The Ca²⁺-Calmodulin-Activated Protein Phosphatase Calcineurin Negatively Regulates Egf Receptor Signaling in Drosophila Development

Corresponding author: Kathleen M. C. Sullivan, Department of Molecular and Cell Biology, University of California, Berkeley, CA 94720., kmcs{at}uclink4.berkeley.edu (E-mail)

Communicating editor: K. GOLIC

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Calcineurin is a Ca²⁺-calmodulin-activated, Ser-Thr protein phosphatase that is essential for the translation of Ca²⁺ signals into changes in cell function and development. We carried out a dominant modifier screen in the Drosophila eye using an activated form of the catalytic subunit to identify new targets, regulators, and functions of calcineurin. An examination of 70,000 mutagenized flies yielded nine specific complementation groups, four that enhanced and five that suppressed the activated calcineurin phenotype. The gene canB2, which encodes the essential regulatory subunit of calcineurin, was identified as a suppressor group, demonstrating that the screen was capable of identifying genes relevant to calcineurin function. We demonstrated that a second suppressor group was sprouty, a negative regulator of receptor tyrosine kinase signaling. Wing and eye phenotypes of ectopic activated calcineurin and genetic interactions with components of signaling pathways suggested a role for calcineurin in repressing Egf receptor/Ras signal transduction. On the basis of our results, we propose that calcineurin, upon activation by Ca²⁺-calmodulin, cooperates with other factors to negatively regulate Egf receptor signaling at the level of sprouty and the GTPase-activating protein Gap1.

THE only protein phosphatase regulated by both Ca²⁺ and calmodulin, calcineurin is a key player in Ca²⁺ signal transduction from yeast to humans and has been implicated in a wide array of processes, from disease progression to development (for a general review, see RUSNAK and MERTZ 2000 ). In the immune system, calcineurin is essential for T-cell activation and is the target of immunosuppressant drugs such as cyclosporin (LIU et al. 1991 ; CLIPSTONE and CRABTREE 1992 ). An abundant neuronal protein, calcineurin has been implicated in various forms of synaptic plasticity (reviewed in YAKEL 1997 ). In addition, calcineurin is involved both in the development of cardiac valves (RANGER et al. 1998 ) and in hypertrophy of cardiac muscle following disease or injury (MOLKENTIN et al. 1998 ).

The enzyme consists of an 60-kD catalytic subunit, calcineurin A (canA), bound to the regulatory subunit, calcineurin B (canB), a 19-kD EF-hand Ca²⁺-binding protein (reviewed in KLEE et al. 1998 ). CanB is essential for phosphatase activity and can be dissociated from canA only by denaturants. CanA has two variable regions at the N and C termini, a highly conserved catalytic domain, and a regulatory region. The regulatory region consists of a binding site for Ca²⁺-calmodulin and a short autoinhibitory domain that blocks substrate access to the active site in the absence of Ca²⁺-calmodulin.

A Ca²⁺-calmodulin-independent, constitutively active phosphatase is made by deleting the canA regulatory region (O'KEEFE et al. 1992 ). Studies from a number of different organisms indicate that, aside from a small degree of Ca²⁺ sensitivity mediated by canB, activated calcineurin functions identically to the full-length, Ca²⁺-calmodulin-activated form (MENDOZA et al. 1996 ; SHIBASAKI et al. 1996 ; WINDER et al. 1998 ).

Calcineurin is activated by a sustained increase in intracellular Ca²⁺ levels that can result from the opening of intracellular Ca²⁺ channels in response to phosphoinositide (PI) signaling (reviewed in BERRIDGE 1993 ). PI signaling is initiated by the activation of a phosphatidylinositol-specific phospholipase C, either PLCß by G-protein-coupled receptors (GPCR) or PLC by receptor tyrosine kinases (RTK). PI-PLCs cleave phosphatidylinositol 4,5-bisphosphate (PIP₂) to yield inositol 1,4,5-trisphosphate (InsP₃), which then activates the InsP₃ receptor Ca²⁺ channel.

GPCRs and RTKs activate an integrated signaling network that includes the Ras/mitogen-activated protein (MAP) kinase cascade, PI3-kinase, and the small GTPase Rho. Depending upon the cellular context, these pathways can antagonize or cooperate with each other and with PI signaling. For example, T-cell activation requires the activation of both NFAT, which is transduced to the nucleus upon dephosphorylation by calcineurin, and AP1, which acts downstream of Ras and MAP kinase (CRABTREE 1999 ).

Conversely, PI signaling has been found to antagonize the Ras pathway in Drosophila. The Egf receptor and Ras/MAP kinase cascade are essential for formation of wing veins and photoreceptor (R) cells in the eye. Mutations in the single phospholipase C gene, small wing (sl), cause the formation of extra R7 cells and wing vein material (THACKERAY et al. 1998 ) and also genetically interact with Egf-receptor-signaling components (THACKERAY et al. 1998 ; POWE et al. 1999 ). A recently proposed model for sl-mediated repression of Egf receptor signaling was based on the identification of the GTPase-activating protein Gap1 as an InsP₄ receptor. PLC-generated InsP₃ is converted to InsP₄, which then activates Gap (POWE et al. 1999 ). Gap converts the active form of Ras, Ras-GTP, to the inactive form, Ras-GDP.

The Drosophila genome contains three canA genes and two canB genes that are 75 and 88% similar to the vertebrate genes, respectively. To date, no mutants have been described for any of the five genes. To study calcineurin function in Drosophila, we expressed a constitutively active form of canA during imaginal development and examined the resulting phenotypes. The activated calcineurin rough eye phenotype was used to perform a genetic modifier screen. We were able to successfully isolate and characterize specific enhancers and suppressors and identified two suppressors as canB2 and sprouty. The activated calcineurin rough eye was also tested extensively for genetic interactions with an array of signaling cascades. Taken together, the genetic evidence is consistent with calcineurin functioning as a negative regulator of Egf receptor/Ras signaling during imaginal development, possibly in the same pathway as PLC.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Stocks:
The following published stocks were used in this study: Egfr^f2 (NUSSLEIN-VOLHARD et al. 1984 ); Egfr^E1 (BAKER and RUBIN 1989 ); drk^e0A, sos^e2H, Ras85D^e1B, and Ras85D^e2f (SIMON et al. 1991 ); Gap1^PB and Gap1^B1 (GAUL et al. 1992 ); sty^D64 and sty^D5 (HACOHEN et al. 1998 ); sl¹ and sl² (THACKERAY et al. 1998 ); pnt^7825D77 and pnt^3680D21 (O'NEILL et al. 1994 ); spi¹ (KLAMBT et al. 1991 ); spi^s3547 (PERRIMON et al. 1996 ); argos⁰⁵⁴⁸⁵ (SCHMUCKER et al. 1997 ); cbl^F165 (PAI et al. 2000 ); dos^2.42 and dos^1.46 (HERBST et al. 1996 ); sos^e46 (YAGI et al. 1998 ); ksr¹⁰⁸⁰, ksr⁶²⁷, phl^1-29, phl^HM7, Dsor1^XS520, rl^s135, rl^XS-324, phyl^s-221, and aop^XE-18 (KARIM et al. 1996 ); sina³ and sina⁴ (CARTHEW and RUBIN 1990 ); phyl²²⁴⁵ (CHANG et al. 1995 ); aop¹ (ROGGE et al. 1995 ); and Rho1⁷²⁰ and Rho1^k07903 (STRUTT et al. 1997 ).

Calcineurin constructs:
The constitutively active form of canA, canA^act, was made from the canA gene Pp2B-14D (BROWN et al. 1994 ) by inserting a stop codon at amino acid residue 456 (out of 570). PCR was carried out using primers that incorporated a 3' stop codon and flanking restriction sites. A full-length Pp2B-14D construct was made by using the same 5' primer and a 3' primer at the end of the open reading frame (ORF). The Pp2B-14D cDNA LD04578 (RUBIN et al. 2000 ) was used as a template for both. Full-length canB was cloned by PCR using wild-type genomic DNA as a template, as the ORF of canB-4F is contained within a single exon. All constructs were confirmed by sequencing. For expression in transgenic animals, the constructs were inserted into pUAST and pGMR, which contains glass response elements (MOSES and RUBIN 1991 ).

Modifier screen:
The canA^act.gl transgene was inserted onto TM3 (TCAG) and CyO (CCAG) by crossing canA^act.gl;TM3 to CyO/Sp;Ki,T^hs.PR2-3. The canA^act.gl/CyO;TM3/Ki,T^hs.PR2-3 progeny were outcrossed to a w¹¹¹⁸ line and individual CyO, or TM3, Ki⁺ males with rough eyes were selected. Upon confirmation that canA^act.gl segregated with TM3 or CyO, TCAG/e ftz ry or CGCA/Adv stocks were established. The presence of canA^act.gl on TM3 was also confirmed by PCR and by Western blot using canA monoclonal antibodies (K. SULLIVAN, unpublished results). TCAGB was made and confirmed similarly, except that canB^gl/CyO;TCAG was used in the starting cross. The isogenic line used in the screen was prepared from a w¹¹¹⁸ stock by using standard procedures (GREENSPAN 1997 ). X-ray and EMS mutagenesis were carried out as described (KARIM et al. 1996 ). Meiotic and deficiency mapping of lethal complementation groups were carried out by using standard procedures (GREENSPAN 1997 ). During meiotic mapping the complementation group alleles were detected in each generation on the basis of their ability to modify TCAG.

Adult eye microscopy:
Scanning electron microscopy (SEM) was carried out on adult flies with an environmental SEM in wet mode. Samples were prepared as described (WOLFF 2000 ), except that the flies were not fixed before dehydration and critical point drying. All micrographs are displayed at the same magnification (x150). Fixing, embedding, and sectioning of adult eyes were done as described (WOLFF 2000 ).

Characterization of the canB2 suppressor group:
Genomic DNA was made from the chromosome 2, suppressor group 1 allele 87 by isolating mutant first instar larvae from a CS2-1⁸⁷/CyO-GFP^arm line (SULLIVAN et al. 2000 ) and from the parent isogenic line. To map the inversion breakpoint by PCR, primers were made at base pair positions -426, +60, and +1582 relative to the start of canB2 transcription.

Sequencing sprouty:
Primers were designed to amplify 700- to 800-bp overlapping fragments that spanned the sty ORF, which is contained within a single exon. Genomic DNA was isolated from the isogenic parent line and from embryos homozygous for the CS3-3 EMS allele ECE(3)518 by using an ECE(3)518/CyO-GFP^arm line. After PCR amplification, the fragments were purified by using the QIAquick PCR purification kit and then directly sequenced. The sty ORF from the isogenic line was identical to the sequence reported in ADAMS et al. 2000 , which differs slightly in the predicted amino acid sequence from that previously published (HACOHEN et al. 1998 ). The EMS allele contained a C-to-T mutation at position 748 relative to the start of the ORF sequence.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Activated calcineurin constructs:
We selected the canA gene Pp2B-14D for our experiments because it is expressed throughout development, including in eye discs (BROWN et al. 1994 ; K. SULLIVAN, unpublished results). The adjacent canA gene at 14F1 (gene designation CG9819; ADAMS et al. 2000 ) encodes a protein that is 83% identical to Pp2B-14D; however, canA-14F is not represented in the expressed sequence tag collection and the expression pattern has not been characterized. The canA gene at 100B4, which was incorrectly localized to 21B, is undetectable by Northern analysis (GUERNINI et al. 1992 ) and appears to have a highly restricted expression pattern (K. SULLIVAN, unpublished results).

An activated form of Pp2B-14D, canA^act, was made by deleting the autoinhibitory and calmodulin-binding domains (O'KEEFE et al. 1992 ; MENDOZA et al. 1996 ). The canA^act construct was expressed in Drosophila under the control of glass response elements, which induce transcription uniformly in cells posterior to the morphogenetic furrow in the eye imaginal disc (MOSES and RUBIN 1991 ).

Flies carrying one copy of the canA^act.gl transgene had mild rough eyes compared to wild type (Fig 1A and Fig B), and the eyes of flies carrying two copies exhibited a stronger phenotype (data not shown). Consistent with observations in other systems, both full-length canA and activated canA without a functional canB-binding domain did not cause any detectable phenotypes when expressed throughout development (data not shown).

View larger version (200K): In this window In a new window Download PPT slide   Figure 1. The canAact.gl rough eye is sensitive to the dose of glass and canB-4F. SEMs are from flies of the following genotypes: (A) w1118; (B and D) TCAG; (C) TCAG/glJ60; (E) Df(1)JC70/Y; TCAG; and (F) TCAGB. (A, B, C, and F) Eyes from males; (D and E) eyes from females. Compared to a normal eye (A), the presence of canAact.gl resulted in a rough eye phenotype (B and D). The rough eye in B is from a line in which the canAact.gl transgene is inserted on the TM3 balancer (TCAG). The TCAG rough eye was significantly suppressed by removing one copy of glass (C). One copy of canB-4F was removed by introducing the deficiency Df(1)JC70, which resulted in suppression of the TCAG rough eye (E). Increasing canB levels by inserting a canBgl transgene onto TCAG (TCAGB) increased the severity of the rough eye (F). The whitish patch near the center of the eye in F is necrotic tissue.

Removing one copy of glass by introducing the null allele gl^60J (MOSES and RUBIN 1991 ) strongly suppressed the canA^act.gl rough eye phenotype (cf. Fig 1C). This demonstrates that the canA^act.gl rough eye phenotype is dependent on glass and is not caused by the insertion site of the transgene or by some other factor.

Reducing the dosage of canB-4F or canB2 by introducing deficiencies that uncover the 4F or 43E genomic region resulted in suppression of the canA^act.gl rough eye (Fig 1E and data not shown). Western blots confirmed that canB protein is present in the eye disc (data not shown); however, it is not known whether the protein is derived from one or both canB genes. Consistent with the effect of reduced canB levels, canB4F^gl, which alone has no phenotype (data not shown), increased the severity of the canA^act.gl rough eye (cf. Fig 1F).

Because expression occurs throughout the later stages of eye development, glass-dependent transgenes can affect many different processes. On the basis of our observations, activated calcineurin may have multiple effects on the differentiation and morphology of photoreceptor and other cell types (data not shown; see also Fig 3F). However, we did not observe an effect of canA^act.gl on cell proliferation or cell death (data not shown).

View larger version (23K): In this window In a new window Download PPT slide   Figure 2. Outline of the canAact.gl dominant modifier screen. The strategy for isolating modifiers of chromosomes 2 and 3 is displayed. Isogenic w1118 males were mutagenized with X rays or EMS and crossed to TCAG/e ftz ry virgin females. The e ftz ry chromosome was used to maintain a stable line of TCAG and to estimate the rate of mutagenesis in the screen. Individual F1 males and females that showed modification of TCAG were backcrossed to the original TCAG/e ftz ry line to determine whether the modification bred true. If all or nearly all TCAG/e+ F2 progeny were modified, the mutation was balanced over TCAG. If no more than half of the TCAG/e+ F2 flies were modified, the mutation was balanced over CCAG (CyO-canAact.gl).

View larger version (134K): In this window In a new window Download PPT slide   Figure 3. Rough eye phenotypes of specific modifier groups isolated in the canAact.gl screen. SEMs and thin sections are from male flies of the following genotypes: (A and E) w1118; (B and F) TCAG; (C and G) TCAG/CE3-3466; (D and H) TCAG/CS3-1465. For comparison, eyes from the parent line are displayed in A and E. The TCAG rough eye (B) had aberrant ommatidial morphology and organization in sections (F). CE3-3 significantly increased the exterior roughness (C) and disruption of ommatidia (G) over TCAG. CS3-1 suppressed both the outer rough eye (D) and the ommatidial defects (H) of TGCA.

Dominant modifier screen:
The canA^act.gl rough eye phenotype was modifiable; i.e., it was sensitive to transgene dose and was specifically modified by canB, which is essential for canA function. Thus, the rough eye was a good candidate for a dominant modifier screen. We prepared an isogenic wild-type stock and inserted canA^act.gl onto the chromosome 3 balancer TM3 (Fig 1B and Fig D), which carries the dominant visible marker Sb. We screened 70,000 progeny of TM3-canA^act.gl (TCAG) females and EMS- or X-ray-treated males (Fig 2). Each individual F₁ with an enhanced or suppressed TCAG rough eye was backcrossed to TCAG to confirm the modification and to determine the chromosomal location. About 21% of the modifiers initially isolated bred true, and stable lines of the confirmed modifiers were established over either TCAG or the chromosome 2 balancer CCAG (CyO-canA^act.gl). Because chromosome 2 balancers, including CyO, harbor one or more suppressors of the TCAG phenotype (data not shown), chromosome 2 suppressors were difficult to balance and are thus underrepresented in the final tally. Modifiers on chromosome 1 were also underrepresented, in part because the balancers carry a dominant eye mutation, Bar, that significantly interfered with scoring TCAG modification. A total of 5 viable and 123 lethal modifiers were isolated in the screen (Table 1).

Of the 123 lethal modifiers, 62 fell into 11 complementation groups and the rest (61) were single hits (Table 2). Examples of the suppression and enhancement of canA^act.gl by the complementation groups are illustrated with SEMs and sagittal eye sections (Fig 3). Compared to TCAG alone (Fig 3B and Fig F), CE3-3 (calcineurin enhancer, chromosome 3, group 3)/TCAG eyes had increased exterior roughness (Fig 3C) and increased disruption of the number and organization of the photoreceptor and support cells (Fig 3G). CS3-1 (calcineurin suppressor, chromosome 3, group 1) was the largest suppressor group, and both SEMs and eye sections of CS3-1/TCAG alleles (Fig 3D and Fig H and data not shown) revealed that, as expected, it reverted the TCAG phenotype toward wild type (Fig 3A and Fig E). Other suppressor groups had similar effects on the TCAG rough eye (Fig 5 and data not shown).

View larger version (15K): In this window In a new window Download PPT slide   Figure 4. Alleles of the suppressor group CS2-1 have lesions in canB2. The genetic lesions of CS2-1128 and CS2-187 in the cytological divisions 43 and 44 on chromosome arm 2R are displayed. The lettered subdivisions of 43 and part of 44 are indicated. The open rectangle below the chromosome subdivisions indicates the region deleted in CS2-1128, and the arrows connected by a line indicate the breakpoints of CS2-187. A close up of the canB2 region shows the right breakpoint of CS2-187, which is within the bracketed region. The arrows above canB2 and cn, the upstream gene, indicate the direction of transcription.

View larger version (167K): In this window In a new window Download PPT slide   Figure 5. Suppressor group CS3-3 is sprouty. SEMs and thin sections are from eyes of the following genotypes: (A and D) TCAG; (B and E) TCAG/CS3-3518; (C) w1118. The TCAG rough eye (A and D) is suppressed by the EMS allele CS3-3518 (B and E). The CS3-3518 sty gene was sequenced and found to have a C-to-T nonsense mutation in the codon corresponding to glutamine 250, which is indicated by an asterisk (F). The position of the mutation is indicated relative to the sprouty homology (shaded) and C-rich regions (solid). The sprouty protein sequence (520-amino-acid residues) is to scale.

Modifier groups that act directly on the glass enhancer do not specifically modify canA^act.gl. Nonspecific groups were identified by testing whether any of the complementation groups modified rough eyes caused by unrelated, glass-dependent transgenes. Two of the enhancer groups, CE3-1 and CE2-1, modified rough eye phenotypes caused by other glass-dependent transgenes, such as sina^gl, a Ras pathway component, and Rho1^gl (Table 2). All complementation groups were additionally tested with Ras^v12, phyllopod, yan^act, and reaper glass-dependent transgenes (data not shown), but only CE3-1 and CE2-1 modified the rough eye phenotypes caused by these transgenes.

We further separated the specific modifier groups into two classes by determining whether they modified the canA^act.gl rough eye phenotype caused by TCAGB (TM3-canA^act.gl,canB^gl). Class I genes, such as Ca²⁺ signaling components or dephosphorylation targets, act downstream of calcineurin and will modify the rough eye phenotype of TCAG and TCAGB. However, class II groups, which act at the level of canB, such as canB or factors that regulate its expression, will modify TCAG but not TCAGB. Only two groups, CE3-3 and CS2-1, failed to modify TCAGB (Table 2). Class I and class II modifier groups were mapped by meiotic recombination and by failure to complement deficiencies. The results from both methods were used to estimate the cytological map position of each group (Table 2).

Suppressor group CS2-1 is canB2:

Meiotic mapping localized the class II group CS2-1 to 44A;50B, and deficiencies refined the region to 42B3;43E18. Polytene chromosome analysis revealed a large deficiency in CS2-1¹²⁸ that uncovered 43E6;44B1 (Fig 4). CS2-1⁸⁷ was an inversion with breakpoints at 43A1-2 and 43E13-18 (Fig 4). The left breakpoint of CS2-1⁸⁷ fails to complement two independent alleles of pk, a gene in 43A1 that is required for tissue polarity in the wing, haltere, and notum (GUBB et al. 1999 ). However, the other CS2-1 alleles complemented pk alleles, the pk mutation in CS2-1⁸⁷ was viable, and pk did not modify TCAG (data not shown). The right inversion breakpoint, 43E13-18, was lethal when uncovered by deficiencies in the 43E region, and these deficiencies also failed to complement other CS2-1 alleles. Thus, the right breakpoint of the CS2-1⁸⁷ inversion corresponds to the CS2-1 TCAG suppressor.

One of the canB genes, canB2, maps to 43E16 and was a strong candidate for CS2-1. The CS2-1¹²⁸ deficiency uncovered canB2, since CS2-1¹²⁸/CyO,cn flies were cn, indicating that the deficiency breaks to the left of cn. PCR on genomic DNA from homozygous CS2-1⁸⁷ flies revealed that the right breakpoint of the insertion occurred between base pair positions -452 and +60, relative to the canB2 start of transcription (Fig 4, data not shown). The gene on the left side of the breakpoint, cn, was not disrupted in CS2-1⁸⁷, because CS2-1⁸⁷/CyO,cn flies were cn⁺. Additionally, Western blots of homozygous CS2-1⁸⁷, CS2-1¹²⁸, and CS2-1¹⁸⁰ larvae that were probed with canB antibodies revealed that, compared to similarly staged controls, total canB protein levels were reduced in CS2-1 alleles (data not shown).

Rescue was carried out by using the UAS-GAL4 system (BRAND and PERRIMON 1993 ) on CS2-1¹⁸⁰, which had no detectable lesions aside from canB2 and had a late larval/pupal lethal stage (data not shown). CS2-1¹⁸⁰/CyO;GAL4^hs flies were crossed to CS2-1¹⁸⁰/CyO;canB-4F^UAS, and the GAL4^hs/canB-4F^UAS progeny were screened for CS2-1¹⁸⁰ (i.e., Cy⁺) animals. Heat shock was not used because basal GAL4^hs activity at 25° induces UAS transgenes at a low level (data not shown). In three independent crosses, the percentage of CS2-1¹⁸⁰ adults was increased from <0.2% to an average of 11% (data not shown). Thus, ectopic canB-4F successfully rescued the lethality associated with CS2-1¹⁸⁰.

Suppressor CS3-3 is sprouty:

Deficiency mapping localized CS3-3 (Fig 5) to 63C6;63E, and the X-ray allele CS3-3¹⁵⁴ had a deletion spanning 63C2-5;63E1-4 (data not shown). The EMS allele CS3-3⁵¹⁸ (Fig 5B and Fig E) failed to complement sty⁵ and sty⁶⁴ (HACOHEN et al. 1998 ), which are hypomorphic alleles of sprouty, a gene in 63D1. In addition, the sty alleles were able to suppress TCAG (data not shown). Sequencing the sty ORF from CS3-3⁵¹⁸ revealed that the codon corresponding to glutamine residue 250 was mutated into a stop codon (Fig 5F). This mutation removes one of the two sty homology domains and the cysteine-rich region (HACOHEN et al. 1998 ) and is predicted to be nonfunctional.

Sprouty is a negative regulator of RTK signaling in Drosophila, including Fgf receptor and Egf receptor signaling (REICH et al. 1999 ). Sprouty protein can bind to the E3 ligase cbl (WONG et al. 2001 ), the adaptor protein Drk, and Gap1 and has been proposed to facilitate Gap1 inactivation of Ras (CASCI et al. 1999 ). A single gene in flies, sprouty has at least five homologs in mammalian genomes.

Calcineurin and Egf receptor signaling:
Increased signaling through the Egf receptor, caused by either the presence of ectopic Egfr/Ras signaling components or hypomorphic mutations in negative regulators, results in the development of extra photoreceptors and wing vein material (KARIM and RUBIN 1998 ; REICH et al. 1999 ). Conversely, a decrease in Egf receptor signaling reduces both wing vein formation and the number of photoreceptor cells (KARIM and RUBIN 1998 ; REICH et al. 1999 ). Ectopic canA^act caused defects in eye and wing vein development consistent with repression of Egfr/Ras signaling. Misexpression of canA^act in the posterior compartment of the developing wing by using GAL4^en resulted in a truncation of wing veins and a decrease in compartment size (Fig 6A and Fig B). GAL4^sev drives expression in the presumptive R3, R4, and R7 photoreceptor cells, as well as in cone cells. Sections of GAL4^sev/canA^act.UAS eyes revealed a decrease in the number of photoreceptor cells per ommatidium (cf. Fig 6D to Fig 6C). Usually the missing cell was R7, but occasionally R3 or R4 was also absent (Fig 6D, circled ommatidia). In addition, we examined TCAG and TCAGB discs for a decrease in active MAP kinase levels by using an antibody specific for the diphosphorylated, active form of MAP kinase (Sigma, M8159). However, posterior to the furrow the levels of active MAP kinase were too low to reliably detect any effect of activated calcineurin on signaling output from the Ras pathway (data not shown).

View larger version (128K): In this window In a new window Download PPT slide   Figure 6. Ectopic expression of canAact in eye and wing discs suppresses photoreceptor and wing vein formation. The wings (A and B) and eye sections (C and D) displayed are of the following genotypes: (A) GAL4en; (B) GAL4en;canAact.UAS; (C) GAL4sev; (D) GAL4sev/canAact.UAS. Compared to the control (A), expression of canAact in the posterior wing compartment by GAL4en (B, bracket indicates area of expression) reduced compartment size and inhibited vein formation (B, arrowheads). The expression of canAact.UAS in R cells 3, 4, and 7 by GAL4sev reduced the number of R cells per ommatidium by one or two at 8% penetrance (D). Examples of ommatidia lacking R cells are circled.

To examine the interaction between calcineurin and individual components of the Egfr pathway, we tested the ability of mutations in these components to modify the activated calcineurin phenotype (Table 3 and data not shown). While most downstream components of the pathway did not modify activated calcineurin, hypomorphic mutations in Egfr, Ras, pnt, sty, Gap1, and sl did (Fig 7). TCAGB (Fig 7A) was enhanced by removing one copy of Egfr, Ras, or pnt (Fig 7, B–D) and was suppressed by Gap1 and sl (Fig 7E and Fig F). Both TCAGB and TCAG suppressed the rough eye caused by hypermorphic Egfr alleles: flies that have one copy of Egfr^E1 (Fig 7G) and TCAGB (Fig 7A) have a rough eye that closely resembles that of TCAGB alone (Fig 7H). TCAG was not detectably modified by hypomorphic Egfr, Ras, or pnt alleles (data not shown). Aside from CS3-3, none of the modifier groups corresponded to Egf receptor/Ras signaling components that genetically interacted with TCAG (data not shown). However, it is possible that these genes are present among the 61 single hits, which have not been characterized. We also tested for genetic interactions with other signaling pathways by crossing to two Rho1 alleles, as well as mutants in components of the Notch and wingless pathways, but no convincing interactions with TCAGB or TCAG were detected (data not shown).

View larger version (88K): In this window In a new window Download PPT slide   Figure 7. A subset of RTK signaling genes genetically interact with activated calcineurin. SEMs are from flies of the following genotypes: (A) TCAGB; (B) Egfrf2/+;TCAGB; (C) Ras85De1B/TCAGB; (D) pnt782577/TCAGB; (E) sl1/Y;TCAGB; (F) Gap1PB/TCAGB; (G) EgfrE1/+; (H) EgfrE1/+;TCAGB. Removing one copy of Egfr (B), Ras (C), or pnt (D) enhanced the rough eye phenotype of activated calcineurin (A). In contrast, removing one copy of sl (E) or Gap (F) significantly suppressed the TCAGB rough eye. The hypermorphic allele EgfrE1 had a rough eye phenotype in one copy (G). Flies carrying activated calcineurin and one copy of EgfrE1 had an eye phenotype no stronger than activated calcineurin alone (H).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Specificity of activated calcineurin in Drosophila development:
Several lines of evidence demonstrate that phosphatase activity is required for ectopic canA^act phenotypes in Drosophila. Genetically raising or lowering the level of canB, which is essential for activity, respectively enhanced or suppressed the phenotype of canA^act. Transgenic flies expressing a form of canA^act that lacks an intact canB-binding site were indistinguishable from wild-type flies. Finally, the full-length phosphatase, which is inactive in the absence of Ca²⁺-calmodulin, did not have a detectable phenotype when overexpressed throughout development.

Activated calcineurin has been used reliably to identify physiologically relevant functions of calcineurin in a number of systems (FRUMAN et al. 1995 ; MENDOZA et al. 1996 ; MOLKENTIN et al. 1998 ; WINDER et al. 1998 ). However, it remains formally possible that calcineurin does not normally function in the cells or at the stage of development in which canA^act has a detectable phenotype, even though Pp2B-14D appears to be ubiquitously expressed. The role of calcineurin must be confirmed by mutational analysis of the canA and canB genes. Despite this caveat, the genetic screen presented here can still be used to identify physiological targets of calcineurin in Drosophila, as well as to provide insight into the roles calcineurin may play in development.

Activated calcineurin A modifier screen:
The canA^act.gl screen yielded 11 complementation groups, 9 of which failed to modify rough eyes caused by other glass-induced transgenes. This demonstrates that the majority of our modifier groups do not act through the glass enhancer. We then divided the nine specific modifiers into class I genes, which act downstream of calcineurin, and class II genes, which act at the level of canB.

Consistent with this classification, the class II group CS2-1 is canB2. The allele CS2-1⁸⁷ had an inversion that breaks within 400 bp of the canB2 start of transcription. CS2-1⁸⁷ and CS2-1¹⁸⁰ had decreased protein levels compared to similarly staged controls. Finally, the lethality of CS2-1¹⁸⁰ was partially rescued by canB-4F. The ability of canB-4F to rescue the canB2 lesion suggests that the canB-4F protein can at least partially substitute for canB2. More importantly, isolation of the calcineurin regulatory subunit in the canA^act.gl modifier screen demonstrates that the screen is capable of identifying genes that are required for calcineurin function.

The class I modifier group CS3-3 failed to complement the hypomorphic sprouty alleles sty⁵ and sty⁶⁴; both sty⁵ and sty⁶⁴ also suppressed canA^act.gl; and the sty gene from CS3-3⁵¹⁸ harbored a nonsense mutation (Q250Stop). Therefore, we conclude that the CS3-3 complementation group is sprouty. The fact that sty falls into the class II group suggests that sprouty functions downstream of calcineurin and/or in a parallel pathway.

Cross-talk between calcineurin and Egf receptor signaling:
Two lines of evidence suggest that calcineurin is a negative regulator of Egf receptor/Ras signaling. First, a negative regulator of RTK signaling, sprouty, was isolated as a suppressor of the canA^act.gl rough eye phenotype in the dominant modifier screen. Both sprouty and canA^act suppressed wing vein formation and reduced the number of photoreceptor cells per ommatidium (CASCI et al. 1999 ). Egf receptor/Ras signaling is essential for both wing vein and R-cell formation.

A thorough examination of genetic interactions between canA^act and components of RTK and other signaling pathways confirmed that canA^act specifically represses the Egf receptor/Ras pathway and that it acts upstream in the pathway. The lack of convincing genetic interactions with other signaling pathways in the imaginal eye disc does not rule out a role for calcineurin in these pathways in other developmental contexts. With the exception of pnt, activated calcineurin was not modified by components downstream of Ras and was modified only by a subset of genes that act between the Egf receptor and Ras. While Gap1 and sty alleles modified activated calcineurin, drk and cbl did not. Thus calcineurin may act downstream of, or parallel to, drk and cbl. The more downstream components of the Ras/MAP kinase pathway may not interact with activated calcineurin because they are too far removed from the point(s) of intersection between calcineurin and the pathway. Alternatively, these components may not be limiting, so that reduction of gene dose, which is the basis of a dominant modifier screen, would have no appreciable effect.

The hypermorphic allele Egfr^E1 inhibits Ras signaling (BAKER and RUBIN 1989 ); thus it might be expected to enhance activated calcineurin. However, low levels of inappropriate Egf receptor activity in eye development are thought to increase secretion of the Egf receptor antagonist argos (LESOKHIN et al. 1999 ). The argos protein inhibits subsequent Egf receptor signaling that is required for photoreceptor determination. Thus, suppression of the Egfr^E1 rough eye by canA^act.gl may be the result of activated calcineurin inhibiting inappropriate Egf receptor signaling.

Consistent with our findings, PLC is a negative regulator of Egf receptor/Ras signaling in eye and wing development (THACKERAY et al. 1998 ; POWE et al. 1999 ). However, PLC was identified in this study as a strong suppressor of activated calcineurin, although biochemically PLC has been placed upstream of calcineurin in the PI signaling pathway. One explanation is that PLC acts on one of the other canA genes. Another possibility is that the signaling pathways activated by PLC parallel to calcineurin are required for calcineurin function.

In a recent model, PLC was proposed to inhibit Egf receptor/Ras signaling via the activation of Gap1 by InsP₄ (POWE et al. 1999 ). Our results suggest that PLC is also acting through calcineurin. The genetic evidence that we present indicates that calcineurin intersects with the Ras pathway at roughly the same point that PLC does, and thus we propose a modified model for the function of PI signaling in Drosophila development (Fig 8). Additionally, the fact that calcineurin can be activated by any sustained Ca²⁺ flux suggests a mechanism by which other signaling pathways, such as GPCRs acting via PLCß, can modulate Egf receptor signaling.

View larger version (18K): In this window In a new window Download PPT slide   Figure 8. A model for the negative regulation of Egf receptor/Ras signaling by calcineurin and phosphoinositide signaling during Drosophila eye development. A simplified schematic illustrates upstream Egf receptor signaling components in an eye disc cell. PLC is activated by the Egf receptor and cleaves PIP2 to yield InsP3. PLC is proposed to negatively regulate Egf receptor signaling through InsP4, which is generated from InsP3 by an InsP3-3 kinase (dotted lines). Gap1 is then activated by InsP4, which results in the inhibition of Ras. Sprouty, which may be linked to the Egf receptor by the adaptor protein drk, may facilitate the inactivation of Ras by Gap. In our model, we propose that PLC also acts via Ca2+ and calcineurin (dashed lines). Genetic evidence suggests that calcineurin acts at the level of sty and Gap1, although it should be noted that calcineurin may act further upstream, e.g., at the level of InsP4. In addition, it is possible that calcineurin is activated by other Ca2+ signaling pathways.

In conclusion, we have demonstrated that a dominant modifier screen can be used successfully to isolate mutations in genes involved in calcineurin function. The mutations in the calcineurin B gene that we isolated in the screen will help determine the roles of calcineurin in Drosophila development. In addition, we have obtained compelling genetic evidence that calcineurin negatively regulates the Egf receptor/Ras signaling pathway at the level of Gap1 and sprouty. Calcineurin may act directly by dephosphorylating one or more signaling components, or it may target a transcription factor and act indirectly through changes in gene expression. More work will be needed to elucidate the molecular mechanism, and the modifiers isolated in the canA^act.gl screen should prove valuable in this endeavor. Furthermore, given the conservation of signal transduction between fruit flies and vertebrates, it is likely that the signaling network we have identified is employed in other organisms.

	ACKNOWLEDGMENTS

We thank Trudi Schüpbach, Justin Thackeray, Mike Simon, Herman Meisner, and Mark Krasnow for generous gifts of fly stocks. We express our gratitude to Todd Laverty for polytene chromosome analyses, Amy Tang for help with sectioning eyes, Tom Serano for assistance with SEM, and Felix Karim and Laurie Lee for advice on modifier screens and mapping. We also thank Kate Beckingham, Eric Lai, Laurie Lee, and Paul Wes for comments on the manuscript. K.M.C.S. was supported in part by a Postdoctoral Research Fellowship from the Cancer Research Fund of the Damon Runyon-Walter Winchell Foundation and by a Young Investigator Award from the National Alliance for Research on Schizophrenic and Affective Disorders.

Manuscript received October 17, 2001; Accepted for publication February 11, 2002.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ADAMS, M. D., S. E. CELNIKER, R. A. HOLT, C. A. EVANS, and J. D. GOCAYNE et al., 2000  The genome sequence of Drosophila melanogaster. Science 287:2185-2195[Abstract/Free Full Text].

BAKER, N. E. and G. M. RUBIN, 1989  Effect on eye development of dominant mutations in Drosophila homologue of the EGF receptor. Nature 340:150-153[Medline].

BERRIDGE, M. J., 1993  Inositol trisphosphate and calcium signaling. Nature 361:315-325[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].

BROWN, L., M. X. CHEN, and P. T. W. COHEN, 1994  Identification of a cDNA encoding a Drosophila calcium/calmodulin regulated protein phosphatase, which has its most abundant expression in the early embryo. FEBS Lett. 339:124-128[Medline].

CARTHEW, R. W. and G. M. RUBIN, 1990  seven in absentia, a gene required for specification of R7 cell fate in the Drosophila eye. Cell 63:561-577[Medline].

CASCI, T., J. VINOS, and M. FREEMAN, 1999  Sprouty, an intracellular inhibitor of Ras signaling. Cell 96:655-665[Medline].

CHANG, H. C. Y., N. M. SOLOMON, D. A. WASSARMAN, F. D. KARIM, and M. THERRIEN et al., 1995  phyllopod functions in the fate determination of a subset of photoreceptors in Drosophila. Cell 80:463-472[Medline].

CLIPSTONE, N. A. and G. R. CRABTREE, 1992  Identification of calcineurin as a key signaling enzyme in T-lymphocyte activation. Nature 357:695-697[Medline].

CRABTREE, G. R., 1999  Generic signals and specific outcomes: signaling through Ca²⁺, calcineurin, and NF-AT. Cell 96:611-614[Medline].

FRUMAN, D., S. PAI, S. BURAKOFF, and B. BIERER, 1995  Characterization of a mutant calcineurin A alpha gene expressed by EL4 lymphoma cells. Mol. Cell. Biol. 15:3857-3863[Abstract].

GAUL, U., G. MARDON, and G. M. RUBIN, 1992  A putative Ras GTPase activating protein acts as a negative regulator of signaling by the Sevenless receptor tyrosine kinase. Cell 68:1007-1019[Medline].

GREENSPAN, R. J., 1997 Fly Pushing: The Theory and Practice of Drosophila Genetics. Cold Spring Harbor Laboratory Press, Plainview, NY.

GUBB, D., C. GREEN, D. HUEN, D. COULSON, and G. JOHNSON et al., 1999  The balance between isoforms of the prickle LIM domain protein is critical for planar polarity in Drosophila imaginal discs. Genes Dev. 13:2315-2327[Abstract/Free Full Text].

GUERNINI, D., C. MONTELL, and C. B. KLEE, 1992  Molecular cloning and characterization of the genes encoding the two subunits of Drosophila melanogaster calcineurin. J. Biol. Chem. 267:22542-22549[Abstract/Free Full Text].

HACOHEN, N., S. KRAMER, D. SUTHERLAND, Y. HIROMI, and M. A. KRASNOW, 1998  sprouty encodes a novel antagonist of FGF signaling that patterns apical branching of the Drosophila airways. Cell 92:253-263[Medline].

HERBST, R., P. M. CARROLL, J. D. ALLARD, J. SCHILLING, and T. RAABE et al., 1996  Daughter of sevenless is a substrate of the phosphotyrosine phosphatase Corkscrew and functions during sevenless signaling. Cell 85:899-909[Medline].

KARIM, F. D. and G. M. RUBIN, 1998  Ectopic expression of activated Ras1 induces hyperplastic growth and increased cell death in Drosophila imaginal tissues. Development 125:1-9[Abstract].

KARIM, F. D., H. C. CHANG, M. THERRIEN, D. A. WASSARMAN, and T. LAVERTY et al., 1996  A screen for genes that function downstream of Ras1 during Drosophila eye development. Genetics 143:315-329[Abstract].

KLAMBT, C., J. R. JACOBS, and C. S. GOODMAN, 1991  The midline of the Drosophila central nervous system: a model for the genetic analysis of cell fate, cell migration, and growth cone guidance. Cell 64:801-815[Medline].

KLEE, C. B., H. REN, and X. WANG, 1998  Regulation of the calmodulin-stimulated protein phosphatase, calcineurin. J. Biol. Chem. 273:13367-13370[Free Full Text].

LESOKHIN, A. M., S. Y. YU, J. KATZ, and N. E. BAKER, 1999  Several levels of EGF receptor signaling during photoreceptor specification in wild-type, Ellipse, and null mutant Drosophila. Dev. Biol. 205:129-144[Medline].

LIU, J., J. D. FARMER, JR., W. S. LANE, J. FRIEDMAN, and I. WEISSMAN et al., 1991  Calcineurin is a common target of cyclophilin-cyclosporin A and FKBP-FK506 complexes. Cell 66:807-815[Medline].

MENDOZA, I., F. J. QUINTERO, R. A. BRESSAN, P. M. HASEGAWA, and J. M. PARDO, 1996  Activated calcineurin confers high tolerance to ion stress and alters the budding pattern and cell morphology of yeast cells. J. Biol. Chem. 271:23061-23067[Abstract/Free Full Text].

MOLKENTIN, J. D., J.-R. LU, C. L. ANTOS, B. MARKHAM, and J. RICHARDSON et al., 1998  A calcineurin-dependent transcriptional pathway for cardiac hypertrophy. Cell 93:215-228[Medline].

MOSES, K. and G. M. RUBIN, 1991  Glass encodes a site-specific DNA-binding protein that is regulated in response to positional signals in the developing Drosophila eye. Genes Dev. 5:583-593[Abstract/Free Full Text].

NUSSLEIN-VOLHARD, C., E. WIESCHAUS, and H. KLUDING, 1984  Mutations affecting the pattern of the larval cuticle in Drosophila melanogaster. I. Zygotic loci on the second chromosome. Roux Arch. Dev. Biol. 193:267-282.

O'KEEFE, S. J., J. TAMURA, R. L. KINCAID, M. J. TOCCI, and E. A. O'NEILL, 1992  FK-506- and CsA-sensitive activation of the interleukin-2 promoter by calcineurin. Nature 357:692-694[Medline].

O'NEILL, E. M., I. REBAY, R. TJIAN, and G. M. RUBIN, 1994  The activities of two Ets-related transcription factors required for Drosophila eye development are modulated by the Ras/MAPK pathway. Cell 78:137-147[Medline].

PAI, L. M., G. BARCELO, and T. SCHÜPBACH, 2000  D-cbl, a negative regulator of the Egfr pathway, is required for dorsoventral patterning in Drosophila oogenesis. Cell 103:51-61[Medline].

PERRIMON, N., A. LANJUIN, C. ARNOLD, and E. NOLL, 1996  Zygotic lethal mutations with maternal effect phenotypes in Drosophila melanogaster. II. Loci on the second and third chromosomes identified by P-element-induced mutations. Genetics 144:1681-1692[Abstract].

POWE, A. C., D. STRATHDEE, T. CUTFORTH, T. D'SOUZA-CORREIA, and P. GAINES et al., 1999  In vivo functional analysis of Drosophila gap1: involvement of Ca²⁺ and IP₄ regulation. Mech. Dev. 81:89-101[Medline].

RANGER, A. M., M. J. GRUSBY, M. R. HODGE, E. M. GRAVALLESE, and F. C. DE LA BROUSSE et al., 1998  The transcription factor NF-ATc is essential for cardiac valve formation. Nature 392:186-190[Medline].

REICH, A., A. SAPIR, and B. SHILO, 1999  Sprouty is a general inhibitor of receptor tyrosine kinase signaling. Development 126:4139-4147[Abstract].

ROGGE, R., P. J. GREEN, J. URANO, S. HORN-SABAN, and M. MLODZIK et al., 1995  The role of yan in mediating the choice between cell division and differentiation. Development 121:3947-3958[Abstract].

RUBIN, G. M., L. HONG, P. BROKSTEIN, M. EVANS-HOLM, and E. FRISE et al., 2000  A Drosophila complementary DNA resource. Science 287:2222-2224[Abstract/Free Full Text].

RUSNAK, F. and P. MERTZ, 2000  Calcineurin: form and function. Physiol. Rev. 80:1483-1521[Abstract/Free Full Text].

SCHMUCKER, D., H. JACKLE, and U. GAUL, 1997  Genetic analysis of the larval optic nerve projection in Drosophila. Development 124:937-948[Abstract].

SHIBASAKI, F., E. R. PRICE, D. MILAN, and F. MCKEON, 1996  Role of kinases and the phosphatase calcineurin in the nuclear shuttling of transcription factor NF-AT4. Nature 382:370-373[Medline].

SIMON, M. A., D. D. L. BOWTELL, G. S. DODSON, T. R. LAVERTY, and G. M. RUBIN, 1991  Ras1 and a putative guanine nucleotide exchange factor perform crucial steps in signaling by the sevenless protein tyrosine kinase. Cell 67:701-716[Medline].

STRUTT, D. I., U. WEBER, and M. MLODZIK, 1997  The role of RhoA in tissue polarity and Frizzled signalling. Nature 387:292-295[Medline].

SULLIVAN, K. M. C., K. SCOTT, C. S. ZUKER, and G. M. RUBIN, 2000  The ryanodine receptor is essential for larval development in Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 97:5942-5947[Abstract/Free Full Text].

THACKERAY, J. R., P. C. GAINES, P. EBERT, and J. R. CARLSON, 1998  small wing encodes a phospholipase C that acts as a negative regulator of R7 development in Drosophila. Development 125:5033-5042[Abstract].

WINDER, D. G., I. M. MANSUY, M. OSMAN, T. M. MOALLEM, and E. R. KANDEL, 1998  Genetic and pharmacological evidence for a novel, intermediate phase of long-term potentiation suppressed by calcineurin. Cell 92:25-37[Medline].

WOLFF, T., 2000 Histological techniques for the Drosophila eye. Part II: Adult, pp. 229–244 in Drosophila Protocols, edited by W. SULLIVAN, M. ASHBURNER and R. S. HAWLEY. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

WONG, E. S. M., J. LIM, B. C. LOW, Q. CHEN, and G. R. GUY, 2001  Evidence for direct interaction between sprouty and cbl. J. Biol. Chem. 276:5866-5875[Abstract/Free Full Text].

YAGI, Y., T. SUZUKI, and S. HAYASHI, 1998  Interaction between Drosophila EGF receptor and vnd determines three dorsoventral domains of the neuroectoderm. Development 125:3625-3633[Abstract].

YAKEL, J. L., 1997  Calcineurin regulation of synaptic function: from ion channels to transmitter release and gene transcription. Trends Pharmacol. Sci. 18:124-134[Medline].

This article has been cited by other articles:

				H. Yeo, L. H. Beck, S. R. Thompson, M. C. Farach-Carson, J. M. McDonald, T. L. Clemens, and M. Zayzafoon Conditional Disruption of Calcineurin B1 in Osteoblasts Increases Bone Formation and Reduces Bone Resorption J. Biol. Chem., November 30, 2007; 282(48): 35318 - 35327. [Abstract] [Full Text] [PDF]

				H. L. Nichols, N. Zhang, and X. Wen Proteomics and genomics of microgravity Physiol Genomics, September 14, 2006; 26(3): 163 - 171. [Abstract] [Full Text] [PDF]

				K. Gajewski, J. Wang, J. D. Molkentin, E. H. Chen, E. N. Olson, and R. A. Schulz Requirement of the calcineurin subunit gene canB2 for indirect flight muscle formation in Drosophila PNAS, February 4, 2003; 100(3): 1040 - 1045. [Abstract] [Full Text] [PDF]