Independent regulation of initiation and maintenance phases of Hoxa3 expression in the vertebrate hindbrain involve auto- and cross-regulatory mechanisms
Miguel Manzanares1,*,
Sophie Bel-Vialar1,
Linda Ariza-McNaughton1,
,
Elisabetta Ferretti2,
Heather Marshall1,4,
Mark M. Maconochie3,
Francesco Blasi2 and
Robb Krumlauf1,4,
1 Division of Developmental Neurobiology, MRC National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 1AA, UK
2 Molecular Genetics Unit, DIBIT, Università Vita-Salute S. Raffaele, via Olgettina 58, 20132 Milan, Italy
3 Mammalian Genetics Unit, MRC, Harwell, Oxon OX11 0RD, UK
4 Stowers Institute for Medical Research, 1000 East 50th Street, Kansas City, MO 64110, USA
* Present address: Department of Developmental Neurobiology, Instituto Cajal, CSIC, Av. Doctor Arce 37, 28002 Madrid, Spain
Present address: The Sanger Centre, Wellcome Trust Genome Campus, Hinxton, Cambridgshire CB10 1SA, UK
View larger version (79K):
[in a new window]
|
Fig. 1. Differential maintenance of Hoxa3 and Hoxb3 expression is conserved in mouse and chick embryos. (A-D) Dorsal views of flat-mounted mouse hindbrains (A,B) and chick hindbrains (C,D) hybridized with Hoxa3 (B,D) or Hoxb3 (A,C). (E-F) Lateral views of mouse (E,F) and chick embryos probed with Hoxa3 (F,H) or Hoxb3 (E,G). All mouse embryos and tissue were at stage 10.5 dpc, and chick embryos and flat-mounts at HH stage 16. The black arrowheads indicate the r4/r5 boundary. Note that in both species, Hoxb3 is downregulated in r5 and r6, while Hoxa3 is maintained in the anterior segments.
|
|
View larger version (72K):
[in a new window]
|
Fig. 2. Identification and activity of a chick Hoxa3 r5/r6 enhancer in mouse and chick embryos. (A) Diagram of the upstream region of the chick Hoxa3 locus and fragments used to make constructs for chick and mouse transgenic analysis. On the right is indicated the construct number (#) and also the number (N) of transgenic embryos, all reproducibly expressing the constructs in r5/r6 in the mouse analysis. (B,E) Reporter expression at HH16 in the chick neural tube electroporated on the right side with construct c1. Anterior expression reproducibly (n=12) maps to r4/r5. (C,F) Dorsal (C) and lateral (F) views of transgene expression in 10.0 dpc mouse embryos carrying construct c1. Note that the pattern and anterior boundary is the same as that detected in chick embryos (see B,E). (D,G) Dorsal (D) and lateral (G) views of reporter expression in r5/r6 directed by a region in construct c1 conserved with the mouse Hoxa3 gene (see Fig. 3). nc, neural crest; ov, otic vesicle. The black arrowheads indicate the r4/r5 boundary. All mouse embryos shown are at 10.0 dpc.
|
|
View larger version (34K):
[in a new window]
|
Fig. 3. Sequence alignment of a region conserved between the chick, mouse, human and horn shark Hoxa3 5' flanking regions. The kreisler-binding site KrA previously identified in the mouse enhancer (Manzanares et al., 1999a) is boxed, as are the putative bipartite HOX/PBC sites A and B, and the Prep/Meis consensus motif. The unbroken black lines above the sequence indicate the double-stranded oligonucleotides A3-PP2, A3-PH1 and A3-PHP1 used in binding and competition assays (Fig. 6). Dashes represent an identity and dots a missing nucleotide.
|
|
View larger version (19K):
[in a new window]
|
Fig. 4. The mouse Hoxa3 locus: transgenic constructs tested for regulatory activity. At the top is a diagram of the genomic region and below an expanded view illustrating the wild-type and mutated genomic fragments linked to a lacZ vector used for stimulating reporter activity in transgenic analysis. The gray box indicates the conserved region whose sequence is shown in Fig. 3. The KrA-binding site is marked by a red box, the HOX/PBC-A site by a blue circle, the HOX/PBC-B site by a purple circle and the Prep/Meis site by a white circle. The black triangle (Post) represents elements that direct posterior expression in mesoderm and neural tissue; they function independently of the segmental elements. Black X in white squares or circles mark mutated sites. The construct numbers (#) for each fragment and a summary of their domains of expression in r5, r6 and posterior (p) regions is indicated at the right of each fragment. Details of the timing and numbers of expressing embryos are provided in Table 1. Restriction sites are RV, EcoRV; A, AvaI; S, SmaI; P, PstI; and N, NotI.
|
|
View larger version (75K):
[in a new window]
|
Fig. 5. A kreisler-independent activity in the conserved block regulates late segmental expression. (A-C) A large (A,B) and a minimal (C) fragment containing the conserved block from mouse mediates both early (8.25 dpc) and late (9.5 dpc) reporter expression in r5/r6. (E-G) Mutation of the KrA sites eliminates the early but not the late phase of r5 expression in transgenic mouse embryos. (D) A fragment containing the conserved block from chick reproducibly (n=8) directs reporter expression in r5/r6 at late stages (HH14) when electroporated in ovo into the right side of the neural tube. (H) Similar to the result in the mouse, mutation of the KrA site in the chick enhancer (n=11) does not abolish its ability to mediate later stage expression. The relevant constructs used in each case are indicated below each panel with the stages assayed listed above.
|
|
View larger version (63K):
[in a new window]
|
Fig. 6. Consensus bipartite HOX/PBC sites in the enhancer and their properties. (A) List of characterized HOX/PBC sites and Meis/Prep sites found in target genes aligned with those detected in Hoxa3. Mutated sequences used in binding and transgenic assays are indicated. (B,C) Reporter expression in 10.0 dpc mouse embryos carrying a construct with five copies of the HOX/PBC-A site linked to lacZ. Note strong expression in r5/r6; arrowhead marks posterior neural expression. (D,E) Electrophoretic mobility shift assays where a labeled double-stranded oligonucleotide containing the Hoxb2 HOX/PBC site and its associated Prep/Meis-binding site (B2-PP2)(Ferretti et al., 2000) has been mixed with the combinations of Pbx1a, Prep1 and Hoxb1 proteins (noted above the panels) in the absence or presence of varying amounts of cold competitor oligonucleotides spanning the HOX/PBC-A site (A3-PP2; E) or the HOX/PBC-B site (A3-PHP1; D). MUTA and MUTB are mutant forms of the competitors carrying the changes noted in A. Arrows at the sides indicate shifted complexes interacted with dimeric and trimeric combinations. (F) Gel shift assay where a labeled double-stranded oligonucleotide containing the Hoxa3 HOX/PBC-B site (A3-PH1; Fig. 3) has been mixed with the combinations of Pbx1a, Hoxb3, Hoxa3 or Hoxd3 proteins (noted above the panels) in the absence or presence of a 100 times excess of cold competitor oligonucleotides containing the wild-type (A3-PH1) or mutated form of the HOX/PBC-B site (MUTB). The addition of anti-Pbxa antibodies ( -Pbxa) inhibits complex formation.
|
|
View larger version (73K):
[in a new window]
|
Fig. 7. Transgenic assay indicates HOX/PBC sites are necessary for enhancer activity in the hindbrain. (A,F) Dorsal (A) and lateral (F) views of reporter expression in 10.0 dpc embryos directed by a fragment that spans the conserved block from the mouse Hoxa3 locus in which the KrA site has been mutated (construct 7, Fig. 4). The activity of this fragment is dependent upon the late control elements. (B,C,G,H) When the HOX/PBC-A (B,G) or the HOX/PBC-B (C,H) sites are mutated (constructs 8 and 9, respectively), expression in r5/r6 is reduced. (D,I) Reporter staining is completely abolished when both HOX/PBC sites (construct 10) are mutated. (E,J) Mutations in the Prep/Meis motif (construct 11) have no effect on reporter activity. All embryos are at 10.0 dpc. The respective constructs and mutated sites are noted at the bottom.
|
|
View larger version (20K):
[in a new window]
|
Fig. 8. Model for regulation of Hoxa3 and Hoxb3 and common role for auto/cross-regulatory mechanisms in maintaining of Hox gene expression in the developing hindbrain. (A) Hox genes are grouped in paralogous relationships and listed under the column ‘Domain/gene’. The specific rhombomeric domain(s) of expression for each set of Hox genes is listed on the left. Under the column ‘Initiation’, is shown the types of factors or sites that are involved in triggering the initiation of segmental expression of the respective Hox genes in the nervous system. The ovals indicate binding sites for retinoic acid receptor elements (RARE, black); kreisler (Kr, orange); Ets factors (yellow); and Krox20 (Krox, brown). Under ‘Maintenance’ is illustrated the distribution of bipartite Hox/Pbx (HOX/PBC, blue ovals) and Meis/Prep/Hth (Meis/Hth, green box) -binding motifs that form the Hox auto or cross-regulatory elements. The purple oval for Hoxb4 indicates a homeodomain (HD) binding site necessary for auto-regulatory activity in which the factor binding has not been identified. Note that, to date, only one paralog from each group has a Hox auto/cross-regulatory response element, while they often share common types of initiation elements. For references on the identification and roles of the sites see Materials and Methods. (B) Model comparing the regulatory interactions leading to similar, yet distinct patterns in the initiation and maintenance of the segmental expression of the paralogous Hoxa3 and Hoxb3 genes. Early expression of kreisler (red) at 8.0-8.25 dpc triggers segmental expression of Hoxa3 (blue) and Hoxb3 (purple) from 8.25-9.0 dpc, via the presence of kreisler-binding sites (KrA, Kr1 and Kr1) in the genes. Unlike the r5/r6 domain of Hoxa3, Hoxb3 expression is restricted to r5, by cooperation of kreisler with Krox20 and Ets proteins. In later stages (9.0-11.5 dpc), when kreisler is downregulated, segmental expression of Hoxa3 is maintained in r5 and r6 by Hox and Pbx factors through two HOX/PBC sites that are not present in Hoxb3 control regions. Hoxb3 is expressed more posteriorly in r7 of the hindbrain via the action of a shared enhancer, with Hoxb4 itself also dependent upon Hox /Pbx interactions.
|
|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati 
Twitter What's this?
© The Company of Biologists Ltd 2001
