INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

AN INCREASE IN EXTRACELLULAR pH is associated with photosynthetic activity of aquatic plants. This pH increase may be primarily extracellular, resulting from an influx of CO₂ at a rate faster than air equilibration can supply it (23) or from carbonic anhydrase-mediated conversion between and CO₂ followed by subsequent CO₂ uptake (4). It may also be primarily intracellular, after influx. The subsequent use of CO₂ by ribulose bisphosphate carboxylase-oxygenase results in the formation of OH inside the cells and its subsequent efflux (or influx of H⁺). In this way, a constant intracellular pH on a one-to-one basis with uptake is maintained. An electrogenic /OH antiport or an H⁺- cotransport system may mediate this efflux (18, 24). As early as 1933, Arens (3) demonstrated that leaves of submerged plants Potamogeton and Elodea were polarized. Only the lower leaf surface absorbed exogenous and became acidic, whereas the upper surface became alkaline, presumably due to OH excretion. Functional polarity and the use of have now been extended to various plant species, including both freshwater and marine algae (18, 19, 21, 22). Elzenga and Prins (8) demonstrated that in Potamogeton, acidification at the lower side of leaves was induced by the activity of a plasma membrane-bound H⁺-adenosinetriphosphatase (ATPase). This acidification shifts the equilibrium between and CO₂ to the formation of CO₂ and its subsequent diffusion into the leaf cells where it is assimilated. Simultaneously, H⁺ uptake at the upper surface compensated for the alkalinizing effect of ATPase activity on cytoplasmic pH. The functional tissue polarity of aquatic plants is assumed to play a major role in utilization for photosynthesis (22).

In a morphological sense, oral epithelial layers of symbiotic anthozoans, which harbor the photosynthetic dinoflagellates Symbiodinium sp. generally called zooxanthellae, are polarized structures, because endosymbionts are localized only within endodermal cells facing the coelenteric cavity. These cells are separated from the surrounding seawater by the mesoglea and the ectodermal cells. Recently, Bénazet-Tambutté et al. (6) have shown that the major source of dissolved inorganic carbon (DIC) for photosynthesis of the sea anemone Anemonia viridis endosymbionts is external seawater. These authors together with Furla et al. (9) suggested that ectodermal cells may play an important role in supplying DIC for photosynthesis of endosymbionts.

The objective of the present study was therefore to study OH (H⁺) fluxes and transport mechanisms by the two faces of the oral epithelial layers of the sea anemone Anemonia viridis to determine if the tentacles present a functional polarity linked to photosynthesis of dinoflagellate endosymbionts. For this purpose, we used two types of experimental preparations described previously (5, 6): the tentacle bag and the Ussing chamber. Further, to validate the use of the tentacle bag as a model, we also carried out measurements under in vivo conditions. Evolution of pH and fluxes of OH (H⁺) were measured by titration in media facing both ectodermal and endodermal cells. A functional polarity similar to that previously described in aquatic plants is demonstrated in the sea anemone, and its role in assimilation is discussed.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Biological material. Specimens of the Mediterranean sea anemone Anemonia viridis (Forskål) were collected in Villefranche-sur-mer, France, and were maintained in an open-circuit seawater aquarium supplied with Mediterranean seawater pumped at a depth of 50 m. Light was provided by a metal halide lamp (HQI-TS 400 W; Philips, Somerset, NJ), with a photosynthetic photon flux density of 125 µmol · m² · s¹ and a 12:12-h light-dark photoperiod. Unless otherwise indicated, all experiments were carried out under a constant saturating irradiance of 300 µmol photons · m² · s¹ (6) (HQI-TS 400 W, Philips) and a constant temperature of 16.0 ± 0.5°C maintained with a thermostatic circulator (Lauda RM20).

Seawater preparation. Filtered seawater (FSW) was obtained by filtering Mediterranean seawater through a 0.45-µm Millipore membrane. -free seawater (0- SW) was prepared according to Bénazet-Tambutté et al. (6). The pH was adjusted to 6.6 or 8.2 with CO₂-free NaOH (6). pH was buffered with either 1 mM piperazine-N,N'-bis(2-ethanesulfonic acid) (pH 6.6) or 1 mM tris(hydroxymethyl)aminomethane (Tris, pH 8.2). This buffer concentration did not interfere with titration measurement (results not shown). The total alkalinity (TA) was measured according to Gran (13) using a DL25 Mettler titrator; the concentration was calculated from the pH and TA. FSW adjusted to pH 9 was prepared with 1 N NaOH. Sodium-free artificial seawater (0-Na⁺ ASW) was prepared according to Bénazet-Tambutté et al. (6).

Preparation and use of tentacle bags. A portion of isolated tentacle was filled with 200 µl of the appropriate experimental medium to form a bag (6). In some experiments, the tentacle was everted (inside-out tentacle bag) as described previously (6). The media inside and outside the bag were called "internal" and "external," respectively. Internal pH was monitored with a flexible esophageal pH electrode (1.2 mm OD; Diamond General, Ann Arbor, MI) introduced in the bag and tied. External pH was measured with a pH microelectrode (Tacussel, France, ref. XC 161). Either internal and external pH or external pH and O₂ evolution were measured simultaneously in 5 ml of FSW in an oxygen chamber (see below).

Measurement of photosynthetic rates. The photosynthetic rate of tentacle bags was measured as the rate of net O₂ production (6) using microrespirometers (Rank Brothers, Cambridge, UK). The rate of O₂ evolution was determined by slope analysis from recorder tracings of the O₂ content vs. time. The chamber (5 ml) was closed by a stopper with a hole to allow the passage of the pH microelectrode. O₂ stratification was avoided using a magnetic stir bar.

Measurement of internal pH in whole sea anemones. Sea anemones (5- to 6-cm pedal disk diameter) were first anesthetized in 100 ml of FSW-0.37 M MgCl₂ (1:1), after which a flexible esophageal pH electrode was introduced in one tentacle and tied. The surrounding medium was then substituted with FSW.

Ussing chambers. Pieces of tentacles were mounted in a Ussing chamber (area of exposed tissue 0.2 cm²) as previously described (5). The volume of solution in each half chamber was 700 µl. The solution in each half chamber was mixed by a small magnetic stir bar. The compartment facing the ectoderm was called the "ectodermal compartment," and that facing the endoderm was called the "endodermal compartment". In some experiments, the endodermal cells were removed by scraping them away with a scalpel under a binocular microscope (6); freshly isolated zooxanthellae (FIZ, ~0.8 × 10⁶ cells/ml) obtained according to Goiran et al. (12) were added to the endodermal compartment.

Measurement of OH (H⁺) fluxes. Two kinds of titrations were performed. Real time titration was used for tentacle bags, and delayed titration was used in Ussing chambers. For the latter, samples of both ectodermal and endodermal media were taken every 15 min after a 10-min equilibration period. The changes in H⁺ concentration permitted the calculation of net OH (H⁺) fluxes. In both cases, H⁺ concentration was measured by titration using an automatic titrator (Tacussel TTP-2). Titration was performed with a 5 mM HCl solution. The pH end point was the initial pH of the samples (30). To obtain reproducible measurements, the sample was thoroughly mixed during titration. The volume of HCl added allowed the calculation of the acid (base) net fluxes during the incubation period. In this paper, we refer to the pH increase of the medium as OH efflux and the pH decrease as H⁺ efflux. It should be noted that this might equally have been caused by H⁺ (OH) influx.

Pharmacology. Pharmacological experiments were carried out in a Ussing chamber. The anion carrier inhibitor 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS) and the carbonic anhydrase inhibitors acetazolamide (N-[5-sulfamoyl-1,3,4-thiadiazol-2-yl]acetamide) (AZ, Diamox) and ethoxyzolamide (EZ) were dissolved in dimethyl sulfoxide (DMSO) and buffered with 1 M Tris to pH 8.2. The H⁺-pump inhibitor diethylstilbestrol (DES) was dissolved in absolute ethanol and buffered with 1 M Tris to pH 8.2. Amiloride, a potent and relative specific inhibitor of the Na⁺/H⁺ antiporter, was dissolved in DMSO. Although preliminary experiments demonstrated that concentrations of DMSO or ethanol up to 1% (vol/vol) caused no significant effect on DIC flux (results not shown), we used DMSO or ethanol at a concentration 0.5%. All chemicals were obtained from Sigma (St. Louis, MO). The permeability of inhibitors was tested with a spectrophotometer (Uvikon 931, Kontron).

Measurement of DIC incorporation into photosynthates. Inorganic carbon incorporation into photosynthates was measured in a Ussing chamber and in tentacle bags. The internal medium was either normal FSW (pH 8.2) or FSW adjusted to pH 9.0; the external medium was FSW. [¹⁴C] (specific activity 2.0 × 10¹² Bq/mol; NEN, Boston, MA) was added to either the external or internal medium to a final specific activity of 2.20 × 10¹⁰ Bq/mol. [¹⁴C] labeling lasted for 30 min (Ussing chamber) or 15 min (tentacle bag). At the end of the incubation period, the tentacle was thoroughly rinsed with FSW to remove any adhering inorganic ¹⁴C and was then sonicated in distilled water. Protein content was measured immediately on an aliquot. After acidification of the sample to pH 4.0 and equilibration for in air 2 h, inorganic ¹⁴C incorporated into photosynthates was measured. Radioactive samples (100 µl) were counted with a scintillation counter (Packard 1600 Tricarb; Packard, Downers Grove, IL) after addition of 4 ml Aqualuma Plus. To compare the different experimental protocols, results were expressed as percent of total ¹⁴C incorporation (i.e., ¹⁴C incorporation into photosynthates when [¹⁴C] is added in the external medium plus ¹⁴C incorporation into photosynthates when [¹⁴C] is added in the internal medium in the same conditions).

Measurement of TA of the internal medium of tentacle bags. After 1 h of incubation in saturating light condition, the internal medium of 8-10 tentacle bags was collected and the pH was rapidly measured. The TA was measured by potentiometric titration of 700-µl samples (Mettler titrator DL25) using 0.01 N HCl (Merck Titrisol, 9974). Samples, HCl, and the titration vessel were maintained at 25°C. TA was computed according to Gran (13). These values were compared with the initial alkalinity of FSW.

Presentation of results. At least four replicates were conducted for each experiment. Except for those experiments involving pH measurements, in which only representative data are reported, results are given as means ± SD. Results were standardized by the protein content of tentacle. Protein determination was carried out according to the method of Lowry, with bovine serum albumin as a standard using an autoanalyzer (Alliance Instruments). Results are presented as OH (H⁺) flux expressed in nanoequivalents per minute per milligrams of protein or as percentage of control flux.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Effect of light on rate of O₂ evolution and pH change of external and internal media. The light dependence of the rate of O₂ evolution and pH change of both the internal and external media of a tentacle bag of the sea anemone Anemonia viridis is presented in Fig. 1. Increase of internal pH occurred only in the light, whereas acidification was observed in the dark. Although the rate of O₂ production was constant during the 45-min light period, the internal pH increased rapidly from pH 7.4 to around 8.9 ± 0.2 within 20-30 min and then ceased. In contrast, the external pH changed slightly (~0.1 pH units). The gradient of pH between the external seawater and the internal medium was ~0.8 pH units. To verify if the bag could be considered representative of in vivo physiological conditions, internal pH was monitored on a tentacle of a whole sea anemone. Figure 2 shows a representative pH evolution. As in the tentacle bags, we observed a light-induced pH increase of the internal medium, with a similar plateau around pH 9.0, and acidification was observed in darkness. To determine the specific role of the ectodermal and endodermal layers, a piece of tentacle was mounted in the Ussing chamber. In this case, both tissue layers were bathed with the same volume of seawater. Figure 3 shows that although pH variations were <0.1 pH units in the ectodermal compartment, an increase of 1.6 pH units (from 7 to 8.6) was measured in the endodermal compartment. Again, the evolution of pH reached a plateau after 20 min.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Light-dependent pH variations in the internal () and external () medium of the Anemonia viridis tentacle bag. O2 content in seawater was also shown (+). Light irradiance was 300 µmol photons · m2 · s1.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Light-dependent pH variations in the internal medium of Anemonia viridis tentacle as measured in vivo on a whole sea anemone.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Light-dependent pH variations in the endodermal () or ectodermal () compartment of Ussing chamber.

Light-sensitive OH flux. Figure 4 shows that when the measurement of light-sensitive OH flux was performed in the external medium of a bag, a lag phase of 24 ± 5 min was observed before the onset of titration. On an inside-out tentacle bag, this lag phase decreased to 2.7 ± 1.5 min, suggesting that the primary site of OH excretion is the endodermal layer. To confirm the cellular origin of OH, titration experiments were conducted in the Ussing chamber. A 15-min duration was chosen for the titration experiments to avoid any diffusion of OH from the endodermal toward the ectodermal compartment. In darkness, an H⁺ efflux was measured in both compartments, with higher excretion observed in the endodermal compartment (0.74 ± 0.02 and 1.14 ± 0.35 neq · min¹ · mg protein¹ in ectodermal and endodermal compartments, respectively). This higher rate of H⁺ efflux may be due to the higher respiratory rate related to the presence of zooxanthellae. Under saturating light irradiance, no net OH excretion was measured in the ectodermal compartment, whereas a net OH excretion (3.55 ± 1.48 neq · min¹ · mg protein¹) was measured in the endodermal compartment, showing that OH came primarily from the endodermal cell layer. If we assume that the rate of H⁺ efflux was constant in the light, then an OH excretion into the ectodermal compartment (0.74 neq · min¹ · mg protein¹) was induced by light, whereas the total rate of OH excretion into the endodermal compartment was 4.69 neq · min¹ · mg protein¹. In subsequent experiments, we only measured the net rate of OH excretion in the Ussing chamber.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Time course of light-dependent OH efflux in tentacle bag. OH efflux was recorded in extratentacular medium of either a normal (solid line) or everted (i.e., inside-out) tentacle (dashed line).

Pharmacology of light-sensitive OH flux. The ion transport properties of the oral epithelial layers were characterized by exploring the effect of putative inhibitors of specific ion transport on OH excretion by both ectodermal and endodermal cell layers. OH movement may result either from carrier-mediated anion exchange, carbonic anhydrase activity, or H⁺-pump activity. Consequently, we tested inhibitors known to interfere with each of these pathways: DIDS (400 µM) (1), Diamox (AZ) and EZ (600 µM) (1), and DES (100 µM) (1). For all inhibitors tested, no effect was measured on OH (H⁺) fluxes by the ectodermal cell layer (results not shown). Consequently, Fig. 5 shows inhibitor effects on OH excretion by the endodermal cell layer only. DIDS added to the ectodermal or endodermal compartment decreased OH excretion by 35%. When added to both compartments, the inhibition was greater, although the difference does not appear to be significant. Diamox and EZ added to the ectodermal compartment inhibited OH excretion by ~50 and 20%, respectively. A 50% inhibition was also observed when Diamox was added to the endodermal compartment or in both compartments, showing that the effect of Diamox was not additive. When added to the endodermal compartment (or in both compartments) EZ almost totally inhibited OH excretion. The proton-pump inhibitor DES decreased OH secretion when added to the ectodermal compartment (~35% inhibition). The percentage of inhibition was 50% when DES was added to the endodermal compartment. Its effect was not additive when added to both compartments.

View larger version (43K): [in this window] [in a new window]   Fig. 5.   Pharmacology of light-dependent OH efflux in endodermal compartment. Fluxes are expressed as %control value (i.e., 3.55 ± 1.48 neq · min1 · mg protein1) as measured in filtered seawater (FSW) during a 15-min period. Effect of an anion transport inhibitor, DIDS (400 µM), carbonic anhydrase inhibitors Diamox (600 µM) and ethoxyzolamide (EZ, 600 µM), and an H+-ATPase inhibitor, diethylstilbestrol (DES, 100 µM). Inhibitors were added (+) to endodermal and/or ectodermal compartment as indicated at top of figure.

Effect of pH and availability. The effects of pH and 0- SW were tested on OH excretion. No effect of any protocol tested was observed on OH excretion by the ectodermal cell layer, whereas OH excretion by the endodermal cell layer was modulated by action on the opposite compartment. Figure 6 shows that elimination of from the ectodermal compartment at pH 8.2 reduced by ~30% OH excretion by the endodermal cell layer, whereas a reduction of 90% was observed when was eliminated from the endodermal compartment. A similar pattern of inhibition of OH excretion was observed when the endodermal layer of the tentacle was immersed in FSW adjusted to pH 9.0. Elimination of from both endodermal and ectodermal compartments at pH 6.6 decreased OH excretion within the endodermal compartment by ~75%. Elimination of from endodermal or ectodermal compartments at pH 6.6 led to a smaller inhibition of OH excretion compared with that observed at pH 8.2 (25 instead of 90% and 10 instead of 28%, respectively).

View larger version (30K): [in this window] [in a new window]   Fig. 6.   and external pH dependence of light-dependent OH efflux in endodermal compartment. Fluxes are expressed as %control value (i.e., 3.55 ± 1.48 neq · min1 · mg protein1) as measured in FSW containing 1 mM Tris (pH 8.2) during a 15-min period. -free seawater (0- SW) or FSW was added to endodermal and/or ectodermal compartments as indicated at top of figure.

Dependence on external Na⁺ and amiloride-sensitive H⁺ excretion. Na⁺-dependent H⁺ excretion is well known in vertebrate and invertebrate cells (2). To study the Na⁺ dependence of OH excretion, we tested the effect of 0-Na⁺ ASW. Figure 7 shows that incubation of the ectodermal cell layer in 0-Na⁺ ASW triggered a net OH excretion by this cell layer. This stimulation was totally cut by 500 µM amiloride, suggesting that the OH excretion observed in 0-Na⁺ ASW was due to the reversal of the Na⁺/H⁺ exchange (leading to H⁺ uptake) after modification of the Na⁺ electrochemical gradient which energizes this antiport. The same effect was observed when the endodermal cell layer was incubated in 0-Na⁺ ASW. In this case, an OH excretion triggered by 0-Na⁺ ASW was superimposed on the light-dependent OH excretion, leading to a twofold increase in this parameter. This OH excretion was sensitive to amiloride. When 0-Na⁺ ASW bathed the two compartments, an amiloride-sensitive OH excretion by the two layers of cells was triggered. This set of experiments demonstrated the presence of a Na⁺/H⁺ exchange on both layers of cells. This exchange did not apparently play a major role in the light-dependent OH excretion.

View larger version (33K): [in this window] [in a new window]   Fig. 7.   External Na+ dependence of OH efflux in ectodermal (hatched bars) or endodermal (open bars) compartments. Fluxes are expressed as %control value (i.e., 3.55 ± 1.48 neq · min1 · mg protein1). Na+-free artificial seawater (0 Na+), FSW, and/or amiloride (Amil, 500 µM) was added to compartment facing endodermal and/or ectodermal compartment as indicated at top of figure.

Inorganic carbon incorporation into photosynthates. To confirm the external supply of inorganic carbon, we carried out measurements of DIC incorporation into photosynthates. Experiments were carried out in both Ussing chambers and tentacle bags. Figure 8 shows that in the Ussing chamber, when the pH of the endodermal compartment was that of natural seawater, i.e., 8.2, ~75% of inorganic carbon came from the endodermal compartment. When the pH of the endodermal compartment increased to 9.2, only 25% of inorganic carbon was supplied from this compartment. In the tentacle bag, a similar pattern was observed with a more pronounced effect: the inorganic carbon supply from the endodermal compartment decreased from ~40% for an initial pH of 8.2 to 6% for a pH of 9.2. 

View larger version (18K): [in this window] [in a new window]   Fig. 8.   Inorganic 14C incorporation into photosynthates. Experiments were carried out in Ussing chamber or in tentacle bag, with initial endodermal pH of either 8.2 or 9.2. For each condition, ectodermal (open bars) or endodermal (closed bars) supply was expressed as %total incorporation (i.e., for initial pH of 8.2 and 9.2 in Ussing chamber, 1.94 ± 0.35 and 1.42 ± 0.06 nmol · min1 · mg protein1, respectively; in tentacle bag, 3.47 ± 0.39 and 2.48 ± 0.33 nmol · min1 · mg protein1, respectively).

Alkalinity of the internal medium of tentacle bags. TA of the internal medium of tentacle bags exposed to light for 1 h increased from 2.626 ± 0.003 meq/l (TA of FSW pH 8.2) to 3.353 ± 0.591 meq/l. We could not measure any significant change of TA in the external medium due to its high volume related to the internal medium.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Our results demonstrate that under light conditions, the tentacle of the sea anemone Anemonia viridis generates a pH gradient of ~0.8 units across its epithelial layers, with the endodermal face being alkaline. Measurements in a bag or in the Ussing chamber were corroborated by the direct measurement of internal pH within a whole sea anemone. Our results showed that in vivo, the endodermal cell layer was subjected to cyclic variations of pH, from acidic in darkness to alkaline in light conditions. Furthermore, this paper confirms previous results (6, 9) showing that inorganic carbon is absorbed from the seawater pool by the ectodermal cells and then transferred to endodermal cells where a net OH secretion (or H⁺ uptake) occurs. One of the functions of the epithelial layers of the tentacle of the sea anemone is then to transport across the epithelial barrier as renal tubular or gastric epithelia.

OH secretion occurred originally in the internal medium of the tentacle, as demonstrated by the titration experiment carried out in the Ussing chamber. This net secretion of OH was further confirmed by the light-dependent increase of TA in the internal medium. Indeed, if both CO₂ absorption and H⁺/OH fluxes affect pH, only H⁺/OH (or , ) fluxes affect alkalinity (27). As the pH gradient across the tentacle became maximal (i.e., after 20 min of saturating light), a slight diffusion of OH took place from the internal toward the external compartment, as shown by the lag phase of 24 ± 5 min before the onset of titration in this medium. Although the increase of pH in the internal medium plateaued at ~9 after 20 min, the rate of O₂ evolution remained linear. An inhibition of photosynthesis when extracellular pH becomes alkaline has been recorded in various photosynthetic organisms, including marine microalgae (26), cyanobacteria (14), and in the symbiotic association between corals and dinoflagellates (12). Such inhibition has been attributed to low levels of at this pH. In contrast to these results, only one side of the photosynthetic cell (i.e., the apical membrane of the endodermal cell) of the sea anemone tentacle was bathed in alkaline pH. The ectodermal side of the tentacle remained around pH 8.2, thus allowing the normal supply of DIC. Such functional polarization was found to an even larger extent in various macroalgae and angiosperms. In Potamogeton, the pH at the upper side increased to a value of ~11, whereas at the lower side the pH decreased to a value as low as 4.0 (21). Although the function of polarity in these organisms is still unclear, it has been suggested that a pH or proton motive force is needed to drive uptake or -to-CO₂ conversion (22). In the giant hydrothermal vent tubeworm Riftia pachyptila, symbiotic with intracellular carbon-fixing sulfide-oxidizing bacteria, it has been recently demonstrated that the maintenance of an alkaline pH within the extracellular fluids of the tubeworms acts to concentrate inorganic carbon (11). However, the rate of transepithelial diffusion which usually supplies ~16% of symbiont photosynthesis (9) did not present any significant difference when the tentacle was perfused with FSW at pH 8.2 or 9.0 (unpublished data). The generation of a transepithelial pH gradient in sea anemones is therefore probably a part of the carbon-concentrating mechanism (6) rather than being the carbon-concentrating mechanism itself.

If a net OH secretion seemed to occur only in the endodermal compartment, our results show unambiguously that this light-sensitive excretion may be modulated by inhibitors (Fig. 5) or by use of -free SW (Fig. 6) within the ectodermal compartment. This inhibition resulted specifically from an effect on ectodermal cells and not from diffusion of inhibitors through the tentacle. The requirement for external was further supported by the measurement of dissolved inorganic ¹⁴C incorporation into photosynthates. Whatever the biological preparation, our results demonstrate that at least a part of the carbon incorporated into photosynthates was supplied by ectodermal cells. The extent of this supply changed according to the preparation and the initial internal pH conditions (Fig. 8). For an initial pH of 8.2, the difference observed between the Ussing chamber and tentacle bag was likely the consequence of DIC availability (6). In the Ussing chamber, the availability of DIC between the two compartments being the same, the uptake of DIC was performed mostly by endodermal cells. In tentacle bags, which are supposed to be closest to in vivo conditions, the internal volume was 25-fold lower than the external one, and the internal DIC availability was not high enough to supply symbiont photosynthesis, leading to a predominant uptake from the external compartment. The extent of ectodermal supply was a function of endodermal pH (which governs carbon speciation). The endodermal supply decreased to nearly zero as light-dependent pH increase occurred, suggesting that the transport capacity of the ectodermal layer was stimulated to compensate for any decrease of direct endodermal supply. These results demonstrate that the epithelial layers of the tentacle displayed a great plasticity according to DIC availability.

Bénazet-Tambutté et al. (6) demonstrated that part of the DIC used for symbiont photosynthesis was supplied by a DIDS-sensitive DIC uptake site located in the membrane of ectodermal cells. Similarly, the present results, using either the anion inhibitor DIDS or -free SW, demonstrated a dependence of OH secretion by endodermal cells on availability in the ectodermal compartment. These results demonstrate that a carrier is present in the membrane of ectodermal cells. The involvement of a secondary active transport of in sea anemone tentacles is consistent with the fact that such transports are generally involved in transepithelial acid transport in leaky epithelia (7) such as the tentacle of the sea anemone (5).

uptake may also be involved in membrane-bound carbonic anhydrase-catalyzed dehydration of followed by CO₂ uptake, leading to extracellular pH increase (4). A Diamox and EZ-sensitive process was found in ectodermal cells; however, EZ, which is thought to be more membrane permeable than Diamox (1), led to a lower inhibition of OH excretion by endodermal cells. A possible explanation for this paradoxical result is a nonspecific effect of Diamox. Diamox is known to inhibit both -ATPase in the blue crab gill and a Cl pump in intestine (10, 17). Almost 50% of OH secretion was insensitive to Diamox, suggesting that at least one-half of DIC was supplied either by direct uptake of followed by intracellular dehydration or by chemical conversion. When Diamox was added to both compartments, its effect was not additive, as previously shown (6) for O₂ evolution, suggesting either a common target on both faces located at the interface of the two layers of cells or at least two targets in series on each epithelial cell layer. EZ added within the endodermal compartment led to a total inhibition of OH secretion, suggesting, as previously shown (1, 6, 29), that carbonic anhydrase is located at the interface of the two symbionts and provides a unidirectional flux of CO₂ by dehydrating into CO₂.

Chemical conversion between and CO₂ may be mediated by H⁺ excretion. A luminal Na⁺/H⁺ exchanger was demonstrated to mediate transcellular flux in the absorptive direction by intraluminal CO₂ generation in various absorptive epithelia, such as renal proximal tubule (2), Necturus gallblader (28), and rabbit intestine (16). We demonstrated the presence of an amiloride-sensitive Na⁺/H⁺ exchanger in the tentacle of the sea anemone that seems to play no role in transport. H⁺ excretion may also be mediated by an H⁺-ATPase. Experiments carried out with DES suggest that such an enzyme was involved in transport. We observed a similar inhibition (~50%) when DES was added to the internal medium or both the internal and external media, suggesting the involvement of two H⁺-ATPases in series within each epithelial cell layer. It is possible that these H⁺-pumps facilitate the diffusion of DIC between the two cell layers by acidifying the mesoglea.

The process involving transport at the membrane level implies that the dehydration reaction of into CO₂ is intracellular. Numerous transport systems have been suggested to account for OH secretion: an electrogenic /OH antiport, H⁺- cotransport system (15, 18), or Cl/OH antiport (25). The fact that OH secretion was inhibited by 90% when was lacking from the endodermal compartment whereas O₂ evolution was only inhibited by 40% (6) suggests a possible coupling between OH and . Such an antiport would be neutral (18). When the internal medium becomes alkaline, the OH concentration gradient inhibits the OH efflux, whereas at acidic pH the opposite is observed. At pH 6.6 the OH concentration gradient will favor OH efflux (or H⁺ influx). Surprisingly, even when -free SW was present in both compartments, a light-induced OH efflux was still recorded, whereas no change was recorded in the ectodermal compartment. This demonstrates that this OH efflux was specific to the endodermal cell layer, suggesting a higher OH (H⁺) cell permeability of endodermal cells than ectodermal cells. Such a different permeability may play a role in intracellular pH regulation during photosynthesis.

Our results demonstrate that the epithelial layers of the sea anemone tentacle carried out transepithelial transport and OH secretion like numerous other epithelia. However, mechanisms underlying this process appeared to be different from other animals, thus providing a new model for studying transport. A pH gradient across the tentacle, resulting from the activity of a carbon-concentrating mechanism, was generated by light. Such a functional polarity has been previously shown to play a role in utilization in macrophytes (22). This role can now be extended to symbiotic anthozoans. It is likely that this phenomenon also occurs in scleractinian corals, in which the oral epithelial layers are similar to those of sea anemones. If this is the case, an alkaline coelenteric space may facilitate diffusion of H⁺ produced by the calcium carbonate precipitation following the reaction which consequently results in enhancement of calcification. A role of H⁺ in coupling photosynthesis and calcification has been previously proposed (20). However, in contrast to this earlier study, our model does not allow calcification to stimulate photosynthesis. Thus the polarized oral epithelial layers of anthozoans may play some role in light-enhanced calcification by coupling OH production by endodermal cells and H⁺ production by calcification.

Perspectives