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

This Article Abstract Full Text (PDF) 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 Kimura, A. Articles by Takahashi, T. Search for Related Content PubMed PubMed Citation Articles by Kimura, A. Articles by Takahashi, T. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 34, 24047-24053, August 20, 1999

Structure and Localization of the Mouse Prolyl Oligopeptidase Gene^*

Atsushi Kimura^§, Ikuya Yoshida^¶, Nobuo Takagi^¶, and Takayuki Takahashi^**

From the  Division of Biological Sciences, Graduate School of Science, the ^¶ Research Center for Molecular Genetics, and the  Division of Bioscience, Graduate School of Environmental Earth Science, Hokkaido University, Sapporo 060-0810, Japan

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

We have cloned and characterized the genomic structure of the mouse gene for prolyl oligopeptidase that is mapped to chromosome 10B2-B3. The gene is about 92 kilobases in size and contains 15 exons. All exon-intron junction sequences conform to the GT/AG rule. Comparison with the presumed domain structures of the mouse prolyl oligopeptidase indicates that the propeller domain of the enzyme is encoded by exons 3-10, whereas the catalytic domain is encoded by exons 1-3 and 10-15. The catalytic triad residues are encoded by two exons (Ser⁵⁵⁴ on exon 13 and His⁶⁸⁰ and Asp⁶⁴² on exon 15). The 5'-flanking region of the mouse prolyl oligopeptidase gene has structural features found in housekeeping gene promoters, including a GC-rich segment and an absence of TATA and CAAT boxes. A primer extension assay showed the presence of multiple sites for the initiation of transcription. Transient transfection analysis demonstrated that the 5'-flanking region of the gene can direct efficient expression in COS1 cells. Deletion studies revealed that the downstream 125-base pair sequence of the region is required for promoter activity in the cells.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Prolyl oligopeptidase, previously called prolyl endopeptidase or post-proline cleaving enzyme, is a peptidase that cleaves the peptide bonds at the carboxyl group of proline residues. Prolyl oligopeptidase belongs to the serine peptidase family, but the enzyme is distinct from trypsin and subtilisin in the order of the primary sequence of catalytic triad residues (1). A very recent study by Fülöp et al. (2) has established the unique three-dimensional structure of porcine muscle prolyl oligopeptidase and may have solved the perplexing problem of why the enzyme only hydrolyzes peptides consisting of no more than 30 amino acid residues (3). Prolyl oligopeptidase is found in various organisms, including mammals (3-12), insects (13), plants (14), mushrooms (15, 16), and bacteria (17-19). Such a wide distribution suggests the general importance of this enzyme.

The biological function of prolyl oligopeptidase is not yet fully understood, but its roles in various biological events have been implicated. Because prolyl oligopeptidase is capable of rapidly degrading many peptide hormones and neuropeptides in vitro (20, 21), the enzyme is presumed to be involved in their maturation and degradation. In this connection, the enzyme has attracted special attention from a pharmaceutical viewpoint in the light of the finding that specific enzyme inhibitors reverse scopolamine-induced amnesia in rats (22-25). In addition, its involvements in depression (26), blood pressure through the metabolism of bradykinin and angiotensins I and II (27), DNA synthesis (28, 29), the generation of amyloid  protein (30-32), and cell proliferation and differentiation (13) have been suggested. In order to define the roles of this enzyme, further studies including a functional characterization of the gene are necessary.

Prolyl oligopeptidase has been cloned from several sources including Flavobacterium meningosepticum (33), porcine brain (1), human lymphocytes (34, 35), mouse brain (29), bovine brain (36), and Sarcophaga peregrina (37). However, no gene structure of prolyl oligopeptidase from eukaryotes is available to date. As an initial approach to examining the mechanisms regulating prolyl oligopeptidase gene expression in mammals, we have cloned and characterized the mouse gene encoding prolyl oligopeptidase. In this paper, we describe the structure and chromosome location of the mouse prolyl oligopeptidase gene. In addition to the gene structure, we have characterized the 5'-flanking region of the gene. The results suggest that the gene has a "housekeeping promoter," characterized by the lack of functional TATA and CAAT boxes as well as the presence of a functional GC-rich promoter region.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Isolation and Characterization of the Mouse Prolyl Oligopeptidase cDNA and Gene-- Poly(A) RNA was prepared from mouse (C57Black/6, female, 8 weeks) brain using the guanidine isothiocyanate-cesium chloride method (38) followed by oligo(dT)-cellulose column chromatography. The adapter-ligated cDNA was inserted into gt10-EcoRI vector (Takara, Japan) and packaged with GIGAPACK II GOLD (Stratagene, La Jolla, CA). About 3.4 × 10⁵ clones were screened using the full-length rat prolyl oligopeptidase cDNA previously isolated in our laboratory (the sequence data available from the DDBJ/EBI/GeneBank^TM Data Bank with the accession number AB012759) as a probe by plaque hybridization. Four positive plaques were isolated and sequenced. The complete sequence of the 3'-end was obtained by the 3'-rapid amplification of cDNA ends method (39) using the 3'-Full RACE Core Set (Takara).

About 5 × 10⁶ clones of the mouse genomic library (C57Black/6, liver, female, 1 year, Stratagene) were screened with the ³²P-labeled mouse prolyl oligopeptidase cDNA mixture of the five total clones described above. Twelve positive plaques were isolated, and digestion with several restriction enzymes revealed that these plaques included four kinds of clones (P13, P23, P24, and P54). We determined the restriction enzyme maps of these four clones and sequenced the restriction fragments containing exons using ABI automatic sequencers, models 373 and 377 (Perkin-Elmer-Applied Biosystems, Foster City, CA). The gaps among these four clones were filled by the long and accurate polymerase chain reaction (LA-PCR)¹ method (40, 41).

Mouse (C57Black/6, female, 8 weeks) liver was homogenized and treated with 20 µg/ml RNase A and then with 100 µg/ml proteinase K (Roche Molecular Biochemicals). After repeated extraction with phenol, genomic DNA was isolated by ethanol precipitation. The PCR was performed under the following conditions: 1 min at 94 °C, 30 cycles of 10 s at 98 °C and 15 min at 66 °C, and 10 min at 72 °C. A 50-µl portion of the reaction mixture contained 500 ng of genomic DNA, 1 × LA-PCR Buffer II (Mg²⁺-free), 2.5 mM MgCl₂, 0.4 mM dNTP mixture, 10 pmol paired primers, and 2.5 units of LA Taq (Takara). The sense and antisense primer pairs used were as follows: LA21 (filling the gap between P13 and P23), 5'-GCATCCTGCCTGGTCTCAGCTTTAATTCTG-3' (identical to 1471-1501 bp downstream of exon 2), and 5'-TAGGGGCAGCTTCACTATTGTGGAGCTTTC-3' (complementary to 10.7 kb upstream of exon 3; LA78 (filling the gap between P23 and P24), 5'-AAAGCATGAACACCCCAGAGATCAGTTTCG-3' (identical to 223-253 bp downstream of exon 6), and 5'-CATCTTGGTGATAGTCAACGAGGAGTACACAACC-3' (complementary to 185-219 upstream of exon 9); and LA105 (filling the gap between P24 and P54), 5'-GGTTTACAATCACTGACGCTTAGGACCTCCTG-3' (identical to 3088-3120 bp downstream of exon 10), and 5'-GAGTGGCTGACTTGGCTGGATTCTTTTCTC-3' (complementary to 1074-1104 bp upstream of exon 11). The PCR products were analyzed by restriction enzyme mapping and sequencing.

Primer Extension-- 1 µg of mouse brain poly(A) RNA or 50 µg of Escherichia coli transfer RNA was hybridized at 75 °C for 90 min in a total 20 µl of 10 mM Tris-HCl (pH 8.3), 1 mM EDTA, and 0.25 M KCl with 2 pmol of the oligonucleotide probes, PE1, 5'-CGTGCGAGCGGGCGAGCGG-3' (+306 to +288), PE2, 5'-ACAAGAGGGAGCGGAGTCGC-3' (+182 to +163), or PE3, 5'-TGAGCAACACTGTTCCTGAG-3' (320 to 339), which had been endo-labeled with T4 polynucleotide kinase (Takara) and [-³²P]ATP at 37 °C for 30 min. The reactions were then allowed to cool for 90 min at room temperature. After adding 46 µl of 65 mM Tris-HCl (pH 8.3), 4.3 mM MgCl₂, 15 mM dithiothreitol, and 0.72 mM dNTP mixture, 200 units of Superscript II reverse transcriptase (Life Technologies, Inc.) was introduced into the reaction mixture, which was then incubated at 42 °C for 60 min. The reactions were phenol-extracted, ethanol-precipitated, and electrophoresed in a denaturing 6% polyacrylamide sequencing gel along with a sequence ladder generated with the unlabeled primers using [-³²P]dCTP and Sequenase version 2.0 (U.S. Biochemical Corp.). The dried gel was visualized by autoradiography.

S1 Nuclease Mapping-- A 914-bp EcoRI-BamHI fragment probe (nucleotides 541 to +373) was 5'-endo-labeled with T4 polynucleotide kinase (Takara) and [-³²P]ATP. The denatured probe (2 × 10⁵ cpm) was hybridized to 50 µg of mouse brain total RNA or E. coli transfer RNA in 80% formamide, 0.4 M NaCl, 40 mM PIPES, and 1 mM EDTA at 62 °C for 16 h. After hybridization, the mixture was digested with 800 units of S1 nuclease (Takara) at room temperature for 20 min. The protected fragments were analyzed on a 5% polyacrylamide sequencing gel. The 1-kb ladder (Life Technologies, Inc.) was endo-labeled and used as a size marker.

Genomic Southern Blot Analysis-- Mouse genome DNA was extracted as described above. 5 µg of the genomic DNA was completely digested with EcoRI, HindIII, BamHI, and XbaI. The DNA was fractionated on a 0.7% agarose gel and alkaline-transferred to a Nytran membrane (Schleicher & Schuell). The blot was hybridized at 60 °C for 16 h in 6× SSPE, 5× Denhardt's solution, 1% SDS, 10% dextran sulfate, and 100 µg/ml denatured herring sperm DNA with a ³²P-labeled 262-bp HindIII-BamHI fragment of mouse prolyl oligopeptidase cDNA, which is included in exon 15. The membrane was washed at 60 °C in 0.1× SSC/0.1% SDS and exposed to Kodak Biomax film.

Chromosomal Localization of Mouse Prolyl Oligopeptidase by Fluorescence in Situ Hybridization-- Metaphase spreads were prepared from concanavalin A-stimulated splenocytes of normal male mice after bromodeoxyuridine incorporation. R-banding was obtained by exposure of the slides to ultraviolet light after staining with Hoechst 33258. A 200-ng sample of probe mixture (P13, P23, P54, LA78, and LA105) was labeled with biotinylated-16-dUTP (Roche Molecular Biochemicals) by nick translation. Prior to hybridization, the chromosome slides were denatured in 70% formamide in 2× SSC at 75 °C and immediately chilled in 70% ethanol (0 °C) and dehydrated. The probe was mixed with 2 µg of mouse Cot-1 DNA (Life Technologies, Inc.) and denatured in formamide at 75 °C. The hybridization was carried out at 42 °C for 48 h in 4× SSC, 4 mg/ml bovine serum albumin, and 20% dextran sulfate. The signal was detected by sequential incubation with 3 µg/ml anti-biotin goat IgG (Vectashield Vector laboratory, Burlingame, CA) and 40 µg/ml fluorescein isothiocyanate-conjugated anti-goat IgG (American Qualex, La Mirada, CA). Fluorescein isothiocyanate signals were further amplified with 40 µg/ml Alexa 488 conjugated anti-fluorescein isothiocyanate (Molecular Probes, Eugene, OR).

Plasmid Construction-- To construct promoter and reporter fusion constructs, different portions of the 5'-flanking region sequences were synthesized by PCR. The PCR was conducted in a total 50-µl volume of 100 ng of template plasmid clone containing the 5'-flanking region, 1× Ex Taq buffer, 0.2 mM dNTP mixture, 10 pmol paired primers, and 2.5 units Ex Taq (Takara). The reactions were performed under the following conditions: 1 min at 94 °C and 30 cycles of 30 s at 98 °C, 30 s at 60 °C, and 3 min at 72 °C. The downstream primers were 5'-CGCGGATCCTACACGTCGGGGTACTGGAA-3' (+406 to +387), 5'-CGCGGATCCGCGCCAAGCAGGAAGACGCT-3' (+249 to +230), 5'-CGCGGATCCACAAGAGGGAGCGGAGTCGC-3' (+182 to +163), and 5'-CGCGGATCCACCAGCCACTCTGTATATGG-3' (62 to 81). The upstream primers were 5'-CCGGAATTCGAGGCCCATGGATAGAGAAG-3' (422 to 403), 5'-CCGGAATTCAAGTTCTCTCACCTCGCTCC-3' (256 to 237), 5'-CCGGAATTCGGCTGGTGGCAGGCAGGCGG-3' (68 to 49), and 5'-CCGGAATTCTCCCTCTTGTCGCCGCTTGG-3' (+173 to +154). The PCR products were sequenced and subcloned into pEGFP-1 vector (CLONTECH, Palo Alto, CA) at EcoRI-BamHI site. The whole promoter sequence (541 to +373) was subcloned from the plasmid clone containing the 5'-flanking region at EcoRI-BamHI site.

Cell Culture and Transfection-- The African green monkey kidney SV40 transformed cell line (COS1, RCB0143) was purchased from Riken Gene Bank (Tsukuba, Japan). The cells were cultured in Dulbecco's modified Eagle's medium supplied with 10% fetal bovine serum (Hyclone, Logan, UT), 0.3 mg/ml L-glutamine, 100 units/ml penicillin, and 0.1 mg/ml streptomycin (Life Technologies, Inc.). When the cells in 75-cm² flasks become confluent, they were trypsinized with 0.25% trypsin/0.5 mM EDTA and transferred into 10-cm dishes in the same medium at an initial concentration of 1 × 10⁶ cells/dish. The cells were cultured for 24 h and transfected by a calcium phosphate precipitation procedure (42) with 20 µg of plasmids and 20 µg of pSV--Gal (Promega, Madison, WI). The transfected cells were further cultured for 48 h. The harvested cells were washed twice with phosphate-buffered saline, suspended in 2 ml/dish of phosphate-buffered saline, and disrupted by the freeze-thawing method. After centrifugation at 8000 × g for 10 min, the supernatant was collected for use in the following assays. The green fluorescent protein (GFP) activity was measured by spectrofluorometry using an excitation wavelength of 488 nm and an emission wavelength of 507 nm. The -galactosidase assay was carried out as described by standard protocols (43). The GFP activity was normalized by -galactosidase activity.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Cloning and Nucleotide Sequencing of the Mouse Prolyl Oligopeptidase Gene-- We isolated four clones of P13 (13.0 kb), P23 (19.0 kb), P24 (11.0 kb), and P54 (20.9 kb) by screening the genomic library. Because these four clones were found to neither encode the 7th and 8th exon nor overlap one another, the LA-PCR method was used to fill the gaps between them. Three LA-PCR products, LA21 (7.0 kb), LA78 (21.9 kb), and LA105 (18.2 kb), were generated for the respective gaps. Consequently, these overlapping seven clones spanned a length approximately 99.1 kb long containing 91.9 kb of the complete prolyl oligopeptidase gene (Fig. 1). The genome contained 15 exons ranging from 75 (2nd exon) to 767 bp (15th exon) in size and 14 introns ranging from 1535 to 22.3 kb. All donor and acceptor splicing sequences had consensus GT and AG dinucleotides, respectively (Table I).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Restriction maps and structural organization of mouse prolyl oligopeptidase gene. The characteristic -propeller (each domain indicated by an oval) and peptidase domains (hatched boxes) are represented at the top. The exon organization of the cDNA is shown below the domain structure. A coding region is indicated by open boxes, and noncoding regions are indicated by horizontal lines. Exon numbers and the positions of the three active-site residues (S, D, and H) are shown. At the bottom, recombinant phage inserts (P13, P23, P24, and P54) and LA-PCR products (LA21, LA78, and LA105) as well as the genomic organization are indicated. The exons are represented on the upper side with numbers, and the restriction maps are on the lower side. Sites for XbaI (large bars), HindIII (small bars), and EcoRI (black circles) are indicated.

A very recent study by Fülöp et al. (2) has established the tertiary structure of porcine muscle prolyl oligopeptidase. The enzyme contains a peptidase domain with an / hydrolase fold, and the active site is covered by the central tunnel of a unique -propeller (2). Although the three-dimensional structure of the mouse prolyl oligopeptidase has not yet been determined, it should be very similar to that of the porcine enzyme based on the remarkable similarities in the amino acid sequences of the two enzymes (29). Therefore, we have considered the relationship between the protein and gene structures for the mouse prolyl oligopeptidase, assuming that the mouse enzyme has basically the same catalytic and noncatalytic domain organization (Fig. 1). The catalytic domain (or peptidase domain) is composed of residues 1-72 and 428-710. This domain is encoded by exons 1-3 for the amino-terminal portion and exons 10-15 for the carboxyl-terminal portion. Exon 15, the longest exon encoding the carboxyl-terminal 98 amino acid residues and the 3'-untranslated regions, contains two catalytic triad residues (Asp⁶⁴² and His⁶⁸⁰) of the mouse prolyl oligopeptidase. The active site serine residue (Ser⁵⁵⁴) is encoded by exon 13. On the other hand, the noncatalytic domain corresponding to residues 73-427 is encoded by exons 3-10. A structural feature of this domain is a 7-fold repeat of four-stranded antiparallel -sheets forming a -propeller domain with seven blades (2). These repeating units at the peptide level are not encoded by individual exons. The mouse prolyl oligopeptidase gene contains two large introns, intron 2 (19.1 kb) and intron 10 (22.3 kb), resulting in a distribution of the exons in three clusters (Fig. 1). In a general way, the first and last clusters of the exons are responsible for the peptidase domain, whereas the middle one is responsible for the -propeller domain. Such an exon-intron organization may indicate that prolyl oligopeptidase has evolved from the insertion of the propeller domain into the NH₂-terminal region of the peptidase domain.

Determination of the Transcription Initiation Site-- Ishino et al. (29) have recently reported the nucleotide sequence of mouse prolyl oligopeptidase cDNA. We also cloned the cDNA, and our results confirmed their data, except that the sequence in the 5'-untranslated region was totally different. To explain this conflict and to establish a basis for primer extension experiments and promoter analysis of the gene, we conducted reverse transcriptase-PCR and LA-PCR using different sets of primers. The results showed that the PCR product was amplified only when using the synthesized primer sets based on the sequences of our clone (data not shown). The real proof of the accuracy of our cDNA sequence was its correspondence to the genomic sequence.

The precise location of the transcription initiation site of the mouse prolyl oligopeptidase gene was determined by the primer extension analysis. Using the primer PE1, one major band and two minor bands were detected (Fig. 2a). From this result, one of the minor sites was determined to be thymine (+139). Because the major and the other minor bands were far upstream of the PE1 primer site, we newly designed two primers, PE2 and PE3, respectively, for further analysis. The results of experiments with these primers indicated that the major start site is the thymine (+1) (Fig. 2b) and the minor site is the guanine (408) (Fig. 2c). To further confirm these findings, we conducted an S1 nuclease mapping experiment. As shown in Fig. 3, a major protected band was found in a position corresponding to 235 bp, confirming that thymine (+139) is an important transcription initiation site. This result along with the results of the primer extension analyses indicates that both thymines (+1 and +139) are important initiation sites in the transcription of the mouse prolyl oligopeptidase gene (Fig. 4).

View larger version (78K): [in this window] [in a new window]   Fig. 2.   Determination of 5'-ends of mouse prolyl oligopeptidase transcripts by primer extension analysis. 1 µg of mouse brain poly(A) RNA (lanes 1, 3, and 5) or 50 µg of E. coli transfer RNA (lanes 2, 4, and 6) were annealed with 32P-labeled PE1 (lanes 1 and 2), PE2 (lanes 3 and 4), or PE3 (lanes 5 and 6) primers and extended with reverse transcriptase. The extended products (indicated by arrows) were resolved by sequencing gels in parallel with four sequencing reactions. The positions of primers and the determined transcription initiation sites are shown in Fig. 4.

View larger version (42K): [in this window] [in a new window]   Fig. 3.   Identification of the major transcription initiation site by S1 nuclease mapping. A 914-bp EcoRI-BamHI fragment was 5'-endo-labeled and hybridized to mouse brain total RNA (lane 3) or E. coli transfer RNA (lane 4). After S1 nuclease digestion, the products were fractionated by electrophoresis and visualized by autoradiography. The endo-labeled size marker (lane 1) and the probe (lane 2) were also electrophoresed. The marker sizes are shown on the left. The upper arrow indicates the position of the probe, and the lower arrow indicates a single protected band at 235 bp.

View larger version (49K): [in this window] [in a new window]   Fig. 4.   DNA sequence of the 5'-flanking region of mouse prolyl oligopeptidase gene. The positions are numbered from the more upstream of the two main transcription initiation sites identified, which is numbered +1. The transcription initiation sites are indicated by bent arrows. The primers used in the primer extension analysis (PE1, PE2, and PE3) are indicated by horizontal arrows. Possible regulatory elements are underlined or overlined. The BamHI sites are shown in italic type, and the translation initiation codon (ATG) is in bold type.

Although there is a TATAAT motif 915 bp upstream of ATG, this motif could not work as a TATA box because the upstream minor transcription initiation site is too distant from the motif. Consistent with these observations is the presence of a consensus motif GCTCC(C/G) at four positions in the sequence covering the nucleotides from 408 to the ATG codon (Fig. 4). This motif is known as a multiple start site element downstream (MED-1), which is detectable in many TATA-less promoters with multiple start sites (44). This region is rich in G+C and contains potential binding sites for several transcription factors such as Sp1, Ap2, and E4TF.

Chromosome Localization of Mouse Prolyl Oligopeptidase Gene-- With the Southern hybridization analysis, a single hybridization-positive band was detected when mouse genomic DNA was digested with four independent restriction enzymes (Fig. 5). Fluorescence in situ hybridization analysis showed a single positive signal on chromosome 10B2-B3, and no other signal was detected (Fig. 6). This result is consistent with the mapping of the human prolyl oligopeptidase gene to human chromosome 6q22 (45), because this region of mouse chromosome 10 shares a region of synteny with human chromosome 6. These results indicate that the mouse prolyl oligopeptidase is a single-copy gene located on chromosome 10B2-B3.

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Genomic Southern blot analysis of mouse prolyl oligopeptidase gene. 5 µg of mouse genomic DNA were used in each digestion. The restriction enzymes selected are indicated. The blot was hybridized with a 32P-labeled 262-bp HindIII-BamHI fragment of mouse prolyl oligopeptidase cDNA, which is included in exon 15. The membrane was washed with a final stringency of 0.1 × SSC/0.1% SDS at 60 °C before autoradiography. Molecular size markers are indicated to the left.

View larger version (56K): [in this window] [in a new window]   Fig. 6.   Chromosomal localization of mouse prolyl oligopeptidase gene. A, metaphase chromosomes hybridized with a mouse prolyl oligopeptidase genomic probe mixture. Hybridization signals were detected at band B2-B3 on both chromosomes 10 (arrows). Of the 73 informative cells, the signals were observed on all four chromatids in all 73 cells. B, G-banding pattern of the same chromosome.

Characterization of the Promoter-- The mouse prolyl oligopeptidase gene isolated in this study contained only 900 nucleotides in the 5'-flanking region. This size seemed somewhat too short for promoter analysis. We tried to obtain the sequence further upstream of the gene by screening an additional 2 × 10⁶ genomic clones, but this attempt was not successful. Therefore, we performed the promoter analysis of the gene with the above region.

The mouse prolyl oligopeptidase gene contains neither a TATA box nor a CAAT box in the region adjacent to the transcription start site (Fig. 4). Interestingly, the 5'-region immediately upstream of the ATG codon is a GC-rich sequence (77.1% between 68 and +377). To examine whether the 5'-flanking region, including the GC-rich sequence, of the gene contained a functional promoter, we constructed a series of 5'- or 3'-deletion mutants linked to the GFP reporter gene (Fig. 7). The prolyl oligopeptidase activity was assayed for the extracts of cultured cells such as COS1, mouse B16 melanoma, and Lewis lung carcinoma cells. We observed that the COS1 cells exhibited the greatest activity (data not shown) and therefore used this cell line for the promoter analysis. As shown in Fig. 7, the 5'-deletion from 541 to 422 appeared to cause a slight reduction (26%) in promoter activity, but this reduction was not significant. Further deletions from 422 to +173 led to no change in activity, suggesting that the region from 541 to +173 was dispensable in the cell. In contrast, transfection by the reporter plasmids with 3'-deletions had a remarkable effect on promoter activity. Deletion from 62 to +373 retained only 7.3% of the promoter activity as compared with that of the complete construct. This reduced activity was not restored even by the 3'-deletion mutant lacking only the 125-bp downstream region from +249 to +373. These results strongly suggest that the region from +249 to +373 is required for high basal promoter activity in COS1 cells.

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Deletion analysis of the mouse prolyl oligopeptidase promoter activity. The left panel indicates schematic drawings of various fragments of the 5'-flanking region subcloned into the upstream of the GFP reporter plasmid pEGFP-1 at EcoRI-BamHI sites. The numbers indicate positions of the 5' or 3' terminus of the constructs. The possible cis-element binding sites are shown at the top. The reporter constructs are transfected into COS1 cells by the calcium phosphate method. The measured GFP activity was shown in the right panel. The data represent the averages ± S.E. of at least four independent experiments. The GFP activity of the promoter-less construct is significantly higher than that of some other constructs, presumably because of the presence of an Ap2-binding motif between EcoRI and BamHI of the pEGFP-1 multi-cloning site.

The 5'-flanking region of the mouse prolyl oligopeptidase gene is characterized by the absence of a TATA motif and the presence of several GC motifs. These structural features are often found in promoters of the housekeeping genes (46, 47). The previous results of Northern blot analyses attempting to identify prolyl oligopeptidase mRNA in various pig (1) and human (34) tissues are compatible with the idea that the prolyl oligopeptidase gene is a housekeeping gene. However, further detailed investigation is necessary to determine in what way this gene is a housekeeping gene. It is interesting to note that the dipeptidyl peptidase IV gene, which belongs to the same prolyl oligopeptidase family, also has a 5'-flanking region characteristic of a housekeeping gene promoter (48).

Comparison with Dipeptidyl Peptidase IV-- Because the current study provides for the first time the genomic structure of mammalian prolyl oligopeptidase, it is worthwhile to compare this structure with the gene structure of related peptidases. The only member of the prolyl oligopeptidase family whose genomic structure has been analyzed so far is mouse and human dipeptidyl peptidase IV (CD26) (49, 50). When the amino acid sequences of mouse prolyl oligopeptidase (29) and dipeptidyl peptidase IV (51) were compared, a moderate similarity was seen throughout the sequences (identical residues were 15.1% in the alignment taken with the Clustal W program (52); data not shown). A comparable level of similarity (identical residues were 14.9%) was observed between the amino acid sequences of the two enzymes even in the COOH-terminal third of the alignment. The exon-intron organization of these mouse genes is quite different, however. Despite the comparable number of amino acid residues in the mouse enzymes (prolyl oligopeptidase, 710 residues; dipeptidyl peptidase IV, 760 residues), the number of exons in the two genes is different; the prolyl oligopeptidase gene is composed of 15 exons, whereas the dipeptidyl peptidase IV gene 26 exons. Because both genes encompass a similar number of nucleotides (91.9 kb for the prolyl oligopeptidase gene and more than 90 kb for the dipeptidyl peptidase IV gene), the average exon of the prolyl oligopeptidase gene is approximately 1.6 times longer than that of dipeptidyl peptidase IV gene. This greater length may indicate that, evolutionally, new intron insertions have occurred more frequently in the dipeptidyl peptidase IV gene than in the prolyl oligopeptidase gene. Further comparative analysis revealed that none of the exon-intron junctions of the two genes correspond with each other. In addition, the catalytic triad residues of mouse prolyl oligopeptidase are encoded by two exons (Ser⁵⁵⁴ on exon 13 and Asp⁶⁴² and His⁶⁸⁰ on exon 15), whereas each residue of mouse dipeptidyl peptidase IV is separately encoded by different exons (Ser⁶²⁴ on exon 22, Asp⁷⁰² on exon 24, and His⁷³⁴ on exon 26) (49). Although it is clear that prolyl oligopeptidase and dipeptidyl peptidase IV are related to each other (53, 54), a comparison of their gene structures did not give very much valuable information as to the evolution of these enzymes.

Conclusions-- In this study, we have cloned and characterized the genomic structure of the mouse prolyl oligopeptidase gene. Our present data clearly show that a single-copy gene in chromosome 10B2-B3 encodes mouse prolyl oligopeptidase. We have previously reported an extracellular enzyme having properties indistinguishable from intracellular prolyl oligopeptidase in the porcine ovary (55). Subsequent studies have suggested that these intracellular and extracellular enzymes are the same products of a single gene (56). However, existing evidence points to the presence of other types of prolyl oligopeptidase in mammalian tissues (57-60), although structural information on these enzymes has not been available to date. We presume that some of the enzymes with enzymatic properties very similar to those of prolyl oligopeptidase are perhaps the same gene products but have undergone different posttranslational modification(s). An alternative processing event(s) involving internal exons of the prolyl oligopeptidase gene could also produce enzymes with modified properties. Another possibility is that a yet unidentified gene(s) encodes a distinct enzyme(s) having prolyl oligopeptidase-like enzyme activity. Further studies are definitely required to clarify the relation between prolyl oligopeptidase and such enzymes. The current data on the genomic structure of prolyl oligopeptidase should be helpful in solving this problem.

Many studies have been carried out with prolyl oligopeptidase since its discovery by Walter et al. (4) in 1971. Various physiological functions for the enzyme have been suggested to date, but definitive evidence for its discrete biological role remains to be provided. The current investigation should provide a foundation for future studies of prolyl oligopeptidase, including loss or gain of gene function studies.

	Note Added in Proof

After this paper was accepted, Williams et al. (61) have reported the structure of the Dictyostelium prolyl oligopeptidase gene, and demonstrated that loss of the gene caused an increased concentration of inositol (1,4,5)-triphosphate, offering a novel mechanism that links the enzyme activity to the intracellular signaling.

	FOOTNOTES

^* This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan.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 GenBank^TM/EMBL Data Bank with accession number(s) AB022047, AB022048, AB022049, AB022050, AB022051, AB022052, and AB022053.

^§ Supported by a Research Fellowship of the Japan Society for the Promotion of Science.

^** To whom correspondence should be addressed: Div. of Biological Sciences, Graduate School of Science, Hokkaido University, Sapporo 060-0810, Japan. Tel.: 81-11-706-2748; Fax: 81-11-706-4851; E-mail: ttakaha@sci.hokudai.ac.jp.

	ABBREVIATIONS

The abbreviations used are: LA-PCR, long and accurate polymerase chain reaction; PIPES, piperazine-N,N'-bis[2-ethanesulfonic acid]; GFP, green fluorescent protein; kb, kilobase(s); bp, base pair(s).

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				A. L. Y. Pang, H. C. Taylor, W. Johnson, S. Alexander, Y. Chen, Y. A. Su, X. Li, N. Ravindranath, M. Dym, O. M. Rennert, et al. Identification of Differentially Expressed Genes in Mouse Spermatogenesis J Androl, November 1, 2003; 24(6): 899 - 911. [Abstract] [Full Text] [PDF]

				H. Matsui, A. Kimura, N. Yamashiki, A. Moriyama, M. Kaya, I. Yoshida, N. Takagi, and T. Takahashi Molecular and Biochemical Characterization of a Serine Proteinase Predominantly Expressed in the Medulla Oblongata and Cerebellar White Matter of Mouse Brain J. Biol. Chem., April 6, 2000; 275(15): 11050 - 11057. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) 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 Kimura, A. Articles by Takahashi, T. Search for Related Content PubMed PubMed Citation Articles by Kimura, A. Articles by Takahashi, T. Social Bookmarking