Marine organisms rely on chemical cues - most as scents - to obtain information (i.e. infochemicals) about their environment. Volatile organic compounds (VOCs) from plants represent one group of infochemicals that can shape ecological interactions and the structure of the ecosystems through the influence of behaviour of receivers. However, there is still little evidence about the ecological importance of VOCs on the structure and interactions of mesograzers community within ecosystems like the Mediterranean seagrass Posidonia oceanica.
The studies presented in this thesis were designed, using behavioural choice assays, to investigate the effects of epiphytes-borne VOCs from Posidonia oceanica on the associated benthic invertebrates and further clarify how their responses to these putative infochemicals could affect the spatial organization of this stable community. The objectives of this thesis were: 1) to standardize a behavioural choice method for benthic invertebrates (e.g. static chambers vs. flumes and the determination of the species-specific minimum number of replicates); 2) to define the existence of associative behavioural patterns of invertebrates (i.e. molluscs and decapods) to epiphyte-borne VOCs at the community level; 3) to identify the roles of VOCs as infochemicals (e.g. food or toxins cues); and 4) to investigate the behavioural responses to infochemicals within the range of pH levels predicted for the end of this century. Species-specific behaviours of mesograzers to VOCs depended on the method of investigation, the concentration of VOCs tested, body constraints and animal ecology.
Overall, the contribution of this thesis to knowledge is that, within ecosystems like the seagrass P. oceanica, co-evolutionary patterns of infochemicals create microhabitats and the coexistence of herbivores on a single plant leaf. These findings support a better understanding of the entire ecosystem for further coastal management also in prevision of near-future climate changes.
TABLE OF CONTENTS
ABSTRACT
ACKNOWLEDGEMENTS
LIST OF FIGURES
LIST OF TABLES
CHAPTER 1
1 Introduction
1.1 Overview
1.2 The general context of marine chemical ecology
1.3 The current knowledge on infochemicals
1.4 Volatile Organic Compounds from marine plants and their functions as infochemical
1.5 The method: behavioural choices in the investigation of infochemicals
1.6 Behavioural traits and mechanisms of chemoreception in marine invertebrates
1.7 The seagrass Posidonia oceanica, its epiphytes and associated invertebrates
1.8 Epiphytes-herbivores interactions: the role of infochemicals within Posidonia oceanica ecosystem27
1.9 Research aims and objectives
CHAPTER 2 VOCs of epiphytes from Posidonia oceanica trigger infochemicals based communication for benthic inverterbrates
2.1 INTRODUCTION
2.2 MATERIALS AND METHODS
2.2.1 Isolation and culture of benthic epiphytes from the seagrass Posidonia oceanica
2.2.2 Extraction of algal odours: VOCs
2.2.3 Study organisms: animal sampling and biology
2.2.4 Experiment 1: choice assay in static chambers
2.2.5 Experiment 2: choice assay in straight flow-through flume
2.2.6 Baseline choice assays
2.2.7 Distribution of VOCs in static chambers
2.2.8 Distribution of VOCs in straight flow-through flume
2.3 STATISTICAL ANALYSIS
2.3.1 Standardization of behavioural data
2.3.2 Effect of time in behavioural responses of invertebrates
2.3.3 Recognition of algal odours and associative patterns
2.3.4 Distribution of odours in static chambers and in straight flow-through flumes
2.4 RESULTS
2.4.1 Kriging analysis for the diffusion in static chambers
2.4.2 Kriging analysis for the flow in straight flow-through flume
2.4.3 Behavioural responses of invertebrates to VOCs in static chambers
2.4.4 Behavioural responses of invertebrates to VOCs in flumes
2.4.5 Behavioural associative patterns to VOCs
2.5 DISCUSSION
2.6 CONCLUSION
CHAPTER 3 Epiphyte-borne infochemicals rule the community structure of mesograzers in
seagrasses
3.1 INTRODUCTION
3.2 MATERIALS AND METHODS
3.2.1 Study organisms: animal sampling and biology
3.2.2 Production of epiphytes: seaweeds and diatoms
3.2.3 VOCs extraction
3.2.4 Gel preparation and concentrations
3.2.5 Behavioural choice tests
3.2.6 Standardization of behavioural choice test: controls
3.3 STATISTICAL ANALYSIS
3.3.1 Standardization of controls
3.3.2 Behavioural choices to VOCs, effect of concentration and time of exposure
3.3.3 Multivariate analysis: taxonomy vs. environmental hypothesis
3.4 RESULTS
3.4.1 Control experiments: the need for standardization of behavioural data
3.4.2 Responses to VOCs, concentrations and time of exposure for each mesograzer
3.4.3 Responses of the community of mesograzers to infochemicals
3.5 DISCUSSION
3.5.1 The mesograzers’ standardized behaviour
3.5.2 Volatile epiphyte-bome infochemicals structure mesograzers’ microhabitat selection upon the leaves of Posidonia oceanica
3.6 CONCLUSION
CHAPTER 4 Decoding the messages beyond epiphyte-borne volatiles: food kairomones, alarm and habitat cues
4.1 INTRODUCTION
4.2 MATERIALS AND METHODS
4.2.1 Animal collection
4.2.3 Preparation of artificial foods
4.2.4 Feeding test and survival
4.2.5 Statistical analysis
4.3 RESULTS
4.3.1 Putative toxicity
4.3.2 Feeding preferences
4.4 DISCUSSION
4.5 CONCLUSION
CHAPTER 5 Ocean acidification effects on the behaviour of mesograzers to epiphyte-borne
infochemicals
5.1 INTRODUCTION
5.2 MATERIALS AND METHODS
5.2.3 Animal collection and epiphyte cultures
5.2.4 Simulated ocean acidification treatment
5.2.5 Experimental design and behavioural choice tests
5.2.6 VOCs extractions and gels preparations
5.2.7 Statistical analysis
5.3 RESULTS
5.3.1 Responses of molluscs to volatile infochemicals at different pH
5.3.2 Responses of decapods to volatile infochemicals at different pH
5.3.3 Reactions of mesograzers’ community to infochemicals at different pH
5.4 DISCUSSION
5.5 CONCLUSION
CHAPTER 6 Discussion and Conclusion
6.1 General discussion
6.2 Conclusions and future directions
6.2.1 Epiphyte-odours are infochemicals for benthic invertebrates
6.2.2 Epiphyte-borne infochemicals structure the benthic community within the Posidonia oceanica ecosystem
6.2.3 OA disrupts behaviour to infochemicals and indirectly the structure of ecosystems
Appendix 1_Results on Full Factorial ANOVA for Chapter
Appendix 2_Table of pH measurements within Petri-dishes for Chapter 5
REFERENCES
ABSTRACT
Marine organisms rely on chemical cues- most as scents- to obtain information (i.e. infochemicals) about their environment. Volatile organic compounds (VOCs) from plants represent one group of infochemicals that can shape ecological interactions and the structure of the ecosystems through the influence of behaviour of receivers. However, there is still little evidence about the ecological importance of VOCs on the structure and interactions of mesograzers community within ecosystems like the Mediterranean seagrass Posidonia oceanica.
The studies presented in this thesis were designed, using behavioural choice assays, to investigate the effects of epiphytes-borne VOCs from Posidonia oceanica on the associated benthic invertebrates and further clarify how their responses to these putative infochemicals could affect the spatial organization of this stable community. The objectives of this thesis were: 1) to standardize a behavioural choice method for benthic invertebrates (e.g. static chambers vs. flumes and the determination of the species-specific minimum number of replicates); 2) to define the existence of associative behavioural patterns of invertebrates (i.e. mollusc and decapods) to epiphyte-borne VOCs at the community level; 3) to identify the roles of VOCs as infochemicals (e.g. food or toxins cues); and 4) to investigate the behavioural responses to infochemicals within the range of pH levels predicted for the end of this century. Species-specific behaviours of mesograzers to VOCs depended on the method of investigation, the concentration of VOCs tested, body constraints and animal ecology. Behaviours of mesograzers appeared more coherent with the concentrations of VOCs within the static chambers compared with those recorded in flumes. When sympatric from Posidonia oceanica and generalist epiphytes were considered, mesograzers showed to fine-tune their behaviours to VOCs according to their ecology responding to infochemicals faced in their own microhabitat like the leaf axis of P. oceanica. However, at lower pH the invertebrates switched their behavioural patterns across the VOC bouquets replacing their “natural ecological preferences” with “taxonomic preferences”, leading to a simplification of chemical relationships within P. oceanica. ecosystem.
Overall, the contribution of this thesis to knowledge is that, within ecosystems like the seagrass P. oceanica, co-evolutionary patterns of infochemicals create microhabitats and the coexistence of herbivores on a single plant leaf. These findings support a better understanding of the entire ecosystem for further coastal management also in prevision of near-future climate changes.
ACKNOWLEDGEMENTS
I would like to thank the Open University and Stazione Zoologica Anton Dohm for giving me the opportunity and fellowship to pursue my Ph.D. In particular, I would like to thank my thesis defense committee members, Dr. Mark Briffa, Dr. Paolo Sordino, and Dr. Christophe Brunet for their brilliant comments, questions and suggestions, which made my defense one of the best moment of my Ph.D.
I would like to thank my supervisors for their advice and support during this project.
Thanks Valerio, for conceding me the great freedom in carrying out the research and to have shared throughout these years your knowledge on the maintenance of animals and algae under laboratory conditions.
Thanks Eric, although the kilometers of distance, you have been always present with your constructive criticism and comments, which helped me remaining focused during the writing-up of this thesis. I will never forget the way that you listened to me, and your pat on my shoulder.
Thanks Elisabetta for your kind encouragements over these years. I am thankful for your way of appreciating the smallest of my results.
Special thanks go to all the scientists and students with whom I had the luck to collaborate with during my studies: Dr. Patrick Fink (University of Cologne) because his expertise on volatiles, their extraction, his constant and kind presence and his precious suggestions, as only a supervisor does, have been much appreciated.
Dr. Maria Cristina Buia (SZN) for her invaluable suggestions on epiphytes and Posidonia oceanica; Lucia Porzio (Università ‘Federico II’ di Napoli) for all the times I asked her to help with the identification of epiphytes. She always helped me with a smile! Dr. Francesco Paolo Patti (SZN) and Antonia Chiarore (Università ‘Federico II’ di Napoli) for their help in the identification of Molluscs and their suggestions; to Dr. Maria Cristina
Gambi (SZN) and Alexia Massa Gallucci (SZN) for the taxonomy identification and help with Polychaetes; to Dr. Giovanna Romano (SZN) for her kindness of human being and precious help with the decadienal and sea urchins experiments; to Prof. Michele Scardi (Università ‘Tor Vergata’ di Roma) for the long afternoon giving me suggestions on statistics and graphs.
Very special thanks go to my previous supervisor, Sam Dupont (The Sven Lovén Centre for Marine Sciences, University of Gothenburg) and my best lab mates of ever and now friends, Narimane Dorey (The Sven Lovén Centre for Marine Sciences, University of Gothenburg), Triranta Sircar (University of Gothenburg), Macarena S. Valiñas (Universidad Nacional de Mar del Plata, Argentina), Isabel Casties, Meike Stumpp and Marian Yong-An Hu (IFM-GEOMAR, Kiel, Germany), who taught me all the tips on how to work with pH and larvae. I will never forget the long months of the experiment and our funny talks on The Walking Dead and of course, sometimes science, mackerel season and barbecue nights in Kristineberg.
Next, it will be a long list! I would like to thank all the members of Stazione Zoologica of Ischia! Coffee breaks (those rare times- that all of you know - I came for cakes) and lunchtime were the most important parts of my long days in lab.
The most special thank goes to Rosanna Messina. You gave me “life” advice, reminding me to eat and drink and to take care of my health. I miss already the internal calls for “coffee” breaks: <<Manu, come down, today there is cake! >>, or the glass of water filled up by you... What else I could say. You have been my adoptive mom throughout these years in Ischia-thank you! Thanks Amit Kumar, for our long days spent in lab and our way walking back home, talking about science problems and solutions and giving each other support. Only a good friend does that, so thank you!
Thanks to all my other lab mates: Antonia Chiarore, Giulia Valvassori, Sara Fioretti, Emanuela Di Meglio, Nuria Teixido for our girls’ chatting; Martina Mulas for our long talks about everything and especially on rock music bands. Our first concert together was awesome! To Daniele Grech for the wild spring fennel gift harvested on mount Epomeo during his Sunday walks. To Maria Cristina Buia for her amazing way to say always what she thinks. I liked her from the first welcoming moment! To Maurizio Lorenti for his constant presence and immediate help when exhausted autoclave and pHmeter fought against me. To Anna Rando for her way to remind at eleven o’clock: <<CAFFÉÉÉ!>> and to see life like a mom; To “The Captain” Vincenzo Rando for his funny whistling way and, of course, for his tasty Tarallini pugliesi always ready on the boat after sampling. To Bruno Iacono for his very analytical way to describe people.
The sounds of your voice and laughs, the smell of ready hot coffee in the kitchen every day, made me feeling like home and you as my adoptive family. So, just thank you!
Last but not least, I would thank my family and the two Francesco making me being who I am. To my dad and mom, because they thought me what Freedom is. I will always say what I think and I will always follow my heart. To my tall brother to be my anchor in life shaking my nerd brain in his rude way every time I need, and to my short sister in law for being the sister I have never had. To my first Francesco, my almost four years old nephew who every time reminds me what the unconditional love is!
And thanks to my second unexpected love, Francesco, for his patience, for all the times I said: <<I cannot, I have to work! Go if you want>>. He never went. Thanks to him, that makes me smiling to life every day, reminding me that something beautiful always might be there. Right there, in the front spot during a very annoying seminar.
LIST OF FIGURES
CHAPTER 1
Figure 1.1 The Venn diagram illustrates chemical ecology as an integrative science at the intersection of the four main approaches of chemistry, ethology, neuroscience, and ecology. Figure from Derby and Sorensen, (2008).
Figure 1.2 Chemical-ecological interactions driven by chemical defences and infochemicals produced by higher animals and primary producers in freshwater and marine ecosystems. Figure from Brönmark and Hansson, (2012).
Figure 1.3 The terminology of infochemicals given according to Dicke and Sabelis (1988). E is the Emitter and R is the Receiver, the symbol + indicates which one between emitter and/or receiver benefits from the chemical message.
Figure 1.4 The planktonic crustacean Daphnia sp. shows several defence strategies against predators’ chemical cues. (A) Scanning electron micrographs showing morphological defences against several predators in various Daphnia species (Photograph by C. Laforsch, from Chemical ecology in aquatic systems. Brönmark & Hansson eds. 2012). (B) Reproduction of Daphnia showed in the absence of predators (middle panel) and defence strategies with predators. In the left panel, behavioural (i.e. diel vertical migrations) and physiological changes (e.g. earlier sexual maturity at a smaller size with increased number of offspring of reduced size) are present in response to fish predation. In the right panel are showed morphological changes (e.g. spines in the neck) and increased body size but decreased number of offspring in the presence of the Chaoborus larvae as predators (Illustration by L. Weiss, from Chemical ecology in aquatic systems. Brönmark & Hansson eds. 2012).
Figure 1.5 (a) Dynamic web of interactions and constraints driven by infochemicals (Figure modified from Klaschka, 2008); (b) examples of infochemical relationships within terrestrial ecosystem with specific insects-plant interactions and (c) with a generalist herbivore (Figure modified from Ueda et al. 2012).
Figure 1.6 (a) Schematic description of the planktonic use of dimethyl sulphide (DMS) that is a foraging kairomone attracting birds to patches of food (Figure modified from Steinke et al. 2002). (b) Schematic description about the tri-trophic interaction described in Coleman et al. (2007a). Point 1 indicates the herbivore Littorina obtusata feeding on the brown algae Ascophyllum nodosum in a rocky pool. Point 2 indicates the release of VOCs from grazed A. nodosum, acting as SOS signal that attracts as foraging infochemical the predator green crab Carcinus maenas that in point 3 “rescues” the algae by eating the herbivore.
Figure 1.7 Photo of a meadow of Posidonia oceanica (L.) Delile with visible epiphytes on the leaves (left) and on the right (d) an image of a shoot of P.oceanica with the rhizome covered by scales (remaining of the old leaf bases) showing also (a) adult leaf, (b) intermediate leaf and (c) juvenile leaf. (Image modified from Buia et al. 2004).
Figure 1.8 Illustration of the spatial succession of algal epiphytes over a leaf axis of Posidonia oceanica that shows the increase in species and abundance from the basal towards the apical zone. (Figure modified from Mazzella et al. 1994 (eds.), Le praterie sommerse del Mediterraneo).
Figure 1.9 The diagram shows the complex trophic relationships within a meadow of Posidonia oceanica against the ‘seagrass grazing paradigm’ (Figure from Mazzella and Zupo, 1995).
Figure 1.10 Example of some mesograzers associated to Posidonia oceanica: (left) the shrimp Hippolyte inermis, (centre) the hermit crab Cestopagurus timidus, and (right) the mollusc Bittium latreillii.
CHAPTER 2
Figure 2.1 The sampling site in Lacco Ameno (40°45’ N/13°50’ E) at the north east of Ischia Island (Tyrrhenian Sea, Italy) is indicated by a grey square.
Figure 2.2 Experimental arena used in static choice experiments indicating (a): different sectors (from Jüttner et al. 2010); (b) Photo of the experimental setup (April, 2013).where each arena was positioned with the (+) targets opposite, to randomize directional effects.
Figure 2.3 Straight flow-through flume system designed according to Atema et al. (2002): (A) project in 3D (gently courtesy from M. Mutalipassi). Overall, the chamber size was 21 cm x 14 cm x 10 cm (length x width x height) in overshadowing glass, 4 mm thick. The arrows indicate the test area and the two inflow compartments. (B) The scheme shows the choice flume system from the upside view with: c) water inflow compartments (with arrows); d) packed drink straws to reduce the turbulence; e) barrier-separated channels to further laminate flow; f) upstream fine mesh (0.5 mm) net to contain animals; g) test area; h) starting point for the acclimation; i) downstream containment net and j) drain area with the central drain opening.
Figure 2.4 Flumes are visualized here in the (a) upside and (b) lateral view for photographic purposes using 0.25 g L"1 of Methylen blu dye dissolved in ambient seawater.
Figure 2.5 Kriging reconstruction of the distribution of the 2-trans-4-trans-decadienal volatile compound at (a): 5; (b): 10; (c): 15; and (d): 20 min after the start of the experiment. Isolines indicate areas with identical concentrations of this volatile organic compound according to the intensity scale on the right (pg mL-1) while the scales on the left and at the bottom indicate the diameter of the static choice arena (cm).
Figure 2.6 Kriging reconstruction of the distribution of Methylen blue dye at (a): 3; (b): 4; (c): 5 min after the start of the experiment. Isolines indicate areas with identical concentrations of the dye according to the intensity scale on the right (pg mL-1) while the scales on the left and at the bottom indicate the dimension of the flume chambers 21 cm x 14 cm (length x width).
Figure 2.7 Mean (± SE) of the score weight preferences for the 9 species of invertebrates at the different concentrations of VOC bouquets of (a) Enteromorpha prolifera, (b) Colaconema daviesii and (c) Cyanobacteria in static chambers. Only the significant choices from one sample t-test analysis (p<0.05) compared to baseline controls are indicated with respective score values signed on the bars while above in the text also the 95% confident interval of the difference is indicated.
Figure 2.8 Mean (± SE) of the score weight preferences for the 6 species of invertebrates at the different concentrations of VOC bouquets of (a) Enteromorpha prolifera, (b) Colaconema daviesii and (c) Cyanobacteria in flumes. Only the significant choices from one sample t-test analysis (p<0.05) compared to baseline controls are indicated with respective values on the bars.
Figure 2.9 nMDS ordination plot of behavioural patterns of mesograzers considering all the three VOCs and concentrations tested in static and flumes experiments with controls. No overall differences between the two methods tested are visible (Stress: 0.14).
Figure 2.10 nMDS ordinations of behavioural patterns to all VOCs (indicated by Pearson’s correlation vectors in bold: Ent = Enteromorpha prolifera, Col = Colaconema daviesii, Cya = Cyanobacteria) for each method of investigation, taxonomic group, and concentration tested. nMDS plots for: (a) static chambers considering the taxonomic groups; (b) flumes considering the taxonomic groups; (c) static chambers and (d) flumes considering also the concentrations of VOCs tested. The plots are based on Euclidean distance on normalized data for all the replicates after removing controls.
CHAPTER 3
Figure 3.1 The variation of differences (■) is expressed as the mean integer index (Jüttner et al. 2010) obtained from the four time end-points for each invertebrate species plotted against the total number of replicates performed in controls and with the best-fit curve with nonlinear regression (solid line). The Plateau (indicated in the graph as dashed lines and an arrow showing the 95% of confidence band of the best-fit curve) at the significance level of α=0.05 represents the minimum number of replicates in order to compare real choices under VOCs treatment experiments with the random walking behaviour for each species.
Figure 3.2 Total number of individuals (± S.E.) within each species of (A) molluscs: Alvania lineata, Rissoa italiensis and Bittium latreillii and (B) decapods: Hippolyte inermis, Cestopagurus timidus and Clibanarius erythropus across the three sectors of choice (Towards, Zero and Away) during the time intervals (5, 10, 15 and 20 min) of the control standardization experiment.
Figure 3.3 The mean choice index (± SE) for the sea snail Alvania lineata after 10 and 20 min of exposure to VOC bouquets of different concentrations (Low, Medium and High) compared to controls. Ctl as controls (n=30); Ent as Enteromorpha prolifera; Dic as Dictyota implexa var. linearis; CN as Diploneis sp. and C2 as Cocconeis scutellum var.parva, (with standardized sample n= 20). Asterisks indicate significant differences by the 4-cells Z-test of proportions between VOC treatments and controls.
Figure 3.4 The mean choice index (± SE) for the sea snail Rissoa italiensis after 10 and 20 min of exposure to VOC bouquets of different concentrations (Low, Medium and High) compared to controls. Ctl as controls (n= 27); Ent as Enteromorpha prolifera; Dic as Dictyota implexa var. linearis; CN as Diploneis sp. and C2 as Cocconeis scutellum var.parva, (with standardized sample n= 11). Asterisks indicate significant differences by the 4-cells Z-test of proportions between VOC treatments and controls.
Figure 3.5 The mean choice index (± SE) for the sea snail Bittium latreillii after 10 and 20 min of exposure to VOC bouquets of different concentrations (Low, Medium and High) compared to controls. Ctl as controls (n=30); Ent as Enteromorpha prolifera; Dic as Dictyota implexa var. linearis; CN as Diploneis sp. and C2 as Cocconeis scutellum var.parva, (with standardized sample n= 18). Asterisks indicate significant differences by the 4-cells Z-test of proportions between VOC treatments and controls.
Figure 3.6 The mean choice index (± SE) for the hermit crab Clibanarius erythropus after 10 and 20 min of exposure to VOC bouquets of different concentrations (Low, Medium and High) compared to controls. Ctl as controls (n=30); Ent as Enteromorpha prolifera; Dic as Dictyota implexa var. linearis; CN as Diploneis sp. and C2 as Cocconeis scutellum var.parva, (with standardized sample n= 24). Asterisks indicate significant differences by the 4-cells Z-test of proportions between VOC treatments and controls.
Figure 3.7 The mean choice index (± SE) for the hermit crab Cestopagurus timidus after 10 and 20 min of exposure to VOC bouquets of different concentrations (Low, Medium and High) compared to controls. Ctl as controls (n=30); Ent as Enteromorpha prolifera; Dic as Dictyota implexa var. linearis; CN as Diploneis sp. and C2 as Cocconeis scutellum var.parva, (with standardized sample n= 17). Asterisks indicate significant differences by the 4-cells Z-test of proportions between VOC treatments and controls.
Figure 3.8 The mean choice index (± SE) for the shrimp Hippolyte inermis after 10 and 20 min of exposure to VOC bouquets of different concentrations (Low, Medium and High) compared to controls. Ctl as controls (n=30); Ent as Enteromorpha prolifera; Dic as Dictyota implexa var. linearis; CN as Diploneis sp. and C2 as Cocconeis scutellum var. parva, (with standardized sample n= 13). Asterisks indicate significant differences with the 4-cells Z-test of proportions between VOC treatments and controls.
Figure 3.9 (A) Non-metric multidimensional scaling (nMDS) ordination bi-plot on the basis of Euclidean distances and Pearson’s correlations for the four VOCs with the cross factors habitat of mesograzers long over the leaf of P. oceanica according to the division of the main zones in the leaf axis - apical, intermediate and basal zones- proposed by Mazzella et al. (1994) and the feeding guilds for those species proposed by Gambi et al. (1992) indicated as HeDF = herbivore deposit feeders; Om= omnivorous; DF= deposit feeders; He= herbivores. (nMDS stress: 0); (B) Group-average clustering on the Euclidean species dissimilarities from the mean of concentrations of the choice index data for each volatiles. The nMDS bi-plot and the group-average cluster are based on the averaged raw data of choice indices at 20 minutes of exposure.
CHAPTER 4
Figure 4.1 Mean of survival in percentage for the six mesograzers under the different food treatments (a) control agar, (b) Enteromorpha prolifera, (c) Dictyota implexa var. linearis, (d) Diploneis sp., (e) Cocconeis scutellum var. parva. (Mean ± Standard Error, n=7). The asterisks indicate significant differences at α = 0.05, from Kruskal-Wallis pairwise comparisons test.
Figure 4.2 Percentage of the total consumption of agar-based foods in feeding assays of decapods (a) Clibanarius erythropus, (b) Cestopagurus timidus and (c) Hippolyte inermis. Treatments were compared by Welch’s one-way ANOVA or Kruskal-Wallis test. Error bars represent standard error and asterisks above bars denote significant differences from Games-Howell or Pairwise comparisons post hoc tests at α = 0.05.
Figure 4.3 Percentage of the total consumption of agar-based foods in feeding assays of molluscs (a) Bittium latreillii, (b) Alvania lineata and (c) Rissoa italiensis. Treatments were compared by Welch’s one-way ANOVA or Kruskal-Wallis test. Error bars represent standard error.
CHAPTER 5
Figure 5.1 Sampling stations in Ischia Island, (Gulf of Naples) Italy. The stations are marked with black dots at Lacco Ameno and at the Castello Aragonese of Ischia.
Figure 5.2 nMDS ordination biplots of the community structure on raw averaged data of the choice index for the concentrations of each volatile bouquets at 20 minutes of exposure after prior normalisation of each variable to have zero mean and unit standard deviation for (a) control (pH 8.1) and (b) acidified (pH 7.7) seawater depending on VOCs bouquet tested indicated as vectors (Ent: Enteromorpha prolifera; Dic: Dictyota implexa var. linearis; CN: Diploneis sp.; C2: Cocconeis scutellum var.parva).
Figure 5.3 Group-average clustering on Euclidean distance matrices on raw averaged data of the choice index for the concentrations of each volatile bouquets at 20 minutes of exposure after prior normalisation of each variable to have zero mean and unit standard deviation at (a) pH 8.1 where the microhabitat preferences of mesograzers determines their behaviours to epiphyte-borne volatile infochemicals and (b) pH 7.7 where the responses of mesograzers to the same volatiles are grouped according their taxonomic relatedness.
LIST OF TABLES
CHAPTER 1
Table 1.1 Example of selection of volatile metabolites taken from Rowan (2011) in which the VOCs are arranged in order of increasing boiling point and chosen to illustrate the different chemical properties of this class of metabolite.
CHAPTER 2
Table 2.1 List of invertebrates used in behavioural assays and their taxonomy, microhabitat preference, and feeding guild according to Gambi et al. (1992). DF= deposit feeder that feeds on surface detritus; HeDF= herbivore-deposit feeders, which feed on plant epiphytes and trapped organic material; He= herbivores graze on micro- and macroalgae; Ca= carnivores.
Table 2.2 Repeated measures ANOVA for the score weights in static chambers within the related time groups: 5 min, 10 min, and 15 min in the baseline control experiment. Significant values if present (α = 0.05) are in bold and Greenhouse-Geisser correction is indicated as (+).
Table 2.3 Repeated measures ANOVA for the score weight in flume chambers within the related time groups: 3 min, 4 min, and 5 min in the baseline control experiment. Significant values if present (α = 0.05) are in bold.
Table 2.4 Permutational multivariate analysis of variance (PERMANOVA) of the differences between methods of investigation, taxonomic groups of invertebrates and concentrations of the three VOC bouquets on the behavioural patterns of invertebrates’ community. Me = method; Ta = taxonomy; pperm = probability values from permutations. Significant values (α = 0.05) are in bold.
Table 2.5 Pairwise comparisons from the multivariate analysis of behavioural patterns of mesograzers to VOCs. Significant values (α = 0.05) are in bold. Me = methods (static and flumes); Ctl = control; L = low concentration; H = high concentration; pperm = probability values from permutations; perms = number of permutations made.
CHAPTER 3
Table 3.1 List of the invertebrate species used in behavioural assays. In the table is indicated their taxonomy, microhabitat preference, and feeding guilds according to Gambi et al. (1992). DF: deposit feeder that feeds on the surface detritus; HeDF: herbivore-deposit feeders, which feed on plant epiphytes and trapped organic material; He: herbivores that graze on micro- and macro-algae; Ca= carnivores.
Table 3.2 Taxonomy, habitat, and references for the epiphyte species used as VOC bouquets in behavioural assays with mesograzers.
Table 3.3 Total volumes of ethanol (vol of EtOH in gl) added in 200 mL of agarose 0.06% for each species of algae and diatoms VOCs and respective stripped controls (FSW) in order to obtain the three final concentrations of bouquets (LOW: 6 mg, MEDIUM: 60 mg and HIGH: 600 mg; mg/dry extracted weight).
Table 3.4 Minimum numbers of replicates with the total number of individuals for each invertebrate used in VOCs experiments in order to reset the natural random walking behaviour. The minimum number of replicates was calculated with α=0.05 using the bestfit values (Span, K and Plateau) from the inverse of the one phase exponential decay equation (Eq.3; Leike, 2002) applied to the cumulative curves obtained as variation in differences of mean of integer index calculated as indicated in Jüttner et al. (2010).
Table 3.5 Pearson’s correlation values at 20 minutes for the four VOCs with the two main nMDS axes (MDS 1 and MDS 2). Ent: Enteromorphaprolifera, Dic: Dictyota implexa var. linearis, CN: Diploneis sp.; C2: Cocconeis scutellum var. parva.
CHAPTER 4
Table 4.1 Summary of results with comparisons between behavioural trends to VOCs and feeding ranking preference for the invertebrates for the different epiphytes and agar food (CTL: agar control food, Ent: Enteromorpha prolifera; Dic: Dictyota implexa var. linearis; C2: Cocconeis scutellum var. parva and CN: Diploneis sp.). The asterisks indicate the significant differences at α= 0.05. The asterisk in responses to VOCs indicate when the behavioural choices were significant according to the 4 cells z test on proportions with the baseline control experiment while the asterisks in feeding preference ranking indicate differences with pairwise comparisons post-hoc analysis (* = p < 0.05; ** = p < 0.01; ***= p<0.001).
CHAPTER 5
Table 5.1 Mean ± SD seawater physical parameters measured throughout the 6 day period in the aquarium. Total alkalinity (TA 2,559 pmol kg m" h" , G Rilov, pers comm., September 2014) and salinity (38 PSU, measured in June 2014) were point measurements. The parameters were calculated with CO2SYS (Lewis and Wallace, 1998) using the pHNBS, temperature, total alkalinity and salinity values with dissociation constants from Mehrbach et al. (1973) refit by Dickson and Millero (1987) and KSO4 using Dickson (1990).
Table 5.2 Summary of the results of behavioural overall patterns obtained for molluscs taking into account the significance of differences indicated by 4-cells Z test for single VOCs bouquet, concentration (Low, Medium and High) and time of exposure (10 and 20 min) at the control mean pH 8.1 and at the experimental mean pH 7.7. At pH 8.1, towards and away percentages on proportions were compared to controls without any “odours” (standardized controls). At pH 7.7, towards and away percentages on proportions from 4- cells Z test were tested against results at pH 8.1. The symbols ‘+’, indicates attraction for the VOCs source, ‘-‘indicates repulsion and ‘0’ indicates absence of choice by molluscs. The asterisks indicate the significance of difference (*p< 0.05, **p< 0.01, or ***p< 0.001) with the exact p-values reported in the test. The abbreviations for VOCs in the table are Ent = Enteromorpha prolifera; Dic = Dictyota implexa var. linearis; CN = Diploneis sp.; C2 = Cocconeis scutellum var. parva).
Table 5.3 Summary of the results of behavioural overall patterns obtained for decapods obtained taking into account the significance of differences indicated by 4-cells Z test for single VOCs bouquet, concentrations (Low, Medium and High) and time of exposure (10 and 20 min) at the control mean pH 8.1 and at the experimental mean pH 7.7. At pH 8.1, towards and away percentages on proportions were compared to controls without any “odours” (standardized controls). At pH 7.7, towards and away percentages on proportions from 4-cells Z test were tested against pH 8.1. The symbols ‘+’, indicates attraction for the VOCs source, ‘-‘indicates repulsion and ‘0’ indicates absence of choice by decapods. The asterisks indicate the significance of difference (*p< 0.05, **p< 0.01, or ***p< 0.001) with the exact p-values reported in the test. The abbreviations for VOCs in the table are Ent = Enteromorpha prolifera; Dic = Dictyota implexa var. linearis; CN = Diploneis sp.; C2 = Cocconeis scutellum var. parva).
Table 5.4 Pearson’s correlation values at 20 min for the four VOCs with the 2 main nMDS axes (MDS 1 and MDS 2) with the two pH treatments (8.1 vs 7.7). (Abbreviations as Ent= Enteromorpha prolifera; Dic= Dictyota implexa var. linearis; CN= Diploneis sp.; C2= Cocconeis scutellum var. parva).
APPENDIX 1
Table A.1.1 Output for Test of Normality for the factors: concentrations, epiphytes, invertebrate species and methods.
Table A.1.2 Levene’s test of Equality of Error Variances.
Table A.1.3 Tests of Between-Subjects Effects.
APPENDIX 2
Table A.2.1 Mean ± SD seawater carbonate chemistry measured in Petri-dishe for each mesograzer under different VOC treatments (n=9). Salinity (38 PSU) and total alkalinity (TA 2,560 pmol kg-1) and were point measurements taken in June 2014. The carbonate parameters were calculated using CO2 SYS spreadsheet (Lewis and Wallace, 1998) using the pHNBS, temperature, total alkalinity and salinity values with dissociation constants from Mehrbach et al. (1973) refit by Dickson and Millero (1987) and KSO4 using Dickson (1990). Temp in °C, pCO2 in patm.
CHAPTER 1 Introduction
1.1 Overview
Marine organisms rely on chemical cues to obtain information about their environment. The most important group of chemical cues carrying information (i.e. infochemicals) is composed mainly of volatiles, and organisms perceive them as “scents”. Volatile organic compounds (VOCs) released from plants represent one of the most important groups of infochemicals able to shape- through the influence onto behaviours of receivers- the ecological interactions and the structure of ecosystems. In this thesis, I will explore how VOCs from algal epiphytes can be putative “food and microhabitat infochemicals” influencing the behavioural choices of benthic invertebrates and their micro-scale patterns of movements within complex marine ecosystems like Posidonia oceanica beds.
1.2 The general context of marine chemical ecology
Chemical signals represent the words of the ‘language’ of life in the sea (Kittredge et al. 1974; Atema, 1995; Hay, 1996) and marine chemical ecology refers to the study of these signals involved into the biotic and abiotic interactions among marine organisms and their environment (Hay, 2009). Thus, the main purpose of chemical ecology is to translate this language from chemistry to ecology in order to understand better how to manage the structure and functions of natural ecosystems (Kubanek, 2014; Hay, 2014).
Since the 1980s, chemical ecology was considered as an integrative science between the field of chemistry and ecology (Pawlik, 1993; Ianora et al. 2011). Indeed, at least four different fields of science convey in chemical ecology: (i) chemistry, that identifies molecules acting as chemical cues; (ii) ethology, that clarifies the effects of these chemicals on animal behaviour; (iii) neuroscience, that determines the mechanisms involved in the reception of these molecules and (iv) ecology, that recognizes the central role of chemicals in driving the structure of populations, communities and evolutionary processes (Fig. 1.1 from Derby and Sorensen, 2008).
illustration not visible in this excerpt
Figure 1.1 The Venn diagram illustrates chemical ecology as an integrative science at the intersection of the four main approaches of chemistry, ethology, neuroscience, and ecology. Figure from Derby and Sorensen, (2008).
The initial progresses of chemical ecology focused mainly on the discovery of ‘unusual’ chemical metabolites (e.g. more than 16,000 compounds; Ianora et al. 2006) from a variety of marine organisms without exploring their ecological functions (Hay and Fenical, 1996; Hay, 2014). However, several of these marine compounds so far have found their application in biomedical utility (Hay, 1996), in aquaculture or even as integrated pest management and in chemical industries (Ianora et al. 2011; Sorensen Forbey et al. 2013).
Moreover, ecologists have highlighted that the classical studies on food web dynamics and on nutrient limitation models fail when they are based on classical isolated predator-prey interactions without considering also behavioural responses to chemical information (see reviews from Van der Stap et al. 2009; Ianora et al. 2011; Turner and Peacor, 2012 and references therein). The research by Kvitek et al. (1991) can provide an appropriate exhaustive example integrating the presence of chemical metabolites and the ecosystem function. These authors clarified how the historical fluctuations of sea otters, Enhydra lutris, along the West coast of North America were “indirectly” linked to the presence of the toxic dinoflagellate red tide bloom forming, Protogonyalux spp. throughout the butter clams Saxidomus giganteus. Sea otters, in fact, preferentially feed on butter clams that sequester extremely high concentration of Paralytic shellfish poisoning toxins (PSPT) and then determine high mortality in sea otters, fish and birds which forage on them. The authors considered that PSPT affected the regional distribution of sea otters since they recorded this species was historically absent where dinoflagellate blooms were common and vice versa. If keystone consumers (Paine, 1966) like sea otters are missing from kelp forests, herbivores as sea urchins can alter the ecosystem structure lowering the biodiversity (‘alternative stable state’; Connell, 1961). This represents only an example regarding the centrality of chemical ecology in understanding environmental fluctuations and ecosystem processes.
Thus, the subjects of marine chemical ecology in aquatic environments concern mainly two aspects of chemically-mediated interactions: (i) the chemical defenses of prey against consumers (with the production of secondary metabolites) and, (ii) the use of these chemical compounds as carrier of information. The latter group of compounds are defined as infochemicals, chemicals that carry information that mediates an interaction between two organisms (generally different species) that ends up in an adaptive response in the receiver but either the sender or the receiver or even both may benefit from this information (Brönmark and Hansson, 2012; Fig. 1.2).
Differently from humans that perceive their environment by sight and hearing, marine organisms lack or have underdeveloped these two senses and they assess threats by relying on chemical cues (Hay, 2009). Indeed, most of the aquatic organisms produce and respond to a variety of chemical compounds which are used for protection and competition (i.e. chemical defenses), for food finding, mating, kin recognition and hierarchy, habitat finding, predation risk (i.e. infochemicals), both in freshwater and marine ecosystems (reviewed in Vos et al. 2006; Hay, 2009), in plankton and benthos realms (McClintock et al. 2010; Ianora et al. 2011).
The aim of this chapter is to introduce the reader in the chemical ecology of infochemical-mediated interactions between marine plants and mesograzers. I will provide a picture regarding the knowledge and the gaps about the use by mesograzers of volatiles from marine diatoms and algae as infochemicals and their effects at the community level within a specific ecosystem such as the Mediterranean seagrass Posidonia oceanica. For this reason, I will only give few examples regarding the broader field of “planktonic marine chemical ecology” without going deeper in details and, I will refer mainly to benthic marine “plants” as the main producers of infochemicals, including micro- and macro-algae as epiphytes of marine seagrasses. For further information on specific topics, I would suggest the detailed overviews in McClintock and Baker, (2001), Brönmark and Hansson, (2012). Additionally, for exhaustive overviews regarding the advances in plankton chemical ecology, I further suggest reviews by Van Donk et al. (2007), Ianora et al. (2011) and Poulson et al. (2013) and references therein contained.
illustration not visible in this excerpt
Figure 1.2 Chemical-ecological interactions driven by chemical defences and infochemicals produced by higher animals and primary producers in freshwater and marine ecosystems. Figure from Brönmark and Hansson, (2012).
1.3 The current knowledge on infochemicals
All marine animals and marine primary producers, purposely or not, release chemicals from their bodies (Breithaupt and Hardege, 2012) making their own environment suffused of chemical stimuli. These chemicals conveying messages, can be just “ballast” products from the main metabolism (Harper et al. 2001) or can be “baggage” evolved under natural grazing/predation pressure (Ianora et al. 2006).
The use of metabolites carrying information, so-called “infochemicals” is wide spread within and between microorganisms, plants, invertebrates, and vertebrates. In this context, the terminology accepted to describe the different classes of infochemicals is on cost-benefit analyses rather than on the origin (Dicke and Sabelis, 1988). Infochemicals are divided into two focal categories, depending on their use within the same species (i.e. pheromones) or among different species (i.e. allelochemicals; Dicke and Sabelis, 1988). Pheromones, according to their specific functions can be subdivided as territory marking pheromones, alarm pheromones, trail-marking pheromones, mating and sex pheromones (Derby and Sorensen, 2008; Breithaupt and Hardege, 2012 and references therein).
Differently, allelochemicals play several ecological roles and are defined as allomones, kairomones and synomones according to whom between senders and/or receivers will benefit from the message (Dicke and Sabelis, 1988; see Fig. 1.3 for the differences).
A communication based on infochemicals starts when a receiver perceives them (Zimmer and Butmann, 2000) and utilizes these messages in several interactions with different functions (von Elert, 2012). Lobsters and crayfish release urine during fights to establish the social status and dominance (Atema, 1995; Thiel and Duffy, 2007). The bloom forming phytoplankton Phaeocystis globosa recognizes the smell of predators, if copepods (that generally feed on larger foods) or ciliates (that prefer smaller food) and then shifts its shape growing as single cells or colonies, respectively (Long et al. 2007).
illustration not visible in this excerpt
Figure 1.3 The terminology of infochemicals given according to Dicke and Sabelis (1988). E is the Emitter and R is the Receiver, the symbol + indicates which one between emitter and/or receiver benefits from the chemical message.
Generally, infochemicals can be specific, used only by one species, and not recognized by other organisms- not even of the same genus-, or may be very common and captured by different unrelated species (Klaschka, 2008). The receiver can respond in many different ways. The planktonic crustacean Daphnia sp., for instance, shows several defence strategies against predators (Fig. 1.4), such as behavioural changes (e.g. diel vertical migration; Loose et al. 1993), physiological adaptations (e.g. suppressed growth and delayed hatching; Stibor and Lüning, 1994) and morphological evolution (e.g. formation of helmets and spines; Agrawal et al. 1999). These chemical cues involve the detection by the sense of smell and we generally refer to them as odorants (Atema, 1995). These odours are usually made by unique compounds or blends and complex mixtures within nanomolar or even lower range of effective concentrations (Wolfe, 2000) and act in a complicated and dynamic web of interrelationships and constraints. In this web of messages, a sender is capable of releasing more infochemicals at different concentrations and intervals of time, and consequently a receiver should be able to detect many infochemicals at once and promptly respond only to the interesting ones (Fig. 1.5a; Klaschka, 2008).
Among allelochemicals, kairomones represent the most studied group of infochemicals released by primary producers. They induce immediate behavioural responses in the target organism, working as attractants or repellents with different ecological functions (Ruther et al. 2002). Moreover, among kairomones, volatile organic compounds (i.e. VOCs) (Fink, 2007; Rowan, 2011) released from marine plants constitute a very promising category of infochemicals.
illustration not visible in this excerpt
Figure 1.4 The planktonic crustacean Daphnia sp. shows several defence strategies against predators’ chemical cues. (A) Scanning electron micrographs showing morphological defences against several predators in various Daphnia species (Photograph by C. Laforsch, from Chemical ecology in aquatic systems. Brönmark & Hansson eds. 2012). (B) Reproduction of Daphnia showed in the absence of predators (middle panel) and defence strategies with predators. In the left panel, behavioural (i.e. diel vertical migrations) and physiological changes (e.g. earlier sexual maturity at a smaller size with increased number of offspring of reduced size) are present in response to fish predation. In the right panel are showed morphological changes (e.g. spines in the neck) and increased body size but decreased number of offspring in the presence of the Chaoborus larvae as predators (Illustration by L. Weiss, from Chemical ecology in aquatic systems. Brönmark & Hansson eds. 2012).
1.5 Volatile Organic Compounds from marine plants and their functions as infochemicals
Volatiles include organic compounds with low molecular weight of 50-200 Daltons, low or moderate hydrophilicity, and high vapour pressure under ambient conditions (VOCs; Rowan, 2011). These features make VOCs readily dissolved into the gas phase at air-water interfaces and perceptible as odours to both terrestrial and aquatic organisms (Fink, 2007). They arise by several biosynthetic pathways from amino and fatty acids, and from terpene pathway (Rowan, 2011). Thousands of different volatile compounds occur in nature as naturally occurring compounds and because of human activities. For humans and animals, volatiles are important as scents and generally they are responsible for the flavor of food (Rowan, 2011). Volatiles arise by a variety of biosynthetic pathways but principally from amino and fatty acids, and terpene biosynthetic pathways (Pichersky et al. 2006) and include a wide range of chemical classes (hydrocarbons, aromatics, alcohols, aldehydes, acids, esters, amines and thiols) with a range of physical properties from gases at room temperature (ethylene, see Table 1.1 ) to higher molecular weight compounds such as skatole and androstenone which possess sufficient vapor pressure and biological activity to be clearly perceived also by humans (Rowan, 2011).
Table 1.1 Example of selection of volatile metabolites taken from Rowan (2011) in which the VOCs are arranged in order of increasing boiling point and chosen to illustrate the different chemical properties of this class of metabolite.
illustration not visible in this excerpt
In terrestrial ecosystems, VOCs are important infochemicals purposely liberated by plants from flowers to attract pollinators in order to ensure themselves the reproduction (e.g. plants and arthropods, Dicke, 2000) or after an induction event, such as an attack by herbivores. In this latter case, VOCs serve to recruit help from carnivore insects (i.e. ‘crying for help’, Kessler and Baldwin, 2001; or ‘tri-trophic interaction’ Dicke and Baldwin, 2010) and/or to communicate the presence of herbivores to neighbour plants that might switch their defences on (Fig.1.5b, c; Ueda et al. 2012).
Differently from terrestrial plants where the ecological roles and the characterization of VOCs have been well understood, in aquatic systems the research regarding their roles and how they evolved is still at the beginning (von Elert, 2012). Marine environments are a source of numerous trage gases that have impact on atmospheric processes and climate and have their role as infochemicals (Steinke et al. 2002; Pohnert et al. 2009). Over the last 25 years, particular attention has been given to the non-methane hydrocarbons including isoprene, halocarbons and most of all to the dimethyl-sulfide (DMS) and its role in climate regulation via oceanic phytoplankton sulphur emission (Wolfe et al. 1997). The major source of DMS is the ocean and all marine micro- and macro-algae studied to date also emit isoprene to protect themselves against high salinity and associated environmental stresses, such as high ultraviolet radiation (Rinnan et al. 2014).
In marine plants, the liberation of these volatile compounds requires cell breaking, being VOCs mostly considered as “herbivore-released chemicals” (Fratini et al. 2001; Van Donk et al. 2011). For marine diatoms, Pohnert (2000) demonstrated that generally volatiles derive from an enzymatic pathway activated few seconds after cell wounding by herbivores. The overall enzymatic mechanism seems to be the same across marine micro- and macroalgae (Pohnert and Boland, 1996; Jüttner, 2001; Fink et al. 2006a, b; Akakabe et al. 2007). A phosholipase triggers the wound-activated chemical defenses and the cleavage of the eicosanoic fatty acids (C20) into polyunsaturated volatile aldehydes (PUAs) through the lypoxygenase and hydroperoxidelyase cascade (Pohnert, 2000; Leflaive and Ten-Hage, 2009).
Despite of the many ecological roles of VOCs identified from the aquatic primary producers, only few are the examples of known compounds (Von Elert, 2012).
Indeed, several chemical cues can induce the same response in the receiver and their activity might be a synergistic effect of a blend further than of a single compound (Wiesemeier et al. 2007). In addition, the bioassay-guided fractionation technique for their identification is a time-consuming process that together with the other features of these VOCs, slows their identification down (Pohnert et al. 2007).
In particular, depending on the main ecological role and the kind of responses of the receivers, VOCs from primary producers can be sexual infochemical-kairomones, foraging or aggregating infochemical-kairomones and enemy-avoidance or defensive infochemical- kairomones (Ruther et al. 2002; Fink, 2007).
The sexual kairomones are generally volatiles inducing attraction behaviour in the receiver that uses these compounds to find partners for reproductive purpose. The brown seaweeds (e.g., Ectocarpus, Laminaria, Fucus and Dictyopteris) are able to use volatile pheromones, which are compounds with from eight to eleven carbon atoms (reviewed in Pohnert and Boland, 2002). Females release these compounds at picomolar concentrations in order to work as reproductive-synchronizers and attractants for male gametes (Boland, 1995). These sexual kairomones showed conserved structures among different brown algae genera and similar volatile pheromones were found in freshwater diatoms. The diatom Asterionella formosa produces fucoserratene (i.e. the pheromone of Fucus vesiculosus) and the diatom Gomphonemaparvulum produces hormosirene (i.e. the pheromone of Analipus japonicas; reviewed in Fink, 2007).
illustration not visible in this excerpt
Figure 1.5 (a) Dynamic web of interactions and constraints driven by infochemicals (Figure modified from Klaschka, 2008); (b) examples of infochemical relationships within terrestrial ecosystem with specific insects-plant interactions and (c) with a generalist herbivore (Figure modified from Ueda et al. 2012).
In aquatic systems, most of the studies have focused on the food-finding chemical cues for fish and invertebrates like decapods and molluscs. Generally, camivourous invertebrates like decapods respond to small cues with nitrogen group like for instance amino acids, peptides, amines and nucleotides while herbivores like molluscs or omnivores species like fish prefer sugars (Derby and Sorensen, 2008; Von Elert, 2012). The foraging and the aggregation kairomones guide herbivores towards their high nutrient value food items and towards their preferential habitats. The freshwater snail Lymnea stagnalis, for instance, recognizes the good quality of food and maximizes its foraging behaviour only towards VOCs from the green algae Ulothrix fimbriata that was reared under high nutrient conditions (PO4 and NO3; Moelzner and Fink, 2014).
In aquatic systems, the bulk of the researches on the ecological functions of VOCs pointed out on their main role as at the same time, foraging and aggregating infochemicals (e.g., food-finding habitat) in different habitats, similarly to what found for the terrestrial plants-herbivores interactions (von Elert, 2012). For benthic and slow-moving grazers such as snails, it is important to rely on chemical cues released by algae in order to reduce their chance to encounter predators during their foraging activities (Norton et al. 1990). The freshwater snail Radix ovata is attracted by a multi-component odour of the damaged green algae Ulothrix fimbriata (Fink et al. 2006a) and by volatile C8-compounds released upon artificial lysis of the benthic freshwater diatom Achnanthes biasolettiana (Fink et al. 2006b), which were the snail diet-derived chemical signals. Responses to a blend of VOCs more than to single compounds are well demonstrated. The nematode Bursilla monohystera responded to VOCs from cyanobacterial biofilms (e.g. Plectonema sp. and Calothrix sp.) when they were tested as mixtures more than as single volatiles (Höckelmann et al. 2004). Similarly, only the mixture of VOCs from the brown alga Dictyota dichotoma containing the trimethylammine (TMA) and dimethyl sulphide (DMS), repelled the herbivore amphipod Amphithoe longimana (Wiesemeier et al. 2007). Jüttner et al. (2010) demonstrated that a mixture of C7 and C5 volatile compounds, released after the cell lyses of the benthic diatom Cocconeis scutellum parva, was attractive for several benthic invertebrates.
In addition, also the active grazing of invertebrates (e.g. the radula of the snail that scrapes the substrate or claws of crabs cutting filamentous algae) determines the release of VOCs that would further attract conspecifics consumers towards these patches of food. The intertidal snail Terebralia palustris (Fratini et al. 2001) and the freshwater snail Lymnea stagnalis (Moelzner and Fink, 2015a) followed gradients of diet-chemical signals after that conspecifics were grazing on mangroves and green algae, respectively.
In the open ocean, the non-homogeneous distribution of resources may explain the reason why organisms adapted on chemical cues to guide themselves towards patches of food (Steinke et al. 2002). DMS is released after viruses or herbivorous such as copepods and protists zooplankton (Oxyrrhis marina) forage on blooms of the haptophytes Emiliania huxleyi or Phaeocystis pouchetti (Dacey and Wakeham, 1986; Wolfe et al. 1997). DMS and other volatiles that usually act as chemical defences, repelling local planktonic grazers (Wolfe et al. 1997), become foraging/aggregating infochemicals for organisms of higher trophic levels. Nevitt, et al. (1995) demonstrated that DMS over a spatial scale of kilometres is a‘predictor of krill grazing’ (Hay, 2009) for several bird species (Savoca and Nevitt, 2014). According to these findings, DMS works as indirect defence for phytoplankton in a tri-trophic interaction similar to those interactions occurring in terrestrial systems, where the plants upon herbivores attacks release volatiles that inform carnivorous species about the presence of their preferred prey (Steinke et al. 2002; Vos et al. 2006; Fig. 1.6a).
So far, Coleman et al. (2007a) described the only example of a benthic tri-trophic interaction driven by VOCs. In marine rocky pools, the brown seaweed Ascophyllum nodosum grazed by the herbivore Littorina obtusata releases infochemicals, which positively influence the foraging behaviour of predators such as the crab Carcinus maenas and the fish Lipophryspholis (Coleman et al. 2007b; Fig. 1.6b).
illustration not visible in this excerpt
Figure 1.6 (a) Schematic description of the planktonic use of dimethyl sulphide (DMS) that is a foraging kairomone attracting birds to patches of food (Figure modified from Steinke et al. 2002). (b) Schematic description about the tri-trophic interaction described in Coleman et al. (2007a). Point 1 indicates the herbivore Littorina obtusata feeding on the brown algae Ascophyllum nodosum in a rocky pool. Point 2 indicates the release of VOCs from grazed A. nodosum, acting as SOS signal that attracts as foraging infochemical the predator green crab Carcinus maenas that in point 3 “rescues” the algae by eating the herbivore.
Contrarily to the first three groups of kairomones, the latter group of enemyavoidance/defensive infochemicals repels the receiver. In this case, prey uses these signals to escape from potential predators or enemies. However, it has to be still proven whether the VOCs from aquatic primary producers can have also this function (Fink, 2007) and so far, to my knowledge, no exhaustive examples can be given regarding this subject. Overall, informations as VOCs jump trophic levels influencing not only single organisms. They are involved also in population-level processes, community organization and ecosystem functions (e.g., Hay and Kubanek 2002; Vos et al. 2006; Pohnert et al. 2007; Nevitt, 2011; Lewis et al. 2012). VOCs from marine plants attract herbivores towards patches of their favourite food source (Moelzner and Fink, 2015b). They attract carnivores towards their herbivore prey (Coleman et al. 2007a; Savoca and Nevitt, 2014). Moreover, they maintain a balanced number of primary producers and grazers (e.g. polyunsaturated aldheydes from diatoms: Miralto et al. 1999; Ianora et al. 2011) and move to the atmosphere controlling the global climate (e.g. DMS, involved in cloud condensation nuclei; Nevitt, 2011; Exton et al. 2015).
1.6 The method: behavioural choices in the investigation of infochemicals
As highlighted in the previous paragraph, several studies have shown the ability of animals to detect and modify their behaviours towards or away from the presence of predators or of food items using chemoreception. Keppel and Scrosati (2004), for instance, demonstrated how benthic invertebrates like Littorina scutulata could also show “behavioural gradient avoidance responses” to predators considering a risk hierarchy.
For research regarding pheromones and chemical cues, the right choice of bioassay is crucial (Wyatt, 2010). Nowadays, a great effort in determining the exact identity of molecules (and thus infochemicals) with new technique (e.g. metabolomics; Pavia et al. 2012) might allow also quantifying the natural concentrations and the rates of release. However, so far, the simplest way to test whether a chemically mediated effect exists and what might be its ecological role is still the use of behavioural assays (Vet, 1999; von Elert, 2012; Pohnert, 2012).
Behavioural researches passed through a transition stage, from purely descriptive studies to a new era in which behaviour is also considered to verify hypotheses and clarify functional consequences (Altmann and Altmann, 2003; de la Haye et al. 2012). Additionally, the use of behavioural choice assays to investigate ecological interactions driven by infochemicals, has the potential to reveal also the repellent activity of these signals at lower and more relevant ecological concentrations than classical toxicity or inhibition assays where most of test organisms do not have the choice to avoid the compounds supplied (von Elert, 2012). This idea was, indeed corroborated by the study of Maibam et al. (2014) that demonstrated how the level of recognition of volatiles from diatoms by invertebrates is inversely correlated with the toxicity of wound-activated compounds released by diatoms. Therefore, immediate responses regarding how the organisms perceive the environmental stimuli, indeed, can come by observing their behavioural choices.
1.7 Behavioural traits and mechanisms of chemoreception in marine invertebrates
Benthic invertebrates are capable to diplay relevant behavioural traits according to different life moments like mating and reproduction, homing, foraging or even to hide themselves to predators. All these activities are generally linked to the chemoreception ability of the animals to perceive their environment (Brönmark and Hansson, 2012).
Chemoreception is the process by which organisms perceive via their olfactory and gustatory organs, chemical cues that describe their environment (Bradbury and Vehrencamp 1998).
Typical intertidal animals like molluscs generally display two important behavioural traits like foraging and homing. Foraging is the series of behaviours the animals generally display when searching for foods items. Most research to date has mainly examined foraging for gastropods and echinoids since they are the most abundant benthic grazers in rocky intertidal habitats. Marine animals generally influence the distribution and abundance of algal cover feeding on algae in different ways (Chapman and Underwood, 1992; Gacia et al. 2009; Vergés et al. 2011). Many species are permanent members of the epiphyton being microalgal grazers which scrape or sweep the substratum collecting more microalgae like diatoms; other species are more transient although their movements over long periods appear to be more linked to the vertical range and they are generally considered macroalgal grazers that feed on pieces of macroalgae which they scrape, bite or tear off the adult plants (Chapman and Underwood, 1992; Williams and Seed, 1992). Therefore, chemical cues are thought to be important in the location and choice of preferred food. Chitons and limpets, for instance, display homing behaviour after feeding excursions. The chiton Acanthopleura gemmata, migrates up to 3 m away from its “scar spot” in the rock to feed on algal ground. Chelazzi et al. (1987) demonstrated that each individual of this chiton species follows its own trail making a loop and return back to its spot after feeding. Marine snails like the intertidal periwinkle Littorina littorea although do not show homing behaviour (Norton et al. 1990), are able to follow pheromones embedded in the mucus trails of females as well as to discriminate trails of conspecifics for foraging purpose (Edwards and Davies, 2002).
Gasteropods have the ability of distant chemoreception probably through the sensory cells at the cephalic tentacles or in the mantle cavity, while contact chemoreception might be linked to the presence of sensory cells in the mantle frindge and the foot (Croll, 1983; Stabell, 2012; Atema, 2012). Similar to molluscs, crustacean decapods (e.g., some shrimps, lobsters, crayfish, hermit crabs and crabs) are able to display specific behavioural traits according to the processes they are dealing with. Crayfish release urine during fights conveying information about the signaller’s fighting ability (Breitaupt and Eger 2002) and hermit crabs use these cues during shell investigation behaviour (Gherardi and Atema, 2005; de la Haye et al. 2012). Sensory organs that detect information in crustaceans are generally different subsets of chemosensors present on most body surfaces including appendages such as the first antennae (i.e. antennules), second antennae, legs and mouthparts, but also body regions like the cephalothorax, abdomen and telson. The chemosensors are organized in sensilla, which are cuticular extensions of the body surfaces that are innervated by the dendrites of chemosensory neurons. The aesthetasc sensilla are unimodal sensors and are the receptors for the olfactory pathway. All the other known chemosensilla are bimodal, being innervated by both chemosensory and mechanosensory neurons and they represent the non-olfactory chemosensory pathways (Derby and Zimmer, 2012).
The first step of chemoreception involves the reversible binding of the chemical stimulus to the receptor surface. This receptor is generally a protein that changes, either in shape or in chemical composition and depolarises; this depolarisation triggers nerve impulse carrying information to the brain (Breer, 2003). Studies on the peripheral nervous system of the American lobster Homarus americanus, have shown that olfactory receptor neurons (ORNs) are able to to transmit and emphasize the odor cues and neurophysiological studies have shown that flicking behaviour enhances the sensitivity of the ORNs in the antennules (Grasso and Basil, 2002).
The integration of a signal is an interpretation of activated neurons in higher brain areas and the cholinergic and gabaergic nervous system have been shown to be involved in signal integration and in fine-tuning of vital responses (Weiss et al., 2012). It was found that physostigmine modulates the response of kairomone stimulation in Daphnia pulex (Barry, 2002). Physostigmine is an inhibitor of the enzyme acetylcholine esterase that degrades and terminates the stimulation of acetylcholine bounded at the postsynaptic side. Therefore, an inhibition of the acetylcholine esterase results in a protracted bounding of acetylcholine with its receptor resulting in increased postsynaptic rates. In Daphnia pulex this process determine the necktheet expression (Barry, 2002). The use of GABA agonists that simulates the inhibitory gabaergic actions decreased the expression of necktheet. Therefore, it was speculated that cholinergic stimulation is not the only component in this transduction process, probably also several stimuli from multiple odours may require cross-talk between different signal transduction pathways. Very recently, for example, the mechanism accounting for a loss of olfaction and chemosensory abilities in marine fishes exposed to acidified conditions appeared to be linked to a disruption of the normal Cl" and/or HCO3- gradients over neuronal membranes, causing some GABA-A receptors to become excitatory (depolarizing) rather than inhibitory (hyperpolarizing) (Nillsson et al. 2012). The GABA-A receptor, which is the major inhibitory neurotransmitter receptor in vertebrate brains, normally hyperpolarizes neuronal membranes by opening a channel that leads to an influx of Cl- and HCO3- (Ledùc et al. 2013).
These are only few examples of the neuronal signal integration after odour perception and the exact mechanisms for different species and for specific chemical stimuli remain unknown. As long as we do not know the chemicals involved in specific behavioural traits, we lack fundamental information in all respects, from the ecology to the molecular level (Weiss et al. 2012).
1.8 The seagrass Posidonia oceanica, its epiphytes and associated invertebrates
Seagrasses are marine phanerogams, the only group of higher plants that have developed the capabilities to survive in a completely submerged marine environment (Den Hartog 1970). They are 60 species that form extensive ‘meadows’ along the coastlines of continents (Hemminga and Duarte, 2000). Seagrasses are also considered ‘ecosystem engineers’ because influence local hydrodynamics, enhance resources and space for several organisms and provide huge productivity (reviewed in Buia et al. 2004 and reference therein).
Along with enhance the marine biodiversity, the primary and secondary production, seagrasses are also important for several ecosystem processes (Orth, 1992). They reduce the coastal erosion by the stabilization of sediment that damp off the flow and wave velocity (Orth, 1997). They improve the water quality by removing nutrients from the water column, increasing the light penetration, and reducing nutrients for phytoplankton growth (Hemminga and Duarte, 2000). Moreover, seagrasses are important in global carbon and nutrient cycling, indeed their biomass ends up in the sediments as detritus acting as sink of biogenic carbon (Buia et al. 2004) and they are important in the production of marine resources since they host juvenile stage of commercially important species of offshore fish and invertebrates (Orth, 1992).
Posidonia oceanica (L.) Delile, (Fig. 1.7) is the endemic seagrass of the Mediterranean basin and provides the associated fauna with food, refuges from predation, and nursery for larvae. It is one of the most productive among all the marine ecosystems (Orth, 1992; Mazzella et al. 1992; Scipione et al. 1996).
The extention, the architecture and the associated communities of flora and fauna in the meadows of P. oceanica depend on the environmental characteristics such as quality of the bottom and water as well as hydrodynamics (Buia et al. 2004). It is the longest-lived species among phanerogams with two main compartments, the leaf canopy and the root- rhizome layer (Piazzi et al. 2002). The leaves grow from the centre of the shoot (juvenile leaves in the centre and older leaves outside) and have a long turnover of almost 300 days, allowing the development of different associated communities of epiphytes and animals (Mazzella et al. 1992; Alcoverro et al. 1997; Bologna and Heck, 1999).
The associated communities change over the seasons, environmental conditions and along the leaf axis (Mazzella and Ott, 1984; Mazzella et al. 1994; Heck and Valentine, 2006).
The epiphytes of Posidonia oceanica can be generally autotrophic (e.g. cyanobacteria, diatoms, crustose or filamentous algae) and heterotrophic (e.g. bryozoans and hydroids) organisms which need a place to settle and grow (van Montefrans et al. 1984). The epiphytes on P. oceanica can grow on the leaves and on rhizomes (Piazzi et al. 2002). In general, a succession and zonation of algal epiphytes is found along the leaf axis according to the age-gradient (Alcoverro et al. 1997), seasons, and depths (Mazzella et al. 1992; Buia et al. 2000; Fig.1.8). The ephemeral or filamentous species, growing mainly from spring to late summer, are associated to the leaves while the perennial species are more common on rhizomes (Piazzi et al. 2002; Nesti et al. 2009; Mabrouk et al. 2014).
illustration not visible in this excerpt
Figure 1.7 Photo of a meadow of Posidonia oceanica (L.) Delile with visible epiphytes on the leaves (left) and on the right (d) an image of a shoot of P.oceanica with the rhizome covered by scales (remaining of the old leaf bases) showing also (a) adult leaf, (b) intermediate leaf and (c) juvenile leaf. (Image modified from Buia et al. 2004).
In the basal part of the juvenile leaves (already after one week), the dominant component are mainly cyanobacteria (e.g., Anabaena sp., Merismopedia sp. and Oscillatoria sp.; Mazzella, 1983; Maibrouk et al. 2011). In the middle part (after ~100 days), diatoms constitute the dominant group with Naviculaceae and Cocconeis sp., which account for the highest number of species, together with an encrusting layer of red and brown macroalgae (e.g. Corallinaceae Pneophyllum sp. and Fosliella sp. and brown alga Myrionema orbicularis; Cinelli et al. 1984; Mazzella et al. 1994). In the apical part (after ~300 days), the erected ephemeral epiphytes grow as macro-algal layers upon the previous diatoms and encrusting species (Mazzella et al. 1992). Generally, the ephemeral species at the tips of the leaves constitute the ‘Herbier de Posidonie’ (Cinelli et al. 1984) among several meadows in the entire Mediterranean basin (detailed tables on epiphytes species are in Cinelli et al. 1984; Prado et al. 2007; Balata et al. 2009, Garrard, 2013; Mabrouk et al. 2014). The algal assemblage on rhizomes belong to environment with reduced conditions of light and are generally turf-forming with only few species as common (e.g. Dictyota sp.,; Mazzella et al. 1992; Piazzi et al. 2002; Mabrouk et al. 2014).
The epiphytic stratum on the leaves of Posidonia oceanica ranges from a maximum of 210 mg/shoot in biomass and 1.1 mg/cm of covering in summer, to a minimum of 9.9 mg/shoot and 0.07 mg/cm2 in winter (Mazzella and Ott, 1984; Buia et al. 2000). The epiphytic algae contribute from 20% up to 60% of the total productivity of P. oceanica canopy (Orth, 1992; Mazzella et al. 1992) and mesograzers strongly modify the patterns of epiphytic assemblages associated to the meadow (Bologna and Heck, 1999; Prado et al. 2007; Gacia et al, 2009; Vergés et al. 2011). The epiphytes of P. oceanica change the entire ecosystem providing food and habitats (Bologna and Heck, 1999) for the associated animal community (Mazzella and Russo, 1989; Russo et al. 1992; Buia et al. 1992; Gambi et al. 1992).
illustration not visible in this excerpt
Figure 1.8 Illustration of the spatial succession of algal epiphytes over a leaf axis of Posidonia oceanica that shows the increase in species and abundance from the basal towards the apical zone. (Figure modified from Mazzella et al. 1994 (eds.), Le praterie sommerse del Mediterraneo).
The associated fauna community of Posidonia oceanica is one of the richest both in terms of species and individuals, as compared to others coastal ecosystems (Buia et al. 2000). Posidonia beds host diversified groups of invertebrates like echinoderms and cephalopods, as well as fish such as Labridae and the herbivore sparid Sarpa salpa (Mazzella et al. 1992). However, the highest biodiversity within meadows is due to greater abundance and number of species of mesograzers (Orth, 1992).
The architecture of Posidonia oceanica and the microclimatic differentiation among the various compartments of the meadows determine several microhabitats giving shelter and food items to different taxonomic groups of benthic invertebrates (Bianchi et al. 1989; Gambi et al. 1992; Scipione et al. 1996).
The leaf stratum is rich of small size molluscs and crustaceans, which show strong diel migrations within the meadows (Russo et al. 1984a). Mesograzers such as the molluscs, Gibbula ardens, G. umbilicaris, Rissoa italiensis, R. violacea, R. auriscalpium, Alvania lineata (Russo et al. 1984b) and decapods like the shrimp Hippolyte inermis and the hermit crab Cestopagurus timidus (Gambi et al. 1992), represent the ‘fundamental stock’ of mesograzers of P. oceanica (Russo et al. 1984b; Garrard, 2013). These species show depths preferences. The shrimp Hippolyte inermis (Zupo, 1994) and Rissoa italiensis (Gambi et al. 1992) are usually more abundant in shallow beds (0-4 m) while the gastropod Bittium latreillii is almost exclusive of deeper meadows (15-30 m; Gambi et al. 1992; Russo et al. 2002). Differently, the hermit crab Cestopagurus timidus is generally abundant at all depths (Zupo et al. 1985; Gambi et al. 1992).
Trophic structure reveals that herbivory is the most frequent trophic habit among several taxa of vagile fauna (Mazzella et al. 1992; Garrard, 2013; Fig. 1.9) though the ‘seagrass grazing paradigm’ considered the detritus pathway as the most important (Heck and Valentine, 2006). The herbivores within P. oceanica meadows in shallow stations account for more than 71 % of the total fauna and are positively correlated to the increase
of epiphytes (Gambi et al. 1992). This epiphyte-grazer relationship might explain the paradox in terms of energy transfer over higher trophic levels (Mazzella and Zupo, 1995; Bologna and Heck 1999; Gacia et al. 2009; Ricevuto et al. 2015; Fig.1.9).
illustration not visible in this excerpt
Figure 1.9 The diagram shows the complex trophic relationships within a meadow of Posidonia oceanica against the ‘seagrass grazing paradigm’ (Figure from Mazzella and Zupo, 1995).
1.9 Epiphytes-herbivores interactions: the role of infochemicals within Posidonia oceanica ecosystem
The epiphyte-grazer relationships and the role of herbivory in seagrasses are fundamental subjects in researches (Mazzella et al. 1992; Heck and Valentine, 2006) to guarantee the maintaining of biodiversity with appropriate coastal management.
The intrinsic features of living Posidonia oceanica (e.g. high content of structural carbohydrates, high C/N ratio, and phenolic compounds) and the huge amount of detritus have brought to the idea of the inconsistent ‘seagrass grazing paradigm’ (Heck and Valentine, 2006). Although only few species such as the sparid fish Sarpa salpa, and the sea urchin Paracentrotus lividus are considered capable of feeding upon living P. oceanica leaves, (Mazzella et al. 1992), the paradox is that herbivores and herbivores-deposit feeders constitute the dominant feeding categories within P. oceanica. Therefore, this paradox suggested the role of epiphytes as main food source for grazers and the potential existence of co-evolutionary processes involving epiphyte, grazers, and seagrasses as main players in exporting energy throughout higher trophic levels (Mazzella and Zupo, 1995; Fig. 1.9).
The features of Posidonia oceanica, the algal and fauna associated communities represent stable and persistent “interactive” ecosystems over the time (Mazzella et al. 1992; Orth et al. 2006; Garrard, 2013). However, though the four decades of researches on P. oceanica ecosystem, to date little is known regarding the mechanisms such as whether (and how) plant defences or chemical cues control grazing and larval settlement within these marine habitats (Orth, 1992; Mazzella et al. 1992; Sieg and Kubanek, 2013).
In laboratory experiments, when offered leaves of Posidonia oceanica with epiphytes as a food source, two congeneric species of the snail Gibbula showed difference in preferences. Gibbula ardens showed preference for cyanobacteria and diatoms associated to juvenile leaves of P. oceanica while by contrast, G. umbilicaris showed preference for macroalgae (Mazzella and Russo, 1989). These feeding habits were in accordance with the arrangement of the two species along the leaves (Russo et al. 1984a) and with their temporal distribution in the meadows (Russo et al. 1984b). The findings by Mazzella and Russo (1989) brought to the idea that active habitat selection (at least microscale patterns) could be determined by some chemical cues from epiphytes (Orth, 1992) but this hypothesis was never tested.
Volatile organic compounds that play a role for the location of suitable habitats by several aquatic organisms in freshwater benthic systems (Höckelmann et al. 2004; Fink et al. 2006a, b; Moelzner and Fink, 2014; 2015a, b) may modulate co-evolutionary processes similar to those fdescribed for terrestrial habitats (Kessler and Baldwin, 2001) also within Posidonia oceanica beds.
Preliminary results of the researches conducted by Jüttner et al. (2010), confirmed that VOCs extracted after cell lysis from the benthic marine diatom Cocconeis scutellum parva, (Mazzella, 1983; De Stefano et al. 2008), triggered different and sometimes unexpected reactions of attraction and repulsion in the wide spectrum of naturally associated invertebrates. The reactions at different concentrations of VOCs from C. scutellum parva suggested the important role of the chemical communication within the seagrass meadows. Moreover, a more recent research conducted by Maibam et al. (2014), suggested that volatiles extracted from three diatoms epiphytes of the seagrass P. oceanica can both elicit infochemical or toxic activity on invertebrates living in the same community. According to their findings, invertebrates seem likely to recognize as infochemicals and toxins VOCs from epiphytes that are stable components of their environments more than VOCs from species of other habitats although closely related (Maibam et al. 2014).
Despite these recent findings, however a full understanding to what extent the “chemical ecology” governs plant-epiphytes and epiphytes-animals interactions and to what extent these interactions may shape and structure the entire seagrass community, is still highly required (review by Sieg and Kubanek, 2013 and references therein) also in prevision of the future global climate change (Maibam et al. 2015).
Therefore, some critical questions should be addressed:
- Do the epiphytes of Posidonia oceanica have their own “odour” in terms of volatile organic compounds and to what extent the concentration of this volatiles matter? May epiphytes’ VOCs work as infochemicals?
- Do associated mesograzers of Posidonia oceanica “respond” to VOCs of epiphytes based on their taxonomy relatedness? Otherwise, do they respond to VOCs according trophic habits?
- Is the selection of microhabitat by mesograzers based on infochemicals derived from the epiphytes?
- How these putative infochemicals shape the community of mesograzers within P. oceanica meadows and to what extent the global climate change will affect their roles as infochemicals in this ecosystem?
illustration not visible in this excerpt
Figure 1.10 Example of some mesograzers associated to Posidonia oceanica: (left) the shrimp Hippolyte inermis, (centre) the hermit crab Cestopagurus timidus, and (right) the mollusc Bittium latreillii.
1.10 Research aims and objectives
Purpose of this thesis is to investigate the effects of volatile organic compounds from the epiphytes of the seagrass Posidonia oceanica on the behaviour of naturally associated benthic invertebrates and further clarify how their behavioural responses to these putative infochemicals may affect the spatial organization of this stable community and the ecosystem functions.
Here, a series of experiments investigate, via choice assays, the responses of several benthic mesograzers to the extracted VOC bouquets (that may be considered as “odours” in the aquatic environment, according to Fink, 2007 and Jüttner et al. 2010) of epiphytes under naturally expected ecological concentrations. In order to evaluate whether invertebrates’ behavioural responses to algal volatiles could refer to their taxonomy or to their trophic ecology, several invertebrates ascribed to different taxonomic groups, and characterized by a range of trophic preferences, were considered. Simultaneously, I evaluated the efficacy of two different methods for choices tests and I further devised a species-specific standardization to avoid misleading results due to random walking behaviour. I then considered, using a crossed experimental design with sympatric epiphytes species and invertebrates from P. oceanica and from rocky pools, whether the behavioural responses to epiphyte-borne volatiles could be considered as driven by environmental plasticity more than taxonomic relatedness. In addition, to clarify whether the VOCs from epiphytes could be really considered as food-finding infochemicals or they could indicate the presence of putative ‘toxins’ (Maibam et al. 2014), I carried out a feeding experiment using the whole algae. Moreover, in order to attempt whether any ‘info-disruption’ of behaviour could be determined by lower pH- high pCO2 and how this stressor could affect the responses of invertebrates’ community to infochemicals of P. oceanica, I also tested the crossed experimental design indicated above within the range of expected values of pH for the end of this century.
In detail, the objectives addressed in each chapter are summarized as following:
In Chapter 2, I carried out an experiment in which several invertebrates’ responses (i.e. molluscs, decapods and polychaetes) to VOC bouquets of generalist epiphytes isolated from P. oceanica, were evaluated by means of two different methods of assays at two concentrations, and at different times. This experiment was established in order to identify: (i) whether extracted VOCs determine behavioural choices on benthic invertebrates depending on concentrations tested, making volatiles as putative ‘infochemicals’, (ii) which method of investigation between static choice chambers and straight flow-through flumes is appropriate to evaluate behavioural choices of small benthic invertebrates within fixed time and concentration gradients, and (iii) which invertebrates can be considered more effective as models for further investigations (due to high numerical abundance and easy handling under laboratory conditions). Moreover, considering generalist epiphytic volatiles but different taxonomic groups and trophic habits of invertebrates, I evaluated to what extent behavioural responses to odours refer to the taxonomy or to trophic ecology.
In Chapter 3 according to the results obtained from the previous experiment, I considered only molluscs and decapods as study organisms taking into account their body features within the recording time of behavioural choices and I increased the number of replicates and total individuals according to a species-specific behavioural standardization as control. Moreover, I used here only static chambers as an appropriate method for choice investigations with small benthic invertebrates and I increased the concentrations of different epiphyte-borne volatiles. The core aim of this experiment was to clarify, by means of choices records, how the volatile infochemicals could affect the organization of P. oceanica at the community level. For this reason, I tried to assess whether the responses of invertebrates to VOCs as infochemicals were linked to the environmental infochemicals that animals could known or whether their responses were more related to their taxonomy. Specifically, referring to two general concepts of analogy and homology from comparative biology (Hall, 1994; Sanderson and Hufford, 1996; Scotland and Pennington, 2000), I also expected that if taxonomically related mesograzers responded in the same way to VOCs, irrespective of their preferred microhabitat, a likely phylogenetic explanation could be involved in the recognition of infochemicals by the associated mesograzers of P. oceanica ecosystem (I referred to this as “Taxonomic hypothesis”). On the contrary, if mesograzers responded directionally to VOCs from the sympatric epiphytes compared to those from allopatric epiphytes, an environmental ability in the recognition of infochemicals could be involved (I referred to this as “Environmental hypothesis”). Therefore, I considered as sympatric from P.oceanica and generalists from rocky shore the preferential habitats of both the epiphytes and the invertebrates. Moreover, within decapods and molluscs, I also considered their taxonomic relatedness choosing three different species, two of which were more taxonomically related each other than the third one.
In Chapter 4, the experiment was built upon the results from chapter 3, in order to clarify whether the behavioural responses exhibited by mesograzers to epiphyte VOCs were related to their perception as food infochemicals (i.e. attraction behaviour) or whether their avoidance behaviour was due to the presence of noxious compounds. Therefore, in order to identify “the messages beyond the odours”, I experimentally investigated, through the use of separate-offer forced feeding experiment: (i) the presence of any toxic compound by measuring the survival in different food treatments and (ii) I ranked food preferences for whole algae without any odour (after lyophilisation process) compared to agar-control items.
Additionally, in Chapter 5 I investigated within the range of predicted pH levels for the end of this century, using the same experimental protocol and design as in chapter 3, whether the ocean acidification (i) has “direct effect” by changing behavioural responses of mesograzers to putative infochemicals and, (ii) how (as “indirect effect”) it will modify the structure of chemical relationships of P. oceanica-associated benthic community.
CHAPTER 2 VOCs of epiphytes from Posidonia oceanica trigger infochemicals based communication for benthic inverterbrates
2.1 INTRODUCTION
In marine habitats, many constraints (e.g. scarcity of light, depth, and long distance) influence communication and in this case animals can better rely on olfactory cues (Derby and Sorensen, 2008). These chemical cues are carrier of information (defined infochemicals; reviewed in Dicke and Sabelis, 1988) and they play ecological roles (Hay, 2009). Infochemicals are, by definition, extracellular substances and mainly they are “odorants” (Klaschka, 2008).
Odours from both animals and plants, influence foraging (Fratini et al. 2008; Nevitt, 2008; Moelzner and Fink, 2015a,b) and feeding (Zimmer et al. 1999; Finelli et al. 2000; Wiesemeier et al. 2007; Moelzner and Fink, 2014) as well as predation and competition (Stabell et al. 2003; Keppel and Scrosati, 2004; Coleman et al. 2007a, b; McClintock et al. 2010). They direct individuals to mate (Hardege et al. 1996) and reproduce (Pearce and Scheib ling, 1990), and determine habitat selection (Atema et al. 2002; Gerlach et al. 2007; James et al. 2008).
Among these odours, volatile organic compounds (VOCs) with low-molecular and low hydrophilicity, often produced upon wounding of algae and cyanobacteria, are responsible of unpleasant odours (e.g. dimethyl sulphide and geosmin; Watson, 2004). Although the processes involved in the production of these odour bouquets have been widely investigated in planktonic and benthic freshwater cyanobacteria and marine diatoms (Jüttner, 1995, 2005; Pohnert, 2000; Pohnert et al. 2002; Leflaive and Ten-Hage, 2009), only a reduced number of studies has clarified their biological and ecological functions (reviewed by Fink, 2007).
The chemical communication via volatiles can be thought as a complicated dynamic web (Vos et al. 2006) where the behaviour and the physiology of organisms are strongly influenced by the composition (e.g. single compounds or unique blends of several substances in mixtures), the concentration, and the hydrodynamic transport of these compounds (Zimmer and Butman, 2000; Atema, 2012). Indeed, the specific chemical composition of these odour blends may have the role of infochemicals only for one species and not for other organisms, even closely related and congeneric (Höckelmann et al. 2004; Fratini et al. 2008). Moreover, the functions of these compounds can vary at different concentrations. Too high concentration of food odours can repel animals while too low concentrations cannot be even detected (Atema, 1995). Besides, hydrodynamic physical limitations as for instance, the molecular diffusion or the turbulent flow may control the delivery of these chemical cues and consequently influence the chemical signalling processes (Weissburg and Zimmer-Faust, 1994; Finelli et al. 2000; Mead et al. 2003; Moore and Crimaldi, 2004).
Specifically, the chemical features of VOCs from primary producers make them perceptible both in water and at air-water interfaces (Fink, 2007) often forming more concentrated and stable microzones of these volatiles in benthic ecosystems (Höckelmann et al. 2004; Fink et al. 2006a; Fratini et al. 2008) compared to the planktonic ones (Blackburn et al. 1998; Steinke et al. 2002).
Within highly bio-diversified benthic habitats such as the seagrass Posidonia ocenica (Delile L.), invertebrate grazers like molluscs, crustaceans, and polychaetes control the patterns of epiphytic assemblages and indirectly the primary production associated to this ecosystem (Bologna and Heck, 1999; Prado et al. 2007). The epiphytes of P. oceanica are layers of micro- (e.g. cyanobacteria and diatoms) and macroalgae (e.g. encrusting and filamentous seaweeds) coating seagrass leaves (Mazzella and Ott, 1984). They represent food sources for several animal taxa (Gambi et al. 1992; Mazzella et al. 1995; Scipione et al. 1996).
The use of chemical cues, for instance, in foraging behaviour of marine benthic grazers has been widely demonstrated (e.g. gastropods: Chapman and Underwood, 1992; Fink et al. 2006a, b; Coleman et al. 2007a; decapods: Brawley, 1992; Finelli et al. 2000; Gherardi et al. 2005; Toth and Pavia, 2007; polychaete worms: Hardege et al. 1996; Höckelmann et al. 2004). However, within P. oceanica ecosystem, only two papers investigated the putative role of VOCs from benthic diatoms as putative infochemicals and toxins (Jüttner et al. 2010; Maibam et al. 2014). These authors showed that VOCs from the wounded benthic diatoms, Cocconeis scutellum parva (Grunow) Cleve and Diploneis sp. (Ehrenberg) Cleve, two of the most abundant epiphytes upon the leaves of P. oceanica, triggered unpredictable behaviours in grazer invertebrates belonging to the same habitat according to the animal ecology and to the concentrations of VOCs tested. However, to date, it has not been investigated yet how epiphyte-borne chemical cues, might be involved within the observed patterns of seagrass-animal-epiphyte interactions at the community level (Orth and Van Montfrans, 1984; Orth, 1992).
To clarify how these chemical cues may influence the patterns of animal-epiphyte interactions within P. oceanica, we hypothesize that VOCs released upon cell disintegration from epiphytic seaweeds can be used differently as infochemicals by several invertebrates belonging to different taxonomic groups living in the same habitat. Therefore, we investigated through behavioural choice assays (Vet, 1999), the responses of nine species of invertebrates, deliberately chosen across three taxonomic groups (i.e. gastropods, decapods and polychaetes) to VOCs from three epiphytic seaweeds (i.e. Enteromorpha prolifera, (O.F. Müller) J. Agardh, Colaconema daviesii (Dillwyn) Stegenga, and a strain of unidentified Cyanobacteria) isolated from leaves of P. oceanica.
Moreover, according to the assumptions regarding the concentration and the hydrodynamic transport of VOCs, we considered two different experimental estimation of concentrations (i.e. 1% and 10% by volume), and two developed standardized assays (i.e. ‘no-flow’ choice chambers and straight flow-through flume; Jüttner et al. 2010; Atema et al. 2002) evaluating also the existence of gradients within these experimental devices.
2.2 MATERIALS AND METHODS
2.2.1 Isolation and culture of benthic epiphytes from the seagrass Posidonia oceanica
Three species of epiphytes isolated from Posidonia oceanica leaves, were used for the present study: Enteromorpha prolifera, Colaconema daviesii, and an unidentified species of Cyanobacteria.
Leaves of Posidonia oceanica were collected by scuba divers, at 5 m depth in Lacco Ameno (Ischia Island, Naples, Italy, Fig.2.1) during the spring 2012. The leaves were examined under a stereomicroscope (Leica MZ6), and different fragments of epiphytes were collected. These fragments were moved into sterile 12-multiwell plates with sterile filtered seawater (SFSW; Whatman 0.22 pm) enriched with Provasoli culture medium (Prov50, Bigelow Lab.; Provasoli et al. 1957) and with germanium dioxide (GeO2) at the final concentration of 6 mg L-1.
Germanium dioxide was added to the culture medium for the first two weeks in order to reduce the growth of diatoms and bacteria, which could impair the development of the epiphytes (Markham and Hagmeier, 1982). After two months and several transfers in new sterile medium, pure isolates were obtained of the above-mentioned species. To obtain a sufficient biomass, multiple cultures of each strain were grown in Petri wells (12 cm diameter and 7 cm deep) containing SFSW with Provasoli, in thermostatic chamber at 18°C with 12:12 photoperiod under Sylvania GroLux 36W phytostimulant fluorescent lamps producing on the whole a light intensity of 140 μmol photons m" s' . Freshly prepared medium was replaced every week until sufficient biomasses were obtained. Each biomass was frozen at -20°C until the extraction of the algal odours.
All these procedures were conducted under sterile conditions to reduce the chance of bacterial and diatoms contamination.
2.2.2 Extraction of algal odours: VOCs
Dr. P. Fink (Aquatic Chemical Ecology Department, University of Cologne, Germany) carried out the extractions of VOCs from the three epiphytes. To simulate the wounding of algal cells and determine the release of the VOC bouquets, the frozen algal biomasses (3 g, 3.57 g and 3.92 g fresh weight for each algal species E. prolifera, C. daviesii and Cyanobacteria, respectively) were disintegrated by a cycle of freezing and thawing in 40 mL of FSW each (Whatmann 0.22 pm). This process of physiological disintegration leads to the activation of the lipoxygenase cascade and the formation of volatile compounds (Pohnert, 2002; Jüttner, 2005).
The samples were then transferred in 100 mL round bottom flasks with 25% of NaCl and VOCs were extracted and concentrated by closed-loop stripping at 22°C for 45 min (Jüttner, 1988; Fink et al. 2006a). After this time interval, most VOCs were absorbed on a Tenax TA cartridge (Fink et al. 2006a) which was then eluted with 6 mL of diethyl ether. Thus, the ether was removed by evaporation using nitrogen gas (N2, grade 5.0) and the residues were dissolved into 300 pl of pure ethanol (EtOH), obtaining approximately 100 pl of extract per each gr of fresh weight. These odours re-solubilised in ethanol, were then dissolved in agarose gel at two experimental concentrations (see below for preparation of gels).
Control extracts were obtained applying the same procedures on FSW without algae and they were used as control treatments in the behavioural assays. All samples were stored at -80°C in gas-tight vials to avoid any loss of VOCs (Fink et al. 2006a) until gels preparation.
2.2.3 Study organisms: animal sampling and biology
Animal sampling was carried out in April 2013, in a Posidonia oceanica meadow in Lacco Ameno (Ischia, Italy; Fig. 2.1). This meadow is settled on a matte and extends from 1 to 32 m in depth (Mazzella et al. 1989).
The benthic invertebrates for the experiments were collected by a circular plankton net (1 m frame diameter; 100 μm of mesh size), trawled above the surface of the leaves of P. oceanica from a boat at a depth between 4-11 m, with a recorded temperature of 17.7°C (standard deviation, ± 0.21, n = 4).
The fauna associated to Posidonia oceanica was then transferred to the Benthic Ecology Laboratory ‘Villa Dohm’ (Ischia, Italy) where it was sorted under a stereomicroscope (Leica MZ6) into 9 final species, considering only species for which at least 30 individuals with roughly the same size were available. In particular, species belonging to different taxonomic groups and feeding habits ranging from detritivores to omnivores were considered: 4 gastropods, 2 decapods, 3 polychaetes (see Table 2.1. for the ecology of species).
All the animals collected at different depths were pooled and transferred into separated aerated drip trays with seawater re-circulating flow at field temperature, salinity, and pH.
No animals were injured during the experiments and after a day in flowing seawater for acclimation, they were released back to the sea.
illustration not visible in this excerpt
Figure 2.1 The sampling site in Lacco Ameno (40°45’ N/13°50’ E) at the north east of Ischia Island (Tyrrhenian Sea, Italy) is indicated by a grey square.
illustration not visible in this excerpt
Table 2.1 List of invertebrates used in behavioural assays and their taxonomy, microhabitat preference, and feeding guild according to Gambi et al. (1992). DF= deposit feeder that feeds on surface detritus; HeDF= herbivore-deposit feeders, which feed on plant epiphytes and trapped organic material; He= herbivores graze on micro- and macroalgae; Ca= carnivores.
illustration not visible in this excerpt
2.2.4 Experiment 1: choice assay in static chambers
We conducted odour choice tests using static chambers according to the method described in Jüttner et al. (2010). Choice tests were carried out under controlled temperature (18°C) and light (low, ambient), using glass Petri dishes (14 cm diameter and a total volume of ~ 200 mL).
For the experiment, we used filtered seawater (Whatman 0.22 pm) to reduce the chance of having other “odours” present. Moreover, to exclude any effect due to temperature and pH, they were maintained at the same as in the field (mean ± standard deviation, 18.73 ± 0.27 °C; pHNBs= 8.08 ± 0.06 units, n=6).
According to the experimental setup described in Jüttner et al. (2010), the arenas contained: (i) one side with the symbol of “+” ranked as +2 or +1 based on the proximity to the odour source, (ii) one zone with a central circle, used for the initial deployment of individuals for the acclimation and, (iii) one side with the symbol “-“ without any odour ranked -2 and -1 (Fig. 2.2a).
The position of “+” targets in each chamber was opposed (Fig. 2.2b) in order to exclude directional effects introduced by the experimental setup. Using this setup, external factors that might have influenced the movements of the animals (e.g. light, magnetism) were randomized amongst the replicates.
The algal odours (VOCs) were included in agarose gels. The agarose gels for tests and controls were prepared before the odour choice assays. In detail, we prepared 8 different gels (2 concentrations x 4 VOCs treatments, considering both the controls and the algal VOCs extracts). We dissolved at 80°C, 1.2 gr of low melt agarose (Sigma A9045) in 200 mL of sterile FSW in flasks of 400 mL with 3.3mL of 0.1M NaOH to maintain the pH around values of 8.2-8.4. Once the agarose was dissolved, we waited until the temperature decreased reaching 28°C in order to add the different volumes of extracts without losing them due to evaporation or heat degradation.
We used two different concentrations (indicated as low and high) of volatiles to have 0.02 and 0.20 pL mL-1 of agarose representing 60 mg and 600 mg respectively of the initial fresh weight in 300 pL. Therefore, tests were performed for each alga at two different odour concentrations: 1% and 10% by volume.
To prepare the two concentrations of control gels, 5 and 45 pL of EtOH containing the control extracts obtained as described above were added in two flasks containing the agarose solutions. Similarly, for the experimental treatments almost 5 and 50 pl of volumes of EtOH for each VOCs extract were added into different flasks. Therefore, the agarose solutions were poured into Petri dishes and gelled in a refrigerator (5 °C) for 1 h prior to cutting the gel into blocks of 0.5 cm with a sterile cover slip. These blocks were inserted at the tips of 14 cm length glass capillary pipettes to avoid confusion with readings. Before the beginning of the experiment, animals were located in the central zone for 2 min of acclimation and ultimately, the capillary pipette was placed in the Petri dish according to the target (+ if the block contained the VOC and - if the block was the control).
The position of animals (number of individuals in each sector of the arena) was recorded after 5, 10, 15 min in 4 replicates with 7 individuals in each for all the nine invertebrates. Each of the three temporal records from the start of the experiment represented a pseudo-replicate. To avoid the dependence of data for further statistical analysis, before testing animals at different concentrations and at the three VOCs, we placed them back into the drip trays with seawater re-circulating flow, in order to allow them to re-acclimatize for 1hour at least.
illustration not visible in this excerpt
Figure 2.2 Experimental arena used in static choice experiments indicating (a): different sectors (from Jüttner et al. 2010); (b) Photo of the experimental setup (April, 2013).where each arena was positioned with the (+) targets opposite, to randomize directional effects.
2.2.5 Experiment 2: choice assay in straight flow-through flume
To evaluate the efficacy of two different methods for choice tests, we compared the behavioural responses of the invertebrates in two different systems. Thus, we conducted behavioural tests also in a straight flow-through flume system (Voigt and Atema, 1996).
To understand if our invertebrates were able to respond to volatiles, we used a choice flume chamber similar to the one devised by Atema et al. (2002). All the experiments were conducted in a choice system maintaining two parallel plumes not mixing until the downstream exit (Fig. 2.3). The choice chambers were in overshadowing glass, 4 mm thick. The grey colour was to avoid that any light or any operator’s movement could affect behavioural choices, while the glass material was to ensure that no other volatiles could be dissolved into water.
This apparatus was designed to conduct pairwise choice experiments with benthic invertebrates able to freely choose between two sides of the flume: one side containing the algal odour and the other containing a plume of filtered seawater. The aquaria size was adapted to the animals’ size we expected to use in our experiment, as suggested in Voigt and Atema, (1996). The flume had a working test area of 8 cm x 14 cm (length x width) and was filled with seawater to a depth of 3 cm. The plumes were generated from identical point sources, two different water containers at the upstream end of the flume. The two water streams were kept separated and parallel by a central panel and to reduce turbulence each stream flowed through glued (with no toxic silicone for aquaria and maintained under seawater flow for two days before using) drinking straws of slightly green colour to avoid any attraction due to the use of sigh. Prior to any experiments, the mean current velocity was set and measured visually with Methylen blue dye (0.25 g L-1 dissolved in ambient seawater). In the centre, the mean current velocity of the flume was ~7 cm min-1 and the plumes remained separate along the entire length of the flume (Fig. 2.4; Kroon and Housefield, 2003). The two plumes were canalized through an upstream fine mesh (0.5 mm of diameter) into the test area, and then flowed through a downstream containment mesh out in one opening to keep the water level around 3 cm and have a total volume inside the aquaria of ~882 mL.
Seawater solutions containing VOCs from each of the algal species were used as an odour attractant for experiments. We prepared 8 odour stock solutions (2 concentrations x 4 VOCs treatments, algal and controls). In particular, we used two concentrations of VOCs (also here indicated as low and high) in order to have 0.07 and 0.70 pL ml-1 assuming 0.8 mg and 8.8 mg respectively of the initial fresh weight compared to ~200 mL of volume of static chamber and proportioned to the volume that was eluted in 3 min of flume (~800 ml). These odour solutions were stored at 4°C and brought up to ambient temperature only prior to use. Both the solutions were delivered through the water inflow compartments at the upstream end with two plastic drippers placed at the top of each side.
We swapped halfway the cue sources through the trials to avoid any directional effects and as long as for the static experiment, we used 4 replicates with 7 individuals for each species. However, due to animal’s physiologic and ethologie peculiarities, it was not possible to employ all nine invertebrates, and we excluded the polychaete species from this system. These species, in fact, immediately after their release in the area test for the acclimation, passed over the upstream fine mesh and entered inside the packed drinking straws not allowing the operators to record their position.
To start the experiment, the animals were placed in the central rectangular zone of 5 cm x 2 cm (length x width, see labelled point g in Fig. 2.3B) inside the test zone and were acclimatized to the flow with simple control FSW for 3 minutes until the flow was stabilized with no drops coming out from the downstream opening. Then, the two plastic drippers were filled up with odour stock solution (+) and with control FSW solution (-). The current velocity of 7 cm min-1 allowed recording the behavioural choices at the end of the 3rd, the 4th and 5th min after the starting of the plumes. Data were recorded considering the number of animals in each side (i.e. towards and away) while the animals in the central acclimation zone were not considered.
Each of the three temporal records represented a pseudo-replicate reading. To avoid the dependence of data for further statistical analysis, before testing animals at different concentrations and at the three VOCs, we kept the animals into the drip trays with seawater re-circulating flow, at least for 1 hour, to ensure their acclimation.
illustration not visible in this excerpt
Figure 2.3 Straight flow-through flume system designed according to Atema et al. (2002): (A) project in 3D (gently courtesy from M. Mutalipassi). Overall, the chamber size was 21 cm x 14 cm x 10 cm (length x width x height) in overshadowing glass, 4 mm thick. The arrows indicate the test area and the two inflow compartments. (B) The scheme shows the choice flume system from the upside view with: c) water inflow compartments (with arrows); d) packed drink straws to reduce the turbulence; e) barrier-separated channels to further laminate flow; f) upstream fine mesh (0.5 mm) net to contain animals; g) test area; h) starting point for the acclimation; i) downstream containment net and j) drain area with the central drain opening.
illustration not visible in this excerpt
Figure 2.4 Flumes are visualized here in the (a) upside and (b) lateral view for photographic purposes using 0.25 g L-1 of Methylen blu dye dissolved in ambient seawater.
2.2.6 Baseline choice assays
In order to check the behaviour of the benthic invertebrates in the absence of odours, we performed a baseline control experiment. In particular, we devised the choice tests with the benthic invertebrates as explained above for the two main tests. However, in the case of static chambers both agarose gels were prepared without any odour while for the flume system in both the sides FSW was used. For each method and invertebrates, we used the same number of replicates (n = 4) with the same number of individuals in each replicate (7 individuals).
Our aim was to check if, in the absence of odours, the benthic invertebrates showed any side bias. Therefore, a preferred side was individually determined and accounted for, in the statistical analysis. The data obtained by these tests were analyzed as for the main data sets and results represented our μο for comparisons with results obtained from VOCs assays in one sample t-test (Höckelmann et al. 2004; see below paragraph 2.3.3 for details).
2.2.7 Distribution of VOCs in static chambers
The static experimental arenas and the position of the agarose blocks, as above specified, were set in order to produce a gradient of the VOCs concentration during the tests. In order to verify that volatiles could be stratified according to the linear function of odour concentration gradients (Sutton, 1953), we performed an additional measurement to track the distribution of odours in the arena. For this purpose, agarose blocks were prepared with a known volatile compound (2-trans-4-trans-decadienal by Sigma-Aldrich, an aldehyde produced upon wounding by planktonic diatoms; Wichard et al. 2005).
We dissolved 150 μΗ of decadienal in 4 mL of methanol and then we added these into 200 mL of freshly prepared agarose, as described above. The mixture was then refrigerated and we cut gel into 0.5 cm blocks, as already described for the experimental VOCs choice tests, in order to have a final concentration of 240 μg mL-1. The blocks were then positioned in the (+) targets of the experimental arena. A special grid was drawn and was placed under the experimental arena with 1 cm nodes and 30 sampling points randomly assigned (Maibam et al. 2015). A set of three independent replicates (collected in replicate vessels) was sampled every 5 min by means of an automatic pipette, and 1 mL of the medium was collected, exactly in each sampling point over the grid. These collections were repeated at 5, 10, 15 and 20 min. These sample solutions were then analyzed using a spectrophotometer (Hewlett Packard 8453 spectrophotometer) and the concentration of decadienal was calculated taking into account its molar extinction (epsylon) in methanol that is 31,000. All concentrations were recorded in a matrix indicating the position of samples and the averages of the three replicated samples at each time in each point, were calculated. These values were computed using the Kriging technique (Matheron, 1969) that allows for a spatial representation of the concentration, considered as a stationary phenomenon.
2.2.8 Distribution of VOCs in straight flow-through flume
In order to verify that volatiles were spatially distributed according to the predictions with the water flow (Vickers, 2000), we performed a similar additional measurement as described above, to track the distribution of odours in the flume system. For this purpose, we prepared a Methylen blue dye solution with a known concentration (Voigt and Atema, 1996; Dixson et al. 2015) and we settled the current velocity as the same of the experimental test (i.e. 7 cm min-1) assuming that the “odour” was not flowed out within the 5 min of test. In order to have the same concentration as the decadienal solution, we dissolved 248 mg L-1 of Methylen blue dye in ambient seawater (Fig.2.4).
[...]
-
-
-
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X. -
-
¡Carge sus propios textos! Gane dinero y un iPhone X.