SEARCH

SEARCH BY CITATION

Keywords:

  • smooth muscle α-actin gene;
  • contractile proteins;
  • heart muscle;
  • in situ hybridization;
  • Rana catesbeiana;
  • Xenopus laevis

Abstract

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

A gene encoding a putative homologue of the avian and mammalian vascular smooth muscle α-actin was isolated from an amphibian, Rana catesbeiana, and characterized in terms of its sequence, organization, and expression pattern. To assess the expression of this gene during amphibian embryonic development, a cDNA encoding the Xenopus homologue of this mRNA was isolated and characterized by in situ hybridization. The expression of this gene was not detected in the enteric smooth muscle cells or, unlike its avian and mammalian homologues, in the somites/skeletal muscle of the Xenopus embryos/tadpoles. Its initial expression coincides with the onset of cardiac muscle differentiation and is coincidental with the expression of the cardiac α-actin mRNAs in the heart-forming region of the stage 26/27 embryo. As development proceeds, transcripts from this gene are expressed throughout the developing heart until the formation of the heart chambers is completed and, thereafter, its expression becomes restricted to the outflow tract of the tadpole heart. The subsequent restricted expression of this gene to the vascular system in both of these amphibians identifies it as the amphibian homologue of the avian and mammalian vascular smooth muscle α-actin. Developmental Dynamics 233:1546–1553, 2005. © 2005 Wiley-Liss, Inc.


INTRODUCTION

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

Actin is a protein that is a major component of the organized contractile apparatus of muscle cells. In birds and mammals, muscle actin is composed of a family of four distinct isoforms of this protein. Two of these isoforms, the α-vascular and γ-enteric, are primarily expressed in smooth muscle cells, whereas the remaining two, the α-cardiac and α-skeletal, are restricted to striated muscle (Vandekerckhove and Weber, 1978, 1984; McHugh et al., 1991). In Amphibia, however, there appear to be at least three actin isoforms in striated muscle (Stutz and Spohr, 1986, 1987) and only a single isoform in the smooth muscle cells (Vandekerckhove and Weber, 1984). This latter point has led to speculation regarding the evolutionary time in which duplication events occurred that resulted in the separation of an ancestral smooth muscle actin gene into the smooth muscle α-vascular and γ-enteric isoforms present in mammals, birds, and some reptiles (see Vandekerckhove and Weber, 1984; White and Crother, 1999). The concept that this duplication event occurred during the evolution of primitive amphibia into the first reptiles or among the stem reptiles (Vandekerckhove and Weber, 1984) was supported by studies (Saint-Jeannet et al., 1992) conducted to provide a marker for studying mesoderm regionalization in Xenopus. In their studies, Saint-Jeannet and colleagues used a monoclonal antibody directed against a decapeptide corresponding to the NH2-terminal peptide of a mammalian vascular smooth muscle α-actin and demonstrated that it recognized a smooth muscle α-actin present in both the visceral and vascular systems of Xenopus. Their results led them to conclude that there is only one smooth muscle actin in Xenopus and that it serves the function of both the smooth muscle vascular α-actin and enteric γ-actin isoforms present in birds and mammals.

On the basis of the supposition that Xenopus and other amphibians do not contain a gene encoding a smooth muscle α-actin that is specific for and limited to the vascular system, we were quite surprised when we isolated a cDNA from an amphibian, Rana catesbeiana, adult heart cDNA library that appeared to encode a homologue of the avian and mammalian vascular smooth muscle α-actin. This observation, coupled with the notion that a similar gene is not thought to be present or expressed in Xenopus (Vandekerckhove and Weber, 1984; Saint-Jeannet et al., 1992), prompted us to characterize the R. catesbeiana gene encoding this mRNA and reexamine the possibility that Xenopus might, in fact, express a similar gene. We report here the characterization of a R. catesbeiana gene encoding a vascular smooth muscle α-actin as well as the characterization and developmental expression pattern of an mRNA encoding a vascular smooth muscle α-actin in Xenopus laevis.

RESULTS AND DISCUSSION

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

Characterization of a cDNA Encoding a Putative Smooth Muscle α-Actin Isoform in the Heart of R. catesbeiana

We screened an adult R. catesbeiana heart library with an X. laevis cardiac α-actin cDNA and obtained several clones that, when sequenced, appeared to encode three different isoforms of α-actin. Although the putative open reading frame (ORF) in each of these cDNAs was the same size, the nucleotide sequences in the putative untranslated regions (UTRs) were different. One of these clones, Rsmα-actin (Fig. 1A; GenBank accession no. AY986488), corresponds to a 1,380 nucleotide (nt) mRNA having an ORF of 1,134 nt, a 5′-UTR of 71 nt, and a 3′-UTR of 175 nt. A BLAST search of the nucleotide sequences in the ORF confirmed that it encoded a member of the actin family and that it shared equally high identity (86%) with the X. laevis α-cardiac actin isoform, and the mammalian and avian smooth muscle α-actins. A comparison of the derived amino acid sequence of the protein encoded from the ORF of this cDNA with the reported X. laevis sarcomeric α-actins (Stutz and Spohr, 1986, 1987) indicates that the amino acid sequence in this actin (Fig. 2) is highly conserved (98-99% identity). However, by virtue of particular amino acids at sites that have been used to distinguish the smooth muscle actin isoforms from the other actin isoforms (i.e., a cysteine at position 19 and a serine at position 91; Vandekerekhove and Weber, 1984; Rubenstein, 1990), we tentatively identified (see White and Crother, 1999) Rsmα-actin as an mRNA encoding a smooth muscle actin. Moreover, that the size of the protein derived from this mRNA is different from the γ-actins (376 amino acids) but identical to the avian and mammalian smooth muscle α-actins (377 amino acids) suggested that this mRNA encodes a protein that might be the R. catesbeiana homologue of the human and avian vascular smooth muscle α-actins.

thumbnail image

Figure 1. A: The nucleotide sequence of a Rana catesbeiana cDNA, Rsmα-actin, that encodes a vascular smooth muscle α-actin. Uppercase lettering is used to identify nucleotides within the open reading frame, whereas lowercase lettering denotes nucleotides in the untranslated regions of the mRNA. The nucleotide sequences that are underlined represent the sequences to which primers were made for use in the reverse transcriptase-polymerase chain reaction experiments. B: The location of each intron within the Rsmα-actin gene is indicated, relative to the finished processed transcript, by a triangle. Numbers within the boxes represent the size of each of the introns (in base pairs).

Download figure to PowerPoint

thumbnail image

Figure 2. Alignment of the cDNA-derived amino acid sequence in the Rana catesbeiana (Rsmα-actin) and Xenopus laevis (Xsmα-actin) vascular smooth muscle α-actins with the amino acid sequences reported in the X. laevis cardiac (Xcα-actin1; Stutz and Spohr, 1986) and skeletal (Xskα-actin2 and Xskα-actin3; Stutz and Spohr, 1986, 1987) muscle α-actins. Conserved residues are indicated by dashes, and amino acids assumed to be diagnostic for the smooth muscle actins (Vandekerekhove and Weber, 1984; Rubenstein, 1990) are in bold type and underlined.

Download figure to PowerPoint

Organization and Characterization of the R. catesbeiana Gene Encoding Rsmα-Actin

The gene that encodes Rsmα-actin (GenBank accession no. AY986489) contains eight introns: one in the 5′-UTR and seven in the ORF. The sites of the introns in the ORF of the Rsmα-actin gene (see Fig. 1B) correspond to the sites at which introns are found in the avian (Carroll et al., 1986) and mammalian (Ueyama et al., 1984) smooth muscle α-actin genes. The 12,384 bp of the Rsmα-actin gene that we sequenced contains 1,578 bp of the putative promoter, 33 bp of the 5′-UTR, the first intron of 4,218 bp, an exon containing the next 167 bp of the transcribed DNA (38 bp of the 5′-UTR and 129 bp of the ORF), a second intron of 1,187 bp, a 129-bp exon, a third intron of 1,647 bp, a 111-bp exon, a fourth intron of 966 bp, an 85-bp exon, a fifth intron of 249 bp, a 162-bp exon, a sixth intron of 93 bp, a 192-bp exon, a seventh intron of 319 bp, a 182-bp exon, an eighth intron of 761 bp, and a downstream region (304 bp) containing 144 bp of the ORF and 160 bp of the 3′-UTR.

A comparison of the nucleotide sequence upstream from the TATA-box (i.e., −35 to −225) reveals that this region shares 72–77% identity with the promoters of the genes encoding the chicken and human vascular smooth muscle α-actins. Moreover, two of the four CArG-box motifs present in the putative promoter of this gene are located in this region and are at sites that correspond with CArG-box elements present in the chicken and human vascular smooth muscle α-actin promoters (Fig. 3A). These elements, as well as the specific spacing of one CArG-box element to the other, are required for the expression of this gene in both birds and mammals (Nakano et al., 1991; Mack and Owens, 1999; Mack et al., 2000). That these elements, as well as the spacing between them, are conserved in the promoter of this amphibian gene suggests that its expression might be regulated in a manner similar to the avian and mammalian genes. Of interest, a BLAST search of the sequences in the introns of the gene encoding Rsmα-actin revealed a 93-bp region in the first intron that shared 81 and 70% identity (Fig. 3B) with a conserved sequence present in the first intron of both the avian and mammalian vascular smooth muscle α-actin genes, respectively (Nakano et al., 1991; Mack and Owens, 1999). This highly conserved sequence contains a CArG-box that is thought to act as a smooth muscle-specific enhancer-like element in both birds and mammals (Mark and Owens, 1999). Collectively, these similarities, in both the promoter region and first intron of this Rana gene to the avian and mammalian vascular smooth muscle α-actin genes, support the notion that Rsmα-actin mRNAs are transcribed from a gene that is the amphibian homologue of the avian and mammalian vascular smooth muscle α-actin genes.

thumbnail image

Figure 3. Transcriptional regulatory elements present in the promoter and first intron of the avian and mammalian smooth muscle α-actin–encoding genes are present in the gene encoding Rsmα-actin. A: An alignment of the sequence upstream from the TATA-box (−225 to −35) in the gene encoding Rsmα-actin with those reported for the genes encoding the chicken (Carroll et al., 1986) and human (Ueyama et al., 1984) smooth muscle α-actins reveals that it shares 72–77% identity with the chicken and human promoters. B: An alignment of a 93-bp sequence in the first intron of the gene encoding Rsmα-actin that corresponds with a CArG-box–containing enhancer sequence reported in the first intron of the avian and mammalian smooth muscle α-actin–encoding genes (Nakano et al., 1991; Mack and Owens, 1999). Asterisks denote the location of conserved nucleotides between all three organisms.

Download figure to PowerPoint

Expression of the Gene Encoding Rsmα-Actin in the Tissues of R. catesbeiana

To determine the functional identity of the gene encoding Rsmα-actin, we initially examined its expression in the liver, brain, intestine, heart, and skeletal muscle of premetamorphic tadpoles (stage X) and adult frogs by Northern blot hybridizations. The results (not shown), using a 3′-UTR probe specific for this actin isoform, indicated that this mRNA is present in the heart but not detectable in the liver, brain, intestine or skeletal muscles of these animals. These results suggested that this mRNA might encode a cardiac actin; however, the observation that the amino acid sequence of the protein derived from this mRNA is characteristic of the vascular smooth muscle α-actins, coupled with the high identity shared by the sequence in the putative promoter region of this gene with the promoters of the mammalian and avian vascular smooth muscle α-actins, raised the possibility that the vascular tissue associated with the heart (i.e., the truncus arteriosus), rather than the heart muscle, might be expressing this gene. To evaluate this possibility, RNA was isolated from an adult frog heart with the truncus arteriosus attached, from a heart with the truncus arteriosus removed, and from the excised truncus arteriosus. These RNAs were used for reverse transcriptase-polymerase chain reaction (RT-PCR) analyses with primer pairs specific for Rsmα-actin and the R. catesbeiana cardiac α-actin mRNAs (Rcα-actin1 and Rcα-actin2; GenBank accession nos. AY986486 and AY986487).

The results shown in Figure 4 demonstrate that the primer pairs prepared for amplifying the Rcα-actin1, Rcα-actin2, and Rsmα-actin mRNAs are specific for each of these actins (Fig. 4A), that the expression of the Rcα-actin1 and Rcα-actin2 mRNAs is confined to the heart musculature, and that the Rsmα-actin mRNAs are expressed primarily in the truncus arteriosus (Fig. 4B). This latter point further supports the contention that the Rsmα-actin gene encodes a vascular smooth muscle α-actin mRNA.

thumbnail image

Figure 4. A: The specificity of primer pairs prepared from Rcα-actin2 (lane 1), Rcα-actin1 (lane 2), and Rsmα-actin (lane 3) is demonstrated by the fact that the cDNA-specific primer pairs, using 1 ng of plasmid DNA (pDNA), amplify polymerase chain reaction (PCR) products of the appropriate size only from the plasmid from which they were derived. B: Reverse transcriptase-PCR (RT-PCR) analyses establish the functional identity of Rsmα-actin mRNAs. Ethidium bromide–stained gels of the products generated from RT-PCR analyses of RNA isolated from an adult frog heart with the truncus arteriosus attached (H+TA) and with the truncus arteriosus removed (H-TA), and of RNA isolated from the excised truncus arteriosus (TA). In each case, lanes 1, 2, and 3 show the PCR products generated from primers pairs specific for Rcα-actin2, Rcα-actin1, and Rsmα-actin mRNAs, respectively. The size of the products was determined by coelectrophoresed standards (Std.).

Download figure to PowerPoint

Characterization of a cDNA (Xsmα-Actin) Encoding a X. laevis Homologue of the R. catesbeiana Smooth Muscle α-Actin mRNA, Rsmα-Actin

We used the R. catesbeiana Rsmα-actin ORF sequence to search the X. laevis expressed sequence tag (EST) database and identified numerous potentially vascular smooth muscle alpha-actin sequences within adult Xenopus lung and testis cDNA libraries. The sequence of the Xsmα-actin clone that we isolated by PCR from an adult frog heart cDNA library (GenBank accession number AY986490) corresponds to a 1,385-nt mRNA having an ORF of 1,134 nt, a 5′-UTR of 29 nt, and a 3′-UTR of 222 nt. A comparison of the nucleotide sequence in the Xsmα-actin with the R. catesbeiana Rsmα-actin mRNA sequence, and the mRNA sequences reported for the X. laevis striated muscle (Stutz and Spohr, 1986, 1987) and cytoplasmic (Mohun and Garrett, 1987) actin isoforms (not shown) disclosed that it shares more identity with the Rsmα-actin (85%) than it does with any of the other reported Xenopus actin isoforms (68–72%). This observation, coupled with the fact that the cDNA-derived protein from Xsmα-actin (see Fig. 2), like the protein encoded from Rsmα-actin, contains amino acids in its amino terminal end that are characteristic of the smooth muscle actin isoforms from other organisms (Vandekerekhove and Weber, 1984; Rubenstein, 1990) supports the contention that this mRNA encodes a smooth muscle actin in X. laevis.

To further assess this notion, we constructed a phylogenetic tree that is derived from a comparison of the nucleotide sequence in the ORFs of the Xsmα-actin, Rsmα-actin, Rcα-actin1, and Rcα-actin2 mRNAs with those reported in the ORFs of mRNAs encoding other X. laevis actin isoforms and in the chicken and human smooth muscle α-actin mRNAs. That the evolutionary dendrogram constructed from this comparison (Fig. 5) demonstrates that Xsmα-actin and Rsmα-actin share a terminal node on this evolutionary tree and form a clade with the smooth muscle α-actins from other organisms indicates that they represent the products of orthologous genes.

thumbnail image

Figure 5. An evolutionary dendrogram derived from the nucleotide sequence in the open reading frame (ORF) of mRNAs encoding the putative Xenopus laevis and Rana catesbeiana vascular smooth muscle α-actins (Xsmα-actin and Rsmα-actin), the human and chicken vascular smooth muscle α-actins, Rcα-actin1 and Rcα-actin2 from Rana catesbeiana,and other actin isoforms present in X. laevis. The scale bar represents 1% divergence, and numbers at the nodes are bootstrap percentages for 1,000 iterations. The abbreviations used to designate the particular isoforms of the X. laevis actins and their GenBank accession numbers are as follows: cardiac muscle (Xcα-actin1; X04669), skeletal muscle (Xskα-actin2; X05393), skeletal muscle (Xskα-actin3; X12525), cytoplasmic Type 8 (XcytT8-actin; M24770), cytoplasmic Type 5 (XcytT5-actin; M24769), and cytoplasmic β (Xcytβ-actin; AF079161). The abbreviations used to designate the human and chicken smooth muscle α-actins and their GenBank accession numbers are Hsmα-actin (J05192) and Csmα-actin (M13756), respectively.

Download figure to PowerPoint

Expression of Xsmα-Actin in Developing X. laevis Embryos

The expression of Xsmα-actin transcripts was assessed by in situ hybridization and transcripts encoding this mRNA, unlike their mammalian and avian homologues (Woodcock-Mitchell et al., 1988; Ruzicka and Schwartz, 1988; Sawtell and Lessard, 1989), were never detected in the somites or skeletal muscle of developing embryos. Xsmα-actin mRNAs were first detected in a bilateral pair of triangular patches that are separated by the ventral midline and lie posterior to the cement gland of the late tail bud embryo (stages 26/27; Fig. 6A,B). The bilateral appearance of the Xsmα-actin mRNA in this heart-forming region of the embryo coincides with the onset of cardiac muscle differentiation (Mohun et al., 2000) and the location of its expression is similar to that observed for transcripts encoding the X. laevis cardiac muscle-specific proteins myosin heavy chain-α, myosin light chain 2a (Mohun et al., 2000), and cardiac troponin I (Drysdale et al., 1994). Interestingly, the expression of this mRNA in the heart-forming region also coincides with the expression of the cardiac (Fig. 6C,D) and skeletal muscle (Mohun et al., 1984; Hemmati-Brivanlou et al., 1990; Logan and Mohun, 1993) α-actin isoforms in this region of the Xenopus embryo. The apparent, coordinated expression of these three actin isoforms in early cardiogenesis is in contrast to the situation in mammals and birds, where the developing heart appears to sequentially expresses mRNAs encoding the vascular smooth muscle α-actin, cardiac muscle α-actin, and skeletal muscle α-actin (Woodcock-Mitchell et al., 1988; Ruzicka and Schwartz, 1988; Sawtell and Lessard, 1989). As development proceeds the expression of Xsmα-actin transcripts extends across the ventral midline and is detectable throughout the heart up to and including the onset of chamber formation (Fig. 6E–G; see Kolker et al., 2000; Mohun et al., 2000). After the heart chambers are formed (stage 46) and before metamorphic climax (stage 61), the expression of Xsmα-actin transcripts becomes confined to the outflow tract of the tadpole heart (Fig. 6H,I). The observation that the expression of this gene is never detected in the visceral tissues and is confined to the vascular system once the heart chambers are formed clearly distinguishes it from a more visceral-like smooth muscle α-actin isoform previously identified and thought to be the only smooth muscle actin isoform in Xenopus (Saint-Jeannet et al., 1992). In fact, based on functional criteria, the subsequent restricted expression of this gene to the vascular system identifies it as one that encodes the Xenopus homologue of the avian and mammalian vascular smooth muscle α-actin.

thumbnail image

Figure 6. Expression of Xsmα-actin in developing Xenopus laevis embryos. AD: Lateral and ventral views of late tail bud embryos (stage 26/27) after whole mount in situ hybridization to detect smooth muscle α-actin (Xsmα-actin; A,B) and cardiac α-actin (Xcα-actin1; C,D) gene expression. Although transcripts from both genes are colocalized in the bilateral domains of the heart-forming (h) region in these embryos, Xsmα-actin mRNA, unlike Xcα-actin1 mRNA, is not present in the somites (s). E: A lateral view of a stage 31 Xenopus embryo showing that Xsmα-actin expression is throughout the developing heart tube. F,G: Lateral views of the anterior portion of Xenopus embryos at stages 34 (F) and 41 (G) showing that Xsmα-actin transcripts are throughout the heart, up to and including the onset of chamber formation. H: In situ hybridization of a cross-section through the heart of a stage 59 tadpole showing that after the heart chambers [atria (a) and ventricle (v)] are formed Xsmα-actin transcripts are confined to the outflow tract (oft) of the heart. I: A higher magnification image of the outflow tract seen in H.

Download figure to PowerPoint

EXPERIMENTAL PROCEDURES

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

Animal Experimentation

R. catesbeiana tadpoles and adult frogs were obtained from WARD'S Natural Science and maintained in dechlorinated, aged tap water at 20°C. The tadpoles were fed tadpole food (Borel Laboratories), whereas adults were fed chopped liver. X. laevis frogs were obtained from NASCO, maintained in aged, dechlorinated tap water at 18°C, and fed frog brittle (NASCO). The developmental stages of spontaneously metamorphosing R. catesbeiana tadpoles were determined based on the morphological criteria of Taylor and Kollros (1946), whereas the X. laevis embryos/tadpoles were staged according to Nieuwkoop and Faber (1967). Tissues were removed after an animal was anesthetized in 0.1% 3-aminobenzoic acid ethyl ester (Sigma-Aldrich) solution and killed by severing the truncus arteriosus. All animals were treated in accordance with the guidelines outlined by the Canadian Council on Animal Care.

Isolation of RNA and Genomic DNA

Total RNA was extracted from various R. catesbeiana tissues using TRIzol Reagent (Invitrogen) and its associated protocol. The final RNA pellets were washed with 75% ethanol, dried, and resuspended in RNA Storage Solution (Ambion) or Millipore filtered, double-distilled water. In some cases, poly (A)+ mRNAs were isolated from the total RNA using a Dynabeads mRNA Direct Kit (Dynal) according to the supplier's protocols. Genomic DNA was isolated from the liver of a R. catesbeiana frog using a genomic DNA purification kit from QIAGEN. The final DNA preparation was solubilized in Tris–ethylenediaminetetraacetic acid (TE) buffer (pH 8.0) and quantified by spectrophotometry.

Screening of cDNA Libraries Prepared From the Heart of Adult R. catesbeiana

A cDNA library was made from poly (A)+ mRNAs isolated from the heart of a R. catesbeiana frog. Oligo (dT)18 was used for the first-strand synthesis, and the library was constructed in the EcoRI site of a lambda ZAPII vector (Stratagene). The library was screened using a [α-32P]dCTP-labeled 1,739 bp AccI/PstI fragment from a cardiac α-actin cDNA from X. laevis (pX1cA2a; Mohun et al., 1984). A total of 5 × 105 recombinants were screened. Hybridizations were performed as previously described (Chen and Atkinson, 1994) and positive clones were identified, removed from the plates, and screened twice more. After the third screening, positive clones were isolated and ligated into pBluescript II SK phagemid (Stratagene), and the cloned inserts were digested with EcoRI and sized by gel electrophoresis. Sequence analyses of the cloned DNA inserts revealed that they contained sequences from three different actin mRNAs, designated as Rcα-actin1, Rcα-actin2, and Rsmα-actin.

Determination of Transcription Start Sites by Rapid Amplification of cDNA Ends

Rapid amplification of cDNA ends (RACE) libraries were prepared from RNAs obtained from R. catesbeiana tadpole heart muscle (stage X) using a GeneRacer Kit (Invtrogen) according to the manufacture's recommendations. A primer specific for the Rsmα-actin mRNA and the GeneRacer 5′ Primer, a specialized 5′ primer located in the ligated adaptor, were used in a PCR to obtain the 5′-end of the Rsmα-actin mRNA. The product, separated on a 1.5% agarose gel, was isolated using a QIAquick Gel Extraction Kit (QIAGEN) and ligated into the pCR 2.1 TA vector (Invitrogen) for sequencing.

Genomic Walking and Genomic PCRs

Genomic DNA, isolated from the liver of a R. catesbeiana frog with a QIAGEN Genomic-tip kit (QIAGEN, Inc.), and primers, prepared from the Rsmα-actin cDNA sequence, were used with a Universal Genome Walker Kit (BD Biosciences Clontech) to obtain portions of the gene encoding this mRNA. In addition, primers were designed from the sequences we obtained from genomic walking and genomic fragments were generated by additional genomic walking and/or by use of the PCR. In all cases, Advantage Genomic Polymerase Mix (BD Biosciences Clontech) was used for the amplifications and the resulting PCR products were gel purified and cloned into a pCR 2.1 TA vector. The clones were initially sequenced with primers for M13 forward and M13 reverse and, subsequently, completely sequenced in both directions with sequencing primers (not shown) designed from the sequenced portions of the clones.

RT-PCR Analyses

RT-PCR analyses were preformed by reverse transcribing 5 μg of total RNA with PowerScript Reverse Transcriptase (BD Biosciences Clontech) and an oligo (dT)18 primer followed by PCR using 4 μl of the first-strand reaction in a 100-μl PCR reaction with Platinum Taq DNA Polymerase (Invitrogen) and the appropriate primers. All of the RT-PCRs conducted made use of RNA isolated from the tissues/organs of a single animal and were performed at the same time using the same reaction premixes. Moreover, each experiment was repeated three or more times on RNA isolated from tissues of different animals. The primers pairs used were as follows: Rcα-actin1, 5′-TGTGGCTTTGGACTTTGAGAATGAGA-3′ (forward) and 5′-TAAATATGATGCTTGAGTCGGAGGTA-3′ (reverse); Rcα-actin2, 5′-AAGATCCCTGCTACGTTCAAC-3′ (forward) and 5′-AGAAAGAAGTCCCCAAGTTATGTA-3′ (reverse); Rsmα-actin (5′-AGGCGGTGCTATCCCTTTAT-3′ (forward) and 5′-GGCCATGTACTTTTTGAA-3′ (reverse). To prevent coamplification of other actin isoforms, the Rcα-actin2 primer pair was designed to sequences within the 5′- and 3′-UTR of the Rcα-actin2 mRNA while the Rcα-actin1 and Rsmα-actin primer pairs were designed with the reverse primer complementary to a sequence within the 3′-UTR of the mRNA to be amplified. Moreover, each primer pair was designed such that at least one intron was present in the corresponding gene and in the event of genomic DNA contamination would produce a considerably larger fragment. The samples from the PCRs were separated by agarose gel electrophoresis, stained with ethidium bromide, and visualized on an ultraviolet transilluminator.

Isolation of a X. laevis Smooth Muscle α-Actin cDNA

A BLAST search of the X. laevis EST database using the coding region of Rsmα-actin revealed the presence of a potentially orthologous sequence within Xenopus. A gene-specific primer (5′-GGAAAGGTTTTAGCAAGTAGCAATAC-3′), made to a portion of the 5′-UTR of these corresponding EST sequences, and a T7 primer were used in a PCR to isolate a cDNA (Xsmα-actin) from an adult X. laevis heart library (Ji et al., 1993). An additional primer (5′-TTTGTATCGAATCAGAGACTAAGA-3′), designed from the 3′-UTR sequence in this clone, and a T3 primer were used in a similar PCR to generate a cDNA and verify the 5′-UTR sequence of the transcript. The resulting PCR fragments were gel purified and cloned into pBluescript II SK.

In Situ Hybridization

Whole-mount in situ hybridization was performed using a modification of the protocol by Harland (1991). A digoxigenin-UTP–labeled antisense riboprobe for Xsmα-actin was prepared by linearizing the Xsmα-actin cDNA with Acc I (a restriction enzyme site located within the Xsmα-actin cDNA) and transcribing with T7 RNA polymerase (Ambion) to generate a probe of 322 nucleotides in length that corresponds to the entire 3′-UTR and the last 84 nucleotides of the ORF. Longer probes for Xsmα-actin were also tested but each exhibited a much broader expression profile, including expression in the developing somites (result not shown), and are not included in this study as they likely represent cross-reaction of the probe with other members of the actin gene family. The antisense X. laevis cardiac α-actin probe has been described previously (Hemmati-Brivanlou et al., 1990).

Computer Analysis

Sequence alignments were done using ClustalW (Higgins and Sharp, 1988). The evolutionary dendrogram was created using the TREECON program (Van de Peer and De Wachter, 1994). In this program, tree topology is inferred by the neighbor joining method (Saitou and Nei, 1987) using 1,000 bootstrap iterations.

Acknowledgements

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

B.G.A. was funded by a NSERC research grant, A.S.W. was the recipient of a NSERC Postgraduate Scholarship, while L.Z. was supported by an Ontario Graduate Scholarship.

REFERENCES

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