Members of the Popeye domain containing gene family were identified independently by two groups using a subtractive hybridization approach aiming at the isolation of novel heart-restricted cDNAs in the chick embryo (Reese et al., 1999; Andrée et al., 2000). Orthologues were identified in other vertebrates (Hitz et al., 2002; Ripley, 2004), lower chordates (Davidson and Levine, 2003; Davidson et al., 2003), as well as insects (Lin et al., 2002). In invertebrates (Drosophila), a single gene seems to be present; in lower chordates (Ciona intestinalis), two genes have been identified; and three genes, popdc1 (also designated as Pop1, Bves), popdc2, and popdc3, constitute the vertebrate Popeye gene family (Brand, 2005). Popdc1 and popdc3 genes are present on the same chromosome (chromosome 6q21 in man, 10 in mouse, and chromosome 3 in chick) and are organized as a tandem; both genes are separated by approximately 17 kB in mouse and man and 9 kB in the chick (Andrée et al., 2000). The popdc2 gene is on a separate locus on chromosome 3 in man, chromosome 16 in mouse, and chromosome 1 in the chick genome (Andrée et al., 2000).
The Popdc proteins have a novel structure lacking any known protein domain; however, one typical signature of this family is a 70 amino acid long hydrophobic domain close to the N-terminus, which by computer algorithms is predicted to form three transmembrane helices. Recently, it was shown that these putative transmembrane domains are indeed functional (Knight et al., 2003). In addition to this transmembrane domains, a conserved 150 amino acid long sequence is present in each Popdc protein and was termed the Popeye domain (pfam04831). Analysis of the membrane topology of the Popdc1 protein established that the amino terminus of Popdc1 protein is extracellular and the carboxyl terminus is cytoplasmic (Knight et al., 2003). Recent evidence suggests that Popdc1 protein can form homodimers and the dimerization motif is part of the Popeye domain (Knight et al., 2003; Vasavada et al., 2004). It has been proposed that Popdc proteins might act as a modulator of cell adhesion; however, its biochemical interaction partner or its exact function presently is unknown (Reese et al., 1999; Wada et al., 2001, 2003). A null mutation for Popdc1 in mice has been reported (Andrée et al., 2002). In these mice, there was no embryonic lethality and normal viability during postnatal life. However, null mutants displayed an impaired ability to regenerate skeletal muscle (Andrée et al., 2002).
The cloning of chick popdc1 and popdc2 and recent analysis in the case of the mouse popdc3 gene revealed the presence of several different isoforms that are generated by alternative splicing (Andrée et al., 2000; Breher et al., 2004). These splice isoforms encode proteins that differ at the carboxyl terminus. In case of the mouse popdc3 gene, alternative splicing also occurs in exons that encode the transmembrane domains and the Popeye domain (Wiegand et al., unpublished). Thus, each popdc gene generates multiple splice isoforms.
The Popdc1 and Popdc2 transcripts are strongly expressed in myocardium and skeletal muscle as visualized by Northern blot and in situ hybridization (Andrée et al., 2000; Breher et al., 2004). By reverse transcriptase-polymerase chain reaction (RT-PCR), Popdc1 transcripts at low levels are also detectable in stomach, gut, brain, kidney, lung, and spleen (Andrée et al., 2000). On the basis of the LacZ staining pattern in tissues of mice with a β-galactosidase knockin into the popdc1 locus, expression in the lung is confined to smooth and cardiac muscle cells lining pulmonary veins (Fleige et al., unpublished observations). Expression in stomach and gut is confined to the smooth muscle cell layer of the digestive tract. Searching of the expressed sequence tag (EST) database as well as results of SAGE data consistently revealed expression in cell types other than striated and smooth muscle cell types, including for example the embryonic pancreas or melanocytes. A polyclonal antiserum (D033) directed against a conserved peptide of Popdc1 detected in the chick embryo a protein that is prominently expressed in the proepicardial organ as well as smooth muscle cells of the coronary arteries (Reese et al., 1999). The expression of Popdc1 in proepicardial cells was corroborated further by the demonstration of expression of Popdc1 both at mRNA and protein level in a rat epicardial cell line (Wada et al., 2003). However, until today expression in the proepicardium and later in the epicardium remains controversial. No expression of popdc genes was found in the proepicardium of the chick embryo by RT-PCR analysis (Breher et al., 2004). Moreover, only in the case of bone morphogenetic protein (BMP) 2–induced trans-differentiation of proepicardial cells to the myocardial cell lineage, popdc gene expression became detectable in this cell type (Schlüter et al., manuscript in preparation). A monoclonal antiserum directed against the carboxyl-terminus of Popdc1 yielded no evidence for expression of Popdc1 in the epicardium with the exception of embryonic day 6 heart (DiAngelo et al., 2001; Vasavada et al., 2004). Moreover, no evidence was found for coronary artery expression using this antibody (Vasavada et al., 2004). Likewise, the knockin of LacZ into the first coding exon of the popdc1 gene in the mouse revealed no evidence for Popdc1 expression in the proepicardium and epicardium or any other nonmuscle cell type of the heart at any time point during embryonic development (Andrée et al., 2002). Another monoclonal antibody (B0846) directed against Popdc1 detected several epithelial structures in the early chick embryo (Wada et al., 2003; Osler and Bader, 2004). This result was surprising because the mRNA expression of Popdc1 has been reported to start at Hamburger Hamilton (HH) stage 11 (Andrée et al., 2000).
To study Popdc1 expression during early chick embryogenesis and to find out whether expression of Popdc1 protein and mRNA do coincide, we analyzed the expression of Popdc1 protein using a monoclonal Popdc1 antibody (DiAngelo et al., 2001). To analyze and compare protein expression pattern with that of the mRNA, a full-length chick Popdc1A probe was used for whole-mount in situ hybridization analysis. In the heart, Popdc1 mRNA was detectable approximately 7.5 hr earlier than the protein. Thus, mRNA and protein expression differed transiently, which suggests a possible posttranslational control of Popdc1 protein. Both, mRNA and protein analysis revealed a consistently weaker expression level in the newly added anterior and posterior heart segments than in segments that were derived from the primary heart fields (Kelly and Buckingham, 2002; Abu-Issa et al., 2004). Consistent with our previous reports, we found no evidence for Popdc1 expression in the proepicardium, epicardium, or in the coronary vasculature. Surprisingly, analysis of mRNA expression revealed novel early noncardiac expression domains of Popdc1; however, no expression was found at the protein level. Thus, Popdc1 mRNA and protein expression differ in the early chick embryo; however, in the heart, Popdc1 mRNA and protein localization was similar.
Recently, the characterization of a monoclonal antibody directed against Popdc1 has been reported (DiAngelo et al., 2001; Vasavada et al., 2004). The initial description of Popc1 protein expression using this antibody, however, failed to analyze Popdc1 protein expression during early chick embryonic development. We first characterized the antigen specificity of this antibody. Western blot analysis with protein extracts of bacteria expressing the cytoplasmic part of chick Popdc1 as a fusion protein with maltose-binding protein revealed that this monoclonal antibody recognized the cytoplasmic domain of Popdc1 protein (Fig. 1A). In addition, detergent extracts of various organs of embryonic day 7 chick embryos were subjected to Western blot analysis (Fig. 1B). We found exclusive Popdc1 protein expression in the heart sample. Two immunoreactive protein bands were visible, a major protein band of 58 kDa and minor protein band of approximately 55 kDa. After substituting detergent extraction by an extraction buffer containing urea, immunoreactive bands were also seen in case of the skeletal muscle samples (data not shown), suggesting that popdc1 protein in skeletal muscle is crosslinked to some other protein complex that cannot be dissolved by detergents only, as has been previously reported (Vasavada et al., 2004). To determine the developmental time of first Popdc1 protein expression, chick embryos of HH stage 4 to 12 were subjected to Western blot analysis using the Popdc1 antibody (Fig. 1C). Popdc1 protein expression was found to start at HH stage 11. We next compared the timing of protein expression with Popdc1 mRNA expression by RT-PCR analysis using a set of primer that will amplify all known Popdc1 splice isoforms. In contrast to the Popdc1 protein expression, Popdc1 mRNA expression was detectable already at HH stage 4 (Fig. 1D). The expression level increased at HH stage 5 and subsequently decreased until HH stage 7. At HH stage 8, the expression level rose again and reached its maximum at HH stage 12 (Fig. 1D). The monoclonal Popdc1 antibody detects protein expression at HH stage 11, whereas the Popdc1 mRNA is already detectable by HH stage 4.
Since we have reported previously that Popdc1 is first expressed at HH stage 11 in the chick embryo, we made use of a full-length Popdc1A (Popdc1 splice isoform A, previously named Pop1A) cRNA probe to perform whole-mount in situ hybridization of chick embryos between HH stage 4 and HH stage 11 (Fig. 2). Consistent with the PCR data, we saw expression in Hensen's node already at HH stage 4 (Fig. 2A,E). The expression domain within Hensen's node became asymmetric at HH stage 5, being consistently stronger on the right side of the node (Fig. 2B,F). In addition, expression was found in the anterior pharyngeal endoderm in the head process. At HH stage 7 (2 somites), expression in Hensen's node still persisted; however, it was symmetrical at this time of development (Fig. 2C). Of interest, Popdc1 was also expressed in the forming notochord. Expression was found to be asymmetric being consistently stronger on the right side (Fig. 2C,G). At this time of development, expression was also present in the ventral foregut. At HH stage 8, expression persisted in the notochord (Fig. 2D,H). At HH stage 9 (7 somites), an asymmetric expression domain in right heart field was observed (Fig. 2I,M). At HH stage 9 (9 somites), when the tubular heart has formed, Popdc1 expression was confined to the presumptive left ventricular segment (Fig. 2J,N). The expression domain on the right side of tubular hearts at HH stage 9/10 (Fig. 2J,K) was consistently stronger and extended over a longer distance along the anteroposterior (A/P) axis. At HH stage 10 (11 somites), expression expanded rostrally and caudally (Fig. 2J,O). At HH stage 11 (13 somites), the Popdc1 expression domain included the right ventricular and atrial segments, however, myocardium of the conus as well as the sinus venosus did not show Popdc1 expression (Fig. 2L,P).
To compare the pattern of expression of the mRNA with the protein, we subjected chick embryos between HH stage 4 and 20 to whole-mount immunohistochemistry. No expression was seen in embryos younger than HH stage 10 (Fig. 3A and data not shown). At HH stage 10, faint expression in the embryonic heart was visible (Fig. 3B–D). At HH stage 11, robust Popdc1 expression within the cardiac myocytes of the atrial and ventricular segment of the tubular heart was seen (Fig. 3E,K). Between HH stage 12 and 15, a rostrad extension of the expression domain was observed and included first the proximal and subsequently also the distal part of the outflow tract (Fig. 3F–H,O). However, Popdc1 was not expressed in the sinus venosus at this time of development (Fig. 3L). At HH stage 18, myocardium of the outer curvature showed strong expression, while the inner curvature myocardium had little to none expression (Fig. 3I,M). At HH stage 20, the entire myocardium was strongly labeled by the Popdc1 antibody; however, upon sectioning through the ventricular myocardium, the compact layer showed slightly higher expression than the trabecular layer (Fig. 3J,N). By this time of development, the distal outflow tract myocardium also showed Popdc1 expression (Fig. 3P).
The presence of Popdc1 in the proepicardium and in the epicardium is a controversial issue (Andrée et al., 2000, 2003). We have shown previously that microdissected proepicardium does not express any of the three popdc genes, when analyzed by RT-PCR (Breher et al., 2004). We used the Popdc1 antibody to address this controversy. Staining of frozen sections of a HH stage 19 embryo revealed the absence of Popdc1 protein in the proepicardium as well as in mesothelial cells that have made contact to the ventricular myocardium (Fig. 4A,B). Also at the mRNA level, we found no evidence for proepicardial expression of Popdc1 (Fig. 4C,D).
We also analyzed expression of Popdc1 in later stages of heart development. At day 7, Popdc1 expression was present in atrial and ventricular cardiac myocytes but was not found in the epicardium, endocardium, or the atrioventricular cushions (Fig. 5A). Of interest, at this time of development, atrial myocardium expressed higher levels of Popdc1 protein than ventricular myocardium. Costaining of Popdc1 and cytokeratin revealed exclusive detection of Popdc1 in cardiac myocytes and absence of expression in the epicardium at day 6, 7, and 9 of development (Fig. 5B–D). Previous reports claimed that Popdc1 is expressed in the coronary vasculature. We therefore analyzed Popdc1 expression in a day 18 heart and found no evidence for expression in the coronary vessels by costaining of Popdc1 and smooth muscle α-actin (Fig. 5E). These data demonstrate that Popdc1 protein as recognized by this monoclonal antibody was exclusively present in cardiac myocytes.
This study analyzed Popdc1 expression during early chick embryogenesis. In the heart, Popdc1 mRNA and protein expression were temporally separated by approximately 7.5 hr. Both, mRNA and protein analysis revealed a consistent weaker expression level in the newly added anterior and posterior heart segments, suggesting that the level of Popdc1 expression is correlated with the extent of myocardial differentiation. Consistent with our previous reports, Popdc1 expression was not found in the proepicardium or later in the epicardium. Surprisingly, analysis of mRNA expression revealed novel expression domains of Popdc1 at HH stage 4 in Hensen's node and, subsequently, in the notochord. At the protein level, no expression was found in these structures. Thus, Popdc1 mRNA and protein expression differ in the early chick embryo.
Popdc1 Displays Left–Right Asymmetry in Hensen's Node
We found asymmetric expression in Hensen's node at HH stage 5. At this time of development, many genes such as fibroblast growth factor (FGF) 8 (Boettger et al., 1999), Sonic hedgehog (Shh; Levin et al., 1995), ActRIIa (Levin et al., 1995), N-Cadherin (Garcia-Castro et al., 2000), BMP4 (Monsoro-Burq and Le Douarin, 2001), Pcl2 (Wang et al., 2004), and NCX-1 (Linask et al., 2001) display left–right (L-R) asymmetric expression domains in Hensen's node in the chick embryo. These asymmetries, although subtle, are of functional significance, because any alteration in their expression pattern results in aberrant L-R axis formation (reviewed in Brand, 2003). ActRIIa, BMP4, FGF8, Pcl-2, and N-Cadherin are all expressed on the right side of the Hensen's node; however, each of these genes also display expression in the primitive streak. In contrast, Popdc1 and NCX-1 expression is confined to Hensen's node (Linask et al., 2001). At the moment, it is unknown whether any of the signaling or regulatory factors asymmetrically expressed on the right side up-regulate Popdc1 expression, or whether Popdc1 asymmetry is generated by a down-regulation of Popdc1 expression on the left side. The null mutant for Popdc1 did not provide any evidence for an involvement of this gene in setting up the L-R axis (Andrée et al., 2000, 2003). Of significance in this regard is our recent finding that Popdc2 in the mouse is expressed in the node (Froese et al., unpublished observations). Nonetheless, coexpression of Popdc1 with N-Cadherin in Hensen's node is an interesting finding, because Popdc1 has been implicated in modulating cell–cell interaction (Wada et al., 2001; Andrée et al., 2003). Recently, it has been described that, not only molecular asymmetries are present during HH stages 4–7, but that Hensen's node and its descendent structures such as the forming notochord also display morphological asymmetries (Dathe et al., 2002). Whether Popdc1 might be involved in modulating epithelial polarity in this context is an interesting question and should be analyzed further. We also found expression of Popdc1 in the forming notochord, which is a structure that is generated by a convergent extension mechanism (Domingo and Keller, 1995). In this regard, it is noteworthy that treatment with a morpholino against Popdc1 interferes with convergent extension in Xenopus embryos (Ripley, 2004).
Tubular Hearts Diplay L-R Asymmetric Expression of Popdc1
L-R asymmetry was also observed within the heart field. Between HH stage 9 and stage 10, the right heart field displayed stronger Popdc1 expression than the contralateral side. Of interest, Popdc2 in the chick also showed asymmetric expression, but it was the left side that showed stronger expression (Breher et al., 2004). Many genes display L-R asymmetric expression during early heart formation. The transcription factor Pitx2 is expressed in the left portion of the cardiac crescent and in the left side of the heart tube (Campione et al., 2001). In the precardiac mesoderm, the extracellular matrix molecules hLAMP and flectin on the left and JB3 on the right side are asymmetrically distributed (Smith et al., 1997). In frogs and zebrafish, BMP4 is predominantly expressed on the left side (Breckenridge et al., 2001). Consistent with our observation of asymmetric expression in the presumptive left ventricular segment is the observation that the left and right heart field have unequal contributions to the tubular heart along the A/P axis. The right heart field has a greater contribution to the rostral end (presumptive left ventricle) of the tubular heart and less to the caudal end (Stalsberg, 1969). In addition, there are some differences in temporal progression of cardiac differentiation between the left and the right side (Satin et al., 1988). Significantly, the first contractions in the tubular heart are observed in the right margin of the ventricular portion at the late 9- to 10-somite stage (Sakai et al., 1996). Whether and how this relates to asymmetric Popdc1 expression in the heart requires further analysis.
Popdc1 Expression in the Tubular Heart Seems to Be Associated With Differentiated Cardiac Myocytes
Popdc1 also displayed a restricted expression pattern along the A/P axis. Initially, expression at the mRNA level was confined to the left ventricular segment at HH stage 9. Subsequently, expression gradually extended both caudally to include the atrium and rostrally to include first the future right ventricle, the conus, and finally the truncal region of the outflow tract. The stepwise addition of positively stained cardiac segments is also seen at the protein level. When heart looping progressed, there was a higher level of Popdc1 expression in the myocardium of the outer curvature and less in the inner curvature, consistent with our previous description at the mRNA level (Andrée et al., 2000, 2003). It has been proposed that the myocardium of the outer curvature is the future chamber myocardium, whereas the inner curvature myocardium represents remnants of primary myocardium (Christoffels et al., 2004). Genes such as Nppa, Chisel, and Cx40 in the mouse heart are representatives of a chamber-specific gene expression program, which is under the control of various transcription factors of the Tbx gene family (Habets et al., 2002; Singh et al., 2005). To what extent Popdc1 expression at the post-looping stage is under the control of this chamber-specific gene expression program is an unresolved question. At embryonic day 7, myocytes in the compact layer apparently had a higher level of expression than trabecular cardiac myocytes and the atrium displayed higher levels of expression than the ventricular myocardium. It is well known that gene expression in the compact layer is modulated by signaling factors derived from the epicardium.
Popdc1 Is Not Expressed in the Proepicardium or in the Epicardium
Previously, Popdc1 has also been named bves (blood vessel/epicardial substance), based on its expression in the proepicardium, the epicardium, and its derivatives the coronary arteries (Reese et al., 1999). We have reported previously that none of the three Popdc cDNAs were detected by PCR amplification of mRNA isolated from microdissected proepicardia (Breher et al., 2004). We now show that that there is no evidence for Popdc1 expression in the proepicardium, or subsequently in the epicardium, both at the mRNA and protein level. Moreover, we show here that Popdc1 is not found to be expressed in the smooth muscle layer of the coronary vessels. Expression of Popdc1 in the proepicardium has been observed previously by immunohistochemical staining using antibodies directed against conserved peptide sequences within the cytoplasmic domain of the Popdc1 protein (Reese et al., 1999; Wada et al., 2001, 2003; Osler and Bader, 2004). At present, we do not know whether these monoclonal antibodies detect only Popdc1 or whether these antibodies crossreact with some other antigen that is expressed in the proepicardium. In addition, it was proposed recently that Popdc1 is expressed in the early gastrula in the epiblast and later in various epithelial structures such as neural tube and somites (Osler and Bader, 2004). Our RNA localization does not provide any evidence for expression in the epiblast at HH stage 4 but rather very specific expression in Hensen's node and subsequently expression in the forming notochord and in the pharyngeal endoderm. Thus, the easiest explanation for these divergent results obtained with the D033 and B0846 antibodies, generated and characterized by the Bader laboratory (Reese et al., 1999; Wada et al., 2001, 2003; Osler and Bader, 2004), is that these antibodies do crossreact with another peptide unrelated to Popdc1. Alternatively, there might exist Popdc1 isoforms generated by alternative splicing that are not recognized with the antibody used in this study.
In a previous study of Popdc1 protein expression using the same monoclonal antibody, it was reported that Popdc1 was transiently expressed at day 6 in the epicardium (Vasavada et al., 2004). We were unable to confirm this finding and, rather, show here that Popdc1 is exclusively expressed in the cardiac myocyte and is not expressed in the epicardium, particularly not at day 6 of development. These differing results are probably due to some technical reason.
Popdc1 Antibody Does Not Recognize the Early Noncardiac Expression Domains of popdc1 mRNA
There are some differences between mRNA localization and protein detection in our data. The early expression of Popdc1 at the mRNA level in Hensen's node and in the notochord as well as in pharyngeal endoderm is not seen by antibody staining. One possibility is that the Popdc1 mRNA is not translated before the heart reaches a functional state. There are some precedents for this: both smooth muscle α-actin and Troponin T mRNA are detected already in the early heart field, whereas translation only occurs shortly before contraction starts (Colas et al., 2000; Antin et al., 2002). An alternative explanation would be that the protein produced by the mRNA that is present in early gastrulating embryos is not detected by the antibody used here. The isoforms for Popdc1 that are known presently are generated through alternative splicing and differ at the carboxyl terminus (Andrée et al., 2000, 2003). In case of the related gene popdc3 in the mouse, we have obtained evidence recently for the presence of an alternative splicing event within the Popeye domain. These splice products may not be recognized by the monoclonal antibody used here. To distinguish between these alternatives, we will characterize the splice isoform production in case of the Popdc1 gene during chick development.
Fertilized chicken eggs (White Leghorn, Gallus gallus) were obtained from Lohmann, Cuxhafen. Eggs were incubated at 38°C and 75% relative humidity. Staging of the embryos was performed according to Hamburger and Hamilton (Hamburger and Hamilton, 1951).
For immunohistochemical detection of Popdc1, a monoclonal antibody (3F11-D9-E8) developed by Melinda Duncan (DiAngelo et al., 2001) was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by The University of Iowa, Department of Biological Sciences (Iowa City, IA 52242). Chick embryos were fixed for 2 hr with 4% paraformaldehyde, sucrose infiltrated, and embedded in OCT. Ten-micrometer sections were taken on Superfrost slides. The sections were first pretreated with 0.3% hydrogen peroxide in MeOH/Aceton (1:1) for 30 min at room temperature to block endogenous peroxidase, washed for 2–3 min with water, and incubated with blocking solution (5% horse serum in 1 × phosphate buffered saline [PBS]). This procedure was followed by an incubation with the 3F11-D9-E8 monoclonal antibody at a 1:10 dilution. After several washes with PBS, the sections were incubated with a 1:200 dilution of a horseradish peroxidase conjugated horse anti-mouse antibody (Vector Laboratories). After several washes with PBS and 0.05 M Tris (pH 7.6), the immunoreaction was color developed using diaminobenzidine. Some sections were counterstained with Mayer's hematoxylin.
For immunofluorescent detection of Popdc1 and costaining of Popdc1 with smooth muscle α-actin (clone 1A4, mouse monoclonal smooth muscle α-actin antibody SIGMA) or cytokeratin (clone LU-5, mouse monoclonal pan-cytokeratin antibody Acris antibodies), the frozen sections were incubated with blocking solution and subsequently with antibodies directed against Popdc1 (1:10 dilution), cytokeratin (1:200), or smooth muscle α-actin (1:200). Popdc1 was detected with a donkey anti-mouse antibody conjugated with Alexa Fluor 555 (Molecular Probes) and binding of the cytokeratin and smooth muscle α-actin antibodies were detected with a donkey anti-mouse Fab fragment conjugated to fluorescein isothiocyanate (Dianova). After the final wash, nuclei were counterstained with 4′,6-diamidine-2-phenylidole-dihydrochloride.
Embryos were washed three times in PBS and fixed in a methanol/dimethyl sulfoxide (DMSO) mixture (4:1) at 4°C overnight. Endogenous peroxidase was blocked by incubating the embryos with methanol/DMSO/30% H2O2 mixture (4:1:1) for 2 hr at room temperature. Unspecific binding sites were blocked by incubating the embryos 2 × 1 hr in 2% BSA, 0.1% Triton X-100 in PBS at room temperature. Embryos were incubated overnight in a 1:200 dilution of the 3F11-D9-E8 antibody in TBST (0.8 g of NaCl, 20 mg of KCl, 25 mM Tris-Cl, pH7.5, 1% Tween-20) containing 1% horse serum. After 5 washes for 1 hr with TBST, the embryos were incubated with a 1:200 dilution of a horseradish peroxidase conjugated horse anti-mouse antibody (Vector Laboratories). After 5 washes for 1 hr with TBST, the immunoreaction was color developed using diaminobenzidine. For cryostat sectioning, the embryos were infiltrated overnight at 50°C with a 7.5% gelatin/15% sucrose solution and snap-frozen in dry ice–cooled isopentane.
Whole-Mount In Situ Hybridization
Whole-mount in situ hybridization was carried out as described (Andrée et al., 1998). For expression analysis of Popdc1, a 1.4-kb full-length cDNA clone (ChEST59k12) was identified in the ChickEST Database (Boardman et al., 2002) and obtained from the MRC geneservice. For cryostat sectioning, the embryos were infiltrated overnight at 50°C with a 7.5% gelatin/15% sucrose solution and snap-frozen in dry ice–cooled isopentane. The section shown in Figure 4D was counterstained with Fast Red.
For biochemical analysis, tissues were lysed in RIPA buffer (1 ×PBS, 1% Igepal Ca-630 [Sigma], 0.5% Natriumdesoxycholate, 0.1% sodium dodecyl sulfate [SDS], 10 μl/ml of a protease inhibitor cocktail; Roche). Each sample (40 μg of protein) was run on a 10% SDS polyacrylamide gel and electroblotted to a nitrocellulose membrane (Bio Trace; Pall). The membrane was blocked in 5% skim milk, 1 × TBS, and 0.1% Tween-20 and then incubated with a 1:1,000 dilution of the monoclonal 3F11-D9-E8 Pop1 antibody. The membrane was washed three time with 1 × TBS-0.1% Tween-20 solution and then incubated with a 1:1,000 dilution of a horseradish peroxidase–conjugated horse anti-mouse antibody (Vector Laboratories). in 1 × TBS, 0.1% Tween-20, 5% milk solution. The membrane was again washed three times, immersed in substrate solution (ECL, Amersham) for 1 min, and then exposed to X-ray film.
For RNA isolation, chick embryos of the indicated stages, from which the extraembryonic tissue had been removed were used. cDNA was synthesized from DNase treated total RNA by using AMV reverse transcriptase. PCR was performed by using the primer pairs: Popdc1fwd, 5′-TTGCTCACCGTAGGATGTGC-3′; and Popdc1rev, 5′-CGGTTCATCTGAGTTGATCG-3′. cDNA input was controlled by amplification reactions with glyceraldehydes-3-phosphate dehydrogenase (GAPDH) primer: GAPDHfwd 5′-ACGCCATCACTATC TTCCAG-3′, GAPDHrev 5′-CAGGCCTTCACTACCCTCTTG-3′. The PCR products were size separated on 1% agarose gels.
Production of a Maltose Binding Protein–Popdc1 Fusion Protein
The carboxyl-terminus of Popdc1 was amplified by Ex-Taq polymerase (Cambrex Bioscience) using primers: Pop1MBPA, 5′-GATAATTCAGCCTGTACAAGAGA ATGTTTGAACCACTC-3′; Pop1MBPB, 5′-GAATCTAGATCAAGGCAGCCGCT GCAGCTCAAGCTTTTC-3′. The resulting fragment was restricted with EcoRI and XbaI and subcloned into pMal-2 vector (New England Biolabs). The resulting plasmid was transformed into BL21 cells. For the production of Popdc1 fusion protein, the recombinant bacterial cells at logarithmic growth were induced with IPTG for 2 hr. For control purposes, cells that had not been induced by IPTG (0 hr) were used. In each case, 500-μl cell suspension was centrifuged and the resulting cell pellet was taken up in SDS sample buffer and size-separated on a 12% SDS-PAGE. The resulting gel was processed for Western blot analysis.
This project was funded by Deutsche Forschungsgemeinschaft, BR1218/9-4, “Functional Characterization of the Popeye gene family” and GRK1048 “Organogenesis.”