Nitrification controls on fluxes and isotopic composition of nitrate from Florida Keys sponges

Abstract

We performed a combination of field and laboratory experiments designed to investigate sponge-hosted microbial nitrification processes, plus a survey of the prevalence of nitrification among sponge species at patch and outer reefs, Florida Keys, USA. Results from dissolved inorganic nitrogen (DIN) measurements, chemical inhibition, and ¹⁵N isotope tracer experiments with Caribbean sponges show that 9 of 12 species tested hosted nitrification. Of the species in our study that are reported to host internal microbial populations, all but one has been found to host this process. Nitrification occurred with an apparent N isotopic fractionation factor of 11 ± 2.6‰ (standard error) and was inhibited by nitrapyrin. Both of these characteristics are consistent with bacterially mediated ammonium oxidation; however, ammonium oxidation by crenarchaeota is also possible. The isotopic composition of nitrate expelled from sponges in situ had lower δ¹⁵N values than nitrate from the ambient water column, suggesting active utilization of the nitrate product after it is released. The frequency of the association between nitrifying microbial communities and sponges that harbor large microbial communities suggests that nitrification is an integral part of the metabolic function of many sponge species. Given the abundance of these species on Conch Reef and the magnitude of their nitrification rates, the majority of benthic nitrification on the Florida Keys outer reef tract probably occurs in sponges. Sponge population size and composition could therefore strongly influence the concentration and speciation of DIN in the reef water column, and possibly other ecosystems where sponges are abundant.

Introduction

Nitrate fluxes as high as 12 ± 2.2 mmol m^− 2 d^− 1 have been reported from sponges on Caribbean reefs (Corredor et al., 1988). This implies a nitrification rate that far exceeds those measured in other marine benthic systems such as microbial mats (1 mmol m^− 2 d^− 1Bonin and Michotey 2006), coral reef sediments (1.68 mmol m^− 2 d^− 1Capone et al., 1992), and temperate continental shelf sediments (2.1 mmol m^− 2 d^− 1Hopkinson et al., 2001). To date, weight-specific rates of nitrification have been reported for five tropical (Corredor et al., 1988, Diaz and Ward, 1997) and six temperate sponge species (Jimenez and Ribes, 2007) and benthic fluxes of nitrate have been calculated for only 3 species (Corredor et al., 1988, Diaz and Ward, 1997). These reported rates and the abundance of sponges on some coral reefs (up to 3.62 L m^− 2, Southwell, 2007) suggest that sponge-hosted nitrification may strongly influence DIN speciation in coral reef waters. However, surprisingly little is known about this process, such as the frequency of occurrence among the species, environmental controls on rates, and the fractionation of stable isotopes that is expected to occur during nitrification (Mariotti et al., 1981).

It was once thought that ammonium oxidation, the first step in nitrification, was performed only by beta- or gamma-proteobacteria. However, recent studies have shown that crenarchaeota from Marine Group 1 are capable of oxidizing ammonium, and may constitute a large portion of the oceanic ammonium oxidizing community (Karner et al., 2001, Francis et al., 2005, Francis et al., 2007). Both bacterial nitrifiers and Marine Group 1 archaea have been detected in sponges, along with a wide variety of other microbial organisms (Taylor et al., 2007, and references therein). Thus, it is presently unclear whether archaea or bacteria (or both) might be responsible for the impressive rates of ammonium oxidation observed in sponges. Although both bacterial and archaeal ammonium oxidizers appear to perform the same biogeochemical function, there may be important differences in the rates, carbon fixation mechanisms, environmental controls, or competitive ability between these groups. Therefore, the identity of the microbial partner may, in fact, influence the biogeochemical processes that occur as a result of this association.

The release of nitrate from sponges has been interpreted as evidence for microbially-mediated nitrification (Corredor et al., 1988, Diaz and Ward, 1997, Jimenez and Ribes, 2007) because animals typically release inorganic metabolic waste in the form of ammonium rather than nitrate. Although nitrate release from sponges was first observed almost 20 years ago, no studies exist, to our knowledge, that directly test this empirical evidence or that determine the prevalence of nitrification among the large number of species (up to 300) found on tropical reefs (Diaz and Rutzler, 2001, and references therein). Furthermore, nothing is known about nitrogen isotopic fractionation that we expect to occur during sponge-hosted nitrification (Mariotti et al., 1981). Such investigations could provide information about microbial communities performing the process because the magnitude of isotopic fractionation partially depends on the enzymes employed in the reaction (Casciotti et al., 2003). However, these effects may be confounded by environmental factors such as oxygen depletion within the sponge tissue (Hoffmann et al., 2005). Stable isotopic analysis of sponge-produced nitrate may also provide clues to concurrent biogeochemical processes that have been speculated to occur within sponges, such as assimilatory or dissimilatory uptake of nitrate or nitrite (Southwell, 2007, Taylor et al., 2007).

Sponges can be a large source of DIN on coral reefs where they are abundant (Corredor et al., 1988, Diaz and Ward, 1997, Jimenez and Ribes, 2007), and the partitioning of these nutrients between benthic and pelagic primary producers could be an important aspect of nutrient recycling and benthic–pelagic coupling. Stable isotopic compositions have been used extensively to trace sources of nutrients in marine organisms (Marguillier et al., 1997, Sammarco et al., 1999, Heikoop et al., 2000) and to indicate specific transformations of inorganic nitrogen species (Sutka et al., 2004, Sigman et al., 2000). Therefore, characterizing the isotopic composition of the nitrate expelled by sponges is a logical first step for investigating the reaction. If the isotopic composition can be successfully constrained, then the relative importance of sponges as a source of nitrate for the coral reef food web could also be investigated using isotope mass balance. This study presents 1) an updated list of nitrate or nitrite production in sponges that expands the number of known species hosting this process to twenty-one, 2) multiple lines of direct evidence for microbial conversion of ammonium to nitrate within sponges, and 3) isotopic analysis of sponge nitrate from both in situ and laboratory incubations. These results are used to generate models of N-cycling processes based on isotope mass balance of nitrate and to provide new insights into the provenance and fate of nitrate produced in sponges.