QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

doi: 10.1242/10.1242/dev.00207

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lunde, K. Articles by Bier, E. Search for Related Content PubMed PubMed Citation Articles by Lunde, K. Articles by Bier, E.

Activation of the knirps locus links patterning to morphogenesis of the second wing vein in Drosophila

¹ Section of Cell and Developmental Biology, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0349, USA
² Albert-Ludwigs-Universität Freiburg, Institut für Biologie I, Hauptstrasse 1, D-79104 Freiburg, Germany
³ Department of Biology, Dickinson College, PO Box 1773, Carlisle, PA 17013, USA
⁴ Max-Planck-Institut für biophysikalische Chemie, Abt. Molekulare Entwicklungsbiologie, Am Fassberg 11, D-37077 Göttingen, Germany

View larger version (45K): [in a new window]   Fig. 1. Map of knirps locus and ri alleles. (A) A wild-type adult male wing. Longitudinal wing veins (L1-L5), and margin (M) are indicated. (B) A kniri[1] homozygous adult male wing, with an incomplete L2 vein. Note that this phenotype is somewhat milder than that of Df(3L)kniri[XT2] mutants (C). (C) A Df(3L)kniri[XT2] homozygous adult male wing. Note the virtual absence of the L2 vein. (D) A kniri[53j] homozygous adult male wing. (E) Diagram of region 77E1 in the left arm of the third chromosome showing kni and knrl transcripts, the regions deleted in Df(3L)kniri[XT2] and Df(3L)kniFC82 as well as lacZ test constructs (EcoRI fragments) E (red), F (green), S (orange), R (brown), and Q (pink). The fragment labeled abd-lacZ (white box) has been shown to contain an embryonic enhancer (yellow box) driving expression in abdominal segments (Pankratz et al., 1992). This fragment does not drive gene expression in the wing primordium. Genomic DNA fragments between E and S were not tested in lacZ fusions. The 4.8 kb fragment E drives reporter gene expression in L2. abd (yellow box), embryonic abdominal enhancer (Pankratz et al., 1992); CG13252, a predicted protein-coding transcript that is expressed ubiquitously in both wild-type and kniri[53j] wing discs (data not shown). The knrl transcript, which comprises 23 kb of genomic DNA (Rothe et al., 1992), is not drawn to scale. The 5' end of the knrl gene is at -74.8 kb on this map. (F) Expanded map of 4.8 kb fragment E, showing the 1.4 kb minimal L2 driver (black), 252 bp kniri[1] deletion (dark blue), C to A base substitution at 0.596 kb in E in kniri[53j] (light blue), and the region of insertion(s) or rearrangements in kniri[M3] (turquoise), from 0.098 to 0.686, and from 1.99 to 3.44 kb. E, EcoRI; B, BamHI; C, HincII; X, XhoI; S, SalI; H, HindIII. (G) lacZ transgenic constructs made to assay expression driven by a region containing the kniri[1] deficiency substituted into E (Eri[1]), and successively smaller 3' subfragments of E: EX, 1.4 kb; and EC, 0.69 kb.

View larger version (63K): [in a new window]   Fig. 2. Expression of wild-type and mutant L2 enhancer constructs. (A) kni expression in a wild-type mid-third instar larval wing disc. The L2 primordium (L2) and margin (M) are indicated. (B) lacZ expression driven by the E-lacZ construct in a mid-third instar larval wing disc. Note the lower levels of lacZ expression in a diffuse stripe in the posterior region of the wing pouch (arrow). (C) E-lacZ expression relative to endogenous kni expression in a mid-third instar larval wing disc double-stained for kni mRNA expression (blue) and anti-ß-galactosidase protein (brown). Note the high degree of coincidence of these two expression patterns. (D) E-lacZ expression relative to salm expression in a mid-third instar larval wing disc double-stained for anti-ß-galactosidase protein (brown) and salm expression (blue). Note the abutting and mutually exclusive expression of these two genes. (E) EX-lacZ expression driven by fragment EX (1.4 kb 3' subfragment of E; see Fig. 1G) in a mid-third instar larval wing disc. (F) EC-lacZ expression driven by fragment EC (0.69 kb 3' subfragment of E; see Fig. 1G) in a mid-third instar larval wing disc. Note the broad anterior and posterior domains of ectopic lacZ expression relative to that driven by the E-lacZ and EX-lacZ constructs. (G) E-ri[1]-lacZ expression driven by the fragment E-ri[1], which is deleted for the same 252 bp segment that is missing in the ri[1] allele, in a mid-third instar larval wing disc. Note the entire absence of patterned staining, including that observed in the posterior region of the wings discs (e.g. compare with staining in panel D). (H) EX(ri53j)-lacZ expression driven by the fragment EX(ri53j), which contains the same single base pair alteration present in the kniri[53j] allele, in a mid-third instar larval wing disc. Note that L2 staining is selectively lost while expression in extreme anterior and posterior domains of the wing pouch is retained (arrowheads: compare with staining in G).

View larger version (71K): [in a new window]   Fig. 3. The L2-enhancer element is activated by Sd. (A) Sequence of the EcoRI-HincII fragment (EC) of E, which contains sites necessary for activation of the minimal L2 enhancer (0 to 0.691 kb of fragment E). (B) Sequence of the HincII-XhoI fragment, which contains sequences required to repress ectopic activity of the L2 enhancer element (0.692 to 1.361 kb of fragment EX). Key: The deletion in the kniri[1] mutant is indicated by blue text and the nucleotide mutated in the kniri[53j] mutant (C to A at 596) highlighted in turquoise. Sd binding sites in fragment EC (underlined in brown) were determined empirically by footprinting and gel shifts. A tandem doublet of binding sites begins at 271 [TACATTTGTCGCATAGTT], while the sites beginning at 463 [TGTATGTAT] (opposite strand), 523 [AAAATGTCG], 570 [GAAATGCGT], and 640 [ACTATTTCT] (opposite strand) are single sites. These experimentally determined sites define a consensus [(A/T)(A/G)NAT(G/T)TNT], which matches well with the consensus [WRVWATKYR] derived for Sd binding to other wing enhancers that require vestigial and sd function such as bs=DSRF (Halder et al., 1998), cut, sal and the vg quadrant element (Guss et al., 2001). Other predicted DNA binding sites are indicated based on matches to known consensus sequences. Engrailed (En) [TAATTA — yellow type]: (Ades and Sauer, 1994); Spalt-related (Salr) [TTATGa/tAa/cT — pink type]: (Barrio et al., 1996); Brinker (Brk) [c/tGCCAg/c — green type]: (Rushlow et al., 2001; Zhang et al., 2001). No predicted DNA binding sites were identified in fragment EX for Mad (Kim et al., 1997), Ci (Kinzler and Vogelstein, 1990), or Kni (Small et al., 1996). (C-F) Reporter gene expression driven by the 4.8 kb element E is dependent upon sd function. (C) A mitotic clone homozygous for the hypomorphic sd58 allele located in the wing pouch is marked by the absence of the MYC epitope tag and is outlined in red. (D) Expression of the E-lacZ reporter gene is eliminated or reduced by the reduction of sd function, except in clones along the wing margin, where it is unaffected (data not shown). The yellow arrowhead indicates the position of the sd- clone. (E) The position of cells is marked by the nuclear dye TOPRO. (F) Merged image of panels A-C. (G) Gel mobility shift assays of oligonucleotide probes spanning the tandem binding sites at 271, and the single sites at 570 and 640, with the Sd TEA domain (left panels of each pair). Mutations introduced at these sites (right panels of each pair), reduce Sd binding. F, free probe; 1, probe with one Sd TEA domain bound; 2, probe with two Sd TEA domains bound. (H) A wing disc from one of the transformant lines in which the 271 doublet, and 570, 640 single Sd sites were mutated in the context of fragment EC.

View larger version (29K): [in a new window]   Fig. 7. A model for L2 initiation: global activation, regional repression and local induction. (A) Activation of kni expression in the L2 primordium as a consequence of the action of patterning genes initiates morphogenesis of the L2 vein. Analysis of the kni/knrl locus L2 enhancer element reveals that this element is activated in a narrow stripe by a combination of global wing specific activation, regional repression, and localized signal mediated activation. Global wing-specific activation is provided at least in part by Sd/Vg (this study). Regional repression is mediated either directly or indirectly by Salm/Salr (Sal) in central wing cells (Lunde et al., 1998), and by a repressor(s) acting in peripheral wing cells (green domain) — possibly Brk (this study). In addition, a short range signal X (blue), which is produced by Sal-expressing cells (pink domain) is required to induce kni/knrl expression (blue line) in cells just anterior to sal-expressing cells. Since kni/knrl expression is activated in cells just anterior to the sal expression domain, but not in posterior cells adjacent to the sal domain (dashed blue line), there must be either a repressor (e.g. En?) acting in the posterior compartment (hatched) or an activator acting in the anterior compartment together with X' which provides this AP specificity in kni/knrl activation. The AP compartment border is indicated (solid black line). (B) The kni/knrl L2 enhancer can be subdivided into separate domains mediating at least certain components of activation (black region) and repression (red region). Global wing-specific activation, which requires Sd/Vg function (this study), is mediated by some combination of the empirically determined Sd binding sites (brown ovals) of which there are four single sites and one double site in the activation region. The action of the hypothetical signal X is presumably mediated by a transcription factor X', which binds to sequences in the activation domain. The binding site for factor X' (blue circle) may include sequences containing the single nucleotide mutated in the kniri[53j] mutant (C596A) since L2 enhancer activity is selectively lost in the L2 primordium of these flies. This site is also eliminated in the 252 bp deletion in the kniri[1] mutant (dark blue region), as are additional sequences required for activity of the enhancer throughout the wing (e.g. four single Sd sites). The repression domain contains five predicted binding sites for the repressor Brk (green diamonds), two sites for En (hatched yellow triangles), and one site for Salr (pink diamond). Since the reporter gene expression driven by the activation region alone is repressed in central cells, this fragment must also contain some yet unidentified binding sites for Salm, Salr, or some other factor mediating the repressive effect of Salm/Salr.

View larger version (64K): [in a new window]   Fig. 4. Rescue of L2 in kniri[1] by targeted gene expression in the L2 primordium. (A) GAL4 expression in an L2-GAL4 mid-third instar larval wing disc. (B) Rescue of the L2 vein (arrow) in an L2-GAL4>UAS-kni; kniri[1] male adult wing. (C) Rescue of the L2 vein in an L2-GAL4>UAS-knrl; kniri[1] male adult wing. (D) Rescue of the L2 vein in an L2-GAL4>UAS-rho; kniri[1] male adult wing.

View larger version (93K): [in a new window]   Fig. 5. Misexpression of non-L2 vein genes in the L2 primordium alters L2 development. (A) An L2-GAL4>UAS-Dl adult female wing. The distal tip of the L2 vein is typically missing (arrow). (B) An L2-GAL4>UAS-Dl adult male wing. The L2 vein is consistently truncated to this severe extent. (C) rho expression in a wild type mid third larval instar disc. Vein primordia L2-L5 and the margin are indicated. (D) rho expression in a 2X L2-GAL4>UAS-Dl mid-third instar larval wing disc is lost selectively in the L2 primordium (arrow). (E) An L2-GAL4>UAS-sc adult male wing. Ectopic bristles are largely confined to the L2 vein as well as to a broad posterior domain, which presumably corresponds to cells in L2-GAL4 wing discs expressing high (L2) or moderate (posterior domain) levels of GAL4 (e.g. see Fig. 4A). (F) An L2-GAL4>UAS-ara adult male wing. Note that the width of the L2 vein is reduced in the middle (arrowhead). This thinning of L2 by misexpression of ara is a consistent phenotype that we have not observed when we misexpress the related gene caup in the same pattern (not shown).

View larger version (106K): [in a new window]   Fig. 6. Use of the L2-GAL4 expression system to assay gene function in the wing. (A) Rescue of the L2 vein in an L2-GAL4>UAS-eg; kniri[1]/kniri[XT2] male adult wing. Note the ectopic sensory bristles along the reinstated L2 vein. (B) An L2-GAL4>UAS-top adult male wing. (C) An L2-GAL4>UAS-vein adult male wing. Note the minor degree of ectopic venation (e.g. anterior to L2 and at the junction of L2 with the margin). (D) An L2-GAL4>UAS-secreted-spitz adult female wing. Note the ectopic vein running parallel and anterior to L2. (E) An L2-GAL4>UAS-rho+UAS-Star wing. Note patches of ectopic vein material anterior to L2. (F) An L2-GAL4>UAS-wingless adult male wing. Note change in shape of the margin, ectopic vein material both anterior and posterior to L2 and ectopic bristles near the junction of L2 with the margin (magnified in insert).