Interactive effects of warming and eutrophication on population dynamics and stalk production of epiphytic diatoms in transitional waters

Journal Pre-proof Interactive effects of warming and eutrophication on population dynamics and stalk production of epiphytic diatoms in transitional waters M.D. Belando, P.M. Martínez, M. Aboal PII:

S0272-7714(19)30437-8

DOI:

https://doi.org/10.1016/j.ecss.2019.106413

Reference:

YECSS 106413

To appear in:

Estuarine, Coastal and Shelf Science

Received Date: 3 May 2019 Revised Date:

31 August 2019

Accepted Date: 8 October 2019

Please cite this article as: Belando, M.D., Martínez, P.M., Aboal, M., Interactive effects of warming and eutrophication on population dynamics and stalk production of epiphytic diatoms in transitional waters, Estuarine, Coastal and Shelf Science (2019), doi: https://doi.org/10.1016/j.ecss.2019.106413. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Ltd.

1

Interactive effects of warming and eutrophication on population dynamics and

2

stalk production of epiphytic diatoms in transitional waters

3

M.D. Belando, P.M. Martínez, M. Aboal*

4

Departamento de Biología Vegetal, Facultad de Biología, Universidad de Murcia,

5

30100 Murcia, Spain.

6

*Corresponding author: [email protected]

7

Abstract

8

The interactive effects of temperature and nitrogen:phosphorus stoichiometry were

9

assessed by an experiment that involves cultures of the stalked epiphytic diatom

10

Licmophora colosalis at different N:P ratios (5, 10, 16, 21 and 42) and temperatures

11

(26, 31 and 36ºC) to simulate predictions of global warming effects on a Mediterranean

12

coastal lagoon. At 26ºC, which seemed optimal for L. colosalis, it grew well at the N:P

13

ratios within the 16-21 range, but a different stoichiometry reduced cell growth. The

14

5ºC increase had deleterious effects on diatom fitness as growth reduced, mortality

15

increased and carbohydrates accumulation amplified under N or P deficiency. In

16

contrast, this temperature stress was better overcome at N:P=16. The 36ºC temperature

17

proved lethal. Stalk production was specifically stimulated when growth was P-limited

18

at both 26ºC and 31ºC, which suggests that a high proportion of stalks, which would

19

form macroscopic colonies, may serve as an indicator of warming and high N:P ratios in

20

transitional waters.

21

Keywords: epiphytic diatoms, biomonitoring, carbohydrates, N:P stoichiometry,

22

temperature, transitional waters.

1

23

Introduction

24

Marine ecosystems are being affected by global climate change as rising sea levels,

25

increased frequency and intensity of storms, and warmer ocean temperatures show

26

(IPCC 2007). At the same time, coastal lagoons and transitional waters may be

27

considered one of the most vulnerable marine environments, and they might suffer from

28

a wide variety of pressures deriving from human activities (e.g., delivery of nutrients

29

from surrounding areas), which may exacerbate climate-driven impacts (Lloret et al.

30

2008). Mediterranean regions have been identified as one of the most prominent “hot

31

spots” in climate change predictions (Giorgi 2006), where climate anomalies have

32

exponentially increased in recent years (Danovaro et al. 2009). Predictions of rising

33

temperature effects indicate changes in species distribution/abundance according to

34

their thermal tolerances and ability to adapt (Harley et al. 2006, and reference therein).

35

In the last few decades, several mucilage overproduction episodes of both planktic and

36

benthic microorganisms have already been observed on eastern Mediterranean coasts

37

(e.g., De Philippis et al. 2005; Schiaparelli et al. 2007; Sartoni et al. 2008), which have

38

been related to anomalous episodes of extremely high water temperatures (Danovaro et

39

al. 2009; Aktan et al. 2011; Mangialajo et al. 2011).

40

This mucilaginous snow consists mainly in macroalgal and microscopic organisms,

41

mainly diatoms and cyanobacteria, embedded in a matrix of extracellular polymeric

42

substances (EPS) (De Philippis et al. 2005). Most empirical studies have focused on the

43

links between EPS overproduction and nutrient supply, but the role of increasing

44

temperature has scarcely been investigated. Several authors have already indicated that

45

nutrient-limited conditions favour the release of EPS in many phytoplankton (e.g.

46

Myklestad 1995), epipelic species (Alcoverro et al. 2000; Staats et al. 2000), and

2

47

benthic intertidal diatoms (Smith and Underwood 2000; Underwood et al. 2004).

48

However nutrient ratios, mainly the N:P ratio of water, can influence also EPS

49

production (Myklestad et al. 1972; Myklestad 1995), but the response to high or low

50

N:P ratios may differ among species.

51

EPS may play a relevant ecological role for benthic diatom species as this excretion is

52

essential for their motility, and for attachment among cells and to substratum (Daniel et

53

al. 1987). Some trade-based approaches have pointed out that stalked diatoms are good

54

candidates for discriminating different trophic statuses in freshwater ecosystems, but the

55

relation of stalks excretion with environmental factors, such as nutrients, temperature

56

and light, has also been suggested (Passy 2007; Berthon et al. 2011). The response of

57

freshwater Didymosphenia geminata (Lyngbye) M. Schmidt to nutrient environmental

58

conditions and stalk overproduction with limited growth under phosphorus-deficiency

59

conditions in rivers has also been proved (e.g. Kilroy and Bothwell 2011; 2012). In

60

marine ecosystems, these relationships have been investigated only for the stalked

61

diatom Licmophora flabellata (Greville) C. Agardh (Ravizza and Hallegraeff 2015).

62

These authors indicated that strong light intensity was the factor that stimulated stalk

63

formation, while nutrients had no effect. The variability in response between different

64

species from marine and freshwater environments makes the prediction of mucilaginous

65

benthic blooms under changing nutrient and temperature conditions difficult.

66

In a global climate change context, the prediction of rising temperature impacts on

67

marine unicellular organisms has focused generally on the cell size changes of

68

phytoplankton communities (e.g., Adams et al. 2013; Peter and Sommer 2015). It is

69

frequently suggested that warming will shift the distribution of phytoplankton size

70

towards smaller individuals (Olson et al. 1986; Peter and Sommer 2012). However, this

71

has led to controversy because some studies have also indicated the contrary (Durbin 3

72

1977; Thompson et al. 1992) or shown no clear trends (Yoder 1979; Verity 1981).

73

Indeed, it has also been suggested that other factors, like nutrients, may even play a

74

more relevant role in cell size change than the direct effect of temperature (Peter and

75

Sommer 2012; Deng et al. 2014; Peter and Sommer 2015). However, further research is

76

recommended.

77

The relatively scarce and sometimes contradictory published data make predictions of

78

global warming impacts on marine benthic communities difficult. To investigate the

79

potential interactive effects of warming and the N:P stoichiometry of waters on benthic

80

microalgal species, we selected a stalked benthic diatom, Licmophora colosalis

81

Belando, Aboal and Jiménez that is widely distributed from temperate to subtropical

82

waters (Belando et al. 2016), grows exponentially in summer in the Mar Menor coastal

83

lagoon (Spanish Mediterranean Sea) and covers large areas attached to macrophytes

84

leaves at the bottom. Some Licmophora species are also common fouling components in

85

fish farm netting in Australia (Ravizza and Hallegraeff 2015) and are among the

86

dominant diatoms that constitute the benthic mucilaginous aggregates in the Tyrrhenian

87

Sea (De Philippis et al. 2005).

88

The main aim of the study was to assess the interactive effects of temperature and N:P

89

stoichiometry on populations of a marine stalked-forming diatom, and to evaluate the

90

suitability of stalk overproduction as an indicator of warming and unbalanced N:P

91

stoichiometry. The population dynamics of L. colosalis (cell growth, mortality rate, cell

92

size, biovolume) and stalk and carbohydrate productions were monitored at different

93

N:P ratioss (5, 10, 16, 21, 42) and at three temperatures (26°C, 31°C, 36°C) by

94

considering the temperature variations predicted for global warming at a Mediterranean

95

coastal lagoon. We hypothesised that variations in the N:P ratio could modulate the

4

96

impacts of increasing temperatures, and that phosphorous limitation could induce mass

97

stalks excretion.

98

Material and Methods

99

Habitat and model organism

100

Mar Menor (Murcia, SE Spain) is one of biggest coastal lagoons in the Mediterranean

101

Sea where a fairly high salinity (42-47 ‰ ) and a warm temperature remain almost all

102

year (Pérez-Ruzafa et al. 2005). It is a semi-enclosed system subjected to several

103

pressures deriving from human activities (Velasco et al. 2006), particularly the release

104

of vast quantities of nutrients (mainly nitrates from surrounding agriculture fields) that

105

enhance the N:P stoichiometry of water to go above 16 and resulted in an abrupt

106

increase of chlorophyll a concentration and loss of water transparency (eutrophication

107

event) at the end of 2015 (Pérez-Ruzafa et al. 2019). For this study design, we took the

108

current temperature variations in summer and maximum temperatures predicted in

109

global warming scenarios. Water temperature varies between 10°C in January to 31°C

110

in August, and can undergo variations of 5°C on a daily basis (Terrados 1991).

111

In the lagoon, the colonies of L. colosalis were observed for several years (from 2009 to

112

2015) spreading rapidly and covering almost the entire surface of macrophytes during

113

summer months (later July to August) with temperature ranging from 25 to 31ºC

114

(Belando et al. 2016).

115

Culture conditions and experimental design

116

Monoclonal cultures of L. colosalis were obtained from the natural populations of the

117

Mar Menor coastal lagoon in June 2015 and maintained in Petri dishes (14 cm diameter)

118

to allow growth on the bottom. Initially one diatom cell was isolated to a Petri dish with

5

119

f/2 culture (SAG Göttingen, Germany) using a micropipette under an inverted

120

microscope to ensure monoclonality. Stock cultures were maintained in modified f/2

121

medium at N:P=16, 44 ‰ salinity, pH 8, in an incubation chamber at 25°C with a 16:8

122

light:dark cycle (75 µmol m-2 · s-1 PAR). Cells were harvested in the exponential phase

123

of stock culture (when cells remained isolated and no formation of aggregates or stalks

124

was observed) to be incubated during the different treatments. Cells were harvested by

125

scraping the bottom of the dish. After pellet sedimentation a subsample of 2 ml, with an

126

initial concentration of 698±190 cells · ml-1, was inoculated in 50 ml of the modified f/2

127

medium for each replicate.

128

We conducted factorial short-term nutrient and temperature experiments to test the

129

individual and interactive effects of N:P stoichiometry and the increase in temperature

130

on population dynamics (cell growth, mortality rate, cell size, biovolume), stalk

131

production and carbohydrate accumulation. Nutrient treatments consisted in cultures of

132

the modified f/2 medium with a gradient of N:P ratios, including different levels of

133

nitrogen and phosphorus limitation. Five nutrient treatments were applied: highly

134

nitrogen-limited (HNlim, N:P=5), nitrogen-limited (Nlim, N:P=10), balanced (Bal,

135

N:P=16), phosphorus-limited (Plim, N:P=21) and highly phosphorus-limited (HPlim,

136

N:P=42). To obtain the specific N:P ratio for each treatment, the total nitrogen (NT) and

137

total phosphorous (PT) concentrations of the f/2 medium were modified by using

138

different combinations of NaH2PO4 (1.6-3.03 mg · l-1) and NaNO3 (10.61-67.7 mg · l-1).

139

The NT and PT concentrations were checked at the beginning and the end of the

140

experiment using plasma optical emission spectroscopy (ICP-OES, AGILENT 7500CE)

141

for PT and Elemental Analyzer (SIMAZU) for NT.

142

The five levels of N:P treatments were combined with three temperatures (26, 31 and

143

36°C) in a full factorial design, which resulted in 15 treatment combinations. The 6

144

cultures incubated at 26°C were used as a control by simulating the beginning of

145

summer in coastal lagoons of the area, and two different warm treatments were applied:

146

31°C and 36°C. The 31°C temperature is the maximum value currently observed in the

147

lagoon, and 36°C simulated the 4.5°C increase following the climate global change

148

prediction (De Pascalis et al. 2012). The 31°C treatment consisted in a 5°C increase

149

(from 26°C to 31°C) on the first day of the experiment. For the 36°C treatment, 5°C was

150

increased on the first day (from 26°C to 31°C) and on the third day (from 31°C to

151

36°C). Three incubation chambers were used, one per temperature, and the increase in

152

temperature was gradually applied for 8 h during the light period on each treatment day.

153

Environmental (e.g. light, temperature) conditions inside the chambers were monitored

154

every day, to ensure that temperature was the only factor of variability. At the same

155

time to reduce the effects of minor differences inside the chamber, the position of petri

156

dishes were changed randomly every day. Cultures were checked each day of the

157

experiment, which finished when stalk production and/or cell death was observed. The

158

31°C treatment lasted 9 days. The experiment was finished when stalk production was

159

observed in any nutrient treatment to ensure that this response was related with

160

treatment and not with the limitation of nutrients over time. The 36°C treatment lasted 5

161

days as drastic cell death was observed on day 4 of the experiment. Each N:P treatment

162

for 31 and 36°C were replicated 5 times (5 petri dishes per N:P treatment and per

163

temperature), and the N:P treatments of the control temperature 10 times (10 petri

164

dishes per N:P treatments at 26°C), because the experiment was sequentially removed.

165

The total number of petri dishes was 100.

166

Sampling and analysis

167

Population dynamics analysis: morphometric measures and cell counts.

7

168

At the beginning and end of the experiments, the counts and morphometric

169

measurements of cells and stalks were non-invasively taken and an analysis of at least

170

15 images of 1.21 mm2 per replicate was run. Image acquisition was systematically

171

recorded by an inverted microscope (Eclipse TE2000-U, Nikon Instruments Europe

172

B.V.) and a camera (Nikon DS-5M). To count and measure cells and stalk size, the

173

Image J software was used.

174

For each replicate treatment, all the cells per image were counted with a mean of 431

175

and a maximum of 1,372 cells per replicate. Fewer cells were counted in some cases (a

176

minimum of 100 cells per replicate) when densities were low (36°C treatment). Then

177

cell growth was obtained by:

ℎ=

no. cell − no. cell 1− 0

178

where no. cell was the number of cells · ml-1 and t was the time at which measures were

179

taken. Living and dead cells were also differentiated to calculate the mortality rate (%).

180

Cell dimensions were recorded for all the cells counted in each replicate, and the total

181

biovolume was calculated according to the gomphonemoid shape of the geometric

182

models proposed by Sun and Liu (2003). The length and width of cells were directly

183

measured on the captured images, and thickness (transapical view) was obtained from

184

the scanning electron micrographs (scanning electron microscope, JEOL-6100, Oxford

185

Instrument) of extra material. As the warm treatments were removed at the various

186

experimentation times, two different stalk development stages were distinguished. As

187

the 36°C treatments did not last long, only the initial development stage of stalks (pads)

188

was observed and counted (Fig. 1: a, b). In the 31°C treatments, long stalks with a clear

189

entity were observed, which were counted, and length was measured (Fig. 1c, d). The

8

190

responses related with population dynamics, such as cell growth, mortality and

191

accumulated carbohydrates, were referred to as total cells. As the mortality rate was

192

high in some treatments, the responses related with individuals' state, such as cell size,

193

biovolume and stalk production, referred to the responses of live cells.

194

Carbohydrate analysis.

195

At the end of the experiments, cells were scraped from the bottom of dishes, and the

196

resultant cell suspension was filtered with pre-combusted GF/F filters (for 1 h at

197

500°C). Five filters were used to analyse the total bound carbohydrate per treatment,

198

which included cell particulate carbohydrates and EPS. The concentration was

199

determined following the modified phenol-sulphuric acid method (Pacepavicius et al.,

200

1997). Filters were digested with a mixture of water, phenol 10% and sulphuric acid

201

(1:1:3, v:v:v). Samples were centrifuged and the carbohydrates in solution were

202

spectrophotometrically determined and expressed as pg of glucose equivalents per cell

203

(pg glu-equiv · cel-1).

204

Statistical analysis.

205

The effects of the two warm treatments, 31°C and 36°C, were independently compared

206

with their controls given the differential experimentation time. A two-way ANOVA

207

(two factors: N:P ratio and temperature) was performed to assess the global effect of

208

temperature, nutrient limitation, and the interaction of both factors, on cell size and

209

accumulated carbohydrates. All the other data sets were analysed using GzLMs with a

210

quasi-Poisson distribution of errors (Zuur et al. 2009). If the N:P ratio x temperature

211

interaction was not significant, one-way GzLM or ANOVA analyses were performed

212

with each factor independently. Model checking included visually inspecting the

213

variance-to-mean relationship and linearity of predictors. If the homoscedasticity and 9

214

normality of the model residuals were not met, a more conservative approach was

215

employed using a p-value of 0.01 for the global analysis and post hoc analyses with

216

Bonferroni correction to identify any significant differences among treatments

217

(Underwood 1997; Rutherford 2001). The ANOVAs and GzLMs were performed with

218

the R statistical software (R Core Team 2015).

219

Results

220

The results of the GlzMs and ANOVAs showed that the N:P ratio and temperature

221

significantly modified all the analysed variables (Table 1). The increase in temperature

222

to 31°C significantly modified the responses to the gradient of the N:P ratios

223

(significant interaction), except for cell growth (significant p-values should be <0.01)

224

and stalk production (Fig. 2, Table 1). At 36°C, all the cells died within 24 h and the

225

parameters related with live cells (cell size, biovolume, stalked cells) could not be

226

measured. No interactive effects between both factors were found (p>0.05) at 36°C.

227

Cell growth and accumulated carbohydrates were significantly modified by the N:P

228

ratio and temperature (p<0.05), while the mortality rate was significantly only affected

229

by temperature (p<0.05). Mortality at 36ºC was also confirmed in axenic cultures.

230

Population dynamics: morphometric measurements and cell counts.

231

For all the temperature treatments, cell growth (no. cells · ml-1) was greater in the Bal

232

N:P treatments than under the nutrient-limited conditions (Fig. 2). At 26°C, maximum

233

growth corresponded to the Bal treatment, which significantly reduced in the N-limited

234

(Nlim and HNlim) and HPlim treatments (Table 1). The increase up to 31°C

235

significantly diminished growth in all the nutrient treatments, and the patterns of

236

variation were similar to those found at 26°C (Bal=Plim> HPlim=Nlim=HNlim). At

10

237

36°C, only the cultures grown under the Bal N:P conditions increased in abundance

238

compared to the initial abundance.

239

The two warm treatments significantly increased the percentage of dead cells versus the

240

control at 26°C (Fig. 2, Table 1). The increase up to 31°C induced higher mortality rates

241

in the nutrient-limited treatments (59-83%) than in the Bal treatments (21%, Fig. 2).

242

The increase of temperature up to 36°C led to drastic cell death (100%) after 24 h, while

243

only 4% of the total cells died in its control.

244

The unique significant change in cell size was observed in 31°C treatments under

245

nutrient-limited conditions (Fig. 2, Table 1), that promoted a significant increase in

246

length compared with the control treatment (26°C). All the cells treated with the 36°C

247

or their control (26°C) treatments were similar in length (120.5±4.9 µm long,

248

mean±SD).

249

The highest total biovolume (µm3) was observed in the Bal N:P treatments at 26°C and

250

tended to decrease under the nutrient-limited conditions (Fig. 2), but significantly

251

reduced in the HNlim treatment. Increasing the temperature up to 31°C significantly

252

reduced the total biovolume in all cases compared to the control at 26°C (Table 1).

253

The percentage of cells attached to stalks showed the same response patterns to the N:P

254

treatments at the three analysed temperatures. In general, the percentage of stalked cells

255

was similar in the treatments with N:P values of 16 or below, while the P-limited

256

treatments promoted a significant increase in stalk production. At 26°C, only the

257

HPLim treatment induced a significant increase in the percentage of stalked cells

258

compared to all the other N:P treatments (Fig. 2). The increase up to 31°C significantly

259

stimulated stalk production in all the N:P treatments versus the 26°C treatments, except

260

for the HNlim treatment (Table 1). Besides, both Plim and HPLim treatments led to a 11

261

significantly increase in the percentage of stalked cells compared to all the other N:P

262

treatments at this temperature. Despite drastic cell death occurring at 36°C, we observed

263

numerous stalks in an incipient phase (pads) and found consistent responses during the

264

gradient N:P treatments (Supplementary Figure 1). A higher percentage of cells with

265

pads was also observed in the HPlim treatment (3%) compared to all the other

266

treatments (<1%).

267

Total carbohydrates. The accumulated carbohydrates (CH) concentration showed the

268

same patterns of response at all three temperatures. The CH production increased from

269

the Bal N:P treatments to the nutrient-limited conditions (N and P). At a higher N or P

270

limitation level, more CH accumulated, but significant differences among treatments

271

(Fig. 2, Table 1) were observed only at the 31°C. These patterns were consistent in the

272

cultures incubated at 36°C, but the response was less evident due to the short

273

experimentation time (Supplementary Figure 1).

274

Discussion

275

The results of this study revealed that the effect of temperature on epiphytic diatom

276

populations was modulated by the N:P stoichiometry of water, which highlights the

277

strongly negative effect of global warming with the unbalanced N:P ratios characteristic

278

of eutrophicated waters.

279

Our data suggested that the optimal growth temperature for the model species L.

280

colosalis remained close to 26°C, and that increasing temperature may have deleterious

281

effects, with 36°C being lethal. It seemed to better adapt to environments with an N:P

282

ratios between 16 and 21 as its growth and total biovolume were negatively affected if

283

the N:P ratio went beyond this range. The strongest impacts at growth level were

284

observed under the N-depleted conditions. The nitrate-enriched waters of the Mar 12

285

Menor lagoon, (≈4.23 µmol·l-1 N-NO3 in Ruzafa et al. 2019), may explain L. colosalis

286

blooming and its dominance on benthic communities in early summer months, when

287

temperatures come close to, or are slightly higher than 26°C (≈28.5°C in July).

288

The main effect of the temperature increase from 26°C to 31°C was the decline in

289

abundance and biovolume, and a rise in mortality. These effects were more evident in

290

the unbalanced N:P treatments, which promoted higher mortality rates (58-82%) than

291

under the balanced conditions (20%). These responses could explain why the L.

292

colosalis populations in the Mar Menor coastal lagoon tended to disappear in August

293

when the water temperature reached 31°C. These results also highlighted that

294

continuous heat waves linked to global climate change may alter the populations

295

dynamics of this species, and the structure of benthic communities.

296

The decline in the total biovolume with increasing temperature was consistent with the

297

biomass responses of phytoplankton (Peter and Sommer 2012; 2015) and some benthic

298

diatom communities (Svensson et al. 2014). This response is generally related with both

299

the small cell size of the constituting diatom species, and decreasing abundance. Peter

300

and Sommer (2012) also highlighted that large species can clearly be at disadvantage of

301

coping with heat stress than smaller species, and suggested that the body size of large

302

diatoms could be further reduced. In our study, reduction in biovolume with increasing

303

temperature matched with the decrease in abundance. However, cells of L. colosalis

304

(with a cell size range from129-335µm long in natural populations, in Belando et al.

305

2016) increased in size when rising temperature was combined with nutrient limitation

306

(N and P), while remained similar in size under N:P balanced conditions. The increase

307

in size with warm has also been observed in other diatom taxa (Sommer et al. 2015;

308

Adams et al. 2013). ,Our results support the hypothesis of Adams et al. (2013) that the

309

reduction in body size with warm can not be universally assumed for diatoms, as other 13

310

abiotic factors as ratio N:P of waters may modulate temperature effects. Therefore,

311

generalisations may be premature especially if based in a still fairly small dataset.

312

Overall, the responses related with population dynamics indicated that temperature

313

could have a stronger impact when growth was nutrient-limited. Moreover, L. colosalis

314

accumulated higher amounts of carbohydrates when growth was severely disrupted by

315

both nutrient deprivation and heat stress. It is well-known that nutrient depletion favours

316

the release of extracellular polymeric substances in diatoms (e.g., Myklestad et al. 1972;

317

Myklestad 1995, Obernosterer and Herndl 1995; Alcoverro et al. 2000; Staats et al.

318

2000; Abdullahi et al. 2006). This phenomenon is explained as a result of

319

photosynthetic overflow production when insufficient nutrients are available to sustain

320

cell division, and the level of light promotes photosynthetic activity (Staats et al. 2000).

321

Under these conditions, a slower growth rate means that more carbon would be

322

available and the synthesis of extracellular polymeric substances should be promoted

323

(Staats et al. 2000; Underwood and Paterson 2003). The responses of L. colosalis

324

seemed to follow these patterns as there were always more carbohydrates under the N-

325

or P-limited conditions, which obtained lower growth rates.

326

A 5ºC increase (from 26ºC to 31°C) enhanced carbohydrate production which highlights

327

the strong impact that global warming could have on the overproduction of benthic

328

mucilaginous aggregates. These kinds of aggregates, usually composed of macroalgae

329

and diatoms, have already been observed in some Mediterranean Sea areas in spring-

330

summer months (e.g., De Philippis et al. 2005; Sartoni et al. 2008). L. colosalis did not

331

produce EPS to coat its frustules in sheaths or capsules, as has been indicated for other

332

Licmophora species (De Philippis et al. 2005). We observed that surface occupied by

333

amorphous mucilage attached to the substratum (Fig. 1a,b,c) was maximum at 31°C and

334

under nutrient-depleted conditions (data not shown), which agrees with the 14

335

carbohydrate accumulation patterns. These findings suggest that at least part of the

336

extracellular metabolism overflow production can be excreted this way but further

337

research is required to confirm this.

338

L. colosalis clearly produced the largest amounts of stalks when growth was limited by

339

phosphorus. The link between the number of stalked cells and P-limited growth was

340

consistent at all temperatures, and became even more evident when the growth rate

341

significantly lowered at 31°C. Some studies, which have essentially focused on stalked

342

freshwater species, have already suggested that photosynthetic EPS are channelled in

343

the form of polysaccharide stalk material under unfavourable growth conditions

344

(Perkins 2010; Kilroy and Bothwell 2011). However, as far as we know the nutrient and

345

stalk production relation in marine systems has been investigated only in Licmophora

346

flabellata, and stalk formation in this species was essentially stimulated by high light

347

intensity (233 µmol m-2 s-1), whereas nutrients or temperature influenced neither growth

348

rates nor stalk formation/length (Ravizza and Hallegraeff 2015). These results did not

349

match the responses of L. colosalis and other marine and freshwater stalked-diatoms

350

(Lewis et al. 2002; Perkins 2010; Bothwell and Kilroy 2011).

351

Our findings agree with the stalk formation patterns of Didymosphenia geminata, which

352

blooms in oligotrophic rivers, and stalk overproduction was clearly related with water’s

353

P-depleted conditions (Bothwell and Kilroy 2011; Kilroy and Bothwell 2011; 2012).

354

These proliferations consist mainly in stalk material (Bray 2014), which has been

355

interpreted as an adaptive mechanism to maximise phosphate supply in chronic

356

phosphorus-limited environments (Bray et al. 2017). In this species, phosphatase

357

activity has been localised on or within stalks (Ellwood & Whitton 2007; Aboal et al.

358

2012). Organic phosphorus from the surrounding mucilage matrix is hydrolysed in the

359

tubular structure before being passed to cells in its inorganic forms (Kirkwood et al. 15

360

2007; Aboal et al. 2012). The structure of the stalks of L. colosalis (Fig. 1e) and D.

361

geminata (Aboal et al. 2012) seem similar, but a more in-depth microscopic study is

362

needed. In both species a high percentage of mucilaginous mass consisted in stalk

363

material (Bray 2014) (Fig. 3c, d). It is quite likely that L. colosalis stalks could also play

364

a significant role in adaptation to marine coastal systems with a high N:P stoichiometry

365

of water and even in alkaline phosphatase activity but further research is needed to

366

confirm these hypothesis.

367

Conclusions

368

The effects of rising temperature on L. colosalis populations are modulated by the N:P

369

ratio of water. A 5°C increase would strongly impact the populations living in waters

370

with unbalanced N:P ratios as it could reduce cell growth and promote high mortality

371

rates. The sharp drop in growth under nutrient-limited conditions (N or P) promoted

372

EPS overproduction, and the highest stalk production occurred only under P-limited

373

conditions.

374

The consistency of stalk overproduction at both 26°C and 31°C and the high N:P ratio

375

of water suggest its suitability as a good morphologic indicator of rising temperature,

376

and that this species (and probably also other stalked diatoms) might be proposed for

377

monitoring warming and nutrient contamination in marine transitional systems. At the

378

same time, the use of epiphytic diatoms in biomonitoring should be further explored as

379

they prove useful in shallow lakes (Cejudo-Figuieras et al. 2010). The effects of global

380

warming on the body size of this large benthic microalgae are driven not only by

381

temperature but also by the combined effects with stoichiometry of water, and seem to

382

promote short-term body size enlargement. These observations imply that much more

16

383

research is needed to fully understand the trends in changing sizes of microalgal

384

communities with global climate change.

385

L. colosalis is probably commoner and worldwide distributed than currently known,

386

especially in transitional waters, and has been already reported in two localities in

387

Florida Bay, USA (Little Rabbit Kay and Duck Key) and in a locality from Red Sea

388

(Jeddah), Saudi Arabia (Belando et al. 2016). The mean annual and mean summer

389

temperatures of these three localities are similar to those recorded at Mar Menor

390

(NOAA, National Centre for Environmental Information, www.seatemperature.org) but

391

even when no species dynamics data are available they are quite likely to be similar and

392

may potentially generate similar problems.

393

The current summer temperature in transitional water systems in warm areas are close

394

to the thermal tolerance of L. colosalis and probably to that of other benthic diatoms. If

395

climate change predictions materialise, the population dynamics and seasonal patterns

396

of the species and of whole benthic communities will substantially alter.

397

17

398

Acknowledgments

399

We thank Mª Ángeles Caravaca (Departament of Vegetal Biology-UM), Mª del Mar

400

Santiago and Almudena Gutiérrez (Service of Agroforestal Experimentation, SEAF-

401

UM) for technical support in several steps of this work. Antonia D. Asencio helped in

402

the preliminary phases of the experiments. This text has been proofread by Helen

403

Warburton, (H & A), Valencia, Spain.

404

References

405

Abdullahi, A.S., Underwood, G.J., Gretz, M.R., 2006. Extracellular matrix assembly in

406

diatoms (Bacillariophyceae). V. Environmental effects on polysaccharide synthesis in

407

the model diatom, Phaeodactylum tricornutum. Journal of Phycology 42, 363-378.

408

Aboal, M., Marco, S., Chaves, E., Mulero, I., A. García-Ayala., A.,2012. Ultrastructure

409

and function of stalks of the diatom Didymosphenia geminata. Hydrobiologia, 695,

410

17-24.

411

Adams, G.L., Pichler, D.E., Cox, E.J., O’Gorman, E.J., Seeney, A., Woodward, G.,

412

Reuman, D.C., 2013. Diatoms can be an important exception to temperature–size

413

rules at species and community levels of organization. Global Change Biology 19,

414

3540-3552.

415

Aktan, Y., Topaloğlu, B., 2011. First record of Chrysophaeum taylorii Lewis & Bryan

416

and their benthic mucilaginous aggregates in the Aegean Sea (Eastern Mediterranean).

417

Journal of Black Sea/Mediterranean Environment 17, 159-170.

418

Alcoverro, T., Conte, E., Mazzella, L., 2000. Production of mucilage by the Adriatic

419

epipelic diatom Cylindrotheca closterium (Bacillariophyceae) under nutrient

420

limitation. Journal of Phycology 36, 1087-1095. 18

421

Belando, M.D., Aboal, M., Jiménez, J.F., Marín, A., 2016. Licmophora colosalis sp.

422

nov. (Licmophoraceae, Bacillariophyta), a large epiphytic diatom from coastal waters.

423

Phycologia 55, 393-402.

424

Berthon, V., Bouchez, A., Rimet, F., 2011. Using diatom life-forms and ecological

425

guilds to assess organic pollution and trophic level in rivers: a case study of rivers in

426

south-eastern France. Hydrobiologia 673, 259-271.

427

Bothwell, M.L., Kilroy, C., 2011. Phosphorus limitation of the freshwater benthic

428

diatom Didymosphenia geminata determined by the frequency of dividing

429

cells. Freshwater Biology 56, 565-578.

430 431

Bray, J.P., 2014. The invasion ecology of Didymosphenia geminata. PhD thesis, University of Canterbury, 195 pp.

432

Bray, J., O’Brien, J., Harding, J.S., 2017. Production of phosphatase and extracellular

433

stalks as adaptations to phosphorus limitation in Didymosphenia geminata

434

(Bacillariophyceae). Hydrobiologia 784, 51-63.

435

Cejudo-Figueiras, C., Álvarez-Blanco, I., Bécares, E., Blanco, S., 2010. Epiphytic

436

diatoms and water quality in shallow lakes: the neutral substrate hypothesis

437

revisited. Marine and Freshwater Research 61, 1457-1467.

438

Daniel, G.F., Chamberlain, A.H.L., Jones, E.B.G., 1987. Cytochemical and electron

439

microscopical observations on the adhesive materials of marine fouling diatoms.

440

British Phycological Journal 22, 101-118.

441

Danovaro, R., Umani, S.F., Pusceddu, A., 2009. Climate change and the potential

442

spreading of marine mucilage and microbial pathogens in the Mediterranean Sea.

443

PLoS One 4, e7006. 19

444

De Pascalis, F., Pérez-Ruzafa Y., A., Gilabert, J., Marcos, C., Umgiesser, G., 2012.

445

Climate change response of the Mar Menor coastal lagoon (Spain) using a

446

hydrodynamic finite element model. Estuarine, Coastal and Shelf Science 114, 118-

447

129.

448

De Philippis, R., Faraloni, C., Sili, C., Vincenzini, M., 2005. Populations of

449

exopolysaccharide-producing cyanobacteria and diatoms in the mucilaginous benthic

450

aggregates of the Tyrrhenian Sea (Tuscan Archipelago). Science of the total

451

environment 353, 360-368.

452

Deng, J., Qin, B., Paerl, H.W., Zhang, Y., Wu, P., Ma, J., Chen, Y.,2014. Effects of

453

nutrients, temperature and their interactions on spring phytoplankton community

454

succession in Lake Taihu, China. PloS One 9, e113960.

455

Durbin, E.G., 1977. Studies on the autecology of the marine diatom Thalassiosira

456

nordenskioeldii. II. The influence of cell size on growth rate, and carbon, nitrogen,

457

chlorophyll a and silica content. Journal of Phycology 13, 150-155.

458

Ellwood, N.T.W., Whitton, B.A., 2007. Importance of organic phosphate hydrolyzed in

459

stalks of the lotic diatom Didymosphenia geminata and the possible impact of

460

atmospheric and climatic changes. Hydrobiologia 592, 121-133.

461

Giorgi, F., 2006. Climate change hot‐spots. Geophysical Research Letters 33, L08707.

462

Harley, C.D., Randall Hughes, A., Hultgren, K.M., Miner, B.G., Sorte, C.J., Thornber,

463

C.S., Rodriguez, L.F., Tomanek, L., Williams, S. L., 2006. The impacts of climate

464

change in coastal marine systems. Ecology letters 9, 228-241.

465

IPCC, 2007. Climate Change 2007: Impacts, Adaptation and Vulnerability.

466

Contribution of Working Group II to the Fourth Assessment Report of the 20

467

Intergovernmental Panel on Climate Change. Ed. Parry, M.L., Canziani, O. F.,

468

Palutikof, J.P., van der Linden, P.J., Hanson, C.E., Cambridge University Press,

469

Cambridge, UK.

470

Kilroy, C., Bothwell, M.L., 2011. Environmental control of stalk length in the bloom‐

471

forming,

freshwater

benthic

diatom

472

(Bacillariophyceae). Journal of Phycology 47, 981-989.

Didymosphenia

geminata

473

Kilroy, C., Bothwell, M.L., 2012. Didymosphenia geminata growth rates and bloom

474

formation in relation to ambient dissolved phosphorus concentration. Freshwater

475

Biology 57, 641-653.

476

Kirkwood, A.E., Shea, T., Jackson, L.J., McCauley, E., 2007. Didymosphenia geminata

477

in two Alberta headwater rivers: an emerging invasive species that challenges

478

conventional views on algal bloom development. Canadian Journal of Fisheries and

479

Aquatic Sciences 64, 1703-1709.

480

Lewis, R.J., Johnson, L.M., Hoagland, K.D., 2002. Effects of cell density, temperature,

481

and light intensity on growth and stalk production in the biofouling diatom

482

Achnanthes longipes (Bacillariophyceae). Journal of Phycology 38, 1125-1131.

483

Lloret J.,Marin, A., Marín-Guirao, L., 2008. Is coastal lagoon eutrophication likely to

484

be aggravated by global climate change?. Estuarine Coastal and Shelf Science 78,

485

403-412.

486

Mangialajo, L., Ganzin, N., Accoroni, S., Asnaghi, V., Blanfuné, A., Cabrini, M.,

487

Cattaneo-Vietti, R., Chavanon, F., Chiantore, M., Cohu, S., Costa, E., Fornasaro, D.,

488

Grossel, H., Marco-Miralles, F., Masó, A. Reñé, M., Rossi, A.M., Sala, M.M.,

21

489

Thibaut, T., Totti, C., Vila, M., Lemée, R., 2011. Trends in Ostreopsis proliferation

490

along the Northern Mediterranean coasts. Toxicon 57, 408–20.

491 492

Myklestad, S.M., 1995. Release of extracellular products by phytoplankton with special emphasis on polysaccharides. Science of the total Environment, 165: 155-164.

493

Myklestad, S., Haug, A., Larsen, B., 1972. Production of carbohydrates by the marine

494

diatom Chaetoceros affinis var. willei (Gran) Hustedt. II. Preliminary investigation of

495

the extracellular polysaccharide. Journal of Experimental Marine Biology and

496

Ecology 9, 137-144.

497

Obernosterer, I., Herndl, G.J., 1995. Phytoplankton extracellular release and bacterial

498

growth: dependence on the inorganic N: P ratio. Marine Ecology Progress Series 116,

499

247-257.

500 501

Olson, R.J., Vaulot, D., Chisholm, S.W., 1986. Effects of environmental stresses on the cell cycle of two marine phytoplankton species. Plant Physiology 80, 918-925.

502

Pacepavicius, G., Lau, Y.L., Liu, D., Okamura, H., Aoyama, I., 1997. A rapid

503

biochemical method for estimating biofilm mass. Environmental Toxicology 12, 97-

504

100.

505

Passy, S.I., 2007. Diatom ecological guilds display distinct and predictable behavior

506

along nutrient and disturbance gradients in running waters. Aquatic Botany 86, 171-

507

178.

508

Pérez-Ruzafa, A., Fernández, A.I., Marcos, C., Gilabert, J., Quispe, J.I., García-

509

Charton, J.A. 2005. Spatial and temporal variations of hydrological conditions,

510

nutrients and chlorophyll a in a Mediterranean coastal lagoon (Mar Menor,

511

Spain). Hydrobiologia 550, 11-27. 22

512

Pérez-Ruzafa, A., Campillo, S., Fernández-Palacios, J.M., García-Lacunza, A., García-

513

Oliva, M., Ibañez, H., Navarro-Martínez, P.C., Pérez-Marcos, M., Pérez-Ruzafa, I.M.,

514

Quispe-Becerra, J.I., Sala-Mirete, A., Sánchez, O., Marcos, C., 2019. Long term

515

dynamic in nutrients, chlorophyll a and water quality parameters in a coastal lagoon

516

during a process of eutrophication for decades, a sudden break and a relatively rapid

517

recovery. Frontiers in Marine Science, 6, 26.

518 519

520 521

522

Perkins, K., 2010. Diatom fouling in Tasmanian Hydro Canals. Unpublished PhD thesis, University of Tasmania, Hobart, 118 pp. Peter, K.H., Sommer, U., 2012. Phytoplankton cell size: intra-and interspecific effects of warming and grazing. PloS One 7, e49632. Peter, K.H., Sommer, U., 2015. Interactive effect of warming, nitrogen and phosphorus

523

limitation on phytoplankton cell size. Ecology and evolution 5, 1011-1024.

524

R Core Team., 2015. R: A language and environment for statistical computing.

525

Ravizza, M., Hallegraeff, G., 2015. Environmental conditions influencing growth rate

526

and stalk formation in the estuarine diatom Licmophora flabellata (Carmichael ex

527

Greville) C. Agardh. Diatom Research 30, 197-208.

528 529

Rutherford, A., 2001. Introducing ANOVA and ANCOVA a GLM Approach. SAGE Publications, London, U.K.

530

Sartoni, G., Urbani, R., Sist, P., Berto, D., Nuccio, C., Giani, M., 2008. Benthic

531

mucilaginous aggregates in the Mediterranean Sea: Origin, chemical composition and

532

polysaccharide characterization. Marine Chemistry 111, 184-198.

23

533

Schiaparelli, S., Castellano, M., Povero, P., Sartoni, G., Cattaneo‐Vietti, R., 2007. A

534

benthic mucilage event in North‐Western Mediterranean Sea and its possible

535

relationships with the summer 2003 European heat wave: short term effects on littoral

536

rocky assemblages. Marine Ecology 28, 341-353.

537

Smith, D.J., Underwood, G. J. 2000. The production of extracellular carbohydrates by

538

estuarine benthic diatoms: the effects of growth phase and light and dark treatment.

539

Journal of Phycology 36, 321-333.

540

Sommer, U., Paul, C., Moustaka-Gouni, M., 2015. Warming and Ocean Acidification

541

Effects on Phytoplankton-From Species Shifts to Size Shifts within Species in a

542

Mesocosm

543

Doi.10.1371/journal.pone.0125239.

Experiment.

PLoS

One

10(5),

e=125239.

544

Staats, N., Stal, L.J., Mur, L. R., 2000. Exopolysaccharide production by the epipelic

545

diatom Cylindrotheca closterium: effects of nutrient conditions. Journal of

546

Experimental Marine Biology and Ecology 249, 13-27.

547 548

Sun, J., Liu, D. , 2003. Geometric models for calculating cell biovolume and surface area for phytoplankton. Journal of Plankton Research, 25, 1331-1346.

549

Svensson, F., Norberg, J., Snoeijs, P., 2014. Diatom cell size, coloniality and motility:

550

trade-offs between temperature, salinity and nutrient supply with climate change. PloS

551

One 9, e109993.

552 553

Terrados, J., 1991. Crecimiento y producción de las praderas de macrófitos del Mar Menor, Murcia. PhD thesis, University of Murcia, 229 pp.

24

554

Thompson, P.A., Guo, M.X., Harrison, P.J., 1992. Effects of variation in temperature. I.

555

On the biochemical composition of eight species of marine phytoplankton. Journal of

556

Phycology 28, 481-488.

557

Underwood, A.J., 1997. Experiments in Ecology: Their Logical Design and

558

Interpretation Using Analysis of Variance. Cambridge University Press, Cambridge,

559

U.K.

560

Underwood, G.J., Boulcott, M., Raines, C. A., Waldron, K., 2004. Environmental

561

effects on exopolymer production by marine benthic diatoms: dynamics, changes in

562

composition, and pathways of production. Journal of Phycology 40, 293-304.

563 564

Underwood, G.J.C, Paterson, D.M., 2003. The importance of extracellular carbohydrate production by marine epipelic diatoms. Advanced Botany Research 40,184–240.

565

Velasco, J., Lloret, J., Millán, A., Marín, A., Barahona, J., Abellán, P., Sánchez-

566

Fernández, D., 2006. Nutrient and particulate inputs into the Mar Menor lagoon (SE

567

Spain) from an intensive agricultural watershed. Water, Air, & Soil Pollution 176, 37-

568

56.

569

Verity, P.G., 1981. Effects of temperature, irradiance, and daylength on the marine

570

diatom

Leptocylindrus

danicus

Cleve.

I.

Photosynthesis

and

571

composition. Journal of Experimental Marine Biology and Ecology 55, 79-91.

cellular

572

Yoder, J.A., 1979. Effects of temperature on light-limitation growth and chemical

573

composition of Skeletonema costatum (Bacillariophyceae), Journal of Phycology 15,

574

362-370.

575 576

Zuur, A.F., Ieno, E.N., Walker, N., Saveliev, A.A., Smith, G.M., 2009. Mixed Effects Models and Extensions in Ecology with R. Springer 25

Table captions: Table 1. GzLM and ANOVA results for the effects of the N:P ratio and temperature (Temp.) on all measured parameters. Independent variables

Dependent variables Growth

Mortality

Cell size

Stalked cells

Biovolume

Carbohydrates

<0.001 <0.001 0.044

<0.001 <0.001 <0.001

<0.001 <0.001 <0.001

<0.001 <0.001 n.s.

<0.001 <0.001 <0.001

<0.001 <0.001 <0.001

Chisq/p-value

Chisq/p-value

Chisq/p-value

Main test N:P Temp. N:P X Temp.

Pair wise (factor level) N:P fixed

Temp.

Chisq/p-value

Chisq/p-value

Chisq/p-value

N:P=5

26 °C x 31 °C

47.28/<0.001

187.1/<0.001

31.1/<0.001

n.s.

36.7/<0.001

23.61/<0.001

N:P=10

26 °C x 31 °C

31.77/<0.001

362.9/<0.001

35.6/<0.001

12.1/<0.01

47.4/<0.001

n.s.

N:P=16

26 °C x 31 °C

61.3/<0.001

127.9/<0.001

n.s.

8.1/<0.05

42.1/<0.001

n.s.

N:P=21

26 °C x 31 °C

59.4/<0.001

272.1/<0.001

73.1/<0.001

21.3/<0.001

61.6/<0.001

18.3/<0.001

N:P=42

26 °C x 31 °C

49.85/<0.001

335.5/<0.001

23.3/<0.001

34.9/<0.001

58.4/<0.001

19.9/<0.001

Temp. fixed

N:P ratio

26 °C

16 X 5

46.2/<0.001

n.s.

n.s.

n.s.

12.4/<0.01

n.s

26 °C

16 X 10

65.2/<0.001

n.s.

n.s.

n.s.

n.s.

n.s

26 °C

16 X 21

10.2/<0.05

n.s.

n.s.

n.s.

n.s.

n.s

26 °C

16 X 42

49.3/<0.001

n.s.

n.s.

62.2/<0.001

n.s.

n.s

26 °C

5 x 10

n.s.

n.s.

n.s.

n.s.

n.s.

n.s

26 °C

20 x 41

15.7/<0.01

n.s.

n.s.

n.s.

n.s.

n.s

31 °C

16 X 5

37.6/<0.001

146.6/<0.001

27.1/<0.001

n.s.

13.4/<0.01

65.9/<0.001

31 °C

16 X 10

23.4/<0.001

176.5/<0.001

35.4/<0.001

n.s.

12.4/<0.01

46.9/<0.001

31 °C

16 X 21

17.37/<0.001

184.9/<0.001

64.3/<0.001

12.5/<0.01

n.s.

54.2/<0.001

31 °C

16 X 42

45.4/<0.001

145.8/<0.001

31.2/<0.001

20.1/<0.001

22.22/<0.001

81.7/<0.001

31 °C

5 x 10

n.s.

n.s.

n.s.

n.s.

n.s.

n.s

10.1/0.046

n.s.

n.s.

10.7/<0.05

n.s

31 °C 20 x 41 not significant=n.s. (p>0.05)

Figure captions:

Figure 1. Micrographs of the cultures of L. colosalis. a, b) An amorphous mucilage and initial stage of the stalks. c) Amorphous mucilage and stalks, d) Colony attached to the substratum by numerous long stalks, e) Detail of an upper part of the stalk.

Figure 2. L. colosalis responses to the gradient of N:P ratios at 26ºC (boxplot in white) and 31°C (boxplot in grey). Capital letters indicate significant differences among N:P treatments at 31°C, and regular letters at 26°C.

Figure 3. 1Macroscopic and microscopic views of the epiphytic mats of Licmophora colosalis in summer months at the Mar Menor coastal lagoon. a) Macroscopic colonies on the leaves of Cymodocea nodosa in August 2015. b) Macroscopic proliferations of L. colosalis upon Acetabularia sp. in June 2015. c) Macroscopic arborescent colonies attached to slides in July 2008 with binocular stereomicroscope d) Detail of the colonies and the large L. colosalis stalks with light microscope.

Highlights -Diatoms in transitional waters are modulated by temperature and N: P ratios. -A 5ºC increase with unbalanced N:P ratios enhance mortality rates. -Unbalanced nutrient ratios promote EPS overproduction. -The highest stalk production occurred under P-limited conditions. -Stalks are useful morphologic indicators of rising temperature and high N:P ratios.