FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2021818985> ?p ?o ?g. }

• W2021818985 endingPage "2318" @default.
• W2021818985 startingPage "2308" @default.
• W2021818985 abstract "Article15 April 1998free access A phosphorylation site in the Ftz homeodomain is required for activity Jianli Dong Jianli Dong Banting and Best Department of Medical Research, University of Toronto, C.H.Best Institute, Toronto, Ontario, Canada, M5G 1L6 Search for more papers by this author Ling-Hong Hung Ling-Hong Hung Banting and Best Department of Medical Research, University of Toronto, C.H.Best Institute, Toronto, Ontario, Canada, M5G 1L6 Search for more papers by this author Robert Strome Robert Strome Banting and Best Department of Medical Research, University of Toronto, C.H.Best Institute, Toronto, Ontario, Canada, M5G 1L6 Search for more papers by this author Henry M. Krause Corresponding Author Henry M. Krause Banting and Best Department of Medical Research, University of Toronto, C.H.Best Institute, Toronto, Ontario, Canada, M5G 1L6 Search for more papers by this author Jianli Dong Jianli Dong Banting and Best Department of Medical Research, University of Toronto, C.H.Best Institute, Toronto, Ontario, Canada, M5G 1L6 Search for more papers by this author Ling-Hong Hung Ling-Hong Hung Banting and Best Department of Medical Research, University of Toronto, C.H.Best Institute, Toronto, Ontario, Canada, M5G 1L6 Search for more papers by this author Robert Strome Robert Strome Banting and Best Department of Medical Research, University of Toronto, C.H.Best Institute, Toronto, Ontario, Canada, M5G 1L6 Search for more papers by this author Henry M. Krause Corresponding Author Henry M. Krause Banting and Best Department of Medical Research, University of Toronto, C.H.Best Institute, Toronto, Ontario, Canada, M5G 1L6 Search for more papers by this author Author Information Jianli Dong1, Ling-Hong Hung1, Robert Strome1 and Henry M. Krause 1 1Banting and Best Department of Medical Research, University of Toronto, C.H.Best Institute, Toronto, Ontario, Canada, M5G 1L6 *Corresponding author. E-mail: [email protected] The EMBO Journal (1998)17:2308-2318https://doi.org/10.1093/emboj/17.8.2308 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The Drosophila homeodomain-containing protein Fushi tarazu (Ftz) is expressed sequentially in the embryo, first in alternate segments, then in specific neuroblasts and neurons in the central nervous system, and finally in parts of the gut. During these different developmental stages, the protein is heavily phosphorylated on different subsets of Ser and Thr residues. This stage-specific phosphorylation suggests possible roles for signal transduction pathways in directing tissue-specific Ftz activities. Here we show that one of the Ftz phosphorylation sites, T263 in the N-terminus of the Ftz homeodomain, is phosphorylated in vitro by Drosophila embryo extracts and protein kinase A. In the embryo, mutagenesis of this site to the non-phosphorylatable residue Ala resulted in loss of ftz-dependent segments. Conversely, substitution of T263 with Asp, which is also non-phosphorylatable, but which successfully mimics phosphorylated residues in a number of proteins, rescued the mutant phenotype. This suggests that T263 is in the phosphorylated state when functioning normally in vivo. We also demonstrate that the T263 substitutions of Ala and Asp do not affect Ftz DNA-binding activity in vitro, nor do they affect stability or transcriptional activity in transfected S2 cells. This suggests that T263 phosphorylation is most likely required for a homeodomain-mediated interaction with an embryonically expressed protein. Introduction Homeodomain proteins constitute a large family of eukaryotic transcription factors that share a common 60 amino acid DNA-binding domain referred to as the homeodomain (reviewed in Gehring et al., 1994a; McGinnis, 1994; Sharkey et al., 1997). In all higher eukaryotes, homeodomain proteins play a major role in patterning the embryonic body plan. Loss-of-function or gain-of-function mutations cause transformations in regional identity that can be quite spectacular. In Drosophila, for example, antennae can be transformed into legs, eyes or mouth parts (Schneuwly et al., 1987; Mann and Hogness, 1990; Lin and McGinnis, 1992; Zeng et al., 1993; Halder et al., 1995; Aplin and Kaufman, 1997). These same antennal transformations can be induced by mis-expressing the vertebrate homologs of the same genes (Malicki et al., 1990; McGinnis et al., 1990; Zhao et al., 1993; Halder et al., 1995), emphasizing the functional as well as structural conservation of these proteins. Homeodomain proteins are thought to control patterning processes by coordinating the expression of specific sets of target genes. Indeed, the ability of homeodomain proteins to bind specific DNA sequences in vitro, and to regulate gene expression in cultured cells, has been well documented (Jaynes and O‘Farrell, 1988; Han et al., 1989; Krasnow et al., 1989; Winslow et al., 1989; Gehring et al., 1994b). The specificity of action that each homeodomain protein exhibits in vivo, however, has not yet been duplicated in vitro. In fact, homeodomain proteins tend to bind short A/T-rich consensus sites that are often indistinguishable from one another (reviewed in Hayashi and Scott, 1990; Kalionis and O'Farrell, 1993; Gehring et al., 1994b). A concerted effort has been made in the past few years to determine how specificity of action is achieved in vivo. One potential source of specificity is through interactions with other DNA-binding transcription factors. Indeed a number of DNA-binding partners and cofactors have been identified in the past few years (reviewed in Mann and Chan, 1996). Some of these partners change DNA-binding specificity whereas others alter the ability to activate or repress adjoining promoters. Another potential source of specificity is the covalent addition of phosphate groups. Phosphorylation has been shown to affect a variety of transcription factor properties including structure, subcellular localization, DNA-binding affinity and specificity, and the ability to activate transcription (reviewed in Hunter and Karin, 1992; Karin, 1994; Hill and Treisman, 1995). All homeodomain proteins examined thus far are phosphorylated (Gay et al., 1988; Krause et al., 1988; Krause and Gehring, 1989; Odenwald et al., 1989; Tanaka and Herr, 1990; Gavis and Hogness, 1991; Kapiloff et al., 1991; Lopez and Hogness, 1991; Segil et al., 1991; Wall et al., 1992; Ronchi et al., 1993; Bourbon et al., 1995; Caelles et al., 1995; Coqueret et al., 1996; Zannin et al., 1996; Jaffe et al., 1997; Zwilling et al., 1997), indicating a likelihood for similar types of phosphorylation-induced changes in activity. Here, we use the Drosophila homeodomain protein Fushi tarazu (Ftz) as a model to examine the importance of phosphorylation on homeodomain protein function and specificity. Ftz is expressed during three stages of embryogenesis: first in alternating segmental primordia, then in differentiating neurons in the central nervous system (CNS), and finally in portions of the gut and posterior epidermis (Hafen et al., 1984; Carroll and Scott, 1985; Krause et al., 1988). ftz mutant embryos lack alternate segmental regions (Wakimoto et al., 1984) and exhibit transformations in neuronal identities (Doe et al., 1988). Ftz was one of the first homeodomain proteins shown to be phosphorylated (Krause et al., 1988; Krause and Gehring, 1989). Phosphorylation occurs on equivalent numbers of serine and threonine residues, with as many as 16 phosphates per molecule. Interestingly, phosphorylation also occurs in a tissue- and stage-specific fashion (Krause and Gehring, 1989). In this study, we report that Ftz residue threonine 263 (T263), which is in the N-terminal arm of the homeodomain, is phosphorylated in vitro in a cAMP-dependent fashion by Drosophila embryo extracts and by purified cAMP-dependent protein kinase (PKA). To test whether this phosphorylation event is required for Ftz activity, we mutated T263 to Ala and Asp to mimic the unphosphorylated and constitutively phosphorylated states, and then subcloned the mutant DNAs into a 10 kb fragment of ftz genomic DNA, capable of rescuing ftz− embryos to adulthood (Hiromi et al., 1985). The reconstituted genes were transformed into flies by P-element-mediated germline transformation and tested for their ability to rescue ftz mutant embryos. Ftz T263A mutant constructs failed to rescue ftz− flies to adulthood whereas wild-type and T263D Ftz constructs did. Phenotypic defects caused by the T263A mutation are described. The results indicate that T263 is phosphorylated when Ftz is in its active form, and that phosphorylation of T263 probably affects a protein–protein interaction. Results Ftz phosphorylation by embryo extracts The Ftz protein, expressed in embryos, is heavily phosphorylated (Krause and Gehring, 1989). In order to label and map sites, and to identify the kinases responsible, we added bacterially expressed Ftz to staged Drosophila extracts in the presence of [γ-32P]ATP. Figure 1A shows labeling of Ftz by extracts under a variety of conditions. Moderate labeling was achieved in extracts adjusted to pH 7.2 and 25 mM MgCl2 (lane 2 versus lane 1). This low level labeling was largely blocked by the addition of inhibitors of protein kinase C (PKC) and calmodulin-dependent protein kinase (Cam PK) (lane 3). However, enhanced labeling could once again be achieved by the addition of cAMP (lane 6), indicating the presence of a PKA type kinase with specificity for Ftz. Using a series of deleted Ftz polypeptides, the sites of cAMP-dependent phosphorylation were mapped to the Ftz homeodomain (Figure 1B). Deletion of the Ftz homeodomain removed all detectable labeling (lane 3), whereas the homeodomain by itself (lane 4) incorporated label as efficiently as the intact Ftz protein. Figure 1.Labeling of Ftz by embryo extracts. Extracts of 0–6 h embryos were used to phosphorylate Ftz polypeptides in vitro in the presence of [γ-32P]ATP. Following labeling, proteins were electrophoresed on PAGE gels and the gels autoradiographed. (A) Stimulation of Ftz phosphorylation by cAMP. Lane 1, extract alone; lane 2, extract + Ftz; lane 3, extract + Ftz + PKC/Cam PK inhibitors; lane 4, extract + PKC/Cam PK inhibitors + Ftz; lane 5, extract + PKC/Cam PK inhibitors + cAMP; lane 6: extract + PKC/Cam PK inhibitors + Ftz + cAMP. (B) Mapping of Ftz cAMP-dependent phosphorylation sites. Full-length, deleted and T263A Ftz polypeptides were labeled in the presence of embryo extract, PKC and Cam PK inhibitors and cAMP. Lane 1, − Ftz; lane 2, + Ftz; lane 3, + FtzΔHD (homeodomain deleted). The protein migrates just below the full-length protein and above the major labeled band just below (determined by Western, not shown). Lane 4, Ftz homeodomain (amino acids 254–314); lane 5, Ftz homeodomain with T263A substitution. Arrows indicate the positions of full-length and homeodomain Ftz polypeptides. Download figure Download PowerPoint Figure 2 shows that homeodomain-specific labeling could also be achieved with purified PKA. As with the cAMP-dependent extract labeling, the homeodomain was found to be both required and necessary for efficient labeling. In order to map the phosphorylated residue(s), extract and PKA-labeled homeodomains were subjected to tryptic digestion, thin-layer chromatography and solid-phase Edman degradation of isolated peptides (Figure 3). Ftz homeodomain labeled using embryo extract yielded two spots in the tryptic fingerprint (Figure 3A). The same two spots were labeled by PKA (Figure 3B and C). Figure 2.Phosphorylation of Ftz by PKA. Full-length and deleted Ftz polypeptides were phosphorylated in vitro in the presence of purified PKA. Labeled polypeptides were separated on PAGE gels and autoradiographed. Lane 1, no Ftz added; lane 2, full-length Ftz; lane 3, FtzΔHD; lane 4, Ftz homeodomain; lane 5, Ftz homeodomain with T263A substitution. Arrows indicate the positions of the full-length and homeodomain Ftz polypeptides. Download figure Download PowerPoint Figure 3.Mapping cAMP- and PKA-dependent phosphorylation sites. Ftz homeodomain (HD), labeled in vitro by embryo extracts or purified PKA, was subjected to trypsin digestion, separated by thin-layer chromatography and autoradiographed. (A) Ftz HD labeled by embryo extract; (B) Ftz HD labeled by PKA and (C) an equal mix of extract- and PKA-labeled proteins. Spot #1 from (A) co-migrates with spot #1 from (B), as do the two #2 spots. (D) T263A HD labeled by embryo extract. Spot #1 is absent. (E) Manual sequencing of tryptic peptide #1. Label eluted at cycle 2. T263 is the only phosphorylatable residue within a tryptic peptide of the Ftz homeodomain that should be released at cycle 2. (F) Manual sequencing of undigested Ftz homeodomain. Label eluted at cycle 11. T263 is the seventh residue of the homeodomain, and the homeodomain construct used has four additional amino acids at its N-terminal end (confirmed by DNA sequencing; data not shown). Download figure Download PowerPoint Edman degradation of the two eluted peptides yielded 32P-labeled amino acid peaks at cycles 2 (Figure 3E) and 3 (not shown) for spots one and two respectively. A tryptic peptide near the N-terminus of the Ftz homeodomain is the only peptide that has a Ser or Thr residue at the second position (T263). Indeed, phosphorylation of T263 was confirmed by Edman degradation of undigested homeodomain and release of label at cycle 11 as expected (Figure 3F). Further confirmation of phosphorylation at T263 was obtained by changing the site to an alanine residue. Figure 1B (lane 5) shows that labeling of the mutant T263A homeodomain by embryo extract was markedly reduced, and Figure 3D shows that spot #1 in the tryptic fingerprint of the T263A homeodomain disappeared as expected. The same results were obtained when labeling was performed using purified PKA (Figure 2, lane 5, and data not shown). Edman degradation analysis suggests that spot #2 corresponds to the homeodomain residue T269. Both sites, T263 and T269, are within PKA consensus sites. Notably, T263 is also in the analogous position to mapped PKA kinase sites in the POU class homeodomains of Pit-1 and Oct-1 (Kapiloff et al., 1991; Segil et al., 1991). Other protein kinases, including PKC, casein kinase I (CKI), casein kinase II (CKII), mitogen-activated protein kinase (MAPK), S6 protein kinase (S6K) and glycogen synthase kinase-3 (GSK-3), failed to phosphorylate this site. Phosphorylation by PKA, on the other hand, was equal in efficiency to phosphorylation of a PKA consensus site (heart muscle kinase peptide; data not shown). In vivo analysis of T263 mutants In order to test whether T263 is phosphorylated in vivo, and to determine the role of phosphorylation, Ftz T263A and T263D mutant expression constructs were injected into flies, and the transgenic constructs tested for their ability to functionally replace the endogenous ftz gene. The ftz open reading frame, along with 6.1 kb of 5′ and 2 kb of 3′ DNA, is sufficient to direct normal ftz expression and to fully rescue ftz mutant embryos (Hiromi et al., 1985). Rescue constructs were introduced into the genome by P-element-mediated germline transformation (Spradling and Rubin, 1982). For each mutation, four transformant lines, all on the second chromosome, were used for subsequent analyses. These, along with lines carrying wild-type ftz transgenes, were crossed to flies carrying the ftz alleles ftz9H34 and ftzw20, both of which are protein null. The resultant lines, P[ftz]; ftz9H34/ftzw20, were then tested for viability. Table I shows the survival index (SI = No. of rescued flies obtained/No. of rescued flies expected) values for flies carrying wild-type, T263A and T263D transgenic ftz genes in the ftz9H34/ftzw20 mutant background. The wild-type control construct rescued flies to adulthood with SI values ranging from 21 to 71%. These values are in the same range as those found in previous studies (Hiromi et al., 1985; Furukubo-Tokunaga et al., 1992). These observed variations in rescuing capacity are likely to be due to reduced or ectopic ftz expression brought about by adjoining sequences at the various sites of P-element insertion. Table 1. Survival indices of Ftz T263 mutant offspring Line Linkage No. rescued SI WT 1 II 132; n = 559 0.708 2 II 36; n = 521 0.207 3 II 65; n = 531 0.367 4 II 75; n = 470 0.479 T263A 1 II 0; n = 278 0 2 II 0; n = 278 0 3 II 0; n = 495 0 4 II 0; n = 443 0 T263D 1 II 3; n = 401 0.022 2 II 16; n = 379 0.126 3 II 63; n = 744 0.254 4 II 19; n = 477 0.119 SI values (No. of flies with appropriate phenotype/No. expected) are given for flies homozygous mutant for the endogenous ftz gene and carrying two copies of either a wild-type, T263A or T263D transgenic ftz gene. For each transgene, four independent lines were examined, all with inserts on the second chromosome (linkage group II). n = No. of flies scored. In contrast to the wild-type ftz construct, the T263A mutant construct showed no ability to rescue: all four transformant lines gave SI values of zero. On the other hand, the T263D construct was able to rescue mutant flies to adulthood, although less effectively than the wild-type construct (SI = 2–25% versus 21–71% for wild-type). Thus, the T263D protein probably exhibits minor irregularities in function or expression. Cuticle patterns The rescue of flies to adulthood is a comprehensive measure of all ftz gene activities. Because ftz has several stage-specific developmental roles, reductions in SI values could result from either general or stage-specific defects. In order to test for segmentation-specific defects, we looked at the cuticles of larvae obtained from rescued embryos. We also examined the expression patterns of putative target genes required for ftz-dependent segmentation. ftz mutant cuticles exhibit pairwise deletions of even-numbered parasegments (Furukubo-Tokunaga et al., 1992, Schier and Gehring, 1993; and Figure 4B). Cuticle preparations of T263A embryos also exhibited pair-wise deletions of even-numbered parasegments (Figure 4C). Interestingly, however, the penetrance of the phenotype was quite variable, with some cuticles exhibiting complete deletion of ftz-dependent parasegments, while others showed few or no detectable defects. Although variable from cuticle to cuticle, defects in each ftz-dependent parasegment occurred with a characteristic frequency. These ranged from the most to least frequently affected as follows: parasegment (PS) 2 (97% affected)>PS6 (82%)>PS10, 4 (70%)>PS12, 14 (58%)>PS8 (36%) (n = 482). Figure 4.Cuticle preparations of T263 mutant larvae. Cuticles were prepared from P[ftz T263] rescued larvae and visualized by dark field microscopy. (A) Wild-type larval cuticle; (B) homozygous ftz mutant cuticle; (C) typical Ftz T263A rescued cuticle; (D) typical Ftz T263D rescued cuticle. Download figure Download PowerPoint In contrast to T263A cuticles, cuticles from T263D flies were indistinguishable from wild-type (Figure 4D). A few cuticles (<1%) had minor abnormalities, but these were variable and also seen when wild-type rescue constructs were used. The ability of the T263D protein, and not the T263A protein, to rescue ftz-dependent segmentation suggests that T263 is normally in the phosphorylated state when the protein is active. ftz autoregulation To trace the causes of cuticular defects observed with T263A mutant embryos, we examined the expression of ftz target genes. The ftz gene itself is a well characterized target of Ftz activity. Ftz facilitates its own expression via binding and activation of an upstream enhancer element (Hiromi et al., 1985; Hiromi and Gehring, 1987; Pick et al., 1990; Schier and Gehring, 1992). Figure 5A and B shows typical patterns of ftz transcripts at two different stages of embryogenesis. In the T263A mutants, stripes of expression were weaker and narrower than normal (Figure 5E and F), except perhaps at the earliest stages of initiation (not shown) which are thought to be ftz-independent. The frequency and severity of defects were similar at the two later stages shown. Patterns of protein expression (not shown) were similar to those of the transcripts, with the same variations seen at the same stages. Protein was localized properly to the nucleus in all transgenic lines. These observations suggest, but do not prove, that it is the ability of Ftz to autoregulate its own promoter, and not its synthesis, stability or localization, that is affected by the mutation. Figure 5.ftz gene autoregulation in Ftz T263 mutants. ftz gene expression was examined by in situ hybridization at two different stages to test for defects in autoregulation. (A and B) Wild-type ftz expression; (C and D) homozygous ftz mutant embryo; (E and F) typical ftz T263A expression; (G and H) typical ftz T263D expression. In (E–H), homozygous ftz− embryos were identified based on lack of β-galactosidase expression (see Materials and methods). Download figure Download PowerPoint As with the cuticle preparations, the severity of defects in ftz expression patterns was highly variable. Some embryos were devoid of detectable expression while others were essentially normal. Certain stripes were more prone to defects than others. Although stripe defects varied from embryo to embryo, the average frequency of defects occurred in the following order: stripe 1>3>5, 2>6, 7>4. This correlates well with the prevalence of defective ftz-dependent parasegments in the T263A cuticle preparations (i.e. PS2>PS6>PS10, 4>PS12, 14>PS8). The T263D embryos, in contrast, produced normal patterns of ftz mRNA (Figure 5G and H) and protein (not shown). en and wg expression Ftz is thought to be a direct transcriptional activator of alternate stripes of the segment polarity gene engrailed (en) (DiNardo and O'Farrell, 1987; Kassis, 1990; Florence et al., 1997). Conversely, transcription of alternate wingless (wg) stripes appears to be repressed by Ftz (Ingham et al., 1988; Copeland et al., 1996). Altered expression patterns of en and wg in T263A mutant embryos (Figure 6) correlated well with the defects observed in ftz expression patterns (Figure 5). ftz-dependent en stripes were generally weak or missing. Conversely, the wg stripes that normally are repressed by Ftz were expanded. As with ftz expression, the penetrance of these defects varied greatly from embryo to embryo and showed similar frequencies of segment-specific defects. No defects in en and wg stripes were observed in T263D embryos. Figure 6.en and wg expression in Ftz T263 mutants. Patterns of en mRNA expression are shown on the left (A–D) and patterns of wg mRNA expression on the right (E–H). From top to bottom, embryos shown are wild-type (A and E), ftz− (B and F), Ftz T263A (C and G) and Ftz T263D (D and H). ftz-dependent en stripes are partially missing in the T263A mutant embryo and wg stripes are partially expanded. In contrast, en and wg stripes are normal in the T263D embryo. Download figure Download PowerPoint DNA-binding activity The results thus far indicate that the T263A mutation reduces Ftz activity, and that the effects of the T263D substitution are extremely minor. Since T263 resides in the DNA-binding homeodomain, we tested whether DNA-binding activity is affected by the mutations. Electrophoretic mobility shift assays (EMSAs) were performed using consensus Ftz-binding sites (Percival-Smith et al., 1990; Pick et al., 1990; Florence et al., 1991). Wild-type, T263A and T263D homeodomains were expressed in reticulocyte lysates. Figure 7 shows that inherent DNA-binding activity of the homeodomain was not affected by either of the two mutations. The same result was achieved using a different Ftz-binding site and different concentrations of protein (data not shown). Thus, the effect of the T263A mutation on Ftz activity does not appear to be exerted at the level of protein–DNA recognition. Figure 7.Binding of Ftz T263 mutant homeodomains to DNA. (A) Wild-type and T263 mutant Ftz homeodomains were expressed in a reticulocyte lysate system and bound to a 32P-labeled Ftz consensus DNA-binding site GGGAAGCAATTAAGGAT (Percival-Smith et al., 1990; Pick et al., 1990). Complexes were resolved on non-denaturing polyacrylamide gels and visualized by autoradiography. Lane 1, no protein added; lane 2, wild-type homeodomain; lane 3, T263A homeodomain; lane 4, T263D homeodomain. All three polypeptides showed equivalent DNA-binding activity. (B) Western blot of in vitro expressed homeodomain polypeptides used in (A). Proteins were separated on 15% PAGE gels, transferred to nitrocellulose and detected using a polyclonal Ftz antiserum. Lane 1, no protein added; lane 2, wild-type homeodomain; lane 3, T263A homeodomain; lane 4, T263D homeodomain. All three polypetides were expressed at similar levels. Download figure Download PowerPoint Transcriptional activity in cultured S2 cells A tissue culture co-transfection assay was used to characterize further the functional properties of Ftz T263 mutants. Wild-type, T263A and T263D Ftz expression constructs were co-transfected individually into Drosophila S2 cells together with the Ftz-responsive reporter plasmid 3′K TATA-CAT, which contains concatamerized Ftz-binding sites derived from a portion of the en promoter (Han et al., 1989). All three Ftz expression constructs were able to activate the reporter gene with similar efficiency: ∼15- to 16-fold over basal levels (Figure 8). Furthermore, co-transfection of the catalytic subunit of PKA (hatched columns in Figure 8), or treatment of cells with PKA stimulators or inhibitors (not shown), had no effect on the activities of the three Ftz proteins. Other Ftz-responsive reporter plasmids, such as Ubx-CAT (Krasnow et al., 1989) and NP6-CAT (Jaynes and O'Farrell, 1988) also gave equivalent responses to the wild-type and mutant Ftz polypeptides (not shown). These results indicate that T263 phosphorylation does not affect the general ability of Ftz to bind and activate simple reporter genes in S2 cells. Rather, the effects of T263 phosphorylation are likely to depend on factors specifically expressed in the Drosophila embryo. Figure 8.Transcriptional activity of Ftz T263 mutants in cultured cells. Wild-type and T263 mutant ftz cDNAs were placed under control of the actin 5C promoter and transfected into cultured Drosophila S2 cells together with the Ftz-dependent reporter gene 3K' TATA-CAT (open columns). Duplicate transformations contained in addition a construct that expresses the catalytic subunit of PKA (cross-hatched columns). Levels of CAT expression were determined immunologically and then corrected for minor variations in levels of Ftz expression. Levels of CAT expression are given as levels relative to the level of basal promoter activity, which was assigned a value of one. All three Ftz proteins had equivalent levels of activity, and activity was not altered by co-expression of PKA. Download figure Download PowerPoint Discussion Evidence for T263 phosphorylation in vitro and in vivo We have shown that T263, in the N-terminal arm of the Ftz homeodomain, is phosphorylated specifically and efficiently in vitro by embryo extracts and PKA. Mutation of this site to alanine, which is generally a conservative substitution (Cunningham and Wells, 1989), resulted in complete loss of Ftz-rescuing activity in vivo. This lethality was attributed mainly to defects in ftz-dependent segmentation. In contrast, mutation of T263 to aspartate, a far less conservative substitution than alanine, but one which has been shown in functional and structural studies effectively to mimic threonine phosphate (see, for example, Cowley et al., 1994; Kowlessur et al., 1995; Napper et al., 1996; Peverali et al., 1996; Boekhoff et al., 1997; Jaffe et al., 1997), returned Ftz activities to near wild-type levels. This strongly suggests that T263 is phosphorylated in vivo, and that phosphorylation of this site is required for normal protein activity. Stage specificity of T263 mutant defects The majority of T263A mutants died as embryos. Segmental defects, most prevalent in the head, are probably responsible for this lethality, as defects were observed in close to a quarter of the cuticles (97% of homozygous larvae). However, it is also possible that pattern abnormalities in the developing CNS or gut, where ftz is expressed later (Carrol and Scott, 1985; Krause et al., 1988), also contributed to lethality. Defects incurred at these later stages could also account for the reduced rescuing capacity of the T263D mutant construct relative to a wild-type construct, since cuticle patterns of these animals appeared to be normal, yet SI values were significantly lower than for wild-type ftz constructs. Both gain-of-function and loss-of-function defects are possible, as cofactors and DNA targets may vary from tissue to tissue in their preference for phosphorylated or unphosphorylated Ftz. The phosphorylation state of T263 may also vary from tissue to tissue. Variability in the T263A phenotype An interesting aspect of the T263A segmental phenotype was its variability. Cuticle patterns and Ftz target gene patterns ranged from wild-type in appearance to patterns that resembled ftz nulls. Also, while certain ftz-dependent segments were more likely to be affected than others, the segments affected within any given embryo were highly variable. Similar observations were made by Furukubo-Tokunaga et al. (1992) in a study" @default.
• W2021818985 created "2016-06-24" @default.
• W2021818985 creator A5009406416 @default.
• W2021818985 creator A5024164747 @default.
• W2021818985 creator A5045799176 @default.
• W2021818985 creator A5089205955 @default.
• W2021818985 date "1998-04-15" @default.
• W2021818985 modified "2023-09-23" @default.
• W2021818985 title "A phosphorylation site in the Ftz homeodomain is required for activity" @default.
• W2021818985 cites W1541492444 @default.
• W2021818985 cites W1900262957 @default.
• W2021818985 cites W1966946991 @default.
• W2021818985 cites W1969461355 @default.
• W2021818985 cites W1972693622 @default.
• W2021818985 cites W1973444729 @default.
• W2021818985 cites W1978637982 @default.
• W2021818985 cites W1979767925 @default.
• W2021818985 cites W1982778642 @default.
• W2021818985 cites W1984751884 @default.
• W2021818985 cites W1987718027 @default.
• W2021818985 cites W1989480480 @default.
• W2021818985 cites W1990958396 @default.
• W2021818985 cites W1992323126 @default.
• W2021818985 cites W1993766073 @default.
• W2021818985 cites W1994355174 @default.
• W2021818985 cites W1995117997 @default.
• W2021818985 cites W1995267165 @default.
• W2021818985 cites W1996045820 @default.
• W2021818985 cites W2002809491 @default.
• W2021818985 cites W2007011466 @default.
• W2021818985 cites W2008657400 @default.
• W2021818985 cites W2013679258 @default.
• W2021818985 cites W2013906846 @default.
• W2021818985 cites W2018184392 @default.
• W2021818985 cites W2028269059 @default.
• W2021818985 cites W2028545666 @default.
• W2021818985 cites W2035378222 @default.
• W2021818985 cites W2038013843 @default.
• W2021818985 cites W2038423754 @default.
• W2021818985 cites W2042655455 @default.
• W2021818985 cites W2047466377 @default.
• W2021818985 cites W2053320483 @default.
• W2021818985 cites W2058773589 @default.
• W2021818985 cites W2058877341 @default.
• W2021818985 cites W2060541402 @default.
• W2021818985 cites W2062579188 @default.
• W2021818985 cites W2064202227 @default.
• W2021818985 cites W2065434570 @default.
• W2021818985 cites W2066306270 @default.
• W2021818985 cites W2067318229 @default.
• W2021818985 cites W2068634624 @default.
• W2021818985 cites W2075879868 @default.
• W2021818985 cites W2077210765 @default.
• W2021818985 cites W2078461921 @default.
• W2021818985 cites W2078856910 @default.
• W2021818985 cites W2079544213 @default.
• W2021818985 cites W2082797919 @default.
• W2021818985 cites W2083355932 @default.
• W2021818985 cites W2084221802 @default.
• W2021818985 cites W2084815177 @default.
• W2021818985 cites W2089368521 @default.
• W2021818985 cites W2090961125 @default.
• W2021818985 cites W2095416319 @default.
• W2021818985 cites W2097145517 @default.
• W2021818985 cites W2100349230 @default.
• W2021818985 cites W2100509364 @default.
• W2021818985 cites W2102589238 @default.
• W2021818985 cites W2103080800 @default.
• W2021818985 cites W2111341106 @default.
• W2021818985 cites W2118615111 @default.
• W2021818985 cites W2126012806 @default.
• W2021818985 cites W2132172516 @default.
• W2021818985 cites W2135726266 @default.
• W2021818985 cites W2142729708 @default.
• W2021818985 cites W2144393631 @default.
• W2021818985 cites W2147683285 @default.
• W2021818985 cites W2149200293 @default.
• W2021818985 cites W2154627422 @default.
• W2021818985 cites W2154892337 @default.
• W2021818985 cites W2155700176 @default.
• W2021818985 cites W2160512275 @default.
• W2021818985 cites W2168473076 @default.
• W2021818985 cites W2171405564 @default.
• W2021818985 cites W2183059413 @default.
• W2021818985 cites W2387292660 @default.
• W2021818985 cites W4229918205 @default.
• W2021818985 cites W4254829742 @default.
• W2021818985 cites W569680583 @default.
• W2021818985 cites W99394036 @default.
• W2021818985 doi "https://doi.org/10.1093/emboj/17.8.2308" @default.
• W2021818985 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/1170574" @default.
• W2021818985 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/9545243" @default.
• W2021818985 hasPublicationYear "1998" @default.
• W2021818985 type Work @default.
• W2021818985 sameAs 2021818985 @default.
• W2021818985 citedByCount "20" @default.
• W2021818985 countsByYear W20218189852012 @default.
• W2021818985 countsByYear W20218189852013 @default.

• next