' ~ - I PARTICULATE PHYTOCHELATINS AS AN INDICATION: l ""' - ~ I -. -- r -· 'I -··· OF METAL STRESS IN NATURAL WATERS

by

SILVIA KEIKO KAWAKAMI

A thesis submitted to the University of Plymouth in partial fulfilment for the degree of

DOCTOR OF PHILOSOPHY

School of Earth, Ocean and Environmental Sciences Faculty of Science

September 2004 University of P ymouth Library PARTICULATE PHYTOCHELATINS AS AN INDICATION OF

METAL STRESS IN NATURAL WATERS

Silvia Keiko Kawakami

ABSTRACT

The aim of this study was to investigate the metal stress response of phytoplankton in the

Fal, Plym and Tamar Estuaries (SW England, UK) by field measurements of particulate phytochelatms (PCs) and glutathione (GSH) and their relationships with Cu speciation. A high performance liquid chromatographic method was optimised for the determination of particular~

PCs and GSH from estuarine waters. Dissolved Cu speciation was determined using well established cathodic stripping voltammetric methods. Glutathione was found throughout the

1 estuaries ranging from 0.6 to 274 J.lmol (g chl a)" , and PCs were detected mostly in the Fal samples

1 at concentrations up to 36 J.lmol (g chl a)" • Elevated PC production was observed in Restronguet

Creek (Fal Estuary) and Gunnislake (Tamar Estuary), the most metal mine impacted sites. For the

2002 survey in the Fal Estuary, GSH and PCs presented strong positive correlations with total

2 dissolved Cu and free Cu + (as log values). For this survey, it was observed that PC production was faster than GSH production under increasing total dissolved Cu concentrations. This indicated that

PC production is a more specific response to metal stress than GSH production in natural waters.

High variabilities in the particulate GSH and PC concentrations were likely caused by the heterogeneous phytoplankton composition within the estuaries and effects of metal interactions.

Combined exposure of Cd and Zn, for instance, caused antagonistic effects on PC production by the laboratory diatom culture Phaeodactylum lricomutum, probably due to competition for cellular binding sites. Other mechanisms may be involved in the phytoplankton defence system against metal concentrations, as not all species reported in those areas appear to be able to synthesise PCs.

11 13 Dissolved Cu-organic complexes with high conditional stability constants (K = I 0 - I 0 ) dominated Cu speciation in most of the samples from the Fal, Tamar and Thau Lagoon (France). forming an important mechanism to reduce metal toxicity.

- 2 - Table of Contents

LIST OF FIGURES...... 6 LIST OF TABLES...... 9 LIST OF ABBREVIATIONS...... 11 ACKNOWLEDGEMENTS...... 12 AUTHOR'S DECLARATION...... 14 I INTRODUCTION...... 16 2 PHYTOCHELATINS AS INDICATION OF METAL STRESS IN NATURAL WATERS...... 17 2 .I . Abstract...... I 8 2.2. Introduction...... 19 2.3. Thiols as intracellular metal-binding peptides...... 23 2.3.1. Glutathione...... 23 2.3.2. Phytochelatins...... 26 2.3.2.1. Phytochelatins and metal tolerance...... 27 2.4. Stimulation of phytochelatin production in laboratory phytoplankton monocultures...... 28 2.4.1. Phytochelatin production by different phytoplankton species...... 29 2.4.2. Effects of various metals on phytochelatin production...... 32 2.5. Phytochelatins and glutathione produced by phytoplankton assemblages in natural waters...... 33 2.5.1. Field measurements ofphytochelatins and glutathione...... 34 2.6. Factors affecting the production of glutathione and phytochelatins in natural waters...... 37 2.6.1. Influence of trace metal speciation on phytochelatin production.... 38 2.6.2. Influence of light and nutrient supply on glutathione and 38 phytochelatin production ...... 2.7. Application ofphytochelatins for environmental purposes...... 39 2.7.1. Use ofphytochelatins as biomarkers of metal stress...... 39 2. 7 .1.1. The ratio PC:GSH in natural waters as an indication of metal stress ...... 40 2.7.2. Other applications for phytochelatins...... 41 2.8. Conclusions and final remarks...... 4L 2.9. References...... 4~

3 DETERMINATION OF PARTICULATE PHYTOCHELATINS AND GLUTATHIONE IN NATURAL WATERS USING HIGH PERFORMANCE LIQUID CHROMATOGRAPHY WITH FLUORESCENCE DETECTION...... 48 3. 1. Abstract...... 48 3.2. Introduction...... 49 3.3. Analytical methods for particulate GSH and PCs in natural waters...... 53 3.3.1. Samples: collection, handling and preparation...... 53 3.3.1.1. Reagents and laboratory ware...... 55 3.3.1.2. Sample collection...... 55 3.3.1.3. Filtration...... 55 3.3.1.4. Extraction ofPCs and GSH...... 56 3.3.1.5. Reduction reactions...... 57 3.3.1.6. Derivatisation orPCs and GSH...... 58 3.3.2. Sample analysis...... 61 3.3.2.1. GSH and PC standards...... 6 i 3.3.2.2. Phytochelatins produced by diatom cultures (Phaeodactylum tricornutum)...... 61 - 3 - 3.3.2.3. Stability ofGSH and PCs in the derivatised form...... 62 3.3.2.4. Chromatographic conditions...... 63 3.3.2.5. Reagent blank and samples...... 64 3.3.2.6 Attempts to identify PCs by LC-ESI-MS...... 67 3.3.2.7. Quality assurance...... 69 3.3.3. Application of the methods to natural waters...... 69 3.4. Conclusions...... 72 3. 7. References...... 74

4 EFFECTS OF METAL COMBINATIONS ON PHYTOCHELATINS AND GLUTATHIONE PRODUCED BY PHAEODACTYLUM TRICORNUTUM...... 78 4.1. Abstract...... 78 4.2. Introduction...... 79 4.3. Experimental...... 80 4.3.1. Culturing medium...... 80 4.3.2. Stock culture of Phaeodactylum tricornutum...... 83 4.3.3. Short-term metal exposure experiments...... 8<1 4.3.4. Analysis of phytochelatins and glutathione produced by P. tricornutum...... 85 4.3.4.1. Reduction with dithiothreitol (DOT)...... 85 4.3.4.2. Reduction with tributylphosphine (TCEP)...... 86 4.3.4.3. HPLC analysis...... 86 4.4. Results and discussion...... 87 4.4.1. Reproducibility of the short-term metal exposure experiments...... 87 4.4.2. Comparison between reducing reagents 88 4.4.3. Effects of combined metal additions on phytochelatins and glutathione production by P. tricornutum...... 90 4.4.3.1. Phytochelatins...... 91 4.4.3.2. Glutathione...... 93 4.4.3.3. Ratios PC:GSH...... 95 4.5. Conclusions and final remarks...... 97 4.6. References...... 98

5 PARTICULATE METAL-BINDING PEPTIDES AND COPPER COMPLEXATION IN THE FAL ESTUARY AND PLYMOUTH SOUND (UK)...... 101 5.1. Abstract...... I 01 5.2. Introduction...... 102 5.3. Study areas- environmetal setting...... I 03 5.3.1. Fa! Estuary...... 104 5.3.2. Plymouth Sound- Tamar and Plym Estuaries...... 106 5.3.3. Other contaminants and environmental quality of the Fal Estuary and Plymouth Sound...... 107 5.4. Experimental...... 109 5.4.1. Sample collection...... I 09 5 .4.2. Particulate phytochelatins and glutathione...... I 09 5.4.3. Total dissolved copper...... 110 5.4.4. Natural organic ligands and copper species...... 111 5.4.5. Chlorophyll a...... Ill 5.5. Results and discussion...... 112 5.5.1. Fa! Estuary...... 112 - 4- 5.5.1.1. Phytoplankton composition ofthe Fal Estuary...... 112 5.5.1.2. Particulate metal-binding peptides and total dissolved Cu in the Fa! Estuary...... 115 5.5.1.3. PC:GSH ratios...... 122 5.5.1.4. Copper complexation in the Fa! Estuary waters...... 124 5.5.1.5. Other dissolved metals and effects on the production of metal-binding peptides in the Fa! Estuary...... 129 5.5.2. Plymouth Sound- Tamar and Plym Estuaries...... 131 5.5.2.1. Phytoplankton composition of the Tamar Estuary...... 131 5.5.2.2. Particulate metal-binding peptides and copper complexation in the Tamar Estuary...... 132 5.5.2.3. Other metals in the Tamar Estuary...... 134 5.5.2.4. Particulate metal-binding peptides in the Plym Estuary and Plymouth Sound...... 135 5.6. Summary: metal-binding peptides in the Fa! Estuary, Plymouth Sound and other natural waters...... 138 5.7. Conclusions and final remarks...... 140 5.8. References...... 141

6 COPPER COMPLEXATION IN THE THAU LAGOON (FRANCE), A SHELLFISH FARMING AREA...... 145 6.1. Abstract...... 145 6.2. Introduction...... 146 6.3. Theory...... 148 6.4. Experimental...... 152 6.4.1. Sampling ...... 152 6.4.2. Determination of copper complexation in the Thau Lagoon...... 153 6.4.3. Determination of eh! a in the Thau Lagoon...... 153 6.4. Results and discussion...... 153 6.5. Conclusions...... 158 6.6. References...... 158

7 CONCLUSIONS AND FUTURE WORK...... 161 7 .1. The use of phytochelatins in environmental quality assessment...... 163

- 5 - List of Figures

Figure 2.1. Structure of glutathione 23

Figure 2.2. Biosynthesis of GSH from its constituent amino acids. Glu, glutamate; Cys, cysteine; Ser, serine; Gly, glycine (Norton & Foyer, 1998). 24

Figure 2.3. Processes involving GSH in eukaryotic algal cells (adapted from Ahner et al. 2002). 25

Figure 2.4. Basic structure of phytochelatins, (y-glutamyl-cysteinyl),-glycine, n = 1 for glutathione and n = 2-11 for phytochelatins. 26

Figure 3.1. Possible routes for the determination of thiol compounds (adapted from Lock & Oavis, 2002). 52

Figure 3.2. A summary of an analytical method for particulate phytochelatins and glutathione (oxidised + reduced forms) in natural waters (adapted from Rijstenbil & Wijnholds, 1996). 54

Figure 3.3. Stability of GSH and PCs in their bimane derivative form. Bars span the range from the average for triplicate injections. 63

Figure 3.4. Chromatograms for a) the reagent blank of the method; b) an estuarine water sample (Fa! Estuary, Aug-2002); and c) an extract of the marine diatom P. tricornutum (after exposure to 1.4 ~-tM Cu2+). 65

Figure 3.5. Mass spectra of GSH standard (5 ~-tM). a) without and b) with post- 68 column addition of 50% acetic acid in acetronitrile. Figure 4.1. Phaeodactylum tricornutum cells. 80

Figure 4.2. Growth curve for stock culture P. tricornutum in Aquil medium, May 2002. 83

Figure 4.3. Scheme of the short-term metal exposure experiment using P. tricornutum under single Cu, Zn and Cd additions and combination of these metals. 84

Figure 4.4. Calibration curves for a) glutathione; b) phytochelatin PC2. 87

Figure 4.5. Chromatograms of extracts of P. tricornutum a) control treated 2 with TCEP; b) from exposure to 4.9 ~-tM Cd + and treated with TCEP; c) control treated with OTT; d) from exposure to 4.9 ~-tM Cd2+ and treated with OTT. 89

Figure 4.6. Phytochelatins produced by P. tricornutum under metal exposure. Bars span the range from the average for duplicate analyses. 91

Figure 4.7. Glutathione and total phytochelatins produced by P. tricornutum under metal exposures, and respective PC 10 ~a 1 :GSH ratios. Mean values for duplicate analyses. 94

- 6 - Figure 5.1. Fa! Estuary (UK). Sampling site: I. Truro/Fal; 2. St Just; 3. St Mawes (Percuil system); 4.Falmouth Yacht Marina; 5. Between Mylor Creek and Pemyn; 6. Mylor Creek; 7. Between Mylor and Restronguet Creek; 8. Restronguet Creek; 9. Camon River. C I, C2, C3, C4, C5: sampling sites in the Camon River. 105

Figure 5.2. Plymouth Sound. Sampling sites in the Tamar Estuary (28th March 2003): 1. Gunnislake; 2. Morwelham; 3. Calstock; 4. Cotehele Quay; 5. Pink house; 6. Pentillie; 7. Weir Quay; 8. Lynher; 9. South ofTorpoint; 10 Sound near Devon's Point; 11. Drake Island; 12. Coxside Pier (2001); A. Halton Quay; B. Cargreen; C. Saltash; D. Tinside. Sampling sites in the Plym Estuary (2001-2002): Plym I, Plym2, Plym3, Plym4, Oreston. 106

Figure 5.3. Salinity and chlorophyll a for the Fa! Estuary surveys. a) August 2002 and October 2002 (marked with *); b) April 2003; c) July 2003. 113

Figure 5.4. Particulate PC2, GSH and dissolved Curotal in the Fa! Estuary. a) PC2 vs Curotal for Aug-2002 survey; b) GSH vs Curotal for the Aug-2002 survey; c) PC2 vs Curotal for April-2003 survey; d) GSH vs Curotal for April-2003; e) PC2 vs Curotal for July-2003; ./) GSH vs Curotal for July-2003. * collected in October 2002. Bars span the range from the average for duplicate analyses. Dotted histograms are below detection limit. 117

Figure 5.5. Fa! Estuary. a) GSH vs Cur0131 for the 2002 survey; b) PC2 vs Curotal for 2002; c) GSH vs Curotal for 2003; d) PC2 vs Curotal for 2003; e) GSH vs Log[Cu2+] for the 2002 survey; ./) PC2 vs Log[Cu2+] for the 2002; g) GSH vs Log[Cu2+] for 2003; h) PC2 vs Log [Cu2+] for the 2003 survey. Triangles correspond to data from April2003. 119

Figure 5.6. Concentrations of GSH and PC2, together with pH values and salinity for the samples from Camon River. Cl, C2 and C3 were collected in April 2003 and C4 and C5 in July 2003. 121

Figure 5.7. Comparison between PC2 concentrations and PC2:GSH ratios for the Fa! Estuary samples. a) August 2002; b) April 2003; c) July 2003. 123

Figure 5.8. a) Typical titration curve for the estuarine samples (Mylor Creek, July 2003); b) van den Berg linearization for the titration data. CuTotal = Cu in sample plus Cu added for the titration; CuL = Cu complexed by natural organic ligands; Cu1abile = Cu complexed by SA. 125

Figure 5.9. Dissolved organic ligands and chlorophyll a concentrations for the Fa! Estuary samples. a) August 2002; b) April2003; c) July 2003. 127

Figure 5.10. Salinity and eh! a concentrations together with PC2:GSH ratios for the Tamar transect, 25th March 2003. 132

Figure 5.11. Particulate GSH and PC2 concentrations for the Tamar transect,

- 7 - 1 25 h March 2003. Open symbols for PC2 indicate values below detection limit. 133

Figure 5 .12. Concentrations of particulate GSH and eh! a in the Plymouth Sound (including Oreston) for 2001-2002, and Plym Estuary 1 (Plym I, Plym2, Plym3 and Plym4) for 24 h September 2002. 136

Figure 5.13. Relationship between particulate GSH and eh! a for the samples from the Plymouth Sound and Plym Estuary. 13 7

Figure 5.14. Relationships between particulate GSH and eh! a for the samples from the Tamar transect. 13 7

Figure 6.1. Thau Lagoon (France). Sampling sites are indicated by numbers. I. 146 Sewage spring; 2. Crique de L' Angle; 3. Pointe de Balaruc; 4. Thau 4; 5. Thau 5; 6. Canal; 7. Middle of the lagoon; 8. Oyster bed.

Figure 6.2. A view of the oyster farm in the Thau Lagoon (southern France). 148

Figure 6.3. a) Typical titration curve for the estuarine samples (Sewage spring, site I, March 2003); b) van den Berg linearization for the titration data. Curotal = Cu in the sample plus Cu added for the titration;

CuL = Cu complexed by natural organic ligands; Cu1abile = Cu complexed by SA !54

Figure 6.4. Copper speciation in the Thau Lagoon, showing effects of the relation Curotal : organic ligands !56

- 8 - List of Tables

Table 2.1. Compounds related to protective mechanisms against metal toxicity/metal-induced oxidative stress in algae, their target or catalysed reactions. 21

Table 2.2. Eukaryotic phytoplankton species used in short-term (22-24h) laboratory metal exposure experiments. Concentration ranges of GSH and PCs are given in Jlmol g- 1 chl a unless indicated otherwise. 30

Table 2.3. Concentrations of PC2 and GSH in the particulate phase of natural 2 waters, together with concentrations of aqueous Cu +. 35

Table 3.1. Characteristics of the instrumental methods for the determination of thiols. DL = detection limit; NM = not mentioned; CLC = capillary liquid chromatography; SEC = size-exclusion chromatography; CE = capillary electrophoresis. 50

Table 3.2. Reagents for derivatisation of thiols, their respective products, and the detection methods. UV = Ultraviolet; FL = fluorescence. 60

Table 3.3. Particulate PCs and GSH in natural waters and the parameters of the analytical methods employed. DL = detection limit of compound between parentheses; nm = not mentioned. 71

Table 4.1. Composition of the culture medium Aquil and stock solutions used for the short-term metal exposure experiments. 82

Table 4.2. Reproducibility of the incubation experiments showing chromatographic peak areas and chl a content. 88

Table 4.3. Concentrations of ionic metals and EDTA-complexes calculated using MINEQL+ in the short-term metal exposure experiments. Cell numbers before and after incubation period. 90

Table 4.4. Production of PC10,.1 (sum of y-Glu-Cys units of the oligomers) and GSH, together with thiol ratios for laboratory P. tricornutum cultures under metal stress. pMetal = -Log[Meta1 2+]. 96

Table 5.1. Summary of contaminants, physical factors, sources, affected areas, and most vunerable featureslbiota in the Fal Estuary and Plymouth Sound and estuaries (adapted from Langston et al. 2003a, b). EQS = Environmental Quality Standard. I 08

Table 5.2. Common dominant phytoplankton species reported in the Fal Estuary (Environment Agency data, August 2002). * Autumn 1989 data from 1 Rijstenbil et al. 1990; +present abundantly (6-50 cells mL- ). 114

Table 5.3. Copper complexation in the Fal Estuary. Dissolved copper species, organic ligands and conditional stability constants for the Cu-organic complexes. %CuL = ([CuTotai]- [Cu2+]- [Cu']/ [CuTotai] x I 00 126

- 9- Table 5.4. Dissolved metal concentration ranges (nM) in the Fa! Estuary (data frotn Environment Agency 1990-2001, estimated from graphs in Langston et al. 2003b; EQS = Environmental Quality Standard for annual average). 130 Table 5.5. Copper complexation in the Tamar Estuary. Dissolved copper species, organic ligands and conditional stability constants for the Cu-organic 2 complexes. %CuL = ([CuTotal]- [Cu +]- [Cu']/ [CUTotal] x 100 133

Table 5.6. Dissolved metals (nM) in freshwater and tidal waters of Tamar (data from Langston et al., 2003a). Data from 2001 except Halton Quay, Warren Point, Hamoaze (1997); and Tamar offLynher (1993). 134

1 Table 6.1. Copper complexation in the Thau Lagoon (l9 h May 2003). 2 Concentrations of dissolved CuTotal. organic ligands (L), Cu +, Cu', together with the conditional stability constants of the Cu-organic complexes (CuL), salinity (S) and chlorophyll a contents. 155

- 10- List of abbreviations

Source/maker DTNP 2,2 '-dithiobis(2-nitrobenzoic acid) DTPA diethylenetriaminepentaacetic acid Sigma (99%)

OTT Dithiothreitol Sigma (99%)

EDTA Ethylenediaminetraacetic acid BDH EQS Environmental Quality Standard GSH Glutathione Sigma (98%)

GSSG Oxidised glutathione HEPES N-2-hydroxyethylpiperazine-N' -2-ethansulphonic BDH (99%) acid HPLC High performance liquid chromatography Merck-Hitachi LC-ESI-MS Liquid chromatography-electron spray ionisation­ Thermo mass-spectrometry mBrB 3, 7-dimethyl-4-bromomethyl-6-methyl-1 ,5- Fluka (>95%) diazobicyclo-[3.3.0] octa 3,6-diene-2,8-dione or monobromobimane

Me4 8 tetramethylbimane MSA Methanesulfonic acid solution Fluka (99.0%) NPM N-( 1-pyrenyl)maleimide OPA o-phthalaldehyde

PC2 Phytochelatin n = 2 (dimer) Pepsin Co. (Liverpool, UK) PC,. Phytochelatin RS-H Reduced thiol R-S-mB Derivatised thiol in the bimane form RS-SR Oxidised thiol SBD-F Ammonium 7-fluorobenzo-2-oxa-1 ,3-diazole-4- Fluka (99.0%) sulfonate TBP Tributylphosphine Acros Organics (95.0%) TCEP tris (2-carboxyethyl) phosphine hydrochloride Sigma TFA Trifluoracetic acid BDH

- I 1 - Acknowledgements

I would like to thank Dr Eric Achterberg and Dr Martha Gledhill for the supervision and patience throughout this work. I am indebted to Dr Jose Barriada, Dr Charly Braungardt, Rob

Harvey, Adrian Hopkins, Sayed El Badr and Sally Madwick for their great job during the field work in the Fal, Tamar and Plym estuaries. I am grateful to Prof Geoffrey Millward for his comments on this project and inspiring professionalism. Many thanks to Dr Alan Tappin for kindly revising this thesis and for his valuable comments. I also need to thank Dr Paul McComarck for his kind assistance with the LC-MS analyses and Cathryn Money for her help with the titration. Thanks also to lan Doidge, Andrew Tonkin and Andy Arnold for putting up with me for almost every day during the labwork years! I also would like to thank Derek Henon for helping me with the initial phytoplankton cultures. Many thanks to the French researchers for their assistance during the Thau

Lagoon campaign and for providing enjoyable French evenings. Thanks to Dr Franc;;oise Elbaz­

Poulichet, Dr Jean-Luc Seidel, Dr Franc;;ois Prevot and Jane Sunjuan for helping with the field work.

I am already missing the hard workers of Davy 504: Sayed El Badr, Prof Yousif

Elhassaneen, Turki AI-Said, Xi Pan, Jason Truscotl (Mi Madre!), Cathryn Money, and Kaori

Watanabe. Their friendly and supportive company made this work a multinational, pleasant and very fruitful experience. Thanks to Richie Sandford, Ying Kavinseksan, Nui Rattanachongkiat,

Angie Milne, Mike Foulkes, and again Cathryn Money and dear habibis (Turki, Yousif, Sayed and

Orif) for making the busy Portland Square 509 a comfy place during my writing up stage. I am very grateful to Gilberto Fillman, Elisa Fernandes and Eduardo Siegle for the prompt help during the hard times.

I need to thank my family for their constant support and understanding my need to "fly away again". Thanks to them all, especially to my father Zensei, I can have this easy life! I am very grateful to my sister Beth for her daily e-mails, and also to my nephews, Kiko and Tom, for updating me with their latest adventures and jokes. Very special thanks to my mother Yoshiko and my brother Nao, from whom I can still learn so much despite the barriers of memory. No PhD degree would be worthwhile without the things they taught me.

This research was made possible with the scholarship from Conselho Nacional de

Pesquisa Cientifica e Tecno/6gica do Brasil (CNPq 200.252/00-3). Financial support was also provided by the European IMTEC Project (EVK3-CT-2000-000036).

- 12- "A rapadura e doce, mas niio e mole ... "

(Rapadura is sweet, but not soft .. .)

Brazilian proverb

- 13 - AUTHOR'S DECLARATION

At no time during the registration for the degree of Doctor of Philosophy has the author been registered for any other University award.

This study was financed with the aid of a studentship from the Brazilian Government Research Council (CNPq- Conselho Brasileiro de Pesquisa Cientijica e Tecnol6gica) and canied out in collaboration with the European Union IMTEC Project.

Relevant scientific seminars and conferences were regularly attended at which work was presented and papers prepared for publication.

Publications to be submitted in October/November 2004:

• Determination of particulate glutathione and phytochelatins in natural waters using high performance liquid chromatography and fluorescence detection. Si/via K. Kawakami, Eric P. Achterberg and Martha Gledhill. Talanta • Copper complexation in the Thau Lagoon (France)- a shellfish farming area. Si/via K. Kawakami, Eric P. Achterberg, Martha Gledhill, Fram;:oise E/baz-Poulichet and Jean-Luc Seidel. Marine Pollution Bulletin • Effects of metal combinations on phytochelatin and glutathione production by Phaeodactylum tricornutum. Si/via K. Kawakami, Eric P. Achterberg and Martha Gledhill. Biometa/s • Phytochelatins produced by phytoplankton in coastal waters. Si/via K. Kawakami, Mart ha Gledhill and Eric P. Achterberg. Journal of Phycology

- 14- Presentation and Conferences Attended:

• Marine Biological Association Seminar Series- Plymouth, UK (09/2004) oral: Phytoche/atins as metal stress indication in the Fa/ Estuary (UK). Si/via K. Kawakami, Eric P. Achterberg and Martha Gledhi/1. • 14'h Meeting of the Society of Environmental Toxicology and Chemistry, SETAC Europe - Prague, Czech Republic (04/2004), oral: Production of intracellular metal-binding peptides and copper contamination in the Fa/ Estuary (UK). Si/via K. Kawakami, Eric P. Achterberg and Mart ha Gledhill. • Annual Meeting of the Society for Experimental Biology - Edinburgh, UK (03/2004), poster: Production of metal-binding peptides by phytoplankton in metal contaminated estuarine waters. Si/via K. Kawakami, Eric P. Achterberg and Martha Gledhill. 1 • 4 b European Meeting on Environmental Chemistry - Plymouth, UK ( 12/2003), poster: Natural organic ligands and dissolved metal species in the Thau Lagoon (France) - a shellfish farming area. Si/via K. Kawakami, Eric P. Achterberg, Mart ha G/edhi/1, Franr;oise Elbaz-Pou/ichet and Jean-Luc Seidel. • V Progress in Chemical Oceanography - Norwich, UK (07/2003), poster: Phytoche/atins as a biochemical response to metal exposure in the Fa! Estuary (UK). Si/via K. Kawakami, Eric P. Achterberg and Mart ha G/edhi/1. • Jnl European Meeting on Environmental Chemistry - Geneva, Switzerland (12/2002), poster: Production of metal-binding peptides in metal contaminated estuarine waters. Si/via K. Kawakami, Eric P. Achterberg and Martha G/edhi/1. • Challenger Centenary Conference in Marine Science - Plymouth, UK (09/2002), poster: Metal-binding polypeptides produced by the diatom Phaeodactylum tricornutum under Cu, Cd and Zn exposure. Si/via K. Kawakami, Eric P. Achterberg and Mart ha G/edhi/1. • 32 International Symposium on the Environment & Analytical Chemistry - Plymouth, UK (06/2002), poster: effects of metal combination on phytochelatin produced by the diatom Phaeodactylum tricornutum. Si/via K. Kawakami, Eric P. Achterberg and Martha Gledhi/1. • IV Progress in Chemical Oceanography - Bangor, UK (09/2001 ), oral: Glutathione and phytochelatins produced by marine phytoplankton in cadmium exposure experiments and in waters from Plymouth Sound (UK). Si/via K. Kawakami, Eric P. Achterberg and Mart ha Gledhill.

s;gn,d ... £+: ...... ~......

Date ______-??.~.--.9:-~----~~±......

- 15 - Chapter 1

1. Introduction

Phytoplankton are well-known to provide a major, direct food source for animals in the water column and sediments and are of paramount importance in the marine food web.

Aquatic systems are subject to enhanced inputs of toxic substances from natural (rivers, atmosphere, hydrothermal venting) and anthropogenic sources (sewage, mines, industry, agriculture, anti-fouling paints from boats). It is possible that deleterious effects on phytoplankton composition and distribution occur as a result of the toxicity of these compounds. In spite of this, studies addressing the capacity of phytoplankton to cope with a mixture of contaminants, such as metals, in natural waters are scarce.

The main emphasis of this research is to examine the production of phytochelatins and their precursor glutathione by phytoplankton in metal contaminated estuarine waters.

Phytochelatin production is one of the mechanisms adopted by a number of phytoplankton species against metal toxicity. The relationships between copper complexation in the dissolved phase and the production of particulate phytochelatins and glutathione were also investigated. The effects of combinations of toxic metals (Cu, Cd and Zn) on phytochelatin production were studied in short-tem1 experiments using a laboratory phytoplankton monoculture.

This thesis is organised in chapters that complement each other, but can also be read and interpreted as individual pieces of research. Thus, in order to facilitate readability, some repetition was necessary to describe or explain the results.

Chapter 2 provides an overview of the evidence for the production of phytochelatins and glutathione as important intracellular metal-binding peptides m phytoplankton for metal detoxification mechanisms. Metal speciation in natural waters and

- 16- laboratory experiments to stimulate PC production in phytoplankton cultures are discussed, as they represent a valuable tool to further information on thiol production in the field.

Chapter 3 examines the current analytical methods for particulate phytochelatins in phytoplankton cells. It provides considerations on the practical problems and limitations encountered with the application of the methods to field studies. In the present work, a method was optimised and applied to phytoplankton from estuarine waters.

In Chapter 4 short-term metal exposure experiments using the manne diatom

Phaeodactylum tricornutum to stimulate phytochelatin production are outlined. The main objective of this chapter is to discuss the effects of metal combinations (Cu, Cd and Zn) on phytochelatin and glutathione production.

Chapter 5 discusses the field measurements ofphytochelatins and glutathione in the

Fa! Estuary and Plymouth Sound (SW England, UK). It describes and compares the results in terms of the production of metal-binding peptides, distribution of trace metals and other ancillary parameters, and relates the production of metal-binding peptides and metal contamination, with special reference to copper species.

In Chapter 6 a short study on copper speciation in the Thau Lagoon (France), an important oyster farming area, is presented. This study provides baseline measurements of copper species and their possible implications to phytoplankton species, which are an important food source for the oysters.

Finally, Chapter 7 presents the conclusions and final remarks of this research. The subjects that need further research and the recommendations for the use of phytochelatins as metal stress indicators in field assessment are highlighted.

This study also contributed to the European Union Project IMTEC (In-situ automated Monitoring of Trace metal speciation in Estuaries and Coastal zones, in relation with the biogeochemical processes, EVK3-CT-2000-000036).

- 17 - Chapter 2

2. Phytochelatins produced by phytoplankton as an indication of metal stress in natural waters

2.1. Abstract

Phytoplankton cope with metal toxicity by a variety of strategies including the production of glutathione and phytochelatins. Glutathione and phytochelatins are efficient metal­ binding peptides produced by eukaryotic algae for several intracellular functions, notably as a protection against oxidative stress, a metal detoxification mechanism and metal homeostasis. The production ofphytochelatins can provide an indication of metal stress for phytoplankton at a sub-lethal level. Despite these essential biological roles, very little is known about the field conditions capable of influencing the production of these compounds by natural phytoplankton assemblages. This chapter describes what is currently known about the chemistry of the metal-binding peptides, particularly phytochelatins, their biological roles and distribution in natural waters, and some of the environmental conditions thought to cause variations in the production of these peptides. The potential applications of phytochelatins for environmental monitoring and remediation are also outlined.

- 18 - Chapter 2 Phytochelatins as an indication ofmetal stress

2.2. Introduction

Many trace metals, including Cd, Co, Cu, Fe, Ni, Mn, Mo and Zn, are important micronutrients and are able to influence productivity and species composition of marine algal communities (Sunda 1988; Butler 1998). However, some trace metal nutrients, such as Cd, Cu, Ni and Zn become extremely toxic to marine algae at elevated free aqueous ion concentrations. These trace metals can complex non-specifically with coordination sites of important biological molecules, such as enzymes, and alter normal metabolic functions

2 3 (Sunda 1988). Trace metal ions such as Cu + and Fe +, for instance, can interfere in the photosynthetic electron chain, generating reactive oxygen species responsible for the oxidative damage of proteins, lipids, and nucleic acids (Pinto et al., 2003 and references

2 2 2 therein). Other metal ions, such as Cd +, Pb + and Hg +, can enhance the pro-oxidant status of cells by reducing the antioxidant component pools, activate calcium-dependent systems, and affect iron-mediated processes (Okamoto et al., 2001 and references therein).

Increased exposure to these metal ions can cause profound effects on organisms, resulting in inhibition of growth, reduced fecundity and even death (Bryan & Langston 1992).

ln coastal waters, the concentrations of trace metals are often enhanced as a result of urban and industrial waste water discharges, run-off from operating and disused metalliferous mines, and leaching of anti-fouling paints from vessels. However, no evidence of metal toxicity on phytoplankton assemblages in the field has been reported (Ahner et al. 1997).

Several studies have demonstrated that a number of phytoplankton species respond to metal toxicity using protective mechanisms which involve induction of several antioxidative compounds (Rijstenbil et al. 1994; Okamoto et al. 2001; Sigaud-Kutner et al.

2002; Sunda et al. 2002), exudation of organic Iigands to complex the toxic metal ions

(Moffett et al. 1997; Croot et al. 2000; Gordon et al. 2000), and production of intracellular

- 19- Chap1er 2 Phylochelalins as an indicalion of me/a/ s/ress

metal-binding peptides (Ahner et al. 1995; 1997). The compounds related to these

protective mechanisms are naturally present or produced in most marine algae and can scavenge reactive oxygen species, promote reduction conditions for essential biomolecules, and/or decrease metal toxicity by forming stable metal-organic complexes

(Grill et al. 1985; Ahner et al. 1995; Pinto et al. 2003).

A summary of compounds related to the mechanisms against metal toxicity and/or metal­ induced oxidative stress, with their corresponding target and catalysed reactions is shown in Table 2.1. All these compounds, apart from the algal exudates and dimethylsulphoxide and its metabolites, have been suggested as biomarkers of metal stress in aquatic organisms

(Stegeman et al. 1992). It has been acknowledged that a suite of biomarkers, rather than a single one, should provide a better indication of stress and be able to give a forewarning of ecosystem-level damage (Adarns & Greeley 2000).

The production of metal-binding peptides by fission yeast and higher plants treated with metals has been investigated since the late 1970s (Rauser 1990). In vitro experiments have shown that metal-binding peptides, known as phytochelatins, protected metal-sensitive enzymes from inactivation and restored the activity of metal-poisoned enzymes (Kneer &

Zenk 1992). Only a limited number of research groups, however, have undertaken field studies to investigate the production of these peptides by natural phytoplankton assemblages and still little is known about the environmental conditions capable of causing variations in their levels (Ahner et al. 1997; 1998; Knauer et al. 1998; Matrai & Yetter

1988; Tang et al. 2000; Wei et al. 2003). In this study, special attention has been given to examining the production of these peptides by phytoplankton in natural waters. This chapter describes what is currently known about the chemistry of the metal-binding peptides in phytoplankton, their biological roles and occurrence in natural waters.

- 20- Table 2.1. Compounds related to protective mechanisms against metal toxicity/metal-induced oxidative stress in algae, their target or catalysed reactions.

Compound Target Comment References Algal exudates Metals Extracellular complexation of metals Gledhill et al., 1997 Croot et al. 2000

1 Ascorbate 0 2 ( 1.\g), OH, Oz , HOz Vitamin C; electron donor to free radicals Noctor & Foyer, 1998 Pinto et al., 2003 Dimethylsulphoniopropionate OH Most abundant sulphur compound in Sunda et al., 2002 (DMSP) marine algae DMS, acrylate, DMSO, MSNA OH Enzymatic cleavage products of DMSP Sunda et al., 2002 Flavonoid OH and HOC! Higher plants N Glutathione Non-specific Eukaryotes; multifunctional metal­ Rijstenbil et al., 1994 binding peptide Noctor & Foyer, 1998 Metallothionein ·oH, metals Bacteria, animals and higher plants Langston et al., 2002 Phytochelatin Metals Plants, algae, some fungi; polypeptide Ahner et al., 1995 a-tocopherol ROz Vitamin E Matsukawa et al., 2000 Pinto et al., 2003 13-carotene Oz ( 1L.\g), ROz Phytoplankton pigment Matsukawa et al. 2000 Pinto et al., 2003

Conlinued on lhe nexl page 9 Table 2.1. continuation -§ Enzyme Reaction catalysed Comment References -, ""-' Ascorbate peroxidase H202 + ascorbate ~ H20 + Eukaryotes Noctor & Foyer, 1998 monodehydroascorbate Pinto et al., 2003

Catalase 2 H202 + 2 H+ ~ 2 H202 +02 Eukaryotes Pinto et al., 2003

Glutathione peroxidase H202 + 2 GSH ~ 2 H20 + GSSG Eukaryotes Pinto et al., 2003

ROOH + 2 GSH ~ ROH + GSSG

Glutathione reductase GSSG + NAD(P)H + H+ ~ 2 GSH + Eukaryotes Pinto et al., 2003 NAD(Pt

Superoxide dismutase 2 02- + 2 H+ ~ H202 +02 Eukaryotes Sigaud-Kutner et al., 2002 N N Thioredoxin Prot-S2 + Prot'(SH)2 ~ Prot(SH)2 + Eukaryotes Pinto et al., 2003 Prot'-S2

1 0 2 ( 6g), oxygen singlet; R02, organic radical; DMS, dimethylsulphide; DMSO, dimethylsulphoxide; MSNA, methane sulphinic acid; GSH, glutathione; GSSG, oxidised glutathione; ROOH, organic acid; Prot-S2, oxidised sulphur protein; Prot'(SH)2, reduced sulphur protein Chapter 2 Phytochelatins as an indication of metal stress

2.3. Thiols as intracellular metal-binding peptides

Phytochelatins and glutathione are metal-binding peptides classified as sulphydryl compounds or thiols - analogous to alcohols in which the oxygen is replaced by sulphur, fom1ing the SH functional group (Buckberry & Teesdale-Spittle 1996; Morrison & Boyd

1997). These peptides chelate metals through coordination with the SH groups.

Phytochelatins and glutathione are considered low molecular weight compounds as they comprise structures of around I 0 kD or less (Grill et al. 1985).

2.3.1. Glutathione

Glutathione (GSH), a tripeptide formed by glutamate (Giu), cysteine (Cys) and glycine

(Gly), is the major thiol produced by animals, plants, algae, and bacteria (Ahner et al.

2002). The structure of GSH is presented in Fig. 2.1.

------1-·-·-·-·-·-·-·-·,------y-Giutamate 1 Cysteine Glycine SH I 01 0 i I I NH 11 I I ~OH I I I i 0 I I ·------_l._. -·-·-·-·-·-·-

Figure 2.1. Structure of glutathione.

The pathways of GSH synthesis, presented in Fig. 2.2, has been elucidated and appears to be common to all organisms that contain GSH (Noctor & Foyer 1998). Two steps catalysed by the enzymes y-glutan1ylcysteine synthetase and glutathione synthetase, lead to the sequential formation of y-Giu-Cys and GSH, respectively. The utilization of

- 23- Chapter 2 Phytochelatins as an indication ofm etal stress photorespiration to provide biosynthetic intermediates for GSH synthesis has been considered less significant. GSH can be synthesised in the cytosol and chloroplast, where its constituent amino acids and required enzymes have been detected, and transported to other cell compartments (Noctor & Foyer 1998).

GSH t Glutathione synthetase

y-Giu-Cys Gly \ r t \ y-glutamylcysteine Glycollate synthetase

Glu Cys

l ~ Ser N s Photo- assimilation assimilation respiration

Figure 2.2. Biosynthesis of GSH from its constituent amino acids. Glu, glutamate; Cys, cysteine; Ser, serine; Gly, glycine (Norton & Foyer, 1998).

The maJor processes involving GSH in eukaryotic algae are summarised in Fig. 2.3.

Glutathione is related to several functions, such as the maintenance of reducing conditions for amino acids and proteins, protective functions against oxidative and radiation damage, and against enhanced concentrations of metals and xenobiotic organic compounds, and

-24- Chapter 2 Phytochelatins as an indication ofme tal stress

production of phytochelatins. The reduced form of GSH exists interchangeably with the oxidised form, GSSG. The multifunctionality of GSH is still not fully understood.

(Buckberry & Teesdale-Spittle 1996; Noctor & Foyer 1998; Ahner et al. 2002).

Interests in the determination of GSH in biological systems have increased in recent years, as the role of sulphydryl compounds in physiological processes has become more evident.

Another potential relevant role for this tripeptide is as a dissolved metal ligand in surface seawater, possibly as phytoplankton exudates (Le Gall & van den Berg 1998; Vasconcelos

& Leal 2001 ). Further studies, however, need to be undertaken to determine the chemical stability and sources of the dissolved GSH in natural waters, as the intracellular GSH produced by phytoplankton are at lower concentrations for the high rates of GSH efflux required to explain the dissolved GSH levels in seawater (Ahner et al. 2002).

Glycine + y-Giu-Cys

Sulfhydryl buffering GSH synthetase \ 1 ~S-t""''•""

1 GSH reductase Degradation

Figure 2.3. Processes involving GSH in eukaryotic algal cells (adapted from Ahner et al., 2002).

- 25- Chapter 2 Phytochelatins as an indication ofm etal stress

2.3.2. Phytochelatins

Phytochelatins (PCs) are produced by plants, algae, and some fungi and have similar

functions as metallothioneins in animals and cyanobacteria (Ahner et al. 1995).

Phytochelatins are also known as class Ill metallothioneins, although they are not proteins

as the metallothioneins (Rauser 1990). Phytochelatins are synthesised by the constitutive

enzyme phytochelatin synthase which catalyses the transfer of y-Glu-Cys from GSH to

another GSH molecule. This process requires a metal ion for activity (Cobbett 2000).

These peptides present the general formula (y-Glu-Cys)11 -Gly with the Glu and Cys

residues linked through a y-carboxylamide bond, and n ranging from 2 to 11 (PC2 to

PC 11 ) (Grill et al. 1985). Phytochelatin polypeptide chains with n = 2, 3 and 4 are

predominant in phytoplankton (Ahner et al. 1995; Rijstenbil & Wijnholds 1996). A simplified structure of GSH and PCs is presented in Fig. 2.4.

n ~------~~------~ ; -.-.-.Y=

0 I 0 I l NH ,, ! 1 ~ OH : I

I I I I I I I . - .- .- .- .- .- . -'------,_ .- .- '- .- .- .- . _.. ------~

~------______G_ l_u_ta_th_io_n_e______~ ~

Phytochelatin n = 2

Figure 2.4. Basic structure of phytochelatins, (y-glutamyl-cysteinyl),-glycine, n = I for glutathione and n = 2-1 I for phytochelatins.

- 26- Chapter 2 Phycochelatins as an indication ofmetal stress

The enzymatic pathway of PC biosynthesis seems to occur in the cytosol, and its complete elucidation still needs further study (Cobbett 2000). The transport of potentially toxic metals from one cell compartment to another, for example, from the membrane to the vacuole, appears to be facilitated by complexation with PCs (Rauser 1990).

2.3.2.1. Phytochelatins and metal tolerance

Because of the high affinity of the SH groups for metals, both GSH and PCs are efficient metal complexing ligands inside cells. The conditional stability constants for the various

PC-metal complexes have not yet been determined, but PCs have proved to be more efficient for sequestering metals and to reactivate metal sensitive plant enzymes than other biological metal chelators such as GSH and citrate (Kneer & Zenk 1992). Phytochelatins have been related to protective mechanisms against enhanced metal concentrations and also to homeostasis process, i.e. regulation of metal concentrations inside cells (Grill et al.

1985; Ahner & More! 1995; Ahner et al. 1995). Longer PC polypeptide chains should form more stable metal-complexes due to the advantageous conformational structure and number of SH groups (Mehra et al. 1995).

The kinetics of PC production have been investigated mainly in laboratory diatom cultures

( Thalassiosira weissjlogii, Thalassiosira pseudonana and Phaeodactylum tricornutum) under metal exposure. Phytochelatins in T weissjlogii are readily synthesised within I h after exposure to pCd = I 0 (pCd = - Log [Cd2+]). The removal of the conditions of Cd stress, by resuspension of the cells in a medium free of Cd ions, resulted in a rapid restoration of the low basal levels of intracellular PCs, which indicates that PC production is tightly regulated by metal ions (Ahner & More! 1995; Lee et al. 1996; Morelli &

Scarano 200 I).

- 27- Chapter 2 Phytachelatins as an indication of metal stress

For most of the phytoplankton spec1es investigated in laboratory metal exposure experiments, PC production increased with increasing dissolved metal concentrations in the medium. Furthermore, the increase in PC production was observed before any essential physiological parameters, such as growth rate, were affected (Ahner et al. 1995; 1997).

However, the intracellular stoichiometry ofPCs to metal ions has apparently no single rule.

Some phytoplankton species seem to present a ratio between PC and intracellular Cd of

2: I, but others produce PCs at concentrations 80 times greater than the required concentration to chelate the intracellular Cd (Ahner et al. 1995; 2002).

Although a high PC-producing capacity is not necessarily a feature of tolerant plants

(Deknecht et al. 1995), early, rapid formation and overproduction of PCs and PC-metal complexes could account for an enhanced metal tolerance (Rauser 1990). lndeed, phytoplankton species under Cd exposure exhibiting high ratios between PC production and intracellular Cd content had little or no effects on growth rate. ln contrast, species more sensitive to Cd presented low PC to intracellular Cd ratios and a significant growth rate decline (Ahner et al. 1995; Rijstenbil & Wijnholds 1996).

2.4. Stimulation of phytochelatin production in laboratory phytoplankton monocultures

Most of the valuable information about PCs produced by phytoplankton has come from short-term metal exposure experiments conducted under controlled laboratory culturing conditions. These experiments consist of incubation of phytoplankton cells, usually monocultures, with metal ions, in an artificial medium with a well defined composition to allow a thorough calculation of metal speciation (Rijstenbil & Wijnholds 1996; Ahner et al. 1995; 2002).

- 28- Chap/er 2 Phylochelalins as an indicalion of me/a/ s/ress

2.4.1. Phytochelatin production by different phytoplankton species

The phytoplankton species investigated together with the metal ions to which they have been exposed are presented in Table 2.2. The concentrations of the metals ranged from low to growth inhibiting levels. The concentrations of GSH and PCs are given in different unit formats in literature which hamper a direct phytoplankton interspecies comparison. The preference for one normalisation or another usually depends on the experimental conditions applied. For incubation experiments at high dissolved metal concentrations

1 (J.Lmol L" ), for example, normalisation with chl a is not recommended because metal toxicity affects pigment content, especially for Cu (Cid et al. 1995; Rijstenbil & Wijnholds

1996). Normalisations of PCs using chl a, however, have been preferred for comparisons between field data or for incubations at low metal concentrations. In a number of laboratory studies, PCs are expressed using the concentration of SH groups and are given in biovolume (cell number x cell volume) to allow comparisons between different phytoplankton species (Rijstenbil & Wijnholds 1996; Ahner et al. 2002).

Despite the different units, it can be observed that chlorophytes and prymnesiophytes produce PCs, with the prymnesiophyte Emiliania huxfeyi showing a pronounced thiol production under Cu and Cd exposure (Ahner et al. 2002). The only two reported dinoflagellate species, Heterocapsa pygmaea and Prorocentrum micans, showed low PC production under Cd exposure, and no production under Cu exposure, respectively, suggesting that they have developed adaptive detoxification mechanisms other than PC production (Ahner et al. 1995; Lage et al. 1996).

- 29- Table 2.2. Eukaryotic phytoplankton species used in short-term (22-24h) laboratory metal exposure experiments. Concentration ranges of GSH and PCs are given in Jlmol (g Chi a)" 1 unless indicated otherwise.

Species GSH PC2 PC3 PC4 PCS PC6 Metal exposure References Bacillariopbyte (diatoms) Ditylum brightwellii 80- 270" 0- 40" 0- 10" Cd,Cu Rijstenbil and (Coastal) Wijnholds, 1996 Skeletonema costatum 400- 640" 15- 170" 0 -14" 0- 26" Cd,Cu (Coastal) 25- 8QQA 75 -7QQA Cd, Cu, Zn Ahner et al. 1997 47- 190 Cd, Cu, Zn (mixture) Wei et al. 2003 Phaeodactylum tricornutum 1970-3380" 64- 570" 0- 240" 0- 515" 0- 270" 0-67" Cd,Cu Rijstenbil and (Estuarine) Wijnholds, 1996 500- 148Gb" 2.2- 160" 3.6- 87° 0.7-8.1" Cd,Cu,Pb,Zn Ahner et al. 2002 58- 120b 20 -72b Cd,Pb,Zn Morelli and Scarano 200 I w 0 Thalassiosira pseudonana 800- 1300" 0- 290" 0- 190" o- so· Cd,Cu Rijstenbil and (Coastal) Wijnholds, 1996 10 Cu, Zn (mixture) Wei et al. 2003 Thalassiosira weissjlogii 1000- 2000" 3.2-730 0.8-500 Cd, Cu, Zn, Ag, Pb, Ahner et al. 1995a (Coastal) Co, Ni, Hg IS- 1200" 12- 690" 5.9-160" Ahner et al. 2002 ~ Thalassiosira oceanica 4- 112 2-60 Cd Ahner et al. 1995b ~ (Oceanic) a·~ ~ § :;· [ §" ~ 3 continued on next page "'§_ "' ~ Table 2.2. continuation S~ecies GSH PC2 PC3 PC4 PCS PC6 Metal ex~osure Reference 9 Chlorophyte (green algae) .g Dunaliella tertiolecta 5- 19 2-66 2-56 Cd Ahner et al., 1995a ""' (Estuarine) "' 2-6 Cu, Cd, Zn (mixture) Wei et al. 2003 0.5-8.5 Cu, Cd, Zn Wei et al. 2003 Scenedesmus subspicatus 23° Cu Knauer et al. 1998 (freshwater) Tetraselmis macu/ata 12.5-425 3-363 Cd, Cu, Zn, Pb Ahner et al. 1995 b (Estuarine) 6- so 2-23 1-2 Ahner et al. 1995a Tretaselmis tetrathele 0 0 0 0 0 Hg,Cd Satoh et al. 1999 Stigeoclonium tenue 100-S&Od 100 -700d so- 300d Cd, Pb, Zn (mixture) Pawlik- (Freshwater) Skowroitska 200 I Stigeoclonium sp 500- 2300d I00-1450d 100-300d Cd, Pb, Zn (mixture) (Freshwater) Prymnesiophyte (coccolithophores) Pleurochrysis carterae 0-84 0-28 Cd Ahner et al. 1995b \;.J (Coastal) Emiliania huxleyi soo- 24oo• 46-4303 58- 4903 17-5603 Cd, Cu, Zn, Pb Ahner et al. 2002 (Coastal/Oceanic) 25-450 13-375 Cd, Cu, Zn, Pb Ahner et al. 1995b Pavlova lutheri 2- 17 Cd (Estuarine) ..., Dinoflagellate ~c Heterocapsa pygmaea S-9 9- 13 Cd Ahner et al. 1995b ,.." (Estuarine) "' Prorocentrum micans oB oB oB os oB oB Cu Lage et al. 1996 "~- (Estuarine) ""' oc oc oc oc oc oc " Cd,Cu Pistocchi et al. ":;· 2000 ;:;·"- 1 1 e. • Jlmol SH per !iter biovolume; b amol cel1" (total PC as y-Glu-Cys pools); c total PC concentration; d nmol SH g· dry weight; Aincubation for 48 h; c;· "'-." Bincubation for 3 h, with no PC synthesis; cincubation for 7 to 21 days. :; "e. :;-"' "'t:: Chap1er 2 Phywchelalins as an indicGiion ofmelal slress

The diatoms Thalassiosira weissflogii and Phaeodactylum tricornutum showed a high PC production in metal exposure experiments (Rijstenbil & Wijnholds 1996; Ahner et al.

2002). These species have been extensively used by many workers to assess the effects of a range of metals on PC production (Cid et al. 1995; 1997; Ahner et al. 1995; Ahner &

More! 1995; Morelli & Scarano 1995; 2001; Rijstenbi1 & Wijnholds 1996; Lee et al. 1996;

Torres et al. 1997; 1998) because of their well-known physiology and relative ease of culturing.

Regarding GSH production, relatively constant concentrations (0.8-2.8 mM) have been observed in different phytoplankton monocultures upon long-term exposure to ambient Cu and Cd concentrations (Ahner et al. 2002). Glutathione was produced at similar concentrations by cultures of T pseudonana and Dunaliella sp. even at different physiological states. Generalisations, however, cannot be made as other species, such as E. hu:r:leyi, produced significantly higher PCs and GSH concentrations in the exponential growth phase than in the senescent phase. Acute metal exposure can cause a decrease in

GSH levels. For instance, P. tricornutum underwent a decrease in GSH levels over the growth phase, which was linked to a biochemical cost for leading to an enhanced PC production (Rijstenbil & Wijnholds 1996). The concentrations of y-Glu-Cys and cysteine, the precursors of GSH, increased significantly in some metal-stressed cultures in response to metal exposure, suggesting a thigh modulation in the synthesis of these thiols (Ahner et al. 2002; Wei et al. 2003).

2.4.2. Effects of various metals on phytochelatin production

The range of free aqueous metal ion concentrations used by Ahner et al. (1995) in short­ term metal exposure experiments spanned from pM to nM, and reflected relevant ambient

- 32- Chapter 2 Phytochelarins as an indication of m era/ stress

metal ion concentrations. Most studies have employed metal ions at concentrations as high as J.!M (Rijstenbil & Wijnholds 1996; Morelli & Scarano 200 l ). The highest PC induction for some representative coastal phytoplankton species (T weisjlogii, T maculata, E. hu.xleyi) has been obtained with Cd and to a lesser extend Cu, Zn and Pb (Ahner & More!

1995; Ahner et al. 1997). The combination of these metals has caused antagonistic and suppressing effects on PC production, probably as a result of competition for cellular binding sites (Pawlik-Skowronska 2001; Wei et al. 2003). Synergistic effects on PC production have also been observed for combinations of Cd, Cu and Zn, but the reasons for such effects have not been fully elucidated (Wei et al. 2003).

For the experiments involving trace metals at low concentrations, it has been necessary to expose some phytoplankton species to the low metal levels for about 3 to 18 generations before PCs could be detected (Ahner et al. 2002; Wei et al. 2003). This continuous metal exposure simulates more closely the chronic metal contamination in the field.

2.5. Phytochelatins and glutathione produced by phytoplankton assemblages in natural waters

Intracellular phytochelatin concentrations produced in natural waters are over an order of magnitude lower than those in laboratory phytoplankton cultures under exposure to the same metal concentrations. This indicates that the interaction between phytoplankton and metals should not be the only factor involved in the thiol production in the field. Thus, the ability to replicate field conditions in the laboratory and the transfer of techniques to the field are essential to further our understanding of thiol production, particularly PCs, in natural waters. It is important to emphasise that environmental considerations from incubation experiments are best addressed using ambient metal concentrations as well as a

- 33- Chapter 2 Phytochelatins as an indication ofmetal stress

range of phytoplankton species which are naturally present in marine ecosystems, as each species responds differently to metal exposure. It is likely that organisms developed from metal-tolerant strains have an enhanced capability to synthesise PCs. Strains of the same phytoplankton species collected from metal-polluted and unpolluted sites produced different levels of PCs (Pawlik-Skowronska 2003). Thus a more realistic approach adopted in some laboratory studies considers the effects of combinations of metals on PC production by phytoplankton communities isolated from study areas. Results from these studies have helped to clarify findings in the field (Ahner et al. 1994; 1997; 1998; Knauer et al. 1998).

2.5.1. Field measurements of phytochelatins and glutathione

Only a few studies have been undertaken to determine the production of PCs and GSH by phytoplankton in natural waters. A summary of the field data for GSH and PC2 in the

2 particulate phase of natural waters subject to varying free Cu + concentrations is presented in Table 2.3. Particulate PCs and GSH in these studies corresponded to the phytoplankton fraction retained on 0.7-0.8 11m pore size filters. The sampling areas comprised diverse environmental conditions, from pristine oceanic waters to highly metal polluted coastal/estuarine systems. Cyanobacterial blooms occurred during the collection of samples in the upper Galveston Bay (Tang et al. 2000). Little information regarding phytoplankton composition is available for the other sites that could be linked to PC production.

- 34- 2 Table 2.3. Concentrations of PC2 and GSH in the particulate phase of natural waters, together with concentrations of aqueous Cu +.

Sample location GSH PC2 Cu + Comments References

J.Lmol (g Chi a)" 1 M

New England, USA 2-50 1o· Various harbours Ahner et al., 1997 9 Equatorial Pacific 2-50 0.3 X 10" Ahner et al., 1998 Switzerland and Italy 4-17 10-16 _ 10.1o Various freshwater lakes Knauer et al., 1998 Galveston Bay, USA 10-40 Up to 6.3 I o-14 - I o-13 Coastal area not strongly Tang et al., 2000 metal contaminated Blooms of cyanobacteria C.J VI Saanich Inlet, USA 38-200 Anoxic fjord Matrai and Vetter, 1988 Southern California Bight, 100-250 Coastal area USA Elizabeth River Estuary, I - 178 I - 19 10.14 _ 10.12 Wei et al., 2003 USA Eel Pond, USA 33-83 0.7-4.5 10.11 _ 10-10 Chapter 2 Phytochelatins as an indication of metal stress

In general, the dominant phytoplankton groups in coastal and estuarine waters are diatoms and dinoflagellates, while other smaller but important groups include cryptophytes, chlorophytes (green algae), and chrysophytes (cyanobacteria) (Fogg & Thake 1987). It is then expected that most of the contribution on PC production in these studies has come from the dominance of diatoms, prymnesiophytes, and green algae, as the few dinoflagellate species investigated so far apparently have a different mechanism against metal toxicity than PC production.

From the field data, it can be observed that PC production provides a general indication of

2 metal stress for phytoplankton. In areas with higher Cu + concentrations, the production of

PC2 was also higher. In small harbours in southeastern New England (USA), for example,

1 PC2 concentrations ranged between 2 to 50 J..Lmol (g chl ar , increasing systematically with

2 14 10 increasing free aqueous Cu + concentrations, from 10" to 10" M. In Massachusetts Bay,

PC2 concentrations were higher in the vicinities of harbours which received sewage and riverine inputs, and decreased seawards, farther from anthropogenic metal sources (Ahner et al. 1994). Phytochelatins have also been observed in freshwater phytoplankton in a number of metal contaminated Swiss lakes (Knauer et al. 1998), at comparable concentrations to those reported in marine phytoplankton. For the remote oceanic region of the Equatorial Pacific, however, where essential metals occur at limiting concentrations,

PCs have also been reported at concentrations comparable to those from polluted coastal sites (Ahner et al. 1998). In such oceanic areas, PC production seems to function as a storage and supply mechanism of essential metals for some phytoplankton species and is stimulated during metal-limiting conditions, notably depletion in Zn which can be ameliorated by Cd through uptake involving PC induction (Ahner et al. 1998; Lee et al.

1996). This has been supported by results from laboratory experiments indicating that Cd can substitute Zn in some Zn-limited diatoms, chlorophytes and prymnesiophytes (Lee &

- 36- Chapter 2 Phytochelatins as an indication of metal stress

More! 1995). High variabilities in the particulate PC and GSH concentrations for a same sampling site, at different times, have been associated to differences in the phytoplankton composition and nutrient limitations (Wei et al. 2003; Tang et al. 2000).

Phytochelatin polypeptide chains longer than n = 3 have not been reported for natural waters. This may be due to the limitations of the population and/or cellular constraints of natural phytoplankton assemblages. The dynamic nature of coastal ecosystems with rapid phytoplankton turn-over rates and short estuarine and coastal flushing times should also influence the production and stability of longer polypeptides. In addition, lack of sensitivity of the current analytical methods puts constraints on the determination of PCs, as PCs with longer chains typically occur at lower concentrations than the dimer PC2.

2.6. Factors affecting the production of GSH and PCs in natural waters

From the above considerations, it can be emphasised that the production of thiols, particularly PCs, is dependent upon phytoplankton species, the degree of toxicity of the trace metal ions, and competition among metals for cellular binding sites. A limited number of studies have addressed the environmental conditions influencing thiol production in natural waters. Metal speciation and dissolved organic matter in natural waters greatly influence the bioavailability and toxicity of metals to biota (Campbell

1995), and consequently should affect the intracellular GSH and PC production. Light intensity, nutrient supply or a combination of both also appear to play a role in the concentrations of the metal-binding peptides in the field (Matrai & Vetter 1988; Rijstenbil et al. 1998).

- 37- Chapter 2 Phytochelatins as an indication ofmetal stress

2.6.1. Influence of trace metal speciation on PC production

A variety of chemical species of trace metals are present in natural waters, differing in their bioavailability and toxic effects on organisms (Sunda 1988). Ionic metal species are thought to be more bioavailable and toxic because they can pass through cell membranes, in contrast to metal-organic complex forms (Campbell 1995). In laboratory experiments, a

2 direct relation between free aqueous Cu + species and Cu toxicity in marine algae has been observed (Ahner & More! 1995; Gledhill et al. 1999). The production of PCs has shown to be proportional to the free aqueous metal ions rather than to the total dissolved metal concentration (Ahner et al. 1997). Based on laboratory observations, the metals most likely to induce PC production in coastal waters are Cu, Cd and Zn, since other toxic metals (Pb,

Ag and Hg) typically occur at levels below the concentration which stimulates PC production (Ahner et al. 1997). Although the organic ligands dominate Cu and Zn

2 2 speciation in coastal waters, free Cu + and Zn + ions can be found at elevated concentrations in areas subjected to anthropogenic influences, and acid mine run-off, where total dissolved Cu and Zn concentrations exceed the buffering capacity of the natural organic ligands (Kozelka & Bruland 1998).

2.6.2. Influence of light and nutrient supply on GSH and PC production

As the intracellular GSH is involved in the antioxidant system, oxidative stressors such as high light, UV exposure and metal intoxication can cause an increase in GSH levels, as a cellular acclimation response. Extreme exposure to these stressors, however, has lead to a decrease in GSH levels (Noctor & Foyer 1998; Okamoto et al. 2001). Phytoplankton exposed to lower than ambient light intensities showed a decrease in GSH concentrations, when nitrate levels were low (Matrai & Vetter 1988). These combinations of effects on

- 38- Chap/er 2 Phyloclrelalins as an indicalion of me/a/ s1ress

GSH concentrations could possibly result in a reduced PC production, since GSH is required for the PC synthesis. However, low GSH concentrations do not seem to affect PC production in the field, suggesting that an alternative biosynthesis pathway of PCs could occur through intracellular polymerization ofy-Giu-Cys (Wei et al. 2003).

A low PC production was observed in the diatom T pseudonana under nitrate limitation, which was associated to the fact that inorganic nitrogen is required for the peptide synthesis (Rijstenbil et al. 1998). A limitation in nitrogen was also found to affect sulphate assimilation and hence GSH biosynthesis (Ahner et al. 2002). Some phytoplankton species were unable to produce PCs in the senescent phase (when the culturing medium becomes depleted in nutrients), also indicating possible effects of nutrient limitation (Ahner et al.

2002).

2. 7. Application of PCs for environmental purposes

Intracellular PCs have been suggested as metal stress biomarkers for phytoplankton in coastal waters as they provide an early measure of metal toxicity (Ahner et al. 1997). The use of synthetic PCs has recently been considered as sensitive biosensors for toxic metals and for bioremediation.

2.7.1. Use of phytochelatins as biomarkers of metal stress

The concept of the biomarker involves the use of a biochemical, cellular, physiological or behavioral parameter as diagnostic screening tools in environmental monitoring (Stegeman et al. 1992; Sanders 1990). The use of the biomarker approach has the advantage to be more sensitive, less variable, and easier to measure in comparison to stress indices such as

- 39- Chapter 2 Phytochelatins as all i11dication of metal stress inhibition of growth, changes in rate of development, and reduced reproductive potential.

The disadvantage is that it can be more difficult to relate the biomarker responses to the health of the organism and to adverse effects on the population (Sanders 1990).

Recently, a suite of biomarkers to assess metal stress has been proposed to be more adequate than the use of a single biomarker, as the effects of adverse environmental conditions can mask some biomarker responses (Adams & Greeley 2000). Thus the use of phytochelatins as biomarkers, as a complement to the biomarkers referred to in Table 2.1, allows us to identify subtle biochemical perturbations caused by a mixture of bioavailable toxic metals, as proposed for similar compounds, metallothioneins, in environmental monitoring (Langston et al. 2002; Mouneyrac et al. 2002).

2.7.1.1. The ratio PC:GSH in natural waters as a metal stress indication

The field data of PCs and GSH are usually presented in terms of eh! a. Chlorophyll a content provides a general indication of phytoplankton biomass and is relatively easy to measure. However, eh! a content varies with light intensity and phytoplankton species composition and this can compromise its use for data normalisation (Wei et al. 2003).

Furthermore, prokaryotes produce eh! a but not PCs, and hence could cause an underestimation of PC normalised concentrations. Ratios of PC to total low molecular weight thiols may provide more accurate information on the physiological status of the phytoplankton assemblages regarding metal stress, than PC concentrations alone (Wei et al. 2003). As GSH is required for a number of biochemical functions, in addition to its use in PC synthesis, it is thought that the PC:GSH ratio is maintained at a stable level in

-40- Chapter 2 Phytochelatins as 011 i11dicatioll of metal stress

healthy phytoplankton cells (Rijstenbil & Wijnholds 1996; Ahner et al. 2002; Tang et al.

2000).

Based on the production of thiols of metal-stressed diatoms (Rijstenbil & Wijnholds 1996),

PC:GSH ratios between 0.018-0.26 were considered an indication of healthy phytoplankton cells in Galveston Bay (USA), with enhanced ratios indicating metal stress

(Tang et al. 2000). As highlighted before, a better understanding of the environmental effects on the thiol production is necessary for the adequate application of such ratios.

Furthermore, a complicating factor is that bacteria, zooplankton and prokaryotic phytoplankton also produce GSH and therefore can interfere in the PC:GSH ratio approach.

2.7.2. Other applications for phytochelatins

Synthetic phytochelatins have recently been applied as sensitive biosensors for a range of heavy metal ions (Hg2+, Cd2+, Pb2+, Cu2+ and Zn2+) at concentrations in the fM to mM range (Bontidean et al. 2003). Genetically engineered long phytochelatin chains (n = 20) could provide an attractive strategy for developing high affinity bioadsorbents suitable for heavy metal remediation (Bae et al. 2000).

2.8. Conclusions and final remarks

I. Phytoplankton cope with metal stress in natural waters by usmg a range of

protective mechanisms. These mechanisms can complement themselves or be more

specific for situations of enhanced metal toxicity, such as with the production of the

- 41 - Chapter 2 Phytochelatins as a11 i11dicatio11 of metal stress

metal-binding peptides phytochelatins. Coastal waters with elevated free metal ion

concentrations have shown high particulate phytochelatin concentrations.

2. From laboratory experiments, it has been outlined that factors such as

phytoplankton species composition, metal competition and effects of metal

combinations in natural waters are to be considered when comparing laboratory and

field data, as they affect phytochelatin production.

3. Despite the range of variables that can affect phytochelatin synthesis, the

production of such metal-binding peptides by phytoplankton has been suggested as

a valuable biomarker of metal stress.

4. It can be anticipated that a thorough characterisation of the environmental

conditions under which phytochelatins and glutathione are produced by

measurements of nutrients, phytoplankton species composition, trace metal

speciation, and chlorophyll a should help assessing the effectiveness of

phytochelatin production as a metal detoxification mechanism and also as a stress

biomarker.

2.9. Reference List

ADAMS, S.M. & GREELEY, M.S. (2000) Ecotoxicological indicators of water quality: Using multi-response indicators to assess the health of aquatic ecosystems. Water, Air and Soil Pollution, 123, pp. 103-115.

AHNER, B.A., KONG, S. & MOREL, F.M.M. (1995) Phytochelatin Production in Marine-Algae .1. An Interspecies Comparison. Limnology and Oceanography, 40, pp. 649- 657.

AHNER, B.A., LEE, J.G., PRICE, N.M. & MOREL, F.M.M. (1998) Phytochelatin concentrations in the equatorial Pacific. Deep-Sea Research Part !-Oceanographic Research Papers, 45, pp. 1779-1796.

- 42- Chapter 2 Phytochelatins as an indication ofmetal stress

AHNER, B.A. & MOREL, F.M.M. (1995) Phytoche1atin Production in Marine-Algae .2. Induction by Various Metals. Limnology and Oceanography, 40, pp. 658-665.

AHNER, B.A., MOREL, F.M.M. & MOFFETT, J.W. (1997) Trace metal control of phytochelatin production in coastal waters. Limnology and Oceanography, 42, pp. 601- 608.

AHNER, B.A., PRICE, N.M. & MOREL, F.M.M. (1994) Phytochelatin Production by Marine-Phytoplankton at Low Free Metal-Ion Concentrations - Laboratory Studies and Field Data from Massachusetts Bay. Proceedings of the National Academy of Sciences of the United States ofAmerica, 91, pp. 8433-8436.

AHNER, B.A., WEI, L.P., OLESON, J.R. & OGURA, N. (2002) Glutathione and other low molecular weight thiols in marine phytoplankton under metal stress. Marine Ecology­ Progress Series, 232, pp. 93-103.

BAE, W., CHEN, W., MULCHANDANI, A. & MEHRA, R.K. (2000) Enhanced bioaccumulation of heavy metals by bacterial cells displaying synthetic phytochelatins. Biotechnology and Bioengineering, 70, pp. 518-524.

BONTIDEAN, I., AHLQVIST, J., MULCHANDANI, A., CHEN, W., BAE, W., MEHRA, R.K., MORTARI, A & CSOREGI, E. (2003) Novel synthetic phytochelatin-based capacitive biosensor for heavy metal ion detection. Biosensors & Bioelectronics, 18, pp. 547-553.

BRYAN, G.W. & LANGSTON, W.J. (1992) Bioavailability, accumulation and effects of heavy-metals in sediments with special reference to United-Kingdom estuaries- a review. Environmental Pollution, 16, pp. 89-131.

BUCKBERRY, L.D. & TEESDALE-SPITTLE, P.H. (1996) Biological interactions of sulfur compounds. S.Mitchell, Taylor & Francis, London.

BUTLER, A ( 1998) Acquisition and utilization of transition metal Ions by man ne organisms. Science, 281, pp. 207-210.

CAMPBELL, P.G.C. (1995) Metal speciation and bioavailability in aquatic systems. pp. 45-101. Wiley, New York.

CID, A, HERRERO, C., TORRES, E. & ABALDE, J. (1995) Copper Toxicity on the Marine Microalga Phaeodactylum- Tricomutum - Effects on Photosynthesis and Related Parameters. Aquatic Toxicology, 31, pp. 165-174.

CID, A, TORRES, E., HERRERO, C. & ABALDE, J.E. (1997) Disorders provoked by copper in the marine diatom Phaeodactylum tricomutum in short-time exposure assays. Cahiers de Biologie Marine, 38, pp. 201-206.

COBBETT, C.S. (2000) Phytochelatins and their roles in heavy metal detoxification. Plant Physiology, 123, pp. 825-832.

CROOT, P.L., MOFFETT, J.W. & BRAND, L.E. (2000) Production of extracellular Cu complexing ligands by eucaryotic phytoplankton in response to Cu stress. Limnology and Oceanography, 45, pp. 619-627.

- 43- Chapter 2 Phytochelatins as an indication of metal stress

DEKNECHT, J.A., VANBAREN, N., TENBOOKUM, W.M., SANG, H.W.W.F., KOEYOETS,P.L.M., SCHAT,H. & VERKLEIJ,J.A.C. (1995) Synthesis and degradation of phytochelatins in cadmium-sensitive and cadmium-tolerant Silene vulgaris. Plant Science, 106, pp. 9-18.

FOGG, G.E. & THAKE, B. (1987) Phytoplankton in its natural environment. ln: Algal cultures and phytoplankton ecology pp. 81-121. The University of Wisconsin Press, London.

GLEDHILL, M., NIMMO, M. & HILL, S.J. (1999) The release of copper-complexing ligands by the brown alga Fucus vesiculosus (Phaeophyceae) in response to increasing total copper levels. Journal of Phycology, 35, pp. 501-509.

GORDON, AS., DONAT, J.R., KANGO, RA., DYER, B.J. & STUART, L.M. (2000) Dissolved copper-complexing ligands in cultures of marine bacteria and estuarine water. Marine Chemistry, 70, pp. 149-160.

GRILL, E., WINNACKER, E.L. & ZENK, M.H. (1985) Phytochelatins - the Principal Heavy-Metal Complexing Peptides of Higher-Plants. Science, 230, pp. 674-676.

KNAUER, K., AHNER, B., XUE, H.B. & SIGG, L. (1998) Metal and phytochelatin content in phytoplankton from freshwater lakes with different metal concentrations. Environmental Toxicology and Chemistry, 17, pp. 2444-2452.

KNEER, R. & ZENK, M.H. ( 1992) Phytochelatins protect plant enzymes from heavy­ metal poisoning. Phytochemistry, 31, pp. 2663-2667.

KOZELKA,P.B. & BRULAND,K.W. (1998) Chemical speciation of dissolved Cu, Zn, Cd, Pb in Narragansett Bay, Rhode Island. Marine Chemistry, 60, pp. 267-282.

LAGE, O.M., PARENTE, AM., VASCONCELOS, M.T.S.D., GOMES, C.A.R. & SALEMA,R. (1996) Potential tolerance mechanisms of Prorocentrum micans ( dinophyceae) to sublethal levels of copper. Journal of Phycology, 32, pp. 416-423.

LANGSTON, W.J., CHESMAN, B.S., BURT, G.R, POPE, N.D. & MCEYOY, J. (2002) Metallothionein in liver of eels Anguilla anguilla from the Thames Estuary: an indicator of environmental quality? Marine Environmental Research, 53, pp. 263-293.

LE GALL, A.C.A. & V AN DEN BERG, C.M.G. (1998) Folic acid and glutathione in the water column of the north east Atlantic. Deep-Sea Research Part !-Oceanographic Research Papers, 45, pp. 1903-1918.

LEE, J.G., AHNER, B.A. & MOREL, F.M.M. (1996) Export of cadmium and phytochelatin by the marine diatom Thalassiosira weissflogii. Environmental Science & Technology, 30, pp. 1814-1821.

LEE, J.G. & MOREL, F.M.M. (1995) Replacement of Zinc by Cadmium in Marine­ Phytoplankton. Marine Ecology-Progress Series, 127, pp. 305-309.

MATRAI, P.A. & VETTER, RD. (1988) ?articulate Thiols in Coastal Waters- the Effect of Light and Nutrients on Their Planktonic Production. Limnology and Oceanography, 33, pp. 624-631.

- 44- Chapter 2 Phytochelatins as an indication ofmetal stress

MATSUKA W A, R., HOTT A, M., MASUDA, Y., CHlliARA, M. & KARUBE, I. (2000) Antioxidants from carbon fixing Chlorella sorokiniana. Journal of Applied Phycology, 12, pp. 263-267.

MEHRA, R.K., KODATI, V.R. & ABDULLAH, R. (1995) Chain Length-Dependent Pb(li)-Coordination in Phytochelatins. Biochemical and Biophysical Research Communications, 215, pp. 730-736.

MOFFETT, J.W., BRAND, L.E., CROOT, P.L. & BARBEAU, K.A. (1997) Cu speciation and cyanobacterial distribution in harbors subject to anthropogenic Cu inputs. Limnology and Oceanography, 42, pp. 789-799.

MORELLI, E. & SCARANO, G. (1995) Cadmium-Induced Phytochelatins in Marine Alga Phaeodactylum- Tricomutum - Effect of Metal Speciation. Chemical Speciation and Bioavailability, 7, pp. 43-47.

MORELLI, E. & SCARANO, G. (2001) Synthesis and stability of phytochelatins induced by cadmium and lead in the marine diatom Phaeodactylum tricomutum. Marine Environmental Research, 52, pp. 383-395.

MORRISON, R. & BOYD, R. (1997) Organic Chemistry. pp. 1389. Allyn & Bacon, Inc., Massachusetts.

MOUNEYRAC, C., AMIARD, J.C., AMIARD-TRIQUET, C., COTTIER, A., RAINBOW,P.S. & SMITH,B.D. (2002) Partitioning of accumulated trace metals in the talitrid amphipod crustacean Orchestia garnmarellus: a cautionary tale on the use of metallothionein-like proteins as biomarkers. Aquatic Toxicology, 57, pp. 225-242.

NOCTOR, G. & FOYER, C.H. (1998) Ascorbate and glutathione: Keeping active oxygen under control. Annual Review of Plant Physiology and Plant Molecular Biology, 49, pp. 249-279.

OKAMOTO, O.K., PINTO, E., LATORRE, L.R., BECHARA, E.J.H. & COLEPICOLO, P. (2001) Antioxidant modulation in response to metal-induced oxidative stress in algal chloroplasts. Archives of Environmental Contamination and Toxicology, 40, pp. 18-24.

PAWLIK-SKOWRONSKA, B. (2001) Phytochelatin production in freshwater algae Stigeoclonium in response to heavy metals contained in mining water; effects of some environmental factors. Aquatic Toxicology, 52, pp. 241-249.

PA WLIK-SKOWRONSKA, B. (2003) When adapted to high zinc concentrations the periphytic green alga Stigeoclonium tenue produces high amounts of novel phytochelatin­ related peptides. Aquatic Toxicology, 62, pp. 155-163.

PINTO, E., SIGAUD-KUTNER, T.C.S., LEITAO, M.A.S., OKAMOTO, O.K., MORSE, D. & COLEPICOLO, P. (2003) Heavy metal-induced oxidative stress in algae. Journal of Phycology, 39, pp. 1008-1018.

PISTOCCHI, R., MORMILE, M.A., GUERRINI, F., ISANI, G. & BONI, L. (2000) Increased production of extra- and intracellular metal-ligands in phytoplankton exposed to copper and cadmium. Journal ofApplied Phycology, 12, pp. 469-477.

- 45- Chapter 2 Phytochelatins as an indication of metal stress

RA USER, W.E. (1990) Phytochelatins. Annual Review of Biochemistry, 59, pp. 61-86.

RIJSTENB1L, J.W., DEHAIRS, F., EHRLICH, R. & WIJNHOLDS, J.A. (1998) Effect of the nitrogen status on copper accumulation and pools of metal-binding peptides in the planktonic diatom Thalassiosira pseudonana. Aquatic Toxicology, 42, pp. 187-209.

RIJSTENBlL, J.W., DERKSEN, J.W.M., GERRINGA, L.J.A., POORTVLIET, T.C.W., SANDEE, A., VANDENBERG, M., VANDRIE, J. & WlJNHOLDS, J.A. (1994) Oxidative Stress-Induced by Copper - Defense and Damage in the Marine Planktonic Diatom Ditylum-Brightwellii, Grown in Continuous Cultures with High and Low Zinc Levels. Marine Biology, 119, pp. 583-590.

RIJSTENBIL, J.W. & WIJNHOLDS, J.A. (1996) HPLC analysis of nonprotein thiols in planktonic diatoms: Pool size, redox state and response to copper and cadmium exposure. Marine Biology, 127, pp. 45-54.

SANDERS, B. (1990) Stress proteins: potential as multitiered biomarkers. In: Biomarkers of environmental contamination (Ed. by J .F.McCarthy & L.R.Shugart), pp. 165-191. Lewis Publisher, CRC Press, Florida.

SATOH, M., KARAK.I, E., KAKEHASHI, M., OKAZAKI, E., GOTOH, T. & OY AMA, Y. (1999) Heavy-metal induced changes in nonproteinaceous thiollevels and heavy-metai binding peptide in Tetraselmis tetrathele (Prasinophyceae). Journal of Phycology, 35, pp. 989-994.

SIGAUD-KUTNER, T.C.S., PINTO, E., OKAMOTO, O.K., LATORRE, L.R. & COLEPICOLO, P. (2002) Changes in superoxide dismutase activity and photosynthetic pigment content during growth of marine phytoplankters in batch-cultures. Physiologia Plantarum, 114, pp. 566-571.

STEGEMAN, J.J., BROUWER, M., DJ GIULIO, R.T., FORLIN, L., FOWLER, B.A., SANDERS, B. & V AN VELD, P.A. ( 1992) Molecular responses to environmental contamination: enzyme and protein systems as indicators of chemical exposure and effect. ln: Biomarkers. Biochemical, physiological, and histological markers of anthropogenic stress (Ed. by R.J.Huggett, R.A.Kimerle, P.M.Jr.Mehrle & H.L.Bergman), Lewis Publishers, London.

SUNDA, W. (1988) Trace metal interactions with manne phytoplankton. Biological Oceanography, 6, pp. 411-442.

SUNDA, W., KIEBER, D.J., KIENE, R.P. & HUNTSMAN, S. (2002) An antioxidant function for DMSP and DMS in marine algae. Nature, 418, pp. 317-320.

TANG, D.G., HUNG, C.C., WARNKEN, K.W. & SANTSCHI, P.H. (2000) The distribution of biogenic thiols in surface waters of Galveston Bay. Limnology and Oceanography, 45, pp. 1289-1297.

TORRES, E., CID, A., FIDALGO, P., HERRERO, C. & ABALDE, J. (1997) Long-chain class Ill metallothioneins as a mechanism of cadmium tolerance in the marine diatom Phaeodactylum tricomutum Bohlin. Aquatic Toxicology, 39, pp. 231-246.

- 46- Chapter 2 Phytochelatins as all indicatioll of metal stress

TORRES, E., CID, A, HERRERO, C. & ABALDE, J. (1998) Removal of cadmium ions by the marine diatom Phaeodactylum tricomutum bohlin accumulation and long-term kinetics of uptake. Bioresource Technology, 63, pp. 213-220.

VASCONCELOS, M.T.S.D. & LEAL, M.F.C. (2001) Adsorption and uptake of Cu by Emiliania huxleyi in natural water. Environmental Science & Technology, 35, pp. 508-515.

WEI, L.P., DONAT, J.R., FONES, G. & AHNER, B.A. (2003) Interactions between Cd, and Cu, and Zn influence particulate phytochelatin concentrations in marine phytoplankton: Laboratory results and preliminary field data. Environmental Science & Technology, 37, pp. 3609-3618.

- 47- Chapter 3

3. Determination of particulate glutathione and phytochelatins in natural waters using HPLC with fluorescence detection

3.1. Abstract

Phytochelatins and glutathione are present at low concentrations in natural waters. Very little is known about the variations in the concentrations of glutathione and phytochelatins produced by natural phytoplankton assemblages in the field. The current analytical methods for particulate thiols in natural waters are based on high performance liquid chromatography (HPLC) with detection of fluorescent derivatives, and comprise expensive time-consuming multi-step protocols. The aim of this chapter is: (i) to review the common analytical approaches, the limitations and main problems related to the current methods for the determination of GSH and PCs in phytoplankton from natural waters; (ii) to identify key strategies for the determination of PCs in the field based on the review; (iii) to discuss the application of a method based on fluorescence detection of derivatised GSH and PCs.

An analytical method was optimised and is exemplified here for estuarine waters. The need for suitable analytical strategies to provide quality assurance and further the use of PCs as biomarkers is highlighted.

- 48- Chapter 3 Determination ofparticulate glutathione and phytochelatins

3.2. Introduction

Phytochelatins and GSH have been extensively investigated m plant cells and phytoplankton monocultures (Ahner et al. 1995; Cobbett 2000). However, there 1s a paucity of data on the variations in the concentrations of these compounds in natural phytoplankton assemblages. The determination of GSH is well-known for clinical applications and has been the subject of a number of recent reviews (Lock & Davis 2002;

Rosenfeld 2003). A range of techniques is available for the determination of thiols

(including PCs, GSH, cysteine, homocysteine, metallothioneins) in clinical and environmental samples. The performance characteristics of some of the analytical techniques for the determination ofthiols are summarised in Table 3.1.

Methods based on electrochemical detection, such as cathodic stripping voltammetry and polarography (Nyberg & Zhou 1995; Winters et al. 1995; Zhang et al. 2002; Yosypchuk et al. 2003), are able to measure thiols directly without the need of a derivatisation step, but they lack in specificity. Sophisticated hyphenated techniques, such as HPLC with inductively coupled plasma-mass spectrometry (ICP-MS) and electrospray-mass spectrometry (ES-MS) (Vacchina et al. 1999a; 1999b; 2000) have provided information on the structure of PCs as metal-complexes and other thiols in plasma from different animal origins. HPLC and capillary electrophoresis can be coupled to different detectors and allow selectivity of a range of thiols in different redox states (Zhou et al. 1994; Kleinman &

Richie 1995; Nozal et al. 1997; Chassaing et al. 1999; Ercal et al. 2001). Nevertheless, the detection limits of these methods are in the nM-11M range and not sufficiently low for the determination of thiols in the field, particularly PCs produced by natural phytoplankton assemblages.

- 49- Table 3.1. Characteristics of the instrumental methods for the determination ofthiols. DL =detection limit; NM =not mentioned; CLC =capillary liquid Q chromatograEh;t; SEC =size-exclusion chromatograEh;t; CE = caEillary electroEhoresis .g ;;;... Separation Detection Matrix Analyte DL (f..LM) Comments Reference <... Polarography P. tricornutum PCs, metallothioneins NM Cu electrode Nyberg & Zhou, (diatom) and Agrostis 1995 cappilaris (graminae) Cathodic stripping Standard solutions GSH, PC2, PC3 0.002 Cu solid amalgam electrode Yosypchuk et al. voltammetry 2003 CLC Amperometry Urine GSH, cysteine, 0.2-0.5 Carbon electrode Zhang et al. 2002 dopamine, thiopurine HPLC Electrochemical Tissue Various 0.008 Dual Hg electrode Kleinman & Richie, 1995 V> HPLC UVNis Rabbit eye tissue GSSG 0.08 Post-column DTNP Nozal et al. 1997 0 ~ derivatisation ~ ~· SEC ICP-MS Plants PCs 10 PCs as Cd-complexes Vacchina et al. S· "' 1999b .,~ " HPLC ICP-MS ES- Plants PCs NM PCs as Cd-complexes Vacch in a et al. ~- MS/MS 2000 " ";:;- HPLC Fluorimetry Various tissues GSSG 0.005 thiolGlo I Pyrene labels Ercal et al. 200 I "";; CE Electrochemical Urine GSSG 6 Ru oxide electrode Zhou et al. 1994 ~ ~ CE Fluorimetry Plasma Total thiols 3 Fluorescein label Chassaing et al. .,"5.. 1999 ~ ...,0 "'" "'§:" :::· <:; Chapter 3 Determination ofparticulale glulalhione and phylochelalins

Selective and very sensitive analytical methods are required, as PCs are at concentrations in the pM range in natural waters. The general routes that can be followed for the determination ofthiol compounds are presented in Fig. 3.1.

The aim of this chapter is: (i) to review the common analytical approaches, the limitations and main problems related to the current methods for the determination of GSH and PCs in phytoplankton from natural waters; (ii) to identify key strategies for the determination of

PCs in the field based on the review; (iii) to discuss the application of a reverse-phase

HPLC method based on fluorescence detection of derivatised GSH and PCs. An analytical method was optimised and is exemplified here for estuarine waters. The need for suitable analytical strategies to provide quality assurance and further the use of PCs as biomarkers is highlighted.

- 5 I - Chapter 3 Determination ofparticulate glutathione and phytochelatins

R-S-S-R

R-SH I Reduction or direct measurement

I x-s·

TCEP or I BHi or R-S-S-X Na(Hg) or + Rs· S03 " eN· Zn I x-s· 1 1 I RSS03• RSCN X-S-S-X + + 1 + Rs· Rs· 2 Rs· 2Rs· I I I lt I Electrochemical R-SH detection '!1 Derivatisation or direct separation '------....J I I mBrB or I SBD-F or Disulphide---~ OPA exchange

1- ______l ______l~ I HPLC or

I I Capillary electrophoresis ·------I ------1 Spectroscopic detection

I UVNIS I

Electrochemical detection I Fluorescence I

Mass spectroscopy: UVNIS ICP-MS; ES-MS I I

Figure 3.1. Possible routes for the determination of thiol compounds (adapted from Lock & Davis, 2002). R-S-SR =disulphide. R-SH =reduced thiol.

-52- Chapter 3 Determination ofparticulate glutathione and phytochelatins

3.3. Analytical methods for particulate GSH and PCs in natural waters

In this work, particulate PCs and GSH refer to the fraction retained on 0.7-0.8 J..lm pore size filters. The few current analytical methods for particulate PCs and GSH in natural waters comprise multi-step procedures: 1) sampling, 2) sample filtration, 3) extraction of thiols, 4) reduction reactions, 5) derivatisation, and 6) analysis of thiol-derivatives by HPLC (Ahner et al. 1997; Tang et al. 2000; 2003).

All steps should be carried out in a careful manner due to the instability of thiol-peptides in aqueous solutions and their rapid oxidation (Winters et al. 1995). A general scheme of an analytical method for particulate PCs and GSH applied to estuarine water samples is summarised in Fig. 3.2. This method was originally optimised for thiols produced by phytoplankton monocultures (Rijstenbil & Wijnholds 1996), grown under controlled laboratory conditions, for which sufficient phytoplankton biomass can be obtained for PC detection.

3.3.1. Sample collection, handling and preparation

Phytochelatins and GSH are present at low concentrations in phytoplankton from natural assemblages and are subject to rapid oxidation once isolated from cells. Therefore, clean techniques and rapid sample preparation to avoid degradation and losses of the particulate compounds should be undertaken as part of the analytical protocol.

-53- Chapter 3 Determination ofparticulate glutathione and phytochefatins

SAMPLE COLLECTION

Acid cleaned polycarbonate bottles (2-4 L) I FILTRATION

Nitrate cellulose filter (47 mm; 0.8 J.lm pore size) gentle vacuum pressure (< 5 psi) I EXTRACTION

Addition of 1.2 ml of 0.1 M HCI in 5 mM DTPA in 1.5 ml microtube Ultrasonication (0°C; 5 min) Centrifugation (20 min; 13000 g; 4°C) I REDUCTION

Buffered with 650 Ill of 200 mM HE PES in 5 mM DTPA (pH = 8.2) Addition of 25 Ill of 20 mM TCEP (Reduction: R-S-S-R + TCEP = 2 RSH) l DERIVATISATION

Addition of 10 Ill of 100 mM mBrB in acetonitrile (R-SH + mBrB = R-S-mB + HBr) Acidification with 100 Ill of 1 M MSA l HPLC ANAL VS IS

Reverse phase, C-18 column Fluorescence detection (470 nm emission; 380 nm excitation)

Figure 3.2. A summary of the optimised analytical method for particulate phytochelatins and glutathione (oxidised + reduced forms) in natural waters, adapted from Rijstenbil & Wijnholds ( 1996).

-54- Chapter 3 Determination ofparticulate glutathione and phytoche/Qiins

3.3.1.1. Reagents and laboratory ware

As organic and trace metal contamination can promote oxidation of reagents and thiol compounds, all laboratory ware were metal- and organic-free. Glassware and plastic bottles were soaked in I M HCI (Aristar) for at least 24 h, and then rinsed 3 times with de­

1 ionised water(> 18.2 MQ cm- ). Reagents and solvents were HPLC or Aristar grade. All solutions were prepared with de-ionised water unless otherwise stated. Metal contamination was minimised during preparation and handling of the reagents and samples by working in a class 100 laminar flow hood.

3.3.1.2. Sample collection

Acid cleaned polycarbonate bottles were used to collect water samples. Large volumes of water samples (2-4 L) were needed to ensure collection of enough biomass (phytoplankton cells). Samples were typically collected from a depth of 0.3 m with the use of small boats.

Samples were placed in cool boxes until return to laboratory and filtered within 8 h tv minimise interferences from biological processes. Sub-samples were taken for chlorophyll a analysis, as this parameter is important to normalise PCs and GSH concentrations.

3.3.1.3. Filtration

The use of a manifold filtration system is highly recommended if a large number of samples are to be processed. In the method optimised here, samples were filtered under gentle vacuum pressure(< 5 psi) to avoid cell breakage and loss of material. The filtration step was carried out slowly to avoid back-pressure, which could clog the filter and lead to cell lysis.

-55- Chapter 3 Determination ofparticulate glutathione and phytochelatins

Nitrocellulose filters (Whatman) of 47 mm diameter and 0.8 J..lm pore size were preferred here for the estuarine water samples. It was observed that 0.45 J..lm pore size filters clogged easily with particles other than phytoplankton, not allowing enough biomass

(phytoplankton cells) for PC analysis. Other workers have employed GF/F filters (0. 7 J..lm pore size) (Ahner et al. 1997). The use of nitrate cellulose filters has provided better recoveries than glass fiber filters for extraction based on ultrasonication (Rijstenbil &

Wijnholds 1996). It may be necessary to combine 2 or more filters from the same sampling site in order to achieve sufficient biomass for particulate PC analysis (Wei et al. 2003).

This is especially worthwhile, again, because the filters can easily clog with less than l.S.L of estuarine water sample by particulate matter other than phytoplankton.

The filters containing the phytoplankton cells were then placed in microtubes (1.5 mL,

Eppendorf) for the extraction process. Filters can be either stored under liquid nitrogen

(Ahner et al. 1997) or immediately submitted to the extraction process.

3.3.1.4. Extraction of PCs and GSH

During the preparation of the phytoplankton extracts, the effects of enzymatic degradation of thiols and losses due to oxidation should be avoided (Fahey & Newton 1987). The use of HCl for extraction promotes denaturation of proteases, and the metal chelating reagent diethylenetriaminepentaacetic acid (DTP A) minimises oxidation of the SH groups. Therefore, the extraction of PCs and GSH from phytoplankton cells was carried out by addition of 1.2 mL of 0.1 M HCl containing 5 mM of DTP A. Phytoplankton cells were disrupted by ultrasonication (0°C, 5 m in) and the cell extract was centrifuged at high speed ( 13000 g, 20 m in, 4°C}. Alternatively, the extraction can be performed using 10 mM methanesulphonic acid (MSA) at 70°C for 2 min, the sample can then be homogenised in a grinding tube and

-56- Chapter 3 Determination ofparticulate glutathione and phytochelatins

centrifuged (Almer et al. 1994). The resulting supematant was immediately submitted to reduction and derivatisation reactions as described below.

3.3.1.5. Reduction reactions

Oxidised GSH and PCs (i.e. RS-SR, R = peptide with residual cysteic group) can be converted to free thiols (RS-H) by reduction using tris (2-carboxyethyl) phosphine hydrochloride (TCEP) or dithiothreitol (OTT) (Fahey & Newton 1987; Rijstenbil &

Wijnholds 1996; Getz et al. 1999). The reduction of thiols by TCEP is indicated in reaction

I.

TCEP oxidised thiol

The reducing reagent TCEP is highly corros1ve and is more expensive than OTT.

However, the use of TCEP has advantages over OTT, as the latter is a thiol itself and therefore can also react with the derivatising reagents producing interfering chromatographic peaks for PCs (Almer et al. 1995; Rijstenbil & Wijnholds 1996). In addition, the reaction time for the disulphide reduction using TCEP is shorter compared with OTT (Rijstenbil & Wijnholds 1996; Getz et al. 1999). Tributylphoshine (TBP) has also been successfully used to break disulphide bridges (Tang et al. 2000), but it requires careful handling due to its volatility. The use of fresh reducing agent solutions is recommended as oxidation of the reagent may occur during long-term storage (Getz et al.

1999).

-57- Chapter 3 Determination ofparticulate glutathione and plzytoclzelarins

An alkaline pH of 8-9 is maintained by buffer solutions during the reduction reactions because the pKa of the thiol moiety is approximately 9 (Nekrassova et al. 2003). It means that at this pH, the predominant species is the thiolate anion Rs·, a good nucleophile, which facilitates the subsequent derivatising reaction. The reagent HEPES is a commonly used pH buffer for biological analyses and can be used for this purpose. Chelating agents such as EOT A and OTP A can also be added in the reaction medium to minimise SH oxidation by metals (Rijstenbil & Wijnholds 1996; Getz et al. 1999).

In the method exemplified here (Fig. 3.2), GSH and PCs were obtained as the total thiol concentrations (oxidised plus reduced form). HEPES was added to 250 J.lL aliquot of the extract (final concentration of 0.1 M, pH = 9) together with TCEP or OTT (final concentration of 0.5 mM in the reaction medium). The reduction reaction was carried out for 5 min (TCEP) or 1 h (OTT). The concentrations of oxidised thiols can be determined by omitting the reduction step for a sample aliquot and subtracting the result from the total thiol concentration. Thus, the thiol redox ratios used to measure metal-induced oxidative stress, [RS-H]/([RS-H] + 0.5 x [RS-SR]) can be determined (Rijstenbil & Wijnholds

1996).

3.3.1.6. Derivatisation of PCs and GSH

The derivatisation of PCs and GSH enhances sensitivity. The application of a derivatising reagent for PCs and GSH which leads to a minimum of interfering reagent peaks is required. A range of potential derivatising reagents for PCs and the corresponding thiol­ derivatives are listed in Table 3.2. From these reagents only mBrB and SBO-F have been used for derivatisation of PCs produced by phytoplankton. Oerivatising reagents for UV detection, such as DTNP, are not sensitive enough for determination of PCs produced by

-58- Chapter 3 Determination ofparticu/ate glutathione and phytochelatins phytoplankton, although they have been successfully applied for PCs produced by higher plants (Fahey & Newton 1987; Rijstenbil & Wijnholds 1996). MBrB is one of the most commonly used reagents, however the chromatographic methods for bimane-derivatives are time-consuming because of the need to eliminate fluorescent materials and to re­ equilibrate the system. MBrB reacts preferentially with thiols, but amines, phosphates, and carboxylates at millimolar concentrations react slowly with mBrB and can interfere with the analysis (Fahey & Newton 1987).

Derivatisation of PCs using SBD-F requires very sensitive fluorescence detectors, with adequate optical filters (Tang 2000, pers. comm). OPA-derivatives are stable only for short periods (- 3 h) and side-reactions of OP A with the amino groups of the peptides can occur

(Mapper & Delmas 1984), limiting the application ofOPA. The use ofNPM, a maleimide, as a pre-column derivatisation may provide detection limits in the nM range for GSH, but can produce diastereomers (Winters et al. 1995).

The derivatisation of PCs and GSH with mBrB (in acetonitrile, final concentration of 1 mM) was carried out in dim light at room temperature. The derivatising reagent was in excess (1-2 mM stoichiometric excess) to ensure rapid and quantitative reaction (Fahey &

Newton 1987). After 15 min, the reaction products were stabilised with MSA (final concentration of 0.1 M). The resulting derivatives were stored at 4°C, in the dark, until

HPLC analysis (Rijstenbil & Wijnholds 1996). The blank of the method consisted of the same volume of the extracting medium, instead of sample extract, using the same analytical conditions.

-59- Chapter 3 Determination ofparticulate glutathione and phytochelatins

Table 3.2. Reagents for derivatisation of thiols, their respective thiol products and the detection methods. UV = ultraviolet; FL = fluorescence.

Reagent type Structure Thiol derivative and other Detection roducts Disulphide exchange, RS-o-N0 2 DTNP uv (Vairavamurthy 02N-Q-S-S-o-N0 2 and Mopper, 1990)

lsoindole, OPA CHO SR FL (Mopper and ~ ojN-CH2CH20H + H;,O CC CHO Delmas, 1984)

Haloacetamide, MBrB FL (Faye and Newton, 1987)

H H Maleimide, 0~0 HtlSR 0 N 0 NPM FL (Winters et al. .Q 1995) &9.Q &9

Benzoxadizole, F SR

SBD-F FL HF ,..._ / ,..._ / + (Tang et al. 2000) 9:\N

-60- Chap/er 3 Delerminalion ofpar/iculale glulalhione and phywchelalins

3.3.2. Sample analysis

The difficulties in detennining particulate PCs in natural phytoplankton assemblages are clearly reflected by the few field data reported so far. Some of the factors that pose limitations to the sample analyses include the acquisition of high quality standards of PCs, the time spent with the current HPLC analyses and lack of procedures to assess analytical perfonnance.

3.3.2.1. GSH and PC standards

Glutathione standard (GSH, 98%) and phytochelatins (PC2, n = 2) were used for calibration purposes. GSH standard can be obtained from a number of companies at different purity levels, whereas commercial PC standards are not readily available. The syntheses and purification of PC standards are at request from a limited number of biochemistry laboratories, and are usually expensive. In the present study, fresh stock standard solutions and working solutions were prepared at concentrations of 0.01 M and

10-100 J..LM, respectively, in a solution containing 0.12 M HCI and 5 mM DTPA in order to minimise oxidation. Alternatively, standard solutions can be prepared in MSA (Fahey &

Newton 1987).

3.3.2.2. Phytochelatins produced by diatom cultures (Phaeodactylum tricornutum)

Some workers have used the GSH calibration to quantify the concentrations of PCs, assuming that the fluorescence response is directly proportional to the number of SH groups (Ahner et al. 1995; Rijstenbil & Wijnholds 1996). As an alternative source for the

- 61 - Chapter 3 Determination ofparticulate glutathione and phytochelatins

PC standards synthesised in laboratory, PCs produced by algae or plants can be used for compound identification in samples. The marine diatom Phaeodactylum tricornutum is an efficient producer of PCs, presents high metal-tolerance and is relatively easy to culture.

Furthermore, several studies have characterized PCs produced by this diatom under metal stress (Cid et al. 1996; Torres et al. 1997; Morelli & Pratesi 1997; Morelli & Scarano 2001;

Morelli et al. 2002; Scarano & Morelli 2002).

Phytochelatins with short chains (PC2, PC3, PC4 and PC5, n = 2 to 5) have been obtained

2 by extraction from P. tricornutum cultures exposed to single additions of Cd +, Cu2+ or

2 Pb + (Ahner et al. 1995; Rijstenbil & Wijnholds 1996; Scarano & Morelli 2002). For this purpose, cultures of P. tricornutum in the synthetic medium Aquil (Price et al. 1988) in exponential growth phase were exposed to total I 0 11M Cd (or Cu, Pb) for 24 h under continuous light, at 15°C. Duplicate culture samples of 500 mL were filtered and the analysis of thiols from the diatom cells proceeded as described for natural water samples.

Control experiments, without Cd addition, were performed in parallel to check for basal levels of PCs and GSH. If necessary, a thorough characterization of the medium conditions with calculation of metal speciation can be carried out using the software MINEQL+

(Schecher & Mcavoy 1992).

3.3.2.3. Stability of GSH and PCs in the derivatised form

After the labeling reaction with mBrB, oxidative loss of thiols can still occur (Fahey &

Newton 1987). Therefore, it is important to assess the effects of storage on sample integrity if derivatised samples are to be stored for later HPLC analyses. SBD-F and mBrB derivatives have been reported to be stable for 10 days and 6 weeks, respectively (4°C, in the dark) (Rijstenbil & Wijnholds 1996; Tang et al. 2000).

- 62- Chapter 3 Determination ofparticulate glutathione and phytochelatins

In our laboratory conditions, the stability of a sample (algal extract) was checked under the same HPLC conditions. GSH as bimane derivative (Table 3.2) underwent 30% degradation after storage in the fridge ( 4°C) for 15 days. Phytochelatins as bimane derivatives were stable over a 30 day-period (at 4°C) without significant losses (Fig. 3.3).

0 GSH ~ PC2 ~ PC3 • PC4

0 20000 40000 60000 80000 1 00000 120000 peak area

Figure 3.3. Stability of GSH and PCs in their bimane derivative form. Bars span the range from the average for triplicate injections.

3.3.2.4. Chromatographic conditions

The separation of thiol-bimane derivatives can be performed using a reverse-phase HPLC system coupled to columns such as C-18. The use of an ion-pairing reagent, such as tetraoctyl ammonium bromide, in the mobile phase can help separation of PCs from many bimane interfering compounds by prolonging the retention times of PCs (Ahner et al.

1995). However, this does not allow separation of GSH, which co-elutes with other compounds at the beginning of the chromatogram. The separation of OPA-derivatives requires a guard-column (Mopper & Delmas 1984; Vairavamurthy & Mopper 1990),

- 63- Chapter 3 Determination ofpartict1/ate glutathione and phytochelatins

because precipitation/flocculation of by-products occurs and can cause obstruction of the

analytical column.

The chromatograms shown in Fig. 3.4 were obtained using an HPLC consisting of two

pumps (Merck-Hitachi Models L-6200 and L-6000), a Rheodyne injection valve with a

I 00 J.LL loop and a Model FD-300 fluorescence detector (Dionex) operating at 380 nm

(excitation) and 470 nm (emission) wavelengths. The separation was carried out using a

150 x 4.6 mm C-18 (Econosphere) HPLC column with 3 J.Lm particle size. Solvent A was

0.1% trifluoroacetic acid (TFA) and solvent B was acetonitrile. The flow rate of the solvent

1 was 1.0 mL min- • The gradient adapted from Rijstenbil & Wijnholds (1996) was: 0 to 13

min, 10 to 21% B; 13 to 33 min, 21 to 35% B; 33 to 40 min, 35 to 100% B; 40 to 50 min,

isocratic 100% B; 50 to 65 m in, 100 to 10% B. This chromatographic condition allows the

separation of derivatised GSH and PCs produced by phytoplankton in a single run.

Calibration curves for the bimane derivatives of GSH and PC2 were obtained using

standard solutions at concentrations ranging from I 0 to 110 and 0.5 to 10 pmol,

respectively. The detection limits for GSH and PC2, which were calculated using three

times the standard deviation of the standard compound with the lowest concentrations,

were 1.0 and 0.9 pmol for a I 00 J.LL injection loop, respectively.

3.3.2.5. Reagent blank and samples

Although the current reverse-phase HPLC methods can provide good separation and

relative high sensitivity for particulate PCs and GSH, some analytical problems were

encountered. For the chromatograms exemplified here (Fig. 3.4), the broad peak at -12.5

min, presenting an intense fluorescence response, was identified as tetramethylbimane

(Me4B) by using chromatography of a tetramethylbimane standard. This compound was

-64-

, ' f I Chapter 3 Determination ofparticulate glutathione and phytochelatins used as part of the synthesis of mBrB. The peaks at the beginning of the chromatogram were the result of fluorescent breakdown products.

a) /

F L u 0 R b) E s GSH c E N c E

c) GSH PC2 PC3 PC4

5 10 15 20 25 30

RETENTION TIME (min)

Figure 3.4. Chromatograms for a) the reagent blank of the method; b) an estuarine water sample (Fat Estuary, Aug-2002); and c) an extract of the marine diatom P. tricornutum (after exposure to I .4 ).!M Cu2T).

- 65- Chapter 3 Determination ofparticulate glutathione and phytochelatins

The chromatogram for a blank of the method is presented in Fig. 3.4a. The blank of the method contained all reagents apart from the sample extract, which was substituted by the same amount of the extracting medium (HCI/DTP A solution). The addition of cysteine to react with the excess of mBrB may improve the blank (Ahner et al. 2002). The most problematic chromatographic peak, however, was caused by Me4B, which does not react with cysteine and accumulates in the system, thus long elution times were needed to wash

Me4B off the column. A 200 J.lL sample injection loop has been used to increase the injected amount of analyte (Ahner et al. 1997) but this also leads to an increase of the interfering reagent peaks. Thorough rinsing with de-ionized water/acetonitrile was also necessary to clean the HPLC injector valve and sample microsyringe prior to the next sample injection.

?articulate GSH, PC2 and some non-identified peaks were observed in a water sample of a metal impacted estuary (Fig. 3.4b), in which PC production was highly pronounced.

Longer PC chains have not been detected in natural phytoplankton assemblages, which was likely due to the influence of the environmental conditions on thiol production rather than analytical problems. Most laboratory phytoplankton monocultures under metal stress, for instance, produced PC2 at higher concentrations than longer PCs.

The method presented here is sensitive for PCs produced in metal impacted waters and by laboratory phytoplankton monocultures (Fig. 3.4c) as they are at optimal culturing conditions (light supply, nutrient and vitamin concentrations) and provide higher biomass than the natural phytoplankton assemblages present in the water samples.

-66- Chapter 3 Determination ofparriculare glurarhione and phyrochelarins

3.3.2.6. Attempts to identify phytochelatins produced by P. tricornutum using LC-ESI-MS

Trifluoracetic acid (TF A) is the most commonly used mobile phase to separate peptides and proteins by reverse-phase liquid chromatography (LC). It forms ion-pairs with positive charged and polar groups on peptides and proteins to protect these sites from polar interactions and bring them to the hydrophobic reserved-phase surface of the column.

However, the use of TFA in reversed-phase electrospray-mass spectrometry (ESI-MS) analysis for peptides has some incompatibilities. Trifluoracetic acid is a signal supressor because the strong ion pairs between TF A and the peptides are not broken apart by the conditions in the electrospray ionisation (Garcia et al. 2002). A post-column addition of

50% acetic acid (or formic acid) in acetonitrile helps to minimise the signal supression by dilution of TFA (Garcia et al. 2002; Petritis et al. 2002). The use of acetic acid or formic acid at an equivalent concentration of TF A seems to be enough to keep the peptide side­ chain carboxylates protonated (Apffel et al. 1995).

To investigate the possibility to identify PCs produced by P. tricornutum, extracts of this diatom exposed to Cd (10 J..LM) were analysed using LC-ESI-MS. The same reverse-phase

HPLC system described above was coupled to the electrospray-mass-spectrometer

(Thermo). A standard solution ofGSH (5 J..LM) was also injected in the LC-ESI-MS system with and without post-column addition of 50% acetic acid in acetonitrile. A signal m/z =

497.7-498.7 was observed for GSH standard in the derivatised bimane form only after post-column addition of acetic acid (Fig. 3.5b). Phytochelatins were not identified in the algae extracts due to lack of sensitivity of the instrument. In order to identify PCs extracted

- 67- Chapter 3 Determination ofparticulate glutathione and phytochelatins from microalgae by the LC-ESI-MS method, larger volumes of phytoplankton cultures

(biomass) are needed for the incubation experiments.

e:\sivia\0811 01 '9sh02 08111101 14;06:59 100ul RI : 0.00 - 20.00 SM: 78 NL: 5.14E7 a) TIC MS Gsh01

~--~~~~=------~~------~~~~======~ L: 2.55E8 TIC MS gsh02 b)

mJz = 497.7-498.7 /

Figure 3.5. Mass spectra for GSH standard (5.0 ).LM). a) without and b) with post-column addition of 50% acetic acid in acetronitrile.

- 68- Chapter 3 Determination ofparticulate glutathione and phytochelatins

3.3.2.7. Quality assurance

The analytical errors, given the low concentrations and multi-step analytical processes involved, have to be properly quantified. The collection of samples in duplicate is usually performed and the analyses of duplicate samples provide I 0-20% variation in the concentration of thiols from the average (Tang et al. 2000; Wei et al. 2003). Triplicate samples are not carried out because of the length of the analysis. The addition of a thiol as internal standard to compensate for oxidation or other chemical reactions, and mechanical losses during sample processing is not adequate owing to differing rates of reaction for different thiols (Fahey & Newton 1987).

For a reliable application of PCs as biomarkers, important analytical aspects are still to be improved as a quality assurance, such as the development of adequate internal standards and certified reference materials for the concentrations of particulate GSH and PCs encountered over a range of natural waters. At present, the measurements of PCs and GSH have to rely on the analyses of replicates.

3.3.3. Application of the methods to natural waters

The current analytical methods for particulate PCs and GSH in the field consider the determination of total thiols (oxidised plus reduced). This may limit the interpretation of oxidative stress, but is acceptable because the thiol redox ratios do not appear to be a useful index for Cu-induced oxidative stress in several marine diatoms upon Cu exposure

(Rijstenbil & Wijnholds 1996). Phytoplankton species respond differently to a mixture of metals, and the production of thiols, more specifically an increase in the concentration of particulate PCs, provides a general indication of metal stress in situ. The concentrations of

- 69- Chapter 3 Determination ofparticulate glutathione and phytochelatins particulate GSH and PC2 in natural waters together with the analytical parameters of the methods applied are summarised in Table 3.3. Particulate GSH is found at concentrations in the nM range in natural waters and there are no major problems with the detection by the current methods. Particulate PCs, however, are present at pM concentrations and therefore require sensitive detection.

The analytical method exemplified here (Fig. 3.2) was successfully applied to water samples from the Fa! Estuary (UK), a heavily metal impacted estuarine system. However, this method lacks in sensitivity for particulate PCs from pristine waters (where the production of PCs as a metal stress response is less pronounced) and is time-consuming.

Another variation of this method, applied to waters from metal polluted harbours in New

England (USA) (Ahner et al. 1997), produces lower detection limits for PCs by employing a 200 11L injection sample loop, but does not allow separation of GSH. Further optimisation of this method, also employing bimane derivatives, have separated other relevant particulate thiols, such as cysteine and y-Glu-Cys. The quantification of such compounds may also be useful for field studies, as they are the precursors for the synthesis of GSH (Noctor & Foyer 1998). A more rapid method that takes 30 min for the chromatographic elution of GSH and PCs employs derivatives based on SBD-F (Tang et al. 2000); it is sensitive but not easy to reproduce as explained in section 3.5. A laborious method comprising derivatisation with DTNP followed by a second derivatisation with

OPA has been developed for thiols (but not for PCs) in water samples (Yairavamurthy &

Mopper 1990). In this method, the DTNP derivatives are suggested as a confirmation reaction for the compounds and also a mean to store thiols, as the OPA derivatives are not stable for long periods but enhance the sensitivity of the analysis.

- 70- Table 3.3. Particulate PCs and GSH in natural waters and the parameters of the analytical methods employed. DL = detection limit of compound g between parentheses; nm = not mentioned. {; ;;;...... Sample location GSH PC2 DL Injection Reduction Derivatisation Time Comments References pM (pmol) (J.LL) (min) New England 7- 120 0.3 (PC2) 200 OTT mBrB 80 Do not separate Ahner et al. (USA) GSH 1997 Galveston Bay 65- 720 7- 73 0.01 (GSH) 150 TBP SBD-F 30 Requires Tang et al. (USA) 0.08 (PC2) adequate 2000 optical filters Black Sea nm 0.05-0.1 f.!M 20 TBP DTPN+OPA 15- 30 Separates GSH Varaivamurthy and other sulfur (several thiols) & Mopper, compounds, -....J but no PCs, 1990 pre- ~ concentration ~ :;· in Sep-Pak columns "g· Elizabeth 10-2390 3-98 nm nm OTT mBrB 40 Separates GSH, Wei et al. ~ PCs, cysteine Estuary (USA) 2003 "'"::t and y-Giu-Cys 2' Fal Estuary 850- 2061 31 - 179 1.0 (GSH) 100 TCEP mBrB 75 Not sensitive This study ;;; for all kind of ";: (UK) 0.9 (PC2) "" natural waters "~ ~ "i5.. "'~ ":::- "'§: ~- Chapter 3 Determination ofparticulate glutathione and phytochelatins

The use of solid-phase extraction, with cartridges such as Sep-Pak C-18, to pre-concentrate

PCs from the dissolved phase, or just after their extraction from phytoplankton cells, may be advantageous. Nevertheless, very little analytical details regarding pre-concentration and stability of PCs on such columns have been provided (Lee et al. 1996). Cartridge­ adsorbed DTNP derivatives (not including PCs) have been reported to be stable at 0-5°C for long periods, and elution of these compounds from the cartridges using I mL methanol has provided quantitative results (Vairavamurthy & Mopper 1990). A combination of pre­ concentration with the derivatisation reaction is ideal as it minimises the analytical steps, but can lead to problems with interfering/problematic reagent peaks. A method involving solid-phase extraction ofthiols as bimane-derivatives (Tang et al. 2003), for example, has a detection limit for PC2 of 0.06 nM, but gives many chromatographic peaks of fluorescent breakdown products, and hence results in long HPLC chromatographic runs.

3.4. Conclusions

There is still a lack of standardised protocols for the analysis of thiols, particularly PCs, in natural waters despite their important roles in metal detoxification mechanisms and potential use as biomarkers of metal stress. Here, a method for the simultaneous determination of PCs and GSH was optimised and applied for estuarine waters. The use of filters of adequate pore size (0.7-0.8 J.Lm) is required for the collection of sufficient phytoplankton biomass from estuarine waters, as the presence of particles other than phytoplankton in turbid waters can rapidly clog the filters of smaller pore sizes (0.4 J.lm).

For the same reason, the combination of more filters from the same sampling site may be necessary in order to have enough biomass for PC analysis. The main analytical challenges identified were the low concentrations of these compounds in natural samples, their

-72- Chap/er 3 Delerminalion ofparliculale glulalhione and phywchelalins

instability in aqueous solutions, and requirement of multi-step procedures (filtration, extraction, reduction and derivatisation reactions prior to HPLC analysis). In addition, commercial PC standards were expensive as was the case for the reagents required (mBrB,

TCEP).

The current analytical methods are time consuming, and limit the number of samples to be analysed. A cost-effective, rapid and sensitive analytical method is a crucial aspect to

further the applications of PCs as indicators of metal stress in coastal waters. This is especially relevant if we consider that large sets of samples for environmental studies/monitoring are frequently needed, and time should also be required for determination of ancillary parameters such as chlorophyll a, trace metal speciation, phytoplankton composition, and nutrient levels in order to allow a full interpretation of the field data.

The need to further improve the analytical methods for particulate PC and GSH for field studies was highlighted. The aspects that need improvements and should enhance sensitivity are the pre-concentration step and derivatisation reaction. Sample enrichment using solid-phase extraction represents a viable alternative not yet properly optimised for

PC analysis. Some of the derivatising reagents still not explored for particulate PC analysis should provide better reagent blanks with less interfering chromatographic peaks. The use of PCs produced by a well known phytoplankton species (P. tricornutum) can circumvent problems with the high costs of PC standards. Suitable procedures to provide quality assurance for the determination of low levels of particulate GSH and PCs are to be developed for their effective application in environmental monitoring.

- 73- Chapter 3 Determination ofparticulate glutathione and phytochelatins

3.5. References

AHNER, B.A., KONG, S. & MOREL, F.M.M. (1995) Phytochelatin Production in Marine-Algae .I. An Interspecies Comparison. Limnology and Oceanography, 40, pp. 649- 657.

AHNER, B.A., MOREL, F.M.M. & MOFFETT, J.W. (1997) Trace metal control of phytochelatin production in coastal waters. Limnology and Oceanography, 42, pp. 601- 608.

AHNER, B.A., PRICE, N.M. & MOREL, F.M.M. (1994) Phytochelatin Production by Marine-Phytoplankton at Low Free Metal-Ion Concentrations - Laboratory Studies and Field Data from Massachusetts Bay. Proceedings of the National Academy of Sciences of the United States ofAmerica, 91, pp. 8433-8436.

AHNER, B.A., WEI, L.P., OLESON, J.R. & OGURA, N. (2002) Glutathione and other low molecular weight thiols in marine phytoplankton under metal stress. Marine Ecology­ Progress Series, 232, pp. 93-103.

APFFEL, A., FISCHER, S., GOLDBERG, G., GOODLEY, P.C. & KUHLMANN, F.E. (1995) Enhanced Sensitivity for Peptide-Mapping with Electrospray Liquid­ Chromatography Mass-Spectrometry in the Presence of Signal Suppression Due to Trifluoroacetic Acid-Containing Mobile Phases. Journal of Chromatography A, 712, pp. 177-190.

CHASSAING, C., GONIN, J., WILCOX, C.S. & WAINER, I.W. (1999) Determination of reduced and oxidized homocysteine and related thiols in plasma by thiol-specific pre­ column derivatization and cappilary electrophoresis with laser-induced fluorescence detection. Journal of Chromatography B, 735, pp. 219-227.

CID, A., HERRERO, C. & ABALDE, J. (1996) Functional analysis of phytoplankton by flow cytometry: A study of the effect of copper on a marine diatom. Scientia Marina, 60, pp. 303-308.

COBBETT, C.S. (2000) Phytochelatins and their roles in heavy metal detoxification. Plant Physiology, 123, pp. 825-832.

ERCAL, N., Y ANG, P. & A YKIN, N. (2001) Determination of biological thiols by high­ performance chromatography following derivatization with ThioGlo maleimide reagents. Journal of Chromatography B, 753, pp. 287-292.

FAHEY,R.C. & NEWTON,G.L. (1987) Determination of Low-Molecular-Weight Thiols Using Monobromobimane Fluorescent Labeling and High-Performance Liquid­ Chromatography. Methods in Enzymology, 143, pp. 85-96.

GARClA, M.C., HOGENBOOM, A.C., ZAPPEY, H. & IRTH, H. (2002) Effect of the mobile phase composition on the separation and detection of intact proteins by reversed­ phase liquid chromatography - electro spray mass spectrometry. Journal of Chromatography A, 957, pp. 187-199.

- 74- Chapter 3 Determination ofparticulate glutathione and phytochelatins

GETZ, E.B., XIAO, M., CHAKRABARTY, T., COOKE, R. & SELVIN, P.R. (1999) A comparison between the sulfhydryl reductants tris(2- carboxyethyl)phosphine and dithiothreitol for use in protein biochemistry. Analytical Biochemistry, 273, pp. 73-80.

KLEINMAN, W.A. & RICHIE, J.P. (1995) Determination of thiols and disulfides using high-performance liquid chromatography with electrochemical detection. Journal of Chromatography B, 672, pp. 73-80.

LEE, J.G., AHNER, B.A. & MOREL, F.M.M. (1996) Export of cadmium and phytochelatin by the marine diatom Thalassiosira weissflogii. Environmental Science & Technology, 30, pp. 1814-1821.

LOCK, J. & DAVIS, J. (2002) The determination of disulphide species within physiological fluids. Trac-Trends in Analytical Chemistry, 21, pp. 807-815.

MOPPER, K. & DELMAS, D. (1984) Trace Determination of Biological Thiols by Liquid­ Chromatography and Precolumn Fluorometric Labeling with Ortho- Phthalaldehyde. Analytical Chemistry, 56, pp. 2557-2560.

MORELLI, E., CRUZ, B.H., SOMOVIGO, S. & SCARANO, G. (2002) Speciation of cadmium - gamma-glutamyl peptides complexes in cells of the marine microalga Phaeodactylum tricomutum. Plant Science, 163, pp. 807-813.

MORELLI, E. & PRA TESI, E. ( 1997) Production of phytochelatins in the marine diatom Phaeodactylum tricomutum in response to copper and cadmium exposure. Bulletin of Environmental Contamination and Toxicology, 59, pp. 657-664.

MORELLI, E. & SCARANO, G. (2001) Synthesis and stability of phytochelatins induced by cadmium and lead in the marine diatom Phaeodactylum tricomutum. Marine Environmental Research, 52, pp. 383-395.

NEKRASSOV A, 0., LAWRENCE, N.S. & COMPTON, R.G. (2003) Analytical determination of homocysteine: a review. Talanta, 60, pp. 1085-1095.

NOCTOR, G. & FOYER, C.H. (1998) Ascorbate and glutathione: Keeping active oxygen under control. Annual Review of Plant Physiology and Plant Molecular Biology, 49, pp. 249-279.

NOZAL, M.J., BERNAL, J.L., TORIBIO, L., MARINERO, P., MORAL, 0., MANZANAS, L. & RODRIGUES, E. (1997) Determination of glutathione, cysteine and N-acetylcysteine in rabbit eye tissues using high-performance liquid chromatography and pos-column derivatization with 5,5'-dithiobis(2-nitrobenzoic acid). Journal of Chromatography A, 778, pp. 347-353.

NYBERG, S. & ZHOU, L.Z. (1995) Polarography As A Tool in Peptide and Protein­ Analysis - Studies on Metal-Chelating Substances Induced by Cadmium in the Algae Pheodactylum-Tricomutum and the Graminae Agrostis- Capillaris. Ecotoxicology and Environmental Safety, 32, pp. 147-153.

PETRITIS, K., BRUSSAUX, S., GUENU, S., ELFAKIR, C. & DREUX, M. (2002) Ion­ pair reversed-phase liquid chromatography-electrospray mass spectrometry for the analysis of underivatized small pep tides. Journal of Chromatography A, 957, pp. 173-185.

- 75- Chapter 3 Determination ofparticulate glutathione and plrytoclrelatins

PRICE, N.M., HARRISON, G.I., HERING, J.G., HUDSON, R.J., NIREL, P.M.V., PALENKl, B. & MOREL, F.M.M. (1988) Preparation and chemistry of the artificial algal culture medium Aquil. Biological Oceanography, 6, pp. 443-461.

RIJSTENBIL, J.W. & WIJNHOLDS, J.A. (1996) HPLC analysis of nonprotein thiols in planktonic diatoms: Pool size, redox state and response to copper and cadmium exposure. Marine Biology, 127, pp. 45-54.

ROSENFELD, J.M. (2003) Derivatization in the current practice of analytical chemistry. Trac-Trends in Analytical Chemistry, 22, pp. 785-798.

SCARANO, G. & MORELLI, E. (2002) Characterization of cadmium- and Jead­ phytochelatin complexes formed in a marine microalga in response to metal exposure. Biometals, 15, pp. 145-151.

SCHECHER, W.D. & MCAVOY, D.C. (1992) Mineql+- A Software Environment for Chemical-Equilibrium Modeling. Computers Environment and Urban Systems, 16, pp. 65- 76.

TANG, D.G., SHAFER, M.A., V ANG, K., KAMER, D.A. & AMSTRONG, D.E. (2003) Determination of dissolved thiols using solid-phase extraction and liquid chromatographic determination of fluorescently derivatized thiolic compounds. Journal of Chromatography A, 998, pp. 31-40.

TANG, D.G., WEN, L.S. & SANTSCHI, P.H. (2000) Analysis ofbiogenic thiols in natural water samples by high- performance liquid chromatographic separation and fluorescence detection with ammonium 7-fluorobenzo-2-oxa-1,3-diazole-4- sulfonate (SBD-F). Analytica Chimica Acta, 408, pp. 299-307.

TORRES, E., CID, A., FIDALGO, P., HERRERO, C. & ABALDE, J. (1997) Long-chain class Ill metallothioneins as a mechanism of cadmium tolerance in the marine diatom Phaeodactylum tricornutum Bohlin. Aquatic Toxicology, 39, pp. 231-246.

VACCHINA, V., CHASSAIGNE, H., OVEN, M., ZENK, M.H. & LOBINSKI, R. (1999a) Characterisation and determination of phytochelatins in plant extracts by electrospray tandem mass spectrometry. Analyst, 124, pp. 1425-1430.

VACCHINA, V., LOBINSKI, R., OVEN, M. & ZENK, M.H. (2000) Signal identification in size-exclusion HPLC-ICP-MS chromatograms of plant extracts by electrospray tandem mass spectrometry (ES MS/MS). Journal of Analytical Atomic Spectrometry, 15, pp. 529- 534.

VACCHINA, V., POLEC, K. & SZPUNAR, J. (1999b) Speciation of cadmium in plant tissues by size-exclusion chromatography with ICP-MS detection. Journal of Analytical Atomic Spectrometry, 14, pp. 1557-1566.

V A IRA V AMURTHY, A. & MOPPER, K. (1990) Field Method for Determination of Traces ofThiols in Natural- Waters. Analytica Chimica Acta, 236, pp. 363-370.

WEI, L.P., DONAT, J.R., FONES, G. & AHNER, B.A. (2003) Interactions between Cd, and Cu, and Zn influence particulate phytochelatin concentrations in marine

- 76- Chapter 3 Determination ofparticulate glutathione and phyrochelatins

phytoplankton: Laboratory results and preliminary field data. Environmental Science & Technology, 37, pp. 3609-3618.

WINTERS, R.A., ZUKOWSKI, J., ERCAL, N., MATTHEWS, R.H. & SPITZ, D.R. ( 1995) Analysis of Glutathione, Glutathione Disulfide, Cysteine, Homocysteine, and Other Biological Thiols by High-Performance Liquid-Chromatography Following Derivatization by N-( 1- Pyrenyi)Maleimide. Analytical Biochemistry, 227, pp. 14-21.

YOSYPCHUK, 8., SESTAKOVA, I. & NOVOTNY, L. (2003) Voltammetric determination of phytochelatins using copper solid amalgam electrode. Talanta, 59, pp. 1253-1258.

ZHANG, S., HUANG, F., ZHAO, J.W., WEN, L.J., ZHOU, F. & Y ANG, P.Y. (2002) Determination of thiols in urinary sample by capillary-column liquid chromatography with amperometric detection at a carbon electrode. Talanta, 58, pp. 451-458.

ZHOU, J.X., OSHEA, T.J. & LUNTE, S.M. (1994) Simultaneous detection of thiols and disulfides by cappilary electrophoresis electrochemical detection using a mixed-valence ruthenium cyanide-modified microelectrode. Journal of Chromatography A, 680, pp. 271- 277.

- 77 - Chapter 4

4. Effects of metal combinations on the production of phytochelatins and glutathione by Phaeodactylum tricornutum

4.1. Abstract

Copper, Zn and Cd have potential toxic effects on most phytoplankton species and can be found at elevated concentrations in contaminated coastal waters. In this chapter, the effects of these metals on the production of phytochelatins and glutathione by the marine diatom

Phaeodactylum tricornutum in laboratory cultures are examined. Single additions of Cu and Cd (I 0 11M total concentration) to the culture induced the production of short-chained phytochelatins (n = 2-5) in P. tricornutum, whereas single addition of Zn (I 0 11M total concentration) did not stimulate phytochelatins production. Combination of Zn with Cu resulted in similar phytochelatin production as for single Cu addition. Simultaneous exposure to Zn and Cd led to an antagonistic effect on phytochelatin production, which was probably caused by metal competition for cellular binding sites. Glutathione concentrations were affected only upon exposure to Cd (85% increase) or in combination with Zn (65% decrease), in relation to the control experiment. Ratios of phytochelatins to glutathione indicated a more pronounced metal stress for exposures to Cu and Cd combined with Zn. Although P. tricornutum is not a representative species in coastal waters, these findings indicate that variabilities in phytochelatin and glutathione production in the field can be explained in part as a result of metal competition for cellular binding sites.

- 78- Chapter 4 Effects ofcombined metals on phytochelatins and gllllathione

4.2. Introduction

Laboratory phytoplankton cultures have been useful for providing information on

phytochelatin and glutathione production under exposure to different trace metals, and also

for optimising the multi-step analytical methods for sulphydryl compounds. Experiments

using phytoplankton monocultures under stress of one single metal form a valuable tool to

assess the primary implications and toxic metal effects on phytoplankton cells, but give a

limited scenario for the interpretation of field observations. A range of laboratory

experiments have indicated that phytochelatin production is dependent upon the phytoplankton species, the degree of toxicity of metal ions and interactions among metals

(Ahner et al. 1995; Ahner & More11995; Knauer et al. 1998; Wei et al. 2003).

The toxic effects of metals on the marine diatom Phaeodacty/um tricornutum have been

1 well documented. Exposure to Cu and Cd at concentrations of 0.1-0.5 mg L- ( 1.6-7.9 J.!M) and 5-25 mg L- 1 (0.4-200 J.!M), respectively, can cause a strong decrease in growth, photosynthetic rate, chlorophyll a content, and intracellular A TP concentration in P. tricornutum cells (Cid et al. 1995; Torres et al. 2000). Phaeodactylum lricornulum is shown in Fig. 4.1. Its dimensions and shape depend on the culturing conditions. This diatom is considered one of the most metal-tolerant diatoms, presenting an EC50 (Effective

Concentration: concentration which reduces growth to 50% in relation to control) value for

200 J.!M Cd (Torres et al. 1997). This diatom species is an efficient producer of phytochelatins, capable of readily synthesising a wide range of phytochelatins under metal stress (Rijstenbil & Wijnholds 1996). Several studies have been undertaken to characterise phytochelatins produced by P. tricornulum (Morelli & Scarano 1995; Rijstenbil &

Wijnholds 1996; Morelli & Pratesi 1997; Torres et al. 1997; Scarano & Morelli 2002). The effects of combinations of toxic metals on phytochelatin and glutathione production by this

- 79- Chapter 4 Effects ofcombined metals on phyrochelatins and glutathione

diatom, however, have not yet been investigated. In coastal waters, a range of potential toxic metals can be present at varying concentrations which could cause different effects on organisms. Thus the study of the toxic effects of a mixture of metals is more relevant for environmental purposes than the effects of one single metal.

In this study, short-term metal exposure experiments using P. tricornutum cultures were performed with the aim to investigate the effects of single additions of Cu, Zn and Cd and combinations of these metals on phytochelatin and glutathione production. Algal extracts from Cd-treated cultures were furthermore used to assess the effectiveness of two reducing reagents for phytochelatin analysis.

Figure 4.1. Phaeodactylum tricornutum cells. http://marine.rutgers.edu/ebme/htm I_ does/staff/a quigg.htm ~"{~ .,-~-~-'.-·.~ 75-160 !-1m3 '""'"''­~~ -- '

4.3. Experimental

4.3.1. Culturing medium

Culturing medium was based on the synthetic medium Aquil, which was specially designed for studies examining the effects of trace metals on algae physiology (More! et al.

- 80- Chapter 4 Effects ofcombined metals 011 phytoclrelatins a11d glutathio11e

1979; Price et al. 1988). This medium comprises the major salts, nutrients, trace metals, vitamins and chelating agents ( ethylenediaminetraacetic acid, EDT A) at sufficient low concentrations to avoid precipitations and allows an accurate calculation of metal speciation. The need for such a medium arose as it became apparent that the bioavailability and toxicity of trace metals were functions of their chemical speciation, specifically their free ion activities (Morel et al. 1979; Price et al. 1988; Campbell 1995). Natural seawater can successfully replace the synthetic ocean water provided that it is collected from areas relatively free of metal impurities and organic matter, such as the Gulf Stream or Sargasso

Sea (Price et al. 1988).

In the present study, a stock solution of 20 L of synthetic ocean water (SOW) containing the major salts (AR, reagent grade) was prepared using de-ionised water Millipore (18.2

1 MQ cm' ). Stock solutions of vitamins, nutrients, the chelating agent EDTA, and trace metals were prepared separately in de-ionised water and added to the SOW in order to form the Aquil medium. All solutions were stored in acid cleaned high density polyethylene bottles. The concentrations of the components in the stock solutions and the final concentration in the culture medium are presented in Table 4.1. The culturing medium and bottles were microwaved prior to use (I min at high setting) in order to sterilise them. Trace metals were added after sterilisation and then the medium was allowed to equilibrate overnight before being seeded with diatom cells. The pH of the medium was 8.1.

- 81 - Chapter 4 Effects ofcombined metals orz phytochelatins and glutathione

Table 4.1. Composition of the culture medium Aquil and stock solutions used for the short-term met a I exposure expenments.. Substances Stock solutions (M) Final concentration in culture medium (M) Reagent grade salts 1 1 NaCI 4.2 X 10' 2.1 X 10' 2 2 Na2SO. 2.9 X 10' 1.5 X 10- 3 KC! 9.4 X J0' 4.7 X 10-J 3 3 NaHC03 2.4 X 10- 1.4 X 10- 2 2 MgCI2. 6 H20 5.5 X 10' 2.8 X 10' 2 2 SOW CaCI2. 2 HzO J.J X 10' 0.6 X 10' Microelements

KBr 8.4 X 10-4 4.2 X JO-' 5 KI 2.4x 10-4 2.4 X 10'

H3BOJ 4.9x 10-4 2.5 X 10-4 5 5 NaF 7.l.x 10- 3.6 X )0' 5 5 SrCI2 . 6 H20 6.4 X 10' 3.2x 10· Nutrients 5 NaH2P04 . H20 20 X 10-J 2.0 X )0' 3 NaN03 20 X 10' 2.0 X 10-4 3 Na2Si03 . 9 H20 10 X 10- !.Ox 10_. Trace metals 3 6 FeCI 3 . 6 HzO 1.0 X 10' 0.5 X 10- 6 8 znso•. 7 H20 23 X 10' 2.3 X )0- 6 8 MnCI2 . 4 H20 9) X 10' 9.) X 10' 6 8 CoCI2. 6 HzO 25.5 X 10' 2.6 X )0' 6 8 Cuso•. 5 H20 4.7 X 10' 0.5 X J0' 6 8 (NH.)6 Mo102• . 4 H20 10.4 X 10' I 0.4 X 10' 6 Na2Se03 10.0 X 10' 1.0 X 10-S Chelating agent 6 Na2EDTA 5.0 X 10-J 5.0 X 10- Vitamins 7 1 812 5.5 X 10' g L' 7 1 Biotin 5.0 X 10' g L' 4 1 Thiamine HCI 1.0 X 10' g L'

- 82- Chapter./ Effects ofcombined metals on phytochelatins and glwathione

4.3.2. Stock culture of Phaeodactylum tricornutum

Cells of the diatoms Phaeodactylum tricornutum Bohlin were obtained from the collection of the Marine Biological Association (UK). This species was collected from the English

Channel (off Plymouth Sound, UK) and isolated by E. J. Alien in 1910. Stock cultures of

P. tricornutum were grown in the Aquil medium (salinity of the SOW= 17), at I5°C under continuous light, until the late exponential growth phase and a cell density of about 8.0x I 08

1 cells L' . The culture was manually shaken once a day. Cells were counted daily using an

Olympus microscope and an Improved Neubauer haemocytometer. A growth curve for the stock culture P. tricornutum is shown in Fig. 4.2.

1 E+06l Figure 4.2. C> C> 0 C> Growth curve for stock culture P. 8E+05l tricornutum in Aquil medium, May 2002 (average of triplicate :_, E 6E+05l counts). V> Qi u 4E+05 l

2E+05 -1 C> I I' OE+OO c-<>-<> 0 2 4 6 8 10 Days

- 83- Chapter .f Effects ofcombined metals on phytochelatins and glutathione

4.3.3. Short-term metal exposure experiments

The conditions of the metal exposure experiments and concentrations of the metals employed in this study were based on Rijstenbil & Wijnholds ( 1996) and Morelli &

Scarano (200 I). Ionic metal concentrations were calculated from the total concentrations in the culture medium using the thermodynamic speciation programme MINEQL+ (Schecher

& Mcavoy 1992). The experiments were performed in duplicate. Aliquots of 200 mL of the stock culture (at late exponential growth phase) were transferred to I L polycarbonate bottles, and diluted with the Aquil medium (salinity of SOW = 17) to 800 mL total volume, to allow a continuous cell growth for the short-term metal exposure experiments.

The cultures were spiked with metal solutions, except the control experiment, as follows:

I) control- no addition of metals; 2) 10 11M Cd(N03h, 3) 10 11M Cu(N03h; 4) 10 11M

Zn(N03)2. The incubations were conducted under continuous light at l5°C for 24 h. After this period, the cultures were filtered and prepared for the analysis of phytochelatins and glutathione on the same day of filtration. A scheme of the incubation experiments is presented in Fig. 4.3.

P. tricornutum (stock) I I I I I I I Control- no Cu(ll) 10 (.lM Zn(ll) 10 (.lM Cd(ll) 10 (.lM Cu(ll) I0 (.lM Cd(ll) 10 (.lM metals Zn(ll) 10 (.lM Zn(ll) 10 (.lM

Figure 4.3. Scheme of the short-term metal exposure experiments using P. tricornutum under exposure to single and combined metal. A stock culture was equally divided into I L polycarbonate bottles, diluted with Aquil medium and spiked with metal ions.

- 84- Chapter 4 Effects of combined metals 011 phytochelatins and glutathione

4.3.4. Analysis of phytochelatins and glutathione produced by P. tricornutum

The analysis of thiols produced by the metal-treated diatom cultures was based on

Rijstenbil & Wijnholds (1996). The source of chemicals, abbreviations and analytical details are explained in Chapter 3. Briefly, 500 mL of cultures were filtered using 0.45 J.Lm nitrocellulose membrane filter, under gentle vacuum pressure to avoid cell breakage. The filters were placed in an Eppendorf microtube and the extraction of thiols from the phytoplankton cells was carried out by adding 1.2 mL solution of 0.1 M HCl and 5 mM diethylenetriamine pentaacetic acid (DTPA), at 0°C by ultrasonication. The extraction was followed by centrifugation at high speed (1300 g I 20 min at 4°C). Aliquots of the extracts from the Cd-treatment experiments and control were submitted to two different reduction reactions in order to compare the reagent blanks.

4.3.4.1. Reduction witb ditbiotbreitol (DTT)

Reagents were added to an Eppendorf tube ( 1.5 mL) according to the sequence: 250 J.I.L algal extract (treated with 10 J.LM Cd(II)); 625 J.LL of 200 mM HEPES containing 5 mM

DTPA at pH 9.0; and 25 J.LL 20 mM DTT in 200 mM HEPES/5 mM DTPA. After 1 hour of reduction, 10 J.LL 100 mM mBrB in acetonitrile was added and the derivatisation step was carried out in a dark room at room temperature for 15 min. The reaction was stopped by addition of 100 J.LL 1 M methanesulfonic acid and the samples were stored in the dark at

4°C until HPLC analysis.

- 85- Chapter 4 Effects ofcombined metals on phytochelatins and glutachione

4.3.4.2. Reduction with tributylphosphine (TCEP)

Aliquots of 250 J.lL algal extract (from incubation experiment with 10 J.lM Cd(II)) were treated with 25 J.lL 20 mM TCEP in 0.12 M HCI containing 5 mM DTP A, instead of DTT, for 5 min. Then 160 J.lL 200 mM HEPES/5 mM DTPA at pH 9.0 was added. After another

5 m in, I 0 J.lL I 00 mM mBrB and 465 J.lL HEPES/DTP A were added and the protocol continued as described for DTT.

4.3.4.3. HPLC analysis

The HPLC system consisted of two pumps (Merck-Hitachi Models L-6200 and L-6000), a

Rheodyne injection valve with a 20 J.lL loop and a Model FD-300 fluorescence detector

(Dionex) operating at 380 run (excitation) and 470 run (emission) wavelengths. Separation was carried out using a 150 x 4.6 mm C-18 (Econosphere) HPLC column with 3 J.lm particle size. Solvent A was 0.1% trifluoroacetic acid (TF A, in de-ionised water) and

1 solvent B was acetonitrile (ACN). The flow rate of the solvent was 1.0 mL min· • The gradient, adapted from Rijstenbil & Wijnholds ( 1996) was: 0 to 13 m in, I 0 to 21% B; 13 to

33 min, 21 to 35% B; 33 to 40 min, 35 to 100% B; 40 to 50 min, isocratic 100% B; 50 to

65 min, 100 to 10% B.

- 86- Chapter -1 Effects ofcombined metals on phytochelatins and glutathione

4.4. Results and Discussion

Typical calibration curves for GSH and PC2 are presented in Fig. 4.4. The ratio between

the slopes of the linear regressions is 1. 7, indicating an approximately double fluorescence

response for PC2 (which contains two SH groups) in relation to GSH (which contains one

SH group). Quantification ofphytochelatins PC3, PC4 and PC5 produced in the short-term

Cd exposure experiments was based on the PC2 calibration curve, assuming that the

fluorescence response for the polypeptides was proportional to the number of SH groups

(Ahner et al. 1995; Rijstenbil & Wijnholds 1996).

a) b) 100000 30000 - 0 l 25000 0 .. 80000 .. ! c 0 8. g_ 20000 / ~ 60000 ~ 0 ..u ~ 15000 i c / 40000 •u •!!' 10000 0 / I::s ::s G: G: 20000 y = 1964.6x y = 3364.4x 2 5000 2 o" R = 0.9916 / R = 0.9931 0 0 0 10 20 30 40 50 0 2 4 6 8 10 Glutathione (pmol) PC2(pmol)

Figure 4.4. Calibration curves for: a) glutathione; b) phytochelatin PC2.

4.4.1. Reproducibility of the short-term metal exposure experiments

The reproducibility of the short-term metal exposure experiments was assessed using Cd at total concentrations of I 0 11M, for 5 replicates. Relative standard deviations (RSD)

between 16 to 24% were observed for the chromatographic peak areas of GSH and PC2,

- 87- Chapter 4 Effects ofcombined metals on phytochelatins and glwathione

PC3, PC4 and 30% for PC5 (Table 4.2). This shows that the incubation experiments

resulted in a reproducible thiol production.

Table 4.2. Reproducibility of the 5 replicate incubation experiments showing chromatographic peak areas.

Replicate GSH PC2 PC3 PC4 PC5 I 59206 20786 10022 17302 2736 2 45335 12370 7368 11175 1556 3 40389 19580 8581 13572 1506 4 53067 13395 10964 15120 2602 5 35161 14503 8292 16364 1690 Mean 46632 16127 9045 14707 2018 RSD% 20.7 23.6 15.9 16.5 29.7

4.4.2. Comparison between reducing reagents

Thiols, such as glutathione and phytochelatins, undergo rapid oxidation forming disulphide

bonds (Getz et al. 1999). Therefore, the use of reducing reagents is important to break down such bonds and keep reducing conditions for the thiols during analysis. The reducing reagent DTT was initially used in the experiments; however, the reagent blank of the method was not satisfactory for phytochelatin analysis. An alternative reducing reagent,

TCEP, has been reported to provide efficient recovery of oxidised thiols, faster reactions and less interfering chromatographic peaks than DTT (Rijstenbil & Wijnholds 1996; Getz et al. 1999). The interferences caused by the different reducing reagents (TCEP and DTT) in the analysis ofphytochelatins can be observed in the chromatograms shown in Fig. 4.5.

The control experiment treated with DTT (Fig. 4.5c) showed several chromatographic peaks with similar retention times expected for PC3 and PC4. As outlined in Chapter 3,

DTT is a thiol itself and can react with the derivatising reagent mBrB, forming fluorescent

- 88- Chapter 4 Effects ofcombined metals on phytochelatins and glutathione

products. Therefore, the use of TCEP as a reducing reagent was preferred throughout this

work.

GSH GSH a) c) ~ 0c.." ~ u .,u u" "'~ 0 u:"

I 11 1 11 11 I 1 11 I 11 I I I I I I I I I I I I I I I I I I 11 1 I I I I I I I I I I I I 1 I I I I I I I I I I I ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ N N t? N N t")

GSH GSH b) d)

"' "c..0 e"' .,"'~ u PC2 "' PC2 0e PCJ PC4 u::::1 PCS

I I I I I I I 11 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I ~ ~ ~ ~ ~ ($) ~ ~ ~ ~ ~ ~ N N I") N N t'l

Retention rime {minutes)

Figure 4.5. Chromatograms of extracts of P. tricornutum a) control treated with TCEP; b) 2 exposure to 4.9 J..t.M Cd + and treated with TCEP; c) control treated with OTT; d) exposure to 4.9 2 )..LM Cd + and treated with OTT.

- 89- Chapter 4 Effects of combined metals on phytochelatins and glutathione

4.4.3. Effects of combined metal additions on phytochelatin and glutathione production by Phaeodactylum tricornutum

The concentrations of the ionic metals and metal-EDTA complexes in the culture media were calculated from the total concentration of metals using the thermodynamic equilibrium speciation programme MINEQL+, and are presented in Table 4.3. The chelating agent EDT A forms metal complexes that are not taken up by phytoplankton, thus serving to control metal speciation and bioavailability in the medium (Price et al. 1988).

2 2 The concentrations of Cu + and Cd + used in the experiments causes a 30% decrease in growth rate according to Rijstenbil & Wijnholds ( 1996), however, such a decrease was not observed in the present study, as the cell numbers in the metal-treated and control experiments did not show significant differences after the 24 h incubation (< 2% difference).

Table 4.3. Concentrations of ionic metals and EDT A-complexes calculated using MINEQL + in the short-term metal exposure experiments. Cell numbers were counted before and after the incubation period.

Metal species Metal exposure experiments {M} Control Zn Cu Cu +Zn Cd Cd+Zn znZ+ 1.57E-09 3.40E-06 1.62E-08 7.71 E-06 7.77E-09 5.07E-06

Cu2+ 1.60E-12 1.69E-ll 1.43E-06 1.43E-06 4.96E-09 4.96E-09 Cdl+ 9.64E-20 S.ISE-19 9.50E-19 5.15E-19 3.17E-06 5.32E-06 EDTA 4.70E-06 4.54E-07 6.73E-07 Zn-EDTA 4.48E-06 1.52E-08 2.78E-06 Cu-EDTA 4.64E-06 4.97E-06 4.85E-06 Cd-EDTA 4.23E-06 1.99E-06

cells L. 1 before 2.00E+08 1.92E+08 1.58E+08 1.92E+08 1.25E+08 2.70E+08 incubation

cells L" 1 after 2.00E+08 1.92E+08 1.58E+08 1.92E+08 1.25E+08 2.75E+08 incubation

- 90- Chapter 4 Effects ofc ombined metals on phytochelatins and glutathione

4.4.3.1. Phytochelatins

The concentrations of phytochelatins produced by P. tricornutum in the experiments were

1 18 1 normalised to cell numbers (amol cell- = 10 mol celr ) (Fig. 4.6). The use of chl a to normalise the data is not recommended for incubation experiments performed at high metal concentrations ( ~-tM) , as metal toxicity affects pigment content (Cid et al. 1995 ; Rijstenbil

& Wijnholds 1996). The production of phytochelatins with n ranging from 2 to 5 (PC2 -

1 PC5) were induced in all metal exposure experiments, ranging from 0.1 to 3.6 amol cell- •

Low concentrations of phytochelatins were also found in the control, which could be a result of metal contamination in the incubations or basal levels of intracellular phytochelatins.

4 0 PC2 0 PC3 ~ PC4 GJ PC5 Figure 4.6. Phytochelatins produced by P. 1l3 tricornutum under metal exposure. 0 Bars span the range from the average E for duplicate analyses. ~ ~ 2 :p

B 1 fo~~j - dl_

Control Cu Cu+Zn Zn Cd Cd+Zn

The dimer PC2 was found at concentrations higher than the other polypeptides in cells treated with Cd only and in combination with Zn. Exposure to Cd only resulted in a more pronounced phytochelatin production compared to exposure to Cu only. Exposure to Zn did not induce phytochelatin production in P. tricornutum, being similar to the control. The

- 91 - Chapter 4 Effects ofcombined metals on phytoclzelatins and gl11tathione combination of Zn and Cu resulted in concentrations of phytochelatins similar to the induction found by the addition of Cu only. This could mean that no competitive interaction between Cu and Zn occurred at the established metal concentrations.

Cultures treated with Cd only produced 50% more PC2 than the combination of Zn and

Cd. Such an antagonistic effect of Zn on the induction of PCs by Cd is probably due to competition between these two metals for cellular binding sites. Cadmium was the most effective inducer of PC2, which is in agreement with previous studies including other phytoplankton species (Ahner et al. 1995; Ahner & More! 1995; Lee et al. 1996). A decrease in effects of Cd on PC production in some diatoms has been observed at high

2 2 levels of Zn + and/or Mn + (Wei et al. 2003). A plausible explanation for this is that Cd uptake takes place through the same channels as Zn and Mn, thus Zn and Mn competitively

2 inhibits the induction of PC by Cd (Sunda & Huntsman 1996). Although Zn + is able to activate PC synthase, it is a weak inducer of PC production in studied algae, plants and

fungi (Morelli & Scarano 2001 and references therein). For example, in a 3 day exposure experiment, the growth rate of P. tricornutum was not affected by intracellular accumulation of Zn (Morelli & Scarano 2001 ). Deleterious effects of Zn on other physiological parameters of P. tricornutum have not been reported.

The synthesis of PCs in P. tricornutum cells have been reported to occur after 15 m in of exposures to either 10 f.!M Cd or 10 f.! M Pb (total concentration, at salinity 35) (Morelli and Scarano 2001 ). In these experiments, PCs reached concentrations comparable to the present study in 7 h (approximately 6 amol cell· 1 for each of the oligomers n = 2, 3 and 4, calculated from graph), with concomitant decrease in glutathione concentrations (from 90

1 to 40 amol cell· , obtained from graph).

-92- Chapter 4 Effects of combined metals on phytochelatins and glutathione

Such rapid formation of PCs in P. tricornutum cells, together with their capability to release GSH-Iike substances and surfactants (Croot et al. 2000) could account for the high metal tolerance of this diatom species. In addition, a recent study reported that P. tricornutum cells can respond rapidly to metal-induced oxidative stress by activation of a suite of antioxidant enzymes (Morelli & Scarano 2004). In contrast, metal sensitive phytoplankton species such as Skeletonema costatum have growth inhibited at I 00 times lower Cd concentrations (Torres et al. 1997). This species has also shown less pronounced

PC production than P. tricornutum (Rijstenbil & Wijnholds 1996), but is able to release

Cu-chelators with weak binding strengths under Cu stress (Croot et al. 2000), which could minimise metal toxicity in its surrounding medium.

The results presented here corroborates with those reported by Wei et al. (2003). They observed a range of effects on PC production, such as antagonism, synergism and suppression, caused by combinations of Zn, Cu and Cd (at concentrations between 10" 10 and 10-7 M) on a natural phytoplankton assemblage and laboratory phytoplankton monocu !tu res.

4.4.3.2. Glutathione

The concentrations of GSH produced by P. tricornutum in the incubation experiments ranged from 4. 7 to 25.3 amol cell- 1 (Fig. 4. 7) and were higher than the total concentrations of phytochelatins (PCTma1). GSH was produced in the control experiment, to which no enrichment with metals was made, and also occurred at similar concentrations in all experiments, except for incubations with added Cd only and with Cd in combination with

Zn. These metal treatments caused an 85% increase and 65% depletion in GSH concentration, respectively, relative to the control.

- 93- Chapter -1 Effects ofco mbined metals on phytochelatins and glutathione

c:::::J GSH -PC total --<>--- PC totai:GSH

35 3.0 Figure 4.7. Glutathione and total phytochelatins produced by P. 30 ------tricornutum under metal

~:::: 25 exposure, and PC10131: GSH ratios. 2.0 -o ~ () Mean values of duplicate 0 20 0 E ![ analyses. .e 15 (;) CJ) en 0 1.0 I £ 10

5

0.0 Control Cu Cu+Zn Zn Cd Cd+Zn

For various phytoplankton spectes, the intracellular GSH was found at similar concentrations upon Cd or Cu exposure at ambient/low levels (Ahner et al. 2002). It has been considered that healthy phytoplankton cells maintain intracellular GSH at a constant level for essential functions, in addition to its use for PC production (Tang et al. 2000;

Ahner et al. 2002). However, there appear to be two situations regarding metal stress that cause variations in GSH concentrations in phytoplankton cells. For example an increase in intracellular GSH concentration can be a result of the cellular acclimation response for an initial metal stress (Okamoto et al. 2001 ). This could explain the higher GSH concentration observed here for the exposure to Cd only in relation to the control. The second situation is related to acute metal stress, when the synthesis of PCs by P. tricornutum could imply a biochemical cost with depletion in GSH concentration (Rijstenbil & Wijnholds 1996;

Morelli & Scarano 2004). This acute stress situation could explain the lower GSH concentration observed for the exposure to Cd combined with Zn in comparison to the control. However, the lower GSH concentration for the exposure to Cd combined with Zn contradicts the effects of the exposure to Cd only. It was expected that a competition

-94- Chapter 4 Effects ofcombined metals on phytochelatins and glutathione between the two metals would minimise Cd toxicity, leading to less effects on intracellular

GSH concentration.

4.4.3.3. Ratios PCtotai:GSH

The ratios of the concentrations of PC10ta 1 to GSH, in which PC1otal corresponds to the sum of the y-Giu-Cys units of the PC oligomers, have been suggested to provide an indication of the extent of metal stress (Tang et al. 2000). In the present experiments, the ratios ranged from 0.11 to 1.95 (Fig. 4.8). The lowest PC1ata1:GSH ratio was observed for the control experiment and the highest ratio for the combination of Cd with Zn. According to these ratios, exposure to Cu caused stronger stress than exposure to the combination of Cu with Zn or single Cd. Despite the competition between Cd and Zn for binding sites, which

should arneriolate Cd toxicity, a higher PCtata 1:GSH value was observed for the exposure to the combination of Cd with Zn than to the single exposure to Cd. Wei et al. (2003) hypothesised that at certain Cd and Zn concentration thresholds, Cd could inhibit Zn uptake and the cell, experiencing limitation in Zn, would promote more intracellular Zn transporters. These transporters would allow for greater Cd accumulation and thus lead to an enhanced stress condition. As intracellular metal concentrations were not determined in that study nor in the present one, direct evidence is required to support this hypothesis.

A companson of GSH and PC concentrations, together with PC1ata1:GSH ratios for P. tricornutum under different metal exposure conditions is presented in Table 4.4. Cultures grown in a medium similar to the one employed here and exposed to separate additions of

0.4 J..!M Cu and 9 J..LM Cd (Rijstenbil & Wijnholds 1996) resulted in PC 10 ~a 1 :GSH values comparable to the present study. P. tricornutum cultures grown at higher salinity and

- 95- Chap/er 4 Ejfec/s ofcombined melals on phylochelalins and glwailrione

subjected to a shorter incubation period (7 h) with individual 10 J..LM Cd and 10 J..LM Pb

produced the highest PC10ra1:GSH values (Morelli & Scarano 2001).

Table 4.4. Production ofPCrotal (sum ofy-Glu-Cys units of the oligomers) and GSH, together with thiol ratios for laborato!l:: P. tricornutum cultures under metal stress. EMetal = -Log[Metaf+]. Metal GSH PC total PC,.,.t:GSH Comments Reference concentration PCs and GSH This study Control 13.7 1.81 0.13 concentrations are given as y-Glu-Cys pCu = 5.8 10.9 11.0 1.01 amol celr1 pCu = 5.8 12.8 9.09 0.71 pZn = 5.0 salinity = I 7 24 h incubation pZn = 5.5 16.0 1.73 0.11 pCd = 5.5 25.3 18.0 0.71 pCd = 5.3 4.68 9.12 1.95 pZn = 5.3

PCs and GSH Ahner et al. Control 780 14.6 0.019 concentrations are 2002 given as j.tM pCd = 11 560 2.90 0.052 1 (~-tmol cell volume. ) pCd = 10 680 53.4 0.079 24 h incubation pCd= 9 710 51.6 0.073 pCu = 11 260 81.1 0.31 pCu = 10 860 116 0.14 pCu= 9 2100 247 0.12 PCs and GSH Morelli and Control concentrations are Scarano, 200 I given as y-Glu-Cys [Cd] = 10 ~-tM 40 65 1.6 amol celr1 (obtained from graphs) [Pb] = 10 j.tM 45 80 1.8 salinity= 35 [Zn] = 10 ~-tM 7 h incubation

PCs and GSH Rijstenbil and Control 3385 79 0.023 concentrations are in Wijnholds, SH ~-tM (biovolume) 1996 [Cu] = 0.4 ~-tM 2359 379 0.16 salinity = 17 24 h incubation [Cd] = 9.0 ~-tM 2173 1424 0.66

- 96- Chapter 4 Effects of combined metals 011 phytoche/ati11s a11d glutathio11e

Incubation exposure experiments usmg low ambient metal concentrations (pM to nM range) produced PC:GSH ratios around one order of magnitude lower than those for metals in the J-lM range (Ahner et al. 2002). In these low metal concentration exposure

experiments, Cu-treated cells produced higher PC1otaJ:GSH ratios than Cd-treated cells.

Such lower ratios are consistent with the role attributed to PCs in metal detoxification mechanisms, with higher metal concentrations inducing a more pronounced PC production.

It is worth noting that the P. tricornutum employed in the present study has been cultured for more than 90 years in the collection facilities of the Marine Biological Association.

Thus, in addition to the natural high metal tolerance of P. tricornutum (Torres et al. 1997), the biological responses to metal toxicity could be altered through evolution of even more tolerant strains over these years. The different strains of P. tricornutum could also explain the variabilities in the production of GSH and PCs by the different workers.

4.5. Conclusions and final remarks

I. Phaeodactylum tricornutum cultures under exposure to Cu, Cd and Zn, singly and

in pairs, synthesised a range of short-chained phytochelatins with n = 2-5.

Simultaneous determination of GSH and PCs can be more useful to assess the

physiological status of phytoplankton under metal stress. However, the proposed

PCtotal:GSH ratios did not explain the level of metal stress within the experiments.

Exposure to Cd only, for example, caused less stress than the Cd combined with

Zn.

- 97- Chapter 4 Effects ofcombined metals 011 phytochelatins and glutathione

2. An antagonistic effect of Zn on PC induction by Cd was observed, which was

likely due to competition between these two metals for cellular binding sites. Zinc

did not induce PC production nor affect GSH concentrations. Copper combined

with Zn induced similar PC concentration to the Cu exposure on its own. Although

P. tricornutum is not considered a representative phytoplankton species for

environmental studies, the results indicated that interaction among metals should

influence PC production by natural phytoplankton assemblages, and probably cause

variations in PC concentrations in the field.

4.6. References

AHNER, B.A., KONG,S. & MOREL, F.M.M. (1995) Phytochelatin Production in Marine­ Algae .1. An lnterspecies Comparison. Limnology and Oceanography, 40, pp. 649-657.

AHNER, B.A. & MOREL, F.M.M. (1995) Phytochelatin Production in Marine-Algae .2. Induction by Various Metals. Limnology and Oceanography, 40, pp. 658-665.

AHNER, B.A., WEI, L.P., OLESON, J.R. & OGURA, N. (2002) Glutathione and other low molecular weight thiols in marine phytoplankton under metal stress. Marine Ecology­ Progress Series, 232, pp. 93-103.

CAMPBELL, P.G.C. (1995) Metal speciation and bioavailability in aquatic systems. pp. 45-101. Wiley,NewYork.

CID, A., HERRERO,C., TORRES, E. & ABALDE, J. ( 1995) Copper Toxicity on the Marine Microalga Phaeodactylum- Tricomutum - Effects on Photosynthesis and Related Parameters. Aquatic Toxicology, 31, pp. 165-174.

CROOT, P.L., MOFFETT, J.W. & BRAND, L.E. (2000) Production of extracellular Cu complexing ligands by eucaryotic phytop1ankton in response to Cu stress. Limnology and Oceanography, 45, pp. 619-627.

GETZ, E.B., XIAO, M., CHAKRABARTY, T., COOKE, R. & SELVIN, P.R. (1999) A comparison between the sulfhydryl reductants tris(2- carboxyethyl)phosphine and dithiothreitol for use in protein biochemistry. Analytical Biochemistry, 273, pp. 73-80.

KNAUER, K., AHNER, B., XUE, H.B. & SIGG, L. (1998) Metal and phytochelatin content in phytoplankton from freshwater lakes with different metal concentrations. Environmental Toxicology and Chemistry, 17, pp. 2444-2452.

- 98- Chapter 4 Effects ofcombined metals on phytochelatins and glr~rathione

LEE, J.G., AHNER, B.A. & MOREL, F.M.M. (1996) Export of cadmium and phytochelatin by the marine diatom Thalassiosira weissflogii. Environmental Science & Technology, 30, pp. 1814-1821.

MOREL, F.M.M., RUETER, J.G., ANDERSON, D.M. & GUILLARD, R.R.L. (1979) Aquil: a chemically defined phytoplankton culture medium for trace metal studies. Journal ofPhycology, 15, pp. 135-145.

MORELLl, E. & PRA TESI, E. ( 1997) Production of phytochelatins in the marine diatom Phaeodactylum tricomutum in response to copper and cadmium exposure. Bulletin of Environmental Contamination and Toxicology, 59, pp. 657-664.

MORELLl, E. & SCARANO, G. (1995) Cadmium-Induced Phytochelatins in Marine Alga Phaeodactylum- Tricomutum - Effect of Metal Speciation. Chemical Speciation and Bioavailability, 7, pp. 43-47.

MORELLI, E. & SCARANO, G. (2001) Synthesis and stability ofphytochelatins induced by cadmium and lead in the marine diatom Phaeodactylum tricomutum. Marine Environmental Research, 52, pp. 383-395.

MORELLI, E. & SCARANO, G. (2004) Copper-induced changes of non-protein thiols and antioxidant enzymes in the marine microalga Phaeodactylum tricornutum. Plant Science, 167, pp. 289-296.

OKAMOTO, O.K., PlNTO, E., LATORRE, L.R., BECHARA, E.J.H. & COLEPICOLO, P. (200 I) Antioxidant modulation in response to metal-induced oxidative stress in algal chloroplasts. Archives of Environmental Contamination and Toxicology, 40, pp. 18-24.

PRICE, N.M., HARRISON, G.I., HERING, J.G., HUDSON, R.J., NIREL, P.M.V., PALENKI,B. & MOREL,F.M.M. (1988) Preparation and chemistry of the artificial algal culture medium Aquil. Biological Oceanography, 6, pp. 443-461.

RIJSTENBIL, J.W. & WIJNHOLDS, J.A. (1996) HPLC analysis of nonprotein thiols in planktonic diatoms: Pool size, redox state and response to copper and cadmium exposure. Marine Biology, 127, pp. 45-54.

SCARANO, G. & MORELLI, E. (2002) Characterization of cadmium- and lead­ phytochelatin complexes formed in a marine microalga in response to metal exposure. Biometals, 15, pp. 145-151.

SCHECHER, W.D. & MCA VOY, D.C. ( 1992) Mineql+ - A Software Environment for Chemical-Equilibrium Modeling. Computers Environment and Urban Systems, 16, pp. 65- 76.

SUNDA, W.G. & HUNTSMAN, S.A. (1996) Antagonisms between cadmium and zinc toxicity and manganese limitation in a coastal diatom. Limnology and Oceanography, 41, pp. 3 73-387.

TANG, D.G., HUNG, C.C., WARNKEN, K.W. & SANTSCHI, P.H. (2000) The distribution of biogenic thiols in surface waters of Galveston Bay. Limnology and Oceanography, 45, pp. 1289-1297.

- 99- Chapter 4 Effects ofcombined metals 011 phytochelati11s and glutathione

TORRES, E., CID, A., FIDALGO, P., HERRERO, C. & ABALDE, J. (1997) Long-chain class Ill metallothioneins as a mechanism of cadmium tolerance in the marine diatom Phaeodactylum tricomutum Bohlin. Aquatic Toxicology, 39, pp. 231-246.

TORRES, E., CID, A., HERRERO, C. & ABALDE, 1. (2000) Effect of cadmium on growth, A TP content, carbon fixation and ultrastructure in the marine diatom Phaeodactylum tricomutum Bohlin. Water Air and Soil Pollution, 117, pp. 1-14.

WEI, L.P., DONAT, J.R., FONES, G. & AHNER, B.A. (2003) Interactions between Cd, and Cu, and Zn influence particulate phytochelatin concentrations in marine phytoplankton: Laboratory results and preliminary field data. Environmental Science & Technology, 37, pp. 3609-3618.

- 100- Chapter 5

5. Particulate phytochelatins and glutathione in the Fal Estuary and Plymouth Sound (UK)

5.1. Abstract

The aim of this chapter is to report field measurements of particulate phytochelatins (dim er

PC2) and glutathione (GSH), together with copper complexation in two metal contaminated coastal systems in the UK: (i) the Fal Estuary and (ii) the Tamar and Plym estuaries (Plymouth Sound). Glutathione was found throughout the estuaries, but showed a

1 wide of concentration range from 0.6 to 274 ~mol (g chl a)" • Phytochelatins were

1 observed mainly in the Fal Estuary at concentrations up to 36 ~mol (g chl a)" , and were below the detection limit in most of the Plymouth Sound. Highest particulate PC2 and

GSH concentrations were observed in the mine impacted sites Restronguet Creek (Fal

Estuary) and Gunnislake (Tamar Estuary). In Restronguet Creek concentrations of total

2 dissolved Cu exceeded those of Cu-complexing ligands. Free Cu + and total dissolved Cu concentrations were positively correlated with PC2 and GSH concentrations in the Fal

Estuary during 2002. High variabilities in GSH and PC2 concentrations (5-200 fold) could be due to the heterogeneous phytoplankton composition within the estuaries and antagonistic and suppressing effects on PC2 and GSH production caused by metal interactions. The production of PC2 showed to be faster than that of GSH in the Fal

Estuary (2002) under increasing total dissolved Cu concentrations, suggesting that PC2 is a more specific response to metal stress than GSH in natural waters.

- 101 - Chapter 5 Field measurements ofparticulate plrytoclrelatins and glutatlrione

5.2. Introduction

Metal bioavailability and toxicity in aquatic systems are related to the kinetically labile species, including free ionic metals, inorganic hydroxide, chloride, and carbonate metal complexes, and also metals complexed by organic ligands with weak binding strengths.

The degree of toxicity of a metal species depends on the metal itself and the organism, but it is usually higher for the free ionic metal species (Gledhill et al. 1997). 1n this respect, dissolved organic ligands ubiquitously found in natural waters play a key role in modifying the metal bioavailability and toxicity. These ligands are thought to be derived from natural sources (algae exudates and humic substances) and anthropogenic inputs (e.g. sewage) and

12 have been classified according to their conditional stability constants as strong (K1 = I 0 -

15 7 11 10 ) and weak (K2 = 10 -10 ) ligands (Moffett et al. 1997). Strong organic ligands complex metals forming non-bioavailable species, and therefore minimise the potential for the toxic effects of metals.

In aquatic systems subject to acid mme run-off, agricultural, industrial and urban discharges, and leaching of metals from antifouling paints, the total metal concentrations can approach or exceed those of the organic ligands, resulting in high aqueous inorganic metal concentrations. It has been suggested that even low ambient concentrations of inorganic metal species in the aquatic environment could have adverse effects on the distribution and composition of phytoplankton communities (Sunda 1988; Rijstenbil et al.

1991 ). The effects are difficult to assess as phytoplankton composition and spatial distribution are functions of various environmental factors, including salinity, turbidity, nutrients, turbulence, and depth (Fogg & Thake 1987a). Furthermore, these variables can be perturbed by natural forcing or human activities. The production of intracellular metal-

- 102- Chapter 5 Field measurements ofparticu/ate phytochelatins and glmathione

binding peptides, more specifically phytochelatins, is one of the mechanisms used to reduce the toxic effects of metals by many phytoplankton species, and is possibly significant in metal affected environments. Phytochelatins have been suggested to provide a sub-lethal measure of metal stress to phytoplankton in natural waters (Ahner et al. 1997).

Few field studies, however, have been undertaken to support this hypothesis.

The aim of this chapter is to describe and interpret field measurements of particulate metal­ binding peptides in two metal impacted systems in the UK: (i) the Fal Estuary and (ii) the

Tamar and Plym estuaries (Plymouth Sound). Only a handful of studies have focused on the adaptive mechanisms against metal toxicity by organisms living in these metal impacted areas (Langston et al. 2003b). Baseline levels of particulate metal-binding peptides have not yet been determined in these systems. Here, the relationships between metal-binding peptides with dissolved (< 0.45 Jlm) copper speciation are considered.

Copper is essential for photosynthetic organisms playing a vital role in electron transport in photosynthesis, but it is also one of the most toxic metals to marine algae (Giedhill et al.

1997; Pinto et al. 2003). Copper is found at elevated concentrations in the Fal and

Plymouth estuarine systems.

5.3. Study areas - environmental setting

The Fal Estuary and Plymouth Sound and its estuaries, are located in the SW England

(UK) and are candidates for special areas of conservation. Langston et al. (2003a; b) have recently reviewed the water and sediment quality of these areas, in order to provide a baseline for assessing the present and future effects of anthropogenic activities on these

- 103- Chapter 5 Field measurements ofparticulate phytochelatins and glmathione environments. Most of the following discussions on metal distribution in these areas are based on these reviews.

5.3.1. Fal Estuary

The Fal Estuary lies in a highly mineralised area and is considered one of the most metal impacted estuaries in the UK (Fig. 5.1 ). Mining activities for deposits of tin, copper, lead, iron, arsenic, tungsten, uranium, and silver reached their apogee in the 19th century, together with ore processing in the Camon Valley (Bryan & Gibbs 1983; Pirrie et al.

1997). Most studies in this area have concerned trace metal association with sediments, and the effects of long-term metal exposure on benthic organisms (Bryan & Gibbs 1983;

Warwick 2001). Restronguet Creek is the most metal affected branch of the Fal Estuary as it is directly fed by the Camon River, which receives drainage from inactive metalliferous mines with high concentrations of Zn, Mn, Cu and Cd (Bryan & Gibbs 1983; Rijstenbil et al. 1991 ). Metal contaminated sediments and mine tailings (sources of metals to the water column) are transported from Restronguet Creek to Mylor, Pill (adjacent to Restronguet) and St Just creeks, where they accumulate (Langston et al 2003b).

- 104- Chapter 5 Field measurements ofparliculate phytochelatins and glutathione

Carrick Roads (middle of estuary) N

Penryn River I

Falmouth Dockyard

3 1 km~ I

Figure 5.1. Fal Estuary (UK). Sampling sites: I. Truro/Fal; 2. St Just· 3. St Mawes (Percuil system); 4. Falmouth Yacht Marina; 5. Between Mylor and Falmouth; 6. Mylor Creek; 7. Between Mylor and Restronguet; 8. Restronguet Creek; 9. Camon River. C l, C2, C3, C4, CS sampling sites in the Camon River.

- 105- Chapter 5 Field measurements ofparticulat e phytochelatins and glutathione

5.3.2. Plymouth Sound-Tamar and Plym estuaries

The Plymouth Sound and its estuaries (Fig. 5.2), also located in the SW England, are subject to several anthropogenic influences such as shipping activities, including naval, commercial and pleasure craft (Langston et al. 2003a).

Tamar

PlymJ

Plym2

0 2 km

English Channel

Figure 5.2. Plymouth Sound. Sampling sites in the Tamar and Plym estuaries. 1. Gunnislake; 2. Morwellham; 3. Calstock; 4. Cotehele Quay; 5. Pink house; 6. Pentille; 7. Weir Quay; 8. Lynher; 9. South of Torpoint; I 0. Sound near Devil 's Point; 11. Drake Island; 12. A. Halton Quay; B. Cargreen; C. Saltash; D. Tinside. Sampling sites in the Plym Estuary (200 1-2002): Plym I, Plym2, Plym3, Plym4

The Tamar and Plym Valleys have around nine centuries of mining activities for deposits of tin, copper, lead, silver, iron, arsenic, zinc, tungsten, and manganese. The catchment area of the upper Tamar Estuary today is influenced by agriculture and old mines, whilst the lower estuary is subjected to substantial urban and industrial development. As a

- 106- Chapter 5 Field measurements ofparticulate phytochelatins and glutathione consequence, the waters and sediments of the Tamar Estuary are enriched with metals and

nutrients through point and diffuse sources. The quality of the waters of the Plym Estuary

is subjected to the influence of the two large consented discharges by volume (excluding

sewage treatment works), which could have effects on other parts of the system (Langston et al. 2003a).

5.3.3. Other contaminants and environmental quality of the Fal Estuary

and Plymouth Sound

Other contaminants and physical changes have also been introduced in the Fa! Estuary and

Plymouth Sound mainly through sewage and industrial discharges and agricultural run-off

from the catchment area. Table 5.1 summarises the major contaminants, physical factors, the most affected areas and consequences to the water quality of the Fa! Estuary and

Plymouth Sound in relation to the Environmental Quality Standard (EQS) values. The EQS was based on laboratory ecotoxicological studies using a range of organisms. Values above the EQS indicate that deleterious effects on biota!feature might occur, although there is a paucity of data on sub-lethal effects to guarantee a safe threshold for these values.

Furthermore, it is possible that synergistic and/or antagonistic effects of a mixture of the various contaminants present in these estuaries occur, altering the expected biological responses for one stressor.

Specific responses to metal impact have been reported for a limited number of organisms

in the Fa! and Tamar estuaries. These include adaptive mechanisms such as production of the metal-binding proteins metallothioneins in gills of mussels and crabs (Pedersen &

Lundebye 1996; Langston et al. 2003b ), changes in permeability in macroalga to reduce

- I 07- Chapter 5 Field measurements ofparticulate phytochelatins and glutathione metal uptake (Langston et al. 2003b), and cyto- and genotoxic biomarkers m molluscs

(Cheung & Jha 2004).

Table 5.1. Summary of contaminants, physical factors, sources, affected areas, and most vunerable featureslbiota in the Fa! Estuary and Plymouth Sound and estuaries (adapted from Langston et al. 2003a; b). EQS =Environmental Quality Standard. Contaminant/ Source Affected area Most vulnerable Water quality species/feature Physical factor Organotin Falmouth Docks, Widespread in both systems, Molluscs Chronic inputs shipping especially Fa! mouth Docks not sediments, systematically sewage investigated discharges Metals (As, Cu, Historic mining Restronguet Tamar and Invertebrates, Exceed EQS Cd, Fe, Zn) activities, and Mylor Tavy birds and fish, values in sediments, Creeks, Estuaries species Restronguet, discharges upper Fa! composition Mylor, Camon and in Also industry in sediments from the Plymouth Tamar and Sound Tavy estuaries Nutrients Sewage Upper Fa! Upper Invertebrates, No systematic discharges, run- (Truro area), Tamar, fish, seabirds, monitoring to off from land, He! ford Yealm, mammals provide mine drainage Lynher comparable data with EQS values Turbidity Sewage Upper Fa! Tamar Benthic discharges, (Truro, communities, aggregate Tresillian fish extraction area) Microbiological Sewage Upper Fa! Lynher, Bivalves, birds, Exceed EQS parameters discharges, farm (Tresillian) Tamar, mar me values animals (run-off Penryn Plym organisms from land) River (flushing) Hydrocarbon Discharges, Lower Fa! Tamar and Bivalves, fish, EQS values not oils, polycyclic shipping, urban and Helford Plymouth seagrasses and defined, but aromatic run-off, sound shoreline tainting in hydrocarbons atmospheric sediments communities, shellfish and deposition species fish have been composition observed Pesticides, Agricultural run- Poorly Plym, Invertebrates, More herbicides and off and sewage defined, silty Tamar, food chain investigations other synthetic discharges sediments Plymouth bioaccumulation on organics are probably Sound accumulation main and distribution reservoir are needed

- 108- Chapter 5 Field measurements ofparticulate phytochelatins and glutathione

5.4. Experimental

5.4.1. Sample collection

Surface water samples were collected from different branches of the Fa! Estuary and some points of the Camon River (Fig. 5.1) during surveys on 081h August 2002, 30th April 2003 and I 51 July 2003, using I or 2 L acid washed polycarbonate bottles. Axial transects along

Plym and Tamar estuaries (Fig. 5.2) were undertaken on 24th September 2002 and 25th

April 2003, respectively. All samples were collected from a depth of around 0.3 m with the use of small boats. The samples were placed in a cool box until returned to laboratory

(within 8 h). Samples for trace metal and chlorophyll a analyses were collected separately using 250 mL acid cleaned high density polyethylene bottles. Sub-samples were collected for the determination of salinity. In-situ measurements of pH, temperature and conductivity were also taken using portable instruments (Hanna instruments).

5.4.2. Particulate phytochelatins and glutathione

The analysis of the particulate metal-binding peptides phytochelatins (PCs) and glutathione

(GSH) was based on Rijstenbil & Wijnholds (1996) with a few modifications, and is fully described in Chapters 3 and 4. The water samples collected from selected sites in Plymouth

Sound and Plym Estuary were filtered using 0.45 ~m pore size nitrocellulose filters, instead of 0.8 ~m, as part of the initial adaptations of the analytical method for field application.

- 109- Chapter 5 Field measurements ofparticula/e phytochelatins and glutathione

5.4.3. Total dissolved copper

Determination of total dissolved Cu (CuTotai) in water samples was based on adsorptive cathodic stripping voltammetry (AdCSV), usmg salicylaldoxime (SA, 2- hydroxybenzaldehyde oxime) as added ligand (Campos & van den Berg 1994). Samples were filtered using acid washed polycarbonate membrane filters (0.4 flm, Whatman), then acidified to pH- 2 using quartz distilled HCl and irradiated with ultraviolet light at 70°C in quartz tubes for 6 h using a 400 W mercury vapour lamp. The pH of the sample was neutralized with ammonia and subsequently a 10 mL aliquot was pipetted into the voltammetric cell. The buffer N-2-hydroxyethylpiperazine-N' -2-ethansulphonic acid

(HEPES, pH = 8.2 adjusted with NH3) and SA solutions were added to the cell to achieve

I 0 mM and 25 flM (final concentrations), respectively.

The measurements were carried out using a potentiostat (Autolab Ecochemie) interfaced with an automated hanging mercury drop electrode (Model V A 663, Metrohrn), with potentials measured against an Ag/AgCl/K.Cl reference electrode. The sample solution was purged for 4 min, and the CuSA complex was deposited on a fresh Hg drop at -1.1 V for

30 s. The potential was scanned from -0.1 to -0.6 V with Cu reduction peak occurring at -

0.35 V (square-wave voltammetry, 50Hz, pulse amplitude 0.025 V, scan rate 25 mV s·\

The concentration was measured by addition of Cu from 1o- 5 M stock standards. Duplicate

Cu measurements produced less than 2% difference, and analysis of certified reference material (SLEW2) were in good agreement with certified values for Cu. Sample handling was carried out in a Class 100 laminar flow hood.

- llO- Chapter 5 Field measurements ofparticulate phytochelatins and glwathione

5.4.4. Natural organic ligands and copper species

2 Determination of natural organic ligands (L), free Cu + and inorganic Cu (Cu ') concentrations in filtered samples was carried out using competitive ligand titrations with

SA as described in Campos and van den Berg (1994). Eleven sample aliquots of 10 mL each were pi petted into 30 mL polystyrene cups together with 2 J..lM SA and 10 mM

HEPES (pH = 8.5). Copper solution was added in ten of the cups in increasing increments

(100-800 nM Cu for samples from Restronguet Creek and Camon River, and 10-200 nM

Cu for the other sample locations), and then allowed to equilibrate for at least 15 h at room temperature. The voltammetric analysis was performed using the same conditions as for the determination of CuTotal· Triplicate titrations resulted in standard deviations of 0.7 and

0.2 for the ligand concentrations and Log KcuL, respectively.

5.4.5. Chlorophyll a

Chlorophyll a analysis was performed usmg a well-established fluorimetric method

(Parsons et al. 1984). Water samples were filtered onto 0.8 J..lm nitrocellulose filters, under gentle vacuum. The filters with retained particles were wrapped in aluminium foil and kept frozen until extraction. The extraction of chl a from the particles was carried out using

90% acetone (HPLC grade) in MilliQ water and ultrasonication in an ice bath for 15 min, followed by overnight storage (4°C, protected from light). Fluorescence of the samples was measured using a spectrofluorimeter (Merck-Hitachi) at 680 nm (emission) and 436 nm

(excitation) wavelengths. A linear response was observed for eh! a standard (Sigrna)

1 solutions (made up in 90% acetone) between 2 to 500 J..lg L' • Pheophytine concentrations were determined after acidification of the samples by addition of 0.1 M HCI to the sample cuvette ( 4 mL) and discounted from the eh! a measurements.

- Ill - Chapter 5 Field measurements ofparticulate phytoclrelatins and gllllatlrione

5.5. Results & Discussion

5.5.1. Fal Estuary

5.5.1.1. Pbytoplankton composition of tbe Fat Estuary

Chlorophyll a concentrations and salinities for the Fa! Estuary samples are presented in

Fig. 5.3. Chi a concentrations ranged from 3.1 to 69.1 J..lg L- 1 and indicated the occurrence

1 of phytoplankton blooms for a number of sample locations (> 10 J..lg L- ). Salinities were relatively similar and higher than 23 for most of the sampling sites. The only exceptions were some of the Camon River and Restronguet Creek samples which were under a greater riverine influence and showed the lowest eh! a concentrations.

The phytoplankton species composition for August 2002 is summarised in Table 5.2

(Environment Agency, pers. comm.). Not all the sampling sites chosen by the Environment

Agency coincided with those from the present study, and data for phytoplankton composition for 2003 was not available. Apart from this data set, recent studies concerning phytoplankton composition/distribution in the Fa! Estuary are scarce. A study conducted in the Fa! Estuary in 1989 showed that the phytoplankton species composition in the metal contaminated Restronguet Cre~k deviated from that in other branches (Rijstenbil et al.

1991 ). Rijstenbil et al. (1991) observed that in the riverine parts of Restronguet Creek, the dinotlagellate E. mutabilis occurred at J..lM Cu and Zn, whereas in clean waters presenting nM Cu and Zn, Chlamydomonas sp. and Oocystis sp. were the common species.

- 112 - Chapter 5 Field measurements ofparticulate phytoche/atins and glwathione

a)

---Chi a --Salinity 40 20 16 () :?;- 30 2: ·c: 12 Q) (ij 20 'E' <0 (/) 8 10 4 ' .:.. 0 0 (ij (;) m Qi ...... t- Q) £ 0 • b Q) u_ ~ ~ ~ 0 0 m ..., 3: 0 >. Cl >. Q) ~ C1l ~ c >. Cl e E ~ ~ 0:: c 2 U5 (ij _g ~ I- ~ u_ m 0 g Q) ~ m - ~ C1l >. Q) 0:: ~ Q) 0:: a..

b)

40 12 10 30 () 8 2: ~ Q) :§ 20 I C1l 6 'E' <0 (/) 4 10 ~' · 2 0 0 (ij (;) m N (") Q) £ 0 Qi 1,!::: ~ ~ ~ c c c ..., 3: 0 >. Cl 0 0 0 e U5 C1l E ~ c E E E 2 ~ (ij _g C1l C1l C1l I- u_ m 0 0 0 - ~ Q) 0:: Q) a..

c)

40 80

30 60 ~ -~ c Q) (ij 20 40 'E' (/) <0 r 10 20 ~.:..

0 0 m Li) (;) Q) £ 0 Qi ~ ~ ~ "'"c c ..., 3: 0 >. Cl 0 0 U5 C1l E ~ c E E ~ (ij _g C1l C1l u_ m 0 0 ~ Q) ~ 0:: a..Q)

Figure 5.3. Salinity and chlorophyll a for the Fa! Estuary surveys. a) 8111 August 2002 and 7 111 October 2002 (latter samples marked with *); b) 30th April 2003; c) 151 July 2003.

- 113- Table 5.2. Common dominant phytoplankton species reported in the Fal Estuary (Environment Agency data, August 2002). "'Autumn 1989 data from 1 1 Rijstenbil et al. 1990; +indicates 6-50 cells mL" ; ++indicates> 50 cells mL· • Species River St Just St Mawes Falmouth Mylor Restronguet Carnon Carrick Falffruro (Percuil) Yacht Marina Creek Creek River Roads Diatoms Dactylioso/en fragilissima + + + + Pseudonitzschia spp. + + + + Rhizosolenia setigera + + + + Minidiscus sp. ++ ++ ++ Skeletonema sp. + + + ++ Thalassiosira sp. ++ + + + Dinoflagellates

_. Kerenia mikimotoi ++ + + ++ ++ + Prorocentrum micans + + + ++ + + + Prorocentrum minimum + + + + Prorocentrum triestinum ++ ++ ++ ++ ++ ++ ++ Dinophysis acuminate + + Euglena mutabi/is* + + Flagellate Rhaphidophyte ++ Green algae Criptomonas spp* + + + Chapter 5 Field measurements ofparticulate phytochelatins and glutatilione

Blooms of the dinoflagellates K. mikimotoi, P. triestinum and P. micans, and the diatoms

Minidiscus sp., Thalassiosira sp. and S. costatum, were common in the estuary during

2002. Among these phytoplankton species, only the diatoms S. cos/alum and Thalassiosira

spp., the dinoflagellate P. micans and the protozoa E. mutabilis have been exposed to

metals under laboratory conditions to evaluate their metal detoxification responses. S.

costatum and Thalassiosira pseudonana were found to be metal tolerant and good

producers of PCs under metal stress (Ahner et al. 1995; Rijstenbil & Wijnholds 1996). P.

micans did not produce PCs as a metal detoxification mechanism, and instead, responded

with a Cu efflux system and Cu sequestration via polymeric substances within the cell

(Lage et al. 1996). There are no literature data for metal-binding peptides produced by the

metal tolerant flagellate E. mutabilis. This species has been reported as present in acidic

mine drainage waters, and can tolerate extremely high concentrations of As without apparent production of methylated species, a common pathway for As detoxification in

most algae (Casiot et al. 2004).

Freshwater phytoplankton spec1es have been reported to produce PCs at similar concentrations as marine species (Knauer et al. 1998; Pawlik-Skowronska 200 I). Thus, it can be expected that the phytoplankton species most likely to contribute to the production of particulate PCs are the numerous diatoms and green algae, either marine or freshwater species, as the abundant dinoflagellates in the Fal system have not yet been reported to produce PCs.

5.5.1.2. Particulate metal-binding peptides and total dissolved Cu in the Fa! Estuary

The chl a-normalised concentrations of particulate GSH and PC2, together with the concentrations of dissolved Curotal for the different branches of the Fa! Estuary, are

115 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione presented in Fig. 5.4. It was preferred to plot the data for metal-binding peptides using the geographical locations (sampling sites) instead of salinity, because the salinity values were relatively similar for most of the branches. In addition, the branches were subject to different metal inputs and a comparison among them was more appropriate to highlight the relations between the metal contamination and the production of metal binding peptides. In this study, only Cu was determined because of 3 aspects: 1) Cu is one of the most toxic metals to phytoplankton; 2) Cu is present at high concentrations in the Fal Estuary; 3) and time constrains to analyse other metals and metal speciation.

Phytochelatin polypeptide chains longer than n = 2 were not observed in the Fal samples.

The only field measurement of PC3 (n = 3) was reported by Ahner et al. (1997) at concentrations approximately three times lower than those of PC2. Longer chained PCs, produced by some phytoplankton species under controlled and prolonged laboratory metal exposure experiments (Ahner et al. 1995; 1997; Rijstenbil & Wijnholds 1996; Morelli &

Pratesi 1997), should be able to bind metals more efficiently due to the number of SH groups and advantageous conformational structures. The dynamic nature of the estuarine system with rapid phytoplankton turnover rates and estuarine flushing times are likely to influence the production and stability of such longer polypeptides. Thus, the absence of longer chained PCs in the samples could be due to population and/or cellular constraints of natural phytoplankton assemblages living in these dynamic environments, in addition to limitations in the sensitivity of the analytical method.

116 Chapter5 Field measurements ofparticulate phytochelatins and glutathione

a) August 2002 b) August 2002

1i1600 ~ 300 :c 0 () ~Cl 400 200 _c 0 i 1200 100 s J: ~ ~ 0 ...... l.!!il:::-:~-dM:::-~1 0 0 ~ .E LL"' I ~ 0::

c) April 2003 d) April 2003

<0 8 [ = PC2 ~ Cu-total 600 -i : ;1] [ 400 l I:::r ..:!- 2 200 ~ 200 ~ N ~ 0 L_~==~====~==~----L-~~_1 0 0 c;; 1ii Qi -, 0> ~ "' c"' 2 u; g 1- 11) a:Cl>

e) July 2003 f) July 2003

<0 0.3 iJ -:0> 0.2 C"·~ I:: r 01 1:{ 20 ~ ~ 20 ~ N 0 ~ 00 ..~ 0 Cii 1ii 11) L£ Qi Cl> 0 ...., I!: 3: Cl"' e u;"' "'~ ~ c 2 ~"' Cii g 1- :; u.. 11) Cl> ~ a: Cl> c..

Figure 5.4. Particulate PC2, GSH and dissolved CuTotal in the Fat Estuary. a) PC2 vs CuTotal for Aug-2002 survey; b) GSH vs CuTotal for Aug-2002 survey; c) PC2 vs CuTotal for April-2003 survey; d) GSH vs CuTotal for April-2003 survey; e) PC2 vs CuTotal for July-2003 survey; f) GSH vs CuTotal for July-2003 survey. *collected in October 2002. Bars span the range from the average for duplicate analyses. Dotted hi stogram columns are below detection limit.

117 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione

For the first survey (August 2002), GSH and PC2 were observed throughout the estuary at

1 1 concentrations ranging from 58 to 274 J.Lmol (g chl a)" and 2 to 36 J.Lmol (g chl a)" , respectively. The highest GSH and PC2 concentrations were found in Restronguet Creek

(site 8). A significantly higher production of metal-binding peptides, particularly PC2, was also found in Mylor Creek sample (site 6). Phytochelatins were not detected in St Mawes

(site 3), situated at the marine end of the estuary and further from the influence of the mine discharges.

A relatively high dissolved CuTotal concentration was measured (250 nM) in Restronguet

Creek during October 2002 (samples marked with * in Fig. 5.4a), at low tide, but PC2 and

GSH concentrations did not show enhanced levels accordingly. With the exception of this last sample, strong correlations were observed between the concentrations of dissolved

CuTotal and metal-binding peptides (Fig. S.Sa and S.Sb). Even stronger positive

2 relationships were observed for free aqueous Cu + (as log values) with GSH and PC2 (Fig.

S.Se and S.Sf). Dissolved CuTotal and Cu2+ concentrations also showed a close and significant correlation (correlation coefficient= 0.94; P < 0.01; r2 = 0.86).

For the other surveys (April and July 2003), GSH concentrations were found between 0.6

1 and 9 J.Lmol (g chl a)" , and PC2 concentrations were below detection limit for many samples. A maximum PC2 concentration of7 J.Lmol (g eh! a)" 1 was observed in Restronguet

Creek for April 2003 (Fig. 5.4c). High correlations between the concentrations of metal­

2 binding peptides and dissolved CuTotal and Cu + were not observed in the data set from

2003 (Fig.S.Sc,d,g,h), in spite of the similar Cu levels observed in the 2002 surveys.

118 Chapter 5 Field measurements ofparticulat e phytochelatins and glutathione

a) August 2002 b) August 2002 300 40 • ~ 0 :c 30 :c 0 ~ 200 .!:!! 0 0 20 y • 0 41 x - 36 ~- 1 . 6x + J08 E E ' Corr coef• 0.78 ::0. Corrcoef - 0 91 .:100 1 P < 0.05: R' • 0.54 N 10 o P < OO J.R • 080 J: u 0 0 (/) c.. Cl 0 0 - 0 20 40 60 80 100 0 20 40 60 80 100 Curolll (nM) Curolll (nM)

cl) April-July 2003 c) April-July 2003 -~ 3 -;___ 10 <0 0 "' :E 0 iJ 8 • 0 .!:!! 6 • • ~2 y a 0081x - I I • 1 ~ 4 • R • 0.36 ::0. • • t1 J: 2 • N (/) u Cl 0 - • c.. 0 0 20 40 60 0 L20 40 60 CUr...,(nM) CUroco~(nM)

e) August 2002 f) August 2002 "-300 40 "' 0 . y = 19.4x + 393.6 {) ;c30 • Corr coe f • 0. 78 '"0 ,9200 1 P < 0 OS; R = 0.54 0> y a 4.8x + 67 .3 0 :;520 E • Corr coef • 0.96 E -2-1oo • P < O.OI, R1 • 0.91 :c • .:10 " uN 0 ~ ~ 0 c.. 0 - ?12 -5 -7 -9 -11 -13 -15 -5 -7 -9 -11 -13 -15 2 Log [CJ] Log [Cu ·]

g) April-July 2003 h) April-July 2003 :_12.0 3.0 0 :E"' 0 0 0 .!:!! 8.0 '"~ 2.0 0 • E • .!:!! ::1. 0 E ;: 4.o _:.1 .0 (/) • • N Cl • u 0 • c.. 0 0.0 - 0.0 0 ~ -5 -7 -9 -11 -13 -15 -5 -7 -9 -1 1 -13 -15 2 2 Log [Cu · ] Log [Cu ']

Figure 5.5. Fal Estuary a) GSH vs CuTotal for 2002; b) PC2 vs CuTotal for 2002; c) GSH vs CuTotal for 2003 ; d) PC2 vs CuTotal for 2003, circled values are below detection limit (not included in the regression equation); e) GSH vs Log[Cu2+] for the 2002 survey;j) PC2 vs Log[Cu2+] for the 2002; g) GSH vs Log[Cu2+] for the 2003; h) PC2 vs Log[Cu2+] for the 2003 survey. Triangles correspond to data from April 2003.

119 Chapter 5 Field measurements afparticulate phytochelatins and glutathione

The interaction between the dissolved and the particulate phases was observed by the positive relationships between the concentrations of dissolved Cu species and particulate metal-binding peptides. The y-intercept in the graph presented in Fig. S.Sa shows a relatively high residual value for GSH, which indicated that 108 11mol (g chl a)" 1 of GSH were produced for a theoretical zero concentration of dissolved CuTotal· It is consistent with the fact that GSH is a naturally thiol produced by all eukaryotes (Noctor & Foyer 1998).

From the equations of the curves shown in Fig. S.Sa and S.Sb, it can be noted that for the

2002 survey the production of PC2 was faster than that of GSH. For example, 5 times increase in dissolved CuTotal concentration resulted in 8 times increase in PC2 production in

2 comparison to the 2 times increase in GSH production. Although free Cu + is the most toxic Cu species for phytoplankton (Giedhill et al. 1997), considerations regarding the

2 relationship between the concentrations of free Cu + and metal-binding peptides are more

2 limited here, as the Cu + concentrations were transformed into log values.

Samples collected between Carnon River and Restronguet Creek at varying salinities (I to

28) during the 2003 surveys also showed low concentrations of particulate GSH and PC2,

1 1 ranging from 3.1 to 8.3 11mol (g chl a)" and 0.2 to 3.3 J..lmol (g chl ar , respectively (Fig.

5.6), in comparison to the 2002 surveys. Dissolved CuTotal concentrations for samples C4

2 and CS were about 300 nM. Free aqueous Cu + and organic ligands were not determined for these samples due to saturation of the electrodes.

For the 2002 surveys, particulate PC2 concentrations in the Fal Estuary were comparable to those in New England (USA), where PC2 concentrations above 10 J..lmol (g chl a)" 1 were typical in heavily metal polluted harbours (Ahner et al. 1997). Contrasting results, however, were observed for the 2003 surveys for which the metal-binding peptides were at

120 Chapter 5 Field measurements ofparticulate phytochelatins and gllllathione

5 to 200-fold lower concentrations, in spite of the similarities in Cu complexation (see

further on). The most metal impacted site (Camon River) did not exhibit high production of metal-binding peptides during any survey. Restronguet Creek showed low concentrations of metal-binding peptides for some surveys. No statistically significant relationships were observed between GSH and PC2 concentrations for any of the surveys

(P > 0.05). Furthermore, neither PC2 nor GSH presented strong correlation with chl a, contrary to what has been reported for less metal affected coastal systems (Matrai & Vetter

1988; Rijstenbil et al. 1998; Tang et al. 2000).

I=GSH =PC2-- salinity ~chla --pH I 10.0 • ...... ____ • 30 Figure 5.6. Concentrations ofGSH .., I" and PC2, together with pH values 0 !!!. ~ 7.5 ' 5" and salinity for samples collected .!1' • 20 -'l. 0 n between Carnon River-Restronguet E ;:,: ~ m Creek. C I, C2 and C3 were N 5.0 (.) "E" collected in April 2003. C4 and CS a. ~ "0c 10 -; were collected in July 2003. .. c.~ I 2.5 .., VJ :I: (!) __ [L_,o C1 C2 C3 C4 CS CamonRiwr

High variabilities in the concentrations of metal-binding peptides were also observed within Elizabeth River Estuary (Wei et al. 2003) and in heavily metal contaminated lakes in Switzerland (Knauer et al. 1998) and were associated with a heterogeneous phytoplankton species composition of the sampling areas. The heterogeneity in phytoplankton composition in the Fa! Estuary may explain the lack of correlation between metal-binding peptides and chl a concentrations. For instance, dinoflagellate species common in the Fal Estuary (P. micans) and Carnon River (E. mutabilis) seem to have alternative metal detoxification mechanisms to PC production; consequently their high abundances could have lead to relative low levels of PC production. The assumption that

121 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione low concentrations of particulate PCs in the field indicate a Jack of metal stress should therefore be taken with caution. Furthermore, as the Fa! Estuary has been contaminated with metals for centuries, the biological responses to metal toxicity could be altered through evolution of metal tolerant strains with detoxification mechanisms other than PC production.

Another contributing factor to the high variabilities in the concentrations of the metal­ binding peptides for the surveys could be a limitation in nutrients, particularly nitrogen, as it is required for peptide synthesis (Rijstenbil et al. 1998; Wei et al. 2003). Nutrient levels in the Fa! system, however, have lead to regular episodes of algal blooms, including toxic ones (Langston et al. 2003b), and therefore were unlikely to be affecting the production of thiols in the Fa! system.

5.5.1.3. PC:GSH ratios

The PC2:GSH ratios together with the PC2 concentrations for the different branches of the

Fa! Estuary are plotted in Fig. 5.7. Restronguet Creek (April2003) and Carnon River (Cl,

April 2003) showed the highest PC2:GSH values, 0.89 and 0.40, which were more indicative of metal stress than their respective PC2 concentrations alone. Samples from the

Camon River, C2 and C3, (July 2003) were collected at lower salinities, therefore the dominating phytoplankton were likely dinoflagellate species, which have not yet been reported to produce PCs. This could explain the low PC2 concentrations and low PC2:GSH values for these metal contaminated samples.

122 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione

aJ

I--- PC2:GSH --+- PC2 I

~ !- m u; Q) £ 0 a; 'L 'L :l "' :l :l 0 0 b Q) L!:: ...., ;;: 0 >- Cl Q) :l <1l >- >- "' Cl !2 (i) E ::il c: ::il ::il c: 2 m g ~ IS I- ~ u.. Q) 0 g "§ "'Q) >- c:: =<1l ::il "'Q) Q) c:: CL

~ N ("") m u; Q) £ 0 a; :l "' :l :l c: c: c: L!:: ...., ;;: 0 >- Cl 0 0 0 e <1l ::il c: c: (i) E E E ~ 2 <1l <1l <1l <1l I- ~ u.. u u u :l ~Q) ~ Q) c:: CL

~ I!) m u; Q) £ 0 a; v :l "' :l :l c: c: L!:: ...., ;;: 0 >- Cl 0 0 e <1l ::il c: :l (i) E E E ~ ~ m g <1l <1l I- ·:; u.. u u "'Q) ~ Q) c:: CL

Figure 5.7. Comparison between PC2 concentrations and PC2:GSH ratios for the Fal Estuary samples. a) August 2002; b) April 2003; c) July 2003.

123 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione

Most of the PC2:GSH ratios in the present study were lower than 0.05, yet within the same range found in Elizabeth River (Wei et al. 2003). Higher ratio values were observed for a number of sample locations in Galveston Bay (0.018-0.52), a coastal system not considered metal polluted (Tang et al. 2000). Apart from the Restronguet Creek and

Camon River samples, these ratios suggest that phytoplankton in the Fa! Estuary branches are not under strong metal stress. It is important to highlight again that PC production is phytoplankton species dependent and the development of tolerant strains could mask the biological responses to metal stress. Furthermore, the disadvantage of using the PC2:GSH ratios is the limitation caused by the lack of data on how GSH concentrations vary in the field. For instance, some studies have associated variations in GSH production with parameters other than metal stress, such as the presence of xenobiotics, light availability, nitrate and phosphate levels (Matrai & Vetter 1988; Rijstenbil et al. 1998; Tang et al.

2000). The Fal Estuary has been subjected to several other contaminants (Table 5.1) that could have effects on the intracellular GSH concentrations, by means of GSH S­ transferases in the detoxification of xenobiotics (Noctor & Foyer 1998). Furthermore, zooplankton and bacteria, also retained in the particulate phase, can cause an overestimation of GSH production by phytoplankton. Nevertheless, the advantage in using

PC:GSH ratios to examine metal stress relies primarily on the elimination of the uncertainties caused by eh\ a normalisation, as chl a levels vary greatly as a function of light intensities, stage of phytoplankton life cycle, and phytoplankton species (Wei et al.

2003).

5.5.1.3. Copper complexatioo in the Fat Estuary waters

Representative Cu-ligand titration and linearization curves are shown in Fig. 5.8. The curvature at the beginning of the titration curve (Fig. 5.8a) indicates the presence of natural

124 Chapter 5 Field measurements ofparti c11fate phytochelatins and gl11tathione

Cu-complexing ligands in the water sample. Linear treatment of the titration data (Fig.

5.8b) produced a straight line for all samples, indicating that Cu-complexation was controlled by a single class ofligands (Campos & van den Berg 1994).

a) b)

200 4.0 0

150 0 3.0 0 ...J .s~ :::1 .!! 100 ~'; 2.0 -i 5 5 u u 50 1.0 I oO 0 oP 0 0 0.0 ~ 0 100 200 300 0 50 100 150 200 Curotal (nM) CUtablle (nM)

Figure 5.8. a) Typical titration curve for the estuarine samples (Mylor Creek, July 2003); b) van den Berg linearization for the titration data. CuToral = Cu in the sample plus Cu added for the titration; CuL = Cu complexed by natural organic ligands; Cu1abile = Cu complexed by SA.

2 The concentrations of dissolved CuTotah organic ligands (L), free Cu +, inorganic Cu (Cu'), together with the conditional stability constants of the Cu-organic complexes (Log KcuL) and the degree of Cu-complexation (%CuL) for the separate field campaigns, are shown in

Table 5.3. The dissolved organic ligands were at concentrations higher than or close to those of CuTotal for most of the samples. The high conditional stability constants for the Cu- organic complexes found in the water samples (Log KcuL = 10.8-13.5) were in the range for the strong ligands. Copper was complexed by strong Jigands (1 00% CuL), as the major species for most of the sites. Complete saturation of the organic ligands by Cu was only observed in Restronguet Creek during low tide ([CuTotal] > [L]). For this sample, the titration curve produced a straight line, without the initial curvature shown in Fig. 5.8a.

2 Such ligand saturation resulted in elevated free Cu + and Cu' concentrations.

125 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione

Table 5.3. Copper complexation in the Fa! Estuary waters. The concentrations of CuTo~al are means 2 obtained from duplicate measurements. %CuL = ([CuTo~al]- [Cu +]- [Cu'])/[CuTo~al])x!OO.

nM M

CUTotal L Log KcuL CuZ+ Cu' %CuL 08-Aug-01 Fal/Truro 13.4 ± 1.5 73.4 ± 2.0 13.5 ± 0.9 6.7E-15 9.3E-15 100.0 St Just 10.9±0.7 34.8 ± 3.8 12.5 ± 0.6 1.3E-13 1.9E-13 100.0 St Mawes 6.7 ± 0.5 24.4 ± 0.4 13.2 ± 0.5 2.2E-14 3.0E-14 100.0 Falmouth Yacht 19.2 ± 1.5 65.1±2.3 13.1±0.7 3.6E-14 5.0E-14 100.0 Mylor 36.2 ± 1.5 95.1±7.9 13.0 ± 0.4 6.0E-14 8.5E-14 100.0 Mylor/Restronguet* 51.2 ± 1.0 52.3 ± 1.5 12.3±0.1 2.0E-11 2.8E-ll 99.9 Restronguet I 81.9 ± 1.2 Restronguet2 100.0 ± 1.2 63 ± 30 10.8 ± 0.03 2.7E-08 3.8E-08 35.6

30-Apr-03 Fal/Truro 33.9 ± 0.5 39.9 ± 0.8 12.9 ± 0.3 7.5E-13 l.IE-12 100.0 St Just 30.0 ± 6.0 48.4 ± 1.1 12.6 ± 0.2 3.8E-13 5.3E-13 100.0 Percuil/Mawes 28.1 ± 2.2 41.6 ± 0.3 13.0±0.1 2.0E-13 2.8E-13 100.0 Falmouth Yacht 17.0±2.6 27.7 ± 2.4 13.1±0.8 1.2E-13 1.6E-13 100.0 Restronguet 500 620 ± 18 11.7 ± 0.2 6.2E-12 8.6E-12 100.0

01-Ju/-03 Fal/Truro 31.8 ± 0.6 32.3 ± 1.2 12.3±0.1 2.8E-ll 4.0E-ll 99.8 St Just 20.8 ± 0.4 21.4±2.4 11.9±0.1 4.0E-ll 5.7E-11 99.5 StMawes 25.2 ± 0.7 36.0 ± 1.0 13.4 ± 0.5 9.4E-14 1.3E-13 100.0 Falmouth Yacht 25.8 ± 0.7 37.0 ± 4.0 12.1±0.1 2.0E-12 2.8E-12 100.0 Mylor 43.7±1.4 70.5 ± 4.0 12.6 ± 0.5 4.4E-13 6.2E-13 100.0 Restronguet 38.5 ± 0.9 44.7±1.4 12.8 ± 0.2 l.OE-12 1.4E-12 100.0

Higher concentrations of dissolved strong ligands observed for some of the branches

(mostly during the 2002 survey) were likely due to anthropogenic inputs of organic matter

and run off of humic substances from the catchment area. No obvious relationship was observed between the concentrations of dissolved organic ligands and eh! a for the surveys

(Fig. 5.9). This is consistent with the fact that the dissolved organic ligands in coastal

waters are derived from a range of sources and are not only associated to phytoplankton

exudates, as is likely the case for oceanic waters (Croot et al. 2000).

126 Chapter 5 Field measurements ofparticulate phytochelatins and glmathione

a)

I --<>-organic ligands _,._ chl a I ~ 5. 100 j I 20 {l 80 - - 16 g. ~ 60 I 12 ; ~ 40 ~ [ 8 ~ ·c: 20 Il 4 ....;-r: ~ 0 l ·0 0 rJ) ~ n; u; Q) £ 0 a; ::J ::J ~ ::J 1!:: -, :;: 0 :;. Q) Cl ro ::2: <: e ii5 ::2: E ~ 2 ::::: n; 0 g 1- LL. :;. rJ) ::J Q) ~ ::2: Q) a:: c.

b)

::2: -;- 8oo 1 I 1o '"' ~600-1 ~---- ~-8 ~ ~ 4oo I --- ~ i ·c; 200 - - 2 r.:. ~ 0 J--~==~====~====~------01 0 rJ) n; u; Q) £ 1!:: -,::J :;: ::J ro 0 e ii5 E 2 ~ n; 1- ·s LL. ~ Q) c.

c)

ro £ ::J 0 ~ (Jl E 2 "iii 1- LL.

Figure 5.9. Dissolved organic ligands and chlorophyll a concentrations for the Fa! Estuary samples. a) August 2002; b) April 2003; c) July 2003.

127 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione

The highest eh! a content in Mylor Creek for July 2003, indicative of a phytoplankton bloom, coincided with the highest ligand concentration. Again, it could be simply due to inputs of organic matter and nutrients (favouring phytoplankton blooms) associated to anthropogenic sources and run off from the catchment area. The Cu complexation in the

Fal Estuary followed the trends observed in other aquatic systems for which the molar ratios of strong organic ligands to Curotal were approximately I: 1. This relation is believed to be tightly modulated by the concentrations of dissolved Curotal (Kozelka & Bruland

1998). Weak dissolved Cu-complexing ligands, not detected by the established analytical conditions, should be at concentrations higher than the strong ligands (Moffett et al. 1997) and could also be important for the total metal buffering capacity of the system, despite their weaker binding strengths.

The abundant presence ofphytoplankton species such asP. micans and S. costatum, able to produce exudates with different binding strengths (Croot et al. 2000), may also be a key characteristic for phytoplankton adaptation in such metal affected systems. This feature is ecologically relevant as the released ligands could benefit other metal-sensitive zoo- and phytoplankton species by decreasing the concentration of bioavailable toxic metals in the surrounding waters, and balancing the low production ofphytochelatins.

The structure and exact composition of the dissolved Cu complexing ligands remam unknown in spite of several attempts at elucidating their identity (Gordon et al. 1996;

2000; Ross et al. 2003). Potential metal complexing ligands include sulphide and thiols like GSH and cysteine (Gordon et al. 1996; 2000; Lea! & van den Berg 1998). The

2 124 conditional stability constants for GSH and cysteine, on the basis of Cu +, are 10 and

13 0 I 0 , respectively (Lea! & van den Berg 1998), which are within the range for the strong

128 Chapter 5 Field measurements ofparticulate plzytoclzelatins and glwathione

ligands found in the present study. Dissolved GSH, determined by voltammetric techniques, was found at concentrations up to I 00 nM in the Mersey Estuary (Le Gall & van den Berg 1993). Dissolved GSH and PC2, determined by chromatography, were observed ranging from 0.2 to 6.2 nM and 1.2 to 3.0 nM, respectively, in Galveston Bay

(Tang et al. 2000). Tang et al. found that dissolved GSH and PC2 were at I 0 and I 00 fold higher concentrations than in the particulate phase, respectively. It seems unlikely that PCs participate in the bulk of the dissolved metal-organic complexes, as PC-metal complexes undergo rapid dissociation and degradation once exported from cells (Lee et al. 1996). A high rate of GSH efflux would be necessary considering the lower levels of GSH in phytoplankton cells (Ahner et al. 1997; Ahner et al. 2002). Therefore, dissolved GSH in natural waters is likely originating from other sources in addition to phytoplankton.

5.5.1.4. Other dissolved metals and effects on tbe production of metal-binding peptides in tbe Fal Estuary

Apart from the Restronguet Creek and the Camon River, the rest of the Fa I system has presented dissolved metal concentrations in the water column that mostly fall within the

Environmental Quality Standard (EQS) values (Table 5.4). The tidal regime and degree of water stratification are responsible for the great variabilities observed for the metal concentrations in Restronguet Creek (Langston et al. 2003b). Zinc and Cu concentrations in other parts of the Fal system occasionally exceed the EQS values due to the influence of the sewage treatment works in the upper Fal Estuary and activities in the Falmouth dockyard area (Langston et al. 2003b).

129 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione

Table 5.4. Dissolved metal concentration ranges (nM) in the Fa! Estuary (data from Environmental Agency 1990-200 I, estimated from graphs in Langston et al. 2003b; EQS = Environmental Quality Standard for annual average). Fall Percuil Fal off Penryn Falmouth Fal off Restronguet EQS Mylor Dock Marina Restronguet mouth Truro Cu 32-47 80 80 80 79-157 80 425 16-441 792d

Zn 77-918 77-382 275-1223 290- 275-459 183-6500 122-1911 11 306 6122d Cd 2.7 1.8-80 8.911 222d 1 As 13-67 444 d 3342d

European Community Dangerous Substances Directive for freshwater 2 European Community Dangerous Substances Directive for saltwater d =dissolved (i.e. involving filtration through 0.45 JJ.m membrane filter before analysis) t =total (i.e. without filtration)

As discussed in the previous chapter, there is evidence that combinations of the free ions

M n 2+ , Pb2+ , C u 2+ , z n 2+ an d Cd2+ cause antagomstJc,. . synergistJc, . . an d suppressiOn . e f",ects on

PC production, by metal competition for cellular binding sites and mechanisms still not fully elucidated (Pawlik-Skowronska 2001; Wei et al. 2003). Free Cd2+ has been shown to be the most effective inducer of PC production for many phytoplankton species under laboratory conditions (Ahner & More) 1995). However, the high levels of dissolved Zn and

Mn in Restronguet Creek could have decreased Cd2+ effects on PC production, as Cd2+ uptake takes place through the same cellular channels as Zn and Mn (Sunda & Huntsman

1998). Thus, the variabilities observed in the concentrations of metal-binding peptides for samples from Restronguet Creek and Carnon River could be in part a result of the metal interactions.

130 Chapter 5 Field meas11rements ofpartie~~late phytochelatins and gl11tathione

Despite the complexity of the estuarine systems regarding to the phytoplankton composition and its variety of possible strategies to cope with metal toxicity, enhanced production of metal-binding peptides, particularly PC2, were observed at the sites with high dissolved Curatal and free Cu2+ concentrations. This supports the role attributed to the particulate metal-binding peptides in protective mechanisms against metal toxicity and as a sub-lethal response in impacted aquatic systems.

5.5.2. Plymouth Sound: Tamar and Plym Estuaries

5.5.2.1. Phytoplankton composition of the Tamar Estuary

Chlorophyll a concentrations were similar for most of the samples from the Tamar

1 transect, ranging from 5.2 to 14.1 IJ.g L- , and indicated phytoplankton blooms only for the two samples at the saline end of the transect (Fig. 5.10). As for the Fal Estuary, little information concerning phytoplankton composition is available for the Tamar and Plym estuaries. A compilation of data by Langston et al. (2003b) indicates that the upper Tamar

Estuary can be dominated by freshwater phytoplankton species, including the diatom

Cyclotella atomus, green algae Nannochloris sp., and dinoflagellates Heterocapsa triquetra and Amphidinium sp. (with dinoflagellates contributing less to the biomass). In addition, marine species such as the diatoms Rhizosolenia delicatula, Nitzchia closterium, and Skeletonema costatum, and the dinoflagellate Prorocentrum micans, can be transported by the flood tide from Plymouth Sound into the brackish waters within the Tamar to survive in the short term. Low light availability, as a result of the high turbidity of the

Tamar estuarine waters (Langston et al. 2003a), could limit the development of phytoplankton species that are more light-dependent, such as diatoms.

131 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione

. -o- chl a --salinity -6- PC2:Gs8J Figure 5.10. Salinity and chl a 40 -- concentrations together with 0.20 __. I PC2:GSH ratios for the Tamar 1 transect, 25 h March 2003. ~ 30 - ~ 0.15 Cl -.::a. ""0 ::;::"' 0 N ~ 20 l 0.10 Q c (/) "' I :it:'s (ij 10 (/) J005 0 0.00 Ql Ql Ql >- E -"" tO >- Qj c c -"" tO 0 Q) :::> tO .c ·c; ·c; tO £ .c "':::> :::> c Cii ~ .c0 a a >- e- c... ·c: Q) (ij ~ ...... J 0 0 -"" ·c c ~ () () 1- "' :::> c =* 0 (!) 0 0::: c ~ > ~ ""Ql Ql c... 0

A continuous mcrease in phytoplankton diversity can be expected down estuary, with progressive gain in marine species (Fogg & Thake 1987b; Langston et al. 2003a).

Therefore, as for the Fal Estuary, the phytoplankton species most likely to contribute to the production of metal-binding peptides in the Tamar and Plym estuaries are diatoms and green algae, either marine or freshwater species.

5.5.2.2. Particulate metal-binding peptides and Cu complexation in the Tamar

Estuary

The chl a-normalised concentrations of particulate PC2 were below the detection limit for most of the samples of the Tamar transect (Fig. 5.11 ). The highest PC2 concentration, 16.5 llmol (g chl ar1 was observed in Gunnislake at the tidal limit of the Tamar. Particulate

GSH was found at concentrations ranging from 18.8 to 244 llmol (g chl ar1 with highest value in Morwelham. The PC2:GSH ratios were the highest in Gunnislake, which receives inputs of metals from old mine workings, spoil heaps and adits (Langston et al. 2003a).

132 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione

---. GSH ---- PC2 Figure 5.11. Particulate GSH 300 24 and PC2 concentrations for the Tamar transect, 25 111 March 2003. Open symbols for PC2 indicate {) 200 . values below detection limit.

0 E ..5 V5 100 . ~

0 ,., 0 Cl> E Cl> ,., Q; ro ~ ro ro 'E 'E ""ro Qi ::I ::I ::I .J::: 'Ci 'Ci 'iii ,!: .J::: 0 0 c: 0.. .J::: 0 ,., e- i ~ ·c; Qi (ij ~ 11 ...J 0 1/) c: ~ (.) (.) ""c: 1- "c: ::I 0 'P 0 ~ c: ~ > ::::!; a: Cl> Cl> 0.. 0

2 Dissolved CuTotal. free Cu +, inorganic Cu (Cu'), and organic ligand concentrations were determined for selected sites in the Tamar transect from a separate field survey. Dissolved

CuTotal concentrations were between 34.0 and 91.0 nM, and slightly lower than those of

2 11 12 organic ligands (34.4 to 92.3 nM), resulting in free Cu + concentrations of 10" to 10" and

Cu-organic complexes with conditional stability constants from 10 12 to 10 13 (Table 5.5).

2 These data indicate a decrease in Cu + concentrations down estuary with increasing salinity. As for the Fal Estuary and other aquatic systems (Kozelka & Bruland 1998), an approximately 1 : 1 molar ratio was observed for the Tamar samples.

Table 5.5. Copper complexation in the Tamar Estuary. Dissolved copper species, organic ligands 2 and conditional stability constants for the Cu-organic complexes. %CuL = ([Cu10131] - [Cu +] - [Cu ' ])/[CuTotarD* I 00 281 March nM M % 2003 CuTotal L Log KcuL Cu2+ Cu' CuL Halton Quay 34.0 ± 0.6 34.4 ± 1.8 12.9 ± 0.2 3.0E-11 4.2E-ll 100.0 Cargreen 91.0 ± 1.6 92.3 ± 1.4 12.8 ± 0.08 1.1E-11 1.6E-11 100.0 Saltash 71.2±1.1 72.5 ± 1.4 13.0 ± 0.2 1.4E-1 2 1.9E-1 2 100.0 Devil's Point 30.0 ± 0.6 35.3 ± 7.4 12.5 ± 0.3 7.7E-12 2.2E-1 1 100.0

133 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione

5.5.2.3. Other metals in the Tamar and Plym estuaries

As in the Fal Estuary, other potential toxic metals such as Cd, Zn and As can be found at elevated concentrations in the Tamar Estuary (Table 5.6). In general, dissolved metal concentration averages were below the EQS values, thus, acute biological and ecological effects were not expected to occur (Langston et al. 2003). Nevertheless, in the present study PC production was highly pronounced in Gunnislake, indicating a sub-lethal metal effect on phytoplankton. Another phytoplankton detoxification response reported for these estuarine waters was linked to arsenic with the presence of dissolved biomethylated arsenic species (monomethylarsenic and dimethylarsenic) at the seaward end (Howard et al. 1988).

These results also indicated that the phytoplankton in the Tamar Estuary were under metal stress.

Table 5.6. Dissolved metal concentration ranges (nM) in freshwater and tidal waters of Tamar Estuary. Data from 2001 , except Halton Quay (1997) and Tamar offLynher (1993). STW =sewage treatment work; EQS = Environmental Quality Standard for annual average (all data from Langston et al. (2003a) obtained from graphs, except EQS). Gunnislake Halton Ernesettle Shellfish Off Plymouth Plym EQS Quay STW water Lynher STW (Marsh (nM) Mills) 1 Cu 79-220 43-79 8-39 32-87 8-79 24-87 16-441 d 792d

Zn 77-275 31-214 61-122 61-184 31-92 122-612 122- 1911 11

2 612 d

1 Cd 0.3-0.7 0.4-!.2 1.3 0.9 1.3 1.2 8.9 I 222d

1 As 27-80 27-67 27-71 33-40 20-33 27-67 444 d 2 334 d 1 European Community Dangerous Substances Directive for freshwater 2 European Community Dangerous Substances Directive for saltwater d = dissolved (i.e. involving filtration through 0.45 ).!m membrane filter before analysis) t = total (i.e. without filtration)

134 Chapter 5 Field measurements ofparticulat e phytoche/atins and glutathione

Whether the dissolved metal concentrations in the Tamar Estuary are at a threshold that causes metal interactions capable of affecting PC production is an issue that requires further investigation.

5.5.2.4. Particulate metal-binding peptides in the Plym Estuary and Plymouth Sound

Chlorophyll a and particulate GSH concentrations in the Plymouth Sound were found to be

1 1 in the range 1.9 and 14.4 J.lg L- and 111 and 419 J.lmol (g chl af , respectively (Fig. 5.12).

1 Chlorophyll a concentrations in the Plym Estuary ranged from 1.5 to 3.0 J.lg L- , with highest value of37.5 J.lg L-1 at the marine end ofthe estuary (Oreston, 2001), indicative of a phytoplank:ton bloom. Particulate GSH concentrations in the Plym Estuary increased seaward from 18.0 to 130 J.lmol (g chl a) -t. The low chl a values were consistent with other data (Langston et al. 2003b), and were probably low due to the high turbidity of the waters.

Particulate PCs were not detected in any of the samples collected in the Plym and

Plymouth Sound. As part of the initial adaptations of the analytical method for thiols, these samples were filtered using 0.45 J.lffi pore size filters, instead of 0.8 J.lm, which limited the collection of sufficient biomass, as the filters clogged with less than 1 L of the filtered sample, probably by particles other than phytoplank:ton cells.

135 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione

c::J GSH --- chl a

500 40 Figure 5.12. Concentrations of particulate GSH and chl a in 400 Plymouth Sound (including 30 Oreston) for 2001-2002, and i3 ";'Cl 300 ~ Plym Estuary (Plym I, Plym2, Q) 1 20~ Plym3 and Plym4) for 24 h ~ CO ..=> 200 September 2002. J: ~ (/) ~ 10 100

0 0 c:::L.. 0 !!! Q) t: c I"> 0 .9 E E Ql ~ Q; 0.. en 'E>- "'>- > e>- c ""~ ~0: 0 ~ a: a: a: a: 0 u > 0 Ql 0

A linear regression for the particulate GSH and eh! a concentrations showed a strong positive linear correlation for the Plymouth Sound samples (Fig. 5.13). Particulate GSH and chl a concentrations for the Tamar transect, however, did not show any obvious relationship (Fig. 5.14). These contrasting results can reflect the heterogeneity composition of the samples from the Tamar transect (likely composed of a mixture of diatoms and dinoflagellates, either marine or freshwater species) in relation to the Plymouth Sound

(dominated by marine diatoms) and Plym Estuary.

As outlined in section 5.5.1.3, GSH can be produced by most eukaryotic organisms and production may be affected by high light intensity, nutrient supply and organic xenobiotics

(Matrai & Vetter 1988; Rijstenbil et al. 1998; Noctor & Foyer 1998). Lack in inorganic nitrogen supply, however, should not have affected GSH production in the Tamar and

Plym estuaries, as the upper estuaries are subject to nutrient enrichment from agricultural run-off and sewage discharges (Langston et al. 2003a). The lack of correlation between

GSH and Chl a concentrations could also indicate that the production of particulate GSH in the Tamar transect was not entirely derived from eukaryotic phytoplankton sources.

136 Chapter 5 Field measurements ofparticulat e phytochelatins and glutathione

Figure 5.13. Relationship between 30 particulate GSH and chl a for the 25 samples from the Plymouth Sound 0 and Plym Estuary. 2.0 ~ c ::; 1 5 (/) y = 0.071x 2 Cl = 1 0 R 0.71 0 Correlation Coef. = 0.87 0.5 p < 0.05

o.o oi:9 0 10 20 30 40 Chi a (119 L·'l

2.0 0 Figure 5.14. Relationship between particulate GSH and chl a for the 1 5 0 samples from the Tamar transect. ~ c: i 10 0 (/) C) 0 0 05 0 00 y = 1.2 - 0.049x 2 0 0 R = 0.08 Correlation Coef. = -0.27 00 p > 0.05 0 5 10 15 1 Chi a (llg L" )

Metal speciation was not determined for the Plymouth Sound and Plym Estuary. Recent literature data from the Environment Agency (Langston et al. 2003) indicate that the dissolved metal concentrations for the Plym Estuary (Marsh Mills) are within the EQS values (Table 5.3). It is important to note, however, that these EQS values do not guarantee a safe threshold for the sub-lethal chronic effects on biota and their consequences.

Therefore, the determination of biological responses such as PC production to provide stress indication could be an additional tool to the environmental assessment of the water quality.

137 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione

5.6. Summary: particulate metal-binding peptides in the Fal Estuary,

Plymouth Sound and other natural waters

The results presented here exemplify the challenges to investigate biological responses in complex environments such as estuaries, for which the dynamic nature involving tidal systems has great effects on the distribution of organisms and contaminants. The river/tidal flow can influence phytoplankton production and taxonomic distribution in an estuarine system through a number of mechanisms, including changes in: nutrient inputs, dilution rates or advection of phytoplankton cells out of the estuary, and light availability through stratification, gravitational circulation, and longitudinal positioning of turbidity maximum

(Day et al. 1989).

The few field studies undertaken so far attributed the variations in the concentration of the metal-binding peptides not only to metal stress but also to the heterogeneity in phytoplankton species composition (Knauer et al. 1998; Tang et al. 2000; Wei et al. 2003).

The sum of the estuarine abiotic and anthropogenic factors can hamper the interpretation of the variabilities in PC production, and the direct use of PCs as a metal stress indication to phytoplankton. Knauer et a! (1998), for example, investigated PC production in metal contaminated Swiss lakes, which have different hydrodynamic characteristics than estuaries, and did not observe particulate PCs in the most metal polluted lake. In contrast,

Tang et al. (2000) found that PCs were ubiquitous in the estuarine waters off of Galveston

Bay (USA), showing different patterns (e.g. different slopes for PC2 vs eh! a) for the upper and lower bay. Such patterns were possibly a result of the occurrence of different phytoplankton blooms and/or phytoplankton species abundances. Likewise, Wei et a!

(2003) found high variabilities in particulate GSH and PC2 concentrations for a same

138 Chapter 5 Field measurements ofparticu/ate phytochelatins and glwathione

sampling site during different sampling surveys. They also suggested interactions among metals (Cu, Zn and Cd) as a contributing factor to the variabilities in the production of thiols.

The Fal and the Plymouth Sound estuaries have long metal contamination histories and are subject to other anthropogenic contaminants that also pose deleterious effects on phytoplankton health (Table 5.1 ). The most common important aspects of the Fa I and

Plymouth Sound estuaries in relation to metal contamination noted here are summarised below:

(i) the EQS values were rarely exceeded, except at the most metal affected sites,

Restronguet Creek (Fal Estuary) and Gunnislake (Tamar Estuary) (Langston et

al. 2003a; b);

(ii) the concentration of dissolved Cu-complexing organic ligands were higher than

the concentrations of total dissolved Cu for most of the sites, meaning that the

Cu-complexing capacity of the waters probably reduced Cu bioavailability, and

thus Cu toxicity to phytoplankton;

(iii) particulate GSH and PC2 were found at high concentrations at the most metal

2 affected sites, where the concentrations of the free Cu + species were high and

the dissolved organic ligands were saturated with Cu.

The high production of PCs and GSH in these estuaries was indicative of areas where Cu species were present at high concentrations in the bioavailable form (Cu 2+) for most phytoplankton species. These results indicate that the production of metal-binding peptides is an important mechanism against metal toxicity, especially at the most affected sites

139 Chapter 5 Field measurements ofparticulate phytoclrelatins and glutathione

(Restronguet and Gunnislake). Again, phytoplankton species composition should be considered, as not all species present in the riverine parts of Restronguet Creek appear to be able to produce PCs. It can be concluded that the production of particulate PCs in the

Fal and Plymouth Sound estuaries represents a general stress indication for a mixture of dissolved metals.

5. 7. Conclusions and final remarks

• This study contributed to establish the first field measurements ofparticulate metal­

binding peptides in European estuaries, using the Fal Estuary and the Plymouth

Sound (Tamar and Plym Estuaries) in the UK as exemplars.

• The surveys in the Fal Estuary for two consecutive years showed distinct results for

the production of metal-binding peptides and its relation with Cu speciation. The

2002 surveys presented higher concentrations of particulate PC2 and GSH than the

2003 surveys. For the 2002 surveys, the production of PC2 showed to be a faster

response than that of GSH under increasing dissolved CuTotal concentrations.

Although other dissolved toxic metals could have influenced the production of

metal-binding peptides, only Cu speciation was investigated in these systems due to

constraints to develop this project within the available time.

• High production of PC2 and GSH in the most metal impacted sites, Restronguet

Creek (Fal Estuary) and Gunnislake (Tamar Estuary), indicates that these metal­

binding peptides are part of an important adaptive mechanism against metal stress.

However, phytochelatin concentrations did not reach high levels in Restronguet

Creek and Camon River for all surveys. These could be mainly due to:

140 Chapter 5 Field measurements ofparticulate phytochelatins and glutathione

i. phytoplankton spectes composition which were dominated by non-

phytochelatin producers

11. antagonistic and/or suppressing effects on PC production caused by the

interaction among the metals present at high concentrations

• At present, a combination of ancillary parameters, including metal speciation,

determination of chlorophyll a, an estimation of phytoplankton species composition

in the samples, are required for a full interpretation for the variations in PC and

GSH concentrations in the field.

5.8. References

AHNER, B.A., KONG, S. & MOREL, F.M.M. (1995) Phytochelatin Production in Marine-Algae .1. An Interspecies Comparison. Limnology and Oceanography, 40, pp. 649- 657.

AHNER, B.A. & MOREL, F.M.M. (1995) Phytochelatin Production in Marine-Algae .2. Induction by Various Metals. Limnology and Oceanography, 40, pp. 658-665.

AHNER, B.A., MOREL, F.M.M. & MOFFETT, J.W. {1997) Trace metal control of phytochelatin production in coastal waters. Limnology and Oceanography, 42, pp. 601- 608.

AHNER, B.A., WEI, L.P., OLESON, J.R. & OGURA, N. (2002) Glutathione and other low molecular weight thiols in marine phytoplankton under metal stress. Marine Ecology­ Progress Series, 232, pp. 93-103.

BRYAN, G.W. & GffiBS, P.E. (1983) Heavy metals in the Fal Estuary, Cornwall: a study of long-term contamination by mining waste and its effects on estuarine organisms. Marine Biological Association ofthe United Kingdom - Occasional publication, 2, pp. 1-111.

CAMPOS, M.L.A.M. & VAN DEN BERG, C.M.G. (1994) Determination of copper complexation in sea water by cathodic strippng voltarnmetry and ligand competition with salicylaldoxime. Analytica Chimica Acta, 284, pp. 481-496.

CASIOT, C., BRUNEEL, 0., PERSONNE, J.C., LEBLANC, M. & ELBAZ­ POULICHET, F. (2004) Arsenic oxidation and bioaccumulation by the acidophilic protozoan, Euglena mutabilis, in acid mine drainage (Camoules, France). Science of the Total Environment, 320, pp. 259-267.

141 Chapter 5 Field measurements ofparticlilate phyrachelatins and glwathione

CHEUNG, V. V. & JHA, AN. (2004) An evaluation of the relative sensitivity of cyto- and genotoxic biomarkers in two marine bivalve mollusc species. Comparative Biochemist!)' and Physiology Part A, 137, pp. S123.

CROOT, P.L., MOFFETT, J.W. & BRAND, L.E. (2000) Production of extracellular Cu complexing ligands by eucaryotic phytoplankton in response to Cu stress. Limnology and Oceanography, 45, pp. 619-627.

DAY, J.W.J., HALL, C.A.S., KEMP, W.M. & YANEZ-ARANCIBIA, A (1989) Estuarine Phytoplankton. In: Estuarine Ecology (Ed. by J.W.J.Day, C.AS.Hall, W.M.Kemp & AYanez-Arancibia), pp. 147-187. John Wiley & Sons, New York.

FOGG, G.E. & THAKE, B. (1987a) Conclusions. In: Algal cultures and phytoplankton ecology pp. 209-222. The University ofWiscosin Press, London.

FOGG, G.E. & THAKE, B. (1987b) Phytoplankton in its natural environment. In: Algal cultures and phytoplankton ecology pp. 81-121. The University of Wisconsin Press, London.

GLEDHILL, M., NIMMO, M., HILL, S.J. & BROWN, M.T. (1997) The toxicity of copper(II) species to marine algae, with particular reference to macroalgae. Journal of Phycology, 33, pp. 2-11.

GORDON, AS., DONAT, J.R., KANGO, R.A, DYER, B.J. & STUART, L.M. (2000) Dissolved copper-complexing ligands in cultures of marine bacteria and estuarine water. Marine Chemistry, 70, pp. 149-160.

GORDON, AS., DYER, B.J., KANGO, R.A & DONAT, J.R. (1996) Copper ligands isolated from estuarine water by immobilized metal affinity chromatography: Temporal variability and partial characterization. Marine Chemistry, 53, pp. 163-172.

HOWARD, A.G., APTE, S.C., COMBER, S.D.W. & MORRIS, R.J. (1988) Biogeochemical control of the summer distribution and speciation of arsenic in the Tamar Estuary. Estuarine Coastal and Shelf Science, 27, pp. 427-443.

KNAUER, K., AHNER, B., XUE, H.B. & SIGG, L. (1998) Metal and phytochelatin content in phytoplankton from freshwater lakes with different metal concentrations. Environmental Toxicology and Chemistry, 17, pp. 2444-2452.

KOZELKA, P.B. & BRULAND, K.W. (1998) Chemical speciation of dissolved Cu, Zn, Cd, Pb in Narragansett Bay, Rhode Island. Marine Chemistry, 60, pp. 267-282.

LAGE, O.M., PARENTE, AM., VASCONCELOS, M.T.S.D., GOMES, C.A.R. & SALEMA, R. (1996) Potential tolerance mechanisms of Prorocentrum micans ( dinophyceae) to subletha11evels of copper. Journal of Phycology, 32, pp. 416-423.

LANGSTON, W.J., CHESMAN, B.S., BURT, G.R., HAWKINS, S.J., READMAN, J. & WORSFOLD, P. (2003a) Characterisation of European Marine Sites. Plymouth Sound and Estuaries (candidate) special area of conservation. Marine Biological Association, Occasional publication, 9, pp. 1-202.

142 Chapter 5 Field measurements afparticulate plzytoclzelatins and glwatlzione

LANGSTON, W.J., CHESMAN, B.S., BURT, G.R., HAWKINS, S.J., READMAN, J. & WORSFOLD, P. (2003b) Characterisation of European Marine Sites. The Fal and Helford (candidate) Special area of conservation. Marine Biological Association, Occasional publication, 8, pp. 1-160.

LE GALL, A.-C. & VAN DEN BERG, C.M.G. (1993) Cathodic stripping voltammetry of glutathione in natural waters. Analyst, 118, pp. 1411-1415.

LEAL, M.F.C. & VAN DEN BERG, C.M.G. (1998) Evidence for strong copper(I) complexation by organic ligands in seawater. Aquatic Geochemistry, 4, pp. 49-75.

LEE, J.G., AHNER, B.A. & MOREL, F.M.M. (1996) Export of cadmium and phytochelatin by the marine diatom Thalassiosira weissflogii. Environmental Science & Technology, 30, pp. 1814-1821.

MATRAI, P.A. & VETTER, RD. (1988) ?articulate Thiols in Coastal Waters- the Effect of Light and Nutrients on Their Planktonic Production. Limnology and Oceanography, 33, pp. 624-631.

MATSUKAWA, R., HOTTA, M., MASUDA, Y., CHIHARA, M. & KARUBE, I. (2000) Antioxidants from carbon fixing Chlorella sorokiniana. Journal of Applied Phycology, 12, pp. 263-267.

MOFFETT, J.W., BRAND, L.E., CROOT, P.L. & BARBEAU, K.A. (1997) Cu speciation and cyanobacterial distribution in harbors subject to anthropogenic Cu inputs. Limnology and Oceanography, 42, pp. 789-799.

MORELLI, E. & PRA TESI, E. (1997) Production of phytochelatins in the marine diatom Phaeodactylum tricomutum in response to copper and cadmium exposure. Bulletin of Environmental Contamination and Toxicology, 59, pp. 657-664.

NOCTOR, G. & FOYER, C.H. (1998) Ascorbate and glutathione: Keeping active oxygen under control. Annual Review of Plant Physiology and Plant Molecular Biology, 49, pp. 249-279.

PARSONS, T.R., MAlTA, Y. & LALLI, C.M. (1984) A manual of chemical and biological methods for seawater analysis. Pergamon Press, Oxford.

PAWLIK-SKOWRONSKA, B. (2001) Phytochelatin production in freshwater algae Stigeoclonium in response to heavy metals contained in mining water; effects of some environmental factors. Aquatic Toxicology, 52, pp. 241-249.

PEDERSEN, S.N. & LUNDEBYE, A.K. (1996) Metallothionein and stress protein levels in shore crabs (Carcinus maenas) along a trace metal gradient in the Fal estuary (UK). Marine Environmental Research, 42, pp. 241-246.

PlNTO, E., SIGAUD-KUTNER, T.C.S., LEIT AO, M.A.S., OKAMOTO, O.K., MORSE, D. & COLEPICOLO, P. (2003) Heavy metal-induced oxidative stress in algae. Journal of Phycology, 39, pp. 1008-1018.

143 Chap/er 5 Field measurements ofparticulaie phytochelatins and glutathione

PIRRIE, D., CAMM, G.S., SEAR, L.G. & HUGHES, S.H. (1997) Mineralogical and geochemical signature of mine waste contamination, Tresillian river, Fa! estuary, Cornwall, UK. Environmental Geology, 29, pp. 58-65.

RIJSTENBIL, J.W., DEHAIRS, F., EHRLICH, R. & WIJNHOLDS, J.A. (1998) Effect of the nitrogen status on copper accumulation and pools of metal-binding peptides in the planktonic diatom Thalassiosira pseudonana. Aquatic Toxicology, 42, pp. 187-209.

RIJSTENBIL, J.W., MERKS, A.G.A., PEENE, J., POORTVLIET, T.C.W. & WIJNHOLDS, J.A. (1991) Phytoplankton composition and spatial distribution of copper and zinc in the Fa! Estuary (Cornwall, UK). Hydrobiological Bulletin, 25, pp. 37-44.

RIJSTENBIL, J.W. & WIJNHOLDS, J.A. (1996) HPLC analysis of nonprotein thiols in planktonic diatoms: Pool size, redox state and response to copper and cadmium exposure. Marine Biology, 127, pp. 45-54.

ROSS, A.R.S., IKONOMOU, M.G. & ORIANS, K.J. (2003) Characterization of copper­ complexing ligands in seawater using immobilized copper(Il)-ion affinity chromatography and electrospray ionization mass spectrometry. Marine Chemistry, 83, pp. 47-58.

SUNDA, W. (1988) Trace metal interactions with marine phytoplankton. Biological Oceanography, 6, pp. 411-442.

2 SUNDA, W.G. & HUNTSMAN, S.A. (1998) Interactions among Cu2+, Zn +, and Mn2+ in controlling cellular Mn, Zn, and growth rate in the coastal alga Chlamydomonas. Limnology and Oceanography, 43, pp. 1055-1 064.

TANG, D.G., HUNG, C.C., W ARNKEN, K.W. & SANTSCHI, P.H. (2000) The distribution of biogenic thiols in surface waters of Galveston Bay. Limnology and Oceanography,45,pp. 1289-1297.

WAR WICK, R.M. (2001) Evidence for the effects of metal contamination on the intertidal macrobenthic assemblages of the Fa! Estuary. Marine Pollution Bulletin, 42, pp. 145-148.

WEI, L.P., DONAT, J.R., FONES, G. & AHNER, B.A. (2003) Interactions between Cd, and Cu, and Zn influence particulate phytochelatin concentrations in marine phytoplankton: Laboratory results and preliminary field data. Environmental Science & Technology, 31, pp. 3609-3618.

144 Chapter 6

6. Copper complexation in the Thau Lagoon (France), a shellfish farming area

6.1. Abstract

The quality of the waters in the Mediterranean Thau Lagoon (south of France) is of great concern as this is an important area for oyster farming. The northern portion of the lagoon is subject to metal inputs from urban and industrial activities. Organic ligands can bind metals and thus ameliorate metal toxicity to biota by decreasing the concentrations of bioavailable metals. In this chapter, the implications of Cu complexation in the surface waters of the Thau Lagoon are briefly considered. Copper complexed with organic ligands presented conditional stability constants in the range for the strong ligands (Log KcuL =

12.5 - 13.5). The concentrations of the strong ligands (16.2 - 30.5 nM) were close to those

2 of the total dissolved Cu (13.2 - 25.5 nM), resulting in free Cu + concentrations typical for

11 13 coastal areas (10- - 10- M). Weaker ligands, not detectable by the established conditions, should also play an important role in the metal buffering capacity of the waters in the Thau Lagoon. The oyster farms appeared to influence Cu speciation, as higher

2 concentration of free Cu + was observed in the vicinities of the oyster beds. The effects of the farmed oysters on Cu speciation and the alterations induced in phytoplankton composition by Cu toxicity demand further investigations, as phytoplankton form the main food source for the bivalves.

- 145- Chapter 6 Copper complexation in the Thau Lagoon

6.2. Introduction

The water quality of the Mediterranean Thau Lagoon, in the southern France, is of great concern as this is an important shellfish farming area, particularly for oysters. This area accounts for an estimated standing stock of the oyster Crassostrea gigas of around 40,000 tons, which is one of the highest production in Europe (Dupuy et al. 2000). The lagoon, shown in Fig. 6.1 , is 15 km long, 5 km wide and has an average of 4 m depth. It comprises a catchment area of 280 km2 and is drained by various small streams with intermittent flows (Plus et al. 2003).

/

- Cities c:::::::J Shellfish farms o 1 2 3 km I · • · I

Figure 6.1. Thau Lagoon (France). Sampling sites are indicated by numbers. 1. Sewage discharge; 2. Crique deL' Angle; 3. Pointe de Balaruc; 4. Thau 4; 5. Thau 5; 6. Canal; 7. Middle of the lagoon (C4); 8. Oyster bed (CS).

Most of the studies in the Thau Lagoon are concerned with metal exchanges between the

water column and sediments (Amouroux et al. 2003), the influence of the oyster farming

- 146- Chapter 6 Copper complexation in the Thau Lagoon

activities on the water column (Souchu et al. 2001), the impact of the watershed and interactions with the Mediterranean sea (Plus et al. 2003). The Thau Lagoon is characterised by a lack of tides and sediment resuspension, with the shellfish farming area acting as a sink of particulate matter and source of nutrients (Souchu et al. 200 I). Large interannual variation in precipitation (from 200 to 1000 mm per year) and strong wind

( 118.5 days per year above Beaufort force 5) are important for the lagoon hydrodynamics

(Plus et al. 2003).

The lagoon is subjected to anthropogenic activities and receives inputs of metals, nutrients and organic matter from agriculture (vineyards), fertilizer industries, and urban discharges

(Pena & Picot 1991). A view ofthe oyster farm is shown in Fig. 6.2. Metals do not appear to cause deleterious effects on cultivated bivalves but can accumulate in their flesh, and render them unpalatable and unhealthy for consumption. Depuration of the oysters is a process that can efficiently reduce such metal problems (Laing & Spencer 1997). The composition and physiology of phyto- and zooplankton assemblages, however, can be affected by metal inputs (Moffett et al. 1997). Although many metals, such as copper, are considered trace nutrients, they can be toxic to phytoplankton at elevated concentrations

(Gledhill et al. 1997; Okamoto et al. 2001). Therefore, metal contamination could compromise the bivalve's food source, as the oysters feed mainly by filtering phytoplankton species and organic detritus during all growth stages.

In this respect, dissolved organic matter ubiquitously found in natural waters play a significant role in controlling metal speciation and toxicity to biota. Several studies have shown that Cu and other metals are reversibly bound by dissolved organic ligands

(Kozelka & Bruland 1998; Kogut & Voelker 2003), forming metal-complexes which are less bioavailable to phytoplankton. These organic ligands are thought to consist of a

- 147- Chapter6 Copper complexation in the Thau Lagoon

continuous spectrum of ligands spanning complexing properties with I :I co-ordination

sites to polyfunctional chelators (Gerringa et al. I995). Such organic ligands include fulvic

and humic acids and phytoplankton and bacterial exudates, organic breakdown products of

these organisms, and constituents of sewage effluents (Kozelka & Bruland I998). These

9 11 11 14 ligands are classified as weak (K = I0 - I0 ) and strong (K = I0 - I0 ) , according to the

conditional stability constants of the metal-complexes formed.

Figure 6.2. A view of the oyster farm in the Thau Lagoon (southern France).

The aim of this study was to determine Cu complexation in the Thau Lagoon and nearby the oyster beds using competitive ligand exchange-adsorptive cathodic stripping voltarnmetry (CLE-ACSV). This technique is commonly used to examine the speciation of

Cu in natural waters because of its sensitivity (Moffett et al. I997).

6.3. Theory

A titration technique is employed in order to quantify the free ionic Cu concentration, estimate the system's capacity to complex additional Cu inputs and determine the strength

- I48- Chapter 6 Copper complexation in the Thau Lagoon

of the complexes involved. The titration data is then transfonned by a linearisation method such as van den Berg/Ruzic. For this purpose, the following relationships are used:

[L] = [L'] + [CuL] (6.1)

where L is the ligand that fonn non-labile complexes, L' is the ligand not complexed with

Cu and CuL is the Cu-organic complex. The conditional stability constant for the fonnation of the complex CuL (Kcud is defined as:

2 KcuL = [CuL]/ [Cu +] X [L] (6.2)

A linear relationship is obtained by substitution of L' in equation (6.1) with equation (6.2), providing that Cu complexation is predominantly controlled by a single class of organic ligand (Campos & van den Berg 1994):

([Cu2+]/ [CuL]) = {[Cu2+]/ [L]) + {1/ KcuL x [L]) {6.3)

The titration of a sample with the metal of interest (Cu), and in the presence of the added ligand (for example, salicylaldoxime, SA), yields a series of labile Cu concentrations,

[ Culabile].

[CuL] = [Curotal]- [Culabile] (6.4)

where [Cur0131 ] is the total dissolved Cu concentration in the respective sample aliquot, which is calculated from the total dissolved Cu concentration in the original sample (after

- 149- Chapter 6 Copper complexation in the Thau Lagoon

UV irradiation of the sample) plus de added Cu concentration during the titration. ln the presence of the added ligand SA, [CuTotal] is:

[CuTotaJ] = [Cu'] + [CuSA) + [CuL] (6.5)

where [Cu'] is the inorganic Cu concentration. The [Culabile] is that which equilibrates with

the added ligand (SA), and this is measured by CSV. Combining equations (6.4) and (6.5)

results in:

[Culabile] = [Cu'] + [CuSA] (6.6)

The CSV peak height is related to the [Culabile] via the sensitivity, S (S =peak current I Cu

concentration (nA/nM)).

lp = S X [CUJabile) (6.7)

S is calibrated by standard additions ofCu to the sample. The [CulabiJe] includes [Cu'], as a

small constant of the added Cu remains uncomplexed by SA. [Cu2+] is directly related to

the [Culabile] by a':

(6.8)

a' is the overall a-coefficient, excluding complexation by L (Campos & van den Berg

1994). Substitution for [Cu2+] in equation (6.3) using equation (6.6) results in:

- 150- Chapter 6 Copper complexation in the Thau Lagoon

([Culabile] I [CuL]) = ([Culabile] I [L]) +(a' I ([L] x KcuL) (6.9)

This equation is used for the van den Berg/Ruzic plot as it is [CulabiJe] which is measured

by cathodic stripping voltammetry (CSV) rather than [Cu2+]. a' is obtained from:

a'= a'cu' + UcuSA (6. 10)

2 where a' cu· is the a coefficient for inorganic complexation of Cu + and acuSA is the a coefficient for the complexation ofCu2+ by SA:

acuSA = P'cuSA [SA') ( 6. 11)

where [SA'] is the concentration of SA not complexed by Cu, and p'cusA is the conditional stability constant of CuSA in seawater:

P'cusA = [CuSA) I [Cu2+) [SA') (6.12)

values for acuSA were obtained from Campos & van den Berg (1994).

The values for [Culabile] in equation (6.9) are obtained directly from the CSV peak current as [Culabile] = ipiS (6.7), and the concentration of CuL from (6.4). The sensitivity S is obtained from the linear portion of the titration where all the ligands were saturated; ligand saturation is verified by comparison with the slope obtained from further Cu standard additions to a sample-aliquot in which the Cu-complexing ligands were previously saturated by a Cu increment greater than the ligand concentration. The sensitivities

- 151 - Chapter 6 Copper complexation in the Thau Lagoon

obtained by these two methods should be equal, providing a means to estimate whether the end of the titration has been reached. Values of [L] and KcuL are calculated by linear least­ squares regression from 1/slope and a'/(y-intercept x [L]) of a plot of[Cuiabiic] I [CuL] as a function of [Cu1abiiel Many titrations of natural samples produce a linear van den

Berg/Ruzic plot, suggesting the presence of only one group of ligands, while a curvature would indicate the presence of more than one class of complexing sites (Campos & van den Berg 1994). The range of detectable ligands is restricted by the detection window of the method applied. The detection window is set by the relative magnitudes of the a coefficient of the added ligand (Campos & van den Berg 1994).

6.4. Experimental

6.4.1. Sampling

Surface water samples were collected during spring 2003 in the Thau Lagoon (Fig. 6.1 ), using clean sampling techniques. All samples were collected using Niskin bottles from a depth of around 0.3 m with the use of a small boat. Samples were transferred to acid cleaned 250 mL low density polyethylene bottles (Nalgene). Sample handling and preparation were carried out using a portable laminar flow hood. All samples were filtered using acid clean 0.4 11m polycarbonate filters (Whatman). Samples for the determination of total dissolved Cu were acidified with quartz distilled HCI to pH 2. All samples were double bagged and frozen. Analyses were performed on return to laboratory in Plymouth.

Samples for phytochelatins and glutathione analysis were also collected and filtered as described in chapters 3 and 4. The filters were frozen and transported to Plymouth.

Glutathione and phytochelatins were not observed in any of the samples, probably because

- 152- Chapter 6 Copper complexation in the Thau Lagoon

of degradation of the compounds as the filters were thawing when they arrived m

Plymouth.

6.4.2. Determination of copper complexation in the Thau Lagoon

Determination of total dissolved Cu (CuTotaJ) and Cu speciation was performed usmg adsorptive cathodic stripping voltammetry as described in Campos & van den Berg ( 1994).

A detailed description of the analysis and instrumentation was given in Chapter 5, items

5.4.3 and 5.4.4.

6.4.3. Determination of chl a and salinity in the Thau Lagoon

Determination of chl a in the samples was based on Parsons et al. (1984), as described in

Chapter 5, item 5.4.5. Salinity was determined using a salinometer.

6.5. Results and Discussion

Salinities were between 32.8 and 33.7 and pH values for surface waters were generally 8.2.

Chlorophyll a contents ranged from 2.9 to 7.9 )lg L" 1 and were higher than the literature

1 values for the system(< 2 )lg L" ) (Dupuy et al. 2000). The higher chl a levels found in this study were consistent with the occurrence of dinoflagellate blooms (Alexandrium sp.) during the sampling season. Low chl a contents in the lagoon has been associated with the abundance of picophytoplankton species, which is considered a paradox because of the high growth rates of the oysters (Dupuy et al. 2000).

- 153 - Chapter 6 Copper comp/exation in the Tha11 Lagoon

Ligand titrations were carried out in order to determine the Cu species, natural organic ligands, and the conditional stability constants of the Cu-organic complexes.

Representative Cu-ligand titration and linearisation curves for the Thau Lagoon samples are shown in Fig. 6.3. The curvature at the beginning of the titration curve (Fig. 6.3a) indicates the presence of natural Cu-complexing ligands. The straight line produced by the van den Berg linearisation treatment of the titration data (Fig. 6.3b) indicates that the complexation was controlled by a single class of ligands (Camp os & van den Berg 1994).

a) b)

30 2.5. - ----~l 25 2.0 . ::J ::;; 20 :J .=. u 1.5 . i 15 . :s: D ~ :J .. 1.0 . u" 10 :J // !do 5 0.5. y / 0 0.0 ·. 0 10 20 30 40 50 0 5 10 15 20 25 30 Curotal (nM) Cu,aolle (nM)

Figure 6.3. a) Typical titration curve for the estuarine samples (Sewage discharge, site I, March 2003); b) van den Berg linearization for the titration data. Curotal = Cu in the sample plus Cu added for the titration; CuL = Cu complexed by natural organic ligands; Cu1abile = Cu complexed by SA.

The parameters obtained from the titration and linearisation, together with salinity and eh!

a concentrations are presented in Table 6.1. The highest dissolved Cur01a1 concentration was found in the Canal sample (25.5 nM, site 6), which was probably due to the proximity to the urban and industrial sources of metals. Copper-organic complexes presented high conditional stability constants (Log KcuL = 12.5-13.5), in the range for the strong ligands.

The concentrations of these ligands (16.2-30.5 nM) were close to the concentrations of the dissolved Curotal (13.2-25.5 nM) for most of the sampling sites, approaching a molar ratio

- I 54- Chapter 6 Copper complexalion in the Thau Lagoon

of I : I. This is in accordance to the trends observed in other aquatic systems (Kozelka &

Bruland 1998). Thau 5 and Canal samples (sites 5 and 6, respectively) showed somewhat higher concentrations of ligands compared to the dissolved Curotal concentrations.

Anthropogenic organic matter and run-off of humic substances from the small streams draining the surrounding area could have contributed to the bulk of dissolved organic ligands. Indeed, both extremities of the lagoon (East-West) have been reported to be rich in organic matter that lead to a risk of anoxia during summer (Plus et al. 2003).

Table 6.1. Copper complexation in the Thau Lagoon (19th May 2003). Concentrations of dissolved Curooah total organic ligands (L), Cu2+, Cu', together with the conditional stability constants of the Cu-organic complexes (CuL), salinity (S) and chlorophyll a contents. nM Log M "g L-

Curoral L KcuL Cu2+ Cu' s Chi a Sewage discharge 13.2 ± 0.3 16.2 ± 1.0 13.5 ± 0.5 1.36E-13 1.90E-13 33.3 2.9 ± 0.1 Cri que de L 'Angle 18.6 ± 0.4 20.2 ± 0.9 13.4 ± 0.4 4.32E-13 6.05E-13 33.5 5.4 ± 0.3 Pointe de Balaruc 17.4 ± 0.3 18.3±2.1 12.5 ± 0.2 6.06E-12 8.48E-12 33.3 Middle of lagoon 17.4 ± 0.3 18.9±0.6 12.7 ± 0.1 2.15E-12 3.02E-12 32.8 7.9 ± 0.4 Near oyster bed 17.7 ± 0.4 17.7± 1.0 12.8 ± 0.2 4.71E-11 6.60E-11 33.7 5.2 ± 0.3 Thau4 22.2 ± 0.4 24.4 ± 0.8 13.5 ± 0.4 3.01E-13 4.21E-13 4.8 ± 0.2 Thau5 22.5 ± 0.5 30.5 ± 0.8 12.8±0.1 4.43E-13 6.21E-13 6.3 ± 0.3 Canal 25.5 ± 0.5 29.8 ± 0.8 12.8 ± 0.1 9.99E-13 1.40E-12 33.5 4.3 ± 0.2

2 13 The Cu + concentration range (I o- -I o·ll M) found in this area is typical for coastal waters

2 11 (Giedhill et al. 1997; Kozelka & Bruland 1998). The highest Cu + concentration (I o- M) was observed near the oyster beds (site 8), where the strong organic ligand concentration was close to the dissolved Curatal concentration. A saturation of organic ligands implies

2 that small increases in the dissolved Curatal results in important increases in Cu + and Cu'

2 concentrations. This is further illustrated in Fig. 6.4 with a maximum Cu + concentration for the sample collected near the oyster beds. Nevertheless, the results indicated that Cu

- !55 - Chapter 6 Copper complexation in the Thau Lagoon

2 was complexed by strong ligands (%CuL = 100, %CuL = [Curotal] - [Cu +] - [Cu'] I

[Cur01a1]) x 100) as the major form within the lagoon. Other dissolved organic ligands, not detectable by the established conditions, may have some importance in the metal buffering capacity of the waters. Weaker organic ligands are usually at concentrations higher than the strong ligands (Moffett et al. 1997). These ligands could be detected by varying the detection window of the method (by either using other added ligand or changing the concentration of the added ligand SA) (Campos & van den Berg 1994) .

~'...J- 40 ?.OE- 11 Cl ..: .,... ns 6.0E-11 CD CD ::c 0 u 30 c: "'0 S.OE-11 N c: + ns Ill ---Organic ligands I :I i" 4.0E-11 CL -o- Chlorophyll a .:.. 20 :;· --cu-total ;; 0 0 - 3.0E-1 1 ea --{I- Free Cu2+ 5 Ill 0 :I _.,._ Inorganic Cu' 2.0E-11 i'i' ~ 10 0 c: c:_ ns ~ 1.0E-11 ~ .!:! c: ns 0 ~ O.OE+OO 0> 0 v 10 (ij e> c: -8..9.! ~ ::l ::l c: 0 · ~ 41 g> ..9.! 8 ., ., ., ::l<( "' "' .c. .c. 0 "' .g- :..J :8~ 1- 1- 8, ~- ~:8 ., 0 41 z ~ (/)

Figure 6.4. Copper speciation in the Thau Lagoon, showing effects of the relation CuTotat : Organic ligand.

2 11 Even the low Cu + concentrations (1 o- M) found in this study could cause deleterious effects on the growth rate of some larvae of copepods, cyanobacteria and planktonic ciliates, as indicated by laboratory studies (Stoecker et al. 1986; Moffett et al. 1997). It has been suggested that the dominance of eukaryotic picophytoplankton over prokaryotic picophytoplankton in the Thau Lagoon could be linked to the tolerance to Cu (Vaquer et al.

- 156 - Chapter 6 Copper complexation in the Thau Lagoon

1996). A defense response from most organisms can be expected for chronic or elevated

Cu stress in natural waters. Mechanisms involving the production of intracellular metal­ binding peptides, such as phytochelatins and glutathione (Ahner et al. 1997) and exudation of organic ligands (Croot et al. 2000) have been reported for natural phytoplankton assemblages under metal stress. For instance, phytoplankton species such as the diatoms

Skeletonema costatum and Pseudo-nitzschia, and the dinoflagellate Prorocentrum sp, common in the Thau Lagoon (Dupuy et al. 2000), were able to release metal-complexing ligands under Cu stress (Lage et al. 1996; Rue & Bruland 2001; Croot et al. 2000), with characteristics comparable to the dissolved ligands determined in the present study.

In the present study, the oysters appeared to affect the dissolved Cu speciation in the surface waters of the lagoon. Indeed, natural population of bivalves are known to control phytoplankton composition, reduce total suspended solids through filter feeding, and recycle and remove organic nutrients in the water column (Pietros & Rice 2003). Filter feeders are efficient vehicles of particulate matter from the water column to sediments by filtration coupled with biodeposition, leading to accumulation of organic matter in sediments (Souchu et al. 2001). It is then possible that the alterations caused by the farmed oysters in their surrounding water column were responsible for the increase in the

2 concentrations of free Cu + and inorganic Cu'. It is important to note, however, that the hydrodynamics of the lagoon is highly influenced by wind (Plus et al. 2003), which could have altered the composition of the water column during sampling. Therefore, more studies are required in order to link dissolved Cu species and oyster activities.

A shift of phytoplankton species dominance from one species to another in response to the feeding rate of the oysters is unlikely to cause large variations in the total dissolved ligand

- 157- Chapter 6 Copper complexation in the Thau Lagoon

concentrations, and with consequent changes in Cu complexation. This is supported by results from laboratory experiments which demonstrated that phytoplankton blooms did not impact Cu speciation (Beck et al. 2002). At present, further investigations are required in order to establish how dissolved Cu species in the Thau lagoon could be affected by the farmed oysters and how Cu species could affect the phytoplankton composition.

6.6. Conclusion

This study provided baseline measurements for dissolved Cu species in the Thau Lagoon,

2 an important shellfish farming area. The concentrations of total dissolved Cu and free Cu +

2 in the lagoon were in the range observed for other coastal areas. Higher free Cu + and inorganic Cu' concentrations were observed in the vicinities of the oyster beds in comparison to the other sites. This could imply that the oysters were affecting the dissolved Cu speciation. The relationships between the effects of oysters on Cu speciation and the alterations in phytoplankton composition due to Cu toxicity demand further investigations, as phytoplankton form the main food source for the bivalves.

6.7. References

AHNER, B.A., MOREL, F.M.M. & MOFFETT, J.W. (1997) Trace metal control of phytochelatin production in coastal waters. Limnology and Oceanography, 42, pp. 601- 608.

AMOUROUX, D., MONPERRUS, M., POINT, D., TESSIER, E., BAREILLE, G., DONARD, O.F.X., CHAUVAUD, L., THOUZEAU, G., JEAN, F., GRALL, J., LEYNAERT, A., CLAVIER, J., GUYONNEAUD, R., DURAN, R., GONI, M.S. & CAUMETTE, P. (2003) Transfer of metallic contaminants at the sediment-water interface in a coastal lagoon: Role of the biological and microbial activity. Journal de Physique fv, 107, pp. 41-44.

- 158- Chapter 6 Copper complexation in the Thau Lagoon

BECK, N.G., BRULAND, K.W. & RUE, E.L. (2002) Short-term biogeochemical influence of a diatom bloom on the nutrient and trace metal concentrations in South San Francisco bay microcosm experiments. Estuaries, 25, pp. I 063-1076.

CAMPOS, M.L.A.M. & VAN DEN BERG, C.M.G. (1994) Determination of copper complexation in sea water by cathodic strippng voltammetry and ligand competition with salicylaldoxime. Analytica Chimica Acta, 284, pp. 481-496.

CROOT, P.L., MOFFETT, J.W. & BRAND, L.E. (2000) Production of extracellular Cu complexing Iigands by eucaryotic phytoplankton in response to Cu stress. Limnology and Oceanography, 45, pp. 619-627.

DUPUY, C., VAQUER, A., LAM-HOAI, T., ROUGIER, C., MAZOUNI, N., LAUTIER, J ., COLLOS, Y. & LE GALL, S. (2000) Feeding rate of the oyster Crassostrea gigas in a natural planktonic community of the Mediterranean Thau Lagoon. Marine Ecology­ Progress Series, 205, pp. 171-184.

GERRINGA, L.J.A., HERMAN, P.M.J. & POORTVLIET, T.C.W. (1995) Comparison of the Linear Vandenberg Ruzic Transformation and A Nonlinear Fit of the Langmuir Isotherm Applied to Cu Speciation Data in the Estuarine Environment. Marine Chemislly, 48, pp. 131-142.

GLEDHILL, M., NIMMO, M., HILL, S.J. & BROWN, M.T. (1997) The toxicity of copper(II) species to marine algae, with particular reference to macroalgae. Journal of Phycology, 33, pp. 2-11.

KOGUT, M.B. & VOELKER, B.M. (2003) Kinetically inert Cu m coastal waters. Environmental Science & Technology, 37, pp. 509-518.

KOZELKA, P.B. & BRULAND, K.W. (1998) Chemical speciation of dissolved Cu, Zn, Cd, Pb in Narragansett Bay, Rhode Island. Marine Chemistry, 60, pp. 267-282.

LAGE, O.M., PARENTE, AM., VASCONCELOS, M.T.S.D., GOMES, C.A.R. & SALEMA, R. ( 1996) Potential tolerance mechanisms of Prorocentrum micans (dinophyceae) to sublethal levels of copper. Journal of Phycology, 32, pp. 416-423.

LANG, I. & SPENCER, B. E. Bivalve cultivation: criteria for selecting a site. 1-43. 1997. Lowestoft, The Centre for Environment, Fisheries & Aquaculture Science.

MOFFETT, J.W., BRAND, L.E., CROOT, P.L. & BARBEAU, K.A. (1997) Cu speciation and cyanobacterial distribution in harbors subject to anthropogenic Cu inputs. Limnology and Oceanography, 42, pp. 789-799.

OKAMOTO, O.K., PINTO, E., LATORRE, L.R., BECHARA, E.J.H. & COLEPICOLO, P. (2001) Antioxidant modulation in response to metal-induced oxidative stress in algal chloroplasts. Archives of Environmental Contamination and Toxicology, 40, pp. 18-24.

PARSONS, T.R., MAlTA, Y. & LALLI, C.M. (1984) A manual of chemical and biological methods for seawater analysis. Pergamon Press, Oxford.

PENA, G. & PICOT, B. (1991) Trace-metals in Thau Lagoon sediments (French Mediterranean Coast). Oceanologica Acta, 14, pp. 459-472.

- !59- Chapter 6 Copper complexation in the Thau Lagoon

PIETROS, J.M. & RICE, M.A. (2003) The impacts of aquacultured oysters, Crassostrea virginica (Gmelin, 1791) on water column nitrogen and sedimentation: results of a mesocosm study. Aquaculture, 220, pp. 407-422.

PLUS, M., CHAPELLE, A., LAZURE, P., AUBY, 1., LEVAVASSEUR, G., VERLAQUE, M., BELSHER, T., DESLOUS-PAOLI, J.M., ZALDIVAR, J.M. & MURRA Y, C.N. (2003) Modelling of oxygen and nitrogen cycling as a function of macrophyte community in the Thau lagoon. Continental Shelf Research, 23, pp. 1877- 1898.

RUE, E. & BRULAND, K. (2001) Domoic acid binds iron and copper: a possible role for the toxin produced by the marine diatom Pseudo-nitzschia. Marine Chemistry, 76, pp. 127- 134.

SOUCHU, P., VAQUER, A., COLLOS, Y., LANDREIN, S., DESLOUS-PAOLI, J.M. & BIBENT, B. (2001) Influence of shellfish farming activities on the biogeochemical composition of the water column in the Thau Lagoon. Marine Ecology-Progress Series, 218, pp. 141-152.

STOECKER, D.K., SUNDA, W.G. & DAVIS, L.H. (1986) Effects of Copper and Zinc on 2 Planktonic Ciliates. Marine Biology, 92, pp. 21-29.

V AQUER, A., TROUSSELLIER, M., COURTIES, C. & BIBENT, B. (1996) Standing stock and dynamics of picophytoplankton in the Thau Lagoon (northwest Mediterranean coast). Limnology and Oceanography, 41, pp. 1821-1828.

- 160- Chapter 7

7. Conclusions and Future Work

This research contributed to establish the first baseline levels for the production of the particulate metal-binding peptides phytochelatins and glutathione in European estuaries.

The production of phytochelatins by phytoplankton showed to be a general indication of metal stress in the Fa! and Tamar estuaries (southwest of the UK). Particulate phytochelatins were present at higher concentrations in Restronguet Creek (Fa! Estuary) and Gunnislake (Tamar Estuary), the most metal affected sites. For the 2002 survey in the

Fa! Estuary, it could be observed that the production of phytochelatins was a faster response than the production of glutathione for same increases in the total dissolved Cu concentrations. This is an indication that the production of phytochelatins IS a more specific response to metal stress in the field than the production of glutathione.

The estuarine conditions that cause variations m the concentrations of thiols, as exemplified by the data set presented here, require further investigation. There is still a paucity of laboratory and field data addressing this issue. The short-term metal exposure experiments using the diatom culture Phaeodactylum tricornutum indicated that phytochelatin production is influenced by metal combinations, probably due to competition among metals for cellular binding sites. Effects on phytochelatin production can be expected, as a mixture of metals such as Cd, Zn and Cu are usually found in contaminated waters. At the present stage, a characterisation of the study area, with the determination of a range of parameters, including eh! a content, phytoplankton species composition.

- I 61 - Chapter 7 Conclusions and Fmure Work nutrients, metals and organtc xenobiotics, ts needed for a full interpretation of the

production ofthiol compounds.

In addition to field work, laboratory metal exposure experiments usmg natural phytoplankton assemblages or representative species will be helpful for the control and assessment of the variables that influence the production of thiols. Further investigations on the effects of a mixture of contaminants on representative phytoplankton species, particularly species isolated from Restronguet Creek and Gunnislake, could provide a better understanding on the chronic stress response and the variations in phytochelatin and glutathione production.

It has been acknowledged that the measurements of biological responses of metal stress provide a better indication of the environmental quality than the measurements of total dissolved metals or even metal speciation data. This is explained by the fact that adverse or toxic effects produced by a chemical on a biological system do not manifest, unless that chemical or its biotransformation products reach specific sites in the organism, at a certain concentration and for a sufficient length of time. There is a vast field to be explored regarding the biological responses caused by a mixture of contaminants in estuarine systems. Work on these responses would then further our understanding on how organisms such as phytoplankton cope with a mixture of stressors, and consequently would help the assessment of the environmental quality.

The current methods for the simultaneous determination of phytochelatins and glutathione comprise multi-step time-consuming protocols that pose constraints to routine analyses.

This is a relevant aspect to take into account due to the large sets of samples required for environmental studies/monitoring. Future work on the development of cost-effective, rapid

- 162 - Chapter 7 Conclusions and Future Work and sensitive analytical methods is a crucial aspect to further the applications of phytochelatins in field assessment. Work on the application of a pre-concentration step and an alternative derivatising reagent for the analysis of particulate PCs, which provides less interfering compounds, is needed in order to improve the sensitivity of the analytical method.

The study undertaken in the Thau Lagoon provided baseline levels for dissolved Cu species. Previous data on dissolved metal speciation and metal complexing ligands were unavailable for the system during this investigation. The present results indicated that the oyster farm could be affecting the dissolved Cu species in the surface waters. The relationships between the effects of oysters on Cu speciation and the alterations in phytoplankton composition due to Cu toxicity demand further investigations.

Understanding the factors that influence phytoplankton composition is an important aspect in the Thau Lagoon system, as phytoplankton form the main food source for the farmed bivalves.

7.1. The use of phytochelatins in environmental quality assessment

The research requirements to assess the environmental quality of the Fal Estuary and

Plymouth Sound were recently reviewed by Langston et al. (2003a, b), and included the need for the determination of metal stress indicators. The use of phytochelatins as a metal stress indication for environmental monitoring has not been applied yet. This is mainly because biological responses to toxicants in dynamic aquatic systems, such as estuaries, are difficult to interpret and a stress indication can be masked, as observed for phytochelatin

- 163- Chapter 7 Conclusions and Fu111re Work production during some field situations. Therefore, as highlighted above and in Chapters 2 and 5, the use of a suite of stress indicators is believed to be the best approach to assess biological impacts in individuals. In this respect, the determination of particulate phytochelatins can represent an additional tool to provide a stress indication for mixtures of toxic metals.

The general recommendations for the use of phytochelatins m environmental quality assessment can be summarised:

I. As a first step for a feasible application of phytochelatins in field monitoring,

improvements to the analytical method for phytochelatins are needed (emphasised

above and in Chapter 3);

2. An adequate sampling strategy is necessary for screening the system for potential

sources of contaminants to allow identification of the most affected areas. This can

be then followed by a refined monitoring of the system, with the determination of

particulate phytochelatins as a metal stress indication;

3. Elevated particulate phytochelatin concentrations indicate where dissolved metals

are more bioavailable for most of the phytoplankton species (and/or where

dissolved organic ligands are saturated). It is important to take into account the

limitations of this approach, as some phytoplankton species do not employ

phytochelatin production as a metal detoxification mechanism. This could provide

a better understanding on the extent of the major impacts on phytoplankton and the

phytoplankton capacity to cope with metal stress.

- 164- Word count: 37,987 words