skittles, a Drosophila Phosphatidylinositol 4-Phosphate 5-Kinase, Is Required for Cell Viability, Germline Development and Bristle Morphology, But Not for Neurotransmitter Release

Corresponding author: Hugo J. Bellen, Howard Hughes Medical Institute, Baylor College of Medicine, Room T630, 1 Baylor Plaza, Houston, TX 77030., hbellen{at}bcm.tmc.edu (E-mail).

Communicating editor: R. S. HAWLEY

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The phosphatidylinositol pathway is implicated in the regulation of numerous cellular functions and responses to extracellular signals. An important branching point in the pathway is the phosphorylation of phosphatidylinositol 4-phosphate by the phosphatidylinositol 4-phosphate 5-kinase (PIP5K) to generate the second messenger phosphatidylinositol 4,5-bis-phosphate (PIP2). PIP5K and PIP2 have been implicated in signal transduction, cytoskeletal regulation, DNA synthesis, and vesicular trafficking. We have cloned and generated mutations in a Drosophila PIP5K type I (skittles). Our analysis indicates that skittles is required for cell viability, germline development, and the proper structural development of sensory bristles. Surprisingly, we found no evidence for PIP5KI involvement in neural secretion.

PHOSPHOINOSITOL lipids have been postulated to play important roles in various cellular processes including growth, differentiation, and vesicular secretion. The phosphatidylinositol pathway consists of a series of conversions of phosphatidylinositol into singly, doubly, and triply phosphorylated products (reviewed in CARPENTER and CANTLEY 1990 , CARPENTER and CANTLEY 1996 ; DIVECHA and IRVINE 1995 ). An important branching point in the pathway occurs when phosphatidylinositol 4-phosphate (PtdIns[4]P) is phosphorylated to become phosphatidylinositol 4,5-bis-phosphate (PtdIns[4,5]P₂ or PIP2), a step catalyzed by phosphatidylinositol 4-phosphate 5-kinase (PIP5K; BORONENKOV and ANDERSON 1995 ; ISHIHARA et al. 1996 ). There are two types of PIP5Ks (PIP5KI and PIP5KII) with distinct biochemical and immunohistochemical properties, but they both catalyze the conversion of PtdIns[4]P into PIP2 (reviewed in LOIJENS et al. 1996 ). The hydrolysis of PIP2 by phospholipase C (PLC) produces the second messengers diacylglycerol (DAG) and inositol tris-phosphate (IP3). DAG is an activator of protein kinase C (PKC), and IP3 plays an important role in the release of intracellular calcium (RANA and HOKIN 1990 ). In addition, PIP2 is converted into phosphatidylinositol 3,4,5-tris-phosphate, which activates some PKC isoforms (TOKER et al. 1994 ).

PIP2 is itself a second messenger that has been implicated in the modulation of the function of cytoskeletal regulatory proteins such as profilin, cofilin, fascin, and gelsolin (JANMEY 1994 ). There is also evidence that phosphoinositide metabolism is involved in signal transduction and cytoskeleton regulation via the interaction with the Rho family of small G proteins (CHONG et al. 1994 ; REN et al. 1996 ). Other work has suggested an interaction between phosphoinositides and receptor tyrosine kinases (COCHET et al. 1991 ). It has also been suggested that PIP5K function may be associated with, or required for, DNA synthesis and cell proliferation (DIVECHA et al. 1993 ). Finally, PIP5KI was shown to be required for vesicular secretion in PC12 cells (HAY et al. 1995 ), while PIP5KII appears to be involved in vesicular trafficking in the budding yeast (YAMAMOTO et al. 1995 ). Most of our understanding of how PIP5Ks function to regulate cellular processes is derived from in vitro data. Whether the various, and apparently distinct, functions in which PIP5K is thought to be involved are related remains unknown. It also remains to be established whether these postulated roles of PIP2 are relevant in vivo and how the modulation of PIP5K levels affects development in animals.

The recent identification of a PIP5KI [skittles (sktl)] in Drosophila (KNIRR et al. 1997A ) makes the genetic and developmental analysis of the in vivo requirements of this gene possible and allows us to understand the role(s) played by phosphoinositides in various tissues and cell types. Our data show that sktl is essential for cell and organism viability and that it is required for cytoskeletal regulation during sensory structure development. We also find that sktl is required for germline development.

Finally our analysis resolves an issue pertinent to the function of another gene, inscuteable (insc). sktl maps to the first intron of insc, whose function is required for cell fate determination during neuronal and myogenic lineage development (KRAUT et al. 1996 ; KNIRR et al. 1997B ; RUIZ-GOMEZ and BATE 1997 ; CARMENA et al. 1998 ). To date, all studies on neuronal insc function have been carried out using deletion alleles that remove or affect both genes, allowing for the possibility that the described phenotypes may be in part due to the loss of sktl or from the combined loss of sktl and insc. In this study we show that the loss of sktl is not responsible for the insc phenotype.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Molecular biology:
cDNA isolation, sequencing, and Northern analysis were done as described in SAMBROOK et al. 1989 .

Genetics:
Mutations in sktl were generated by the imprecise excision of a revertible P-element insertion in the fata morgana (fam) complementation group that affects peripheral nervous system development (KANIA et al. 1995 ; note that except where a genotype is given, the P-element insertion will be referred to as fam^k07505). fam^k07505 was mobilized to generate imprecise excisions as follows: yw; famP{lacZ,w+}/CyO males were crossed to yw; Ki,2-3/TM3 females. yw; famP{LacZ,w+}/+; Ki,2-3/+ males were individually crossed to yw; Bl/CyO females. Individual white eyed yw; fam/CyO males were crossed back to yw; Bl/CyO females to establish stocks. Stocks lacking non-Curly flies were kept for further analysis.

Due to the presence of two genes, and several independently isolated alleles of each within the locus, some nomenclature issues must be clarified. All skittles alleles are denoted using the name of the gene (sktl). inscuteable alleles are referred to in this article using the gene name (insc); however, they are allelic to the nem mutations described by KNIRR et al. 1997B . Thus nem and insc are interchangeable. fata morgana (fam) is a complementation group described by KANIA et al. 1995 that affects both genes simultaneously and therefore cannot be used to describe either gene independently. The following mutant strains were used in this study:

• sktl alleles: yw; sktl⁵/CyO, yw; sktl¹⁵/CyO, and yw; sktl²⁰/CyO (this study)

• insc alleles: insc⁸/CyO and insc²²/CyO (BURCHARD et al. 1995 ; KNIRR et al. 1997B )

• Alleles that affect both sktl and insc: famP{lacZ,w+}/CyO, (KANIA et al. 1995 ), a P-element insertion in the sktl locus; this study, and P49/CyO (KRAUT and CAMPOS-ORTEGA 1996 ), a deletion creating null alleles in both sktl and insc.

Lethal phase identification:
To identify stages of larval lethality we balanced yw; sktl alleles with CyO, y+. Homozygous mutant larvae are therefore yellow and are identified by the color of their mouth hooks.

Clonal analysis:
The cross used to generate mitotic clones in the wing disc was as follows: yw, hsFLP; P{ry+neoFRT42D},P{y+,ry+44B}/CyO X yw/yw; P{ry+neoFRT42D}, sktl^{5 or 5}/CyO. The parental FRT stock (yw/yw; P{ry+neoFRT42D},P{y+,ry+44B}/CyO) was used to generate control flies. Flies were allowed to lay for 24 hr, removed, and the eggs aged for another 24 hr. Heat shock (37° 1 hr) was applied 48 and 53 hr after egg laying to maximize FLP activation. White non-Curly females were examined for the presence of yellow bristles.

The crosses used to generate females with a sktl mutant germline and clones in the eye imaginal disc were carried out as follows: yw/yw; P{w+FRT42B}, sktl¹⁵/CyO, y+ X yw, hsFLP; P{w+FRT42B}, P{w+ovoD1}2R1, P{w+ovoD1}2R2/CyO. Flies were allowed to lay for 24 hr and removed, and the eggs were aged for another 24 hr. Flies were heat shocked as above. Yellow, White non-Curly females produced by the above cross were examined for eye clones and mated to wild-type Canton S males or sktl¹⁵ males to examine the phenotypes of the progeny if any were produced. Control crosses in which the parental FRT stock was used instead of the FRT, sktl, were as follows: yw, hsFLP; P{w+FRT42B}, L/CyO X yw/yw; P{w+}, P{w+ovoD1}2R1, P{w+ovoD1}2R2/CyO.

Overexpression of sktl:
To generate UAS-sktl flies, yw flies were transformed with pUAST vector (BRAND and PERRIMON 1993 ) carrying sktl cDNA as insert and were selected using w+ as marker. Overexpression was carried out by crossing UAS-sktl to various Gal4 driver flies as described in the results section.

In situ hybridization:
In situ hybridization was carried out as described (TAUTZ and PFEIFLE 1989 ). A 2.9-kb sktl cDNA fragment in pBluescript was used to generate antisense riboprobes.

Immunohistochemistry:
The following antibodies were used as described: rabbit anti-PROSPERO (PROS; 1:1000; VAESSIN et al. 1991 ), MAb22C10 (1:100; CANAL and FERRUS 1986 ), Rabbit anti-INSCUTEABLE (INSC; 1:1000; KRAUT and CAMPOS-ORTEGA 1996 ), and Mab anti-EMBRYONIC LETHAL VISION DEFECTIVE (ELAV; 1:50; ROBINOW and WHITE 1991 ). We made numerous attempts to raise antibodies to the SKTL protein using bacterially expressed fusion proteins and unique peptide sequences but were unsuccessful. We also tested mouse antibodies (LOIJENS et al. 1996 ) for cross-reactivity with Drosophila embryos without success. For 4',6-diamidino-2-phenylindole (DAPI) staining, ovaries were fixed in 4% formaldehyde in phosphate-buffered saline and 0.1% Tween (PBT). Ovaries were washed in PBT and mounted in 75% glycerol in 100 nM Tris pH 7.5 with 1 mg/ml DAPI.

Electrophysiology at the third instar larval neuromuscular junction:
sktl¹⁵ and fam^k07505 flies were balanced with a translocation balancer in which the Curly (Cy) and Tubby (Tb) markers segregate together [T(2;3) sktl¹⁵/SM5; TM6B, Tb and T(2;3) fam^k07505/SM5; TM6B, Tb], thus allowing second chromosome mutations to be identified during larval stages by the absence of Tb (sktl¹⁵/T(2:3)Cy, Tb x fam^k07505/T(2:3)Cy, Tb). Non-Tubby third instar larvae were grown at 24°. Dissections, nerve stimulation, and recordings were performed as described (JAN and JAN 1976 ) in HL-3 solution (STEWART et al. 1994 ) at 2 mM Ca²⁺. Nerves innervating the ventral body wall were cut at the ventral ganglion and stimulated using a suction electrode. Nerves were stimulated for 0.2 msec at a voltage 2.5 times greater than the threshold voltage. For neuropeptide secretion analysis (large dense core vesicle release) nerves were stimulated at 30 or 50 Hz for a train of 250 ms. Four mutant larvae were examined. Recordings were performed at muscle fiber 6 in abdominal segments 4-6. Spontaneous release [miniature excitatory junctional potential (mEJP)] was measured for ~5 min. All dissections and recordings were performed at 24°.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

sktl encodes a putative PIP5KI:
sktl was identified as a transcription unit nested in the first (10 kb) intron of inscuteable (insc; Figure 1A; KNIRR et al. 1997B ). Both genes map to 57B on the second chromosome. The structure and sequence of the insc/sktl genomic region, and the sizes of the sktl exons (Figure 1A) indicated that the protein sequence described by KNIRR et al. 1997A may not correspond to the full-length SKTL protein. We cloned the sktl cDNA by plasmid rescuing a lethal P-element insertion from the fata morgana complementation group (fam^k07505; KANIA et al. 1995 ) that maps to the sktl locus. We used the flanking genomic fragments to screen an embryonic cDNA library (ZINN et al. 1989 ). A 2.9-kb cDNA was isolated with a single open reading frame encoding a 700-amino-acid protein (GenBank Accession number 3288870) with 59% identity to mouse and 58% identity to human PIP5KI over the entire length of the respective proteins. In contrast, the homology to human and mouse PIP5K type II isoforms was limited to 25% identity and 45% similarity. Therefore, sktl encodes a Drosophila homologue of vertebrate PIP5KI proteins. Interestingly, the first 100 amino acids are unique to SKTL and do not show homology to any amino-acid sequences in databases.

View larger version (13K): In this window In a new window Download PPT slide   Figure 1. Schematic representation of the skittles (sktl) locus. (A) sktl () is located within the first intron of inscuteable (insc; ). sktl has two exons with a single open reading frame () encoding a predicted 700-amino-acid protein. A P element (fam7505) is inserted in the intron of sktl and causes mutations in both genes. Mutations in sktl were created by imprecise excisions of the fam P element. Three excisions are shown represented by black bars. The boundaries of these excisions were not precisely mapped. Numbers indicate kilobases. P49 is a deletion that uncovers the entire sktl gene and undetermined regions of insc (KRAUT et al. 1996 ). No sktl RNA or INSC proteins are detectable P49 mutants. (B) sktl encodes a single 3.8-kb message that is invariant throughout development (larval stages not shown). E, embryonic; A, adult.

sktl has a dynamic expression pattern during development:
Northern analysis shows that sktl encodes a single 3.8-kb message (Figure 1B) that is invariant in size during development (data not shown). In situ hybridization shows that during embryogenesis sktl is expressed at all stages, but there is a very dynamic pattern of regulation in various developing tissues. At all stages there is a basal level of expression in all cells. At stage 5 (stages according to CAMPOS-ORTEGA and HARTENSTEIN 1985 ) strong expression is seen in the procephalic neuroectoderm (Figure 2A). During gastrulation, expression is elevated in the invaginating cells of the ventral and cephalic furrows (Figure 2B). At stage 11 all central nervous system (CNS) and peripheral nervous system (PNS) precursor cells express high levels of sktl (Figure 2C). At stage 13 most developing tissues (heart, gut, muscles, CNS, and PNS) express high levels of skittles (not shown). By the end of embryogenesis (stage 17) expression is prominent in a few CNS cells and the gut (Figure 2D). This expression pattern, particularly in the nervous system, is remarkably similar, if not identical, to that of insc (KRAUT and CAMPOS-ORTEGA 1996 ; KNIRR et al. 1997B ). This suggests that the two genes may share common regulatory elements and raises the question of whether they interact during nervous system development. Alternatively, sktl may be under the control of insc enhancers and may serve an unrelated function.

View larger version (98K): In this window In a new window Download PPT slide   Figure 2. sktl is expressed in a dynamic pattern throughout embryonic and larval development. sktl RNA is provided maternally (not shown). At all stages, basal levels of sktl are expressed in all cells. At stage 5 (A) sktl transcript is highly enriched in procephalic neuroblasts. During gastrulation (B) sktl continues to be expressed in the developing brain in addition to elevated levels of expression in the ventral and cephalic furrows. By stage 11 (C) sktl transcript is detectable in all neuronal precursors in the CNS and PNS as well as in migrating germ cells. Later in embryonic development sktl is expressed at low levels in all cells and elevated in the developing CNS, PNS, dorsal vessel, gut, and mesoderm (not shown). By stage 17 (D) sktl is very strongly expressed in the gut, the pharynx, and a small subset of CNS cells. In third instar larvae sktl is expressed in most cells of the wing (E) and eye-antennal (F) discs. In the wing disc expression is elevated in the precursors of anterior wing margin sense organs (SOPs) and along or next to the anterior-posterior axis (AP). Little expression, if any, is detectable along the dorsoventral axis (DV). In the eye-antennal disc expression is elevated in the precursors of R8 photoreceptors (arrows). In the brain sktl transcript is detectable in the optic lobe (OL), and several groups of cells in the midbrain (G), and many cells (arrows) in the ventral ganglion (H).

During third instar larval development sktl is expressed widely in all imaginal discs. In the leg disc expression is ubiquitous and uniform (data not shown). In the wing disc, expression is elevated in the precursors of the anterior wing margin sensory organs and along the anterior-posterior axis (Figure 2E). Expression is very low or absent along the dorso-ventral axis. In the eye disc, expression is elevated in the row of cells anterior to the morphogenetic furrow from which the R8 photoreceptors will differentiate (Figure 2F). In the third instar larval brain, sktl is expressed widely but not ubiquitously (Figure 2G). Areas of expression include the outer proliferation center of the optic lobes, several patches of cells in the midbrain, and subsets of cells in the ventral ganglion (Figure 2H).

Imprecise excisions of fam^k07505 result in sktl mutations:
The fam complementation group was defined by the revertible fam^k07505 insertion in the sktl locus. Homozygous fam^k07505 embryos have strongly reduced levels of sktl transcript, suggesting that fam^k07505 is a sktl allele. In addition, fam^k07505 fails to complement two EMS alleles of insc, insc⁸ and insc²². Therefore fam^k07505 represents a mutation in both sktl and insc. To create mutations that affect only sktl we generated imprecise excisions of fam^k07505. We screened for sktl mutations by complementation analysis with insc⁸, insc²², and fam^k07505. Table 1 shows the results of this screen. Nine homozygous lethal excisions were tested for complementation with insc⁸, insc²², and fam^k07505. Six excisions complemented insc⁸ and insc²² but failed to complement fam^k07505 (sktl⁵, sktl⁹, sktl¹⁵, sktl²⁰, sktl²⁴, and sktl³¹). One excision (sktl¹³) failed to complement insc⁸, insc²², and fam^k07505. Two excisions complemented insc⁸, insc²², and fam^k07505 (sktl²⁵, sktl²⁹; data not shown). Because insc⁸ and insc²² affect insc, and fam^k07505 affects both sktl and insc, it follows that excisions that complement the insc⁸ and insc²² but fail to complement the fam^k07505 are potentially mutations in sktl alone. Therefore sktl⁵, sktl⁹, sktl¹⁵, sktl²⁰, sktl²⁴, and sktl³¹ are good candidates for mutations that specifically affect sktl. To verify that these alleles do not affect insc we examined the expression of the insc RNA and protein in these mutants. We observed no detectable difference between mutant and control embryos (data not shown). Next we determined the molecular lesions associated with the excision of the P element in each of these alleles (Table 1). All alleles are deletions ranging between 0.7 and 4 kb. None of these deletions affects insc exons, and the normal pattern of insc expression in these alleles suggests that they do not affect any putative intronic insc enhancers. In contrast, all alleles showed a marked decrease in sktl expression (Figure 3). Therefore, imprecise excisions of the fam^k07505 P element result in mutations that affect sktl but not insc, allowing us to study the function(s) of the two genes independently.

View larger version (136K): In this window In a new window Download PPT slide   Figure 3. sktl alleles have reduced levels of sktl RNA. Panels represent lateral views of a wild-type embryo (A) and a homozygous sktl20 embryo (B) at stage 11. Strong reduction in expression is seen in sktl20 embryos in which a large portion of sktl is deleted, supporting the genetic evidence that suggests that sktl20 is either a stong hypomorphic allele or a null mutation. Other alleles show degrees of message reduction consistent with the size of their respective deletions (not shown).

sktl mutations result in embryonic lethality:
To determine the function of sktl during embryonic development, we examined the phenotypes associated with the loss of sktl function using the smallest (sktl¹⁵), an intermediate (sktl⁵), and the largest (sktl²⁰) excisions. sktl¹⁵ larvae are late first larval instar lethal and have no gross morphological abnormalities. Similarly, sktl⁵ mutants are early first instar lethal and have no obvious defects. sktl²⁰ mutants are late embryonic lethal and also show no gross morphological defects as revealed by light microscopy (see also next section). sktl¹⁵/sktl²⁰ and sktl⁵/sktl²⁰ transheterozygotes are early first instar lethal while sktl²⁰/P49 transheterozygotes are identical in stage of lethality to sktl²⁰ homozygous embryos. While these genetic data suggest that sktl⁵ and sktl¹⁵ are strong hypomorphs and sktl²⁰ is a null allele, in situ hybridization shows that the sktl message, although severely reduced, is not completely abolished in sktl²⁰ embryos (Figure 3). These observations can be reconciled by the fact that the deletion in sktl²⁰ uncovers at least part of the open reading frame (ORF) (Figure 1A) and therefore may lack a functional SKTL protein. However, in the absence of a sktl antibody we are unable to confirm this prediction.

Zygotic sktl expression is not required for embryonic nervous system development:
Because sktl is expressed at high levels in most if not all nervous system precursors, we were interested in determining the consequences of the loss of sktl on nervous system development. We used sktl²⁰ mutant embryos to determine if sktl is required for nervous system development. Nervous system development was examined using anti-ELAV (ROBINOW and WHITE 1991 ) and Mab 22C10 (CANAL and FERRUS 1986 ) antibodies to detect neurons and anti-PROS to detect neuronal precursors during early neurogenesis and glial cells during late neurogenesis (VAESSIN et al. 1991 ). No detectable defects were seen by stage 16 with these markers (data not shown). The absence of an insc phenotype in sktl mutants demonstrates that the phenotypes reported by KRAUT and CAMPOS-ORTEGA 1996 and KRAUT et al. 1996 , i.e., the loss of nervous system cells due to mislocalization of Numb and Prospero during neuronal lineage development, are indeed caused by the absence of insc and not by the loss of function of either sktl alone or both sktl and insc.

sktl is not required for neurotransmitter release at the neuromuscular junction:
PIP5KI was shown to be required for Ca²⁺-dependent neuropeptide secretion from PC12 cells (HAY et al. 1995 ). In addition, several lines of evidence suggest that PIP5Ks play a crucial role in regulating membrane trafficking (reviewed in DE CAMILLI et al. 1996 ; MARTIN 1997 ; see DISCUSSION). Furthermore, sktl is expressed in many cells in the ventral ganglion of third instar larvae. Some of these cells may correspond to motor neurons innervating the larval body wall muscles. The Drosophila third instar larval neuromuscular junction is an excellent system for measuring neurotransmitter release from motor neurons. To test the requirement of sktl for neurotransmitter release we used third instar larvae transheterozygous for the sktl¹⁵ and fam^k07505 alleles. This combination results in late second instar lethality with a few third instar escapers. Electrophysiological recordings at the neuromuscular junction (JAN and JAN 1976 ) were done to examine spontaneous release as well as evoked release. The size, shape, and frequency of evoked and spontaneous responses were examined in control (heterozygous for either allele alone) and transheterozygous mutant larvae. Evoked response was measured after either single or repetitive stimulation (Figure 4). We also tested for nerve fatigue by repetitive stimulation. No detectable defects were observed in any of the above measurements, suggesting that sktl is not required for glutamate neurotransmitter release, at least at the larval neuromuscular junction. It is therefore unlikely to play a role in regulating vesicular trafficking at that junction.

View larger version (13K): In this window In a new window Download PPT slide   Figure 4. Evoked excitatory junctional potentials (EJPs) in third larval body-wall muscles. EJPs were evoked by either a single stimulus (left) or a short train of stimuli at 30 Hz (right). Larval neuromuscular junction preparations were dissected as described (JAN and JAN 1976 ) and incubated in 1 mM CaCl2 HL-3 solution (STEWART et al. 1994 ). The rapid and transient glutamatergic responses (i.e., individual EJPs) are indistinguishable between control larvae (A; sktl15/TM6 or famk07505/TM6) and the mutant larvae (B; sktl15/famk07505). The slow ramp depolarization underlying the train of stimuli is thought to be mediated in part by the peptide PACAP (ZHONG and PENA 1995 ). This ramp potential appears to be unaltered in sktl15/famk07505 larvae. The resting potential was -64 mV.

ZHONG and PENA 1995 showed that repetitive stimulation at 20 Hz or higher results not only in a fast evoked response but also in a slow, neuropeptide-dependent, depolarization. Neuropeptides are released by dense core vesicles from the nerve terminal. HAY et al. 1995 showed that PIP5KI was required for dense core vesicle secretion from PC12 cells. To test the requirement of sktl for dense core vesicle secretion we stimulated the motor nerves at 30 and 50 Hz. The resulting slow depolarization profile in sktl mutants was indistinguishable from that of heterozygous controls or wild-type larvae (Figure 4). Therefore sktl does not appear to play a role in regulating dense core vesicle secretion at least at the larval neuromuscular junction. It is not possible to exclude the possibility that the transheterozygous combination used in these experiments, while being strong enough to cause larval lethality, is not strong enough to have an effect on peptide release. However, this is unlikely because even viable hypomorphic mutations of numerous proteins involved in neurotransmission, such as synaptotagmin and RAS opposite (ROP), show severe electrophysiological defects (LITTLETON et al. 1994 ; WU et al. 1998 ).

sktl is required for germline development:
KNIRR et al. 1997A reported that sktl is expressed in germ cells during oogenesis. At stage 6 (stages according to SPRADLING 1993 ), sktl expression is restricted to the future oocyte. By stage 9, expression is initiated in the nurse cells. At the end of oogenesis, large amounts of sktl transcript are present in the mature egg. To examine the function of sktl in germline development we generated sktl germline clones using one of the excision alleles (sktl¹⁵). Negative control crosses in which no recombination was induced resulted in 100% of the females being sterile. The ovaries of these females, carrying the dominant sterile ovo^D1 marker, were severely atrophic and showed very early arrest of egg chamber development. Positive control crosses showed that 47% of the females were fertile (Table 2). In contrast, all sktl recombinant females were sterile (Table 2). Ovaries were dissected and stained with DAPI to reveal the nuclei. Oogenesis in these females was arrested after stage 10, and very few eggs were fully developed. Arrested egg chambers showed defects in nurse cell nuclei at and after stage 10 (compare Figure 5B and Figure C), but no defects in nuclear morphology were seen before that stage (arrows in Figure 5C). The affected nuclei appeared very small and fragmented, suggesting that sktl is required for nurse cell viability, and therefore proper egg chamber development. In contrast, the oocyte nucleus did not appear to be affected. The few eggs that developed were smaller than the eggs produced by control flies and showed defects in their dorsal appendages (compare Figure 5D and Figure E). Generally, the dorsal appendages of sktl mutant eggs were short and thick. The sterility associated with the partial loss of sktl in the female germline precludes the determination of the consequences of the loss of sktl in embryos.

View larger version (67K): In this window In a new window Download PPT slide   Figure 5. Loss of function of sktl suggests that sktl is required for germline development and bristle differentiation. Clonal analysis in the wing imaginal disc results in very few, small, clones each covering one or, rarely, two bristles. These bristles have a variety of structural defects ranging from wavy shape (not shown) to sharp bends. (A) Reduction of sktl levels (using sktl15 allele) in the female germline results in sterile females. (B) A normal egg chamber at stage 10 of oogenesis stained with DAPI to reveal nuclei of all cells. Nurse cell nuclei (arrowheads) are normal in size and shape. In contrast, nurse cells with reduced levels of sktl (C) contain some small and abnormal nurse cell nuclei (arrowheads). Mutant egg chambers do not show any defects in nuclear morphology before stage 10 (arrows in C). Very few mutant egg chambers complete oogenesis. Those that do (E) are small and have short and thick dorsal appendages when compared to control eggs (D).

sktl is required for cell viability and bristle differentiation:
As discussed above, sktl is expressed widely in the wing disc. To examine the function of sktl in wing disc development we generated sktl mutant clones using the sktl⁵ and sktl¹⁵ alleles (using the FLP-FRT system, see MATERIALS AND METHODS). The absence of the yellow marker was used to identify the clones. Approximately 76% of the control flies had yellow bristle clones on their notum, legs, and wing margin (Table 2). These bristles showed no defects. In contrast, only 1–2% of the recombinant flies had yellow bristles on the notum. In addition the size of these clones was markedly reduced in comparison to control flies: each clone consisted of a single, and rarely of two bristles. The few mutant bristles recovered showed structural abnormalities ranging from a wavy shape (in most cases) to sharp bends (in a few cases; Figure 5A), suggesting cytoskeletal defects. No mutant bristles were observed on the wing margin or the legs. The very small number of clones obtained combined with the small size of each clone suggest that sktl is required for either cell viability, proliferation, or both during wing disc development.

To determine if loss of sktl is required in the eye disc, where it is abundantly expressed, we also generated sktl mutant clones. In control experiments, 45% of the recombinant flies had white eye clones of variable sizes. In contrast, no sktl mutant clones were observed (Table 2), supporting the conclusion that sktl function is required for cell viability or cell division in imaginal discs. It should be noted that third instar larvae transheterozygous for the sktl¹⁵ and fam^k07505 alleles show no defects in the sizes of the imaginal discs and the brain (data not shown). Therefore it is more likely that the absence of sktl clones results from an effect on cell viability. While we favor this hypothesis, we cannot exclude the possibility that the sktl¹⁵/fam^k07505 combination, while being lethal, is not severe enough to reveal a role for sktl in cell proliferation.

Overexpression of sktl affects bristle number and morphology:
To further characterize the function of sktl, we carried out overexpression studies. We generated flies with a UAS-sktl construct and used it to overexpress sktl using a variety of Gal4 drivers (BRAND and PERRIMON 1993 ). The neuronal-specific elav-Gal4 driver (DE SIMONE and WHITE 1993 ) resulted in no detectable phenotypes and gave rise to fertile adults. Overexpression using the ubiquitous daughterless-Gal4 driver (WODRAZ et al. 1995 ) and the heat shock-Gal4 driver resulted in no obvious phenotypes during embryogenesis but caused early larval lethality. Expression with the ubiquitous imaginal disc driver T80-Gal4 resulted in third instar larval lethality. The lethality associated with the ubiquitous expression of sktl, which is itself very widely expressed, precluded the use of the UAS-sktl construct to rescue sktl mutants. Overexpression with the dpp-Gal4 driver (STAEHLING-HAMPTON et al. 1994 ; expressed in the morphogenetic furrow of the eye disc and along the anterior-posterior boundary in the wing disc) showed no detectable phenotypes in the eye, but caused ectopic bristle formation on the notum and wing blade. The appearance of ectopic bristles on the wing blade was seen in ~20% of the flies, but the majority of the flies (70%) showed ectopic bristles on the notum. These results prompted us to examine the effects of generalized overproduction of sktl in the wing disc. We used the 32B-Gal4 driver that is expressed ubiquitously at high levels in the third instar wing disc. Overexpression of sktl resulted in ectopic bristles (macrochaetae and microchaetae) on the notum (Figure 6B) and the wing blade (data not shown). On the notum, both the ectopic and normal bristles showed severe structural defects (Figure 6C). The structural defects of the bristles are consistent with a role for sktl in regulating cytoskeletal components. All ectopic bristles were associated with socket cells, suggesting that the production of ectopic bristles may be the result of the specification of extra sense organ precursors rather than the transformation of a socket cell into a bristle cell.

View larger version (99K): In this window In a new window Download PPT slide   Figure 6. Overexpression of sktl causes severe defects during larval development. Wild-type notum is shown in A. Overexpression in the entire wing disc (B and C) causes larval and pupal death with a few pharate adult escapers. Eclosed (B) and dissected (C) adults show ectopic microchaetae (1.7x normal number) and macrochaetae (3–6 per animal). Both normal and ectopic bristles display bends and twists (arrows), suggesting aberrant cytoskeletal architecture. Note that, contrary to wild-type flies (A), bristles on transgenic flies are in different focal planes (B), possibly reflecting problems in notum development. Some flies show severe abnormalities in the notal structure (C).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

PIP5Ks are thought to be involved in a variety of cellular processes including cytoskeletal regulation and neuronal secretion. We generated mutations in the Drosophila PIP5K sktl and undertook the analysis of its function during fly development. We find that the loss of sktl function affects several aspects of Drosophila development and that it does not affect other aspects in which PIP5K function has been implicated.

sktl and vesicular trafficking:
Several studies link PIP5K activity to the regulation of vesicular trafficking (reviewed in DE CAMILLI et al. 1996 ; MARTIN 1997 ). It is thought that PIP2 metabolism may regulate both endocytosis and exocytosis through interactions with key proteins required for both processes, namely dynamin, AP-2, AP-3, synaptotagmin, and CAPS. (CHEN et al. 1991 ; SMYTHE et al. 1992 ; GAIDAROV et al. 1996 ; SALIM et al. 1996 ; HAO et al. 1997 ; MARTIN 1997 ). It has also been shown that a PIP5KI is required for Ca²⁺-regulated dense core vesicle release from neuroendocrine cells (HAY et al. 1995 ). These data suggest that PIP2 plays a crucial role in regulation of membrane trafficking. On the basis of these observations, it is expected that mutations in a PIP5KI would result in defects in evoked neurotransmitter release. The expression of sktl in the larval ventral ganglion (Figure 2H) also suggested that sktl may play a role in neural secretion. However, we find no defects in neurotransmitter release at the third larval instar neuromuscular junction in sktl mutants. While we did not find defects in peptide secretion, we cannot exclude minor defects not visible in our assays. The lack of a vesicular secretion phenotype may be due to the activity of other PIP5Ks in Drosophila. Drosophila has a PIP5K type II that maps to the tip of chromosome 4 and that appears, from preliminary expression analysis, to be expressed specifically in the late embryonic CNS (B. HASSAN, unpublished results). It remains to be established whether this form of PIP5K functions in neuronal secretion. Finally, as it appears that sktl may function in regulating cytoskeletal processes (see below), we examined whether sktl mutants showed any gross morphological defects in the motor nerve terminals. No such defects were observed by light microscopy.

sktl and germline development:
The reduction in the level of sktl transcript results in female sterility, indicating that sktl is necessary for germline development. When examined with DAPI, sktl mutant egg chambers show what appear to be degenerating nurse cell nuclei. Nurse cells normally act to provide the developing oocyte with RNA and proteins necessary for various aspects of embryonic development. If sktl is required for the function of nurse cells, reducing its levels would be expected to result in smaller eggs, which is what is observed (Figure 5E). Interestingly, no defects were observed in the oocyte nucleus, despite the fact that sktl is transcribed in the oocyte even before its expression in nurse cells (KNIRR et al. 1997A ).

sktl and bristle development:
The strongest evidence for PIP5K requirement in the regulation of the actin cytoskeleton comes from studies in platelets. Upon thrombin receptor activation the small G-protein Rac is activated. Rac activation is thought to induce PIP5K activity that results in increased PIP2 synthesis (HARTWIG et al. 1996 ). PIP2 is known to have strong binding affinities to several actin interacting proteins such as fascin, profilin, gelsolin, and -actinin (FUKAMI et al. 1996 ; HARTWIG et al. 1996 ; SCHAFER et al. 1996 ). It is thought that the interaction between these proteins and PIP2 results in the inhibition of their interaction with monomeric actin and therefore allows the nucleation of new actin filaments (POLLARD 1986 ; POLLARD and RIMM 1991 ). This model is probably applicable to many, if not most, cells undergoing cytoskeletal rearrangements. The Drosophila adult peripheral nervous system has cells that extend sensory bristles rich in actin, providing an excellent system for the study of the defects in cytoskeleton assembly (TILNEY et al. 1995 , TILNEY et al. 1996 ). Mutations in the Drosophila actin interacting proteins profilin (chickadee) and fascin (singed) show severe defects in bristle morphology (CANT et al. 1994 ; VERHEYEN and COOLEY 1994 ). Loss of function of either protein results in bristles that lack actin filament integrity, causing bending and branching during extension (VERHEYEN and COOLEY 1994 ; CANT and COOLEY 1996 ). Our analysis indicates that the alteration of PIP5KI levels results in structural defects in sensory bristles, providing genetic evidence for the involvement of PIP5KIs in cytoskeletal regulation. The phenotypes observed in sktl clones are similar but not identical to those observed in singed and chickadee mutants (VERHEYEN and COOLEY 1994 ; CANT and COOLEY 1996 ). Specifically, while the loss- and gain-of-function of sktl results in bent and wavy bristles (Figure 4A; Figure 5B and Figure C), we did not observe branching bristles. Finally, it is interesting to note that mutations in all three genes (chickadee, singed, and sktl) result in female sterility. Therefore, both germline and bristle development present model systems in which to study the interactions between PIP5K and the actin cytoskeleton.

Does sktl have an early role in the adult PNS?
The appearance of extra bristles associated with socket cells as a result of the overexpression of sktl in the wing disc can be explained in one of two ways: extra cell division, or specification of an extra precursor cell. The sterility resulting from the removal of the maternal component and the failure of somatic clones to survive does not allow us to correlate the ectopic production of bristles with a loss-of-function phenotype. However, it is interesting to note that cytoskeleton-associated proteins like INSC (KRAUT and CAMPOS-ORTEGA 1996 ; KRAUT et al. 1996 ) and sanpodo (a Drosophila tropomodulin homologue; DYE et al. 1998 ) play a significant role in cell fate specification in the nervous system.

In conclusion, the great diversity of functions proposed for PIP5Ks and phoshatidylinositol kinases in general makes the in vivo analysis of their function difficult. However, the identification of these genes in a multicellular organism amenable to genetic manipulation like Drosophila (LEEVERS et al. 1996 ; KNIRR et al. 1997A ; this study), allows us to test genetically the biochemical and cell biological data available and to uncover new roles for PIP5Ks in development. Mutations in sktl allowed us to test which of the many roles proposed for PIP5KI proteins are carried out by sktl and which may be the function of other isoforms and/or types of PIP5Ks. For instance, it appears that sktl is involved in regulating the cytoskeleton and possibly cell survival. In contrast, sktl does not appear to play a role, or plays a redundant role, in neurotransmitter or neuropeptide release, and in cell proliferation. The distinction between redundancy and lack of requirement awaits the genetic and functional characterization of other PIP5Ks in Drosophila.

	ACKNOWLEDGMENTS

The authors thank Yuchun He and Ruth Hyland for excellent technical assistance, Harald Vaessin for Prospero antibody, William Chia for INSC antibody and cDNA, and Just Loijens and Richard Anderson for mouse PIP5KI antibodies. We thank Richard Atkinson and Mark Wu for comments on the manuscript. B.H. thanks Michael Zavortink for sharing unpublished information. S.B. thanks Boehringer Ingelheim for support. This research was funded in part by a National Institutes of Health grant to H.J.B. B.H. is a Postdoctoral Associate of the Howard Hughes Medical Institute (HHMI), B.Z. is a fellow of the American Cancer Society, and H.J.B. is an Associate Investigator of the HHMI.

Manuscript received June 10, 1998; Accepted for publication August 28, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BORONENKOV, I. V. and R. A. ANDERSON, 1995  The sequence of phosphatidylinositol 4-phosphate 5-kinase defines a novel family of lipid kinases. J. Biol. Chem. 270:2881-2884[Abstract/Free Full Text].

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].

BURCHARD, S., A. PAULULAT, U. HINZ, and R. RENKAWITZ-POHL, 1995  The mutant not enough muscles (nem) reveals reduction of the Drosophila embryonic muscle pattern. J. Cell Sci. 108:1443-1454[Abstract].

CAMPOS-ORTEGA, J. A., and V. HARTENSTEIN, 1985 The Embryonic Development of Drosophila melanogaster. Springer-Verlag, Berlin.

CANAL, I. and A. FERRUS, 1986  The pattern of early neuronal differentiation in Drosophila melanogaster. J. Neurogenet. 3:293-319[Medline].

CANT, K. and L. COOLEY, 1996  Single amino acid mutations in Drosophila fascin disrupt actin bundling function in vivo.. Genetics 143:249-258[Abstract].

CANT, K., B. A. KNOWLES, M. S. MOOSEKER, and L. COOLEY, 1994  Drosophila singed, a fascin homologue, is required for actin bundle formation during oogenesis and bristle extension. J. Cell Biol. 125:369-380[Abstract/Free Full Text].

CARMENA, A., B. MURUGASU-OEI, D. MENON, F. JIMENEZ, and W. CHIA, 1998  inscuteable and numb mediate asymmetric muscle progenitor cell divisions during Drosophila myogenesis. Genes Dev. 12:304-315[Abstract/Free Full Text].

CARPENTER, C. L. and L. C. CANTLEY, 1990  Phosphoinositide kinases. Biochemistry 29:11147-11156[Medline].

CARPENTER, C. L. and L. C. CANTLEY, 1996  Phosphoinositide kinases. Curr. Opin. Cell Biol. 8:153-158[Medline].

CHEN, M. S., R. A. OBAR, C. C. SCHROEDER, T. W. AUSTIN, and C. A. POODRY et al., 1991  Multiple forms of dynamin are encoded by shibirie, a Drosophila gene involved in endocytosis. Nature 351:583-586[Medline].

CHONG, L. D., A. TRAYNOR-KAPLAN, G. M. BOKOCH, and M. A. SCHWARTZ, 1994  The small GTP-binding protein Rho regulates a phosphatidylinositol 4-phosphate 5-kinase in mammalian cells. Cell 79:507-513[Medline].

COCHET, C., O. FILHOL, T. PAYRASTRE, T. HUNTER, and G. N. GILL, 1991  Interaction between the epidermal growth factor receptor and phosphoinositide kinases. J. Biol. Chem. 266:637-644[Abstract/Free Full Text].

DE CAMILLI, P., S. D. EMR, P. S. MCPHERSON, and P. NOVICK, 1996  Phosphoinositides as regulators in membrane traffic. Science 271:1533-1539[Abstract].

DE SIMONE, S. M. and K. WHITE, 1993  The Drosophila erect wing gene, which is important for both neuronal and muscle development, encodes a protein which is similar to the sea urchin P3A2 DNA binding protein. Mol. Cell. Biol. 13:3641-3649[Abstract/Free Full Text].

DIVECHA, N. and R. F. IRVINE, 1995  Phospholipid signaling. Cell 80:269-278[Medline].

DIVECHA, N., H. BANFIC, and R. F. IRVINE, 1993  Inositides and the nucleus and inositides in the nucleus. Cell 74:405-407[Medline].

DYE, C. A., J.-K. LEE, R. C. ATKINSON, R. BREWSTER, and P.-L. HAN et al., 1998  The Drosophila sanpodo gene controls sibling cell fate and encodes a tropomodulin homolog, an actin/tropomyosin-associated protein. Development 125:1845-1856[Abstract].

FUKAMI, K., N. SAWADA, T. ENDO, and T. TAKENAWA, 1996  Identification of a phosphatidylinositol 4,5-bisphosphate binding site in chicken skeletal muscle -actinin. J. Biol. Chem. 271:2646-2650[Abstract/Free Full Text].

GAIDAROV, I., Q. CHEN, J. R. FALCK, K. K. REDDY, and J. H. KEEN, 1996  A functional phosphatidylinositol 3,4,5-trisphosphate/phosphoinositide binding domain in the clathrin adaptor AP-2 subunit. J. Biol. Chem. 271:20922-20929[Abstract/Free Full Text].

HAO, W., Z. TAN, K. PRASAD, K. K. REDDY, and J. CHEN et al., 1997  Regulation of AP-3 function by inositides: identification of phosphatidylinositol 3,4,5-trisphosphate as a potent ligand. J. Biol. Chem. 272:6393-6398[Abstract/Free Full Text].

HARTWIG, J. H., S. KUNG, T. KOVACSOVICS, P. A. JANMEY, and L. C. CANTLEY et al., 1996  D3 phosphoinositides and outside-in integrin signaling by glycoprotein IIb-IIIa mediate platelet actin assembly and filopodial extension induced by phorbol 12-myristate 13-acetate. J. Biol. Chem. 271:32986-32993[Abstract/Free Full Text].

HAY, J. C., P. L. FISETTE, G. L. JENKIN, K. FUKAMI, and T. TAKENAWA et al., 1995  ATP-dependent inositide phosphorylation is required for Ca²⁺-activated exocytosis. Nature 374:173-177[Medline].

ISHIHARA, H., Y. SHIBASAKI, N. KIZUKI, H. KATAGIRI, and Y. YAZAKI et al., 1996  Cloning of cDNAs encoding two isoforms of 68kD type I phosphatidylinositol-4-phosphate 5-kinase. J. Biol. Chem. 271:23611-23614[Abstract/Free Full Text].

JAN, L. Y. and Y. N. JAN, 1976  Properties of the larval neuromuscular junction in Drosophila melanogaster.. J. Physiol. 262:189[Abstract/Free Full Text].

JANMEY, P. A., 1994  Phosphoinositides and calcium as regulators of cellular actin assembly and disassembly. Annu. Rev. Physiol. 56:169-191[Medline].

KANIA, A., A. SALZBERG, M. BHAT, D. D'EVELYN, and Y. HE et al., 1995  P-element mutations affecting embryonic peripheral nervous system development in Drosophila melanogaster.. Genetics 139:1663-1678[Abstract].

KNIRR, S., A. SANTEL, and R. RENKAWITZ-POHL, 1997a  Expression of the PI4P 5-kinase Drosophila homologue skittles in the germline suggests a role in spermatogenesis and oogenesis. Dev. Genes Evol. 207:127-130.

KNIRR, S., S. BREUER, A. PAULULAT, and R. RENKAWITZ-POHL, 1997b  Somatic mesoderm differentiation and the development of a subset of pericardial cells depend on the not enough muscles (nem) locus, which contains the inscuteable gene and the intron located gene, skittles.. Mech. Dev. 67:69-81[Medline].

KRAUT, R. and J. A. CAMPOS-ORTEGA, 1996  Inscuteable, a neural precursor gene of Drosophila, encodes a candidate for a cytoskeleton adaptor protein. Dev. Biol. 174:65-81[Medline].

KRAUT, R., W. CHIA, L. Y. JAN, Y. N. JAN, and J. A. KNOBLICH, 1996  Role of inscuteable in orienting asymmetric cell divisions in Drosophila.. Nature 383:50-55[Medline].

LEEVERS, S. J., D. WEINKOVE, L. K. MACDOUGALL, E. HAFEN, and M. D. WATERFIELD, 1996  The Drosophila phosphoinositide 3-kinase Dp110 promotes cell growth. EMBO J. 15:6584-6594[Medline].

LITTLETON, J. T., M. STERN, M. PERIN, and H. J. BELLEN, 1994  Calcium dependence of neurotransmitter release and rate of spon(a)taneous vesicle fusions are altered in Drosophila synaptotagmin mutants. Proc. Natl. Acad. Sci. USA 91:10888-10892[Abstract/Free Full Text].

LOIJENS, J. C., I. V. BORONENKOV, G. J. PARKER, and R. A. ANDERSON, 1996  The phosphatidylinositol 4-phosphate 5-kinase family. Adv. Enzyme Regul. 36:115-140[Medline].

MARTIN, T. F. J., 1997  Phosphoinositides as spatial regulators of membrane traffic. Curr. Opin. Neurobiol. 7:331-338[Medline].

POLLARD, T., 1986  Rate constants for the reactions of ATP- and ADP-actin with the ends of actin filaments. J. Cell Biol. 103:2747-2754[Abstract/Free Full Text].

POLLARD, T. and D. RIMM, 1991  Analysis of cDNA clones for Acanthamoeba profilin-I and profilin-II shows end to end homology with vertebrate profilins and a small family of profilin genes. Cell Motil. Cytoskeleton 20:169-177[Medline].

RANA, R. S. and L. E. HOKIN, 1990  Role of phosphoinositides in transmembrane signalling. Physiol. Rev. 70:115-164[Free Full Text].

REN, X.-D., G. M. BOKOCH, A. TRAYNOR-KAPLAN, G. H. JENKINS, and R. A. ANDERSON et al., 1996  Physical association of the small GTPase rho with a 68 kd phosphatidylinositol 4-phosphate 5-kinase in Swiss 3T3 cells. Mol. Biol. Cell 7:435-442[Abstract].

ROBINOW, S. and K. WHITE, 1991  Characterization and spatial distribution of the ELAV protein during Drosophila melanogaster development. J. Neurobiol. 22:443-461[Medline].

RUIZ-GOMEZ, M. and M. BATE, 1997  Segregation of myogenic lineages in Drosophila requires numb. Development 124:4857-4866[Abstract].

SALIM, K., M. J. BOTTOMLEY, E. QUENFURTH, J. ZVELEBIL, and I. GOUT et al., 1996  Distinct specificity in the recognition of phosphoinositides by the pleckstrin homology domains of dynamin and Brutons tyrosine kinase. EMBO J. 15:6241-6250[Medline].

SAMBROOK, J., E. F. FRITSCH and T. MANIATIS, 1989 Molecular Cloning: A Laboratory Manual, Ed. 2. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SCHAFER, D. A., P. B. JENNINGS, and J. A. COOPER, 1996  Dynamics of capping protein and actin assembly in vitro: uncapping barbed ends by polyphosphoinositides. J. Cell Biol. 135:169-179[Abstract/Free Full Text].

SMYTHE, E., L. L. CARTER, and S. L. SCHMID, 1992  Cytosol and clathrin dependent stimulation of endocytosis in vitro by purified adaptors. J. Cell Biol. 119:1163-1171[Abstract/Free Full Text].

SPRADLING, A. C., 1993 Developmental Genetics of Oogenesis. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

STAEHLING-HAMPTON, K., P. D. JACKSON, M. J. CLARK, A. H. BRAND, and F. M. HOFFMANN, 1994  Specificity of bone morphogenetic protein-related factors: cell fate and gene expression changes in Drosophila embryos induced by decapentaplegic but not 60A. Cell Growth Differ. 5:585-593[Abstract].

STEWART, B. A., H. L. ATWOOD, J. J. RENGER, J. WANG, and C.-F. WU, 1994  Drosophila neuromuscular preparations in haemolymph-like physiological salines. J. Comp. Physiol. A Sens. Neural Behav. Physiol. 175:179-191[Medline].

TAUTZ, D. and C. PFEIFLE, 1989  A non-radioactive in situ hybridization method for the localization of specific RNAs in Drosophila embryos reveals translational control of the segmentation gene hunchback.. Chromosoma 98:81-85[Medline].

TILNEY, L. G., M. S. TILNEY, and G. M. GUILD, 1995  F actin bundles in Drosophila bristles. I. Two filament cross-links are involved in bundling. J. Cell Biol. 130:629-638[Abstract/Free Full Text].

TILNEY, L. G., P. CONNELLY, S. SMITH, and G. M. GUILD, 1996  F-Actin bundles in Drosophila bristles are assembled from modules composed of short filaments. J. Cell Biol. 135:1291-1308[Abstract/Free Full Text].

TOKER, A., M. MEYER, K. K. REDDY, J. R. FALCK, and R. ANEJA et al., 1994  Activation of protein kinase C family members by the novel polyphosphoinositides PtdIns-3,4-P₂ and PtdIns-3,4-P₃. J. Biol. Chem. 269:32358-32367[Abstract/Free Full Text].

VAESSIN, H., E. GRELL, E. WOLFF, E. BIER, and L. Y. JAN et al., 1991  prospero is expressed in neuronal precursors and encodes a nuclear protein that is involved in the control of axonal outgrowth in Drosophila. Cell 67:941-953[Medline].

VERHEYEN, E. M. and L. COOLEY, 1994  Profilin mutations disrupt multiple actin-dependent processes during Drosophila development. Development 120:717-728[Abstract].

WODRAZ, A., U. HINZ, M. ENGELBERT, and E. KNUST, 1995  Expression of crumbs confers apical character on plasma membrane domains of ectodermal epithelia of Drosophila.. Cell 82:67-76[Medline].

WU, M. N., J. T. LITTLETON, M. BHAT, A. PROKOP, and H. J. BELLEN, 1998  ROP, the Drosophila Sec1 homolog, interacts with syntaxin and regulates neurotransmitter release in a dose dependent manner. EMBO J. 17:127-139[Medline].

YAMAMOTO, A., D. B. DEWALD, I. V. BORONENKOV, R. A. ANDERSON, and S. D. EMR et al., 1995  Novel PI(4)P 5-kinase homologue, Fab1p, essential for normal vacuole function and morphology in yeast. Mol. Biol. Cell 6: