MARINE ECOLOGY PROGRESS SERIES Vol. 169: 295-301, 1998 Published August 6 Mar Ecol Prog Ser p Ammonium contribution from boring bivalves to their coral host -a mutualistic symbiosis? Ofer ~okady'.',Yossi ~oya~,Boaz ~azar~ 'Institute for Nature Conservation Research and 'Department of Zoology, George S. Wise Faculty of Life Sciences, and the Porter Super- Center for Ecological and Environmental Studies, Tel-Aviv University, Tel-Aviv 69978, Israel Institute of Earth Sciences and Moshe Shilo Minerva Center for Marine Biogeochemistry, The Hebrew University of Jerusalem. Jerusalem 91904, Israel ABSTRACT: The mytilid bivalve Lithophaga simplex is found within its tissue, offers an example for such a nitrogen- to inhabit the scleractinian coral Astreopora myriophthalma recycling system (Hawkins & Klumpp 1995). in high densities. This boring bivalve, living inside the CaC03 Aspects of ammonium uptake have been studied in skeleton of the coral, produces considerable amounts of ammonium as a nitrogenous waste product. Ammonium pro- many algae-invertebrate associations (e.g. Wilkerson duction rate by the bivalves and consumption rate by the & Trench 1986, Fitt et al. 1993). The association be- coral (via the symbiotic algae) were measured in laboratory tween hermatypic corals (Order Scleractinia) and the expenments. The population density of L. simplex bivalves dinoflagellate endosymbiont Symbiodinium sp. is a In A. myri~phth~lrnacorals was surveyed in the Nature Reserve Reef, Eilat, Red Sea, Israel. Ammonium production very common symbiotic relationship in tropical and rate by the bivalves, inhabiting the coral at a density of 0.22 +_ subtropical marine environments (Falkowski et al. 0.11 bivalves cm-', is calculated to be 8.2 i 3.8 and 3.5 1984). In these symbioses the algae (zooxanthellae) 1.6 nmol (cm2 coral)-' h-' during daytime and nighttime, serve as primary producers that translocate organic respectively. Under conditions of low ammonium concentra- carbon to their coral host (Muscatine et al. 1989). They tion (0.2 to 1.2 pM)the consumption rate of the coral ranged between 5 and 22 nmol cm-' h-' Thus, under naturally occur- also take up dissolved inorganic nutrients, which, ring levels of ammonium (<0.15FM), recycling of nitrogenous together with nutrients obtained from consumption of waste produced by the bivalves (ammonium)may account for zooplankton by the coral host, are retained and recy- a significant portion of the needs of the coral/zooxanthellae. cled within the association (Falkowski et al. 1984, In contrast to the generally accepted view of boring bivalves as parasites of their coral hosts, it is hypothesized that the Rahav et al. 1989). Mechanisms of ammonium uptake association between L. slmplex and A. myriophthalma may have been studied both in the coral host (Sumn~onset also be an example of mutualistic symbiosis. The results indi- al. 1986) and in the zooxanthellae (Dudler & Miller cate a possible pathway in the biogeochemical cycle of nitro- 1988). It has been shown that assimilation of inorganic gen in the coral reef environment. nitrogen compounds is done by the zooxanthellae (Muscatine & D'Elia 1978, Muscatine et al. 1979, Burris KEY WORDS: Ammonium Boring bivalves . Symbiosis - Nitrogen cycle Reef ecology 1983, Wilkerson & Trench 1986). The involved bio- chemical pathways have been addressed both in iso- lated zooxanthellae and in the intact symbiosis (e.g. In oligotrophic waters typical of coral reef environ- Gunnersen et al. 1988, Bythell 1990, Wafar et al. 1993, ments, levels of dissolved N in the water column are Yellowlees et al. 1994). Physiological effects of ammo- very low (Entsch et al. 1983). Thus, biological systems nium enrichment on corals and their symbionts have that conserve nitrogen and/or recycle metabolized also been investigated in many studies (e.g. Hoegh- nitrogen, such as ammonium, have an advantage over Guldberg & Smith 1989, Hoegh-Guldberg 1994, systems that do not (Szmant-Froelich & Pilson 1977). Muller-Parker et al. 1994, Stambler et al. 1994). The giant clam Tridacna gigas, hosting algal symbionts Excretions of nitrogenous waste products by marine animals may serve as nitrogen sources for organisms that are capable of utilizing them. Meyer et al. (1983) recorded the high level of ammonium excreted by 0 Inter-Research 1998 Resale of full article not permitted 296 Mar Ecol Prog Sc haemul~dfish schools resting over coral colonies. The [<0.15pM as measured by Korpal (1991)and <0.1 pM excretions of the sea urchin Diadema antillarum were from our own monitori.ngl. Shortly before the expen- suggested to provide ammonium-N for algal turfs upon ments, the experimental vessels (beakers or aquarla) which the urchin grazes, thus increasing the nutrient- were thoroughly rinsed with 10% HC1 and double- limited primary productivity of the algae (Williams & distilled water and then rinsed with FSW. Carpenter 1988). Spotte (1996)investigated the supply Daytime experiments were conducted under labora- of regenerated nitrogen to sea anemones by their sym- tory illumination (ca 10 pE m 'S-'), at 26 to 28OC. Night biotic shrimps. Dense bivalve populations were shown experiments were carried out in the dark at 24 + 1°C. to play an important role In nitrogen regeneration In all the experiments described below, analysis of (Prosch & McLachlan. 3.984). samples for ammon.i.um concentration followed Strlck- Species of the mytilid bivalve Lithophaga are found land & Parsons (1972). Br~efly,duplicate samples of boring inside the skeletons of many species of live scle- 25 m1 were taken with pre-rinsed syringes at each ractlnian corals (reviewed by Morton 1983). These sampling time and kept in dark-stained bottles for bivalve-coral associations are very common in the olig- ammonium analysis. The reagents for ammonium otrophic waters of the reefs of the northern Gulf of analysis were immediately added to the samples, Eilat, Red Sea (Loya 1981). The bivalve L. simplex and sample bottles were kept for 2 h in the dark for Iredale (1939) forms such an association with the mas- color development. The reaction results in production sive coral Astreopora myriophthalma (Lamarck, 1816). of a blue color, the intensity of which is dependent This coral forms large colonies, often carrying dense upon the concentration of ammonium. Absorbency at populations of L. simplex. The bivalves are completely 640 nm was read with a Beckman DU~-~spectropho- enclosed by the coral skeleton, with only the ends of tometer equipped with a 10 cm long optical cell. The their siphons reaching out to the surrounding seawater average standard deviation between duplicate analy- (see Fig. 1). Known benefits for the bivalves from thls ses was 0.13 yM (typically in the range of 1 to 5%).The association include protection against predators and amount of ammonium present in the experimental the nutritional use of coral mucus (Shafir S1 Loya. 1983). vessels (in pmol) at each sampling time was calculated However, possible gains for the coral are less clear at by multiplying the ammonium concentration (in PM) present. We have conducted experiments to examlne by the volume of FSW (in liters) at the time of sam- the possibility that L. simplex contributes ammonium pling. to its coral host A. myriophthalma (to be utilized by Three groups of 8 Lithophaga simplex each, care- endosymbiotic zooxanthellae), a.nd that the coral- fully removed from Astreopora myriophthalma corals, bivalve sym.biosis is mutualistlc. were used to measure the rate of ammonium excretion Materials and methods. Astreopora myriophthalma by the bivalves. Each group was placed in a beaker corals containing Lithophaga simplex bivalves were filled with 250 m1 FSW. Bivalves quickly relaxed in the collected in the northern part of the Gulf of Eilat, Red experimental vessels, opening their valves and extrud- Sea, at a depth of 6 to 12 m. Bivalves were extracted ing the siphons and mantle-margins within a few min- from the coral colonies by gently breaking the coral utes. Day and night experiments lasted 6 h and sam- skeleton. The resulting coral fragments, used in the ples were taken at the beginning of the experiment experiments below, were inspected for infestation by and every 2 h thereafter Excretion rates were calcu- other organisms (including L. sirnplex biva.lves). The lated from regression slopes of the total ammonium dead bottom parts of all fragments were partially cov- contents in the experimental vessels versus time. Fol- ered (<10% area) by encrusting organisms (sponges, lowing the experiments, bivalve tissue was removed tunicates, bryozoans). In most cases it was impossible from the shell, dried and weighed (each group sepa- to extract 2 or 3 of the bivalves without excess break- rately). age (see column 1 in Table 1; corals I and I1 used in the The relationship between bivalve ammonium supply spiking experiments contained 3 and 4 bivalves, rate and coral consumption rate was determined in respectively). Animals were allowed 12 to 24 h for vitro using ammonium excretion by live bivalves and acclimation and recovery in aquaria with running sea- ammonium supply controlled, with a peristaltit pump. water prior to the experiments. Experiments were typ- The pump enabled us to increase (double) the bivalve ically performed during a period of 3 to 5 d, and the supply rate without extracting more bivalves from corals were returned to the sea (a flat surface at a corals. Live bivalve experiments were conducted depth of 9 m) for long-term storage (several weeks). by incubating each of 4 Astreopora myriophthalma Laboratory experiments were conducted with surface colonies with a group of 10 to 14 Lithopbaga slmplex seawater sampled close to the marine laboratory and bivalves (a total of 4 independent experiments) in 750 filtered through a 0.22 pm Millipore filter (FSW).This to l000 m1 FSW. Ammonium production rate was mea- seawater contains very low ammonium concentrations sured for the bivalve groups serving as an ammonium Mokady et al.