| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
b,
Christophe Hitte4,a, and
Leonard Rabinowa,c
Corresponding author: Leonard Rabinow, Laboratoire d'Embryologie Moléculaire, Bâtiment 445, Université de Paris XI, 91405 Orsay Cedex, France., lenny.rabinow{at}emex.u-psud.fr (E-mail)
Communicating editor: J. A. BIRCHLER
 | ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Activity of the Darkener of apricot (Doa) locus of Drosophila melanogaster is required for development of the embryonic nervous system, segmentation, photoreceptor maintenance, normal transcription, and sexual differentiation. The gene encodes a protein kinase, with homologues throughout eukaryotes known as the LAMMER kinases. We show here that DOA is expressed as at least two different protein isoforms of 105 and 55 kD throughout development, which are primarily localized to the cytoplasm and nucleus, respectively. Doa transcripts and protein are expressed in all cell types both during embryogenesis and in imaginal discs. Although it was recently shown that DOA kinase is essential for normal sexual differentiation, levels of both kinase isoforms are equal between the sexes during early pupal development. The presence of the kinase on the cell membrane and in the nuclei of polytene salivary gland cells, as well as exclusion from the nuclei of specific cells, may be indicative of regulated kinase localization. Mosaic analysis in both the soma and germline demonstrates that Doa function is essential for cell viability. Finally, in contrast to results reported in other systems and despite some phenotypic similarities, genetic data demonstrate that the LAMMER kinases do not participate in the ras-MAP kinase signal transduction pathway.
THE LAMMER protein kinase family possesses homologues in all eukaryotes, including Drosophila, mice, Saccharomyces cerevisiae, and humans (
LAMMER kinases autophosphorylate on Ser/Thr and Tyr residues ("dual specificity";
Understanding of LAMMER kinase function within the context of a developing organism depends largely upon analysis of the Drosophila family member, encoded at the Darkener of apricot (Doa). Doa is an essential gene whose mutations were isolated during screens for dosage-sensitive regulatory loci (
In addition to suppression of copia insertion phenotypes, Doa mutants possess pleiotropic defects, although homozygotes derived from heterozygous females develop normally and die during early larval stages (
Failure of the embryonic nervous system to elaborate and the specific degeneration of retinal photoreceptors in mutants possessing residual Doa activity, along with loss of adult cuticular structures such as bristles and occasionally ocelli, suggest the existence of a requirement for higher Doa activity in the nervous system relative to other tissues. Since even homozygotes dying as embryos possess residual maternally contributed Doa activity, we have hypothesized a general role for LAMMER kinases in the initiation of differentiation. Furthermore, the degeneration of retinal photoreceptors in heteroallelic animals demonstrates a continued requirement for the gene's activity for cellular viability. Thus, we believe that LAMMER kinases' function is essential for cellular viability and differentiation.
Support for the latter role interpretation is derived from the identification of PK12, a tobacco LAMMER kinase, whose transcription and activity are induced in response to the hormone ethylene (
The LAMMER kinase-encoding locus, KNS1, from the yeast S. cerevisiae, would be amenable to genetic analysis. However, it is not esssential for vegetative growth, meiosis, or sporulation (
The only LAMMER kinase substrates so far identified in vivo are the SR proteins. This conserved eukaryotic protein family influences RNA splice site selection and other aspects of RNA metabolism (for reviews, see
To further characterize the function of Doa and its products, we have examined its expression patterns, as well as requirements for its activity in mosaic animals. Here we report that at least two different isoforms of DOA protein are expressed in all tissues and developmental stages examined. One of these is cytoplasmic while the other is primarily nuclear, as revealed in part through subcellular fractionation experiments. Analysis of Doa transcript and protein levels and expression patterns during embryogenesis and in imaginal discs reveals ubiquitous expression, while both somatic and germline mosaic analyses demonstrate that Doa activity is essential for cellular viability. Finally, although several pleiotropic phenotypes of Doa mutations resemble those of mutations in components of the ras-mitogen-activated protein kinase (MAP-K) signal transduction pathway, no genetic interactions were observed in double mutants, suggesting that Doa acts independently of this well-characterized signal transduction cascade.
| MATERIALS AND METHODS |
|---|
| TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Drosophila stocks and crosses:
Drosophila melanogaster were raised on standard cornmeal medium at 25°. Doa mutant strains and heteroallelic combinations were previously described (
The deficiency chromosome Df(3R) 3450 (98E3:99A6–8) was examined as a possible deletion of the Doa locus, and the lacZ-marked P-element-induced recessive lethal l(3)01705 at 98F1–4 from the Berkeley Drosophila Genome Project (SPRAD- LING et al. 1995) was examined as a possible Doa allele. ß-Galactosidase expression was examined in embryos, imaginal discs, and larval tissues using assays for activity (
Generation of somatic mosaic clones:
Stocks for the induction of somatic and germline mosaics are listed (Table 1). Somatic clones of three Doa alleles (Doa
3B, DoaEMS2, and DoaI5) were generated using FRT/FLP site-specific recombination (
|
Three alleles were chosen for the generation of mosaic clones on the basis of availability of cytological, genetic, and molecular data, as follows. Doa3B is a null, because mutants possess only 50% of wild-type Doa 2.7-kb mRNA levels encoding an active kinase and because it only poorly complements two Doa alleles (DoaMsu2 and DoaRemA), which complement the vast majority of other Doa mutations (C. SO, B. YUN and L. RABINOW, unpublished results). The 3B allele is due to a small cytologically visible inversion and possesses a breakpoint directly in the gene, upstream of the region encoding the kinase catalytic domain (
These three Doa alleles (98F) were recombined onto a third chromosome carrying a FRT at 82B. Mutant cell clones were marked in repulsion with white+ in eyes and were linked to ebony. Clones were produced by collecting eggs for 12 hr, aging them for between 24 hr and 6 days, and incubating them in a 38° water bath for 60 min. Cultures were then returned to 25°. Homozygous mutant eye clones were identified as patches lacking pigment, and cuticle clones were identified by homozygosity for ebony. A large sample of individuals in which clones had been induced was examined (>50 for each genotype). These experiments were repeated several times, with clones induced at various developmental stages. In examination of mosaic imaginal discs, 6-day-old larvae were heat shocked, returned to 25° for 2–3 hr, and dissected on ice.
Generation of germline mosaic clones:
y w-; P[FRT], e Doa3B/TM3 female virgins were crossed with w-; P[w+, ovoD1,11.3b]/+ males, a gift of Maryvonne Ninio (3B/P[w+, ovoD1,11.3b] virgins were crossed with y w; Doa3B/TM3 in 21 vials. As a control, wild-type females were crossed with male w-, ovo0: P[w+, ovoD1,11.3b] and F1 irradiated as above. A total of 279 irradiated y w/w; +/P[w+, ovoD1,11.3b] virgins were crossed with +/+ males. Females in 13/21 vials (134 flies) laid eggs, indicating the successful induction of germline clones.
In situ hybridization to whole embryos:
The procedure of Pfeifle and Tautz was followed for nonradioactive in situ hybridization to transcripts in whole mount embryos (
Bacterial fusion proteins and antibody generation:
Antibodies against the affinity-purified 85-kD catalytic domain DOA fusion protein (
Anti-DOA fusion sera were affinity purified on Affi-gel 10 and Affi-gel 15 according to instructions (Bio-Rad, Hercules, CA). Nonspecific antibodies were removed by chromatography through a mixture of Affi-gel 10 and 15 coupled with crude bacterial extracts in which the pMAL vector had been expressed. Antibodies specific to the maltose-binding protein (MBP)-DOA fusion protein were bound on Affi-gel 10 and eluted with 0.1 M glycine, pH 2.5.
An octameric peptide (
Histology and immunocytochemistry:
Immunohistochemical staining of dissected imaginal discs and embryos was performed as described (
Protein preparation and pupal nuclear extracts:
Embryos, third instar larvae, 0- to 1-day-old pupae, and 0- to 2-day-old adults were collected and washed in phosphate-buffered saline. Sexed pupal extracts were prepared by sorting third instar larvae prior to pupariation and aging them. Individual tissues were hand dissected. Samples were homogenized in SDS gel loading buffer and boiled. Pupal nuclear extracts were prepared according to
SDS gel electrophoresis, protein transfers, and immunoblots:
SDS gel electrophoresis, protein transfers, and immunoblots were performed as described (
-tubulin monoclonal antibody DM1A (Sigma, St. Louis). Antibody dilutions for immunoblot analyses were similar to those described for immunocytochemical analyses.
| RESULTS |
|---|
| TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Immunoblot analysis of DOA expression during development reveals at least two protein isoforms:
Affinity-purified anti-DOA antibodies and crude antipeptide sera directed against the LAMMER motif reveal the size and developmental expression of reacting proteins on immunoblots. Proteins of 55 and 105 kD are detected at all developmental stages (Fig 1A and Fig B), including 0- to 24-hr embryos, third instar larvae, 0- to 24-hr pupae, and adults, as well as in extracts of 0- to 4-hr embyros, larval eye-antennal imaginal discs, salivary glands, fat bodies, adult heads, and ovaries (not shown). These proteins are recognized by both crude and affinity-purified anti-DOA serum (Fig 1A and Fig B), as well as the anti-LAMMER serum (Fig 1D, lanes 3 and 4). For unclear reasons, the affinity-purified anti-DOA serum produces relatively stronger signal against the 105-kD isoform than against the 55-kD isoform (Fig 1A; compare lanes 1–3 with 4–6). Additional minor bands are also recognized by these sera, including the affinity-purified one. We believe that some, if not all, of these are degradation products, due to their variability among transfers, although it remains possible that these bands represent minor DOA isoforms.
|
The 55-kD protein is consistent in size with that of the kinase deduced from the sequence of cDNA clones deriving from a 2.7-kb Doa transcript expressed throughout development (
The authenticity of the 105-kD protein as an independent protein isoform, as opposed to being due to post-translational modification, was examined by treating pupal extracts with urea (final concentration 6 M), in addition to SDS, to exclude the possibility that the 105-kD protein is a dimer of the 55-kD protein resistant to SDS and reducing conditions. No differences with control samples were observed (not shown). Also, treatment of embryonic and pupal extracts with either calf-intestinal phosphatase or potato-acid phosphatase for 30 min at 37° did not produce any differences in levels or mobility of the 55- or 105-kD isoforms (not shown). However, this analysis does not eliminate the possibility that these apparently different proteins are not due to other post-translational modifications, for example carbohydrate modification.
Doa mutations reduce levels of the 55- and 105-kD proteins:
To confirm that the 55- and 105-kD proteins are both encoded by Doa, we examined their levels in extracts of mutant pupae. Pupae were used because selection of rare homozygous or heteroallelic adults expressing higher than average DOA levels might be responsible for the survival of the rare individuals escaping recessive lethality. This phenomenon was in fact observed, since pupal extracts of the 105/HD genotype possess significantly less DOA protein than wild type, while adults of this genotype escaping lethality possess more nearly normal levels (Fig 1B, lane 3 vs. lane 4).
Levels of both the 55- and 105-kD proteins are visibly reduced in at least two of the three Doa mutant combinations relative to all three wild-type strains examined (Fig 1B). It is not surprising that DOA protein is detected in the mutant extracts, since activity of the gene is required for viability, and these alleles are necessarily hypomorphic. Since affinity-purified anti-DOA antibodies detect the 105-kD protein, and two of three mutant Doa genotypes examined affect its quantity, we conclude that both the 55- and 105-kD proteins are authentic DOA isoforms.
Expression of DOA is equivalent between the sexes:
As mentioned, Doa alleles subtly alter sexual differentiation, as well as alter the splicing of mRNA from the dsx locus, a crucial component of somatic sex determination (
Differential subcellular localization of the 55- and 105-kD isoforms:
Immunoblot analysis of nuclear and whole pupal cell extracts was used to determine the intracellular localization of the two DOA isoforms, since our antisera do not differentiate between them. Normalization of the loading of each sample was accomplished via two methods, in which (1) the total amount of protein loaded was equivalent between both samples, as well as (2) the amount of the 55-kD signal was normalized between the two.
The results using crude anti-DOA, affinity-purified anti-DOA, and anti-LAMMER sera show that the 105-kD protein is found only in whole-cell extracts, while it is missing in nuclear fractions (Fig 1D). In contrast, the 55-kD isoform is found in both nuclear and whole-cell extracts. Only 
5 µg of nuclear extract was needed to detect the 55-kD protein, whereas 20 µg of total protein was required to detect the cytoplasmic form, suggesting that the bulk of the kinase is nuclearly localized. This analysis does not exclude the possibility that some of the 55-kD protein is cytoplasmically localized in all cells, or all of it is cytoplasmic in certain cell types.
Doa transcript accumulation during embryogenesis and in imaginal discs:
The pattern of Doa transcription was examined during embryogenesis and in imaginal discs by in situ hybridization using a 2.0-kb cDNA probe (
|
In the stage 1 precellular blastoderm (stages as per
Light but relatively uniform staining is observed in the eye and wing imaginal discs (Fig 2E and Fig F). Apparently higher localized transcript concentrations in the figure are due to folds in the monolayer of disc cells, doubling the apparent Doa transcript signal in these regions.
DOA protein expression during embryogenesis and in imaginal discs:
DOA protein expression was examined via immunohistochemical staining with affinity-purified anti-DOA antiserum. Consistent with the pattern of transcript expression and pleiotropic phenotypes, DOA protein is widely expressed (Fig 3) in patterns nearly identical with the transcription patterns observed (e.g., compare Fig 2C and Fig 3B, and Fig 2D and Fig 3C). Unlike the segmental pattern of transcript accumulation imposed on a virtually ubiquitous background (Fig 2C), no segmental variation in the localization of DOA protein was observed (Fig 3B). At stage 17, expression of DOA in the CNS and brain is noticeably heavier than surrounding tissue (Fig 3D). Early embryos (precellular blastoderm) stained with affinity-purified antibodies reveal essentially cytoplasmic staining (not shown), but the higher affinity of the purified sera for the cytoplasmic 105-kD DOA isoform demonstrated above (Fig 1) may significantly reduce nuclear staining. Virtually identical embryonic staining patterns were observed using the crude anti-DOA serum, as well as the antipeptide (anti-LAMMER) serum described above, while preimmune sera revealed no labeling of embryos or discs (not shown), demonstrating the specificity of the signals as due to DOA.
|
DOA protein is ubiquitously expressed in imaginal discs (Fig 3E). Slightly higher DOA protein levels are observed in the differentiating third instar eye imaginal disc posterior to the morphogenetic furrow (Fig 3F), consistent with Doa's demonstrated role in photoreceptor differentiation and maintenance. It should also be noted that relatively higher levels of DOA expression are observed in Bolwig's nerve (Fig 3F, arrow), which innervates the larval visual organ, demonstrating both that relatively higher levels of DOA are found in neural cells as well as the presence of the kinase in axonal cytoplasm.
Fig 3G shows a portion of a late third instar salivary gland stained with affinity-purified anti-DOA. Clear labeling of nuclei, the cell surface, and a cytoplasmic meshwork is observed, easily visible in a higher magnification view of a salivary gland cell (Fig 3H). This mesh may correspond to a component of the cytoskeleton, and the same pattern is revealed when these cells are labeled with antitubulin (R. FARKA and L. RABINOW, unpublished results). However, these cells are producing high levels of glue protein at this stage, which is being stored in vacuoles prior to secretion. Thus the entire cytoplasm of the polytene salivary gland cells may be restricted to the "mesh" observed, and all its components might be erroneously thought to co-localize. Further experiments will be required to determine the origin of this mesh.
In contrast to the polytene glue-producing cells, the cytoplasm at the anterior end of the salivary gland in the imaginal ring is intensely and uniformly labeled. Staining in these cells is excluded from the nuclei, suggesting that the intracellular localization of DOA is regulated.
Identification of deletion and enhancer-trap Doa alleles:
The deficiency chromosome Df(3R) 3450 (98E3:99A6-8) was examined as a possible deletion of the Doa locus, while the lacZ-marked P-element-induced recessive lethal l(3) 01705 at 98F1–4 from the Berkeley Drosophila Genome Project (
|
ß-Galactosidase activity in embryos and imaginal discs of the l(3)01705 line was detected in only a small set of tissues. However, immunohistochemical staining with anti-ß-galactosidase revealed widespread expression (not shown), consistent with the labeling observed with anti-DOA serum.
We also tested an enhancer of glass, E(gl)3C, which maps by recombination to the same location as Doa (
An essential role for Doa in the soma:
To analyze the effects of complete loss of Doa function on the survival or differentiation of individual cell types and to determine whether Doa mutations act cell autonomously, homozygous mutant Doa clones were generated by mitotic recombination in the soma. Clonal analysis was performed using three alleles, Doa3B, DoaEMS2, and DoaI5 (Table 1), which were chosen for reasons explained in MATERIALS AND METHODS.
When somatic recombination was induced in Doa3B/+ heterozygotes 25–36 hr after egg laying (AEL), no mutant clones were detected in either eyes or cuticles, although +/+ clones were recovered in mosaic eyes. The morphology of eyes and cuticles was normal in these individuals. Cell clones were therefore induced, but the cells did not produce adult cuticle. To determine whether homozygous cell clones would survive and affect the development of specific cell types if induced later during development, mitotic recombination was induced every 12 hr from early first instar to early pupal periods (7 days AEL). When somatic recombination was induced in 4- to 6-day-old larvae (late second through third instar), 3B mosaic eyes containing +/+ clones were often rough. Sagittal sections of these mosaic eyes revealed randomly missing photoreceptors and pigment cells, although no loss of specific cell types was observed. Even in individuals with apparently normal eyes, sectioning revealed occasional defects in the organization of the retina, presumably due to death of homozygous mutant cells (Fig 4A, black arrow). Ommatidia near homozygous wild-type clones for Doa were often missing random photoreceptors (Fig 4A, white arrow). Flies were also recovered with one or more bristles missing in the triple row of the anterior wing margin (Fig 4B), suggesting that genetic ablation of one or more of the progenitor cells in the lineage leading to these bristles had occurred on clone induction. Thus, no Doa3B homozygous clones were detected in mosaic eyes or cuticles, indicating that mutant cell clones induced even late in third instar periods die.
|
To determine how quickly the putative cell death occurred in Doa3B/Doa3B homozygous cells, development in mosaic eye discs was examined using a monoclonal antibody recognizing the nuclear neuronal antigen ELAV, which is expressed immediately posterior to the morphogenetic furrow in eye imaginal discs (3B/Doa3B mitotic clones, one or more photoreceptor precursor cells were missing in four (Fig 4C, arrow), one or two rows behind the morphogenetic furrow (Fig 4C, arrowhead), as revealed by lack of ELAV expression interrupting the ordered array of differentiating neuronal cells. The ordered array of photoreceptor clusters was seen without any such defects in an equal number of heat-shocked wild-type eye disc controls (not shown). This finding suggests that homozygous mutant Doa photoreceptor precursor cells died almost immediately following clone induction, or were unable to differentiate as neuronal photoreceptor precursors, since the morphogenetic furrow requires roughly 2 hr to generate each row.
When mitotic recombination was induced in 5- to 6-day-old larvae heterozygous for DoaI5, no white- (i.e., homozygous Doa-) clones were detected in the eye. Moreover, mosaic eyes containing +/+ clones were rough. Sagittal eye sections of these animals show disruptions in the ommatidial pattern (Fig 4E, arrow), near homozygous wild-type cell clones (Fig 4E, arrowhead), identical to the defects observed in Doa heteroallelic flies (
In contrast to the results with the 3B and I-5 alleles, homozygous cell clones of the EMS2 allele were recovered in eyes and cuticles when mitotic recombination was induced during first instar and later larval periods. Sagittal eye sections containing these clones show essentially normal retinal organization (e.g., Fig 4D, arrow), adjoining homozygous wild-type clones (Fig 4D, arrowhead). Cuticular EMS2 clones marked with ebony were also recovered (not shown), demonstrating survival of EMS2 homozygous cells without developmental repercussions. On the basis of the normal appearance of clones both in the cuticle and eye, we conservatively conclude that this allele produces homozygous viable and normal mutant Doa clones, probably due to being a hypomorphic as opposed to a null allele.
Synthesizing the data obtained with the 3B, I5, and EMS2 alleles, Doa appears necessary for the proper development of all cuticular structures, including the eye, wing, and thorax. This result suggests that the gene is essential for cell viability and is consistent with its ubiquitous expression in embryos and imaginal discs.
Doa is essential in the female germline:
A maternal contribution of Doa function to the embryo had been deduced from the observation that reciprocal crosses between Doa alleles yielding heteroallelic adults produce different percentages of trans-heterozygous adult escapers (
To analyze the effects of homozygosity for Doa mutations, we induced germline clones to eliminate all maternal contributions (3B heterozygotes. Progeny of this cross were irradiated as first instar larvae. Adult female Doa3B/P[w+, ovoD1,11.3b] deriving from the irradiated progeny were crossed with Doa3B/TM3 males. Among 290 females crossed in 21 vials, females in only 1 vial laid eggs, and these developed normally. These progeny were probably due to leakiness of the ovoD1,11.3b allele (3B/P[w+, ovo D1,11.3b] females revealed four individual ovaries even more rudimentarily developed than the minimal development observed in P[w+, ovoD1,11.3b] heterozygotes, presumably due to the existence of Doa clones and the necessity of their activity for development of the germline. Therefore, stem cell clones homozygous for Doa3B were generated by X-ray irradiation, but were unable to produce eggs due to a requirement for Doa function.
This finding was confirmed by the Perrimon laboratory with a second Doa allele during screens for zygotic lethals with maternal-effect phenotypes (3B allele and demonstrates that even residual levels of Doa activity are insufficient to support normal oogenesis, since the survival of some heteroallelic Doa combinations with l(3) 01705 demonstrates it is a hypomorph (Table 2).
Doa acts independently of the ras-MAP kinase pathway:
Adult heteroallelic and homozygous Doa mutants of various classes display a number of phenotypes similar to those of mutations in components of the ras-MAP-K pathway (
Ras-MAP-K pathway components tested included hypomorphic or null alleles of ras, rolled (MAP kinase), sevenless, seven-up, and sina. Other hypomorphic or null alleles tested included glass and a deficiency including Roughened, a GTP-binding protein homologous to mammalian rap (
| DISCUSSION |
|---|
| TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The ubiquitous expression of Doa mRNA and DOA protein is consistent with the essential role for activity of the gene demonstrated in clonal analysis of both somatic and germline tissues. The finding of low but detectable levels of the kinase in all cells of the eye-imaginal disc, with increased expression posterior to the morphogenetic furrow, is also consistent with previously described phenotypes of Doa mutants. These observations both lend support to the hypothesis that DOA kinase is essential for the initiation and maintenance of cellular differentiation, but the exact pathway(s) implicated remain to be identified.
Developmental profiles on immunoblots demonstrate that DOA protein is expressed as 55- and 105-kD isoforms throughout development in all tissues examined. One or both isoforms are reduced in quantity in various mutants, verifying their origin as the Doa locus. These isoforms are not interconvertible by treatment with urea or phosphatase, suggesting that they possess different primary structures. The 55-kD protein is essentially nuclear, while the 105-kD protein is cytoplasmically restricted, as demonstrated on immunoblots of fractionated pupal extracts. Expression of the 105-kD protein in early embryos and ovaries suggests that DOA protein as well as RNA is maternally contributed to the developing oocyte. Further studies are required to examine the dynamics of subcellular localization of the 55-kD isoform during development and in different tissues. Determination of the structure and kinase activity of the 105-kD isoform will provide further insight into DOA function, but even this level of analysis will not reveal where and when the kinase is catalytically active. The observation that expression of both kinase isoforms is essentially equal between the sexes suggests that post-translational regulation of Doa activity may occur to influence sex-specific splicing, since Doa alleles affect sexual differentiation and splicing of dsx.
In support of our observations of the existence of multiple DOA isoforms, the Berkeley Drosophila Genome Project (BDGP; see http://www.fruitfly.org/) has recovered multiple DOA cDNAs (clot 2594), as well as sequenced the entire gene (70 amino acids at the N terminus of the protein. Whether the mRNAs corresponding to this subclot account for one or more of the additional signals occasionally seen on our immunoblots requires further analysis. However, based on the size of the potential protein product, it would seem unlikely that the mRNA defining subclot 1 is responsible for the 105-kD protein observed. The third subclot listed at BDGP is represented by a single, apparently partial cDNA, since alignment of its putative translation product demonstrates that it lacks virtually all sequences N-terminal to the kinase catalytic domain.
Preliminary analysis of the Berkeley genomic Doa sequence has identified several putative ORFs within the large introns of the gene, which is at least 40 kb in size, on the basis of the span of exons of the mRNA encoding the 55-kD protein. At least one of these appears to be expressed, since we recently recovered a cDNA including one of these alternative exons (L. CHANEY, B. YUN and L. RABINOW, unpublished reults).
Our studies suggest that the kinase is localized to specific cytoplasmic structures, i.e., a mesh and the cell surface in the polytene cells of the salivary gland. DOA is excluded from the nuclei of specific cells in the salivary gland, further suggesting the intriguing possibility that intracellular localization of the kinase is regulated. The generation of antibodies specific for the two different DOA isoforms, as well as additional staining of embryonic and larval tissues, will be necessary to further examine the significance of these observations. One obvious implication is that DOA participates in the transmission of signals from the cytoplasm to the nucleus.
Murine Clk1, which encodes a LAMMER kinase, expresses multiple transcripts (
The differential function(s) of the two DOA isoforms, if any, have yet to be identified. It is clear that LAMMER kinases phosphorylate SR proteins, both in vitro (
The similarities in phenotypes between Doa alleles and mutations in various loci encoding proteins in the MAP kinase cascade, as well as a report suggesting that the murine LAMMER kinase Clk1 influenced MAP kinase pathway activity (
In summary, our data reveal the necessity of DOA activity for cell survival and differentiation. The existence of multiple protein isoforms with localization to specific intracellular compartments and structures raises the distinct possibility that DOA and the other LAMMER kinases possess multiple functions in the relay of as yet unidentified signals during development, to regulate vital cellular processes in addition to the regulation of alternative splicing of pre-mRNAs.
| FOOTNOTES |
|---|
1 These authors contributed equally to this work. 
2 Present address: Department of Biological Sciences, Basic Science Institute, Dankook University, Cheonan City, Chungnam 330-714, South Korea.
3 Present address: The Fox Chase Cancer Center, 7701 Burholme Ave., Philadelphia, PA 19111.
4 Present address: CNRS UPR41 Recombinaisons Génétiques, 2, rue du Pr. Leon Bernard, 35043 Rennes Cedex, France.
| ACKNOWLEDGMENTS |
|---|
We especially thank Lee Simon for generous help and instruction in tissue sectioning and Dick Leidich for his expertise and assistance with the X-ray machine. Thanks are also due to Maryvonne Mevel-Ninio, who generously provided the P[ovoD] on 3R stock prior to publication; Norbert Perrimon for communicating unpublished data; Marie-Laure Samson for the gift of anti-ELAV antibody and helpful suggestions throughout this work; Ernst Hafen, Kevin Moses, and Yosh Hiromi for Drosophila stocks; and Dale Dorsett for the pupal nuclear preparation protocol. We also benefited from the availability of both genomic and EST clone sequences made accessible by the Berkeley Drosophila Genome Project. Cheng Du generously provided the Western blot of DOA protein in extracts of sexed Drosophila pupae. Additional Drosophila stocks were also provided by the Bowling Green and Indiana Drosophila stock centers. This work was supported by grants to L.R. from the National Science Foundation, the Margaret Colyer Research Fund of the Fight for Sight—Research Division of Prevent Blindness America, and the New Jersey Commission on Cancer Research. R.F. was supported in part by grants VEGA 95/5305/272 and SO 95/5305/043 from the Slovak Grant Agency for Science and a supplement from the office for Central European Programs, U.S. National Science Foundation.
Manuscript received April 3, 2000; Accepted for publication June 21, 2000.
| LITERATURE CITED |
|---|
| TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE 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 287:2185-2194
ASHBURNER, M., 1989a Drosophila: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
ASHBURNER, M., 1989b Drosophila: A Laboratory Handbook. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
AUSUBEL, F. M., R. BRENT, R. E. KINGSTON, D. D. MOORE, J. G. SEIDMAN et al., 1989 Current Protocols Molecular Biology. Greene/Wiley Publishing, New York.
BEN-DAVID, Y., K. LETWIN, L. TANNOCK, A. BERNSTEIN, and T. PAWSON, 1991 A mammalian protein kinase with potential for serine/threonine and tyrosine phosphorylation is related to cell cycle regulators. EMBO J. 10:317-325[Medline].
BRIGGS, W. H., K. H. ZAVITZ, B. DICKSON, A. VAN DER STRATEN, and D. BRUNNER et al., 1994 The Drosophila rolled locus encodes a MAP kinase required in the sevenless signal transduction pathway. EMBO J. 13:1628-1635[Medline].
BRUNNER, D., N. OELLERS, J. SZABAD, W. H. I. BIGGS, and S. L. ZIPURSKY et al., 1994 A gain of function mutation in Drosophila MAP kinase activates multiple receptor tyrosine kinase signalling pathways. Cell 76:875-888[Medline].
CAMPOS-ORTEGA, J. A., and V. HARTENSTEIN, 1997 The Embryonic Development of Drosophila melanogaster. Springer, Berlin.
CAO, W., S. F. JAMISON, and M. A. GARCIA-BLANCO, 1997 Both phosphorylation and dephosphorylation of ASF/SF2 are required for pre-mRNA splicing in vitro. RNA 3:1456-1467[Abstract].
COLWILL, K., L. L. FENG, J. M. YEAKLEY, G. D. GISH, and J. F. CACERES et al., 1996a SRPK1 and Clk/STY protein kinases show distinct substrate specificities for serine/arginine-rich splicing factors. J. Biol. Chem. 271:24569-24575
COLWILL, K., T. PAWSON, B. ANDREWS, J. PRASAD, and J. L. MANLEY et al., 1996b The Clk/STY protein kinase phosphorylates SR splicing factors and regulates their intranuclear distribution. EMBO J. 15:265-275[Medline].
COX, K. H., D. V. DELEON, L. M. ANGERER, and R. C. ANGERER, 1984 Detection of mRNA in sea urchin embryos by in situ hybridization using asymmetric RNA probes. Dev. Biol. 101:485-502[Medline].
DIAZ-BENJUMEA, F. and E. HAFEN, 1994 The sevenless signaling cassette mediates Drosophila EGF receptor function during epidermal development. Development 120:569-578[Abstract].
DICKSON, B., F. SPRENGER, D. MORRISON, and E. HAFEN, 1992 Raf functions downstream of RAS1 in the sevenless signal transduction pathway. Nature 360:600-603[Medline].
DICKSON, B. J., A. VAN DER STRATEN, M. DOMINGUEZ, and E. HAFEN, 1996 Mutations modulating raf signaling in Drosophila eye development. Genetics 142:163-171[Abstract].
DORSETT, D., 1990 Potentiation of a polyadenylation site by a downstream protein-DNA interaction. Proc. Natl. Acad. Sci. USA 87:4373-4377
DU, C., M. E. MCGUFFIN, B. DAUWALDER, L. RABINOW, and W. MATTOX, 1998 Protein phosphorylation plays an essential role in the regulation of alternative splicing and sex determination in Drosophila. Mol. Cell 2:741-750[Medline].
DUNCAN, P. I., B. W. HOWELL, R. M. MARIUS, S. DRMANIC, and E. M. J. DOUVILLE et al., 1995 Alternative splicing of STY, a nuclear dual specificity kinase. J. Biol. Chem. 270:21524-21531
GUI, J. F., W. S. LANE, and X. D. FU, 1994a A serine kinase regulates intracellular localization of splicing factors in the cell cycle. Nature 369:678-682[Medline].
GUI, J. F., H. TRONCHERE, S. D. CHANDLER, and X. D. FU, 1994b Purification and characterization of a kinase specific for the serine and arginine-rich pre mRNA splicing factors. Proc. Natl. Acad. Sci. USA 91:10824-10828
HARIHARAN, I. K., R. W. CARTHEW, and G. M. RUBIN, 1991 The Drosophila Roughened mutation: activation of a rap homolog disrupts eye development and interferes with cell determination. Cell 67:717-722[Medline].
HOWELL, B. W., D. E. H. AFAR, J. LEW, E. M. J. DOUVILLE, and P. L. E. ICELEY et al., 1991 STY, a tyrosine-phosphorylating enzyme with sequence homology to serine/threonine kinases. Mol. Cell. Biol. 11:568-572
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].
KUROYANAGI, N., H. ONOGI, T. WAKABAYASHI, and M. HAGIWARA, 1998 Novel SR-protein-specific kinase, SRPK2, disassembles nuclear speckles. Biochem. Biophys. Res. Commun. 242:357-364[Medline].
