HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 276, Issue 11, 8044-8050, March 16, 2001

This Article Abstract Full Text (PDF) All Versions of this Article: 276/11/8044    most recent M009091200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Dass, B. Articles by MacDonald, C. C. Search for Related Content PubMed PubMed Citation Articles by Dass, B. Articles by MacDonald, C. C.

The Gene for a Variant Form of the Polyadenylation Protein CstF-64 Is on Chromosome 19 and Is Expressed in Pachytene Spermatocytes in Mice^*

Brinda Dass, K. Wyatt McMahon, Nancy A. Jenkins^§, Debra J. Gilbert^§, Neal G. Copeland^§, and Clinton C. MacDonald^¶

From the  Department of Cell Biology and Biochemistry and ^¶ Southwest Cancer Center at University Medical Center, Texas Tech University Health Sciences Center, Lubbock, Texas 79430 and ^§ Mouse Cancer Genetics Program, NCI, Frederick Cancer Research and Development Center, Frederick, Maryland 21702

Received for publication, October 4, 2000, and in revised form, December 4, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Many mRNAs in male germ cells lack the canonical AAUAAA but are normally polyadenylated (Wallace, A. M., Dass, B., Ravnik, S. E., Tonk, V., Jenkins, N. A., Gilbert, D. J., Copeland, N. G., and MacDonald, C. C. (1999) Proc. Natl. Acad Sci. U. S. A. 96, 6763-6768). Previously, we demonstrated the presence of two distinct forms of the M_r 64,000 protein of the cleavage stimulation factor (CstF-64) in mouse male germ cells and in brain, a somatic M_r 64,000 form and a variant M_r 70,000 form. The variant form was specific to meiotic and postmeiotic germ cells. We localized the gene for the somatic CstF-64 to the X chromosome, which would be inactivated during male meiosis. This suggested that the variant CstF-64 was an autosomal homolog activated during that time. We have named the variant form " CstF-64," and we describe here the cloning and characterization of the mouse CstF-64 cDNA, which maps to chromosome 19. The mouse CstF-64 protein fits the criteria of the variant CstF-64, including antibody reactivity, size, germ cell expression, and a common proteolytic digest pattern with CstF-64 from testis. Features of mCstF-64 that might allow it to promote the germ cell pattern of polyadenylation include a Pro  Ser substitution in the RNA-binding domain and significant changes in the region that interacts with CstF-77.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Polyadenylation is the process of eukaryotic mRNA processing in which 3' end cleavage occurs, followed by the addition of as many as 250 adenosine residues (1, 2). Messenger RNA polyadenylation is important for cellular processes such as transcription termination (3-6), splicing (7, 8), mRNA transport (9), translation (10-12), and mRNA stability (13, 14). Polyadenylation requires at least five protein complexes, including the cleavage and polyadenylation specificity factor (CPSF),¹ the cleavage stimulation factor (CstF), two cleavage factors (CFI and CFII), and the poly(A) polymerase (1, 2). Other factors, including the poly(A)-binding protein II (which mediates poly(A) tail length) (15), the U1A small nuclear ribonucleoprotein protein (which interacts with both CPSF and the poly(A) polymerase) (16, 17), and DSEF-1 (which binds G-rich auxiliary elements) (18), also contribute to efficient polyadenylation.

In somatic cells, the sequence AAUAAA is required for accurate and efficient cleavage and polyadenylation, and must reside within 30 nucleotides upstream of the cleavage site in the pre-mRNA (19-23). Ninety percent of all sequenced genes have a canonical AAUAAA sequence (1). Recently, we noted that many mRNAs expressed in male germ cells lack AAUAAA, yet are efficiently polyadenylated (24).² To explain proper polyadenylation of mRNAs lacking AAUAAA signals in germ cells, we hypothesized the presence of a testis-specific variant of a known polyadenylation protein. In examining the expression of the M_r 64,000 protein of CstF (CstF-64 (25)), we discovered a variant form of this protein in germ cells that is a candidate for this function (24). CstF-64 interacts directly with another subunit of the complex, CstF-77 (26), linking it to CstF-50 to form the CstF trimer (25, 27). CstF-64 is the RNA-binding component of CstF that binds to U- or GU-rich elements downstream of the cleavage site during polyadenylation (28, 29). As such, CstF-64 is essential for polyadenylation (25), for cell viability (30), and has been shown to interact with RNA polymerase II, coupling polyadenylation and transcription (31).

In our previous work, we showed that in mice the principal M_r 64,000 form of CstF-64 is found in all somatic cells, and in premeiotic and postmeiotic (but not meiotic) germ cells (24). This form of CstF-64 is recognized by the monoclonal antibody 3A7 (25). A variant M_r 70,000 form of CstF-64 is found in meiotic and postmeiotic cells in the testis and to a smaller extent in brain. We have named this form of CstF-64 " CstF-64." In mice, CstF-64 is recognized by the monoclonal antibody 6A9 (24, 25).

The gene for CstF-64 has recently been renamed CSTF2 in human and Cstf2 in mouse by authorities of The Genome Data base (32, 33), and we will adhere to that nomenclature here. Cstf2 or CSTF2 mapped to the X chromosome in either mouse or human (24), supporting the hypothesis that inactivation of the X chromosome during male meiosis results in inactivation of the X-linked Cstf2 in mouse spermatocytes. This inactivation of the essential Cstf2 led us to hypothesize the requirement of an alternative CstF-64 protein in spermatocytes, which we proposed was encoded by an autosomal gene.

We describe here the cloning and characterization of mCstF-64, a cDNA that encodes CstF-64 from mouse pachytene spermatocytes. The protein encoded by mCstF-64 is similar to other isoforms of CstF-64 from several species and retains protein motifs such as the RNA binding, MEAR(A/G) repeats, and C-terminal domains. mCstF-64 is capable of encoding a protein of about M_r 70,000 that is recognized by the 6A9 monoclonal antibody. Partial protease digestion shows that peptide fragments from the protein encoded by mCstF-64 are identical to those of the variant CstF-64 protein obtained from mouse testis. We also show that the RNA-binding domain of the protein encoded by mCstF-64 is functional in vitro. Finally, Cstf2t, the gene for mCstF-64 maps to an autosome, chromosome 19 in mouse. These data suggest that mCstF-64 encodes the variant form of CstF-64 seen in meiotic and postmeiotic male germ cells.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Antibody Interaction Mapping-- Full-length cDNA encoding human CstF-64 (34) was cloned in frame into the pGEX2T vector (Amersham Pharmacia Biotech). 3' end truncations were made by limited enzymatic digestion of the above construct to obtain versions that terminated at amino acids 531, 425, and 325. A fourth construct was made by introducing a 290-bp BamHI fragment of CstF-64 encoding amino acids 316-412 into pGEX2T. Constructs were introduced into Escherichia coli DH5 cells by transformation, grown to mid-logarithmic phase, and induced for 3 h at 37 °C with 1 mM isopropyl -D-thiogalactoside. Bacterial extracts were prepared in SDS-PAGE loading buffer by sonication and boiling and prepared for immunoblotting with either the 3A7 or 6A9 monoclonal antibody as described previously (24).

Complementary DNA Isolation-- An adult mouse pachytene spermatocyte cDNA library in Uni-Zap XR vector (35) was screened using the 290-bp BamHI cDNA fragment of CstF-64 encoding amino acids 316-412. Filters were denatured in 0.5 M NaOH, 1.5 M NaCl, neutralized in 0.5 M Tris-HCl, pH 7.4, 1.5 M NaCl, and hybridized with [-³²P]dCTP-labeled cDNA probe in hybridization solution (10× SSC, 0.05 M sodium phosphate, pH 6.5, 5× Denhardt's solution, 0.1% Na₄P₂O₇, 0.5% SDS, 0.05 mg/ml salmon sperm DNA) overnight at 65 °C. Filters were washed at a final stringency of 2× SSC, 0.1% SDS at 65 °C, and exposed to film at 80 °C with an intensifying screen. Positive plaques were purified by two additional rounds of screening. Plasmid rescue into pBluescript SK-was according to the manufacturer's directions (Stratagene, La Jolla, CA). Of two million plaques screened initially, 24 hybridized to the probe, all of which represented the same mRNA transcript (see "Results"). The longest clone (3612 bp) was designated mCstF-64 and was sequenced by a combination of primer walking and subcloning, using the Sequenase 2.0 kit (U. S. Biochemical Corp.). Sequences were aligned and grouped into contigs using the SeqMan analysis program (DNAStar).

In Vitro Transcription and Translation-- Polypeptides corresponding to human (hCstF-64 (34)), mouse (mCstF-64 (36)), and the mouse CstF-64 (mCstF-64) cDNAs were prepared in vitro using the T3 TNT Coupled Reticulocyte Lysate System (Promega, Madison, WI). Products from the transcription/translation reactions were separated on a 10% SDS-PAGE and immunoblotted with either the 3A7 or 6A9 monoclonal antibody (24).

Peptide Mapping by Limited Proteolysis-- Radiolabeled proteins corresponding to mCstF-64 and mCstF-64 were made in vitro using the T3 TNT reticulocyte lysate system (Promega) with [³⁵S]methionine (PerkinElmer Life Sciences). Translated products were separated on 10% SDS-PAGE, and the band corresponding to full-length mCstF-64 protein was excised and eluted overnight at 4 °C in Cleveland buffer (0.125 M Tris-HCl, pH 6.8, 0.5% SDS, 1% glycerol, 0.0001% bromphenol blue (34, 37)).

Radiolabeled CstF-64 was made in vivo by incorporation of [³⁵S]methionine during short term culture of mouse seminiferous tubules. Testes from 8 CD-1 mice (Charles River Breeding Laboratories) were decapsulated and washed several times in cold PBS to remove interstitial cells. Tubules were then washed in prewarmed DMEM lacking methionine (Cellgro, Mediatech, Inc) followed by incubation in DMEM containing 10 mM methionine, 2 mM glutamine, and 1.25 mCi/ml Tran³⁵S-label (ICN) for 7 h at 32 °C (38). Following incubation, tubules were washed in DMEM, resuspended in RIPA (150 mM NaCl, 1% Nonidet P-40, 0.5% deoxycholate, 0.1% SDS, 50 mM Tris, pH 8.0), and sonicated on ice. After preclearing, samples were immunoprecipitated at 4 °C using the 6A9 monoclonal antibody and protein A-Sepharose beads as described (29). Following immunoprecipitation, beads were washed in RIPA, boiled in Laemmli buffer (39), and digested with 0.1 or 1.0 µg of V8 protease (Sigma) in the presence of 5 µg of bovine IgG protein (Bio-Rad) for 30 min at 37 °C (34, 37). Polypeptide fragments were separated by 15% SDS-PAGE, followed by fluorography. The image was captured on x-ray film exposed in the presence of an intensifying screen at 80 °C for the times indicated in Fig. 3.

Interspecific Mouse Backcross Mapping-- Interspecific backcross progeny were generated by mating (C57BL/6J × Mus spretus) F1 females and C57BL/6J males as described (24, 40). A total of 205 N2 mice were used to map the Cstf2t locus. DNA isolation, restriction enzyme digestion, agarose gel electrophoresis, and Southern blot analysis were performed as described (41). The probe, a 514-bp DraI/XhoI fragment of mCstF-64 from the 3'-UTR, was labeled with [-³²P]dCTP using a nick translation primed labeling kit (Roche Molecular Biochemicals); washing was done to a final stringency of 0.8× SSCP (120 mM NaCl, 5 mM sodium citrate, 20 mM sodium phosphate, pH 6.8), 0.1% SDS at 65 °C. A fragment of 0.5 kb was detected in TaqI-digested C57BL/6J DNA, and a fragment of 1.8 kb was detected in TaqI-digested M. spretus DNA. The presence or absence of the 1.8-kb TaqI M. spretus-specific fragment was followed in backcross mice. A description of the probes and RFLPs for the loci linked to Cstf2t including Gnaq and Fas has been reported previously (42, 43). Recombination distances were calculated using Map Manager, version 2.6.5 (Roswell Park Cancer Institute, Buffalo, NY). Gene order was determined by minimizing the number of recombination events required to explain the allele distribution patterns.

Recombinant Protein Purification-- RNA-binding domains (RBDs) from human (hRBD) and mouse CstF-64 (mRBD) and mouse CstF-64 (mRBD) were prepared as fusion polypeptides with N-terminal hexahistidine tags. Construction of the plasmid hRBD (formerly known as rHis64247) was described previously (34). The RNA-binding domains of mCstF-64 (nucleotides 53-897) and mCstF-64 (nucleotides 77-598) were cloned in frame with the hexahistidine tag of the pQE9 vector (Qiagen, Valencia, CA) to make mRBD and mRBD, respectively. hRBD, mRBD, and mRBD plasmid DNAs were transformed into UltraMAXDH5-FT cells (Life Technologies, Inc.), grown to mid-logarithmic phase, and induced at 37 °C for 3 h by the addition of isopropyl -D-thiogalactoside to 1.5 mM. His-tagged recombinant proteins were isolated as described (34) and dialyzed against buffer D (20 mM HEPES, pH 7.9, 0.1 M KCl, 0.2 mM EDTA, 1.5 mM MgCl₂, 10% glycerol, 0.5 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride (44)) overnight at 4 °C. Recombinant proteins were quantified by comparison to bovine serum albumin standards after staining of 12.5% SDS-PAGE gels with Coomassie Brilliant Blue R-250.

UV Cross-linking-- 0.3 µg each of hRBD, mRBD, mRBD, or bovine IgG (Bio-Rad) was mixed with ³²P-labeled SVL substrate (3 × 10⁴ cpm (45)) in buffer D for 30 min at 30 °C. Reaction mixtures were exposed to 10⁷ µJ/cm² of ultraviolet light in a CL-1000 Ultraviolet Cross-linker (Ultraviolet Products, Upland, CA). Control reactions were processed without exposure to UV. Reaction mixtures were incubated with 10 units of RNaseONE (Promega, Madison, WI) at 37 °C for 15 min. SDS-PAGE loading buffer was added, and the samples were boiled and RNA-cross-linked polypeptides separated on 12.5% SDS-PAGE. The gel was stained with Coomassie Blue to ensure equal loading of the recombinant proteins, destained, dried, and subjected to autoradiography at 80 °C with an intensifying screen.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Epitopes for the Monoclonal Antibodies 3A7 and 6A9 Map to Two Distinct Regions of CstF-64 cDNAThe monoclonal antibodies 3A7 and 6A9, obtained by using human CstF purified from HeLa cells, can distinguish the somatic (3A7) and variant (6A9) forms of CstF-64 in mice (24). In human, however, both antibodies recognize the somatic form of CstF-64 ((25) see Fig. 2). Therefore, to map the recognition sites of each antibody, polypeptides corresponding to different regions of human CstF-64 were expressed as fusions to glutathione S-transferase in bacteria and immunoblotted with either the 3A7 or 6A9 antibody (data not shown, summarized in Figs. 1 and 6).

View larger version (43K): [in this window] [in a new window]   Fig. 1.   Comparison of mCstF-64 protein sequence to known CstF-64 homologs. A, protein sequence alignment of human CstF-64 (top line) and mCstF-64 (bottom line). Alignment was by the ClustalV method (61). Numbering is according to the mCstF-64 sequence. Amino acids that differ from the human CstF-64 are boxed. Significant features (RBD, 17-92; MEARA repeats, 425-464; C-terminal domain, 589-630) are boxed in gray. Inserted segments in mCstF-64 relative to hCstF-64 are indicated by a black overline. The region thought to interact with CstF-77 (26) is indicated by a gray overline. Boxed segments of the human sequence represent the regions of interaction for the 6A9 (332) and 3A7 (441) antibodies. B, sequence identities (in percent) of known CstF-64 homologs. Shown are sequences from human (hCstF-64 (34)), mouse (mCstF-64 (36)), mouse CstF-64 (mCstF-64), human brain (KIAA0689 (48)), Xenopus laevis (xCstF-64 (51)), Caenorhabditis elegans (ceCstF-64 (52)), Drosophila melanogaster (dCstF-64 (50)), and Saccharomyces cerevisiae (yRNA15 (27)). Pairwise identities were determined from the alignments as above. C, tree diagram showing the relative similarities of the human, mouse, Xenopus, Caenorhabditis, Drosophila, and yeast homologs of CstF-64.

By using this approach, we determined that the region where the 3A7 antibody interacted with CstF-64 lay between amino acids 426 and 531 (441-583 in Fig. 1). Similarly, the region of interaction for the 6A9 antibody lie on a BamHI fragment between amino acids 316 and 412 (333-427 in Fig. 1). Since CstF-64 was recognized by the 6A9 antibody but not the 3A7 antibody (24), we chose to screen a mouse pachytene spermatocyte cDNA library with the BamHI fragment of the human CstF-64 that includes the 6A9 epitope to isolate its cDNA.

Isolation of a cDNA for CstF-64 from Mouse Pachytene Spermatocytes-- An adult mouse pachytene spermatocyte library was the kind gift of John McCarrey (Southwest Foundation for Biomedical Research, San Antonio, TX (35)). Our previous results suggested that cells during this period of spermatogenesis expressed CstF-64 exclusively of the somatic CstF-64 (24). Two million plaques were screened using the 290-bp BamHI cDNA fragment of human CstF-64 that was described above (see Fig. 1). Twenty four positive clones were identified and rescued into pBluescript SK. Each of these plasmids represented nearly identical transcripts that differed in length at the 5' ends. None of the cDNAs represented the mouse somatic CstF-64 (data not shown, see Ref. 36). The plasmid containing the longest insert was designated mCstF-64 (GenBank^TM accession number AF322194) and was chosen for further characterization.

Sequence analysis of mCstF-64 revealed a cDNA of 3596 bp size and 16 bp of 3' poly(A). A single open reading frame of 1890 bp was deduced that encoded a 630-amino acid protein with an estimated molecular mass of 65,893.8 Da and an isoelectric point of 7.10 (Fig. 1). The putative translation initiation codon ATG at nucleotide 77 is in good translational consensus (46, 47). The mCstF-64 cDNA had 76 bp of 5'-UTR and 1640 bp of 3'-UTR. There is a canonical polyadenylation sequence AATAAA at nucleotides 3576-3581.

The protein encoded by the mCstF-64 cDNA is similar to other known forms of CstF-64 and is 69.8 and 71.6% identical, respectively, to the mouse and human somatic forms of CstF-64 (Fig. 1, B and C). Interestingly, the protein encoded by mCstF-64 is more highly related (85.6% identical) to the protein encoded by KIAA0689, a cDNA uncovered in a survey of long open reading frames expressed in human brain (48). mCstF-64 also has two peptide inserts (amino acids 213-231 and 498-555) relative to human, mouse, and Xenopus; these inserts are shared by KIAA0689. These data suggest the possibility that KIAA0689 is the human ortholog of mCstF-64 (see "Discussion"). The inserts also probably account for the larger apparent molecular size of the variant CstF-64 protein on SDS-PAGE (see Ref. 24 and see Fig. 2). The downstream insert (498) contains 12 imperfect repeats of the 5-amino acid motif MQG(A/G)G; two such repeats are seen in human CstF-64. Although it is unlikely that these repeats form a stabilized structure such as the MEAR(A/G) repeats (Ref. 49 and see below), it is possible they perform a similar but undefined function.

View larger version (60K): [in this window] [in a new window]   Fig. 2.   Antibody recognition of the protein encoded by mCstF-64. Coupled transcription-translation extracts were programmed with cDNA clones of hCstF-64 (lanes 1), mCstF-64 (lanes 2), mCstF-64 (lanes 3), or pBluescript SK-(control, lanes 4). Proteins were separated by 10% SDS-PAGE and immunoblots incubated with either the 3A7 (A) or 6A9 (B) monoclonal antibodies. Approximate sizes of CstF-64 protein products are indicated at the left.

Fig. 1A shows an alignment of the mCstF-64 protein with human CstF-64 (34). The mCstF-64 protein shares features of known CstF-64 proteins as follows: an N-terminal RBD of the RNA recognition motif type (amino acids 17-92), proline- and glycine-rich regions (amino acids 198-425 and 464-579), a highly conserved C terminus (amino acids 589-630 (50)), and eight imperfect repeats of the amino acids MEAR(A/G) (amino acids 425-464 (49)) that are repeated 12 times in human (34), mouse (36), and chicken CstF-64 ((30) Fig. 1A). The RBD of mCstF-64 is identical to that of CstF-64 from human, mouse, and Xenopus (51) except for a serine replacing the proline at position 41 of the protein.

Antibody Reactivity of the mCstF-64 Protein-- cDNA clones for CstF-64, mCstF-64, and mCstF-64 were transcribed and translated in vitro using the rabbit reticulocyte lysate system, and proteins were tested for reactivity with the 3A7 and 6A9 monoclonal antibodies (Fig. 2). As expected, CstF-64 from human reacted with both the 3A7 and 6A9 antibodies (Fig. 2, lanes 1). CstF-64 from mouse reacted with 3A7 but not 6A9 (lanes 2, compare A and B). This is in agreement with our earlier assessment that the somatic form of CstF-64 from mouse is recognized by the 3A7 but not the 6A9 monoclonal antibody (24). In contrast, the protein encoded by mCstF-64 reacted with 6A9 but not 3A7 (lanes 3, compare A and B). Furthermore, the protein recognized by 6A9 had a larger apparent molecular weight than either the mouse or human somatic CstF-64 proteins. This suggests that mCstF-64 has the same antibody reactivity as the variant form of CstF-64 found in mouse testis (24). The slowest migrating band of about 70 kDa in the human CstF-64 sample (Fig. 2A, lane 1) may be due to posttranslational modification, possibly phosphorylation (34). The small amount of immunoreactivity in control samples that were incubated with vector DNA is probably due to endogenous CstF-64 protein in the rabbit lysates (lanes 4).

Peptide Maps of the mCstF-64 Protein Are Identical to the Pattern of the Variant CstF-64 from Testis-- To determine whether the protein encoded by mCstF-64 is the same as that present in mouse testis, we compared partial protease digestion patterns of the two proteins. Radiolabeled mCstF-64 protein was produced in vitro in rabbit reticulocyte lysate translation extracts in the presence of [³⁵S]methionine. Radiolabeled CstF-64 was isolated in vivo by short term culture of mouse seminiferous tubules in medium containing [³⁵S]methionine followed by immunoprecipitation of the variant CstF-64 with the 6A9 antibody. Both proteins were treated identically with either 0.1 or 1.0 µg of Staphylococcus aureus V8 protease and analyzed by 15% SDS-PAGE and fluorography (Fig. 3).

View larger version (86K): [in this window] [in a new window]   Fig. 3.   Partial V8 protease mapping of the protein encoded by mCstF-64 (in vitro) and the variant CstF-64 protein (in vivo). Either protein made in the reticulocyte lysate system (lanes 1-3, in vitro) or immunoprecipitated from adult mouse seminiferous tubule extracts (lanes 4-6, in vivo) were treated with the indicated amounts of V8 protease, subjected to 15% SDS-PAGE, and visualized by fluorography (see "Experimental Procedures"). Dots indicate bands common to the in vivo and in vitro samples. Lanes 1-3 were exposed to film for 10 days and lanes 4-6 were exposed for 105 days.

By comparing the partial protease digestion profiles of the protein synthesized in vitro (Fig. 3, lanes 1-3) with the protein synthesized in vivo (lanes 4-6), the two profiles appear to share many common peptides. This suggests strongly that the two proteins share the same primary structure. In contrast, the protein encoded by the cDNA for the mouse somatic form of CstF-64, mCstF-64, has a distinctly different partial protease digestion profile, indicating that it has a different primary structure (data not shown).

Cstf2t Is on Chromosome 19 in Mouse-- The mouse chromosomal location for the mCstF-64 gene (Cstf2t) was determined by interspecific backcross analysis using progeny derived from matings of ((C57BL/6J × M. spretus)F₁ × C57BL/6J) mice. This interspecific backcross mapping panel has been typed for over 2900 loci that are well distributed among all the autosomes as well as the X chromosome (40). C57BL/6J and M. spretus DNAs were digested with several enzymes and analyzed by Southern blot hybridization for informative restriction fragment length polymorphisms (RFLPs) using a mouse cDNA probe from mCstF-64. The 1.8-kb TaqI M. spretus RFLP (see "Experimental Procedures") was used to follow the segregation of Cstf2t in backcross mice. The mapping results indicated that Cstf2t is located in the central region of mouse chromosome 19 linked to Gnaq and Fas. Although 120 mice were analyzed for every marker and are shown in the segregation analysis (Fig. 4), up to 167 mice were typed for some pairs of markers. Each locus was analyzed in pairwise combinations for recombination frequencies using the additional data. The ratios of the total number of mice exhibiting recombinant chromosomes to the total number of mice analyzed for each pair of loci and the most likely gene order are as follows: centromere---Gnaq---24/167---Cstf2t---1/122---Fas. The recombination frequencies (expressed as genetic distances in centimorgans ± S.E.) are as follows: centromere---Gnaq---14.4 ± 2.7---Cstf2t---0.8 ± 0.8---Fas.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   The mCstF-64 gene, Cstf2t, maps to mouse chromosome 19. Left, columns represent the chromosome identified in backcross progeny that inherited from the (C57BL/6J × M. spretus) F1 parent. Shaded boxes represent a C57BL/6J and white boxes a M. spretus allele. Number of offspring inheriting each type of chromosome is listed at the bottom. Right, partial linkage map of chromosome 19 showing Cstf2t in relation to linked genes. Recombination distances between loci are shown to the left, and positions of loci in human chromosomes are shown in parentheses. Human loci cited can be obtained from Genome Data Base, a computerized data base maintained by The William H. Welch Medical Library of The Johns Hopkins University (Baltimore, MD).

RNA Binding of the mCstF-64 Protein RBD-- The RBDs of all vertebrate CstF-64s are identical except for mCstF-64, which contains a single amino acid difference (amino acid 41). Therefore, we wanted to determine whether the mCstF-64 RBD was functional in binding RNA in a UV cross-linking assay. The RNA-binding domains of human CstF-64 (amino acids 1-247), mouse CstF-64 (amino acids 1-247), and mCstF-64 (amino acids 1-174) were incubated with ³²P-labeled RNA in vitro and subjected to cross-linking with UV light (29, 34, 45) (Fig. 5). Under the conditions used, each RBD-containing polypeptide bound covalently to RNA only in the presence of UV light (lanes 2, 4, and 6). Minor differences in the intensity of the RNA cross-linked protein bands were not reproducible (data not shown). In contrast, a non-RNA binding protein (IgG) did not bind to RNA irrespective of UV radiation (lanes 7 and 8). This suggests that all three RNA-binding domains are functional to bind a complex RNA substrate, although it does not address the question of RNA-binding specificity.

View larger version (32K): [in this window] [in a new window]   Fig. 5.   RNA binding of RBDs from different CstF-64 proteins. 32P-Labeled RNA was mixed with synthetic RBD domains from human (hRBD, lanes 1 and 2), mouse (mRBD, lanes 3 and 4), mouse variant (mRBD, lanes 5 and 6), or with bovine IgG (IgG, lanes 7 and 8). Mixtures were allowed to incubate 30 min at 30 °C, exposed to UV light (lanes 2, 4, 6, and 8) to cross-link RNA to protein, treated with RNase, then subjected to 10% SDS-PAGE, and autoradiography. Control samples were not exposed to UV light (lanes 1, 3, 5, and 7). Approximate mobilities of RBDs are indicated at left.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We recently described two forms of CstF-64, which are expressed in distinctive patterns in mouse male germ cells (24). One form was clearly the somatic CstF-64 that is found in nearly every cell and tissue. The other had a more limited range and is designated CstF-64. We hypothesized that, since the gene for the somatic form of CstF-64 (Cstf2) was located on the X chromosome and most X-linked genes are inactivated in male meiosis, CstF-64 was expressed from an autosomal paralog of CstF-64. In this paper, we describe the cloning and characterization of a cDNA for CstF-64 from adult mouse pachytene spermatocytes that has all the properties of the gene we hypothesized (Fig. 6). Specifically, the cDNA, designated mCstF-64, (i) is found in a pachytene spermatocyte cDNA library, (ii) encodes a protein with an apparent mobility on SDS-PAGE of about M_r 70,000, (iii) protein derived from this cDNA is recognized by the 6A9 but not the 3A7 monoclonal antibody, (iv) is encoded by the gene Cstf2t that is on an autosome, chromosome 19, (v) encodes a protein that has a partial peptide map that is identical to that of CstF-64 from testis, and (vi) is most similar to KIAA0689 (48), a cDNA from brain where CstF-64 is also found. We further showed that the RBD of mCstF-64 functions in RNA binding despite having a Pro  Ser substitution at amino acid 41 of the RBD.

View larger version (12K): [in this window] [in a new window]   Fig. 6.   Domains of interest in human, mouse, and mouse variant CstF-64 proteins. Diagrams are of hCstF-64 (577 amino acids), mCstF-64 (580 amino acids), and mCstF-64 (630 amino acids) with features as indicated. Shown are the RNA-binding domains (light gray), the region of interaction with CstF-77 (dark gray), MEARA repeats (hatched), and the conserved C-terminal domain (dark gray). The region of mCstF-64 corresponding to the CstF-77 interaction domain is hatched to indicate its divergence from hCstF-64. Inserted domains relative to hCstF-64 are indicated as black.

The mCstF-64 protein is clearly the product of a different gene than is the mouse CstF-64 protein. The cDNAs are only 69.8% identical, leading to a number of amino acid substitutions throughout the protein, rather than inclusion or exclusion of individual exons. Furthermore, mouse backcross analysis determined that Cstf2t is on chromosome 19 (Fig. 5) and not the X chromosome as is Cstf2 (24). A human cDNA clone, KIAA0689 (48), is quite similar to mCstF-64 and is probably the human ortholog of this gene.³ These similarities between mCstF-64 and the human and mouse somatic CstF-64s suggest that mCstF-64 is the result of a duplication or retroviral insertion of the CstF-64 gene that occurred prior to the divergence of primates and mice (Fig. 1C).

As has been noted elsewhere (50), the C termini (amino acids 589-630) of all known CstF-64 homologs are remarkably conserved (27, 34, 36, 48, 50-52), suggesting an essential function for that region. Also conserved is the RNA recognition motif type RNA-binding domain at the N terminus, which is identical in all vertebrate CstF-64s examined (human (34), mouse (36), Xenopus (51), and chicken (53)), except mCstF-64, which has a single proline  serine change at amino acid 41. RBDs of the RNA recognition motif type have a well defined motif (54-56), and serine 41 is part of the second loop following the first -helix in the RBD that might alter RNA substrate specificity (57). Therefore, mCstF-64 might have a different RNA binding specificity than CstF-64, which would contribute to the differences seen in germ cell polyadenylation.

Many of the amino acid substitutions in mCstF-64 relative to CstF-64 are in regions that have as yet unknown functions, including the Gly/Pro-rich regions (Fig. 1A, amino acids 198-425 and 464-579 (34)). Interestingly, a number of substitutions and a 19-amino acid insertion occur in the region thought to interact with CstF-77 (Fig. 1A, amino acids 180-260, summarized in Fig. 6), which bridges CstF and CPSF, and symplekin, a protein whose function in the nucleus is not known (26). This suggests the possibility that mCstF-64 interacts differently than CstF-64 with other proteins of the polyadenylation complex, and the interaction might affect mCstF-64 function in germ cell polyadenylation.

The MEAR(A/G) repeat region of mCstF-64 is significantly different than the 12 repeats in CstF-64 (34, 49). The region in mCstF-64 contains only eight recognizable repeats (425), one of which is incomplete (441), some of which have proline substitutions (430, 438), and none of which precisely match the consensus. In CstF-64, the MEAR(A/G) repeats likely form a stable, monomeric -helix that might serve as a rigid structural element in polyadenylation (49). Perhaps the degenerate MEAR(A/G) region in mCstF-64 forms a shorter structural variant or is dispensable, as it is in Xenopus which lacks MEAR(A/G) (51). However, the second insert in mCstF-64 (498) includes 12 repeats of the amino acids MQG(A/G)G that might substitute for the MEAR(A/G) function.

How might mCstF-64 be involved in promoting polyadenylation of non-AAUAAA-containing mRNAs in male germ cells? One possibility is that the Pro  Ser substitution at amino acid 41 alters the RNA binding affinity of the variant CstF-64 (58, 59), allowing binding to a different downstream sequence element (29). An altered affinity of CstF for a downstream sequence element could then influence the binding of CPSF to an upstream element, which might or might not match the AAUAAA consensus. Our RNA binding experiments (Fig. 5) suggest that mammalian CstF-64 RBDs have similar affinities for nonspecific RNAs. However, the RNA binding specificity of CstF-64 is quite different in isolation than in complex with CstF-77, CstF-50, and CPSF (29, 60). Therefore, other regions of mCstF-64 and other interacting proteins might have a strong influence on RNA substrate specificity. In light of this, mCstF-64 contains a number of amino acid differences in the site of protein-protein interaction with CstF-77 and symplekin. Changes in this region (Fig. 1A, amino acids 108-229) could disrupt binding of CstF-64 to CstF-77 or even to symplekin, thus dramatically altering CstF interaction with the pre-mRNA and with CPSF.

Finally, we note the similarity of mCstF-64 to the human KIAA0689 cDNA from brain (48), suggesting the possibility that KIAA0689 is the human ortholog of mCstF-64. Our recent cloning of the variant CstF-64 from a human testis cDNA library confirms this possibility.³ This strongly suggests that the phenomenon of non-AAUAAA polyadenylation occurs in human germ cells, as well as in mouse. This possibility is currently under investigation.

	ACKNOWLEDGEMENTS

We thank John McCarrey for cDNA libraries; A. Michelle Wallace-Shannon, Eman Attaya, and Andreé Reuss for technical help; Susan San Francisco for sequencing; and S. Sridhara and Stuart Ravnik for critical reading of the manuscript.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grant 1 R01 HD37109-01A1, the South Plains Foundation (to C. C. M.), the NCI, Department of Health and Human Services, National Institutes of Health (to N. A. J. and N. G. C.), and a Helen Hodges Educational Charitable Trust and Raymond Green Scholarships (to B. D.).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.

To whom correspondence should be addressed: Dept. of Cell Biology and Biochemistry, Texas Tech University Health Sciences Center, 3601 4th St., Lubbock, TX 79430. Tel.: 806-743-2703; Fax: 806-743-2990; E-mail: cbbccm2@ttuhsc.edu.

Published, JBC Papers in Press, December 11, 2000, DOI 10.1074/jbc.M009091200

² J.-L. Redondo and C. C. MacDonald, unpublished work.

³ B. Dass and C. C. MacDonald, manuscript in preparation.

	ABBREVIATIONS

The abbreviations used are: CPSF, cleavage and polyadenylation specificity factor; CstF, cleavage stimulation factor; RBD, RNA-binding domain; RRM, RNA recognition motif; UTR, untranslated region; bp, base pair; kb, kilobase pair; PAGE, polyacrylamide gel electrophoresis; DMEM, Dulbecco's modified Eagle's medium; RFLPs, restriction fragment length polymorphisms.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

This article has been cited by other articles:

				B. Dass, S. Tardif, J. Y. Park, B. Tian, H. M. Weitlauf, R. A. Hess, K. Carnes, M. D. Griswold, C. L. Small, and C. C. MacDonald Loss of polyadenylation protein {tau}CstF-64 causes spermatogenic defects and male infertility PNAS, December 18, 2007; 104(51): 20374 - 20379. [Abstract] [Full Text] [PDF]

				D. Liu, J. M. Brockman, B. Dass, L. N. Hutchins, P. Singh, J. R. McCarrey, C. C. MacDonald, and J. H. Graber Systematic variation in mRNA 3'-processing signals during mouse spermatogenesis Nucleic Acids Res., January 12, 2007; 35(1): 234 - 246. [Abstract] [Full Text] [PDF]

				A. V. Evsikov, J. H. Graber, J. M. Brockman, A. Hampl, A. E. Holbrook, P. Singh, J. J. Eppig, D. Solter, and B. B. Knowles Cracking the egg: molecular dynamics and evolutionary aspects of the transition from the fully grown oocyte to embryo Genes & Dev., October 1, 2006; 20(19): 2713 - 2727. [Abstract] [Full Text] [PDF]

				Q. P. Vong, Y. Li, Y.-F. C. Lau, M. Dym, O. M. Rennert, and W.-Y. Chan Structural Characterization and Expression Studies of Dby and Its Homologs in the Mouse J Androl, September 1, 2006; 27(5): 653 - 661. [Abstract] [Full Text] [PDF]

				K. J. Delaney, R. Xu, J. Zhang, Q. Q. Li, K.-Y. Yun, D. L. Falcone, and A. G. Hunt Calmodulin Interacts with and Regulates the RNA-Binding Activity of an Arabidopsis Polyadenylation Factor Subunit Plant Physiology, April 1, 2006; 140(4): 1507 - 1521. [Abstract] [Full Text] [PDF]

				Z. HUBER, R. R. MONAREZ, B. DASS, and C. C. MacDONALD The mRNA Encoding {tau}CstF-64 Is Expressed Ubiquitously in Mouse Tissues Ann. N.Y. Acad. Sci., December 1, 2005; 1061(1): 163 - 172. [Abstract] [Full Text] [PDF]

				A. M. Wallace, T. L. Denison, E. N. Attaya, and C. C. MacDonald Developmental Distribution of the Polyadenylation Protein CstF-64 and the Variant {tau}CstF-64 in Mouse and Rat Testis Biol Reprod, April 1, 2004; 70(4): 1080 - 1087. [Abstract] [Full Text] [PDF]

				J. J. Emerson, H. Kaessmann, E. Betran, and M. Long Extensive Gene Traffic on the Mammalian X Chromosome Science, January 23, 2004; 303(5657): 537 - 540. [Abstract] [Full Text] [PDF]

				M. I. Zarudnaya, I. M. Kolomiets, A. L. Potyahaylo, and D. M. Hovorun Downstream elements of mammalian pre-mRNA polyadenylation signals: primary, secondary and higher-order structures Nucleic Acids Res., March 1, 2003; 31(5): 1375 - 1386. [Abstract] [Full Text] [PDF]

				P. J. Wang and D. C. Page Functional substitution for TAFII250 by a retroposed homolog that is expressed in human spermatogenesis Hum. Mol. Genet., September 15, 2002; 11(19): 2341 - 2346. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 276/11/8044    most recent M009091200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager