Research ArticleDevelopmental Biology

Self-Induced Patched Receptor Down-Regulation Modulates Cell Sensitivity to the Hedgehog Morphogen Gradient

See allHide authors and affiliations

Science Signaling  24 Aug 2010:
Vol. 3, Issue 136, pp. ra63
DOI: 10.1126/scisignal.2001059


Morphogens form signaling gradients that control patterning processes during development. Responding cells must perceive and interpret the concentration-dependent information provided by the morphogen to generate precise patterns of gene expression and cell differentiation in developing tissues. Generally, the absolute number of activated, ligand-bound receptors determines cell perception of the morphogen. In contrast, cells interpret the morphogen Hedgehog (Hh) by measuring the ratio of bound to unbound molecules of its receptor Patched (Ptc). This ratio depends on both the Hh concentration and the absolute number of Ptc molecules. Here, I describe a posttranscriptional process that controls the absolute amount of Ptc present in a cell, which regulates gradient interpretation, wherein self-induced receptor down-regulation that is independent of ligand binding dictates the cell response to a morphogen gradient.


Morphogens play a key role in pattern formation of multicellular organisms. They spread from localized sites of production, forming a gradient that specifies distinct cellular outcomes at different concentrations. Because of their important role in development, many mechanisms regulate their production, movement, gradient formation, and temporal dynamics (15). In terms of their interpretation, however, for most morphogens, cells sense the graded signal by a simple mechanism—measuring the absolute number of activated, ligand-bound receptors. In these systems, gradient interpretation is independent of the number of inactive, unbound receptors (6, 7). The evolutionarily conserved Hedgehog (Hh) pathway is different. Cells receive and interpret extracellular Hh concentration through a mechanism mediated by two transmembrane proteins: its receptor Patched (Ptc) and the Hh signal transducer Smoothened (Smo) (8). Ptc is a 12-transmembrane domain protein with an overall structure similar to the resistance-nodulation-division (RND) family of bacterial proton–driven transmembrane molecular transporters (914). Ptc is active in the absence of Hh and blocks the expression of Hh target genes by inhibiting Smo, a seven-transmembrane receptor of the G protein (heterotrimeric guanosine triphosphate–binding protein)–coupled receptor family (1517). Hh binding to Ptc abrogates this inhibitory effect, unleashing Smo activity and inducing target gene expression (810, 13, 14, 18). Therefore, the active form of the receptor is the free, unbound Ptc. However, the inactive, bound form of Ptc also functions in the interpretation of the Hh gradient because cells measure the ratio of bound to unbound Ptc to respond appropriately to different concentrations of Hh (19).

In insects and vertebrates, one of the first responses to Hh is the transcriptional up-regulation of Ptc itself. In Drosophila wing imaginal discs, Hh is produced and secreted in the posterior compartment and spreads toward the anterior compartment (Fig. 1A), where it induces the expression of target genes, such as ptc and decapentaplegic (dpp) (20, 21). Hh signaling does not occur in posterior compartment cells because they do not express Ptc and other critical components of the Hh pathway. In contrast, all anterior compartment cells have sufficient Ptc to repress Smo in the absence of Hh. However, in those cells located close to the anteroposterior (AP) compartment boundary, Hh triggers pathway activity and, consequently, an increase in the transcription of the Ptc-encoding gene (9, 10, 2124). This transcriptional stimulation of Ptc results in the sequestration of Hh and restriction of Hh movement (10). However, in these cells close to the AP boundary, the increase in Ptc modifies the ratio between bound and unbound Ptc, thus posing the question of how these cells adapt to this dynamic environment to respond accurately to the Hh gradient.

Fig. 1

Ptc is posttranscriptionally down-regulated at the AP boundary. (A) Wing disc diagram. Blue represents the posterior (P), white represents the anterior (A), and red indicates the AP boundary. The rectangle marks the area shown in (D) to (H). (B) Sample GFP intensity plot of wing discs expressing L>Ptc-GFP (red) from data shown in (E) or L>PtcΔ2-GFP (green) from data shown in (F). A vertical line separates the anterior (left) and posterior (right) compartments. (C) Quantification of Ptc at the posterior and AP boundary of the wing imaginal discs. All differences between the intensity at the AP boundary and posterior compartment are statistically significant from the intensity of the Ptc-GFP signal of the same transgene in the anterior compartment far from the AP boundary based on Wilcoxon test (P < 0.001). Average and 95% confidence interval are shown. (D) Immunostaining of endogenous Ptc and β-galactosidase, showing dpp-lacZ, in a wild-type disc. (E to G) Wing discs of flies with the indicated transgenes showing GFP (green, transgene intensity), endogenous Ptc immunostaining (red), and dpp expression (blue, monitored by the dpp-LacZ transgene). (H) Wing disc of flies with the indicated transgene showing GFP (green, transgene) intensity and clones expressing membrane-attached Hh (UAS DSR/UAS CD2-Hh) (red). Anterior clone “1” and posterior clone “2” reveal down-regulation of PtcΔ2-GFP only in the anterior compartment. Anterior compartment boundary is marked by the sharp border of dpp-lacZ (blue) and Ptc (red) in (D) to (F) and by a white line in (G) and (H).

This work proposes the existence of a previously unknown mechanism of a self-induced decrease in Ptc abundance that lowers the total amount of Ptc at the cell surface. This mechanism requires the Ptc C-terminal domain (CTD), is independent of Hh binding, and is therefore distinct from the Hh binding–dependent Ptc endocytosis previously reported (2, 13, 2528). This self-induced decrease of Ptc at the cell surface may play a role in adjusting the cell’s sensitivity to extracellular Hh.


Ptc is posttranscriptionally down-regulated at the AP boundary of the wing disc

To follow Ptc dynamics, I generated a transgene encoding a wild-type form of Ptc tagged with green fluorescent protein (GFP) (L>Ptc-GFP; Table 1) and quantified its abundance in various regions of the wing imaginal disc (Fig. 1, B and C). From this transgene, low amounts of Ptc-GFP are produced ubiquitously throughout the organism. The fusion protein has GFP fused to the C terminus of Ptc. In wing discs from flies with L>Ptc-GFP, the abundance of Ptc-GFP was decreased at the AP boundary [defined by the sharp border of Ptc and Dpp (monitored with a dpp-lacZ transgene) expression (Fig. 1D)], precisely where there is transcriptional induction of the endogenous Ptc-encoding gene (Fig. 1E). To quantify the decrease in abundance, I assigned an arbitrary value of 100% to the GFP intensity in the region of the anterior compartment far from the AP boundary. At the AP boundary, GFP intensity was ~58% (Fig. 1, B, C, and E). The decrease in Ptc-GFP abundance occurred posttranscriptionally, because in situ hybridization with a probe for the GFP portion of the transcript did not show a decrease at the AP boundary (fig. S1). A similar decrease in Ptc-GFP abundance was also detected at the AP boundary in haltere and leg imaginal discs (fig. S2), suggesting that it is a general phenomenon.

Table 1

The various Ptc constructs used to evaluate Ptc regulation. Differences between constructs come from the amount of protein expressed, which was low (L), high (H), or very high (UAS). None of these transgenes were expressed from promoters that are responsive to Hh signaling, and therefore, the abundance of the produced proteins is controlled through posttranscriptional mechanisms. Transgenic proteins differ in their activity toward Smo. Normal activity means that binding of Hh alleviates Smo repression by Ptc and, in the absence of Hh, Ptc inhibits Smo.

View this table:

Ptc down-regulation at the AP boundary also occurs in the absence of Hh binding

Hh binding targets Ptc for degradation (2, 13, 2528). This presumably also occurs with the Ptc-GFP protein, because the intensity of GFP in the posterior compartment, where Hh is readily available, was lower than that in the anterior compartment far from the boundary, where there is no Hh (Fig. 1, B, C, and F). To analyze whether Hh binding was also responsible for the decrease in Ptc-GFP abundance at the AP boundary, I examined wing discs from flies with a transgene that expresses a mutant form of Ptc that does not bind Hh (L>PtcΔ2-GFP; Table 1) (18, 19). The abundance of PtcΔ2-GFP in these discs was increased in the posterior compartment, where Hh is present (Fig. 1, B, C, and F). However, the GFP intensity of PtcΔ2-GFP in the discs of flies with the L>PtcΔ2-GFP transgene was still reduced at the AP boundary, similar to the reduction in discs from flies with L>Ptc-GFP (Fig. 1, B, C, and F), indicating a Ptc degradation mechanism that is independent of Hh binding.

Ptc induces its decrease in abundance

I wanted to determine whether signaling through the Hh pathway was necessary for the reduced abundance of Ptc-GFP at the AP boundary. Some observations suggested that the reduction in Ptc-GFP was enhanced by Hh pathway activity. The regions of decreased abundance of Ptc-GFP and PtcΔ2-GFP coincided with the expression domain of the Hh target gene dpp-lacZ (Fig. 1, E and F). The decrease in Ptc-GFP abundance was graded in L>Ptc-GFP and L>PtcΔ2-GFP discs, with the strongest reduction in the cells located next to the AP boundary, where Hh signaling is maximal (Fig. 1B). Anterior clones that had high amounts of PtcΔ2-GFP protein from the H>PtcΔ2-GFP transgene (Table 1) suppressed Hh signaling (note the absence of dpp-lacZ expression in the anterior clone of Fig. 1G, which indicates inhibition of the Hh pathway), and the abundance of Ptc was not decreased. GFP intensity in both the AP boundary and the posterior compartment was similar [~103% in the posterior compartment versus ~100% in the AP boundary; Fig. 1G]. Ectopic activation of the pathway by overexpression of a membrane-tethered form of Hh, UAS CD2-Hh (29, 30), induced a decrease in abundance of PtcΔ2-GFP in anterior compartment clones of wing discs from L>PtcΔ2-GFP flies (clone 1, Fig. 1H), but not in posterior compartment clones where the Hh pathway cannot be activated (clone 2, Fig. 1H).

However, two results suggest that Ptc itself, rather than the activation of the Hh pathway, is responsible for the decrease in Ptc-GFP abundance at the AP boundary. The addition of GFP to the Ptc C terminus does not affect Ptc’s ability to repress Smo activity in the absence of Hh or its ability to relieve such repression in response to Hh (27). Accordingly, Ptc-GFP in the L>Ptc-GFP transgenic fly discs repressed dpp-lacZ far from the AP boundary in mitotic ptc clones (Fig. 2A) and allowed dpp-lacZ expression at the boundary where Hh signaling occurs (Fig. 2B). If activation of the Hh pathway was required to mediate the decrease in Ptc abundance, one would expect the intensity of GFP from Ptc-GFP to be the same inside and outside ptc clones far from the boundary, where the absence of Hh keeps the pathway “off.” Instead, the intensity of GFP inside the ptc clone was higher than in surrounding cells (Fig. 2A), suggesting that endogenous Ptc, even in the absence of Hh, is needed to induce a decrease in Ptc-GFP abundance. In wing discs of L>Ptc-GFP flies containing ptc clones at the AP boundary, the abundance of Ptc-GFP was not decreased in the ptc regions, even though this is where, as a result of the presence of high concentrations of Hh, the pathway is active as indicated by the expression of target genes, except for ptc in these genetically deficient cells (Fig. 2B). These results suggest that Ptc itself induces its own down-regulation and that the correlation with the domain of Hh pathway activation may reflect the transcriptional stimulation of Ptc abundance in response to Hh, which leads to a higher degree of Ptc degradation. Consistent with this model of Ptc-induced Ptc down-regulation, clones with high amounts of Ptc from the UAS-Ptc transgene (Table 1) (30) exhibited reduced GFP intensity in both the anterior and the posterior compartments of wing discs of L>PtcΔ2-GFP flies (Fig. 3A).

Fig. 2

Ptc induces Ptc down-regulation. (A) Low abundance of Ptc-GFP (by the transgene L>Ptc-GFP) in the anterior compartment represses dpp-lacZ (blue) in ptc clones (marked by the absence of CD2, red). In this disc, GFP intensity is higher inside the clone (marked by a white line) than in surrounding cells. (B) Ptc-GFP intensity is not reduced in a ptc clone, marked by the absence of endogenous Ptc up-regulation (red), located just at the AP boundary, even though the presence of dpp-lacZ (blue) indicates that the pathway is active.

Fig. 3

Regulation of Smo signaling is not required, but the Ptc CTD is required for Ptc down-regulation. Green indicates the abundance of the ubiquitously expressed transgene L>PtcΔ2-GFP; red shows the presence of abundant Ptc from the strong-expressing flip-out clones (indicated along the left). (A to E) PtcΔ2-GFP is down-regulated when wild-type (A), constitutively active (B), or dominant-negative (C) forms of Ptc are overexpressed, but is stable when the overexpressed Ptc lacks its CTD (D) or when the CTD has been replaced with the human LDL receptor intracellular domain (E).

Ptc requires its CTD to induce its own down-regulation

The next experiments address whether the repression of Smo activity by Ptc was required for Ptc self-induced down-regulation. Ptc with a deletion of the second extracellular loop (PtcΔ2) cannot bind Hh and is therefore constitutively active (always repressing Smo independently of Hh presence) (18, 19), whereas the point mutation D583N generates a form of Ptc (PtcS2) that behaves as a dominant-negative mutant that does not repress Smo (31, 32). Overexpression of either of these two mutants in flies with the UAS-PtcΔ2 transgene (Table 1 and Fig. 3B) or the UAS-PtcS2 transgene (Table 1 and Fig. 3C) resulted in reduced abundance of PtcΔ2-GFP in both the anterior and the posterior compartments of wing discs of L>PtcΔ2-GFP flies. These results showed that, when overexpressed, Ptc induces a decrease in PtcΔ2-GFP regardless of its capacity to repress Smo.

The Ptc CTD plays a major role in regulating Ptc internalization and turnover (33, 34). The abundance of PtcΔ2-GFP in the wing discs of L>PtcΔ2-GFP flies was not decreased by overexpression of a form of Ptc that lacks its CTD (PtcΔCTD, from the UAS PtcΔCTD transgene) (Table 1 and Fig. 3D). Because PtcΔCTD accumulates at the plasma membrane (33, 35), the lack of a decrease in the abundance of PtcΔ2-GFP may reflect deficient Ptc endocytosis. To analyze whether internalization was necessary for Ptc down-regulation or whether some other function of the CTD was needed, I replaced the CTD with the human low-density lipoprotein (LDL) receptor intracellular domain and expressed it from the transgene UAS PtcLDL (Table 1). This Ptc fusion protein was endocytosed and sequestered Hh, but did not repress Smo activity (fig. S3). The abundance of PtcΔ2-GFP in the wing discs of L>PtcΔ2-GFP flies was not decreased by overexpression of PtcLDL (Fig. 3E), suggesting that the CTD plays a specific role in inducing down-regulation independently of its role in endocytosis. The abundance of PtcΔ2-GFP in the wing discs of L>PtcΔ2-GFP flies was also not decreased by overexpression of Dispatched, a protein structurally related to Ptc that is required for Hh secretion (36) (fig. S4), suggesting that the effect observed is specific for Ptc.

The NPXY motif present at the CTD is required for self-induced Ptc down-regulation

Ptc contains a NPXY motif in the CTD that is required to bind to the ubiquitin ligase Nedd4 (33, 34). This and other results have led to the suggestion that monoubiquitination in the CTD is a signal that targets Ptc to the lysosomes, although it is not required for movement to early endosomes (33). Ptc also forms stable trimers (33). Therefore, a possible mechanism for Ptc self-induced reduction in abundance could be that ubiquitination of one molecule within a trimer induces the degradation of the whole complex.

To analyze the role of the NPXY motif in Ptc self-induced down-regulation, I generated the transgenes UAS PtcAAAA (producing large amounts of PtcAAAA) and L>PtcAAAA-GFP (producing low amounts of PtcAAAA-GFP); these produced a Ptc in which the PPXY motif has been mutated to AAAA (Table 1). The NPXY motif is not required for signal transduction (33) and, consistently, L>PtcAAAA-GFP rescued ptc mitotic clones, repressing dpp-lacZ in clones located far from the AP boundary but allowing dpp-lacZ production in clones located at the boundary where Hh signaling occurs (fig. S5). The ubiquitous presence of low amounts of PtcAAAA-GFP in the wing discs of L>PtcAAAA-GFP flies resulted in a weaker decrease in GFP signal at the AP boundary (~75%; Figs. 1C and 4A) and in the posterior compartment (~88%; Figs. 1C and 4A) than that observed for Ptc-GFP in the L>Ptc-GFP flies (Fig. 1, C and E), suggesting that both Hh binding–dependent and Hh binding–independent degradation mechanisms are reduced in the absence of the NPXY motif. Posterior clonal overexpression of PtcAAAA by the UAS PtcAAAA transgene did not induce a strong reduction in the abundance of PtcAAAA-GFP in L>PtcAAAA-GFP flies (Fig. 4B). This result suggests that when the complex is formed only by molecules that do not contain the NPXY motif, it is not subjected to the Ptc-induced down-regulation. In contrast, the presence of just one ubiquitination-competent molecule in the complex may be sufficient to target the whole complex for down-regulation, because wild-type Ptc (from the UAS-Ptc transgene) induced a reduction in PtcAAAA-GFP (from the L>PtcAAAA-GFP) (Fig. 4C), and UAS PtcAAAA induced a reduction in the GFP intensity of PtcΔ2-GFP (from the L>PtcΔ2-GFP transgene) (Fig. 4D). These results suggest a role for the Ptc NPXY motif in controlling the total amount of Ptc present in the cell and, consequently, how the Hh gradient is interpreted. In experimental conditions, the self-induced decrease in the abundance of Ptc happens throughout the entire wing disc, suggesting that there are no spatial restrictions or that factors involved in the process may be ubiquitous.

Fig. 4

At least one Ptc with an NPXY motif is required for maximal down-regulation of Ptc. (A) Wing discs of L>PtcAAAA-GFP flies with GFP intensity (green), endogenous Ptc (red), and dpp expression (blue, dpp-lacZ) show less down-regulation of the transgene than in wing discs of flies with the wild-type Ptc transgene (see Fig. 1C for quantitation). (B and C) PtcAAAA-GFP (green) is down-regulated by wild-type Ptc (red, UAS-Ptc) but not a form of Ptc lacking the NPXY motif (red, UAS PtcAAAA). (D) The abundance of the constitutively active Ptc (green, transgene L>PtcΔ2-GFP) is reduced by overexpression of Ptc lacking the NPXY motif (red, UAS PtcAAAA).

To analyze whether the addition of GFP at the Ptc C terminus affected the stability of Ptc, I compared the activity of both tagged and untagged forms of PtcΔ2. In the presence of endogenous Ptc, low amounts of PtcΔ2 or PtcΔ2-GFP did not repress Hh pathway target genes next to the AP boundary (fig. S6, A and B) (19). In contrast, transgenes that produced high amounts of PtcΔ2 or PtcΔ2-GFP fully repressed Hh target genes (fig. S6, C and D). These results suggest that the addition of GFP does not prevent the interaction with endogenous Ptc and that Ptc oligomers behave similarly in the presence or absence of the GFP tag.

Decreased abundance of Ptc adjusts the cell’s response to intermediate concentrations of Hh

The self-induced, Hh binding–independent Ptc down-regulation mechanism may adjust the total number of Ptc molecules present in a cell and, consequently, control the cell’s sensitivity to the extracellular concentration of Hh. Because cells read the ratio between bound and unbound Ptc to measure Hh extracellular concentration (19), an increase in the total number of Ptc molecules would change this ratio and therefore change the cellular response to the same amount of Hh. To test this hypothesis, I used a mutant form of Ptc, Ptc14, which has a Leu-Gln change (L83Q) and produces a protein that is not internalized but regulates Smo activity (27). To analyze whether the Hh binding–independent Ptc down-regulation mechanism was impaired by the ptc14 mutation, I generated a transgene that produced high amounts of Ptc14-GFP (H>Ptc14-GFP; Table 1). GFP intensity in wing discs ubiquitously expressing Ptc14-GFP was similar in the posterior compartment (~90%), in the AP boundary (~91%), and in mitotic ptc14 clones (~86%) located next to the AP boundary (Fig. 5C). These clones produce only this mutant form of the protein under the control of the native Hh-responsive promoter, and therefore it is transcriptionally up-regulated by Hh (Fig. 5A). This result indicates that the Hh-regulated increase in Ptc14 did not decrease the abundance of Ptc14-GFP (from the H>Ptc14-GFP transgene), and therefore that Ptc14 mutant protein is unable to induce its own down-regulation.

Fig. 5

The Hh gradient is misinterpreted in the absence of Ptc down-regulation. (A) When expressed ubiquitously from the transgene H>Ptc14-GFP (Table 1), the GFP intensity of Ptc14-GFP is similar at the posterior compartment, at the AP boundary (white line), and in ptc14 mitotic clones (marked by the absence of DsRed) located at the AP boundary, suggesting that the self-induced down-regulation of Ptc14 is low. Hh pathway activity is indicated by dpp-lacZ expression (blue). (B) The ptc14 clone 1, located at the anterior compartment far from the AP boundary (thick white line) where there is no Hh, shows repression of dpp-lacZ expression, indicating that Ptc14 repressed Smo. dpp-lacZ (green) is present in wild-type amounts in clone 2, located just at the AP boundary, indicating that Hh alleviated the Ptc14-mediated Smo inhibition. Clone 3, located a few cell diameters away from the AP boundary, shows lower amounts of dpp-lacZ compared to neighboring cells (magnified in B′; compare dpp-lacZ abundance in the cells marked with asterisks). A white square marks the magnified region in (B′). (C) Quantification of the Ptc-GFP intensity in regions of the discs with different Ptc backgrounds. Average and 95% confidence interval or the values of a single replicate are shown.

To analyze whether the lack of Ptc down-regulation affects the interpretation of Hh gradient, I generated mitotic ptc14 clones. When located far from the AP boundary, clones expressing Ptc14 repressed dpp-lacZ expression (Fig. 5B, clone 1), but did not when located near the AP boundary (Fig. 5B, clone 2) (27). However, ptc14 clones located a few cell diameters away from the AP boundary showed weaker dpp-lacZ expression when compared to neighboring cells (Fig. 5B, clone 3, magnified in B′). This result suggests that, in the absence of Ptc self-induced degradation, cells accumulate more Ptc molecules and are therefore less sensitive to Hh. When Hh is abundant (next to the AP boundary), accumulation of Ptc14 protein has no effect on the cellular response. However, a few cell diameters away, where extracellular Hh is less abundant, accumulation of Ptc14 reduces cellular sensitivity to Hh and, consequently, cells behave as if they were located in a more anterior position, farther away from the Hh source.


The results shown here indicate that self-induced Ptc down-regulation contributes to setting the cellular sensitivity to extracellular concentrations of Hh (Fig. 6). Ptc has two functions—to sequester Hh and to regulate Smo activity—and these actions seem to have contradictory requirements. Wherever it has been examined, Hh induces the expression of the gene encoding Ptc, resulting in cells adjacent to the Hh source with highly abundant Ptc (8). This up-regulation has a crucial role in binding and sequestering Hh (10, 18). Hh binding to Ptc induces its endocytosis and degradation at the lysosome, thus shaping the Hh gradient and restricting its range of action (27, 28). Ptc is also a negative regulator of Hh signaling, because increasing Ptc concentration progressively decreases the activity of the Hh pathway as a result of the increase in unoccupied Ptc molecules that can continue to inhibit Smo (10, 14, 19, 37). The results shown here suggest that Ptc fine-tunes its own accumulation at the cell membrane, keeping a steady state that allows cells to accurately measure and respond to the Hh gradient.

Fig. 6

Model diagram. Self-induced Ptc down-regulation and Hh binding–dependent Ptc degradation adjust the number of Ptc molecules at the cell membrane. Therefore, both mechanisms modulate cellular responsiveness to extracellular Hh. In the absence of Ptc self-induced down-regulation, cells become less responsive to Hh as the accumulation of Ptc molecules shifts the ratio of bound to unbound Ptc.

The sorting of signals and their receptors to different membrane-bound compartments plays a critical role in modulating the amount and localization of signaling during development (38, 39). In the Hh pathway, endocytosis seems to regulate signaling through two different mechanisms, one ligand-dependent and the other ligand-independent. The ligand-dependent endocytosis mechanism would target Ptc and Hh for degradation, thus shaping the Hh gradient. Furthermore, binding of Hh is thought to promote sorting of Ptc and Smo into an endosomal compartment, resulting in preferential recycling of Smo and net pathway activation from the plasma membrane (26). The ligand-independent mechanism of Ptc down-regulation would increase the cell’s sensitivity to the Hh extracellular gradient. Many cells that respond to gradients can adjust their sensitivity up and down as a function of exposure to a specific ligand. For example, signaling receptors may undergo rapid ligand-induced endocytosis after ligand-induced activation, thereby reducing the number of receptors present on the cell membrane and attenuating cellular responsiveness to extracellular ligand. In contrast, Ptc degradation, regulated indirectly by Hh, would increase the cellular responsiveness, because a lower amount of Ptc protein makes the cell more sensitive to Hh by increasing the ratio of ligand-bound to unbound receptors.

The CTD of Ptc seems to play an essential role in the regulation of Ptc abundance at the cell membrane. The CTD is important for the formation of trimeric complexes and the regulation of the localization of the protein, because a mutant form of Ptc that lacks the CTD remains at the plasma membrane (33). Also, the CTD contains an NPXY motif that has been shown in other systems to be essential for binding the ubiquitin ligase Nedd4 (33), leading to the suggestion that Ptc could be ubiquitinated. Therefore, even though ubiquitination of Ptc has not been shown in vivo, one possible interpretation of the results shown here is that a ubiquitination-mediated internalization of a trimeric Ptc complex modulates the total amount of Ptc present at the cell membrane. It is unclear, however, how the rate of ubiquitination or internalization, or both, could be influenced by the increase in Ptc protein. If Ptc was subjected to degradation only when forming part of a trimeric complex, one possibility would be that the increase in the number of Ptc molecules would increase the percentage of these molecules that form trimers and therefore would be subjected to degradation.

Finally, a balance of ubiquitination and deubiquitination of the Wingless (Wg, also known as Wnt) receptor Frizzled determines the cellular responsiveness to Wnt both in mammalian cells and in Drosophila (40). This suggests that mechanisms that control the cell surface abundance of receptors might be more general, controlling cellular responsiveness to morphogens.

Materials and Methods


Immunostaining was performed with standard techniques and antibodies against Ptc (a gift from I. Guerrero), GFP (Molecular Probes), and β-galactosidase (Cappel).

Ptc mutations, fusion proteins, and transgenes

The following mutations and transgenes have been previously described: UAS Ptc (10), hsp70-flp (20), FRT42D ptc14 (27), UAS CD2-Hh (29), UAS Disp (36), dpp-lacZ10628 (41), and Tubα1>Gal80>Gal4 (42). The stocks containing UAS DsRed, FRT42D Tubα1>DsRed, and FRT42D Tubα1>Gal80; Tubα1>Gal4 were gifts from G. Struhl. Low (L) and high (H) expression were accomplished as described in (19). Ptc-GFP was generated by fusing the GFP sequence five amino acids before Ptc stop codon. PtcAAAA-GFP (P1219A, P1220A, Y1222A) and Ptc14-GFP (L83Q) were generated by polymerase chain reaction (PCR) and cloned into Ptc-GFP. UAS PtcΔCTD was generated by changing Leu1105 to a stop codon and cloning it into the pUAST vector. To generate UAS PtcLDL, I used a complementary DNA clone of human LDL receptor as a template to amplify its intracellular domain by PCR (primers used were 5′-AAGCTTCTAAGAACTGGCGGCTTAAG-3′ and 5′-ACTAGTTCACGCCACGTCAT CCTC-3′). The PCR product was cloned into Ptc at the Leu1105 position.


The following Drosophila genotypes were used:

Fig. 1: y w hsp70-flp; dpp-lacZ10628/+; rp49>Ptc-GFP (or PtcΔ2-GFP )–hsp70 3UTR or Tubα1>CD2, y+>PtcΔ2-GFP–Tubα1 3UTR/+ and y w hsp70-flp UAS DsRed; dpp-lacZ10628 Tubα1>Gal80, y+>Gal4/+; rp49>PtcΔ2-GFP–hsp70 3UTR/UAS-CD2-Hh.

Fig. 2: y w hsp70-flp; FRT42D ptcIIW sha/ dpp-lacZ10628 FRT42D Tubα1>Gal80 CD10y+; rp49>Ptc-GFP–hsp70 3′UTR/Tubα1>Gal4.

Fig. 3: y w hsp70-flp UAS DsRed; dpp-lacZ10628 Tubα1>Gal80, y+>Gal4/+; rp49>PtcΔ2-GFP–hsp70 3UTR/ UAS Ptc (or UAS PtcΔ2 or UAS PtcS2 or UAS PtcΔCTD or UAS PtcLDL).

Fig. 4: y w hsp70-flp; dpp-lacZ10628/+; rp49>PtcAAAA-GFP hsp70 3′UTR/+ and y w hsp70-flp UAS DsRed (or y w hsp70-flp); dpp-lacZ10628 Tubα1>Gal80, y+>Gal4/+; rp49>PtcΔ2-GFP (or PtcAAAA-GFP)–hsp70 3′UTR/UAS Ptc (or UAS PtcAAAA).

Fig. 5: y w hsp70-flp;FRT42Dptc14/FRT42DTubα1>DsRed; Tubα1>Ptc14-GFP–Tubα1 3′UTR (or +)/+.

Flp-out and mitotic clones were induced by 1-hour heat shocks at 37°C to 24- to 48-hour larvae.

Quantification of protein abundance

Protein abundance was quantified as previously described (19). Briefly, the intensity of GFP was determined by averaging the intensity of five samples each within (i) the anterior compartment far from the AP boundary, (ii) the posterior compartment, and (iii) the anterior compartment next to the AP boundary. The GFP intensity in (i) was assigned a 100% value, and the intensities of (ii) and (iii) were calculated according to it. All measurements were made within the wing primordium. GFP intensity plots were generated with ImageJ 1.34s (W. S. Rasband, Statistical analysis was performed with the Wilcoxon signed-rank test.


Acknowledgments: I am in special debt to G. Struhl and J. Casanova for ideas and support. I also thank A. Adachi (G. Struhl laboratory) and E. Caminero (Plataforma de Transformación de Drosophila, Consolider Ingenio 2007) for injections; G. Darras for the in situ hybridization; members of J. Casanova’s laboratory for assistance and support; D. Rosell for help with statistical analysis; and K. Campbell, J. Casanova, J. Font, X. Franch-Marro, D. Shaye, M. Strigini, and G. Struhl for discussion and advice on the manuscript. This work was mostly done in J. Casanova’s laboratory. Funding: Supported by the European commission (MIRG-CT-2007-44921) and the Spanish Ministerio de Educación y Ciencia (BFU2007-60663/BMC). Competing interests: The author declares that he has no competing interests.

Supplementary Materials

Fig. S1. In situ hybridization to detect GFP-positive transcripts.

Fig. S2. PtcΔ2-GFP (from the L>PtcΔ2-GFP transgene) is down-regulated in leg and haltere discs.

Fig. S3. PtcLDL is endocytosed and sequesters Hh, but does not repress Smo.

Fig. S4. Overexpression of Dispatched (Disp) does not induce degradation of Ptc-GFP (from the L>Ptc-GFP transgene).

Fig. S5. PtcAAAA-GFP regulates Smo activity in the absence of endogenous Ptc.

Fig. S6. The GFP tag does not affect the ability of PtcΔ2 to interact with endogenous Ptc.

References and Notes

View Abstract

Stay Connected to Science Signaling

Navigate This Article