Vacuolar-type H -ATPases (V-ATPases), which are composed of at least ten different subunits, can generate a proton-motive force by hydrolyzing ATP and acidify the contents of various intracellular organelles. Subunits A and B of V-ATPase have been detected immunologically in ascidian blood cells, predominantly in signet ring cells (vanadocytes), which accumulate vanadium in their vacuoles. The action of V-ATPase in ascidian blood cells has been demonstrated by the fact that bafilomycin A1, a specific inhibitor of V-ATPases, inhibits the acidification of the vacuoles of vanadocytes. As the next step in studying the function of V-ATPase in vanadocytes, we isolated cDNAs encoding subunits A and B of V-ATPase from the blood cells of an ascidian, Ascidia sydneiensis samea. The nucleotide sequences of the cDNAs for subunits A and B encoded proteins of 619 and 509 amino acids, respectively, both of which were highly conserved among organisms.
INTRODUCTION
The blood cells of ascidians, especially those in the family Ascidiidae in the suborder Phlebobranchia, accumulate extremely high levels of vanadium from seawater (Henze, 1911). At the same time, the homogenate of the blood cells has been revealed to be extremely acidic (Henze, 1911, 1912, 1913, 1932). Most of the vanadium ions in the vanadocytes are in the +3 oxidation state (Hirata and Michibata, 1991). These unusual phenomena have attracted the interest of investigators because of the possible role of the highly acidic environment in changing or maintaining the redox potential.
Previously, we found that there is a correlation between the concentration of vanadium(III) ions and the pH within the vacuole (Michibata et al., 1991). In Ascidia gemmata, which is known to contain the highest concentration of vanadium at 350 mM, the vacuoles have the lowest pH of 1.86. Vacuoles of A. ahodori containing 60 mM vanadium have a pH of 2.67, and those of A. sydneiensis samea containing 13 mM vanadium have a pH of 4.20 (Michibata et al., 1991).
Immunocytological studies, using antibodies against subunits A and B of the vacuolar-type H+-ATPases (V-ATPases) originated from bovine chromaffin granules, have shown that V-ATPases are localized in the vacuolar membranes of vanadocytes (Uyama et al., 1994). A specific inhibitor of VATPases inhibits the proton pump in the vacuoles of the vanadocytes, neutralizing the vacuoles' contents (Uyama et al., 1994). Thus, one definite function of V-ATPases is to accumulate protons in the vanadocytes. However, it is difficult to explain the extremely low pH observed in ascidian vacuoles by only the action of V-ATPases, since the maximum ΔpH that a V-ATPase can generate under typical physiological conditions is around 4 pH units, based on measured H+/ATP stoichiometry (Rea and Sanders, 1987).
We have proposed that two mechanisms may be responsible for the proton accumulation in vanadocytes. One is the hydrolysis of the water molecules coordinating with the vanadium(III) ions. In our recent study, we showed that an extremely low pH could be achieved by hydrolysis of the water molecules coordinating with vanadium(III) ions (Kanamori et al., unpublished data). The other mechanism involves the extremely tight coupling of ATP hydrolysis and proton pumping by V-ATPase in the vanadocytes. As a first step to assess the latter possibility, we isolated and analyzed cDNA of subunits A and B of V-ATPase from the blood cells of the vanadium-rich ascidian, Ascidia sydneiensis samea.
MATERIALS AND METHODS
PCR amplification
A cDNA library of the blood cells of A. sydneiensis samea has been constructed in Uni-Zap XR vector (Stratagene, USA) (Uyama et al., 1998). Lambda phage DNA was extracted from the amplified cDNA library. Degenerate primers were designed from conserved regions of subunits A and B of V-ATPase. For subunit A, the forward primer was 5′- AT(A/C/T) GT(A/C/G/T) TT(C/T) GC(A/C/G/T) GC(A/C/G/T) ATG -3′ and the reverse primer was 5′- (A/G)TC (A/C/G/T)GT (A/G) TA CAT (A/G)TA (A/C/G/T)CC -3′. For subunit B, the forward primer was 5′- AT(A/C/T) GT(A/C/G/T) TT(C/T) GC(A/C/G/T) GC(A/C/G/T) ATG -3′ and the reverse primer was 5′- (A/G)TC (A/C/G/T)GT (A/G) TA CAT (A/G)TA (A/C/G/T)CC -3′. PCR conditions were as follows: 0.2 μg lambda phage DNA, 200 pmole each primer, 0.2 mM dNTP, 25 mM Tris-HCl (pH 8.3), 50 mM KCl and 1.5 mM MgCl2. The reaction volume was 50 μl. 30 cycles of PCR were run; each cycle consisted of 94°C for 45 sec, 50°C for 45 sec and 72°C for 45 sec. This was followed by a final extension at 72°C for 5 min. Amplified DNA fragments of the expected length (126 bp for subunit A and 306 bp for subunit B) were cloned into pBluescript SK(−) vector (Stratagene) and sequenced.
Screening of a cDNA library
The same cDNA library was screened with digoxigenin-labeled random-primed DNA probes derived from the cloned PCR fragments. Hybridization was done at 50°C (subunit A) or 65°C (subunit B) for at least 16 hr in 0.25 M disodium hydrogenphosphate buffer (Na-Pi buffer, pH 7.2), 7% SDS, 1 mM EDTA and 10 ng/ml probes. The membranes were washed three times in 20 mM Na-Pi buffer (pH 7.2), 1% SDS at 40°C (subunit A) or 65°C (subunit B). They were treated with anti-digoxigenin-AP conjugate (Boehringer Mannheim, Germany). The signals were detected by chemiluminescence using CSPD substrates (Boehringer Mannheim) and Hyper-ECL film (Amersham Pharmacia Biotech, Sweden). Positive plaques were screened again before they were subcloned. The cDNAs were excised as plasmids in vivo according to the manufacturer's protocol. Nucleotide sequences were determined using an ABI 377 automated sequencer (Perkin-Elmer Applied Biosystems, USA) or an ALFexpress automated sequencer (Amersham Pharmacia Biotech).
5′-RACE of subunit A cDNA
Oligonucleotides, 5′- TAG TTG GAG CGA AAT CCC AC -3′ (R1), complementary to the 5′ region of cDNA clone #104 of subunit A of VATPase (VATA) were synthesized. The lambda phage DNA from the cDNA library was again used for amplifying a longer 5′-stretch of VATA cDNA. cDNA fragments corresponding to the 5′ region of VATA cDNA were amplified by PCR with the specific primer R1 and a common primer T3, but they did not reach the expected 5′-end of the cDNA. Therefore, another oligonucleotide, 5′- CAG TTC GGA ATG GCC TAC AC -3′ (R2), was designed complementary to the cDNA fragment amplified from the cDNA library. The 5′-RACE was performed with a 5′/3′ RACE Kit (Boehringer Mannheim) according to the manufacturer's protocol. Briefly, total RNA was extracted from blood cells except for giant cells as described (Uyama et al., 1998). Two micrograms of total RNAs were reverse-transcribed with R1 primer. The cDNA products were used as templates for PCR with R2 and anchored dT primers supplied by the manufacturer. The PCR products were cloned into pBluescript SK(−) vector and sequenced.
RESULTS
Cloning cDNA for V-ATPase subunit A
We amplified 126-bp cDNA fragments corresponding to V-ATPase subunit A from the cDNA library constructed from whole blood cells of the ascidian A. sydneiensis samea. Three clones of subunit A were isolated, and were revealed to contain almost identical fragments. Within the amplified region, the ascidian and bovine amino acid sequences were identical at 23 residues out of 30. Using one of the PCR fragments as a probe, we isolated two cDNA clones for subunit A out of 6 × 104 phages from the cDNA library of Ascidia sydneiensis samea blood cells. We designated the corresponding gene as the ascidian VATA gene. The two cDNA clones had almost identical nucleotide sequences. Comparisons of the open reading frame (ORF) of one of the cDNA clones (clone number #104) with those of VATA genes from other organisms showed that cDNA clone #104 seemed to lack a 5′-region corresponding to roughly one-hundred amino acids.
In order to isolate the missing region, we designed two specific primers complementary to the cDNA, and amplified this region directly from the cDNA library and a blood cell cDNA pool (see Materials and Methods). The 3′-part of the amplified fragments was identical to the 5′-part of clone #104. We combined the amplified sequences with that of cDNA clone #104 to generate a cDNA sequence for the VATA gene (Fig. 1).
Fig. 1
Nucleotide and amino acid sequences of VATA cDNA assembled from cDNA clone #104 and two 5′-RACE products from Ascidia sydneiensis samea. The regions corresponding to the forward and reverse primers are shown by dotted lines. The P-loop motif is underlined. Clone #104, isolated by the cDNA library screening, starts at the 376th nucleotide of this sequence, the first round 5′-RACE product at the 124th nucleotide, and the second round product at the 1st nucleotide. This sequence is stored in the DDBJ, GenBank, and EMBL nucleotide databases under accession number AB016483.
The assembled VATA cDNA was composed of 2,253 nucleotides including a 21-bp poly(A) tail. The cDNA contained a single, long ORF of 1,860 nucleotides including the termination codon. It encoded a protein of 619 amino acids. The calculated molecular mass of the predicted protein was 69.0 kDa, which showed a good correlation to the expected molecular weight of 70 kDa determined by western blot analyses using antibodies against the 72K subunit (subunit A) derived from bovine chromaffin granules (Uyama et al., 1994). The deduced amino acid sequence showed a striking similarity to the known 72K subunit of V-type ATPases from various organisms (Fig. 2). There was 81.7% amino acid identity between the ascidian and the bovine sequence and 68.1% identity between the ascidian and yeast sequence, excluding the intein that is spliced into a DNA endonuclease (Gimble and Thorner, 1992). The nucleotide-binding P-loop motif, GXXXXGKT (Saraste et al., 1990), was also found in the ascidian VATA cDNA (Figs. 1, 2).
Fig. 2
Alignment of the amino acid sequences of subunit A of V-ATPases from various organisms. Identical residues are boxed. The P-loop motif is underlined. The solid triangle shows the position of protein splicing, which generates a DNA endonuclease (Gimble and Thorner, 1992) in yeast. Abbreviations: AS, Ascidia sydneiensis samea (this study); BO, bovine (Puopolo et al., 1991); SC, budding yeast (Hirata et al., 1990).
Cloning the cDNA of subunit B
We used a similar procedure to clone subunit B cDNA. We cloned and determined the sequences of eight PCR fragments amplified from the same blood cell cDNA library. The identity of the nucleotide sequences between the fragments was at least 93% (data not shown). Within the amplified region, the ascidian and bovine amino acid sequences were at least 95% identical. Using one of the PCR fragments as a probe, two positive clones were isolated out of 6 × 104 phages. Both were almost identical and contained a nearly full-length cDNA. We designated the corresponding gene as VATB and determined the complete nucleotide sequence of one of the cDNA clones (#105, Fig. 3).
Fig. 3
Nucleotide and amino acid sequence of VATB cDNA clone #105 isolated from Ascidia sydneiensis samea. The regions corresponding to the PCR primers are shown by dotted lines. This cDNA sequence is stored in the DDBJ, GenBank, and EMBL nucleotide databases under accession number AB016484.
The ascidian VATB cDNA clone had an insert of 2,017 nucleotides. The insert contained a single, long open reading frame of 1,530 nucleotides including the termination codon, encoding a protein of 509 amino acids. Its calculated molecular mass was 56.7 kDa, which also showed a good correlation to the molecular weight of the antigen detected by anti bovine subunit B antibody (Uyama et al., 1994). The deduced amino acid sequence of the VATB cDNA was very similar to known sequences of V-ATPase subunit B from various organisms (Fig. 4).
Fig. 4
Alignment of the amino acid sequences of subunit B of V-ATPases from various organisms. Identical residues are boxed. Abbreviations: AS, Ascidia sydneiensis samea (this study); B1, bovine brain-type isoform (Puopolo et al., 1992); B2, bovine renal-type isoform (Nelson et al., 1992); SC, budding yeast (Nelson et al., 1989).
DISCUSSION
V-ATPases that can generate a proton-motive force by hydrolyzing ATP are known to acidify the contents of various intracellular organelles including clathrin-coated vesicles, endosomes, lysosomes, Golgi-derived vesicles, multivesicular bodies and chromaffin granules that belong to the central vacuolar system, and to make an electrochemical gradient of H+ linked to the uptake of certain small molecules (Forgac, 1989, 1992; Nelson, 1992). We are focusing our attention on the close relationship between the high concentration of protons and the high level of vanadium in vanadocytes (Michibata et al., 1991), which may provide a key to clarify the energetics of the accumulation of vanadium in ascidians.
In this experiment, we successfully isolated full-length cDNAs of subunits A and B of the V-ATPase from the vanadium-rich ascidian Ascidia sydneiensis samea. Since antibodies against subunits A and B from bovine chromaffin granules detected antigens specific to vanadocytes (Uyama et al., 1994), we suppose that these cDNAs were derived from the transcripts in the vanadocytes. The subunit composition and cDNA sequences of V-ATPases in various living organisms have been reviewed (Finbow and Harrison, 1997) and some differences in these V-ATPases related to intracellular components or the species of living organisms have been reported. In general, V-ATPase is composed of at least five different subunits, denoted as subunits A to E. Of these, direct chemical labeling and sequence homology studies show that subunits A and B play an important role in binding and catalyzing ATP. The highly conserved amino acid sequences of subunits A and B in living organisms imply that they have important roles. In some organisms, more than one isoform has been reported for each subunit. There are two isoforms of subunit A in the human and chick, and two of subunit B in the human and bovine. The ascidian subunit B isolated in this study is more closely related to the vertebrate homologs than the diverged isoform from bovine renal cells. So far, we have not found any evidence of other isoforms in this ascidian species.
On the one hand, it is difficult to explain the very low pH values observed in ascidian vacuoles by the action of V-ATP-ase alone because, in general, the maximum ΔpH that a VATPase can generate is around 4 pH units (Rea and Sanders, 1987). We have found that the extremely low pH of the vanadocyte vacuoles is partly explained by the release of protons with the hydrolysis of water molecules coordinated to vanadium(III) ions (Kanamori et al., unpublished data). However, it is still possible that the V-ATPase in the vanadocytes has an unusually tight coupling of ATPase activity and proton pumping and can generate such low pH values in the vacuole. Recently, the subunit composition of V-ATPase in the lemon juice sac, whose vacuole has a pH of 2.5, was reported to differ from that in other organs, and the authors suggested that it may be responsible for the low pH (Müller et al., 1997). To determine whether V-ATPase actually functions in vanadocytes and how much proton-pumping activity the ascidian V-ATPase has, the subunit composition and the role of each subunit in vanadocytes must be clarified and will be our future subjects of study.
Acknowledgments
We would like to thank Mr. K. Morita and other staffs at the Otsuchi Marine Center, Otsuchi, Iwate, Japan, for their help in collecting adult ascidians. We also thank Prof. Noriyuki Satoh and his colleagues at Kyoto University, Kyoto, Japan, for their help in determining the DNA sequences. Thanks are also due to Mr. N. Abo in our laboratory, who collected some of the animals and kept them healthy in an aquarium. This work was supported in part by Grants-in-Aid from the Ministry of Education, Science, Sports and Culture of Japan (#09440278, #09874178 and #09839017).
