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

J. Biol. Chem., Vol. 280, Issue 21, 20204-20215, May 27, 2005

This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: 280/21/20204    most recent M501678200v1 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 Furutani, Y. Articles by Hirose, S. Search for Related Content PubMed PubMed Citation Articles by Furutani, Y. Articles by Hirose, S. Social Bookmarking                      What's this?

Identification, Evolution, and Regulation of Expression of Guinea Pig Trappin with an Unusually Long Transglutaminase Substrate Domain^*

Yutaka Furutani¶, Akira Kato, Azzania Fibriani, Taku Hirata, Ryoji Kawai, Ju-Hong Jeon||, Yasuhisa Fujii**, In-Gyu Kim||, Soichi Kojima, and Shigehisa Hirose

From the Molecular Cellular Pathology Research Unit, RIKEN, Wako-shi, Saitama 351-0198, Japan, the Department of Biological Sciences, Tokyo Institute of Technology, Midori-ku, Yokohama 226-8501, Japan, the ^||Department of Biochemistry and Molecular Biology, Aging and Apoptosis Research Center, Seoul National University College of Medicine, Seoul 110-799, Korea, and the ^**Department of Urology and Reproductive Medicine, Tokyo Medical and Dental University Graduate School, Bunkyo-ku, Tokyo 113-8519, Japan

Received for publication, February 14, 2005

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Trappins are found in human, bovine, hippopotamus, and members of the pig family, but not in rat and mouse. To clarify the evolution of the trappin genes and the functional significance of their products, we isolated the trappin gene in guinea pig, a species belonging to a rodent family distinct from rat and mouse. Guinea pig trappin was confirmed to encode the same domain structure as trappin, consisting of a signal sequence, an extra large transglutaminase substrate domain, and a whey acidic protein motif. Northern blot analysis and in situ hybridization histochemistry as well as immunohistochemistry demonstrated that guinea pig trappin is expressed solely in the secretory epithelium of the seminal vesicle and that its expression is androgen-dependent. We confirmed that guinea pig trappin is cross-linked by prostate transglutaminase and that the whey acidic protein motif derived from guinea pig trappin has an inhibitory activity against leukocyte elastase. Genome sequence analysis showed that guinea pig trappin belongs to the family of REST (rapidly evolving seminal vesicle transcribed) genes.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Trappins are a family of secreted proteins composed of two domains: a transglutaminase substrate (TGS)¹ domain and a whey acidic protein (WAP) motif (1). The TGS domain serves as an anchoring sequence through which trappins become covalently trapped at the site of action (2, 3). Covalent anchoring is mediated by transglutaminase, a Ca²⁺-dependent enzyme that catalyzes cross-linking of proteins between Lys and Gln side chains (4–6). The WAP motif comprises eight Cys residues that form four disulfide bonds in a conserved arrangement. This four-disulfide core structure was first described in whey acidic protein in mouse milk and thus was named the "WAP motif" (7). Members of the family of WAP motif proteins contain one to three WAP motifs, many of which have proteinase inhibitor activity and/or antibacterial activity (8, 9). Trappins are unique in having TGS domains in their N termini among diverse family members of WAP motif-containing proteins that include toxins, pollen allergens, serine proteinase inhibitors, growth inhibitors, calcium transport inhibitors, and others whose functions have not yet been determined (10–17).

The two-domain structure of trappins suggested that such a combined structure might have evolved by exon shuffling between ancestral genes for a TGS protein and a WAP motif protein. Determination of the structures of trappin genes by us (18–20) and those of the semenogelin or seminal vesicle clotting protein (SVP) genes (21–23) indeed revealed a significant similarity between the trappin genes and the semenogelin or SVP genes that encode TGSs and are collectively called REST (rapidly evolving seminal vesicle transcribed) genes. Furthermore, phylogenetic analysis of the WAP motif sequences indicated that the WAP signature sequence of trappins is most closely related to the second WAP motif of the secretory leukocyte proteinase inhibitor (SLPI), which has two WAP motifs in tandem (18). These findings led to the hypothesis that the trappin genes evolved through exon shuffling from the ancestral REST and SLPI genes (1, 18).

The presence of trappins has been demonstrated in only a limited number of species, including pig, wart hog, collared peccary, bovine, hippopotamus, and human (18, 20, 24–27). The use of small animals is advantageous for studying the functional significance of trappins. However, attempts by DNA data base searching and homology screening to identify rat and mouse homologs of trappins have so far been unsuccessful. Now, the lack of trappin genes in mouse has been established by the completion of mouse genome sequencing projects. So, we extended the search to guinea pig, a rodent species considered to be distantly related to rat and mouse in the Rodentia order. In guinea pig, a WAP motif protein called caltrin II have been identified and shown to have the WAP motif by protein sequencing (16), but its N-terminal amino acid sequence was not known. We therefore first considered the possibility that caltrin II may be a guinea pig homolog of trappins and decided to determine its complete amino acid sequence by cDNA cloning. Contrary to our expectations, however, sequencing of a full-length caltrin II cDNA revealed the absence of the TGS domain (28). We next tried homology screening and successfully identified a guinea pig trappin. Here, we report its mRNA structure, properties, tissue distribution, androgen-dependent regulation of expression, and gene structure, which contribute to clarifying the evolution and function of the trappin family. The identification and characterization of the trappin gene in guinea pig demonstrate that the guinea pig trappin gene has a unique structure that evolved after speciation. This is consistent with the characteristics of the trappin gene family: the duplication and variation occurred in each mammalian species after speciation, and the numbers and compositions of trappin genes are varied among species. Previous studies have functionally established protective roles of trappins in the skin (trappin-2), circulation and hemostatic plug (trappin-1), respiratory tract (trappin-2), and digestive tract (trappin-2) (for review, see Ref. 1); trappin-2 is an elastase inhibitor and is also called SKALP/elafin. In addition to these functions, the results from this study suggest a potential protective role of trappin in the vaginal plug.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Materials—Guinea pigs (male, 7 weeks old) were obtained from Tokyo Laboratory Animal Science Co., Ltd. (Tokyo, Japan). The mRNA purification kit, Hybond H⁺ nylon membranes, [-³²P]dCTP, and the Ready-To-Go DNA labeling kit were from Amersham Biosciences (Buckinghamshire, United Kingdom). T4 polynucleotide kinase, restriction enzymes, and the DNA ligation kit (version 2) were from Takara (Kyoto, Japan). pBluescript II SK(–), ZAP II, and a guinea pig kidney genomic DNA library in FIX II were from Stratagene. Nitrocellulose filters were from Schleicher & Schüll (Dassel, Germany). The SequiTherm Long-Read cycle sequencing kit for LiCor and MaxPlax Lambda packaging extract were from Epicentre Technologies Corp. (Madison, WI). The FirstChoice RLM-RACE kit was from Ambion Inc. (Austin, TX). The SuperScript Choice system was from Invitrogen.

Preparation of RNA and Construction of a cDNA Library—Seminal vesicles were removed from five adult guinea pigs (8 weeks old), and total RNA was isolated by the guanidinium thiocyanate/CsCl method (29). Poly(A)-rich RNA was then obtained by column chromatography on oligo(dT)-cellulose (Amersham Biosciences). A cDNA library was constructed in the ZAP II vector using an oligo(dT) primer and the SuperScript Choice system for cDNA synthesis according to the manufacturer's instruction. The ligated cDNA library was packed into phage particles using MaxPlax Lambda packaging extract.

cDNA Library Screening and Sequencing—Approximately 300,000 independent phage plaques were screened using a random-primed [-³²P]dCTP-labeled probe corresponding to nucleotides 1–734 of hippopotamus trappin cDNA (18). The hybridization conditions were 20% formamide, 6x SSPE (0.9 M saline, 60 mM sodium phosphate (pH 7.7), and 6 mM EDTA), 1% SDS, and 5x Denhardt's solution (0.1% polyvinylpyrrolidone, 0.1% bovine serum albumin, and 0.1% Ficoll) at 42 °C for 16 h. Filters were sequentially washed in 1x SSC (0.15 M saline and 15 mM sodium citrate (pH 7.0)) and 0.1% SDS at 50 °C and exposed to Eastman Kodak X-Omat film for 36 h. The positive clones were isolated and converted to pBluescript II SK(–) plasmid DNA by f1 helper phage-mediated in vivo excision and recircularization using Escherichia coli XL1-Blue and ExAssist helper phage according to the manufacturer's instructions. The subcloned DNA was sequenced by the dideoxynucleotide chain termination method using the SequiTherm Long-Read cycle sequencing kit for LiCor. More than 10 independent clones were sequenced on both strands.

CAP Site Rapid Amplification of cDNA Ends (RACE)—A partial cDNA sequence of guinea pig trappin was extended by 5'-RACE using the FirstChoice RLM-RACE kit. Poly(A)⁺-selected RNA was treated with calf intestinal phosphatase to dephosphorylate 5'-truncated transcripts, followed by tobacco acid pyrophosphatase to remove CAP from the remaining full-length mRNAs. The synthetic RNA adapter from the kit was then ligated to the full-length 5'-monophosphate-containing transcript population using T4 RNA ligase, followed by first-strand cDNA synthesis with random decamers as primers. For the initial PCR, the conditions were as follows: 0.4 µM outer 5'-primer (5'-GCTGATGGCGATGAATGAACACTG-3') provided in the kit, 0.4 µM gene-specific primer sp1 (5'-AATGTGACCGATCGTGGGATC-3'), and 1 µl of the first-strand cDNA as template. Amplification by the Expand Long Template PCR system (Roche Applied Science) was carried out for 35 cycles at 95 °C for 30 s, 60 °C for 60 s, and 72 °C for 4 min and for one cycle at 72 °C for 10 min with 1.75 mM MgCl₂, with the reaction mixture being heated to 95 °C for 1 min before addition of the DNA polymerase mixture. The PCR mixture was resolved on a 1% agarose gel, and bands of 3.6, 2.0–1.0, 0.8–0.5, and 0.5–0.2 kb from the amplification with primer sp1 were separated. The 3.6-kb band was isolated and purified using QIAquick gel extraction kit II (Qiagen Inc.) in a 50-µl solution. One µl of 500-fold diluted solution was then re-amplified by nested PCR (35 cycles at 95 °C for 30 s, 60 °C for 60 s, and 72 °C for 4 min and one cycle at 72 °C for 10 min with 1.75 mM MgCl₂) using 0.4 µM inner 5'-primer (5'-CGCGGATCCGACACTCGTTTGCTGGCTTTGATG-3') from the kit and primer sp1. The defined 3.6-kb band was isolated, phosphorylated, and cloned into EcoRV-digested and dephosphorylated pBluescript II SK(–), and 10 clones were sequenced.

Genomic Library Screening—A guinea pig genomic DNA library constructed in FIX II was plated at 50,000 plaques/plate (15 x 10 cm). About 1,000,000 phage plaques were screened using the longest cDNA and CAP site RACE product as probes under stringent conditions: hybridization in 40% formamide, 6x SSPE, 1% SDS, and 5 x Denhardt's solution and a final wash in 1x SSC and 0.1% SDS at 55 °C. Two positive clones were obtained and then sequenced.

PCR Amplification of the Guinea Pig Trappin Gene—Genomic DNA was isolated from guinea pig liver according to the standard method (30). Four PCR primers were designed based on the cDNA sequence of guinea pig trappin. The sequences of the primers were as follows: 5'-ATGAAGCCTACCGTCTTCCTC-3' and 5'-TTAACCCAAATGGAACCGATCAT-3' for PCR product 1 and 5'-ATGGAGTGTTTGATCCCCGAGTGA-3' and 5'-TTGACTCGCATGCGACCTTTCAT-3' for PCR product 2 (see Fig. 1A). PCR was carried out with 10 ng of the genomic DNA as template, 2 µM primers, and the Expand Long Template PCR system. PCRs were performed for 35 cycles under the following conditions: denaturation for 30 s at 95 °C, annealing for 1 min at 65 °C, and extension for 3 min at 72 °C with 1.75 mM MgCl₂. The defined band was isolated, phosphorylated, cloned into EcoRV-digested and dephosphorylated pBluescript II SK(–), and sequenced.

Nucleotide Sequence Analyses—The nucleotide sequences of guinea pig trappin and the other related proteins were aligned using ClustalX software (31), and best fit/gap placement was confirmed manually. Phylogenetic analysis was performed using the Neighbor-Joining method (32) within MEGA Version 2.1 software (33). The nucleotide substitution rates, i.e. Jukes-Cantor distance (34), were calculated using MEGA Version 2.1. Distance values for synonymous and non-synonymous substitutions were calculated by the method of Nei and Gojobori (52) using MEGA Version 2.1. Harr plot analyses were performed at a 14/20 nucleotide stringency using Genetyx-win (Genetyx Co., Tokyo).

Animal Treatment—Hartley guinea pigs were housed under a constant 12-h light/12-h dark cycle and were allowed free access to standard food and water. Adult male guinea pigs (7 weeks old) were separated into four groups and subjected to the following treatments (n = 3): (i) injection of vehicle alone (200 µl of olive oil); (ii) injection of 2 mg of 17-estradiol; (iii) castration, followed by injection of vehicle; and (iv) castration, followed by treatment with 2 mg of testosterone propionate. Injections were performed subcutaneously every other day (on days 1, 3, 5, 7, 9, and 11), and the animals were killed on day 12.

Northern Blot Analysis—Total RNA was isolated from various tissues of 11-week-old male guinea pigs by the acid guanidinium thiocyanate/phenol/chloroform method. For Northern analysis, 5 µg of total RNA was separated on formaldehyde-containing 1% agarose gel and transferred to Hybond H⁺ nylon membrane by vacuum blotting. A subcloned cDNA fragment covering the intronic sequences in exon 2 (nucleotides 4698–4932) of guinea pig trappin cDNA was used as a probe. Guinea pig -actin cDNA (nucleotides 1–368; GenBank^TM accession number AF508792 [GenBank] ) and glyceraldehyde-3-phosphate dehydrogenase cDNA (nucleotides 1–370; GenBank^TM accession number U51572 [GenBank] ) were also used as control probes. The probes were ³²P-labeled by random priming using the Ready-To-Go DNA labeling kit and hybridized to the RNA filters in 40% formamide, 6x SSPE, 1% SDS, and 5 x Denhardt's solution for 16 h at 42 °C. After hybridization, the filters were washed in 0.1x SSC at 65 °C, exposed to an imaging plate for 48 h, and then analyzed with a BAS 2500 imaging analyzer (Fujifilm, Tokyo). Gels were calibrated with RNA molecular mass markers (Novagen, Madison, WI).

In Situ Hybridization—Seminal vesicles isolated from guinea pig were fixed in 4% paraformaldehyde at 4 °C for 15 h. Cryostat sections (7 µm) were cut and attached to slides coated with Vectabond (Vector Laboratories, Burlingame, CA). In situ hybridization was performed with digoxigenin-labeled sense or antisense RNA probes complementary to the 3'-fragment of guinea pig trappin cDNA (nucleotides 4293–5087). Digoxigenin-labeled RNA probes were synthesized from guinea pig trappin cDNA using T3 or T7 RNA polymerase, and digoxigenin-labeled nucleotides were prepared using a digoxigenin RNA labeling kit (Roche Applied Science) according to the manufacturer's instructions. Sections were post-fixed in 4% paraformaldehyde in diethyl pyrocarbonate-treated phosphate-buffered saline (PBS) for 30 min, washed twice in PBS with 0.1% active diethyl pyrocarbonate for 15 min, and equilibrated in diethyl pyrocarbonate-treated 5x SSC. Prehybridization was carried out in a damp chamber at 58 °C for 2 h in 50% formamide and 5x SSC. Hybridization with the probe (final concentration of 400 ng/ml) was carried out overnight at 58 °C in a damp humidified chamber. Sections were then sequentially washed in 2x SSC for 30 min at room temperature, in 2x SSC for 1 h at 65 °C, and in 0.1x SSC for 1 h at 65 °C; equilibrated in 0.1 M Tris-HCl (pH 7.5) and 0.15 M NaCl for 5 min; and incubated with alkaline phosphatase-conjugated anti-digoxigenin antibody. Excess antibody was washed away, and the color substrates (0.34 mg/ml nitro blue tetrazolium salt and 0.18 mg/ml 5-bromo-4-chloro-3-indolyl phosphate) were added. Slides were allowed to develop in the dark, and the color was visualized under a light microscope until maximum levels of staining were achieved. The slides were washed in TE buffer (10 mM Tris-HCl (pH 8.0) and 1 mM EDTA) for 15 min to stop the reaction, rinsed in 95% ethanol for 1 h to remove nonspecific background, rinsed in water for 15 min to dissolve and remove potential crystals due to TE buffer, dehydrated, and cover-slipped. As a control, parallel incubations with a digoxigenin-labeled RNA sense strand were performed under identical conditions.

Production and Purification of Recombinant Proteins—To produce recombinant proteins representing part of the TGS domain plus a WAP motif of guinea pig trappin, the products were used for determining the potential of guinea pig trappin as a TGS, for raising antiserum, and for determining its anti-proteinase activity. The following expression vectors were constructed: pTgs-Gpt (where Gpt is guinea pig trappin), pTgs-Wap-Gpt, and pWap-Gpt. pTgs-Gpt was designed to encode part of the TGS domain corresponding to amino acids 1403–1513 in Fig. 1E and was amplified with primers 5'-AAGGATCCCAGCTCAAAGAACACAATTTC-3' (containing a BamHI site) and 5'-TGAAGCTTTCAAGGTTTACTGAAACTTCGC-3' (containing a HindIII site). pTgs-Wap-Gpt was designed to encode part of the TGS domain and a WAP motif corresponding to amino acids 1403–1557 in Fig. 1E and was similarly amplified with primers 5'-AAGGATCCGAGGGTCAAAGGACAAAATTTC-3' (containing a BamHI site) and 5'-TGAAGCTTTCACTCGGGGATCAAACACT-3' (containing a HindIII site). pWap-Gpt was designed to encode a WAP motif corresponding to amino acids 1507–1557 in Fig. 1E and was similarly amplified with primers 5'-AAGGATCCGAGGCGAAGTTTCAGTAAACCT-3' (containing a BamHI site) and 5'-TGAAGCTTTCACTCGGGGATCAAACACT-3' (containing a HindIII site). The PCR products were digested with BamHI and HindIII and cloned into pRSET-B (Invitrogen). The constructs were confirmed by sequencing.

E. coli BL21(DE3) (Invitrogen) was transformed with the plasmid. Bacteria were grown overnight at 37 °C, and isopropyl -D-thiogalactopyranoside was added to a final concentration of 1 mM. Induction was continued at 30 °C for 4 h; bacteria were harvested by centrifugation at 3000 x g; and the cell pellet was suspended in 30 ml of buffer composed of 20 mM sodium phosphate (pH 7.8) and 0.5 M NaCl containing one tablet of Complete EDTA-free protease inhibitor mixture (Roche Applied Science). After sonication, the suspension was centrifuged at 4000 x g for 30 min at 4 °C. The soluble fraction was bound overnight at 4 °C to nickel-nitrilotriacetic acid-agarose (Qiagen Inc.). The agarose was loaded onto the column and washed three times with 5 bed volumes of wash buffer (20 mM sodium phosphate (pH 6.0) and 0.5 M NaCl). The column was subsequently washed in a solution composed of 6 volumes of isopropyl alcohol to remove bacterial endotoxins and 5 volumes of wash buffer containing 10 mM imidazole, and the protein was eluted with 3 ml of wash buffer containing stepwise concentrations of 50 and 300 mM imidazole. The recombinant proteins that were purified from E. coli transformed by pTgs-Gpt, pTgs-Wap-Gpt, and pWap-Gpt were designated r-Tgs-Gpt, r-Tgs-Wap-Gpt, and r-Wap-Gpt, respectively. The concentrations of the eluted proteins (r-Tgs-Gpt, r-Tgs-Wap-Gpt, and r-Wap-Gpt) were determined using the BCA protein assay kit (Pierce).

Refolding of recombinant proteins was performed according to the method described by Tsumoto et al. (35) and Tsumoto and co-workers (36). Briefly, the eluted proteins were diluted to a concentration of 7.5 µM with 50 mM Tris-HCl (pH 8.0), 0.2 M NaCl, and 6 M guanidine hydrochloride and were reduced overnight with 375 µM 2-mercaptoethanol. The eluted proteins were subsequently dialyzed against 50 mM Tris-HCl (pH 8.0), 0.2 M NaCl, and stepwise reducing concentrations of 6, 3, and 2 M guanidine hydrochloride and then against 50 mM Tris-HCl (pH 8.0), 0.2 M NaCl, stepwise reducing concentrations of 1 and 0.5 M guanidine hydrochloride, 375 µM oxidized glutathione, and 0.4 M L-arginine to allow refolding of the proteins. The folded proteins were finally dialyzed three times against PBS and then centrifuged at 12,000 x g for 20 min at 4 °C to remove unfolded or aggregated proteins. The supernatants were checked by SDS-PAGE under nonreducing conditions, and the concentrations of recombinant guinea pig trappin were determined using the BCA protein assay kit.

Preparation of Antiserum and Western Blotting—Antiserum to guinea pig trappin was prepared in a New Zealand White rabbit. Affinity-purified recombinant trappin (residues 1403–1557, r-Tgs-Wap-Gpt) was dialyzed in 0.15 M NaCl, emulsified with TiterMax Gold adjuvant (CytRx Co., Norcross, GA), and injected intramuscularly three times at 3-week intervals with 0.2 mg of the protein. A test bleed was taken at 1 week after the second injection to check for anti-guinea pig trappin antibody activity, and a final bleed (50 ml) was collected 1 week after the final injection. The final bleed was used in all subsequent studies. For Western blotting, guinea pig seminal vesicles were cut into small pieces (2–5 mm), suspended in 20 ml of PBS containing 5 mM EDTA and one tablet of Complete EDTA-free protease inhibitor mixture, and centrifuged at 8000 x g for 10 min. The supernatant was used as a luminal fluid-rich fraction. The proteins in the luminal fluid-rich fraction were separated by SDS-PAGE (7% (w/v) polyacrylamide) and electroblotted onto Immobilon-P membrane. After blocking with PBS containing 0.05% Tween 20 and 5% nonfat milk, the membrane was incubated with anti-guinea pig trappin antiserum (1:2000). The resulting immunocomplexes were visualized with horseradish peroxidase-conjugated secondary antibody using the ECL Plus Western blotting detection system (Amersham Biosciences).

Immunohistochemistry—Tissue specimens were obtained from 8-week-old Hartley guinea pigs and frozen after fixation by perfusion with a fixative containing 4% paraformaldehyde in PBS under anesthesia. For immunohistochemistry, 7-µm-thick sections were cut, fixed in 4% paraformaldehyde in PBS for 10 min, blocked with PBS containing 10% fetal bovine serum for 30 min, and then treated overnight at 4 °C with anti-guinea pig trappin antiserum (1:2000). The slides were treated with EnVision (Dako Corp., Glostrup, Denmark) for 1 h and then developed with a liquid diaminobenzidine substrate/chromogen system (Dako Corp.).

Purification of Guinea Pig Trappin from the Seminal Vesicle—For preparation of an immunoaffinity column, 4 ml of anti-guinea pig trappin antiserum was applied to a column containing 0.5 ml of protein A-Sepharose 4B (Amersham Biosciences) pre-equilibrated with PBS. The column was washed tree times with 5 ml of PBS containing 0.1% Tween 20, washed once with 5 ml of 0.1 M boric acid (pH 9.0) and 0.5 M NaCl, and washed once with 5 ml of 0.2 M triethanolamine (pH 8.2). The bound anti-guinea pig trappin antibody was cross-linked with 40 mM dimethyl pimelimidate in the same buffer at room temperature for 2 h. The column was incubated with 0.2 M ethanolamine (pH 8.2) at 4 °C for 1 h to block unoccupied dimethyl pimelimidate, washed in 0.1 M boric acid (pH 9.0) and 500 mM NaCl, washed in 0.1 M glycine HCl (pH 2.8), and immediately neutralized with 1 M Tris-HCl (pH 9.5). Luminal fluid-rich fractions (5 ml) of guinea pig seminal vesicles were applied to the column, which had pre-equilibrated with PBS containing 0.1% Tween 20. The column was washed three times in PBS containing 0.1% Tween 20. The bound proteins were eluted with 0.1 M glycine HCl (pH 2.8) and immediately neutralized with 1 M Tris-HCl (pH 9.5). The eluted proteins were separated by 7% SDS-PAGE and visualized by Coomassie Brilliant Blue staining. The purity of guinea pig trappin in the eluate was determined to be 93.5% by gel analysis using NIH Image software. The concentration of the eluate was determined using the BCA protein assay kit. About 1 ml (81 µg/ml) of affinity-purified guinea pig trappin was obtained.

Cell Culture—The human prostate cancer cell line LNCaP was obtained from American Type Culture Collection (Manassas, VA) and maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum, 100 units/ml penicillin, and 100 µg/ml streptomycin at 37 °C under 5% CO₂. LNCaP cells were plated at 0.5 x 10⁶ cells/100-mm dish with 10 ml of the medium. Two days after the plating, the medium was changed, and the cells were cultured another 2 days. The medium was then collected, treated with 5 mM EDTA, and used for prostate TGS assay.

Reverse Transcription-PCR—Total RNA was extracted from the LNCaP cells using an RNeasy mini kit (QIAGEN Inc). First-strand cDNA was synthesized from 5 µg of the total RNA using SuperScript II reverse transcriptase (Invitrogen) according to the manufacturer's instruction. Expression of human prostate transglutaminase, tissue transglutaminase, and glyceraldehyde-3-phosphate dehydrogenase was detected by reverse transcription-PCR using 0.2 µM primers: 5'-CAAGACCTACATCAACAGCC-3' and 5'-TGGGATTGGATGGTCTCAC-3' for prostate transglutaminase, 5'-GTACCTGCTCAACCTCAAC-3' and 5'-CCATTCTCACCTTAACTTCCTC-3' for tissue transglutaminase, and 5'-ACCCAGAAGACTGTGGATGG-3' and 5'-CCCTGTTGCTGTAGCCAAAT-3' for glyceraldehyde-3-phosphate dehydrogenase. The cDNA equivalent of 250 ng of total RNA was used as template for PCR using the Expand High Fidelity PCR system (Roche Applied Science). PCRs were performed for 20, 25, and 30 cycles under the following conditions: denaturation for 10 s at 94 °C, annealing for 20 s at 60 °C, and extension for 30 s at 72 °C with 1.5 mM MgCl₂. Mammalian expression vectors encoding human prostate transglutaminase and tissue transglutaminase were used as controls for PCR amplification efficiency of the primers.

Prostate TGS Assay—The activity of prostate transglutaminase for guinea pig trappin was analyzed by a modified method of Slaughter et al. (37) using 5-(biotinamido)pentylamine as an acyl acceptor substrate. The wells of a 96-well microtiter plate were coated with 50 µl of immunoaffinity-purified guinea pig trappin (20 µg/ml) and bovine serum albumin (BSA; control) as acyl donor substrates and stored overnight at 4 °C. The wells were blocked with 3% BSA in PBS for 1 h at 37 °C and washed three times in Tris-buffered saline (50 mM Tris-HCl (pH 7.5) and 0.15 M NaCl). The reaction was initiated by adding 90 µl of 20 mM CaCl₂, 1 mM dithiothreitol, and 0.5 mM 5-(biotinamido)pentylamine and 10 µl of the spent medium of LNCaP cells. After a 60-min incubation at 37 °C, the reaction was stopped with 50 mM EDTA. The wells were washed three times in PBS containing 0.1% Tween 20. Alkaline phosphatase-conjugated mouse anti-biotin monoclonal antibody (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) was diluted 1:3000 with PBS containing 0.1% Tween 20, added to the wells, and incubated at 37 °C for 1 h. The wells were washed three times in PBS containing 0.1% Tween 20. The bound anti-biotin antibodies were quantified from the absorbance at 405 nm after colorimetric reaction using 100 µl of one tablet of p-nitrophenyl phosphate (Sigma) dissolved in 5 ml of 1 M diethanolamine (pH 9.8) as a substrate.

Liver TGS Assay—Histidine-tagged green fluorescent protein (His₆-Xpress-GFP), which has recently been demonstrated by us (38) to be a good substrate for transglutaminase, was used to assess the potential of the TGS domain of guinea pig trappin to serve as a substrate also for type 2 transglutaminase. The rationale of this approach is that if the protein of interest is a substrate for transglutaminase, it can be covalently cross-linked to His₆-Xpress-GFP, and the cross-linked products can be detected by monitoring the sizes of fluorescent bands by SDS-PAGE. r-Tgs-Gpt (1 µg) prepared as described above was incubated with 0.5 µg of His₆-Xpress-GFP and 0.31 µg of guinea pig liver transglutaminase (Oriental Yeast Co., Tokyo) in transglutaminase assay buffer containing 50 mM Tris-HCl (pH 7.5), 1 mM dithiothreitol, and 5 mM CaCl₂ in the presence and absence of 50 mM EDTA, a transglutaminase inhibitor. After a 1-h incubation at 37 °C, the reaction was stopped with 50 mM EDTA, and the samples were separated by SDS-PAGE on a 15% gel without heat treatment. Fluorescent bands in the gel were directly visualized with an FLA-2000 bio-imaging analyzer (Fujifilm Japan).

Measurement of the Elastase Inhibitor Activity of Recombinant Guinea Pig Trappin—The elastase inhibitory activities of guinea pig trappin and elafin were assayed according to a modification of the method of Wiedow et al. (26) using the fluorogenic substrate N-methoxysuccinyl-L-alanyl-L-alanyl-L-prolyl-L-valine 4-methylcoumaryl-7-amide (Peptide Institute, Inc., Osaka, Japan). r-Wap-Gpt at final concentrations of 1.67, 0.84, 0.42, and 0.21 µM; r-Tgs-Wap-Gpt at final concentrations of 3.61, 1.80, 0.90, and 0.45 µM; r-Tgs-Gpt at final concentrations of 1.89, 0.95, 0.47, and 0.24 µM; and recombinant elafin at final concentrations of 0.92, 0.46, 0.23, and 0.12 µM (28) were preincubated with porcine pancreatic elastase (20 ng/ml). r-Wap-Gpt at final concentrations of 0.42, 0.21, 0.10, and 0.05 µM; r-Tgs-Wap-Gpt at final concentrations of 0.90, 0.45, 0.23, and 0.11 µM; r-Tgs-Gpt at final concentrations of 0.47, 0.24, 0.12, and 0.06 µM; and recombinant elafin at final concentrations of 23.0, 11.5, 5.7, and 2.9 nM were preincubated with human leukocyte elastase (2.5 milliunits/ml; Sigma) and 30 µl of assay buffer containing 50 mM HEPES (pH 7.5), 0.15 M NaCl, and 0.1% polyethylene glycol 6000 at 37 °C for 30 min. The remaining elastase activities were measured by adding 50 µl of assay buffer containing final concentrations of 100, 50, 33, and 25 µM substrate for porcine pancreatic elastase and final concentrations of 50, 33, 25, and 16.5 µM substrate for human leukocyte elastase. Controls were run using PBS in place of recombinant proteins. Fluorescence intensities were monitored at 465 nm with excitation at 380 nm for 30 min at 1-min intervals using a SPECTRAFluor-E microtiter plate reader (Tecan Switzerland AG, Maennedorf, Switzerland). Dixon analysis was used for determining the type of inhibition and for calculating K_i values (39).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Identification of Guinea Pig Trappin—As mentioned in the Introduction, we first thought that caltrin II found by Coronel et al. (16) in the guinea pig seminal vesicle may be a guinea pig homolog of trappins because its WAP motif sequence shares a considerably high degree of similarity with those of trappins (18). However, cloning and sequencing of its full-length cDNA indicated that caltrin II consists of only the WAP motif and lacks the TGS domain (28). A promising result was obtained when a Northern filter that had been used for determining the tissue distribution of the caltrin II message was reprobed with a hippopotamus trappin cDNA. The trappin probe hybridized to a band (>5 kb) that was clearly distinct from the caltrin II band (0.5 kb). Northern blot analysis also indicated that the seminal vesicle is the richest source of the candidate molecule. We therefore screened, using hippopotamus trappin cDNA as a probe, the guinea pig seminal vesicle cDNA library that had been constructed for the isolation of caltrin II cDNA and obtained several positive clones, including 12 clones corresponding to nucleotides 3184–5087 (three clones), 3504–5070 (one clone), 4054–5087 (one clone), 4315–5087 (two clones), 4432–5087 (three clones), and 4452–5087 (two clones) of guinea pig trappin cDNA (Fig. 1D). Their sequencing revealed the presence of both a repetitive sequence rich in Lys and Gln residues, a characteristic feature of the TGS domain of trappins, and the WAP motif, indicating that they are indeed guinea pig trappin cDNA clones. The isolated cDNA clones did not cover the entire coding sequence; even the longest clone lacked the 5'-untranslated region and an adjacent portion of the 5'-coding region (Fig. 1, C and D). We therefore performed CAP site RACE to determine the complete open reading frame. Ten clones of the longest CAP site RACE product (3.7 kb) were sequenced, and all contained 22 nucleotides of the 5'-untranslated region, the initiator Met codon, and a long open reading frame encoding the signal sequence and part of the TGS domain (Fig. 1, C and D). The amino acid sequence deduced from guinea pig trappin cDNA is shown in Fig. 1E.

Structure and Properties of Guinea Pig Trappin—Sequence analysis of the cDNA clones and CAP site RACE products (Fig. 1D) described above indicated that guinea pig trappin has the typical structure of trappin, consisting of a signal sequence, a potential TGS domain, and the WAP motif (Fig. 1, C and E). The potential TGS domain is composed of 1500 amino acids (Fig. 1E). This number is quite larger than the number of amino acids in the corresponding region of other trappins (30–150 residues). The majority of the TGS domain comprises short, weakly conserved repetitive sequences rich in Lys and Gln, as clearly seen by the distribution pattern of the Lys residues (Fig. 1E, indicated by dots) and the Gln residues (Fig. 1E, underlined). Many of the repeats contain the semiconserved hexapeptide sequence KGQXXX.

At the C terminus, there is a WAP motif sequence of 45 residues that contains eight conserved Cys residues. This C-terminal WAP motif shares significant sequence similarity with those of other trappin family members, suggesting that the WAP motif region of guinea pig trappin acts as a proteinase inhibitor. The alignment and phylogenetic analysis of the WAP motif of guinea pig trappin are shown in Fig. 6. The variable region (V2, part of the proteinase-binding site) of guinea pig trappin is different from those of other WAP motif proteins.

View larger version (45K): [in this window] [in a new window]   FIG. 1. Schematic representation of the guinea pig trappin gene and cDNA and its deduced sequences. A, genomic DNA clones isolated from a guinea pig genomic library and PCR products amplified from guinea pig genomic DNA were used for determination of the exon-intron organization of the guinea pig trappin gene. The 5'-flanking region and part of exon 2 were amplified by PCR. B, shown is the exon-intron organization of the guinea pig trappin gene. The intron is indicated by the horizontal line, and exons 1 and 2 are indicated by boxes. C, shown is the structure of guinea pig trappin cDNA. The presequence (Pre), transglutaminase (TGase) substrate domain, and WAP motif are indicated by white, light gray, and black boxes, respectively. The 5'- and 3'-noncoding regions are indicated by dark gray boxes. The nucleotide numbering is relative to Met (position +1) of the first in-frame ATG codon. 3' UTR, 3'-untranslated region. D, the largest cDNA clones isolated from a guinea pig seminal vesicle cDNA library were used to determine the cDNA structure. CAP site RACE products were used to determine the whole structure of guinea pig trappin cDNA and the transcription initiation site. E, the amino acid sequence of guinea pig trappin was deduced from its cDNA. Glutamine and lysine residues that can act as acyl donors and acceptors are shaded and indicated by underlining and dots, respectively. The translation initiation site (Met) identified by CAP site RACE is boxed and shaded. Asterisks indicate the positions of cysteines in the WAP motif. F, shown is the splice site mutation (GT GC) that led to inclusion of intronic sequence (white box) in exon 2 of the guinea pig trappin gene. Asterisks indicate the stop codons.

View larger version (100K): [in this window] [in a new window]   FIG. 2. Tissue distribution of guinea pig trappin mRNA. A, Northern blot analysis. Total RNA (5 µg/lane) from the indicated tissues was subjected to formaldehyde-agarose gel electrophoresis, blotted onto a nylon membrane, and hybridized with 32P-labeled guinea pig trappin (upper panel) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (lower panel) cDNA probes. B, in situ hybridization histochemistry of guinea pig trappin. Sections of guinea pig seminal vesicles were probed with antisense and sense digoxigenin-labeled riboprobes of guinea pig trappin. L, lumen; SE, secretory pseudostratified epithelium; SM, smooth muscle. Original magnifications were x10. Scale bar = 100 µm.

Gene Structure—Fig. 1B illustrates the structure of the guinea pig trappin gene, the entire sequence of which was determined by sequencing the two genomic DNA clones and two PCR products shown in Fig. 1A. The gene spans 11 kb and consists of two exons. Exon 1 codes for the 5'-noncoding region and the signal sequence, and exon 2 codes for the TGS domain, the WAP motif, and the 3'-noncoding region (Fig. 1B). This two-exon structure is different from the three-exon structure of the previously determined trappin genes of other animals, in which the third exon codes almost exclusively for the 3'-noncoding region (19, 20). Because the 3'-untranslated region of exon 2 shares significant sequence identity with intron 2 and exon 3 of other trappin genes (for example, GenBank^TM accession numbers AB161364 [GenBank] and D13156 [GenBank] ) (data not shown), the two-exon structure of the guinea pig trappin gene is therefore most likely due to skipping the splice site in the second intron in the guinea pig lineage (Fig. 1F).

Seminal Vesicle-restricted Expression—The size and tissue distribution of guinea pig trappin mRNA were determined by Northern blot analysis, which showed a dominant band of 5 kb found only in the seminal vesicle (Fig. 2A). To determine the type of cells expressing trappin in the guinea pig seminal vesicle, we performed in situ hybridization histochemistry using a digoxigenin-labeled riboprobe. The trappin transcripts were strongly detected in the secretory epithelium of the highly folded mucous membrane (Fig. 2B). A similar expression pattern was obtained at the protein level by immunohistochemistry (see Fig. 4B).

View larger version (30K): [in this window] [in a new window]   FIG. 3. Androgen-dependent expression of trappin mRNA in guinea pig seminal vesicles. A, total RNA (20 µg) isolated from seminal vesicles of castrated (lane 1), 17-estradiol-treated (lane 2), control (lane 3), and castrated plus testosterone-supplemented (lane 4) guinea pigs was analyzed by Northern blotting. The probes used were 32P-labeled guinea pig trappin cDNA (upper panel) and -actin cDNA (lower panel). B, the radioactivities of the bands were measured using a BAS 2000 imaging analyzer and normalized to the -actin level. Error bars indicate S.D. (n = 3).

Size of Guinea Pig Trappin Estimated by Western Blot Analysis—To determine the size of mature trappin in the guinea pig seminal vesicle, we first raised antiserum against recombinant guinea pig trappin and used it in Western blot analysis. When seminal vesicle fluid that had been collected by slicing the tissues was analyzed, major bands of 182 and 196 kDa were detected (Fig. 4A, lane 2, arrows) together with minor bands of 85–162 kDa (arrowheads). The prominent 182- and 196-kDa bands and the inconspicuous 85–162-kDa bands seen in the fluid-rich fraction may represent the secreted mature form and the proteolytically degraded products of trappin in the guinea pig seminal vesicle, respectively.

Functional Characterization of the TGS Domain—At the N terminus of mature trappin of the guinea pig seminal vesicle, there is a long Lys- and Gln-rich region, which is characteristic of TGSs (Fig. 1, C and E). Of nine transglutaminases, prostate transglutaminase is a prostate-specific enzyme and is secreted in prostate fluid (42). Thus, to determine whether guinea pig trappin can actually serve as a prostate TGS, we prepared (i) guinea pig trappin from the seminal vesicle and (ii) prostate transglutaminase from prostate cancer cells (LNCaP), which selectively express prostate-type transglutaminase (Supplemental Fig. S1). When these two preparations were mixed in the presence of 5-(biotinamido)pentylamine, an acyl acceptor substrate, significant incorporation of 5-(biotinamido)pentylamine into guinea pig trappin occurred in a Ca²⁺-dependent manner (Fig. 5), demonstrating that guinea pig trappin serves as a substrate for prostate transglutaminase.

To determine whether the TGS domain serves as a general substrate for transglutaminase, we produced a recombinant protein containing part of the TGS domain (r-Tgs-Gpt); incubated it with another fluorescent substrate (His₆-Xpress-GFP) in the presence of type-2 transglutaminase (prepared from guinea pig liver); and analyzed the reaction products by SDS-PAGE, followed by fluorography. As reported previously (38), His₆-Xpress-GFP served as a substrate and yielded products that were cross-linked intramolecularly (Supplemental Fig. S2, lane 1, asterisks) and intermolecularly (arrowheads). The TGS domain-containing recombinant protein r-Tgs-Gpt was found to be cross-linked to His₆-Xpress-GFP, yielding higher molecular mass products (Supplemental Fig. S2, lane 3, arrows) that were not seen in the case of His₆-Xpress-GFP alone (lane 1), demonstrating a broad specificity of the TGS domain of guinea pig trappin.

Elastase Inhibitor Activity of Guinea Pig Trappin—For functional studies, we prepared recombinant guinea pig trappin containing the WAP motif (r-Wap-Gpt), the TGS domain and WAP motif (r-Tgs-Wap-Gpt), and the TGS domain (r-Tgs-Gpt) and human trappin-2 containing the WAP motif (recombinant elafin) as described under "Experimental Procedures." Although less effective compared with native elafin, our recombinant elafin preparation strongly inhibited elastase (Table I), indicating that a sufficient amount of the recombinant protein was correctly refolded and disulfide-bonded intramolecularly. Table I summarizes the inhibitory effects of r-Wap-Gpt, r-Tgs-Wap-Gpt, r-Tgs-Gpt, and recombinant elafin against porcine pancreatic elastase and human leukocyte elastase. Three independent inhibition assays were performed using a fluorogenic substrate, and then the type of inhibition and inhibition constants (K_i) were determined by Dixon analysis. r-Wap-Gpt exerted an inhibitory activity against human leukocyte elastase (K_i = (2.33 ± 0.46) x 10^–7 M) (Table I), although its inhibitory activity was weaker compared with recombinant elafin (K_i = (1.27 ± 0.51) x 10^–8 M). Only a much weaker inhibitory activity was observed against porcine pancreatic elastase ((1.96 ± 0.96) x 10^–5 M); and furthermore, the potency was not significantly higher compared with the r-Tgs-Gpt control. In the case of human leukocyte elastase, weak activation by r-Tgs-Wap-Gpt and r-Tgs-Gpt was observed, as represented by the negative values in Table I, but this is probably due to their stabilization effect on leukocyte elastase. The type of inhibition by r-Wap-Gpt and recombinant elafin was competitive (data not shown).

View larger version (40K): [in this window] [in a new window]   FIG. 4. Western blot analysis and immunohistochemical localization of trappin in guinea pig seminal vesicles. A, Western blot analysis. The luminal fluid-rich fraction of the seminal vesicle that contained secreted proteins was immunostained with preimmune serum (lane 1) and anti-guinea pig trappin (Anti-gpTrappin) antiserum (lane 2). Arrows and arrowheads indicate major bands representing 182- and 196-kDa secreted forms and minor bands representing 85–162-kDa proteolytically degraded products of guinea pig trappin, respectively. B, immunohistochemical localization of guinea pig trappin. Sections of guinea pig seminal vesicles were probed with preimmune serum (panel a) and immune serum (panel b). L, lumen; SE, secretory pseudostratified epithelium; SM, smooth muscle. Original magnifications were x40. Scale bar = 50 µm.

Rapid Evolution of the TGS Domain-coding Region of the Guinea Pig Trappin Gene—The TGS domains of SVPs are known to consist of repeats of 24 amino acids (43). Comparison of the amino acid sequences of trappin and SVPs demonstrated that the 24-amino acid repeating units of SVP are composed of four tandem repeats of the semiconserved hexapeptide sequences of trappins (data not shown). Furthermore, we found that guinea pig trappin also contains the 24-amino acid repeats, which are spaced with the semi-conserved hexapeptide (Fig. 8A). The TGS domains of SVPs are composed mainly of 24-amino acid repeats (Fig. 8B).

View larger version (19K): [in this window] [in a new window]   FIG. 5. Prostate transglutaminase activities with guinea pig trappin as a substrate. A 96-well microtiter plate was coated with 20 µg/ml immunoaffinity-purified guinea pig trappin from seminal vesicles and BSA. The plate was blocked with 3% BSA and incubated with 5-(biotinamide)pentylamine and the spent medium of LNCaP cells with or without Ca2+ at 37 °C for 1 h, followed by detection of bound 5-(biotinamide)pentylamine with alkaline phosphatase-conjugated anti-biotin antibody. Bars represent the absorbance at 405 nm subtracted from the absorbance of BSA without Ca2+ condition. Error bars indicate S.D. (n = 4).

We next examined whether the substitution rate of the nucleotide sequences of the TGS domain-coding region is accelerated or not. The substitution rate of TGS domain-coding regions was calculated and compared with those of other regions. However, the resulting score was not significantly higher than those of other regions (Fig. 7F, dg). The rates for synonymous (dgS) and non-synonymous (dgN) substitutions within TGS domain-coding regions were found to be only slightly different, suggesting that there is no strong positive selection. These results demonstrate that the rapid evolution of the TGS domains is due to the individually different elongation patterns of repeating units, but not due to hypermutation by positive selection.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Using hippopotamus trappin cDNA as a probe, we isolated a clone from a guinea pig seminal vesicle cDNA library predicting a protein of 1557 amino acids (165 kDa). In terms of the domain structure and conserved sequence elements, there are several features that show it to be a member of the trappin family. 1) Its precursor is composed of a signal sequence, a TGS domain, and a WAP motif; 2) the N-terminal TGS domain is highly repetitive in nature and is rich in Gln and Lys residues; and 3) the C-terminal WAP motif consists of a conserved pattern of eight Cys residues, and its recombinant protein has elastase inhibitory activity.

Noteworthy are the length and sequence characteristics of the TGS domain, the potential of which to serve as a substrate for prostate transglutaminase is demonstrated by cross-linking experiments. This similarity strongly supports the previous suggestion that the TGS domains of trappins may share a common ancestry with the REST genes, which was proposed by Hagstrom et al. (22) based on the common exon-intron organization (three-exon structure) of the human trappin and REST genes and significant similarities (60%) in their nucleotide sequences, including the 5'- and 3'-flanking regions, introns, and third noncoding exons. However, the similarity of the second coding exon was found to be relatively low in a previous comparison of the human and rat sequences of seminal vesicle secretory proteins (44). Our sequence analyses of the guinea pig trappin and REST genes revealed that the nucleotide similarity of the 5'- and 3'-flanking regions, introns, TGS domains, and third noncoding exons is relatively high, providing direct evidence for their evolutionary relationship, whereas the similarity of the WAP motifs is relatively low. Furthermore, the TGS domain of guinea pig trappin has rapidly elongated, so the similarity of the entire region of the TGS domain becomes much lower than the similarity of the 5'- and 3'-flanking regions, introns, and third noncoding exons, even though the similarity of part of the TGS domain is relatively high. The number of exons in the guinea pig trappin gene was reduced from three to two because of mutation of the splicing donor site of intron 2 (Fig. 1F). These characteristics of the guinea pig trappin gene fit well with those of the REST genes, strongly indicating that guinea pig trappin is a member of REST gene family.

View larger version (58K): [in this window] [in a new window]   FIG. 6. Alignment and phylogenetic analysis of the WAP motifs of trappins, caltrin II, and SLPI. A, comparison of the amino acid sequences of the WAP motifs of guinea pig trappin and other WAP motif proteins. Conserved and semiconserved residues are indicated by dark and light gray shading, respectively. Eight conserved Cys residues constituting the WAP motif signature sequence are shaded in black and indicated by asterisks. The variable V2 region, which is thought to determine the specificity of WAP motifs, is boxed. SLPI consists of two WAP motifs; the N- and C-terminal motifs are indicated as 1st and 2nd, respectively. B, phylogenetic analysis of WAP motifs of trappins, caltrin II, and SLPI. A phylogenetic tree was constructed by comparing the nucleotide sequences encoding WAP motifs. Bars indicate 10% replacement of an amino acid per site.

Sequence analysis of the guinea pig trappin gene offers an explanation for evolution of the guinea pig REST genes (Fig. 9, A–F). An ancestral trappin gene became fused with an ancestral semenogelin gene at the end of exon 2 of the semenogelin gene, possibly by gene conversion, yielding an intermediate form (Fig. 9B). The semenogelin-derived exon 2 and trappin-derived exon 1 lost their functions and became pseudoexons as illustrated by the X'd gray boxes in Fig. 9C. Following duplication of this prototypic gene, three different genes were created by the following events: (i) exon and intron skipping and elongation of exon 2, yielding the guinea pig trappin gene (Fig. 9D); (ii) loss of the WAP motif and elongation of the TGS domain, yielding the guinea pig SVP-1/-3/-4 gene (Fig. 9E); and (iii) loss of the TGS domain, yielding guinea pig caltrin II (Fig. 9F). Thus, trappin and trappin-derived genes acquired diversity in their structures and functions by deletion or elongation of the functional domains.

View larger version (62K): [in this window] [in a new window]   FIG. 7. Characterization of the guinea pig trappin gene by comparison with other related genes. A–D, Harr plot analyses of the guinea pig trappin gene in comparison with the guinea pig caltrin II gene (28), the guinea pig SVP-1/-3/-4 gene (22), the human trappin-2 gene (19), and the human semenogelin I gene (23). The SVP-1/-3/-4 gene encodes a precursor from which SVP-1, SVP-3, and SVP-4 derive. A diagram of the exon-intron organization of each gene is shown along the axes. The homologous regions of the TGS domain are circled (g). Other homologous regions are also marked (a–e and h–j). E, schematic representations of the guinea pig trappin gene and the results of their structural comparison. Exons 1 and 2 are indicated by white boxes. The intron is shown by the horizontal line. Bars indicate regions homologous to the human gene for semenogelin I (a–c) or trappin-2 (d–j). The sequence position of each region is given. F, nucleotide substitution rates of guinea pig (gp) trappin and other related genes. Jukes-Cantor distances were separately computed for each region (da–dj) (34). Synonymous (dgS and dhS) and nonsynonymous (dgN and dhN) distances of the TGS domain and WAP motif-coding regions were computed by the method of Nei and Gojobori (52). Regions that show relatively faster nucleotide substitution rates are shaded. hu, human; semg 1, semenogelin I. G–I, phylogenetic analyses of exons 1 and 3 and TGS domain-coding regions of REST genes (44). The GenBankTM accession numbers and the sequence positions used for the analyses are indicated in parentheses. The numbers for interior branches refer to bootstrap values for 5000 replications. In I, the TGS-coding regions of the human trappin-2 and pig trappin-3 genes are much shorter than those of other genes; thus, bootstrap values are low because the presence of short sequence decreases the statistical significance of the analysis. The values for some branches increased when we eliminated those sequences from the analysis (indicated by italic numbers). 3'-UTR, 3'-untranslated region.

View larger version (47K): [in this window] [in a new window]   FIG. 8. Evolution of the TGS domain of the guinea pig trappin gene family. A and B, schematic representations of the TGS domains of guinea pig trappin and SVP-1/-3/-4 gene products. The 24-amino acid repeats are indicated by black boxes. 6-Amino acid repeats that do not conform to the 24-amino acid repeating units are indicated by dots. The positions of the first amino acid of the 24-amino acid repeats are indicated. t and s indicate the repeating units of the trappin and SVP-1/-3/-4 genes, respectively. Pre, presequence. C, phylogenetic relationship among the coding regions for the 24-amino acid repeating units of the guinea pig trappin and SVP-1/-3/-4 genes. Five branches are indicated by a–f. Branches representing the repeating units of the guinea pig trappin and SVP-1/-3/-4 genes are indicated by thin and thick lines, respectively.

We wanted to clone mouse trappin to create knockout mice, but failed. Data base searches of whole genome sequences revealed that the trappin gene is not present in rat and mouse, although they have genes for other TGS (semenogelin family) and WAP (SLPI, WDNM1, and HE4) proteins. The data base searches performed were (i) BLAST searches using nucleotide sequences of trappin genes as queries for whole genome sequence and expressed sequence tag data bases of mouse and rat (BLASTN) and (ii) BLAST searches using amino acid sequences of trappins as queries for whole genome sequence and expressed sequence tag translated data bases of mouse and rat (TBLASTN). In human chromosome 20q12–13, the trappin gene is clustered with genes for many WAP motif proteins (SLPI, HE4, eppin, WAP2, WAP12, etc.), TGS proteins (semenogelins), and transglutaminases (18, 49). In mouse chromosome 2 and rat chromosome 3, the orthologs of those genes are also clustered, but no trappin genes are found within those loci (data not shown). Thus, we conclude that the trappin gene is deleted in rat and mouse. This deletion may be due to rapid evolution of the trappin gene in the Rodentia lineage, viz. by frequent gene conversions or mutations, the trappin gene may have been deleted or lost its function in rodents (Fig. 10). Interestingly, however, transgenic mice expressing human trappin-2 have been demonstrated to show superior anti-inflammatory activities against infection with encephalomyocarditis virus compared with wild-type mice (50). The evolutionary information obtained in a previous study (1) and in this study suggests that trappin genes were multiplicated independently in pig, bovine, and guinea pig lineages (Fig. 10) and acquired a variety of functions.

Northern and Western blot analyses revealed high level expression of guinea pig trappin in seminal vesicles, which produce viscous alkaline fluid that helps to neutralize the acidity of the vagina and contains fructose, an energy source for motile sperm, and the REST gene products or the clotting proteins that coagulate after ejaculation by the action of prostate-derived and androgen-regulated transglutaminase (42). The guinea pig trappin identified here has a long TGS domain like the clotting proteins. It is therefore considered to be a clotting protein that acquired a proteinase inhibitor domain (WAP motif) as a C-terminal extension. The trappin molecules covalently incorporated into the vaginal clots may serve as inhibitors of bacterial and acrosomal proteinases. This possible mode of action is reminiscent of that of pig trappin-1, which is secreted from small intestinal crypt cells into the circulation and thought to be incorporated into the hemostatic plug following injury (46).

View larger version (27K): [in this window] [in a new window]   FIG. 9. Expected model for evolution of the guinea pig trappin gene family. In the guinea pig lineage, an ancestral trappin gene became fused with an ancestral semenogelin gene possibly by gene conversion (A), yielding an intermediate form (B). The semenogelin-derived exon 2 and trappin-derived exon 1 lost their functions and became pseudoexons (C). Following duplication of the prototypic gene, three genes were generated and evolved by loss of domain-coding regions, elongation of TGS domains, and inactivation of intron 2 (D–F). Exons are boxed and numbered. TGS domains and WAP motifs are indicated by hatched and black boxes, respectively. Horizontal dotted and solid lines show semenogelin- and trappin-derived introns, respectively. Pseudoexons are indicated by gray boxes that have been crossed through with an X. Zigzagged lines indicate androgen-driven promoter regions.

View larger version (24K): [in this window] [in a new window]   FIG. 10. Birth and deletion of the trappin gene. Trappin genes are indicated with a phylogenetic tree of the examined animals. The phylogenetic tree of the mammalian species was constructed according to Murphy et al. (53). The birth and deletion of the trappin gene are shown by arrows and an arrowhead, respectively.