells within epithelia, including dorsal closure in Drosophila and in mammals. Interestingly, JNK activity is required for the synthesis of MMP2 by v-Src-transformed mammalian cells (Vidal, 2006).
JNK activity can be triggered by several upstream signaling factors, including the small GTPases of the Rho family, and genetic data provide a link between dCsk, dJnk, and Rho1. Rho family proteins are key regulators of cell shape and motility. They also promote the cytoskeletal rearrangements required for epithelial-to-mesenchymal transitions (EMTs), and it is noted that dCsk boundary cells show a number of features that are reminiscent of EMTs. In Drosophila, Rho1 was found to induce an 'invasive' phenotype in wing disc cells, but, in this study, it was demonstrated that, similar to dCsk boundary cells, ptc>Rho1 misexpressing cells also undergo apoptotic death. Most importantly, halving the genetic dose of Rho1 strongly suppresses discrete loss of dCsk, but does not appreciably affect broad loss. Thus, Rho1 activity is linked to dCsk, and activation of Rho1 is sufficient to phenocopy both the apoptotic and migratory phenotypes of dCsk cells located near wild-type tissue (Vidal, 2006).
Previous work in mammalian cell culture has provided direct links between Src and P120-catenin, between cadherins and P120-catenin, and between RhoA and P120-catenin; the latter two interactions have been reported in Drosophila tissue culture systems as well. This study further supports links between these factors in dCsk boundary cells. Interestingly, although normal levels of both dP120ctn and Rho1 were required for the efficient removal of dCsk boundary cells, they were not required for the phenotypes resulting from broad loss of dCsk. The requirement for p120ctn specifically in boundary cells may explain why, although it is the only ortholog present in Drosophila, dP120ctn (Drosophila p120-catenin) is not required for organism viability (Vidal, 2006).
Both Src and P120-catenins are known to directly interact with cadherins, and, in fact, a role was demonstrated for dE-cadherin/Shotgun in the removal of dCsk cells. A model is postulated in which the loss of dCsk results in the remodeling of the zonula adherens, presumably by the phosphorylation of catenins and dE-cadherin itself by Src. Src activation is known to switch cadherin from a strong adhesive state to a weak one, providing one potential explanation for why dCsk retinal cells displayed reduced cell adhesion in situ. One critical question regarding cadherins is whether they have signaling roles that are independent of their adhesive properties. Perhaps relevant to this point, it was surprising to find that reducing dE-cadherin function leads to a suppression of the effects of dCsk-IR at the boundary. A simple dCsk-IR-mediated reduction in dE-cadherin adhesion would be enhanced by further reducing dE-cadherin activity, suggesting that dE-cadherin may provide an active signal that promotes boundary cells' release from the epithelium. If such a signal does exist, neighboring wild-type cells must trigger it, either through their own endogenous dE-cadherin or through a separate, local signal. Why are multiple (3-4) rows affected? The results are consistent with the creation of a successive new boundary as the previous row of cells descends, although other longer-range signals cannot be ruled out (Vidal, 2006).
It is noted that reducing dCsk activity by itself is not sufficient to direct stable tumor overgrowth, supporting the importance of a 'two-hit' model in Drosophila. Loss of the junction protein Scribble showed similar phenotypes to dCsk, including apoptosis, but was found to confer survival and metastatic-like behavior to cells in the presence of an activated Ras isoform. Interestingly, coexpression of dE-cadherin prevents this metastatic behavior (Vidal, 2006).
Finally, how can dP120ctn and Rho1 promote release of dCsk near wild-type boundaries but not act similarly with other dCsk cells? One source of information is the cadherins themselves: the boundary creates an interface of cadherins that have been exposed to different levels of Csk and, presumably, Src activity. This unusual interface may generate the needed dE-cadherin signal. Importantly, recent work has noted a change in the subcellular localization of P120-catenin and E-cadherin specifically at the border of human tumor tissues. At the time that ptc>dCsk-IR boundary cells lose their apical profiles, this study found that dP120ctn is relocalized to the cytoplasm. These results again emphasize the possibility that cells at tumor boundaries pose a special risk of undergoing epithelial-to-mesenchymal-like transitions and metastatic behavior. Metastasis is often the most serious complication of progressing tumors. Targeting therapies to this aspect of cancer may benefit from considering boundary cells and their potentially distinctive properties (Vidal, 2006).
Rho GTPase and its upstream activator, guanine nucleotide exchange factor 2 (RhoGEF2), have emerged as key regulators of actin rearrangements during epithelial folding and invagination (Nikolaidou, 2004). A Rho-GTPase-signaling pathway is used reiteratively in epithelial folding and potentially selects the outcome of Rho activation. This study shows that Drosophila 18 wheeler (18W), a Toll-like receptor protein, is a novel component of the Rho-signaling pathway involved in epithelial morphogenesis. 18w mutant embryos have salivary gland invagination defects similar to embryos that lack components of the Rho pathway, and ubiquitous expression of 18W results in an upregulation of Rho signaling. Transheterozygous genetic interactions and double mutant analysis suggest that 18W affects the Rho-GTPase-signaling pathway not through Fog and RhoGEF2, but rather by inhibiting Rho GTPase activating proteins (RhoGAPs). RhoGAP5A and RhoGAP88C/Crossveinless-c (CV-C) are required for proper salivary gland morphogenesis, implicating them as potential targets of 18W (Kolesnikov, 2007).
Based on microarray experiments, 18w is downregulated in Scr mutants, implicating it in salivary gland development. To follow 18w expression in more detail, RNA in situ hybridizations were performed. The results show that in wild-type embryos 18w RNA is not maternally contributed but is expressed in salivary gland cells prior to and throughout their invagination. Within the salivary placode, 18w expression first becomes evident at stage 11 as a small spot in the dorsal posterior region. During early stage 12, following the beginning of invagination, 18w expression spreads throughout the placode. During the remainder of stage 12, 18w transcripts can be detected in both gland cells that have yet to invaginate as well as those that have already internalized, albeit at a reduced intensity. At stage 13, 18w transcripts cease to be expressed in salivary gland cells but are evident in salivary duct cells. In addition to salivary glands and ducts, 18w RNA is also detected in other tissues undergoing morphogenesis, including the tracheal placodes and the hindgut. As anticipated from microarray experiments, 18w transcripts are absent exclusively from parasegment two in Scr mutants while expression in the rest of the embryo remains unaltered. Overall, performing microarray experiments with Scr mutant embryos has proven to be an effective method for identifying salivary gland genes (Kolesnikov, 2007).
While 18w is expressed in several tissues undergoing morphogenesis, embryonic defects have yet to be identified in 18w mutant embryos. The striking 18w expression within the invaginating salivary gland prompted a more carefully investigation of the role of 18w in embryonic salivary gland development. In examining a null allele of 18w, 18wΔ21, it was discovered that initiation of invagination in the dorsal posterior region of the placode appears normal. However, during the next phase of invagination, in which the remaining cells of the salivary placode normally internalize in a strict sequential order, defects become apparent. In 18w mutants invagination is less synchronized; too many placode cells invaginate simultaneously rather than sequentially, thereby causing a wider lumen in 66% of 18w mutants when compared to wild-type embryos. Thus, it is the timely progression of invagination, not the invagination process itself, that is affected in 18w mutants. Due to the lack of proper coordination in 18w, the internalization of the gland cells appears to be delayed causing the cells that internalize last to remain near the ventral surface instead of reaching their final, more dorsal, position within the embryo. Eventually all of the salivary gland cells do internalize in 18w mutants, but the proximal part of the gland, which remains abnormally close to the ventral surface, becomes caught in the anterior movement of the ectoderm during head involution. As a result, the glands end up much closer to the anterior end of the embryo than in wild-type embryos. Similar defects are seen in mutants homozygous for the loss of function 18w allele, 18wΔ7–35, and in 18wΔ21, 18wΔ7–35 transheterozygous mutant embryos. Based on these results, 18w is an important component in coordinating salivary gland invagination (Kolesnikov, 2007).
Activation of the Rho-signaling pathway results in the phosphorylation of the myosin II regulatory light chain, encoded by the spaghetti squash (sqh) gene. The phosphorylated form of Sqh can interact with actin and cause actomyosin-based contractility at the apices of cells. To verify that 18W is part of the Rho pathway, whether overexpression of 18W results in an upregulation of Rho signaling, evident as an increase in phosphorylation of Sqh, was checked. Immunoblot analysis on embryonic extracts using anti-phospho-Sqh antibody reveals that overexpressing 18W throughout the embryo results in a two-fold increase of P-Sqh when compared to wild type. Similar results are seen when a constitutively active form of Rho is overexpressed ubiquitously in the embryo. Moreover, introducing one copy of a phosphomimetic sqh transgene, sqhE20E21, into an 18w mutant rescues the 18w invagination defects, indicating that 18W acts upstream of Sqh phosphorylation. Thus, both genetic and biochemical evidence indicate that 18W is a novel component of the Rho signaling pathway (Kolesnikov, 2007).
To determine whether 18W negatively regulates the RhoGAP branch of the Rho pathway, it was necessary to identify the RhoGAPs involved in salivary gland development. Of the 20 distinct RhoGAPs encoded by the Drosophila genome, 17 UAS-RhoGAP dsRNA lines were examined. Each of these lines was crossed to flies containing a salivary gland-expressing driver, scabrous-GAL4, and the progeny were screened for salivary gland defects. Only RhoGAP5A dsRNA expression resulted in salivary gland defects. Invagination defects are evident from more anteriorly placed glands, while migration defects resulted in wavy glands. Similar defects are seen upon expressing high levels of constitutively active Rho in the salivary gland with the scabrous-GAL4 driver. Overexpressing low levels of Rho in the gland results in mostly migratory defects, indicating that gland migration is more sensitive to the levels of Rho signaling than is the process of invagination. Invagination defects are seldom accompanied by migration defects, presumably because glands that have not properly internalized do not reach the visceral mesoderm upon which they normally migrate (Kolesnikov, 2007).
Since the GAL4/UAS dsRNA system may not result in a complete loss-of-function phenotype, strong alleles of two GAPs, RhoGAP68F and RhoGAP88C/Cv-c, which will be referred to simply as Cv-c. These have previously been shown to regulate Rho activity but lack salivary defects using the dsRNA interference technique. While neither rhoGAP68F nor cv-c mutations cause invagination defects, the cv-c mutants do display migration defects similar to those seen in embryos expressing RhoGAP5A dsRNA. To determine whether RhoGAP5A and Cv-c act redundantly within the salivary gland to regulate Rho activity, RhoGAP5A dsRNA within the gland was examined in a cv-c mutant. These embryos should have reduced activity of both RhoGAPs. They have gland invagination and migration defects that are more severe and penetrant than embryos that lack just one of them, indicating that these RhoGAPs are, in fact, partially redundant during salivary gland development (Kolesnikov, 2007).
In situ hybridization in wild-type embryos shows that cv-c RNA is not maternally contributed but is expressed in several tissues undergoing morphogenesis. In addition to its expression within the developing trachea and mesoderm, cv-c, similar to 18w, is expressed in the salivary glands prior to and during their invagination. Unlike 18w, however, cv-c expression does not appear to originate at the initial invagination site but rather initiates expression throughout most of the placode. During the onset of invagination at stage 12, cv-c expression intensifies and continues to be expressed within cells that have internalized until the conclusion of invagination at stage 13 (Kolesnikov, 2007).
To examine whether 18W regulates Rho activity by inhibiting Cv-c, several overexpression and genetic interaction experiments were performed. As might be expected if 18W negatively regulates Cv-c, overexpressing 18w within the salivary gland results in migratory defects similar to those seen in cv-c mutant embryos. In addition to migratory defects, however, some embryos overexpressing 18w also exhibit invagination defects, suggesting that Cv-c may not be the only RhoGAP negatively regulated by 18W. Moreover, lowering the dose of cv-c enhances the defects caused by 18w overexpression and suppresses the 18w mutant invagination defects, further supporting the role of 18W as a negative regulator of RhoGAP signaling. Therefore, genetic interaction experiments indicate that 18W regulates Rho signaling in the salivary gland by inhibiting at least one known RhoGAP (Kolesnikov, 2007).
Previous studies have shown that the Fog ligand activates RhoGEF2 through an as yet unidentified receptor, leading to the apical constriction of cells that form the ventral furrow and posterior midgut. Similar to salivary gland cells in 18w mutants, cells of the ventral furrow and posterior midgut in fog mutants do eventually invaginate but in an uncoordinated and delayed fashion. Since 18w and fog mutants have similar invagination defects, and 18W is a receptor protein that activates Rho signaling, whether 18W might be the FOG receptor was examined. This seems unlikely, however, because FOG overexpression within the salivary gland rescues the 18w mutant salivary gland defects. Since fog mutations do not completely eliminate apical constriction during ventral furrow and posterior midgut formation but RhoGEF2 mutations do, it has been argued that additional pathways must regulate apical constrictions via RhoGEF2. However, since 18w RhoGEF2 double mutants have more severe defects than either of the single mutants, 18W is not one of the additional upstream activators of RhoGEF2. Although neither present downstream of FOG nor upstream of RhoGEF2, 18W does appear to be positioned upstream of Sqh phosphorylation since the 18w salivary gland mutant phenotype can be rescued by introducing one copy of a phosphomimetic allele of sqh (Kolesnikov, 2007).
One possible way that 18W might activate Rho signaling is by negatively regulating RhoGAPs. Two RhoGAPs, RhoGAP5A and Cv-c, were identified that function partially redundantly during salivary gland morphogenesis. Embryos defective for both RhoGAPs exhibit invagination and migration defects similar to those observed when 18W is overexpressed within the salivary glands. Comparable defects are also seen upon expression of activated Rho, supporting the role of both RhoGAPs and 18W in Rho signaling (Kolesnikov, 2007).
Although overexpression and genetic interaction data demonstrate that 18W does indeed work in opposition to Cv-c activity, whether 18W actually negatively regulates RhoGAPs or if it controls Rho signaling through an alternate and unknown pathway has yet to to be deciphered. Another RhoGAP, RhoGAPp190, is regulated by the Src family of tyrosine kinases in both mammals and Drosophila. Depending on the site of phosphorylation, mammalian RhoGAPp190 can be either activated or inhibited by Src, while the Drosophila RhoGAPp190 appears to be only negatively regulated by the Drosophila Src homolog, Src64B. Genetic interactions and double mutant analysis with 18w and either Src64B or the other Drosophila Src gene, Src42A, however, suggest 18W does not regulate RhoGAPs via Src kinases in Drosophila (Kolesnikov, 2007).
Considering that 18W is a member of the Toll family of receptors, it might signal through the pathway used by Toll itself. Upon activation by its ligand, Spätzle, Toll signals via the cytoplasmic proteins MyD88, Tube, and Pelle to promote the degradation of the Cactus protein. This degradation releases the sequestered transcription factor Dorsal, allowing it to enter the nucleus and activate transcription. Although both 18w and Toll are expressed in the salivary gland, no evidence was found to suggest that 18W signals through the Toll-pathway or that it functions redundantly with Toll. Zygotic tube, pelle, MyD88, or Toll mutant embryos do not have salivary gland defects and MyD88 does not physically interact with any of the Toll-like receptors except for Toll itself. Similarly, there are no obvious genetic interactions between mutant alleles of 18w and Toll based both on lethality and salivary gland abnormalities (Kolesnikov, 2007).
Similar to the Toll family of receptors, many RhoGAPs and RhoGEFs are found in both mammals and flies. The Drosophila genome encodes 21 RhoGAPs and 20 RhoGEFs but only seven Rho-family GTPases. Since a specific RhoGTPase can be regulated by multiple RhoGAPs, there may be some redundancy in the function of the RhoGAPs. This appears to be the case during salivary gland development. Of the 17 RhoGAPs analyzed by RNAi, by available alleles, or by both, two resulted in distinct defects in the salivary glands. Mutant embryos that lack both of these RhoGAPs have more severe and penetrant gland defects than embryos that only lack one, indicating that the two have redundant roles during gland development (Kolesnikov, 2007).
Since 18W is expressed in several tissues undergoing morphogenesis, it will be interesting to establish whether it is important for the development of additional tissues other than the salivary gland. It will also be interesting to determine whether 18W functions in opposition to the particular RhoGAPs that are active within these other tissues. Overall, since very little is known about pathways controlling RhoGAP activity during apical constriction, identifying additional genes that interact with 18W may prove to be important not only in elucidating RhoGAP regulation but also in understanding the process of epithelial invagination (Kolesnikov, 2007).
A Drosophila gene (properly termed Rho-type guanine exchange factor - FlyBase ID: FBgn0015803)
has been cloned that has substantial sequence homology to a distinct class of
proto-oncogenes: these include DBL, VAV, Tiam-1, ost and ect-2. It has predicted Rho or Rac guanine exchange factor (Rho/RacGEF) and pleckstin homology (PH) domains with the PH immediately downstream of the Rho/RacGEF. Rho/RacGEFs are known to catalyze the dissociation of GDP from the Rho/Rac subfamily of Ras-like GTPases, thus activating the target Rho/Rac. Members of the Rho/Rac subfamily regulate organization of the actin cytoskeleton, which controls the morphology, adhesion and motility of cells. Message from this gene is found throughout oogenesis and embryogenesis. Of particular interest, message is most abundant in furrows and folds of the embryo where cell shapes are changing and the cytoskeleton is likely to be undergoing reorganization (Werner, 1997).
Mammalian Rho-kinase/ROK alpha, one of the targets of Rho, has been shown to bind to Rho in GTP-bound form and to phosphorylate the myosin light chain (MLC) and the myosin-binding subunit (MBS) of myosin phosphatase, resulting in the activation of myosin. Thus, Rho-kinase/ROK alpha has been suggested to play essential roles in the formation of stress fibers and focal adhesions. A two-hybrid analysis demonstrates that Drosophila Rho-associated kinase interacts with the GTP-bound form of the Drosophila Rho1 at the conserved Rho-binding site. Rok can phosphorylate MLC and MBS, preferable substrates for bovine Rho-kinase, in vitro. These results suggest that Rok is an effector of Rho1 (Mizuno, 1999).
Exchange factors for Rho proteins are known to activate their downstream targets through direct binding to Rho GTPases. Hence, the molecular basis for the observed genetic interaction between pebble and Rho1 may be a physical interaction between Pebble and Rho1 proteins. To test this hypothesis, a yeast two-hybrid assay was carried out. Fusions of Pbl (full-length and amino-terminally truncated) as well as Drosophila Rho proteins, Rho1, Rac1, and Cdc42, were constructed with both the GAL4 DNA-binding domain (DBD) and GAL4 activation domain (AD). Plasmids were transformed in various combinations, and the resulting colonies were tested for the ability to activate HIS3 and lacZ reporters. Only colonies that carry plasmids expressing both Pbl (full-length or DeltaPbl325-853, amino-terminally truncated Pbl) and Rho1 are able to grow on a medium that lacks His, Leu, and Trp, suggesting that the HIS3 gene is induced. Therefore, it is suggested that Pbl and Rho1 proteins interact in vivo and that this interaction is specific for Rho1, but not for Rac1 or Cdc42. Furthermore, the DH domain, but not the amino terminus of Pbl (containing BRCT domains and NLS), is essential for this interaction, because it is abolished by a small deletion within the DH domain, but not by the amino-terminal truncation of Pbl. These results indicate that Pbl and Rho1 form a protein complex in vivo, and that the basis for the genetic interaction between pbl and Rho1 may be a direct interaction between the two proteins (Prokopenko, 1999).
To test the physical interaction between Drosophila Rok and various Rho-GTPases, a pull-down assay using GST-GTPase fusion proteins and in vitro translated Rok was used. Rok binds to the constitutively active form of Drosophila RhoA, but not to constitutively active Rac1 or Cdc42. Mutating a key amino acid within the effector-binding domain (T37A) abolishes the interaction with Rok. These results suggest that Rok is an effector specific for Drosophila RhoA (Winter, 2001).
The Rho GTPases regulate many different cellular and developmental processes, and activation of Rho GTPase signaling is mediated through interaction with the Dbl homology (DH) protein domain. The expression pattern is described of DrhoGEF3 (cytological position 61B1-B3), which encodes a new member of the DH domain protein family from Drosophila and is a homolog of the human protein hPEM-2. During gastrulation and germ band extension, DrhoGEF3 exhibits a segmented expression pattern. DrhoGEF3 is subsequently expressed in the visceral mesoderm, at the sites of muscle attachment and in specific groups of sub-epidermal cells. The possible function of such a dynamically expressed signaling molecule is discussed (Hicks, 2001).
Initially, during the blastoderm stage, DrhoGEF3 expression occurs in a pattern of circumferential stripes. Upon gastrulation a close association between these stripes and several early morphogenetic events become apparent. At the anterior trunk region DrhoGEF3 is initially expressed in the cells flanking the cephalic furrow, which later move into the furrow during its invagination. DrhoGEF3 expression is also detectable in cells anterior to the invaginating posterior midgut that undergo extensive shape changes. As the germ band extends, DrhoGEF3 becomes expressed in a segmented pattern that is similar to genes known to be required for segmentation, such as engrailed. Indeed, the stripes of DrhoGEF3 expression, which are 3-4 cell diameters wide, partially overlap the posterior edge of the stripe of engrailed expression in each segment. Upon the completion of germ band extension, DrhoGEF3 is specifically expressed in scattered groups of sub-epidermal cells within each segment. At this stage DrhoGEF3 is also detectable in the visceral mesoderm, and expression within this tissue persists as the germ band retracts. Finally, following the completion of germ band retraction and dorsal closure, DrhoGEF3 expression is detected at the sites of muscle attachment (Hicks, 2001).
Mutations in dpix (rho-type guanine exchange factor) were recovered from a large-scale screen in Drosophila for genes that control synaptic structure. dpix encodes dPix, a Rho-type guanine nucleotide exchange factor (RtGEF) homologous to mammalian Pix. dPix plays a major role in regulating postsynaptic structure and protein localization at the Drosophila glutamatergic neuromuscular junction. dpix mutations led to decreased synaptic levels of the PDZ protein Dlg, the cell adhesion molecule Fas II, and the glutamate receptor subunit GluRIIA, and to a complete reduction of the serine/threonine kinase Pak and the subsynaptic reticulum. The electrophysiology of these mutant synapses is nearly normal. Many, but not all, dpix defects are mediated through dPak, a member of the family of Cdc42/Rac1-activated kinases. Thus, a Rho-type GEF and Rho-type effector kinase regulate postsynaptic structure (Parnas, 2001).
spire is a maternal effect locus that affects both the dorsal-ventral
and anterior-posterior axes of the Drosophila egg
and embryo. It is required for localization of determinants
within the developing oocyte to the posterior pole and to
the dorsal anterior corner. During mid-oogenesis, spire
mutants display premature microtubule-dependent
cytoplasmic streaming, a phenotype that can be mimicked
by pharmacological disruption of the actin cytoskeleton
with cytochalasin D. spire has been cloned by transposon
tagging and is related to posterior end mark-5, a gene from
sea squirts that encodes a posteriorly localized mRNA.
Spire mRNA is not, however, localized to the posterior
pole. Spire also contains two domains with similarity to
the actin monomer-binding WH2 domain, and Spire binds to actin in the interaction
trap system and in vitro. In addition, Spire interacts with
the rho family GTPases RhoA, Rac1 and Cdc42 in the
interaction trap system. This evidence supports the
model that Spire links rho family signaling to the actin
cytoskeleton (Wellington, 1999).
Previous work has shown that in spir mutants, the
microtubules bundle at the cortex prematurely, during stage
8, and this bundling of the microtubules is accompanied by
rapid, microtubule-dependent swirling of the cytoplasm. Both the bundling of
microtubules and cytoplasmic streaming are normally seen
later in stage 10 wild-type oocyte. A bi-directional signaling process occurs between the oocyte
and the posterior follicle cells to establish the posterior pole of
the egg.
Phenotypes indicative of a defect in this signaling process
include transformation of the posterior follicle cells into an
anterior follicle cell fate; misorganization of the microtubules
at stage 6; localization of Oskar mRNA to the center of the
oocyte; localization of Bicoid mRNA to the posterior pole, and
premature cytoplasmic streaming. The posterior follicle cell fates are established correctly in
spir. In contrast to a previous report, no central spot of Oskar mRNA
staining is observed in spir mutants. Finally, Bicoid mRNA localization
appears relatively normal in spir. These results
suggest that signaling between the posterior follicle cells and
the oocyte is not abnormal in spir mutants. In spir mutant oocytes, microtubules sometimes show abnormal
distributions during stage 6, but it is believed that
this probably reflects an earlier manifestation of the known spir
microtubule defect (Wellington, 1999 and references therein).
WH2 domains, like those found in Spir, have been found
in the Wiskott-Aldrich syndrome protein (WASP), verprolin,
Scar-1 (see Drosophila SCAR), and a number of other proteins of unknown function. The WH2 domains of N-WASP and of Scar1
have been shown to bind directly to G-actin in vitro. In
addition, Spir binds to unpolymerized
actin in vitro.
Although Spir is capable of binding to actin monomers
through its WH2 domain, no defects in the
actin cytoskeleton have been observed in spir mutants, suggesting a number of
possibilities. The defects may be in actin structures that are
difficult to observe, such as the cell cortex. In fact, spir phenotypes can be mimicked by treatment with
cytochalasin D, a drug that affects the polymerization state of
actin; defects in the actin
cytoskeleton in cytochalasin D-treated oocytes have not been observed. Alternatively,
spir may act downstream of the actin cytoskeleton and, thus,
not change it. Finally, in vitro experiments have shown that, in
the absence of the neighboring cofilin homology and acidic
domains, the two WH2 domains of N-WASP have either no
effect on or slowly depolymerize filamentous actin. Since Spir is lacking the cofilin homology
and acidic domains, Spir may have only minor or no
effect on filamentous actin (Wellington, 1999 and references therein).
It is becoming more apparent that a relationship exists
between the actin cytoskeleton and premature microtubule-dependent
cytoplasmic streaming. The premature cytoplasmic streaming phenotype of spir can be mimicked by the addition of cytochalasin D, a drug which depolymerizes actin filaments. Additional evidence that the actin cytoskeleton is involved in repressing
microtubule bundling and streaming comes from analysis of
mutant phenotypes of genes linked to the actin cytoskeleton.
In addition to spir, mutants in chickadee, which encodes profilin and capu, which is thought to bind to profilin, also exhibit these microtubule
behaviors. While phenotypes for cdc42 and rac1 have been
described during oogenesis, their effects on patterning during
oogenesis are unknown. The finding that Spir interacts with rho family GTPases suggests
that at least one of the rho family GTPases is functioning in
patterning. Further genetic and biochemical studies will be
required to determine the nature of Spir's interaction with rho
family GTPases in vivo. The analysis of spir suggests that rho family GTPAses and
actin function with Spir in patterning the Drosophila oocyte.
Further studies on spir should elucidate the role of rho family
GTPases and the actin cytoskeleton in patterning during oogenesis (Wellington, 1999).
Among the putative Rho/Rac effector targets in mammals
are the protein kinase N/protein kinase C-related kinase (PKN/PRK) family of
serine/threonine kinases. PKN (also referred to as PRK1) and the closely related protein PRK2 together account for the vast majority of Rho-binding autokinase activity detected in most mammalian tissues. The carboxy-terminal catalytic domains of these kinases are highly homologous to the PKC family kinases, but they
possess unique amino-terminal regulatory sequences, including three
leucine zipper-like repeats shown to be important for the interaction with the Rho GTPase. These proteins also interact detectably with the Rac GTPase, suggesting that they may be shared effector targets of the Rho and Rac GTPases. Despite the identification of closely related PKN homologs in several organisms, the precise biological function of these putative Rho targets remains unknown. The ability of the Rho GTPases to regulate cell morphology and motility
suggests that these proteins and their associated signaling pathway
components are likely to perform functions essential to the normal
morphogenesis of developing multicellular organisms. Dorsal closure (DC) does
not require new cell divisions, and appears to depend solely on
dramatic cell shape changes within a subset of epidermal cells. These
shape changes are initially restricted to two symmetric rows of
epidermal cells, known as the leading edge (LE) cells, and are followed
by the stretching of the more lateral epidermal cells, ultimately
resulting in the meeting of the two rows of LE cells at the dorsal midline.
Three classes of genes have been implicated in DC; namely, the Rho
family of GTPases; the c-Jun amino (N)-terminal kinase (JNK) cascade
components, including the decapentaplegic (dpp) signaling
pathway genes, and several membrane-associated proteins. The recently described Drosophila loss-of-function Rho1 mutant is defective for DC, and homozygous mutant embryos exhibit an obvious hole in the
dorsal-anterior portion of the larval epidermis. Although Rac1 loss-of-function mutants have yet to be reported, overexpression of Rac1N17, a dominant-negative form of Rac1, in developing Drosophila embryos, results in a DC defect. Similar results have also been reported for Cdc42N17 (Lu, 1999 and references).
Four Drosophila homologs of the mammalian JNK pathway genes [hemipterous (hep; JNKK), basket (JNK), Djun, and kayak (Dfos)] are all required for DC. Mutations in any of those genes result in a dorsal-open phenotype very similar to that seen in either the Rho1 mutants or embryos overexpressing dominant-negative Rac1 or Cdc42. Furthermore, disruption of the JNK pathway in hep mutants abolishes the expression of two independent downstream target genes of Djun, dpp and puckered (a MAP kinase phosphatase), which are also required for DC. The hep mutation also blocks an increase in dpp expression in the LE cells induced by expression of activated Rac1V12 in those cells. These results have led to a model for
DC in which Rac1 (and possibly Cdc42) signals through the JNK pathway
to activate the expression in LE cells of Dpp, a secreted ligand of the
TGF-beta receptor. In turn, Dpp relays an instructive signal to
initiate stretching of the more lateral epidermal cells. The signaling role of Rho1 or downstream effector targets of Rho1 in this process is unknown. A Drosophila homolog of the mammalian PKN family kinases, Drosophila Pkn is shown to bind specifically to both Rho1 and Rac1 GTPases in a GTP-dependent manner, and its kinase activity is promoted
by both interactions, suggesting that Rho1 and Rac1 GTPases can
utilize Pkn as a downstream effector target. Both Rho1 and
Rac1 bind Pkn through the amino-terminal region of the protein. Significantly, it appears that Rho1 and Rac1 interact with Pkn
through distinct binding sites. A loss-of-function mutation in the Drosophila Pkn gene leads specifically to a DC defect during embryogenesis. However, this Pkn-mediated DC pathway is
independent of the Rac-JNK-Dpp pathway; rather, the Pkn-mediated DC pathway appears to act
coordinately with the Rac-JNK-Dpp pathway to regulate epidermal cell shape changes
during morphogenesis of the Drosophila embryo (Lu, 1999 and references).
The predicted Drosophila Pkn protein sequence is closely
related to both human PKN and PRK2 (60% overall identity to both) and has all of the conserved features found in other PKN family members, including the amino-terminal negative
regulatory pseudosubstrate motif, the three
leucine zipper repeats (HR1a, HR1b, and HR1c) that mediate GTPase
binding, a central conserved
region of unknown function found in the PKC-eta kinases, and the carboxyl-terminal PKC-like kinase domain (Lu, 1999).
The expression pattern of the Drosophila Pkn gene is highly
dynamic during embryogenesis. An in situ hybridization analysis of wild-type embryos reveals that Pkn mRNA is
abundant at the blastodermal stage, suggesting that it is maternally
loaded. At stage 13, when DC is normally initiated, the most prominent expression is seen in the dorsal LE cells and in two pairs of discontinuous stripes on the epidermis of each segment. However, the
expression becomes more restricted in later stages and can only be
detected in the anterior and posterior spiracles, the pharynx, and the
mouth tip at stage 16 (Lu, 1999).
Although homozygous Pkn mutant embryos do not exhibit
obvious developmental defects prior to stage 13, ~10% of them die
as embryos with an obvious hole in the anterior region of the dorsal
epidermis. This dorsal-open phenotype closely resembles the
DC defects observed previously with loss of function of the
Rho1 gene and several components of the Rac-mediated
JNK cascade. To examine the requirement for maternally loaded Pkn mRNA,
germ-line clone (GLC) mutants of Pkn were generated:
~50% of the mutant embryos derived from these clones are found to
display the same DC defect. Phenotypic analysis of these
embryos with various histological markers reveals that they are
correctly patterned and do not exhibit any detectable defects of the
central or peripheral nervous system or the somatic musculature, suggesting that Pkn is not required prior to DC (Lu, 1999).
According to the current working model of DC, the Rac1 and Cdc42
GTPases activate the downstream JNK cascade kinases to induce the
expression of several genes in the LE cells, including dpp. Thus, DPP mRNA expression in the LE cells is a convenient assay of activation of the JNK cascade in
vivo. Because Pkn functions biochemically as a Rac1 effector target,
and mutations in Pkn cause a DC defect similar to that seen
with JNK pathway mutants, it was of interest to determine whether Pkn is
required for the previously established Rac-JNK-Dpp pathway. Therefore, the expression of DPP mRNA was examined in
Pkn mutant embryos. Mutations in the JNK cascade gene hep
result in embryos that fail to express detectable levels of
DPP mRNA in the LE cells. In contrast to the loss of dpp
expression seen in hep mutant embryos, expression of
dpp in the LE cells of Pkn mutants is not detectably
affected, indicating that Pkn is not
required for the previously reported Rac-JNK-Dpp pathway. This
conclusion was also verified by examination of the expression of beta-galactosidase from a puckered-lacZ enhancer trap, which
has been shown previously to be eliminated in a hep mutant
background. Pkn mutant embryos
that exhibit an obvious dorsal-open phenotype retain normal levels of
puckered-lacZ expression, confirming that the
Rac-JNK pathway leading to transcriptional activation is not
detectably affected by the absence of Pkn (Lu, 1999).
The shapes of epidermal cells of Pkn mutant
embryos were examined. All epidermal cells adopt an unstretched polygonal shape following an initial apparently normal LE
cell stretching, similar to that seen in the JNK pathway
mutants, such as hep. Thus, it appears that while
both Rac-JNK- and Pkn-mediated signals are required for the stretching
of epidermal cells required for DC, they are associated with distinct
pathways. Because the expression of Pkn mRNA is enriched specifically
in the LE cells prior to DC, the possibility that Pkn
expression is regulated by the JNK pathway was examined. However, Pkn
expression is unaffected in hep mutant embryos (Lu, 1999).
Accumulating evidence suggests that the Rac1 GTPase, as well as some
of the JNK pathway components, performs a function in the LE cells that
may be distinct from the regulation of gene expression, but is
necessary for the stretching of the LE cells. Because Pkn is a biochemical effector of the
Rac1 GTPase that mediates DC, but is not required for dpp or
puckered gene expression, the possibility was tested that the
Pkn mutant interacts genetically with the JNK pathway
component, basket. Removal of one copy of the
basket gene from a Pkn mutant GLC background
significantly increases the frequency of dorsal-open embryos, suggesting that both Pkn and JNK activities converge at some point to affect a related aspect of the DC process (Lu, 1999).
The Rho1 GTPase has also been implicated in DC.
Null alleles of Rho1 exhibit a DC defect that closely resembles that seen in Pkn mutants although the relevant Rho1-mediated pathway in this process has not been established. Because
the cell shape changes in the LE cells appear to initiate the DC
process, and Pkn expression is enriched in the LE cells just prior to
DC, the possibility was explored that a Rho1-Pkn signal
mediates shape changes in those cells. To start, the
requirement for Rho1 activity was examined specifically in the LE cells. A dominant-negative form of Rho1
(Rho1N19) is expressed in the LE cells of wild-type embryos. More than 60% of these
embryos display a dorsal-open phenotype very similar to that seen in
the Rho1, Pkn, and JNK pathway mutants. Similar to Rho1 and
Pkn mutants, but not to a hep mutant, embryos
expressing Rho1N19 in the LE cells exhibit normal levels of
DPP mRNA expression in the LE cells. Moreover, all
of the epidermal cells in these embryos ultimately adopt an unstretched
polygonal shape following a normal initial LE cell stretching at early
stage 13. This result demonstrates clearly that expression of the dominant-negative Rho1 in the LE cells does not cause a dorsal-open phenotype by nonspecifically blocking the Rac-JNK-Dpp pathway and
suggests that a Rho1-mediated second instructive signal is generated in
the LE cells, which together with Dpp, is required for the stretching
of the more lateral-ventral cells (Lu, 1999).
These results are
consistent with a role for a Rho1-Pkn signal that contributes to the
stretching of the LE cells. To establish a specific requirement for Pkn
in the LE cells, a transgenic line was established in which a putative
dominant-negative form of Pkn (kinase deficient) is expressed under the
control of a UAS element. By crossing this line to a line harboring the LE-GAL4 driver, it was determined that a small percentage of embryos exhibit a DC defect that is indistinguishable from that seen
in Pkn or Rho1 loss-of-function mutants. Genetic interaction between Rho1 and Pkn was examined. Removal of one copy of
the Rho1 gene (a null allele) from the Pkn mutant GLC
background results in an increase in the frequency of dorsal-open
embryos from 55% to 68%. The fact that a reduction of Rho1
activity enhances the frequency of DC defects observed in a Pkn-null
background suggests that Rho1 most likely interacts with at least one
additional downstream target that is also required for DC. Consistent
with this hypothesis, it was observed that heat shock-induced
Pkn expression is not sufficient to rescue the Rho1
mutant DC defect (Lu, 1999).
Currently, little is known regarding the biological role of the
mammalian PKN family proteins. Expression of a dominant-negative (kinase-deficient) form of human PRK2 in microinjected fibroblasts results in the disruption of actin stress fibers, suggesting a normal
role for PRK2 in regulating Rho-mediated actin reorganization. Both human PKN and PRK2 undergo partial
caspase-mediated proteolysis during apoptosis, suggesting a potential role for these kinases
in the morphological changes in cells during programmed cell death.
Significantly, both Rho and Rac GTPases have been implicated previously
in apoptosis. A closely
related homolog of PRK2 in starfish is highly expressed and
specifically phosphorylated in oocytes during meiotic maturation,
suggesting a possible role in that process.
Human PKN, which is highly expressed in brain, has been found to be
enriched in the neurofibrillary tangles associated with Alzheimer's
disease and can phosphorylate the Tau protein as
well as the Neurofilament L protein, suggesting a
potential role for PKN in neuronal degeneration. Despite these
observations, the precise biological function of these prominent
Rho/Rac effector targets remains unclear. The results described here provide evidence for a biological function
for this putative Rho/Rac-effector target in embryonic development and indicate a specific role for Drosophila Pkn in regulating cell shape changes required for tissue morphogenesis. Previous studies have revealed a role for the Drosophila Rho1
GTPase in several developmental processes, including gastrulation,
establishment of tissue polarity, and DC. However, Pkn is required only for DC, indicating
that Pkn is not required for all Rho1-mediated activities related to
morphogenesis.
Thus, it seems likely that additional Rho-associated downstream
effector targets mediate the function of Rho in these other
developmental processes. Possibly, the ability of Pkn to serve as a
target of both activated Rho and Rac GTPases accounts for a specialized role for this protein in DC, which appears to require both Rho- and
Rac-mediated signaling pathways. Significantly, gastrulation, a
morphogenetic process that is also largely dependent on cell shape
changes, requires Rho1 but not Rac1 GTPase activity and is unaffected by the absence of Pkn (Lu, 1999 and references).
The fact that homozygous zygotic Pkn mutants that undergo
normal DC ultimately die during larval development suggests that Pkn
may mediate additional functions of Rho/Rac signaling
that are utilized in post-embryonic development. However, no
obvious developmental defects are found in somatic homozygous Pkn mutant clones in the adult eye, and the loss-of-function Pkn mutant
does not suppress developmental defects associated with overexpression of Rho1 or Rac1 in the developing fly eye. Together with the observation that these GTPases appear to
play a role in establishing normal ommatidial polarity, these results suggest that it is unlikely that Pkn is
mediating the tissue polarity functions of Rho and Rac in eye development. However, a role for Pkn in additional morphogenetic processes that take place during larval development cannot be excluded (Lu, 1999).
Although the precise role of Pkn in DC is not clear, for the
following reasons, it appears likely that Pkn is transducing a
Rho-dependent signal in the LE cells: (1) Pkn expression is enriched in
the LE cells of stage 13 embryos; (2) the activated Rho1 GTPase binds
to and activates the Pkn kinase in vitro; (3) loss-of-function
Rho1 and Pkn mutants exhibit a very similar
dorsal-open phenotype that is associated with a defect in stretching of
the LE cells; (4) in both Rho1 and Pkn mutant
embryos, expression of DPP mRNA in the LE cells is apparently
normal; and (5) expression of the Rho1N19 and the
PknKD mutants specifically in the LE cells (where Pkn
expression is highly enriched just prior to DC) results in dorsal-open
phenotypes that are indistinguishable from the Pkn mutant phenotype.
The fact that expression of Rho1N19 specifically in the LE
cells leads to a cell nonautonomous stretching defect in the more lateral epidermal cells (despite normal dpp expression in the LE cells), suggests that Rho1 mediates a second, Dpp-independent, instructive signal. Possibly, this signal is
also mediated by a Rho1-Pkn interaction. However, the finding that
removal of one copy of Rho1 from a Pkn mutant GLC
background (a genetic null) increases the frequency of dorsal-open
embryos suggests that Rho1 is probably performing an additional,
Pkn-independent, function in DC. In support of this hypothesis, it was
found that the heat shock-Pkn transgenic construct is unable
to rescue the DC defect in Rho1 mutant embryos. What thus far remains unknown is the Rho effector target that mediates
the ability of Rho to regulate gene expression via activation of the
serum response factor, which is a known downstream target of Rho activation in
mammalian cells. Hence, it is possible
that this second instructive signal is induced by a Rho-mediated
transcriptional pathway, which is clearly distinct from the
Rac-mediated transcriptional pathway. It is worth noting that these
results do not exclude the possibility that the cell shape changes
associated with the lateral epidermal cells require an additional
direct (cell-autonomous) role for the Rho1-Pkn pathway in those cells
as well. It is clear, however, that the role of Rho1 in the LE cells
during DC is distinct from that of Rac1 (Lu, 1999).
In light of the fact that Drosophila Pkn interacts equally
well with the activated Rac GTPase, it is possible that a Rac-Pkn interaction also contributes to DC. However, the lack of a
Rac1 loss-of-function mutant in Drosophila makes it
difficult to examine the specific role of that interaction. Because the
JNK pathway mutants are also associated with a defect in stretching of
the LE cells, it has been suggested that components of the JNK pathway may mediate a Rac-dependent cell stretching signal that is unrelated to
transcriptional regulation. It is difficult to imagine how Pkn
could transduce a signal from Rac to this JNK-mediated cell shape
change pathway and yet not be required for the Rac-JNK transcriptional
pathway. However, it is possible that Pkn can transmit a Rac signal
independent of the JNK-Dpp pathway. Indeed, recent evidence suggests
that the Drosophila gene Myoblast city, which is
required for DC, encodes a Rac-specific activator that does not appear to regulate dpp expression
(Nolan, 1998). This observation suggests that Rac may perform
multiple functions in dorsal closure (Lu, 1999).
Significantly, there does seem to be some cross-talk between the
Pkn-mediated signaling pathway and the JNK pathway. Removal of one copy of
basket from a Pkn mutant GLC background significantly
increases the frequency of dorsal-open embryos. This result suggests
that some component of JNK cascade signaling is sensitive to the
activity of Pkn. Taken together with the fact that Rho1 generates a
JNK-Dpp independent signal in the LE cells that is required for DC, it is clear from these studies that distinct but coordinated signaling pathways mediated by the Rho and Rac GTPases within the LE cells are
essential for normal DC, and that Pkn is a strong candidate for an
effector that mediates signals downstream of both GTPases (Lu, 1999).
Plexins are neuronal receptors for the repulsive axon guidance molecule Semaphorins. Plexin B (PlexB) binds directly to the active, GTP-bound form of the Rac GTPase. A seven amino acid sequence in PlexB is required for RacGTP binding. The interaction of PlexB with RacGTP is necessary for Plexin-mediated axon guidance in vivo. A different region of PlexB binds to RhoA. Dosage-sensitive genetic interactions suggest that PlexB suppresses Rac activity and enhances RhoA activity. Biochemical evidence indicates that PlexB sequesters RacGTP from its downstream effector PAK. These results suggest a model whereby PlexB mediates repulsion by coordinately regulating two small GTPases in opposite directions: PlexB binds to RacGTP and downregulates its output by blocking its access to PAK and, at the same time, binds to and increases the output of RhoA (Hu, 2001).
Several lines of evidence suggest that RhoA is involved in PlexiB signaling. Clustering of the vertebrate PlexB in Swiss 3T3 cells leads to stress fiber formation, indicative of Rho activation. The response can be blocked by inhibitors of Rho or of its downstream effector Rho kinase. Genetic data also indicate that RhoA mediates part of Plexin B signaling in embryonic axon guidance. It was of interest, then, to enquire whether RhoA may also directly associate with PlexB (Hu, 2001).
PlexBDelta, a larger piece of the PlexB cytoplasmic domain (1617 through 1827) binds to RhoA. In contrast to a preferential binding to GTPgammaS-bound Rac, PlexBDelta binds to the GTPgammaS and GDP-bound forms of RhoA equally well. The binding requires the last 40 amino acids of PlexBDelta. The seven amino acid internal deletion that eliminates PlexBDelta binding to Rac does not affect its binding to RhoA. Thus, two independent regions in PlexB cytoplasmic domain have been defined that are important for PlexB association with Rac and RhoA, respectively. Cdc42, another Rho family GTPase, does not bind to PlexBDelta (Hu, 2001).
The role of RhoA in PlexB signaling was examined by reducing RhoA gene dosage with two different RhoA mutant alleles, Rhorev220 and RhoAl(2)k07236. Instead of enhancing the PlexB gain-of-function phenotypes as the Rac deficiency does, partially removing RhoA suppresses the PlexB gain-of-function phenotypes. This result suggests that RhoA acts antagonistically to Rac and, moreover, that RhoA partially mediates Plexin B signaling (Hu, 2001).
The results presented here allow a confirmation and extension of a current model concerning the role of GTPases in axon guidance. This model suggests that attractive guidance cues locally activate Rac or Cdc42 in the growth cone while repulsive guidance cues locally activate RhoA. It is argued that what is important is the relative balance in the output of Rac versus RhoA. An example is provided in which the PlexiB receptor mediates repulsive axon guidance by downregulating RacGTP output and simultaneously upregulating RhoA output. A coordinate regulation of these two small GTPases may allow the receptor to have a finer control over actin regulatory machinery. Semaphorin signaling can be converted from repulsion to attraction by changes in cGMP level. It would be interesting to test whether and how the cGMP signaling can affect this Rac/Rho balance (Hu, 2001).
Drosophila has two Plexins: A and B. Both Plexin A and B are highly expressed in the central nervous system. The two proteins share high sequence similarity in their cytoplasmic domain, indicating a similar mode of signaling shared by the two. A direct physical association of RacGTP with PlexB but not with PlexA has been demonstrated. However, genetic interactions have been found between Rac and both Plexins. For example, increasing PlexA also enhances the Rac dominant-negative phenotype as does PlexB. In COS cell and DRG neurons, Rac shows coclustering with PlexA upon Sema3A ligand treatment. It is likely that ligand binding to PlexA causes Rac binding (and subsequent inactivation of Rac) just as with PlexB, but it may be that PlexA requires an unknown third protein to help mediate or facilitate this physical interaction. From a genetic perspective, they both appear to function in the same way, mediating repulsion at least in part by inactivating Rac (Hu, 2001).
The transmembrane protein OTK associates with Plexin A and contributes to the Sema 1a/Plexin A signaling pathway. Mammalian Plexin B1 also coimmunoprecipitates with OTK. In the future, it will be interesting to test whether PlexB also interacts with OTK in vivo and to what degree the Rac/Rho GTPases and OTK signaling pathways function together or in parallel downstream of Plexins (Hu, 2001).
The key role of the Rho family GTPases Rac, Rho, and CDC42 in regulating the
actin cytoskeleton is well established. Increasing evidence suggests that the
Rho GTPases and their upstream positive regulators, guanine
nucleotide exchange factors (GEFs), also play important roles in the
control of growth cone guidance in the developing nervous system. The
identification and molecular characterization of a novel Dbl family
Rho GEF, GEF64C, is presented that promotes axon attraction to the central nervous
system midline in the embryonic Drosophila nervous system. In
sensitized genetic backgrounds, loss of GEF64C function causes a
phenotype where too few axons cross the midlin