INTRODUCTION

From the end of 19th century through the first decades of this century, numerous advances were made in many scientific fields. In the field of analytical chemistry, the techniques for emission spectrometry and colorimetric analysis became established, and analyses of the elementary composition of living organisms were popular. A German chemist, Martin Henze (Fig. 1) was one of the pioneers of such bioinorganic chemistry. In 1902, he was employed at the Stazione Zoologica di Napoli, which had been founded by Dr. Anton Dohrn in 1874 and is now one of the oldest marine biological laboratory in the world, when the Department of Physiological Chemistry was established. In 1911, Henze discovered high levels of vanadium in the blood cells (coelornic cells) of an ascidian, known as the sea squirt (Henze, 1911).

Fig. 1.

Dr. Friedrich Wolfgang Martin Henze (1873∼1956) who first found out high levels of vanadium in the blood cells (coelomic cells) of an ascidian in 1911 at Stazione Zoologica di Napoli.

After Henze's interesting discovery, many investigators, including not only analytical chemists but also physiologists, bioinorganic chemists and chemists of natural products, joined in attempts to characterize the vanadium in living organisms, in part because of the initial interest in its very presence in ascidians and in part because of the considerable interest in its possible involvement in as an oxygen-carrier in addition to iron and copper. However, studies on the mechanism of the accumulation and the physiological significance of vanadium made very slow progress since the problems required interdisciplinary cooperation to be solved.

Nevertheless, levels of vanadium in various ascidians were determined by many analytical chemists, who often were little interested in the phytogeny of the ascidians that they were studying. Furthermore, as discussed below, the blood cell that contains high levels of vanadium was misidentified initially as the morula cell (Webb, 1939; Endean, 1960; Kalk, 1963a, b; Kustin et al., 1976), since the color of the morula cell was thought to be that of a vanadium complex. A tunichrome that was proposed to participate in the reduction of vanadium was reported to be present in morula cells and not in signet ring cells (Robinson et al., 1984). The problems to be resolved had been approached in a one-sided manner only by chemists, and the absence of cooperation with biologists hampered appropriate interdisciplinary studies.

Concerning the accumulation of vanadium by ascidians, although many reviews have been published (Goodbody, 1974; Biggs and Swinehart, 1976; Kustin et al., 1983; Boyd and Kustin,1985; Michibata, 1989, 1993; Michibata and Sakurai, 1990; Wever and Kustin, 1990; Smith et al., 1995; Kustin and Robinson, 1995), we intend to focus on the stream of this field in the present paper.

DETERMINATION OF THE VANADIUM CONTENTS OF SEVERAL TISSUES

During the past several decades, the vanadium in ascidians has been analyzed by a variety of analytical methods, such as colorimetry, emission spectrometry and atomic absorption spectrometry (Cantacuzène and Tchekirian, 1932; Vinogradov, 1934; Kobayashi, 1935; Webb, 1939; Noddack and Noddack, 1939; Bertrand 1950; Boeri, 1952; Lybing, 1953; Boeri and Ehrenberg, 1954; Webb, 1956; Levine, 1961; Bielig et al., 1954, 1961a, b, c, 1966; Kalk, 1963a, b; Ciereszko et al., 1963; Rummel et al, 1966; Carlisle, 1968; Swinehart et al., 1974; Danskin, 1978; Botte et al, 1979a, b; Hawkins et al., 1980a). These techniques vary widely in sensitivity and in the precision with which the quantification of vanadium can be performed. Moreover, data were reported in terms of dry weight, wet weight, ash weight, inorganic dry weight, or amount of protein. Thus, early data could not be compared directly, presenting problems to those attempting to pursue physiological studies of the accumulation of vanadium by ascidians.

To bring some order to the field, we decided to re- determine the contents of vanadium in several tissues of ascidians by neutron-activation analysis, which is an extremely sensitive method for the quantification of vanadium. We collected many species of ascidians that belonged to two of three suborders, Phlebobranchia and Stolidobranchia, mainly from the waters around Japan and the Mediterranean. Specimens were dissected into eight samples for analysis, namely, blood cells (coelomic cells), plasma, tunic, mantle (muscle), branchial basket, stomach, hepatopancreas, and gonad. The samples were dried and weighed and then they were mineralized at 500°C and submitted to neutron-activation analysis in the nuclear reactor at the Institute for Atomic Energy, Rikkyo University, Yokosuka, Japan. Some of the stable vanadium in each sample was converted to the radioactive nuclide, ⁵²V, which emits γ-rays at 1,434 keV (Fig. 2). Since the frequency of γ-rays emitted depends on the level of stable vanadium in the sample, the original amount of vanadium in the tissues could be calculated (Michibata, 1984; Michibata et al., 1986).

Fig. 2.

γ-ray spectrogram of the blood cells of Ascidia ahodori. Some of the stable vanadium contained in the samples were activated to radioactive nuclides, ⁵²V, which emitted γ-ray at 1,434 keV after irradiation of thermal neutrons in the TRIGA MARK II nuclear reactor at Rikkyo University.

The data that we obtained are summarized in Table 1. Although vanadium was detectable in samples from almost all species examined, the ascidian species belonging to the suborder Phlebobranchia apparently contained higher levels of vanadium than those belonging to the Stolidobranchia. We also confirmed that blood cells especially contained the highest amounts of vanadium among the tissues examined. Levels of iron and manganese, determined simultaneously, did not vary much among the members of the two suborders. Webb (1939) first proposed the hypothesis that ascidians are animals that represent a transition form between users of vanadium and users of iron and that the relative concentrations of vanadium and iron reflect phylogeny. This hypothesis was based on earlier reports that species in the suborders Phlebobranchia and Aplousobranchia contained high levels of vanadium, whereas the evolutionarily more advanced species in the Stolidobranchia contained smaller quantities of vanadium but retained large quantities of iron. However, we found little difference among levels of iron in specimens from the two suborders and only the vanadium content varied substantially depending on the suborder (Michibata et al., 1986). Furthermore, the highest concentration of 350 mM vanadium was found in the blood cells of Ascidia gemmata belonging to the suborder Phlebobranchia (Michibata et al., 1991a), whose concentration corresponds 10⁷ times higher than that in seawater (Cole et al., 1983; Collier, 1984).

Table 1.

Concentrations of vanadium in tissues of several ascidians (mM)

CELL SEPARATION BY DENSITY-GRADIENT CENTRIFUGATION

Our analytical study confirmed that ascidian blood cells contain the highest amounts of vanadium among the tissues examined. In general, ascidians have nine to eleven different types of blood cells that can be grouped into six categories on the basis of their morphology: hemoblasts, lymphocytes, leucocytes, vacuolated cells, pigment cells and nephrocytes (cf. Wright, 1981). The vacuolated cells can be further divided into at least four different types: morula cells, signet ring cells, compartment cells and small compartment cells. The morula cells were thought for many years to be the so-called vanadocytes (Webb, 1939; Endean, 1960; Kalk, 1963a, b; Kustin et al., 1976). The pale-green color of morula cells resembles that of a vanadium complex dissolved in aqueous solution and the dense granules in morula cells that can be observed under the electron microscope after fixation with osmium tetroxide were assumed to be deposits of vanadium. However, at the end of the 1970's, with the increasing availability of scanning transmission electron microscopes equipped with an energy disperse X-ray detector, it became possible to address the question of whether or not morula cells are the vanadocytes with greater confidence. An Italian group was the first to demonstrate that the characteristic Xray due to vanadium was not detected from morula cells but from granular amoebocytes, signet ring cells and compartment cells and, moreover, that vanadium was selectively concentrated in the vacuolar membranes of these cells, with vanadium granules being present inside the vacuoles (Botte et al., 1979b; Scippa et al., 1982, 1985; Rowley, 1982). Robinson et al. (1984) criticized these results, stating that metals must have been lost during the fixation and dehydration procedures required for such X-ray microanalysis and that it is unlikely that each type of blood cells would lose metals to the same extent. Identification of the true vanadocytes became a matter of the highest priority to those concerned with the mechanism of accumulation of vanadium by ascidians.

To put end to the controversy about the identification of the true vanadocytes, we attempted to settle the problem not by X-ray microanalysis but by a combination of density gradient centrifugation, for the isolation of specific types of blood cells and neutron-activation analysis, for the quantification of the vanadium contents of the isolated subpopulations of blood cells (Michibata et al., 1987). Ascidia ahodori, one of the vanadium-rich ascidians, was used as the material for this analysis. Blood cells were separated from the serum by centrifugation and then loaded onto a discontinuous gradient that consisted of four different concentrations of Ficoll in artificial seawater, prepared without Ca²⁺ and Mg²⁺ to prevent clotting, and the gradient was centrifuged at 100 x g. The blood cells were partitioned into four discrete layers and the subpopulation recovered from the each layer was submitted to neutron-activation analysis. The subpopulation of cells in layer 4, where signet ring cells were dominant, contained the highest level of vanadium. The pattern of distribution of vanadium coincided with that of signet ring cells but not with that of morula cells or that of compartment cells, as shown in Fig. 3. These results proved that the signet ring cells were the true vanadocytes (Fig. 4). The same experiment was repeated with other ascidians, namely, A. sydneiensis samea and A. gemmata, and again signet ring cells were revealed to be the true vanadocytes (Michibata et al., 1990, 1991a; Hirata and Michibata, 1991).

Fig. 3.

Distribution patterns of subpopulations of blood cells and content of vanadium after density-gradient centrifugation. Histograms revealed that the pattern of distribution of vanadium coincided with that of signet ring cells but not with that of morula cells or that of compartment cells. Namely, signet ring cells were identified as vanadium-containing blood cells (vanadocytes) in ascidians (Michibata et al., 1987). L.1: layer 1, L.2: layer 2, L.3: layer 3, and L4: layer 4 from the top. to bottom of discontinuous density-gradient.

Fig. 4.

Morula cells (A), misidentified initially as vanadocytes, and signet ring cell (B), identified newly as vanadocyte, in the ascidian, Ascidia ahodori. Scale bar indicates 10 μm.

CHEMICAL SPECIES OF VANADIUM

Vanadium is a multivalent transition metal. Vanadium ions under ordinary aqueous conditions are limited to the oxidation states, +2, +3, +4 and +5 and to only the +3, +4 and +5 oxidation states under physiological conditions (cf. Chasteen, 1981, 1983; Kustin et al., 1983; Boas and Pessoa, 1987). An E vs. pH diagram and redox potentials are shown in Fig. 5. Vanadium ions in the +3 oxidation state [V(lll)] are usually unstable towards air oxidation, and V(lll) ions are hydrolyzed to V(OH)²⁺ which tend to dimerize to (VOV)⁴⁺ at pH 2.2 and over. in neutral and alkaline conditions, simple soluble V(lll) compounds without any strong ligands have never been reported. As described below, vanadium ions in the blood cells of ascidians are stable under the strongly acidic conditions in these cells. It is of interest to coordination chemists to determine whether any ligands participate in the stabilization of vanadium ions in the +3 oxidation state in the ascidian blood cells.

Fig. 5.

Redox potential versus pH diagram for the vanadium-water system. Vanadium is a multivalent transition metal belonging to the first transition series in the periodic table. The chemical species of vanadium are strongly influenced by the oxidation/reducing properties of the metallic center and by pH conditions. Vanadium ions are, therefore, limited to only the +3, +4 and +5 oxidation states under physiological conditions. Although vanadium ions in the +3 oxidation state are unstable at physiological pH and in the presence of oxygen, yet they are the predominant in ascidian blood cells (vanadocytes) under extremely low pH conditions. The chemical species in the meshed area are in precipitated (cf. Boas and Pessoa, 1987).

Vanadium ions in the +5 oxidation state give a pale yellow solution that is due to the presence of V0₂⁺ under strongly acidic conditions and a colorless solution that is due to the presence of V0₄³⁻ under strongly alkaline conditions. Vanadium ions in the +4 oxidation state are paramagnetic and give a blue solution of oxo-ions, V0²⁺ (vanadyl ions), under moderately acidic conditions. Above pH 4, the chemical species, VO(OH)⁺ or [(VO)₂(OH)₂]²⁺ are formed. At neutral and alkaline pH, vanadyl ions are polymerized to [VO(OH)₃]_n⁻ (cf. Boas and Pessoa, 1987).

OXIDATION STATE OF VANADIUM IN ASCIDIANS

It is believed that vanadium is dissolved in the +5 oxidation state in seawater (McLeod et al, 1975) but this possibility remains to be confirmed (Sugimura et al., 1978). With regard to the oxidation state of vanadium in ascidian blood cells, Henze (1911) was the first to suggest the existence of vanadium in the +5 oxidation state. Later, Lybing (1953), Bielig et al. (1954), Boeri and Ehrenberg (1954), and Webb (1956) reported the +3 oxidation state of vanadium. More recently, noninvasive physical methods, including electron spin resonance spectrometry (ESR), extended X-ray absorption spectrometry (EXAFS), X-ray absorption spectrometry (XAS), nuclear magnetic resonance spectrometry (NMR), and superconducting quantum interference device (SQUID), have been used to determine the intracellular oxidation state of vanadium. Such studies indicated that the vanadium ions in ascidian blood cells were predominantly in the +3 oxidation state, with a small amount being in the +4 oxidation state (Carlson, 1975; Tullius et al., 1980; Dingley et al., 1981; Frank et al., 1986; Lee et al., 1988; Brand et al., 1989).

These results were, however, derived not from the vanadocytes but from the entire population of blood cells. Thus, some questions remained to be answered. In particular, does vanadium exist in two oxidation states in one type of blood cells, or is each state formed in a different cell type? After separation of the various types of blood cells of A. gemmata, we made noninvasive ESR measurements of the oxidation state of vanadium in the fractionated blood cells under a reducing atmosphere (Hirata and Michibata, 1991). Only the +4 oxidation state of vanadium is detectable by ESR spectrometry. As shown in Fig. 6, weak signals due to the +4 oxidation state were recorded when the subpopulation of vanadocytes was submitted to ESR spectrometry at 77K under nitrogen (a). Then, when the sample was lysed by thawing and submitted again to ESR spectrometry at 77K in a reducing atmosphere, the heights of peaks in the ESR spectrum increased (b). Bubbling of oxygen gas into the lysate dramatically increased peak height, with a 7-fold increase 2 hr later (c) and a 13-fold increase 24 hr later (d). However, addition of hydrogen peroxide to the lysate caused the peaks that had been due to oxo-vanadium to disappear. These results indicate that the oxidation state of vanadium in vanadocytes is predominantly the +3 oxidation state, with a small amount of vanadium being in the +4 oxidation state. The ratio of chemical species of vanadium in the two states is 97.6:2.4. No other organisms apart from a polychaeta, Pseudopotamilla occelata (Ishii et al., 1993), are known at present to contain high levels of vanadium in the +3 oxidation state, the most reduced chemical species of vanadium.

Fig. 6.

ESR (electron spin resonance) spectrometry of vanadium. Vanadium ions in the +4 oxidation state are paramagnetic and give a blue solution of oxo-ions, V0²⁺ (vanadyl ions), under moderately acidic conditions. Only the +4 oxidation state of vanadium is detectable by ESR spectrometry . Weak signals due to the +4 oxidation state of vanadium were observed in the intact vanadocytes (a), indicating that the vanadocytes contain a small amount of vanadium in the +4 oxidation state. Lysing the sample increased the signal intensity (b) and bubbling of oxygen gas into the lysate dramatically increased the signals, with a 7-fold 2 hr later (c) and a 13-fold 24 hr later (d), meaning that vanadium in the +3 oxidation state was oxidized to that in the +4 oxidation state. Addition of hydrogen peroxide to the lysate caused the signal to disappear (e), showing that vanadium in the +4 oxidation state was further oxidized to that in the +5 oxidation state. That is to say, it was revealed that vanadium contained initially in the vanadocytes is predominantly in the +3 oxidation state and based on the data, 97.6% of vanadium was calculated to be in the +3 oxidation state in vanadocytes (Hirata and Michibata, 1991). ×50 means magnifications of signal intensity and mT does milli tesra (= gs⁻²A^−-1).

COMPOUNDS THAT REDUCE VANADIUM

Henze (1911, 1932) was the first to report the coexistence of sulfate with vanadium in ascidian blood cells. Vanadium was thought to bind to a nitrogenous compound and sulfate to form a complex, designated haemovanadin, that acted as a respiration pigment in ascidian blood cells (Califano and Boeri, 1950; Webb, 1956). Bielig et al. (1966) suggested that haemovanadin could reduce vanadium. However, the chemical structure of haemovanadin was not determined even though model compounds were proposed. Kustin's group proposed that haemovanadin was actually an artifact formed by airoxidation and disputed the possibility that it might be involved in respiration (Macara et al., 1979a). They isolated a low-molecular compound that they thought might be involved in the reduction of vanadium from the blood cells of Ascidia nigra and Ciona intestinalis and named it tunichrome (Macara et al., 1979b, c). Tunichrome was subsequently revealed to be a compound composed of three pyrogallol moieties by Nakanishi and his co-workers (Bruening et al., 1985). However, since it was demonstrated that tunichrome had no special binding sites for vanadium, since it seemed unlikely that tunichrome could reduce vanadium in the +5 oxidation state to that in the +3 oxidation state (Bulls et al., 1990), and since tunichrome was absent from the vanadium-containing blood cells, namely, the signet ring cells (Michibata et al., 1988, 1990), the possibility could be ruled out that tunichrome is involved in the reduction of vanadium in ascidian blood cells. Recently, Ryan et al. (1992) observed the reduction of vanadium in the +5 and +4 oxidation states by a tunichrome, designated Mm-1, in buffer solution at pH 7 in vitro. It is, however, unclear whether such a reaction could occur in ascidian blood cells.

SULFATE IN VANADOCYTE

A considerable amount of sulfate has always been found in association with vanadium in ascidian blood cells (Henze, 1932; Califano and Boeri, 1950; Bielig et al., 1954; Levine, 1961; Botte et al., 1979a, b; Scippa et al., 1982, 1985, 1988; Bell et al., 1982; Pirie and Bell, 1984; Lane and Wilkes, 1988; Frank et al., 1986, 1987, 1994, 1995; Anderson and Swinehart, 1991), suggesting that sulfate might be involved in the biological function and/or the accumulation and reduction of vanadium. However, Frank et al. (1987) suggested the existence of a non-sulfate sulfur compound, such as an aliphatic sulfonic acid, in ascidian blood cells. As the first step towards an analysis of the possible correlation between the accumulation and/or reduction of vanadium and sulfate, we determined the ratio of the level of sulfate to that of vanadium in the blood cells of the ascidian Ascidia gemmata by Raman spectroscopy, as shown in Fig. 7. The ratio obtained was approximately 1.5, as would be predicted if sulfate ions were present as the counter ions of vanadium ions in the +3 oxidation state. We also found evidence that an aliphatic sulfonic acid was present in the blood cells (Kanamori and Michibata, 1994).

Fig. 7.

Raman spectra of the blood cells of Ascidia gemmata. High levels of sulfate and/or sulfur compounds have found in association with vanadium in ascidian blood cells. Raman spectrometry can determine noninvasively the amounts of sulfate and vanadium in the vanadocytes. A band at 983 cm⁻¹ and a shoulder at 995 cm⁻¹ are derived from S0₄²⁻ symmetric stretching vibration and V=0 stretching vibration, respectively. Based on these intensities of the peaks, the concentration ratio of S0₄²⁻ and V³⁺ was calculated to be 1.47 (Kanamori and Michibata, 1994).

EXTREMELY LOW pH IN VANADOCYTES

Henze (1911, 1912, 1913, 1932) was the first to report that a homogenate of ascidian blood cells is extremely acidic. Almost all subsequent investigations have been supported his observation. Recently, however, Kustin's group disputed the earlier reports and claimed that the previously used methods gave spurious results when applied to ascidian blood cells because the interior of blood cells that contained high levels of vanadium and iron was probably a strongly reducing environment (Dingley et al., 1982; Agudelo et al, 1983). In their view, the possibility remained that it was the intracellular redox potential, and not the pH, that had been measured. They reported that the intracellular pH was neutral on the basis of measurements made by a new technique, which involved the transmembrane equilibrium of ¹⁴C-labeled methylamine. Hawkins et al. (1983) and Brand et al (1987) also reported nearly neutral values for the pH of the interior of ascidian blood cells, which they determined noninvasively from the chemical shift of ³¹P-NMR. However, Frank et al (1986) demonstrated that the interior of the blood cells of Ascidia ceratodes has a pH of 1.8, basing their results on the new finding that the ESR line width accurately reflects the intracellular pH. Thus, the reported pH inside ascidian blood cells has excited considerable controversy, as summarized in Table 2.

Table 2.

Reported pH inside ascidian blood cells

We consider that the main reason for the extreme variations is that the measurements of pH were made with entire populations of blood cells and not with the subpopulation of vanadocytes specifically. Thus, one or two specific types of blood cells might have a highly acidic solution within their vacuoles, in which vanadium would be present in a reduced state. With this possibility in mind, we designed an experiment in which we combined the separation of each type of blood cells, measurement of pH with a microelectrode under anaerobic conditions, to avoid air-oxidation, and ESR spectrometry as a noninvasive method for the measurement of pH to confirm the results obtained with the microelectrode (Michibata et al., 1991a). Three species of vanadium-rich ascidians, Ascidia gemmata, A. ahodori, and A. sydneiensis samea, were used as the materials. Blood cells drawn each species were fractionated by density-gradient centrifugation, as described above. A comparison of the patterns of distribution of each type of blood cells, the measured pH values after conversion to concentrations of protons ([H⁺]) and levels of vanadium in each layer of cells are shown in Fig. 8. It is clear from Fig. 8 that the patterns of distribution of protons and vanadium were similar. Thus, the signet ring cells contained high levels of vanadium and had a low intracellular pH in all three species. In A. sydneiensis samea one other type of blood cells in addition to signet ring cells was also found to be acidic. These cells were large and spherical, measuring 30 to 80 μm in diameter and having a single fluid-filled vacuole. This type of cell is probably analogous to the so-called nephrocyte, as judged by the criteria proposed by Wright (1981). However, we propose to call these cells as “giant cells” until their function is revealed. Giant cells have the lowest specific gravity of all the blood cells, they contain no vanadium but have a very low pH of 1.48 (Michibata et al., 1991a) The contents and function of these cells remain to be clarified.

Fig. 8.

Determination of acidic blood cells of Ascidia gemmata.

Histograms depict the number of each type of blood cells and the concentrations of H⁺ and vanadium in three different layers (layers 1, A and F; cf. Michibata et al., 1991a) of blood cells which were fractionated by density-gradient centrifugation in Percoll. MC, morula cells; SRC, signet ring cells; CC, compartment cells; [H⁺], concentration of protons; V, vanadium

ESR spectrometry was also used for noninvasive measurements of the intracellular acidity of blood cells (Michibata et al., 1991a). The method is based on the phenomenon that the ESR line width due to oxo-vanadium [VO(IV)] ions increases (cf. Frank et al., 1986) almost linearly from pH 1.4 to pH 2.3, as shown in Fig. 9. We confirmed that the low pH values obtained with a microelectrode were not artifacts when we applied this method to the measurement of the pH of the contents of signet ring ceils from A. gemmata. We note, to avoid any misunderstanding, that the acidic solution is assumed to be contained in the vacuole of each signet ring cells. However, most of each signet ring cell is, in fact, occupied by the vacuole itself, so the contents of the vacuole are almost equivalent to the contents of the cell.

Fig. 9.

Calibration curve for pH values estimated by the half-width of (-7/2)// line given by ESR spectrometry. pH values in the blood cells of vanadium-rich ascidians are noninvasively measurable over the range from 1.4 to 2.3 (Michibata et al., 1991 a).

ENERGETICS OF THE ACCUMULATION OF VANADIUM

Our comparison of pH values and levels of vanadium in the signet ring cells of three different species, as shown in Table 3, suggested that there might be a close correlation between a higher level of vanadium and lower pH, namely, a higher concentration of protons. It is well known that H⁺-ATPases can generate a proton-motive force by hydrolyzing ATP. This enzyme plays a role in pH homeostasis in various intracellular organelles, including clathrin-coated vesicles, endosomes, lysosomes, Golgi-derived vesicles, multivesicular bodies and chromaffin granules that belong to the central vacuolar system (cf. Forgac, 1989, 1992; Nelson, 1992).

Table 3.

Correlation between concentrations of vanadium and pH values in ascidian blood cells

Therefore, we examined the signet ring cells of the ascidian Ascidia sydneiensis samea for presence of H⁺-ATPase (Uyama et al., 1994). The vacuolar-type H⁺-ATPase is composed of several subunits, and subunits of 72 kDa and 57 kDa have been reported to be common to all eukaryotes examined. Antibodies prepared against the 72 kDa and 57 kDa subunits of a vacuolar-type H⁺-ATPase from bovine chromaffin granules did indeed react with the vacuolar membranes of signet ring cells. Immunoblotting analysis confirmed that the antibodies reacted with specific antigens in ascidian blood cells. Furthermore, addition of bafilomycin A₁ a specific inhibitor of vacuolar-type H⁺-ATPases (Bowman et al., 1988), inhibited the pumping function of the vacuoles of signet ring cells, with resultant neutralization of the contents of the vacuoles, as shown in Fig. 10.

Fig. 10.

Acidity of vacuoles of vanadocytes and inhibition of the acidification by bafilomycin A₁ Signet ring cells (vanadocytes) were revealed to emit a brilliant vermilion indicating acidic phi after incubation of blood cells of Ascidia sydneiensis samea with 2μM acridtne orange for 1 hr but the other types of blood cells did not indicate acidic pH. However, addition of 1 μM bafilomycin A₁, a specific inhibitor of vacuolar H⁺ATPase, caused vanadocytes neutralized (showing green fluorescence) with resultant inhibiting pump function of the H⁺-ATPase. No changes in color of autonomous fluorescence emitted from morula cells were found with or without bafilomycin A₁ (Uyama et al., 1994). S, signet ring cells (vanadocytes); M, morula cells. Scale bar indicates 10 μ.m.

The acidification of vacuolar compartments in eukaryotic cells has been shown to have a number of important functions in neuronal and endocrine systems (Moriyama et al., 1992) and to be important for the degradation of proteins (Yoshimori et al., 1991). There appears to be a strong possibility that the activity of the enzyme is linked to the accumulation of vanadium in signet ring cells. We are trying to obtain direct evidence for such an association.

PREPARATION OF MONOCLONAL ANTIBODIES AGAINST ASCIDIAN BLOOD CELLS

The identification of vanadocytes has been a subject of controversy, in part because of difficulties associated with morphological discrimination between several types of blood cells and in part because of our inadequate knowledge of cell lineages from the so-called stem cells to the peripheral cells. For example, Scippa et al. (1988) reported that the vacuolated and granular amoebocytes and a variety of compartment cells in Phallusia mammillata could be considered to be vanadocytes, in addition to the signet ring cells.

The establishment of reliable cell markers for the recognition of different types of blood cells seems, therefore, to be a necessary step if we are to clarify not only the function but also the lineage of each type of cell. We prepared a monoclonal antibody, which we hoped might serve as a powerful tool for solving these problems, using a homogenate of the subpopulation of signet ring cells from Ascidia sydneiensis samea as the antigen (Uyama et al., 1991). The monoclonal antibody obtained, S4D5, was shown to react specifically with the vanadocytes not only from A, sydneiensis samea but also from two additional species, A. gemmata and A. ahodori. Immunoblotting analysis showed that this antibody recognized a single polypeptide of about 45 kDa from all three species. Recently, an other monoclonal antibody, C2A4, was prepared and it reacted specifically with the vacuolar amoebocytes and recognized a single band of a protein of about 200 kDa (Kaneko et al., 1995), as shown in Fig. 11.

Fig. 11.

A monoclonal antibody, C2A4, reacted specifically with the vacuolar amoebocytes among several types of blood cells of Ascidia sydneiensis samea (the left figures) and the antigen recognized by the monoclonal antibody was subjected to immunoblot analysis (the right figure). AC, vacuolar amoebocytes; MC, morula cells; CC, compartment cells. Scale bar indicates 10μm.The antibody recognized a single band of a protein about 200 kDa, after treatment with endoglycosidase F (Kaneko et al., 1995).

There are several reports on the localization of blood cells in ascidian (Kalk, 1963a; Smith, 1970a; Ermak, 1975, 1976). According to these earlier reports, haematogenic activity is observed in three main areas of ascidians, 1) in the connective tissues around alimentary canal, 2) in the pharyngeal wall and transverse vessels of the branchial basket, and 3) in discrete nodules located in the body wall. Employing these monoclonal antibodies and the autonomous fluorescence emitted by each type of cell as cell markers (Wuchiyama and Michibata, 1995), we tried to identify the haematopoietic sites of vanadocytes. We found vanadocytes in the connective tissues around the alimentary canal, separate from morula cells, compartment cells and amoebocytes. It seems that the precursor cells of vanadocytes are formed in the connective tissues, while other types of blood cells are formed in the other sites (Kaneko et al., 1995).

ACCUMULATION OF VANADIUM DURING EMBRYOGENESIS

Monoclonal antibodies are also useful tools with which to determine the time at which the accumulation of vanadium starts during embryogenesis. Since the amount of vanadium stored in embryos is beneath the limits of detection of conventional analytical methods, such as atomic absorption spectrometry, there have been no reports of the direct determination of vanadium accumulated during ascidian embryogenesis. Using neutron-activation analysis and an immunofluorescence method, we found that the amount of vanadium per individual increased dramatically two weeks after fertilization. Within two months, the amount accumulated in larvae was about 600,000 times greater than that in the unfertilized eggs of A. sydneiensis samea (Michibata et al., 1992), as shown in Fig. 12. A vanadocyte-specific antigen, recognized by a monoclonal antibody specific to the signet ring cells, first became apparent in the body wall at the same time as the first significant accumulation of vanadium (Uyama et al., 1993).

Fig. 12.

Accumulation of vanadium during embryogenesis of the ascidian, Ascidia sydneiensis samea. To determine the time at which the accumulation of vanadium commences during embryogenesis, eggs and embryos were submitted for neutron-activation analysis. The levels of vanadium began to increase and the amount in larvae reached 2.3 )μg/individual, which was about 600,000 times higher than the amount in unfertilized eggs (Michibata et al., 1992). Furthermore, a vanadocyte-specific antigen, first became apparent in the body wall at the same time as the first significant accumulation of vanadium (Uyama et al., 1993).

VANADIUM-BINDING PROTEINS

With respect to the pathway for the accumulation of vanadium from seawater, it seems likely that some proteins might participate, even though the results reported to date seem to indicate that vanadium is present in a free, noncomplexed form or is associated with low-molecular-weight components in ascidians. We are now trying to extract proteins that bind to or associate with vanadium in ascidians, assuming that there should be at least three types of proteins, namely, vanadium-transfer, vanadium-receptor and vanadium-channel proteins, as shown in Fig. 13.

Fig. 13.

Schematic representation of the pathway of accumulation of vanadium by ascidians. Vanadium in the +5 oxidation state dissolved in seawater is thought to be incorporated into the interior of the body via branchial baskets, where vanadium is reduced to the +4 oxidation state. Vanadium in the +4 oxidation state is further reduced to the +3 oxidation state and stored in the vacuole of vanadocyte. Some proteins, namely, vanadium transfer, vanadium receptor and vanadium channel proteins, are thought to be involved in the accumulation of vanadium during the course.

The route for the accumulation of vanadium ions from seawater in the blood system has not yet been revealed. The uptake of vanadium ions was studied with radioactive vanadium ions (⁴⁸V). Previous studies were commonly designed to clarify the direct uptake of vanadium ions from the surrounding seawater and were, therefore, limited to an examination of how much vanadium was incorporated into some tissues (Goldberg et al., 1951; Bielig et al., 1963; Dingley et al., 1981; Michibata et al., 1991b) except a few reports (Hawkins et al., 1980b; Roman et al., 1988). The majority of the vanadium incorporated by ascidians was thought to be dissolved as ionic species or associated with the low-molecular-weight substances rather than proteins (cf. Kustin and Robinson, 1995).

In general, heavy metal ions incorporated into the tissues of living organisms are known to bind to macromolecules such as proteins. We have, therefore, searched for vanadium-binding proteins in the blood cells of ascidians. Using a combination of SDS-PAGE and flameless atomic absorption spectrometry, we succeeded in isolating at least four different types of vanadium-binding proteins. Monoclonal antibodies against some of these proteins have been prepared as part of our efforts to determine whether the proteins are associated with vanadium-containing blood cells (Wuchiyama et al., unpublished data).

PHYSIOLOGICAL ROLES OF VANADIUM IN ASCIDIANS

Although the unusual phenomenon whereby some ascidians accumulate vanadium to levels more than ten million times higher than those in seawater has attracted researchers in various fields, the physiological roles of vanadium remain to be explained. Endean (1955a, b, c, 1960) and Smith (1970a, b) proposed that the cellulose of the tunic might be produced by vanadocytes. Carlisle (1968) suggested that vanadium-containing vanadocytes might reversibly trap oxygen under conditions of low oxygen tension. The hypothesis has also been proposed that vanadium in ascidians acts to protect them against fouling or as an antimicrobial agent (Stoecker, 1978; Rowley, 1983). However, most of the proposals put forward do not seem to be supported by sufficient evidence.

Recently, we observed an unexpected phenomenon of great interest. We found that the number of vanadocytes increased when ascidians were immersed in a solution that contained 10 mM or 20 mM NH₄CI (Hayashi et al., 1996). The increase in size of the population of signet ring cells might be interpreted as a self-defense response. Of course, the reason why the number of signet ring cells increased in response to NH₄CI is still a matter of conjecture. However, attempts to characterize this phenomenon can be expected to promote more information about the unusual accumulation of vanadium by one class of marine organisms.