Metabolomics Reveals Cryptic Interactive Effects of Species

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Metabolomics reveals cryptic interactive effects of species interactions and environmental stress on nitrogen and sulfur metabolism in seagrass Harald Hasler-Sheetal, Max C.N. Castorani, Ronnie N. Glud, Don E. Canfield, and Marianne Holmer Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b04647 • Publication Date (Web): 12 Oct 2016 Downloaded from http://pubs.acs.org on October 13, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27

Environmental Science & Technology

3

Metabolomics reveals cryptic interactive effects of species interactions and environmental stress on nitrogen and sulfur metabolism in seagrass

4 5

Harald Hasler-Sheetal* 1,2,3, Max C.N. Castorani4, Ronnie N. Glud,1,2, 5, 6 , Donald E. Canfield2, Marianne Holmer1

1 2

6 7 8 9 10 11 12 13

1

Department of Biology, University of Southern Denmark, Denmark. Nordic Center for Earth Evolution (NordCEE), University of Southern Denmark, Denmark. 3 VILLUM Center for Bioanalytical Sciences, University of Southern Denmark, Denmark. 4 Marine Science Institute, University of California, Santa Barbara, USA. 5 Scottish Association for Marine Science, Oban PA37 1QA, UK 6 University of Aarhus, Arctic Research Centre, Aarhus, Denmark 2

14 15

ABSTRACT

16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Eutrophication of estuaries and coastal seas is accelerating, increasing light stress on subtidal marine plants and changing their interactions with other species. To date we have limited understanding of how such variations in environmental and biological stress modifies the impact of interactions among foundational species and eventually affect ecosystem health. Here we used metabolomics to assess the impact of light reductions on interactions between the seagrass Zostera marina, an important habitat-forming marine plant, and the abundant and commercially important blue mussel Mytilus edulis. Plant performance varied with light availability but were unaffected by the presence of mussels. Metabolomic analysis, on the other hand, revealed an interaction between light availability and presence of M. edulis on seagrass metabolism. Under high light, mussels stimulated seagrass nitrogen and energy metabolism. Conversely, in low light mussels impeded nitrogen and energy metabolism, and enhanced responses against sulfide toxicity, causing inhibited oxidative energy metabolism and tissue degradation. Metabolomic analysis thereby revealed cryptic changes to seagrass condition that could not be detected by traditional approaches. Our findings suggest that coastal eutrophication and associated reductions in light may shift seagrass-bivalve interactions from mutualistic to antagonistic, which is important for conservation management of seagrass meadows.

ACS Paragon Plus Environment

1


Page 2 of 27

32

INTRODUCTION

33

Among the most integral tasks for ecologists are to investigate and understand the way species

34

interact with one another (e.g., mutualism, antagonism, competition) and the strength of these

35

interactions. Species that physically or chemically modify environmental stressors or resource

36

availability, such as ecosystem engineers1, may simultaneously exacerbate and alleviate several

37

stressors2. This results in highly complex species interactions that vary depending upon

38

environmental conditions3. However, understanding the role of environmental context in

39

mediating species interactions has historically been complicated by difficulties in quantifying

40

sublethal physiological stress4. Fortunately advances in ecological metabolomics are providing

41

novel solutions to this problem because stress-related effects are instantly reflected in the

42

metabolic functioning of key organisms5, providing new, practical analytical methods to assess

43

sublethal stress in ecosystem engineers6. Nevertheless, to date there have been no empirical

44

evaluations of the potential for environmental conditions to mediate the metabolic mechanisms

45

underlying species interaction.

46

Estuaries are ideal ecosystems to test this hypothesis because they support abundant and

47

important ecosystem engineers, such as suspension-feeding bivalves, that can have context-

48

dependent impacts on their surrounding community3,7,8. For example, bivalves may have either

49

positive or negative impacts on co-occurring seagrasses depending on environmental conditions

50

(reviewed in Castorani et al.9). Seagrasses are submerged marine angiosperms essential for

51

coastal ecosystem function by creating habitat for many species of invertebrates and fishes,

52

providing food for larger resident and migratory species, increasing secondary production,

53

enhancing biodiversity, stabilizing sediments, reducing coastal erosion, and recycling

54

nutrients10,11.


2

Page 3 of 27


55

When plants face unfavorable environmental conditions, environmental and biological stress

56

perturbs plant metabolism. Consequently, the metabolic network needs to be reprogrammed to

57

adapt to the prevailing conditions and maintain essential metabolic pathways. Metabolomics

58

aims to profile the entire set of low molecular weight metabolites—the metabolome—which in

59

fact is determined by the physiological, developmental, or environmental state of an organism12.

60

Therefore metabolomics represents an exceptional tool to assess effect of environmental end

61

biological stress on organisms by quantifying the capacity to react, acclimatize, and/or adapt to

62

changing environments5,13,14. Metabolomics techniques such as liquid chromatograpy–high

63

resolution mass spectroscopy (LC-MS) and gas chromatography–accurate-mass mass

64

spectroscopy (GC-MS) measure hundreds to potentially thousands of metabolites15. Consequent

65

data analysis of these metabolites allow rapid and accurate separation of samples according to

66

their environmental exposure, yielding characteristic metabolic fingerprints directly related to the

67

prevailing phenotype15. Metabolic fingerprinting allows a rapid and holistic classification of the

68

samples according to their origin and environmental or biological exposure14, and does not

69

require the identification of every metabolite present. The data can, however, also be screened

70

for specific metabolic patterns or pathway types through exploratory data analysis

71

(chemometrics)15, which prerequisites identification of the metabolites and provides in depth

72

information about prevailing metabolic responses14.

73

Despite the analytical power of metabolomics to assess and understand the effect of

74

environmental conditions on organisms, the field of seagrass metabolomics is still in

75

development and only two studies have applied metabolomics to assess temperature16 and

76

hypoxia6 stress on seagrasses, respectively. To date it is unclear what effect bivalves and light

77

availability have on the metabolic functioning of seagrasses. Determining the influence of


3


Page 4 of 27

78

bivalves on seagrasses is important because seagrass distributions are estimated to be declining

79

~7% per year globally due in large part to light stress caused by coastal eutrophication10,11. In

80

oligotrophic systems, bivalves can facilitate seagrasses by increasing nutrient availability

81

through deposition of feces and pseudo-feces17. However, bivalves can also inhibit seagrass

82

growth by enriching sediments with organic matter and increasing concentrations of toxic

83

sulfides18. These variable interaction effects are modulated by environmental drivers such as

84

light, hydrodynamics, and temperature and cannot be easily disentangled by traditional (non-

85

metabolomic) approaches19,9,20. Using traditional plant metrics (e.g., growth, survival), a short-

86

term study demonstrated a predictable influence of light on seagrass condition but failed to detect

87

any effect of bivalves9. However, cryptic seagrass-bivalve interactions might be discerned using

88

new metabolomic tools. Thus, here we tested whether light - the primary factor of seagrass

89

productivity9,21 - modulates bivalve-seagrass interactions through metabolic mechanisms. We

90

used metabolomics to test for independent and interactive effects of light availability and bivalve

91

presence on the seagrass metabolome and to relate environmental parameters to primary energy

92

and nitrogen metabolism, and to early apoptosis.

93 94

EXPERIMENTAL METHODS

95

Study system

96

The epibenthic suspension-feeding blue mussel, Mytilus edulis L., frequently co-occurs with

97

Zostera marina L., in estuaries and shallow coastal water of the temperate North Atlantic Ocean,

98

North Sea, and Baltic Sea22,23. In this study we used sediment, water, mussels and plants from the

99

Danish straits connecting the Baltic Sea with the Nord Sea. In this region Z. marina and M.


4

Page 5 of 27


100

edulis are widely distributed and often co-occuring18. Danish coastal waters in this region are

101

often eutrophic and turbid resulting in fluctuating light availability24.

102

Specimen collection and exposure

103

To assess the role of light availability in mediating habitat modification by blue mussels Mytilus

104

edulis and impacts on Zostera marina, we manipulated light availability (high vs. low) and

105

mussel abundance (present vs. absent) in a factorial design in an indoor mesocosm experiment.

106

Briefly, sediment, seawater, seagrass, and mussels were collected from coastal field sites in

107

Denmark, transplanted into mesocosms, and maintained in a controlled recirculating seawater

108

system (see Hasler-Sheetal et al.6 and Castorani et al.9 for details). Mesocosmos (N=24) were

109

established and the temperature (14.4 °C), salinity (13.4 ± 0.8), water column oxygen saturation

110

(100%) and water column nutrients (11.5 ± 7.7 µmol NH4+ L−1; 0.293 ± 0.026 µmol NO3- L−1)

111

were maintained constant and similar in all mesocosms. To mimic low light conditions in

112

eutrophic estuaries, half of the mesocosms were exposed to light levels slightly below 100 µmol

113

photons s-1 m-2 (the light saturation level for Z. marina is ~100 µmol photons s-1 m-2) limiting Z.

114

marina growth25. The other half of the mesocosms received high light levels ~550 µmol photons

115

s-1 m-2) that do not limit Z. marina growth. During night (12 h) all systems were completely dark.

116

To half of the mesocosms we added M. edulis at densities observed in the field (891 m-2)18. After

117

21 days of exposure the 24 mesocosms were harvested and processed as decribed in Hasler-

118

Sheetal et al. In brief plants were randomly harvested, separated in leaves, rhizome and roots

119

(total of 72 samples), snap frozen in liquid nitrogen lyophilized for 48h and homogenized for

120

further analysis. Elemental sulfur in tissues, a proxy for sediment sulfide intrusion into seagrass

121

tissues26, was measured following Frederiksen et al.27. A detailed experimental description is

122

presented in the supplementary material.


5


Page 6 of 27

123

Metabolomics

124

Metabolites were extracted from 72 snap-frozen, lyophilized and homogenized seagrass samples

125

(methanol/water 5:1 [v/v]), the extracts were dried and subsequently resuspended for LC-MS or

126

derivatised prior GC-MS analysis, respectively. Metabolites were separated by reverse phase

127

(RP), hydrophobic interaction chromatography (HILIC), and gas chromatography (GC), and

128

detected by quadrupole time-of-flight mass-spectroscopy (qTOF-MS) in ESI+/- (Electrospray

129

ionization) and electron ionization, respectively. In total we used five different analytical

130

procedures to resolve the metabolome in leaves, rhizome, and roots of Z. marina reverse: RP,

131

HILIC both in ESI+/- qTOF-MS and GC-qTOF-MS. Data inspection, data mining, annotation,

132

and interpretation were done in MassHunter, Profinder, and Mass Profiler Professional (Agilent

133

Technologies, Santa Clara, California, USA). The LC-MS data was used to assess metabolic

134

fingerprints and the GC-MS data to assess pathway analysis. Ceramide (validated as d18:1/12:0)

135

with a mass of 481.4495)28 and sphigosine-1-phosphate (S1P) were assessed by RP-LC-MS and

136

confirmed by comparison with standards. A detailed description of the analytical setup and the

137

metabolic profiling is given in the supplementary material.

138 139

Data analysis

140

Peak areas were standardized for sample weight and to the internal standard, missing values were

141

imputed by k-nearest neighbour method29 and later log2(x+1) transformed and baselined by unit

142

scaling (mean-centered and divided by the standard deviation of each variable). To exclude false

143

positives, only entities with a coefficient of variation (CV) less than 35% and present in at least

144

80% of the quality control samples (QC) samples were used for analysis. The effects of light

145

availability and mussel presence were compared using analysis of variance (ANOVA; α = 0.05)


6

Page 7 of 27


146

and Tukey’s post-hoc test. We applied a false discovery rate correction using the Benjamini–

147

Hochberg30,31 method, an adjusted p value < 0.05 was considered significant. To graphically

148

visualize the data we used heatmaps illustrating Ward clustering of the Euclidian distances

149

between the treatment groups and metabolites, respectively. In addition, we used principal

150

component analysis (PCA) to visualize and summarize the metabolic response of Z. marina to

151

light availability and mussel presence. To identify the most influential metabolites in separating

152

the samples along the principal components (PC), Covariance vs. Correlation plots (CC-plots) of

153

all 3 components were inspected32. We mined the metabolite matrix for correlations with

154

ecological parameters like sediment biogeochemistry and plant performance by Spearman

155

correlation. An alpha value of 0.05 was applied consistently.

156 157

RESULTS AND DISCUSSION

158

Metabolic profiling and multivariate analysis

159

We detected 92,478 mass spectral features in Z. marina roots, rhizomes and leaves and of these

160

5,541 passed our quality control filters (present in 80% of the quality control (QC) samples and

161

with a CV < 35%). The 5,541 reproducibly detected metabolite entities represents a large

162

number33 and suggest a good and robust coverage of the Z. marina metabolome and thus were

163

used for metabolomic fingerprinting without further annotation. The visualization of the filtered

164

and unannotated metabolome in a heatmap showed clear treatment-specific differences (Figure

165

1). Light intensity grouped the samples in two clusters and in each of these clusters mussel

166

presence formed two sub-clusters (x-axis in Figure 1). The clustering of metabolites showed

167

distinct and treatment specific metabolite clusters (y-axis in Figure 1). Light modified the effect

168

of mussel presence on the metabolome indicated by the distinct condensed and mussel specific


7


Page 8 of 27

169

clustering under either low or high light conditions, suggesting interactive physiological

170

responses.

171

To visualize treatment specific effects, a PCA was conducted and showed a clear treatment

172

related clustering of the samples in all tissues and analytical conditions (Figure 2). Light

173

exposure was the main variance observed between the samples, indicated by a light-depended

174

separation of samples on component 1 (accounting for 34-39% of the metabolic variation in the

175

data set) and governed by metabolites of the energy metabolism. In particular, energy source and

176

storage carbohydrates responded to light reduction (Figure 2; Table S1, Figure S1).

177

These findings are in line with previous studies showing major effects of light intensity on Z.

178

marina energy metabolism25. Remarkably, light exposure modified the effect of mussels on the

179

metabolome as indicated by mussel dependent grouping of samples on component 2 under high

180

light or respective component 3 under low light, explaining 27-24% and 22-16% of the variation

181

in the data set respectively (Figure 2). Different metabolite sets were governing the separation on

182

each component (Figure 2), illustrating divergent physiological responses of Z. marina to the

183

presence of mussels under low and high light availability. Metabolites governing mussel-

184

dependent separation under high light were associated with nitrogen metabolism (Figure 2; Table

185

S1) where as under low light the mussel treatment was associated with metabolites of glycolysis

186

and TCA cycle (Figure 2; Table S1). These light-dependent metabolic responses to mussel

187

presence were not reflected in traditional plant performance metrics such as leaf growth rate,

188

shoot density, and soluble sugars (data presented in Castorani et al.9; Table S3). The plant

189

performance metrics were depressed by low light availability but, in contrast to the metabolome,

190

were unaffected by mussels. This suggests that either the metabolic shift caused by mussel

191

presence did not affect Z. marina performance or that the duration of the experiment (21 days)


8

Page 9 of 27


192

was not long enough to manifest traditional performance metrics. The latter explanation is likely

193

because indeed high levels of sulfide intrusion26,34 were observed in Z. marina grown in low light

194

with mussels9 (Table S2) that typically lead to sulfide toxicity and impaired plant

195

performance9,26. To better understand the potential negative effects of mussels on Z. marina

196

under low light, we explored the primary energy metabolism, associated nitrogen assimilation

197

pathways, and entities of apoptosis.

198 199

Growth regulation and early apoptosis

200

Several metabolites of the primary energy metabolism (glycolysis, TCA, and the associated

201

amino acid pathways) were affected by one and two-way interactions of light and mussels

202

(Figure 3). The general decrease of glucose and fructose under light depletion (Figure 3)

203

indicates energy deprivation, most likely due to lower rates of photosynthesis. This was

204

previously described in plants after prolonged periods of decreased photosynthesis35. The even

205

greater decrease of carbohydrates in Z. marina under low light and mussel presence (Figure 3)

206

could further be related to increased glycolysis depleting glucose and fructose levels. Increased

207

pyruvate and the decreased citrate levels (Figure 3) indicate that pyruvate is not sufficiently

208

feeding the TCA cycle under low light and mussel presence, thus limiting the carbon flux via

209

pyruvate, acetyl-CoA, and citrate into the TCA cycle, and ultimately leading to energy

210

deprivation. The concomitant increase of lactate (Figure 3) indicates the presence of lactic

211

fermentation leading to phytotoxic cytosolic acidification6,36. These findings suggest that (1)

212

light-dependent energy deprivation in low light and (2) inhibition of oxidative energy

213

metabolism in seagrasses under low light and co-occurred with the presence of bivalves.

214

However, it has been proposed that seagrasses can compensate for energy deprivation by (1)


9


Page 10 of 27

215

increasing the glycolytic flux and (2) utilization of glutamine-derived carbon via the GABA

216

shunt6,37, which also shunts excess pyruvate to alanine and mitigates cytosolic acidification6. In

217

agreement with these mechanisms, we observed an associated increase of lactate and pyruvate as

218

fructose and glucose both decreased with changes in alanine, GABA and glutamine levels

219

(Figure 3). This indeed implies that increased glycolytic flux drained sugar pools, a non-

220

functioning TCA-cycle caused energy deprivation, and a GABA shunt compensated for energy

221

deprivation and mitigated cytosolic acidification38.

222

The presence of mussels under low light induced accumulation of several amino acids

223

(alanine, proline, threonine, valine) and depletion of the TCA cycle derived amino acids

224

(glutamine, serine, tyrosine and glycine) in eelgrass tissues (Figure 3). This reprograming of the

225

amino acid profile is another indicator of the inhibition of oxidative energy metabolism39,40.

226

Overall, decreased levels of carbohydrates, increased lactic fermentation, and presence of

227

mitigation pathways clearly indicate impaired energy metabolism of seagrasses that co-occur

228

with mussels under low light.

229

Nitrogen assimilation in plants is highly complex, being controlled by hormones and levels of

230

sugars, organic acids, and amino acids41. However, glutamate is a key source for amino acid

231

synthesis, and acts as hub for nitrogen metabolism by shuttling reduced nitrogen into nitrogen

232

assimilation, and is the primary product of ammonium and nitrate assimilation. We found that

233

the presence of mussels increased glutamate and glutamine levels in high light, and decreased

234

them in low light (Figure 3). This indicates that light can modify the effect of bivalves on

235

seagrass nitrogen metabolism, enhancing nutrient uptake capacity with high light and reducing

236

root nitrogen uptake with low light. This was clearly reflected in the levels of metabolites

237

associated with nitrogen metabolism (i.e., serine, tyrosine, glycine; Figure 3).


10

Page 11 of 27


238

When screening the metabolomic data for metabolites involved in apoptosis and cell

239

degradation, we found increased levels of ceramide and decreased levels of sphingosine-1-

240

phosphate (S1P) in Z. marina grown with mussels under low light (Figure 4). Ceramide and S1P

241

play a central role in cellular processes governing growth, differentiation, apoptosis, and survival

242

in eukaryotic organisms42. While ceramide mainly induces apoptosis, reduces growth and

243

increases in response to stress42, S1P promotes survival and growth42. The relative ratio between

244

ceramide and S1P (referred as sphingo-lipid rheostat) determine the cell development43 and

245

apoptosis. The latter is related with decreases in S1P levels and increases in ceramide44. In this

246

experiment mussels presence under low light caused a shift in the relative ratios between

247

ceramide and S1P from values around 1.7 to 0.6 (Figure 4). In addition, we found that low light

248

increased phytol levels in the leaves and that mussels exacerbated this effect (Figure 4). Phytol is

249

a chlorophyll degradation product accumulating under plant senescence45-47. Based on the

250

function of the sphingo-lipid rheostat and phytol in other organisms, we suggest that increased

251

levels of ceramide, decreased levels of S1P, shifts in ceramide:S1P ratios and accumulation of

252

phytol in seagrasses (Figure 4) under bivalve presence and low light indicate the onset of tissue

253

degradation and propagated cell death.

254 255

Integration of metabolomics and environmental parameters

256

To investigate the effect of sulfide intrusion on seagrass metabolism, we explored the

257

metabolomic data matrix for relations between tissue elemental sulfur levels and metabolite

258

concentrations. Elemental sulfur in the roots originates from oxidation of intruding gaseous

259

sediment sulfide26 and is a strong proxy for sulfide intrusion27 into seagrass tissues26,34. Sulfide

260

intrusion modulated (|R| > 0.6) levels of 77 metabolites (Table S1), suggesting Z. marina


11


Page 12 of 27

261

metabolism is strongly related to sulfide intrusion under stress from low light and mussels. This

262

is a new discovery that corroborates prior work demonstrating the omnipresence and importance

263

of sediment sulfides and sulfide intrusion in seagrass systems26,34. Among the metabolites related

264

to sulfide intrusion many play a role in metabolism under impaired oxidative energy metabolism.

265

Lactate, pyruvate, GABA, alanine, glutamate, β-sitosterol, linoleic acid, and α tocopherol were

266

positively correlated to sulfide intrusion, whereas sucrose, proline, L-DOPA, glutamine, and

267

gluconate were negatively correlated to sulfide intrusion (Table S1). This metabolic response is

268

most likely related to sulfide toxicity and/or sediment hypoxia. Both would trigger similar

269

metabolic responses by inhibiting the oxidative energy metabolism48,49. Constant aeration of the

270

water of the mesocosm throughout the experiment prevented hypoxic conditions in the water

271

column9. However high respiration rates related to mussel presence (Table S2) likely caused

272

local hypoxic conditions near the sediment surface, promoting sulfide intrusion into the plants50.

273

Sulfide toxicity inhibits oxidative energy metabolism49 but sulfide-induced formation of reactive

274

oxygen species (ROS) formation can regulate sulfide toxicity51. Eghbal et al. 51 showed that ROS

275

scavenging metabolites reduce sulfide toxicity and another study suggested that ROS formation

276

in general stimulates plant antioxidant defense52. This should be manifested in increased ROS

277

scavenging antioxidants during stress acclimation52,53. Indeed the ROS scavenging antioxidants

278

α-tocopherol54, β-sitosterol37, L-DOPA55 and proline54 were correlated to apparent sulfide

279

intrusion in our study (Table S1). These compounds also stabilize membranes and protect

280

pigments, proteins, and fatty acids from oxidative damage54,56. Consequently, the correlation of

281

proline, β-sitosterol, L-DOPA, and α-tocopherol with sulfide intrusion may indicate a sulfide

282

detoxification reaction in seagrasses. Indeed, linoleic acid was correlated to the level of sulfide

283

intrusion (Table S1) and represents an important component of the cell membrane that is


12

Page 13 of 27


284

particularly susceptible to oxygenation by ROS and related to stress from membrane

285

degradation54,57. Hence, the correlation with sulfide intrusion may indicate increased membrane

286

degradation caused by sulfide toxicity. Overall, the increased sulfide detoxification and

287

concomitant increase in membrane degradation suggest sulfide toxicity in seagrasses was co-

288

occurring with bivalves in low-light environments.

289

Using novel applications of metabolomics, we have demonstrated that bivalves and light

290

interactively structure seagrass metabolism. Under high light conditions, mussels stimulated

291

nitrogen metabolism, but Z. marina performance did not benefit because growth was not limited

292

by nitrogen (as shown in Castorani et al.9, Table S2 and S3). Low light reduced the resilience of

293

Z. marina by reducing growth and diminishing pools of stored sugars (as shown in Castorani et

294

al.9, Table S2 and S3). Consequently, Z. marina did not benefit from the increased nutrients.

295

Strongly-respiring mussels drained oxygen levels close to the sediment surface, thereby

296

enhancing the potential for sulfide intrusion and leading to sulfide toxicity in less resilient, low

297

light seagrass58.

298

Eutrophication is accelerating in temperate estuaries worldwide, reducing light availability and

299

increasing hypoxic conditions for seagrasses and other marine plants10,59,60. Our results suggest

300

that continued degradation of coastal and estuarine water quality will enhance sulfide stress on Z.

301

marina and other seagrasses, especially where co-occurring with filter-feeding bivalves. This

302

finding might to some extent contrast with two recent studies that predicted sulfide oxidation by

303

bacterial symbionts within lucinid clams that could be critically important to seagrass

304

persistence61,62. In light of ongoing and future eutrophication of temperate estuaries, our results

305

indicate a need for careful evaluation of suggestions that seagrasses and bivalves are strictly


13


Page 14 of 27

306

mutualistic. In some systems, the antagonistic effects of bivalves on seagrasses might overpower

307

mutualistic interactions.

308

Our study also demonstrates that metabolomics may be an important tool for early

309

identification of plant stress in coastal environments. Our approach (1) provides information on

310

system shifts through heatmaps and PCA, (2) illuminates pathways affected by environmental

311

drivers, (3) reveals in depth and otherwise hidden effects of environmental stress on seagrasses,

312

and (4) provides potential candidates for early-warning and stress-specific biomarkers (i.e., bio-

313

indicators), although such must be thoroughly validated63. Our novel metabolomics-based

314

approach should be adapted for other systems to detect sublethal environmental stress far in

315

advance of the physiological manifestations that are the focus of traditional methods. Continued

316

developments and applications of metabolomics to community ecology will shed further light on

317

the influence of environmental stress on species interactions.

318 319

ASSOCIATED CONTENT

320

Supporting Information.

321

Table S1, Table S2, Table S3 and Figure S1 are available free of charge via the Internet at

322

http://pubs.acs.org. The raw data is available free of charge from figshare.com:

323

10.6084/m9.figshare.3823890

324 325

AUTHOR INFORMATION

326

Corresponding Author


14

Page 15 of 27


327

Email: [email protected], phone/fax: +45 65 50 84 66, Nordic Center for Earth Evolution,

328

Department of Biology, University of Southern Denmark, 5230 Odense, Denmark

329

Author Contributions

330

All authors contributed equally to this study. The study was designed by contribution of all

331

authors. The experimental work and data analysis was performed by MC and HHS,

332

metabolomics were conducted by HHS. The manuscript was written through contributions of all

333

authors. All authors have given approval to the final version of the manuscript.

334

Funding Sources

335

HHS was supported by the European Research Council Advanced Grant program through the

336

OXYGEN project (no. 267233-ERC). MCNC was supported by a U.S. National Science

337

Foundation (NSF) Graduate Research Fellowship and a NSF–Danish National Research

338

Foundation (DNRF) Nordic Research Opportunity Fellowship.

339

Notes

340

The authors declare no competing financial interest.

341

ACKNOWLEDGMENT

342

The authors thank E. Laursen, S. Møller, B. Christensen, M. Flindt, A. Glud, K. Hancke, R.O.

343

Holm, K.C. Kirkegaard, and S.W. Thorsen for research assistance. We thank the three

344

anonymous reviewers for helpful and constructive comments.

345 346


15


Page 16 of 27

347

REFERENCES

348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390

1. Jones, C. G.; Lawton, J. H.; Shachak, M., Organisms as Ecosystem Engineers. In Ecosystem Management: Selected Readings, Springer New York: New York, NY, 1996; pp 130147, http://dx.doi.org/10.1007/978-1-4612-4018-1_14 2. Norkko, A.; Hewitt, J. E.; Thrush, S. F.; Funnell, G. A., Conditional outcomes of facilitation by a habitat-modifying subtidal bivalve. Ecology 2006, 87, (1), 226-234, http://dx.doi.org/10.1890/05-0176 3. Bertness, M. D.; Hacker, S. D., Physical Stress and Positive Associations Among Marsh Plants. Am Nat 1994, 144, (3), 363-372, http://dx.doi.org/10.2307/2462950 4. Romero, L. M., Physiological stress in ecology: lessons from biomedical research. Trends Ecol Evol 2004, 19, (5), 249-255, http://dx.doi.org/10.1016/j.tree.2004.03.008 5. Jorge, T. F.; Rodrigues, J. A.; Caldana, C.; Schmidt, R.; van Dongen, J. T.; ThomasOates, J.; Antonio, C., Mass spectrometry-based plant metabolomics: Metabolite responses to abiotic stress. Mass Spectrom Rev 2016, 35, (5), 620-49, http://dx.doi.org/10.1002/mas.21449 6. Hasler-Sheetal, H.; Fragner, L.; Holmer, M.; Weckwerth, W., Diurnal effects of anoxia on the metabolome of the seagrass Zostera marina. Metabolomics 2015, 11, (5), 1208-1218, http://dx.doi.org/10.1007/s11306-015-0776-9 7. Haven, D. S.; Morales-Alamo, R., Aspects of biodeposition by oysters and other invertebrate filter feeders. Limnol Oceanogr 1966, 11, (4), 487-498, http://dx.doi.org/10.4319/lo.1966.11.4.0487 8. Kautsky, N.; Evans, S., Role of biodeposition by Mytilus edulis in the ciculation of matter and nutrients in a Baltic coastal ecosystem. Mar Ecol Prog Ser 1987, 38, 201-212, http://dx.doi.org/10.1016/0022-0981(93)90141-A 9. Castorani, M. C. N.; Glud, R. N.; Hasler-Sheetal, H.; Holmer, M., Light indirectly mediates bivalve habitat modification and impacts on seagrass. J Exp Mar Biol Ecol 2015, 472, 41-53, http://dx.doi.org/10.1016/j.jembe.2015.07.001 10. Orth, R. J.; Carruthers, T. J. B.; Dennison, W. C.; Duarte, C. M.; Fourqurean, J. W.; Heck, K. L.; Hughes, A. R.; Kendrick, G. A.; Kenworthy, W. J.; Olyarnik, S.; Short, F. T.; Waycott, M.; Williams, S. L., A Global Crisis for Seagrass Ecosystems. Bioscience 2006, 56, (12), 987-996, http://dx.doi.org/10.1641/0006-3568(2006)56%5B987:agcfse%5D2.0.co;2 11. Waycott, M.; Duarte, C. M.; Carruthers, T. J.; Orth, R. J.; Dennison, W. C.; Olyarnik, S.; Calladine, A.; Fourqurean, J. W.; Heck, K. L., Jr.; Hughes, A. R.; Kendrick, G. A.; Kenworthy, W. J.; Short, F. T.; Williams, S. L., Accelerating loss of seagrasses across the globe threatens coastal ecosystems. Proc Natl Acad Sci U S A 2009, 106, (30), 12377-81, http://dx.doi.org/10.1073/pnas.0905620106 12. Oliver, S. G.; Winson, M. K.; Kell, D. B.; Baganz, F., Systematic functional analysis of the yeast genome. Trends Biotechnol 1998, 16, (9), 373-378, http://dx.doi.org/10.1016/S01677799(98)01214-1 13. Goulitquer, S.; Potin, P.; Tonon, T., Mass spectrometry-based metabolomics to elucidate functions in marine organisms and ecosystems. Mar Drugs 2012, 10, (4), 849-880, http://dx.doi.org/10.3390/md10040849 14. Sardans, J.; Peñuelas, J.; Rivas-Ubach, A., Ecological metabolomics: overview of current developments and future challenges. Chemoecology 2011, 21, (4), 191-225, http://dx.doi.org/10.1007/s00049-011-0083-5


16

Page 17 of 27

391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435


15. Weckwerth, W.; Morgenthal, K., Metabolomics: from pattern recognition to biological interpretation. Drug Discov Today 2005, 10, (22), 1551-1558, http://dx.doi.org/10.1016/s13596446(05)03609-3 16. Gu, J.; Weber, K.; Klemp, E.; Winters, G.; Franssen, S. U.; Wienpahl, I.; Huylmans, A. K.; Zecher, K.; Reusch, T. B.; Bornberg-Bauer, E.; Weber, A. P., Identifying core features of adaptive metabolic mechanisms for chronic heat stress attenuation contributing to systems robustness. Integr Biol 2012, 4, (5), 480-93, http://dx.doi.org/10.1039/c2ib00109h 17. Carroll, J.; Gobler, C.; Peterson, B., Resource-restricted growth of eelgrass in New York estuaries: light limitation, and alleviation of nutrient stress by hard clams. Mar Ecol Prog Ser 2008, 369, 51-62, http://dx.doi.org/10.3354/meps07593 18. Vinther, H.; Norling, P.; Kristensen, P.; Dolmer, P.; Holmer, M., Effects of coexistence between the blue mussel and eelgrass on sediment biogeochemistry and plant performance. Mar Ecol Prog Ser 2012, 447, 139-149, http://dx.doi.org/10.3354/meps09505 19. Reusch, T. B. H.; Williams, S. L., Variable responses of native eelgrass Zostera marina to a non-indigenous bivalve Musculista senhousia. Oecologia 1998, 113, (3), 428-441, http://dx.doi.org/10.1007/s004420050395 20. Wagner, E.; Dumbauld, B. R.; Hacker, S. D.; Trimble, A. C.; Wisehart, L. M.; Ruesink, J. L., Density-dependent effects of an introduced oyster, Crassostrea gigas, on a native intertidal seagrass, Zostera marina. Mar Ecol Prog Ser 2012, 468, 149-160, http://dx.doi.org/10.3354/meps09952 21. Zimmerman, R. C., Light and Photosynthesis in Seagrass Meadows. In Seagrasses: Biology, Ecology and Conservation, Larkum, A. W. D.; Orth, R. J.; Duarte, C. M., Eds. Springer Netherlands: Dordrecht, Nederlands, 2006; pp 303-321, http://dx.doi.org/10.1007/978-1-40202983-7_13 22. Reusch, T.; Chapman, A.; Groger, J., Blue mussels Mytilus-edulis do not interfere with eelgrass Zostera-marina but fertilize shoot growth through biodeposition. Mar Ecol Prog Ser 1994, 108, (3), 265-282, http://dx.doi.org/10.3354/meps108265. 23. Bologna, P. A. X.; Fetzer, M. L.; McDonnell, S.; Moody, E. M., Assessing the potential benthic–pelagic coupling in episodic blue mussel (Mytilus edulis) settlement events within eelgrass (Zostera marina) communities. J Exp Mar Biol Ecol 2005, 316, (2), 117-131, http://dx.doi.org/10.1016/j.jembe.2004.10.009 24. Nielsen, S. L.; Sand-Jensen, K.; Borum, J.; Geertz-Hansen, O., Phytoplankton, nutrients, and transparency in Danish coastal waters. Estuaries 2002, 25, (5), 930-937, http://dx.doi.org/10.1007/bf02691341 25. Dennison, W. C.; Alberte, R. S., Role of daily light period in the depth distribution of Zostera marina (eelgrass). Mar Ecol Prog Ser 1985, 25, (1), 51-61, http://dx.doi.org/10.3354/meps025051 26. Hasler-Sheetal, H.; Holmer, M., Sulfide intrusion and detoxification in the seagrass Zostera marina. PLoS One 2015, 10, (6), e0129136, http://dx.doi.org/10.1371/journal.pone.0129136 27. Frederiksen, M. S.; Holmer, M.; Díaz-Almela, E.; Marba, N.; Duarte, C., Sulfide invasion in the seagrass Posidonia oceanica at Mediterranean fish farms: assessment using stable sulfur isotopes. Mar Ecol Prog Ser 2007, 345, 93-104, http://dx.doi.org/10.3354/meps06990 28. Markham, J. E.; Jaworski, J. G., Rapid measurement of sphingolipids from Arabidopsis thaliana by reversed-phase high-performance liquid chromatography coupled to electrospray


17


436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481

Page 18 of 27

ionization tandem mass spectrometry. Rapid Commun Mass Spectrom 2007, 21, (7), 1304-14, http://dx.doi.org/10.1002/rcm.2962 29. Steuer, R.; Morgenthal, K.; Weckwerth, W.; Selbig, J., A Gentle Guide to the Analysis of Metabolomic Data. In Metabolomics: Methods and Protocols, Weckwerth, W., Ed. Humana Press: Totowa, NJ, 2007; pp 105-126, http://dx.doi.org/10.1007/978-1-59745-244-1_7 30. Benjamini, Y.; Hochberg, Y., On the Adaptive Control of the False Discovery Rate in Multiple Testing With Independent Statistics. J Educ Behav Stat 2000, 25, (1), 60-83, http://dx.doi.org/10.3102/10769986025001060 31. Benjamini, Y.; Hochberg, Y., Controlling the False Discovery Rate - a Practical and Powerful Approach to Multiple Testing. J Roy Stat Soc B Met 1995, 57, (1), 289-300, http://dx.doi.org/10.2307/2346101 32. Wiklund, S.; Johansson, E.; Sjöström, L.; Mellerowicz, E. J.; Edlund, U.; Shockcor, J. P.; Gottfries, J.; Moritz, T.; Trygg, J., Visualization of GC/TOF-MS-Based Metabolomics Data for Identification of Biochemically Interesting Compounds Using OPLS Class Models. Anal Chem 2008, 80, (1), 115-122, http://dx.doi.org/10.1021/ac0713510 33. Sana, T. R.; Gordon, D. B.; Fischer, S. M.; Tichy, S. E.; Kitagawa, N.; Lai, C.; Gosnell, W. L.; Chang, S. P., Global mass spectrometry based metabolomics profiling of erythrocytes infected with Plasmodium falciparum. PLoS One 2013, 8, (4), e60840, http://dx.doi.org/10.1371/journal.pone.0060840 34. Holmer, M.; Hasler-Sheetal, H., Sulfide intrusion in seagrasses assessed by stable sulfur isotopes – A synthesis of current results. Front Mar Sci 2014, 1:64, http://dx.doi.org/10.3389/fmars.2014.00064 35. Baena-Gonzalez, E.; Sheen, J., Convergent energy and stress signaling. Trends Plant Sci 2008, 13, (9), 474-82, http://dx.doi.org/10.1016/j.tplants.2008.06.006 36. Smith, R. D.; Pregnall, A. M.; Alberte, R. S., Effects of anaerobiosis on root metabolism of Zostera marina (eelgrass): implications for survival in reducing sediments. Mar Biol 1988, 98, (1), 131-141, http://dx.doi.org/10.1007/Bf00392668 37. Obata, T.; Fernie, A., The use of metabolomics to dissect plant responses to abiotic stresses. Cell. Mol. Life Sci. 2012, 69, (19), 3225-3243, http://dx.doi.org/10.1007/s00018-0121091-5 38. Bouche, N.; Fromm, H., GABA in plants: just a metabolite? Trends Plant Sci 2004, 9, (3), 110-5, http://dx.doi.org/10.1016/j.tplants.2004.01.006 39. Kreuzwieser, J.; Hauberg, J.; Howell, K. A.; Carroll, A.; Rennenberg, H.; Millar, A. H.; Whelan, J., Differential response of gray poplar leaves and roots underpins stress adaptation during hypoxia. Plant Physiol 2009, 149, (1), 461-73, http://dx.doi.org/10.1104/pp.108.125989 40. van Dongen, J. T.; Fröhlich, A.; Ramírez-Aguilar, S. J.; Schauer, N.; Fernie, A. R.; Erban, A.; Kopka, J.; Clark, J.; Langer, A.; Geigenberger, P., Transcript and metabolite profiling of the adaptive response to mild decreases in oxygen concentration in the roots of arabidopsis plants. Ann Bot 2009, 103, (2), 269-280, http://dx.doi.org/10.1093/aob/mcn126 41. Chellamuthu, V. R.; Ermilova, E.; Lapina, T.; Luddecke, J.; Minaeva, E.; Herrmann, C.; Hartmann, M. D.; Forchhammer, K., A widespread glutamine-sensing mechanism in the plant kingdom. Cell 2014, 159, (5), 1188-99, http://dx.doi.org/10.1016/j.cell.2014.10.015 42. Alden, K. P.; Dhondt-Cordelier, S.; McDonald, K. L.; Reape, T. J.; Ng, C. K.; McCabe, P. F.; Leaver, C. J., Sphingolipid long chain base phosphates can regulate apoptotic-like programmed cell death in plants. Biochem Biophys Res Commun 2011, 410, (3), 574-80, http://dx.doi.org/10.1016/j.bbrc.2011.06.028


18

Page 19 of 27

482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526


43. Spiegel, S.; Milstien, S., Sphingosine-1-phosphate: an enigmatic signalling lipid. Nat Rev Mol Cell Biol 2003, 4, (5), 397-407, http://dx.doi.org/10.1038/nrm1103 44. Hannun, Y. A., Functions of ceramide in coordinating cellular responses to stress. Science 1996, 274, (5294), 1855-59, http://dx.doi.org/10.1126/science.274.5294.1855 45. Ischebeck, T.; Zbierzak, A. M.; Kanwischer, M.; Dormann, P., A salvage pathway for phytol metabolism in Arabidopsis. J Biol Chem 2006, 281, (5), 2470-7, http://dx.doi.org/10.1074/jbc.M509222200 46. Zhang, W.; Liu, T.; Ren, G.; Hortensteiner, S.; Zhou, Y.; Cahoon, E. B.; Zhang, C., Chlorophyll degradation: the tocopherol biosynthesis-related phytol hydrolase in Arabidopsis seeds is still missing. Plant Physiol 2014, 166, (1), 70-9, http://dx.doi.org/10.1104/pp.114.243709 47. Lippold, F.; vom Dorp, K.; Abraham, M.; Holzl, G.; Wewer, V.; Yilmaz, J. L.; Lager, I.; Montandon, C.; Besagni, C.; Kessler, F.; Stymne, S.; Dormann, P., Fatty acid phytyl ester synthesis in chloroplasts of Arabidopsis. Plant Cell 2012, 24, (5), 2001-14, http://dx.doi.org/10.1105/tpc.112.095588 48. Wang, R., Physiological implications of hydrogen sulfide: a whiff exploration that blossomed. Physiol Rev 2012, 92, (2), 791-896, http://dx.doi.org/10.1152/physrev.00017.2011 49. Lamers, L. P. M.; Govers, L. L.; Janssen, I. C. J. M.; Geurts, J. J. M.; Van der Welle, M. E. W.; Van Katwijk, M. M.; Van der Heide, T.; Roelofs, J. G. M.; Smolders, A. J. P., Sulfide as a soil phytotoxin A review. Front. Plant Sci. 2013, 4, 1-14, http://dx.doi.org/10.3389/fpls.2013.00268 50. Pedersen, O.; Binzer, T.; Borum, J., Sulphide intrusion in eelgrass (Zostera marina L.). Plant Cell Environ 2004, 27, (5), 595-602, http://dx.doi.org/10.1111/J.1365-3040.2004.01173.X 51. Eghbal, M. A.; Pennefather, P. S.; O’Brien, P. J., H2S cytotoxicity mechanism involves reactive oxygen species formation and mitochondrial depolarisation. Toxicology 2004, 203, (1– 3), 69-76, http://dx.doi.org/10.1016/j.tox.2004.05.020 52. Brain, R. A.; Cedergreen, N., Biomarkers in Aquatic Plants: Selection and Utility. In Rev Environ Contam Toxicol, Whitacre, D. M., Ed. Springer New York: New York, NY, 2009; pp 49-109, http://dx.doi.org/10.1007/978-0-387-09647-6_2 53. Babu, T. S.; Tripuranthakam, S.; Greenberg, B. M., Biochemical responses of the aquatic higher plant Lemna gibba to a mixture of copper and 1,2-dihydroxyanthraquinone: Synergistic toxicity via reactive oxygen species. Environ Toxicol Chem 2005, 24, (12), 3030-3036, http://dx.doi.org/10.1897/05-073R.1 54. Das, K.; Roychoudhury, A., Reactive oxygen species (ROS) and response of antioxidants as ROS-scavengers during environmental stress in plants. Front environ sci 2014, 2, http://dx.doi.org/10.3389/fenvs.2014.00053 55. Soares, A. R.; Marchiosi, R.; Siqueira-Soares Rde, C.; Barbosa de Lima, R.; Dantas dos Santos, W.; Ferrarese-Filho, O., The role of L-DOPA in plants. Plant Signal Behav 2014, 9, (4), e28275, http://dx.doi.org/10.4161/psb.28275 56. Blokhina, O.; Virolainen, E.; Fagerstedt, K. V., Antioxidants, oxidative damage and oxygen deprivation stress: a review. Ann Bot 2003, 91 Spec No, 179-94, http://dx.doi.org/10.1093/aob/mcf118 57. Dietz, K. J.; Turkan, I.; Krieger-Liszkay, A., Redox- and Reactive Oxygen SpeciesDependent Signaling into and out of the Photosynthesizing Chloroplast. Plant Physiol 2016, 171, (3), 1541-50, http://dx.doi.org/10.1104/pp.16.00375


19


527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546

Page 20 of 27

58. Holmer, M.; Frederiksen, M. S.; Mollegaard, H., Sulfur accumulation in eelgrass (Zostera marina) and effect of sulfur on eelgrass growth. Aquat Bot 2005, 81, (4), 367-379, http://dx.doi.org/10.1016/j.aquabot.2004.12.006 59. James, E. C., Our evolving conceptual model of the coastal eutrophication problem. Mar Ecol Prog Ser 2001, 210, 223-253, http://dx.doi.org/10.3354/meps210223 60. Howarth, R.; Chan, F.; Conley, D. J.; Garnier, J.; Doney, S. C.; Marino, R.; Billen, G., Coupled biogeochemical cycles: eutrophication and hypoxia in temperate estuaries and coastal marine ecosystems. Front Ecol Environ 2011, 9, (1), 18-26, http://dx.doi.org/10.1890/100008 61. de Fouw, J.; Govers, L. L.; van de Koppel, J.; van Belzen, J.; Dorigo, W.; Sidi Cheikh, M. A.; Christianen, M. J.; van der Reijden, K. J.; van der Geest, M.; Piersma, T.; Smolders, A. J.; Olff, H.; Lamers, L. P.; van Gils, J. A.; van der Heide, T., Drought, Mutualism Breakdown, and Landscape-Scale Degradation of Seagrass Beds. Curr Biol 2016, 26, (8), 1051-1056, http://dx.doi.org/10.1016/j.cub.2016.02.023 62. van der Heide, T.; Govers, L. L.; de Fouw, J.; Olff, H.; van der Geest, M.; van Katwijk, M. M.; Piersma, T.; van de Koppel, J.; Silliman, B. R.; Smolders, A. J.; van Gils, J. A., A threestage symbiosis forms the foundation of seagrass ecosystems. Science 2012, 336, (6087), 1432-4, http://dx.doi.org/10.1126/science.1219973 63. Forbes, V. E.; Palmqvist, A.; Bach, L., The use and misuse of biomarkers in ecotoxicology. Environ Toxicol Chem 2006, 25, (1), 272-280, http://dx.doi.org/10.1897/05257R.1

547 548 549 550 551 552 553 554 555 556 557 558 559 560 561


20

Page 21 of 27


562 563 564 565 566

Figure Captions:

567 568 569 570 571 572 573 574 575

Figure 1: Heatmaps of clusters associated with the effect of light availability and Mytilus edulis presence in Zostera marina leaves (panels A,D), rhizomes (panels B,E) and roots (panels C,F), based on apolar (panels A-C) and polar (panels D-F) metabolites. Euclidean distance was used as distance measure and Ward as clustering algorithm. Columns labels indicate low and high light availability, and mussel presence (+) and mussel absence (–). The color gradient from green to red indicates lower to higher metabolite levels, respectively. The colored clusters indicate a cluster threshold of 20. The data is presented as group average (N = 6 samples per treatment) for the sake of clarity, however the clustering algorithm is based on all samples. Only metabolites that passed the quality control filters are included.

576 577 578 579 580 581 582 583 584 585

Figure 2: Figure 2: PCA scores plots of polar (left panel) and apolar (right panel) compounds in Zostera marina leaves exposed to differing light availability and mussel presence. The first row shows PC1 vs PC2 and the second column shows PC1 vs PC3. Squares indicate samples under high light intensities and triangles samples under low light intensities; blue colored samples indicate mussel presence and red colored samples indicate mussel absence. The Venn diagram indicates the number of metabolites with high influence on sample separation on the respective PC (obtained from the CC-plot); metabolites with the highest influence are presented inside the circles. Only metabolites that passed the quality control filters are included. (plots for rhizomes and roots are shown in Figure S1).

586 587 588 589 590 591 592 593 594 595 596 597 598 599

Figure 3: Figure 3: Pathway analysis of metabolites in the leaves of the seagrass Zostera marina exposed to varying light and the blue mussel Mytilus edulis. Visualized are the primary energy metabolism (glycolysis and TCA cycle, left panel) and the associated nitrogen metabolism (right panel). Metabolites are denoted as the ratio in each treatment relative to high light conditions in the absence of mussels. The shaded area indicates low light conditions. Metabolites framed in green indicate significant (p < 0.05) increases, red indicate significant decreases, and black indicate no change relative to eelgrass grown under high light without mussels. Red bars indicate lower metabolite levels and blue bars indicate higher metabolite levels relative to eelgrass grown under high light without mussels. Solid arrows illustrate enzymatic reactions; dotted lines indicate the same metabolite presented in different and linked pathways. Abbreviations for experimental treatments are as follows: high light with mussels (H+); low light without mussels (L-); Low light with mussels (L+). Modified and used with permission from Hasler-Sheetal et al6.

600


21


601 602 603 604 605

Page 22 of 27

Figure 4. Relative and normalized levels of metabolites involved in apoptosis. The left panel shows ceramide and sphigosine-1-phosphate (S1P) levels and the right panel shows phytol levels inZ. marina leaves as a function of light availability and mussel presence. The numbers in boxes indicate the ceramide:S1P ratio. Levels not sharing the same letter indicate significant differences (ANOVA; p < 0.05; Tukey Post Hoc test, p < 0.05).


22

Page 23 of 27


TOC TOC 84x46mm (300 x 300 DPI)


Leaves

Low

High

+ -

+ -

B

Low

High

+ -

+ -

C

Page 24 of 27

Low

High

+ -

+ -

Apolar Compounds

A

Environmental Science & Technology Roots Rhizomes

High

Low

Low

High

+ -

+ -

E

Low

High

+ -

+ -

F

Low

+ -

Polar Compounds

D


High

+ -

Page 25 of 27


Figure 2: PCA scores plots of polar (left panel) and apolar (right panel) compounds in Zostera marina leaves exposed to differing light availability and mussel presence. The first row shows PC1 vs PC2 and the second column shows PC1 vs PC3. Squares indicate samples under high light intensities and triangles samples under low light intensities; blue colored samples indicate mussel presence and red colored samples indicate mussel absence. The Venn diagram indicates the number of metabolites with high influence on sample separation on the respective PC (obtained from the CC-plot); metabolites with the highest influence are presented inside the circles. Only metabolites that passed the quality control filters are included. (plots for rhizomes and roots are shown in Figure S1). Figure 2 120x75mm (300 x 300 DPI)



Figure 3: Pathway analysis of metabolites in the leaves of the seagrass Zostera marina exposed to varying light and the blue mussel Mytilus edulis. Visualized are the primary energy metabolism (glycolysis and TCA cycle, left panel) and the associated nitrogen metabolism (right panel). Metabolites are denoted as the ratio in each treatment relative to high light conditions in the absence of mussels. The shaded area indicates low light conditions. Metabolites framed in green indicate significant (p < 0.05) increases, red indicate significant decreases, and black indicate no change relative to eelgrass grown under high light without mussels. Red bars indicate lower metabolite levels and blue bars indicate higher metabolite levels relative to eelgrass grown under high light without mussels. Solid arrows illustrate enzymatic reactions; dotted lines indicate the same metabolite presented in different and linked pathways. Abbreviations for experimental treatments are as follows: high light with mussels (H+); low light without mussels (L-); Low light with mussels (L+). Modified and used with permission from from Hasler-Sheetal et al6 Figure 3 131x96mm (300 x 300 DPI)


Page 26 of 27

227 of Ceramide PageEnvironmental 27 Science & Technology Phytol

d

d d

0

a a

a

-1 c

Normalised Concentration

Normalised Concentration

1

c

b

S1P

1

0

b a

a

-1

ACS Paragon Plus Environment 1.8 1.7 1.6 0.6 -2 None Mussels None Mussels

High Light

Low Light

None Mussels None Mussels

High Light

Low Light

Metabolomics Reveals Cryptic Interactive Effects of Species

Recommend Documents