SEARCH

SEARCH BY CITATION

Keywords:

  • AP-2;
  • transcription factor;
  • mouse embryogenesis;
  • gene expression

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

AP-2 proteins comprise a family of highly related transcription factors, which are expressed during mouse embryogenesis in a variety of ectodermal, neuroectodermal, and mesenchymal tissues. AP-2 transcription factors were shown to be involved in morphogenesis of craniofacial, urogenital, neural crest-derived, and placental tissues. By means of a partial cDNA fragment identified during an expressed sequence tag search for AP-2 genes, we identified a fifth, previously unknown AP-2–related gene, AP-2ϵ. AP-2ϵ encodes an open reading frame of 434 amino acids, which reveals the typical modular structure of AP-2 transcription factors with highly conserved C-terminal DNA binding and dimerization domains. Although the N-terminally localized activation domain is less homologous, position and identity of amino acids essential for transcriptional transactivation are conserved. Reverse transcriptase-polymerase chain reaction analyses of murine embryos revealed AP-2ϵ expression from gestational stage embryonic day 7.5 throughout all later embryonic stages until birth. Whole-mount in situ hybridization using a specific AP-2ϵ cDNA fragment demonstrated that during embryogenesis, expression of AP-2ϵ is mainly restricted to neural tissue, especially the midbrain, hindbrain, and olfactory bulb. This expression pattern was confirmed by immunohistochemistry with an AP-2ϵ–specific antiserum. By using this antiserum, we could further localize AP-2ϵ expression in a hypothalamic nucleus and the neuroepithelium of the vomeronasal organ, suggesting an important function of AP-2ϵ for the development of the olfactory system. Developmental Dynamics 231:128–135, 2004. © 2004 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

AP-2α was identified due to its ability to bind to the SV40 and the human metallothionein IIa gene promoters and initially was considered to represent a unique transcription factor without any homology to other transcriptional regulators (Mitchell et al., 1987). In 1995, a second homologous gene, AP-2β, was cloned (Moser et al., 1995) and, subsequently, two further AP-2 genes, AP-2γ (also known as AP-2.2) and AP-2δ, enlarged the family of AP-2 transcription factors (Bosher et al., 1996; Chazaud et al., 1996; Oulad-Abdelghani et al., 1996; Zhao et al., 2001). All currently known AP-2 proteins share a modular structure consisting of an N-terminal proline- and glutamine-rich transactivation domain, followed by a positively charged α-helical DNA binding region and a helix-span-helix motif, which mediates homo- and heterodimerization of AP-2 proteins (Williams and Tjian, 1991a, b; Bosher et al., 1996). The C-terminal domains of AP-2 proteins share the highest degree of similarity and were also highly conserved during evolution, as cloning of the Drosophila AP-2 protein revealed 68% identity with AP-2α in this region (Bauer et al., 1998; Monge and Mitchell, 1998). In contrast, the N-terminal transactivation domain appears to be structurally more flexible and, therefore, less conserved between the individual AP-2 proteins, although, with the exception of AP-2δ, certain critical residues and motifs involved in transcriptional activation are conserved. A limited number of only 36 critical amino acid residues was mapped previously and is believed to interact with coactivators of the transcription machinery (Wankhade et al., 2000).

Expression analyses in mammals, birds, and amphibians showed that AP-2 genes are involved in the formation of craniofacial, neuroectodermal, ectodermal structures, limb buds, and urogenital tissues (Mitchell et al., 1991; Snape et al., 1991; Moser et al., 1995, 1997a; Shen et al., 1997; Epperlein et al., 2000). The functional relevance of AP-2 proteins for mammalian development was demonstrated by the lethal phenotypes of AP-2α, AP-2β, and AP-2γ mouse mutants. Of interest, the phenotypes of these mutants differ significantly, despite that the expression patterns of AP-2α, AP-2β, and AP-2γ genes overlap in many tissues during mouse development. AP-2α knockout mice exhibit severe craniofacial defects and defective closure of the anterior neural tube and body wall (Schorle et al., 1996; Zhang et al., 1996; Nottoli et al., 1998; Brewer et al., 2002). AP-2β–deficient mice complete embryogenesis, but die shortly after birth due to impaired kidney function (Moser et al., 1997b, 2003). Finally, AP-2γ mutant embryos die very early after gastrulation as a result of defective placental development (Auman et al., 2002; Werling and Schorle, 2002).

Recently, germ line missense mutations in the human AP-2β gene have been linked to Char syndrome, characterized by patent ductus arteriosus, facial malformations, and hand anomalies. AP-2β mutations in patients with Char syndrome result in a transdominant negative AP-2β protein with defective DNA binding properties (Satoda et al., 1999, 2000). These results clearly indicate that AP-2 genes execute also essential nonredundant functions during human development.

In this study, we describe a new fifth member of the AP-2 transcription factor family, AP-2ϵ, which is mainly expressed in the central nervous system during murine embryogenesis. We additionally describe here the generation and characterization of an AP-2ϵ–specific antiserum that can be used for Western blots, gel mobility shift assays, and immunohistochemistry.

RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

Identification of the Mouse AP-2ϵ cDNA

By a BLAST search for AP-2 genes, we identified a partial cDNA clone, which encoded a predicted peptide with significant homology to the central region of AP-2 proteins (AA414551; IMAGE 778986). By using this partial cDNA sequence, an extended expressed sequence tag (EST) BLAST search identified several overlapping EST clones (described in the Experimental Procedures section). Because all of these independent ESTs encoded the same predicted peptide with significant homology to the DNA binding and dimerization domains of the four previously known AP-2 proteins, we assumed that they were derived from a distinct, fifth AP-2 protein, which we designated AP-2ϵ. Unfortunately, all these clones lacked the 5′ region of the new AP-2ϵ gene. Specific primer pairs were chosen to amplify a 551-bp cDNA fragment by reverse transcriptase-polymerase chain reaction (RT-PCR) that was subsequently used as a probe to screen an embryonic day (E) 14 mouse embryo cDNA library. The cDNA library screen identified one positive lambda phage clone and the identity of the AP-2ϵ cDNA was verified by sequencing. Again, this cDNA clone only covered the 3′ region of the murine AP-2ϵ cDNA. Finally, a BLAST search to human EST clones identified a full-length AP-2ϵ cDNA clone (IMAGE 5786430). In addition to the coding sequence, this clone contained 232-bp 5′- and 664-bp 3′-untranslated regions (UTRs) and the insert was used to generate a cytomegalovirus (CMV) promoter-driven AP-2ϵ expression plasmid.

By using the NIX software package, the murine genomic sequence from Celera and the NCBI database, we determined the genomic organization of the murine AP-2ϵ gene. The overall structure of the mAP-2ϵ gene is highly conserved and, as known from AP-2α, β, and γ genes, consists of seven exons spanning approximately 20,000 bp. The genomic organization suggests that these genes were most likely derived by gene duplication of a single ancestor. The exon–intron boundaries follow the conserved splice donor and acceptor sites (Table 1). The NCBI database maps the AP-2ϵ gene to human chromosome 1 and mouse chromosome 4. Table 2 summarizes the chromosomal loci of all five human and murine AP-2 genes.

Table 1. Overview of Exons and Exon–Intron Junction of AP-2ε
Exon numberExon size (bp)Sequence at exon–intron junction
3′splice acceptor5′splice donor
1>144 TCC GCC ATG gtg agt
2573ctt ctg cag GAG CGC CCCGAA TTG CAG gta agc
352tat ttg cag GCG ATA GATTCA AGA AAG gta agg
4223gtt cca cag TCC CCA TTCCCT CCG AAG gta gga
5119tac gtt aag GGC CAA GTCTGG TGG AAG gta agc
6142ttt ccc cag GAG AAG CTGGGC TGC CAA gtg agt
7>283ccc cac cag GCA ǴAT CTGCAT CGG AAG taa ctg
Table 2. Chromosomal Locations of Human and Murine AP-2 Genes
AP-2Chromosome location
HumanMouse
α6p2413 A5-B1
β6p121 A2-A4
γ20q13.22 H3-H4
δ6p12.11 A3
ε1p34.34 D2.2

The AP-2ϵ gene encodes a predicted peptide of 434 amino acids harboring all modular domains shared by the classic AP-2 genes, AP-2α, β, and γ. Highly conserved DNA binding and dimerization motifs homologous to the other murine AP-2 proteins are present in the C-terminal half starting from amino acid 208 (Fig. 1). The transcriptional activation domain located in the N-terminal half is much less conserved between the different AP-2 proteins. However, functionally important residues, including the PY motif at positions 47 to 51 in AP-2ϵ, the aspartic acid residue corresponding to position 52 in AP-2α and the two leucine residues at positions 107 and 108 of AP-2α (Wankhade et al., 2000), are present in AP-2ϵ and in all other AP-2 isoforms, except for AP-2δ (Fig. 1). Of interest, AP-2ϵ lacks a short N-terminal peptide, a feature that has been also observed in Xenopus AP-2 (Winning et al., 1991).

thumbnail image

Figure 1. Clustal W alignment of murine AP-2ϵ and the previously known AP-2α, β, γ, and δ proteins. Identical amino acid residues are boxed in black with white lettering, whereas similar residues are shown in gray boxes. Gaps between amino acids are filled with a dash. Functional AP-2 domains are indicated above by horizontal bars.

Download figure to PowerPoint

Tissue-Specific Expression of AP-2ϵ During Murine Embryogenesis and in Adult Tissues

To determine the expression profile of AP-2ϵ during murine embryogenesis, RT-PCR was performed with AP-2ϵ–specific primers chosen from the 3′ end and the 3′ UTR of the AP-2ϵ cDNA (see Experimental Procedures section). Gel electrophoresis revealed bands on ethidium bromide–stained agarose gels for all analyzed embryonic stages from E7.5 to E17.5. The expression signal strikingly increases from E8.5 to E14.5 and declines thereafter. Of interest, expression levels at E7.5 were higher compared with later stages, probably due to the fact that these RNA samples were prepared from whole conceptuses, suggesting that AP-2ϵ, as well as AP-2α and AP-2γ (Moser et al., 1997a; Auman et al., 2002), is also expressed in extraembryonic structures (Fig. 2A).

thumbnail image

Figure 2. AP-2ϵ expression during murine embryogenesis. A: Reverse transcriptase-polymerase chain reaction with an AP-2ϵ–specific primer set reveals AP-2ϵ expression during embryogenesis from embryonic day (E) 7.5 until E17.5. As a control, the quality and amount of cDNA preparations were tested by amplification of a β-actin fragment. B–G: Whole-mount in situ hybridization of E8.5 (B,C), E9.0 (D,E), and E9.5 (F,G) mouse embryos. AP-2ϵ–specific signals are present in a restricted region of the neural folds of the future midbrain at E8.5 (marked by an arrow in B and C). At E9.0 and E9.5, AP-2ϵ is expressed in the midbrain region (arrowheads) the hindbrain (arrow in G) and spinal cord (arrowhead in G).

Download figure to PowerPoint

Next, we transcribed sense and antisense digoxigenin-labeled riboprobes from an AP-2ϵ–specific cDNA fragment obtained from the 3′ UTR to determine embryonic expression patterns by whole-mount in situ hybridization. Specificity was verified by hybridizing the probe to a Southern blot with full-length cDNAs of AP-2α, β, γ, and δ and the partial murine AP-2ϵ cDNA clone (data not shown). Whole-mount in situ hybridization of E8.5 embryos revealed AP-2ϵ expression in a distinct patch of cells in the neural folds of the prospective midbrain region (Fig. 2B,C). This region increased in size, when neural folds started to fuse at E9.0 (Fig. 2D,E). From E9.5 on, AP-2ϵ signals were abundant in the hindbrain primordial anterior of the fourth ventricle and in the spinal cord. In addition, strong signals were detected in the midbrain and midbrain–hindbrain junction (Fig. 2F,G). Of interest, a pair of distinct patches of cells in the anterior midbrain became evident at E10.5 (Fig. 3A,B). Thus, AP-2ϵ expression partially overlaps with expression of other AP-2 genes in these regions. In particular, AP-2α and AP-2β were previously detected in the midbrain, the midbrain–hindbrain junction, the primordia of the cerebellum and the spinal cord (Moser et al., 1995, 1997a). Additionally, AP-2δ expression has also been localized to the midbrain region at this time point (Zhao et al., 2003). Of interest, AP-2ϵ expression was also observed in the developing olfactory bulb, in which no other AP-2 genes are expressed at this early stage of development (Fig. 3D–G). Sectioning of these embryos confirmed the restricted AP-2ϵ expression in the neuroepithelium and did not reveal any signals in neural crest cells (data not shown).

thumbnail image

Figure 3. Whole-mount in situ hybridization of embryonic day (E) 10.5 (A–C), E11.5 (D,E), and E12.5 (F,G) mouse embryos hybridized with an AP-2ϵ–specific probe. A,C: AP-2ϵ is expressed in the midbrain and hindbrain. B: Note the two lateral regions of AP-2ϵ expression in the anterior midbrain. C: Backside view indicates staining anterior of the fourth ventricle (long arrow) and the spinal cord (arrow). D–G: Arrows indicate AP-2ϵ expression in the primordium of the olfactory bulbs of E11.5 and E12.5 embryos. Control hybridizations performed with a sense probe did not reveal any signals (data not shown).

Download figure to PowerPoint

To analyze AP-2ϵ expression in the brain at later stages, we dissected brains from E12.5 to E15.5 embryos and performed whole-mount in situ hybridizations. Figure 4 shows different perspectives of an E14.5 brain hybridized with the AP-2ϵ probe. This stage reveals very strong expression of AP-2ϵ in the olfactory bulb and a much weaker signal in the midbrain. In addition, signals could also be detected in distinct nuclei of the pons and medulla oblongata and the spinal cord (Fig. 4A–D). In parallel, we performed radioactive in situ hybridizations on sections of E12 to E16.5 embryos. Of interest, we were not able to detect AP-2ϵ in the midbrain any longer at E16.5, suggesting a down-regulation of AP-2ϵ in this structure, whereas expression in the olfactory bulb was still detectable (data not shown). Finally, we were unable to detect any AP-2ϵ expression in the brain of 1.5- and 3-month-old mice by in situ hybridization (data not shown). The decrease in AP-2ϵ expression was also confirmed by RT-PCR analyses from adult tissue preparations showing only very weak signals in the brain and ovary (data not shown).

thumbnail image

Figure 4. Whole-mount in situ hybridizations of brain from embryonic day 14.5 embryos hybridized with an AP-2ϵ–specific cDNA probe. A–D: Lateral (A), backside (B), ventral (C), and frontal (D) view of the same brain showing AP-2ϵ expression in the olfactory bulb and the mesencephalon. Additional sites of AP-2ϵ expression are distinct regions in the pons (arrowhead in A), medulla oblongata (arrows in A, B, and C), and spinal cord (arrowhead in C).

Download figure to PowerPoint

Generation of AP-2ϵ–Specific Antiserum and Immunohistochemical Analysis of AP-2ϵ Expression

To analyze AP-2ϵ protein expression in more detail, we aimed to generate an AP-2ϵ–specific polyclonal antiserum (see Experimental Procedures section). AP-2ϵ–transfected NIH3T3 cells showed strong nuclear staining when incubated with the antiserum raised against an N-terminal peptide of the AP-2ϵ protein (data not shown). Specificity was tested by Western blot analysis of lysates from HepG2 cells that were transfected with cDNAs of all five AP-2 isoforms (Fig. 5A). In addition, we performed gel mobility shift assays of in vitro translated AP-2 isoforms together with the optimized hMtIIa binding sequence (Mohibullah et al., 1999). Here, a supershift was only formed with the AP-2ϵ–DNA complex (Fig. 5B). Both assays confirmed that the antiserum specifically recognizes the AP-2ϵ protein without any cross-reactivity.

thumbnail image

Figure 5. Characterization of an AP-2ϵ–specific antiserum. A: Western-blot analysis from RIPA extracts of HepG2 cells transfected with expression plasmids of all five AP-2 isoforms probed with the anti–AP-2ϵ antiserum. Equal loading was confirmed by anti-tubulin staining. B: Gel mobility shift assays of in vitro translated (IVT) AP-2 proteins incubated with radioactively labeled optimized hMtIIa and the AP-2ϵ–specific antiserum shows the formation of a supershifted AP-2ϵ–DNA antibody complex.

Download figure to PowerPoint

Finally, we used this antiserum for immunohistochemistry of paraformaldehyde-fixed sections of E12.5 and E15.5 embryos. Again, the specificity of this antiserum was proven by the nuclear staining and the staining pattern itself, which clearly confirmed the data obtained from in situ hybridizations. At E12.5, we detected labeled cells in the midbrain and the most cranial region of the telencephalon (Fig. 6A–C). Of interest, AP-2ϵ–positive cells are not limited to the roof of the midbrain, where three other members of the AP-2 family, AP-2α, AP-2β, and AP-2δ, are expressed at that time point in more differentiated neuronal cells (Moser et al. 1995; Zhao et al. 2003; own unpublished data). In contrast, AP-2ϵ–positive cells are distributed throughout the neuroepithelium in neuroblasts that have left the proliferative ventricular zone (Fig. 6B). These data indicate that (1) AP-2ϵ is expressed already earlier during the maturation process of neuronal cells and (2) its expression does not overlap at the cellular level with the other members of the AP-2 family in the midbrain. Although, we were able to show an interaction between AP-2ϵ with other members of the AP-2 gene family in vitro, by performing pull down analysis and gel shift assays, our data here do not support any in vivo relevance of this interaction. Therefore, this point needs further examinations (data not shown).

thumbnail image

Figure 6. A–C: Immunohistochemical staining with the AP-2ϵ–specific antiserum of sagittal sections of the head region of embryonic day (E) 12.5 embryos. B and C show magnifications of the boxed area in A. Single AP-2ϵ–positive cells in the midbrain (arrowheads in B) and the most cranial telencephalon (arrow in C). D,E: Coronal and sagittal sections of the olfactory bulb from E15.5 embryos, respectively. AP-2ϵ is expressed in mitral cells of the olfactory bulb and the rostral accessory olfactory bulb (encircled by a dashed line in E). F: AP-2ϵ is further expressed in the ventrolateral nucleus of the hypothalamus (indicated by arrows). G,H: Transverse (G) and coronal (H) sections of E15.5 and newborn mice of the vomeronasal organ, respectively. AP-2ϵ–positive cells are found in the sensory neuroepithelium of the vomeronasal organ. AOB, accessory olfactory bulb; III, third ventricle; MC, mitral cells; NS, nasal septum; OE, olfactory epithelium; OV, olfactory ventricle; VNO, vomeronasal organ.

Download figure to PowerPoint

From E12.5 onward, strongest AP-2ϵ expression is seen in the olfactory system. At E15.5, no AP-2ϵ–positive cells were stained in the midbrain. In contrast AP-2ϵ protein expression could be detected in the mitral cell layer of the olfactory bulb and the rostral portion of the accessory olfactory bulb (Fig. 6D,E).

Chemosensation in most mammals is achieved by at least two distinct nasal tissues: the main olfactory epithelium and the vomeronasal organ. The axonal processes of both structures project to mitral cells of the olfactory bulb and the accessory olfactory bulb, respectively. In both structures, AP-2ϵ is expressed. Of interest, AP-2ϵ is also expressed in the vomeronasal organ, which represents the sensory organ of social chemical stimuli, like pheromones, whereas no expression is found in the olfactory epithelium. In the vomeronasal organ, AP-2ϵ is only present in the crescent-shaped sensory epithelium and not in the laterally located nonsensory epithelium at E15.5 and in newborn animals (Fig. 6G,H). The vomeronasal neuroepithelium is divided into an apical and basal zone of sensory epithelium. Axons of apical vomeronasal neurons terminate on mitral neurons in the rostral zone of the accessory olfactory bulb, where AP-2ϵ is also expressed (Zufall et al., 2002; reviewed in Halpern and Martinez-Marcos, 2003). At the moment, we cannot claim, that AP-2ϵ–positive cells only belong to the apical zone of the vomeronasal neuroepithelium, it rather seems that it is expressed in neuroepithelial cells of both layers. Finally, pheromonal stimuli induce endocrinological changes in the animal, which are mainly regulated by certain hypothalamic nuclei controlling the secretion of hormones by the hypophysis. Of interest, AP-2ϵ expression could be also detected in a ventrolateral nucleus of the hypothalamus (Fig. 6F). Whether these cells control hypophyseal function needs further investigation.

Taken together, our expression data reveal a restriction of AP-2ϵ expression to neuronal cells of the midbrain, hindbrain, olfactory bulb, and vomeronasal neuroepithelium, which clearly differs with respect to time and spatial pattern from all other members of the AP-2 family. We therefore speculate that AP-2ϵ fulfills essential and nonredundant functions for embryonic development in these organs.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

Identification of AP-2ϵ and Cloning of a Partial Murine AP-2ϵ cDNA Clone

Computer searches using the BLAST algorithm identified a murine cDNA clone with high homology to the AP-2 gene family (AA414551; IMAGE 778986). An extended database search using the NCBI-BLAST software revealed several EST clones containing partial cDNA clones of the murine AP-2ϵ gene (BF464155, BI134735, BF461668, BE987978, AI50520).

The predicted cDNA sequences were used to design primer sets to amplify an AP-2ϵ probe (mAP-2ϵ sense1: GTC TTC CAG GAT TGG CGG AG; mAP-2ϵ anti2: GGC TGG AAA CTC AGT CTC AC). As a template, we used total cellular RNA extracted from a mouse embryo at gestational stage E14. The resulting cDNA fragment of 551 bp was then used to screen a murine embryonic E14 cDNA library (AMS Biotechnology Europe LTD, Lugano, Switzerland). Handling of lambda phages and plaque lifting was performed as described previously (Moser et al., 1995). After screening 4.5 × 105 lambda phages, one positive clone was identified, which contained the 3′ part of the AP-2ϵ cDNA (see Fig. 1).

Both the lambda clone and, in parallel, the IMAGE cDNA clone (IMAGE 778986) were sequenced on both strands using an ABI377 DNA sequencer. Finally, a full-length human EST clone (IMAGE 5786430) containing 5′ and 3′ UTRs was also obtained, sequenced, and used for further experiments. The genomic organization of the mouse AP-2ϵ gene was mapped by using the NIX software package and the NCBI database, together with a comparison of the cDNA sequences.

RT-PCR and Whole-Mount In Situ Hybridization

Total cellular RNA from mouse embryos of different stages and tissues from adult C57BL/6 mice were isolated by using Trizol (Invitrogen, Karlsruhe, Germany) following the manufacturer's instructions. cDNA synthesis was performed with 2 μg of total RNA using the Superscript II cDNA synthesis kit (Invitrogen). For RT-PCR studies, the following primer set were taken: mAP-2ϵ spec sense: CAA GCA TCG GAA GTA ACT GGC; mAP-2ϵ spec anti: CAC CTC TGA TGT GTT ATC AGC.

Whole-mount in situ hybridization was performed essentially as described (Wilkinson, 1993). As a probe, a 526-bp fragment obtained from the 3′ UTR region of the murine AP-2ϵ cDNA was excised with StuI/SalI from the murine IMAGE clone (IMAGE 778986) and subcloned into pBluescript. Digoxigenin-labeled cRNA sense and antisense probes were in vitro transcribed from the linearized plasmid as described previously (Moser et al., 1995). The same fragment was gel-purified and labeled with [32P]dCTP by using a random primed labeling kit (Amersham, Braunschweig, Germany). Southern-blot hybridization to linearized plasmids of the AP-2α, AP-2β, AP-2γ, AP-2δ, and AP-2ϵ cDNAs indicated that the probe hybridized specifically to AP-2ϵ but not to any other AP-2 gene.

Generation of AP-2ϵ–Specific Antiserum and Western Blotting

Polyclonal AP-2ϵ–specific antiserum was raised in rabbits by using peptide immunogens fused to KLH. The AP-2ϵ–specific peptide was MERPDGLGGAAAGGTR, which represents the N-terminus of the protein. For Western-blot analysis, 20 μg of RIPA cell lysates of HepG2 cells transfected with each of the five AP-2 expression plasmids were loaded per well, separated on 10% sodium dodecyl sulfate-polyacrylamide gel and transferred to nitrocellulose. A 1:5,000 dilution of the AP-2ϵ antiserum was used before specific antibody binding was detected with the ECL system (Amersham).

Immunohistochemical Analysis

Paraffin sections of E12.5, E15.5, and newborn mice were rehydrated in alcohol solutions, endogenous peroxidase activity was blocked with 3% H2O2/methanol for 15 min, followed by antigen retrieval by heat treatment in 10 mM sodium citrate buffer (pH 6.0). After incubation with 10% goat serum/3% bovine serum albumin, sections were incubated with AP-2ϵ–specific antiserum (at a dilution of 1:2,500) at 4°C overnight. Positive signals were developed with peroxidase-conjugated secondary antibody using diaminobenzidine followed by counterstaining with methyl green.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

M.M., A.K.B., and R.B. were funded by grants from the Deutsche Forschungsgemeinschaft.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES