JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Makeyev, A. V.
Right arrow Articles by Liebhaber, S. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Makeyev, A. V.
Right arrow Articles by Liebhaber, S. A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

J Biol Chem, Vol. 274, Issue 35, 24849-24857, August 27, 1999


A Set of Highly Conserved RNA-binding Proteins, alpha CP-1 and alpha CP-2, Implicated in mRNA Stabilization, Are Coexpressed from an Intronless Gene and Its Intron-containing Paralog*

Aleksandr V. Makeyev, Alexander N. Chkheidze, and Stephen A. LiebhaberDagger

From the Howard Hughes Medical Institute and the Departments of Genetics and Medicine, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104

    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Gene families normally expand by segmental genomic duplication and subsequent sequence divergence. Although copies of partially or fully processed mRNA transcripts are occasionally retrotransposed into the genome, they are usually nonfunctional ("processed pseudogenes"). The two major cytoplasmic poly(C)-binding proteins in mammalian cells, alpha CP-1 and alpha CP-2, are implicated in a spectrum of post-transcriptional controls. These proteins are highly similar in structure and are encoded by closely related mRNAs. Based on this close relationship, we were surprised to find that one of these proteins, alpha CP-2, was encoded by a multiexon gene, whereas the second gene, alpha CP-1, was identical to and colinear with its mRNA. The alpha CP-1 and alpha CP-2 genes were shown to be single copy and were mapped to separate chromosomes. The linkage groups encompassing each of the two loci were concordant between mice and humans. These data suggested that the alpha CP-1 gene was generated by retrotransposition of a fully processed alpha CP-2 mRNA and that this event occurred well before the mammalian radiation. The stringent structural conservation of alpha CP-1 and its ubiquitous tissue distribution suggested that the retrotransposed alpha CP-1 gene was rapidly recruited to a function critical to the cell and distinct from that of its alpha CP-2 progenitor.

    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Two highly similar human proteins, alpha CP-1 and alpha CP-2 (also known as heterogeneous nuclear ribonucleoprotein E1/PCBP1 and heterogeneous nuclear ribonucleoprotein E2/PCBP2), belong to a growing family of K homology (KH)1 domain containing RNA-binding proteins that includes Nova-1, an autoantigen in paraneoplastic opsoclonus myoclonus ataxia (1); and FMR1, which is associated with fragile X mental retardation syndrome (2). The alpha CP proteins appear to be multifunctional. Both alpha CP-1 and alpha CP-2 have been identified as components of a ribonucleoprotein complex that assembles on the 3'-UTR of a subset of long-lived mRNAs (3-5) and have been linked to stabilization of human alpha -globin, rat collagen, and rat tyrosine hydroxylase mRNAs (4, 6, 7). alpha CP-1 and alpha CP-2 also function as translational coactivators of poliovirus RNA via a sequence-specific interaction with stem-loop IV of the IRES and promote poliovirus RNA replication by binding to its 5'-terminal cloverleaf structure (8). Finally, these proteins have been implicated in translational control of the 15-lipoxygenase mRNA (9), human Papillomavirus type 16 L2 mRNA (10), and hepatitis A virus RNA (11). Thus, the alpha CP proteins appear to mediate a variety of functions relating to mRNA stability and expression.

The sequences of the human (h) alpha CP-1 and halpha CP-2 mRNAs have been previously established (12, 13), as has that of a murine (m) alpha CP-2 mRNA (heterogeneous nuclear ribonucleoprotein X (14), murine CBP (15)). With the aim of better understanding the potential roles of these related and highly abundant RNA-binding proteins as well as to establish a foundation for delineation of related genetic defects, we extended these analyses by establishing the structures of murine and human alpha CP-1 mRNAs and genes. The data were remarkable in that alpha CP-1 mRNA in both species was an exact colinear copy of the alpha CP-1 gene. This intronless structure of the human alpha CP-1 gene contrasted with the multiexon structure of the alpha CP-2 genes in both species. We have concluded from these and other data that the alpha CP-1 gene was most likely generated by retrotransposition of a fully processed alpha CP-2 mRNA prior to the mammalian radiation and that this processed gene was immediately recruited to serve a critical function distinct from that of its alpha CP-2 progenitor.

    EXPERIMENTAL PROCEDURES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Isolation and Characterization of P1 Genomic Clones-- Genomic clones containing the malpha CP-2 gene (clones 4716 and 4717) and the halpha CP-1 gene (clone 19548) were identified (Genome Systems, Inc., St. Louis, MO) in 129SV mouse ES and human P1 libraries, respectively, by PCR screening with primers derived from the malpha CP-2 cDNA (GenBankTM accession number L19661: 5'-GAC AGG TAC AGC ACA GGC-3' (nucleotides 712-729 of the (+)-strand) and 5'-GTG GTC CCT CCA GGT CAG-3' (nucleotides 773-796 of the (-)-strand)) and the halpha CP-1 cDNA (GenBankTM accession X78137: 5'-GCG ACG CTG TGG GCT AC-3' (nucleotides 635-651 of the (+)-strand) and 5'-GGA ATG GTG AGT TCA TGG GT-3' (nucleotides 871-890 of the (-)-strand)). Genomic clones containing the malpha CP-1 gene (clones 12172 and 12173) were identified in 129SV mouse ES P1 libraries by PCR screening with the same primer set that had been used for isolation of halpha CP-1 genomic clones. Clone inserts were analyzed by Southern hybridization with gene-specific probes amplified from the 3'-UTRs with primers INV2 and MH14 (see Table I and Fig. 2).

Isolation and Cloning of the halpha CP-1 Gene-- Human genomic DNA was isolated from the K562 cell line by the method of Blin and Stafford (16). The coding region of the halpha CP-1 gene was amplified from this DNA by PCR using sets of nested primers: INV6 and INV7 (first PCR) and INV6 and CP17 (second nested PCR). Sequences of the gene-specific primers and their positions on cDNAs are shown in Table I and Fig. 1. Polymerase chain reactions (200 ng of genomic DNA, 25 µM each primer, 1.5 mM MgCl2, 0.2 mM dNTPs, and 2.5 units of Taq polymerase (Perkin-Elmer)) were carried out using a thermocycler (Perkin-Elmer) programmed for 5 min of initial denaturation at 95 °C, followed by 30 cycles for 1 min at 94 °C, 1 min 55 °C, and 3 min at 72 °C. The amplicon of predicted size (1081 bp) was cloned in the pUC19 vector (New England Biolabs, Inc.) and confirmed as the halpha CP-1 gene by sequencing.

RNA Isolation, RT-PCR, and 3'-Rapid Amplification of cDNA Ends-- Total cytoplasmic RNA was isolated from mouse erythroleukemia (MEL) cells, human epithelial cancer (HeLa) cells, and primary mouse tissues (RNeasy Mini kit, QIAGEN Inc.). RNA was treated with 10 units of RNase-free DNase (Ambion Inc.) prior to final isolation. To initiate RT-PCR, 5 µg of total RNA was reverse-transcribed with Superscript II reverse transcriptase (Life Technologies, Inc.) using 2 pmol of a gene-specific primer. The subsequent PCRs were then performed as described above. For selective amplification of alpha CP-1 mRNA, the RT step was carried out with primer CP17, which was specific for the alpha CP-1 mRNA (see Table I and Fig. 1). The subsequent PCR was carried out with primers INV3 and INV4, which were universal for both alpha CP-1 and alpha CP-2. 3'-Rapid amplification of cDNA ends was carried out between a dT primer (5'-CGG AAT TCC T18-3') and nested alpha CP-1-specific primers AM52 (first PCR) and AM50K (second PCR) (see Table I and Fig. 1). The annealing temperature in PCRs with the dT primer was decreased to 42 °C; the number of cycles was increased to 35-40; and a final elongation step (7 min at 72 °C) was added. The final amplification product was cloned in the pUC19 vector and sequenced.

Relative Expression of alpha CP-1 and alpha CP-2 mRNAs by RT-PCR-- 5 µg of total cytoplasmic RNA from different mouse tissues was used for first strand cDNA synthesis with 0.5 µg of dT18 primer using 10 units of avian myeloblastosis reverse transcriptase (Promega). PCR was carried out with the INV2 and MH14 alpha CP-specific primers (see Table I and Fig. 2). Reverse primers were 5'-end-labeled using [gamma -32P]ATP and T4 polynucleotide kinase (New England Biolabs, Inc.). Aliquots were removed every two cycles between cycles 16 and 30 and electrophoresed on a 2.5% MetaPhor-agarose gel (FMC Corp. BioProducts). The gel was dried, and signals were quantified by PhosphorImager analysis (Molecular Dynamics, Inc.) and ImageQuantTM software. Logarithms of 32P activity incorporated into PCR products were plotted against the corresponding PCR cycle, and intercepts on the abscissa corresponding to the ratio of transcripts prior to PCR amplification were calculated (17). Common reaction mixtures were used for all reactions, and exposures were done in parallel.

Southern Blot Analysis-- P1 DNAs were prepared using a high molecular mass plasmid purification kit (MKb-100, Genome Systems, Inc.). Mouse genomic DNA (adult male BALB/c kidney tissue) and human genomic DNA (normal term placenta) were obtained from CLONTECH. 100 ng of P1 DNA and 25 µg of genomic DNA were subjected to overnight restriction cleavage with stated enzymes (New England Biolabs, Inc.). The restriction fragments were separated by 0.8% agarose gel electrophoresis at 20 V for 22 h, transferred to a nylon membrane (Nytran Plus, Schleicher & Schüll) under alkaline conditions, and immobilized by baking for 2 h at 80 °C. DNA templates for the synthesis of alpha CP-1-specific RNA probes were obtained as PCR products of P1 clones 12172 (malpha CP-1) and 19548 (halpha CP-1) with primers T7-INV2 and MH14 (see Table I and Fig. 2). 32P-Labeled RNA probes were synthesized using a MEGAshortscript kit (Ambion Inc.). 50 µCi of [32P]CTP (400 Ci/mmol) was added to 20 µl of transcription reaction mixture. RNA probes (specific activity of 10-20 Ci/mmol) were separated from unincorporated nucleotides on a Sephadex G-50 Quick Spin column (Roche Molecular Biochemicals). Prehybridization and hybridization were performed in 0.5 M NaPO4 buffer (pH 7.2), 1.5 mM EDTA (pH 8.0), and 7% SDS. The probes were hybridized to the membranes at 65 °C overnight.

Antibodies and Western Blot Analyses-- MEL cells were grown to subconfluence, lysed in buffer (20 mM Tris (pH 7.4), 10 mM KCl, 1 mM MgCl2, 1 mM dithiothreitol, 0.1% Triton X-100, 0.1% SDS, and anti-protease mixture (2 µg/ml leupeptin, 1 µg/ml pepstatin, 2 µg/ml aprotinin, and 100 µg/ml phenylmethylsulfonyl fluoride)), sonicated, and then heated at 95 °C for 5 min. After centrifugation, supernatant protein concentrations were estimated using a Bio-Rad Bradford protein assay kit. 50 µg of proteins was separated on each lane of a 12.5% SDS-polyacrylamide gel and electroblotted onto a nitrocellulose membrane (Protrane, Schleicher & Schüll) in buffer containing 25 mM Tris, 192 mM glycine, and 20% methanol. After overnight blocking in 1× phosphate-buffered saline, 5% (w/v) dry milk, and 0.3% Tween, the membrane was incubated with rabbit antisera raised against alpha CP-1 (1:6000 dilution) or alpha CP-2 (1:6000 dilution) (gifts of Dr. G. Gamarnik, University of California, San Francisco, CA) (8). The gels were developed by incubation with horseradish peroxidase-conjugated donkey anti-rabbit IgG (used at 1:7000 dilution; followed by ECL detection of immune complexes Amersham Pharmacia Biotech). Bacterially expressed recombinant alpha CP-1 and alpha CP-2 with His6 tags were prepared and purified as detailed (18).

Chromosome Mapping-- For fluorescence in situ hybridization (FISH), P1 plasmid DNA (clone 12173 for alpha CP-1 and clone 4716 for alpha CP-2) was labeled with digoxigenin-dUTP by nick translation. Labeled probe was combined with sheared mouse DNA and hybridized to metaphase chromosomes derived from mouse embryo fibroblasts in a solution containing 50% formamide, 10% dextran sulfate, and 2× SSC. Specific hybridization signals were detected by incubating the hybridized slides in fluoresceinated anti-digoxigenin antibodies, followed by counterstaining with 4,6-diamidino-2-phenylindole. Chromosome identity was confirmed with probes specific to the telomere (chromosome 6) or centromere (chromosome 15).

Interspecific backcross panels (Jackson Laboratory Mapping Resource) contained 95 progeny ((C57BL/6J × Mus spretus)F1 × C57BL/6J) and 94 progeny ((C57BL/6JEi × SPRET/Ei)F1 × SPRET/Ei) (19). Using primers 5'-TGC TGG AGG TGG GGG TGA T-3' and 5'-TGA ACT ACA ACA CAA CTG CTT T-3', corresponding to one of the malpha CP-2 gene introns, we identified a PCR product length polymorphism (C57BL/6J, approx 220 bp; M. spretus, approx 190 bp) with which to follow segregation of Pcbp2 alleles.

Estimation of Divergence Rates among alpha CP Gene Sequences-- The evolutionary distances between two homologous sequences were computed by "Method 1" of Ina (20), which includes a correction for multiple substitutions at single sites based on the two-parameter model of Kimura. Distances (K) were converted to rates (r) using the equation r = K/(2T), where T is the divergence time between the two species (21).

    RESULTS
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

The malpha CP-1 Gene Lacks Introns-- As part of our ongoing study of the alpha CP proteins, we initiated an analysis of the alpha CP-1 and alpha CP-2 genes. The entire 11-kilobase alpha CP-2 gene has been structurally defined and will be presented in detail elsewhere.2 This gene, which has been partially mapped by others (15), contains at least 12 introns. Two distinct mouse genomic P1 clones containing the alpha CP-1 gene were identified (see "Experimental Procedures"). The full sequences of the malpha CP-1 gene on the two clones were determined and were identical to each other. This malpha CP-1 gene sequence was surprising in that it revealed an uninterrupted 1071-nucleotide open reading frame (ORF) (Fig. 1). The malpha CP-1 ORF was highly similar (94% sequence identity) to that of the previously reported halpha CP-1 mRNA, and the inferred protein structure of malpha CP-1 was identical to halpha CP-1 with the exception of a single valine-for-alanine substitution at position 205 (Fig. 1). Thus, in marked contrast to the multiexon alpha CP-2 gene, the malpha CP-1 gene lacks introns.


View larger version (45K):
[in this window]
[in a new window]
  		Fig. 1.   malpha CP-1 gene ORF and alignment of mouse and human alpha CP-1 and alpha CP-2 cDNA coding regions. The halpha CP-1 (GenBankTM accession number X78137) (13), malpha CP-1 (ORF from P1 clone 12173 (GenBankTM accession number AF139895) and cDNA coding region (GenBankTM accession number AF139894)), halpha CP-2 (GenBankTM accession X78136) (13), and malpha CP-2 (GenBankTM accession number L19661) (14) cDNAs and inferred protein sequences (single-letter code) are shown. Nucleotides and amino acids that are identical among all four genes are indicated by hyphens, and all other sequences are enumerated. The positions of primers used in the study for RT-PCRs are indicated by arrows at appropriate positions on the cDNAs (see also Table I). The positions of the three KH domains are indicated by labeled and overlined.

The "Processed" malpha CP-1 Gene Is Expressed-- To determine whether the processed malpha CP-1 gene that we had identified was a pseudogene or was expressed, we compared its sequence to that of the malpha CP-1 mRNA. A 1.1-kDa fragment encompassing the entire ORF of malpha CP-1 mRNA was generated by RT-PCR from poly(A) mRNA from MEL cells. The sequence of this amplified region was identical to that of the malpha CP-1 gene (Fig. 1). This sequence comparison was extended to include the 3'-untranslated region by 3'-rapid amplification of cDNA ends amplification (see "Experimental Procedures"). The amplified 345-nucleotide 3'-UTR terminated in a poly(A) tail located 14 bases 3' to the second of two putative polyadenylation signals (AAUAAA). The sequence of this entire segment was an exact match to the corresponding region within the malpha CP-1 gene (Fig. 2). A compilation of 55 reported malpha CP-1 cDNA clones in the GenBankTM EST data base failed to reveal any additional or divergent sequences (see the list in Appendix 1).3 These data supported the conclusion that malpha CP-1 mRNA was transcribed from the intronless alpha CP-1 gene.


View larger version (34K):
[in this window]
[in a new window]
  		Fig. 2.   Alignment of alpha CP 3'-UTR sequences. The 3'-UTRs of the following are presented: halpha CP-1 (GenBankTM accession number X78137) (13), malpha CP-1 (GenBankTM accession number AF139894) (this report), halpha CP-2 (GenBankTM accession number X78136) (13) extended using EST clones (GenBankTM accession numbers W60916, AI183557, and N29959), malpha CP-2 (GenBankTM accession number X97982), rat (r) alpha CP-1 and alpha CP-2 (this report; see Appendix 2). Nucleotides that are identical among all mRNAs are indicated by hyphens, and all other sequences that differ from halpha CP-1 (top line) are enumerated. The positions of the TAG termination codons are shown in boldface at the beginning of each sequence; the AATAAA poly(A) addition signal is underlined; and the poly(A) tails are shown. The positions of primers used in RT-PCRs are shown (see also Table I).

Translation of the defined malpha CP-1 mRNA was confirmed by detection of its protein product on Western blot analysis (Fig. 3). Antisera specific for alpha CP-1 revealed a single band in MEL cell extracts that was of slightly higher mobility than the protein identified with the antisera specific for alpha CP-2 (Fig. 3, lanes 1 and 4). These data, which were in full agreement with the predicted molecular masses (37.5 and 38.6 kDa, respectively), demonstrated that processed alpha CP-1 mRNA was translationally active.


View larger version (70K):
[in this window]
[in a new window]
  		Fig. 3.   Western blot detection of mouse alpha CP-1 and alpha CP-2. MEL cell extract (lanes 1 and 4), recombinant alpha CP-1 (lanes 2 and 5), and recombinant alpha CP-2 (lanes 3 and 6) were electrophoresed (SDS-polyacrylamide gel), transferred to a nitrocellulose membrane, and probed with rabbit antisera specific to either alpha CP-1 (lanes 1-3) or alpha CP-2 (lanes 4-6). The specificities of the two antisera were documented by specific interactions with the recombinant proteins as shown. Recombinant alpha CP-1 and alpha CP-2 were expressed from the pET28- and pQE-type vectors, respectively, and contained 34 (alpha CP-1) and 10 (alpha CP-2) additional vector-derived amino acids. The positions of two molecular mass markers are shown on the left.

The Processed alpha CP-1 Gene Originated Prior to the Mammalian Radiation-- The identical sequences of the intronless malpha CP-1 gene and the malpha CP-1 mRNA could be explained by two distinct models. The first model would posit that the processed alpha CP-1 gene was a pseudogene generated so recently that there had not been sufficient time for its sequence to diverge from an originating retrotransposed malpha CP-1 mRNA. This model would suggest that an expressed, intron-containing alpha CP-1 gene had been missed in the cloning survey. The second model would posit that the intronless alpha CP-1 gene was in fact functional and had brought with it or had acquired sufficient gene control elements to support substantial expression. To compare these two models, we estimated how long the processed alpha CP-1 gene had been in existence by searching for it in the human genome. A segment of the halpha CP-1 gene encompassing the entire coding region was amplified from human genomic DNA (see "Experimental Procedures"), and in parallel, a P1 clone containing the alpha CP-1 gene was isolated from a genomic library (see "Experimental Procedures"). The alpha CP-1 gene sequences from both these sources were identical to each other and were perfect matches to halpha CP-1 mRNA (GenBankTM accession number X78137) (12). Thus, the intronless alpha CP-1 gene was present in the human genome as well as the mouse genome. This finding established that the processed alpha CP-1 gene had been generated over 80 million years ago, predating the mammalian radiation (21).

Human and Mouse Cells Contain Similar Sets of alpha CP mRNAs-- Multiple splice variants of the malpha CP-2 transcript have been described. The most extensive alternative splicing of alpha CP-2 exons occurs in the region encoding the segment between the second and third KH domains (Fig. 4, exons B, C, and E) (15).4 In contrast, an intronless alpha CP-1 gene would be expected to yield a single mRNA species. To confirm this prediction, total cytoplasmic mRNA was surveyed for alpha CP-2 and alpha CP-1 mRNAs. An initial RT-PCR analysis with primers universal to all alpha CP mRNAs (Table I and Fig. 1) revealed a complex set of mRNAs in MEL cells. The largest cDNA fragment (360 bp) corresponded to full-length malpha CP-2 cDNA, the next to full-length malpha CP-1 cDNA (330 bp), as well as the alpha CP-2 splice variant lacking the 39 ± 3-bp alternatively spliced segment, and the smallest fragment corresponded to malpha CP-2 cDNA splice variants lacking the 93-bp segment (Fig. 4B, lane 2). Each of these designations was confirmed by sequence analysis (data not shown). In HeLa cells, the same set of amplified cDNAs was observed, as well as several additional smaller fragments of undetermined structure (Fig. 4B, lane 5). The presence of multiple alpha CP-2 mRNA spliced forms was further substantiated by the EST data base, which revealed that 67.5% of 30 malpha CP-2 cDNAs and 70.4% of 71 halpha CP-2 cDNAs were alternatively spliced between the second and third KH domains. We next carried out an amplification with an RT primer specific to alpha CP-1 mRNA. This analysis, which also encompassed the second and third KH domains, detected a single band (Fig. 4B, lanes 3 and 6). The sequence of this band (data not shown) confirmed its assignment as malpha CP-1 mRNA. In agreement with these data, EST data base analysis revealed a single sequence for 55 mouse and 235 human alpha CP-1 cDNA clones. Thus, both mouse and human cells contained multiple spliced forms of alpha CP-2 mRNA and a single species of alpha CP-1 mRNA.


View larger version (46K):
[in this window]
[in a new window]
  		Fig. 4.   Detection of multiple alternatively spliced alpha CP-2 mRNAs, but only a single alpha CP-1 mRNA. A, the region of the alpha CP-2 gene located between the second and third KH domains is shown. The six consecutive exons in this region are arbitrarily labeled A-F (full structure of the alpha CP-2 gene will be detailed elsewhere; see Footnote 2). The alternatively spliced exons or exon segments are shown in solid black. Exon splicing is indicated by the angled lines; exon F has two competing splice acceptor sites. The size of each of the alternatively spliced exons is indicated, as are the positions of the two RT primers (INV7 and CP17) and the two PCR primers (INV3 and INV4) used in the amplification reactions (horizontal arrows). The specificity of each of the RT primers is indicated to the right of the respective arrows. B, shown are the results from the analysis of RT-PCR products. RNAs from mouse (MEL) and human (HeLa) cells were reverse-transcribed using a primer common to alpha CP-1 and alpha CP-2 mRNAs (INV7) (lanes 2 and 5; see Fig. 1) or specific to alpha CP-1 (CP17) (lanes 3 and 6; see Fig. 1) and then amplified between primers INV3 and INV4 (shown in A). The faint lower band in lane 3 was not reproducible. The control reactions lacking RT are shown in lanes 1 and 4, and the molecular size standards (lane M) are shown on the left.

                              
View this table:
[in this window]
[in a new window]
 
Table I
Oligonucleotides used for RT-PCR assays and for cloning
Sequences shown in boldface correspond to restriction sites added to facilitate cloning. The underlined sequence is the core T7 promoter. See Figs. 1 and 2 for locations of each primer on cDNA.

The Tissue Distribution of alpha CP-1 mRNA Parallels That of alpha CP-2-- The relative expression of alpha CP-1 and alpha CP-2 mRNAs was compared in a variety of tissues. RT-PCR was carried out with primers that were universal for human and mouse alpha CP-1 and alpha CP-2 mRNAs and that were positioned so as to amplify a slightly larger fragment for alpha CP-2 than alpha CP-1 (Fig. 5, A and B). The bands representing each of the two mRNAs were quantified, and alpha CP-1/alpha CP-2 mRNA ratios were established for each tissue (Fig. 5C). There were no major differences between the relative levels of alpha CP-1 and alpha CP-2 mRNAs in any of the tissues studied (Fig. 5C). Thus, expression of the processed alpha CP-1 gene paralleled that of the intron-containing alpha CP-2 gene in all tissues examined.


View larger version (26K):
[in this window]
[in a new window]
  		Fig. 5.   Parallel expression of alpha CP-1 and alpha CP-2 mRNAs in different mouse tissues. A, a representative quantitative RT-PCR of alpha CP-1 and alpha CP-2 mRNAs (mouse heart mRNA) is shown. Primer MH14 was used for RT, and cDNA fragments were amplified with primers INV2 and MH14 (Table I and Figs. 1 and 2). The amplified alpha CP-1 and alpha CP-2 cDNA fragments were of the predicted sizes (219 and 234 bp, respectively). Aliquots of the reaction were sampled at the indicated cycle number. B, shown are the amplification kinetics of the RT-PCR amplification shown in A. The relative quantities of alpha CP-1 and alpha CP-2, quantified by PhosphorImager analysis, are shown for each cycle number indicated. The equivalent exponential amplification phases of alpha CP-1 and alpha CP-2 cDNAs validated the accuracy of the measured alpha CP-1/alpha CP-2 mRNA ratio (1:1.6 in the heart). C, shown are the relative levels of alpha CP-1 and alpha CP-2 mRNAs in different mouse tissues. All values are taken from RT-PCRs at exponential amplification phase. rel. units, relative units.

The Genes Encoding alpha CP-1 and alpha CP-2 Are Single Copy, Are Located on Different Chromosomes, and Are in Linkage Groups That Are Conserved between Human and Mouse Genomes-- The formal possibility that a second intron-containing alpha CP-1 gene existed in the genome was tested by Southern blot analysis. The alpha CP-1 and alpha CP-2 loci in the mouse and human genomes were each specifically identified using discriminating probes corresponding to their unique 3'-UTRs (see "Experimental Procedures"). Each probe hybridized to a single band in four different restriction digests of genomic DNA and to an identically sized set of hybridizing bands generated from the corresponding mouse and human alpha CP-1 (P1) clones (Fig. 6 and data not shown). These data demonstrated that the alpha CP-1 and alpha CP-2 genes were each present as unique loci in the mouse and human genomes.


View larger version (65K):
[in this window]
[in a new window]
  		Fig. 6.   Detection of single alpha CP-1 and alpha CP-2 loci in mouse and human genomes by Southern blot analysis. Mouse or human genomic DNA and DNA from P1 clones containing the malpha CP-1 and halpha CP-1 P1 genes (clones 12172 and 19548, respectively) were digested with HindIII (H) or SacI (S), resolved on an agarose gel, and hybridized with a 32P-labeled probe specific to alpha CP-1 (see "Experimental Procedures"). The positions of the DNA molecular size markers are shown on the right.

The chromosome map positions of the malpha CP-1 and malpha CP-2 genes were determined by in situ hybridization and recombination mapping. The mouse alpha CP-2 gene (Pcbp2) was localized by FISH to cytoband 15F1 (Fig. 7C). In full agreement with this assignment, an interspecific backcross analysis (see "Experimental Procedures") demonstrated the Pcbp2 locus as nonrecombinant with the marker D15Mit16, placing Pcbp2 58-61 centimorgans distal to the centromere of mouse chromosome 15. This map position of the malpha CP-2 locus shared an extensive region of synteny with the previously reported map position of the human alpha CP-2 gene (PCBP2) (human chromosome 12q13.12-q13.13; Oxford Grid) (22). The malpha CP-1 gene was localized by FISH analysis to cytoband 6D1 (Fig. 7, A and B). This region was syntenic with human chromosome 2p12-p13 (Oxford Grid), which is the map position of the halpha CP-1 gene (PCBP1) (22). These data demonstrated synteny of corresponding alpha CP-1 and alpha CP-2 loci in mouse and human genomes and confirmed the orthologous nature of their relationships.


View larger version (31K):
[in this window]
[in a new window]
  		Fig. 7.   Chromosomal localization of the murine alpha CP-1 and alpha CP-2 genes. A and D, representative FISH analyses for malpha CP-1 and malpha CP-2 loci, respectively. Red arrows point to gene-specific signals, and white arrows point to chromosome-specific telomeric (A) and centromeric (D) signals (the centromeric signal at chromosome 15 is not well visualized in the reproduction). The chromosomal localization of the mouse alpha CP-2 gene (Pcbp2) was determined by FISH using P1 clone 4716 as a probe. A single predominant signal was obtained on chromosome 15 at a position that corresponds to cytoband 15F1. FISH analysis using the encompassing P1 clone (clone 12173) as a probe localized the malpha CP-1 locus (Pcbp1) to mouse chromosome 6 at a position that corresponds to cytoband 6D1. B and C, idiograms of mouse chromosomes 6 and 15, respectively, indicating the positions of the malpha CP-1 (Pcbp1) and malpha CP-2 (Pcbp2) loci.


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

The present report demonstrates that the highly conserved RNA-binding protein alpha CP-1 is encoded by a processed gene. This conclusion derived from the following three lines of evidence. 1) The sequence of the alpha CP-1 gene was colinear with and identical to its mRNA (Figs. 1 and 2). 2) This processed alpha CP-1 gene was the only alpha CP-1 locus in the mouse and human genomes (Fig. 6). 3) In contrast to the multiple alternatively spliced alpha CP-2 mRNAs, only a single alpha CP-1 mRNA could be identified (Fig. 4 and Appendix 1). This processed alpha CP-1 gene encodes an abundant and functionally active protein (Fig. 3) (8, 9).

Based on the data presented, the most likely origin of the processed alpha CP-1 gene was via retrotransposition of a fully processed alpha CP-2 mRNA. For this gene to be vertically transmitted, such an event would have had to occur in the germ line. The observation that the alpha CP-2 mRNA is widely expressed (Fig. 5) and is present in testis, unfertilized mouse egg, and in two-cell stage mouse embryo (see EST data base data in Appendix 1) supports this model. Characteristic features of retroposons such as a remnant poly(A) tail and/or direct repeats flanking the region of homology between the processed gene and originating gene (23, 24) tend to be present in retrotransposed genes dating less than 40 million years and are uniformly absent in those whose origins preceded avian and mammalian divergence (approx 300 million years ago). The absence of a convincing remnant poly(A) tail and direct repeats associated with the alpha CP-1 gene suggests a remote origin.

We have estimated the date of origin for the alpha CP-1 gene based on the extent of its sequence divergence (Tables II-IV). The accuracy of this approach was increased by supplementing the mouse and human sequence data with the rat alpha CP gene sequences identified from EST data bases (Appendix 2). The expected numbers of substitutions per 100 sites have very similar values for each orthologous pairing (Table III), and this divergence rate was much less than that observed in the paralogous pairings (alpha CP-1 versus alpha CP-2 in mouse, rat, and human) (Table IV). The orthologous relationships of the alpha CP-1 and alpha CP-2 genes in these three species, along with the localization of the mouse and human alpha CP-1 genes within corresponding regions of known synteny (Fig. 7), supported the conclusion that the retrotransposition generating the alpha CP-1 gene predated mammalian radiation.

                              
View this table:
[in this window]
[in a new window]
 
Table II
Coding sequence comparison of mouse and human alpha CP orthologs

                              
View this table:
[in this window]
[in a new window]
 
Table III
3'-UTR sequence comparison of mouse, human, and rat alpha CP orthologs

                              
View this table:
[in this window]
[in a new window]
 
Table IV
3'-UTR sequence comparison of mouse, human, and rat alpha CP paralogs

The molecular clock, based on the divergence rate of the alpha CP-1 3'-UTR sequences subsequent to human/rodent and mouse/rat divergence (80 and 15 million years ago, respectively) (16), predicted that the processed alpha CP-1 gene was generated as much as 400-500 million years ago (Fig. 8). The absolute rate of sequence divergence at the alpha CP-1 and alpha CP-2 loci has been remarkably slow. A statistical analysis of 2820 orthologous rodent and human sequences provided the average value and the distribution range of divergence for both coding sequences and UTRs (25). Very high coding sequence identity (97.6%) placed the alpha CP-2 gene at an extreme position, revealing that it is one of the most highly conserved sequences in the distribution of aligned mouse/human nucleotide identities. The coding sequence identity of the alpha CP-1 gene (94.2%) was also highly significantly above the reported average value (85.2%) (25). In fact, alpha CP-1 and alpha CP-2 were both among the 10 most highly conservative proteins in this extensive comparison group. Unexpectedly, a high level of conservation was also maintained in the 3'-UTRs of human, mouse, and rat alpha CP mRNAs (Table III). Although it is generally assumed that 3'-UTRs of eukaryotic mRNAs are under less selective pressure than coding regions, the rate of sequence divergence within 3'-UTRs of the alpha CP mRNAs was an order of magnitude lower than the average rate of nucleotide substitutions in an extensive survey of orthologous 3'-UTRs (25). This conservation of the 3'-UTR suggests the existence of a unexplained functional constraint on its structure.


View larger version (12K):
[in this window]
[in a new window]
  		Fig. 8.   Evolutionary tree of the alpha CP-1 and alpha CP-2 genes. The time scale in millions of years (Myr) is shown on the left. The positions of the mouse, rat, and human alpha CP-1 and alpha CP-2 genes are indicated. The putative generation of the alpha CP-1 gene by retrotransposition of the alpha CP-2 cDNA is indicated by the curved arrow to distinguish this event from a typical gene duplication event.

Coexpression of functional retroposons and their intron-containing parental genes has been identified and characterized in several studies. Among the first of these sets to be discovered were autosomal processed genes that had originated from constitutively expressed genes located on the X chromosome (phosphoglycerate kinase 2 (Pgk-2) (26, 27), the E1alpha subunit of pyruvate dehydrogenase (Pdha-2) (28, 29), the zinc finger protein (Zfa) (30), and glucose-6-phosphate dehydrogenase (G6pd-2) (31)). Evolutionary fixation of this set of retrotransposons was suggested as a mechanism to escape X inactivation (26, 28). However, this observation could not be generalized as examples of X-linked retrotransposons originating from autosomal precursors were also observed (32). Finally, situations have been reported in which functional processed genes and their originating intron-containing genes demonstrate a similar broad range of expression (the mouse S-adenosylmethionine decarboxylase gene (Amd2) (33) and the human eukaryotic initiation factor 4E2 gene (34)). Thus, it is difficult to support general rules correlating expression patterns of originating genes and retrotransposed gene copies.

Since retroposons appear to integrate randomly (23), the most surprising and improbable event in the generation of a functional processed gene is the acquisition of functional promoter elements. Theoretically, this can result from mutation of sequences adjacent to the insertion site or insertion within or adjacent to a preexisting transcription unit (35). In the later case, the origin of transcription should be 5' to the direct repeat flanking the retrotransposition site (glutamine synthetase) (36). As the fortuitous acquisition of promoters by either of these two mechanisms must be exceedingly rare, other mechanisms may also be involved. One possibility (Pgk-2 gene) (37) is that transcription could be initiated 5' to the normal start site via an aberrant initiation event (38, 39) or from a bona fide minor or alternative promoter. In either case, the resultant transcript could then incorporate proximal (major) promoter elements within the retrotransposed unit (40). Consistent with this model, EST-derived sequence data suggest that the alpha CP-2 gene has a minor promoter 5' to the predominant housekeeping promoter (GenBankTM accession AA317361). This mechanism, which can be further explored, would explain the parallel expression profiles of the retrotransposed alpha CP-1 locus and the originating alpha CP-2 locus (Fig. 5).

The stringent conservation of alpha CP-1 gene structure and coding potential, predating the mammalian radiation, suggests that this processed gene has taken on a nonredundant and essential role. The most prominent structural differences between the alpha CP-1 and alpha CP-2 proteins is that the former contains a 31-amino acid auxiliary domain located between RNA-binding KH domains II and III. This region is excluded from most alpha CP-2 isoforms due to extensive alternative splicing of the alpha CP-2 transcript in this region. Since the alpha CP-1 gene originated as a copy of the fully spliced alpha CP-2 mRNA, the alpha CP-1 retrotransposition served to fix the high level expression of an alpha CP containing this auxiliary domain (Fig. 4A). The fact that alpha CP-1 binds to the 3'-UTR stability motif of alpha 2-globin mRNA in vitro (18) and inhibits translation of erythroid 15-lipoxygenase and Papillomavirus type 16 L2 protein mRNAs in transfected cells establishes its functional capacity. Whether specificity of such actions, or additional actions, of this alpha CP isoform can be related to the region between the two last KH domains remains to be demonstrated.

    ACKNOWLEDGEMENTS

We thank Martin Holcik for contributions to Pcbp2 chromosomal mapping. The manuscript was assembled with the expert secretarial assistance of Jessie Harper.

    FOOTNOTES

* This work was supported in part by National Institutes of Health Grants HL38632-6 and CA72765-01 (to S. A. L.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBankTM/EMBL Data Bank with accession number(s) AF139894.

Dagger Investigator of the Howard Hughes Medical Institute. To whom correspondence should be addressed: Depts. of Genetics and Medicine, Rm. 428 CRB, University of Pennsylvania School of Medicine, 415 Curie Blvd., Philadelphia, PA 19104. Fax: 215-898-1257; E-mail: Liebhaber@mail.med.upenn.edu.

2 A. V. Makeyev, A. N. Chkheidze, and S. A. Liebhaber, manuscript in preparation.

3 For further information, Appendices 1 and 2 may be obtained from the author upon request.

4 A. V. Makeyev, A. N. Chkheidze, and S. A. Liebhaber, unpublished data.

    ABBREVIATIONS

The abbreviations used are: KH, K homology; UTR, untranslated region; halpha CP, human alpha CP; malpha CP, murine alpha CP; RT-PCR, reverse transcription-polymerase chain reaction; bp, base pair(s); FISH, fluorescence in situ hybridization; ORF, open reading frame; EST, expressed sequence tag.

    REFERENCES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

1. Buckanovich, R. J., Posner, J. B., and Darnell, R. B. (1993) Neuron 11, 657-672[CrossRef][Medline] [Order article via Infotrieve]
2. Musco, G., Stier, G., Joseph, C., Castiglione Morelli, M. A., Nilges, M., Gibson, T. J., and Pastore, A. (1996) Cell 85, 237-245[CrossRef][Medline] [Order article via Infotrieve]
3. Wang, X., Kiledjian, M., Weiss, I. M, and Liebhaber, S. A. (1995) Mol. Cell. Biol. 15, 1769-1777[Abstract]
4. Kiledjian, M., Wang, X., and Liebhaber, S. A. (1995) EMBO J. 14, 4357-4364[Medline] [Order article via Infotrieve]
5. Holcik, M., and Liebhaber, S. A. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 2410-2414[Abstract/Free Full Text]
6. Stefanovic, B., Hellerbrand, C., Holcik, M., Briendl, M., Liebhaber, S. A., and Brenner, D. A. (1997) Mol. Cell. Biol. 17, 5201-5209[Abstract]
7. Czyzyk-Krzeska, M. F., and Beresh, J. E. (1996) J. Biol. Chem. 271, 3293-3299[Abstract/Free Full Text]
8. Gamarnik, A. V., and Andino, R. (1997) RNA 3, 882-892[Abstract]
9. Ostareck, D. H., Ostareck-Lederer, A., Wilm, M., Thiele, B. J., Mann, M., and Hentze, M. W. (1997) Cell 89, 597-606[CrossRef][Medline] [Order article via Infotrieve]
10. Collier, B., Goobar-Larsson, L., Sokolowski, M., and Schwartz, S. (1998) J. Biol. Chem. 273, 22648-22656[Abstract/Free Full Text]
11. Graff, J., Cha, J., Blyn, L. B., and Ehrenfeld, E. (1998) J. Virol. 72, 9668-9675[Abstract/Free Full Text]
12. Aasheim, H.-C., Loukianova, T., Deggerdal, A., and Smeland, E. B. (1994) Nucleic Acids Res. 22, 959-964[Abstract/Free Full Text]
13. Leffers, H., Dejgaard, K., and Celis, J. E. (1995) Eur. J. Biochem. 230, 447-453[Medline] [Order article via Infotrieve]
14. Hahm, K., Kim, G., Turck, C. W., and Smale, S. T. (1993) Nucleic Acids Res. 21, 3894[Free Full Text]
15. Funke, B., Zuleger, B., Benavente, R., Schuster, T., Goller, M., Stevenin, J., and Horak, I. (1996) Nucleic Acids Res. 24, 3821-3828[Abstract/Free Full Text]
16. Blin, N., and Stafford, D. W. (1976) Nucleic Acids Res. 3, 2303-2308
17. Smith, C. H. W. J., Gooding, C., Roberts, G. C., and Scadden, A. D. J. (1996) in Laboratory Guide to RNA: Isolation, Analysis, and Synthesis (Krieg, P. A., ed) , pp. 411-440, Wiley-Liss, Inc., New York
18. Chkheidze, A. N., Lyakhov, D. L., Makeyev, A. V., Morales, J., Kong, J., and Liebhaber, S. A. (1999) Mol. Cell. Biol. 19, 4572-4581[Abstract/Free Full Text]
19. Rowe, L. B., Nadeau, J. H., Turner, R., Frankel, W. N., Letts, V. A., Eppig, J. T., Ko, M. S., Thurston, S. J., and Birkenmeier, E. H. (1994) Mamm. Genome 5, 253-274[CrossRef][Medline] [Order article via Infotrieve]
20. Ina, Y. (1995) J. Mol. Evol. 40, 190-226[CrossRef][Medline] [Order article via Infotrieve]
21. Li, W.-H. (1997) Molecular Evolution , pp. 1-497, Sinauer Associates, Inc., Sunderland, MA
22. Tommerup, N., and Leffers, H. (1996) Genomics 32, 297-298[CrossRef][Medline] [Order article via Infotrieve]
23. Vanin, E. F. (1985) Annu. Rev. Genet. 19, 253-272[CrossRef][Medline] [Order article via Infotrieve]
24. Wilde, C. D. (1986) Crit. Rev. Biochem. 19, 323-352[Medline] [Order article via Infotrieve]
25. Makalowski, W., and Boguski, M. S. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 9407-9412[Abstract/Free Full Text]
26. Boer, P. H., Adra, C. N., Lau, Y.-F., and McBurney, M. W. (1987) Mol. Cell. Biol. 7, 3107-3112[Abstract/Free Full Text]
27. Gebara, M. M., and McCarrey, J. R. (1992) Mol. Cell. Biol. 12, 1422-1431[Abstract/Free Full Text]
28. Dahl, H.-H. M., Brown, R. M., Hutchison, W. M., Maragos, C., and Brown, G. K. (1990) Genomics 8, 225-232[CrossRef][Medline] [Order article via Infotrieve]
29. Fitzgerald, J., Hutchison, W. M., and Dahl, H.-H. M. (1992) Biochim. Biophys. Acta 1131, 83-90[Medline] [Order article via Infotrieve]
30. Ashworth, A., Skene, B., Swift, S., and Lovell-Badge, R. (1990) EMBO J. 9, 1529-1534[Medline] [Order article via Infotrieve]
31. Hendriksen, P. J., Hoogerbrugge, J. W., Baarends, W. M., de Boer, P., Vreeburg, J. T., Vos, E. A., van der Lende, T., and Grootegoed, J. A. (1997) Genomics 41, 350-359[CrossRef][Medline] [Order article via Infotrieve]
32. Shashidharan, P., Michaelidis, T. M., Robakis, N. K., Kresovali, A., Papamatheakis, J., and Plaitakis, A. (1994) J. Biol. Chem. 269, 16971-16976[Abstract/Free Full Text]
33. Persson, K., Holm, I., and Heby, O. (1995) J. Biol. Chem. 270, 5642-5648[Abstract/Free Full Text]
34. Gao, M., Rychlik, W., and Rhoads, R. E. (1998) J. Biol. Chem. 273, 4622-4628[Abstract/Free Full Text]
35. Brosius, J. (1991) Science 251, 753[Free Full Text]
36. Bhandari, B., Roesler, W. J., DeLisio, K. D., Klemm, D. J., Ross, N. S., and Miller, R. E. (1991) J. Biol. Chem. 266, 7784-7792[Abstract/Free Full Text]
37. Robinson, M. O., McCarrey, J. R., and Simon, M. I. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 8437-8441[Abstract/Free Full Text]
38. Soares, M. B., Schon, E., Henderson, A., Karathanasis, S. K., Cate, R., Zeitlin, S., Chirgwin, J., and Efstratiadi, S. A. (1985) Mol. Cell. Biol. 5, 2090-2103[Abstract/Free Full Text]
39. Chakrabarti, R., McCracken, J. B., Jr., Chakrabarti, D., and Souba, W. W. (1995) Gene (Amst.) 153, 163-199[CrossRef][Medline] [Order article via Infotrieve]
40. Fourel, G., Transy, C., Tennant, B. C., and Buendia, M. A. (1992) Mol. Cell. Biol. 12, 5336-5344[Abstract/Free Full Text]

Copyright © 1999 by The American Society for