Genetic Evidence for a Protein Kinase A/Cubitus Interruptus Complex That Facilitates Processing of Cubitus Interruptus in Drosophila

Corresponding author: John A. Kiger, Jr., Molecular and Cellular Biology, University of California, 1 Shields Ave., Davis, CA 95616., jakiger{at}ucdavis.edu (E-mail)

Communicating editor: T. SCHÜPBACH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Hedgehog (Hh) activates a signal transduction pathway regulating Cubitus interruptus (Ci). In the absence of Hh, full-length Ci (Ci-155) is bound in a complex that includes Costal2 (Cos2) and Fused (Fu). Ci-155 is phosphorylated by protein kinase A (PKA), inducing proteolysis to Ci-75, a transcriptional repressor. Hh signaling blocks proteolysis and produces an activated Ci-155 transcriptional activator. The relationship between PKA and the Ci/Cos2/Fu complex is unclear. Here we examine Hh target gene expression caused by mutant forms of PKA regulatory (PKAr) and catalytic (PKAc) subunits and by the PKAc inhibitor PKI(1-31). The mutant PKAr*, defective in binding cAMP, is shown to activate Hh target genes solely through its ability to bind and inhibit endogenous PKAc. Surprisingly, PKAcA⁷⁵, a catalytically impaired mutant, also activates Hh target genes. To account for this observation, we propose that PKAc phosphorylation targeting Ci-155 for proteolysis is regulated within a complex that includes PKAc and Ci-155 and excludes PKI(1-31). This complex may permit processive phosphorylation of Ci-155 molecules, facilitating their processing to Ci-75.

EARLY in Drosophila development most cells divide along the anterior-posterior axis of the embryo into groups called compartments, designated alternately "anterior" or "posterior." Some cells in each compartment are fated to form the larval body and others to become imaginal cells that later differentiate to form the adult body. Anterior and posterior cells communicate with each other to pattern first the larval body segments and later the imaginal discs, employing a number of signaling pathways. One of the earliest communications between compartments employs the highly conserved Hedgehog (Hh) pathway. The current state of knowledge on creation of the extracellular Hh signal and its reception and processing by target cells has recently been reviewed (KALDERON 2000 ; MCMAHON 2000 ).

Hedgehog signaling is best understood in the Drosophila wing imaginal disc. Cells of the anterior compartment synthesize the membrane proteins Patched (Ptc) and Smoothened (Smo) and the transcription factor Cubitus interruptus (Ci). Cells of the posterior compartment also synthesize Smo as well as Engrailed (En), which represses synthesis of Ptc and Ci and induces synthesis and secretion of Hh. Extracellular Hh crosses the compartment border and activates the Hh signal transduction pathway by binding its receptor, Ptc, on the surface of anterior cells. In anterior cells at the compartment border, Hh signaling elicits synthesis of En and Decapentaplegic (Dpp) and increases synthesis of Ptc. Newly synthesized Ptc (to which Hh binds) limits the movement of Hh across the anterior compartment, creating a high concentration of Hh (and thereby En and Dpp) near the anterior/posterior boundary. Dpp then diffuses across both the anterior and posterior compartments, patterning the development of the entire wing (BASLER and STRUHL 1994 ; CHEN and STRUHL 1996 ; STRIGINI and COHEN 1997 ).

While Ptc normally restricts movement of Hh into the anterior compartment, all anterior cells can respond to Hh. Ectopic expression of Hh near the anterior margin of the wing disc leads to induction of Hh target genes, producing duplications of wing pattern (TABATA and KORNBERG 1994 ). The discovery of mutants that mimic the effect of ectopic Hh expression in the anterior compartment of the wing disc has implicated several additional proteins in Hh signal transduction: Cos2, a kinesin-related/microtubule-binding protein (ROBBINS et al. 1997 ; SISSON et al. 1997 ); Slimb/ß-TrCP, an F-box/ubiquitin ligase component (JIANG and STRUHL 1998 ; SPENCER et al. 1999 ); and PKAc (protein kinase A catalytic subunit encoded by DC0) (JIANG and STRUHL 1995 ; LEPAGE et al. 1995 ; LI et al. 1995 ; PAN and RUBIN 1995 ; STRUTT et al. 1995 ). Mitotic clones, homozygous for loss-of-function mutations of these genes, produce large wing duplications when located in the anterior compartment. Similar wing duplications can be produced by ectopic expression, under GAL4/UAS control, of PKAr*, a mutant PKA regulatory subunit type I, defective in its ability to bind cAMP and thereby rendered a constitutive inhibitor of PKAc (LI et al. 1995 ).

How activated Ptc interacts with Smo to further transduce the signal is a matter of controversy (KALDERON 2000 ). However, at the opposite end of the Hh signal transduction pathway is Ci, whose role in transducing the cytoplasmic state of the pathway into the nucleus is well established (for reviews, see INGHAM 1998 ; AZA-BLANC and KORNBERG 1999 ). Ci is a complex transcription factor existing in multiple forms. Ci-155 is mostly cytoplasmic and is bound in a large microtubule-associated complex with Cos2 and Fu (ROBBINS et al. 1997 ; SISSON et al. 1997 ). Su(Fu) binds both Ci-155 and Fu and may also be a part of this complex (MONNIER et al. 1998 ). Ci-155 is a substrate for PKAc phosphorylation that, in the absence of Hh signaling, targets it for proteolysis to Ci-75 (AZA-BLANC et al. 1997 ; Y. CHEN et al. 1998 , Y. CHEN et al. 1999 ; C.-H. CHEN et al. 1999 ; METHOT and BASLER 1999 , METHOT and BASLER 2000 ; PRICE and KALDERON 1999 ; WANG et al. 1999 ). Ci-75 enters the nucleus and represses transcription of hh and dpp. Binding of Hh to Ptc at the cell surface leads to dissociation of Cos2 and the rest of the complex from microtubules through the action of Smo (ROBBINS et al. 1997 ; KALDERON 2000 ). Concomitantly, Ci-155 proteolysis is inhibited and an activated form of Ci-155, different from cytoplasmic Ci-155, appears (AZA-BLANC et al. 1997 ; OHLMEYER and KALDERON 1998 ; METHOT and BASLER 1999 , METHOT and BASLER 2000 ; WANG and HOLMGREN 1999 ; WANG et al. 1999 ). Activated Ci-155 enters the nucleus where it activates transcription of dpp, en, and ptc. The nuclear localization of Ci-75 and of activated Ci-155 is controlled in multiple ways (AZA-BLANC et al. 1997 ; OHLMEYER and KALDERON 1998 ; C.-H. CHEN et al. 1999 ; WANG and HOLMGREN 1999 ; METHOT and BASLER 2000 ).

The relationship between PKA and the Ci/Cos2/Fu complex is presently unclear. PKAc is generally thought to be part of a PKA holoenzyme where its activity is inhibited by PKAr and regulated by cAMP (TAYLOR et al. 1990 ). Since Hh signaling leads to inhibition of Ci-155 phosphorylation and proteolysis, one might expect that this would coincide with a drop in ambient cAMP level and inhibition of PKAc activity by PKAr. No change in cAMP levels, however, has been detected in a mouse cell line sensitive to Sonic hedgehog signaling when exposed to Sonic hedgehog (MURONE et al. 1999 ). Experiments addressing this question in Drosophila have employed ectopic expression of PKAc*, a constitutively active mutant mouse catalytic subunit impaired in interaction with PKAr. Low levels of PKAc* expression appear to not affect normal Hh signaling (JIANG and STRUHL 1995 ; LI et al. 1995 ; OHLMEYER and KALDERON 1997 ), while high levels override Hh signaling and promote Ci-155 proteolysis (LI et al. 1995 ; WANG et al. 1999 ).

We have previously shown that ectopic expression of the PKAc inhibitors PKAr* and PKI(1-31) produces quite different responses from Hh target genes: PKAr* mimics the effect of mitotic clones deficient in PKAc (homozygous for loss-of-function DC0 mutants), while PKI(1-31) has no effect, despite its demonstrated ability to inhibit endogenous PKAc (KIGER et al. 1999 ). Here we address the basis for the different effects of PKAr* and PKI(1-31) through studies of mutant forms of PKAr* and PKAc.

We test the possibility that free PKAr* (and free wild-type PKAr) has a target other than PKAc through which it induces Hh signaling. Precedence for such a role can be found in Dictyostelium, where free PKAr binds to, and activates, a cAMP-specific phosphodiesterase (SHAULSKY et al. 1998 ). In this view, a mitotic clone deficient in PKAc would induce Hh signaling because of the free PKAr created therein.

We show that ectopic expression of a catalytic site mutant of PKAc, PKAcA⁷⁵, produces an unexpected bipartite phenotype compared to that of ectopic PKAc. To wit, PKAcA⁷⁵ is null with respect to PKAc phenotypes and is dominant negative with respect to activation of Hh target genes. These observations cast new light on the role of PKAc in Hh signaling and indicate that PKAc is part of a larger complex that includes Ci-155. This insight makes it possible to understand those attributes of PKAr*, lacking in PKI(1-31), that allow PKAr* to block PKAc involved in phosphorylation of Ci-155.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Transgene construction:
UAS-PKAr* mutant transgenes: The vector pUAST-HAPKAr* used to create transgenic strains UAS-PKAr*BDK 33 and 35 was generously supplied by Daniel Kalderon. This vector carries a 2-kb KpnI/XbaI fragment of cDNA containing the mutated Drosophila PKArI gene (KALDERON and RUBIN 1988 ) inserted into the multiple cloning site of pUAST (BRAND and PERRIMON 1993 ). Fused in frame to its 5' end is a synthetic DNA encoding a hemagglutinin (HA) epitope tag followed by a duplication of the first three codons of the PKArI gene. The KpnI/XbaI fragment was excised, cloned into Bluescript to give pBS-PKAr*, and mutated using the Stratagene (La Jolla, CA) QuickChange kit to create pBS-PKAr*G^91,92. Sequencing of pBS-PKAr*G^91,92 confirmed the presence of G^91,92 as well as the mutations in the cAMP binding sites indicated in Fig 1. However, glutamine³⁰⁴ in the published sequence is missing in both pBS-PKAr* and pBS-PKAr*G^91,92; this is evidently a consequence of alternative splice sites separated by one codon (KALDERON and RUBIN 1988 ; D. KALDERON, personal communication). The region encoding the dimerization domain of pBS-PKAr* was deleted by cutting with the blunt-end enzymes MscI and ScaI (Fig 1) and religating to give pBS-PKAr*. Sequencing confirmed the presence of the deletion. The KpnI/XbaI fragments of pBS-PKAr*G^91,92 and pBS-PKAr* were then cloned back into pUAST-HAPKAr* cut with KpnI and XbaI. The resulting vectors were used to create UAS-PKAr*G^91,92 and UAS-PKAr* transgenes.

View larger version (5K): In this window In a new window Download PPT slide   Figure 1. The PKA regulatory subunit type I monomer. Solid boxes identify functional domains discussed in the text. Numbers above the boxes indicate amino acid positions. The letters above the numbers indicate amino acids found in the wild-type protein, and the arrows show the mutational changes discussed. MscI cuts between G and C at codon 6 (GCC/ala), and ScaI cuts between T and A at codon 52 (TAC/tyr) of the cDNA, allowing an in-frame deletion of the dimerization domain to be made.

PKAcA⁷⁵ transgenes: The vector pBS-DC0A (KIGER et al. 1999 ) containing most of the wild-type PKAc cDNA was mutated using the Stratagene QuickChange kit to change lysine⁷⁵ to alanine; the result was confirmed by sequencing. This vector was then used to create UAS-PKAcA⁷⁵ transgenes as described previously (KIGER et al. 1999 ).

Fly strains and crosses:
The GAL4 and UAS strains employed have been described previously (KIGER et al. 1999 ), with the exception of apterous-GAL4, which was obtained from the Bloomington Stock Center. Larvae carrying both apterous-GAL4 and UAS-PKAr*G^91,92 transgenes were identified using green fluorescent protein (GFP)-tagged balancer chromosomes (CASSO et al. 1999 ). All crosses were carried out at 25°.

Immunostaining and microscopy:
The procedures followed have been described previously (KIGER et al. 1999 ). Wing discs were stained for HA-tagged PKAr*G^91,92 using horseradish peroxidase conjugated HA-probe (F-7) from Santa Cruz Biotechnology at a final dilution of 1/200 or 1/400. Peroxidase activity was detected using DAB, and dissected discs were mounted in 70% glycerol (PATEL 1994 ). GFP expression patterns were obtained by dissecting and mounting unfixed wing discs in PBS. Photography was carried out using a Zeiss Axioplan microscope equipped for differential interference contrast and epifluorescence.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The domain structure and functional features of the PKA regulatory subunit type I monomer are shown in Fig 1. Two monomers dimerize in reverse orientation and are covalently linked by disulfide bonds between the pairs of cysteine residues in the dimerization domains, D, to give PKAr. The dimerization domain also contains residues that are essential for PKAr type I binding to dual specificity A-kinase anchoring proteins (D-AKAPs); these residues are conserved between humans and Caenorhabditis elegans (BANKY et al. 1998 ). In the PKA holoenzyme, the substrate binding sites of two PKAc molecules are bound to the pseudosubstrate inhibitory domains, I, of PKAr (see Fig 1). The inhibitory domain is characterized by a pair of arginine residues at positions 91 and 92 that are also characteristic of the preferred phosphorylation substrates of PKAc. The initial docking of PKAc with PKAr is achieved by attachment of the PKAc substrate binding site to the PKAr pseudosubstrate inhibitory site. The specificity of this recognition is strongly dependent on the arginine residues at positions 91 and 92. Replacing these residues with alanine increases the K_d(app) for PKAc and PKAr by more than three orders of magnitude in the presence of MgATP (BUECHLER et al. 1993 ; POTEET-SMITH et al. 1997 ). These replacements also interfere with the binding of PKAr to D-AKAP (BANKY et al. 1998 ). The holoenzyme is highly allosteric; binding of four cAMP molecules in domains A and B is required to dissociate the PKAc subunits from PKAr. PKAr* is mutant at the cAMP binding sites in domains A and B, as indicated in Fig 1, making its binding to PKAc insensitive to cAMP.

PKAr* must bind PKAc to activate Hh target genes:
To test the hypothesis that the Hh signaling activity of PKAr* might be due to a property other than its ability to inhibit PKAc, we mutated the arginines of the inhibitory domain to glycines (Fig 1) and constructed a number of UAS-PKAr*G^91,92 transgenic strains. These strains were tested for their ability to express the transgene using apterous-GAL4 as a driver by staining for HA-tagged PKAr*G^91,92 in wing discs. Four of five strains tested clearly showed staining specifically in the dorsal compartment of the wing disc, as expected for apterous expression (Fig 2).

View larger version (99K): In this window In a new window Download PPT slide   Figure 2. Immunostaining of HA-tagged PKAr*G91,92 in wing imaginal discs. The anterior compartment is to the left, and the wing pouch is to the bottom. (A) The apterous-GAL4 transgene driving expression of UAS-PKAr*BDK33. The ventral compartment of the disc is not stained (extreme bottom near the bar). Note the outgrowth of the anterior compartment caused by PKAr*. (B) The apterous-GAL4 transgene driving expression of UAS-PKAr*GG 24.1. Note the ventral compartment is not stained and the absence of an outgrowth. Bar, 0.10 mm.

These four strains were then crossed to a panel of GAL4 enhancer trap strains previously identified (KIGER et al. 1999 ). Table 1 compares the results obtained for UAS-PKAr* and UAS-PKAr*G^91,92 strains crossed to four of these GAL4 strains. In these crosses one or the other strain was heterozygous for a balancer chromosome and the other strain was homozygous. Thus one-half of all the progeny will carry both the GAL4 and UAS transgenes; the other half will carry a balancer chromosome. The data demonstrate thatUAS-PKAr* transgenes cause complete lethality when expressed by all four GAL4 strains. Death occurs at embryonic or larval stages as previously described (KIGER et al. 1999 ). In contrast, flies carrying both UAS-PKAr*G^91,91 and GAL4 transgenes survive in expected numbers and appear completely normal.

The four UAS-PKAr*G^91,92 strains were also crossed to GAL4 enhancer trap strains known to cause adult or pharate adult flies expressing UAS-PKAr* to exhibit evidence of Hh target gene activation, i.e., wing duplications (GAL4-30A and GAL4-RZ4) or knobby, truncated legs with ectopic bristle columns (GAL4-1J3) (LI et al. 1995 ; WOLFGANG et al. 1996 ; KIGER et al. 1999 ). In all cases flies carrying both GAL4 and UAS-PKAr*G^91,92 transgenes emerged in expected numbers and were completely normal (compare A, B, and C in Fig 3). Thus, the effects caused by PKAr* are dependent on the integrity of arginines 91 and 92. As mentioned above, these residues are required both for direct binding of PKAc and for binding of the dimerization domain to D-AKAPs.

View larger version (67K): In this window In a new window Download PPT slide   Figure 3. Effects of UAS transgenes driven by GAL4-RZ4. (A) A normal wing. (B) UAS-PKAr*BDK33 wing shows the effect of ectopic Hh signaling. (C) UAS-PKAr*GG 24.1 wing is normal. (D) UAS-PKAr*21.4 wing is similar to a wing exhibiting ectopic Hh signaling. (E) UAS-PKAcA75F 46.7 wing exhibits the anterior outgrowth typical of ectopic Hh signaling but has a mass of vein material replacing wing blade in most of the duplicated region. (F) UAS-GFP in a wing imaginal disc. Anterior is to the top, and the wing pouch is to the right. GFP expression is typically strongest at the anterior edge of the wing pouch. Bar in A is 0.40 mm and bar in F is 0.10 mm.

We deleted the dimerization domain of PKAr* as indicated in Fig 1 and created six transgenic strains designated UAS-PKAr*. These strains were tested for activity by crossing to GAL4 strains, and the results are shown in Table 1. Five strains produce no adult progeny like the UAS-PKAr* strains employed. Death occurs either in the embryonic or larval stages, depending upon the UAS-PKAr* strain. On the other hand, UAS-PKAr* 21.3 produces significant numbers of progeny with GAL4-KO5 and GAL4-RK5. These progeny have an abnormal sheen to their cuticle, melanization of the thorax, and crossed scutellar bristles similar to those phenotypes previously described for UAS-PKAr* driven by these GAL4 strains (KIGER et al. 1999 ). When UAS-PKAr* strains are crossed to GAL4-RZ4, all (except for UAS-PKAr* 21.3) allow very few or no adult progeny to emerge. The few surviving flies exhibit wing duplications similar to those produced by UAS-PKAr* (compare B and D in Fig 3). UAS-PKAr* 21.3/GAL4-RZ4 adults emerge in significant numbers, and 50% of the females have a very weak expansion of the anterior wing margin. UAS-PKAr* strains crossed to GAL4-1J3 do not produce pharate adults except for UAS-PKAr* 21.3; these pharate adults exhibit knobby legs like those produced by UAS-PKAr*. Thus, UAS-PKAr* andUAS-PKAr* produce essentially the same effects on Hh target gene expression and differ only in the strength of expression of particular transgenes. Because the dimerization domain is not required for high affinity binding of PKAr to PKAc (HERBERG et al. 1994 ; WEN et al. 1995 ; POTEET-SMITH et al. 1997 ), we conclude that PKAr* must bind PKAc to activate Hh target genes.

Catalytically impaired PKAc elicits Hh target gene expression:
In the course of studies to determine whether the effects of ectopic PKAc expression are due to its catalytic activity, we constructed transgenes that express a PKAc in which the active site lysine⁷⁵ is mutated to alanine. This lysine is invariant in the protein kinase family where it assists in binding the terminal phosphate of MgATP at the catalytic site. Replacements at this site in many protein kinases abolish catalytic and biological activity. Replacing this lysine with alanine in Saccharomyces cerevisiae cAMP-dependent protein kinase causes an 844-fold decrease in k_cat, but does not completely inactivate the enzyme in vivo (GIBBS and ZOLLER 1991 ).

These transgenes, designated UAS-PKAcA⁷⁵F, were tested for activity by crossing to the four GAL4 strains used in Table 1 and previously used to characterize wild-type UAS-PKAcF transgenes (KIGER et al. 1999 ). We expected that this replacement would abolish or ameliorate the effects of PKAc expression. Surprisingly, PKAcA⁷⁵ has strong phenotypic effects; however, it produces effects quite different from those observed for wild-type PKAc. The data on adult survival for five UAS-PKAcA⁷⁵F strains are shown in Table 2. Each GAL4 strain exhibits a characteristic pattern of abnormalities caused by PKAcA⁷⁵; these abnormalities will be compared to those produced by wild-type PKAc (KIGER et al. 1999 ).

GAL4-JW1: PKAcA⁷⁵ expressed by GAL4-JW1 causes death at the pupal stage except for UAS-PKAcA⁷⁵F10.4. With this strain many adults cannot free themselves from the pupal case. Of those that do, a few have wings with large anterior duplications (Fig 4A), others have large wing masses enclosed in the thorax (the duplications are too large to permit the wing blade to evert), and the remainder have duplications of the notum replacing wings. In contrast, depending upon the specific transgene, PKAc causes death prior to the pupal stage at either embryonic, first, or second instar stages.

View larger version (60K): In this window In a new window Download PPT slide   Figure 4. UAS-PKAcA75F transgenes cause ectopic Hh signaling in the anterior compartment of the wing when driven by three GAL4 transgenes that are expressed in the anterior compartment of the wing pouch. GAL4-JW1 drives UAS-PKAcA75F10.4 (A) and UAS-GFP (D). GAL4-KO5 drives UAS-PKAcA75F10.1 (B) and UAS-GFP (E). GAL4-PP3 drives UAS-PKAcA75F10.4 (C) and UAS-GFP (F). Bar in C is 0.40 mm and bar in F is 0.10 mm.

GAL4-KO5: PKAcA⁷⁵ expressed by GAL4-KO5 has no effect on viability. Adults emerge in expected numbers and are normal (UAS-PKAcA⁷⁵F30.1 and 10.4) or exhibit small wing outgrowths from the medial costal region of the anterior wing (UAS-PKAcA⁷⁵F46.7; 68.2 and 10.1; Fig 4B). In contrast, depending upon the specific transgene, PKAc is almost invariably lethal: larvae fail to grow after hatching; pupae melanize and die; pharate adults melanize and die; or many fewer than expected numbers of adults emerge who are unable to open their wings completely.

GAL4-PP3: PKAcA⁷⁵ expressed by GAL4-PP3 causes death at the pupal or pharate adult stage except for UAS-PKAcA⁷⁵F10.4, which causes spread wings and small outgrowths of the proximal costal region of the anterior wing (Fig 4C) in some, otherwise normal, adults. In contrast, PKAc causes larvae to fail to grow after hatching, causes death during larval stages, or causes pupae to blacken and die, depending upon the specific transgene.

GAL4-RK5: PKAcA⁷⁵ expressed by GAL4-RK5 has no effect on viability and all adults are normal. PKAc, on the other hand, generally causes larval death, retarded growth, and small pupae that blacken and die. However, weak transgenes do allow some adults to emerge exhibiting a variety of defects (KIGER et al. 1999 ).

GAL4-RZ4: Crossed to GAL4-RZ4, UAS-PKAcA⁷⁵F transgenes have variable effects on viability. Adults that emerge exhibit large anterior expansions of the wings similar in shape to those caused by PKAr*. However, for most transgenes the duplicated portions of the wings are often blistered and/or have a confused mass of enlarged veins in contrast to the duplications produced by PKAr* (compare B and E in Fig 3). PKAcA⁷⁵ generally causes more extreme deviations from normal wing development than does PKAr*. The nature of this difference was explored as follows.

PKAcA⁷⁵ expression by GAL4-RZ4 has deleterious effects on viability and fertility. However, we have been able to create chromosomes carrying both GAL4-RZ4 and the relatively weak UAS-PKAcA⁷⁵F30.1 transgene. This allows the effects of one and two doses of UAS-PKAcA⁷⁵F30.1 on wing phenotype to be compared. Flies of genotype GAL4-RZ4, UAS-PKAcA⁷⁵F30.1/+ show an effect similar to or weaker than that caused by the example for PKAr* (see the wing in Fig 3B) in 61/88 individuals. In contrast, flies of genotype GAL4-RZ4, UAS-PKAcA⁷⁵F30.1/UAS-PKAcA⁷⁵F30.1 show an effect similar to or stronger than that caused by the example for PKAcA⁷⁵ (see the wing in Fig 3E) in 115/120 individuals. Therefore, the differences in the phenotypes caused by PKAr* and PKAcA⁷⁵ would appear to be of a quantitative nature rather than a qualitative one.

In summary, impairing the catalytic activity of PKAc blocks the phenotypic effects that are observed when wild-type PKAc is expressed ectopically and creates, instead, a new set of effects. Such effects are not caused by ectopic expression of another mutant form, PKAcR²²⁴, which has no phenotypic effect when expressed by the above (or other) GAL4 transgenes (KIGER et al. 1999 ; our unpublished data).

A null allele of ci suppresses the effects of PKAcA⁷⁵:
PKAr* driven by GAL4-30A causes the formation in the costal region of a small anterior mirror-image wing duplication similar to duplications caused by ectopic En expression driven by GAL4-30A (GUILLEN et al. 1995 ). The size of the duplication varies between individuals, PKAr* BDK 35 producing a distribution of larger sizes than that produced by PKAr* BDK 33. Examples of the nonoverlapping extremes of the two distributions are shown in Fig 5A and Fig B.

View larger version (90K): In this window In a new window Download PPT slide   Figure 5. Effects of UAS transgenes driven by GAL4-30A. (A) UAS-PKAr*BDK33. (B) UAS-PKAr*BDK35. (C) UAS-PKAcA75F10.1. (D) UAS-PKAcA75F10.1 and UAS-PKI(1-31) F 5-1. Bar in A is 0.40 mm.

When UAS-PKAcA⁷⁵F 10.1 or 30.1 transgenes are expressed by GAL4-30A, the flies are viable and fertile and some of their wings exhibit a small mirror-image duplication similar to duplications caused by PKAr*. These transgenes are located on the second autosome, permitting single chromosomes carrying both GAL4-30A and UAS-PKAcA⁷⁵F to be created by meiotic recombination and preserved in balanced stocks. We have used these chromosomes, as well as two independently derived GAL4-RZ4, PKAcA⁷⁵F 30.1 chromosomes described in the previous section, to examine how PKAcA⁷⁵ produces its effects.

Because the effects of PKAcA⁷⁵ on wing development resemble those of PKAr*, we tested the effect of ci⁹⁴, a null allele of ci (METHOT and BASLER 1999 ). Females carrying GAL4, UAS-PKAcA⁷⁵F chromosomes were crossed to ci⁹⁴/ey^D males; the frequencies of progeny flies with wing duplications are recorded in Table 3. Heterozygosity for ci⁹⁴ causes strong suppression of the effects caused by PKAcA⁷⁵ expression. We conclude that the effect of PKAcA⁷⁵ on wing development is mediated by an increase in the level of Ci-155, and is therefore a consequence of Hh target gene activation.

PKI(1-31) enhances the effects of PKAcA⁷⁵:
Flies carrying GAL4-30A, UAS-PKAcA⁷⁵F chromosomes were crossed to flies carrying a UAS-PKI(1-31)F transgene to test for interactions. For two UAS-PKAcA⁷⁵F transgenes in two different genetic backgrounds, PKI(1-31) markedly enhances the frequency of wings with duplications (Table 4). In addition to enhancing the frequency of duplications, the sizes of the duplications caused by PKAcA⁷⁵ are increased by PKI(1-31). Examples of the nonoverlapping extremes of the size distributions observed for PKAcA⁷⁵F 10.1 and for PKAcA⁷⁵F 10.1 coexpressed with PKI (1-31)F 5-1 are shown in Fig 5C and Fig D.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Dominant negative effects of PKAcA⁷⁵:
GAL4-driven expression of wild-type PKAc produces a number of distinct phenotypes depending upon where a particular GAL4 strain is expressed (KIGER et al. 1999 ; our unpublished studies). We expected that these phenotypes would be absent, or strongly ameliorated, when PKAcA⁷⁵ was expressed, and this expectation has been met (particularly for GAL4-RK5) as the presentation of the studies outlined in Table 2 indicates. The additional effects of PKAcA⁷⁵ were quite unanticipated. In cases where the adult fly is defective, the phenotypes are similar to those caused by ectopic Hh target gene expression. These effects are suppressed by halving the ci⁺ gene dosage, directly implicating Ci-155.

Because PKAcA⁷⁵ and PKAr* have similar effects, PKAcA⁷⁵ must be inhibiting the action of endogenous wild-type PKAc in repressing Hh target gene expression. In other words, the UAS-PKAcA⁷⁵ F transgene is acting as a dominant negative mutant of DC0. HERSKOWITZ 1987 has examined in detail the ways in which overproducing a mutant protein might cause functional inactivation of a normal gene.

The more extreme effects (masses of enlarged veins) produced by two doses of UAS-PKAcA⁷⁵F 30.1 and by other UAS-PKAcA⁷⁵F transgenes must be due to greater inhibition of endogenous PKAc activity by PKAcA⁷⁵ than is achieved by PKAr*. Overexpressed PKAcA⁷⁵ is evidently competing with endogenous PKAc in a way that blocks phosphorylation of Ci-155. This competition must be specific to phosphorylation of Ci-155. If it were not, we might expect PKAcA⁷⁵ to compete for all PKAc substrates and to mimic phenotypes previously found to be caused by PKI(1-31), as well as those effects on larval growth caused by PKAr*, which it does not do (KIGER et al. 1999 ). This result is even more remarkable since the effect that might be most expected is that PKAcA⁷⁵ would bind PKAr and displace active PKAc from the holoenzyme, leading to an increase in free PKAc activity.

This unexpected behavior can be understood in part by considering what is known about the kinetic mechanism of PKAc and its interaction with one of its two substrates, MgATP. Binding of MgATP to free PKAc induces a conformational change that enables PKAc·MgATP to bind strongly to either PKI(1-31) or to Kemptide (a synthetic peptide substrate used in most enzymatic studies of PKAc) (WHITEHOUSE and WALSH 1983 ; WHITEHOUSE et al. 1983 ). Thus, PKAcA⁷⁵ evidently must not be able to recognize and bind to substrates that PKAc normally recognizes and phosphorylates. In addition, PKAcA⁷⁵ is probably not able to form a holoenzyme with PKAr type I, but it could do so with PKAr type II and displace active PKAc as mentioned above (HERBERG and TAYLOR 1993 ). How then does PKAcA⁷⁵ specifically block phosphorylation of Ci-155 by endogenous PKAc?

We propose that PKAc involved in Ci-155 phosphorylation is locked in close proximity to Ci-155 as part of a complex. Cells must have a limited number of these complexes for competition between PKAcA⁷⁵ and PKAc to be effective. Supporting this proposal are the properties of a dominant negative mutation of the DC0 gene, Pka^DN (PAN and RUBIN 1995 ). The phenotype of Pka^DN is enhanced by a weak loss-of-function DC0 allele and suppressed by a DC0⁺ transgene, suggesting competition between normal and mutant PKAc molecules. Like PKAcA⁷⁵, Pka^DN evidently must have impaired catalytic activity.

In the absence of cAMP, PKAc is generally believed to be bound to PKAr (TAYLOR et al. 1990 ), although other inhibitory proteins have recently been described (ZHONG et al. 1997 ; RAZANI et al. 1999 ). The fact that ectopic expression of PKI(1-31) does not induce Hh target gene expression may indicate that active free PKAc is not present at a level sufficient to promote Ci-155 proteolysis. Therefore, it may be that PKAc enters its complex with Ci-155 bound to an inhibitory protein rather than free and accessible to inhibition by ectopic PKI(1-31). The ability of PKAcA⁷⁵ to enter the complex by competing with endogenous PKAc could be due to the fact that this replacement at position 75 does not completely inactivate the enzyme (as for the S. cerevisiae enzyme), allowing it to maintain a conformation that can bind to the inhibitory protein. PKAcR²²⁴, on the other hand, is phenotypically inactive compared to both PKAc (KIGER et al. 1999 ) and to PKAcA⁷⁵ (data not shown), indicating that it is unable to compete with endogenous PKAc.

The existence of a PKAc/Ci complex casts new light on previous observations. Just as PKI(1-31) is excluded from the complex for steric reasons, free PKAc*, expressed at low levels, would be excluded from the complex and would not be able to block the effect of normal Hh signaling on target gene expression (JIANG and STRUHL 1995 ; LI et al. 1995 ). Higher levels of ectopic PKAc*, on the other hand, would cause phosphorylation of Ci-155 independently of the complex and lead to proteolysis (LI et al. 1995 ; WANG et al. 1999 ). The requirement for high level expression of PKAc* to phosphorylate Ci-155 indicates that normally the complex itself is required to facilitate phosphorylation, supporting the idea that active free PKAc is not present at a level sufficient to promote Ci-155 proteolysis. Phosphorylation of a number of sites on Ci-155 is required for its proteolysis (Y. CHEN et al. 1998 , Y. CHEN et al. 1999 ; C.-H. CHEN et al. 1999 ; PRICE and KALDERON 1999 ; WANG et al. 1999 ; METHOT and BASLER 2000 ), suggesting that phosphorylation could be processive within a complex. If so, Ci-155 must be able to rotate to access the active site of the bound PKAc and other components of the complex must be required to hold it in place. Such steric constraints could prevent PKI(1-31) from finding its PKAc binding site, explaining the inability of this inhibitor to induce Hh target gene expression.

The PKAc/Ci complex:
Determining what other components make up the PKAc/Ci complex will be an important step in understanding Hh signal transduction. Since most cytoplasmic Ci-155 is bound in a Ci/Cos2/Fu complex, PKAc must be part of this complex or part of another complex that is closely tied to it. The association of PKAr type I and PKAc in vitro is exquisitely balanced by the binding of MgATP to the holoenzyme at the PKAc catalytic site. A decrease in MgATP concentration leads to rapid dissociation of the holoenzyme (HERBERG and TAYLOR 1993 ). Since PKAcA⁷⁵ is mutant at a residue required for binding of MgATP, it seems probable that PKAcA⁷⁵ is unable to form a holoenzyme with PKAr type I, just as it is unable to block phosphorylation of most PKAc substrates. Indeed, the fact that PKAr*, a type I subunit, is less effective than PKAcA⁷⁵ in inducing Hh target genes could be due to PKAr type I not being part of the PKA/Ci complex. On the other hand, binding of PKAr type II to PKAc does not require MgATP (HERBERG and TAYLOR 1993 ), suggesting that PKAcA⁷⁵may enter the complex bound to PKAr type II. If PKAr type II is part of the complex, then the presence of an A-kinase anchor protein (AKAP) might also be expected. At least five AKAPs have been described in Drosophila that tether PKA type II to plasma membranes or the cytoskeleton (HAN et al. 1997 ; LI et al. 1999 ). It is also possible that an inhibitor protein might escort PKAc to the complex but not be part of the complex.

Knowing the components of the complex will permit choices to be made between possible models. In one model, in the absence of a Hh signal, ambient cAMP level in the complex would be high enough to dissociate PKAr and PKAc and permit processive phosphorylation of Ci-155. We envision that dissociated PKAr and PKAc would remain bound within the complex. A Hh signal then might act to reduce cAMP level and allow bound PKAc and bound PKAr to reassociate, preserving Ci-155 from phosphorylation. In another model, active PKAc may be bound within the complex, and a Hh signal might activate a phosphatase to convert phosphorylated Ci-155 into a transcriptional activator (C.-H. CHEN et al. 1999 ). Whether or not cAMP is involved in Hh signaling remains an open question.

Synergistic effects of PKI(1-31) and PKAcA⁷⁵ on Ci-155 activity:
An uncleavable activator form of Ci-155 (Ci^U) has been shown to be inactivated by PKAc through phosphorylation of the same residues that induce proteolysis of Ci-155, demonstrating two discrete roles for PKAc in regulating the active forms of Ci (WANG et al. 1999 ). Our observations would support this conclusion. PKAcA⁷⁵ evidently blocks phosphorylation of Ci-155 within the complex preventing proteolysis. Unphosphorylated C1-155 must leave the complex and enter the nucleus to induce Hh target genes. The synergistic effect of PKI(1-31) suggests that after leaving the complex some unphosphorylated Ci-155 may be phosphorylated by free PKAc and inactivated, reducing the level of Hh target gene induction. This synergism indicates that a basal level of free PKAc does exist but that it is too low to cause proteolysis of Ci-155. Proteolysis must require highly phosphorylated Ci-155 created by processive phosphorylation within the complex, and it may be that the complex evolved to assure that processive phosphorylation would occur.

Different effects of PKAr* and PKI(1-31) on Hh target genes:
We initiated these studies to determine how PKAr* induces Hh target gene expression while PKI(1-31) does not. We have tested the hypothesis that free PKAr* might induce Hh target genes independently of its ability to bind free PKAc. Our results indicate the contrary; the activity of PKAr* is due to its ability to bind and inhibit endogenous PKAc catalytic activity. The important differences between PKAr* and PKI(1-31) in activating Hh target genes can now be understood. Since PKAr* is insensitive to cAMP, it will compete with endogenous PKAr for PKAc; most PKAc will eventually be bound by the relatively high concentration of ectopic PKAr* rather than PKAr. Therefore, ectopic PKAr* forms a sink for PKAc, mimicking the loss of PKAc in a clone of DC0 mutant cells. In contrast, PKI(1-31) cannot compete with PKAr in binding PKAc (HERBERG and TAYLOR 1993 ).

	ACKNOWLEDGMENTS

We thank Robert Holmgren for the ci⁹⁴ strain. We are grateful to Dan Kalderon for discussions and for the cDNA clone of PKAr*. This manuscript benefited greatly from the comments of two anonymous reviewers. This work was supported by funds of the Agricultural Experiment Station at UC Davis.

Manuscript received March 2, 2001; Accepted for publication April 18, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AZA-BLANC, P. and T. B. KORNBERG, 1999  Ci, a complex transducer of the Hedgehog signal. Trends Genet. 15:458-462[Medline].

AZA-BLANC, P., F.-A. RAMÍREZ-WEBER, M.-P. LAGET, C. SCHWARTZ, and T. B. KORNBERG, 1997  Proteolysis that is inhibited by Hedgehog targets Cubitus interruptus protein to the nucleus and converts it to a repressor. Cell 89:1043-1053[Medline].

BANKY, P., L. J.-S. HUANG, and S. S. TAYLOR, 1998  Dimerization/docking domain of the type I regulatory subunit of cAMP-dependent protein kinase. J. Biol. Chem. 273:35048-35055[Abstract/Free Full Text].

BASLER, K. and G. STRUHL, 1994  Compartment boundaries and the control of Drosophila limb pattern by hedgehog protein. Nature 368:208-214[Medline].

BRAND, A. H. and N. PERRIMON, 1993  Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118:401-415[Abstract].

BUECHLER, Y. J., F. W. HERBERG, and S. S. TAYLOR, 1993  Regulation-defective mutants of type I cAMP-dependent protein kinase. J. Biol. Chem. 268:16495-16503[Abstract/Free Full Text].

CASSO, D., F.-A. RAMÍREZ-WEBER, and T. B. KORNBERG, 1999  GFP-tagged balancer chromosomes for Drosophila melanogaster.. Mech. Dev. 88:229-232[Medline].

CHEN, C.-H., D. P. VON KESSLER, W. PARK, B. WANG, and Y. MA et al., 1999  Nuclear trafficking of Cubitus interruptus in the transcriptional regulation of Hedgehog target gene expression. Cell 98:305-316[Medline].

CHEN, Y. and G. STRUHL, 1996  Dual roles for Patched in sequestering and transducing Hedgehog. Cell 87:553-563[Medline].

CHEN, Y., N. GALLAHER, R. H. GOODMAN, and S. M. SMOLIK, 1998  Protein kinase A directly regulates the activity and proteolysis of cubitus interruptus. Proc. Natl. Acad. Sci. USA 95:2349-2354[Abstract/Free Full Text].

CHEN, Y., J.-R. CARDINAUX, R. H. GOODMAN, and S. M. SMOLIK, 1999  Mutants of cubitus interruptus that are independent of PKA regulation are independent of hedgehog signaling. Development 126:3607-3616[Abstract].

GIBBS, C. S. and M. J. ZOLLER, 1991  Rational scanning mutagenesis of a protein kinase identifies functional regions involved in catalysis and substrate interactions. J. Biol. Chem. 266:8923-8931[Abstract/Free Full Text].

GUILLÉN, I., J. L. MULLOR, J. CAPDEVILA, E. SÁNCHEZ-HERRERO, and G. MORATA, 1995  The function of engrailed and the specification of Drosophila wing pattern. Development 121:3447-3456[Abstract].

HAN, J.-D., N. E. BAKER, and C. S. RUBIN, 1997  Molecular characterization of a novel A Kinase Anchor Protein from Drosophila melanogaster.. J. Biol. Chem. 272:26611-26619[Abstract/Free Full Text].

HERBERG, F. W. and S. S. TAYLOR, 1993  Physiological inhibitors of the catalytic subunit of cAMP-dependent protein kinase: effect of MgATP on protein-protein interactions. Biochemistry 32:14015-14022[Medline].

HERBERG, F. W., W. R. DOSTMANN, M. ZORN, S. J. DAVIS, and S. S. TAYLOR, 1994  Crosstalk between domains in the regulatory subunit of cAMP-dependent protein kinase: influence of amino terminus on cAMP binding and holoenzyme formation. Biochemistry 33:7485-7494[Medline].

HERSKOWITZ, I., 1987  Functional inactivation of genes by dominant negative mutations. Nature 329:219-222[Medline].

INGHAM, P. W., 1998  Transducing Hedgehog: the story so far. EMBO J. 17:3505-3511[Medline].

JIANG, J. and G. STRUHL, 1995  Protein kinase A and Hedgehog signalling in Drosophila limb development. Cell 80:563-572[Medline].

JIANG, J. and G. STRUHL, 1998  Regulation of the Hedgehog and Wingless signaling pathways by the F-box/WD40-repeat protein Slimb. Nature 391:493-496[Medline].

KALDERON, D., 2000  Transducing the Hedgehog signal. Cell 103:371-374[Medline].

KALDERON, D. and G. M. RUBIN, 1988  Isolation and characterization of Drosophila cAMP-dependent protein kinase genes. Genes Dev. 2:1539-1556[Abstract/Free Full Text].

KIGER, J. A., JR., J. L. EKLUND, S. H. YOUNGER, and C. J. O'KANE, 1999  Transgenic inhibitors identify two roles for Protein Kinase A in Drosophila development. Genetics 152:281-290[Abstract/Free Full Text].

LEPAGE, T., S. COHEN, F. J. DIAZ-BENJUMEA, and S. M. PARKHURST, 1995  Signal transduction by cAMP-dependent protein kinase A in Drosophila limb patterning. Nature 373:711-715[Medline].

LI, W., J. T. OHLMEYER, M. E. LANE, and D. KALDERON, 1995  Function of protein kinase A in Hedgehog signal transduction and Drosophila imaginal disc development. Cell 80:553-562[Medline].

LI, Z., E. A. ROSSI, J. D. HOHEISEL, D. KALDERON, and C. S. RUBIN, 1999  Generation of a novel A Kinase Anchor Protein and a myristoylated alanine-rich C Kinase substrate-like analog from a single gene. J. Biol. Chem. 274:27191-27200[Abstract/Free Full Text].

MCMAHON, A. P., 2000  More surprises in the hedgehog signaling pathway. Cell 100:185-188[Medline].

MÉTHOT, N. and K. BASLER, 1999  Hedgehog controls limb development by regulating the activities of distinct transcriptional and repressor forms of Cubitus interruptus. Cell 96:819-831[Medline].

MÉTHOT, N. and K. BASLER, 2000  Suppressor of Fused opposes Hedgehog signal transduction by impeding nuclear accumulation of the activator form of Cubitus interruptus. Development 127:4001-4010[Abstract].

MONNIER, V., F. DUSSILLOL, G. ALVES, C. LAMOUR-ISNARD, and A. PLESSIS, 1998  Suppressor of fused links Fused and Cubitus interruptus on the Hedgehog signaling pathway. Curr. Biol. 8:583-586[Medline].

MURONE, M., A. ROSENTHAL, and F. J. DE SAUVAGE, 1999  Sonic hedgehog signaling by the Patched-Smoothened receptor complex. Curr. Biol. 9:76-84[Medline].

OHLMEYER, J. T. and D. KALDERON, 1997  Dual pathways for induction of wingless expression by protein kinase A and Hedgehog in Drosophila embryos. Genes Dev. 11:2250-2258[Abstract/Free Full Text].

OHLMEYER, J. T. and D. KALDERON, 1998  Hedgehog stimulates maturation of Cubitus interruptus into a labile transcriptional activator. Nature 396:749-753[Medline].

PAN, D. and G. M. RUBIN, 1995  Protein kinase A and hedgehog act antagonistically in regulating decapentaplegic transcription in Drosophila imaginal discs. Cell 80:543-552[Medline].

PATEL, N. H., 1994  Imaging neuronal subsets and other cell types in whole-mount Drosophila embryos and larvae using antibody probes. Methods Cell Biol. 44:445-487[Medline].

POTEET-SMITH, C. E., J. B. SHABB, S. H. FRANCIS, and J. D. CORBIN, 1997  Identification of critical determinants for autoinhibition in the pseudosubstrate region of type I cAMP-dependent protein kinase. J. Biol. Chem. 272:379-388[Abstract/Free Full Text].

PRICE, M. A. and D. KALDERON, 1999  Proteolysis of Cubitus interruptus in Drosophila requires phosphorylation by Protein Kinase A. Development 126:4331-4339[Abstract].

RAZANI, B., C. S. RUBIN, and M. P. LISANTI, 1999  Regulation of cAMP-mediated signal transduction via interaction of caveolins with the catalytic subunit of protein kinase A. J. Biol. Chem. 274:26353-26360[Abstract/Free Full Text].

ROBBINS, D. J., K. E. NYBAKKEN, R. KOBAYASHI, J. C. SISSON, and J. M. BISHOP et al., 1997  Hedgehog elicits signal transduction by means of a large complex containing the kinesin-related protein Costal2. Cell 90:225-234[Medline].

SHAULSKY, G., D. FULLER, and W. F. LOOMIS, 1998  A cAMP-phosphodiesterase controls PKA-dependent differentiation. Development 125:691-699[Abstract].

SISSON, J. C., K. S. HO, K. SUYAMA, and M. P. SCOTT, 1997  Costal2, a novel kinesin-related protein in the Hedgehog signaling pathway. Cell 90:235-245[Medline].

SPENCER, E., J. JIANG, and Z. J. CHEN, 1999  Signal-induced ubiquitination of IB by the F-box protein Slimb/ß-TrCP. Genes Dev. 13:284-294[Abstract/Free Full Text].

STRIGINI, M. and S. M. COHEN, 1997  A Hedgehog activity gradient contributes to AP axial patterning of the Drosophila wing. Development 124:4697-4705[Abstract].

STRUTT, D. I., V. WIERSDORFF, and M. MLODZIK, 1995  Regulation of furrow progression in the Drosophila eye by cAMP-dependent protein kinase A. Nature 373:705-709[Medline].

TABATA, T. and T. B. KORNBERG, 1994  Hedgehog is a signaling protein with a key role in patterning Drosophila imaginal discs. Cell 76:89-102[Medline].

TAYLOR, S. S., J. A. BUECHLER, and W. YONEMOTO, 1990  cAMP-dependent protein kinase: framework for a diverse family of regulatory enzymes. Annu. Rev. Biochem. 59:971-1005[Medline].

WANG, G., B. WANG, and J. JIANG, 1999  Protein kinase A antagonizes Hedgehog signaling by regulating both the activator and repressor forms of Cubitus interruptus. Genes Dev. 13:2828-2837[Abstract/Free Full Text].

WANG, Q. T. and R. A. HOLMGREN, 1999  The subcellular localization and activity of Drosophila Cubitus interruptus are regulated at multiple levels. Development 126:5097-5106[Abstract].

WEN, W., J. L. MEINKOTH, R. Y. TSIEN, and S. S. TAYLOR, 1995  Identification of a signal for rapid export of proteins from the nucleus. Cell 82:463-473[Medline].

WHITEHOUSE, S. and D. A. WALSH, 1983  Mg^.ATP^2--dependent interaction of the inhibitor protein of the cAMP-dependent protein kinase with the catalytic subunit. J. Biol. Chem. 258:3682-3692[Abstract/Free Full Text].

WHITEHOUSE, S., J. R. FERAMISCO, J. E. CASNELLIE, E. G. KREBS, and D. A. WALSH, 1983  Studies on the kinetic mechanism of the catalytic subunit of the cAMP-dependent protein kinase. J. Biol. Chem. 258:3693-3701[Abstract/Free Full Text].

WOLFGANG, W. J., I. J. H. ROBERTS, F. QUAN, C. O'KANE, and M. FORTE, 1996  Activation of protein kinase A-independent pathways by Gs in Drosophila. Proc. Natl. Acad. Sci. USA 93:14542-14547[Abstract/Free Full Text].

ZHONG, H., H. SUYANG, H. ERDJUMENT-BROMAGE, P. TEMPST, and S. GHOSH, 1997  The transcriptional activity of NF-kB is regulated by the IkB-associated PKAc subunit through a cyclic-AMP-independent mechanism. Cell 89:413-424[Medline].

This article has been cited by other articles:

				D. Ayaz, M. Leyssen, M. Koch, J. Yan, M. Srahna, V. Sheeba, K. J. Fogle, T. C. Holmes, and B. A. Hassan Axonal Injury and Regeneration in the Adult Brain of Drosophila J. Neurosci., June 4, 2008; 28(23): 6010 - 6021. [Abstract] [Full Text] [PDF]

				H. Luan, W. C. Lemon, N. C. Peabody, J. B. Pohl, P. K. Zelensky, D. Wang, M. N. Nitabach, T. C. Holmes, and B. H. White Functional Dissection of a Neuronal Network Required for Cuticle Tanning and Wing Expansion in Drosophila J. Neurosci., January 11, 2006; 26(2): 573 - 584. [Abstract] [Full Text] [PDF]

				T. Sheng, S. Chi, X. Zhang, and J. Xie Regulation of Gli1 Localization by the cAMP/Protein Kinase A Signaling Axis through a Site Near the Nuclear Localization Signal J. Biol. Chem., January 6, 2006; 281(1): 9 - 12. [Abstract] [Full Text] [PDF]

				L. S. Collier, K. Suyama, J. H. Anderson, and M. P. Scott Drosophila Costal1 Mutations Are Alleles of Protein Kinase A That Modulate Hedgehog Signaling Genetics, June 1, 2004; 167(2): 783 - 796. [Abstract] [Full Text] [PDF]

				J. E. Hooper Smoothened translates Hedgehog levels into distinct responses Development, September 1, 2003; 130(17): 3951 - 3963. [Abstract] [Full Text] [PDF]