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

First published online 2 June 2004
doi: 10.1242/dev.01161

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 Pueyo, J. I. Articles by Couso, J. P. Search for Related Content PubMed PubMed Citation Articles by Pueyo, J. I. Articles by Couso, J. P.

Chip-mediated partnerships of the homeodomain proteins Bar and Aristaless with the LIM-HOM proteins Apterous and Lim1 regulate distal leg development

School of Biological Sciences, University of Sussex, Falmer, Brighton BN1 9QG, UK

View larger version (61K): [in a new window]   Fig. 1. The role of Chip and dlmo in leg development, and their genetic relationships with apterous. (A) The distal region of a wild-type prothoracic leg showing the distal part of the tibia (Tb), tarsal segments one to five (t1-t5), and the distalmost organ, the claw, in the pretarsus (c). (B-B''') Each segment of the distal part of the leg is characterised by differential expression of the LIM-HOM Ap and Lim1 and the Prd-HOM Bar transcription factors. The images show a side view of an everting leg imaginal disc. B shows the merged triple staining; B'-B''' show the separate channels. Expression of a Bar reporter gene in tarsus four and five is shown in green (B'); Lim1 protein distribution in the pretarsus is shown in blue (B''); and Ap protein distribution is shown in red (B'''; yellow in overlap in B). (C) Wild-type leg imaginal disc showing Chip protein distributed ubiquitously in the disc epithelium. (D) Dlmo protein distribution in a late third instar leg imaginal disc. Specific staining can be detected in a few cells in the peripodial membrane (arrow). (E) Minute+ Chipe55 clones in leg. The tissue lacking Chip is marked by its yellow (y) phenotype and is outlined in black. Clones in the tibia, femur, coxa and pretarsus show a phenotype similar to strong Lim1 mutants. The fourth tarsal segment fails to develop, as in strong ap mutants. (F) Higher magnification of the tip of the leg shown in E. The majority of the distal part of the leg is y– apart from two bristles that are y+ (asterisks). In the pretarsus no claws develop (arrowhead). In addition, only a remnant part of a joint is observed between the last tarsal segments (arrow). (G) Leg of a DllGal4;UAS-Chip fly. Only four tarsal segments develop and the claw organ is absent, similar to the phenotype of Chip lack of function, which is shown in E. (H) Ap (red) and Lim1 (green) protein expression are normal in a DllGal4;UAS-Chip leg disc. The white dotted line denotes the edge of the distal domain of expression of the DllGal4 line. (I) Leg of an apGal4;UAS-dlmo fly (29°C). Although the LIM-only gene (dlmo) is not expressed in the leg imaginal disc, Dlmo overexpression produces loss of the fourth tarsal segment. (J) Co-expression of UAS-Chip in an apGal4;UAS-dlmo genetic background rescues the loss of the fourth tarsal segment.

View larger version (68K): [in a new window]   Fig. 2. Genetic interactions between LIM-HOM proteins and Ap function. (A) Leg of an apGal4/apUGO35 fly. The fourth tarsal segment is almost completely lost and is fused to the fifth (arrowhead), whereas the other leg segments are normal. (B) Rescue of the apGal4/Df(2)nap1 leg mutant phenotype by overexpression of an ap transgene. The ap mutant phenotype is completely rescued by one copy of the UAS-ap construct (compare with A and Fig. 1A). (C) Rescue of an apGal4/Df(2)nap1 leg mutant phenotype by overexpression of the chimaera protein Chip-Ap, consisting of Chip lacking the LIM interaction domain (LID) linked to Ap lacking the LIM domains. (D) Leg of an apGal4/Df(2)nap1;UAS-Lim3:ap-HD fly. Ectopic expression of a chimaera protein, which consists of the LIM domains of Lim3 and the Ap homeodomain, rescues the ap mutant phenotype. (E) apGal4;UAS-isl leg lacking the fourth tarsal segment. (F) Leg of an apGal4/UAS-ap;UAS-isl fly. Overexpression of Ap does not overcome the dominant-negative effect produced by ectopic Islet expression (compare with E). (G-G'') Leg imaginal disc of an apGal/UASGFP;UAS-islet larva. Ectopic expression of Islet does not affect either Ap or Bar expression. (G) Bar protein distribution (blue). (G') apGal4 expression (red). (G'') Merged image.

View larger version (106K): [in a new window]   Fig. 3. Bar is the factor affected by ectopic LIM-HOM protein expression in the fourth tarsal segment. (A) Leg with Bar mutant clones marked by forked phenotype in the tarsal region. A large clone along the ventral (lower) side of the leg is outlined in black. Cells lacking Bar do not grow properly and the tarsal segments t2-t5 appear fused. Magnification of the distal part of the clone is shown in the inset, showing a remnant joint (arrow) and a forked bristle (arrowhead). (B) Leg with Bar mutant clones similar to those shown in A, except that here they also express Ap using DllGal4 (see Materials and methods). The distal tarsal segments are fused, similar to those observed in Bar clones. Inset shows a magnification of the distal part of the clone, showing a forked bristle (arrowhead) and a remnant joint (arrow). (C) Leg imaginal disc showing Bar-lacZ reporter expression in a ring of cells in the presumptive fourth and fifth tarsal region (green), and Lim1 distribution (red). (D-D'') Ectopic expression of the Lim1 gene represses Bar in a dppGal;UAS-Lim1 leg imaginal disc. (D) Lim1 protein distribution. (D') Bar protein distribution, showing an absence of Bar in the area where Lim1 is present. (D'') Merged image. (E) Leg of an apGal4/UAS-Lim1;UAS-Bar fly. Overexpression of Bar in an apGal4/UAS-Lim1 genetic background partially rescues the ap-like dominant-negative effect of ectopic Lim1 (Pueyo et al., 2000). (F) Overexpression of Bar in an apGal4;UAS-islet genetic background also partially rescues the loss of the fourth tarsal segment (compare with Fig. 2E). Magnification of the distal part is shown in the inset. A remnant joint in the dorsal part of the fused t4-t5 segment is seen (arrowhead). An apical bristle can also be distinguished (out of focus; arrowhead).

View larger version (76K): [in a new window]   Fig. 4. Bar is the limiting factor for the development of the fourth tarsal segment. (A) Leg of an InBM2 mutant. This mutation produces partial loss of Bar function (Kojima et al., 2000). In these mutants, 33% of the legs show a weak fusion between the fourth and fifth tarsal segment, with the joint not properly differentiated (inset, arrowhead). (B) Leg of a InBM2; apUGO /+ mutant. Removal of a copy of ap enhances the mutant phenotype observed in InBM2 mutants. Tarsus four and five are shorter and are fused in 61% of the legs (inset, arrowhead; compare with A). (C) The ap mutant phenotype is also enhanced by reducing Bar function in InBM2; apUGO /apGal4 flies (compare with Fig. 2A). Tarsus four is completely absent (inset), and even the joint between tarsus five and three is affected (arrowhead). (D) Overexpression of Bar does not repress Ap expression in the fourth tarsal segment (arrow). (E) Leg of an apGal4;UAS-Bar fly. Overexpression of Bar prevents the development of the fourth tarsal segment. (F) Overexpression of Ap rescues completely the phenotype caused by overexpression of Bar in the ap domain (compare with E). (G) Ectopic expression of a hybrid molecule, consisting in the LIM domains of Lim3 and the Ap homeodomain, completely rescues the loss of tarsus four phenotype produced by overexpression of Bar (compare with E). (H) Rescue of the apGal;UAS-Bar phenotype is also achieved by overexpression of Chip. (I) Ectopic expression of Chip lacking the LIM interaction domain is not able to rescue the dominant-negative effect produced by Bar overexpression. (J) Ectopic expression of the Chip-Ap hybrid protein partially rescues the apGal4;UAS-Bar phenotype, suggesting that Ap interacts with Chip to form dimers in tarsus four. However, the hybrid protein does not rescue as efficiently as the Ap and Chip wild-type proteins (compare with F, G and H).

View larger version (31K): [in a new window]   Fig. 5. Bar interacts with the Chip and Ap proteins. (A) Representation of different domains in Chip and deleted Chip proteins. Chip contains a proline and glutamine rich (PQ rich) region at the amino-terminal end, followed by a Dimerisation Domain (DD). The LIM interaction domain (LID) is located at the carboxyl-terminal end. The Other Interaction Domain (OID) appears between amino acids 439 and 456, and mediates the interaction with Bicoid. The ChipLID protein lacks the LIM interaction domain, and the ChipOID lacks the OID domain. (B) Sample western blots of the affinity chromatography experiments using leg disc extracts; Gst-Chip fusion proteins and beads used are indicated at the top of the lanes, and the different antibodies used for immunodetection are indicated on the left. The ‘Gst’ and ‘Beads’ lanes show the lack of protein retained by beads with the Gst protein alone, and by the Gluthathione-agarose beads alone, respectively. Other lanes on the top row show an 62 kDa band in the anti-Bar western blot, corresponding with the predicted size of Bar. Bar is able to interact with Chip, and with the Chip LID- and Chip OID-deleted proteins, but it does not interact with Gst or with beads alone. Similarly, in the middle row a 46 kDa band is detected in the anti-Lim1 western blot, showing that Chip interacts with Lim1. However, a decrease of signal of this band is detected in the ChipLID lane, as has also been found with Ap (Torigoi et al., 2000), corroborating that the LID is crucial for the interaction between Ldb and LIM-HOM proteins. The lack of the OID domain does not affect this interaction. Finally, in the bottom row the western blot shows that Al interacts with Chip. An 40 kDa band corresponding with the predicted size of Al is detected. A decrease of the signal is observed in the ChipLID lane and the signal is almost undetectable in the ChipOID lane. Thus, both protein domains are necessary for the proper binding of Al. (C) Representation of the protein domains in Ap and Ap-LIM proteins. The Ap protein contains two LIM domains at the amino-terminal part of the protein followed by a homeodomain. The Ap-LIM protein consists of the amino-terminal end containing the LIM domains. (D) Western blot carried out similar to that shown in B, but with Gst-Ap constructs. The same 62 kDa band was detected using the anti-Bar antibody. Bar interacts with Ap and Ap-LIM, as well as with Chip, but it does not interact with Gst or with beads alone. The increase of signal in the Ap-LIM lane in comparison with in the Ap and Chip lanes is due to the higher molarity of Ap-LIM protein loaded in comparison with Ap and Chip proteins.

View larger version (87K): [in a new window]   Fig. 6. Genetic relationships between the tarsal gene Bar and the pretarsal genes al and Lim1. (A) Pattern of expression of a reporter Bar-lacZ in a wild-type late third instar leg imaginal disc. (B) A Lim1R12.4 mutant leg imaginal disc showing Bar-lacZ reporter expression invading the pretarsal region (arrow, compare with A). (C) Bar-lacZ expression in an al strong mutant (alex/alice), which lacks al and loses Lim1 expression (Pueyo et al., 2000). Bar expression invades the whole pretarsal region at the centre of the disc. (D) Bar-lacZ expression in an alice/alex mutant background expressing Lim1 driven by the dppGal4 driver. Expression of Lim1 is not able to repress Bar expression in the absence of Al function (compare with Fig. 3C,D-D''). (E) Leg from a dppGal4/UAS-Bar fly. Ectopic expression of Bar produces fusion of the proximal segments, such as femur, tibia and the first tarsal segment (arrow), and in the pretarsus one claw is missing (arrowhead). (F) Ectopic Al expression (arrow) produced by dppGal;UAS-Lim1. (G) High magnification of the pretarsal region from a leg imaginal disc with a clone of cells deficient for Bar (outlined in white). Lim1 expression (red) extends into the clone. (H-H''') A dppGal4/UAS-GFP leg disc showing Lim1 protein in blue (H), Al protein in red (H'), and the pattern of Gal4 expression in green (H''), in an otherwise wild-type leg. (H''') Merged image. (I-I''') A dppGal4/UAS-GFP;UAS-Bar leg disc stained as in H. Ectopic expression of Bar represses Lim1 and Al in the pretarsus (compare with H).

View larger version (88K): [in a new window]   Fig. 7. Functional relationships between LIM-HOM, Prd-HOM, Chip and Dlmo proteins in the pretarsus. (A) Ectopic expression of Al using the apGal4 driver produces a loss of the fourth tarsal segment. (B) Co-expression of Bar in an apGal4/UAS-al background partially rescues the dominant-negative effect on the development of the fourth tarsal segment (compare with A). (C) Ectopic expression of Islet using the dppGal4 driver causes fusion of the femur, tibia and first tarsal segment. In the pretarsus only one claw develops (arrowhead). (D,D') A dppGal4/UAS-GFP;UAS-islet leg disc showing Lim1 protein distribution (D, blue) and Lim1 protein distribution plus the Gal4 pattern of expression (D', green). Ectopic expression of Islet does not repress Lim1 expression; therefore the dominant-negative effect on Lim1 function seems to be posttranslational. (E) Ectopic expression of a truncated Chip protein lacking the LIM interaction domain with the DllGal4 driver produces a similar phenotype in the pretarsus to that seen in DllGal4;UAS-Chip flies: lack of the claws (arrowhead), and fusion of the fifth and fourth tarsal segments (compare with Fig. 1G). (F) A DllGal4;UAS-dlmo leg. Ectopic expression of the dlmo gene mimics the Lim1 lack-of-function phenotype. Arrowhead denotes the pretarsus lacking the claws. (G) Ectopic expression of Lim1 driven by the dppGal4 driver disrupts leg development causing the fusion of femur, tibia, and tarsus one to three. In the pretarsus it produces a similar phenotype to that seen in Lim1 mutants, or after ectopic expression of Lim1 antagonists (arrowhead; compare with C). (H) Pattern of Lim1 expression in a wild-type leg imaginal disc. (I) Lim1 protein distribution in a DllGal4;UAS-dlmo leg imaginal disc. The Lim1 protein is detected in a normal number of pretarsal cells, suggesting that the ectopic Dlmo effect on Lim1 function is not transcriptional, although part of the Lim1 domain is disorganised (compare with H). The white dotted line denotes the proximal limit of the DllGal4 pattern of expression.

View larger version (15K): [in a new window]   Fig. 8. Specific developmental functions are carried out by different partnerships between interacting LIM-HOM and HOM proteins. (A) Ap function in the wing is carried out by a complex of Ap and Chip. This unit dimerises to form a tetrameric complex comprising two molecules of Ap bridged by a Chip dimer. The relative stoichiometry of the two proteins is important for the formation of these complexes. Dlmo regulates Ap function by sequestering Chip into non-functional complexes. (B) Ap-Chip complexes are also necessary for the proper development of Ap motoneurones. However, balanced amounts of Chip and Ap are not required for tetrameric complex formation indicating that the limiting factor is Ap. In addition, there is no regulation by Dlmo. (C) In the fourth tarsal segment, Ap function might be achieved by a multiprotein complex, comprising Ap, Bar and Chip proteins. Our experiments indicate that the limiting factor in the formation of functional complexes is Bar, whereas Ap and Chip are more abundant. Bar interacts with Chip but not through the OID domain. This Ap-Chip-Bar functional unit could dimerise to produce a hexamer, or could consist of a molecule of each Ap and Bar bridged by a dimer of Chip. (D) High levels of Bar expression are required for the development of the fifth tarsal segment. As loss of Chip also affects the fifth tarsal segment, it is possible that a heterodimer of Bar and Chip is the functional unit in tarsus five. This unit could dimerise to produce a tetramer. (E) Synergism between Al and Lim1 is required for pretarsal development. Lim1 and Chip interact through their LIM and LID domains, respectively, and Al is also able to interact with Chip. In addition, genetic experiments show that Chip, Al and Lim1 are required in balanced amounts, suggesting that the functional unit in the pretarsus involves these three proteins simultaneously.