Striatum- and testis-specific phosphodiesterase PDE10A

Isolation and characterization of a rat PDE10A

Authors


K. Omori, Discovery Research Laboratory, Tanabe Seiyaku Co. Ltd, 2-50, Kawagishi 2-chome, Toda, Saitama 335-8505, Japan. Fax: + 81 48 4338159, E-mail: k-omori@tanabe.co.jp

Abstract

PDE10A, a phosphodiesterase (PDE) exhibiting properties of a cAMP PDE and a cAMP-inhibited cGMP PDE, was cloned and investigated in detail in rats. PDE10A transcripts were abundant in the brain and testis. In situ hybridization analysis using a PDE10A riboprobe demonstrated the presence of PDE10A transcripts in the neurons of the striatum and the olfactory tubercle regions of the brain. Rat PDE10A cDNAs were isolated from a brain cDNA library and nucleotide sequence analysis revealed several N-terminal variants. The deduced amino-acid sequence of one of the major variant forms contained 794 amino acids, and it was 96% identical to that of the human PDE10A2. The other major form has a distinct N-terminal sequence that is not found in humans. PDE10A was partially purified from rat striatum and testis, and characterized with respect to Km, inhibitor sensitivity and immunoreactivity to an anti-PDE10A serum. These findings indicate that PDE10A functions in these tissues.

Abbreviations
PDE(s)

cyclic nucleotide phosphodiesterase(s)

iBuMeXan

3-isobutyl-1-methylxanthine

EST

expressed sequence tag(s)

Cyclic nucleotide phosphodiesterases (PDEs) catalyze the hydrolysis of cyclic nucleotides such as cAMP and cGMP, which regulate many functions in various tissues as second messengers [1–4]. Many kinds of PDEs participate in the metabolism of cyclic nucleotides. Based on their amino-acid sequence homology, biochemical properties and inhibitor profiles, 10 PDE families have been recognized in mammalian tissues [5,6]: PDE1, Ca2+/calmodulin-dependent, hydrolyzing both cAMP and cGMP; PDE2, hydrolyzes cAMP and cGMP, and the activity is stimulated by cGMP; PDE3, a cGMP-inhibited PDE; PDE4, cAMP-specific and rolipram-sensitive; PDE5, a cGMP-binding and cGMP-specific PDE; PDE6, a photoreceptor cGMP PDE; PDE7, cAMP-specific and rolipram-insensitive; PDE8 and PDE9, cAMP- and cGMP-specific PDEs, respectively, and are insensitive to a nonspecific PDE inhibitor, 3-isobutyl-1-methylxanthine (iBuMeXan) [7–10]; and PDE10, a new member was recently added to the PDE family [11].

We first reported cDNA cloning and characterization of a novel human PDE, PDE10A, which catalyzes the hydrolysis of both cAMP and cGMP [11]. The cDNA was isolated from a human fetal lung library by an approach using bioinformatics. The deduced amino-acid sequence contains 779 amino acids, including a putative cGMP-binding sequence in the N-terminal portion of the molecule, and a catalytic domain that is 16–47% identical in amino-acid sequence to those of other PDE families. PDE10A hydrolyzes cAMP and cGMP with Km values of 0.26 and 7.2 µm, respectively, and has a Vmax for cGMP almost twice that with cAMP. cGMP inhibits hydrolysis of cAMP, and cAMP inhibits cGMP hydrolysis with IC50 values of 14 and 0.39 µm, respectively. Thus, PDE10A exhibits properties of a cAMP PDE and a cAMP-inhibited cGMP PDE. PDE10A transcripts were particularly abundant in the putamen and caudate nucleus regions of the brain and in the thyroid and testis, and in much lower amounts in other tissues.

Genes encoding PDE may be grouped into subfamilies. Within each subfamily there is a high degree of sequence identity. In addition, each subfamily may generate further sequence diversity by alternative RNA splicing. For example the largest subfamily, PDE4, consists of four genes and a large number of alternative splice variants [6,12]. In many cases, different gene products and alternative splice variants in each PDE family show different expression patterns in tissues and different subcellular localization [6,13–21]. Some of alternatively spliced variant forms of PDE were reported to differ in their regulation by kinases and associated proteins [21,22]. Some PDE families are composed of splice variants having distinct N-terminal domains, which are associated with some cofactors such as kinases, and by which their tissue expression patterns and subcellular localization are specified [22–24]. We have recently described splicing variants of PDE5A that are subjected to distinct gene regulation [25], and we have recently isolated a cDNA encoding N-terminal variant form of PDE10A, which we named PDE10A2 [26].

The approach using bioinformatics enabled us to isolate unidentified putative PDE cDNAs. However, no report was made demonstrating the presence of the PDE activity in tissues. We have now cloned PDE10A cDNAs from a rat cDNA library, revealing the presence of several N-terminal variants. An anti-PDE10A polyclonal serum was raised using a synthetic peptide containing a part of the PDE10A amino-acid sequence. We describe the tissue-specific expression patterns of rat PDE10A transcripts and report their cellular localization, which may lead to a better understanding of their physiological role. We also investigated the PDE10A activity in animal tissues in order to compare the activity with those of other PDEs.

Materials and methods

Materials

Restriction endonucleases and DNA-modifying enzymes were obtained from Takara Shuzo (Kyoto, Japan). [α-32P]dCTP and Hybond-N+ nylon membrane were from Amersham Pharmacia Biotech. The plasmid pBluescript II SK(+) and the rat brain cDNA library were from Stratagene. The TA-cloning vector pGEM-T easy was from Promega. The expression vector pFLAG–CMV-2 and the anti-FLAG M5 monoclonal antibody were obtained from Eastman Kodak and Sigma, respectively. Rat brain cDNA and Rat Multiple Tissue Northern Blots were purchased from Clontech. iBuMeXan, dipyridamole and zaprinast were from Sigma. E4021, milrinone and rolipram were synthesized at Tanabe Seiyaku Co. Ltd, Osaka, Japan.

General methods

Nucleotide sequences were determined by the dideoxy-chain-termination method using a BigDye Terminator Cycle Sequencing Reaction kit and an automated DNA sequencer ABI PRISM™ 310 (PE Applied Biosystems). The computer programs genetyx (Software Development, Tokyo, Japan) was used to analyze nucleotide and amino-acid sequence data.

Isolation and analysis of cDNA sequences

The cDNA sequence of the human PDE10A coding region [11] was used as a query to search the databases of expressed sequence tags (EST) with Basic Local Alignment Search Tool (blast) [27]. One EST sequence showing extremely high sequence similarity to the query sequence was obtained by the search. The clone (H32734) was demonstrated to possess a sequence corresponding to a part of the rat PDE10A cDNA.

Screening of cDNA library

To isolate the DNA fragment encoding the H32734 sequence, PCR was carried out with the primer set R51 (5′-GCGCTCTTCCAGGTGGACC-3′) and R31 (5′-CATCTTGAAGTTGTTCTC-3′) and rat brain cDNA as template. PCR was carried out through 30 cycles of denaturation at 94 °C for 1 min, annealing at 55 °C for 1 min and extension at 72 °C for 1 min. The PCR product was cloned into TA-cloning vector pGEM-T easy, resulting in pGEM-R10A, and the nucleotide sequence was confirmed. The inserted DNA fragment (H32734 probe) was radiolabeled using a 32P random primer labeling kit (Takara Shuzo), and then used as a probe for plaque hybridization. Library screening was performed according to a standard procedure [11]. Fifteen plaques that hybridized to the probe were purified by two rounds of replating and rescreening. The inserted DNAs were isolated as plasmids according to the manufacturer’s instructions. The plasmid DNAs were subjected to dot-blot analysis using the 32P-labeled 0.4-kb NotI–KpnI fragment (HNP1 probe, N-terminal part), and the 32P-labeled 0.3-kb NcoI–EcoRI fragment (HCP1 probe, C-terminal part) of the human PDE10A cDNA encoded by pBlue-PDE10A [11]. Hybridization was carried out using Hybond-N+ membrane and the 32P-labeled probes in hybridization buffer at 55 °C for 16 h. The blots were washed in 1 × NaCl/Cit, 0.5% SDS at 55 °C for 10 min and the filters were exposed to X-ray film at −70 °C for 4 h. The plasmids were digested with the restriction endonuclease EcoRI to confirm the size of the inserted DNA. After subcloning, the entire nucleotide sequences were determined.

Reverse transcriptase-PCR analysis

To investigate the presence of the mRNAs coding for rat PDE10A N-terminal variants, PCR was carried out with the rat brain cDNA and primer sets shown below. The 5′ primers (5′-CTGGCACTTCGAAACCGC-3′, 5′-ATCCCAGGTCAACATTGG-3′, 5′-AAGAAAACCCCCAAACCC-3′ and 5′-GGCAATGAAGCAGAAAGG-3′) for RNPDE10A3, RNPDE10A4, RNPDE10A5 and RNPDE10A6, respectively, and the 3′ primer 5′-TTGAGTCAGTTGCTAGGC-3′ were used for first amplification. For second amplification, the 5′ primers (5′-ACCGCACAGACTCTCGGGGG-3′, 5′-GAGACCCCAGAGCATTGG-3′, 5′-CCCCAGGCCCTGGGCAGA-3′ and 5′-GGAGTTTTGTACTGGAGG-3′) for RNPDE10A3, RNPDE10A4, RNPDE10A5 and RNPDE10A6, respectively, and the 3′ primer 5′-GGCAGACATCAGGTCAGG-3′ were employed. Each PCR was carried out in 50 µL using Ex-Taq (Takara) with conditions of 30 cycles of denaturation at 94 °C for 30 s, annealing at 55 °C for 30 s and extension at 72 °C for 2 min. For second amplification, one-tenth of the first PCR product was used as a template with the same conditions of the first amplification. The PCR products were subjected to electrophoresis in a 0.7% agarose gel and then transferred onto Hybond-N+ nylon membrane. Hybridization was performed using the 32P-labeled H32734 probe (nucleotides 1172–1519) as described above, and the membranes were exposed to X-ray film at − 70 °C for 1 h.

Northern blot analysis

Rat Multiple Tissue Northern Blots were used for investigating tissue-specific expression pattern of rat PDE10A. To examine PDE10A expression in a rat brain, various regions in the brain were excised as previously reported [28]. Total RNAs were purified using Isogen solution (Nippon Gene, Japan). Total RNA (10 µg per lane) was subjected to electrophoresis in 1% agarose/0.66 m formaldehyde gels. Gels were stained with ethidium bromide and photographed. The RNA fractions were transferred onto Hybond-N+ nylon membrane and fixed by UV irradiation cross-linker (Stratalinker, Stratagene). The blots were hybridized with the 32P-labeled H32734 probe. Hybridization was performed in 50% formamide, 4 × NaCl/Cit, 0.5% SDS, 5 × Denhardt’s and 100 µg·mL−1 salmon sperm DNA at 42 °C for 16 h. All blots were washed finally in 0.2 × NaCl/Cit and 0.1% SDS at 60 °C for 1 h. The membranes were exposed to X-ray film at −70 °C for 2 weeks with intensifying screen.

In situ hybridization analysis

RNA riboprobes for in situ hybridization of rat PDE10A were prepared by in vitro transcription with digoxigenin-UTP using a DIG RNA Labeling kit (Boehringer Mannheim) according to the manufacturer’s instructions. The plasmid pGEM-R10A was linearized with SphI for the antisense probe and SacI for the sense probe. The antisense and sense probes were synthesized from 1 µg of the linearized plasmid templates with T7 polymerase and SP6 polymerase, respectively.

Whole brains excised from 10-week-old male rats were frozen and stored at −70 °C. Frozen sections were thaw-mounted on silane-coated slides, soaked with 4% formaldehyde, 10 µg·mL−1 proteinase K and 0.1 m triethanolamine/0.25% anhydrous acetic acid and then dehydrated in graded alcohols. After prehybridization at 50 °C for 30 min in 800 µL of hybridization solution (20 mm Tris/HCl, pH 8.0, 300 mm NaCl, 10% sodium dextran sulfate, 50% formamide, 0.2% N-lauroylsarcosine, 100 µg·mL−1 salmon sperm DNA and 1 × Denhardt’s solution), hybridization was performed using digoxigenin-labeled riboprobes in 800 µL hybridization solution at 50 °C for 16 h in a humid chamber. The sections were washed with a high-stringency buffer (50% formamide, 300 mm NaCl, 30 mm sodium citrate) at 60 °C for 30 min, treated with RNaseA (1 µg·mL−1) for 10 min at 37 °C, and then washed with the high-stringency buffer again. They were incubated with polyclonal antibodies reactive to digoxigenin diluted 1: 500 in buffer A (100 mm Tris/HCl, pH 7.5, 150 mm NaCl) containing 1.5% blocking reagent (Boehringer Mannheim) for 5 h at room temperature, and then extensively washed with buffer A. The sections were incubated with a freshly prepared color-substrate solution (100 mm Tris/HCl, pH 9.5, 100 mm NaCl, 50 mm MgCl2) containing NitroBlue Tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate, and color developed in the absence of light at room temperature for 2 days.

High performance anion-exchange chromatography

Striata and testes of 10-week-old male rats were homogenized in 10 mL of ice-cold homogenization buffer (20 mm Tris/HCl, pH 7.4, 2 mm magnesium acetate, 0.3 mm calcium chloride, 1 mm dithiothreitol, 40 µm leupeptin, 1.3 mm benzamidine, 0.2 mm phenylmethanesulfonyl fluoride and 1 mm NaN3) in a Polytron homogenizer for 1 min at a medium speed. The homogenates were centrifuged at 6000 g for 20 min, and then the supernatants were recentrifuged at 100 000 g for 60 min. The cytosolic fractions isolated were applied to a HiTrap-Q Sepharose column (Amersham Pharmacia Biotech) equilibrated in elution buffer (20 mm Tris/HCl, pH 7.4, 1 mm dithiothreitol, 1 mm calcium chloride, 2 µm leupeptin and 5 mm benzamidine). The column was washed with 10 mL of elution buffer, and the proteins were then eluted from the column by running a linear NaCl gradient (0–400 mm, 40 mL and 400–1000 mm, 10 mL) in elution buffer. Fractions (1 mL each) were collected on ice and assayed for PDE activity.

PDE assay

The PDE assay was performed as described previously [11].

Generation of antiserum

A rabbit polyclonal serum was raised against the poly l-lysine-based multiple antigen peptide complex containing a 19-amino-acid synthetic peptide of the sequence DHKNKELYSDLFDIGEEKE (residues 287–305 of rat PDE10A2), which is derived from a portion in common with rat and human PDE10As. The multiple antigen peptide complex was synthesized by the peptide synthesizer PSSM-8 (Shimadzu, Japan) using multiple antigen peptide resin TAKO8-WTGS (Shimadzu), and then mixed with Freund’s complete adjuvant (Sigma) for the first immunization and with Freund’s incomplete adjuvant when boosted. After immunizing two Japanese White rabbits (Kitayama Laboratories, Japan) three times, antiserum was collected [28a]. All experimental protocols for animal studies were approved by the Animal Care and Use Committee of Tanabe Seiyaku Co. Ltd., Japan.

Immunoblotting

Samples were subjected to 7.5% SDS/PAGE, and transferred onto poly(vinylidene difluoride) membranes (Millipore). After blocking with 4% Block Ace (Snow Brand Milk Products, Japan) overnight, the blots were incubated with anti-PDE10A polyclonal serum, biotin-conjugated anti-(rabbit IgG) Ig, and then avidin–biotin peroxidase. Alternatively, the blots were incubated with anti-FLAG M5 monoclonal antibody followed by peroxidase-conjugated anti-(mouse IgG) Ig. Visualization of bands was performed by the ECL Western blotting detection reagents (Amersham Pharmacia Biotech).

Results

cDNA cloning of a rat PDE10A

In order to isolate the rat PDE10A cDNA, we searched the EST database using the nucleotide sequence of the cDNA encoding human PDE10A as the query sequence. A homologous sequence (H32734), which is in a PC12 cell cDNA library, was found in the EST database. The DNA fragment having the H32734 sequence was obtained by PCR with primers designed from the sequence and rat brain cDNA as a template. A specifically amplified DNA fragment of approximately 350 bp was subcloned, sequenced, and confirmed to be homologous to the human PDE10A cDNA sequence and identical to the H32734 sequence. Using the DNA fragment as a probe (H32734 probe), we screened a rat brain cDNA library to isolate a full-length cDNA. More than 50 positive signals were observed in 9 × 105 plaques. Of 15 clones isolated, five clones (clones 2, 4, 8, 10 and 17) hybridized with both the probes HNP1 and HCP1 corresponding to N-terminal and C-terminal sequences of the human PDE10A, respectively. Clones 9, 11 and 12 hybridized to HNP1, and clones 6, 13, 16 and 18 hybridized to HCP1. After subcloning, the entire nucleotide sequences of the clones 4 and 8 were determined. The nucleotide sequence of the full-length cDNA (3427 bp) is shown in Fig. 1. The H32734 sequence was found in the central region (nucleotides 1172–1520). An ORF of 2385 nucleotides starting from the first initiation codon ATG to the termination codon TGA (nucleotides 281–2665) was present. The complete ORF encoded a protein of 794 amino acids with a predicted molecular mass of 90 kDa. The N-terminal sequence perfectly matched that of human PDE10A2 but not PDE10A1 (Fig. 2). The entire amino-acid sequence was 96% identical to the human PDE10A2. The putative polyadenylation site AATAAA [29], was found in the 3′ noncoding region of the cDNA (nucleotides 3378–3383). From the analysis, it was concluded that the ORF encodes the rat PDE10A, designated RNPDE10A2 [30].

Figure 1.

Figure 1.

Nucleotide and deduced amino-acid sequences of the rat PDE10A cDNA. The ORF is boxed. The deduced amino-acid sequences are shown in one-letter code below the nucleotide sequence. The termination codon at the end of the ORF is represented by an asterisk. The underlined AATAAA sequence represents a putative polyadenylation signal. EcoRI sites (GAATTC) derived from the vector are present at each end.

Figure 2.

Figure 2.

Alignment of rat and human PDE10A2 amino-acid sequences. Amino acid sequences are shown in one-letter code. Identical amino-acid residues are boxed.

Presence of novel N-terminal variants in rat PDE10A

Clones 2 and 10 were suspected to encode the rat PDE10A2 from a partial sequence analysis, whereas clone 17 contained an ORF similar to rat PDE10A2 but with a distinct N-terminal sequence (Fig. 3), suggesting the presence of an N-terminal variant in rat PDE10A (RNPDE10A3). Clone 9, encoding the ORF truncated at the C-terminal region, also possessed the N-terminal sequence of PDE10A3. In order to analyse variations at the N-terminus of rat PDE10A, 15 additional cDNA clones were isolated, and were hybridized with both H32734 and HNP1 probes. The partial nucleotide sequences of the clones were determined and the results are summarized in Fig. 3. Five of 15 clones contained the N-terminal sequence identical to PDE10A2. Seven clones possessed the N-terminal sequence identical to PDE10A3. Three clones encoded the rat PDE10A but with distinct N-terminal sequences, designated PDE10A4, PDE10A5 and PDE10A6, respectively (Fig. 3). Reverse transcriptase-PCR using specific primer sets designed from these sequences generated specific products of approximately 2.4 kb, indicating the presence of these variants in transcripts of rat brain (data not shown).

Figure 3.

Figure 3.

Comparison of the N-terminal amino-acid sequences of rat and human PDE10A splicing variants. Open boxes represent the PDE10A sequences in common with PDE10A variants. Predicted amino-acid sequences of PDE10A variants are shown from Met1 in one-letter code. An asterisk indicates a termination codon TGA in the sequence.

In humans, a variant form (PDE10A1) having an insertion of 128 nucleotides in the N-terminal region of the PDE10A2 has been reported [11]. PDE10A1 and PDE10A2 transcripts were detected as DNA fragments with different sizes by PCR [26]. In order to determine the presence of a rat PDE10A1 variant, PCR was performed using a primer set from the 5′ noncoding region and N-terminal region of rat PDE10A2 with the rat brain cDNA as a template. However, only the DNA fragment of approximately 300 bp corresponding to the N-terminal sequence of rat PDE10A2 was amplified by PCR (data not shown). In rats, no cDNAs having an insertion similar to that of the human PDE10A1 were detected. Another difference between rat and human PDE10A cDNAs was C-terminal sequence diversity. The nucleotide sequence analysis of eight clones hybridizing with both the H32734 and RCP1 probes revealed that all the clones tested carried a C-terminal sequence identical to the ORF described above.

Tissue distribution of PDE10A transcripts

Northern blot analysis of multiple rat tissues was performed using the 32P-labeled H32734 probe in order to show the distribution of the PDE10A transcripts (Fig. 4A). PDE10A transcripts were particularly abundant in the brain and testis. A band of approximately 9.5 kb was detected in the brain. In the testis, a band of ≈ 3.5 kb was observed. We investigated distribution of PDE10A transcripts in the brain in detail (Fig. 4B). Total RNAs were prepared from various regions in the rat brain such as cerebrum cortex, cerebellum, hippocampus, striatum, olfactory bulb, midbrain and brain stem. Northern blot analysis demonstrated that PDE10A transcripts were found at high levels in the striatum.

Figure 4.

Figure 4.

Expression of rat PDE10A transcripts in various tissues. Northern blot analysis was carried out with the 32P-labeled H32734 probe. (A) Rat Multiple Tissue Northern Blots loaded poly(A)+ RNA from rat tissues (Clontech) were subjected to Northern blot analysis and exposed to X-ray film for 5 days. The sizes (in kb) and positions of mRNA size markers are as shown. (B) Total RNAs (10 µg per lane) from various regions of 10-week-old rat brains were separated in a 1% agarose/0.66 m formamide gel. The gel was stained with ethidium bromide (lower panel). The RNAs were transferred onto Hybond-N+ nylon membrane. The blots were subjected to Northern blot analysis and exposed to X-ray film for 2 weeks (upper panel). The position of the 28S rRNAs is shown.

In situ hybridization of PDE10A mRNA in the brain

To examine the cellular localization of the PDE10A mRNA in the brain, we performed in situ hybridization analysis. Figure 5A,C shows that positive staining by an antisense riboprobe was observed in the striatal neurons; however, not all of the striatal neurons were stained (Fig. 5D). The neurons in the olfactory tubercle were also stained (Fig. 5E). PDE10A mRNA was not detected in other parts of brain by in situ hybridization. These findings agreed well with the results obtained from Northern blot analysis of the rat brain shown above (Fig. 4B). No specific staining was observed when the sections were hybridized with a sense probe (Fig. 5B,F).

Figure 5.

Figure 5.

Localization of PDE10A mRNA in rat brain. Serial sections were taken from frozen rat brains. (A and C) Positive staining was observed in the striatal neurons using a digoxigenin-UTP-labeled rat PDE10A antisense riboprobe; bar, 25 µm. (D) Close up of the striatum stained with an antisense riboprobe is shown. An arrow and an arrowhead indicate the neurons showing positive and negative signals, respectively; bar, 5 µm. (E) The neurons of olfactory tubercle (ot) were stained with an antisense probe; bar, 25 µm. (B and F) No positive staining was seen using a sense probe in the striatum and olfactory tubercle; bar, 25 µm.

Antigenic specificity of anti-PDE10A serum

A polyclonal antiserum reactive towards the synthetic peptide (amino-acid residues 287–305 of the rat and human PDE10A2 proteins) has been obtained. The antigenic specificity was investigated with COS-7 cells expressing the FLAG–human PDE10A2 fusion protein. COS-7 cells were transfected with pFLAG10A2, which encodes the FLAG–human PDE10A2 fusion protein as described elsewhere [26]. The cytosolic extracts from transfected COS-7 cells were electrophoretically separated, blotted to membranes, and incubated with the anti-FLAG monoclonal antibody and the polyclonal anti-PDE10A serum (Fig. 6). A specific band of 95 kDa was detected in the cytosolic extracts from COS-7 cells transfected with pFLAG10A2 by both the monoclonal antibody to FLAG and the polyclonal anti-PDE10A serum. The signal was diminished after competition with a preadsorbed anti-PDE10A serum. These results showed that the antiserum binds specifically to the PDE10A protein.

Figure 6.

Figure 6.

Antigenic specificity of polyclonal anti-PDE10A serum. The cytosolic extracts from COS-7 cells transfected with mock (lanes 1 and 3) and pFLAG10A2 (lanes 2, 4 and 5) were subjected to 7.5% SDS/PAGE. The gel was analyzed by immunoblotting with anti-FLAG M5 monoclonal antibody (lanes 1 and 2) and polyclonal anti-PDE10A serum (lanes 3, 4). The expected size of the FLAG–human PDE10A2 fusion protein is 95 kDa. The specific band disappeared with polyclonal anti-PDE10A serum preadsorbed with an immunogenic peptide (lane 5). Positions of molecular mass markers are as shown.

Enzymatic properties of PDEs in rat striatum and testis

HiTrap Q-Sepharose column chromatography was carried out in order to separate PDE10A from other PDEs in rat striatum and testis. The cytosolic extracts of rat striatum yielded two peaks (peak I and peak II) of cGMP hydrolytic activity (Fig. 7A). The fractions of peak I hydrolyzed both cGMP and cAMP and the cGMP hydrolytic activity was not stimulated by calmodulin and Ca2+. In contrast, the activity of peak II was stimulated by fourfold to eightfold by calmodulin and Ca2+ (data not shown). The fractions of peak I were subjected to immunoblot analysis with the polyclonal anti-PDE10A serum (Fig. 7A). An immunoreactive band of 88 kDa was found, whereas no signal was observed in the fractions of peak II (data not shown). Therefore, we considered that peak I and peak II corresponded to PDE10A and calmodulin–PDE, respectively. In the testis, peak I and peak II were also observed (Fig. 7B) and their properties were similar to those in the striatum described above. An immunoreactive band of 88 kDa was found in the fractions of peak I but not peak II (Fig. 7B). Thus, the activity of PDE10A was also present in rat testis. In addition, the cytosolic extracts of rat testis yielded a peak III containing cGMP hydrolytic activity. The activity was inhibited by a PDE5 specific inhibitor, E4021 [31] with IC50 value of 14.7 nm. Peak III was considered to include PDE5 activity. However, the fractions of peak III also showed the cAMP hydrolytic activity. The activity may be derived from other PDEs.

Figure 7.

Figure 7.

Elution profile from high-performance anion-exchange chromatography of cytosolic extracts of rat striatum and testis. The cytosolic extracts of rat striatum (A) and testis (B) were loaded onto a HiTrap Q-Sepharose column. The column was washed with buffer lacking NaCl, and the proteins were eluted by running a linear NaCl gradient in elution buffer (dotted line). Each fraction was used for the assay of PDE activity (upper panels). cGMP or cAMP (1 µm) was used as a substrate. Circles and squares represent the cGMP and cAMP hydrolytic activities in the presence of EGTA, respectively. Three peaks in the cGMP hydrolytic activity are represented by peak I, peak II and peak III. Each fraction of peak I was subjected to immunoblot analysis using a polyclonal anti-PDE10A serum (lower panels).

Kinetic properties of the rat PDE10A enzyme

Kinetic properties of peak I activity in rat striatum (Fig. 7A) are shown in Table 1. Km values were derived from Lineweaver–Burk plots [32] of activity with cGMP or cAMP as a substrate. The Km value for cAMP was 0.26 µm and was significantly lower than that for cGMP (9.3 µm). These results agreed with that of the partially purified recombinant human PDE10A [11]. The inhibitory effects of various PDE inhibitors on peak I activity were very similar to those on the recombinant human PDE10A (Table 1) [11]. Peak I of rat striatum (Fig. 7A) exhibited properties of a cAMP PDE and a cAMP-inhibited cGMP PDE similar to what is observed in the recombinant human PDE10A. These findings confirmed the presence of PDE10A enzymatic activity in rat striatum.

Table 1.  Kinetic properties of the PDE10A enzyme in rat striatum Partially purified PDE10A in rat striatum was used for the assay. K m values were derived from Lineweaver–Burk plots of activity with cGMP or cAMP as a substrate. Assays measuring cAMP and cGMP hydrolytic activity were performed at substrate concentrations of 0.1–2 µm for cAMP and 1.3–160 µm for cGMP, respectively. IC50 values were calculated by linear regression at the concentrations of 0.3 µm for cAMP and 9 µm for cGMP. Enzyme concentration and incubation time were optimized to give about 10% hydrolysis of cAMP and cGMP in the absence of inhibitors. Data are the means of duplicate assays. ND, not determined.
 Compound (µm)
Property/compoundcAMPcGMP
K m 0.269.3
IC50
 iBuMeXan1611
 Zaprinast1917
 Dipyridamole0.940.72
 E40216.16.5
 Milrinone> 100> 100
 Rolipram8471
 cAMPND0.79
 cGMP11ND

Discussion

We have recently isolated a cDNA encoding a new PDE, PDE10A, using bioinformatics together with PCR and cloning techniques, and PDE10A has been characterized with recombinant PDE10A transfected and expressed in COS-7 cells [11]. However, PDE10A enzymatic activities in tissues were not investigated because human tissues could not be obtained. The enzymatic activities of new PDEs, including PDE8 and PDE9, have not yet been reported. Therefore, we aimed to identify PDE10A activity in the rat tissues. First, we have revealed the cDNA sequences and the tissue distribution of rat PDE10A. As a result, a polyclonal anti-PDE10A serum was raised, and the antiserum finally enabled us to identify PDE10A enzymatic activities in the brain and testis.

The cDNA cloning of rat PDE10A revealed the presence of several N-terminal variant forms. Two major forms and three minor forms were identified from a rat brain cDNA library. One of the major forms possesses an N-terminal sequence identical to human PDE10A2 [26]. The other major form, PDE10A3, contains an N-terminal sequence different from those of human PDE10A1 and PDE10A2. In humans, PDE10A1 has been isolated, which has an insertion in the N-terminal-coding region of PDE10A2, resulting in a distinct N-terminal sequence [26]. However, a variant corresponding to PDE10A1 was not isolated in the rat brain cDNA library or detected by PCR analysis. Variation of N-terminal sequences occurs at a specific site (at Leu24 in the rat PDE10A2 sequence), suggesting that these variants are derived from RNA splicing. The sequence of the catalytic domain and a putative cGMP-binding sequence are unchanged among the variants. As reported for PDE5A [25], N-terminal variants may be subject to specific regulation of expression, which directs to the tissue-specific expression. In humans, other variants in addition to PDE10A1 and PDE10A2 may exist.

In the brain, PDE10A transcripts were detected in the neurons of striatum and olfactory tubercles by in situ hybridization analysis, and PDE10A enzymatic activity was identified in the striatum. On the other hand, the presence of PDE1B in the striatal neurons has also been reported [13,14]. These results indicated that PDE10A as well as PDE1B may be involved in neurotransduction through the modulation of cyclic nucleotide levels in the brain cells. The Km values of PDE10A for cAMP and cGMP were 0.26 and 9.3 µm, respectively, whereas those of PDE1B were 24.3 and 2.7 µm[17]. Considering the Km values, PDE10A is expected to have a strong preference for hydrolysis of cAMP rather than cGMP in the striatum.

The striatum is known to be involved in the control of major behavioral and motor functions. Many aspects of these functions imply dopaminergic transmission. Pharmacological studies have demonstrated that dopamine action is mediated by the dopamine receptors D1 and D2, which are expressed prominently in the striatal neurons. Dopamine differentially regulates the two striatal output pathways including striatonigral and striatopallidal neurons in the rat. Signals from D1 and D2 receptors are transmitted via cAMP: D1 stimulates the adenylate cyclase and D2 inhibits the enzyme. PDE10A is present in the regions of high dopaminergic innervation and is possibly responsible for controlling cAMP levels there. The distribution of D1 and D2 receptors has been widely described from studies using autoradiography in the striatum. Some reports have demonstrated that these receptor genes are expressed in distinct populations of striatal neurons in the normal adult rat [33,34]. This implies that D1 and D2 are segregated into distinct striatal output neurons. Therefore, it is intriguing to determine which receptor colocalizes with PDE10A. Further analysis, such as double staining of D1 or D2 receptors with PDE10A, will be needed to clarify this.

In the testis, cAMP is known to play a major role in spermatogenesis [35]. PDE4 is a cAMP-specific PDE that is expressed in the testis and involved in the regulation of cAMP levels [36,37]. PDE8 was recently reported as a cAMP-specific PDE shown to be expressed in the seminiferous epithelium by in situ hybridization [7], but its enzymatic activity has not yet been identified in the testis. The enzymatic activity of PDE10A was found in rat testis and confirmed by immunoblot analysis. The Km value of rat PDE4A is 2.4–5.4 µm[18], which is significantly higher than that of PDE10A, suggesting that PDE10A, as well as PDE4, might contribute to spermatogenesis through cAMP hydrolysis.

In conclusion, we have cloned PDE10A from rats and identified its activity in the testis and in the striatum regions of the brain. Although the biological functions of PDE10A remain unclear, pharmacological analysis using selective inhibitors for this enzyme might help reveal specific roles of PDE10A in these tissues.

Acknowledgements

We are grateful to C. W. Mahoney and G. W. Clendennen for a critical reading of the manuscript. We also thank C. Aruga for helpful discussions.

Footnotes

  1. Enzyme: cyclic-nucleotide phosphodiesterase (EC 3.1.4.17).Note: the nucleotide sequences reported in this paper have been submitted to the GenBank™/EBI Data Bank with the accession numbers AB027155 and AB027156.

Ancillary