Potential role of sulfide and ammonia as confounding factors in elutriate toxicity bioassays with early life stages of sea urchins and bivalves

https://doi.org/10.1016/j.ecoenv.2005.12.008Get rights and content

Abstract

This work reports some considerations on the possible contribution of sulfide and ammonia to the toxicity of elutriate samples of sediments from the Venice lagoon, tested with a battery of bioassays using early life stages of the sea urchin Paracentrotus lividus and the oyster Crassostrea gigas. A comparison of ammonia or sulfide concentration in the test matrix, matrix toxicity, and the sensitivity limit of bioassays for ammonia or sulfide were used in evaluating toxicity data. Results highlighted that sperm cell and embryo toxicity of elutriates were not affected by sulfides. Neither was any direct relationship shown between elutriate toxicity and ammonia concentration. Most elutriates had ammonia concentrations below the sensitivity limit of acute test methods, while the more sensitive subchronic toxicity tests were affected by ammonia interference in some samples.

Introduction

In a monitoring program, a first question could be, “Which toxicity test is applicable to this environment?” followed by, “How reliable will the data be?” Evaluation of the applicability of toxicity bioassays to an environment involves different steps, with progressive iterative evaluations of methods and recalibration of procedures that could slightly modify operative protocols. A central phase of this validation process is evaluation of the test applicability to the environmental matrices and of the test's discriminatory power, focusing on identification of possible chemical or physical interference. Some important variables in test matrices (elutriates and, particularly, pore waters), called “confounding factors,” occur naturally and interfere with the biological effects of micropollutants, frequently giving rise to a “false positive” (Giesy and Hoke, 1990; Matthiessen et al., 1998; Morin and Morse, 1999; Postma et al., 2002; Thursby et al., 1997) and labelling a sample as toxic that is not so in reality.

Recognized confounding factors include sediment texture (De Witt et al., 1988) and matrix parameters such as temperature, salinity, and dissolved oxygen, ammonia, and sulfide concentrations (Lapota et al., 2000; Postma et al., 2002). Furthermore, the method for extracting and diluting test matrices may influence sample hypoxic condition and natural toxicant concentrations and, consequently, the final toxicity (Beiras, 2002).

In recent years, we have been studying the applicability to a transitional environment of bioassays using the early life stages of the sea urchin (Paracentrotus lividus) and oyster (Crassostrea gigas), with the Venice lagoon as a case study (Arizzi Novelli et al., 2003a; Volpi Ghirardini et al., 2003).

In these toxicity bioassays, the most important confounding factors are the presence of ammonia and sulfide; during the test, temperature and salinity are kept within the optimal range for organisms and dissolved oxygen has values over 80% of saturation. Many authors have reported recently that the toxicity of elutriates and pore waters from marine and coastal environments was influenced by ammonia when routine tests were used (Ankley et al., 1991; Carr et al., 1996a, Carr et al., 1996b; Postma et al., 2002; Whiteman et al., 1996). No normal ranges of ammonia and sulfide in elutriates and pore waters are specified in the literature. For surface waters, the US EPA proposed a limit of 2 μg/L for sulfide and 35 μg/L for ammonia as quality criteria (Wang and Chapman, 1999).

Moreover, the sensitivity of the test methods used in monitoring towards these substances is often not available. With sea urchin bioassays, for example, ammonia toxicity data are available for Arbacia punctulata (Carr et al., 1996a; Hooten and Carr, 1998) and Strongylocentrotus purpuratus (Bay et al., 1993; Greenstein et al., 1996), and sulfide data are available for S. purpuratus (Knezovich et al., 1996), but no information has been found for a test with the species P. lividus. For this reason our research group recently studied the sensitivity of sperm cell and embryo toxicity test with P. lividus toward ammonia and sulfide (Arizzi Novelli et al., 2003b; Losso et al., 2004a, Losso et al., 2004b).

In transitional environments, such as the Venice lagoon, naturally rich in organic matter and often characterized by anoxic conditions in areas with low water turnover, concentrations of ammonia and sulfide can be defined ad hoc, according to the considerations recently proposed for estuaries (Chapman and Wang 2001).

In polluted environments, a high concentration of ammonia in sediments can be due mostly to human activities directly introducing ammonia into waters or increasing organic matter in sediments. In this case, ammonia is considered a pollutant and not a confounding factor. To the best of our knowledge, no studies on “background” concentrations of ammonia and sulfide in coastal and/or marine environments are available that make it possible to discriminate between anthropogenic and natural ammonia and/or sulfide.

What then can be done with environmental samples at high concentrations of ammonia and/or sulfide? International protocols recommend that concentrations of these substances in tested matrices not exceed the sensitivity limits of bioassays (American Society for Testing and Materials (ASTM), 1998). This means that most lagoon elutriates or pore waters cannot be tested if even a part of the ammonia concentration might be derived from human sources. We also have to consider that test matrices are diluted in several solutions in order to obtain EC50 (Effective Concentration 50) values, so this also means dilution of the ammonia and/or sulfide concentration. For this reason, the best solution may be to test the samples at high concentrations of ammonia and/or sulfide and then discuss the obtained toxicity data, taking into account all available information.

This work aims at the following:

  • Continuing to evaluate the applicability of the test with sea urchins and oysters to the elutriates of the Venice lagoon; the discriminatory power of the test with P. lividus has already been evaluated (Losso et al., 2004b; Volpi Ghirardini et al., 2003).

  • Acquiring new data on elutriate toxicity for the Venice lagoon using sea urchin and bivalves.

  • Evaluating if toxicity data are affected by confounding factors, comparing ammonia and sulfide concentrations in the test matrix, test matrix toxicity, and the sensitivity limit of bioassays for ammonia and sulfide.

Section snippets

Sperm cell and embryo toxicity tests with P. lividus

Adults of P. lividus were collected from a rocky breakwater in the northern Adriatic Sea, near the barrier island of Pellestrina (45°18′50″N; 12°19′24″E), on the seaward side of the Lagoon of Venice, Italy. Sea urchins were kept for up to one month in flowing seawater from the sampling site, maintained on macroalgae and molluscs, at a temperature of 18±1 °C and salinity of 35±1‰, with a natural photoperiod. Adult collection and culture procedures have previously been reported in detail (Volpi

QA/QC for toxicity tests

Tests with P. lividus showed good repeatability, evaluated using copper as a positive control: the sperm cell toxicity test had a mean EC50±SD (standard deviation) of 55±8 μg/L (CV (coefficient of variation)=15%, n=25), which was within the EC50 acceptability range (39–71 μg/L) (Volpi Ghirardini and Arizzi Novelli, 2001); the embryo toxicity test highlighted a mean EC50±SD of 68±9 μg/L (CV=13%, n=9), within the EC50 acceptability range (51–87 μg/L) (Arizzi Novelli et al., 2002).

For the

Conclusion

This study reports toxicity data on elutriates from sediment of the Venice lagoon, tested with a battery of bioassays using the early life stages of sea urchin and oyster. The possible contributions of sulfide and ammonia to toxicity are discussed, as an evaluation of false positives is an important step in method validation and in particular for evaluating the applicability of methods to a given environment.

The general approach in monitoring programs, suggested by the ISO or OECD standards,

Acknowledgment

This work was partly granted by the Consorzio Ricerche Laguna (Co.Ri.La.) of Venice (Italy). Alison Garside revised the English text.

References (33)

  • A. Arizzi Novelli et al.

    Toxicity of heavy metals using sperm cell and embryo toxicity bioassays with Paracentrotus lividus (Echinodermata: Echinoidea): comparisons with exposure concentrations in the Lagoon of Venice (Italy)

    Environ. Toxicol. Chem.

    (2003)
  • A. Arizzi Novelli et al.

    Ammonia as confounding factor in toxicity tests with the sea urchin Paracentrotus lividus (Lmk)

    Toxicol. Environ. Chem.

    (2003)
  • Bay, S., Burgess, R., Nacci, D., 1993. Status and application of Echinoid (Phylum Echinodermata) toxicity test methods....
  • R. Beiras

    Comparison of methods to obtain a liquid phase in marine sediment toxicity bioassays with Paracentrotus lividus sea urchin embryos

    Arch. Environ. Contam. Toxicol.

    (2002)
  • R.S. Carr et al.

    Sediment quality triad assessment survey of the Galveston Bay, Texas system

    Ecotoxicology

    (1996)
  • S.R. Carr et al.

    Sediment quality assessment studies of Tampa Bay, Florida

    Environ. Toxicol. Chem.

    (1996)
  • Cited by (0)

    View full text